Structure and morphology of catalysts produced via oxidation of Mo-Pt alloys

Article abstract of DOI:10.1134/S0020168509030091

A novel approach has been proposed for producing platinum-containing catalysts via oxidation of molybdenum-platinum powders. The effect of alloy composition on the structure and morphology of the oxidation products has been studied by a variety of physico

Full text of DOI:10.1134/S0020168509030091

ISSN 0020-1685, Inorganic Materials, 2009, Vol. 45, No. 3, pp. 264–270. © Pleiades Publishing, Ltd., 2009.
Original Russian Text © S.N. Nesterenko, L.L. Meshkov, P.A. Zosimova, N.S. Nesterenko, 2009, published in Neorganicheskie Materialy, 2009, Vol. 45, No. 3, pp. 306–312.
Structure and Morphology of Catalysts Produced
via Oxidation of Mo–Pt Alloys
S. N. Nesterenko, L. L. Meshkov, P. A. Zosimova, and N. S. Nesterenko
Faculty of Chemistry, Moscow State University, Moscow, 119899 Russia
e-mail: snnesterenko@general.chem.msu.ru
Received June 25, 2008
Abstract—A novel approach has been proposed for producing platinum-containing catalysts via oxidation of
molybdenum–platinum powders. The effect of alloy composition on the structure and morphology of the oxi-
dation products has been studied by a variety of physicochemical methods, and heat-treatment conditions have
been optimized to achieve a uniform distribution of platinum nanoparticles in the resulting molybdenum oxide.
Selective oxidation has been shown to be a viable approach to producing platinum–transition metal oxide cat-
alytic systems.
DOI: 10.1134/S0020168509030091
INTRODUCTION
ultrasonic processor (discrete power levels from 20 to
50 W) and mechanical stirrer.
There is considerable scientific and technological
interest in developing novel approaches for the prepara-
tion of modified Pt-containing catalysts offering
enhanced stability to traditional catalyst poisons [1].
Selective oxidation of Pt-containing alloys makes it
possible to produce catalysts with a small particle size
of the active phase, stable to sintering [2]. The Pt–MoO3
system, stable in the presence of sulfur-containing
compounds, was chosen to study selective oxidation
processes [3, 4].
Heat treatment. The alloys were equilibrated by
annealing at 900 5°ë for 400 h in silica ampules
which were pumped to ~10–2 Pa and mounted in tubular
resistance-heated furnaces.
Characterization techniques. The alloys were
characterized by a variety of physicochemical methods.
X-ray diffraction (XRD) patterns were collected on
a DRON-4 diffractometer (diffracted-beam graphite
monochromator, EXPRESS software, CuKα radiation,
step-scan mode with a step of 0.1° and a counting time
per data point of 3–5 s). The data were indexed and ana-
lyzed using RIETAN-94 [5] and STOE WinXPOW [6]
(Table 2).
The purpose of this work was to optimize proce-
dures for the synthesis of modified catalysts and to
study the relationship between their microstructure and
physicochemical properties.
The crystallite size was evaluated from the width of
diffraction line profiles using standard EXPRESS soft-
ware and the Scherrer equation: D = 0.94α/(βcos
θ)where D is the average crystallite size, α is the x-ray
wavelength, β is the full width at half maximum of the
XRD peak, and θ is the diffraction angle. The error of
determination was within 10%.
EXPERIMENTAL
Alloy preparation. Alloys for this investigation
were prepared from platinum plates (99.99% purity)
and carbothermal molybdenum (99.9%). Weighed sam-
ples (3 g) containing 2, 10, or 30 at % Pt were melted in
an electric arc furnace using a nonconsumable tungsten
electrode under a high purity argon atmosphere at a pres-
sure of 1.1 × 105 Pa. The alloys were remelted four times
for homogenization. The alloy compositions and melting
losses were determined gravimetrically (Table 1).
Thermogravimetric analysis (TGA) was carried out
with an SDT-Q600 system using powder samples (on
the order of 0.5 g) prepared from the alloys. TGA data
Powder samples were prepared by an abrasive pro-
cess. Particle size analyses were performed with a
CILAS 1180 laser analyzer, which insured particle size
determination in the range 40 nm to 2.5 mm. The
immersion liquid was selected so that the powder to be
analyzed was evenly dispersed without dissolution. Iso-
propanol was used as a surfactant to prevent particle
aggregation. The samples were dispersed using an
Table 1. Compositions of the synthesized molybdenum–
platinum alloys
Mo, at %
Pt, at %
Mo, wt %
Pt, wt %
98
90
70
2
10
30
96
4
81.6
53.4
18.4
46.6
264

