Evaluation of iron-containing aluminophosphate molecular sieve catalysts prepared by different methods for phenol hydroxylation

Article abstract of DOI:10.1007/s10562-013-1027-1

Three iron-containing AlPO₄-5 molecular sieve catalysts (i.e. FeAlPO-5-Iono, FeAlPO-5-Hydro and FeAlPO-5-Imp) were prepared by ionothermal synthesis, direct hydrothermal synthesis and impregnation method, and characterized by various techniques. N₂ adsorption/desorption measurements, together with the results of SEM and TEM, showed that FeAlPO-5-Iono was a new type of aluminophosphate molecular sieve with hierarchical micro- and meso-porous structure, whereas both FeAlPO-5-Hydro and FeAlPO-5-Imp were typical microporous materials. DR UV-Vis analysis suggested that the order of the amounts of isolated Fe³⁺ species in these three samples was as follows: FeAlPO-5-Hydro > FeAlPO-5-Iono > FeAlPO-5-Imp. However, the order of their catalytic performances towards phenol hydroxylation was in opposition to that of the amounts of isolated Fe³⁺ species. Detailed analysis revealed that it was textural properties of these catalysts that determined phenol conversion and dihydroxylbenzene selectivity. The catalyst-recycling tests suggested that extra-framework iron clusters were readily leached into the reaction system and had great influence on the hydroxylation of phenol. Due to the sole presence of isolated Fe³⁺ species and enhanced mass transport of reactants and products to/from these active sites, FeAlPO-5-Hydro was found to have the best reusability among these catalysts.

Full text of DOI:10.1007/s10562-013-1027-1

Catal Lett (2013) 143:657–665
DOI 10.1007/s10562-013-1027-1
Evaluation of Iron-Containing Aluminophosphate Molecular
Sieve Catalysts Prepared by Different Methods for Phenol
Hydroxylation
•
Xinhong Zhao Zhongpeng Sun Zhaoqi Zhu
•
•
•
An Li Guixian Li Xiaolai Wang
•
Received: 18 December 2012 / Accepted: 21 May 2013 / Published online: 4 June 2013
Ó Springer Science+Business Media New York 2013
Abstract Three iron-containing AlPO4-5 molecular sieve
catalysts (i.e. FeAlPO-5-Iono, FeAlPO-5-Hydro and FeAl-
PO-5-Imp) were prepared by ionothermal synthesis, direct
hydrothermal synthesis and impregnation method, and
characterized by various techniques. N2 adsorption/desorp-
tion measurements, together with the results of SEM and
TEM, showed that FeAlPO-5-Iono was a new type of
aluminophosphate molecular sieve with hierarchical micro-
and meso-porous structure, whereas both FeAlPO-5-Hydro
and FeAlPO-5-Imp were typical microporous materials.
DR UV–Vis analysis suggested that the order of the amounts
of isolated Fe3? species in these three samples was as
follows: FeAlPO-5-Hydro [ FeAlPO-5-Iono [ FeAlPO-5-
Imp. However, the order of their catalytic performances
towards phenol hydroxylation was in opposition to that
of the amounts of isolated Fe3? species. Detailed analysis
revealed that it was textural properties of these catalysts
that determined phenol conversion and dihydroxylbenzene
selectivity. The catalyst-recycling tests suggested that extra-
framework iron clusters were readily leached into the reac-
tion system and had great influence on the hydroxylation of
phenol. Due to the sole presence of isolated Fe3? species and
enhanced mass transport of reactants and products to/from
these active sites, FeAlPO-5-Hydro was found to have the
best reusability among these catalysts.
Keywords FeAlPO-5 molecular sieve ꢀ Ionothermal ꢀ
Hierarchical ꢀ Phenol hydroxylation
1 Introduction
Hydroxylation of phenol to catechol (CAT) and hydro-
quinone (HQ) is an industrially demanding reaction, as the
products have several important large-scale uses such as
photographic film developers, antioxidants, polymerization
inhibitors, medicines, and perfumes [1]. Therefore much
attention has been devoted to developing high-performance
catalysts for synthesis of these dihydroxy compounds. A
series of heterogeneous catalysts such as TS-1, Ti-MWW,
metal oxides, heteropoly compounds have been studied
from both fundamental and industrial standpoints [2–6].
However, their industrial applications were limited by
higher prices and lower activities of the catalysts [7].
Iron-incorporated microporous or mesoporous molecu-
lar sieves are characterized as suitable catalysts for phenol
hydroxylation reaction because iron shows higher activity
than other transition metals [8]. However, the sole presence
of micropores in microporous crystals imposes diffusion
limitations on the reaction rate that restricts the efficacy of
the porous materials [9]. The problem of mass transfer is
solved successfully in mesoporous molecular sieves,
however, the walls of these mesoporous molecular sieves
are comprised of incompletely cross-linked amorphous
frameworks and hence structural stability is low [10]. For
diminishing the diffusion limitations and improving sta-
bility, hierarchically structured molecular sieves have
attracted ever-increasing attention in recent years [11–13].
