Site-specific catalytic activities to facilitate solvent-free aerobic oxidation of cyclohexylamine to cyclohexanone oxime over highly efficient Nb-modified SBA-15 catalysts

Article abstract of DOI:10.1039/d0cy00479k

The development of highly active and selective heterogeneous catalysts for efficient oxidation of cyclohexylamine to cyclohexanone oxime is a challenge associated with the highly sensitive nitrogen center of cyclohexylamine. In this work, dispersed Nb oxide supported on SBA-15 catalysts are disclosed to efficiently catalyze the selective oxidation of cyclohexylamine with high conversion (>75%) and selectivity (>84%) to cyclohexanone oxime by O₂without any addition of solvent (TOF = 469.8 h^?1, based on the molar amount of Nb sites). The role of the active-site structure identity in dictating the site-specific catalytic activities is probed with the help of different reaction and control conditions and multiple spectroscopy methods. Complementary to the experimental results, further poisoning tests (with KSCN or dehydroxylation reagents) and DFT computational studies clearly unveil that the surface exposed active centers toward activation of the reactants are quite different: the surface -OH groups can catch the NH₂group from cyclohexylamine by forming a hydrogen bond and lead to a more facile cyclohexylamine oxidation to desired products, while the monomeric or oligomeric Nb sites with a highly distorted structure play a key role in the dissociation of O₂molecules beneficial for insertion of active oxygen species into cyclohexylamine. These catalysts exhibit not only satisfactory recyclability for cyclohexylamine oxidation but also efficiently catalyze the aerobic oxidation of a wide range of amines under solvent-free conditions.

Full text of DOI:10.1039/d0cy00479k

View Article Online
View Journal
Catalysis
Science &
Technology
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: W. ding, H. Peng,
W. Zhong, L. Mao and Y. Dulin, Catal. Sci. Technol., 2020, DOI: 10.1039/D0CY00479K.
Volume
Number
21 January 2017
Pages 313-534
7
2
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Catalysis
Science &
Technology
rsc.li/catalysis
Accepted Manuscripts are published online shortly after acceptance,
before technical editing, formatting and proof reading. Using this free
service, authors can make their results available to the community, in
citable form, before we publish the edited article. We will replace this
Accepted Manuscript with the edited and formatted Advance Article as
soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes to the
text and/or graphics, which may alter content. The journal’s standard
Terms & Conditions and the Ethical guidelines still apply. In no event
shall the Royal Society of Chemistry be held responsible for any errors
or omissions in this Accepted Manuscript or any consequences arising
from the use of any information it contains.
ISSN 2044-4761
PAPER
Chaoqiu Chen, Yong Qin et al.
Highly dispersed Pt nanoparticles supported on carbon
nanotubes produced by atomic layer deposition for hydrogen
generation from hydrolysis of ammonia borane
rsc.li/catalysis

Please do not adjust margins
Catalysis Science & Technology
Page 1 of 14
View Article Online
DOI: 10.1039/D0CY00479K
ARTICLE
Site-specific catalytic activities to facilitate solvent-free aerobic
oxidation of cyclohexylamine to cyclohexanone oxime over highly
efficient Nb-modified SBA-15 catalysts
Received 00th January 20xx,
Accepted 00th January 20xx
Wei Ding, Haoyu Peng, Wenzhou Zhong*, Liqiu Mao*, Dulin Yin
DOI: 10.1039/x0xx00000x
The development of highly active and selective heterogeneous catalysts for efficient oxidation of cyclohexylamine to
cyclohexanone oxime is a challenge associated with the highly sensitive nitrogen-center of cyclohexylamine. In this work,
dispersed Nb oxide supported on SBA-15 catalysts are disclosed to efficiently catalyze selective oxidation of
cyclohexylamine with high conversion (>75%) and selectivity (>84%) to cyclohexanone oxime by O2 without any addition of
solvent (TOF = 469.8 h–1, based on the molar amount of Nb sites). The role of active-site structure identity is probed in
dictating the site-specific catalytic activities with the help of different reaction and control conditions and multiple
spectroscopic methods. Complementary to the experiments results, further poisoning tests (with KSCN or dehydroxylation
reagents) and DFT computational studies clearly unveil that the surface exposed active centers toward activation of
reactants are quite different: the surface –OH groups can catch NH2 group from cyclohexylamine by forming a hydrogen
bond and lead to a more facile cyclohexylamine oxidation to desired products, while the monomeric or oligomeric Nb sites
with highly distorted structure play an key role in the dissociation of O2 molecules beneficial for insertion of active oxygen
species into cyclohexylamine. These catalysts exhibit not only satisfactory recyclability for cyclohexylamine oxidation, but
also efficiently catalyze the aerobic oxidation of a wide range of amine under solvent-free conditions.
interesting method has also been proposed that the transformation
involves the hydrogenation of nitrobenzene or nitrocyclohexane
with all the atoms (100%) needed to form the cyclohexanone
Introduction
Cyclohexanone oxime is one of the most important intermediates in
chemical industry, particularly as a key precursor of ε-caprolactam
for the polyester Nylon-6 production.1,2 In 2012, the worldwide
cyclohexanone oxime production capacity was approximately 560
million tons/year, and the demand is projected to grow at an
annual rate of 2~3% in the next decade.3 Currently, most methods
of cyclohexanone oxime production involve the condensation of
cyclohexanone with hydroxylamine sulfates as nitrogen source (Fig.
1). This conventional technology route, however, is via a multi-step
process with low atom-efficiency (63%) and coproduces large
volumes of salt waste.4,5 So far, several new methods for the
preparation of cyclohexanone oxime from cyclohexanone have
been developed to minimize waste. Liquid-phase cyclohexanone
ammoximation to cyclohexanone oxime with ammonia and H2O2
over TS-1 has achieved industrialization.6,7 An attempt with the
cheap O2 as oxidant for ammoximation of cyclohexanone with
ammonia has been examined using CoIICoIIIAlPO-36 catalysts.8
However, the biggest challenge of these methods is the difficulty in
obtaining cyclohexanone with 80% selectivity, which is usually
supplied by aerobic cyclohexane oxidation with a low conversion of
oxime structure.10,11 Alternatively, the use of cyclohexylamine as
the feedstock to cyclohexanone oxime is particularly attractive
route by compared with the conventional methods, as its significant
advantages include easy availability and the elimination of the
inorganic by-products. Cyclohexylamine can be widely obtained
from benzene via nitrobenzene and aniline, nitrocyclohexane,
cyclohexene or cyclohexanol. Particularly, this successful
implementation in atomic-economy will provide a new green
Fig.
1 Existing commercial hydroxylamine-based route (i) and alternative
cyclohexylamine-based route (ii) for cyclohexanone oxime production.
less than 5% by preventing the deep-oxidation.9 Recently, a dimension for the development of new cyclohexanone oxime
preparation route with almost 100% theoretical atom from benzene
via hydrogenation, hydrolysis, amination with NH3 as nitrogen
source and cyclohexylamine oxidation.
a. National & Local United Engineering Laboratory for New Petrochemical Materials
&Fine Utilization of Resources, Hunan Normal University, Changsha 410081, P. R.
China. E-mail: zwenz79@163.com, mlq1010@126.com; Tel: +86-731 88872576;
Fax: +86-731-88872531
Please do not adjust margins

Please do not adjust margins
Catalysis Science & Technology
Page 2 of 14
ARTICLE
Journal Name
Although oxyfunctionalization of cyclohexylamine with industrial
In recent years, niobium-based materials have VbieeweAnrticqleuOicnlkinlye
potential is of great importance throughout the chemical value- emerging as the effective catalysts in the DfiOeIl:d10o.1f0h39e/tDe0roCgYe0n0e4o79uKs
chain for the production of cyclohexanone oxime, they remain catalysis. Owing to its uniform acidity and water tolerant properties,
problematic transformations to achieve, both with respect to the wide research activities are focused on acid-catalyzed reactions
selective formation of desired cyclohexanone oxime and the nature such as hydrolysis23,24 and dehydration25,26. In addition, niobium
of the oxidizing species. This is just one example of how selective materials also show the ability to release oxygen in active peroxo-
cyclohexylamine oxidation can offer interesting and attractive new niobium, either by treatment of H2O2 or on photo-irradiation, and
opportunities. At present, various methods for the transformation act in oxidation of various substrates such as SO227, alkenes28, 5-
of cyclohexylamine to cyclohexanone oxime are limited owing to hydroxymethylfurfural29, thioethers27 and alcohols30. Among these
the highly sensitive nitrogen-center of -NH2 groups. Several studies, however, the work on selective oxidation of amine to the
oxidation procedures using stoichiometric oxidants such as CrVI corresponding oxime with O2 as a cheaper oxidant over niobium-
salts, Dess-Martin, or tert-butyl hydroperoxide have been derived heterogeneous catalysts is still very scarce, possibly
reported.12,13 These processes, however, suffer from the formation because Nb catalysts seem to afford a lower catalytic activity.
