Heteropoly acid/carbon nanotube hybrid materials as efficient solid-acid catalysts

Article abstract of DOI:10.1002/cctc.201402270

A prototype supported solid-acid catalyst composed of a heteropoly acid (HPA) and nitrogen-functionalized carbon nanotubes (NCNT) has been developed. The chemical structure and acid strength of the HPA/NCNT hybrid have been characterized thoroughly by IR spectroscopy, X-ray photoelectron spectroscopy, TEM, and titration measurements. NCNT supports provide the HPA with an ideal hydrophobic environment, which could prevent the deactivation of the acid sites effectively in aqueous media. HPA/NCNT hybrid catalysts tend to adsorb and enrich hydrophobic substrates (ester) and release the product with high polarity (ethanol) in the aqueous phase, and this synergistic effect facilitates the catalytic process and enhances the activity of HPAs for ester hydrolysis reactions. The HPA and NCNT combine through electrostatic interactions, which ensures the relatively high stability and easy regeneration of the HPA/NCNT hybrid catalysts. The present approach provides a promising and universal strategy for the construction of hybrid catalysts based on HPA and nanocarbon with unique acidic or redox properties.

Full text of DOI:10.1002/cctc.201402270

CHEMCATCHEM
FULL PAPERS
DOI: 10.1002/cctc.201402270
Heteropoly Acid/Carbon Nanotube Hybrid Materials as
Efficient Solid-Acid Catalysts
Wei Qi, Wei Liu, Songyun Liu, Bingsen Zhang, Xianmo Gu, Xiaoling Guo, and
Dangsheng Su*[a]
A
prototype supported solid-acid catalyst composed of
the product with high polarity (ethanol) in the aqueous phase,
and this synergistic effect facilitates the catalytic process and
enhances the activity of HPAs for ester hydrolysis reactions.
The HPA and NCNT combine through electrostatic interactions,
which ensures the relatively high stability and easy regenera-
tion of the HPA/NCNT hybrid catalysts. The present approach
provides a promising and universal strategy for the construc-
tion of hybrid catalysts based on HPA and nanocarbon with
unique acidic or redox properties.
a heteropoly acid (HPA) and nitrogen-functionalized carbon
nanotubes (NCNT) has been developed. The chemical structure
and acid strength of the HPA/NCNT hybrid have been charac-
terized thoroughly by IR spectroscopy, X-ray photoelectron
spectroscopy, TEM, and titration measurements. NCNT supports
provide the HPA with an ideal hydrophobic environment,
which could prevent the deactivation of the acid sites effec-
tively in aqueous media. HPA/NCNT hybrid catalysts tend to
adsorb and enrich hydrophobic substrates (ester) and release
Introduction
The development of solid acids is a hot topic in chemical engi-
neering and materials science because of their promising ap-
plications as efficient catalysts.[1] Solid acids normally exhibit
significant advantages of heterogeneous catalysis over tradi-
tional mineral acids, which include high acidity, easy separation
from the reaction medium, reduction of corrosion, and im-
proved regenerability,[1b] that could meet the requirements of
green and sustainable chemistry. Heteropoly acids (HPAs),
a family of nano-sized transition-metal oxide clusters with pro-
tons as counter cations, exhibit a higher acid strength than
typical mineral or solid acids, such as CF3SO3H, H2SO4, g-Al2O3,
and Nafion-H resin.[2] They also have definite and easily tuned
chemical structures,[3] which allows the design of acidic materi-
als on an atomic/molecular level. HPAs are considered to be
ideal solid-acid catalysts and have already been applied in sev-
eral important industrial process, such as the polymerization of
THF.[4] However, there are still several serious drawbacks that
limit the efficient application of HPA catalysts in chemical reac-
tions, particularly in reactions in which water participates as
a reactant, product, or solvent, such as hydrolysis, hydration,
and esterification[5] because of the significant influence of H2O
on the structure and acidity of HPAs. Firstly, the formation of
the HPA secondary structure aggregates with H2O as linkers
decreases the acid strength and surface area of the clusters,[6]
which limits their activity and accessibility in the aqueous
phase. HPA clusters also have mass-transfer problems as cata-
lysts in the aqueous phase, and the compatibility between
HPAs and organic substrates are normally poor[5,7] because of
their high lattice energy as inorganic clusters. In addition, HPA
clusters exhibit remarkable solubility in water, thus they will be
liquid acids and lose their advantages of recyclability as hetero-
geneous catalysts.
