The isopropylation of biphenyl over transition metal substituted aluminophosphates: MAPO-5 (M: Co and Ni)

Article abstract of DOI:10.1016/j.molcata.2015.11.006

The isopropylation of biphenyl (BP) was examined over transition metal substituted aluminophosphates (MAPO-5; M: Co and Ni) with 12-membered (12-MR) oxygen ring pore-entrances of AFI topology. The MAPO-5 samples were synthesized by dry gel conversion method using trimethylamine as a structure directing agent, and their properties were characterized by XRD, XPS, SEM, N₂ adsorption, NH₃-TPD, pyridine adsorption, and o-xylene uptake. They are clear crystals without impurity phases and agglomerates, and found small amounts of Br?nsted acid sites which are expecting active for acid catalysis. The isopropylation of BP over both of Co(5)APO-5 and Ni(5)APO-5 at 250 °C gave the high selectivities for 4,4′-DIPB: 65-75%. 4-IPBP is almost exclusive precursor of 4,4′- and 3,4′-DIPB. 3-IPBP was not significantly concerned even though 3-IPBP was predominant among IPBP isomers at the late stages: the MAPO-5 channels allow preferential access of 4-IPBP, and prevent the access of 3-IPBP due to reactant selectivity mechanism. The selective formation of 4,4′-DIPB occurred by preferential exclusion of bulkier 3,4′-DIPB and other isomers through the steric interaction of transition states with the channels by the restricted transition state selectivity mechanism. MAPO-5 (M: Co and Ni) has the same level of the selectivities for 4,4′-DIPB to SSZ-24 and other MAPO-5 (M: Si, Mg, and Zn), and these selectivities were originated by the AFI channels. The selectivities for 4,4′-DIPB were kept 65-75% at low and moderate temperatures over MAPO-5 (M: Co and Ni); however, they were decreased by the isomerization to stable 3,4′-DIPB with the increase in temperature.

Full text of DOI:10.1016/j.molcata.2015.11.006

Journal of Molecular Catalysis A: Chemical 412 (2016) 117–124
Contents lists available at ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
The isopropylation of biphenyl over transition metal substituted
aluminophosphates: MAPO-5 (M: Co and Ni)
Hui Xing Zhanga,b, Anand Chokkalingama, Priya V. Subramaniama, Stalin Josepha,g
,
Seiji Takeuchic, Ming Deng Weib, Abdullah M. Al-Enizid, Hoi-Gu Jange, Jong-Ho Kime,
Gon Seoe, Kenichi Komuraf, Yoshihiro Sugia,f,∗, Ajayan Vinua,g,∗∗
a Australian Institute of Bioengineering and Nanotechnology, The University of Queensland, Brisbane 4072, QL, Australia
b Institute of Advanced Energy Materials, Fuzhou University, Fuzhou, Fujian 350002, China
c Application Development Center, Sanjyo Factory, Shimadzu Corporation, Kyoto 604-8511, Japan
d Department of Chemistry, Faculty of Science, King Saud University, Riyadh 11451, Saudi Arabia
e School of Applied Chemical Engineering and Research Institute for Catalysis, Chonnam National University, Gwangju 500-757, South Korea
f
Department of Materials Science and Technology, Faculty of Engineering, Gifu University, Gifu 501-1193, Japan
g Future Industries Institute, University of South Australia, Mawson Lake, 5095 SA, Australia
a r t i c l e i n f o
a b s t r a c t
Article history:
The isopropylation of biphenyl (BP) was examined over transition metal substituted aluminophosphates
(MAPO-5; M: Co and Ni) with 12-membered (12-MR) oxygen ring pore-entrances of AFI topology. The
MAPO-5 samples were synthesized by dry gel conversion method using trimethylamine as a structure
directing agent, and their properties were characterized by XRD, XPS, SEM, N2 adsorption, NH3-TPD,
pyridine adsorption, and o-xylene uptake. They are clear crystals without impurity phases and agglom-
erates, and found small amounts of Brønsted acid sites which are expecting active for acid catalysis. The
isopropylation of BP over both of Co(5)APO-5 and Ni(5)APO-5 at 250 ◦C gave the high selectivities for 4,4ꢀ-
DIPB: 65–75%. 4-IPBP is almost exclusive precursor of 4,4ꢀ- and 3,4ꢀ-DIPB. 3-IPBP was not significantly
concerned even though 3-IPBP was predominant among IPBP isomers at the late stages: the MAPO-5
channels allow preferential access of 4-IPBP, and prevent the access of 3-IPBP due to reactant selectivity
mechanism. The selective formation of 4,4ꢀ-DIPB occurred by preferential exclusion of bulkier 3,4ꢀ-DIPB
and other isomers through the steric interaction of transition states with the channels by the restricted
transition state selectivity mechanism. MAPO-5 (M: Co and Ni) has the same level of the selectivities for
4,4ꢀ-DIPB to SSZ-24 and other MAPO-5 (M: Si, Mg, and Zn), and these selectivities were originated by the
AFI channels. The selectivities for 4,4ꢀ-DIPB were kept 65–75% at low and moderate temperatures over
MAPO-5 (M: Co and Ni); however, they were decreased by the isomerization to stable 3,4ꢀ-DIPB with the
increase in temperature.
Received 22 September 2015
Received in revised form 8 November 2015
Accepted 9 November 2015
Available online 9 December 2015
Keywords:
CoAPO-5
NiAPO-5
AFI topology
Biphenyl
Isopropylation
Shape-selective catalysis
© 2015 Elsevier B.V. All rights reserved.
1. Introduction
allow both reactants and products to enter, reside, and leave in the
catalytic process [1].
Shape selective catalyses in confined environments of ordered
solid acids introduce a type of controls for the prevention of bulky
products in their channels, resulting in the formation of the least
bulky products which are not always kinetically and/or thermody-
namically favored [1,2]. Ordered molecular sieves, such as zeolites,
are the most favorable candidates for the purpose because their
pores and channels are uniformly distributed, and their dimensions
Shape-selective catalyses over zeolites are based on the dif-
ferentiation of reactants, products, and/or reaction intermediates
according to shape and size in sterically confined environments [3].