In our previous work, FeAlPO-5 molecular sieve with
hierarchical micro- and meso-porous structure was suc-
cessfully prepared by ionothermal method [14]. This
X. Zhao (&) ꢀ Z. Sun ꢀ Z. Zhu ꢀ A. Li ꢀ G. Li
School of Petrochemical Engineering, Lanzhou University
of Technology, Lanzhou 730050, People’s Republic of China
e-mail: licpzhaoxh@lut.cn
X. Wang
State Key Laboratory for Oxo Synthesis and Selective
Oxidation, Lanzhou Institute of Chemical Physics,
Chinese Academy of Sciences, Lanzhou 730000,
People’s Republic of China
123

658
X. Zhao et al.
2.2 Characterization of Catalysts
material may have great potential in the hydroxylation of
phenol.
Powder X-ray diffraction (XRD) patterns of the synthesized
materials were recorded on a D/Max-2400 Rigaku diffrac-
tometer with Cu Ka radiation operated at 40 kV and 40 mA.
The content of iron in the final product was quantified by a
Rigaku JY38S ICP-AES. Nitrogen adsorption/desorption
studies were conducted on a Micromeritics ASAP 2020
surface area and pore size analyzer at 77 K. Samples were
outgassed at 473 K for 1.5 h prior to measurements. Spe-
cific surface areas of materials were calculated from the
adsorption data obtained at p/p0 between 0.07 and 0.20,
using the Brunauer–Emmett–Teller (BET) equation. The
micropore volumes were determined by the t-plot method.
The mesopore parameters were determined from adsorption
branches by BJH method. Diffuse reflectance UV–Vis
spectra (DR UV–Vis) were measured with a spectrometer
of Shimadzu UV-240, and BaSO4 was an internal standard
sample. In order to compare the relative amounts of various
iron species in the three catalysts, origin 6.0 software and
Gaussian equation were used to analyze the UV bands.
Scanning electron microscope images were obtained from a
SEM (JSM-6701F) instrument. Transmission electron
micrographs (TEM) were obtained on a JEOL JEM-2010
microscope with a point resolution of 0.2 nm at 200 kV. For
TEM, the sample was ground, dispersed in ethanol, and
deposited on a holey carbon film supported on a copper grid.
A large number of documents have expressed that for iron-
based catalysts the nature and distribution of iron species and
pore structure were two vital factors affecting their catalytic
activities in phenol hydroxylation [15–17]. In this work, to
make clear how the two factors determine their catalytic prop-
erties, three iron-containing AlPO4-5 molecular sieve catalysts
(AFI topology) exhibiting different distribution of iron species
and textural properties were prepared by ionothermal synthesis,
direct hydrothermal synthesis and impregnation method, and
characterized by various techniques, and compared with each
other in detail with respect to catalytic performances in the
hydroxylation of phenol. Some useful conclusions were drawn
from the characterization and catalytic results.
2 Experimental
2.1 Preparation of Catalysts
Three iron-containing AlPO4-5 molecular sieve catalysts
prepared by different methods were denoted as FeAlPO-5-
Iono, FeAlPO-5-Hydro and FeAlPO-5-Imp, respectively.
The FeAlPO-5-Iono catalyst was prepared by ionother-
mal method according to following procedure. (1) Synthesis
of eutectic mixture solvent (denoted as EU): The as-bought
succinic acid, choline chloride and tetraethylammonium
bromide were measured out in a molar ratio of 8:8:1 and
ground in a mortar. (2) A round-bottom flask was charged
with the eutectic mixture solvent, aluminum isopropoxide,
phosphoric acid (85 wt% in water), ammonium ferric citrate,
hydrofluoric acid (40 wt% in water) and deionized water.
The molar ratio of initial mixture was 62EU:1.1P2O5:A-
l2O3:1.5HF:0.6Fe2O3. The round-bottom flask containing
the final reaction mixture was placed in a microwave reaction
system and heated to 453 K at about 20 K/min, then kept
under this temperature and ambient pressure for 1 h. After
the synthesis mixture had cooled to room temperature, the
solid product was recovered by diluting the final mixture
with distilled water and subsequent filtration. The resulting
product was washed thoroughly with acetone and distilled
water, and then dried in air to get the as-synthesized sample.
The FeAlPO-5-Iono catalyst, the iron content of which was
0.41 wt% determined by inductively coupled plasma atomic
emission spectrometer (ICP-AES), was obtained by calcin-
ing the as-synthesized sample at 823 K for 6 h.
2.3 Catalytic Reaction
In a standard reaction, 2 g of phenol and 0.10 g of catalyst
were added to 24 mL of deionized water in a two neck flask
equipped with a magnetic stirrer and a reflux condenser.