Effective matrices for the immobilization of HPAs to fabricate
hybrid catalysts is one of the promising solutions for these
problems.[5,8] For example, the incorporation of HPAs into or-
ganic electrolytes or emulsion systems could improve water
tolerance and organic compatibility effectively and thus in-
crease the catalytic activity.[9] Carbon nanotubes (CNTs) exhibit
a relatively high surface area, considerable organic compatibili-
ty, and extraordinary stability (stable in aqueous media at near-
supercritical temperature) and should be a viable option as
a support for HPAs.[10] It has been reported that the combina-
tion of HPAs and CNTs with ammonium dendrimers as a bridge
yielded new anodic materials that exhibited relatively high ac-
tivity in the electrocatalytic oxidation of water.[11] Although
HPA/CNT hybrid catalysts have shown promising catalytic ap-
plications in various reactions, related research in this field is
reported rarely because the combination of an HPA and CNTs
normally requires complex modification procedures to improve
the compatibility of the two components and the stability of
the hybrid. We report here a facile method for the construction
of a prototype hybrid material that combines HPAs and CNTs
for the acidic catalysis of hydrolysis reactions. Nitrogen-func-
tionalized CNTs (NCNTs) were used in the research to provide
an enhanced binding to HPA clusters through electrostatic in-
teractions and an ideal hydrophobic environment, which could
prevent the deactivation and aggregation of HPAs and increase
[a] Dr. W. Qi, W. Liu, S. Liu, Dr. B. Zhang, X. Gu, X. Guo, Prof. D. Su
Shenyang National Laboratory for Materials Science
Institute of Metal Research, Chinese Academy of Sciences
Shenyang, 110016 (China)
E-mail: dssu@imr.ac.cn
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cctc.201402270.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2014, 6, 2613 – 2620 2613

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
the compatibility between organic substrates and inorganic
HPA catalytic centers. The obtained HPA/NCNT hybrid catalysts
exhibit a high catalytic activity and considerable stability in
ester hydrolysis reactions with H2O as the reactant and solvent.
Moreover, the present method is a universal approach applica-
ble to all kinds of HPAs, which allows the design and fabrica-
tion of HPA/NCNT hybrid catalysts with tunable acid and redox
abilities suitable for various requirements.
Results and Discussion
Fabrication of HPA/CNT hybrid catalysts
Figure 2. Zeta potential of NCNTs as a function of pH.
The concept of HPA/NCNT hybrid catalysts and their applica-
tions in hydrolysis reactions are illustrated in Scheme 1. NCNTs
are synthesized through a modified chemical vapor deposition
(CVD) procedure at 9008C that uses imidazole as the C and N
source and with Fe, Mo, Al, etc., as catalysts.[12] The obtained
ties on their surface,[13,14] and the pH of its aqueous dispersion
(2.5 mgmLꢀ1) is around 7.4.
Zeta potential measurements at different pH values provide
surface electrical information for NCNTs (Figure 2). The point of
zero charge (PZC) of NCNTs is
6.4 from the measurements,
which represents the pH if the
electrical charge density on the
NCNT surface reaches zero.
H3PW12O40 (denoted as PWA)
was chosen as a typical HPA
with a Keggin structure. The pH
of the aqueous solution of PWA
used for the fabrication of the
hybrid catalyst is 1.3 (at a con-
centration of 40 mgmLꢀ1), thus
NCNTs have a positively charged
Scheme 1. The overall procedure for the fabrication of PWA/NCNT hybrid and the catalytic hydrolysis reactions on
the hybrid surface.
surface at this pH and will com-
bine with HPA anionic clusters
spontaneously through electro-
NCNTs exhibit a BET surface area of 69.2 m2 gꢀ1 and a broad
size distribution (diameter ꢁ50ꢂ20 nm as shown in the TEM
images). The average N content is around 5.5% based on ele-
mental analysis, and the N species that exist on the NCNT sur-
face are mainly pyridinic, pyrolic, graphitic, quaternary, and ni-
trogen oxides from the deconvolution of the N1s X-ray photo-
electron spectroscopy (XPS) signal (Figure 1).[13] NCNTs are
weakly alkaline as a result of the introduction of N functionali-
static interactions after the two components are mixed togeth-
er. The resulting hybrid precipitates in H2O, and the PWA/NCNT
hybrid catalyst is obtained after thorough washing with water
to remove chemisorbed HPA. The weight loading of PWA is
around 3.8% based on elemental analysis and thermogravi-
metric analysis (TGA) results (Figure S1).
The IR spectra of PWA/NCNT (Figure 3) exhibit characteristic
WꢀO (n˜ =800–1100 cmꢀ1),[15] CꢀN (n˜ =1290 cmꢀ1), and C=C (or
C=N at n˜ =1620 cmꢀ1) vibration modes of Keggin-type PWA
clusters and NCNTs, respectively, which suggests the successful
integration of the two components in the hybrid catalysts.
PWA/NCNT hybrid catalysts and pure PWA clusters exhibit
identical PꢀOa, WꢀOd, and WꢀOcꢀW vibrations (n˜ =1079 vs.
1076, 982 vs. 978, and 910 vs. 913 cmꢀ1, respectively). However,
the WꢀObꢀW vibrations are blueshifted (n˜ =831 cmꢀ1 in PWA/
NCNT vs. n˜ =818 cmꢀ1 in pure PWA) for PWA supported on
NCNTs. This observation is consistent with reported results that
the WꢀObꢀW vibration is sensitive to the identity of HPA coun-
ter cations, and there will be a 15–20 cmꢀ1 blueshift after the
protons of PWA combine with organic amino or pyridinic
groups.[16]
XPS spectra of the PWA/NCNT hybrid exhibit a characteristic
W4f signal (Figure S2), which indicates the successful introduc-
Figure 1. N1s XPS spectra of NCNT (bottom) and PWA/NCNT (top).