If the zeolite pores and channels are small enough to accommodate
both of reactants and products, the reactant conversion and prod-
uct formation probabilities are determined by the molecular sizes
and the configurations of the pores and channels. Only a reactant
molecule smaller than the critical sizes against the pores, can enter
in the pores, and reacts at the catalytic site in the channels. Further-
more, only a product molecule, which can diffuse out through the
pores, appears as a product. Therefore, it is important to fit zeolites
with the type of catalysis. In particular, the shape-selective cataly-
sis appears through interaction of zeolites with reactants, transition
∗
Corresponding author at: B-504, 7-222, Shinmatsudo, Matsudo 270-0034, Japan.
Corresponding author at: Future Industries Institute, University of South
∗∗
Australia, Mawson Lake, 5095 SA, Australia.
E-mail addresses: ysugi@gifu-u.ac.jp (Y. Sugi), ajayan.vinu@unisa.edu.au
(A. Vinu).
http://dx.doi.org/10.1016/j.molcata.2015.11.006
1381-1169/© 2015 Elsevier B.V. All rights reserved.

118
H.X. Zhang et al. / Journal of Molecular Catalysis A: Chemical 412 (2016) 117–124
states, and products by “reactant”, “product”, and “restricted tran-
sition state” selectivities [3].
follows: 2.05 g of aluminum triisopropoxide (5.0 mmol as Al2O3)
was slurred in water (1.85 g). To this slurry, 1.15 g of 85% phos-
phoric acid (5.0 mmol as P2O5) diluted in water (2.0 g) was added
dropwise over a period of 0.5 h with constant magnetic stirring.
To the resulting solution, a 0.107 g of cobalt acetate (0.5 mmol as
CoO) was added and the stirring was further continued for 0.5 h.
The gels thus obtained were dried at 80 ◦C in an oil bath with con-
tinuous magnetic stirring to remove water. When the gel became
thick and viscous, it was homogenized manually using a Teflon rod
and this was continued until it dried. The gels in Teflon cups were
placed in the autoclaves (23 ml, Parr Instrument Co., IL, U.S.A.), and,
finally, a mixture of 384 mg of Et3N (3.8 mmol) and 150 mg of water
was taken at the bottom so as to be supplied as water vapor. The
crystallization was carried out by heating the autoclave statically
at 175 ◦C for 24 h in a convection oven. The autoclave assembled for
the synthesis is shown in our previous paper [22].
After the crystallization, the products were washed with dis-
tilled water, separated by centrifugation, and dried at 100 ◦C
overnight. The as-synthesized samples were calcined in a muf-
fle furnace in a flow of air (flow rate: 50 ml/min) as follows: the
temperature was raised from room temperature to 550 ◦C in 4 h,
kept at this temperature for another 7 h, and finally cooled to room
temperature in ambient conditions.
Ordered microporous aluminophosphates (APO) were first syn-
thesized by Wilson et al. in 1982 [4]. They have frameworks
composed of networks of aluminum and phosphorous with sim-
ilar topologies of zeolites. They are expected to work as molecular
sieves for the catalysis, separation of organic compounds, etc. The
drawback of these materials as catalytic materials, particularly for
acid catalysis, is low acidity because of neutral tetrahedral frame-
work composed of aluminum (three valence) and phosphorous
(five valence); however, the introduction of the other metal cations
to their frameworks enhances the appearance of acidities: typical
examples are silicoaluminophosphates (SAPO) and other metalloa-
luminophosphates (MAPO) by Flanigen and her co-workers [5,6].
The valent imbalance by partial replacement of Al3+ by divalent
cations or P5+ by tetravalent cations gives rise to Brønsted acid
sites [7]. The addition of metal cations will expand the number and
diversity of the APO, and afford new properties to the microporous
materials, such as acid catalysis, oxidation, etc. depending on the
types of metal species [9,10].
The isopropylation of biphenyl (BP) is a typical acid catalysis in
Friedel–Crafts alkylation of aromatic hydrocarbons, which demon-
strates the shape-selective nature of the molecular sieve [10–21].
The selective formation of the least bulky 4,4ꢀ-diisopropylbiphenyl
(4,4ꢀ-DIPB) should be facilitated over ordered microporous mate-
rials if the catalytic sites are properly restricted by their channels
[10–12,13].
NiAPO-5 was synthesized in a similar manner to CoAPO-5 using
nickel acetate as NiO precursor.
The resultant aluminophosphates are expressed as M(x)APO-5,
where M is transition metals, and x is input percentage of substi-
tuted metal (mol) with aluminum (mol) (see Table 1).
In the previous papers, we described that 4,4ꢀ-DIPB has been
selectively produced from BP over dealuminated H-mordenite
(MOR) [12,13]. We also found SSZ-24 with AFI topology has high
potentials for the shape-selective formation of 4,4ꢀ-DIPB [14–16].
However, most of previous works on the alkylation of polynu-
clear hydrocarbons have used zeolites as catalysts, and there are
only a few works using MAPO [18–21]. It is interesting to explore
the possibilities of shape-selective catalysis by large pore MAPO.
Among them, we focused on the MAPO-5 of AFI topology, which are
isostructural with SSZ-24 [22]. We found MAPO-5 (M: Si, Mg, and
Zn) for aluminophosphate version and SSZ-24 for aluminosilicate
version gave the almost same levels of the selectivities for 4,4ꢀ-DIPB
[14–16,18–20]. These results indicate that shape-selective natures
of these materials are due to AFI topology based on their oxygen
rings.
