Acid strength controlled reaction pathways for the catalytic cracking of 1-pentene to propene over ZSM.5

Article abstract of DOI:10.1021/cs501967r

The influence of acid strength was evaluated toward the selectivity to propene on conversion of 1-pentene. For the catalytic cracking of 1-pentene, the main reaction pathways and the molar ratio of propene to ethene (P/E ratio) were controlled by acid strength with the appropriate amount of total acid sites. The results showed that the P/E ratio increased with decreasing amounts of strong acid sites, since the activation energies of individual reaction pathways were influenced by acid strength to a different extent. Strong acid sites could promote pathway I′ (C₅^2- → C₂^2- + C₃^2-) and pathway II′-1 (C₆^2- → C₂^2- + C₄^2-), while weak acid sites preferred pathway II′ (2C₅^2- → C₁₀⁺ → C₄^2- + C₆^2-) and pathway II′-2 (C₆^2- → 2C₃^2-), since pathways II′ and II′-2 underwent some energetically favorable routes (tertiary. secondary, secondary-secondary) of carbenium ion intermediates. By manipulation of the acid strength distribution on ZSM-5, the P/E ratio and selectivity of propene could be significantly improved, suggesting that this can provide an important guideline for improving such a process. In addition, we also designed a coupled process combing butene and pentene coconversion, as pentene and butene could be produced during C₄^2- and C₅^2- catalytic cracking. The coupled process could offer a promising solution to gain high selectivity of propene from C₄ and C₅ olefin cracking.

Full text of DOI:10.1021/cs501967r

Research Article
pubs.acs.org/acscatalysis
Acid Strength Controlled Reaction Pathways for the Catalytic
Cracking of 1‑Pentene to Propene over ZSM‑5
Long F. Lin, Shu F. Zhao, Da W. Zhang, Hui Fan, Yue M. Liu,* and Ming Y. He
Shanghai Key Laboratory of Green Chemistry and Chemical Processes, Department of Chemistry, East China Normal University,
North Zhongshan Road 3663, Shanghai 200062, People’s Republic of China
S
* Supporting Information
ABSTRACT: The influence of acid strength was evaluated toward the
selectivity to propene on conversion of 1-pentene. For the catalytic cracking of
1-pentene, the main reaction pathways and the molar ratio of propene to
ethene (P/E ratio) were controlled by acid strength with the appropriate
amount of total acid sites. The results showed that the P/E ratio increased with
decreasing amounts of strong acid sites, since the activation energies of
individual reaction pathways were influenced by acid strength to a different
extent. Strong acid sites could promote pathway I′ (C52− → C22− + C32−) and
2−
2−
pathway II′-1 (C6+ → C2 + C42−), while weak acid sites preferred pathway
II′ (2C5 → C10 → C4 + C62−) and pathway II′-2 (C6 → 2C32−), since
2−
2−
2−
pathways II′ and II′-2 underwent some energetically favorable routes (tertiary−
secondary, secondary−secondary) of carbenium ion intermediates. By
manipulation of the acid strength distribution on ZSM-5, the P/E ratio and
selectivity of propene could be significantly improved, suggesting that this can
provide an important guideline for improving such a process. In addition, we also designed a coupled process combing butene
2−
2−
and pentene coconversion, as pentene and butene could be produced during C4 and C5 catalytic cracking. The coupled
process could offer a promising solution to gain high selectivity of propene from C4 and C5 olefin cracking.
KEYWORDS: 1-pentene catalytic cracking, ZSM-5, reaction pathways, acid strength, propene
of 1-pentene. In comparison with the catalytic cracking of 1-
butene, more insights need to be gained into the different and
more complex reaction pathways of the catalytic cracking of 1-
pentene.
1. INTRODUCTION
Propene is one of the most indispensable petrochemicals.1 Due
to the rapidly growing demand for propene, the catalytic
cracking of low-value C4 and C5 olefins to propene has drawn
much interest.2−9 In addition, as the fluid catalytic cracking
(FCC)10 and methanol-to-olefin (MTO) processes11 produce
numerous low-value C4 and C5 olefins, the catalytic valorization
of these olefins is invariably anticipated. As this cracking
reaction contains various side reactions, such as dehydrogen-
ation−aromatization,12−15 hydrogen transfer reactions,12,13,16
dealkylation,6,12,13 and coke formation,17−20 the introductions
of phosphorus,21−26 alkali-metal ions,4,27−29 and alkaline-earth
metals30 were adopted to modify the acidity of zeolites and
then to suppress side reactions. Recently, we have systemati-
cally investigated the catalytic cracking of 1-butene to propene
over ZSM-531 and also found that the selectivity of propene
could be improved by reducing the acid strength of the zeolite.
It should be noted that around 21% pentene is a byproduct
with propene in the catalytic cracking of 1-butene.31 Thus, an
investigation of the catalytic cracking of 1-pentene rather than
1-butene is more significant. The side reactions of 1-pentene
catalytic cracking are similar to those of 1-butene catalytic
cracking,31,32 but still there are some essential differences. The
monomolecular cracking of butene is energetically highly
unfavorable,33 but a competition between monomolecular and
dimerization−cracking pathways exists in the catalytic cracking
In the monomolecular cracking pathway, 1-pentene under-
goes protonation and isomerization to form 2-pentyl carbenium
ions and/or 3-methyl-1-pentyl carbenium ions. Then β-scission
produces ethene and propene (Scheme 1).7−9 Apart from this,
the cracking of pentene can proceed through oligomeric
carbenium ion intermediates (C10+).9 Miyaji and co-workers7
have investigated the dimerization−cracking mechanism for the
conversion of pentene over SAPO-5 at the initial stage. First,
2,3,4,5-tetramethyl-2-hexyl carbenium ion (X) and 3,4,5-
trimethyl-3-heptyl carbenium ion (Y) are produced by the
dimerization of pentenes. The experimental results showed that
butenes were the main product, indicating that the β-scission of
decyl carbenium ions favored the production of butenes and
hexenes. After isomerization and cracking of decyl carbenium
ions, the generated hexenes continued to crack to propene or
ethene and butene.34 On the basis of previous studies, a
plausible reaction network of pentene conversion is shown in
Scheme 1. The main reactions include two pathways: pathway
Received: September 15, 2014
Revised: May 22, 2015
© XXXX American Chemical Society
4048
DOI: 10.1021/cs501967r
ACS Catal. 2015, 5, 4048−4059

ACS Catalysis
Research Article
a
Scheme 1. Possible Main and Side Reactions Network for 1-Pentene Cracking over ZSM-5
a
Abbreviations: Prot, protonation; Dep, deprotonation; Dime, dimerization; β-Scis, β-scission; Htf, hydrogen transfer; Mtf, methyl transfer; Isom,
isomerization; Olig, oligomerization; Dehy, dehydrogenation; Cyclis, cyclization; Alk(side), alkylation side chain; Alk(nucl), alkylation aromatic
nucleus.
Scheme 2. Reaction Pathways of Monomolecular Cracking and Dimerization−Cracking of 1-Pentene
2−
2−
2−
+
I′ (C5 → C2 + C32−) and pathway II′ (2C5 → C10
→
According to the main reaction pathways summed up in
Scheme 2, some essential theoretical estimation could be made
on the basis of the carbon balance. For example, if half of the
pentenes underwent monomolecular cracking (i.e., pathway I′)
C42− + C62−). As C62− could continue to crack, pathway II′ has
two subsequent branches: pathway II′-1 (C62− → C22− + C4
)
2−
34−36
and pathway II′-2 (C6 → 2C32−).
The side reactions of
2−
dehydrogenation−aromatization12−15 and hydrogen trans-
fer12,13,16 mostly produce aromatics, alkanes, and hydrogen.
In principle, side reactions should be suppressed, and then the
main pathway II′ and pathway II′-2 should be strengthened to
improve the selectivity of propene and of propene/ethene ratio
(P/E ratio).
2−
2−
with C2 and C3 as the products, then the others would
2−
undergo pathway II′, producing C4 and C62−. In addition, if
2−
the intermediate C6 were further transformed through
pathways II′-1 and II′-2 also in equal proportions, then the
P/E ratio would be 1.2 while the P/B ratio (propene/butene
ratio) would be 2. However, in actual reactions, it is difficult to
directly figure out the proportion each pathway accounts for.
Therefore, we supposed pathway II′ accounts for a, while
pathway I′ accounts for 1 − a. Meanwhile, for the conversion of
intermediate C62−, II′-2 was assumed to account for b and II′-1
for 1 − b. Then on the basis of the carbon balance, we could
improve the P/E ratio by controlling the acid strength of zeolite
according to a and b, as P/E = (1 − a + ab)/(1 − 0.5a −
0.5ab), while P/B = (1 − a + ab)/(a − 0.5ab). Hence, in this
paper, we attempt to control the main reaction pathways to
improve the P/E ratio by means of designing acid strength on
ZSM-5, which has been reported to be inverse proportion to
the reaction activation energy.
Miyaji et al.7 found that a spatial volume of zeolite that was
almost the same as the volume of pentyl cations promoted the
monomolecular cracking of pentene, while a larger zeolite
cavity favored the dimerization−cracking reaction pathway.
With ferrierite, the P/E mole ratio was around 1, because
monomolecular cracking of pentene was dominant. However,
ZSM-5 gave a P/E mole ratio of approximately 1.8, due to the
formation of the decyl carbenium ions.7 Bortnovsky et al.9
achieved similar results by investigating the cracking of 2-
methyl-2-butene using various zeolite and zeotype catalysts. To
the best of our knowledge, there is no report on how to control
the main reaction pathways of catalytic cracking of pentenes via
modifying the acid strength of zeolites, which is one of the key
factors of acid zeolites for improvement of the P/E ratio.
4049
DOI: 10.1021/cs501967r
ACS Catal. 2015, 5, 4048−4059

