Mechanism of byproducts formation in the isobutane/butene alkylation on HY zeolites

Article abstract of DOI:10.1039/c7ra12629h

Submicron-size HY zeolites with a particles size of 200-700 nm were synthesized employing a crystal precipitation method in this study. The catalytic activity for the isobutane/butene alkylation was evaluated. The results indicated that butene conversion was above 90% and the selectivity of expected products (C8) was nearly at 90% within 72 h. The micropores-blocking and coverage of acid sites resulting from high hydrocarbons increased the difficulty for the diffusion of products to the bulk and inhibited the adsorption of reactant on activity sites, which caused deactivation of catalyst. The ultimate C12 content in alkylate oil, stemmed from trimerization of butene, was reduced via the addition reaction with butene to C16 and the cracking to C5-C7. The formation mechanisms and transformation processes of byproducts in alkylate oil revealed that the source of C9-C11 switched from cracking of C16+ to the addition of C5-C7 carbocations with butene when acid sites concentration was reduced by accumulating oligomers.

Full text of DOI:10.1039/c7ra12629h

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Mechanism of byproducts formation in the
isobutane/butene alkylation on HY zeolites
Cite this: RSC Adv., 2018, 8, 3392
a
a
a
b
c
a
Zhiwei Chen, Feng Gao, Kun Ren, Quangui Wu, Yan Luo, Hongjun Zhou,
d
a
Meng Zhang and Quan Xu
*
Submicron-size HY zeolites with a particles size of 200–700 nm were synthesized employing a crystal
precipitation method in this study. The catalytic activity for the isobutane/butene alkylation was
evaluated. The results indicated that butene conversion was above 90% and the selectivity of expected
products (C8) was nearly at 90% within 72 h. The micropores-blocking and coverage of acid sites
resulting from high hydrocarbons increased the difficulty for the diffusion of products to the bulk and
inhibited the adsorption of reactant on activity sites, which caused deactivation of catalyst. The ultimate
C12 content in alkylate oil, stemmed from trimerization of butene, was reduced via the addition reaction
with butene to C16 and the cracking to C5–C7. The formation mechanisms and transformation
processes of byproducts in alkylate oil revealed that the source of C9–C11 switched from cracking of
C16+ to the addition of C5–C7 carbocations with butene when acid sites concentration was reduced by
accumulating oligomers.
Received 21st November 2017
Accepted 14th December 2017
DOI: 10.1039/c7ra12629h
rsc.li/rsc-advances
investigating new solid acid catalysts (super solid acid, hetero-
1
. Introduction
polyacid and zeolites) and decreasing the critical reaction
2
As a consequence of the strict environmental scrutiny and conditions. However, there are numerous limitations at an
upgrading of vehicle technology, methyl-tert-butyl ether (MTBE) industrial scale, ascribed to fast deactivation caused by oligo-
will gradually be replaced as it contaminates groundwater and merization of butene. Thus, solid acid catalysts for alkylation,
harms human health by stimulating the central nervous system. particularly zeolites, are a promising research eld due to
Therefore, environment-friendly and high-octane replacements outstanding advantages in environment protection.
are considered as gasoline blending components. Such
Among the zeolites used in isobutane/butene alkylation,
3
–8
9–11
12–15
13,16
17
replacements mainly consists of isooctane meeting to the which include Y,
requirements of environment and fuel quality and can be MCM, the Y zeolites are the most studied because of their
produced via butene dimerization and subsequent hydrideation considerable activity for hydride transfer and large pore size.
X,
b,
mordenite,
ZSM and
18
19
1
as well as isobutane/butene alkylation. Both these processes However, the controversy regarding the activity of the protonic
require conventionally used homogeneous and heterogeneous form of Y zeolites towards alkylation has raged unabated.
