Lewis acidic strength controlled highly selective synthesis of oxime via liquid-phase ammoximation over titanosilicates

Article abstract of DOI:10.1039/c4ra08799b

The Lewis acidity of titanosilicates determines the selectivity of the oxime in ammoximation. Higher Lewis acidic strength of Ti active sites could promote free H₂O₂ to participate in the highly efficient formation of NH₂OH by lowering the reaction activation energy for the formation of Ti-OOH species, and thus fundamentally suppress the side reactions of deep oxidation.

Full text of DOI:10.1039/c4ra08799b

RSC Advances
View Article Online
View Journal | View Issue
COMMUNICATION
Lewis acidic strength controlled highly selective
synthesis of oxime via liquid-phase ammoximation
over titanosilicates†
Cite this: RSC Adv., 2014, 4, 55685
Received 17th August 2014
Accepted 29th September 2014
*
Zuoxi Zhuo, Lizhi Wu, Lei Wang, Yichun Ding, Xiaoqian Zhang, Yueming Liu
and Mingyuan He
DOI: 10.1039/c4ra08799b
www.rsc.org/advances
The Lewis acidity of titanosilicates determines the selectivity of the titanosilicates can be divided into two parts:8 rst, the activation
oxime in ammoximation. Higher Lewis acidic strength of Ti active sites of H2O2 on framework tetrahedral Ti4+ to form the oxygen
could promote free H2O2 to participate in the highly efficient forma- donating-intermediates of Ti–OOH species, which depends on
tion of NH2OH by lowering the reaction activation energy for the the catalytic activity of titanosilicates, has been proven to be the
formation of Ti–OOH species, and thus fundamentally suppress the most crucial step in titanosilicate-catalyzed oxidation reac-
side reactions of deep oxidation.
tions;6b,9 then, the surrounding NH3 can capture the active
oxygen in Ti–OOH species to give NH2OH intermediates
(S3 ESI†). Thus, in order to fundamentally promote the highly
efficient formation of NH2OH, it is important to accelerate the
activation of free H2O2 via lowering the reaction energy barrier
of Ti–OOH species. Furthermore, it also well known that reac-
tion activation energy is strongly reduced with the increase of
acid strength in hydrocarbon conversion over zeolites.10
Reasonably, titanosilicate catalyst equipped with stronger Lewis
acidity of Ti active sites is inclined to remarkably reduce the
reaction activation energy of Ti–OOH species, thereby activating
free H2O2 to produce NH2OH more efficiently and further
cutting off the side reaction of oxidation. Therefore, we infer
that the Lewis acidity of titanosilicates determines oxime
selectivity in ammoximation by controlling reaction path of free
H2O2 to NH2OH intermediates.
In this communication, to demonstrate that the Lewis acidic
strength of Ti active sites is the decisive factor in the highly
selective synthesis of oxime, a series of ammoximation reac-
tions of linear ketones and acetaldehyde over representative
titanosilicates with varied strength of Lewis acidity (TS-1 < Ti-
MWW < F-Ti-MWW) (S4 ESI†) has been carried out. As antici-
pated, a titanosilicate catalyst with stronger Lewis acidity gives
better selectivity to oxime.
Finding a selective catalyst for the synthesis of a desired product
is very oen the beginning step of catalysis in industry.1 The
successful manufacture of cyclohexanone oxime via liquid-
phase ammoximation over TS-1(MFI) is considered a new
milestone in the eld of zeolite catalysis.2 However, applying
this green chemical process to the production of high-value ne
oximes (e.g. methyl ethyl ketone oxime, acetone oxime and
acetaldehyde oxime) is hampered due to the relatively lower
selectivity for the target product.3 The typical side reaction in
ammoximation can be concluded to be the deep oxidation of
reactant or oxime, which depends greatly on the behavior of free
H2O2 oxidation.3,4 It is well accepted that the ammoximation
reaction proceeds via NH2OH intermediates (S2 ESI†),5 and it
was proven that a higher concentration of NH3 favors NH2OH
formation so that can inhibit side reaction of oxidation.4 Hence,
promoting the highly efficient formation of NH2OH, and thus
suppress free H2O2-induced oxidation reactions would
contribute to the high selectivity of oxime.
