Hydroxyl free radical route to the stable siliceous Ti-UTL with extra-large pores for oxidative desulfurization

Article abstract of DOI:10.1039/c8cc09225g

A facile and mild method was successfully developed to prepare the stable, highly siliceous Ti-UTL zeolite with extra-large pores. The hydroxyl free radical (OH) was introduced for the first time to effectively promote the isomorphous substitution of Si for framework Ge under room temperature and neutral conditions, resulting in a stable titanosilicate with an outstanding catalytic activity for oxidative desulfurization.

Full text of DOI:10.1039/c8cc09225g

ChemComm
View Article Online
View Journal
COMMUNICATION
Hydroxyl free radical route to the stable siliceous
Ti-UTL with extra-large pores for oxidative
desulfurization†
Cite this:DOI: 10.1039/c8cc09225g
Received 20th November 2018,
Accepted 28th December 2018
Dier Shi, Le Xu, * Pohua Chen, Tianqiong Ma, Cong Lin, Xiaoge Wang,
Diming Xu
and Junliang Sun
*
DOI: 10.1039/c8cc09225g
rsc.li/chemcomm
A facile and mild method was successfully developed to prepare the Ge under acidic conditions.9 Later, another approach was
stable, highly siliceous Ti-UTL zeolite with extra-large pores. The developed by low-temperature plasma treatment with TiCl4
hydroxyl free radical (ꢀOH) was introduced for the first time to and SiCl4 precursors.10 Wu and Tuel reported a treatment
effectively promote the isomorphous substitution of Si for frame- method of Ge-rich materials under acidic and high-temperature
work Ge under room temperature and neutral conditions, resulting conditions.11,12 Although the employment of harsh conditions
in a stable titanosilicate with an outstanding catalytic activity for or complex apparatus is required to ensure the efficient removal
oxidative desulfurization.
of Ge from zeolite frameworks, it also results in safety and
environmental concerns. Thus, a facile and mild post-treatment
Zeolites with crystalline inorganic frameworks, unique channel method that can accomplish the stabilization of Ge-rich zeolites
systems, and large surface areas have been widely used as hetero- under room temperature and neutral conditions is the goal for
geneous catalysts in industrial processes.1,2 With the increasing practical applications.
ꢀ
demand for processing bulky reactants, there is much interest in
Herein, hydroxyl free radicals ( OH) were introduced for the
synthesizing stable zeolites with extra-large pores (Z14-ring), for first time to assist the stabilization of the Ge-rich, Ti-containing
faster diffusion of reactants and thus overcoming the diffusion zeolite under mild conditions, leading to a highly siliceous
restriction problem in catalytic processes.3 In the past two titanosilicate with high hydrothermal stability. An interesting
decades, many new zeolites with extra-large pores have been dis- example is the germanosilicate UTL, the first extra-large pore
covered with germanium.4–6 This is because Ge atoms facilitate zeolite that is constructed with a high Ge content. Furthermore,
the construction of specific small building units, such as double the UTL topology with a two-dimensional channel system
4-ring (d4r) units, which are essential for the construction of (14 Â 12-ring) offers a unique opportunity for catalytic reactions
many zeolite frameworks with extra-large pores.7 Although the involving bulky molecules.13
incorporation of Ge leads to lots of new zeolites with extra-large
The parent calcined Ti-UTL is a Ge-rich zeolite (Si/Ge = 3.9,
pores aiming for the catalytic upgrading of bulky substrates, Si/Ti = 105). Its typical interlayer (200) appears at 6.11. The
structural degradation takes place readily due to the low stability Ge-rich UTL zeolite can be treated as two-dimensional silicate
of the Ge–O bonds towards moisture, resulting in a fatal obstacle layers pillared by the Ge-rich d4r bridging units, and it can be
to applications.
removed under acidic conditions according to a previous
Stabilization of germanosilicate can be realized by direct study.14,15 The shift of the (200) peak to 7.81 upon acid treat-
synthesis of zeolites with high Si/Ge ratio. However, even with ment and a further change to 9.31 after calcination confirms
carefully designed organic structure-directing agents, only a few the reduction of the interlayer distance (Fig. S1, ESI†). Despite
germanosilicates with reduced Ge content could be synthe- the relatively low Ge content (Si/Ge = 36.9) from the ‘‘assembly-
sized.8 Thus, the post-treatment of Ge-rich germanosilicate to disassembly-organization-reassembly’’ processed material, the
the stable Ge-poor analogue is an alternative option. Valtchev largest opening pore shrinks into a medium pore due to the
reported an approach that led to Ge-rich BEC zeolite stabili- removal of d4r units.14,16,17
zation by the isomorphous substitution of Al for framework
The stabilization of the Ge-rich zeolite is dynamic upon
chemical post-treatment, involving the breaking of framework
Ge–O bonds, the hydrolysis of the added silica source, and the
reconstruction of stable Si–O bonds in the interlayer region. In
College of Chemistry and Molecular Engineering, Peking University, Beijing, 100871,
China. E-mail: junliang.sun@pku.edu.cn, lexu@pku.edu.cn
ꢀ
this study, the latter two procedures were of more concern. OH
† Electronic supplementary information (ESI) available: Experimental details,
Fig. S1–S14 and Table S1. See DOI: 10.1039/c8cc09225g
was used because it was involved in the zeolite hydrothermal
This journal is ©The Royal Society of Chemistry 2019
Chem. Commun.

