Mixed oxides as highly selective catalysts for the flash pyrolysis of phenacyl benzotriazole: One-pot synthesis of dibenzazepin-7-one

Article abstract of DOI:10.1021/cs3008335

The one-pot synthesis of dibenzo[b,d]azepin-7-one (3) was selectively achieved from 1-phenacyl-1,2,3-benzotriazole (1) using the catalytic flash vacuum pyrolysis (cfvp) methodology. Catalysts with the scheelite structure ABO₄ (A = Ca²⁺, Sr²⁺, Ba²⁺ and B = Mo⁶⁺, W⁶⁺) and fergusonite BiVO₄ were explored in this new catalytic reaction. These oxides promoted high conversion of starting material at lower temperatures than those observed for noncatalytic reactions. The chemical nature of A and B cations in the scheelite structure showed a strong influence on the formation toward the desired azepinone. In addition, the catalyst's morphology had a significant influence on the course of the cfvp reaction.

Full text of DOI:10.1021/cs3008335

Letter
pubs.acs.org/acscatalysis
Mixed Oxides as Highly Selective Catalysts for the Flash Pyrolysis of
Phenacyl Benzotriazole: One-Pot Synthesis of Dibenzazepin-7-one
German Lener,†,‡ Raul E. Carbonio,‡ and E. Laura Moyano*,†
́
‡
†Departamento de Química Organ
́
ica, Departamento de Físico Química, Facultad de Ciencias Químicas, Universidad Nacional de
́ ́
Cordoba, INFIQC (CONICET), Ciudad Universitaria, X5000HUA Cordoba, Argentina
S
* Supporting Information
ABSTRACT: The one-pot synthesis of dibenzo[b,d]azepin-7-one (3) was
selectively achieved from 1-phenacyl-1,2,3-benzotriazole (1) using the
catalytic flash vacuum pyrolysis (cfvp) methodology. Catalysts with the
scheelite structure ABO4 (A = Ca2+, Sr2+, Ba2+ and B = Mo6+, W6+) and
fergusonite BiVO4 were explored in this new catalytic reaction. These
oxides promoted high conversion of starting material at lower temperatures
than those observed for noncatalytic reactions. The chemical nature of A
and B cations in the scheelite structure showed a strong influence on the
formation toward the desired azepinone. In addition, the catalyst’s
morphology had a significant influence on the course of the cfvp reaction.
KEYWORDS: catalytic pyrolysis, catalysis, mixed oxides, dibenzazepinone, phenacyl benzotriazole
oxygen; and the promotion of intramolecular reactions, which
1. INTRODUCTION
do not take place under other conventional thermal conditions.
Cfvp reactions represent a good technique to study catalytic
gas−solid interaction in dynamic conditions. Some experiments
have been developed using alumina, amorphous silicates, and
zeolites.9
The chemistry of heterocycles has been widely explored
because of the known interest in organic, inorganic, and
medicinal chemistry. In this context, nitrogen-containing
heterocycles have received much attention essentially as a
result of a detailed history as significant medicinal pharmaco-
phores.1 In particular, the dibenzazepinone nucleus exhibits a
wide range of biological activities. Thus, for example, some
dibenzo[b,d]azepin-6-one derivatives have displayed a potent
inhibition of γ-secretase, which appears as an important target
for drug discovery in the treatment of Alzheimer’s disease.2
Even though dibenzo[b,d]azepin-7-one compounds have
shown interesting pharmacological properties, conventional
organic syntheses have not been adequately developed, and the
known protocols involve many steps and, hence, low yields.3−6
Taking into account these difficulties in obtaining this fused
ring by conventional synthesis, we applied flash vacuum
pyrolysis (fvp) methodology to afford the dibenzazepinone
ring from a simple benzotriazole precursor. This gas phase
synthesis consists of a nitrogen extrusion reaction of the
starting benzotriazole, followed by rearrangement of the
carbene intermediate to get the desired azepinone.7,8 In the
fvp procedure, the substrate molecules are subjected to high
temperatures (200−600 °C) and low pressures (∼10−3 Torr)
over a short period of time (∼10−2 s) to promote their reaction
in a pyrolysis tube. In addition, solid and thermostable catalysts
can be used in the pyrolysis reactor; the methodology is called
catalytic flash vacuum pyrolysis (cfvp). This technique is an
extension of the conventional homogeneous fvp and keeps
important features and advantages: the possibility to isolate
kinetic as well as unstable products; absence of solvents and
In our group, we have studied the cfvp reactions of many
nitrogen heterocycles using acid and basic catalysts (zeolites,
hydrotalcites, and mesoporous materials).10 In general, the
application of cfvp has decreased the reaction temperature with
respect to the noncatalyzed homogeneous process.8 Moreover,
in the zeolite and mesoporous-catalyzed experiments, the
presence of Lewis acid sites were extremely important in the
selectivity of the thermal process.9,10 With the aim to develop
new catalytic gas phase pyrolysis, we decided to study the
behavior of different mixed oxides in these heterogeneous
systems. Thus, catalysts containing scheelite-type compounds
ABO4 (A = Ca2+, Sr2+, or Ba2+ and B = Mo6+ or W6+) and the
fergusonite-type BiVO4 were investigated.