2.3. The isopropylation of BP over MAPO-5 (M: Co and Ni)
The alkylation of BP was carried out in a 100-ml SUS-316 auto-
clave (Nitto Koatsu, Tsukuba, Japan) under a constant pressure of
propene. Typical procedures of the isopropylation of BP are as fol-
lows: BP (3.84 g, 20 mmol) and Co(5)APO-5 (100 mg) were mixed
in an autoclave, which was flushed with nitrogen followed by
heating to a desired temperature. Propene was then introduced
to the desired pressure (0.8 MPa), and the reaction started by
the agitation. The pressure was maintained throughout the reac-
tion period (4 h). After the reaction, the autoclave was cooled,
and the catalyst was filtered and washed with toluene (10 ml).
The solution (ca. 1.5 ml) was taken from the total bulk products
and diluted with toluene (1.5 ml). The products were identified
by gas chromatography–mass spectroscopy on Shimadzu GC–MS;
QP5000, and analyzed by a Shimadzu 2010plus gas chromatograph
equipped with an Ultra-1 capillary column (25 m × 0.2 mm; film
thickness: 25 ␮m; Agilent Technologies).
In this paper, we examined the isopropylation of BP over tran-
sition metal substituted APO-5 (MAPO-5; M: Co and Ni) in order
to clarify the influence of transition metals in the acid catalysis.
This is the first example succeeding in shape-selective catalysis for
transition metal substituted APO-5.
The conversion of BP and the yield of each product were calcu-
lated on the basis of consumed BP. The selectivities for each DIPB
and isopropylbiphenyl (IPBP) isomer are expressed based on total
amounts of IPBP and DIPB isomers, respectively.
2. Experimental
2.1. Reagents
The products encapsulated in the catalysts were analyzed as
follows: after collection of the bulk products (as mentioned ear-
lier), the catalyst was washed 5 times with 20 ml of acetone, and
dried at 110 ◦C for 12 h. Then 20 mg of resulting catalyst was care-
fully dissolved in 0.5 ml of concentrated hydrochloric acid (35%) at
room temperature. This solution was basified with solid potassium
carbonate, and the organic layer was extracted three times with
2 ml of dichloromethane. After removing of the solvent in vacuo,
the residue was dissolved in 0.5 ml of toluene. GC analysis was
performed similarly to the case of bulk products.
Aluminum triisopropoxide, cobalt(II) acetate tetrahy-
drate (Co(OAc)2·4H2O), and nickel(II) acetate tetrahydrate
(Ni(OAc)2·4H2O), and phosphoric acid (85%), trimethylamine
(Et3N), BP, and other organic chemicals were obtained from
Sigma–Aldrich. 4,4ꢀ-Diisopropylbiphenyl (4,4ꢀ-DIPB) was pur-
chased from Tokyo Chem. Ind., Tokyo, Japan.
2.2. The synthesis of MAPO-5 (M: Co and Ni)
The synthesis of CoAPO-5 was carried out by dry gel conversion
method (vapor phase transfer method) [23]. The gel composition
for the synthesis of Co(5)APO-5 was as follows: 1.0Al2O3-0.03-
0.40CoO-1.0P2O5-0.76Et3N-45H2O. The typical procedures are as
2.4. The characterization of MAPO-5 (M: Co and Ni)
Crystallinity and phase purity of MAPO-5 were confirmed on
a Rigaku Ultima III X-ray diffractometer with Cu K␣ radiation

H.X. Zhang et al. / Journal of Molecular Catalysis A: Chemical 412 (2016) 117–124
119
Table 1
Syntheses of MAPO-5 (M: Co and Ni) and their textual properties.
Input/molar ratio
Output/molar ratio
BET/cm2 g−1
t-plot/cm2 g−1
Pore volume/cm3 g−1
M/P
Co
Al
P
M/AI
(Al + Co)/P
Micro
External
Co(3)APO-5
Co(5)APO-5
Co(10)APO-5
Co(20)APO-5
Co(30)APO-5
Co(40)APO-5
Ni(5)APO-5
0.03
0.05
0.10
0.25
0.43
0.67
0.05
0.15
0.25
0.50
2.00
3.00
4.00
0.25
5.0
5.0
5.0
8.0
7.0
6.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
0.029
0.054
0.103
0.265
0.467
0.532
0.057
1.187
1.150
1.085
1.047
1.093
1.081
1.159
262
254
n.d.
263
220
127
189
202
192
n.d.
192
159
73
60
61
n.d.
72
60
0.079
0.076
n.d.
0.076
0.063
0.032
0.025
55
145
43
(ꢀ = 0.15418 nm). The bulk composition of MAPO-5 was determined
by Bruker S8 Tiger X-ray Fluorescence Spectrometer. XPS spectra
were recorded on MultiLab 2000 spectrometer (VG Systems, LLC,
Co., VA, USA). The observation of morphologies was carried out
using a Hitachi S-4700 Scanning Electron Microscope. N2 adsorp-
tion isotherms were recorded on a Micromeritics 3Flex and Tristar
3020 Analyzers and/or a MireaSI nanoPorosity-XG Analyzer. Pyri-
dine adsorption was recorded on a Shimadzu FT-IR spectrometer,
IRTracer-100, with an MCT detector and a vacuum IR cell (Makuhari
Rikagaku Garasu Inc., Chiba, Japan). IR spectra of adsorbed pyri-
dine on wafer were obtained under the exposure of 0.75–5.5 Torr
of pyridine by rapid scan during 60 s in every 0.19 s (resolution:
4 cm−1) after the evacuation at 400 ◦C during 1 h. Finally, the spectra
of pyridine strongly adsorbed were obtained after the evacuation
for 30 min. under 0.5 Torr at room temperature. Uptakes of o-xylene
on MAPO-5 were measured using a gravimetric adsorption system
equipped with a quartz spring. The sample was first evacuated at
300 ◦C for 1 h, and the mass gain was measured at 90 ◦C during
30 min after exposing 6.1–6.7 Torr of o-xylene.
3. Results and discussion
Fig. 1. XRD patterns of MAPO-5 (M: Co and Ni).
3.1. The characterization of MAPO-5 (M: Co and Ni)
Fig. 2 shows SEM images of MAPO-5 samples. They have well-
ordered plate type clear crystals. The crystal size of Co(5)APO-5
were smaller than APO-5 and Ni(5)APO-5. There are no agglomer-
ates assigned to cobalt and nickel species.