20
catalysts. The drawbacks of butene oligomerization are man- Querinl et al. reported that HY was not active for TMP
ifested in its complex process and lower butene conversion rate. formation due to a strong adsorption of the olens and subse-
In comparison, isobutane/butene liquid acid alkylation has quent oligomerization as a consequence of very fast deactiva-
18
been industrialized on a large scale due to its gentle operation tion. Corma et al. investigated the impact of the unit cell size
conditions, higher butene conversion rate, and the complete of steam-dealuminated Y zeolites and found that oligomeriza-
utilization of the C4 fraction. Traditionally, HF and H SO
2 4
were tion continuously increased with the zeolite unit cell size
the industrial alkylation catalysts, which have negative inu- although the maximum conversion of 2-butene was attained on
ence on environment and humans because of their corrosivity samples with unit cell sizes between 2.435 and 2.450 nm.
and toxicity. Therefore, massive efforts have been made to Although strong Brønsted acid sites that benet alkylation can
3
+
substitute the existing liquid acid catalysts, including be achieved by ion-exchange with La , the alkylation efficiency
on LaY was not satisfactory because of the short lifespan of the
4
zeolite and low selectivity to isooctane. Hence, it is necessary to
a
State Key Laboratory of Heavy Oil Processing, China University of Petroleum Beijing,
further study Y zeolites.
Beijing, China, 102249. E-mail: xuquan@cup.edu.cn
b
It is known that C8 hydrocarbons are the main products in
Dongying Keerte New Material Co., Ltd, Dongying, Shandong, 257081, China
c
Department of Chemical Engineering, West Virginia University, Morgantown, 26505, C5+ alkylate over all types of zeolites and the distribution of
USA
individual components depends on the catalyst categories and
reaction conditions. In order to increase the selectivity of
d
Department of Mathematics, University of North Georgia-Oconee, Watkinsville,
3
0677, USA
3392 | RSC Adv., 2018, 8, 3392–3398
This journal is © The Royal Society of Chemistry 2018

View Article Online
Paper
RSC Advances
desirable alkylate trimethylpentane (TMP), considerable measurement, the samples were pressed into self-supporting
ꢀ
research has been carried out to understand the formation wafers and dried under vacuum at 400 C for 1 h to remove
2
1–24
mechanism of C8.
However, the formation pathways of the water and weakly adsorbed deposits. Then, pyridine was
ꢀ
byproducts were discussed roughly. It is expected that the C5– adsorbed at 30 C in vacuum for 2 h until no changes were
C7 hydrocarbon species are derived from the cracking of higher observed in the spectrum. Aer weakly physisorbed pyridine
hydrocarbons by b-scission, and the C9+ hydrocarbon species was removed, the IR spectra were recorded at 200 C.
ꢀ
are formed via butene oligomerization. However, the interesting
phenomenon that the main byproduct of C9+ over solid acid 2.3 Isobutane/butene alkylation test
catalysts is a C11 chain hydrocarbon rather than C12 has not
Initially, 5 g fresh HY zeolite was loaded in the constant
25
been explained. The group also found that a coke precursor
deposited on the surface of zeolites was comprised of long chain
alkanes in which the number of carbon atoms was odd.
The objective of this study is to demonstrate that HY zeolites
exhibit activity for isobutene/butene alkylation and reveal the
formation mechanism of the main alkylation byproducts over
HY zeolites.
temperature section of a stainless tubular reactor with inner
diameter of 12 mm and length of 50 cm. Prior to reaction, the
ꢀ
catalysts were pretreated at 500 C for 1 h in N at a ow rate of
4
2
ꢁ1
0 mL min in order to remove the water adsorbed over the
2
zeolites. Then, the reactor was cooled by N to the reaction
temperature and pressurized to 2.0 MPa. Finally, the isobutane/
butene mixture with a molar ratio of I/O ¼ 180 was pumped
(
SZWEICO, 2ZB-1L10A) to the reactor at a weight hourly space
ꢁ
1
velocity (WHSV) of 7.5 h . The mixture of butenes consisted of
2. Experimental section
1
8
-butene, 2-butenes, and isobutene at concentrations of 16.3,
3.1, and 0.6 wt%, respectively. The reaction was stopped when
2.1 Synthesis of HY zeolites
Sodium metasilicate, sodium aluminate, and sodium hydroxide the conversion rate of butene was lower than 90% or time on
were dissolved in deionized water according to the mole ratio stream (TOS) reached 72 h.
