Isomerization of 3-phenyl-1-propene (allylbenzene) over base catalysts

Article abstract of DOI:10.1016/S0021-9517(02)93733-4

The isomerization of 3-phenyl-1-propylene (allylbenzene) to 1-phenyl-1-propylene (β-methylstyrene) (cis + trans) was studied as a new test reaction for base catalysts. The injection of pure trans-β-methylstyrene (without catalyst) only yielded 1% of the other isomer (cis-βmethylstyrene). The injection of pure trans-β-methylstyrene, in the presence of catalysts, yielded small quantities of allylbenzene and cis-β-methylstyrene. Poisoning of the catalyst with CO₂ led to a sharp decrease in activity. The trans/cis ratio was ～ six for all the catalysts.

Full text of DOI:10.1016/S0021-9517(02)93733-4

Journal of Catalysis 211, 556–559 (2002)
doi:10.1006/jcat.2002.3733
RESEARCH NOTE
Isomerization of 3-Phenyl-1-Propene (Allylbenzene) over Base Catalysts
Mar´ıa A. Aramend´ıa, Victoriano Borau, Ce´sar Jime´nez, Alberto Marinas,1
Jose´ M. Marinas, and Francisco J. Urbano
Department of Organic Chemistry, University of Co´rdoba, Campus de Rabanales, Edificio C-3, Ctra Nnal. IV, Km. 396, 14014 Co´rdoba, Spain
Received April 10, 2002; revised June 18, 2002; accepted June 27, 2002
base and amphoteric characterization of heterogeneous
catalysts.
Finally, double-bond isomerization processes have also
Several magnesium containing catalysts were tested for the 2-
methyl-3-butyn-2-ol (MBOH) test reaction in a microcatalytic pulse
reactor. All solids showed a selectivity close to 100% toward prod-
ucts formed on basic sites (mainly acetone and acetylene) in the
433–623 K range. However, no correlation was found between mo-
lar conversion per gram of catalyst and basic site density, deter-
mined by temperature-programmed desorption–mass spectrome-
try of CO2 as the probe molecule. In the search for a process
that fulfilled such a correlation, a new model reaction for base
catalysts was found: isomerization of 3-phenyl-1-propene (allyl-
benzene) to 1-phenyl-1-propene (cis + trans). Therefore, a good
correlation (r = 0.986) between molar conversion (per gram of
catalyst) and basic site density was obtained at 523 K for all
c
been used for acid–base characterization. Thus, 1-butene
(16–18), 2-methyl-2-pentene (19), and 2,3-dimethylbut-2-
ene (20, 21) are some examples of the probe molecules.
The present piece of research deals with the use of the
transformation of 3-phenyl-1-propene (allylbenzene) as a
test reaction in order to characterize the basicity of various
solids consisting of one magnesium oxide. several mixed
magnesium–titanium oxides, and a magnesium–zirconium
oxide. Such a process is known to occur in the homogeneous
phase through a carbanionic process (22).
Moreover, Tsuji et al. (23) studied the process at room
temperature, in the liquid phase, for one magnesium oxide
and different zeolitic materials modified with several metal
oxides. Nevertheless they only found isomerization in the
case of pure magnesia, probably due to diffusion problems
inside the zeolitic micropores.
catalysts tested.
ꢀ 2002 Elsevier Science (USA)
Key Words: test reaction; 1-phenyl-1-propene; allylbenzene;
MBOH; 2-methyl-3-butyn-2-ol; acid–base characterization; double-
bond isomerization.
INTRODUCTION
Despite the widely documented literature on the use
of test reactions to characterize acidity in solids (mainly
heterogeneous catalysts), there is still very little devoted
to base characterization. Within this reduced group, aldol
condensation (1, 2) or retrocondensation (3) together with
Knoevenagel (4, 5) or Michael (6) reactions can be men-
tioned. However, it is important to point out that aldol con-
densation can also occur over acid catalysts, through a car-
bocationic process.
Some test reactions, such as alcohol transformation
(7–12) or alkylation of toluene with methanol or formalde-
hyde (13), can be used to determine both the acidity
and the basicity of catalysts. Moreover, there are some
other processes, such as the transformation of 2-methyl-
3-butyn-2-ol (14, 15), that allow us to carry out an acid,
METHODS
Synthesis of the Solids
Catalysts were synthesized using the sol–gel technique,
starting from Mg(NO3)2 · 6H2O, ZrOCl2 · 6H2O (Merck),
and TiO2 (Fluka). Different amounts of such chemicals
were dissolved in distilled water (in the case of TiO2, a sus-
pension was formed). Then NaOH was added until pH 10
was reached. The gel was aged for 72 h and then filtered
and air dried for 24 h and placed in an oven at 383 K. Calci-
nation of gels in a furnace at 873 K for 3 h, either in the air
(AIR) or in oxygen flow (OX), led to the solids in Table 1.
Nomenclature is as follows: oxides present are represented
by the symbol of the metal followed, in the case of mixed
oxides, by the atomic Mg/Zr or Mg/Ti ratio, as determined
by ICP (Perkin–Elmer Optima 3000 SC). Finally “AIR” or
“OX” indicate that calcinations were performed either in
the air or in the oxygen flow.
1 To whom correspondence should be addressed. Fax: +34 957212066.
E-mail: qo2maara@uco.es.
556
0021-9517/02 $35.00
c
ꢀ 2002 Elsevier Science (USA)
All rights reserved.