Fig. 1 shows XRD patterns of MAPO-5. All of them have the pat-
terns of AFI topology. The distinct diffraction of AFI topology was
observed in the wide range of cobalt content (3–30%). There were
no patterns of impurities and agglomerated cobalt species even
for Co(30)APO-5, either. Ni(5)APO-5 also had the patterns of AFI
topology.
XPS observation suggests that cobalt and nickel were trivalent
oxides in the materials (data not shown). These results mean metal
species are in arrays of APO-5, where metal species were present
as mixed phosphate clusters of cobalt and aluminum in the chan-
nels, and the isomorphously substituted M2+ species with Al3+ in
the framework were not observed due to their small amounts as
dicussed below. The formation of cobalt cluster in CoAPO-5 chan-
nels has been suggested by van Breukelen et al. [24].
Nitrogen adsorption isotherms of MAPO-5 are shown in Fig. 3.
Amount of N2 adsorbed in micropores are around 70 cm3/g (stp) for
APO-5 and Co(5)APO-5; however, 30 cm3/g for Ni(5)APO-5. These
differences indicate the Ni(5)APO-5 is lowly crystallized compared
to APO-5 and Co(5)APO-5, although clear XRD patterns and SEM
were observed as in Figs. 1 and 2. Table 1 shows that BET surface
areas and pore volumes were decreased with the increase in cobalt
contents of CoAPO-5. t-Plot calculation indicates that micropores
are decreased with the increase in external surface areas by the
increase of cobalt contents. These results suggest that the crys-
tallinities were decreased with increase in cobalt content, and that
Fig. 2. SEM images of MAPO-5 (M: Co and Ni).

120
H.X. Zhang et al. / Journal of Molecular Catalysis A: Chemical 412 (2016) 117–124
Fig. 3. N2 adsorption isotherms of MAPO-5 (M: Co and Ni). Legend: : Co(5)APO-5;
: Ni(5)APO-5; :APO-5.
Fig. 5. NH3-TPD profiles of MAPO-5 (M: Co and Ni).
Fig. 4. o-Xylene uptake on MAPO-5 (M: Co and Ni).
small particles of mixed aluminum and cobalt phosphate at extra-
framework of CoAPO-5 are choked the channels with the increase
in cobalt content.
Fig. 6. IR spectra of pyridine adsorbed on MAPO-5 (M: Co and Ni).
Uptakes of o-xylene on MAPO-5 are shown in Fig. 4. o-Xylene
can easily enter and accommodate in MAPO-5 channels. All MAPO-
5 rapidly uptake o-xylene within 3 min, and then, uptakes are
slowed down. The amounts after 20 min are in the order: Ni(5)APO-
5 « CoAPO-5 ≤ APO-5. The results also indicate that APO-5 and
Co(5)APO-5 have well-ordered microporous channels in them, and
that Ni(5)APO-5 has less crystallinity with low pore volume.
NH3-TPD profiles of MAPO-5 are shown in Fig. 5. Co(5)APO-5
and Ni(5)APO-5 have three types of peak, although there were only
weak peaks on APO-5. The deconvolution of peaks suggests that
they have three major peaks at 200, 380, and 440 ◦C. Peaks at around
200 ◦C are assigned to physically adsorbed NH3 and peaks around
380 ◦C are assigned to strong acid sites expecting active for the acid
catalysis. The wide peaks at around 440 ◦C are also assigned the
desorbed NH3 from strong acid sites. It is possible that 2nd and
3rd peaks involve Brønsted (B-) and Lewis (L-) acid sites [25,26]. To
know the type of acidities, we examine the observation of adsorbed
pyridine on MAPO-5 by IR spectroscopy. Fig. 6 shows the pyridine
adsorption on MAPO-5, where excess of pyridine was evacuated
at 0.5 Torr for 30 min at room temperature. All MAPO-5 samples
have small peaks assigned to pyridine adsorbed on B-acid sites at
1540 cm−1, and another absorption around 1444 cm−1 assigned to
pyridines adsorbed on Lewis acid (L-acid) sites [27]. Unexpectedly,
these results suggest that the B-acid sites, appeared on MAPO-5 by
the isomorphous substitution of M2+ with Al3+ in the frameworks of
APO-5 [7–9], are only in small amounts, probably due to difficulty of
substitution of large Co2+ cation with Al3+ in the frameworks. How-
ever, we consider that they are enough for the acid catalysis, such
as Friedel–Crafts alkylation. There are the examples of high activ-
ities with very small amounts of B-acid sites in the isopropylation
of BP over highly dealuminated MOR [13,17]. The large amounts
of L-acid sites are considered to exist probably on mixed metal and
aluminum phosphate at extra-framework of APO-5 channels; how-
ever, further studies are necessary on the natures of the phosphate
and L-acid sites on them.
The input and output results of the synthesis of MAPO-5 are
shown in Table 1. Molar ratios of (Al + Co)/P were remained around
1.0–1.2. These results mean that the materials obtained by our syn-
thesis are substituted APO-5, and that each components were well
balanced after the synthesis.
From the results, cobalt and nickel species are considered to
exist either as the isomorphously substituted M2+ with Al3+ in the
framework of APO-5, or as mixed metal and aluminum phosphates
in extra-framework of the channels because all MAPO-5 have sim-
ilar XRD patterns with no particular large agglomerates, and the
surface area and pore volume from nitrogen isotherm were grad-
ually decreased with the increase in cobalt content. However, we
have no information of the properties of the mixed phosphates in

H.X. Zhang et al. / Journal of Molecular Catalysis A: Chemical 412 (2016) 117–124
121
reaction. These results indicate that 4-IPBP was consumed steadily
for the formation of 4,4ꢀ- and 3,4ꢀ-DIPB in the channels, and that
3-IPBP was not concerned on the formation of 3,4ꢀ-DIPB during the
reaction even though 3-IPBP was predominant among IPBP isomers
at the late stages.