After heating the reaction mixture to designated temperature,
2.22 mL of hydrogen peroxide (H2O2, 30 wt% aqueous
solution) was added through a syringe to the phenol solution
containing catalyst (molar ratio of phenol/H2O2 = 1/1). The
product distributions were determined by a LC2000 HPLC
(Tianmei, Shanghai, China) equipped with a reversed phase
C18 column using the methanol/water mixture (40/60, vol-
ume ratio) as the mobile phase at the flow rate of 1.0 mL/min
with UV detection at 277 nm. The used catalyst was filtrated,
dried and calcined at 823 K before being recycled.
3 Results and Discussion
3.1 Characterization of Catalysts
For comparison, FeAlPO-5-Hydro [18] catalyst was
synthesized by direct hydrothermal method and FeAlPO-5-
Imp [19] was prepared by impregnation method based on
the literature. The iron contents of FeAlPO-5-Hydro and
FeAlPO-5-Imp determined by ICP-AES were 0.38 and
0.42 wt%, respectively.
Figure 1 illustrates the XRD patterns of the three samples
(i.e. FeAlPO-5-Iono, FeAlPO-5-Hydro and FeAlPO-5-Imp).
As Fig. 1 displays, these calcined samples presented the
identical characteristic peaks at 2h = 7.38°, 12.84°, 14.86°,
123

Evaluation of Iron-Containing AlPO4-5 Molecular Sieve Catalysts
659
300
250
200
150
100
50
0
5
10 15 20 25 30 35 40 45 50
Pore diameter (nm)
FeAlPO-5-Imp
FeAlPO-5-Hydro
FeAlPO-5-Iono
0
10
20
30
40
50
0.0
0.2
0.4
0.6
0.8
1.0
Relative Pressure (p/po)
2 theta (degree)
Fig. 1 XRD patterns of FeAlPO-5-Iono, FeAlPO-5-Hydro and
FeAlPO-5-Imp
Fig. 2 N2 adsorption/desorption isotherm of FeAlPO-5-Iono (Inset is
the corresponding pore size distribution)
Table 1 N2 physicosorption data of FeAlPO-5-Iono, FeAlPO-5-
Hydro, and FeAlPO-5-Imp
19.74°, 20.94°, and 22.38° consisting with the diffractions
of (100), (110), (200), (210), (002), and (211) of AlPO4-5
[JCPDS: 41-0044], which demonstrated that these samples
have accordant AFI topology structure. The characteristic
XRD patterns of these catalysts are in consonance with
those reported in the literature [20–22]. It is worthwhile to
note that two additional small peaks appeared at 20.4° and
21.5° can be found for the XRD patterns of the sample
FeAlPO-5-Hydro, which could be assigned to the phase of
FeAl2(PO4)2(OH)2 formed [23]. According to the inter-
pretation by Cheng et al. [23] and Wei et al. [24], the
generation of this phase is caused by the excessive amount
of iron present and high temperature calcinations. In our
case, the synthetic conditions of the two catalysts FeAlPO-
5-Iono and FeAlPO-5-Imp are quite different from that of
FeAlPO-5-Hydro. Thus, we speculate that the formation
of additional phase for FeAlPO-5-Hydro, on the other
hand, should be related with its particular hydrothermal
condition.
Sample
BET surface
area (m2/g)
Micropore
volumea
(cm3/g)
Mesopore
volumeb
(cm3/g)
FeAlPO-5-Iono
FeAlPO-5-Hydro
FeAlPO-5-Imp
a
330
285
96
0.11
0.11
0.05
0.22
0.04
0
Calculated by t-plot method
b
Calculated by BJH method
with hierarchical micro- and meso-porous structure,
whereas both FeAlPO-5-Hydro and FeAlPO-5-Imp are
typical microporous materials. Additionally, one can see
that the BET surface area and micropore volume of Fe-
AlPO-5-Imp are much smaller than those of the other two
samples. Such a result may be due to that the ferric salt
partly blocks the pores of support during the impregnation
process.
The N2 adsorption/desorption isotherm and corre-
sponding pore size distribution (PSD) curve for the typical
sample FeAlPO-5-Iono are presented in Fig. 2 (For clarity,
the isotherms of the other two catalysts are not shown). It is
obvious that the sample exhibits a hysteresis loop appear-
ing from relative pressure of p/p0 = 0.4, confirming the
presence of mesopores in the FeAlPO-5-Iono crystals [25].
Moreover, a steep adsorption step at relative pressure of
p/p0 [ 0.8 can be observed, indicating the presence of
mesopores with larger size [26]. As expected, the PSD
curve of FeAlPO-5-Iono (Fig. 2, inset) shows two maxima,
one at 9 nm and other at 30 nm with a broad PSD.