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2014, 6, 2613 – 2620 2614

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
tegrity of NCNTs after the supporting procedure. High-resolu-
tion and high-angle annular dark-field scanning transmission
electron microscopy (HAADF-STEM) images of PWA/NCNT (Fig-
ure 4c and d) demonstrate that PWA is nearly monodispersed
on the NCNT surface with a diameter of 1.1–1.2 nm, which is
consistent with the size of Keggin HPAs obtained from XRD
patterns.[18] Energy-dispersive spectroscopy (EDS) elemental
mapping results (Figure 5) show that N species are nearly
Figure 3. ATR-IR spectra of PWA (c), NCNT (c), and PWA/NCNT (c).
tion of PWA clusters onto the NCNTs. The PWA/NCNT hybrid
exhibits almost the same C1s XPS signal as that of NCNT,
which indicates that the structure of NCNT support is well
maintained after the loading of HPAs. The N1s XPS signals for
NCNT and the PWA/NCNT hybrid are identical in shape. Decon-
volution of the N1s XPS spectra (Figure 1) shows that the rela-
tive content of pyridinic N species decreases by 12% and the
content of quaternary N (pyridine salt, ammonium, quaternary
graphitic N, etc.) increases by 15% after the HPAs are support-
ed, which suggests that some of the pyridine sites on NCNTs
are protonated by PWA, consistent with the IR spectroscopy,
and suggests that the two components combine through elec-
trostatic interactions.
TEM images show the characteristic bamboo-like structure
of both pure NCNT and the PWA/NCNT hybrid (Figure 4a and
b),[14,17] and there is no clear damage to the NCNT surface after
the immobilization of PWAs, which indicates the structural in-
Figure 5. TEM images of a) NCNT and b and c) PWA/NCNT and d) HAADF-
STEM image of PWA/NCNT.
monodispersed on the nanotubes, and PWAs are located
mainly in the N-rich regions, which confirms the XPS results
that pyridinic N species play a vital role to anchor PWA clus-
ters.
The acidity of the PWA/NCNT hybrid was evaluated by am-
monia temperature programmed desorption (TPD). PWA exhib-
its two distinct peaks (Figure 6), of which the one at the lower
temperature (centered at ꢁ1708C) is attributed to the desorp-
tion of NH3 combined with PWA through hydrogen bonding
(NH2ꢀH···O). The peak centered at around 5708C could be as-
signed to the desorption of chemically adsorbed ammonia on
PWA, and the splitting and broadening of the peak is a result
of the thermal decomposition of the ammonium salt of
PWA.[19] Desorption peaks centered at around 120 and 2608C
for the pure NCNT support suggests that ammonia is mainly
weakly chemisorbed.[20] Ammonia released above 4008C is the
result of the partial decomposition of NCNTs, and some N spe-
cies on NCNT are released in the form of NH3, which is consis-
tent with the TGA results (Figure S1). The ammonia TPD pro-
files of PWA/NCNT exhibit characteristic desorption peaks for
both PWA and NCNT with a similar intensity (peak area) ratio
to the two pure components, which suggests the structural in-
tegrity after immobilization. The desorption peak for PWA at
Figure 4. TEM images of a) NCNT and b and c) PWA/NCNT and d) HAADF-
STEM image of PWA/NCNT.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2014, 6, 2613 – 2620 2615

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
NCNTs are neutralized by NaOH during the titration process.
Based on the content of PWA from the TGA results and the ti-
trant (NaOH) consumption at the end point, the total number
of acid sites of PWA/NCNT is 3.3 H+/PWA, which is close to the
stoichiometric value (3.0 H+/PWA) and suggests the structural
integrity of PWA after it is supported on the NCNT surface. Two
kinds of acid sites with different strengths (pKa =4.8 and 7.4)
can be observed from the first derivative of pH as a function
of base consumption (Figure 7), which could be assigned as
free protons (H+) and protons combined with N atoms (NH+),
respectively. The acidity of the PWA/NCNT hybrid is around
ꢀ1
4.0ꢁ10ꢀ5 molH+ ghybrid based on titration, and the free protons
that could be used for catalysis account for 30% of the total
protons. This corresponds to approximately one acidic proton
per PWA cluster on average, which suggests that each PWA
cluster could combine two pyridinic N groups on NCNTs. It is
not realistic that the PWA clusters protonated two N species
on a single NCNT because of the high dispersion and small
amount of N (5.5% weight), and it is more reasonable that one
single PWA cluster may connect multiple individual NCNTs.
This hypothesis is consistent with the N2 adsorption measure-
ment results that reveal that PWA/NCNT exhibits a clear micro-
porous structure caused by the cross linking effect of PWA
clusters. As a result, PWA/NCNT, which has a network structure,
exhibits a larger BET surface area than pure NCNTs (84 vs.
69 m2 gꢀ1, an increase of over 20%; Figure S3), and the relative-
ly high surface area of PWA/NCNT is conducive for catalytic re-
actions.
Figure 6. Ammonia TPD profiles and corresponding detailed deconvolution
of NH3 desorption peaks for PWA (c/a), NCNT (c/a), and PWA/
NCNT hybrid (c/a/a).
5708C broadens after PWA is supported on NCNT, which sug-
gests that the acidity of some of the protons to PWA decreases
because of the interaction with the NCNT support. The phe-
nomenon is consistent with XPS results and suggests that PWA
may combine with NCNT through electrostatic interactions.