ISOMERIZATION OF 3-PHENYL-1-PROPENE OVER BASE CATALYSTS
557
TABLE 1
RESULTS AND DISCUSSION
Specific Surface Area (SBET) and Basic Site Density, as
Determined by TPD of CO2, for the Different Catalysts
Physical and Base Characterization of the Solids
BET surface areas of all the solids are shown in Table 1.
Pure magnesia catalysts (Mg-AIR and Mg-OX) consist of
periclase (bands at 2θ = 37.2, 43.1, and 62.5 in the X-ray
spectra). As far as magnesium–titanium oxides are con-
cerned, no mixed phases are detected except for MgO (per-
iclase) and TiO2 (mainly anatase), individually. The pres-
ence of TiO2 as anatase is confirmed by Raman spectra
(bands at 199, 398, 516, and 640 cm−1). Finally, MgZr27-OX
consists of MgO (periclase) and tetragonal ZrO2 (bands in
the X-ray spectrum at 2θ = 60.2, 50.6, 35.3, and 30.4). Mixed
oxides are not detected.
Basic site density (a.u./m2)
Low
Medium
strength
High
strength
SBET
strength
Catalyst
Mg–AIR
(m2 · g−1
)
(<673 K) (673–873 K) (>873 K) Total
53
47
52
67
78
74
5.3
6.7
7.2
8.8
8.3
6.4
0.9
0.3
2.4
1.1
2.2
3.0
1.4
2.4
2.1
4.0
2.1
6.2
7.6
9.4
11.7
13.9
12.6
15.6
Mg–OX
MgTi5–OX
MgTi19–OX
MgTi31–OX
MgZr27–OX
Results regarding base characterization are also included
in Table 1. An example of the TPD profiles is depicted in
Fig. 1. As can be seen, basic sites can be classified according
to their different strengths (different carbon dioxide des-
orption temperature). Therefore, there are low-strength
(desorption temperature below 673 K), medium-strength
(desorption between 673 and 873 K), and high-strength ba-
sic sites (desorption over 873 K).
Physical Characterization
The textural properties of solids were determined from
nitrogen adsorption–desorption isotherms at liquid nitro-
gen temperature by using a Micromeritics ASAP-2000
instrument. Surface areas were calculated by the BET
method.
X-ray investigation of the solids was carried out using
a Siemens D-500 diffractometer provided with an auto-
matic control and data acquisition system (DACO-MP).
The patterns were run with nickel-filtered copper radiation
(λ = 1.5405 A) at 35 kV and 20 mA; the diffraction angle
Reactivity Tests
Due to their base character all solids are carbonated as
confirmed by XPS and DRIFT. Therefore, prior to catalytic
tests the solids were heated at 823 K (50 K below their calci-
nation temperatures) for 30 min. Moreover, such a process
was monitored by mass spectrometry observing the loss of
water and carbon dioxide.
2θ was scanned at a rate of 2◦ min−1
.
FT-Raman spectra were obtained on a Perkin–Elmer
2000 NIR FT-Raman system with a diode-pumped NdYAG
laser (9394.69 cm−1). It was operated at a resolution of
4 cm−1 throughout the 3600- to 200-cm−1 range to gather
64 scans.
2-Methyl-3-butyn-2-ol (MBOH) test reaction. All solids
tested showed a selectivity close to 100% to the main
Base Characterization of the Solids
Base characterization of the solids was carried out by
temperature-programmed desorption–mass spectrometry
(TPD–MS), using carbon dioxide as the probe molecule,
as described elsewhere (15).
Reactivity Tests
Experiments were conducted in a microcatalytic pulse
reactor described in a previous paper (15). Experimental
conditions are as follows:
• MBOH pulses, 0.5 µL; temperature, 433–623 K; N2
flow, 75 mL min−1; FID detector (523 K); Supelcowax cap-
illary column (30-m length and 0.20-mm i.d.); oven temper-
ature, 333 K (3 min); increase in temperature up to 473 K
(at 8◦C min−1), final temperature kept for 5 min.
• Allylbenzene pulses, 0.1 µL; temperature, 433–623 K;
N2 flow, 75 mL min−1; FID detector (523 K); Supelcowax
capillary column (30-m length and 0.20-mm i.d.) and SPB-
FIG. 1. Temperature-programmed desorption–mass spectrometry
(TPD–MS) profile of catalyst Mg–OX deconvoluted into peaks corres-
50(60-mlengthand0.25-mmi.d.);oventemperature, 393K. ponding to low-, medium-, and high-strength basic sites.