The nitrogen physisorption data of the three samples are
summarized in Table 1. As Table 1 displays, FeAlPO-5-
Iono is a new type of aluminophosphate molecular sieve
The nature and distribution of iron species in Fe-incor-
porated zeolites are commonly assessed by UV–Vis diffuse
reflectance spectroscopy. After suitable deconvolution of
the DR UV–Vis spectra of the three samples (Fig. 3),
several characteristic UV–Vis absorption bands can be
extracted and attributed to the specific ferric ion species,
the proportion of which shown in Table 2 thus can be
estimated. Generally, the two bands at ca. 210 and 270 nm
can be assigned to isolated Fe3? ions in tetrahedral and
octahedral coordination, respectively [27, 28]. One band
located at 336 nm is caused by small oligonuclear Fe3xþOy
clusters in the zeolite channel and surface [27, 28], and the
bands at above 400 nm (centered at 460 and 550 nm) are
assigned to Fe2O3 nanoparticles at the external surface of
123

660
X. Zhao et al.
samples is as follows: FeAlPO-5-Hydro [ FeAlPO-5-
Iono [ FeAlPO-5-Imp.
0.5
0.4
0.3
0.2
0.1
0.0
The SEM micrographs of FeAlPO-5-Iono, FeAlPO-5-
Hydro and FeAlPO-5-Imp are presented in Fig. 4a–c.
Evidently, there is great difference among them in terms of
morphology. FeAlPO-5-Iono is comprised of irregular
column-shaped crystals with rough surface. Particularly,
considerable amounts of interparticle pores can be clearly
observed in the surface of this sample, implying that it is a
mesoporous material. By contrast, both FeAlPO-5-Hydro
and FeAlPO-5-Imp are consisted of regular smooth crys-
tals. No intercrystalline void space can be found for these
two samples. FeAlPO-5-Hydro appears as polycrystalline
aggregates with plate-like morphology. For FeAlPO-5-
Imp, both long needled- and hexagonal column-shaped
crystals can be observed. The TEM micrographs of the
three samples are shown in Fig. 4d–f. For FeAlPO-5-Iono,
bright spots appear due to differences in thickness, which
correspond to mesopores [29]. As comparison, no light
spots are visible for FeAlPO-5-Hydro and FeAlPO-5-Imp,
indicating that no mesopores are formed. The SEM and
TEM results, together with nitrogen physisorption charac-
terization, fully prove that FeAlPO-5-Hydro and FeAlPO-
5-Imp are typical microporous materials, and FeAlPO-5-
Iono possesses hierarchical micro- and meso-porous
structure.
FeAlPO-5-Imp
200
400
600
800
Wavelength (nm)
1.2
1.0
0.8
0.6
0.4
0.2
FeAlPO-5-Hydro
200
400
600
800
3.2 Phenol Hydroxylation Reaction
Wavelength (nm)
For the hydroxylation of phenol by H2O2, iron-incorpo-
rated molecular sieve is one of the most studied catalysts.
The effects of various reaction parameters on phenol
hydroxylation over these catalysts have been thoroughly
investigated [15, 17]. The results suggest that water is the
best solvent and the optimum molar ratio of phenol to
H2O2 is 1:1. Therefore, subsequent investigations will be
carried out under these optimized conditions.
0.6
0.4
0.2
0.0
FeAlPO-5-Iono
Phenol hydroxylation has been known to proceed via a
redox mechanism involving Fe(III)/Fe(II) redox pair as
described in Scheme 1 [16], and the ability of the frame-
work transition metal in zeolitic materials to change oxi-
dation states between (II) and (III) is well-known [30]. In
the present work, phenol hydroxylation was conducted
over three iron-containing catalysts prepared by different
procedures with the same level of Fe content. The exper-
imental findings were interpreted based on the redox
mechanism proposed.
200
400
600
800
Wavelength (nm)
Fig. 3 UV-Vis diffuse reflectance spectra of FeAlPO-5-Iono, FeAl-
PO-5-Hydro, and FeAlPO-5-Imp
zeolite crystal [27, 28]. As can be seen from Table 2, all
iron species in FeAlPO-5-Hydro are isolated Fe3? ions,
whereas FeAlPO-5-Iono and FeAlPO-5-Imp possess large
amounts of oligonuclear Fe3xþOy clusters and Fe2O3
nanoparticles in addition to isolated Fe3? species. The
order of the amounts of isolated Fe3? species in these three
FeAlPO-5-Iono catalyst is selected to investigate the
influence of reaction temperature on its catalytic perfor-
mance in the hydroxylation of phenol (Table 3). At 313 K,
only 2.25 % phenol conversion is obtained on this catalyst,
which indicates that no enough energy can be utilized to
generate OH radicals from H2O2 at this low reaction
123

Evaluation of Iron-Containing AlPO4-5 Molecular Sieve Catalysts
661
Table 2 Percentage of subband areas derived by deconvoluting UV–Vis spectra and corresponding Fe species loading
Sample
Isolated Fe3? a
Oligonuclear
Fe3xþOy A3 (%)b
Fe2O3 nanoparticlesc
Isolated
(wt%)
Oligonuclear
(wt%)
Nanoparticles
(wt%)
A1 (%)
A2 (%)
A4 (%)
A5 (%)
FeAlPO-5-Iono
FeAlPO-5-Hydro
FeAlPO-5-Imp
a
33.9
32.0
29.3
29.8
68.0
27.8
20.7
*0
3.67
*0
11.9
*0
0.26
0.38
0.24
0.08
*0
0.06
*0
26.5
7.32
8.98
0.11
0.07
A1 and A2 represent the percentage of subband area at 210 and 270 nm, respectively
b
c
A3 represents the percentage of subband area at 336 nm
A4 and A5 represent the percentage of subband area at 460 and 550 nm, respectively
(a)
(b)
(c)
(d)
(e)
(f)
Fig. 4 SEM (left) and TEM (right) micrographs of FeAlPO-5-Iono (a, d), FeAlPO-5-Hydro (b, e) and FeAlPO-5-Imp (c, f)
123

