Reactions of butylbenzene isomers on zeolite Hbeta: Methanol-to-olefins hydrocarbon pool chemistry and secondary reactions of olefins

Article abstract of DOI:10.1021/jp020811k

The reactions of n-butylbenzene, isobutylbenzene, sec-butylbenzene, and tert-butylbenzene on zeolite HBeta (SiO₂/Al₂O₃ = 150) were studied at 350 ?°C using a pulse reactor with GC-MS analysis of product gases. Similar experiments also probed the reactions of butylbenzene isomers with excesses of methanol-¹³C. The reactions of tert-butylbenzene were also explored as a function of zeolite acid site density and reactant loading. In the absence of secondary reactions such as oligomerization and cracking, olefin product selectivity is governed by the detailed structure of alkyl side chains; for example, tert- and isobutylbenzene each yielded isobutene, but sec- and n-butylbenzene each gave equilibrium mixtures of 1-butene and cis- and trans-2-butene. The rate of butene elimination from the benzene rings increased with the degree of branching on the carbon ?± to the aromatic ring as well as the extent of in sim ring methylation by co-fed methanol. In the presence of methanol-¹³C, isomerization between 2-butene and isobutene did not occur through a direct pathway; rather skeletal isomerization was observed after chain growth to C₇ or higher olefins followed by cracking. Coking reactions formed volatile alkanes through a mechanism that conserved the carbon skeleton of the precursor olefin. These studies relate to the hydrocarbon pool mechanism of methanol-olefin catalysis (MTO), as well as secondary reactions in MTO chemistry.

Full text of DOI:10.1021/jp020811k

8768
J. Phys. Chem. B 2002, 106, 8768-8773
Reactions of Butylbenzene Isomers on Zeolite HBeta: Methanol-to-Olefins Hydrocarbon
Pool Chemistry and Secondary Reactions of Olefins
Alain Sassi, Mark A. Wildman, and James F. Haw*
Loker Hydrocarbon Research Institute and Department of Chemistry, UniVersity of Southern California,
UniVersity Park, Los Angeles, California 90089-1661
ReceiVed: March 27, 2002; In Final Form: June 2, 2002
The reactions of n-butylbenzene, isobutylbenzene, sec-butylbenzene, and tert-butylbenzene on zeolite HBeta
(SiO2/Al2O3 ) 150) were studied at 350 °C using a pulse reactor with GC-MS analysis of product gases.
Similar experiments also probed the reactions of butylbenzene isomers with excesses of methanol-13C. The
reactions of tert-butylbenzene were also explored as a function of zeolite acid site density and reactant loading.
In the absence of secondary reactions such as oligomerization and cracking, olefin product selectivity is
governed by the detailed structure of alkyl side chains; for example, tert- and isobutylbenzene each yielded
isobutene, but sec- and n-butylbenzene each gave equilibrium mixtures of 1-butene and cis- and trans-2-
butene. The rate of butene elimination from the benzene rings increased with the degree of branching on the
carbon R to the aromatic ring as well as the extent of in situ ring methylation by co-fed methanol. In the
presence of methanol-13C, isomerization between 2-butene and isobutene did not occur through a direct pathway;
rather skeletal isomerization was observed after chain growth to C7 or higher olefins followed by cracking.
Coking reactions formed volatile alkanes through a mechanism that conserved the carbon skeleton of the
precursor olefin. These studies relate to the hydrocarbon pool mechanism of methanol-olefin catalysis (MTO),
as well as secondary reactions in MTO chemistry.
Introduction
the alkyl chain grows, rather it takes the alkyl chain as already
formed, and then follows it to products. This study then relates
to a second important question, and that is how does the structure
of the hydrocarbon pool relate to selectivity for specific olefin
products?
