Flash vacuum pyrolysis over solid catalysts. 1. Pyrazoles over zeolitest

Article abstract of DOI:10.1021/jo001390a

Flash vacuum pyrolysis (fvp) reactions of 1H-pyrazole (1), 3,5-dimethylpyrazole (2), and 3,5-diphenylpyrazole (3) were carried out over zeolites. Reactions were performed using ZCOY-7, NH₄-Y, and Na-Y zeolites. Reaction temperatures of heterogeneous reactions were lower than the corresponding temperatures in the homogeneous system, showing a catalytic effect of the zeolites. Compounds 1-3 afforded nitrogen extrusion in homogeneous fvp reactions while in the heterogeneous ones different reactions were present. Compounds 1 and 2 also afforded nitrogen extrusion; products arising from ring fragmentation were found in reactions of 2 and 3 while an isomeric imidazole was isolated in reactions of 3. Isomerization of 3 is attributed to a transition-state selectivity by the catalyst due to the relation between the size of the molecule and the cavity of the zeolite. This isomerization reaction was present only when zeolites with active Broensted sites were used.

Full text of DOI:10.1021/jo001390a

J . Org. Chem. 2001, 66, 2943-2947
2943
F la sh Va cu u m P yr olysis over Solid Ca ta lysts. 1. P yr a zoles over
Zeolites†
Elizabeth L. Moyano and Gloria I. Yranzo*
Instituto de Investigaciones Fı´sicoquı´micas de Co´rdoba (INFIQC), Departamento de Qu´ımica Orga´nica,
Facultad de Ciencias Quı´micas, Universidad Nacional de Co´rdoba, Ciudad Universitaria,
5000 Co´rdoba, Argentina
yranzogi@dqo.fcq.unc.edu.ar
Received September 19, 2000
Flash vacuum pyrolysis (fvp) reactions of 1H-pyrazole (1), 3,5-dimethylpyrazole (2), and 3,5-
diphenylpyrazole (3) were carried out over zeolites. Reactions were performed using ZCOY-7, NH4-
Y, and Na-Y zeolites. Reaction temperatures of heterogeneous reactions were lower than the
corresponding temperatures in the homogeneous system, showing a catalytic effect of the zeolites.
Compounds 1-3 afforded nitrogen extrusion in homogeneous fvp reactions while in the heteroge-
neous ones different reactions were present. Compounds 1 and 2 also afforded nitrogen extrusion;
products arising from ring fragmentation were found in reactions of 2 and 3 while an isomeric
imidazole was isolated in reactions of 3. Isomerization of 3 is attributed to a transition-state
selectivity by the catalyst due to the relation between the size of the molecule and the cavity of the
zeolite. This isomerization reaction was present only when zeolites with active Bro¨nsted sites were
used.
Sch em e 1
In tr od u ction
For several years, we have been engaged in the study
of flash vacuum pyrolysis (fvp) of nitrogen heterocycles,
pyrazoles being one of our targets. These reactions proved
to be very interesting from both mechanistic and syn-
thetic points of view.1,2 In general terms, pyrazoles afford
two main types of reactions depending on substitu-
tion at N1: loss of nitrogen in N1-unsubstituted com-
pounds and different reactions in N1-substituted pyra-
zoles. Isomerizations, homolytic cleavages, and elimina-
tions take place depending on the nature of this
substituent.1 Nitrogen extrusion reactions take place by
a unique mechanism that involves two isomerization
steps and nitrogen extrusion from a vinyldiazomethane
intermediate to give a vinylcarbene2 as shown in Scheme
1. These reactions seem to be interesting from a mecha-
nistic point of view, since they give information on
vinylcarbene reactivity but are not useful as a synthetic
tool because they have large energies of activation
and should be carried out at high temperatures. We
decided to study the possibility of lowering their energy
of activation and therefore reaction temperatures with-
out changing the type of reaction and chose the use
of solid catalysts, zeolites in this paper, as an alterna-
tive.
