In situcreation of multi-metallic species inside porous silicate materials with tunable catalytic properties

Article abstract of DOI:10.1039/d1cc01797g

Porous metal silicate (PMS) material PMS-11, consisting of uniformly distributed multi-metallic species inside the pores, is synthesized by using a discrete multi-metal coordination complex as the template, demonstrating high catalytic activity and selectivity in hydrogenation of halogenated nitrobenzenes by synergistically activating different reactant moleculesviaNi and Co transition metal centers, while Gd^IIILewis acid sites play a role in tuning the catalytic properties.

Full text of DOI:10.1039/d1cc01797g

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
In situ creation of multi-metallic species inside
porous silicate materials with tunable catalytic
properties†
Cite this: Chem. Commun., 2021,
57, 6185
Received 5th April 2021,
Accepted 21st May 2021
Yang-Yang Liu, Guo-Peng Zhan and Chuan-De Wu
*
DOI: 10.1039/d1cc01797g
rsc.li/chemcomm
Porous metal silicate (PMS) material PMS-11, consisting of uniformly traditional synthesis strategy for porous silicate materials, the
distributed multi-metallic species inside the pores, is synthesized by uniform shapes and sizes of metal–organic templates would be
using a discrete multi-metal coordination complex as the template, able to define the uniform distribution of pore sizes and shapes
demonstrating high catalytic activity and selectivity in hydrogenation in porous silicate materials. Superior to the traditional strategy
of halogenated nitrobenzenes by synergistically activating different for immobilization of metallic species in porous silicate mate-
reactant molecules via Ni and Co transition metal centers, while GdIII rials, the constituents and sizes of in situ generated and
Lewis acid sites play a role in tuning the catalytic properties.
encapsulated redox-active metallic species in porous silicate
materials are systematically tunable to improve the catalytic
The catalytic sites, consisting of multiple redox-active metal properties by tuning the synthesis methods and annealing
species, demonstrate superior catalytic properties in numerous conditions. Because many transition metal ions have similar
reactions, derived from the distinct electron-deficient/rich coordination ability and coordination environments, they
functions of individual metals that could synergistically acti- easily form isostructural complexes, indicating that the metal
vate different reactant molecules.1,2 Because metallic species constituents in porous metal silicate (PMS) materials are arbi-
are easily aggregated, they are traditionally stabilized by organic trarily tunable to optimize the catalytic properties.
ligands or immobilized in porous matrices.3 However, the
As shown in Scheme 1, [Ni2Gd2] is a coordination complex,
mismatched pore sizes and structures, and incompatibility of consisting of two NiII and two GdIII ions connected by hydro-
porous supports to metallic species often results in metal xymethylpyridine and benzoate ligands.7 Because CoII and NiII
aggregation, which would heavily deteriorate the catalytic ions with similar coordination ability easily form isostructural
properties.4
Among numerous porous support matrices for immobiliza-
tion of redox-active metal species, porous silicate materials,
which are very stable in a harsh environment, represent one
class of the most ideal porous matrices for developing indust-
rially applicable catalysts.5 Because the affinity between silica
and redox-active metals is very weak, it is a challenge to
introduce and subsequently stabilize redox-active metal sites
inside porous silicate materials, due to the easy sintering and
agglomeration nature of metallic species.6 Since porous silicate
materials were traditionally synthesized by using organic tem-
plates, it would be a very promising strategy for in situ produc-
tion and subsequent immobilization of metallic species into
porous silicate materials by using discrete metal–organic com-
plexes instead of traditional organic templates. Similar to the
State Key Laboratory of Silicon Materials, Department of Chemistry,
Zhejiang University, Hangzhou, 310027, P. R. China. E-mail: cdwu@zju.edu.cn
Scheme 1 Schematic representation of the synthetic procedures for
† Electronic supplementary information (ESI) available: Experimental details and
PMS-11.
additional figures and scheme. See DOI: 10.1039/d1cc01797g
This journal is © The Royal Society of Chemistry 2021
Chem. Commun., 2021, 57, 6185–6188 | 6185

View Article Online
Communication
ChemComm
carboxylate moieties of benzoate (Fig. S3, ESI†).9,10 The vibra-
tion peaks for organic ligands were gradually weakened when
raising the annealing temperature, while the vibration of the
Si–O bonds becomes stronger, indicating gradual decomposi-
tion of the organic ligands in PMS-11.