16Na O : Al O
2 2 3
: 15SiO
2
2
: 300H O. Subsequently, the mixture was
The components in the liquid-phase product were analyzed
aged at room temperature for 12 h to generate a crystallization using a gas chromatograph GC-9720/FID with a capillary column
directing agent. Then, the above mixture and directing agent were (50 m ꢂ 0.2 mm ꢂ 0.5 mm Agilent-HP-PONA). The GC9790II/FID
ꢀ
transferred into an autoclave and crystallized at 100 C for 12 h with a capillary column (30 m ꢂ 0.530 mm Agilent-GC-Alumina)
2 2 3 2 2
according to the mole ratio of 2.7Na O : Al O : 8.4SiO : 200H O. was employed to determine the components of feedstock and
It is worth noting that Si derived from the guiding agent would not effluent gas. The butene conversion and product selectivity could
be lower than 90%. The obtained solution mixture was ltered be calculated according to eqn (1) and (2).
through vacuum suction ltration and washed with deionized
F
in
y
in ꢁ Fout
y
out
ꢀ
water. The resultant solid was dried at 120 C overnight and
Butene conversion ð%Þ ¼
ꢂ 100%
(1)
F
in
y
in
ꢀ
calcined at 550 C for 4 h to form the parent powder of NaY. HY
zeolite powder was obtained by the ion-exchange between NaY
y
i
ꢁ
1
Product selectivity ð%Þ ¼
X
ꢂ 100%
(2)
powder and 1 mol L NH Cl aqueous solution, drying, and
4
yi
stepwise calcination. The above steps were repeated thrice. The
i¼5

nal catalysts were prepared through a series of steps, including
where Fin, Fout and yin, yout refer to the molar ow and butene
percentage at the reactor inlet and outlet, respectively, and y is
the molar percent of product with i number of carbons in the
effluent.
ꢀ
extrusion with a binder into cylindrical shape, drying at 120 C for
i
ꢀ
ꢀ
12 h, and calcination at 350 C for 2 h and then at 550 C for 3 h.
2.2 Characterization
2
.4 Regeneration of used catalyst
recorded using a D8 Advance (Bruker, Germany) diffractometer The catalyst needed to be regenerated when the time on stream
operating at 40 mA and 40 kV using monochromated Cu Ka
reached 72 h or the butene conversion decreased to 90%. The
X-Ray Powder Diffraction (XRD) spectra of the catalysts were
1
˚
ꢀ
ꢀ
radiation (l ¼ 1.541 A) in the 2q range from 5 to 90 . The used catalysts were regenerated in situ at a temperature of
ꢀ
diffraction data was recorded using continuous scanning at 450 C at atmospheric pressure for 12 h by passing air at a ow
ꢀ
ꢁ1
ꢁ1
a rate of 0.02 s . Scanning electron microscopy (SEM) images rate of 50 mL min . Subsequently, the reaction bed was cooled
were collected on a ZEISS Gemini SEM300 at 5 kV. Powder to 75 C by N at a rate of 40 mL min and then pressurized at
ꢀ
ꢁ1
2
samples extruded mouldings for SEM imaging were placed on 2.0 MPa. The alkylation reaction proceeded by purging feed-
a conductive adhesive. N
2
adsorption–desorption studies were stock at the same reaction conditions as fresh catalysts aer the
ꢀ
conducted at ꢁ196 C aer sample pretreatment under vacuum carbon deposit was wiped off.