There are natural and synthetic zeolites. Natural
zeolites are alumino-silicate structures containing water
or other cations having the general formula3
x[(MI‚MII1/2)‚Al2O3]‚ySiO2‚zH2O
where MI ) Li, Na, K, etc. and MII ) Mg, Ca, Sr, Ba, etc.
All zeolites have a characteristic aluminosilicate frame-
work composed of [SiO4]4- and [AlO4]5- that are linked
together through oxygen bridges. Zeolites A and X/Y have
the sodalite cage, a truncated octahedron that has eight
hexagonal and six square faces, as their basic structure.
In zeolites X and Y (faujasites), the spherical internal
cavity generated when eight sodalite cages are joined is
about 13 Å diameter and is called the R cage. The
entrance of this cavity is about 7-8 Å wide (Figure 1).
Due to their size, several organic molecules can enter the
cavity. Most of the reactions take place inside the cavity
of the zeolites, and this is the main characteristic of their
activity as catalysts when compared with others that
present only surface catalysis. Zeolites have Bro¨nsted and
Lewis acid sites, so they are acid catalysts. It is possible
to inhibit Bro¨nsted sites by changing H by a cation to
study the type of catalysis present. The shape selectivity
observed in zeolite catalysts has been categorized in three
† Dedicated to Prof. Dr. J ose´ Elguero.
(1) (a) Pe´rez, J . D.; Yranzo, G. I.; Phagouape´, L. M. Bull. Soc. Chim.
Fr. 1987, 129. (b) Pe´rez, J . D.; Yranzo, G. I.; Ferraris, M. A.; Claramunt,
R. M.; Lo´pez, C.; Elguero, J . Tetrahedron 1988, 44(20), 6429. (c) Pe´rez,
J . D.; Yranzo, G. I. J . Anal. Appl. Pyrol. 1989, 16, 165. (d) Ferraris,
M. A.; Yranzo, G. I.; Pe´rez, J . D.; Cabildo, P.; Sanz, D.; Claramunt, R.
M.; Elguero, J . Anal. Qu´ım. Int. Ed. 1996, 92, 3.
(2) (a) Moyano, E. L.; Yranzo, G. I.; Elguero, J .; J . Org. Chem. 1998,
63, 8188 and references therein. (b) Yranzo, G. I.; Moyano, E. L.; Rozas,
I.; Dardonville, C.; Elguero, J .; J . Chem. Soc., Perkin Trans. 2 1999,
211.
(3) Schwochow, F.; Puppe, L. Angew. Chem., Int. Ed. 1975, 14(9),
620.
10.1021/jo001390a CCC: $20.00 © 2001 American Chemical Society
Published on Web 04/11/2001

2944 J . Org. Chem., Vol. 66, No. 9, 2001
Moyano and Yranzo
Ta ble 2. Rea ction s of 3,5-Dim eth ylp yr a zole (2)
T
T
(°C) catalyst % 2 products
(°C) catalyst % 2
products
330 none
330 Na-Y
330 NH4
100
25 6, 7
32 6, 7
560 none
100
560 Na-Y
560 NH4-Y
560 ZCOY-7
660 none
660 Na-Y
660 NH4-Y
660 ZCOY-7
15 5, 6, 7, 8, 9
20 5, 6, 7, 8, 9
0
330 ZCOY-7 27 6, 7
5, 6, 7
450 none
450 Na-Y
450 NH4-Y
100
7
100
5, 6, 7
10 5, 6, 7, 8, 9
9
0
11 6, 7
5, 6, 7, 9
5, 6, 7
450 ZCOY-7 97 5, 6, 7
Sch em e 2
F igu r e 1. Schematic representation of faujasite (Y type
zeolite) (from http://drake.che.udel.edu/∼group/feuerstein.html
with permission from the author.)