ACS Catalysis
Research Article
detectors which were used to detect hydrogen and nitrogen.
For simplicity, we grouped all types of pentenes as the overall
feedstock. The composition of products analyzed here by GC
are mole results (the composition of calibration gas used to
standardize GC is mole percent); the total amount is not
constant before and after reactions, but nitrogen does not
always change; thus, we use nitrogen as a reference to calculate
pentene conversion. Then, the mole conversions of pentenes
(equal to mass conversions), the mole selectivities of the
products, the mass selectivities of products, and the mass yields
of products were calculated by eqs 1−4.
2. EXPERIMENTAL SECTION
2.1. Catalyst Preparation. A series of ZSM-5 zeolites with
different Si/Al2 mole ratios were prepared by hydrothermal
synthesis and are denoted as ZSM-5(n), n representing the Si/
Al2 mole ratio detected by ICP. Typically, aluminum
isopropoxide (24.7 wt % Al2O3, Sinopharm Chemical Reagent
Co., Ltd.) was first dissolved in deionized water, into which was
then added tetrapropylammonium hydroxide (TPAOH, 25 wt
%, Sinopec Co., Ltd.) as the structure-directing agent. After the
mixture was stirred at room temperature for 2 h, tetraethyl
orthosilicate (28.4 wt % SiO2, Sinopharm Chemical Reagent
Co., Ltd.) was added, and then the mixture was stirred for
another 2 h, resulting in a gel with a chemical composition of 1
SiO2:x Al2O3:0.25 TPAOH:15 H2O (x was determined by the
desired Si/Al2 mole ratio). The gel was transferred into a 100
mL Teflon-lined stainless steel autoclave which was heated
statically at 170 °C for 48 h. The solid products were filtered,
washed with deionized water, dried overnight at 80 °C, and
finally calcined at 550 °C in air for 6 h.
(nC H )0 − (nC H
)
t
5
10
5 10
conversion =
× 100%
(nC H
)
0
5
10
(n )0 × 1/9.83 − (n )t × (C5H10%)t /(N2%)t
N2
N2
=
(nN )0 × 1/9.83
2
× 100%
ZSM-5 zeolites with different acid strength distributions were
designed and denoted Z-1−Z-5, respectively. Z-1 was prepared
by impregnating 10 g of ZSM-5(43) into 10 mL of phosphoric
acid aqueous solution containing 1.9 wt % phosphorus,
followed by drying at 50 °C for 24 h under vacuum and 120
°C for 4 h and finally calcination at 550 °C for 6 h.21,37 Z-2 with
1.6 wt % phosphorus was obtained from ZSM-5(55) using the
same method as for Z-1. ZSM-5(86) was leached with 6 mol
L−1 HNO3 at reflux for 5 h with continuous stirring, followed
by washing with water, drying, and then calcination at 550 °C
for 6 h.38 After all these procedures, Z-3 was successfully
obtained. Z-5 was achieved from ZSM-5(61) with the same
method, but a relatively lower acid concentration (2 mol L−1
HNO3) was employed to treat the ZSM-5(61). Z-4 was
obtained directly from ZSM-5(142) without any treatment.
2.2. Catalyst Characterization. The catalysts were
characterized by various techniques, including XRD, SEM,
BET, ICP, NH3-TPD, and 27Al and 31P solid-state MAS
NMR.31 Thermogravimetric (TG) analyses were carried out on
a Mettler Toledo TGA/SDTA851e instrument with a ramping
(C5H10%)
t
=100% − 9.83 ×
× 100%
(N2%)
t
(nN )0 = (nN )
t
(1)
(2)
2
2
(CxHy%)
t
mole selectivity (CxHy) =
mass selectivity (CxHy) =
× 100%
∑ (CiHj%)
t
W(CxHy)t
∑ (W(CiHj)t )
M(CxHy) × (nN )t × (CxHy%)t /(N2%)t
2
=
=
× 100%
(nN ) × ∑ (M(CiHj) × (CiHj%)t /(N2%)t)
t
2
M(CxHy) × (CxHy%)
t
∑ (M(CiHj) × (CiHj%) )
t
(3)
(4)
rate of 10 K min−1 in an air flow of 40 mL min−1. The Bronsted
̈
and Lewis acid sites of the samples were investigated by FT-IR
of adsorbed pyridine in an in situ cell with CaF2 windows.
Wafers with a weight of 25 mg and radius of 6.5 mm were
degassed for 1 h under vacuum at 600 °C. Then pyridine was
admitted, and after equilibration, the samples were outgassed
for 0.5 h at increasing temperatures (150, 200, 250, 350, and
450 °C). The spectra were recorded on a Nicolet iS50 FT-IR
spectrometer.
mass yield = conversion × selectivity
When we verified the law of conservation of mass, nitrogen
was also used as a reference. The carbon atom of input (mol
h−1), carbon atom of output (mol h−1), mass of input (g h−1),
and mass of output (g h−1) were calculated by eqs 5−8. Details
of the calculations are given in a discussion of the conservation
of mass and carbon atoms and Tables S1 and S2 in the
Supporting Information.
2.3. Catalytic Testing. Catalytic cracking of 1-pentene over
the zeolites were carried out in a stainless steel continuous-flow
reactor (10 mm i.d.), with a thermocouple in the center of the
catalyst bed. The catalyst was pressed, crushed, and sorted into
grains of 40−60 mesh and then was activated at 550 °C for 3 h
under a nitrogen flow before each reaction run. Then 1-pentene
(98%) in diluted nitrogen (N2/1-C5H10 = 9.83, mole ratio,
analyzed by GC) was passed through the reactor at the desired
temperature. The output products were analyzed online by an
Agilent 7890A gas chromatograph equipped with a Agilent
19095P-M25 column (50 m × 530 μm × 15 μm), a Dikma 2 m
× 1/8 5A column, a Dikma 2 m × 1/8 in. Porapak Q column,
two Dikma 1 m × 1/8 in. Porapak Q columns, and an FID
detector which was used to detect hydrocarbons and two TCD
1
9.83
input carbon atom =
× (nN )0 × 5
2
(5)
(6)
1
9.83
input total mass =
× (nN )0 × 70
2
⎛
⎝
⎞
⎠
(CxHy%)
t
⎜
⎟
⎟
output carbon atom =
× (nN ) × x
∑
⎜
t
2
(N2%)
t
(7)
x
4050
DOI: 10.1021/cs501967r
ACS Catal. 2015, 5, 4048−4059