The long-standing question of which general mechanism
explains the catalytic conversion of methanol to hydrocarbons1-4
on microporous solid acid catalysts such as HZSM-5 and
HSAPO-34 has been answered. The groups of Kolboe5-9 and
Haw10-16 have independently shown, by a variety of methods,
that the correct general idea is an indirect mechanism involving
a “hydrocarbon pool” of intermediates which is methylated by
methanol or dimethyl ether and subsequently eliminates ethyl-
ene, propene, and butenes. Using highly pure reagents, we
recently demonstrated17 that methanol and dimethyl ether are
unreactive on HZSM-5 and HSAPO-34 in the absence of a
hydrocarbon pool, and this finding rules out all direct mecha-
nisms. While a variety of stable cyclic species are capable of
serving as reaction centers for methanol catalysis, the most
important constituents of the hydrocarbon pool are methylben-
zenes. If methylbenzenes can escape from the catalyst (as with
unmodified medium pore zeolite HZSM-5), methanol conversion
is said to take place under methanol-to-gasoline (MTG) condi-
tions. Alternatively, if the methylbenzenes are trapped in the
catalyst (as with the small-pore silico-aluminophosphate HSAPO-
34) and secondary reactions of the first-formed olefins are not
excessive, the result is methanol-to-olefin (MTO) catalysis.
Current research is directed at a more detailed understanding
of the structure and function of the hydrocarbon pool. One
important question is how does a methylbenzene hydrocarbon
pool species extend an alkyl chain that is subsequently
eliminated as an olefin? The two possibilities considered are
the paring reaction, originally proposed by Sullivan and co-
workers,18 and side-chain methylation, originally proposed by
Mole et al.19,20 The present contribution does not consider how
Butenes are the smallest olefins to exist in isomeric forms,
and so we explored the relationships between the structures of
C4 alkyl chains on alkylbenzene precursors and olefin product
distributions. We studied the reactions of all four isomeric
butylbenzenes on the large pore zeolite solid acid HBeta at 350
°C, with and without an excess of methanol-13C. We found that
the activity of a butylbenzene isomer for butene elimination
was related to the degree of branching on the carbon R to the
aromatic ring, and that the structure of the alkyl carbon chain
was strongly conserved in the volatile products even for the
least reactive isomers. For example, n-butylbenzene showed high
selectivity for 1-butene and 2-butene, but isobutene was formed
preferentially by isobutylbenzene. cis- and trans-2-Butene and
1-butene were in equilibrium under reaction conditions, but the
interconversion of linear butenes and isobutene required chain
growth (by oligomerization or reaction with methanol) followed
by cracking.21,22 Disproportionation of olefins to form alkanes
and coke was most significant at higher olefin concentrations
and higher zeolite acid site densities. Alkane formation occurred
with conservation of carbon skeletal structure; e.g., 2-butene
formed n-butane rather than isobutane.
Experimental Section
Materials and Reagents. Methanol-13C (99%) was obtained
from Cambridge Isotope Laboratories. n-Butylbenzene (99+%),
isobutylbenzene (99.8%), sec-butylbenzene (99%), tert-butyl-
benzene (98%), and 1,2,3,4-tetrahydronaphthalene (99%) were
* Corresponding author.
10.1021/jp020811k CCC: $22.00 © 2002 American Chemical Society
Published on Web 07/24/2002

Reactions of Butylbenzene Isomers on Zeolite HBeta
J. Phys. Chem. B, Vol. 106, No. 34, 2002 8769
Figure 2. The four experiments in the previous figure were repeated
using the alumina column to separate all four butene isomers: (a) tert-
butylbenzene, (b) sec-butylbenzene, (c) isobutylbenzene, and (d)
n-butylbenzene.
Figure 1. GC-MS total ion chromatograms (Petrocol DH 150 column)
from analyses of the volatile products exiting a catalytic reactor (300
mg zeolite HBeta, SiO2/Al2O3 ) 150, 350 °C) sampled 1.5 s following
pulsed introduction of 0.123 mmol of various pure butylbenzene
compounds. This loading corresponds to two molecules per acid site
in the catalyst bed. The short retention time regions (olefins and light
alkanes) are scaled as necessary for visualization of these products. (a)
tert-Butylbenzene was the most reactive isomer and it formed isobutene
with high selectivity. (b) sec-Butylbenzene was the second-most reactive
isomer, and it formed 2-butene with higher selectivity that did tert-
butylbenzene. (c) Isobutylbenzene was less reactive still and formed
more isobutene than 2-butene. (d) The reactivity of n-butylbenzene
(shown) was comparable to that of isobutylbenzene, but 2-butene was
formed preferentially in this case.