Scheelites possess interesting physical properties that make
them useful in several technological applications.11−13 These
materials can be synthesized by several methods, with the
possibility to incorporate a large number of different A or B
cations.14 The flexibility in their preparation as well as the
accessible precursors makes this family of oxides a model
system for different catalytic reactions. Some scheelites have
been reported to be efficient in the catalytic oxidation of
Received: December 20, 2012
Revised: March 8, 2013
Published: April 3, 2013
© 2013 American Chemical Society
1020
dx.doi.org/10.1021/cs3008335 | ACS Catal. 2013, 3, 1020−1025

ACS Catalysis
Letter
Figure 1. SEM images of SrMoO4 synthesized by different methods: (a) dp, (b) cm, (c) sg.
hydrocarbons;12 however, the performance of these materials in
more complex catalytic processes have not been explored.
The crystallographic phase of these oxides shows that the A
cations are coordinated by eight O2−, forming a distorted cube,
and the B cations are coordinated by four O2−, giving a slightly
asymmetric tetrahedron. If the A cation has a nonbonding
electron, such as Bi3+, the structure is distorted to monoclinic
fergusonite phase (SG = I 21/b, #15). Typical oxides that
present a transition from scheelite to fergusonite structures are
Pb(Mo,W)O4 and BiVO4.15,16 The latter has been investigated
because of its photocatalytic properties,17,18 solar-driven
hydrogen production,19 and catalytic use in ammoxidation of
alkanes.20 BiVO4 can be synthesized by different preparation
routes.21−24 In this work, we used ultrasound and microwave
irradiation as two alternative approaches to obtain the desired
monoclinic phase of this oxide in short reaction times.
The surface area was determined by a six-point nitrogen
adsorption isotherm according to the Brunauer−Emmett−
Teller (BET) method. Micrometrics Series 680, Mod. Accusorb
2100E-210-00001-01. In these experiments, the samples were
dried and degassed at 110 °C.
2.2. Fergusonite BiVO4 Catalyst. a). Ultrasound Syn-
thetic Method. The experiment was performed using an
Ultrasonic Cleaner Testlab (80 W, 40 kHz). Stoichiometric
amounts of Bi(NO3)3·5(H2O) and NH4VO3 were dissolved in
30 mL of 1 M HNO3. Then an excess of urea was added, and
the mixture was subjected to ultrasound irradiation at 60 °C for
4 h. The yellow-orange precipitate was filtered, washed, and
dried at 80 °C for 12 h. The PXRD−Rietveld analysis indicated
the presence of tetragonal zircon structure (space group I4/m)
and a fergusonite monoclinic structure (space group I21/b) in a
50/50 relationship. The pure desired monoclinic fergusonite
crystallographic phase was achieved after heat treatment at 450
°C for 10 h.
b). Microwave Synthetic Method. Stoichiometric amounts
of Bi(NO3)3·5(H2O) and NH4VO3 were dissolved in 30 mL of
1 M HNO3, then an excess of urea was added. The mixture was
irradiated at 100 °C for 4 min in a 50 mL reactor using an open
vessel CEM-Discovery monomode microwave apparatus work-
ing at 100 W as power supply. The yellow precipitate was
separated, washed, and dried at 80 °C for 12 h. The PXRD−
Rietveld analysis indicated a high purity of monoclinic
fergusonite phase, over 97%. The pure 100% monoclinic
phase was achieved after a thermal treatment at 250 °C for 12
h.
2.3. Fvp and cfvp Experiments. The apparatus for fvp
experiments consists of three main parts: (1) the injection
region; (2) the pyrolysis region; and (3) trapped the
condensation region, where products are condensates in a
glass U-tube immersed in cryogenic liquids. The system is
evacuated to 1 × 10−3 mmHg by a high-capacity vacuum pump.