´
ARAMENDIA ET AL.
558
FIG. 3. Allylbenzene test reaction. Correlation between molar
conversion/gram of catalyst and basic site density (a.u./m2), at 523 K, for
different catalysts.
FIG. 2. 2-Methyl-3-butyn-2-ol (MBOH) test reaction. Results found
for different catalysts expressed in terms of variation of molar conversion
with temperature. Catalyst weight was 20 mg in all cases.
the trans form is the main product, since it is more stable
than the cis one.
products formed on basic sites (acetone and acetylene)
within the 433- to 623-K range, which allows us to classify
them with base catalysts. Figure 2 shows the conversions
found for all the catalysts at four different temperatures.
Catalyst weight remained constant (20 mg). Any attempt
to correlate either molar conversion at low temperatures
with high-strength basic site density or molar conversion at
high temperatures with total basic site density did not lead
to good results. This is consistent with results reported by
Handa et al., who found that MBOH decomposition is not
a sensitive test reaction for estimating the nature of very
strong basic catalysts (20). According to Handa et al., these
results could be explained in terms of the strong adsorption
of acetylene and acetone on the catalyst. Moreover, as tem-
perature increases, some secondary products coming from
aldol condensation of acetone are formed, which contribute
to carbon deposition (11, 24).
Blank reactions (in the absence of the catalyst) did not
lead to isomerization of allylbenzene in the 443- to 623-K
range. Unlike blank reactions, allylbenzene underwent iso-
merizationinthepresenceofanyofthebasicsolids, yielding
β-methylstyrene (cis + trans). Chain isomerization, which
would have yielded α-methylstyrene, did not occur. Fi-
nally, three new experiments were carried out in order
to cast further light on the nature of the process. (i) The
injection of pure trans-β-methylstyrene (without catalyst)
only yielded 1% of the other isomer (cis-β-methylstyrene).
Therefore, the distribution of products is not purely de-
termined by thermodynamic equilibrium. (ii) The injection
of pure trans-β-methylstyrene, in the presence of catalyst,
yielded small quantities of both allylbenzene and cis-β-
methylstyrene. This is evidence of the existence of an equi-
librium among the three species, as expressed in Scheme 1.
(iii) Poisoning of the catalyst with carbon dioxide led to a
sharp decrease in activity. Thus, the process is catalysed by
bases.
Figure 3 depicts the favorable correlation (r = 0.986) be-
tween molar conversion (per gram of catalyst) and total
basic site density for all catalysts, at 523 K. The inclusion of
either only low- and medium-strength basic sites (r = 0.935)
or medium- and high-strength basic sites (r = 0.716) wors-
ened the results. This is hardly surprising since double-bond
isomerization is quite an easy process, not requiring a high
basic strength. At temperatures below 523 K, the different
products are greatly retained, which results in the tailing of
the peaks. On the other hand, at high temperatures (over
523 K) coke formation, and therefore catalyst deactivation,
is very important. The trans/cis ratio was about 6 for all the
catalysts.
3-Phenyl-1-propene (allylbenzene) test reaction. In the
search for a test reaction which allowed us to correlate
basic site densities with molar conversion, we considered
alkene isomerizations. These reactions seem to be a more
effective means to estimate the nature of strong basic sites
than the MBOH one (20). In this sense, isomerization of
3-phenyl-1-propene (allylbenzene) to 1-phenyl-1-propene
(β-methylstyrene), cis + trans (Scheme 1), could yield some
interesting results. According to Scheme 1, 3 (cis + trans
stereoisomers) is more stable than 1, due to the conjuga-
tion of the olefin bond with the aromatic ring. Moreover,
CONCLUSIONS
The 2-methyl-3-butyn-2-ol test reaction allowed us to
classify a series of catalysts as basic solids but no correlation
SCHEME 1. Mechanism for isomerization of allylbenzene in a basic
medium.