It has been conrmed that the active site of titanosilicates is
tetracoordinated framework TiIV,6 and many publications have
supported that the catalytic nature of titanosilicates is attrib-
uted to Lewis acidic sites related to these framework Ti species.7
Thus, it is reasonable to employ Lewis acidity of Ti active sites
to represent the catalytic activity of titanosilicates. In ammox-
imation, the ammonia oxidation by free H2O2 over
Table 1 shows the results of ammoximation of linear ketones
(acetone and MEK) over Ti-MWW and TS-1. To focus on the
issue of product selectivity, the ketone conversion was main-
tained at over 99% by optimizing the reaction conditions, such
as supplying a sufficient catalyst concentration (15 g molꢀ1) and
loading the free H2O2 drop-wise in case of further oxidation
decomposition of NH2OH intermediates (S2 ESI†).5a,11 Ti-MWW
was proven to be a selective catalyst for linear ketone ammox-
imation, and the selectivity of DMKO and MEKO both were
Shanghai Key Laboratory of Green Chemistry and Chemical Processes, Department of
Chemistry, East China Normal University, Shanghai, China. E-mail: ymliu@chem.
ecnu.edu.cn; Fax: +86 21 62232058
† Electronic
supplementary
information
(ESI)
available:
Synthesis,
characterization methods and Fig. S. See DOI: 10.1039/c4ra08799b
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 55685–55688 | 55685

View Article Online
RSC Advances
Communication
Table 1 A comparison of products selectivity in different ammox-
nearly 25% higher than that of TS-1. Thus, the catalytic abilities
of forming NH2OH between Ti-MWW and TS-1 are consistent
with their order of Lewis acidic strength (Fig. 1a) and also agree
with the change in oxime selectivity (Table 1). These positive
correlations fully verify our inference that titanosilicates with
higher Lewis acidic strength are benecial for the acceleration
of the reaction path of NH2OH. Therefore, the above results
support well our opinion that the Lewis acidity of titanosilicates
decides the highly selective synthesis of oxime via affecting the
highly efficient formation of NH2OH intermediates.
imation processesa
DMKb (mol%)
MEKc (mol%)
No.
Catalyst
Conv.
Sel.
Sel.d
Conv.
Sel.
Sel.e
1
2
Ti-MWW
TS-1
99.6
99.3
99.8
94.9
0.2
5.1
99.7
99.7
99.7
95.9
0.3
4.1
a
Reaction conditions: catalyst 0.15 g; ketone 10 mmol, H2O2 12 mmol;
NH3, 21 mmol; temp. 335 K for Ti-MWW (Si/Ti ¼ 42) and 345 K for TS-1
(Si/Ti ¼ 38); time 2 h. b DMK for dimethyl ketone or acetone. c MEK for
methyl ethyl ketone. d The selectivity of 2-nitropropane. e The selectivity
of 2-nitrobutane.
In order to better demonstrate the decisive relations between
the Lewis acidity and oxime selectivity, a series of linear ketone
ammoximation reactions over mixed catalysts of TS-1/Ti-MWW
was further conducted. The results showed that TS-1 possesses
weaker Lewis acidity than Ti-MWW (Fig. 1a), and thus, it is
feasible to prepare a catalyst sample with adjustable Lewis
acidic strength by mixing these two types of titanosilicates in
different proportions. Reasonably, upon increasing the weight
proportion of Ti-MWW, the Lewis acidic strength of the mixed
catalysts is enhanced simultaneously. If oxime selectivity is
decided by the Lewis acidity of titanosilicates, then the selec-
tivity would be changed along with the acid strength. Fig. 2
reports the results of DMK and MEK ammoximation over the
TS-1/Ti-MWW mixed catalysts. As expected, the oxime selectivity
was improved gradually with the increase in the Ti-MWW
proportion. This gradual change in selectivity also excludes
the possibility that the positive effect on oxime selectivity
enhancement is merely caused by the introduced Ti-MWW. No
further variation in oxime selectivity was noted when the Ti-
MWW proportion was higher than 30%, which means this
specic composition of mixed catalysts has enough Lewis acidic
strength to prepare oxime with high selectivity. These ndings
show that the Lewis acidic strength of titanosilicate catalyst is
able to regulate the oxime selectivity in ammoximation. Corre-
spondingly, 2-nitroalkane selectivity also decreased with an
increase in the proportion of Ti-MWW, and the by-product was
nearly eliminated when adopting the mixed sample consisting
of 30% Ti-MWW and 70% TS-1 as the catalyst. This implies that
the deep oxidation of oxime, even in the presence of diffusion
higher than 99.5% (Table 1, no. 1). When it comes to TS-1, the
oxime selectivity was reduced to a certain extent because of the
existence of 2-nitropropane and 2-nitrobutane that originated
from the consecutive oxidation of oxime (Table 1, no. 2).3a
Although Song et al. attributed the decline in linear oxime
selectivity to diffusion constraints induced by TS-1 catalyst,3a
our further study demonstrates that the diffusion constraints of
catalyst merely intensify the side reaction of deep oxidation
instead of deciding the oxime selectivity (S5 ESI†). Thus,
combined with the ammoximation of linear ketones of different
carbon chains (S6 ESI†), the oxime selectivity differences exist-
ing between Ti-MWW and TS-1 conform quite well to our
assumption; namely, Ti-MWW possessing higher Lewis acidic
strength displays superior oxime selectivity compared to that of
TS-1 with relatively weaker Lewis acidity.