The acid strength of the PWA/NCNT hybrid in the aqueous
phase was determined by a titration process with NaOH as the
titrant (Figure 7). The pH of the PWA/NCNT aqueous dispersion
increases with the addition of NaOH, and the titration end
point is at pH 7.4 as indicated by the first derivative of the pH
as a function of NaOH consumption. This value is exactly the
same as the pH of the aqueous dispersion of pure NCNTs,
which suggests that all of the protons of PWA supported on
Catalysis by HPA/CNT hybrid catalysts
Three types of acid-catalyzed reactions (hydrolysis, rearrange-
ment, and alkylation) were used as probe reactions to evaluate
the catalytic activity of the PWA/NCNT hybrid catalysts
(Scheme 2). H2O is both a reactant and the solvent in ethyl
acetate (EA) hydrolysis, thus the reaction is used to evaluate
the acidic catalytic activity of PWA/NCNT in the presence of
water. Ethanol and acetic acid are the only detected products
for the reactions in the present research. On the basis of the
amounts of substrate and product in the reaction mixture sam-
Figure 7. Titration of acid sites on PWA/NCNT hybrid materials with NaOH
aqueous solution (0.01m) as titrant: pH as a function of titrant consumption
and its first derivative.
Scheme 2. Three reactions catalyzed by HPA/NCNT.
ChemCatChem 2014, 6, 2613 – 2620 2616
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
Table 1. Activity of catalysts for EA hydrolysis in water.[a]
Catalyst
Acidity
[10ꢀ5molgꢀ1
Catalytic Catalytic activity
system[b] [mmolEA molcatalystꢀ1 minꢀ1
]
]
PWA/NCNT
PWA
Cs2.5H0.5PW12O40
4.0
100
10
9.1
80
1980
39
260
47
het.
hom.
het.
het.
het.
hom.
het.
het.
het.
374
123
200
275
202
25
70
0
0
[c]
PWA/AP/SBA-15[d]
Nafion-H resin[c]
H2SO4
H-ZSM-5 zeolite[c]
HY zeolite[c]
[c]
g-Al2O3
[a] The reactions were all performed in the aqueous phase at 608C under
similar conditions. [b] het.: Heterogeneous; hom.: homogeneous.
[c] Ref. [21]. [d] PWA supported on SBA-15 from Ref. [5].
because of the deactivation of acid sites by H2O.[21] Nafion-H
resin, Cs2.5H0.5PW12O40, and PWA/AP/SBA-15 exhibit relatively
high activity under similar reaction conditions as they are all
believed to have hydrophobic surfaces that could protect
active sites from H2O deactivation.[5,19] Compared with bare
PWA clusters, significant acid catalytic activity increases could
be observed after hydrophobic modifications by Cs or organic
compounds (123 vs. 200 or 275 mmolEA molcatalystꢀ1 minꢀ1). The
proposed PWA/NCNT hybrid exhibited a catalytic activity of
374 mmolEA molcatalystꢀ1 minꢀ1, which is over 15 times higher
than that of H2SO4 and three times higher than that of bare
PWAs, even though the latter two are homogeneous catalysts.
The reason for the high activity of the PWA/NCNT hybrid
catalyst may be attributed to the high surface area of the
hybrid and the uniform dispersion of PWA clusters on the
NCNT surface. It has been reported that nano-sized HPA clus-
ters could form vesicle-like superstructure aggregates in vari-
ous solvents such as water.[22] This would definitely decrease
the surface area of HPAs, which is very small originally (the sur-
face area of PWA is smaller than 5 m2 gꢀ1)[23], and restrict their
catalytic activity. In our case, PWAs are nearly monodispersed
in the hybrid and have a large surface area (ꢁ84 m2 gꢀ1),
which increases the accessibility and, therefore, the catalytic
activity of the PWA clusters. Another important reason for the
high catalytic activity of PWA/NCNT hybrid catalysts is the hy-
drophobic surface of the CNTs.[24] It is known that most inor-
ganic solid-acid catalysts deactivate quickly in aqueous solu-
tion. However, CNTs provide PWA with an ideal hydrophobic
environment in the PWA/NCNT hybrid catalyst, which is favora-
ble for acid-catalyzed reactions and also enhances the organic
compatibility of HPA clusters. The NCNT support exhibits
a high adsorption ability for EA, which is more than four times
higher than that of ethanol under the same conditions
(Figure 9). The results indicate that NCNT supports are more
able to capture substrates with low polarity (EA) and release
the strongly polar products (ethanol) to the aqueous solution.
The enzyme-like supramolecular interactions between the
PWA/NCNT catalyst, reactants, and products facilitate the mass-
transfer process and play a crucial role to promote the activity
of PWA.
Figure 8. Kinetic profiles: a) TON and b) ꢀln(Ct-EA/C0-EA) of EA hydrolysis reac-
~
&
tions catalyzed by PWA/NCNT ( ), PWA ( ), and H2SO4 (N), respectively.
pled at a certain specified time (t), plots of turnover number
(TON), which is defined as the molar ratio of the converted EA
(NEA-convert) to acid sites (NH+) [Eq. (1)], versus reaction time were
constructed. The kinetics of the hydrolysis reactions catalyzed
by the PWA/NCNT hybrid (heterogeneous), bare PWA (homo-
geneous), and H2SO4 (homogeneous) are depicted in Figure 8,
in which C0-EA and Ct-EA represent the initial EA concentration
and the concentration at time t, respectively. A linear relation-
ship (Figure 8b) between the value of ln(Ct-EA/C0-EA) and reac-
tion time indicates that the reaction follows pseudo-first-order
kinetics, and the rate constants (k) can be calculated as ex-
pressed in Equation (2).