ACS Catalysis
Research Article
corresponding to the total acid amount, whereas the adsorption
amount of pyridine, degassed at relatively high temperature of
350 °C, corresponded with the acid amount of strong acid.37 As
shown in Table 2, the total amounts of B acid sites of these five
⎛
⎝
⎞
(CxHy%)
t
⎜
⎜
⎟
⎟
output total mass =
× (nN ) × M(C H )
∑
t
x
y
2
(N2%)
⎠
t
x
(8)
a
Table 2. Bronsted Acid Properties of Z-1−Z-5
̈
Here (nC H )0, (nC H )t, (nN )0, and (nN )t represent the mole
5
10
5
10
2
2
flows of pentenes in the feed, of pentenes in the output, of
nitrogen in the feed, and of nitrogen in the output, respectively.
W(CxHy)t (x = 0, 1, 2, 3, 5, 6, 7, 8) and W(CiHj)t (i = 0, 1, 2, 3,
5, 6, 7, 8) represent the mass flows of a certain product in the
output and any product in the output, respectively. (C5H10%)t,
(N2%)t, (CxHy%)t, and (CiHj%)t represent the mole
compositions analyzed by GC of pentenes in the output, of
nitrogen in the output, of a certain product in the output, and
any product in the output, respectively. M(CxHy) and M(CiHj)
represent the molecular weights of a certain product and any
product, respectively.
zeolite
weak (au)
strong (au)
total (au)
strong/weak
Z-1
Z-2
Z-3
Z-4
Z-5
0.056
0.047
0.033
0.030
0.026
0.008
0.011
0.035
0.034
0.040
0.064
0.058
0.068
0.064
0.066
0.14
0.23
1.06
1.13
1.54
a
Detected by FT-IR spectroscopy after adsorption of pyridine.
samples were similar and the strong/weak acid ratios also
increased from Z-1 to Z-5. In addition, a semiquantitative study
of the strength of B acid sites was also carried out using FT-IR
spectroscopy with adsorbed pyridine. To evaluate this
parameter, we defined the ratio as the normalized area at
3. RESULTS AND DISCUSSION
3.1. Preparation and Characterization of Z-1−Z-5. In
order to investigate the influence of acid strength on the main
reaction pathways, a series of ZSM-5 zeolites with different acid
strengths were needed. However, as ZSM-5 had a complex
crystal structure with 12 different T sites39 in which the
aluminums had various proton energies corresponding to
different acid strengths,40,41 a distribution of the acid strength
on ZSM-5 instead of a single value is exhibited and could be
characterized by NH3-TPD. A typical NH3-TPD spectrum of
ZSM-5 was obtained with two maximum peaks at low and high
temperatures, corresponding to the weak and strong acid sites
of ZSM-5, respectively.42 Therefore, from the NH3-TPD result,
we can easily get the ratio between the amounts of strong acid
sites and weak acid sites (HC/LC ratio), which represents the
relative acid strength. Through phosphoric acid and nitric acid
modification, five ZSM-5 samples (Z-1−Z-5) were prepared. As
shown in Table 1, from Z-1 to Z-5, the HC/LC ratios increased
gradually from 0.244 to 2.627, and the amount of total acid sites
of each sample is around 0.19 mmol g−1.
temperature T, which is between the Bronsted absorption band
̈
area measured after pyridine adsorption followed by evacuation
at temperature T and at 150 °C. This quantity, calculated for B
acid site absorption bands, has been plotted as a function of
evacuation temperature in Figure 1. The more quickly the
Figure 1. Normalized area relative to Bronsted acid sites versus
evacuation temperature.
̈
Table 1. Acid Properties of Z-1−Z-5 Samples Measured by
NH3-TPD
a
b
LC (mmol
g−1
HC (mmol
g−1
total acid sites (mmol
g−1
HC/
LC
normalized area decreased, the weaker the acid strength.45 As
can be seen from Figure 1, the strength of B acid sites over Z-1
was the weakest, and from Z-1 to Z-5, the strength gradually
increased. Obviously, the results of FT-IR spectroscopy after
adsorption of pyridine fit well with the results of NH3-TPD.
The approaches of modification by phosphoric acid and
nitric acid were employed to obtain Z-1−Z-5 (except Z-4, see
the section 2.1), so that their acid strengths were greatly
different. NH3-TPD was used to characterize the acidities of
ZSM-5 modified by phosphoric acid, which showed that after
treatment the total amount of acid sites decreased and the
number of strong acid sites decreased noticeably while the
number of weak acid sites did not decrease as much.21,25,31 In
addition, the acidities of these samples were measured using
FT-IR spectroscopy with pyridine as the probe molecule. As
shown in Table 3, the amount of B acid sites decreased after
phosphoric acid treatment. More specifically, the number of
weak B acid sites remained almost the same while the number
of strong B acid sites decreased dramatically. On investigation
of the L acid sites, a similar phenomenon appeared. Therefore,
zeolite
)
)
)
Z-1
Z-2
Z-3
Z-4
Z-5
0.160
0.131
0.076
0.058
0.051
0.039
0.074
0.115
0.123
0.134
0.199
0.205
0.191
0.181
0.185
0.244
0.565
1.513
2.121
2.627
a
b
Low-temperature center, attributed to weak acid sites. High-
temperature center, attributed to strong acid sites.
It was generally believed that B acid sites played a key role in
the cracking of olefins.43 FT-IR spectroscopy with pyridine as a
probe molecule was used to characterize B acid sites. Generally,
the vibration band at around 1540 cm−1 was assigned to
pyridine adsorbed on the Bronsted (B) acid sites and the band
̈
around 1450 cm−1 to the Lewis (L) acid sites,44 which
represent the relative amounts of B and L acid sites,
respectively. The pyridine-adsorbed samples were degassed at
150 and 350 °C, respectively. The adsorption amount of
pyridine degassed at a low temperature of 150 °C,
4051
DOI: 10.1021/cs501967r
ACS Catal. 2015, 5, 4048−4059

ACS Catalysis
Table 3. Acidity Properties of ZSM-5 and Modified ZSM-5
Research Article
a
Bronsted acid (∼1540 cm−1
)
Lewis acid (∼1450 cm−1
)
Bronsted and Lewis acid
̈
̈
weak
(au)
strong
(au)
total
(au)
strong/
weak
weak
(au)
strong
(au)
total
(au)
strong/
weak
weak
(au)
strong
(au)
total
(au)
strong/
weak
b
zeolite
ZSM-5(55)
0.059
0.047
0.050
0.033
0.061
0.026
0.094
0.011
0.049
0.035
0.036
0.040
0.153
0.058
0.099
0.068
0.097
0.066
1.59
0.23
0.98
1.06
0.59
1.54
0.025
0.019
0.055
0.017
0.081
0.030
0.008
0.061
0.018
0.073
0.016
0.055
0.027
0.116
0.035
0.154
0.028
1.20
0.42
1.11
1.06
0.90
1.33
0.084
0.066
0.105
0.050
0.142
0.038
0.124
0.019
0.110
0.053
0.109
0.056
0.208
0.085
0.215
0.103
0.251
0.094
1.48
0.29
1.05
1.06
0.77
1.47
ZSM-5(55)-1.6P
ZSM-5(86)
ZSM-5(86)-D6
ZSM-5(61)
ZSM-5(61)-D2
0.012
a
b
Detected by FT-IR spectroscopy after adsorption of pyridine. Legend: ZSM-5(55)-1.6P, modified by phosphoric acid and containing 1.6 wt %
phosphorus, also denoted Z-2; ZSM-5(86)-D6, dealumination from ZSM-5(86) using 6 mol L−1 HNO3, also denoted Z-3; ZSM-5(61)-D2,
dealumination from ZSM-5(61) using 2 mol L−1 HNO3, also denoted Z-5.
a
Table 4. Catalytic Cracking of 1-Pentene at Different Temperatures over ZSM-5(142)
sel (mol %)
T (°C) conv(C52−) (mol %) TOF (h−1
)
C2
C3
C4
C6
4.9
C1 + C2 C3 −C5
H2 arom P/E (mol) P/B (mol)
a
b
2−
2−
2−
2−
0
0
0
0
450
475
500
525
550
93.6
94.6
96.3
97.5
98.5
567.1
573.1
583.4
590.8
596.8
19.3
23.7
28.2
35.2
41.3
36.1
34.2
35.5
36.3
36.5
20.7
16.9
14.9
12.8
10.6
0.5
1.0
1.1
1.5
1.7
13.5
14.6
9.2
1.6
2.5
2.8
3.4
3.4
3.5
4.0
4.0
4.5
2.0
1.9
1.5
1.3
1.0
0.9
1.7
2.0
2.4
2.8
3.4
0.688
0.563
0.467
0.342
0.229
0.688
0.623
0.569
0.379
0.027
3.2
4.3
1.3
0.7
5.1
3.8
a
Reaction conditions: cat., ZSM-5(142), 1.0 g; pressure, 0.1 MPa; 1-pentene flow rate, 0.2 mL min−1; N2 flow rate, 320 mL min−1; WHSV, 7.68 h−1;
TOS, 0.5 h.
the total number of weak acid sites decreased slightly but that
of strong acid sites experienced a remarkable decline, which was
consistent with the results of NH3-TPD.
ZSM-11 zeolites exhibited a high and stable selectivity to C2−
C4 olefins with a high conversion level. In this paper, we chose
ZSM-5 zeolite for the catalytic cracking of 1-pentene. In the
catalytic cracking of 1-butene, the side reactions could be
suppressed by decreasing the number of total acid sites of
zeolites. When ZSM-5(142) was chosen as the catalyst, the
number of total acid sites of which was 0.181 mmol g−1 and the
byproducts were suppressed to around 20%.31 Here, the
influence of temperature on catalytic cracking of 1-pentene was
studied over ZSM-5(142).
In comparison with the phosphoric acid modification, the
phenomenon of nitric acid treatment was rather different. After
treatment with 6 mol L−1 HNO3, the amounts of B acid and L
acid of ZSM-5(86) both decreased. With regard to acid
strength, it was interesting to note that regardless of B acid or L
acid, the numbers of strong acid sites and weak acid sites
decreased at the same step, resulting in almost no change in the
strong/weak acid ratio (Table 3). Previous NH3-TPD results
also showed that the HC/LC ratio remained almost unchanged
(ZSM-5(86), 1.697; ZSM-5(86)-D6, 1.513).31 After leaching
by 2 mol L−1 HNO3, an obvious decline in the number of weak
B acid sites in ZSM-5(61) was seen, whereas the number of
strong B acid sites remained stable. However, the numbers of
weak and strong L acid sites decreased to a similar extent.
Then, the strong/weak acid ratio of total B and L acid sites
experienced some rise from 0.77 to 1.47 (Table 3). Similarly, an
increase from 2.096 to 2.627 of the HC/LC ratio gained from
NH3-TPD was also observed.31 In conclusion, nitric acid
treatment favored the removal of weak acid sites or the removal
of weak and strong acid sites at a similar step, which depended
on the concentration of HNO3 and the parent zeolites, whereas
phosphoric acid modification favored a decrease in the number
of strong acid sites. Thus, using different parent zeolites and
employing different treatment methods, a series of zeolties (Z-
1−Z-5) with various acid strengths but similar numbers of acid
sites was successfully achieved. However, the XRD patterns,
27Al MAS NMR spectra, SEM, and BET showed that the other
physicochemical properties of Z-1−Z-5 were approximately
identical.31
Table 4 shows the reaction results of catalytic cracking of 1-
pentene at different temperatures over ZSM-5(142). When the
reaction temperature was increased, the conversion of pentenes
experienced a slight growth, as did the TOF. The total
selectivity of byproducts was not more than 22.0%, which
meant that the side reactions in the catalytic cracking of 1-
pentene were also suppressed over ZSM-5(142). The selectivity
of ethene showed a stable increase from 19.3% to 41.3% when
the reaction temperature was raised from 450 to 550 °C, while
the selectivity of butene decreased from 20.7% to 10.6% and
the selectivity of propene remained stable at 36%. It was
interesting to note that, at 500 °C, P/E = 1.3 and P/B = 2.4. In
2−
theory, if 50% of C5 underwent pathway II′ (a = 0.5) and
2−
50% of C6 underwent pathway II′-2 (b = 0.5), it would lead
to P/E = 1.2, and P/B = 2. In this experiment, a = 0.467 and b
= 0.569 were calculated from P/E = 1.3 and P/B = 2.4, which
are close to the theoretical values of a and b. Although the
experimental results are highly consistent with the theoretical
analysis, the difference in P/B ratios between the experimental
value (P/B = 2.4) and the theoretical value (P/B = 2) cannot
2−
be ignored. In theoretical analysis, we assumed that all the C4
produced during the cracking reaction were remained
3.2. Influence of Temperature. Bortnovsky et al.9
investigated the cracking of 2-methyl-2-butene by using various
zeolite and zeotype catalysts, including ZSM-5, ZSM-11, ZSM-
12, SAPO-11, ferrierite, mordenite, and beta. Only ZSM-5 and
unchanged, but in fact, a small part of C4 might take part
2−
in the following reactions. This is why the experimental value of
the P/B ratio is slightly higher than the theoretical value. When
the reaction temperature was decreased to below 500 °C, it
4052
DOI: 10.1021/cs501967r
ACS Catal. 2015, 5, 4048−4059