STRUCTURE AND MORPHOLOGY OF CATALYSTS PRODUCED VIA OXIDATION
265
Q3, %
100
Q3, %
100
Q3,%
[+10]
Q3,%
[+10]
80
60
40
20
0
80
60
40
20
0
0.01 0.1
1.0
10.0
Particle size, µm
100.0 1000.0
2500.0
0.01 0.1
1.0
10.0
Particle size, µm
100.0
1000.0
2500.0
Fig. 1. Particle size distribution in the Mo Pt powder.
Fig. 2. Particle size distribution in the Mo Pt powder.
90 10
98
2
were analyzed using the software package supplied persion procedure used in this study gives irregularly
with the instrument [7].
shaped particles of two types: plates and polyhedra.
X-ray microanalysis was performed on a Leo Supra
50 VP scanning electron microscope equipped with an
energy-dispersive x-ray spectrometer. The sampling
volume was 0.1–0.3 µm3, the absolute local sensitivity
10–15 to 10–12 g, and the error of determination 2% [8].
Microstructures of the as-prepared and oxidized alloys
were examined by scanning electron microscopy
(SEM) in secondary electron mode at magnifications of
800x and 25000x. Powder specimens were cemented
to a conductive carbon substrate. The average size and
size distribution of platinum particles were determined
by transmission electron microscopy (TEM) on a JEOL
3000F operated at an accelerating voltage of 300 kV.
Figures 7–10 present TGA results on the oxidation
kinetics of the powdered alloys. The powders were
found to oxidize much more rapidly compared to bulk
samples [9]. For comparison, we also examined the
oxidation of pure molybdenum powder prepared by the
same procedure, including remelting.
The oxidation rate of Mo90Pt10 considerably
exceeded that of the other alloys in all of the TGA runs.
Note that, over the entire temperature range examined,
the early-stage oxidation of the alloys proceeds faster
than that of molybdenum, with the largest difference at
a platinum content of 2 at %.At 500 and 550°C, the oxi-
dation rate of the alloys decreases slightly with increas-
ing platinum content (Figs. 8, 9). In addition to the
reduced oxidation rate, a distinctive feature of the
RESULTS AND DISCUSSION
In studies of the oxidation of powder samples, infor-
mation about the size and shape of the particles is of
key importance. Figures 1–3 show the particle size dis-
tributions of the powders prepared from the Mo–Pt
alloys, in the form of histograms (differential particle
size distributions) and cumulative curves. The histo-
grams present the volume fraction of each size range,
equal to the weight fraction when all the particles have
the same density, which is represented by the q3 param-
eter: the percentage of particles in a given size range.
The cumulative curves present the percentage of parti-
cles whose diameter does not exceed the upper limit of
a given size range, Q3.
These data demonstrate that the average particle size
of the powders increases with platinum content
(Figs. 1–3, Table 3).
The SEM results for the powders (Figs. 4–6) are
consistent with the above particle size distributions.
Molybdenum and the Mo–Pt alloys studied here are
characterized by high hardness and brittleness. The dis-
Q3, %
100
Q3, %
[+10]
80
60
40
20
0
0.01 0.1
1.0
10.0
Particle size, µm
100.0
1000.0
2500.0
Fig. 3. Particle size distribution in the Mo Pt powder.
70 30
INORGANIC MATERIALS Vol. 45 No. 3 2009