of an equivalent amount of organic or toxic waste, which leads to Herein, we report the aerobic selective oxidation of
low atom economy and environmental issues. Given an cyclohexylamine to cyclohexanone oxime using niobium-based
environmentally benign reduction byproduct (H2O), catalytic heterogeneous catalysts in a solvent-free system. By changing the
oxidation of cyclohexyl-amine with H2O2 in the presence of TS-1 or preparation method to adjust the structure of niobium species and
heteropolyacid, gives cyclohexanone oxime, nitrocyclohexane and surface hydroxyl groups on the catalysts, the efficiency of
cyclohexanone.14 Nevertheless, its cost is still prohibitive for less cyclohexylamine oxidation to cyclohexanone oxime was optimized.
value-added cases.12,15 As a result, a catalytic system using O2 has The catalyst with Nb species in extra-framework portion prepared
been desired in view of environmental and economical aspects. by impregnation showed a 75.2% cyclohexylamine conversion at
Silica, γ-Al2O3, WO3/Al2O3 or heteropolyacid are known to be 373 K, with a cyclohexanone oxime selectivity of 84.3%. Our Nb
capable of catalyzing vapor-phase aerobic oxidation of catalysts for turnover frequency (TOF=469.8 h–1) outperform all W
cyclohexylamine at higher temperature (>443 K) to give the or Ti-based catalysts previously employed. The experimental and
corresponding oxime in low yields.16-19 Recently, Ti-based catalysts computational results show that selective site-specific adsorption-
are demonstrated to be efficient for aerobic oxidation of activation of Nb species and surface –OH groups takes place in the
cyclohexylamine to oxime.20,21 The reported method with different reactants to promote the selective oxidation of
employing an organic 1, 1-diphenyl-2-picrylhydrazyl catalyst and cyclohexylamine.
WO3/Al2O3 cocatalyst in CH3CN solvent is found to give oxime in
high yields.22 With regard to the synthesis of cyclohexanone oxime
by aerobic cyclohexylamine oxidation, potentially practical and
Results and discussion
sustainable processes need to employ heterogeneous catalysis,
especially in a solvent-free manner; however, this is extremely
challenging to simultaneously achieve high activities and
selectivities in such transformations.
Characterization of the catalysts
ICP-AES analysis was used to determine the quantitative
Table 1 Textural property of the different samples.
Catalyst
SBETa (m2/g)
Sext (m2/g)
Pore sizeb (nm)
Pore volume (cm3/g)
Nb, wt% as assumed
Nb, wt% ICP
SBA-15
668
661
576
548
527
723
648
634
557
518
509
690
5.87
5.79
5.41
5.33
5.16
7.67
1.02
0.98
0.86
0.78
0.76
1.30
-
-
Nb/SBA-15/1im
Nb/SBA-15/3im
Nb/SBA-15/5im
Nb/SBA-15/7im
Nb-SBA-15/3co
1.0
3.0
5.0
7.0
3.0
0.9
2.8
4.9
6.8
2.7
a BET surface area. b The pore-size distribution determined by the BJH method.
amount of Nb metals present in the samples and the results
are shown in Table 1. The data clearly show that the amount
of metals employed during the catalyst preparation are close
to the expected values. N2 adsorption/desorption isotherms
characterise the difference between the prepared catalysts
(Table 1 and Fig. S1). A wider hysteresis loop (characteristic of
large-pore mesoporous solids31) is shifted to low relative
pressure in the range of 0.54 and 0.80 with increasing Nb-
loading amounts, suggesting a decrease in the pore size for
the Nb/SBA-15/im samples. However, this effect is not
observed for Nb-SBA-15/3co prepared by co-condensation,
which shows a hysteresis loop very similar to that of pristine
SBA-15. Among the samples investigated, Nb-SBA-15/3co
shows the highest specific surface area (SBET, 723 m2/g) with
1.30 cm3/g total pore volume, whereas Nb/SBA-15/im
samples show a lower SBET (527 ~ 661 m2/g) and total pore
volume (0.76 ~ 0.98 cm3/g) compared to the parent SBA-15.
This behavior is in agreement with the fact that niobium
species can be incorporated into the framework of SBA-15
with promoting the Si−O−Nb linkages for Nb-SBA-15/3co; a
2 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 3 of 14
Catalysis Science & Technology
Please do not adjust margins
Journal Name
ARTICLE
regularly decrease in SBET and total pore volumes for the
Nb/SBA-15/im samples is attributed to the increase in the
sample density after impregnation with the higher molecular
weight Nb species.
Compared with the parent SBA-15 support, two less intense
peaks between 1.64° and 1.85° in Low-angle XRD patterns provide
evidence that the mesoporous structure remains intact after
framework or highly dispersed on surfaces of mesoporous silica.
View Article Online
Further, uniform pores can also be obseDrvOeId: 10fr.1o0m39/tDh0eCNY0b0-S4B79AK-
15/3co sample, and their pore size is the larger (d=7.74 nm, 3B2)
compared to Nb/SBA-15/3im (5.46 nm, 3A2), which is well agree
with the average pore diameter from N2 physisorption. This result
reveals that the incorporation of more niobium oxide species into
the framework for Nb-SBA-15/3co might be caused by co-
condensation procedure. Based on the substrate diameters
(<1.1 nm) by DFT calculation, the reaction is not controlled or
limited by pore diffusion. Energy dispersive X-ray spectroscopy
(EDS) of the Nb/SBA-15/3im sample displays the presence of
niobium, oxygen, silicon and copper elements (Fig. 3C), and the
presence of copper element is attributed to the conductive copper
tape used in the EDS measurement.
Raman spectroscopy is employed to investigate the isolated
NbO4 information and extent of oligomer formation in networks of
surface niobium oxides (Fig. 4A). Comparison with pure Nb2O5,
similar bridging Nb–O–Nb bands at approximately 664 and 234
cm−1 are observed in Nb/SBA-15/im samples with high Nb content
(>3%).33 Meanwhile, some remarkable modifications of these
signals with regard to broadness and position toward lower
frequencies are also detected, which can be associated with the
NbO6 distortions for the phonon relaxation energy by alteration of
the niobium structure.34 The new peak appears at approximately
462 cm−1 assigned to the bridging Nb–O–Nb–O–Si sites, confirming
the presence of agglomerized extra framework niobium species.
Meanwhile, all Nb-loaded samples possess a Raman band at about
967 cm−1, which is related to symmetric stretching of terminal
Nb=O bonds on the surface due to highly dispersed Nb species.35
In the 790-810 cm−1 region, tetrahedral NbO4 shows its major
Raman bands,36 and the intensity is higher in the spectrum of Nb-
SBA-15/3co sample. Therefore, two types of surface niobium
species are most likely present in the samples: one is the surface
monomeric or oligomeric NbO6 species, another one is the NbO4
species.
Fig. 2 Low-angle (A) and high-angle (B) XRD patterns of samples.
impregnation (Fig. 2A).32 The positions of diffraction peaks for
Nb/SBA-15/im samples are higher than those of SBA-15 support
which testified that the ordering porous structure of SBA-15 was
partly distorted after the impregnation of Nb species. In case of
Nb-SBA-15/3co, all diffraction signals are shifted to lower values,
which can be attributed to an increase in the unit cell parameter
caused by the incorporation of Nb ions with larger ionic radius
(0.64 Å) in the SBA-15 framework.3 The characteristic reflections
of crystalline niobium oxide (JCPDS card No. 00-007-0061) are not
detected for any catalyst (Fig. 2B), which indicates that Nb ions are
either well dispersed or attained an amorphous oligomeric form
outside of the framework.
TEM images of the samples in Fig. 3 demonstrate the retention
of the periodic structure, which shows a long range periodic
structure and niobium oxide species encapsulated into the
FT-IR spectra of Nb-modified SBA-15 samples show a sharp and
weak peak is observed at ca. 3736 cm–1 (Fig. 4B), which is
characteristic of terminal –OH bonded to distorted NbO6 species
present on the surface.37 Comparison with pure silica support, two
band intensities at 3459 and 1630 cm−1 for Nb/SBA-15/im samples
decrease upon the loading of small amounts of niobia, showing a
drastic decrease of the amount of silanol groups, indicating that
the loading of Nb species on SBA-15 consumes surface Si–OH.
With further increasing the Nb-loading amount, an increase in
absorption, however, is observed likely due to hydrogen-bonding
interactions between silanols and more neighboring Nb-OH
groups. In contrast, these peak intensities for Nb/SBA-15/3co
C
C
Fig. 3 TEM images of Nb/SBA-15/3im sample (A1 and A2) and Nb-SBA-15/3co
sample (B1 and B2); X-ray spectroscopy (EDS) of Nb/SBA-15/3im sample (C).