Ta ble 1. Rea ction s of 1H-p yr a zole (1)
T (°C)
catalyst
% 1
T (°C)
catalyst
% 1
450
450
450
450
560
560
none
100
19
16
13
100
8
560
560
660
660
660
660
NH4-Y
ZCOY-7
none
Na-Y
NH4-Y
ZCOY-7
6
3
100
7
3
0
Na-Y
NH4-Y
ZCOY-7
none
Na-Y
which is not surprising as it is an ultrastable catalyst.
As reactions here described have the same pattern as the
ones carried out in the homogeneous fvp system and
there is not a proton-catalyzed reaction (see results with
Na-Y zeolite in Table 1), there is no reason to think in a
different mechanism so we propose that this extrusion
reaction takes place by the mechanism depicted in
Scheme 1 (R ) H), which is the same one described for
fvp reactions of 1 in the homogeneous system.
F la sh Va cu u m P yr olysis of 3,5-Dim eth ylp yr a zole
(2). Reactions were carried out between 330 and 660 °C
using ZCOY-7, NH4-Y, and Na-Y as catalysts. Results of
these experiments as well as the ones reported earlier
for the homogeneous system7 are depicted in Table 2.
Product composition can only be explained by three
different reactions: nitrogen extrusion (as in the non-
catalyzed fvp) affording 1,3-pentadiene (5), ring frag-
mentation to acetonitrile (6) and propionitrile (7), and a
radical reaction for the formation of 3-(5)-methylpyrazole
(8) and pyrazole (9) (Scheme 2). Ring fragmentation
reaction may be analogous to cracking reactions of
hydrocarbons,8 one of the most common reactions cata-
lyzed by zeolites. Besides, the fact that 8 and 9 arise from
a radical reaction, is demonstrated by the formation of
bibenzyl in reactions when toluene was used as carrier
gas. It is worth to remember that this is a well-known
test of the presence of radicals.
F la sh Va cu u m P yr olysis of 3,5-Dip h en ylp yr a zole
(3). Reactions were carried out between 330 and 660 °C,
and results are shown in Table 3 where it can be seen
that different reactions take place when using these solid
catalysts, i.e., isomerization and ring fragmentation.
Products formed in the fragmentation reaction are ben-
zonitrile (10) and phenylacetonitrile (11), while 2,5-
diphenylimidazole (12) is the isomeric product. At higher
temperatures, closer to the ones of the homogeneous
system,2 2-phenylindene (13) and 3-phenylindene (14) are
formed. These results, summarized in Scheme 3, suggest
that there is no catalysis for nitrogen extrusion reaction
and that ring fragmentation and isomerization are cat-
types:4 (1) reactant selectivity (only reactants with the
appropriate dimensions can enter the cavity); (2) product
selectivity (only products with the appropriate dimen-
sions can leave the cavity); and (3) transition-state
selectivity (only transition states with the appropriate
dimensions can be formed at reasonable rates inside the
cavity).
Thus, zeolites can control the selectivity of a certain
chemical transformation with good yields.
Although zeolites are used in organic synthesis, there
are only a few reports of their utilization in fvp reactions,5
and there are also some reports of fvp reactions over
silica-alumina and clay as catalysts.6
Here we report our results of flash vacuum pyrolysis
of 1H-pyrazole (1), 3,5-dimethylpyrazole (2), and 3,5-
diphenylpyrazole (3) over ZCOY-7, NH4-Y, and Na-Y
catalysts, all of them faujasites. ZCOY-7 is a cracking
catalyst with 25% of zeolite Y while NH4Y is a protonic
zeolite and NaY is a zeolite without protons, both of them
are pure zeolite Y.
Resu lts a n d Discu ssion
F la sh Va cu u m P yr olysis of 1H-P yr a zole (1). Reac-
tions were carried out between 450 and 660 °C using
zeolites ZCOY-7, NH4-Y, and Na-Y as catalysts. As shown
in Table 1, the only reaction product is propyne (4), which
agrees with previously reported results carried out in a
homogeneous system.7 Results show that reaction tem-
peratures are lower than the ones described in the
homogeneous system and that there is no change in
product composition when Na-Y zeolite was used, which
shows that the reaction is catalyzed, but not by protons
as this zeolite has no Bro¨nsted acid sites. It is also clear
that ZCOY-7 catalyst is the most effective of the three,
(4) Sen, S. E.; Smith, S. M.; Sullivan, K. A. Tetrahedron 1999, 12657.