A flow of carrier gas (dry N2) is regulated as 1 mL per 10 s.
Under these experimental conditions, substrate molecules
spend about 1 × 10−2 s in the hot region. Sample amounts
were 18−20 mg with catalyst/substrate ratios equal to 25
(experimental approach I) and 1 (experimental approach II),
respectively.
We report herein the application of different mixed oxides
with scheelite and fergusonite type structures in the catalyzed
gas phase pyrolysis of 1-phenacyl-1,2,3-benzotriazole (1)
toward the formation of 7H-dibenzo[b,d]azepin-7-one (3).
2. EXPERIMENTAL SECTION
2.1. Preparation of Catalysts. Scheelite ABO4 Cata-
lysts. Synthesis of ABO4 oxides by a dissolution−precipitation
(dp) method: Materials were prepared by dissolution of
(NH4)6Mo7O24·4H2O or (NH4)2WO4 in water at room
temperature. The pH of these solutions was increased to 10
by addition of NH3 solution, of A(NO3)2 (A = Ca2+, Sr2+, or
Ba2+) was added dropwise. The white precipitate was filtered
and dried at 100 °C. Some samples were also heat treated at
600 °C for 12 h for comparison with the other preparation
methods.
Synthesis of ABO4 supported on inert ceramic fiber (method
dp−fc): Oxides were prepared following the dp method. The
ceramic fiber was added in the solution before the precipitation
process. Thus, the oxide was uniformly deposited on the
ceramic fiber. After the deposition, the material was dried at
200 °C.
Synthesis of SrMoO4 by a ceramic (cm) method: This oxide
was prepared from SrCO3 and MoO3 solids thoroughly mixed
with acetone in an agate mortar. Then this mixture was heated
at 600 °C under air for 12 h.
In these experiments, the mass balance percentage is
calculated as the ratio between the mass of inputs and the
mass of outputs multiplied by 100. The mass of outputs is the
total mass trapped in the condensation zone of the system.
Catalysts were placed in the middle region of the reactor,
perpendicular to the trajectory of the carrier gas, using ceramic
fiber as an inert support. All catalysts were reused
approximately six times. Products were analyzed by GC/MS
Synthesis of SrMoO4 by a sol gel (sg) method: This oxide
was prepared by dissolution of (NH4)6Mo7O24·4H2O and
Sr(NO3)2 in a saturated solution of citric acid. Then the water
was evaporated at 80 °C until gel formation. After grinding, this
amorphous precursor was calcined at 600 °C for 12 h in flowing
air.
All catalytic materials were characterized by powder X-ray
diffraction (PXRD) carried out with a Panalytical X’Pert Pro
Difractometer (40 mV, 40 mA), Cu Kα (λ = 1.5418 Å).
1
and H and 13C NMR spectroscopy.
1021
dx.doi.org/10.1021/cs3008335 | ACS Catal. 2013, 3, 1020−1025

ACS Catalysis
Letter
vanadate was conveniently prepared during 3 h of irradiation at
180 °C.23
In this work, fergusonite BiVO4 was synthesized by two
approaches: (a) ultrasound assisted synthesis (BiVO4-US) and
(b) microwave-assisted synthesis (BiVO4-MW) (see the
Experimental Section). The morphologies of particles prepared
by these two methods are shown in Figure 2. Ultrasound
3. RESULTS AND DISCUSSION
3.1. Synthesis of ABO4 Scheelite Oxides (A = Ca2+,
Sr2+, and Ba2+; B = Mo6+ and W6+) and Fergusonite
BiVO4. SrMoO4 was the initial oxide evaluated in the cfvp study
of phenacyl benzotriazole 1. To compare the effect of the
morphology of the particles in the catalyst performance, this
scheelite was synthesized by four different methods: (a) dp, (b)
dp−fc, (c) cm, and (d) sg. PXRD analysis indicated that all
catalysts were pure and crystallized with scheelite-type
tetragonal structure (space group I41/a, #88). PXRD showed
intense and well-defined peaks, suggesting a good degree of
crystallization or long-range structural ordering. The exper-
imental lattice parameters and atomic positions were refined
using the Rietveld method25 with the FULLPROF program.