Fig.4 UV-Raman spectra (A) and FT-IR spectra (B) of samples.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 3
Please do not adjust margins

Please do not adjust margins
Catalysis Science & Technology
Page 4 of 14
ARTICLE
Journal Name
prepared by co-condensation, however, are about the same as
those of the silica support. In addition, the band at 960 cm−1 is due
to a Si–O vibrational mode perturbed by the presence of metal
ions in a neighboring position. A similar band has been observed
for titanosilicates, which is assigned to surface titanyl species Ti–O
or tetrahedral TiO4 units.38 Thus, we assign the band at 960 cm−1 in
the IR spectra of Nb-modified samples to the presence of Si–O–Nb
structure by a [SiO4] unit bonded to a niobium ion (O3Si–O–
NbO3).39
UV-Vis spectra of the Nb-modified catalysts are obtained with
SBA-15 and Nb2O5 as references (Fig. 5A). Various Nb/SBA-15/im
samples show a band around 210 nm associated with local Nb–O–
Si bonds in monomeric [NbO4] tetrahedral units,40 which is also
observed for the as-synthesized Nb/SBA-15/3co. This indicates
that part of niobium species introduced by impregnation is
incorporated into the zeolite framework, which is consistent with
the results of Liu et al.41 Comparison with pure Nb2O5, the broad
band centered around 250 nm with a tail extending from 270 nm
can be assigned to distorted NbO6 octahedra, and the intensity
increases with increasing the doping of Nb content. This can be
related to a wider distribution of monomeric Nb structural units,
with the presence of a greater amount of more oligomeric NbO6
species in the high Nb-loaded samples.42 The Nb/SBA-15/3co
sample prepared by co-condensation can be also observed similar
absorptions at 210 and 270 nm, but the intensities are much lower
in the same content. Thus, Nb/SBA-15/im samples prepared by
impregnation is highly dispersed on the surface of the siliceous
support, which is in well agreement with the XPS and X-ray
fluorescence studies that XPS and bulk Nb/Si atomic ratios are
very similar for Nb/SBA-15/im samples.
surveys for all Nb-modified samples clearly show the presence of
View Article Online
Si, Nb, O and trace amounts of adventitioDuOs IC: 1f0r.o10m39p/Du0reCYS0B0A4-7195K.
XPS analysis for Nb3d spectra revealed spin-orbit doublet peaks
centred at ca. 208.0 eV (3d5/2) and ca. 211.0 eV (3d3/2)
demonstrating the presence of Nb5+ in an oxide surrounding.43
However, the shift toward higher values (208.3 eV) than that
typical of Nb2O5 species (207.0 eV) has been attributed to the high
dispersion of niobium oxides on the support and to the change in
the Nb coordination by formation of Nb–O–Si or or Si–O–Nb–O–
Nb linkages. Moreover, the binding energy of Nb 3d follows a
decreased trend towards those observed for crystalline Nb2O5
with increasing the Nb-loading amount, as shown in Fig. 5C. This
can contain contributions arising from different niobium species
such as a spot of small oligomer on the surface of catalysts,44
which is in agreement with what already observed by Raman and
UV-Vis spectra. From quantitative XPS results, the surface Nb
molar concentrations can be determined from the surface Nb/Si
ratios. Table S1 shows the comparison between the surface Nb/Si
atomic ratio values as determined by using XPS and those
corresponding to the bulk atomic ratio by using X-ray fluorescence
techniques. The XPS and bulk Nb/Si atomic ratios are very similar
for Nb/SBA-15/7im prepared by impregnation. However, when the
Nb species loading is prepared by co-condensation, the surface
Nb/Si atomic ratio notably decreases with respect to that of the
bulk Nb/Si for the Nb/SBA-15/3co. This confirms the high ability of
the impregnation method to accommodate host phases, as
niobium oxide, at the surface of the guest silica matrix. Niobium
oxide is well-dispersed over the surface of the support as
previously reported.45 The O1s lineshape region (Fig. 5C) for all the
Nb-loaded catalysts show an almost symmetric peak in the region
532.9–533.1 eV, close to the value observed for O2− of SBA-15
support (533.0 eV),46 which evidenced the presence of
homogeneously dispersed Nb species on the surface of the
mesoporous support.
The qualitative and quantitative study on the niobia presence
at the surfaces are carried out by X-ray photoelectron
spectroscopy (XPS), and are shown in Table S1 and Fig. 5B. XPS
Pyridine adsorption in-situ FTIR spectra is collected for the
Nb/SBA-15/im and Nb/SBA-15/3co samples in the region of
1630~1400 cm−1 (Fig. 6). In the case of Nb/SBA-15/1im samples,
the bands at 1446 and 1597 cm−1 dominate (Fig. 6A), which is
attributable to the pyridine adsorbed via a weak hydrogen-
bonding on surface hydroxyl groups (e.g., Si–OH or Nb–OH).47
Meanwhile, these peak intensities increased with increasing the
Nb content up to 3% and decreased with a further increase, which
is attributable to the concentration distribution of surface OH
groups. On the other hand, several weak peaks for adsorbed
pyridine are also observed at 1577, 1544 and 1491 cm−1. The band
at 1577 cm−1 arises from adsorbed pyridine on Lewis acid sites;40
absorbance features at 1544 cm−1 is assignable to pyridinium ions
originated from the Nb–OH–Nb in polymerized Nb species,47,48 and
peak intensities keep progressive increase with increasing Nb2O5
content. The band at 1491 cm−1 rises from the presence of Si–OH–
Nb or Nb–OH–Nb sites in which the pyridine is thought to bond
simultaneously to Lewis and Brønsted acid sites.49 Compared to
Nb/SBA-15/3im, significant increases of peak intensities at 1444
and 1598 cm−1 are observed for Nb/SBA-15/3co prepared by co-
condensation (Fig. 6A), indicating more surface Si–OH groups.
Further, IR spectra of Nb/SBA-15/3im is recorded after evacuation
at various temperatures. As shown in Fig. 6B, the peaks
C
Fig. 5 UV-Vis spectra (A) of samples; Survey spectrum of samples (B) and XPS spectra
of Nb 3d, O 1s and Si 2p of samples (C).
4 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 5 of 14
Catalysis Science & Technology
Please do not adjust margins
Journal Name
ARTICLE
turnover frequency (TOF) values for cyclohexylamine conversion
View Article Online
DOI: 10.1039/D0CY00479K
based on niobium sites on catalyst are calculated. Generally,
cyclohexylamine conversion increases with increasing niobium
loading up to 5 wt.% Nb; at higher loadings cyclohexylamine
conversion keeps nearly constant (entries 8-11). The TOF of the
lowest content Nb species is very high (740.1 h-1), and TOFs in
oxidation reactions strongly decrease with increasing Nb loading.
As already discussed in the preceding characterization section, the
monomeric or oligomeric niobium species become progressively
polymerized with increasing surface coverage. Therefore, one can
conclude that monomeric or oligomeric niobium species are
intrinsically more active than their high agglomerated
counterparts. Meanwhile, the total selectivity of the main
cyclohexyl-cyclohexylidene-amine and cyclohexanone by-products
over Nb/SBA-15/1im is 17.5% and higher than the selectivity of
14.0% achieved over Nb/SBA-15/7im catalysts with higher
niobium loading, which indicates that further hydrolysis of
cyclohexanone oxime is catalyzed by Lewis acidic sites. This
observation can be explained by the stronger Lewis acidity of
monomeric or oligomeric niobium-sites compared to higher
aggregated ones. On the contrary, the relative selectivity of
nitrocyclohexane by-product increases with an increase in the
niobium loading, which suggests that higher aggregated niobium
species show a higher catalytic activity toward further oxidation of
cyclohexanone oxime to nitrocyclohexane.
Fig. 6 Pyridine adsorption in-situ FTIR spectra of samples (A) and pyridine adsorption
in-situ FTIR spectra of Nb/SBA-15/3im after evacuation at various temperatures.
attributable to the pyridine adsorbed via a weak hydrogen-
bonding obviously decreased after evacuation at 423 K, which is
attributable to a reduction in the number of the surface OH
groups. The bands attributable to weak Lewis-bound pyridine
(1576 cm−1) diminish at
a further increase in evacuation
temperature (473 K), while a new band at 1610 cm−1 origined from
strong Lewis-bounded pyridine appears. The strong Lewis-
bounded reactants is unfavorable to the selectivity of desired
products from the point of view of catalysis.
Catalytic activities
Selective aerobic oxidation of cyclohexylamine to cyclohexanone
oxime catalyzed by the various Nb-based catalysts under solvent-
free conditions is investigated on cyclohexylamine, known to be
highly sensitive to auto-oxidation with main formation of nitro-
compound and ketone-derived products. The catalytic results
compared with the corresponding Nb2O5 and SBA-15 supports are
shown in Table 2. Blank experiment without any catalysts shows
that a remarkably low cyclohexylamine conversion (5.7%) with
only small amounts of cyclohexanone oxime (34.2%) is obtained as
To elucidate whether the niobium-sites linked to the silica
surface is
a key structure in determining the catalytic
performances, we compare the catalytic activity and selectivity of
the reaction performed on the Nb sites incorporated in the
framework for Nb/SBA-15/3co containing the same moles of
niobium. Relatively lower catalytic activity (conversion and TOFs),
however, is obtained under similar conditions (entry 12). This can
be a result of the change in surface speciation of Nb/SBA-15/3co,
and Nb incorporated in the framework is relatively less reactive for
oxidation process. To gain a further insight into the roles of the Nb
species toward the oxygen activation, two sets of poisoning
experiments have been carried out with KSCN involved in the
reactions, which is thought that the metal sites would deactivate
for the O2 activation due to the strong affinity of SCN– ions to Nb
metals by the formation of highly stable metal-ligand
compounds.52,53 One of the experiments is investigated with the
addition of KSCN (15 equiv. relative to Nb) into the catalytic
systems over the Nb2O5 catalyst, the cyclohexylamine conversion
significantly decreases from 15.9% to 1.1%; another
cyclohexylamine oxidation to cyclohexanone oxime over Nb/SBA-
a
result of the thermal auto-oxidation process (entry 1).