(5) (a) Banciu, M. D.; Cira, O.; Petride, A.; Banciu, A.; Draghici, C.
J . Anal. Appl. Pyrol. 1997, 42, 177. (b) Banciu, M. D.; Cira, O.; Petride,
A.; Banciu, A. Rev. Roumanie Chim. 1997, 42(11), 1093.
(6) See, for example: van der Waals, A.; Klundeer, A. J . H.; van
Buren, F. R.; Zwanenburg, B. J . Mol. Catal. A: Chem. 1998, 134, 179.
(7) Pe´rez J . D.; Yranzo, G. I. Bull. Soc. Chim. Fr. 1986, 473.
(8) See for example: Zhang, W.; Smirnotis, P. G. J . Catal. 1999,
182, 400.

Flash Vacuum Pyrolysis over Solid Catalysts
J . Org. Chem., Vol. 66, No. 9, 2001 2945
Ta ble 3. Rea ction s of 3,5-Dip h en ylp yr a zole (3)
Sch em e 4
T (°C) catalyst % 3 % 10 % 11 % 12 % 13 % 14 % othersa
330 none
330 Na-Y
330 NH4-Y 100
330 ZCOY-7 95
100
1
99
5
450 none
450 Na-Y
450 NH4-Y
100
2
66
98
8
2
1
24
38
450 ZCOY-7 48 10
3
560 none
100
12 88
560 Na-Y
560 NH4-Y
560 ZCOY-7
660 none
660 Na-Y
660 NH4-Y
3
4
83
96
13
1
2
47
8
6
53
19
22
87
19
9
32
5
1
660 ZCOY-7 26
64
a
Benzene, anthracene, phenanthrene, and other aromatic
compounds.
Sch em e 3
show isomerization and ring fragmentation. It is surpris-
ing that there is no catalysis for nitrogen extrusion in
reactions of 3 and, instead of this, isomeric 12 is formed.
Thermal isomerization reaction has previously been
described for N-phenylpyrazoles,1a N-benzoylpyrazole,1c
and N-adamantylpyrazoles,1d but this is the first time
this reaction is described for an N-unsubstituted pyra-
zole. Photochemical isomerizations of N-substituted10 and
N-unsubstituted11 pyrazoles to imidazoles has also been
reported. The mechanism of thermal and photochemical
isomerizations involves the formation of a 1-azirine (not
detected in these reactions) and then a ring opening
reaction of the three- membered ring to afford the
corresponding imidazole. We have no reason to suppose
that reactions of 3 take place by a different mechanism.
It is clear that this reaction is only present when zeolites
with Bro¨nsted acid sites are used (see Table 3), which
strongly suggests that protons catalyze this reaction,
which is not surprising since pyrazoles have a basic
nitrogen in the ring. We propose the mechanism depicted
in Scheme 4 with the participation of zeolite’s protons.
In this mechanism, pyrazole is bonded to the zeolite
through a H-bond and after loss of water it is covalently
bonded to the active site. At this point, isomerization to
the 1-azirine takes place, and at this stage of the
mechanism there are two possibilities: (a) isomerization
to imidazole and then leaving of the active site with
participation of a molecule of water or (b) the 1-azirine
leaves the active site with the participation of water and
then it isomerizes to the imidazole. We believe that this
last option is better because the N atom in the 1-azirine
is more free to interact with a molecule of water than
the N atom of the imidazole.
alized reactions. Fragmentation reactions should lead to
the same amounts of 10 and 11 (Scheme 4) but, as can
be seen in Table 3, this is not the case. Compound 11 is
only found in minor quantities than 10, so we supposed
that 11 may decompose under reaction conditions. To
check this, we carried out reactions of 2,2-diphenylac-
etonitrile (with a similar structure of 11) at 400 °C and
found that this compound gave rise to benzene and
condensed aromatic compounds which may be considered
as evidence that 11 has the same reaction.