TABLE 1: Enthalpies of Formation in the Gas Phase of
Reactants and Products
compounds
∆fHg°as (kcal/mol)
1-butene
cis-2-butene
-0.15 ( 0.19
-1.83 ( 0.30
-2.58 ( 0.24
-4.29 ( 0.26
-3.30 ( 0.31
-5.15 ( 0.34
-4.17 ( 0.33
-5.42 ( 0.34
trans-2-butene
isobutene
n-butylbenzene
isobutylbenzene
sec-butylbenzene
tert-butylbenzene
purchased from Aldrich. Zeolyst International supplied the
samples of zeolite HBeta (BEA). These were: CP814E (SiO2/
Al2O3 ) 25), CP811E-75 (SiO2/Al2O3 ) 75), CP811E-150
(SiO2/Al2O3 ) 150), and CP811C-300 (SiO2/Al2O3 ) 300).
Catalysis. We used a benchtop microreactor system16 for all
experiments reported here. All metal components contacting the
catalyst, reactants, or products were stainless steel. He (600
sccm) was used as the carrier gas in all experiments. The reactor
(5 cm long, 0.65 cm internal diameter) was loaded with 300
mg of fresh HBeta for each experiment which was activated in
place immediately prior to use by heating at 500 °C for 1 h in
flowing helium. All tubing downstream of the chamber was
heated. A Valco valve was used to collect a gas sample 1.5 s
after each reagent injection.
Gas chromatography (Agilent 6890 Series GC system) with
mass spectrometric detection (Agilent 5973) was used to analyze
all reaction products. The ionization voltage was 69.9 eV and
the source temperature was 240 °C. In most of the results shown
the products were separated on a 150 m Petrocol DH 150 fused
silica capillary column (0.25 mm diameter, 1.0 µm film
thickness). A temperature program maintained the oven tem-
perature at 35 °C for an initial 25 min followed by a ramp of
20 °C/min to a final temperature of 290 °C. The above procedure
separated all species of interest with the exception of 1-butene
and isobutene, which co-eluted. These compounds have very
similar mass spectra; therefore, we repeated the studies of the
butylbenzene isomers alone using a second column optimized
for butene separations. This was a 30 m J&W scientific GS
alumina column (0.53 mm diameter) operated with a temperature
program starting at 120 °C and increasing to 200 °C with a 20
°C/min ramp. This procedure was used for the chromatograms
shown in Figure 2, and it was also used to correct the isobutene
yields for 1-butene in the data reported in Table 2.
Results
Butylbenzene Isomers Alone. Figure 1 presents GC-MS total
ion chromatograms (Petrocol DH 150 column) characterizing
all of the volatile products from the reactions of the four
butylbenzene isomers alone on zeolite HBeta (SiO2/Al2O3 )
150) catalyst beds at 350 °C. In each case 0.123 mmol of
butylbenzene (ca. 19 µL) was delivered onto a fresh catalyst
bed as a pulse, and the volatile products were sampled 1.5 s
later. tert-Butylbenzene was the most reactive; 96% conversion
was observed in Figure 1a. sec-Butylbenzene was slightly less
reactive; the conversion in this case was only 87% (Figure 1b).
Isobutylbenzene (Figure 1c) and n-butylbenzene (Figure 1d)
were much less active, with conversions of 13 and 11%,
respectively. Cumene (isopropylbenzene) formed as one of the
products of either tert-butyl- or sec-butylbenzene.
Figure 2 reports results from experiments identical to those
described above except using the alumina column to obtain
separations of all four butene isomers, and these chromatograms
most clearly show one of the essential findings of this

8770 J. Phys. Chem. B, Vol. 106, No. 34, 2002
Sassi et al.
TABLE 2: Product Selectivity Data for the Reactions of Butylbenzene Isomers on HBEA at 350 °C
i-C4
C4H8
butylbenzene
isomer
HBEA
butylbenzene
total C4
(SiO2/Al2O3)
conversion (%)
i-C4 + n-C4
selectivity
C4H8 + C4H10
n-
150
150
150
150
25
11
13
83
96
100
99
0.24
0.68
0.47
0.80
0.92
0.75
0.81
0.77
85
88
66
60
71
57
69
80
0.80
0.63
0.69
0.48
0.12
0.57
0.56
0.95
iso-
sec-
tert-
tert-
tert-
tert-
tert-
75
300
97
100
300a
a 9.5 mL of tert-butylbenzene (one-half of normal) was pulsed on 300 mg of catalyst.
