Remarkable effect of iron-substitution for molybdenum in phosphododecamolybdate on oxidative dehydrogenation of 2-propanol

Article abstract of DOI:10.1246/cl.2001.28

Cesium hydrogen salt of mono-iron-substituted phospho-undecamolybdate, Cs_2.8H_1.2PMo₁₁FeO₃₉, was synthesized and the catalyst showed intrinsically higher selectivity to acetone for the title reaction compared with the iron-impregnated Fe³⁺(1.2-5.0 wt%)Cs_3.0PMo₁₂O₄₀.

Full text of DOI:10.1246/cl.2001.28

28
Chemistry Letters 2001
Remarkable Effect of Iron-Substitution for Molybdenum in Phosphododecamolybdate
on Oxidative Dehydrogenation of 2-Propanol
Joon-Seok Min, Makoto Misono, Akira Taguchi, and Noritaka Mizuno*
Department of Applied Chemistry, School of Engineering,The University of Tokyo, Hongo, Bunkyo-ku, Tokyo 113-8656
(Received October 10, 2000; CL-000913)
Cesium hydrogen salt of mono-iron-substituted phospho-
was added to the solution and the solution was filtered. To the
reddish yellow filtrate was added cesium chloride (6.0 mmol) to
yield a yellow precipitate. The precipitate was filtered off,
washed with water, and aspirated to dryness (yield 55%, yellow).
Elemental analysis: Found: Cs, 16.43; P, 1.34; Mo, 46.37; Fe,
2.03%. Calcd for Cs2.8H1.2[PMo11{Fe(H2O)}O39]·6H2O: Cs,
16.43; P, 1.37; Mo, 46.58; Fe, 2.46%. No changes in the anion
structure were observed by IR and TG/DTA analyses after the
thermal treatment at 473 K and after the reaction at the same
temperature. More characterization was carried out for the
phenyltrimethylammonium salt since Cs2.8H1.2PMo11FeO39 was
insoluble in any solvents without the decomposition of the anion
structure. Phenyltrimethylammonium salt of mono-iron-substi-
tuted phosphoundecamolybdate was prepared by adding
phenyltrimethylammonium chloride to the acetonitrile solution of
tetra-n-butylammonium salt.21 Elemental analysis: Found: C,
22.05; H, 2.96; N, 3.52; P, 1.20; Mo, 41.50; Fe, 2.15; Cl, 1.35%.
Calcd for [C6H5(CH3)3N]5[PMo11{Fe(Cl)}O39]·1CH3CN·1H2O:
C, 22.21; H, 2.98; N, 3.31; P, 1.22; Mo, 41.52; Fe, 2.20; Cl,
1.39%. The UV–vis spectrum in v/v 7/3 acetonitrile/dimethyl-
sulfoxide at 296 K showed a broad absorption bands at 307 nm
(ε 29000 M–1cm–1) characteristic of O→Mo charge transfer
bands of α-Keggin structure.22 No signal was observed for the
31P NMR spectrum in acetonitrile at 295 K because of the pres-
ence of P–O–Fe3+ (paramagnetic species) bond in the polyoxo-
metalate structure in agreement with reports.23,24 The cesium
salt of H3PMo12O40 (Cs3.0PMo12O40) was prepared with aque-
ous solution of cesium carbonate and H3PMo12O40 as reported
previously.17 Fe3+(1.2–5.0 wt%)/Cs3.0PMo12O40 was prepared
by impregnating Cs3.0PMo12O40 with an aqueous solution of
ferric nitrate. The iron content of 2.5 wt% corresponded to that
in Cs2.8H1.2PMo11FeO39. It was confirmed by IR and XRD
spectroscopy that the primary and secondary structures of phos-
phododecamolybdate were kept before and after the impregna-
tion.
Reactions were carried out with a flow reactor (Pyrex tube,
10 mm internal diameter) at 453 K under atmospheric pressure.
The feed gas consisted of 17 vol% of 2-propanol, 35 vol% of
O2, and N2 balance. Total flow rates were 24 cm3·min–1. All
the flow lines were heated at 393 K to prevent adsorption of 2-
propanol and products. Prior to the reaction, 0.8 g of each cata-
lyst was treated in an O2 stream (48 cm3·min–1) for 1 h at 473
K. The gases at the outlet of the reactor were taken out inter-