From the results described above for reactions of 1-3,
it is clear that there is an effective catalysis when
different zeolites are used in fvp of NH-pyrazoles, al-
though the final products depend on ring substitution and
in some cases on the type of zeolite employed. It is also
clear from experimental results (see Tables 1-3) that
there is a decrease in the yield of conversion in reactions
at high temperatures (above 640 °C). This effect is more
evident in reactions of 3. To check if the catalyst was
losing its activity, we carried out a reaction of 3 with wet
air as carrier gas, which is the way to recover the
catalyst, and found 100% conversion and some other
products arising from reaction with O2. This experiment
suggests that the catalyst gets poisoned at high temper-
atures, but can be recovered, and that this poisoning is
stronger in reactions of 3 than in the corresponding of 1
or 2.
Concerning reaction products, it was found that in
reactions of 1 only nitrogen extrusion product, i.e.,
propyne, is formed and no products arising from compet-
ing reactions were detected. It is worth to remember here
that propyne is the only product in homogeneous fvp
reactions. On the other hand, reactions of 2 and 3 have
different pathways: nitrogen extrusion at lower temper-
atures than in the classic fvp,9 ring fragmentation and
homolytic cleavage in reactions of 2 while reactions of 3
Ring fragmentation is a competitive reaction in fvp of
2 and 3. It was discussed by Rademacher in his review
on fragmentations of five-membered rings.12 This author
(9) Pe´rez J . D.; Yranzo, G. I. Bull. Soc. Chim. Fr. 1986, 473.
(10) (a) Beak, P.; Messer, W. Tetrahedron 1969, 25, 3287. (b) Pavlik,
J . W. In Organic Photochemistry; Ramamaurthy, V., Schanze, K. S.,
Eds.; Marcel Dekker: New York, 1997; Vol. 1, 57.
(11) Tiefentaler, H.; Do¨rschein, W.; Go¨th, H.; Schmidt, H. Helv.
Chim. Acta 1967, 50, 8, 2244.

2946 J . Org. Chem., Vol. 66, No. 9, 2001
Moyano and Yranzo
F igu r e 2. Comparative sizes of 1, 2, and 3 with faujasite.
Sch em e 5
are shown in Figure 2. As is clear, compounds 1 and 2
are small enough to enter the cavity, and the transition
states and intermediate compounds to the corresponding
vinylcarbenes can be held inside the zeolite. On the other
hand, dimensions of 3 are very close to the dimensions
of the cavity, so a ring expansion reaction is unlikely to
happen. Instead of this, a ring contraction reaction to the
isomeric 1-azirine has more chances to take place. These
events may be interpreted as a transition state selectivity
by zeolite Y.
As a summary of results described in this paper, it is
important to point out that zeolites catalyze fvp reactions
of pyrazoles. When these results are compared with the
ones obtained in homogeneous fvp reactions of the same
compounds,2a,7 it is found that some different reactions
are produced along with nitrogen extrusion in ring
substituted compounds provided that the transition state
has enough capacity inside zeolite’s cavity. As a continu-
ation of this work, reactions of NH-pyrazoles over dif-
ferent solid catalysts are being carried out and will be
published soon.
describes [5 f 3 + 2] fragmentations of different hetero-
cycles such as furanes, pyrroles, thiophenes, pyrazoles,
and isoxazoles in concerted or stepwise mechanisms. It
is worth to mention here that nitrogen extrusion reac-
tions of pyrazoles may be considered as this type of
fragmentations, where the molecular nitrogen is the two-
atoms fragment. In the present case, we report a different
fragmentation pattern of pyrazole ring when solid acid
catalysts are used. This fragmentation corresponds to
N-N and C3-C4 bond fission (see Scheme 5), which were
reported in shock tube reactions of isoxazole13 and flash
vacuum pyrolysis of 3,6-diphenyl-1,4,2,5-dioxadiazine.14
It is not probable that reactions of 2 and 3 are concerted,
since the final products are formed by migrations of H
and R (CH3 or C6H5), and this may probably occur
through three-membered ring intermediates.