N2 sorption measurements revealed that PMS-11 and its pre-
cursor exhibit the typical type IV isotherms consisting of hyster-
esis loops. PMS-11 takes up 465 cm3 gÀ1 N2 at 77 K and 1 bar,
resulting in a microporous BET surface area of 232 m2 gÀ1, which
is much higher than that of the PMS-11 precursor (47 m2 gÀ1). The
micropore size in PMS-11 is mainly around 2.0 nm, which is very
close to the size of [NiCoGd2] in PMS-11 (B1.9 nm), indicating
that the pore characteristics of PMS-11 depend on the metal–
organic template (Fig. 1b and Table S1, ESI†). The SEM image
shows that there are loosely packed nanoparticles in PMS-11,
which are similar to those in its precursor (Fig. 1c and Fig. S4,
ESI†). TEM and STEM-EDS elemental mapping images show the
uniform morphology of the annealed materials at different tem-
peratures, while the metallic species is dispersed in the material
without aggregation (Fig. S5–S8, ESI†).
Fig. 1 (a) PXRD profiles of [NiGd], [NiCoGd2], the PMS-11 precursor and
PMS-11 (the number in the parenthesis represents the annealing tempera-
ture). (b) Nitrogen adsorption/desorption isotherms for the PMS-11 pre-
cursor and PMS-11 (the inset shows the pore size distributions for the
PMS-11 precursor and PMS-11). (c) SEM and (d) TEM images of PMS-11.
The surface chemical states and binding energies of Ni and
Co species in PMS-11 were studied by X-ray photoelectron
spectroscopy (XPS) (Fig. S9, ESI†). The binding energies of Ni0
complexes, the NiII ions could be partially or fully replaced by 2p3/2 and 2p1/2 for PMS-11 are located at 855.6 and 873.3 eV,
CoII ions with any ratio to form a series of isostructural while the Ni2+ 2p3/2 and 2p1/2 peaks are located at 861.5 and
coordination complexes, which have been confirmed by PXRD 879.7 eV, respectively (Fig. 2a).11 The strong peaks at 779.2 and
and ICP–MS analysis (Fig. 1a). Additionally, rare earth metal 795.3 eV are the characteristic binding energies of Co0, and the
ions with unique orbital structures and 4f electrons might be peaks at 784.3 and 801.2 eV correspond to Co2+ (Fig. 2b).
able to improve the catalytic properties by synergistic coopera- Compared with those of the PMS materials whose templates
tion with transition metal centers.8 We report herein a PMS contain NiII, CoII, NiII and GdIII, CoII and GdIII, and NiII and
material, denoted as PMS-11, consisting of Ni0, Co0 and GdIII CoII, the Ni 2p peaks for PMS-11 shifted to the high binding
metal species, which was templated by [NiCoGd2] with a Ni : energy field, while the Co 2p peaks shifted to the low binding
Co : Gd ratio of 1 : 1 : 2 and subsequently annealed at 600 1C energy field, indicating that electron transfer occurred after the
under a N2 atmosphere, demonstrating excellent catalytic introduction of GdIII and Co0 species in PMS-11 (Fig. S10 and
properties in hydrogenation of halogenated nitrobenzenes to S11, ESI†). The above results demonstrate that the introduction
produce halogenated anilines with almost quantitative conver- of Co and Gd species in Ni-based catalysts could tune the
sion and selectivity.