662
X. Zhao et al.
most active one among these catalysts. It is well known
that the framework iron species are found to be more active
than the extra-framework iron oligomer in the hydroxyl-
ation of phenol over iron-incorporated mesoporous
molecular sieve catalysts [7, 17, 31, 32]. In our case, DR
UV–Vis results had suggested that FeAlPO-5-Imp pos-
sessed the maximum amount of extra-framework iron oli-
gomer among these three catalysts. Thus the catalytic
activity of FeAlPO-5-Imp should not be superior to that of
FeAlPO-5-Iono and FeAlPO-5-Hydro. It should be noted
that the pore structure also has much influence on the
catalytic performance of iron-incorporated molecular sieve
catalysts besides the distribution of iron species. In the
light of the characterization results of DR UV–Vis and
nitrogen physisorption, a profile (Scheme 2) illustrating the
sites of iron species in the channel of the three catalysts can
be schemed out and used to explain the present catalytic
results. In fact, the microporous volume of FeAlPO-5-Imp
is very low, as shown in Table 1; it can be defined as a
nonporous catalytic material to some extent. Thus the
catalytic reaction mainly occurs on the external surface or
pore mouth of this catalyst and superior mass transport of
reactants and products to/from active sites can be expected.
In contrast, in the case of FeAlPO-5-Hydro and FeAlPO-5-
Iono with larger pore volume the reactants firstly need to
diffuse into the micropore and hierarchical pore channels,
respectively, and then the catalytic reaction can take place
on the active sites located on the pore channels. However,
as far as liquid-phase reaction happened within the long
narrow channels of molecular sieves is concerned, the
diffusion rate of reactant is often not fast. Diffusion in
microporous solids may be the rate-controlling step.
Hence, the catalytic activities of FeAlPO-5-Iono and Fe-
AlPO-5-Hydro are lower than that of FeAlPO-5-Imp.
Moreover, compared with FeAlPO-5-Hydro, FeAlPO-5-
Iono perhaps can confer a relatively fast diffusion rate of
reactant due to the presence of mesoporosity. Then a rel-
atively fast reaction rate with FeAlPO-5-Iono as a catalyst
is observed. Based on above discussion, the differences in
catalytic activity can be well explained among these three
catalysts. The TOF values for FeAlPO-5-Iono, FeAlPO-5-
Hydro and FeAlPO-5-Imp were calculated based on the
Scheme 1 Mechanism of the phenol hydroxylation [16, 31] cata-
lyzed by iron-containing AlPO4-5 molecular sieves
temperature and the phenol hydroxylation proceeds very
slowly. As the reaction temperature is increased from 313
to 353 K, both phenol conversion and the molar ratio of
HQ/CAT increases gradually. Similar results had been
observed in the work of Li and coworkers [17]. Dihydr-
oxylbenzene (CAT and HQ) selectivity does not change
much with increasing reaction temperature. Except for di-
hydroxylbenzene, large amount of tar is also formed in the
hydroxylation of phenol, resulting in the low selectivity of
dihydroxylbenzene. At temperature above 353 K, conver-
sion of phenol is expected to decrease due to enhanced
H2O2 decomposition as verified by many researchers [7,
17, 31]. Therefore further studies over these catalysts will
be carried out at 353 K.