266
NESTERENKO et al.
Table 2. Phase composition of the synthesized molybdenum–platinum alloys
Alloy
Phase composition
(Mo) = 100
‡, nm
Ò, nm
Mo98Pt2
0.3143 0.0001
–
Mo90Pt10
(Mo) : Mo3Pt2 = 84 : 16
(Mo) : Mo3Pt2 = 43 : 57*
0.3133 0.0002
0.2793 0.0004
–
0.4401 0.0009
Mo70Pt30
0.31549 0.001
0.2780 0.0013
–
0.448 0.002
Note: * Mo3Pt2 = high-temperature phase.
450°C oxidation is that the above trend is broken down.
The XRD data for the Mo70Pt30 alloy oxidized in air
during steplike heating with 20-min isothermal holds at
each temperature (Fig. 11) indicate that the formation
of molybdenum oxides begins at 500°ë. With increas-
ing temperature, the process speeds up, leading to com-
plete structural degradation of the alloy. Increasing the
percentage of the active phase in the alloy or the oxida-
tion temperature increases the intensity of the XRD
peaks from metallic platinum (Fig. 12). The optimal
selective oxidation temperature is 550°ë. The upper
limit of heat treatment is imposed by the significant
åÓé3 volatility starting above 600°ë.
In particular, Mo90Pt10 has the slowest oxidation rate.
Low-temperature nitrogen adsorption measure-
ments showed that the specific surface area of the pow-
ders ranged from 2.0 to 2.5 m2/g. Thus, one can evalu-
ate the oxidation rate constant for our powder samples.
The plots of ∆m/S versus τ1/2 in Fig. 10 demonstrate that
the early stage oxidation kinetics of the powders are
well represented by a parabolic rate law.
Oxidation was run until full conversion of the
molybdenum to oxides. Conversion was evaluated rel-
ative to the weight gain corresponding to the oxidation
of all the molybdenum. The weight gain depended on
the alloy composition and oxide stoichiometry:
SEM results demonstrate that oxidation at 550°ë is
accompanied by significant morphological changes
(Fig. 5). At low Mo contents (≤70%), the alloy particles
do not fully break down, and oxidation leads to the for-
mation of needlelike åÓé3 crystals up to 40 µm in
length on the particle surface (Fig. 6). At Mo contents
above 90%, oxidation causes complete structural deg-
radation due to the increase in sample volume upon
MoO3 formation.
Mo70Pt30 Mo90Pt10 Mo98Pt2
Mo
MoO3
MoO2
26.7%
17.8%
40.8%
27.2%
48%
32%
50%
33.3%
At 550°ë, complete oxidation of the Mo70Pt30 and
Mo90Pt10 powders took 100 min, and the weight gain
was 24 and 41%, respectively (Fig. 9). At 500°ë, these
According to XRD results, the oxidized powders
levels were reached in about 200 min. At 450°ë, com- consisted of two phases: orthorhombic MoO3 and
plete oxidation took a much longer time: even after 24 h
of oxidation, the weight gain (ꢀ11%) in the alloy con-
taining 2 at % platinum was below the theoretical level.
metallic platinum. None of the other molybdenum oxides
described in the literature [10] were detected in this study.
It seems likely that they only appear in intermediate stages
of oxidation. The XRD patterns in Fig. 12 illustrate the
effect of platinum content on the phase composition of
the oxidized powders. The lattice parameters of plati-
num and molybdenum trioxide in our samples agree
well with those reported in the literature (Table 4).
Table 3. Particle size distributions in the Mo–Pt powders
(µm)
Alloy
Mo98Pt2
Mo90Pt10
Mo70Pt30
The average sizes of the platinum particles (coher-
ently scattering domains) evaluated from the XRD data
for the oxidized powders using the Scherrer equation
indicate that the metal-oxide matrix composites pro-
duced in this study contain nanoparticles less than
50 nm in size (Table 5). The present results lead us to
Q3 = 10%
Q3 = 50%
Q3 = 90%
13.4
41.0
77.9
21.6
50.8
95.2
25.2
62.6
138.5
INORGANIC MATERIALS Vol. 45 No. 3 2009