Meanwhile, the bulk Nb2O5 is active for the cyclohexylamine
oxidation and gives good oxime selectivity (>73%) along with the
main cyclohexyl-cyclohexylidene-amine by-product (13.2%),
whereas a low cyclohexylamine conversion of 15.9% is observed
under study (entry 3). One can generally believe that O2
dissociates on Nb sites by transferring charge density from Nb
sites to the vacant π* molecular orbital of adsorbed O2,50,51 which
is produced concomitantly and thereby generates active oxygen
species for the initiation of oxidation reaction. Interesting, its
reactivity is significantly improved by the introduction of SBA-15 as
supports; in striking contrast, about 75.2% of the cyclohexylamine
is converted with Nb/SBA-15/3im under the same conditions, and
the higher selectivity to cyclohexanone oxime (84.3%) is achieved
along with the main nitrocyclohexane by-product (entry 9). This
finding strongly supports that the supported niobium-sites
structure with a different coordination environment is responsible
for the enhanced catalytic activity in the selective oxidation
process. Actually, we also tested the activity of a sample of
mechanically mixed Nb2O5 and SBA-15 (3 wt% of SBA-15), and
the conversion is only one-third of that of Nb/SBA-15/3im
(not shown), with producing a quantity of nitrocyclohexane
by-products. Thus, the differently loaded Nb/SBA-15/im catalysts
are tested for their activity and selectivity in the selective
oxidation of cyclohexylamine with the aim to study the influence
of niobium-site structures on the catalytic performance, and the
15/3im shows
a similar trend in KSCN-based poisoning
experiment. This phenomenon reveals that only Nb sites on the
catalyst surface can be strongly coordinated by SCN– ions and
affect their electron donating ability to the vacant π* molecular
orbital of adsorbed O2. As a result, Nb sites on the catalyst surface
provide a direct catalytic site for the activation of O2, as generally
accepted for O2 dissociation on transition metals such as Au, Pt
and Ru in the literature.54,55
In addition, it should be considered if –OH groups present on
the surface of Nb-loaded SBA-15 samples are in sufficient amount
to account for the high activity. First, pure SBA-15 support is
generally considered a catalytically inactive solid with composed
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 5
Please do not adjust margins

Please do not adjust margins
Catalysis Science & Technology
Page 6 of 14
ARTICLE
Journal Name
of Si–O–Si and Si–OH species, which is investigated as a possible
reference (entry 2). Interesting, a notable catalytic activity (12.4%)
is observed under similar conditions, but cyclohexanone oxime
selectivity is remarkably lower (53.7%) over this support.
Obviously, this catalytic activity is related to the surface Si–OH
groups of support. Cyclohexylamine possesses electronegative
nitrogen due to its –NH2 groups and therefore it is easily adsorbed
on –OH sites by formation of hydrogen bonds,56 resulting in a
lower reaction energy barrier for the oxidation of
cyclohexylamine. This result is further confirmed by the fact that
both Nb/MCM-41/im and Nb/Silicate-1/im at the same Nb loading
containing higher concentration of surface –OH groups exhibit also
high conversion of cyclohexylamine to cyclohexanone oxime
(entries 6 and 7). Nb-loaded Al2O3 also exhibits a relatively high
catalytic activity; nevertheless, the selectivity to cyclohexanone
oxime is lower than that of Nb-loaded silica (entry 5). One can
argue from the product distribution, further hydrolysis of
cyclohexanone oxime occurs along with oxidation process, which
is catalyzed by the stronger Lewis acidity sites from the Al–O–Al or
Al–O–Nb structure of Nb/Al2O3/im due to the strong interaction
between Nb5+ or Al3+ and the C=N bond of oxime. In addition, the
changes on the surface of the catalyst with regard to catalytic
activity are also confirmed by the experiment with the use of
Nb/C/im having no surface Si–OH groups, in which the poor
activity and selectivity (cyclohexylamine conversion, 18.1%;
cyclohexanone oxime selectivity, 50.7%) are observed under the
same reaction conditions. Meanwhile, the surface-OHs of Nb/SBA-
View Article Online
DOI: 10.1039/D0CY00479K
15/3im are dehydroxylated using
triphenylsilylation agent and the formation of oxime decreases,
which supports this conclusion (entries 13, 14). hus, the
a trimethylsilylation or
T
presence of hydroxyl groups is essential to observe the
enhanced catalytic activity for cyclohexylamine oxidation
when using niobium containing catalysts.
Based on above catalysis, one has to consider the changes
of the –OH active sites taking place on the catalyst surface in
determining the catalytic performances. The best performing
catalyst, Nb/SBA-15/3im, is dehydrated at 423, 623, and 823
K respectively and the corresponding catalytic activities are
also investigated under the same conditions. The Nb/SBA-
15/3im catalyst dehydrated at 423
K gives a higher
cyclohexylamine conversion (83.7%) and than that of
untreated Nb/SBA-15/3im catalyst (entry 15). One can
conclude that the removal of weakly adsorbed bulk water to
release more Si–OH and Nb–OH groups at 423 K can improve
the accessibility of cyclohexylamine to surface active sites.
Opposing this result, a dehydration of temperature at 623 K
affords a relatively low cyclohexylamine conversion (54.2%)
Table 2 Selective aerobic oxidation of cyclohexylamine catalyzed by the various Nb-based catalysts under solvent-free conditionsa.
Selectivity (%)c
Yield/%
(oxime)
TOF.d
(h-1)
Entry
Catalyst
Conv.(%)b
Otherse
O
NOH
NO
2
N
1
2
3
4
5
6
7
8
9
10
11
12
13f
14f
15g
16g
17g
18h
19h
20i
21j
None
SBA-15
Nb2O5
Nb/C/3im
5.7
4.7
21.2
5.0
11.2
23.8
4.7
5.4
3.2
2.7
8.6
12.9
7.3
3.7
6.0
2.1
5.4
4.4
2.8
3.1
34.2
27.0
18.2
3.3
13.0
8.7
3.7
15.2
1.8
19.9
0.6
14.2
6.3
4.9
13.8
2.6
1.1
1.9
2.2
2.0
3.0
3.9
2.8
1.7
2.1
2.6
2.4
2.2
1.0
3.1
8.2
7.2
1.9
6.7
11.7
9.2
-
-
12.4
15.9
18.1
36.2
60.8
51.9
39.5
75.2
79.0
80.1
64.3
21.5
34.1
83.7
54.2
38.7
1.1
53.7
73.6
50.7
37.1
74.6
68.4
78.5
84.3
61.1
31.8
67.4
80.9
69.9
81.7
75.1
68.0
18.9
79.9
32.7
49.8
13.2
11.3
27.8
15.9
9.1
14.3
1.3
1.2
1.1
1.8
2.0
0.8
1.0
2.8
4.0
70.1
5.0
99.3
113.1
226.2
379.5
324.3
740.1
469.8
296.7
214.5
401.7
134.4
213.0
531.6
338.4
241.8
6.9
Nb/γ-Al2O3/3im
Nb/MCM-41/3im
Nb/Silicate-1/3im
Nb/SBA-15/1im
Nb/SBA-15/3im
Nb/SBA-15/5im
Nb/SBA-15/7im
Nb-SBA-15/3co
Nb/SBA-15/3im
Nb/SBA-15/3im
Nb/SBA-15/3im
Nb/SBA-15/3im
Nb/SBA-15/3im
Nb2O5
13.4
45.4
35.5
31.0
63.4
48.3
25.5
43.3
17.4
23.8
68.4
40.7
26.3
0.2
9.7
26.1
50.3
20.7
11.7
21.2
12.6
14.3
21.4
7.2
Nb/SBA-15/3im
Nb/SBA-15/3im
Nb/SBA-15/3im
42.3
13.7
4.6
8.9
47.3
12.5
33.8
4.5
2.3
264.3
85.6
28.7
5.9
16.1
5.9
14.4
a
All reactions were done with 0.10 g of catalyst, 10.0 g (101.0 mmol) of cyclohexylamine, 0.8 MPa O2, 373 K, time (5 h).b Conversion (%) based on substrate =
c
{1−[(concentration of substrate left after reaction) × (initial concentration of substrate)−1]} × 100; Product selectivity = content of this product/(adding
cyclohexylamine amount (mmol)-the amount of cyclohexylamine recovered (mmol)) × 100%. Turnover frequency (TOF): number of moles of cyclohexylamine
converted per mole of catalyst (Number in parentheses is based on number of moles of Nb sites only). Others: cyclohexanol, cyclohexene, N, N’-dicyclohexyl-
hydrazine, N-cyclohexyl-N’-cyclohexylidene-hydrazine and small amounts of unidentified products. Using the dehydroxylated Nb/SBA-15/3im by a trimethylsilylation
d
e
f
(entry13) or triphenylsilylation (entry14) agent. g Nb/SBA-15/3im is dehydrated at 423 K (entry15), 623 K (entry16), and 823 K (entry17) respectively. h Using KSCN (15
equiv. relative to Nb) as a poisoning additive for Nb species in the oxidation system. i Using the alkaline Nb/SBA-15/3im by 0.5% NaOH treatment. j Addition of 5 mol%
2,6-di-tert-butyl-p-cresol as a free radical scavenger.