Figure 4. GC-MS total ion chromatograms (Petrocol DH 150 column)
from experiments probing the reactions of butylbenzenes with methanol-
13C. The ratio of methanol-13C to butylbenzene was 5:1 (mol:mol). All
experiments shown were carried out at 350 °C using HBeta with SiO2/
Al2O3 ) 150 and gas sampling at 1.5 s. Methanol conversion was in
all cases less than quantitative. (a) tert-Butylbenzene, (b) sec-butyl-
benzene, (c) isobutylbenzene, (d) n-butylbenzene. Selected isotopic
distribution data from these experiments are reported in Table 3.
Figure 3. GC-MS total ion chromatograms (Petrocol DH 150 column)
from experiments probing the reactions of tert-butylbenzene on 300
mg catalyst beds of HBeta at 350 °C under various conditions. (a) 0.123
mmol on a catalyst with SiO2/Al2O3 of 300. Reducing the reactant
loading a factor of 2 greatly reduced all secondary reactions and
afforded isobutene with a very high selectivity. (b) 0.123 mmol on a
catalyst with SiO2/Al2O3 of 300. The olefin yield was higher with the
lower acid site density. (c) 0.123 mmol on a catalyst with SiO2/Al2O3
of 75. Coking was also extensive with this acid site density. (d) 0.123
mmol on a catalyst with SiO2/Al2O3 of 25. Extensive coking on this
catalyst consumed much of the olefinic products and left isobutane as
the predominant C4 product.
butylbenzene on catalyst beds of HBeta zeolites with various
acid site densities. On the highest acid site density studied (SiO2/
Al2O3 ) 25, Figure 3a) the predominant volatile products were
benzene and isobutane. Clearly, this result was dominated by
secondary reactions leading to the formation of nonvolatile
aromatic coke from some of the primary olefinic products with
hydrogen transfer to form isobutane. Product selectivity using
a catalyst with SiO2/Al2O3 ) 75 (Figure 3b) was very similar
to that noted earlier with SiO2/Al2O3 ) 150 (Figure 1a). A
further modest reduction in secondary reactions occurred when
the acid site density was reduced to the lowest value tested
(SiO2/Al2O3 ) 300, Figure 3c). The largest reduction in alkane
formation was achieved when we reduced the tert-butylbenzene
loading by a factor of 2 on the lowest acid site density catalyst
(Figure 3d). In this case, alkane formation was almost com-
pletely suppressed.
investigation. The olefin isomers formed were largely prede-
termined by the structure of the alkyl chain on the benzene
derivative. Isobutene predominated with tert-butylbenzene and
isobutylbenzene, but linear butenes were obtained in high yields
from n-butybenzene and sec-butylbenzene. Table 1 summarizes
the gas-phase enthalpies of formation of the four butenes and
the four butylbenzenes.23 The stability order for the butenes is
isobutene > trans-2-butene > cis-2-butene > 1-butene. Inde-
pendent of the butylbenzene reactant we observed that the ratio
of trans-2-butene to cis-2-butene was always ca. 1.5, and the
ratio of 2-butene (cis- plus trans-) to 1-butene was always ca.
2.5. The fraction of isobutene was strongly dependent on the
reactant.