Con clu sion s
Some important conclusions may be drawn from the
above results: (1) It is possible to carry out flash vacuum
pyrolysis over solid catalysts. (2) Zeolites are good
catalysts for this purpose, although in the case of pyra-
zoles they derive to different competing reactions. (3)
When the size of the reacting pyrazole is very similar to
the size of the zeolite cavity, the molecule can enter the
catalyst, but an isomerization via a ring contraction
reaction is preferred. This isomerization may be of higher
energy than nitrogen extrusion as this reaction is not
found in small compounds where this last reaction is
found.
To check if 10 and 11 were formed from fragmentation
of 12 as well as from 3, we carried out some reactions of
12 under the same conditions of 3. Compounds 10 and
11 were found along with some other products (aromatic
compounds); with this evidence we cannot discard the
pathway in Scheme 5.
Exp er im en ta l Section
Gen er a l Meth od s. Reactions were carried out in Vycor
glass fvp equipment using a Thermolyne 21100 tube furnace
with a temperature controller device. Oxygen-free dry nitrogen
or a mixture of nitrogen/toluene was used as carrier gas.
Samples to be pyrolyzed were 30-50 mg. Contact times were
around 10-2 s, and pressures of 0.2-0.1 Torr were measured
with a Mc’Leod manometer. In a typical run, 0.75-1.50 g of
fractured catalyst was placed along the reactor (30 cm length,
1 cm diameter) using ceramic wool fiber as inert support.
Now the absence of nitrogen extrusion in reactions of
3 should be discussed. To find an explanation to this fact,
we calculated the size of compounds 1-3 to be compared
with the pore and the cavity sizes of zeolite Y. Results
(12) Rademacher, P. Adv. Heterocycl. Chem. 1999, 72, 361.
(13) Lifshitz, A.; Wohlfeiler, D. J . Phys. Chem. 1992, 96, 11, 4505.
(14) Mitchell, W. R.; Paton, R. M. J . Chem. Res., Synop. 1984, 58.

Flash Vacuum Pyrolysis over Solid Catalysts
J . Org. Chem., Vol. 66, No. 9, 2001 2947
Products were trapped at the liquid air temperature, extracted
with solvent, and submitted to different analyses or separation
techniques. Gas chromatography/mass spectrometry (GC/MS)
analyses were performed in a Perkin-Elmer Q-Mass 910
spectrometer equipped with an SE-30 column, using helium
as eluent gas at a flow rate of 1 mL/min and a heating rate of
40 °C for 5 min and 40-280 °C for 35-40 min. Mass spectra
were obtained in the electron impact mode (EI) using 70 eV
as ionization energy. 1H NMR and 13C NMR spectra were
carried out in Cl3CD in a Bruker 200 FT spectrometer (at 200
MHz). Chemical shifts are reported in parts per million (ppm)
downfield from TMS. Column and thin-layer chromatography
were performed on silica gel. Recovery of material was >90%
in all in fvp experiments. In reactions of 1 and 2, quantification
matography using chloroform/benzene, 50:50 as solvents), and
1
the isolated products were analyzed by H and 13C NMR and
GC/MS. In all cases, 2 as well as 5-9 were detected in different
amounts depending on reaction temperature. The identity of
compounds 6 and 9 were established by comparison with
commercial samples while 5 was compared with the product
obtained in a fvp reaction of 2 in the homogeneous system.