NEAꢀconvert
ð1Þ
ð2Þ
TON ¼
NHþ
ꢀ
ꢁ
CtꢀEA
C0ꢀEA
ln
¼ ꢀkt
The performances of various acid catalysts in aqueous-phase
EA catalytic hydrolysis reactions, which include the PWA/NCNT
hybrid, are listed in Table 1. All the catalytic activity tests were
performed at 608C under similar reaction conditions.[21] As
shown by the results in the table, classic inorganic liquid
(H2SO4) or solid (H-ZSM-5, HY, and g-Al2O3) acids exhibit zero or
limited activity in aqueous-phase catalytic hydrolysis reactions
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2014, 6, 2613 – 2620 2617

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
action system, toluene was used as both the reactant and sol-
vent, and the reactions between toluene and 1-octene give
the 2-phenyl isomer as the main product under the catalysis of
PWA (Scheme 2; selectivity for the 2-phenyl isomer >99%).[26]
Supported PWA catalysts (PWA/NCNT and PWA/MCM-41) ex-
hibited a better activity than bare PWA clusters as PWA is an
insoluble bulk material that has a much lower surface area
than supported PWA catalysts in toluene, which has a low po-
larity (Table 2). Unlike in a polar solvent (water or phenyl ni-
trile), PWA/NCNT exhibits a lower acidic catalytic activity than
PWA/MCM-41 (2470 vs. 5732 mmoltoluene molcatalystꢀ1 minꢀ1),
which suggests that the hydrophobic protection brought by
the CNT support is not a decisive factor, and the acidic catalytic
activity of PWA may only depend on its dispersion state and
the surface area of the support in low-polarity reaction sys-
tems.[26] This proves the importance of hydrophobic protection
to PWA especially in highly polar reaction systems and sug-
gests that polar reaction media are more conducive to the
unique high catalytic activity of the proposed PWA/NCNT
hybrid catalyst.
~
&
Figure 9. Adsorption amount of EA ( ) and ethanol ( ) on NCNT substrates
as a function of time.
The acidic catalytic activity of the PWA/NCNT hybrid was
also evaluated in two other typical acid-catalyzed reactions
(Scheme 2), and the performances of various PWA-based cata-
lysts are summarized in Table 2. The liquid-phase Beckmann re-
The PWA/NCNT hybrid catalysts exhibit a relatively high sta-
bility and are easily recovered as heterogeneous catalysts. The
structure of the recovered PWA/NCNT hybrid catalyst is well
maintained according to XPS and TGA, and they do not show
any significant deactivation after the fourth use (Figure 10).
Table 2. Summary of the catalytic activity of PWA-based catalysts for re-
arrangement and alkylation reactions.
Reaction
Catalyst
Catalytic activity[a]
rearrangement of
PWA
50
1880
190
160
235
cyclohexanone oxime[b]
PWA/NCNT
Cs2.5H0.5PW12O40
H-ZSM-5 zeolite[c]
PWA[e]
[c]
alkylation of toluene[d]
PWA/NCNT
PWA/MCM-41[e]
2470
5732
[a] In [mmolsubstrate molcatalystꢀ1 minꢀ1]. [b] The listed reactions were all per-
formed in phenylnitrile at 1308C for 6 h under similar conditions. [c] From
Ref. [25]. [d] The reactions were all performed at 808C under similar con-
ditions. [e] From Ref. [26].
arrangement of cyclohexanone oxime is an important catalytic
process because caprolactam, the main product of the reac-
tion, is a precursor of Nylon 66, which is a very important in-
dustrial product. The Cs salt of PWA exhibits a higher activity
Figure 10. Comparison of the catalytic activity (TON) for NCNT-supported
and unsupported HPAs.
than bare PWA (190 vs. 50 mmolcyclohexanone oxime molcatalystꢀ1 minꢀ1
)
because of its unique hydrophobic surface[23,25] and its catalytic
activity is even better than that of H-ZSM-5 (190 vs.
160 mmolcyclohexanone oxime molcatalystꢀ1 minꢀ1), which is a common
solid-acid catalyst in the chemical industry (Table 2).
The turnover frequency (TOF) of the PWA/NCNT hybrid
The content of PWA in the hybrid catalysts could be adjusted
to some extent by tuning the feed ratio of the two compo-
nents in the fabrication process. The loading content of PWA
in the hybrid catalysts reaches a plateau after a weight content
above 3.5% (Figure S4), which suggests that the loading of
PWA is limited by the surface concentration of N functional
groups because PWA clusters are anchored mainly on the
NCNT surface through electrostatic interactions with N species
as discussed above. Finally, two other typical Keggin HPAs with
different central or addenda atoms, H4SiW12O40 (SiWA) and
H3PMo12O40 (PMoA), were also used to verify that the present
method is a universal strategy for the fabrication of hybrid
solid acids. TGA and XPS confirmed the structural integrity of
SiWA, PMoA, and the NCNT supports, and the hydrolysis reac-
catalyst
in
rearrangement
reactions
reaches
1880 mmolcyclohexanone oxime molcatalystꢀ1 minꢀ1, which is nearly
10 times higher than that of Cs2.5H0.5PW12O40 because of the
large surface area and hydrophobic protection to the PWA cat-
alytic centers brought by CNT substrates as in the hydrolysis
reactions discussed above.