Figure 3 reports GC-MS total ion chromatograms (Petrocol
DH 150 column) of the volatile products obtained from tert-
Table 2 summarizes olefin and alkane product selectivity data
from the experiments in Figures 1 and 3. The total C4 selectivity

Reactions of Butylbenzene Isomers on Zeolite HBeta
J. Phys. Chem. B, Vol. 106, No. 34, 2002 8771
TABLE 3: 13C Distribution in the Products Formed by Reaction of a 5:1 Excess of 13C-Methanol and the Butylbenzene Isomers
on HBEA (SiO2/Al2O3 ) 150) at 350 °C
system
C3H8
i-C4H8
2-butene (cis-)
2-butene (trans-)
i-C4H10
n-C4H10
Σ C5H10
Σ C5H12
Σ [C6H12 + C6H14]
13C-CH3OH (excess) + n-butylbenzene
Total 13C %
13C0 (%)
13C1 (%)
13C2 (%)
13C3 (%)
13C4 (%)
13C5 (%)
13C6 (%)
18.6
68.3
13.7
11.7
6.2
13.4
69.1
16.1
7.8
4.7
1.9
7.5
81.9
10.6
4.3
1.9
1.3
7.7
81.5
10.6
4.5
29.2
44.8
19.6
16.6
12.5
6.6
10
29.2
16.9
47.5
18
10
5.5
33.9
14.3
45.1
15.9
11.3
8.4
38.9
10.5
20.3
37.7
8.4
9.1
7.2
81.4
2.5
12.7
3.3
0
2.1
1.2
2.1
5
6.7
13C-CH3OH (excess) + isobutylbenzene
Total 13C %
13C0 (%)
13C1 (%)
13C2 (%)
13C3 (%)
13C4 (%)
13C5 (%)
13C6 (%)
16.9
63.1
24.1
11.7
1.1
17.4
61.3
18.4
11.4
7.1
16.0
64.5
15.5
12.5
5.3
15.6
63.5
20.9
7.2
15.8
65.2
16.7
9.8
12.0
73.6
7.9
16.8
1.6
33.5
13.9
39.9
22.4
14
8.3
1.5
28.5
15.4
51.9
14.2
14.4
1.7
41.5
12.8
11.3
35.5
19.5
7
5.5
6.8
1.8
2.1
2.7
1.6
0
2.4
5.1
9.1
13C-CH3OH (excess) + sec-butylbenzene
Total 13C %
13C0 (%)
13C1 (%)
13C2 (%)
13C3 (%)
13C4 (%)
13C5 (%)
13C6 (%)
41.3
25.9
35.4
27.7
11
22.5
50.1
22.4
17.1
8.2
10.5
76.4
11.6
7.4
3.4
1.3
10.5
76.4
11.5
7.6
33.8
24.3
35.3
23.3
14.8
2.2
12
33.1
13.2
40.4
23.4
15
6.3
1.6
33.0
15.4
45.6
22.9
13.3
6.8
36.4
11.1
20.1
33.7
17.6
10.8
5.3
77.7
7.7
14.2
3.9
0
3.4
1.3
2.2
1.3
1.4
13C-CH3OH (excess) + tert-butylbenzene
Total 13C %
13C0 (%)
13C1 (%)
13C2 (%)
13C3 (%)
13C4 (%)
13C5 (%)
13C6 (%)
57.6
10.5
29.3
37.0
23.2
33.3
38.2
19.5
19.9
15.8
6.6
43.5
22.2
22.2
24.9
20.7
10
43.8
22.2
21.4
24.3
21.4
10.3
28.4
50.1
13.4
16.5
12.6
7.3
45.1
25.2
15.5
25.0
22.9
11.5
46.8
6.7
26.2
22.6
22.1
16.3
6.2
46.1
2.6
35.1
20.2
20.3
15.5
6.4
45.6
8.0
13.5
24.9
22.3
17.3
9.1
4.8
and olefin selectivity were inversely correlated with conversion
(or loading), reflecting the greater propensity for dimerization
and cracking as well as coke formation at higher transient olefin
concentrations. 76% of the C4 products from n-butylbenzene
were linear (i.e., 1-butene, 2-butene, and n-butane), and the
balance, 24%, were isobutene and isobutane (Table 2). A higher
fraction of branched products was obtained from sec-butylben-
zene (47%), but branched products clearly predominated with
either isobutylbenzene (68%) or tert-butylbenzene (80% on the
same catalyst used to test the other isomers). The data in Table
2 permit a more quantitative interpretation of the results in
Figure 3d. With a reduced loading of tert-butylbenzene on the
lowest acid site density catalyst, 80% of the volatile products
were C4 and 95% of these were olefins.