ISOMERIZATION OF 3-PHENYL-1-PROPENE OVER BASE CATALYSTS
559
between molar conversion and basic site density could be
drawn. In the search for such a correlation, isomerization
of 3-phenyl-1-propene (allylbenzene) to β-methylstyrene
(cis + trans) was proposed as a new test reaction for base
catalysts. A good correlation (r = 0.986) between molar
conversion (per gram of catalyst) and total basic site density
for all catalysts was obtained at 523 K. However, it must be
taken into account that β-methylstyrene (cis + trans) could
also be obtained over acid sites through a carbocationic pro-
cess. Therefore, allylbenzene isomerization might be used
as a test reaction for either base or acid catalysts but not am-
photeric ones since, in this case, two different mechanisms
(carbocationic + carbanionic) would be involved. Never-
theless, this is hardly surprising since the same can be said
of all alkene isomerizations. Even some other test reactions
4. Cabello, J. A., Campelo, J. M., Garc´ıa, A., Luna, D., and Marinas, J. M.,
J. Org. Chem. 49, 5195 (1984).
5. Corma, A., and Mart´ın-Aranda, R. M., Appl. Catal. 105, 231 (1993).
6. Aramend´ıa, M. A., Borau, V., Jime´nez, C., Marinas, J. M., and Romero,
F. J., Chem. Lett. 574 (2000).
7. Ai, M., J. Catal. 40, 318 (1975).
8. Aramend´ıa, M. A., Borau, V., Jime´nez, C., Marinas, J. M., Ruiz, J. R.,
and Urbano, F. J., Appl. Catal. A 206, 95 (2001).
9. Gervasini, A., and Auroux, A., J. Catal. 131, 190 (1991).
10. Bezouhanova, C. P., and Al-Zihari, M. A., Catal. Lett. 11, 245
(1991).
11. Aramend´ıa, M. A., Borau, V., Garc´ıa, I. M., Jime´nez, C., Marinas, A.,
Marinas, J. M., Porras, A., and Urbano, F. J., Appl. Catal. A 184, 115
(1999).
12. Cutrufello, M. G., Ferino, I., Monaci, R., Rombi, E., Colo´n, G., and
Nav´ıo, J. A., Phys. Chem. Chem. Phys. 3, 2928 (2001).
13. Yashima, T., Sato, K., Hayasaka, T., and Hara, N., J. Catal. 26, 303
(1972).
undeniably useful and extensively used for basic characteri- 14. Lauron-Pernot, H., Luck, F., and Popa, J. M., Appl. Catal. 78, 213
(1991).
zation of solids (such as aldol condensations) can also occur
on acid materials.
15. Aramend´ıa, M. A., Borau, V., Jime´nez, C., Marinas, J. M., Marinas, A.,
Porras, A., and Urbano, F. J., J. Catal. 183, 240 (1999).
16. Stevens, M. G., and Foley, H. C., Chem. Commun. 519 (1997).
17. Hattori, H., Appl. Catal. A 222, 247 (2001).
18. Be´res, A., Pa´linko´, I., Kiricsi, I., Nagy, J. B., Kiyozumi, Y., and
ACKNOWLEDGMENTS
Mizukami, F., Appl. Catal. A 182, 237 (1999).
19. Gao, S., and Moffat, J. B., Catal. Lett. 61, 61 (1999).
20. Handa, H., Fu, Y., Baba, T., and Ono, Y., Catal. Lett. 59, 195 (1999).
21. Ono, Y., and Baba, T., Catal. Today 38, 321 (1997).
22. Isaacs, N. S., “Reactive Intermediates in Organic Chemistry.” Wiley,
New York, 1975.
The authors gratefully acknowledge the financial support of Junta de
Andaluc´ıa and Ministerio de Ciencia y Tecnolog´ıa in the framework of
Project BQU2001-2605 (cofinanced with FEDER funds).
REFERENCES
23. Tsuji, H., Yagi, F., Hattori, H., and Kita, H., in “New Frontiers in Cata-
lysis” (L. Guczi, F. Solymosi, and P. Te´te´nyi, Eds.), p. 1171. Elsevier,
Amsterdam, 1993.
24. Lippert, S., Baumann, W., and Thomke, K., J. Mol. Catal. 69, 199
(1991).
1. Zhang, G., Hattori, H., and Tanabe, K., Appl. Catal. 36, 189 (1996).
2. King, F., and Kelly, G. J., Catal. Today 73, 75 (2002).
3. Campelo, J. M., Garc´ıa, A., Luna, D., and Marinas, J. M., Can. J. Chem.
62, 638 (1984).