The Friedel–Crafts alkylation of aromatic compounds is one
of the most important processes for the synthesis of alkyl aro-
matic compounds on an industrial scale.[26] In our alkylation re-
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2014, 6, 2613 – 2620 2618

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
NCNT at different loadings followed a similar strategy with only the
identity and feed ratio of HPA and NCNTs changed.
tions showed that such catalysts performed as well as PWA/
NCNT (Figure 10), which suggests that the present strategy
offers a general approach for the fabrication of hybrid solid-
acid catalysts that contain common HPAs suitable for various
requirements.
Catalytic activity measurements
Typically, the hydrolysis reaction of EA under the catalysis of PWA/
NCNT was performed as follows. EA (7.5 g) and PWA/NCNT catalyst
(800 mg) were dispersed in H2O (140 mL), and the reaction mixture
was stirred at 300 rpm at 608C. Aliquots (1 mL) of the reaction mix-
ture were withdrawn and filtered at certain times. The filtrates
were collected carefully and their compositions were determined
by GC. GC analysis was performed by using an Agilent 7890C
system with an HP-5 column and a flame ionization detector (FID).
The catalysts were recovered and regenerated by simple filtration,
washing with H2O, and subsequent drying at 1508C for 8 h.
Conclusions
We have achieved a new efficient and water-tolerant solid-acid
catalyst composed of heteropoly acids (HPAs) and N-function-
alized carbon nanotubes (NCNTs), which combine through
electrostatic interactions. NCNTs provide an ideal hydrophobic
environment for the HPA catalyst, which prevent its deactiva-
tion in aqueous media. The HPA/NCNT hybrid catalysts tend to
adsorb and enrich the hydrophobic substrate (ester in hydroly-
sis reactions) in water and release the highly polar product
(ethanol in hydrolysis reactions) to the solution. The synergistic
effect facilitates the catalytic process, and enhances the activity
of HPA for ester hydrolysis reactions. The HPA/NCNT hybrid
catalysts also exhibit an acceptable performance in rearrange-
ment and alkylation reactions. From a comparison with report-
ed catalytic systems based on HPA clusters, we found that the
protection effect brought by the CNT support is clearer in reac-
tions systems with a high polarity, that is, a polar reaction
medium favors the unique high catalytic activity of the HPA/
NCNT hybrid catalyst. The HPA/NCNT hybrid exhibits not only
better activity but also higher stability and easier recovery
than homogeneous catalytic systems. The present study indi-
cates a promising way to fabricate functional materials com-
posed of HPA and CNTs, the acidity and hydrophobicity of
which can be tuned easily. The synergistic effect of the HPA/
NCNT catalyst shown in the hydrolysis reaction also suggests
the importance of the design and control of the catalytic sys-
tems.
Typically, the Beckmann rearrangement reaction of cyclohexanone
oxime under the catalysis of PWA/NCNT was performed as follows.
Cyclohexanone oxime (115 mg) and PWA/NCNT catalyst (50 mg)
were dispersed in phenylnitrile (10 mL). DMF (50 mL) was added as
a GC internal standard. The reaction mixture was stirred at
300 rpm at 1308C. The reaction mixture was cooled to RT, and the
PWA/NCNT catalysts were collected by filtration after 6 h. The fil-
trates were collected carefully and their compositions were deter-
mined by GC. The rearrangement reaction of cyclohexanone oxime
catalyzed by PWA was performed in a similar way using the same
amount (weight) of PWA instead of PWA/NCNT as the catalyst.
Typically, the alkylation of toluene under the catalysis of PWA/
NCNT was performed as follows. 1-Octene (1.5 mL) and PWA/NCNT
catalyst (50 mg) were mixed and dispersed in toluene (8.4 mL).
n-Decane (500 mL) was added as a GC internal standard. The reac-
tion mixture was stirred at 300 rpm at 1208C. The reaction mixture
was cooled to RT, and the PWA/NCNT catalysts were collected by
filtration after 6 h. The filtrates were collected carefully and their
compositions were determined by GC. The alkylation reaction of
toluene catalyzed by PWA was performed in a similar way using
same amount (weight) of PWA instead of PWA/NCNT as the cata-
lyst.
Adsorption measurements
Experimental Section
Typically, the adsorption ability of the NCNT support for the reac-
tant (EA) and product (ethanol) was measured under conditions
similar to those of the hydrolysis reactions, and the detailed experi-
mental procedure is summarized as follows. NCNT (1.0 g) and EA
or ethanol (0.1 g) were dispersed in H2O (50 mL), and the mixture
was stirred at 300 rpm at 608C. Aliquots (0.5 mL) of the reaction
mixture were withdrawn and filtered at certain times. The filtrates
were collected carefully and the compositions of the mixtures
were determined by GC.
