JOURN.
11, OF CATALYSIS 4, 581-585 (1965)
Catalytic Oxidation of 1-Butene over Bismuth
Molybdate Catalysts
1. The System BiD-Moo3
A. C. X. M. BLEIJENBERG, B. C. LIPPENS, AND G. C. A. SCHUIT
From the Department of Inorganic Chemistry, Technological University, Bindhooen, The Netherlands
Received December 10, 1964
In the system B&O, -MOOS the following compounds were ident,ified:
(1) BiZOa,3Mo03 = Bi2(MoO& m.p. 676°C
(2) BisOa.MoOa = B&Moo6 = (BiO)&~oOa m.p. 938°C
(3) 3Bi203.MoOa = BieMoOie m.p. 995°C
These compounds were found by investigation of the phase diagram of the BizOj-
MoOa binary system by means of thermal analysis. X-ray studies of sintered mixtures in
this system confirmed t.he existence of the three compounds. A compound Bi20a.2MoOa
could not be detected.
INTRoDCETI~~V must be formed, presumably on the surface
Oxides of vanadium, molybdenum, and of the oxidation catalyst. When this ally1
related elements find frequent use as oxida- radical has a structure such that by dis-
tion catalysts. The application of V205 for sociation of another H atom a conjugated
the oxidation of aromatics has been known diene can be formed, this reaction appears
for a considerable time: the oxidation of more favored than the oxidation of the
naphthalene and o-xylene to phthalic an- radical to an unsaturated aldehyde. Conse-
hydride and of benzene to maleic anhydride quently 1-butene is converted almost exclu-
have been applied in the industry for some sively to butadiene*
decades. Somewhat more recent is the dis- The organic chemistry of the oxidation
covery of the selective oxidation of olefins. therefore appears relatively well understood.
Veatch, Callahan, Milberger, and Forman The inorganic chemistry seems far less clear.
(1) reported on the oxidation of propylene It has often been observed that binary
to acrolein over a bismuth phosphomolyb- catalysts, such as, for instance, SnOr-
date/silica catalyst while Hearne and Fur- V205, appear to be more active and selective
man (2) showed that butylene was converted than the single components. For the olefin
in good yields to butadiene. It is the latter oxidation this is also true: neit.her MoOa nor
reaction that forms the subject of Paper II. Bi20, are particularly selective. The general
Thanks to the work of Adams and Jen- impression is that the ratio Bi/Mo = 1
nings (3) ; Sachtler (4) ; Adams (5) ; and presents the best catalyst composition avail-
Adams, Voge, Morgan, and Armstrong (6) able but the evidence seems somewhat
considerable information became available meager (1). It has not been investigated
regarding the reaction mechanism of the whether or not this ratio corresponds to the
oxidation of various olefins and in particular existence of a binary compound. The de-
of propylene and butylene. ‘Using D- and pendency of conversion and yield has not
C14-tagged olefins they showed that as an been studied in detail as far as we know.
intermediate stat,e an ally1 radical or ion The experiments reported in this paper are
581
582 BLEIJENBERG, LIPPENS, AND SCHUIT
concerned with the binary system Biz03- heated at 1250°C. The linear temperature
MOOS. It will be followed by a second paper rise of the furnace was 10°C min-l.
in which the relation between activity and The X-ray investigations were carried out
selectivity on the one hand and catalyst with a Philips X-ray diffractometer with
composition on the other hand is discussed. Geiger-Mtiller counter and Ni-filtered Cu
Subsequent experiments have been per- Ka radiation.
formed to st,udy the interaction of hydro-
carbon and inorganic catalyst systems RESULTS
without air being present. They will be Figure 1 shows some of the thermograms.