mittently with the aid of sampler directly connected to the sys-
tem and analyzed by a gas chromatograph with FFAP (1.0 m),
Porapak-Q (4.0 m), and Molecular Sieve 5A (1.2 m) columns.
Selectivities were fractions of the sum of the products and cal-
culated on the C3(2-propanol)-basis. Carbon balance was usu-
ally more than 90%. The conversion and selectivity for the
oxidative dehydrogenation of 2-propanol catalyzed by
Cs2.8H1.2PMo11FeO39 reached an almost constant level after 2 h;
undecamolybdate, Cs2.8H1.2PMo11FeO39, was synthesized and
the catalyst showed intrinsically higher selectivity to acetone
for the title reaction compared with the iron-impregnated
Fe3+(1.2–5.0 wt%)/Cs3.0PMo12O40.
Polyoxometalates provide a good basis for the molecular
design of mixed oxide catalysts because the redox and acidic
properties can be controlled at atomic/molecular levels by
changing the constituent elements. Among polyoxometalates,
the catalysis by Keggin-type anions has most extensively been
investigated because of their rather high thermal stabilities and
ease of synthesis.1–7 During the past decade, there has been a
great interest in the synthesis and oxidation catalysis of transi-
tion metal-substituted polyoxometalates due to their unique and
remarkable influences on the catalytic performances.3,6,8–10 For
the synthesis of transition metal-substituted polyoxometalates,
lacunary (or defect) polyoxometalates are important synthetic
intermediates, and the vacancy provides a pentadentate ligand
for the transition metals. It has been reported that the transition
metals play significant roles in the redox processes as reservoirs
for electrons and active sites for the activation of hydrocarbon
and molecular oxygen.11–16 It has recently been reported that
the addition (“not substitution”) of Fe enhanced the catalytic
performance of Cs2.5H0.5PMo12O40 or Cs2.5H1.5PVMo11O40 for
the selective oxidation of light alkanes11,12,17 and that Fe3+ pro-
moted the H abstraction from alkanes by O2– neighboring
Mo6+.17 It is expected that the introduction of Fe3+ into the
polyoxometalate can result in a large enhancement of catalytic
performance. Therefore, the synthesis of iron-substituted
molybdophosphate and the demonstration of the effectiveness
of the Fe3+–O–Mo6+ for the selective oxidation are important
from the standpoint of the development of heterogeneous cata-
lysts. However, little is known of the oxidation catalysis by
iron-substituted molybdophosphates because there is no reli-
able, reproducible, and selective method of their synthesis.18–20
In this work, we report the synthesis of mono-iron-substituted
phosphoundecamolybdate and demonstrate the remarkable
effect of the iron-substitution on the selectivity for the oxidative
dehydrogenation of 2-propanol.
Cesium hydrogen salt of mono-iron-substituted phospho-
undecamolybdate was synthesized as follows. Phosphododeca-
molybdic acid (H3PMo12O40) was commercially obtained from
Nippon Inorganic Color and Chemical Co., Ltd. and recrystal-
lized from water. H3PMo12O40·28H2O (2.0 mmol) was dis-
solved in water (0.1 mol·dm–3), and the pH of the solution was
adjusted to 4.3 with lithium carbonate. Only one signal was
observed at –0.9 ppm (vs 85% H3PO4) for the 31P NMR spec-
trum, showing the presence of highly pure lacunary species,
PMo11O397–. Then, ferric nitrate Fe(NO3)3·9H2O (2.2 mmol)
Copyright © 2001 The Chemical Society of Japan