6 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Please do not adjust margins
Catalysis Science & Technology
Page 7 of 14
View Article Online
DOI: 10.1039/D0CY00479K
ARTICLE
and 75.1% cyclohexanone oxime selectivity (entry 16), which
is expected to explain by the loss of part –OH groups for the
activation of cyclohexylamine (Fig. S2). A similar effect is observed
at very high dehydration temperatures (823 K) with only 38.7%
cyclohexylamine conversion and 68% cyclohexanone oxime
selectivity (entry 17).
beginning of the reaction. After
3
h, the conversion and
cyclohexanone oxime selectivity increases progressively, indicating
the presence of reaction induction period. A possible explanation
for this phenomenon is attributed to a radical-type autoxidation
process. Further increasing the reaction time (up to 6 h) causes
the decline of oxime selectivity, implying the occurrence of over-
oxidation. In order to further find out the radical-based
intermediates in oxidation process, we take aliquots of the sample
at regular intervals and analysed them by iodometric titration, and
no hydroperoxide is detected. One can suggest that the possible
hydroperoxide intermediates under reaction conditions are
quickly transformed into more stable products. Fig. 8C presents
that initially the conversion of cyclohexylamine increases slightly
with the catalyst weight in the range 0.05~0.4 g, whereas
cyclohexylamine conversion is not affected much by further
addition of catalyst. Obviously, a concomitant decrease in the
selectivity of cyclohexanone oxime product with an increase in the
amount of catalysts, implies a direct participation of Nb species in
catalyst in the rate-determining step of the oxime over-oxidation.
Finally, an increase of reaction pressure from 0.6 MPa to 1.4 MPa
Fig. 7 Effect of the temperature (A), reaction time (B), catalyst amount (C) and
pressure (D) on the conversion of cyclohexylamine and cyclohexanone,
cyclohexanone oxime, nitrocyclohexane, cyclohexyl-cyclohexylidene-amine and
others selectivity over the synthesized Nb/SBA-15/3im catalyst.
results in
a slightly enhancement of the cyclohexylamine
conversion to 81.6% with the oxime selectivity of 56.7% (Fig. 7D).
This phenomenon is attributed to an enhancement of oxygen
solubility making more oxygen accessible in the reaction medium.
From the context of a ‘green’ approach, the solvent-free
In order to further understand the product distribution in the
cyclohexylamine oxidation under solvent-free conditions, the
catalytic performance of Nb/SBA-15/3im is further optimized by
tuning reaction parameters (Fig. 7). The catalyst displays a very
low catalytic activity (8.3%) with only 63.9% oxime selectivity at
368 K, but with increasing temperature from 368 to 373 K the
conversion of cyclohexylamine rapidly increased to 75.2% (Fig.
7A). These results indicate that the higher reaction temperature
favors the activation of oxygen molecule by niobium species as
well as the formation of dissociative chemisorptions of
cyclohexylamine by surface hydroxyl groups in the first reaction
step. Meanwhile, the cyclohexanone oxime selectivity also
increases sharply from 63.9% to 84.3%, which suggests that a
decrease in the strength of adsorption of cyclohexanone oxime on
the surface of catalyst due to the temperature increase leads to
avoid further hydrolysis of cyclohexanone oxime catalyzed by
Lewis acidic sites. Further increase in the reaction temperature
(378~393K) leads to a remarkable increase in the ability to further
oxidation of cyclohexanone oxime (cyclohexanone and
nitrocyclohexane are the main product with the total selectivity of
about 70%, which is caused by the higher oxidizing ability of the
Nb sites), while this effect is less pronounced for cyclohexylamine
conversion. The effect of the reaction time on the aerobic
oxidation of cyclohexylamine is investigated (Fig. 7B), the
conversion to cyclohexanone oxime is very low (9.5%) in the
Fig.
oxidation. All reactions were carried out with 0.10
cyclohexylamine (101.0 mmol), 0.8 MPa O2, 373 K, time (5 h).
8
Recyclability tests of Nb/SBA-15/3im catalyst for the cyclohexylamine
g
of catalyst, 10 of
g
oxidation of cyclohexylamine to cyclohexanone oxime with
dioxygen is performed for a reused Nb/SBA-15/3im catalyst under
the same reaction conditions. After simple separation and washed
thoroughly with ethanol for three times, Nb/SBA-15/3im catalyst
Please do not adjust margins

Please do not adjust margins
Catalysis Science & Technology
Page 8 of 14
ARTICLE
Journal Name
after five successive oxidation cycles do not lead to any significant
decline in its efficiency in terms of conversion and oxime
selectivity (Fig. 8), demonstrating its good recyclability and
deactivation resistance. Owing to Nb-based catalyst with water
tolerant properties, some similar phenomenons are also reported
for aqueous acid-catalyzed processes and epoxidation of alkene
with H2O2,24 proving to be truly of a heterogeneous nature.
Furthermore, after completion of each run, the solid catalyst is
removed from the reaction mixture by hot-filtration. The resulting
filtrate is independently analyzed by ICP-AES analysis for the
presence of niobium, and no niobium ions are present in the
filtrate. The UV-Vis and FT-IR profiles of the spent catalysts are
very similar to those of the fresh catalysts, as shown in Fig. S3. As
a result, its catalytic efficiency and recyclability will make this Nb-
based catalyst attractive for both fundamental research and
practical applications. To prove the universality of this catalytic
oxidation system based on Nb/SBA-15/3im, the selective oxidation
of several other types of amines is further examined under
solvent-free conditions, and the results are listed in Table 3.
Similar to cyclohexylamine, alicyclic primary amines containing
hydrogen atoms at the α-carbon such as cyclopentylamine and
cycloheptanamine give the high catalytic activity and >73% oxime
selectivity (entries 9 and 11). However, aliphatic primary amine
such as 1-propylamine, 1-butylamine 2-butylamine, n-octylamine
and benzyl amine with a hydrogen atom at the α-carbon can also
View Article Online
DOI: 10.1039/D0CY00479K
be converted into the corresponding oxime, but given the poor
selectivity for oxime (entries 1-4 and 8). These aliphatic primary
amines can experience oxidative coupling to their corresponding
imines with high selectivity, implying that the different structures
can affect the product distribution due to the existence of an
electronic effect in this reaction. Meanwhile, the primary diamines
such as ethanediamine are much less accessible owing to the
intrinsic self-coupling properties of the substrates, unfortunately
no oxime selectivity is observed with a low conversion rate of
diamine (entry 5). However, the primary amine lacking hydrogen
atoms at the α-carbon such as aniline can not be oxidized to the
corresponding oxime or imine (entry 7), implying that the α-
hydrogen is involved in the oxime or imine formation. Meanwhile,
the aerobic oxidation of secondary amines such as di-n-butylamine
with hydrogen atoms at the α-carbon gives the 1-butanone oxime
with 44.6% selectivity over Nb/SBA-15/3im catalyst (entry 6).
DFT treatments of oxygen and cyclohexylamine adsorption-
activation
Selective site-specific adsorption-activation takes place as the
8 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Please do not adjust margins
Catalysis Science & Technology
Page 9 of 14
View Article Online
DOI: 10.1039/D0CY00479K
ARTICLE
Table 3 Selective aerobic oxidation of various amines catalyzed by the 3% Nb/SBA-15 catalysts under solvent-free conditionsa.
Conv.
TOF.d
(h-1)
Entry
Substrate
NH2
NH2
Selectivity (%)c
(%)b
13.5
NOH
N
N
1
2
141.5
69.8
26.1
24.0
11.0
203.3
41.4
NOH
53.3
NOH
NH2
3
4
93.2
30.8
NOH
N
69.2
N
9.2
44.1
NH2
10.3
79.1
8.2
N
NH2
5
6
7
2.3
15.5
1.1
23.7
74.4
7.3
H2N
N
100
NH
NOH
N
44.6
16.8
NO2
NH2
N
N
20.9
79.1
83.2
NOH
N
NH2
8
83.5
482.7
NOH
O
NO2
9
NH2
77.9
75.2
566.7
469.7
73.3
7.6
12.8
NOH
NOH
O
NO2
10
NH2
NH2
84.3
2.7
9.7
O
NO2
11
64.3
351.8
81.5
3.8
10.4
a All reactions were done with 0.10 g of catalyst, 10.0 g of reactant, 0.8 MPa O2, 373 K, time (5 h). b Conversion (%) based on substrate = {1−[(concentration of substrate
left after reaction) × (initial concentration of substrate)−1]} × 100. c Main product selectivity = content of this product/(adding reactant amount (mmol)-the amount of
d
reactant recovered (mmol)) × 100%. Turnover frequency (TOF): number of moles of substrate converted per mole of catalyst. Number in parentheses is based on
number of moles of Nb sites only.
first elementary step of catalytic processes, which consists of a link
between the reactants and the catalytic active centers.57,58 Based
on the experimental examination of surface active centers by
Raman, XPS, UV-vis and FT-IR methods, three possible types of
active structures have been imagined (Fig. S4), in which the
monomeric Nb species is grafted onto SBA-15 surface (a), or
incorporated into the framework of SBA-15 (b) and dimeric Nb
species (as an oligomeric model) is grafted onto SBA-15 surface
(c). In these optimized structure models, the framework Nb=O
bond lengths are 1.732~1.736 Å, which agrees well with EXAFS
measurements of Nb=O bond lengths in the Nb species (1.73±
0.02 Å).59 Similar findings have also been obtained by DFT
calculations on Nb species.60-62 To get insight about how the
different Nb structures can activate oxygen, the interactions
between O2 and the Nb sites are optimized in Fig. 9. First, O2
adsorbed on the monomeric (a-o) or dimeric (c-o) Nb sites grafted
onto SBA-15 surface with a distance of 3.365 Å or 2.803 Å
between Nb and O2, is shorter than these of the O2 adsorbed on
the Nb sites (b-o) incorporated into the framework (3.400 Å).