reported in the third paper. Similar experi- They were interpreted in the following way:
ments will be reported on nonstoichiometric (a) As the temperature at which the eutectic
MO oxides. line was passed the temperature at the foot
From literature data it appears that in
the system Biz03-MoOB the existence of
only one of the compounds described is be-
yond doubt, viz. the compound Bi203.Mo03,
which occurs in nature as the mineral
koechlinite. Gattow (7) prepared the mixed
oxide 2Bia03,Mo03 by precipitation from
aqueous solution. Zambonini (8) obtained
tetragonal Biz(Mo04)3 by melting together
B&O3 and MoOB (m.p. 643°C). Sill&n and
Lundberg (9) could not affirm the existence
of this last compound. They found a number
of phases of the type Bil--IMozOl.b(l+z) with
x between 0.05 and 0.15. Belyaev and
Smolyaninov (10) studied the ternary sys-
tem B&OS-Moos-PbO. In the Biz03-
Moos range they found the following com-
pounds: Bi2(Mo0J3, m.p. 648°C; Bi2MoOe,
m.p. 970°C; and BisMoOlz, m.p. 990°C.
EXPERIMENTAL PROCEDURE
As starting materials we used the ox-
ides Moos and B&OS, which were made
by heating in air: ammonium molybdate,
(NHI)6M07024.4H20, E. Merck A.G. Z.A.
1182; and bismuth nitrate, Bi(N0J3.5Hn0, AT -
E. Merck A.G. 1860, at 600°C during 20 hr
in platinum crucibles.
Twenty-one mixtures were made (pure
Biz03 and Moos included) with a difference
of 5 at y. MO in compositions. These mix- 600 600 1000 -"C
tures were sintered in air at 600°C in plati- FIQ. 1. Thermograms of (a) MoOI; (b) Bi20i*
num crucibles during 20 hr, ground in an 3Mo03 ; (c) Bi&*MoOa; (d) 17Biz0,.6M003;
agate ball mill during half an hour, and (e) Bi20s. A!!’ is the temperature difference be-
sintered again under the same conditions. tween sample and reference in arbitrary units.
The thermal analysis experiments were
performed according to the D.D.K. method of the first peak was taken. (b) The tempera-
(11) with the apparatus of the Linseis Co. ture at the top of the last peak was regarded
We used platinum crucibles, Pt-PtlORh the final melting point. The phase diagram
thermocouples and as reference a-Al203 based on the thermogram data is presented
OXIDATION OVER BISMUTH MOLYBDATE. I 583
PHASE DIAGRAM
I
Moo3 ATOMIC RATIO -
FIG. 2. Phase diagram of the system BilOrMoOJ (- --, from ref. 10).
in Fig. 2. As a first approximation we drew Some minor observations may be of in-
(horizontal) eutectic lines, though there are terest for the catalytic work. Firstly, there
indications for the occurrence of crystalline is a peculiar change in the color of the sam-
solutions, as we will discuss later. ples with increasing Bi content. Up to 25 at
The X-ray powder patterns are rather ‘% Bi they are nearly white with a slight
complex. Figure 3 gives a survey of the green haze, as is known for pure Moos; from
strongest lines in the patterns of the various 30 to 70 at $& Bi t,hey are increasingly yel-
samples; numerous weak to very weak lines low; above this composition the cream color
have been omitted from t,his figure. of BLO, is developed. Spe&rophotometer
584 BLEIJENBERG, LIPPENS, AND SCHUIT
measurements show that the yellow color is strongly deformed tetrahedra. One of the
caused by the flank of a very strong charge- three cation sites remains vacant.
transfer band in the far UV. Crystal field According to Sill&n and Lundberg (9) the
bands in the visible and near infrared region structure of compound (2) will have some
could not be detected. The absence of ESR relation to that of (La0)2M004, which con-
signals of the different compositions and sists of layers of La202 alternating with
their diamagnetism also indicate that no layers of Mood tetrahedra. The composition
Mov is present. of compound (2) (the mineral koechlinite)
then can be written as (BiO)z(MoOl). How-
DISCUSSION ever, although there are certain similarities
The phase diagram (Fig. 2) shows that between the powder diagrams of (LaO)nMoOd
there are three well-defined compounds in and of (BiO)sMoOa [e.g., the 001 reflections
the system BisOa- MOO,: of (LaO)nMoOa correspond in intensity to
lines in the (BiO)zMoOl diagram], it is not
(1) Bi203.3MoOa, m.p., 676°C possible to describe the (BiO)zMoO1 lattice
(2) BizOl.MoOa, m.p., 938°C completely with the (LaO)nMoOc lattice.