ACS Catalysis
Research Article
a
Table 5. Catalytic Cracking of 1-Pentene over Z-1−Z-5 at 450 °C
sel (mol %)
zeolite conv(C52−) (mol %) TOF (h−1
)
C2
C3
C4
C6
5.7
C1 + C2 C3 −C5
H2 arom P/E (mol) P/B (mol)
a
b
2−
2−
2−
2−
0
0
0
0
Z-1
Z-2
Z-3
Z-4
Z-5
84.5
86.7
93.5
93.6
94.1
465.9
464.1
537.1
567.1
557.6
12.0
13.7
19.0
19.3
20.5
41.9
40.7
37.2
36.1
31.3
31.6
28.9
21.9
20.7
17.5
0.1
0.2
0.4
0.5
0.7
6.7
8.8
0.4
0.5
1.1
1.6
2.1
1.7
1.6
2.5
3.5
5.0
3.5
3.0
2.0
1.9
1.5
1.3
1.4
1.7
1.7
1.8
0.904
0.858
0.706
0.688
0.618
0.731
0.730
0.694
0.688
0.618
5.6
4.9
4.9
4.3
13.1
13.5
18.5
a
Reaction conditions: cat., 1.0 g; temperature, 450 °C; pressure, 0.1 MPa; 1-pentene flow rate, 0.2 mL min−1; N2 flow rate, 320 mL min−1; WHSV,
7.68 h−1; TOS, 0.5 h.
Table 6. Output Distribution in Catalytic Cracking of Different Feedstocks over Z-1
output distribn (wt %)
2−
2−
2−
2−
2−
0
0
0
0
entry
feedstock
butene
C2
C3
C4
C5
C6
4.3
C1 + C2
C3 −C5
H2
arom
P/E (wt)
a
1
4.8
14.8
7.9
30.3
33.5
27.4
44.3
26.4
39.6
11.8
14.8
17.8
0.1
0.1
0.1
3.6
4.6
3.1
0
0
0
0.9
1.8
1.2
6.3
2.3
3.5
b
2
pentene
4.0
c
4
butene + pentene
2.9
a
Reaction conditions: cat., Z-1, 1.0 g; temperature, 500 °C; pressure, 0.1 MPa; 1-butene flow rate, 40 mL min−1; N2 flow rate, 320 mL min−1; TOS,
b
1.5 h. Reaction conditions: cat., Z-1, 1.0 g; temperature, 500 °C; pressure, 0.1 MPa; 1-pentene flow rate, 0.2 mL min−1; N2 flow rate, 320 mL min−1;
c
TOS, 0.5 h. Reaction conditions: cat., Z-1, 1.0 g; temperature, 500 °C; pressure, 0.1 MPa; 1-butene flow rate, 20 mL min−1; 1-pentene flow rate, 0.1
mL min−1; N2 flow rate, 320 mL min−1; mole ratio of 1-butene and 1-pentene, 1:1; TOS, 0.5 h.
gave rise to P/E > 1.3, and P/B < 2.4. In contrast, P/E < 1.3
and , P/B > 2.4 were found when the reaction temperature was
raised to beyond 500 °C. In addition, the values of a and b
declined when the reaction temperature was increased. As
for different reaction pathways were also increased to a different
extent.46 In this case, how does the acid strength influence the
main reactions of the catalytic cracking of pentenes? As the
main reaction network was complicated and the activation
energy of each step was difficult to evaluate by experiments, we
summed them up to investigate the apparent activation energy.
The steps for cracking of C5 molecules to C2 and C3 are
classified as pathway I′, and the apparent activation energy of
pathway I′ is denoted EI′. Similarly, pathway II′ includes the
dimerization steps of C52− molecules to C10+ and then cracking
to C42− and C62−. Pathway II′-1 contains the steps for cracking
of C62− molecules to C22− and C42−, and pathway II′-2 contains
the steps for cracking of C62− molecules to C32−. The apparent
activation energies of pathways II′, II′-1, and II′-2 are denoted
as EII′, EII′‑1, and EII′‑2, respectively.
2−
2−
pathway I′ (C5 ↔ C2 + C32−) is a reversible reaction,
raising the reaction temperature would benefit both the forward
reaction rate and reverse reaction rate, but with the forward
2−
2−
2−
2−
+
reaction rate being changed more. Pathway II′ (2C5 ↔ C10
→ C4 + C62−) is also a reversible reaction, but its reverse
reaction rate could be increased more than the forward reaction
rate when the temperature was raised. Therefore, raising the
temperature could promote pathway I′ while suppressing
pathway II′: i.e., the a value could be decreased. In conclusion,
the reaction temperature has a significant influence on the
reaction pathway: low temperature is beneficial to promote
pathways II′ and II′-2. Thus, the P/E ratio could be improved
by a reduction in the reaction temperature.
2−
Then there are two possible situations concerning the effect
of acid strength on activation energies of pathways I′ and II′. If
the acid strength displayed greater effect on EI′ than on EII′,
increasing acid strength would result in a faster decrease of EI′
in comparison to that of EII′. As we know, the reaction rate is
indicated by the reaction rate constant (k). Thus, as EI′
decreased more quickly than EII′, the reaction rate constant
of pathway I′ (kI′) would increase more quickly than that of
pathway II′ (kII′) according to the Arrhenius equation (k =
3.3. Influence of Acid Strength. In section 3.2, the values
of a and b were improved to 0.688 through a decrease in the
reaction temperature to 450 °C. In this section, we will
investigate the influence of acid strength on the main reaction
pathways in detail and find out whether it is possible to further
improve the values of a and b by adjusting the acid strength.
As shown in Table 5, catalytic cracking of 1-pentene over Z-
1−Z-5 at 450 °C was performed. With an increase in the acid
strength distribution, the conversion of pentenes increased, and
the TOF also showed an upward trend. A rise in the selectivity
of ethene was observed from 12.0% to 20.5%, while the
selectivity of propene and butenes decreased from 41.9% to
31.3% and from 31.6% to 17.5%, respectively. The P/E ratio
dipped to only 1.5 from 3.5 when the HC/LC ratio was
improved from 0.244 (Z-1) to 2.627 (Z-5); meanwhile, the P/B
ratio experienced a slight growth from 1.3 to 1.8. According to
P/E ratios and P/B ratios, a and b values were determined and
are given in Table 5. It is interesting to note that the a and b
values showed a downward trend with an increase in acid
strength.
Ae−E /RT),47 meaning that the improvement in the reaction rate
a
of pathway I′ is faster than that of pathway II′. Thus, improving
acid strength could promote pathway I′ while decreasing the a
value. In contrast, if the acid strength displayed a smaller effect
on the activation energy of pathway I′ in comparison to that of
pathway II′, the a value would be improved by an increase in
acid strength. Obviously, the experimental result that the a
value decreased with an increase in acid strength of ZSM-5
suggests that the first situation is reasonable. Therefore,
pathway II′ could be promoted by decreasing the number of
strong acid sites over ZSM-5.
For the relationship between pathways II′-1 and II′-2, there
are also two possible effects of acid strength on the apparent
activation energies, which could lead to opposite results. If the
effect of acid strength on the activation energy of pathway II′-1
It is known that reaction activation energies can be strongly
reduced if the acid strength is increased, and notably, the rates
4053
DOI: 10.1021/cs501967r
ACS Catal. 2015, 5, 4048−4059