In order to further clarify that the effect of Lewis acidity of
titanosilicates on the improvement of oxime selectivity is
derived from promotion of the formation of NH2OH interme-
diates, we have correlated the Lewis acidic strength of Ti-MWW
and TS-1 with their corresponding abilities to produce NH2OH.
Fig. 1b compares the yield of NH2OH obtained by Ti-MWW and
TS-1 with different Ti contents according to the relevant results
of cyclohexanone ammoximation. Ti-MWW kept a much better
yield of NH2OH than TS-1 under the same Ti content, and the
maximum yield gained by Ti-MWW was close to 95%, which is
Fig. 2 The ammoximation of DMK (A) and MEK (B) over the TS-1/Ti-
Fig. 1 Correlation between the formation of NH2OH (b) and the Lewis MWW mixed catalysts. (a) Ketone conv., (b) oxime sel., (c) 2-nitroalkane
acidic strength of titanosilicates (a). Reaction conditions: catalyst, sel., (d) H2O2 residual concentration. Reaction conditions: catalyst,
0.05 g; cyclohexanone, 20 mmol; H2O2, 10 mmol; NH3, 10 mmol, 0.45 g; ketone, 30 mmol; H2O2, 36 mmol; NH3, 63 mmol; sol., 10 g,
others see Table 1.
temp., 341 K; time, 2 h.
55686 | RSC Adv., 2014, 4, 55685–55688
This journal is © The Royal Society of Chemistry 2014

View Article Online
Communication
RSC Advances
constraints caused by TS-1, can be mostly suppressed by incorporated uorine species.13 Therefore, F-Ti-MWW was
strengthening the Lewis acidity of the catalyst. Furthermore, the applied to the AA ammoximation to further verify our inference
variation of residual free H2O2 also supports that titanosilicates about Lewis acidity controlling the oxime selectivity. Expec-
with stronger Lewis acidity prefer to activate more free H2O2 tantly, Table 2 (no. 5) shows that the AAO selectivity was further
molecules to form NH2OH. Thus, it is well conrmed that the increased up to 99.6% and it was equal to the linear oxime
Lewis acidity of titanosilicates is the decisive factor in the selectivity achieved with Ti-MWW (Table 1, no. 1). Hence, the
synthesis of oxime with high selectivity.
titanosilicates investigated for AA ammoximation displayed the
In comparison to linear ketones, acetaldehyde (AA) is following order of oxime selectivity: F-Ti-MWW > 99.5% > Ti-
chemically more active, so that the selectivity issue in the AA MWW > 90% > TS-1, which also entirely conforms to their
ammoximation becomes more complicated.3b,12 On the one relevant Lewis acidic strength. This indicates that to highly
hand, it was shown that AA would be further oxidized to acetic selectively obtain chemically more active oxime, stronger Lewis
acid by Ti–OOH species,3a although this may be unlikely in an acidity of titanosilicates is needed. Moreover, by comparing the
actual ammoximation due to the existence of excess NH3. On by-product selectivity between acetic acid and acetamide, we
the other hand, it has been proven that certain amount of acetic nd that acetic acid selectivity is lower than that of acetamide.