Fabrication of HPA/NCNT hybrid catalysts
NCNTs were synthesized through a modified CVD process. Catalysts
(100 mg) composed of
a mixture of Fe, Al, and Mo alloy
nanoparticles were heated to 9008C in a tube furnace. Imidazole
(5 g) was heated to 2508C in a separate tube furnace connected to
that with the catalysts. NH3 (10%) in Ar was subsequently intro-
duced to the linked furnaces from the imidazole side at a flow rate
of 100 cm3 minꢀ1. The two furnaces were cooled to RT after 15 min,
and the obtained raw products were dispersed in concentrated
HCL (50 mLHCl gproductꢀ1) under vigorous stirring for 6 h at RT. The
NCNT products were collected by filtration, washed with H2O until
the pH of the filtrate reached 7, and dried at 1508C.
Characterization
TEM measurements were performed by using a FEI Tecnai F20 mi-
croscope with an accelerating voltage of 200 kV. C1s, N1s, W4f,
and Mo4f XPS spectra were obtained by using a surface analysis
system (ESCALAB 250, Thermo VG, USA) with AlKa X-rays
(1486.6 eV, 150 W, 50.0 eV pass energy). The N1s XPS spectra were
fitted by using mixed Gaussian–Lorentzian component profiles (at
a ratio of 80:20) after the subtraction of a Shirley background by
using XPSPEAK41 software. The fitting was performed by fixing the
peak position for individual species within 0.05 eV and applying
Commercial H3PW12O40 (PWA), H4SiW12O40 (SiWA), and H3PMo12O40
(PMoA) were used directly without any pretreatment. In a typical
immobilization process, an aqueous solution of PWA (0.4–
40 mgmLꢀ1, 25 mL) was added dropwise to NCNT aqueous disper-
sion (25 mL, 4 mgmLꢀ1). The precipitate was collected by filtration
after 4 h of vigorous stirring, and the obtained PWA/NCNT hybrid
was washed with over 2 L of H2O until the pH of the filtrate
reached 7. The fabrication of SiWA/NCNT, PMoA/NCNT, and PWA/
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2014, 6, 2613 – 2620 2619

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
[8] a) R. Gall, C. Hill, J. Walker, J. Catal. 1996, 159, 473–478; b) R. Gall, C.
Hill, J. Walker, Chem. Mater. 1996, 8, 2523–2527; c) D. Zhou, B. Han, Adv.
Funct. Mater. 2010, 20, 2717–2722; d) S. Azoz, J. Jiang, G. Keskar, C.
McEnally, A. Alkas, F. Ren, N. Marinkovic, G. Haller, S. Ismail-Beigib, L.
Pfefferle, Nanoscale 2013, 5, 6893–6900.
[9] a) A. Haimov, R. Neumann, J. Am. Chem. Soc. 2006, 128, 15697–115700;
b) C. Li, Z. Jiang, J. Gao, Y. Yang, S. Wang, F. Tian, F. Sun, X. Sun, P. Ying,
C. Han, Chem. Eur. J. 2004, 10, 2277–2280.
[10] a) N. Kawasaki, H. Wang, R. Nakanishi, S. Hamanaka, R. Kitaura, H. Shino-
hara, T. Yokoyama, H. Yoshikawa, K. Awaga, Angew. Chem. Int. Ed. 2011,
50, 3471–3474; Angew. Chem. 2011, 123, 3533–3536; b) G. Charron, A.
Giusti, S. Mazerat, P. Mialane, A. Gloter, F. Miserque, B. Keita, L. Nadjo, A.
Filoramo, E. Riviere, W. Wernsdorfer, V. Huc, J.-P. Bourgoin, T. Mallah,
Nanoscale 2010, 2, 139–144; c) D. Su, S. Perathoner, G. Centi, Chem.
Rev. 2013, 113, 5782–5816.
[11] F. Toma, A. Sartorel, M. Lurlo, M. Carraro, P. Parisse, C. Maccato, S.
Rapino, B. Gonzalez, H. Amenitsch, T. Daros, L. Casalis, A. Goldoni, M.
Marcaccio, G. Scorrano, G. Scoles, F. Paolucci, M. Prato, M. Bonchio, Nat.
Chem. 2010, 2, 826–831.
a full-width at half-maximum (FWHM) of 1.2–1.5 eV. N2 physisorp-
tion was measured by using a Micrometrics ASAP 2020 instrument.
Attenuated total reflectance (ATR) FTIR spectra were recorded by
using a Varian spectrometer equipped with a liquid-nitrogen-
cooled mercury cadmium telluride (MCT) detector. Spectra were re-
corded at 4 cmꢀ1 resolution in the region of 4000–550 cmꢀ1. Am-
monia TPD patterns were recorded by using a mass spectrometer
connected to a microreactor. Samples (100 mg; HPA, NCNT, or
HPA/NCNT) were placed in a glass reactor and dried at 1508C for
60 min in a He flow. The adsorption of ammonia (10% NH3 bal-
anced with He, 100 mLminꢀ1) was performed for 35 min at 358C.
Then the physically adsorbed ammonia was desorbed by purging
with He (50 mLminꢀ1) for 500 min at 358C. Subsequently, the TPD
profile was recorded upon heating the sample at a rate of
58Cminꢀ1 up to 7008C. HPA/NCNT hybrid (800 mg) was dispersed
in NaCl aqueous solution (140 mL, 0.1m) in a typical titration pro-
cess. The titration of the acid strength was performed by using
a Mettler DL77 autotitrator with NaOH (0.01m) as the titrant.