The TGA curve shows that the weight loss of the PMS-11
precursor that occurred between 200 and 570 1C under a
nitrogen atmosphere, which is similar to the thermal behaviors
of [NiCoGd2], should be ascribed to pyrolysis of the organic
ligands (Fig. S1, ESI†). Accordingly, the PMS-11 precursor was
pyrolyzed at different temperatures. There is no sharp diffrac-
tion PXRD peak for the PMS-11 precursor and the annealed
products at 400, 500 and 600 1C, indicating that metal species
were well-distributed without aggregation in the solid materials
(Fig. 1a and Fig. S2, ESI†). When the temperature was increased
to 700 1C, there appeared the characteristic diffraction peaks
ascribed to silica (JCPDS No. 39-1425).9 Further increasing the
temperature to 800 1C, the crystallization degree of PMS-11 was
highly improved.
In the FT-IR spectrum of the PMS-11 precursor, the peaks at
470, 795 and 1100 cmÀ1 are attributed to the vibration of Si–O
bonds, the peaks at 640 and 720 cmÀ1 are the vibration of the
Fig. 2 (a) High resolution Ni 2p and (b) Co 2p XPS spectra for different
pyridine ring in hydroxymethylpyridine, and the peaks at 1400,
PMS materials (the letters in the parentheses represent the encapsulated
1545 and 1602 cmÀ1 correspond to the stretching vibration of metal species).
6186 | Chem. Commun., 2021, 57, 6185–6188
This journal is © The Royal Society of Chemistry 2021

View Article Online
ChemComm
Communication
electron and orbital properties, which might exhibit unique
catalytic properties.
Halogenated anilines are an important class of intermediates
and raw materials for the synthesis of fine chemicals, which were
traditionally synthesized by hydrogenation of halogenated nitro-
benzenes, catalyzed by noble metal-based catalysts, while dehalo-
genation inevitably occurred.12–14 Even though great effort has been
paid to developing non-noble metal-based catalysts, low activity
and selectivity have heavily frustrated their practical applications.15
It is interesting that PMS-11 exhibited high catalytic activity and
selectivity in hydrogenation of p-chloronitrobenzene (p-CNB) to
produce halogenated aniline p-chloroaniline (p-CAN) with 499%
conversion and 499% selectivity. Detailed experiments revealed
that every metal species and porous silicate matrix played very
important roles in the catalytic properties (Fig. S12, ESI†). The
p-CNB conversion catalyzed by PMS-11 (CoGd) is very low, while a
significant amount of p-CNB was dehalogenated catalyzed by
PMS-11 (NiGd) under catalytic conditions. The Gd species also
played very important roles in the catalytic reaction, which resulted
in decreased conversion and selectivity in the absence of the GdIII
species. Without the protecting silicate matrix, the p-CNB conver-
sion decreased to 9.1%, due to the easy aggregation nature of
metallic species. The annealing temperature also played important
roles in the catalytic results, lower and higher temperature results
in low catalytic activity, and even low selectivity (Table S3, ESI†).
The catalytic properties of PMS-11 are much superior to those
of most transition metal-based catalysts, even noble-metal-based
catalysts, in the literature (Tables S4 and S5, ESI†).
Fig. 3 Reaction rates for selective hydrogenation of p-CNB catalyzed by
different catalysts (the inset shows the apparent activation energies for
different catalysts). Reaction conditions: p-CNB (5 mmol), catalyst (0.05 mol%
based on Ni and Co), H2O (2.5 mL) and ethanol (7.5 mL), 1 MPa H2.