Table 4 presents the effect of reaction time on phenol
hydroxylation over newly prepared FeAlPO-5-Iono, Fe-
AlPO-5-Hydro and FeAlPO-5-Imp catalysts. It is notice-
able that these three catalysts exhibit great differences in
terms of catalytic activity. For FeAlPO-5-Imp, the time
required to arrive at the maximum phenol conversion is
within 1 h, whereas for FeAlPO-5-Iono and FeAlPO-5-
Hydro it is 2 and 4 h, respectively. FeAlPO-5-Imp is the
Table 3 Hydroxylation of phenol catalyzed by FeAlPO-5-Iono at different temperatures
Temperature (K)
Conversion of phenol (%)
Yield (CAT ? HQ) (%)
Product selectivity
(CAT?HQ) (%)
HQ/CAT (molar ratio)
313
333
353
2.25
11.8
20.0
1.36
6.82
11.8
60.3
57.8
59.1
0.48
0.58
0.70
Reaction conditions: 2.0 g of phenol, 0.10 g of FeAlPO-5-Iono, 24 mL of deionized water as solvent, H2O2 (30 wt% aqueous solution) as
oxidant, phenol/H2O2 = 1/1 (molar ratio), reaction temperature from 313 to 353 K, reaction time for 4 h
123

Evaluation of Iron-Containing AlPO4-5 Molecular Sieve Catalysts
663
Table 4 Effect of reaction time on phenol hydroxylation over FeAlPO-5-Imp, FeAlPO-5-Hydro and FeAlPO-5-Iono catalysts
Catalyst
Iron content Reaction time (h) Phenol conversion Yield (CAT?HQ) TOF (h-1 c
)
Product selectivity
(wt%)a
(%)b
(%)b
(CAT?HQ) (%)b HQ/CATb
FeAlPO-5-Imp
0.42
1
2
3
4
1
22.7|24.1|22.0
20.6|20.4|21.7
20.2|22.2|20.4
22.4|22.2|21.9
11.4|10.1|14.2
12.8|11.8|10.0
12.8|12.3|12.3
13.0|12.3|13.8
12.7|11.7|12.4
1.85|3.74|6.55
363
58
56.3|49.1|45.5
62.1|60.5|56.7
64.2|55.3|67.6
56.5|53.2|56.6
16.2|37.0|46.1
0.81|0.76|0.63
0.81|0.75|0.78
0.80|0.80|0.80
0.81|0.78|0.77
0.58|0.48|0.64
FeAlPO-5-
Hydro
0.38
0.41
2
3
4
1
2
3
4
13.3|18.2|19.3
17.7|17.4|18.7
19.0|19.1|19.5
10.8|13.6|20.2
18.5|20.6|22.0
19.5|20.2|24.4
20.0|19.8|21.6
5.24|10.2|11.1
7.58|11.9|12.4
10.1|11.8|11.5
4.24|3.70|2.32
11.7|11.8|11.1
11.7|12.2|11.0
11.8|12.0|11.5
39.4|56.0|57.5
42.8|68.3|66.3
53.0|61.8|59.0
39.3|27.2|11.3
63.0|57.5|50.2
59.8|60.4|45.1
59.1|60.8|53.2
0.61|0.68|0.71
0.64|0.71|0.72
0.69|0.72|0.71
0.51|0.55|0.50
0.71|0.73|0.65
0.71|0.73|0.73
0.70|0.74|0.74
FeAlPO-5-Iono
123
Bold values indicate the differences of catalytic performances for the three catalysts
Reaction conditions: 2.0 g of phenol, 0.10 g of catalyst, 24 mL of deionized water as solvent, H2O2 (30 wt% aqueous solution) as oxidant,
phenol/H2O2 = 1/1 (molar ratio), reaction temperature at 353 K, reaction time from 1 to 4 h
a
Determined by ICP-AES
b
From left to right the three numbers placed in each cell were separated by vertical lines and belonged to fresh catalysts, recycled one time and
recycled two times
c
Mol dihydroxylbenzene formed per mol Fe per hour
yield of dihydroxylbenzene, and shown in Table 4. Obvi-
ously, FeAlPO-5-Imp is the most active one among these
three catalysts, whose TOF value is as high as 363 h-1. As
compared with MFS-n [17] and TS-1 [13] in the hydrox-
ylation of phenol with similar reaction parameters, the TOF
values reported here are lower than those of MFS-n, but
much higher than TS-1.
may have faster diffusion rate than CAT owing to their
different molecular sizes and symmetries, while CAT is
considered to be the most thermodynamically favored isomer
in the phenol hydroxylation [8, 15, 32, 33]. As a result, a low
molar ratio of HQ/CAT can be observed at the initial stage.
The molar ratio increases gradually as the reaction proceeds
due to the superior mass transport of HQ from active centers
relative to CAT.
As illustrated in Table 4, selectivity towards dihydr-
oxylbenzene over fresh FeAlPO-5-Imp is also different
from that over fresh FeAlPO-5-Iono and FeAlPO-5-Hydro
in addition to catalytic activity. For FeAlPO-5-Imp, di-
hydroxylbenzene selectivity and the molar ratio of HQ/
CAT keep near constant in the whole range of reaction time
investigated, whereas for FeAlPO-5-Hydro and FeAlPO-5-
Iono catalysts it is seen that both dihydroxylbenzene
selectivity and the molar ratio of HQ/CAT increase grad-
ually with the reaction time, but starting from a lower level,
and reach their maxima within 4 and 2 h, respectively. It is
known that the pore structure of these catalysts influences
not only the mass transport of reactants to active centers,
but also that of products from active sites. The order of the
diffusion rate of products over these catalysts should be as
follows: FeAlPO-5-Hydro\FeAlPO-5-Iono\FeAlPO-5-Imp.