SEM micrographs were of fundamental importance to study the
morphology and microstructures of scheelite crystals, depend-
ing on the synthetic methods (Figure 1). Thus, the application
of the dp method produces quasispherical particles with 10 μm
diameter composed of a large number of nanocrystallites. In all
cases, our findings by SEM were coincident with the average
size of crystallite obtained by the Scherer’s equation (about 15
nm).26 Samples synthesized by the dp method, but with a heat
treatment at 600 °C for 12 h, showed the same particle size
(see Supporting Information Figure S2), but the PXRD data
showed an increase in the crystallite size to ∼40 nm. On the
other hand, the cm method led to stick-shaped crystals with a
length of 10 μm composed of nanostructured crystallites. Using
the sg method, the assemblies of particles were randomly
organized with the formation of irregular and porous crystals
with a large size (about 600 μm).
Figure 2. SEM images of BiVO4: (a) synthesized by the ultrasound
method and (b) synthesized by the microwave method.
precipitation produced irregular spherical particles between 50
and 100 nm in size, giving place to agglomerates of around 1
μm. Using microwave precipitation, irregular cubic particles of
500 nm in size were formed alongside spherical agglomerates of
around 2 μm.
For cfvp experiments, the catalyst BiVO4 was supported on
inert ceramic fiber. An aqueous suspension of the fergusonite
BiVO4 was mixed with the ceramic fiber, and after the
deposition of the oxide was completed, the impregnated fiber
was dried at 200 °C. Figure 3 shows the SEM image of the
BiVO4-supported fiber, where it can be seen that the
distribution of the oxide particles is uniform.
According to the results of cfvp reactions (see Table 1), the
oxide SrMoO4 synthesized by the dp method displayed the
Table 1. Fvp (homogeneous) and Cfvp Reactions of 1 over
Scheelites ABO4; Experimental Approach I
method of
synthesis
SBET
T
a
b
b
b
catalyst
(m2/g−1
)
(°C) % 1 % 3 % 4
none
none
none
400
500
550
400
400
400
400
400
400
400
400
78
35
0
0
30
10
20
20
20
30
0
0
0
0
SrMoO4
SrMoO4
SrMoO4
CaMoO4
sg
3.3
1.4
2.6
6.5
1.1
2.3
8.7
55
20
0
4
cm
dp
dp
dp
dp
dp
dp
20
4
0
6
c
BaMoO4
SrWO4
0
0
Figure 3. SEM images of oxides deposited on ceramic fiber: (a) BiVO4
synthesized by MW and (b) BaWO4 synthesized by dp method.
0
5
5
CaWO4
0
20
20
c
BaWO4
0.5
0
0
0
It should be noted that after the cfvp experiments, all
catalysts where characterized by PXRD, and the results
indicated that the crystal structure remained unchanged during
the thermal reaction.
a
b
Catalyst/substrate ratio was 25. Values determined by 1H NMR. At
c
temperatures below 400 °C, the conversion was not significant. Main
products formed by fragmentation of 1: benzotriazole, benzonitrile,
benzaldehyde, acetophenone, and aniline.
3.2. Catalytic Flash Vacuum Pyrolysis of 1. In the
development of the scheelite−heterogeneous catalytic pyrolysis,
two experimental approaches (I and II) were evaluated. In
approach I, the bulk solid catalyst was packed inside the middle
section of the reactor tube, covering all the transversal area, and
the inert support (ceramic fiber) was placed in the outer region
of the catalyst section to avoid the emission of solid particles. In
this case, the amount of catalyst was minimized to get an
efficient covered cross section of the reactor.
better catalytic performance. For this reason, the rest of the
ABO4 (A = Ca2+, Sr2+, Ba2+; B = Mo6+, W6+) catalysts were
synthesized by the same synthetic method.
Regarding the synthesis of fergusonite BiVO4, it was
determined that direct coprecipitation followed by heat
treatment at 450 °C allowed obtaining of the monoclinic
fergusonite phase (I 21/b).21 By the ultrasound-assisted
method, synthesis of the monoclinic BiVO4 was also achieved
at 60 °C over 2 h.24 Using microwave irradiation, the same
In procedure II, the catalyst was supported on the ceramic
fiber, which was then placed inside the middle region of the
1022
dx.doi.org/10.1021/cs3008335 | ACS Catal. 2013, 3, 1020−1025

ACS Catalysis
Letter
Scheme 1. Formation of Products in the Cfvp Reactions of 1
reactor tube. This arrangement enabled a significant decrease in
the amount of catalysts exposed to the reaction, increasing their
surface area. To confirm the deposition of the oxides on the
ceramic fiber, SEM images and PXRD of these fiber-supported
catalysts were performed (see Figure 3 for BiVO4 and BaWO4).