To understand the catalytic properties of PMS-11, we studied
the kinetic behaviors in the hydrogenation reaction. The reac-
tion was conducted at 100–130 1C for the calculation of rational
apparent activation energies (Ea), resulting in different Ea
values in the order of PMS-11 (34.1 kJ molÀ1) o PMS-11 (NiCo)
(49.8 kJ molÀ1) o PMS-11 (NiGd) (65.3 kJ molÀ1) o PMS-11
(CoGd) (89.8 kJ molÀ1) (Fig. 3 and Fig. S13–S15, ESI†). These
results clearly demonstrate the correlation between the catalytic
activity and the redox-active centers. It is worth noting that the
Ea value for PMS-11 is very close to those of Pt (42.7 kJ molÀ1
)
and Au-based catalysts (24.9–94.6 kJ molÀ1), indicating the high
catalytic activity of PMS-11.16 PMS-11 with Ni : Co ratio of 1 : 1
also demonstrates the highest turnover frequency (TOF) of
394.9 hÀ1, compared with the control catalysts with different
Ni : Co ratios, indicating that the Ni : Co ratio is one of the
important factors to improve the catalytic properties by syner-
gistically activating different reactant molecules (Fig. S16 and
Scheme 2 Hydrogenation of various halogenated nitrobenzenes to pro-
duce halogenated anilines catalyzed by PMS-11. Reaction conditions:
halogenated nitrobenzene (0.5 mmol), PMS-11 (1.0 mol% based on Ni
and Co), H2O (1.5 mL) and ethanol (4.5 mL).
Table S3, ESI†). PMS-11 also demonstrates general applicabil- selectivity for catalytic hydrogenation of iodonitrobenzenes is
ity, which could hydrogenate various halogenated nitroben- difficultly improved in the literature.17
zenes (Scheme 2). o-, m- and p-CNB and o-, m- and
PMS-11 exhibited excellent stability and reusability, which
p-fluoronitrobenzenes were completely converted into the could be simply recovered by centrifugation, and reused in
corresponding chloroanilines, despite the fact that there exists the next run for ten cycles with 499% p-CNB conversion and
steric hindrance between –NO2 and –X in o-CNB and 499% p-CAN selectivity (Fig. S17, ESI†). The PXRD pattern
o-fluoronitrobenzene. o-, m- and p-bromonitrobenzene and and TEM image of recovered PMS-11 are almost identical to
o-, m- and p-iodonitrobenzenes, containing C–Br and C–I those of the as-synthesized one, indicating that PMS-11 is
bonds, which are more easily dehalogenated, were also trans- stable and the metal species were not aggregated when
ferred into the corresponding halogenated anilines without catalyzing the hydrogenation reaction (Fig. S18 and S19,
dehalogenation. These results are interesting, because the ESI†).
This journal is © The Royal Society of Chemistry 2021
Chem. Commun., 2021, 57, 6185–6188 | 6187

View Article Online
Communication
ChemComm
To comprehensively understand the contribution of differ-
We are grateful for the financial support of the National
ent metal species in catalyzing the hydrogenation reaction, we Natural Science Foundation of China (grant no. 21525312 and
synthesized a series of control catalysts that consist of different 21872122).
metal species by directly pyrolyzing the corresponding coordi-
nation complexes in the absence of protecting silicate matrix,
denoted as [Ni], [Co], [NiCo], [NiGd], [CoGd] and [NiCoGd2]
based on the metal species. Considering that iodonitrobenzene
Conflicts of interest
There are no conflicts to declare.
is more easily dehalogenated, we selected hydrogenation of
o-iodonitrobenzene as a model reaction to more evidently
illustrate the effect of different metal species on the catalytic
results (Table S6, ESI†). The substrate conversion and product
selectivity are 5.5 and 74% catalyzed by [Ni], which are different
to 1.7 and 499% catalyzed by [Co], respectively, indicating that
Ni and Co metallic species played different roles in improving
the catalytic efficiency and selectivity, respectively. The conver-
sion (7.1%) and selectivity (85%) were apparently improved
when a physical mixture of [Ni] and [Co] was used as the
catalyst, indicating that there occurred synergistic work
between Ni and Co redox-active species by activation of differ-
ent reactant molecules. When [NiCo] was used as a catalyst, the
conversion and selectivity were further improved to 8.2 and
93%, respectively, due to the closed contact between Ni and Co
redox-active sites. GdIII species could further improve the
catalytic properties, as demonstrated by the catalytic results
by [NiCoGd2] with 8.5% conversion and 499% selectivity.