Second, the monomeric (a-o) or dimeric (c-o) Nb sites grafted onto
SBA-15 surface shows a higher adsorption energies (E(ads)) of O2
(12.1 kJ/mol, 14.5 kJ/mol) in comparison to the corresponding
values for the Nb sites (b-o) incorporated into the framework
(10.8 kJ/mol). This suggests that the surface Nb sites strengthened
the O2 adsorption and made the adsorbed O2 easier to be
activated.
Please do not adjust margins

Please do not adjust margins
Catalysis Science & Technology
Page 10 of 14
ARTICLE
Journal Name
for Nb–OH) and thus benefiting the catalytic VpieewrAfortricmleaOnnclinee.
DOI: 10.1039/D0CY00479K
:Nb, :Si, :O, :N, :C, :H
Fig. 10 Suggested simple models and electronic structure of optimized models
for the isolated Nb species grafted onto SBA-15 surface with the interactions
between cyclohexylamine and the acidic Si-OH sites (a-n1) or Nb-OH sites (a-
n2).
Fig. 9 Suggested simple models and electronic structure of optimized models
for the isolated Nb species grafted onto SBA-15 surface (a-o), Nb species
incorporated in the framework of SBA-15 (b-o) or oligomeric Nb species
grafted onto SBA-15 surface (c-o) with the interactions between O2 and the
Reaction mechanism
Cyclohexylamine conversion to cyclohexanone oxime involves
several key elementary steps, and each elementary step is thought
to be catalyzed by a certain type of functional active sites (Fig. 11).
The active Nb species and –OH groups in the present bifunctional
Nb/SBA-15 catalysts mainly exist as site-specific catalytic centers
toward the activation of different reactants. Combining the
insights provided by preceding results, cyclohexylamine can be
dissociatively chemisorbed on the –OH groups by hydrogen bonds.
Meanwhile, dissociatively chemisorbed cyclohexylamine is
dehydrated at reaction temperature. This can be confirmed by the
following facts: (i) the reactivity of dehydroxylated Nb/SBA-15 or
SBA-15 is decreased remarkably; (ii) the changes in activity are
consistent with the observed surface hydroxyl-site contents by the
dehydration temperature; (iii) the formation of cyclohexanone
oxime strongly depends on the types of catalyst supports.
Actually, this is also similar to the previous reports that pure silica
shows reactivity for gas-phase cyclohexylamine oxidation at high
temperature (>443 K).18,19 Furthermore, DFT calculation reveals
that the N–H bond for adsorbed cyclohexylamine by hydroxyl-sites
is elongated from 1.016 Å to 1.017 Å, which can facilitate
insertation of oxygen species with electrophilic property.
On the other hand, molecular oxygen can be selective
chemisorbed on the dispersed Nb sites to form probably O2-
superoxo radical species, favoring the selective oxidation of
cyclohexylamine. To verify the evidence for the formation of
such radicals, a free radical scavenger (5 mol% 2,6-di-tert-
butyl-p-cresol) was added (Table 2, entry 21). It was found
that the radical scavenger had a strong inhibitory effect on
the cyclohexylamine oxidation. In addition, our DFT calculations
confirm that molecular oxygen can be dissociated on Nb sites,
which is thought to be produced concomitantly and thereby
generates active oxygen species by transferring charge density
from metal to the vacant π* molecular orbital of adsorbed oxygen.
This evidence is in line with our experimental studies that the
poisoning additive (KSCN) for Nb species has a strong inhibitory
effect on the cyclohexylamine reaction and therefore occurs by an
active oxygen-based mechanism. Further per analogy to Nb-based
catalyst,61 experimental and theoretical studies have found that
the proposed oxygen chemisorption species over Nb sites is very
probable. Therefore, the active oxygen species are inserted into
redox Nb sites.
:Nb, :Si, :O, :H
Furthermore, the activation of O2 in our case can be reflected by
the change of number of electron transfer.63 The NBO charge and
molecular orbital analysis shows that the electron transferring
occurs from one oxygen atom to another oxygen atom for
adsorbed O2, which caused by the surface Nb sites compared with
the NBO charge free O2 (0.000 e, Fig. S5 and S6). Owing to the
electrophilic mechanism of the oxidation process, the lower
electron density of the reactive center (O atom) is more likely to
insert into the N–H bond.
To realize more clearly the different activities of Si–OH and Nb–
OH for the activation of cyclohexylamine (models a-n1 and a-n2,
Fig. 10), the basic geometric and electronic properties of active
sites with adsorbed cyclohexylamine are investigated. Upon
adsorption of cyclohexylamine, the calculated O–H bond length of
active sites increases (Fig. 10 and Fig. S4) and the O–H bond length
in Nb–OH is larger than that in Si–OH. Actually, this increase of O–
H bond length is accompanied by a shortening in the formed N····H
hydrogen-bonding length between the nitrogen of –NH2 groups
and the –OH, and consequently denotes a stronger tendency to
transfer the electron on that site. This is in good agreement with
results obtained by the computational adsorption energies: the
adsorption energy is about 85.8 kJ/mol more stable for adsorption
on the Nb–OH sites than on the Si–OH sites (76.3 kJ/mol). Due to
the change in charge distribution around the N atom, the N–H
bond in the adsorbed cyclohexylamine complex is significantly
elongated (1.017 Å), compared to the free cyclohexylamine
molecule (computational dN-H=1.016 Å, Fig. S5). Therefore, a
cleavage of the weak N–H bond of cyclohexylamine occurs easily
over the coordinated nitrogen atom with the –OH site (particularly
10 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 11 of 14
Catalysis Science & Technology
Please do not adjust margins
Journal Name
ARTICLE
hydroxyl groups of Nb/SBA-15 catalysts. This new dual active
View Article Online
DOI: 10.1039/D0CY00479K
center model with site-specific catalytic activities, which has been
validated by our experimental measurements and computational
results, may provide a new avenue for unraveling the chemical
properties of various metal-modified bifunctional catalysts.
Experimental
Catalyst preparation
Pure SBA-15 was synthesized according to literature65 with
Pluronic P123 triblock-co-polymer (EO20PO70EO20). Reactant
mixtures consisted of tetraethyl orthosilicate (TEOS), Pluronic
P123, hydrochloric acid, and deionized H2O with the molar
ratio: 1 SiO2 : 0.005 Pluronic P123 : 1.45 HCl : 124 H2O. After
the dissolving of Pluronic P123 in 1M hydrochloric acid
solution, TEOS was added dropwise under stirring. The
obtained mixture was continuously stirred at 313 K for 24 h
and transferred into a Teflon-lined stainless steel autoclave
and heated without any stirring at 373 K for 24 h. The solid
product was filtered, washed with deionized water, dried at
373 K for 12 h and calcined at 823 K for 8 h to remove
EO20PO70EO20.
Method 1: Nb-loaded catalysts (Nb/SBA-15) were prepared
by the aqueous impregnation method employing NbCl5 as a
precursor. The prepared SBA-15 (2 g) was poured into the
ethanol solution of NbCl5 to give 3 wt.% niobium (0.032 mol%)
in the final product. After stirring at room temperature for 12
h, 9.0 mL deionized H2O were slowly added to the mixture
and vigorously stirred for 6 h and continually stirred for
another 24 h. The impregnated sample was allowed to
evaporate slowly at 383 K until dry sample was obtained. The
1wt.% (0.011 mol%), 5 wt.% (0.054 mol%) and 7 wt.% (0.075
mol%) Nb-loading samples were prepared by an identical
procedure. The final samples contained 1, 3, 5, and 7 mass %
of Nb (Nb/SBA-15/xim, with x representing the Nb mass %).
Method 2: Nb-incorporated SBA-15 molecular sieves (Nb-
SBA-15) were prepared by co-condensation method with 3
Fig. 11 Possible synergistic mechanism for the oxidation of cyclohexylamine with O2
over the Nb-modified SBA-15 catalysts.
adsorbed amine to form absorbed amino peroxide, which further
reacts with H2O by the hydrolysis of Si–N bond to leave the
catalyst surface and the recovery of the –OH groups. The formed
free amino peroxide in our system is decomposed rapidly to form
intermediate nitrosocyclohexane, in agreement with the results
presented by Corma et al., who showed that it is possible to form
nitroso compound directly.64 If nitrosocyclohexane comes in
contact with a Brønsted acid site, it can be quickly isomerized to
cyclohexanone oxime which is released in the reaction medium.19
This can be confirmed by poisoning experimental studies that
the alkaline Nb/SBA-15/3im catalyst by 0.5% NaOH treatment
gave the poor selectivity for oxime (Table 2, entry 20).
Actually, a effective promotion of reaction rate is also obtained for
amine oxidation with the acid-treated catalyst. As the coexistence
of active oxygen species from Nb-based catalytic site,
nitrosocyclohexane can also react with active oxygen species to
form nitrocyclohexane by-products. Therefore, this underlines the
importance of having strong Brønsted acid sites available at the
catalyst surface (than weak Lewis acid sites). Thus, two different
types of surface active sites should be well balanced to suppress
formation of unwanted by-products.