Chemistry Letters 2001
29
the conversions were 15, 14, 14, and 14% after 0.5, 1.0, 1.5,
and 2.0 h, respectively. The products were acetone, propene,
di-isopropyl ether (DIPE), and CO2.
pared with the traditional impregnation of iron. It was con-
firmed that Cs2.8H1.2PMo11FeO39 showed higher selectivity to
methacrolein than Cs3.0PMo12O40 for the oxidation of isobu-
tane. Further experiments of oxidation of isobutane are in
progress and will be reported in due course. The strategy of the
catalyst design and synthesis would be very important for the
development of new solid catalysts.
This work was supported in part by a Grant-in-Aid for
Scientific Research from the Ministry of Education, Science,
Sports and Culture of Japan. We thank Prof. Masato
Hashimoto (Wakayama University) for the crystallographic
analysis.
References and Notes
Table 1 summarizes the results for oxidative dehydrogena-
1
2
3
I. V. Kozhevnikov, Chem. Rev., 98, 171 (1998).
N. Mizuno and M. Misono, Chem. Rev., 98, 199 (1998).
T. Okuhara, N. Mizuno, and M. Misono, Adv. Catal., 41, 113
(1996).
tion of 2-propanol catalyzed by Cs2.8H1.2PMo11FeO39
,
Fe3+(2.5wt%)/Cs3.0PMo12O40, and Cs3.0PMo12O40 at 453 K.
The respective selectivities to acetone were 80, 38, and 16%, and
that observed for iron-substituted Cs2.8H1.2PMo11FeO39 was much
higher than that of iron-impregnated Fe3+(2.5wt%)/
Cs3.0PMo12O40, while the activities were close to each other. The
acid amounts are also shown in Table 1 and decreased in the
same order as that of selectivities to acetone. This fact shows
that the highest selectivity to acetone for Cs2.8H1.2PMo11FeO39
does not result from the suppression of the production of
propene and DIPE catalyzed by acidic sites, but the iron-substi-
tution. Figure 1 shows changes in selectivities with percent
conversion of 2-propanol. The conversion was changed by
changing W/F, where W is the weight of catalyst and F is the
total flow rate. The selectivity to acetone almost unchanged
with an increase in conversion for Cs2.8H1.2PMo11FeO39.
Similar dependencies of selectivities to propene and CO2 on
conversions were observed. The selectivities to acetone extrap-
olated to 0% conversion for Cs2.8H1.2PMo11FeO39 and Fe3+(2.5
wt%)/Cs3.0PMo12O40 catalysts were ca. 73% and 30%, respec-
tively. These facts show that the selectivity to acetone for
Cs2.8H1.2PMo11FeO39 is intrinsically much higher than that of
Fe3+(2.5 wt%)/Cs3.0PMo12O40. The selectivities to acetone
were in the range of 26–38% and did not much increase when
amounts of iron loaded changed in the range of 1.2–5.0 wt%.
Thus, it is clearly demonstrated that the incorporation of iron
into molybdophosphate much more increases the selectivity for
the oxidative dehydrogenation of 2-propanol (to acetone) com-
4
5
I. V. Kozhevnikov, Catal. Rev. Sci. Eng., 37, 311 (1995).
Y. Ono, in “Perspectives in Catalysis,”, ed. by J. M. Thomas
and K. I. Zamaraev, Blackwell, London (1992), p. 431.
M. T. Pope and A. Müller, Angew. Chem., Int. Ed. Engl., 30,
34 (1991).
6
7
8
9
M. Misono, Catal. Rev. Sci. Eng., 29, 269 (1987).
R. Neumann, Prog. Inorg. Chem., 47, 317 (1998).
C. L. Hill and C. M. Prosser-McCartha, Coord. Chem. Rev.,
143, 407 (1995).
10 M. T. Pope, in “Heteropoly and Isopoly Oxometalates,” ed.
by M. T. Pope, Springer-Verlag, Berlin (1983), p. 93.
11 M. Langpape, J. M. M. Millet, U. S. Ozkan, and M.
Boudeulle, J. Catal., 181, 80 (1999).
12 F. Cavani and F. Trifirò, Stud. Surf. Sci. Catal., 110, 19
(1997); F. Cavani, E. Etienne, M. Favaro, A. Galli, F. Trifirò,
and G. Hecquet, Catal. Lett., 32, 215 (1995).
13 S. Albonetti, F. Cavani, and F. Trifirò, Catal. Rev. Sci. Eng.,
38, 413 (1996).
14 E. Cadot, C. Marchal, M. Fournier, A. Tézé, and G. Hervé, in
“Polyoxometalates: From Platonic Solids to Anti-Retroviral
Activity,” ed. by M. T. Pope and A. Müller, Kluwer,
Dordrecht (1994), p. 315.
15 M. Akimoto, Y. Tsuchida, K. Sato, and E. Echigoya, J.
Catal., 89, 196 (1984).
16 C. Sanchez, J. Livage, J. P. Launay, M. Fournier, and Y.
Jeannin, J. Am. Chem. Soc., 104, 3194 (1982).
17 N. Mizuno, M. Tateishi, and M. Iwamoto, J. Catal., 163, 87
(1996); K. Eguchi, N. Yamazoe, and T. Seiyama, Nippon
Kagaku Kaishi 1981, 336; J. B. Black, and N. J. Clayden, J.
Chem. Soc., Dalton Trans., 1984, 2765.
18 L. A. Combs-Walker and C. L. Hill, Inorg. Chem., 30, 4016
(1991).
19 J. A. Rob van Veen, O. Sudmeijer, C. Emeis, and H. de Wit,
J. Chem. Soc., Dalton Trans., 1986, 1825.
20 K. Nomiya, M. Sugaya, and M. Miwa, Bull. Chem. Soc. Jpn.,
52, 3107 (1979) and private communication with Prof. K.
Nomiya.
21 The cyclic voltammetry of tetra-n-butylammonium salt
shows a reversible Fe2+/Fe3+ oxidation at –490 mV (Fc/Fc+
couple, +70 mV).
22 The crystallographic data show that no iron ions are present
as countercations and will be reported in due course.
23 D. K. Lyon, W. K. Miller, T. Novet, P. J. Domaille, E. Evitt,
D. C. Johnson, and R. G. Finke, J. Am. Chem. Soc., 113,
7209 (1991).
24 X. Chang, Q. Chen, D. C. Duncan, R. J. Lachicotte, and C. L.
Hill, Inorg. Chem., 36, 4381 (1997).