ꢀ
at 250 C for 4 h. The specic surface area was determined by
the Brunauer–Emmett–Teller (BET) method in the P/P range
0
3
. Results and discussion
0
.05–0.2, and the pore size distribution was calculated by the
Barrett–Joyner–Halenda (BJH) method using the desorption
branch of the isotherm. Brønsted and Lewis acid concentra- The powder XRD patterns of the synthesized and used catalysts
3.1 Characterization of catalysts
ꢀ
ꢀ
tions were obtained from pyridine-IR spectra. Before (Fig. 1) exhibited characteristic peaks at 2q ¼ 6.25 , 10.19 ,
This journal is © The Royal Society of Chemistry 2018
RSC Adv., 2018, 8, 3392–3398 | 3393

View Article Online
RSC Advances
Paper
ꢀ ꢀ ꢀ ꢀ ꢀ
1.96 , 15.72 , 18.76 , 20.44 , and 23.73 , in accordance with
1
the reference patterns for HY zeolites in the Joint Committee on
Powder Diffraction Standards (JCPDS) database, indicating that
the synthesized HY zeolites were well-crystallized and pure. The
inferior peak intensities of the spent zeolites when compared to
that of fresh catalyst were due to the carbon deposits on the
catalysts.
Fig. 2 displays the SEM images of the synthesized HY zeolites
with prominent crystal morphology. The particles sizes ranged
from 200 to 700 nm. However, visible agglomeration of the
particles is observed and the entangled particles that are
inserted in each other decrease the efficient surface for reaction.
The agglomeration became more severe aer deactivation
(Fig. 2(d)) when compared to the fresh zeolites (Fig. 2(c)).
The N adsorption–desorption isotherms and pore size
2
distributions of the fresh and used HY zeolites are depicted in
Fig. 3(a) and (b). Both isotherms observed in Fig. 3(a) display
Fig. 1 XRD patterns of fresh and spent catalysts.
type IV behavior with a hysteresis loop above p/p ¼ 0.4, which
0
can be typically observed in mesoporous materials with pore
diameter up to ca. 50 nm. Moreover, a steep increase at p/p <
0
0.02 of the catalysts can be interpreted as microporous lling,
while the other steep increase at 0.4 < p/p is caused by capillary
0
condensation in the mesopores. The pore distribution has an
insignicant change as observed in Fig. 3(b), where apparent
mesopores with a specic diameter of 7.61 nm along with
a narrow distribution are observed. In addition, the amount of
pores with different sizes of catalysts decreased aer the alkyl-
ation reaction. The reason for the above phenomenon could be
that oligomers of olens absorbed to the micro and mesopores
cannot diffuse to the bulk phase. The concrete values of the
textural properties of zeolites, including specic surface area
and pore volume, are listed in Table 1. All the textural indictors
of fresh catalysts declined aer the alkylation reaction. In
particular, the micropore specic surface area and micropore
volume decrease by 18.8% and 19.1%, respectively. This was
possibly resulted from bulk pores and covered surface with
oligomers. The catalysts began to deactivate when large
Fig. 2 SEM images (a and b) HY powder before extrusion; (c) fresh HY;
and (d) spent HY.
2
Fig. 3 (a) N adsorption–desorption isotherm and (b) pore sizes distribution of fresh and spent HY zeolites.
3394 | RSC Adv., 2018, 8, 3392–3398
This journal is © The Royal Society of Chemistry 2018

View Article Online
Paper
RSC Advances
Table 1 Textural properties of fresh and used HY zeolites
Table 2 Brønsted and Lewis acid sites concentrations of the fresh and
used HY zeolites
a
b
c
d
HY
catalyst
S
BET
S
micro
V
P
V
micro
V
meso
2
ꢁ1
2
ꢁ1
3
ꢁ1
3
ꢁ1
3
ꢁ1
ꢁ1
ꢁ1
(m g
)
(m g
)
(cm g
)
(cm g
)
(cm g
)
Catalyst
C
B
(mmol g
)
C
L
(mmol g
)
B/L
Fresh
Spent
588.3
490.6
479.8
389.7
0.360
0.305
0.235
0.190
0.125
0.115
Fresh
Used
0.1019
0.0818
0.0263
0.0254
3.8
3.22
a
b
c
S
micro ¼ t-plot micropore area.
V
p
¼ total pore volume.