The NMR and MS data of 7 and 8 were identical to those
published previously.16a,b
F VP of 3. Compound 3 was prepared as described in the
literature from 1,3-diphenyl-1,3-propanedione and hydrazine.17
After fvp reactions, products were extracted from the cold trap
and separated by column chromatography using benzene:
methanol (90:10) as solvents. The identity of 10 was estab-
lished by comparison with a commercial sample, while 13 and
14 were identified by comparison with samples obtained from
fvp reactions of 3 in homogenoeus fvp experiments as described
1
of starting material was performed by H NMR using 2-3 µL
of nitromethane as internal standard. There is no information
of quantities of reaction products because some of them are
volatile. In reactions of 3 starting material and reaction
products were quantified by 1H NMR using 2-3 µL of ni-
tromethane as internal standard. Calculations were performed
using the Hyperchem program using AM1 for geometrical
optimization.
1
elsewhere.2a Nitrile 11 was identified by its H and 13C NMR
spectra and GC/MS analysis, which were in agreement with
previous observations.18 Isomer 12 was isolated from the
1
reaction crude and characterized by H NMR, 13C NMR, and
GC/MS. Spectral data: 1H NMR (CDCl3) δ (ppm) 3.81 (1H,
broad signal), 7.05-7.76 (7H, m), 7.80-8.51 (4H, m); 13CNMR
(acetone-d6) δ (ppm) 123.5, 124.8, 126.0, 128.3, 129.5, 130.2,
131.2, 133.0, 157.0; GC/MS m/z 220 (100), 205 (9), 191 (8), 165
(27), 96 (19), 89 (38), 77 (31), 63 (39), 51 (61), 39 (46); retention
time 26.34 ( 0.02 min.
Ca ta lytic Ma ter ia l. Zeolites Na-Y (Si/Al ) 4.61; Al/Na )
1.0) NH4-Y (Si/Al ) 8.77), and ZCOY-7 (Si/Al ) 6.5, matrix
area: 52m2/g, zeolite area: 144m2/g, aO: 24.46 Å, La2O3:
0.27%, CeO2: 0.55%, Al2O3: 41.3%) were kindly donated by
Prof. Eduardo Herrero from CITeQ-UTN (Co´rdoba, Argentina).
All zeolites were preactivated in air at 500-550 °C for 4 h
(including heating, 100 °/h) before each reaction. The catalysts
were pressed, fractured, sieved to the desired particle size
fraction of 12-20 mesh, and stored under ambient atmosphere.
Ack n ow led gm en t. We thank CONICOR (Co´rdoba,
Argentina), SECyT-UNC (Co´rdoba, Argentina), and
INFIQC (Argentina) for financial support. E.L.M. is a
grateful recipient of a fellowship from FOMEC (Minis-
terio de Cultura y Educacio´n, Argentina). We also thank
Dr. Elena Basaldella (CINDECA, Argentina) for her
helpful assistance in characterization of the catalysts.
F VP of 1. Compound 1 was commercially available from
SIGMA and purified by sublimation in vacuo. After the fvp
experiment, the reaction crude contained a solid identified as
1 and a gaseous product. Compound 1 was compared with
commercial sample and the gaseous product was dissolved in
1
J O001390A
Cl3CD and analyzed by H NMR. The spectrum showed only
one signal at δ ) 1.85 ppm, attributed to 4 as reported
previously.7
(15) Morgan G. T.; Burges S. H. J . Chem. Soc. 1921, 697.
(16) (a) Takahashi, T.; Tokuda, M.; Itoh, M.; Suzuki, A. Chem. Lett.
1975, 523. (b) Cabildo, P.; Claramunt, R. M.; Elguero, J . Org. Magn.
Reson. 1984, 22(9), 603.
(17) Elguero, J .; J acquier, R. Bull. Soc. Chim. Fr. 1966, 90.
(18) Organic Syntheses; Wiley: New York, 1990; Collect. Vol. VII,
p 27.
F VP of 2. Compound 2 was synthesized by a modification
of the synthesis of 3,5-dimethylisoxazole using 2,4-pentanedi-
one and hydrazine instead of hydroxylamine and then purified
by sublimation in vacuo.15 After the fvp experience, the
reaction crude was submitted to radial and thin-layer chro-