The relative amounts of dihydroxylbenzene in the solution
that can be detected by HPLC also follow the same order as
that of the diffusion rate of products. Then the variation of
dihydroxylbenzene selectivity with reaction time can be well
explained based on above analysis. On the other hand, HQ
In the present study, the catalytic stabilities of FeAlPO-
5-Imp, FeAlPO-5-Hydro and FeAlPO-5-Iono were also
investigated under the same reaction conditions. It can be
seen from Table 4 that using FeAlPO-5-Hydro after 2 h
reaction the conversion of phenol is increased from 13.3 to
18.2 % and 19.3 % after the first and second recycling.
Similar case can be observed with FeAlPO-5-Iono as the
catalyst. Why are the recycled catalysts more active than
the fresh catalysts with respect to phenol conversion? The
reason is very simple and the explanation is as follows.
Firstly, On the basis of the reaction mechanism shown in
Scheme 1, H? is predicted to form during the reactions. On
further measurements it is found that the pH of the reaction
system significantly decreases from 6.0 to ca. 2 after 4 h
reaction. Therefore, it can be conjectured that under such
acidic condition iron clusters located at extra framework
must partly be removed but isolate iron species remain
more or less preserved. A similar result had been reported
by Ahn and coworkers [31]. ICP-AES measurements show
that after three uses the iron contents of FeAlPO-5-Imp and
123

664
X. Zhao et al.
Scheme 2 Sites of iron species
in the channel of FeAlPO-5-
Imp, FeAlPO-5-Hydro and
FeAlPO-5-Iono
FeAlPO-5-Hydro
FeAlPO-5-Imp
Fe2O3 nanoparticles
isolated Fe3+
oligonuclear Fe3+xOy
FeAlPO-5-Iono
FeAlPO-5-Iono have declined from 0.42–0.41 to 0.13–0.35
wt%, respectively. Ahn and coworkers [31] found that
using Fe(NO3)3 salt as a homogeneous catalyst 70 %
phenol conversion can be obtained within less than 5 min,
accompanied by the formation of large amount of tar. This
result implicates that the enhanced phenol conversion for
the recycled FeAlPO-5-Iono catalyst should be due to the
metal leaching. Secondly, ICP-AES analysis confirms that
no iron leaching occurs in FeAlPO-5-Hydro catalyst. Thus
the higher phenol conversion for recycled FeAlPO-5-
Hydro should be attributed to other factors. We speculate
that the pore structure of FeAlPO-5 may be modified by the
acidic environment during the reactions. N2 adsorption/
desorption characterization reveals that after three uses the
microporous volume and mesoporous volume of FeAlPO-5
are changed from 0.11–0.04 to 0.08–0.16 cm3/g, respec-
tively. The accessibility of reactants to the iron species is
thus improved by this modification. Therefore, in com-
parison to fresh FeAlPO-5-Hydro, a higher phenol con-
version can be expected with recycled FeAlPO-5-Hydro as
the catalyst.
indicating that this catalyst has a better reusability. For
FeAlPO-5-Imp and FeAlPO-5-Iono the decrease in the
yields of dihydroxylbenzene can be explained by the metal
leaching because the iron species stepwise leached into the
reaction system are believed to exhibit low selectivity to
dihydroxylbenzene [31]. As for FeAlPO-5-Hydro, it is
reasonable to assume that both the high stability of isolated
iron species and enhanced mass transport of reactants and
products to/from active sites are responsible for its excel-
lent reusability.