The selectivities for DIPB isomers are shown in Fig. 8c. The selec-
tivities for the least bulky 4,4ꢀ-DIPB in bulk products were around
70%; and were slightly decreased to 60% at the late stages. The
selectivities for the second least bulky 3,4ꢀ- DIPB were 30–35%
throughout the isopropylation. The selectivities for 4,4ꢀ-DIPB in
encapsulated products were remained almost constant and in the
same levels as in bulk products as in Fig. 8c . These results show
that shape-selective formation of the least bulky 4,4ꢀ-DIPB occurred
from 4-IPBP by the preferential exclusion of the bulkier 3,4ꢀ-DIPB
in CoAPO-5 channels.
The isopropylation of BP over Ni(5)APO-5 was examined to com-
pare the results of Co(5)APO-5 (Fig. 9). The profiles were quite
similar to each other’s: 4-IPBP is almost exclusive precursor of 4,4ꢀ-
and 3,4ꢀ-DIPB during the reaction. NiAPO-5 channels allow prefer-
ential access of 4-IPBP, and prevent the access of 3-IPBP in spite
of predominance of 3-IPBP at the late stages. The selectivities for
4-IPBP and 4,4ꢀ-DIPB in encapsulated products also indicate that
the isopropylation occurs in the NiAPO channels as discussed for
Co(5)APo-5.
These results suggest that the catalyses over the MAPO-5 (M:
Co and Ni) occur at the internal B-acid sites in resembled circum-
stances in their channels with AFI topology, and that these metals
work as active sites for the acid catalysis. The substitution with
transition metals gave no significant influence for the formation of
by-products at such a moderate temperature as 250 ◦C (Figs. 8 and
9).
Fig. 7. The isopropylation of BP over CoAPO-5 (Co: 5-40). Reaction conditions:
BP: 20 mmol; CoAPO-5: 100 mg; temperature: 250 ◦C; propene pressure: 0.8 MPa;
period: 4 h. Legend:
: 4,4ꢀ-DIPB;
: 3,4ꢀ-DIPB;
: 3,3ꢀ-DIPB;
: 2,x’-
DIPB; ꢁ: Conversion (BP).
MAPO-5 channels. Further investigations are necessary for under-
standing their details of the metal species in the channels.
3.2. The isopropylation of BP over MAPO-5 (M: Co & Ni)
3.2.1. The isopropylation over CoAPO-5
Fig. 7 shows the effects of cobalt content of CoAPO-5 on the
isopropylation of BP. Catalytic activities were not found for AlPO-
5 [19]: they appeared by the introduction of cobalt species. They
were the highest at 3–5% of cobalt content; however, decreased
with further increases in cobalt content. The decreases are due
to the decrease in surface area by choking mixed cobalt and alu-
minum phosphate in microporous APO-5 channels as suggested
in previous section. The selectivities for the least bulky 4,4ꢀ-DIPB
were around 70% among DIPB isomers, and the combined selectiv-
ities for 4,4ꢀ-DIPB and the secondly least bulky 3,4ꢀ-DIPB were over
95%: the channels of MAPO-5 were highly shape-selective for the
least bulky 4,4ꢀ-DIPB and the second least bulky 3,4ꢀ-DIPB among
DIPB isomers. These selectivities are almost in the same level as
SSZ-24 and MAPO-5 (M: Si, Mg, Ca, Sr, Ba, Zn) with AFI topology
[14–16,18–20].
3.2.3. The isomerization of 4,4ꢀ-DIPB during the isopropylation of
BP
Figs. 10 shows the effects of reaction temperature on the selec-
tivity of DIPB isomers in the isopropylation of BP over Co(5)APO-5
under 0.8 MPa of propene pressure. The selectivities for 4,4ꢀ-DIPB
in bulk products were as high as 70% at low and moderate tempera-
tures; however, they were gradually decreased with the increase in
the temperatures from around 250–275 ◦C, and then, rapidly with
further increases up to 350 ◦C. The decrease in the selectivities for
4,4ꢀ-DIPB accompanied concurrent increase in those for 3,4ꢀ-DIPB.
The selectivities for 3,3ꢀ-DIPB were gradually increased with the
increase in the temperature; however, remained in the levels of 10%
even at 350 ◦C. Equilibrium composition of DIPB isomers at 300 ◦C:
3,4ꢀ—33%, 3,3ꢀ—37%, 4,4ꢀ—9%, 3,5—17%, and others 4% [28]. These
values suggest 4,4ꢀ-DIPB rapidly isomerizes to 3,4ꢀ-DIPB; however,
further isomerization to equilibrium composition takes long time
at 270–350 ◦C.
3.2.2. Reaction profiles of the isopropylation of BP
Typical reaction profile of the isopropylation of BP over
Co(5)APO-5 under 250 ◦C and 0.8 MPa of propene pressure is sum-
marized in Fig. 8. Fig. 8a shows the yields of IPBP and DIPB isomers
over Co(5)APO-5. 4-IPBP appeared rapidly at low conversion of BP,
saturated at 40–60%, and then, decreased with further increase in
the conversion. 3-IPBP was minor among the isomers at low con-
version. However, 3-IPBP was spontaneously accumulated with
progress of the reaction because 4-IPBP was chosen as the pref-
erential precursor of DIPB isomers even in the presence of large
amounts of 3-IPBP. Further, 2-IPBP, the bulkiest isomer, was also
spontaneously increased although the yield was less than 3%. The
yields of 4,4ꢀ- and 3,4ꢀ-DIPB were increased with the conversion by
accompanying preferential formation of the least bulky 4,4ꢀ-DIPB
and the second least bulky 3,4ꢀ-DIPB among DIPB isomers.
Fig. 8b shows the selectivities for IPBP isomers during the iso-
propylation. 4-IPBP were predominant over 3-IPBP in the ratio of
around 2 at the early stages. The selectivities for 4-IPBP, the least
bulky isomers, were decreased with the increase of the conversion,
accompanying concurrent increase in the selectivities for 3-IPBP.