V
micro ¼ t-
d
plot micropore volume.
BJH method.
V
meso ¼ mesopore volume determined by
hydrocarbons generated in the micropores fail to diffuse to bulk
9
phase.
Fig. 4 exhibits the Py-IR spectra to qualitatively and quanti-
tatively determine the Lewis acid and Brønsted acid concen-
ꢁ
1
ꢁ1
trations in HY zeolites located at 1545 cm and 1455 cm
respectively.
,
26–28
The surface Brønsted (B) and Lewis (L) acid
sites concentrations were calculated according to the equation
29
published by Emeis and the results are summarized in
Table 2. The Brønsted acid sites are dominant in both fresh and
used HY zeolites, while the amount of Lewis acid sites is small.
The amount of B and L acid types decreased by 19.7% and 3.4%
aer the alkylation reaction, respectively. The ratio of B/L
declined from 3.8 to 3.22, demonstrating that Brønsted acid
Fig. 5 The effects of reaction temperatures on butene conversion.
type was primarily responsible for alkylation and the decrease Reaction condition: the ratio of isobutane to butene (I/O ¼ 180), P ¼
ꢁ1
2.0 MPa, WHSV ¼ 7.5 h
.
of Brønsted acid sites was the main reason for catalyst
deactivation.
for the activation and diffusion of reactants and products as
well as overcoming steric hindrance. However, higher temper-
ature was more convenient for butene oligomerization resulting
3
.2 Alkylation activity of the synthesized catalysts
The activity of the synthesized HY zeolites for isobutane/butene
alkylation was investigated. The butene conversions over time
on stream at various temperatures (70, 75, and 80 C) are shown
in Fig. 5, where conversion gradually decreased from 100% in
the initial 12 h to 91, 92, and 95% at the end of the reactions
aer 70 h, respectively. The minimum decrease in butene
8
in a decline of activity centres. The covering surface and bulk
ꢀ
pore values are summarized in Table 1 and Fig. 2, and the
decrease of Brønsted acidities is illustrated in Fig. 4 and Table 2.
The distribution of individual components in the alkylate at
ꢀ
the optimal temperature of 75 C is shown in Fig. 6, revealing
ꢀ
conversion was achieved at a reaction temperature of 75 C. The
possible reason was that lower temperature was not benecial
Fig. 6 Selectivity of each component in alkylate (C5+) over HY
zeolites. Reaction condition: the ratio of isobutane to butene (I/O ¼
ꢀ
ꢁ1
Fig. 4 Py-IR profiles at the desorption temperature of 200 C.
180), P ¼ 2.0 MPa, WHSV ¼ 7.5 h
.
This journal is © The Royal Society of Chemistry 2018
RSC Adv., 2018, 8, 3392–3398 | 3395

View Article Online
RSC Advances
Paper
that the selectivity of C5–C7 and C8 decreased with TOS, but the
selectivity of C9+ increased. The high selectivity of C8 hydro-
carbon species (above 80%) and the low selectivity of C12+
(lower than 1% at 72 h), observed in Fig. 6, reected that the
synthesized HY zeolites had great stability and also, the oligo-
merization activity producing higher hydrocarbons (over C12)
was limited in this stage due to a highly active catalyst. The
selectivities of the main constituents of C8, namely, trime-
]
thylpentane, dimethylhexane (DMH), and C8 are described in
Fig. 7. The selectivities of TMP and DMH were stable at 75 and
]
1
5%, respectively, while the selectivity of C8 was less than 1%.
In general, C5–C7 species were derived from oligomers, such as
C12+, through cleavage of C–C bonds, while C9+ species was
23
generated from oligomerization of butene. Although both
cracking and oligomerization require acid centers, the former
took place on Brønsted acid centres and the latter primarily
depended on Lewis acid centers. The paralleled selectivity
tendencies of C5–C7 and C8 were in accordance with the
requirement of the same Brønsted acid active sites for alkyl-
Fig. 8 Butene conversion with TOS in the presence of fresh and
regenerated catalyst. Reaction condition: the ratio of isobutane to
ꢁ1
butene (I/O ¼ 180), P ¼ 2.0 MPa, WHSV ¼ 7.5 h
.