4 Conclusions
In summary, FeAlPO-5-Iono, FeAlPO-5-Hydro and FeAl-
PO-5-Imp were prepared by ionothermal synthesis, direct
hydrothermal synthesis and impregnation method. The
characterization results indicated that there are great dif-
ferences among these catalysts in terms of pore parameters
and the distribution of iron species. The comparisons of
their catalytic results for phenol hydroxylation revealed that
phenol conversion and selectivity towards dihydroxylben-
zene were mainly determined by the textural properties. The
catalyst-recycling tests suggested that extra-framework iron
clusters were easy leached into the reaction system and had
great influence on the hydroxylation of phenol. The catalyst
On the other hand, it is interesting to note that for Fe-
AlPO-5-Imp and FeAlPO-5-Iono catalysts the yields of
dihydroxylbenzene within 1 h reaction time gradually
decrease as the numbers of catalysts recycling increase,
whereas for FeAlPO-5-Hydro an opposite tendency is seen,
123

Evaluation of Iron-Containing AlPO4-5 Molecular Sieve Catalysts
665
11. Hua ZL, Zhou J, Shi JL (2011) Chem Commun 47:10536
12. Zhou L, Lu T, Xu J, Chen M, Zhang C, Chen C, Yang X, Xu J
(2012) Micropor Mesopor Mater 161:76
13. Xin H, Zhao J, Xu S, Li J, Zhang W, Guo X, Hensen EJM, Yang
Q, Li C (2010) J Phys Chem C 114:6553
14. Zhao X, Wang H, Dong B, Sun Z, Li G, Wang X (2012) Mi-
cropor Mesopor Mater 151:56
15. Parida KM, Singha S, Sahoo PC (2010) Catal Lett 136:155
16. Liu CB, Shan YJ, Yang XG, Ye XK, Wu Y (1997) J Catal 168:35
17. Xin H, Liu J, Fan F, Feng Z, Jia G, Yang Q, Li C (2008) Mi-
cropor Mesopor Mater 113:231
with the sole presence of isolated iron species, i.e. FeAlPO-
5-Hydro, was found to be the most stable one. As is
expected, those iron-containing catalysts with stable active
iron species that can be readily accessible by reactants and
products are bound to exhibit superior catalytic perfor-
mances in phenol hydroxylation. The results reported here
perhaps provided some valuable information for designing
high-performance iron-based catalysts.
18. Catana G, Pelgrims J, Schoonheydt RA (1995) Zeolites 15:475
19. Cheng D-g, Zhao X, Chen F, Zhan X (2009) Catal Commun
10:1450
20. Hentit H, Bachari K, Ouali MS, Womes M, Benaichouba B,
Jumas JC (2007) J Mol Catal A 275:158
Acknowledgments This work was supported by the National Nat-
ural Science Foundation of China (Grant No. 51263012), the Natural
Science Foundation of Gansu Province, China (Grant No.
1112RJZA020) and Development Program of Lanzhou University of
Technology for excellent teachers (Grant No. Q201113).
21. Chen F, Do MH, Zheng W, Cheng D-g, Zhan X (2008) Catal
Commun 9:2481
22. Shiju NR, Fiddy S, Sonntag O, Stockenhuber M, Sankar G (2006)
Chem Commun 4955
References
23. Cheng D-G, Chen F, Zhan X (2012) Appl Catal A 435:27
24. Wei W, Moulijn JA, Mul G (2008) Micropor Mesopor Mater
112:193
1. Rocha GO, Johnstone RAW, Hemming BF, Pires PJC, Sankey JP
(2002) J Mol Catal A 186:127
2. Tuel A, Moussakhouzami S, Bentaarit Y, Naccache C (1991) J
Mol Catal 68:45
25. Koo J-B, Jiang N, Saravanamurugan S, Bejblova M, Musilova Z,
Cejka J, Park S-E (2010) J Catal 276:327
3. Wang DA, Liu ZQ, Liu FQ, Ai X, Zhang XT, Cao Y, Yu JF, Wu
TH, Bai YB, Li TJ, Tang XY (1998) App Catal A 174:25
4. Zhang HP, Zhang XM, Ding Y, Yan L, Ren T, Suo JS (2002)
New J Chem 26:376
5. Wroblewska A, Wojtowicz G, Makuch E (2011) J Adv Oxid
Technol 14:267
6. Wroblewska A (2012) J Adv Oxid Technol 15:418
7. Wu C, Kong Y, Gao F, Wu Y, Lu Y, Wang J, Dong L (2008)
Micropor Mesopor Mater 113:163
8. Preethi MEL, Revathi S, Sivakumar T, Manikandan D, Divakar
D, Rupa AV, Palanichami M (2008) Catal Lett 120:56
9. Perez-Ramirez J, Christensen CH, Egeblad K, Christensen CH,
Groen JC (2008) Chem Soc Rev 37:2530
26. Kailasam K, Mu¨ller K (2008) J Chomatogr A 1191:125
27. Li LD, Shen Q, Yu JJ, Hao ZP, Xu ZP, Lu GQM (2007) Environ
Sci Technol 41:7901
28. Kumar MS, Schwidder M, Grunert W, Bruckner A (2004) J Catal
227:384
29. Danilina N, Krumeich F, van Bokhoven JA (2010) J Catal 272:37
30. Dai PSE, Petty RH, Ingram CW, Szostak R (1996) Appl Catal A
143:101
31. Choi JS, Yoon SS, Jang SH, Ahn WS (2006) Catal Today
111:280
32. Liu H, Lu G, Guo Y, Guo Y, Wang J (2008) Micropor Mesopor
Mater 108:56
33. Liu H, Lu G, Guo Y, Guo Y, Wang J (2006) Nanotechnology
17:997
10. Kim J, Bhattacharjee S, Jeong KE, Jeong SY, Choi M, Ryoo R,
Ahn WS (2010) New J Chem 34:2971
123