STRUCTURE AND MORPHOLOGY OF CATALYSTS PRODUCED VIA OXIDATION
267
30 µm
30 µm
Fig. 5. Microstructure of the Mo Pt powder oxidized at
550°ë.
70 30
Fig. 4. Microstructure of the as-prepared Mo Pt powder.
70 30
m, %
130
125
120
115
110
105
100
4
1
2
3
1 Mo Pt
70 30
Mo Pt
2
90 10
3 Mo Pt
98
2
Mo
4
0
100 200 300 400 500 600 700 800 900
3 µm
Time, min
Fig. 6. Microstructure of the oxide scale on the Mo Pt
powder oxidized at 550°ë.
Fig. 7. Oxidation kinetic curves for the Mo–Pt powders at
450°ë in air.
70 30
Table 5. Effect of oxidation temperature on the crystallite
size of Pt in the oxidized powders
Table 4. Lattice parameters of the phases present in the al-
loys oxidized at 550°C
Particle size, nm
Alloy
Mo60Pt40 Mo70Pt30 Mo90Pt10 Mo98Pt2
toxidation, °C
Mo60Pt40 Mo70Pt30 Mo90Pt10 Mo98Pt2
Pt a, Å 3.9204(14) 3.9133(3) 3.9196(10) 3.917(6)
MoO3 a, Å 13.918(8) 13.916(7) 13.921(3) 13.920(2)
b, Å 3.699(5) 3.694(2) 3.701(5) 3.700(4)
c, Å 3.961(3) 3.960(2) 3.958(8) 3.968(5)
400
450
500
550
12
19
24
–
16
18
23
37
16
22
28
31
18
14
27
–
INORGANIC MATERIALS Vol. 45 No. 3 2009

268
m, %
NESTERENKO et al.
∆m/s, g/Òm
2
150
145
140
135
130
125
120
115
110
105
100
95
0.16
0.14
0.12
0.10
0.08
0.06
0.04
0.02
(a)
Mo Pt
4
3
1
2
3
70 30
Mo Pt
90 10
3
2
Mo Pt
98
2
1
2
1
4 Mo
1 Mo Pt
4
70 30
Mo Pt
2
3
90 10
Mo Pt
98
2
4 Mo
6
7
8
9
10
1/2
1/2
min
τ ,
0
100 200 300 400 500 600 700 800
2
∆m/s,g/Òm
Time, min
0.42
0.40
0.38
0.36
0.34
0.32
0.30
3
(b)
Fig. 8. Oxidation kinetic curves for the Mo–Pt powders at
500°ë in air.
1
2
3
Mo Pt
0.28
0.26
0.24
0.22
0.20
0.18
0.16
0.14
0.12
0.10
0.08
70 30
Mo Pt
90 10
1
Mo Pt
98
2
m, %
4 Mo
2
155
150
145
140
135
130
125
120
115
110
105
100
5
4
6
7
8
9
10
τ1/2, minÌ
1/2
3
2
2
∆m/s, g/Òm
1 Mo Pt
2 Mo Pt
Mo Pt
Mo Pt
1
70 30
0.36
0.34
0.32
0.30
(c)
90 10
3
2
3
4
5
98
2
4
1
2
3
Mo Pt
70 30
0.28
0.26
0.24
0.22
0.20
0.18
0.16
0.14
0.12
0.10
0.08
0.06
0.04
60 40
Mo Pt
90 10
Mo
Mo Pt
98
2
4 Mo Pt
60 40
Mo
0
100 200 300 400 500 600 700 800
1
5
5
Time, min
4
Fig. 9. Oxidation kinetic curves for the Mo–Pt powders at
550°ë in air.
3.0
3.5
4.0
4.5
5.0
τ1/2, min
1/2
conclude that the size of the Pt particles in the alloys is
a weak function of Pt content but strongly depends on
oxidation conditions. For example, reducing the oxida-
tion temperature from 550 to 450°C markedly reduces
the average size of the Pt particles in oxidized Mo70Pt30
and Mo90Pt10. The likely reason for this is the high-tem-
perature particle migration over the surface of molyb-
denum trioxide and cluster growth.
Fig. 10. Parabolic plots for the oxidation of the Mo–Pt pow-
ders at (a) 450, (b) 500 and (c) 550°C in air.
particles differed in shape, which might lead to selec-
tive oxidation by different mechanisms.
The present results demonstrate that selective oxida-
tion of molybdenum–platinum powders is a viable
approach to producing fine-particle platinum-contain-
ing oxide catalysts. Optimizing oxidation conditions,
one can produce Pt-containing catalytic systems with
a uniform distribution of the active phase, stable to
TEM results (Figs. 13, 14) correlate with the plati-
num particle sizes inferred from XRD data. As seen in
Fig. 13, platinum particles are present not only on the
surface but also in the interior of molybdenum trioxide
crystals. One possible reason for this is that the alloy sintering.
INORGANIC MATERIALS Vol. 45 No. 3 2009