30
ation and cracking. In addition, the acidity strength needed by
each reaction step in the process of alkylation was different:
The regeneration effect is exhibited in Fig. 8, where the
butene conversion recovered to 100% in the initial 10 h and
then decreased at different gradients for the rst and second
regeneration. In addition, the product distribution (Table 3)
showed that while the target TMP decreased from 72.51 to
cracking > alkylation (electrophilic addition of butene to iso-
+
C
4
H
9
) > oligomerization (electrophilic addition of butene to
+
21
4 9
sec-C H ). This indicated that oligomerization still proceeded
stably when acidity centres were less in number and weaker.
Therefore, the inhibiting effect of fewer acid centres on cracking
69.36% for the rst regeneration and 65.59% for the second
21
was more distinct than for alkylation and addition reaction, so
that the selectivity of C12+ increased with time on stream when
the catalyst began to undergo deactivation.
regeneration, the selectivity of the alkylation byproducts
increased, including those for C5–C7 and C9+. The initial
butane conversion was 100% in 12 h for the regenerated cata-
lysts, which was similar to that for the fresh catalysts, inferring
that the regeneration process was efficient. However, the sharp
descent gradient and unsatisfactory product distribution of
spent catalysts aer 15 h indicated that the stability of regen-
erated catalysts was inferior to that of the fresh samples,
possibly because the coke could not be burned out completely.
It can be considered that the used zeolites need to be regen-
erated once the butene conversion was lower than 100%, which
could avoid the formation of intractable carbon deposits and
facilitate the recovery of the activity of the catalyst.
For electrophilic addition of butene, C12 directly derived
from the trimerization of butene was the main component in
C9–C12, and C9–C11 was scarcely included in the alkylate
products. However, the reducing order of selectivity was C11 >
C10 > C9 > C12+, as observed in Fig. 6, which was contrary to the
viewpoint that higher hydrocarbon than C8 was simply formed
14
by butene oligomerizing.
3.3 Formation mechanism of alkylation byproducts
It is clear that both cracking and oligomerization occurred in
the process of butene alkylation (Fig. 9). Theoretically, the
number of carbon atoms in primary oligomers originating from
butene oligomerization is a multiple of four. However, b-
Table 3 Products distribution of the fresh and used catalysts
st
nd
Catalysts
Fresh HY, %
1
regeneration, %
2
regeneration, %
C5–C7
C8
3.64
88.32
72.51
14.47
0.37
6.07
6.79
84.43
69.36
12.93
0.32
80.31
65.59
12.38
0.32
TMP
DMH
Fig. 7 Selectivity of individual C8 component in alkylate (C5+) vs. TOS
]
C8
over HY zeolites. Reaction condition: the ratio of isobutane to butene
ꢁ
1
C9+
8.05
9.51
12.90
was I/O ¼ 180, P ¼ 2.0 MPa, WHSV ¼ 7.5 h
.
3396 | RSC Adv., 2018, 8, 3392–3398
This journal is © The Royal Society of Chemistry 2018

View Article Online
Paper
RSC Advances
Scheme 1 Intramolecular hydride transfer.
Scheme 2 Cracking of long chain carbocation.
maximum. Nevertheless, the decreased content and strength of
acid centers with catalysts deactivation, particularly Brønsted
acid centers, enhanced oligomers but suppressed cracking
more drastically when compared to the addition of butene with
carbonations. This was because stimulation of cracking
Fig. 9 Formation routes of target compound TMP and primary olig-
omers in isobutane/butene alkylation over zeolites.
21,31
20
required stronger acidity.
As shown in Fig. 6, there are no C1–C3 hydrocarbons, indi-
cating that extended cracking of primary crackates scarcely
occurred in the alkylation process. As a result, crackates of C12+
only included C4–C8, in which C4 and C8 were directly revers-
ible reactants. Moreover, the existence of C5–C7 hydrocarbons
indicated the possibility of hydride transfer in C12+ as shown in
Scheme 1, where C12+ was the primary oligomer of 2-butene on
HY zeolites.