These changes correspond well to the yields of IPBP isomers as
shown in Fig. 8a. Further, the selectivities for 4-IPBP in encapsulated
products were remained almost constant at around 70% during the
The selectivities for 4,4ꢀ-DIPB in encapsulated products were
remained in the similar levels as those in bulk products up to 275 ◦C.
However, we could not obtain the clear results of the selectivities
for DIPB isomers in encapsulated products at higher tempera-
tures due to disturbance of unidentified components in GC analysis
formed during the reaction and/or during working up.
Fig. 11 shows the effects of reaction temperature on the selec-
tivities for DIPB isomers in the isopropylation over Ni(5)APO-5
under 0.8 MPa of propene pressure. The selectivities for 4,4ꢀ-DIPB
in bulk products were remained at around 70% up to 300 ◦C, and
the increase in the temperature accompanied the decrease in the
selectivities for 3,4ꢀ-DIPB by the isomerization of 4,4ꢀ-DIPB.
The selectivities for 4,4ꢀ-DIPB in encapsulated over Ni(5)APO-5
were remained high in the level of 80% even with the increase in
the temperature. These results mean the isomerization of 4,4ꢀ-DIPB
occurred at the external acid sites, not in the channels of Ni(5)APO-
5.

122
H.X. Zhang et al. / Journal of Molecular Catalysis A: Chemical 412 (2016) 117–124
Fig. 8. Profiles of the isopropylation of BP over Co(5)APO-5. Reaction conditions: BP: 20 mmol; Co(5)APO-5: 100 mg; temperature: 250 ◦C; propene pressure: 0.8 MPa; period:
0.025–4 h.
3.3. Mechanistic aspects of the isopropylation of BP over MAPO-5
state compared to the second least bulky 3,4ꢀ-DIPB by the steric
interaction with the channels through restricted transition state
selectivity. These synergies of two mechanisms enhance the selec-
tivities for 4,4ꢀ-DIPB in the second step from IPBP isomers to DIPB
isomers as proposed in previous papers [10,11,13].
MAPO-5: CoAPO-5 and NiAPO-5 with transition metal sub-
stitution have AFI topology with straight channels of 12-MR
pore entrances [4–9,22]. They are active for acid catalysis in
Friedel–Crafts alkylation of aromatic hydrocarbons such as the iso-
propylation of BP. The activities are due to the B-acid sites, which
appear by the isomorphous substitution of M2+ with Al3+ in the
neutral frameworks of APO-5 [7–9]. There have been some exam-
ples of high activities by small amounts of acid sites over highly
dealuminated MOR in the isopropylation of BP [13,17]. The other
types of acid sites observed in large amounts are L-acid sites due to
mixed metal and aluminum phosphate clusters at extra-frmaework
in the channels. It is possible the latter types of L-acid sites have the
activities for the solid catalysis; however, we consider that they are
not principal for the isopropylation of BP.
The profiles of the isopropylation of BP gave us the interest-
ing features in the catalysis in the AFI channels of MAPO-5. The
isopropylation of BP occurred in the MAPO-5 channels by two con-
secutive steps at moderate temperatures as 250 ◦C. BP gave 4-
and 3-IPBP in the ratio of around 2 (obtained in the early stage);
however, 2-IPBP was formed less than 3%. The channels allow pre-
dominant formation of both of 4- and 3-IPBP by restricted transition
state selectivity mechanism. In the second step from IPBP to DIPB
isomers, 4-IPBP was chosen as a precursor, and bulkier 3-IPBP was
not much concerned the formation of DIPB isomers in spite of its
predominance among IPBP isomers at the late stages. The selec-
tion of 4-IPBP against 3-IPBP occurred through reactant selectivity
mechanism. Resultantly, 4-IPBP formed DIPB isomers with pref-
erential formation of 4,4ꢀ-DIPB through the least bulky transition
It is interesting in discrepancy between bulk and encapsulated
products in the selectivities for IPBP isomers: the selectivities for
4-IPBP were around 80% in encapsulated products; however, 3-
IPBP was predominant over 4-IPBP in bulk products. These results
also mean that 4-IPBP can preferentially access to the channels and
play the role in the precursor to DIPB isomers, irrespective of the
amounts of 3-IPBP in the bulk products, because of the differences
in bulkiness of two isomers. These results also support that the roles
of 3-IPBP were very limited in the formation of 3,4ꢀ-DIPB in spite
of predominance of 3-IPBP at the late stage. The similar profile was
observed in the isopropylation of BP over MOR [10,11,13].
The high selectivities for 4,4ꢀ-DIPB over Co(5)APO-5 and
Ni(5)APO-5 at low and moderate temperatures well correspond to
the previous results that the microporous molecular sieves with AFI
topology, SSZ-24 and MAPO-5 (M: Si, Mg, Ca, Sr, Ba, Zn) are active
for the shape-selective formation of 4,4ꢀ-DIPB with the almost same
level of the selectivities [18–20]. These results support our pro-
posal that oxygen 12-membered rings of AFI topology afford the
same level of shape-selective natures for the isopropylation, by
the synergy of reactant and restricted transition state selectivities
mechanisms [10,11,13].
The effects of reaction temperature suggest another feature of
the isopropylation of BP over MAPO-5. The high selectivities for
4,4ꢀ-DIPB in bulk and encapsulated products at lower temperatures,
such as 250 ◦C, indicate the shape-selective natures of MAPO-5

H.X. Zhang et al. / Journal of Molecular Catalysis A: Chemical 412 (2016) 117–124
123
Fig. 9. Profiles of the isopropylation of BP over Ni(5)APO-5. Reaction conditions: BP: 20 mmol; Ni(5)APO-5: 100 mg; temperature: 250 ◦C; propene pressure: 0.8 MPa; period:
0.05–8 h.