acid and acetamide can be generated from homogenous deep This proves that Ti–OOH species would not promote the deep
oxidation of AA or oxime with free H2O2 under alkaline condi- oxidation of AA in the presence of excess NH3, and the side
tions (S7 ESI†), which also closely relies on the free H2O2 reactions are still mainly ascribed to those deep oxidations
oxidation actions. Consequently, in order to achieve an excel- induced by free H2O2. These phenomena further indicate that
lent selectivity of acetaldehyde oxime (AAO), a selective titano- the fundamental role of Lewis acidity in oxime selectivity is
silicate with much stronger Lewis acidity is required based on controlling the oxidation behavior of free H2O2. This role was
our present opinion. Table 2 compares the results of linear also supported by the apparent activation energy of Ti–OOH
ketone and AA ammoximation over titanosilicates with different active species formed using TS-1 (32.29 kJ molꢀ1), Ti-MWW
Lewis acidic strength. TS-1 exhibited much poorer selectivity for (24.72 kJ molꢀ1), and F-Ti-MWW (10.02 kJ molꢀ1) (S8 ESI†),
AAO than linear oximes (Table 2, no. 1–3), which indicates a which demonstrates that the increase in Lewis acidic strength
higher chemical activity of AA. Then, AAO selectivity was can remarkably lower the activation barrier of free H2O2 and
signicantly increased to 98.6% when adopting Ti-MWW as the thus promote participation of free H2O2 molecules in the
catalyst because of the notable enhancement in Lewis acidity. formation of NH2OH intermediates.
Nevertheless, the AAO selectivity still failed to reach the parallel
In summary, the Lewis acidic strength of Ti active sites is
level of linear oximes obtained under the same reaction conrmed to be responsible for the high selectivity of oxime
conditions (Table, no. 1). These consequences fully sustain our based on the catalytic nature of titanosilicates and the
speculation that with regard to the chemically more active AA, ammoximation mechanism. A stronger Lewis acid promotes the
achieving high oxime selectivity becomes much harder than formation of NH2OH, while weaker acid strength is less able to
linear ketones by choosing titanosilicates with considerable suppress the free H2O2-induced side reactions. Therefore,
strength of Lewis acidity. Hence, a titanosilicate catalyst with regulating the Lewis acidity of titanosilicates could become a
more superior Lewis acidity than Ti-MWW is needed for the AA crucial guideline for designing or modifying catalysts for highly
ammoximation.
selective synthesis of oximes or even other oxygenated ne
Recently, the Lewis acidic strength of Ti active sites in F-Ti- chemicals.
MWW has been proven to be further enhanced compared to
Ti-MWW due to the strong electron-withdrawing effect of
Acknowledgements
We gratefully acknowledge the Science and Technology
Commission of Shanghai Municipality (12JC1403600) and the
Shanghai Leading Academic Discipline Project (B409).
Table 2 A comparison of acetaldehyde and ketone ammoximation
over titanosilicates with different Lewis acidic strengtha
Notes and references
Sel.(mol%)
By-product
Conv.
1 J. G. de Vries and S. D. Jackson, Catal. Sci. Technol., 2012, 2,
2009.
No.
Reactant
Catalyst
(mol%)
Oxime
Ic
IId
2 (a) P. Roffia, G. Leofanti, A. Cesana, M. A. Mantegazza,
M. Padovan, G. Petrini, S. Tonti and P. Gervasutti, Stud.
Surf. Sci. Catal., 1990, 55, 43; (b) F. Rivetti and R. Buzzoni,
in Liquid phase Oxidation via Heterogeneous Catalysis:
Organic Synthesis and Industrial Application, ed. M.G. Clerici
and O.A. Kholdeeva, The John Wiley & Sons Press, New
York, 2013, p. 462.
1b
2b
3
4
5
MEK
DMK
AA
AA
AA
TS-1(38)
TS-1(38)
TS-1(38)
Ti-MWW(42)
F-Ti-MWW(29)
99.7
99.3
99.1
99.3
99.3
95.9
94.9
86.6
98.5
99.6
4.1
5.1
5.0
0.5
0
8.4
1.0
0.4
a
b
Reaction conditions: see Table 1. The main by-product was
c
d
2-nitroalkane. By-product I was acetic acid. By-product II was
acetamide.
3 (a) F. Song, Y. Liu, L. Wang, H. Zhang, M. He and P. Wu, Appl.
Catal., A, 2007, 327, 22; (b) J. Ding, L. Xu, Y. Yu,
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 55685–55688 | 55687

View Article Online
RSC Advances
Communication
H. W. S. Huang, Y. Yang, J. Wu and P. Wu, Catal. Sci.
Technol., 2013, 3, 2587.
4 (a) A. Cesana, M. A. Mantegazza and M. Pastori, J. Mol. Catal.