Reactions of Methylbenzenes with Methanol. The GC-MS
total ion chromatograms (Petrocol DH 150 column) in Figure
4 show that each butylbenzene isomer was more reactive when
co-injected with a 5:1 mole ratio of methanol-13C. For example,
the conversion of n-butylbenzene increased from 11% to 28%
with addition of methanol. With co-added methanol-13C the
aromatic products included not only benzene but also various
methylbenzenes with up to five 13C-labeled methyl groups. We
also observed some degree of ring methylation with retention
of the butyl group; for example, some of the smaller peaks in
Figure 4d are due to isomers of methyl- and dimethyl-n-
butylbenzene.
methanol-13C on methylbenzene reaction centers, and the
consequences of oligomerization and cracking and other second-
ary reactions. This complex scheme of reactions is reflected in
the carbon isotope distributions in the products. These are
reported in Table 3 for the four experiments in Figure 4.
Distributions are reported for propene, all C4 products except
1-butene, C5 olefins (summed over all isomers), C5 alkanes
(summed over all isomers), and the sum of all C6 olefin and
alkane products. Note that the carbon isotope distributions for
cis- and trans-2-butene were invariably identical.
Table 3 contains many numbers, but the essential results can
be understood by contrasting sec-butylbenzene and tert-butyl-
benzene. For the reactions of sec-butylbenzene with methanol-
13C 77% of the 2-butene and 78% of the n-butane molecules
contained no 13C atoms, but this was true for only 50% of the
isobutene and 38% of the isobutane molecules in that same
experiment. The isobutene result is an upper limit due to the
presence of a small amount of unresolved 1-butene, which would
have the same low 13C content as the 2-butene. A reverse
ordering was seen in the products from tert-butylbenzene and
methanol-13C; here only 22% of the 2-butene was free of 13C
label (25% of n-butane), but 13C0 isotopomers predominated
for the branched products (isobutene, 38%; isobutane, 50%).
In every experiment, the predominant C5 isotopomers had one
13Csthis was true for olefins and alkanes, and the predominant
C6 isotopomers had two 13C atoms. Also, in every case, propene
had a higher total 13C content than did the butenes.
The olefin and alkane products in Figure 4 reflect elimination
of the original butyl groups as butenes, reaction of these butenes
with methanol-13C, some additional olefin synthesis from
Tetrahydronaphthalene. In addition to the four butylbenzene
isomers studied above, we also considered 1,2,3,4-terahy-

8772 J. Phys. Chem. B, Vol. 106, No. 34, 2002
Sassi et al.
SCHEME 1
SCHEME 2
and crack to form butenes,21,22 as well as pentenes plus propene
(but not ethylene and hexenes). Dimerization and cracking no
doubt also accounts for some of the results reported here.
Table 3 shows that in the presence of methanol-13C, butenes
undergo chain growth to pentenes and hexenes with (typically),
one and two 13C atoms, respectively. There is no reason for
this process to stop with C6 olefins; further homologation to C7
olefins followed by cracking would produce isomeric butenes
and propene molecules with mixtures of 12C and 13C atoms.
Scheme 2 illustrated this overall process for the case of isobutene
and methanol-13C reacting to form 2-butene with two 13C atoms
and propene with one 13C.
dronaphthalene, which could react on zeolite HBeta by at least
two routes as described in Scheme 1.
Tetrahydronaphthalene could, in principle, ring open and then
eliminate butadiene, which is very reactive on zeolites.24
Alternatively, or in parallel with butadiene elimination and
reaction, terahydronaphthalene could loose hydrogen and form
naphthalene. 1,2,3,4-Terahydronaphthalene reacted on HBeta
(SiO2/Al2O3 ) 75) at 350 °C with the evolution of almost no
volatile products. We concluded that tetrahydronaphthalene
readily forms coke, and we did not study its chemistry any
further.
This is an important result; it shows that when the concentra-
tion of olefins in the catalyst is sufficiently high, methylation
of these olefins by methanol/DME competes with methylation
of aromatic rings. Aromatic ring methylation leads to ethylene
and propene, through previously reported hydrocarbon pool
routes. Under the chain growth conditions observed here the
olefins must also be considered as reaction centers in the
hydrocarbon pool.
cis- and trans-2-Butene invariably had identical carbon
isotopic distributions, but isobutene had distinct distributions
(Table 3). The butane isomers also had distinct carbon isotope
distributions that mirrored those of the corresponding olefins.
Thus, the formal steps of protonation and hydride transfer to
form butanes from butenes occurred without skeletal isomer-
ization.