Fig. 10. The effects of temperature on the selectivity for 4,4ꢀ-DIPB in the isopropyla-
tion of BP over Co(5)APO-5. Reaction conditions: BP: 20 mmol; Co(5)APO-5: 100 mg;
temperature: 200–350 ◦C; propene pressure: 0.8 MPa; period: 4 h.
Fig. 11. The effects of temperature on the selectivity for 4,4ꢀ-DIPB in the isopropyla-
tion of BP over Ni(5)APO-5. Reaction conditions: BP: 20 mmol; Ni(5)APO-5: 100 mg;
temperature: 200–350 ◦C; propene pressure: 0.8 MPa; period: 4 h.
channels: the formation of 4,4ꢀ-DIPB occurred from 4-IPBP through
the preferential exclusion of bulky transition states by the chan-
nels, and the roles of 3-IPBP was very limited in the formation of
3,4ꢀ-DIPB due to product selectivity as discussed in previous sec-
tion.
equilibrium. Judging from the selectivities for 4,4ꢀ-DIPB in encap-
sulated products over Ni(5)APO-5, 4,4ꢀ-DIPB was formed inside the
channels, and isomerized to 3,4ꢀ-DIPB at the external acid sites;
not inside the channels. Further isomerization of 3,4ꢀ-DIPB to 3,3ꢀ-
DIPB was not observed at high temperatures, even at 350 ◦C in the
isopropylation of BP over Ni(5)APO-5. However, we cannot dis-
cuss where the isomerization of 4,4ꢀ-DIPB occurs over Co(5)APO-5
because of lack of the information on encapsulated products.
The selectivities for 4,4ꢀ-DIPB in the bulk products were
decreased with the increase in reaction temperatures in the iso-
The decrease in the selectivities for 4,4ꢀ-DIPB occurred with the
increase in temperature starting from 250 ◦C for Co(5)APO-5 and
300 ◦C for Ni(5)APO-5. The decrease in the selectivities for 4,4ꢀ-DIPB
with concurrent increase in the selectivities for 3,4ꢀ-DIPB by the
isomerization to 3,4ꢀ-DIPB directed towards the thermodynamic

124
H.X. Zhang et al. / Journal of Molecular Catalysis A: Chemical 412 (2016) 117–124
propylation of BP over all materials with AFI topology, including
SSZ-24 and MAPO-5 (M: Si, Mg, Zn, Co, and Ni), and their decrease
in the selectivities for 4,4ꢀ-DIPB accompanied by the increase in
those for 3,4ꢀ-DIPB by the isomerization towards thermodynamic
equilibrium [14–16,18–20]. However, there are differences among
the materials where the isomerization of 4,4ꢀ-DIPB occurred at the
external and/or internal acid sites judging from the selectivities
for 4,4ꢀ-DIPB in encapsulated products. The selectivities for 4,4ꢀ-
DIPB in encapsulated products were decreased with temperatures
over SSZ-24 and MAPO-5 (M: Si, Mg and Zn) [14–16,18–20]. These
results indicate that the channels of AFI topology allow the accom-
modation of bulkier 3,4ꢀ-DIPB, resulting in the isomerization at
internal acid sites as well as at external acid sites.
From the results, MAPO-5 (M: Co and Ni) have the same level
of the selectivities for 4,4ꢀ-DIPB to SSZ-24 and MAPO-5 (M: Si, Mg,
and Zn), and these selectivities were decided by AFI channels of
12-membered oxygen ring pore-entrances.
We found the small particles of the mixed metal and aluminum
phosphate in the APO-5 channels. They may have high potentials for
advanced materials, such as super capacitors, etc. Further studies
on the aspects are under progress.
Acknowledgements
Y. Sugi appreciates to The University of Queensland for staying
as a visiting Professor. H.X. Zhang also thanks The University for
staying as the Internship student. The authors would like to extend
their sincere appreciation to the Deanship of Scientific Research at
King Saud University, Riyadh, Sauji Arabia for funding this research
group NO. RG#1435-010.
The isomerization of 4,4ꢀ-DIPB over Ni(5)APO-5 occurred only
at external acid sites, and not at internal acid sites judging from the
selectivities for 4,4ꢀ-DIPB in encapsulated products. This is quite
different from other materials with AFI topology [14–16,18–20].
Currently, we cannot understand why cobalt and nickel promote
the difference in the catalysis from the other materials. We expect
further investigation of the properties of these materials will give
the reasons of the differences.
The characterization of Ni(5)APO-5 indicates its low capacity
of the channels; however, catalytic results are quite similar to
Co(5)APO-5. This result means that AFI channels of Ni(5)APO-5 are
very effectively worked in the isopropylation of BP.
References
[1] P.B. Venuto, Microporous Mater. 2 (1994) 297–411.
[2] S. Al-Khattaf, S.A. Ali, A.M. Aitani, N. Zilková, D. Kubicˇka, J. Cejka, Catal. Rev. 56
ˇ
ˇ
(2014) 333–402.
[3] S.M. Csicsery, Zeolites 4 (1984) 202–213.
[4] S.T. Wilson, B.M. Lok, C.A. Messina, T.R. Cannan, E.M. Flanigen, J. Am. Chem.
Soc. 104 (1982) 1146–1147.
[5] B.M. Lok, C.A. Messina, R.L. Patton, R.T. Gajek, T.R. Cannan, E.M. Flanigen, J.
Am. Chem. Soc. 106 (1984) 6092–6093.
[6] E.M. Flanigen, B.M. Lok, R.L. Patton, S.T. Wilson, Pure Appl. Chem. 58 (1986)
1351–1358.
[7] E.M. Flanigen, B.M. Lok, R.L. Patton, S.T. Wilson, Stud. Surf. Sci. Catal. 39 (1988)
103–112.
4. Conclusion
Transition metal substituted aluminophosphates (MAPO-5; M:
Co and Ni) with AFI topology were applied for the isopropylation
of BP. They were synthesized by dry gel conversion method using
trimethylamine as structure directing agent, and characterized by
XRD, XPS, SEM, N2 adsorption, NH3-TPD, pyridine adsorption and o-
xylene uptake. They are clear crystals without impurity phases and
agglomerates of metal species from XRD and SEM observations.