A: Chem., 1997, 117, 367; (b) M. A. Mantegazza, A. Cesana and
M. Pastori, Top. Catal., 1996, 3, 327; (c) J. Le Bars, J. Dakka
and R. A. Sheldon, Appl. Catal., A, 1996, 136, 69.
5 (a) M. A. Mantegazza, G. Leofanti, G. Petrini, M. Padovan,
A. Zecchina and S. Bordiga, Stud. Surf. Sci. Catal., 1994, 82,
541; (b) A. Zecchima, G. Spoto, S. Bordiga, F. Geobaldo,
G. Petrini, G. Leofanti, M. Padovan, M. Mantegazza and
P. Roffia, Stud. Surf. Sci. Catal., 1993, 75, 719; (c) P. Wu,
T. Komatsu and T. Yashima, J. Catal., 1998, 168, 400; (d)
Y. Oumi and A. Miyamoto, J. Phys. Chem. B, 2001, 105,
3493; (c) C. Chu, H. Zhao, Y. Qi and F. Xin, J. Mol. Model.,
2013, 19, 2217.
9 (a) P. E. Sinclair and C. R. A. Catlow, J. Phys. Chem. B, 1999,
103, 1084; (b) J. Zhuang, G. Yang, D. Ma, X. Lan, X. Liu,
X. Han, X. Bao and U. Mueller, Angew. Chem., Int. Ed.,
2004, 43, 6377; (c) L. Wang, G. Xiong, J. Su, P. Li and
H. Guo, J. Phys. Chem. C, 2012, 116, 9122; (d) W. Lin and
H. Frei, J. Am. Chem. Soc., 2002, 124, 9292; (e) S. Bordiga,
A. Damin, F. Bonino, G. Ricchiardi, C. Lamberti and
A. Zecchina, Angew. Chem., Int. Ed., 2002, 41, 4374; (f)
M. G. Clerici and P. Ingallina, J. Catal., 1993, 140, 71.
F. Song, Y. Liu, H. Wu, M. He, P. Wu and T. Tatsumi, J. 10 A. M. Rigby, G. J. Kramer and R. A. van Santen, J. Catal., 1997,
Catal., 2006, 237, 359. 170, 1.
6 (a) B. Notari, Adv. Catal., 1996, 41, 253; (b) P. Ratnasamy, 11 (a) L. Xu, J. Ding, Y. Yang and P. Wu, J. Catal., 2014, 309, 1;
¨
D. Srinivas and H. Knozinger, Adv. Catal., 2004, 48, 1.
(b) L. D. Pozzo, G. Fornasari and T. Monti, Catal. Commun.,
2002, 3, 369.
7 (a) A. Corma and H. Garcia, Chem. Rev., 2003, 103, 4307; (b)
J. Zhuang, D. Ma, Z. Yan, F. Deng, X. Liu, X. Huan, X. Bao, 12 (a) S. Zhao, J. Xu, M. Wei and Y. Song, Green Chem., 2011, 13,
X. Liu, X. Guo and X. Wang, J. Catal., 2004, 211, 670; (c)
G. Yang, J. Zhuang, D. Ma, X. Lan, L. Zhou, X. Liu, X. Han
384; (b) D. Sloboda-Rozner and R. Neumann, Green Chem.,
2006, 8, 679.
and X. Bao, J. Mol. Struct., 2008, 882, 24; (d) J. Zhuang, 13 (a) X. Fang, Q. Wang, A. Zheng, Y. Liu, Y. Wang, X. Deng,
Z. Yan, X. Lan, X. Liu, X. Han, X. Bao and U. Mueller,
Catal. Lett., 2002, 83, 87.
8 (a) A. Zecchina, S. Bordiga, C. Lamberti, G. Ricchiardi,
G. Ricchiardi, D. Scarano, G. Petrini, G. Leofanti and
M. Mantegazza, Catal. Today, 1996, 32, 97; (b) H. Munalata,
H. Wu, F. Deng, M. He and P. Wu, Catal. Sci. Technol.,
2012, 2, 2433; (b) X. Fang, Q. Wang, A. Zheng, Y. Liu,
L. Lin, H. Wu, F. Deng, M. He and P. Wu, Phys. Chem.
Chem. Phys., 2013, 15, 4930.
55688 | RSC Adv., 2014, 4, 55685–55688
This journal is © The Royal Society of Chemistry 2014