Cumene Formation. Industrial processes for cumene syn-
thesis co-feed propene and benzene on zeolite HBeta at
temperatures somewhat lower than those used in this investiga-
tion.26,27 Cumene formed here in those experiments where the
propene concentration was high; in particular, it formed from
tert-butylbenzene and sec-butylbenzene which gave highest
olefin concentrations in the catalyst, and thus propene by way
of olefin equilibration. The enthalpy of reaction for making
Discussion
Equilibration of Reactants and Products. Table 1 shows
the relative stabilities of reactant and product species in the gas
phase. tert-Butylbenzene is the most stable reactant isomer and
this underscores the fact that its greater reactivity has a kinetic
origin. There was no evidence of side chain isomerization prior
to olefin elimination. A very small amount of tert-butylbenzene
was seen in some experiments with sec-butylbenzene, but this
could be accounted for by a sequence of elimination, butene
isomerization, and alkylation, in analogy to the route leading
to cumene (vide infra).
1-Butene is the least stable butene isomer, and it always
formed in lower yield than 2-butene. Previous work has shown
that 1-butene isomerizes to 2-butene on acidic zeolites at
temperatures below 0 °C without H/D exchange between the
olefin and the acid site.25 n-Butylbenzene presumably eliminated
1-butene as the primary product and this rapidly equilibrated
with 2-butene. The constant relative yields of cis- and trans-
2-butene as well as their identical carbon isotope distributions
in all experiments in Table 3 show that these were also in
equilibrium with each other.
Elimination of Butyl Side Chains from Aromatic Rings.
In a recent investigation we showed that cumene was more
reactive than ethylbenzene on zeolite HBeta catalyst.16 The
former eliminated propene and the latter ethylene. Co-injection
of methanol-13C increased the production of propene-13C0 from
cumene and ethylene-13C2 from ethylbenzene by methylating
the aromatic rings. Here we observed the activity order tert- >
sec- > iso- ≈ n- for olefin elimination from butylbenzenes in
zeolite HBeta; this is the same ranking that one would anticipate
based on analogies to chemistry in acidic solutions. Ring
methylation increased the rates of butene elimination, consistent
with classical substituent effects in organic chemistry.
Thus, under MTO reaction conditions the structure of the
alkyl chain was conserved during olefin elimination. In the
absence of secondary reactions, such as dimerization and
cracking, the structure of the butene product (linear or branched)
was predetermined by the structure of the alkyl chain. This result
further supports the connection between the structure of the
hydrocarbon pool and olefin product selectivity.
cumene from benzene and propene in the gas phase is -23.76
23
kcal/mol,
and this is slightly more exothermic than the
corresponding reactions to form butylbenzenes. For example,
the formation of tert-butylbenzene from isobutene and benzene
has a very slightly less favorable ∆H of -20.95 kcal/mol. Thus
under conditions in which secondary reactions lead to compa-
rable concentrations of propene and isobutene in the catalyst,
cumene can be seen under conditions where the equilibrium
concentration of tert-butylbenzene is much lower. Cumene
formation underscores the fact that benzene de-alkylation is
reversible in these experiments.
Conclusions
Under MTO reaction conditions in zeolite HBeta C4 side
chains are eliminated from benzene rings with retention of
carbon skeletal structure, and branching on the carbon R to the
aromatic ring increases the rate of elimination. With a high
concentration of butenes in the catalyst the rate at which
methanol and/dimethyl ether reacted with olefins was competi-
tive with the rate of benzene ring methylation. Thus, the olefins
functioned as components of the hydrocarbon pool. When the
Olefin Chain Growth Followed by Cracking. Butene
isomerization on several zeolite catalysts including Ferrierite
is believed to occur by dimerization to C8 olefins that rearrange

Reactions of Butylbenzene Isomers on Zeolite HBeta
J. Phys. Chem. B, Vol. 106, No. 34, 2002 8773
(9) Arstad, B.; Kolboe, S. J. Am. Chem. Soc. 2001, 123, 8137-8138.
(10) Goguen, P. W.; Xu, T.; Barich, D. H.; Skloss, T. W.; Song, W.;
Wang, Z.; Nicholas, J. B.; Haw, J. F. J. Am. Chem. Soc. 1998, 120, 2651-
2652.