(3) 3Bi20rMo03, m.p., 995’C Below 700°C bismuth oxide occurs as the
monoclinic a-Bi203, which at higher tempera-
in good agreement with the results of ture, is converted into cubic &B&OS. This
Belyaev and Smolyaninov (4). The small &BizOI has a CaFz structure with only
differences in melting temperature may be three-quarters of the oxygen sites occupied
due to the different technique they used. (7). It is stabilized by building in SiOz,
Neither in the X-pay patterns nor in the Moos, or WOrc molecules, which can fill up
phase diagram can indications be found for the vacant oxygen sites. Probably the com-
the existence of a compound Biz03*2Mo03. pound (3) (3Bi203.Mo03) is the end member
Also during visual examination of the melt- of this series of crystalline solutions.
ing process under the microscope the pres- Attempts are being made to obtain single
ence of such a compound in the region of crystals of the three compounds in order to
45-55 at y0 Bi could not be ascertained. establish their structure.
Crystalline solutions seem to be present in a
great part of the system. The X-ray dia- ACKNOWLEDGMENTS
grams reveal that on the Bi-rich side of the
compounds Biz03.3Mo03 and Bi203.Mo03 We are greatly indebted to Prof. Dr. G. D. Rieck
and to Dr. K. Koopmans (Laboratory of Physical
solid solutions occur to an appreciable extent
Chemistry of the Technological University of Eind-
with a strong influence on the lattice dimen- hoven) for putting at our disposal the accommoda-
sions (corresponding lines are shifted to tion of the X-ray laboratory and for their interest in
higher spacings). our work and to Miss M. Kuyer for performing
From the melting phenomena observed the X-ray work.
under the microscope the conclusion has to
be drawn that also on the MO-rich side of the REFERENCES
compound Bi203.Mo03 crystalline solutions
to a limited extend will be present, in this 1. VEATCH, F., CALLAHAN, J. L., MILBERGBR, E. C.,
AND FORMAN, R. W., Actes Congr. Intern.
case without noticeable influence on the
Catalyse, de, Paris, 1960 2, 2647 W60.
lattice constants. 2. HEARNE, G. W., AND FURMAN, K. E., Shell
Only a few speculations concerning the Development Company U.S. Patent 2,995,
structures of the compounds can be made 320, July 4, 1961 (to Shell Development
in view of the complexity of the X-ray Co.).
powder diagrams. Zambonini reports com- 3. ADAMS, C. R., AND JENNINGS, T. T., J.
pound (1) to have a defect scheelite struc- Catalysis 2, 63 (1963).
ture, so the composition should be written 4. SACHTLER, W. M. H., Rec. Trav. Chim. 8%
as BiZ(MoOS3. If so, each Bi ion is sur- 243 (1963); Proc. htern. congr. Catabis
rounded by eight MoO( ions, which are 3r4 Amsterdam, i96,$, p. 252.
OXID.ITIOS OVER BISMUTH MOLYBDATE. I 585
5. -ADAMS, c. R., l’ioc. Idern. Congr. Catalysis 9. SILL&N, L. G., AND LUNDBERG, K., A&u Kemi
3rd, Amsterdam, 1964, p. 240. Min. Geol. Sci. Al7 (1945).
6. ADAMS, c. R., VOGE, H. H., MORG.~N, c. Z., AND 10. BELYAEV, I.N., AND SMOLYANINOV, N.P., Russ.
ARMSTRONG, W, E., J. Catalysis, to be J. Znorg. Chem. (English Trunsl.) 7, (1962),
published. 579.
7. GATTOW, G., Z. Anolg. Allgem. Chem. 298, G4, 11. SCHWIETE, H. E., AND ZIEGLER, G.: “Grundlnge
71 (1959). und Anwendungsgebiete der dynamische
S. ZAMBONIXI, F., Z. fiist. Min. Petrog. 58, 226 Differenzkalorimetrie,” Rer. Deutcrh. Keram.
(1923) ; 59, 563 (1924). Ges. 35, 193 (1958).