ACS Catalysis
Research Article
Table 7. Output Mass and Conversion in Catalytic Cracking of Different Feedstocks over Z-1
output mass (g)
input mass
conv(C42−) (wt
%)
conv(C52−) (wt
%)
2−
2−
2−
2−
2−
0
0
0
0
entry
feedstock
butene
(g)
C2
C3
C4
C5
C6
2.4
C1 + C2 C3 −C5 H2 arom
a
1
56
70
2.7
10.4
13.0
9.9
17.0
23.4
40.4
34.6
24.8
18.5
43.3
49.9
6.6
10.4
17.0
22.5
0
2.0
3.2
5.2
3.9
0
0
0
0
0.5
1.3
1.8
1.6
55.7
b
2
pentene
2.8
5.2
3.6
0.1
0.1
0.1
85.2
75.8
67.9
c
3
butene + pentene
butene + pentene
126
126
22.7
11.1
d
4
a
b
Reaction conditions: the same as those in Table 6, entry 1; total input, 56 g of 1-butene. Reaction conditions: the same as those in Table 6, entry
c
d
2; total input, 70 g of 1-pentene. Hypothetical conditions: entries 1 and 2 in one reactor, but with no interplays between their reactions. Reaction
conditions: the same as those in Table 6, entry 4; total input, 56 g of 1-butene and 70 g of 1-pentene.
Table 8. Production Selectivity in Catalytic Cracking of Mixed Feedstocks over Z-1
sel (wt %)
2−
2−
2−
0
0
0
entry
feedstock
C2
C3
C6
7.9
6.8
C0 + C2
C3 −C5
H2
arom
P/E (wt)
a
3
butene + pentene
butene + pentene
19.8
18.4
61.5
64.3
0.2
0.2
8.0
7.3
0
0
2.7
2.9
3.1
3.5
b
4
a
b
Conditions: the same as those in Table 7, entry 3. Conditions: the same as those in Table 7, entry 4.
was greater than that of pathway II′-2, increasing acid strength
would give rise to a faster decrease of EII′‑1 in comparison to
that of EII′‑2, which means that strong acid sites preferred
pathway II′-1. Therefore, the b value would decrease with
increasing acid strength. In contrast, if the acid strength
displayed a smaller effect on the activation energy of pathway
II′-1 in comparison to that of pathway II′-2, the results would
be opposite. However, the experimental result that the b value
decreased with an increase in acid strength of ZSM-5 was
consistent with the first situation. Thus, weak acid sites
preferred pathway II′-2. In conclusion, when there is a decrease
in the number of strong acid sites of ZSM-5, the P/E ratio can
be improved because of the promotion of pathways II′ and II′-
2. On Z-1 with the lowest number of strong acid sites, a and b
values soared to 0.904 and 0.731, respectively, giving rise to a
value of 3.5 for the P/E ratio (Table 5).
cracking of 1-butene (Table 7, entry 1) and 1-pentene (Table 7,
entry 2), respectively. Then the supposed conversions of
butenes and pentenes could be calculated, which were 22.7%
and 75.8%, respectively, in an apparent decrease of both. This is
caused by the catalytic cracking of 1-butene to produce
pentenes and the catalytic cracking of 1-pentene to produce
butenes. Essentially, the actual capacity of conversion did not
decrease. However, in our practical experiment, if 56 g of 1-
butene and 70 g of 1-pentene were placed in one reactor, the
output mass of butenes and pentenes (Table 7, entry 4) were
higher than the supposed mass in Table 7, entry 3, with a
correspondingly lower actual conversion of butenes and
pentenes. According to Le Chatelier’s principle, an internal
reaction caused by an external disturbance would oppose the
initial change.48 As the catalytic cracking of 1-butene was a
reversible reaction and could produce pentenes if external 1-
pentene was added, on the basis of Le Chatelier’s principle, this
external disturbance would inhibit 1-butene conversion.
Similarly, adding 1-butene to the reactor in the catalytic
cracking of 1-pentene would also suppress its conversion.
Therefore, in the mixed system, the conversion of butenes and
pentenes both decreased.
3.4. Influence of Cofeed. Experimental results have
showed that a considerable amount of pentenes remained31
in the system of 1-butene conversion over Z-1, while numerous
butenes were produced in the conversion of 1-pentene, which
strongly urged us to investigate the influence of cofeed of 1-
butene and 1-pentene.
As shown in Table 6, the P/E ratio was 6.3 with only 1-
butene as the reactant, catalyzed by Z-1 zeolite, while it was 2.3
when 1-pentene was used as the only reactant. However, when
both 1-butene and 1-pentene were the reactants, the P/E ratio
was 3.5, which is between 2.3 and 6.3. This result indicated that
the dimerization of C42− and C52− to C92− may not happen in a
mixed system; otherwise, the P/E ratio would be higher than
6.3. In order to investigate more distinctly whether interplays
exist between the catalytic cracking of 1-butene and 1-pentene
in the mixed system, we attempted to do a quantitative analysis
directly. First, we supposed that 1 mol of 1-butene (56 g) and 1
mol of 1-pentene (70 g) were individually added to reactors
under the same conditions. According to the output
distribution in Table 6, the output mass of each substance
and the conversion of butenes or pentenes can be found (Table
7, entries 1 and 2). If there were no interplays between the
catalytic cracking of 1-butene and that of 1-pentene in the
mixed system, the output mass of each substance (Table 7,
entry 3) with both reactants in one reactor would be easily
calculated, which was just the sum of those in the catalytic
For extraction of the butenes and pentenes and normal-
ization of the production distribution, the selectivity of each
production in entries 3 and 4 is shown in Table 8. It is
interesting to note that the selectivity of each product in entry 3
is similar to the selectivity in entry 4. In addition, the P/E ratios
were 3.1 and 3.5, respectively, which are also similar. This
phenomenon indicated that the reaction pathways of catalytic
cracking of both 1-butene and 1-pentene experienced no
change in the mixed system. In conclusion, a cofeed of 1-butene
and 1-pentene had no significant influence on the catalytic
cracking reaction pathways but had some effects on the
conversion of 1-butene and 1-pentene.
3.5. Possible Reaction Mechanism for the Conversion
of 1-Pentene. To figure out the reasons weak acid sites favor
pathways II′ and II′-2, while pathways I′ and II′-1 require
strong acid sites, the possible reaction mechanism for the
conversion of 1-pentene was investigated.
The structural isomers of pentene consist of n-pentene and
isopentene. On one hand, n-pentene may form three kinds of
pentyl carbenium ions: namely, 1-pentyl carbenium ion, 2-
4054
DOI: 10.1021/cs501967r
ACS Catal. 2015, 5, 4048−4059