Conclusions
mass
% (0.032 mol%) of Nb (Nb-SBA-15/xco, with x
representing the Nb mass %).32 NbCl5 was used as a niobium
source with addition of hydrochloric acid. In this preparation
method, an amount of NbCl5 was first dissolved in 140 mL
ultrapure H2O, followed by the addition of 4 g EO20PO70EO20
template. After dissolution of the template, 9.0 mL (40.7
mmol) of TEOS were added and stirred at 313 K for 24 h. The
obtained suspension was placed into a Teflon-lined stainless
steel autoclave and heated without the stirring at 373 K for
24 h. Then, the solid was filtered, washed and air dried at
room temperature. The dried materials were air-calcined at
823 K for 8 h at a ramp rate of 10 K.
In conclusion, we present the first green examples of Nb-modified
mesoporous SBA-15 catalysts for selective oxidation of various
amines using O2 under solvent-free conditions. Particularly, the
aerobic selective oxidation of cyclohexylamine into cyclohexanone
oxime is thoroughly studied in order to get more insight into the
structure-property correlation and mechanism for the oxidation of
amine. The best Nb/SBA-15 catalyst with a 3% niobium content
exhibits superior catalytic performance (TOF
=
469.8 h–1),
achieving 75% cyclohexylamine conversion and 84%
cyclohexanone oxime selectivity at 373 K. The results, obtained
from the comprehensive experimental studies involving the
characterization, detection, reaction of the different amines and
reaction conditions have demonstrated that the surface dispersed
Nb species and hydroxyl groups are identified as catalytic active
sites for cyclohexylamine. The dispersed Nb species mainly
catalyzed the activation of O2 to active oxygen species, which is
essential for the selective oxidation of cyclohexylamine. Both the
activity and selectivity strongly depend on the exposed surface
Characterization techniques
The crystalline phases of powers were characterized by a
Bruker D8 ADVANCE diffractometer operated at 40 kV and 40
mA, using CuKα radiation with a wavelength of 1.542 Å.
Particle size and morphological properties of samples were
done using scanning obtained on
a Hitachi S-4800
microscope, while the TEM images were collected using a
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 11
Please do not adjust margins

Please do not adjust margins
Catalysis Science & Technology
Page 12 of 14
ARTICLE
Journal Name
JEOL 2010 field emission gun at 200 kV accelerating voltage.
The near surface compositions of Nb and Si species in the
synthesized solids were determined using an energy
dispersive X-ray spectroscopy (EDS) equipped in the electron
microscope JEOL-JSM-6400. The BET surface area and
porosity of samples were obtained by N2 adsorption at 77 K
on a Tristar 3000 sorptometer. Prior to the tests, the samples
were degassed at 473 K under vacuum until a final pressure
of 1×10-5 Torr was reached. Raman spectra were obtained for
solids on an inVia Raman microscope using the excitation
wavelength from a laser at 325 nm. Framework infrared
spectra of samples in KBr pellets were recorded at room
temperature on an AVATAR 370 Thermo Nicolet
spectrophotometer with a resolution of 2 cm-1. UV/Vis
diffuse-reflectance spectra were conducted on a Varian-Cary
5000 spectrometer with an integrating-sphere attachment
using BaSO4 as a reference to assess the nature of Nb species.
The concentrations of Brønsted and Lewis sites in materials
were determined by the infrared spectra of adsorbed
pyridine using a custom-built transmission cell coupled to a
Fourier transform infrared (FT-IR) spectrometer (Bruker,
Tensor 37) with a liquid-N2-cooled HgCdTe detector. The
powders were pressed binder-free into self-supporting
wafers with a density of 4.0~11.0 mg cm-2. Prior to the
adsorption of pyridine the samples were activated ‘in situ’ by
overnight evacuation at 573 K. Pyridine was admitted at 373
K for 1 h in order to reach adsorption equilibrium, followed
by degassing at different temperatures for 0.5 h, and then
the IR spectrum was measured at 373 K. The surface
electronic states were analyzed by X-ray photoelectron
spectroscopy (XPS, Axis Ultra, Kratos Analytical Ltd.) with Al
Ka radiation (1486.7 eV). Binding energies were referenced to
the C1S peak at 284.8 eV, and fit XPS software was used for
curve fitting. Bulk elemental analysis was carried out with
Inductive CouplePlasma Atomic Emission Spectroscopy (ICP-
AES) on a spectroflame D (Spectro Analytic Instrument).
DFT calculations
minima and obtained zero point vibrational energy (ZPE)
corrections to the electronic energies.
View Article Online
DOI: 10.1039/D0CY00479K
The binding energies of oxygen or cyclohexylamine
adsorbates on the catalyst surfaces with different active sites
(BEads) were calculated by eq (1) ： BEads= Eadsorbate+surface
−(Eadsorbate + Esurface), in which Eadsorbate+surface represented the
total energy of the adsorbate interacting with the active sites
of the catalyst, while Eadsorbate and Esurface were the energy of
for the adsorbates in the gas phase and the bare catalyst
surface respectively. One can do this by considering that
there are two different adsorption sites on the surface for
oxygen or cyclohexylamine, with energies Eads-Nb-oxygen and
Eads-OH-cyclohexylamine, each one can be calculated from equation
(1).
Cyclohexylamine selective oxidation reaction
The catalytic selective oxidation of cyclohexylamine with
molecular oxygen in the absence of solvent was carried out in
a Teflon-lined stainless steel autoclave of 100 cm3 capacity
with
a
magnetical stirring. Typically, 10.0
g
of
cyclohexylamine (101.0 mmol) was mixed with 0.1 g of
Nb/SBA-15 catalyst and then raised to the desired
temperature (373 K). Then, oxygen was charged into the
reactor to the desired pressure (0.8 MPa) and reaction was
started by adjusting stirrer speed to 600 rpm. After reaction,
the reactor was quickly cooled down and then depressurized
carefully. The products were dissolved in ethanol and the
solid catalyst was separated from aqueous phase by filtration.
The mixtures were quantitatively analyzed by
a gas
chromatography (GC, Agilent 6890N, 30 m × 0.32 mm × 0.50
µm DB-17 polysiloxane capillary column) equipped with a
flame ionization detector using chlorobenzene as an internal
standard substance. Both the injector and detector
temperatures were 523 K, and the column temperature was
443 K. The components of oxygen formed were analyzed by
gas chromatography with TCD detector. The products were
satisfactorily identified by gas chromatography/mass
spectrometry (GC-MS) and compared with authentic
standard samples. Mass balances were verified.
On the basis of the experimental results, we chose the
smallest geometrical structures with active Nb sites and
surface hydroxyl groups as catalyst surface models for
computation in this work, which provides a straightforward
link between theory and experiment. In our system, Gaussian
09 program package was applied to perform our density
functional theory (DFT) calculations using the B3LYP hybrid
functional with tight self-consistent field (SCF) convergence
and ultrafine integration grids.66 The dispersion corrections
were described by “Empirical Dispersion = GD3”.67 The 6-
311++(d, p) basis set was adopted for the C, H, O, and N
atoms, and the Nb atoms adopted the valence basis
functional (LANL2DZ).56,68 Natural bond orbital (NBO)
analyses were performed using the NBO 6.0 program to gain
insight into the natural charge distribution and bonding
patterns for all the structures. In these models, the
unsaturated bonds are saturated by hydrogen atoms.
Vibrational frequencies were calculated for all obtained
structures to confirm that they corresponded to the energy
Acknowledgements
This
work
was
supported
by
the
Natural Science Foundation of China (Grant No.21576078,
21878074 and 21978078) and Collaborative Innovation
Center of New Chemical Technologies for Environmental
Benignity and Efficient Resource Utilization.
References
1
2
3
4
S. Guo, Z. Du, S. Zhang, D. Li, Z. Li and Y. Deng, Green
Chem., 2015, 280, 670−676.
C. Shen, Y. J. Wang, C. Dong and G. S. Luo, Chem. Eng. J.,
2014, 235, 75−82.
M. Anilkumar and W.F. Hoelderich, Appl. Catal. B: Environ.,
2015, 165, 87–93.
H. Ichihashi and H. Sato, Appl. Catal. A: Gen., 2001, 221,
359–366.
12 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 13 of 14
Catalysis Science & Technology
Please do not adjust margins
Journal Name
ARTICLE
5
6
7
Z. Wang, H. Ling, J. Shi, C. Stampfl, A. Yu, M. Hunger and J.
Huang, J. Catal., 2018, 358, 71–79.
A. Thangaraj, S. Sivasanker and P. Ratnasamy, J. Catal., 1991,
131, 94−400.
R. Chen, H. Mao, X. Zhang, W. Xing and Y. Fa, Ind. Eng.
Chem. Res., 2014, 53, 6372−6379.
R. Prasad and S. Vashisht, J. Catal., 1996, 161, 373−376.
W. Zhong, T. Qiao, J. Dai, L. Mao, Q. Xu, G. Zou, X. Liu, D. Yin
and F. Zhao, J. Catal., 2015, 330, 208−221.
37 A. A. Gabrienko, S. S. Arzumanov, A. V. Toktarev, I. G.
Danilova, I. P. Prosvirin, V. V. Kriventsov, V. I.VZieawikAortvicslekOiin,liDne.