(11) Haw, J. F.; Nicholas, J. B.; Song, W.; Deng, F.; Wang, Z.;
Heneghan, C. S. J. Am. Chem. Soc. 2000, 122, 4763-4775.
(12) Song, W.; Haw, J. F.; Nicholas, J. B.; Heneghan, K. J. Am. Chem.
Soc. 2000, 122, 10726-10727.
(13) Song, W.; Fu, H.; Haw, J. F. J. Am. Chem. Soc. 2001, 123, 4749-
4754.
butylbenzenes were reacted with methanol, butene isomerization
was observed and it occurred by chain growth followed by
cracking. In the absence of methanol, butene isomerization
apparently occurred by oligomerization and cracking, as ob-
served in previous work, and it was suppressed by reducing
the amount of feed pulsed onto the catalyst. Disproportionation
of butenes to form butanes and coke was most rapid on catalysts
with a higher acid site density and it occurred with conservation
of the carbon skeletal structure.
(14) Song, W.; Fu, H.; Haw, J. F. J. Phys. Chem. B 2001, 105, 12839-
12843.
(15) Fu, H.; Song, W.; Haw, J. F. Catal. Lett. 2001, 76, 89-94.
(16) Sassi, A.; Wildman, M.; Ahn, H. J.; Prasad, P.; Nicholas, J. B.;
Haw, J. F. J. Phys. Chem. B 2002, 106, 2294-2303.
(17) Song, W.; Marcus, D. M.; Fu, H.; Ehresmann, J. O.; Haw, J. F. J.
Am. Chem. Soc. 2002, 124, Web Release in March, 2002.
(18) Sullivan, R. F.; Egan, C. J.; Langlois, G. E.; Sieg, R. P. J. Am.
Chem. Soc. 1961, 83, 1156-1160.
Acknowledgment. This work was supported by the National
Science Foundation (CHE-9996109) and the U.S. Department
of Energy (DOE) Office of Basic Energy Sciences (BES) (Grant
No. DE-FG03-93ER14354). Alain Sassi acknowledges the
Deutsche Forschungsgemeinschaft (DFG) for their financial
support.
(19) Mole, T.; Whiteside, J. A.; Seddon, D. J. Catal. 1983, 82, 261-
266.
(20) Mole, T.; Bett, G.; Seddon, D. J. Catal. 1983, 84, 435-445.
(21) Bearez, C.; Chevalier, F.; Guisnet, M. React. Kinet. Catal. Lett.
1983, 22, 405-411.
References and Notes
(1) Chang, C. D. Catal. ReV. 1983, 25, 1-118.
(22) Guisnet, M.; Andy, P.; Gnep, N. S.; Benazzi, E.; Travers C. J.
Catal. 1996, 158, 551-560.
(2) Sto¨cker, M. Microporous Mesoporous Mater. 1999, 29, 3-48.
(3) Keil, F. J. Microporous Mesoporous Mater. 1999, 29, 49-66.
(4) Chang, C. D. Shape-selective catalysis: chemicals synthesis and
hydrocarbon processing. Song, C., Garces, J. M., Sugi, Y., Ed.; ACS
Symposium Series 738, Washington DC, 2000.
(23) NIST Chemistry WebBook (http://webbook.nist.gov/chemistry/).
(24) Richardson, B. R.; Lazo, N. D.; Schettler P. D.; White, J. L.; Haw,
J. F. J. Am. Chem. Soc. 1990, 112, 2886-2891.
(25) Kondo, J. N.; Domen, K.; Wakabayashi, F. Microporous Mesopo-
rous. Mater. 1998, 21, 429-437.
(26) Ward, D. J. U. S. Patent 4,008,290, 1977.
(27) Smith, L. A., Jr.; Arganbright R. P.; Hearn D. U.S. Patent 5,055,627,
1991.
(5) Dahl, I. M.; Kolboe, S. J. Catal. 1996, 161, 304-309.
(6) Dahl, I. M.; Kolboe, S. J. Catal. 1994, 149, 458-464.
(7) Mikkelsen, O.; Ronning, P. O.; Kolboe, S. Microporous Mesopo-
rous Mater. 2000, 40, 95-113.
(8) Arstad, B.; Kolboe, S. Catal. Lett. 2001, 71, 209-212.