To verify that ceramic fiber could be used as an inert support,
this fiber was used alone in the gas phase pyrolysis reactions of
phenacyl benzotriazole 1. Under these conditions, no catalysis
was achieved and formation of products was comparable to the
homogeneous system.
3.2.1. Catalytic Reactions Using a Packed Catalyst
(Experimental Approach I). Cfvp reactions of 1 in the
presence of scheelites ABO4 (A = Ca2+, Sr2+, or Ba2+ and B =
Mo6+ or W6+) were carried out in the temperature range 270−
450 °C (Scheme 1). The ratio betweencatalyst and substrate
was 25. In general, high conversion of 1 (>80%) was achieved
at 400 °C, exceptions being the reactions performed with
SrMoO4 synthesized by the cm and sg methods. These results
as well as the obtained ones in noncatalytic fvp systems are
depicted in Table 1. The analysis of the pyrolysate indicated the
presence of products coming from nitrogen extrusion and other
fragmentations of the starting phenacyl benzotriazole 1. Thus,
the formation of the desired 7H-dibenzo[b,d]azepin-7-one (3)
can be rationalized by nitrogen elimination of 1 to give the
iminocarbene intermediate, which rearranges to 5,6-dihydro-
7H-dibenzo[b,d]azepin-7-one (2), followed by dehydrogen-
ation (Scheme 1, fragmentation a). Although azepinone 2 was
not isolated, this product was detected in the reaction mixture,
providing evidence of the proposed mechanism. Dehydrogen-
ation reactions have been obtained in this kind of pyrolysis
system, and many of transformations involve radicals as
intermediates.27
Product 3 was also formed in homogeneous fvp reactions in
which the greatest yield (30%) was obtained at 500 °C. In that
case, total conversion of 1 was achieved at 550 °C, giving 3 and
1-phenyl-2-(phenylimino)ethanone (5) as the main products.
This imine is proposed to be formed by the same iminocarbene
precursor as 2 after rearrangement involving hydrogen
migration. On the other hand, heterogeneous mesoporous-
catalyzed fvp of 1 led to the synthesis of 3 in very good yields.10
Using SrMoO4 synthesized by the dp method, total
conversion of 1 was achieved at 400 °C. This catalytic system
exhibited a reduction of 150 °C in the reaction temperature
with respect to the noncatalyzed pyrolysis. The same results
were obtained with catalysts that, after being precipitated, were
heat-treated at 600 °C for 12 h, which is reasonable, since
particle size and shape are similar to those not heat-treated (see
Supporting Information Figure S2). In the same way, the
material prepared by the cm method exhibited an acceptable
conversion of 1 (80%). In contrast, the oxide SrMoO4
synthesized by the sg method displayed a low conversion of
1. These results indicated that the surface morphology of the
catalyst prepared by the dp method was more reactive than the
others, even though the sample is heat treated at 600 °C. On
the basis of the results, the regular and small size distribution of
catalyst particles increased the conversion of 1, although the
selectivity of 3 has remained low.
Thus, the catalytic activity of SrMoO4 was strongly
dependent on the preparation method of the catalyst. This
could be attributed to the influence of the preparation method
on the formation of the active sites on the catalyst surface,
promoting different gas−solid interactions, and also to
differences in particle size, with the smaller particles having
the highest conversion.
Cfvp in the presence of CaMoO4 and BaMoO4 also displayed
total conversion of 1 at 400 °C. The yield of the desired 3 was
30% for reactions carried out over CaMoO4, the best result in
the catalytic pyrolysis using this experimental approach. In the
BaMoO4−cfvp systems, other reactions instead of the expected
N2 extrusion of 1 were promoted.
Regarding the tungstate scheelite-catalyzed reactions, total
conversion of 1 was also obtained at 400 °C, but the formation
of azepinone 3 was detected only when CaWO4 was used
(Table 1).
A secondary and competitive reaction was the decarbon-
ylation of 3 to give phenanthridine (4) in a typical
aromatization process. CO elimination usually requires more
than 600 °C in a typical gas phase thermolysis;28 however, this
elimination is commonly favored in cfvp systems.9,10 According
to the results, decarbonylation was mainly promoted in the
reactions using SrMoO4 (prepared by the cm method) and
CaWO4.
Other fragmentations involving rupture of C−N and C−C
bonds in benzotriazole 1 (Scheme 1, fragmentations b and c) as
well as in the azepinones 2 and 3 took place, and undesired
products were obtained. This kind of reaction was favored in
virtually all scheelite-catalyzed pyrolysis, decreasing the
selectivity of the process to the formation of the expected 3.