In addition, the C6 species was generated via the cleavage of
C–C bond as shown in Scheme 2. Similarly, C5 and C7 hydro-
carbons were also formed by the b-scission of other C12+
isomers.
scission that can be easily initiated in longer chain hydrocar-
bons on acid sites, results in the carbon atom number of end
products in alkylate oil ranging from 5 to 12. Cracking of C8
hardly took place in the alkylation conditions, supporting that
13,14
C5–C7 components primarily stem from C12–C16.
More-
over, C9–C11 components were produced through two routes:
13
(
1) cracking of C16+ from tetramerization of butene and (2)
electrophilic addition of C5–C7 carbocation with C H . Hence,
4
8
the formation pathways of alkylation byproducts can be
described as shown in Fig. 10.
As shown in Fig. 6, the selectivity of quantity of C5–C7
decreased gradually, but the selectivity of C9+ species increased
with the deactivation of catalysts. The possible reason was that
the acid strength of the catalyst was sufficient to promote
hydride transfer and enable cracking at the initial stage of
It is speculated that the higher the carbocation, the more
carbon atoms are possessed by the primary products of
cracking. Therefore, the selectivity of cracking products
decreased in the following order: C7 > C6 > C5. Correspond-
ingly, the selectivity decreasing in the order of C9–C11 can be
explained based on two reasons: (1) the more alkyl groups in the
carbocation chain, the faster the electrophilic addition rate
because the products with more alkyl groups were more stable
due to charge dispersion. (2) The higher the concentration of
reactants, the easier the oligomerization reaction. It is necessary
to emphasize that only C12 was detected in C12+ alkylate oil,
possibly because chromatography function was limited and the
32
alkylation. Therefore, the cracking reaction rate, r , of C12+
3
remained high and the selectivity of C5–C7 hydrocarbons was
13
higher carbon number components retained over catalysts.
There was a greater potential for sequential addition of C12+
once it was formed via the addition of C8+ with butene, sug-
gesting accumulation of C12 was few. Moreover, it was possible
that C12 was converted into C5–C7. As a result, the selectivity of
C9–C11 was higher than that of C12 though the relative rate of
oligomerization increased.
4. Conclusions
Isobutane alkylation with butenes was carried out over synthe-
sized HY zeolites in a tubular reactor. HY zeolites exhibited
Fig. 10 Formation routes of alkylation byproducts.
This journal is © The Royal Society of Chemistry 2018
RSC Adv., 2018, 8, 3392–3398 | 3397

View Article Online
RSC Advances
Paper
ideal activity and desirable selectivity of C8 hydrocarbon at
9 C. Sievers, I. Zuazo, A. Guzman, et al., J. Catal., 2007, 246,
315–324.
ꢀ
75 C with a butane conversion rate above 95% and the selec-
tivity of C8 near to 90%. However, the deactivation of catalysts 10 L. Patrylak, O. P. Pertko, S. V. Konovalov, et al., Curr. Catal.,
was unavoidable due to the decrease of surface area and acid 2016, 5, 108–115.
sites concentration. The selectivity of C5–C7 decreased because 11 K. I. Patrilyak, L. K. Patrilyak, Y. G. Voloshina, et al., Theor.
of the weakened cracking ability, resulting from the decreased Exp. Chem., 2011, 47, 205.
acid sites especially Brønsted acid sites. Moreover, the 12 K. Yoo and P. G. Smirniotis, Appl. Catal., A, 2002, 227, 171–
enhanced selectivity of C9–C11 was due to the addition of C5– 179.
C7 carbocations with butane, which was another factor to 13 K. Yoo, E. C. Burckle and P. G. Smirniotis, J. Catal., 2002,
decrease the accumulation of C5–C7. It is indispensable to 211, 6–18.
further study the improvement measurements of particles 14 S. V. Nayak, P. A. Ramachandran and M. P. Dudukovic,
dispersion and pores distribution of HY zeolites used in the
alkylation of isobutane with butene.