ACS Catalysis
Research Article
pentyl carbenium ion, and 3-pentyl carbenium ion. 1-Pentyl
carbenium ion can crack to ethene and 1-propyl carbenium ion
through β-scission. Modes of β-scission are classified by the
types of involved carbenium ions: A (tertiary−tertiary), B
(secondary−tertiary), C (secondary−secondary), D type
(primary−secondary), E (primary−tertiary), and F (primary−
primary).34 Type A is the most energetically favored form of
cracking, and type F, which involves only primary carbenium
ions, is energetically unfavorable. As 1-pentyl carbenium ion
and 1-propyl carbenium ion are primary carbenium ions, this
cracking process undergoes type F β-scission. Similarly, through
β-scission, 3-pentyl carbenium ion can crack to methyl
carbenium ion, which is energetically unstable. On the other
hand, four kinds of pentyl carbenium ions, i.e. 2-methyl-1-butyl
carbenium ion, 2-methyl-2-butyl carbenium ion, 3-methyl-1-
butyl carbenium ion, and 3-methyl-2-butyl carbenium ion, can
be converted from isopentene. The cracking of 2-methyl-1-
butyl carbenium ion undergoes type F β-scission, and the
cracking of 2-methyl-2-butyl carbenium ion and 3-methyl-2-
butyl carbenium ion can produce methyl carbenium ion. To
sum up, there are only two kinds of pentyl carbenium ions (2-
pentyl carbenium ion and 3-methyl-1-butyl carbenium ion)
whose cracking processes are energetically favorable.7−9 The
monomolecular cracking of 1-pentene is shown in Scheme 3
(pathway I′).
that of type D;34 therefore, we suppose that X preferentially
isomerizes to these 10 kinds of decyl carbenium ions (Scheme
S1a in the Supporting Information) whose cracking undergoes
type A, B, or C β-scission. The cracking of these 10 kinds of
decyl carbenium ions to butenes and hexyl carbenium ions or
hexenes and butyl carbenium ions is shown in Scheme 4. In
+
Scheme 4. Possible Forms of β-Scission of C10 (with Four
Methyls) over ZSM-5
+
Scheme 3. Possible Forms of β-Scission of C5 over ZSM-5
Miyaji and co-workers7 have investigated the dimerization
cracking of 2-methyl-2-butene at the initial stage over SAPO-5,
suggesting that first 2,3,4,5-tetramethyl-2-hexyl carbenium ion
(X) and 3,4,5-trimethyl-3-heptyl carbenium ion (Y) were
preferentially produced by dimerization of 3-methyl-2-butyl
carbenium ion with 2-methyl-2-butene at the monosubstituted
olefinic carbon and 2-methyl-1-butene at the unsubstituted
olefinic carbon, respectively.7 Isomerization of pentenes was
sufficiently faster than the cracking and/or dimerization to
maintain an equilibrium.49 Therefore, although the feedstock in
our experiments was 1-pentene, the dimerization reactions were
similar to those using 2-methyl-2-butene as feedstock,
producing decyl carbenium ions X and Y preferentially. Then
X and Y isomerized to other decyl carbenium ions which could
crack to butene and hexene though β-scission. As the
isomerization which followed the protonated cyclopropane
mechanism requires bulky intermediates, X and Y preferentially
isomerizes to the decyl carbenium ions with four and three
methyls, respectively. There are 11 kinds of structural isomers
of decene with four methyls. Correspondingly, 59 kinds of
decyl carbenium ions exist theoretically. Among all of the
forms, there are 24 kinds of decyl carbenium ions that can crack
to either butenes with hexyl carbenium ions or hexenes with
butyl carbenium ions. The cracking of 14 kinds of decyl
carbenium ions undergo type D, E, or F β-scission, which all
involve primary carbenium ions, while only 10 kinds of decyl
carbenium ions crack through type A, B, or C β-scission, which
involve secondary or tertiary carbenium ions. It has been
reported that the reaction rate of type C is 20 times faster than
terms of the decene with three methyls, 16 kinds of structural
isomers existed, from which 118 kinds of decyl carbenium ions
could be produced in theory. Among these, 46 kinds of decyl
carbenium ions could crack to butenes with hexyl carbenium
ions or hexenes with butyl carbenium ions, only 16 kinds of
which could occur through type A, B, or C β-scission (Scheme
5). These 16 kinds of decyl carbenium ions could be formed
through isomerization reactions from Y (Scheme S1b in the
Supporting Information). In total, there are 26 kinds of decyl
carbenium ions that could crack to butene or hexene and were
energetically favorable (pathway II′, Schemes 4 and 5). On
comparison of Scheme 3 with Schemes 4 and 5, it is obvious
that pathway I′ proceeded through a primary cation but
pathway II′ mainly proceeded through a secondary or tertiary
carbenium ion. This is why the Arrhenius activation energy of
the former is higher. Thus, pathway I′ needs strong acid sites to
reduce the relatively high activation energy. This mechanism
could reasonably explain the phenomena that pathway I′ was
promoted (a value decreased) when the HC/LC ratio was
improved.
The hexenes produced from pathway II′ could continue to
crack, and its structural isomers consist of one linear hexene,
4055
DOI: 10.1021/cs501967r
ACS Catal. 2015, 5, 4048−4059

ACS Catalysis
Research Article
+
+
Scheme 5. Possible Forms of β-Scission of C10 (with Three
Scheme 6. Possible Forms of β-Scission of C6 over ZSM-5
Methyls) over ZSM-5
secondary carbenium ion. Therefore, pathway II′-1 needed
strong acid sites, while weak acid sites preferred pathway II′-2.
The phenomenon that pathway II′-1 was promoted (b value
decreased) when the HC/LC ratio was improved could be
accounted for by this mechanism.
2−
3.6. Coupled Technological Route of C42−/C5
Catalytic Cracking. The work presented here could provide
a technological guideline for highly efficient production of
propene from C42−/C52−. Since in the cofeed system the
selectivity of propene and the P/E ratio were not improved
while the conversion of butenes and pentenes decreased, the
catalytic cracking of 1-butene and of 1-pentene should thus be
separated and coupled in order to improve the efficiency of
low-value C4 and C5 olefins.
It has been clearly revealed that the acid strength of zeolites
and reaction temperature influence the main reaction pathways.
Therefore, in order to maximize propene, we should choose the
optimized reaction conditions for the system of 1-butene (step
1) and 1-pentene conversion (step 2). First, with regard to the
acid strength, the results in both previous and present cases
have showed that Z-1 with the fewest amount of strong acid
sites was the best zeolite for producing propene in the catalytic
cracking of 1-butene31 or 1-pentene (Table 5). Definitely, we
chose Z-1 as the catalyst for steps 1 and 2. Second, over Z-1,
the effect of reaction temperature on the catalytic cracking of 1-
butene and 1-pentene is shown in Tables S3 and S4 in the
Supporting Information, respectively. Table S3 shows that with
an increase in temperature, the selectivity of propene remained
around 60%, and the selectivity of ethene experienced an
increase, which led to a decrease in the P/E ratio. Meanwhile, a
noteworthy downward trend was observed for the conversion
of butenes. Therefore, a suitable reaction temperature for step 1
was 450 °C. In addition, as shown in Table S4, the P/E ratio
also decreased with an increase in temperature, but the
selectivity of propene reached the highest point at 450 °C.
Hence, 450 °C was also the optimized temperature for step 2.
Consequently, for the high production of propene, we shall
choose the same reaction conditions (catalyst, Z-1; temper-
ature, 450 °C) for steps 1 and 2.
two methylpentenes, and two dimethylbutenes. Thus, 17 kinds
of hexyl carbenium ions exist theoretically, among which there
are 9 kinds of hexyl carbenium ions cracking through type F β-
scission or producing methyl carbenium ions. Among the
remaining 8 kinds of hexyl carbenium ions, 4-methyl-2-pentyl
carbenium ion, 2-hexyl carbenium ions, and 2,3-dimethyl-1-
butyl carbenium ions could crack to propene. The cracking of
the first undergoes type C β-scission,34 and the cracking of the
last two undergoes type D β-scission. As mentioned above, the
reaction rate of type C is 20 times faster than that of type D.34
Hence, pathway II′-2 mainly undergoes type C β-scission
(Scheme 6). All 5 kinds of hexyl carbenium ions that could
crack to ethene or butene undergo type D or E β-scission34
(Scheme 6, pathway II′-1). Apparently, pathway II′-1 involves
primary carbenium ions, but pathway II′-2 proceeds through
After the investigation of separate experiments for both step
1 and step 2 over Z-1 at 450 °C (Table 9), the process that
introduces pentenes, which are the products of step 1, to be the
reactant of step 2 while guiding the butenes produced in step 2
to be the reactant of step 1 was simulated. Finally, considering
4056
DOI: 10.1021/cs501967r
ACS Catal. 2015, 5, 4048−4059