N2 adsorption and o-xylene uptake indicate that amounts of N2
and o-xylene adsorbed on Ni(5)APO-5 were smaller than those on
Co(5)APO-5. Pyridine adsorption suggests that both of the materials
have only small amounts of B-acid sites in addition to large amounts
of L-acid sites, where B-acid sites are active for the Friedel–Crafts
alkylation.
[8] M. Hartman, L. Kevan, Chem. Rev. 99 (1999) 635–663.
[9] M. Hartman, L. Kevan, Catal. Rev. 28 (2003) 625–695.
[10] Y. Sugi, Y. Kubota, in: J.J. Spivey (Ed.), Catalysis, Specialist Periodical Report,
vol. 13, Royal Soc. Chem., 1997, pp. 55–84, Chapter 3.
[11] Y. Sugi, A. Vinu, Catal. Surv. Asia 19 (2015) 188–200.
[12] T. Matsuzaki, Y. Sugi, T. Hanaoka, K. Takeuchi, H. Arakawa, T. Tokoro, G.
Takeuchi, Chem. Express 4 (1989) 413–416.
[13] Y. Sugi, S. Tawada, T. Sugimura, Y. Kubota, T. Hanaoka, T. Matsuzaki, K.
Nakajima, K. Kunimori, Appl. Catal. A: Gen. 189 (1999) 251–261.
[14] Y. Sugi, Y. Kubota, A. Ito, H. Maekawa, R.K. Ahedi, S. Tawada, S. Watanabe, I.
Toyama, C. Asaoka, H.-S. Lee, J.-H. Kim, G. Seo, Stud. Surf. Sci. Catal. 154 (2005)
2228–2238.
[15] H. Maekawa, A. Ito, H. Kawagoe, K. Komura, Y. Kubota, Y. Sugi, Bull. Chem. Soc.
Jpn. 80 (2007) 215–223.
[16] Y. Sugi, H. Maekawa, A. Ito, C. Ozawa, T. Shibata, A. Niimi, C. Asaoka, K.
Komura, Y. Kubota, J.-Y. Lee, J.-H. Kim, G. Seo, Bull. Chem. Soc. Jpn. 80 (2007)
2232–2242.
[17] G.S. Lee, J.J. Maj, S.C. Rocke, J.M. Garces, Catal. Lett. 2 (1989) 243–247.
[18] M. Bandyopadhyay, R. Bandyopadhyay, S. Tawada, Y. Kubota, Y. Sugi, Appl.
Catal. A: Gen. 225 (2002) 51–62.
[19] H. Maekawa, S.K. Saha, S.A.R. Mulla, S.B. Waghmode, K. Komura, Y. Kubota, Y.
Sugi, J. Mol. Catal. 263 (2007) 238–246.
[20] S.K. Saha, H. Maekawa, S. Komura, Y. Kubota, Y. Sugi, S.J. Cho, Mater. Trans. 46
(2005) 2659–2667.
[21] T. Matsuda, T. Kimura, E. Herawati, C. Kobayashi, E. Kikuchi, Appl. Catal. A:
Gen. 136 (1996) 19–28.
[22] IZA, Structure Commission (2015) http://www.iza-structure.org/databases/.
[23] R. Bandyopadhyay, M. Bandyopadhyay, Y. Kubota, Y. Sugi, J. Porous Mater. 9
(2002) 83–95.
[24] H.F.W.J. van Breukelen, G.J.C. Kraaijveld, L.J.M. van de Ven, J.W. de Haan, J.H.C.
van Hooff, Microporous Mater. 12 (1997) 313–322.
[25] H.-G. Jang, H.-K. Min, S.B. Hong, G. Seo, J. Catal. 299 (2013) 240–248.
[26] F. Hao, J. Zhong, P. Liu, K. You, C. Wei, H. Luo, Chin. J. Catal. 33 (2012) 670–676.
[27] M.I. Zaki, M.A. Hasan, F.A. Al-Sagheer, L. Pasupulety, Colloids Surf. A
Physicochem. Eng. Asp. 190 (2001) 261–274.
[28] G. Takeuchi, H. Okazaki, T. Kito, Y. Sugi, T. Matsuzaki, J. Jpn. Petrol. Inst. 34
(1991) 242–247.
Both of Co(5)APO-5 and Ni(5)APO-5 gave the high selectivities
for 4,4ꢀ-DIPB: 65–75% in both of bulk and encapsulated products
in the isopropylation of BP. 4-IPBP is almost exclusive precursor of
3,4ꢀ- and 4,4ꢀ-DIPB, and 3-IPBP was not significantly concerned in
spite of its predominance among IPBP isomers at the late stages.
The channels of the MAPO-5 allow preferential access of 4-IPBP,
and prevent the access of 3-IPBP due to reactant selectivity mecha-
nism. The selective formation of 4,4ꢀ-DIPB occurred by preferential
exclusion of the second least bulky 3,4ꢀ-DIPB at the transition states
through the steric interaction of transition states against the AFI
channels by the restricted transition state selectivity mechanism.
The selectivities for 4,4ꢀ-DIPB were kept 65–75% at low and mod-
erate temperatures over both of Co(5)APO-5 and Ni(5)APO-5, and
they were decreased with the increase in temperature by the iso-
merization to 3,4ꢀ-DIPB towards thermodynamic equilibrium. The
selectivities for 4,4ꢀ-DIPB in encapsulated products were almost
constant over Ni(5)APO-5, irrespective of temperatures: the iso-
merization of 4,4ꢀ-DIPB occurred at the external acid sites, and not
at the internal acid sites of Ni(5)APO-5. However, we cannot get
the information of encapsulated products in Co(5)APO-5 at higher
temepratures the isomerization occurred, due to the disturbance
of unidentified components in GC analysis formed during the reac-
tion and/or during working up formed during the reaction and/or
during work-up.