Thus, using experimental approach I, the best yields of this
azepinone were obtained in the calcium scheelites-catalyzed
reactions. In addition, these catalytic materials displayed greater
surface areas. This feature, together with the presence of the
Ca2+ ion, were found important in the formation of 3 through
this packed experimental setup.
Using this experimental arrangement, it seems that the total
number of active sites of the catalysts is too high because the
1023
dx.doi.org/10.1021/cs3008335 | ACS Catal. 2013, 3, 1020−1025

ACS Catalysis
Letter
mass balance was over 50−60%, and the high degree of
degradation of starting benzotriazole to volatile compounds as
well as the favored fragmentation processes could be attributed
to the exposure of molecules to a high concentration of
catalytic active sites of catalysts in a very short time.
3.2.2. Catalytic Reactions Using a Supported Catalyst
(Experimental Approach II). To minimize the total number of
active sites and, consequently, undesired fragmentations and to
improve the formation of 3, we have decreased the amount of
catalyst for the experiments, keeping a good surface area to
promote the substrate gas−solid catalyst interaction. Thus, we
evaluated experimental approach II, in which catalysts prepared
by the dp method were deposited on the ceramic fiber support.
This arrangement led to a ratio of catalyst/substrate between 1
and 2. Results obtained in this new catalytic system are shown
in Table 2.
tungstates-catalyzed reactions the empty d orbitals of the B6+
cation seem to be significant in the selectivity of the thermal
reaction. In addition, the selectivity to the formation of
azepinone 3 as a function of the A2+ cation followed the order
Ca2+ < Sr2+ ∼ Ba2+. These metal ions have empty s and p
orbitals (ns0, np0). It has being calculated that these empty
orbitals may accept the electron density from the electro-
negative nitrogen atoms of the phenacyl benzotriazole 1 in the
gas phase interaction (Figure 4), with the energy of adsorption
Table 2. Cfvp Reactions of 1 over Scheelites ABO4;
Experimental Approach II
Figure 4. Distribution of the electronic density of phenacyl
benzotriazole 1. Red orbitals represent a negative electronic density.
a
b
b
catalyst
% 1
% 3
being in the order Ca2+ < Sr2+ ∼ Ba2+.30 Thus, this indicates
that the selectivity toward azepinone 3 is intimately related to
the electron transfer from the substrate 1 to the A cation of the
scheelite.
These catalytic performances were more pronounced for
SrWO4 and BaWO4 since they favored the selective nitrogen
extrusion reaction of 1 and further cyclization. In the case of
cfvp experiments in the presence of fergusonite BaWO4
supported on fiber, the catalytic behavior of the oxide was
dependent on their morphology. Even acceptable yields of 3
were achieved using the vanadate prepared by the ultrasound
technique; the more regular particles obtained by the
microwave precipitation method showed a better catalytic
profile, giving 84% yield of the desired azepinone at 400 °C
(see Table 3). In both cases, the increment of the temperature
SrMoO4
CaMoO4
BaMoO4
0
40
0
20
54
15
65
82
50
87
c
BaMoO4
SrWO4
20
13
15
10
CaWO4
d
BaWO4
a
b
d
Catalyst/substrate ratio was between 1 and 2. Temperature: 400 °C.
Values determined by H NMR. Reused without heat treatment.
Total conversion of 1 was achieved at 450 °C, yielding 3 (45%).
c
1
It was noted that the catalytic activity of SrMO4 did not
change, and yields of 3 remained low. The main products were
formed by multiple and nonselective fragmentations of starting
benzotriazole. For CaMoO4- and BaMoO4-catalyzed reactions,
the formation of 3 was increased (up to 45% yield); however, in
the case of CaMoO4, the conversion of 1 was only 60%. These
results probably indicated that catalytic activity of molybdates
remained high in terms of transformation of 1; however, the
catalysis has not yet been specific toward the desired process.
To prove this assumption, cfvp was performed with reused
BaMoO4 without any activation treatment. Thus, the
conversion of substrate decreased to 80%, but surprisingly,
the yield of 3 increased to 65%, minimizing the formation of
competitive products. This finding would indicate that the
poisoning of the catalyst surface with consequent reduction of
activity makes the process more selective to nitrogen extrusion
reaction to give azepinone 3, reducing other feasible processes.