ACS Catalysis
Research Article
Table 9. Reaction Results of Cracking of 1-Butene (Step 1) and Cracking of 1-Pentene (Step 2) over Z-1 at 450 °C
sel (wt %)
2−
2−
2−
2−
2−
2−
step
conv(C4 or C52−) (wt %)
C2
6.1
7.0
C3
C4
35.7
C5
21.0
C6
C0 + H2 + arom
a
1
64.7
85.2
51.5
35.8
11.0
9.7
10.4
11.8
b
2
a
Reaction conditions: cat., Z-1, 1.0 g; temperature, 450 °C; pressure, 0.1 MPa; 1-butene flow rate, 40 mL min−1; N2 flow rate, 320 mL min−1;
b
WHSV, 6.0 h−1; TOS, 1.5 h. Reaction conditions: cat., Z-1, 1.0 g; temperature, 450 °C; pressure, 0.1 MPa; 1-pentene flow rate, 0.2 mL min−1; N2
flow rate, 320 mL min−1; WHSV, 7.68 h−1; TOS, 0.5 h.
Scheme 7. Coupled Process of 1-Butene Cracking and 1-Pentene Cracking To Produce Propene
Table 10. Simulation Results of Coupled Process of 1-Butene Cracking and 1-Pentene Cracking
a
conditions
sel (wt %)
C0 + H2 + arom
C2 + C3
P/E (wt)
2−
2−
2−
2−
2−
route
step 1
step 2
C2
C3
C6
1
2
3
4
59.8, Z-1, 450
75.2, Z-3, 500
0, Z-3, 500
0, Z-1, 450
0, Z-3, 500
93.3, Z-3, 500
8.2
18.1
22.1
31.6
63.9
55.1
47.6
37.0
14.0
7.2
13.9
19.6
21.2
30.1
72.1
73.2
69.7
68.6
7.9
3.0
2.2
1.2
9.1
0, Z-5, 550
98.0, Z-5, 550
1.3
a
2−
Parameters in the following order: conv(C4 or C52−) (wt %), zeolite, T (°C).
Table 11. Reaction Results of Cracking of 1-Butene (Step 1) and Cracking of 1-Pentene (Step 2) over Z-5 at 550 °C
sel (wt %)
2−
2−
2−
2−
2−
2−
step
conv(C4 or C52−) (wt %)
C2
C3
C4
11.7
C5
2.9
C6
3.0
0.9
C0 + H2 + arom
a
1
85.2
98.3
24.6
28.6
42.4
31.9
27.1
26.9
b
2
a
Reaction conditions: cat., Z-5, 1.0 g; temperature, 550 °C; pressure, 0.1 MPa; 1-butene flow rate, 40 mL min−1; N2 flow rate, 320 mL min−1;
b
WHSV, 6.0 h−1; TOS, 1.5 h. Reaction conditions: cat., Z-5, 1.0 g; temperature, 550 °C; pressure, 0.1 MPa; 1-pentene flow rate, 0.2 mL min−1; N2
flow rate, 320 mL min−1; WHSV, 7.68 h−1; TOS, 0.5 h.
the effect of these two extra feedings, the reaction results of a
coupled technological route were calculated on the basis of
separate experiment data with the detailed calculation process
illustrated in Scheme 7. When the feeding mass of 1-butene was
supposed to be 100 g, there was 35.3 g of butenes unreacted
after the reaction in reactor 1, which would be reused in reactor
1 after separation. Meanwhile, 13.6 g of pentenes was produced
for further reaction in reactor 2 and 51.1 g of other products
were obtained. In reactor 2, no external pentenes were fed in.
Then it was calculated that 16.0 g of pentenes entered reactor 2
and 2.4 g of pentenes underwent backflow. In addition to 8.7 g
of products, 4.9 g of butenes was output, which was introduced
back to reactor 1 for further utilization. Finally, we needed to
introduce 59.8 g of external butenes to this coupled system. In
conclusion, the performance of route 1 after a combination of
products from steps 1 and 2 is shown in Table 10. The
selectivity of propene was 63.9%, and P/E ratio (wt) reached as
high as 7.9.
In route 1, the external pentenes were not fed in, because it
would lead to a decrease in the P/E ratio. On comparison of
the catalytic cracking performance of 1-butene with that of 1-
pentene (Table S5 in the Supporting Information), it was
found that, under the same conditions, the P/E ratio of the
former was higher than that of the latter. This was because part
of the pentenes underwent monomolecular cracking which
would lead to a P/E ratio of 1, but all butenes underwent
dimerization−cracking which led to a higher P/E ratio. Thus,
no external pentenes should be fed in route 1 in order to
achieve the highest P/E ratio.
4057
DOI: 10.1021/cs501967r
ACS Catal. 2015, 5, 4048−4059

ACS Catalysis
Research Article
Scheme 8. Coupled Process of 1-Butene Cracking and 1-Pentene Cracking To Produce Ethene
For designing a coupled technological route for producing
ethene (route 4), pentenes should be chosen as a feedstock,
while external butenes should not. Obviously, Z-5 with the
largest amount of strong acid sites was the best zeolite for
producing ethene from catalytic cracking of 1-butene31 or 1-
pentene (Table 5). In addition, a high reaction temperature was
beneficial to producing ethene (Tables S6 and S7 in the
Supporting Information). As a result, the same reaction
conditions (catalyst, Z-5; temperature, 550 °C) for steps 1
and 2 should be chosen for the high production of ethene.
According to the separated experimental data of steps 1 and 2
over Z-5 at 550 °C (Table 11), we simulated the coupled
process of route 4 (Scheme 8), and the results are shown in
Table 10. The selectivity of ethene was 31.6%, and the P/E
ratio (wt) reached the lowest point at 1.2.
Coupled technological routes with P/E ratios varying
between 1.2 and 7.9 could be realized through choosing a
suitable feedstock, designing a zeolite with suitable acid
strength, and adjusting the reaction temperature. As shown in
Table 10, feeding in 1-butene in route 2 performed over Z-3 at
500 °C gave a P/E ratio of 3.0. However, when accounting for
1-pentene, the P/E ratio decreased to 2.2 under the same
reaction conditions (route 3). From route 1 to route 4, the
selectivity of propene decreased, and the P/E ratio underwent a
sharp decline, but the total selectivity of ethene and propene
remained at around 70%. At present, the best performance in a
single reactor was achieved by using P-modified ZSM-5.22 The
total selectivity of ethene and propene was 74.4%, and the P/E
ratio was 3.4. However, in the coupled technological route 1,
the P/E ratio reached 7.9; meanwhile, the total selectivity of
ethene and propene remained at 72.1%. Thus, this coupled
system could provide advantages lying not only in the
utilization efficiency of low-value C4 and C5 olefins but also
in the P/E ratio.
pathways of catalytic cracking of 1-pentene were directly
controlled by the reaction temperature and acid strength of
ZSM-5 zeolite. High reaction temperature benefited pathway I′,
while pathway II′ could be promoted at low reaction
temperature. Thus, the P/E ratio could be improved through
a decrease in the reaction temperature. When the reaction
temperature was decreased to 450 °C, pathway II′ become
dominant (a = 0.688). Moreover, reaction pathways could be
significantly influenced by the acid strength of ZSM-5 zeolite.
Strong acid sites would promote the reaction pathways that
produced ethene (pathways I′ and II′-1), while weak acid sites
preferred the reaction pathways that formed propene (pathways
II′ and II′-2). When the HC/LC ratio decreased to 0.244 (Z-
1), cracking of 1-pentene mainly took place by pathway II′ (a =
0.904), with b also reaching 0.731 at the same time. Therefore,
it would be a wise choice to use a zeolite with a small number
of strong acid sites in order to get a high selectivity of propene.
In addition, the work presented here provides a guideline for
designing a coupled technological route that could produce
propene and ethene with various P/E ratios. Also, the P/E ratio
(wt) could be controlled between 1.2 and 7.9, with the total
mass selectivity of propene and ethene remaining at around
70% at the same time through adjusting the feedstock, reaction
temperature, and acid strength of zeolite. As the structure of
pentene or butene isomers has no influence on product
distributions when they are the reactants, although in our
experiment the feedstock was pure 1-pentene or 1-butene, the
achievement would still provide extensive applications in
industry. It could be anticipated that the technological process
which couples the two systems of butene and pentene
conversion would be a promising solution for cracking low-
value C4 and C5 olefins to propene with high utilization
efficiency.
4. CONCLUSIONS
In this paper, we have systematically investigated the catalytic
cracking of 1-pentene over ZSM-5. Although the essence of this
reaction that the reaction pathways could be controlled by acid
strength of catalyst through influencing the activation energy of
reaction is the same as that of the catalytic cracking of 1-butene,
the specific reaction pathways of catalytic cracking of 1-pentene
which contain monomolecular cracking and dimerization−
cracking pathways are quite different from the catalytic cracking
of 1-butene, which leads to different results. The main reaction
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/cs501967r.
Conservation of mass and carbon atoms, reaction scheme
of decyl carbenium ion isomerization, and reaction
results (PDF)
4058
DOI: 10.1021/cs501967r
ACS Catal. 2015, 5, 4048−4059