According to the catalytic results and the literature, we
proposed a plausible mechanism for hydrogenation of haloge-
nated nitrobenzenes catalyzed by PMS-11 (Scheme S1, ESI†).
Initially, halogenated nitrobenzene and hydrogen molecules
were chemically adsorbed on the Co0 and Ni0 active sites in
Notes and references
1 (a) M. Sankar, N. Dimitratos, P. J. Miedziak, P. P. Wells and
G. J. Hutchings, Chem. Soc. Rev., 2012, 41, 8009; (b) P. Buchwalter,
J. Rose and P. Braunstein, Chem. Rev., 2015, 115, 28.
2 G.-P. Zhan and C.-D. Wu, Sci. Bull., 2019, 64, 385.
3 (a) J. Liang, Z. B. Liang, R. Q. Zou and Y. L. Zhao, Adv. Mater., 2017,
29, 1701139; (b) L. Jiao, Y. Wang, H.-L. Jiang and Q. Xu, Adv. Mater.,
2018, 30, 1703663; (c) G. Li, S. Zhao, Y. Zhang and Z. Tang,
Adv. Mater., 2018, 30, 1800702; (d) C.-D. Wu and M. Zhao, Adv.
Mater., 2017, 29, 1605446.
4 K. Morgan, A. Goguet and C. Hardacre, ACS Catal., 2015, 5, 3430.
ˇ
5 (a) J. Cejka, R. E. Morris and P. Nachtigall, Zeolites in Catalysis:
Properties and Applications, Royal Society of Chemistry, UK, 2017;
(b) S. Cao, F. Tao, Y. Tang, Y. Li and J. Yu, Chem. Soc. Rev., 2016,
45, 4747; (c) J. Ye, J. Tang, Y.-J. Zhao and C.-D. Wu, Inorg. Chem.,
2020, 59, 767.
6 Y. Liu, Z. Li, Q. Yu, Y. Chen, Z. Chai, G. Zhao, S. Liu, W.-C. Cheong,
Y. Pan, Q. Zhang, L. Gu, L. Zheng, Y. Wang, Y. Lu, D. Wang, C. Chen,
Q. Peng, Y. Liu, L. Liu, J. Chen and Y. Li, J. Am. Chem. Soc., 2019,
141, 9305.
7 W.-R. Yu, G.-H. Lee and E.-C. Yang, Dalton Trans., 2013, 42, 3941.
8 P. Terzieff and K. Lee, J. Appl. Phys., 1979, 50, 3565.
9 Munasir, A. S. Dewanto, D. H. Kusumawati, N. P. Putri,
A. Yulianingsih, I. K. F. Sa’adah, A. Taufiq, N. Hidayat,
S. Sunaryono and Z. A. I. Supardi, IOP Conf. Ser.: Mater. Sci. Eng.,
2018, 367, 012010.
10 X.-H. Guan, G.-H. Chen and C. Shang, J. Environ. Sci., 2007, 19, 438.
PMS-11, respectively. The dihydrogen molecule was subse- 11 A. K. Chaurasia, H. Goyal and P. Mondal, Int. J. Hydrogen Energy,
quently dissociated to form hydrogen (Hꢀ) active species on
2020, 45, 18250.