Freude and A. G. Stepanov, ACS CatalD.,O2I:01107.1,073,91/D801C8Y–010843709K.
38 P. R. H. P. Rao, A. V. Ramaswamy and P. Ratnasamy, J.
Catal., 1992, 137, 225–231.
39 A. M. Prakash and Larry Kevan, J. Am. Chem. Soc., 1998,
120, 13148–13155.
8
9
40 B. Kilos, A. Tuel, M. Ziolek and J. C. Volta, Catal. Today, 2006,
118, 416–424.
10 P. Serna, M. Lopez-Haro, J. J. Calvino and A. Corma, J. Catal.
2009, 263, 328-334.
11 P. Rubio-Marqués, J. C. Hernández-Garrido, A. Leyva-Pérez
and A. Corma, Chem. Commun., 2014, 50, 1645-1647.
12 F. Cavani, N. Ballarini and S. Luciani, Top. Catal., 2009, 52,
935–947.
41 K. Peng, X. Li, X. Liu and Y. Wang, Mol. Catal., 2017, 441,
72–80.
42 N. E. Thornburg, A. B. Thompson and J. M. Notestein, ACS
Catal., 2015, 5, 5077–5088.
43 L. Kong, C. Zhang, J. Wang, W. Qiao, L. Ling and D. Long, Sci.
Rep., 2016, 6, 21177.
13 J. S. Reddy and A. Sayari, Appl. Catal. A, 1995, 128, 231–
242.
14 R. Joseph, T. Ravindranathan and A. Sudalai, Tetrahedron
Lett., 1995, 36, 1903–1904.
15 G. Náray-Szabó and L. T. Mika, Green Chem., 2018, 20,
2171–2191.
16 J. N. Armor, E. J. Carlson, R. Riggitano, J. Yamanis and P. M.
Zambri, J. Catal., 1983, 83, 487−490.
17 K. Rakottyay and A. Kaszonyi, Appl. Catal. A: Gen., 2009,
367, 32−38.
18 W. Zhong, L. Mao, W. Yi, S. R. Kirk, D. Yin, A. Zheng and H.
Luo, Catal. Commu., 2014, 56, 148–152.
19 S. Liu, K. You, J. Jian, F. Zhao, W. Zhong, D. Yin, P. Liu, Q. Ai
and H. Luo, J. Catal., 2016, 338, 239–249.
20 S. K. Klitgaard, K. Egeblad, U. V. Mentzel, A. G. Popov, T.
Jensen, E. Taarning, I.S. Nielsen and C.H. Christensen, Green
Chem., 2008, 10, 419−423.
21 S. Liu, K. You, J. Song, R. Deng, F. Zhao, P. Liu, Q. Ai and H.
Luo, Appl. Catal. A: Gen., 2018, 568, 76−85.
22 K. Suzuki, T. Watanabe and S. I. Murahashi, Angew. Chem.
Int. Ed., 2008, 47, 2079−2081.
23 L. Dong, Y. Xin, X. Liu, Y. Guo, C.-W. Pao, J.-L. Chen and Y.
Wang, Green Chem., 2019, 21, 3081−3090.
24 I. Nowak and M. Ziolek, Chem. Rev., 1999, 99, 3603−3624.
25 K. Nakajima, J. Hirata, M. Kim, N. K. Gupta, T. Murayama, A.
Yoshida, N. Hiyoshi, A. Fukuoka and W. Ueda, ACS Catal.,
2018, 8, 283−290.
44 I. D. Ivanchikova, N. V. Maksimchuk, I. Y. Skobelev, V. V.
Kaichev and O. A. Kholdeeva, J. Catal., 2015, 332, 138–148.
45 C. García-Sancho, I. Sádaba, R. Moreno-Tost, J. Mérida-
Robles, J. Santamaría-González, M. López-Granados and P.
Maireles-Torres, ChemSusChem, 2013, 6, 635–642.
46 Z. Yan, L. Fu, X. Zuo and H. Yang, Appl. Catal. B: Environ.,
2018, 226, 23–30.
47 D. T. Bregante, N. E. Thornburg, J. M. Notestein and D. W.
Flaherty, ACS Catal., 2018, 8, 2995–3010.
48 A. Takagaki, J. N. Kondo, M. Hara, S. Hayashi and K. Domen,
Chem. Mater., 2005, 17, 2487–2489.
49 T. Barzetti, E. Selli, D. Moscotti and L. Forni, J. Chem. Soc.,
Faraday Trans., 1996, 92, 1401–1407.
50 S. Ghosh, S. S. Acharyya, R. Tiwari, B. Sarkar, R. K. Singha, C.
Pendem, T. Sasaki and R. Bal, ACS Catal., 2014, 4,
2169−2174.
51 A. Tabler, A. Häusser and E. Roduner, J. Mol. Catal. A:
Chem., 2013, 379, 139−145.
52 L. Lin, Z. K. Yang, Y. F. Jiang and A. W. Xu, ACS Catal., 2016,
6, 4449–4454.
53 J. Li, G. Liu, X. Long, G. Gao, J. Wu and F. Li, J. Catal., 2017,
355, 53–62.
54 T. Matsushima, Surf. Sci., 1985, 157, 297–318.
55 S. Han and C. B. Mullins, ACS Catal., 2018, 8, 3641−3649.
56 G. Qiu, C. Huang, X. Sun and B. Chen, Green Chem., 2019,
21, 3930−3939.
57 G. K. Koyanagi, D. Caraiman, V. Blagojevic and D. K. Bohme,
J. Phys. Chem. A, 2002, 106, 4581–4590.
58 M. C. Silaghi, C. V. Aleix and C. Copéret, ACS Catal., 2016, 6,
4501−4505.
26 Md. N. Rashed, S. M. A. H. Siddiki, Md. A. Ali, S. K. Moromi,
A. S. Touchy, K. Kon, T. Toyao and K. Shimizu, Green Chem.,
2017, 19, 3238−3242.
27 M. Ziolek, I. Sobczak, P. Decyk, I. Nowak and J. Kujawa,
Micropor. Mesopor. Mat., 2000, 35–36, 195–207.
28 P. Carniti, A. Gervasini, C. Tiozzo and M. Guidotti, ACS Catal.,
2014, 4, 469–479.
29 X. Kong, Y. Zhu, Z. Fang, J. A. Kozinski, I. S. Butler, L.
Xu, H. Song and X. Wei, Green Chem., 2018, 20, 3657–
3682.
59 C. Tiozzo, C. Bisio, F. Carniato, A. Gallo, S. L. Scott, R. Psaro
and M. Guidotti, Phys. Chem. Chem. Phys., 2013, 15,
13354–13362.
60 P. D. Patel, B. B. Laird and W. H. Thompson, J. Mol. Catal. A:
Chem., 2016, 424, 1–7.
61 H. Zhai, X. Zhang, W. Chen, X. Huang and L. Wang, J. Am.
Chem. Soc., 2011, 133, 3085–3094.
30 M. Ziolek, I. Sobczak, P. Decyk, K. Sobańska, P. Pietrzyk and
Z. Sojka, Appl. Catal. B: Environ., 2015, 164, 288−296.
31 Y. Fan, S. Cheng, H. Wang, D. Ye, S. Xie, Y. Pei, H. Hu, W.
Hua, Z. H. Li, M. Qiao and B. Zong, Green Chem., 2017, 19,
2174–2183.
32 D. Zhao, J. Feng, Q. Huo, N. Melosh, G. H. Fredrickson, B. F.
Chmelka and G. D. Stucky, Science, 1998, 279, 548–552.
33 Z. Zhao, X. Gao and I. E. Wachs, J. Phys. Chem. B, 2003, 107,
6333–6342.
62 W. Yan, A. Ramanathan, P. D. Patel, S. K. Maiti, B. B. Laird,
W. H. Thompson and B. Subramaniam, J. Catal., 2016, 336,
75–84.
63 X. Jin, C. Lv, X. Zhou, C. Zhang, Q. Meng, Y. Liu and G. Chen,
Appl. Catal. B: Environ., 2018, 226, 53–60.
64 A. Corma, P. Concepción and P. Serna, Angew. Chem. 2007,
119, 7404-7407.
65 M. Trejda, A. Tuel, J. Kujawa, B. Kilos and M. Ziolek,
Micropor. Mesopor. Mat., 2008, 110, 271–278.
66 C. Lee, W. Yang and R. G. Parr, Phys. Rev. B: Condens
Matter Mater. Phys., 1988, 37, 785–789.
34 G. W. Graham, W. H. Weber, C. R. Peters and R. Usmen, J.
Catal., 1991, 130, 310–313.
35 L. J. Burcham, J. Datka and I. E. Wachs, J. Phys. Chem. B,
1999, 103, 6015.
36 J. M. Jehng and I. E. Wachs, Chem. Mater., 1991, 3, 100–
107.
67 S. Grimme, J. Antony, S. Ehrlich and H. Krieg, J. Chem.
Phys. 2010, 132, 154104-154119.
68 K. Fukui, Acc. Chem. Res., 1981, 14, 471–476.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 13
Please do not adjust margins

Catalysis Science & Technology
Page 14 of 14
View Article Online
DOI: 10.1039/D0CY00479K
Figure-abstract