In the case of tungstate-catalyzed pyrolysis, a great
improvement in the formation of 3 was obtained. Thus, using
SrWO4 and BaWO4, quantitative yields of 3 were achieved
(>80%). Even the conversion of starting benzotriazole was not
complete; the selectivity toward the nitrogen extrusion reaction
followed by the dehydrogenation was substantially improved. In
all cases, decarbonylation of 3 was insignificant, and 4 was
obtained in trace yields (∼1%). Thus, lower amounts of
catalysts exposed to the gas phase substrate minimized the
multiple rupture of this compound, giving selectively the
elimination of molecular nitrogen in a typical [5 → 3 + 2]
fragmentation.29
Table 3. Cfvp Reactions of 1 over BiVO4; Experimental
Approach II
a
b
b
b
catalyst
T (°C)
% 1
% 3
% 4
c
BiVO4
BiVO4
BiVO4
BiVO4
400
450
400
450
35
20
15
0
63
40
84
52
2
0
1
1
c
d
d
a
b
The catalyst/substrate ratio was between 1 and 2. Values
c
d
1
determined by H NMR. Synthesized by US method. Synthesized
by MW method.
decreased the selectivity toward 3, favoring the other
fragmentation process as it was described for scheelite-catalyzed
systems.
The high activity displayed by this fergusonite-type oxide can
also be explained by the charge transfer from the electro-
negative nitrogen atoms of the phenacyl benzotriazole 1 to the
30
p orbitals (s orbitals are full) of Bi3+ that improve the
interaction with the electronegative nitrogen atoms (N2 and
N3) of the benzotriazole 1 (Figure 4).
4. CONCLUSION
In summary, scheelite ABO4 and fergusonite BiVO4 catalysts
constitute good materials to promote the conversion of
phenacyl benzotriazole 1 under cfvp conditions with high
The catalytic behavior can be analyzed in light of some
considerations about the nature of the A and B cations. In
1024
dx.doi.org/10.1021/cs3008335 | ACS Catal. 2013, 3, 1020−1025

ACS Catalysis
Letter
(13) Muktha, B.; Madras, G.; Guru-Row, T. N. J. Photochem.
Photobiol. A 2007, 187, 177−185.
selectivity to 3. These oxides are stable and reusable, and their
syntheses are easy and inexpensive. For cfvp systems, the
catalytic activity was strongly dependent on the morphology of
the oxide as a consequence of the synthetic method.
(14) Kloprogge, J.; Weier, M.; Duong, L.; Frost, R. Mater. Chem.
Phys. 2004, 88, 438−443.
(15) Errandonea, D.; Manjon, F. J. Mater. Res. Bull. 2009, 44, 807−
811.
(16) Wood, P.; Glasser, F. Ceram. Int. 2004, 30, 875−882.
(17) Ke, D.; Peng, T.; Ma, L.; Cai, P.; Jiang, P. Appl. Catal., A 2008,
350, 111−117.
Using the catalytic material supported on the ceramic fiber,
the selectivity of the process toward nitrogen extrusion of 1 was
significantly enhanced. The controlled deposition of particles
on the fiber improved the reactivity of the substrate and the
selectivity of the process. Differences in the catalytic behavior
were observed, depending on the nature of cations located in
the mixed oxide. Thus, tungstates AWO4 and BiVO4 favored
the formation of 3 with high selectivity. The selectivity to the
formation of azepinone 3 as a function of A2+ cation followed
the order Ca2+ < Sr2+ ∼ Ba2+.
In contrast, molybdates favored multiple C−C and C−N
fragmentations of 1. The use of inexpensive mixed oxides as
specific catalysts in cfvp reactions constitutes a new and
efficient “one-pot” procedure to synthesize the compound 3, a
precursor of potential bioactive dibenzazepinones.
(18) Liu, Y.; Huang, B.; Dai, Y.; Zhang, X.; Qin, X.; Jiang, M.;
Whangbo, M. Catal. Commun. 2009, 11, 210−213.
(19) Wash, A.; Yan, Y.; Huda, M.; Al-Jassim, M.; Wei, S. Chem.
Mater. 2009, 21, 547−551.
(20) Dhachapally, N.; Kalevaru, V.; Bruckner, A.; Martin, A. Appl.
̈
Catal., A 2012, 11, 443−444.
(21) Yu, J.; Zhang, Y.; Kudo, A. J. Solid State Chem. 2009, 182, 223−
228.
́
́
(22) Garcia Perez, U.; Sepulveda-Guzman, S.; Martínez-de la Cruz,
́
A.; Ortiz, U.; Mendez, J. J. Mol. Catal. A: Chem. 2011, 335, 169−175.