12 (a) S. Iihama, S. Furukawa and T. Komatsu, ACS Catal., 2016, 6, 742;
the Ni0 site, which could react with the adsorbed nitro group in
(b) F. Zhang, C. Zhao, S. Chen, H. Li, H. Yang and X.-M. Zhang,
halogenated nitrobenzene to form a nitroso compound inter-
mediate. The nitroso compound is highly reactive, which
further reacted with hydrogen (Hꢀ) to produce phenylhydrox-
ylamine, and finally arylamine compound.18,19 The electrone-
gatively charged Co site, as revealed by XPS spectra, could
generate electrostatic repulsion interaction with the electron-
enriched C–X moiety, which would prevent the dehalogenation
reaction to improve the selectivity.15,20,21 The electrophilic GdIII
species could also decrease the electron density of Ni species,
which would improve the binding connectivity between Ni0 and
H2 and subsequent dissociation rate of H2 to improve the
catalytic activity.22
In summary, we developed a practically applicable strategy
for the synthesis of porous metal silicate material PMS-11,
consisting of in situ generated multi-metal sites with control-
lable constituents, structures and properties inside the active
site-adaptive pores. Attributed to the synergistic effect between
Lewis acid GdIII sites and redox-active metallic Ni0 and Co0
species, PMS-11 demonstrates high catalytic activity, selectivity
and stability in hydrogenation of halogenated nitrobenzenes to
produce halogenated anilines with high conversion (499%)
J. Catal., 2017, 348, 212.
13 (a) M. Liu, Q. Bai, H. Xiao, Y. Liu, J. Zhao and W. W. Yu, Chem. Eng.
J., 2013, 232, 89; (b) M. Wu, Y. Xu and J. X. Ju, Nano-Micro Lett., 2016,
11, 315; (c) Q. Wei, Y. S. Shi, K. Q. Sun and B. Q. Xu, Chem. Commun.,
2016, 52, 3026; (d) H. Liu, K. Tao, C. Xiong and S. Zhou, Catal. Sci.
Technol., 2015, 5, 405; (e) P. Zhang, Y. Hu, B. Li, Q. Zhang, C. Zhou,
H. Yu, X. Zhang, L. Chen, B. Eichhorn and S. Zhou, ACS Catal., 2015,
5, 1335; ( f ) J. Li, X. Li, Y. Ding and P. Wu, Chin. J. Catal., 2015,
36, 1995.
14 L. Sho, F. Shinya and K. Tayayuki, ACS Catal., 2016, 6, 742.
15 (a) P. Munnik, P. E. Jongh and K. P. de Jong, J. Am. Chem. Soc., 2014,
136, 7333; (b) F. Li, Appl. Catal., A, 2016, 514, 248; (c) X. Sun,
A. I. Olivos-Suarez, D. Osadchii, M. J. Valero Romero, F. Kapteijn
´
and J. Gascon, J. Catal., 2018, 357, 20; (d) L. Liu, P. Concepcion and
A. Corma, J. Catal., 2016, 340, 1; (e) Y. Cao, K. Liu, C. Wu, H. Zhang
and Q. Zhang, Appl. Catal., A, 2020, 592, 117434; ( f ) Z. Wei, J. Wang,
S. Mao, D. Su, H. Jin, Y. Wang, F. Xu, H. Li and Y. Wang, ACS Catal.,
2015, 5, 4783.
16 (a) Veena L. Khilnani and S. B. Chandalia, Org. Process Res. Dev.,
2011, 5, 257; (b) Z. Hu, S. Tan, R. Mi, X. Li, D. Li and B. Yang,
Catal. Lett., 2018, 148, 1490.
17 S. Iihama, S. Furukawa and T. Komatsu, ACS Catal., 2016, 6,
742.
18 M. Tian, X. L. Cui, M. Yuan, J. Yang, J. T. Ma and Z. P. Dong, Green
Chem., 2017, 19, 1548.
19 K. Shimizu, Y. Miyamoto and A. Satsuma, J. Catal., 2010, 270, 86.
20 B. Zhao and Y.-W. Chen, Mater. Chem. Phys., 2011, 125, 763.
21 L. Huang, Y. Lv and S. H. Liu, Ind. Eng. Chem. Res., 2020, 59, 1422.
and selectivity (499%), and is ready for practical applications. 22 B. L. Ramirez and C. C. Lu, J. Am. Chem. Soc., 2020, 142, 5396.
6188 | Chem. Commun., 2021, 57, 6185–6188
This journal is © The Royal Society of Chemistry 2021