ACS Catalysis
Research Article
(28) Epelde, E.; Gayubo, A. G.; Olazar, M.; Bilbao, J.; Aguayo, A. T.
Ind. Eng. Chem. Res. 2014, 53, 4614−4622.
AUTHOR INFORMATION
■
Corresponding Author
*Y.M.L.: tel, +86-21-6223-2058; fax, +86-21-6223-2058; e-mail,
ymliu@chem.ecnu.edu.cn.
(29) Epelde, E.; Gayubo, A. G.; Olazar, M.; Bilbao, J.; Aguayo, A. T.
Chem. Eng. J. 2014, 251, 80−91.
(30) Wakui, K.; Satoh, K.; Sawada, G.; Shiozawa, K.; Matano, K.;
Suzuki, K.; Hayakawa, T.; Yoshimura, Y.; Murata, K.; Mizukami, F.
Catal. Lett. 2002, 84, 259−264.
Notes
The authors declare no competing financial interest.
(31) Lin, L. F.; Qiu, C. F.; Zhuo, Z. X.; Zhang, D. W.; Zhao, S. F.;
Wu, H. H.; Liu, Y. M.; He, M. Y. J. Catal. 2014, 309, 136−145.
ACKNOWLEDGMENTS
(32) Fottinger, K.; Kinger, G.; Vinek, H. Appl. Catal., A 2003, 249,
̈
■
205−212.
We gratefully acknowledge the National Key Technology R&D
Program (2012BAE05B02), the National Natural Science
Foundation Of China (U1462106), the Science and Technol-
ogy Commission of Shanghai Municipality (12JC1403600),
and the Shanghai Leading Academic Discipline Project (B409).
(33) Sun, Y. X.; Yang, J.; Zhao, L. F.; Dai, J. X.; Sun, H. J. Phys. Chem.
C 2010, 114, 5975−5984.
(34) Buchanan, J. S.; Santiesteban, J. G.; Haag, W. O. J. Catal. 1996,
158, 279−287.
(35) Guo, Y. H.; Pu, M.; Chen, B. H.; Cao, F. Appl. Catal., A 2013,
455, 65−70.
(36) Nawaz, Z.; Tang, X. P.; Zhu, J.; Wei, F.; Naveed, S. Chin. J.
Catal. 2009, 30, 1049−1057.
REFERENCES
■
(1) Koyama, T.; Hayashi, Y.; Horie, H.; Kawauchi, S.; Matsumoto,
A.; Iwase, Y.; Sakamoto, Y.; Miyaji, A.; Motokura, K.; Baba, T. Phys.
Chem. Chem. Phys. 2010, 12, 2541−2554.
(2) Wang, B.; Gao, Q.; Gao, J. D.; Ji, D.; Wang, X. L.; Suo, J. S. Appl.
Catal., A 2004, 274, 167−172.
(3) Zhao, G. L.; Teng, J. W.; Zhang, Y. H.; Xie, Z. K.; Yue, Y. H.;
Chen, Q. L.; Tang, Y. Appl. Catal., A 2006, 299, 167−174.
(4) Zhu, X. X.; Liu, S. L.; Song, Y. Q.; Xu, L. Y. Catal. Lett. 2005, 103,
201−210.
(5) Lu, Y.; He, M. Y.; Shu, X. T.; Zong, B. N. Appl. Catal., A 2003,
255, 345−347.
(6) Zhu, X. X.; Liu, S. L.; Song, Y. Q.; Xu, L. Y. Appl. Catal., A 2005,
288, 134−142.
(37) Blasco, T.; Corma, A.; Martinez-triguero, J. J. Catal. 2006, 237,
267−277.
(38) Moreno, S.; Poncelet, G. Microporous Mater. 1997, 12, 197−
222.
(39) Olson, D. H.; Kokotailo, G. T.; Lawton, S. L.; Meier, W. M. J.
Phys. Chem. 1981, 85, 2238−2243.
(40) Kramer, G. J.; van Santen, R. A.; Emeis, C. A.; Nowak, A. K.
Nature 1993, 363, 529−531.
(41) Rigby, A. M.; Kramer, G. J.; van Santen, R. A.; Aguayo, A. T. J.
Am. Chem. Soc. 1995, 117, 1766−1776.
(42) Post, J. G.; van Hooff, J. H. C. Zeolites 1984, 4, 9−14.
́
(43) Corma, A.; Orchilles, A. V. Microporous Mesoporous Mater. 2000,
35, 21−30.
(7) Miyaji, A.; Sakamoto, Y.; Iwase, Y.; Yashima, T.; Koide, R.;
Motokura, K.; Baba, T. J. Catal. 2013, 302, 101−114.
(8) Jeffrey, H. P.; Redondo, A.; Guo, Y. J. Catal. Today 1999, 50,
517−523.
(44) Emeis, C. A. J. Catal. 1993, 141, 347−354.
(45) Barzetti, T.; Selli, E.; Moscotti, D.; Forni, L. J. Chem. Soc.,
Faraday Trans. 1996, 92, 1401−1407.
(46) Rigby, A. M.; Kramer, G. J.; van Santen, R. A. J. Catal. 1997,
170, 1−10.
(9) Bortnovsky, O.; Sazama, P.; Wichterlova, B. Appl. Catal., A 2005,
287, 203−213.
(47) Arrhenius, S. Z. Phys. Chem. 1889, 4, 226−248.
(48) Liu, Z.; Gren, J.; Hillert, M. Fluid Phase Equilib. 1996, 121, 167−
177.
(10) Gao, X.; Tang, Z.; Zhang, H.; Ji, D.; Lu, G.; Wang, Z.; Tan, Z. J.
Mol. Catal. A: Chem. 2010, 325, 36−39.
(11) Erichsen, M. W.; Svelle, S.; Olsbye, U. J. Catal. 2013, 298, 94−
101.
(49) Maurer, T.; Kraushaar-Czarnetzki, B. J. Catal. 1999, 187, 202−
208.
(12) Choudhary, V. R.; Banerjee, S.; Panjala, D. J. Catal. 2002, 205,
398−403.
(13) Choudhary, V. R.; Banerjee, S.; Panjala, D. Microporous
Mesoporous Mater. 2002, 51, 203−210.
̌
̌ ́ ́
ova, A.; Rojasova, E.; Hudec, P.; Sabo, L. React. Kinet.
(14) Smiesk
Catal. Lett. 2004, 2, 227−234.
(15) Chen, M.; Yang, Y. J. Mol. Catal. (China) 1996, 10, 418−422.
(16) Sazama, P.; Dedecek, J.; Gabova, V.; Wichterlova, B.; Spoto, G.;
Bordiga, S. J. Catal. 2008, 254, 180−189.
(17) Quintana-Solorzano, R.; Thybaut, J. W.; Marin, G. B.; Lødeng,
̌
̌
́
́
́
́
R.; Holmen, A. Catal. Today 2005, 107−108, 619−629.
(18) Guisnet, M.; Magnoux, P. Appl. Catal., A 2001, 212, 83−96.
(19) Yan, L.; Fu, J.; He, M. Y. Acta. Petrol. Sinica 2000, 16, 15−26.
(20) Guisnet, M.; Magnoux, P. Catal. Today 1997, 36, 477−483.
(21) Zhao, G. L.; Teng, J. W.; Xie, Z. K.; Jin, W. Q.; Yang, W. M.;
Chen, Q. L.; Tang, Y. J. Catal. 2007, 248, 29−37.
(22) Xue, N. H.; Chen, X. K.; Nie, L.; Guo, X. F.; Ding, W. P.; Chen,
Y.; Gu, M.; Xie, Z. K. J. Catal. 2007, 248, 20−28.
(23) Zhao, G.; Teng, J.; Jin, W.; Yang, W.; Xie, Z.; Chen, Q. Chin. J.
Catal. 2004, 25, 3−4.
(24) Wang, Z. W.; Jiang, G. Y.; Zhao, Z.; Feng, X.; Duan, A. J.; Liu, J.;
Xu, C. M.; Gao, J. S. Energy Fuels 2010, 24, 758−763.
(25) Jiang, G. Y.; Zhang, L.; Zhao, Z.; Zhou, X. Y.; Duan, A. J.; Xu, C.
M.; Gao, J. S. Appl. Catal., A 2008, 340, 176−182.
(26) Xue, N. H.; Nie, L.; Fang, D. M.; Guo, X. F.; Shen, J. Y.; Ding,
W. P.; Chen, Y. Appl. Catal., A 2009, 352, 87−94.
(27) Xu, R. F.; Liu, J. X.; Liang, C. C.; Jia, W. G.; Li, F. F.; Guo, H. C.
J. Fuel. Chem. Technol. 2011, 39, 449−454.
4059
DOI: 10.1021/cs501967r
ACS Catal. 2015, 5, 4048−4059