(23) Zhu, Z.; Du, J.; Li, J.; Zhang, Y.; Liu, D. Ceram. Int. 2012, 38,
4827−4834.
(24) Liu, W.; Cao, L.; Su, G.; Liu, H.; Wang, X.; Zhang, L. Ultrason.
Sonochem. 2010, 17, 669−674.
ASSOCIATED CONTENT
* Supporting Information
Detailed experimental procedures and characterization analysis
are summarized. This material is available free of charge via the
Internet at http://pubs.acs.org.
■
(25) Rietveld, H. J. Appl. Cryst. 1969, 2, 65−71.
(26) Jenkins, R.; Snyder, R.. In Introduction to X-Ray Powder
Diffractometry; Winefordner, J. D., Ed.; Wiley-Interscience: New York,
1996; Vol. 138, p 47.
S
(27) Pepino, A.; Pelae
2012, 3424−3430.
́
z, W.; Moyano, E.; Arguello, G. Eur. J. Chem.
̈
AUTHOR INFORMATION
Corresponding Author
*E-mail: lauramoy@fcq.unc.edu.ar.
(28) Brown, L. In Pyrolytic Methods in Organic Chemistry;
Wasserman, H., Ed.; Academic Press: New York, 1980, p 164.
(29) Rademacher, P. In Advances of Heterocyclic Chemistry; Katritzky,
A. R., Ed.; Academic Press: London, 1999, Vol. 72, p 361.
■
Notes
(30) Lener, G.; Vel
preparation
́
ez, P.; Leiva, E.; Moyano, E.; Carbonio, R. In
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
Thanks are given to Valeria Fuertes, Ph.D. for her significant
assistance and Prof. Luis Cadus for the surface area (BET)
determinations. Financial assistance from FONCYT, CONI-
CET, and SECyT-UNC is gratefully acknowledged.
■
́
REFERENCES
■
(1) Pozharskii, A.; Soldatenkov, A.; Katritzky, A. In Heterocycles in
Life and Society: An Introduction to Heterocyclic Chemistry, Biochemistry
and Applications, 1st ed.; Wiley: New York, 1997.
(2) Peters, J-U; Galley, G.; Jacobsen, H.; Czech, C.; David-Pierson,
P.; Kitas, E.; Ozmen, L. Bioorg. Med. Chem. Lett. 2007, 17, 5918−5923.
(3) Pegoraro, S.; Lang, M.; Dreker, T.; Kraus, J.; Hamm, S.; Meere,
C.; Feurle, J.; Tasler, S.; Prutting, S.; Kuras, Z.; Visan, V.; Grissmer, S.
̈
Bioorg. Med. Chem. Lett. 2009, 19, 2299−2304.
(4) Ishida, M.; Muramura, M.; Kato, S. Synthesis 1989, 562.
(5) Moore, J.; Theuer, W. J. Org. Chem. 1965, 30, 1887−1889.
(6) Paterson, W.; Proctor, G. J. Chem. Soc. 1962, 3468−3472.
(7) Moyano, E.; Eimer, G.; Lucero, P.; Chanquía, C.; Herrero, E.;
Yranzo, G. Appl. Catal., A 2010, 373, 98−103.
(8) Moyano, E.; Eimer, G.; Lucero, P.; Herrero, E.; Yranzo, G. Org.
Lett. 2007, 9, 2179−2181.
(9) Denis, J.; Gaumont, C. In Gas Phase Reaction in Organic Synthesis;
́
Vallee, J., Ed.; Gordon and Breach Science Publishers: Amsterdam,
1997, 195−204.
(10) Moyano, E.; Yranzo, G. J. Org. Chem. 2001, 66, 2943−2947.
(b) Moyano, E.; del Arco, M.; Rives, V.; Yranzo, G. J. Org. Chem. 2002,
67, 8147−8150.
(11) Angloher, G.; Bucci, C.; Cozzini, C.; Feilitzsch, F.; Frank, T.;
Hauff, D.; Henry, S.; Jagemann, T.; Jochum, J.; Kraus, H.; Majorovits,
B.; Ninkovie, J.; Petricca, F. Nucl. Instrum. Methods Phys. Res., Sect. A
2004, 520, 108−111.
(12) Han, Y.; Ueda, W.; Moro-Oka, Y. J. Catal. 1999, 186, 75−80.
1025
dx.doi.org/10.1021/cs3008335 | ACS Catal. 2013, 3, 1020−1025