STRUCTURE AND MORPHOLOGY OF CATALYSTS PRODUCED VIA OXIDATION
269
Pt
t, °C
600
500
400
300
200
100
25
30
40
50
60
70
80
2θ, deg
Fig. 11. In situ XRD data for the Mo Pt alloy oxidized in air during steplike heating.
70 30
I
MoO
3
Pt
Pt
Pt
Pt
3
2
1
20
28
36
44
52
60
68
76
84
92
2θ, deg
Fig. 12. XRD patterns of the (1) Mo Pt , (2) Mo Pt , and (3) Mo Pt powders oxidized at 550°ë.
98
2
90 10
70 30
INORGANIC MATERIALS Vol. 45 No. 3 2009

270
NESTERENKO et al.
CONCLUSIONS
We studied selective oxidation of molybdenum–
platinum alloys as a novel approach to producing
Pt−åÓé3 catalysts.
The structural and morphological features of the
distribution of fine platinum particles in the resultant
oxide matrix were investigated by TGA, XRD, and
SEM.
The results indicate that the optimal oxidation tem-
perature lies in the range 500–550°ë.
REFERENCES
1. Ponec, V., Alloy Catalysts: The Concepts, Appl. Catal.,
A, 2001, vol. 222, pp. 31–45.
2. Sanz, N., Lomello-Tafin, M., Valmalette, J.-C., et al.,
Preparation and Characterization of Au–ZrO2 Nanopar-
ticles Obtained by Oxidation of Zr–Au Alloy, Mater. Sci.
Eng., C, 2002, vol. 19, pp. 79–83.
100 nm
3. Hardacre, C., Rayment, T., and Lambert, R.M., Plati-
num/Ceria CO Oxidation Catalysts Derived from Pt/Ce
Crystalline Alloy Precursors, J. Catal., 1996, vol. 158,
pp. 102–108.
Fig. 13. Platinum distribution in the åÓ Pt alloy oxi-
dized at 550°ë.
70 30
4. Jennings, J.R., Lambert, R.M., Nix, R.M., et al., Novel
Methanol Synthesis Catalysts Derived from Intermetal-
lic Precursors: CO2 Poisoning and Molecular Mecha-
nism of the Synthesis Reaction, App. Catal., 1989, vol.
50, pp. 157–170.
5. Izumi, F., The Rietveld Method, Young, R.A., Ed.,
Oxford: Oxford Univ. Press, 1993, chapter 13.
6. STOE WinXPOW, Version 1.06, STOE & Cie, 1999.
7. Universal Analysis for Windows 2000/XP, Version 4.1D,
Build 4.1.0.16, Waters LLC, 1998–2004.
8. Gorelik, S.S., Skakov, Yu.A., and Rastorguev, L.N.,
Rentgenograficheskii i elektronno-opticheskii analiz
(X-Ray Diffraction and Electron-Optical Analysis),
Moscow: Mosk. Inst. Stali i Splavov, 2002.
9. Nesterenko, S.N., Meshkov, L.L., Zosimova, P.A., and
Nesterenko, N.S., Composition and Structure of the Oxi-
dation Products of Mo–Pt Alloys, Neorg. Mater., 2009,
in press.
10. Massalski, T.B., Okamoto, H., and Subramanian, P.R.,
Binary Alloy Phase Diagrams, Materials Park: ASM
International, 1990, 2nd ed, vol. 2.
10 nm
Fig. 14. Individual platinum particle in oxidized Mo Pt .
70 30
INORGANIC MATERIALS Vol. 45 No. 3 2009