Chem. Eng. Sci., 2010, 65, 335–342.
15 B. O. D. Costa and C. A. Querini, Appl. Catal., A, 2010, 385,
144–152.
1
1
1
1
6 A. C. Oliveira, N. Essayem, A. Tuel, et al., J. Mol. Catal. A:
Chem., 2008, 293, 31–38.
7 K. Yoo and P. G. Smirniotis, Appl. Catal., A, 2003, 246, 243–
Conflicts of interest
There are no conicts to declare.
2
51.
8 A. Corma, A. Mart ´ı nez and C. Mart ´ı nez, J. Catal., 1994, 146,
85–192.
9 Z. Qin, K. A. Cychosz, G. Melinte, et al., J. Am. Chem. Soc.,
017, 139, 17273–17276.
Acknowledgements
1
We thank the National Key Research and Development Plan
2
(No. 2016YFC0303701), Beijing Novo Program Interdisciplinary
2
2
2
0 C. A. Querinl and E. Roa, Appl. Catal., A, 1997, 163, 199–215.
1 A. Feller, J. Catal., 2003, 220, 192–206.
Studies Cooperative project, Science Foundation of China
University of Petroleum, Beijing (No. 2462014YJRC011, 201604)
for the support.
2 A. Feller, I. Zuazo, A. Guzman, et al., J. Catal., 2003, 216, 313–
323.
2
2
3 L. K. Patrilyak and P. I. Bartosh, Theor. Exp. Chem., 2003, 39,
177–183.
4 L. K. Patrylak and I. A. Repetskyi, Theor. Exp. Chem., 2013, 49,
References
1
2
3
4
5
6
7
8
A. Sarkar, D. Seth, F. T. T. Ng, et al., Ind. Eng. Chem. Res.,
014, 53, 18982–18992.
A. E. Koklin, V. M. K. Chan, V. B. Kazanskii, et al., Kinet.
Catal., 2010, 51, 410–415.
143–157.
2
2
2
5 H. Yigong and H. Yufeng, Chin. J. Catal., 2000, 21, 175–178.
6 H. J. Lee, D.-C. Kang, S. H. Pyen, et al., Appl. Catal., A, 2017,
5
31, 13–20.
7 L. b. Qin, J. Han, Y. Q. Zhan, et al., Energy Fuels, 2016, 30,
44–550.
8 L. b. Qin, J. Han, W. Y. Chen, et al., Appl. Therm. Eng., 2016,
4, 813–818.
R. Klingmann, R. Josl, Y. Traa, et al., Appl. Catal., A, 2005,
2
2
281, 215–223.
5
C. Sievers, J. R. S. Liebert, M. M. Stratmann, et al., Appl.
Catal., A, 2008, 336, 89–100.
J. a. V. Bokhoven, A. L. Roest and D. C. Koningsberger, J.
Phys. Chem. B, 2000, 104, 6743–6754.
9
29 C. A. Emeis, J. Catal., 1993, 141, 347–354.
30 D. H. Park, S.-S. Kim, T. J. Pinnavaia, et al., J. Phys. Chem. C,
L. Yuying, L. Yongxiang, L. Jun, et al., Acta Pet. Sin., 2014, 30,
2
011, 115, 5809–5816.
1 R. Nilton Jr and A. C. J. Mota, J. Braz. Chem. Soc., 2005, 16,
91–694.
2 W. Xinxing, Q. Xiaolei, S. Jingming, et al., Ind. Catal., 2017,
5, 61–65.
765–771.
3
3
T. Juzsakova, A. Csavdari, A. Redey, et al., Environ. Eng.
Manage. J., 2014, 13, 2343–2347.
B. O. D. Costa and C. A. Querini, Chem. Eng. J., 2010, 162,
6
2
829–835.
3398 | RSC Adv., 2018, 8, 3392–3398
This journal is © The Royal Society of Chemistry 2018