[12] A brief description of the synthesis of NCNTs can be found in the Sup-
porting Information.
[13] R. Arrigo, M. Havecker, S. Wrabetz, R. Blume, M. Lerch, J. McGregor, E.
Parrott, J. Zeitler, L. Gladden, A. Knop-Gericke, R. Schlçgl, D. Su, J. Am.
Chem. Soc. 2010, 132, 9616–9630.
Acknowledgements
[14] a) S. van Dommele, K. de Jong, J. Bitter, Chem. Commun. 2006, 4859–
4861; b) A. Villa, J. Tessonnier, O. Majoulet, D. Su, R. Schlçgl, Chem.
Commun. 2009, 4405–4407; c) A. Villa, J. Tessonnier, O. Majoulet, D. Su,
R. Schlçgl, ChemSusChem 2010, 3, 241–245.
[15] W. Qi, H. Li, L. Wu, Adv. Mater. 2007, 19, 1983–1987.
[16] S. Himeno, M. Takamoto, T. Ueda, J. Electroanal. Chem. 1999, 465, 129–
135.
The authors acknowledge the financial support from MOST
(2011CBA00504), NSFC of China (21303226, 21133010, 51221264,
21261160487), “Strategic Priority Research Program” of the Chi-
nese Academy of Sciences, Grant No. XDA09030103, and a Gener-
al Financial Grant from the China Postdoctoral Science Founda-
tion (2012M520651). The authors thank Prof. Dr. Robert Schlçgl
from FHI (Berlin) for fruitful discussions.
[17] C. Chen, J. Zhang, B. Zhang, C. Yu, F. Peng, D. Su, Chem. Commun.
2013, 49, 8151–8153.
[18] G. Brown, M.-R. Noe-Spirlet, W. R. Busing, H. A. Levy, Acta Crystallogr.
Sect. B 1977, 33, 1038–1046.
Keywords: electrostatic
catalysis · nanotubes · supported catalysts · supramolecular
chemistry
interactions
·
heterogeneous
[19] N. Essayem, R. Frety, G. Coudurier, J. C. Vedrine, J. Chem. Soc. Faraday
Trans. 1997, 93, 3243–3248.
[20] W. Xia, Y. Wang, R. Bergstraßer, S. Kundu, M. Muhler, Appl. Surf. Sci.
2007, 254, 247–250.
[21] M. Kimura, T. Nakato, T. Okuhara, Appl. Catal. A 1997, 165, 227–240.
[22] a) H. Li, H. Sun, W. Qi, M. Xu, L. Wu, Angew. Chem. Int. Ed. 2007, 46,
1300–1303; Angew. Chem. 2007, 119, 1322–1325; b) A. Nisar, J.
Zhuang, X. Wang, Adv. Mater. 2011, 23, 1130–1135; c) P. Yin, D. Li, T. Liu,
Chem. Soc. Rev. 2012, 41, 7368–7383.
[23] T. Okuhara, H. Watanabe, T. Nishimura, K. Inumaru, M. Misono, Chem.
Mater. 2000, 12, 2230–2238.
[24] F. Liu, W. Kong, C. Qi, L. Zhu, F. Xiao, ACS Catal. 2012, 2, 565–572.
[25] N. R. Shiju, H. M. Williams, D. R. Brown, Appl. Catal. B 2009, 90, 451–
457.
[1] a) J. Clark, Acc. Chem. Res. 2002, 35, 791–797; b) K. Tanabe, W. F. Hoel-
derich, Appl. Catal. A 1999, 181, 399–434; c) F. Liu, L. Wang, Q. Sun, L.
Zhu, X. Meng, F. Xiao, J. Am. Chem. Soc. 2012, 134, 16948–16950.
[2] T. Okuhara, N. Mizuno, M. Misono, Appl. Catal. A 2001, 222, 63–77.
[3] a) W. Qi, L. Wu, Polym. Int. 2009, 58, 1217–1225; b) J. Macht, R. Carr, E.
Iglesia, J. Am. Chem. Soc. 2009, 131, 6554–6565.
[4] a) I. Kozhevnikov, J. Mol. Catal. A 2007, 262, 86–92; b) I. Kozhevnikov,
Catalysts for Fine Chemicals, Vol. 2: Catalysis by Polyoxometalates, Wiley,
Chichester, 2002.
[5] K. Inumaru, T. Ishihara, Y. Kamiya, T. Okuhara, S. Yamanaka, Angew.
Chem. Int. Ed. 2007, 46, 7625–7628; Angew. Chem. 2007, 119, 7769–
7772.
[26] Y. Liu, L. Xu, B. Xu, Z. Li, L. Jia, W. Guo, J. Mol. Catal. A 2009, 297, 86–92.
[6] A. Micek-Ilnicka, J. Mol. Catal. A 2009, 308, 1–14.
Received: April 24, 2014
[7] W. Qi, Y. Wang, W. Li, L. Wu, Chem. Eur. J. 2010, 16, 1068–1078.
Published online on July 30, 2014
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2014, 6, 2613 – 2620 2620