View Article Online
Communication
ChemComm
reduction of the TEOS/GeO2 molar ratios, 1 or 2, due to the
(200) peak becoming broader and partially shifted to 7.01 and
7.51 in stability tests, respectively. This indicates that the
defects caused by the removal of Ge atoms could not be readily
repaired by Si atoms and therefore the interlayer structure
partially collapsed during the stabilization test.
For comparison, the controlled treatments were carried out
under dark conditions. However, the UTL structure could only
be stabilized with a TEOS/Ge molar ratio as high as 10 (Fig. S3,
ESI†), showing that the UV irradiation promoted the stabilization
process and led to a large reduced amount of TEOS. Thus, it is
Scheme 1 Strategy for the stabilization of Ge-rich zeolites by the iso-
morphous substitution of Ge (green) with Si (yellow) under neutral con-
ditions and room temperature. The Si sources come from the tetraethyl
orthosilicate (TEOS).
ꢀ
reasonable that OH may be the key factor in the stabilization
ꢀ
process under UV conditions due to the fact that OH could be
generated by water or the homolytic dissociation of the O–O bond
of H2O2. A certain amount of H2O2 solution was added into the
treatment under dark conditions. The results indicate that
the parent Ge-rich, Ti-UTL was also well stabilized with a low
TEOS/Ge molar ratio of 2.5 (Fig. S4, ESI†), resulting in a highly
siliceous structure with a Si/Ge ratio of 24.9.
crystallization and has a positive effect in the zeolite nucleation
stage in terms of promoting the formation of an amorphous
gel, the breaking of Si–O–Si bonds to form soluble species, and
the formation of new Si–O–Si bonds.18 We therefore hypothe-
ꢀ
sized that the involvement of OH can accelerate the breaking
SEM images illustrated that the morphology of Ti-UTL was
well preserved after stabilization (Fig. S5, ESI†). The Le Bail fitting
of the PXRD patterns indicated that the stabilized samples had
comparable unit cell parameters b and c to the parent sample,
while the unit cell parameter a shortened slightly (Fig. S6, ESI†).
Since the Ge-rich d4r bridging units were distributed only along
the a-direction in the UTL structure, the substitution of Si by Ge
led to a shorter a unit cell parameter while the b and c parameters
were hardly affected due to the low concentration of Ge in the
layer. The signals of the solid-state 29Si MAS NMR spectra
illustrate different environments of framework Si atoms. For
the stabilized Ti-UTL, the signals at À100.5 to À106.5 ppm,
corresponding to the (SiO)nSi(GeO)4Àn (n = 2 to 4) coordination,
significantly decreased, confirming the substitution of Si for
framework Ge in the d4r units (Fig. S7, ESI†).19 Additionally, the
band around 575 cmÀ1 is related to the vibration of d4r units
in Ti-UTL.20,21 This band still existed but with a blue shift to
596 cmÀ1 after the stabilization treatment (Fig. S8, ESI†). The
blue shift of 21 cmÀ1 is in agreement with the theoretical calcu-
lation by replacing Ge atoms with Si atoms in the d4r units.11
The PXRD patterns based on the UTL structure with the different
substitution degree of Ge in the d4r units were simulated
(Fig. S9, ESI†). As expected, our experimental results were coincident
with the simulation results (Fig. S10, ESI†).
of Si–O–Si bonds from the added silica source via hydrolysis,
and promote the substitution of Si for framework Ge, hence
leading to a germanosilicate zeolite with enhanced stability
under mild conditions (Scheme 1). We carried out experiments
ꢀ
in three different conditions to confirm the positive role of OH
in obtaining a more stable highly siliceous Ti-UTL zeolite. These
ꢀ
three different OH-related conditions are ultraviolet (UV) irradia-
tion conditions, dark conditions (without UV irradiation) and
H2O2 conditions (dark conditions with added H2O2).
Under UV conditions, the Ge-rich UTL structure can be
stabilized in the presence of tetraethyl orthosilicate (TEOS) when
the molar ratio of TEOS/GeO2 (GeO2 contains in the parent
Ti-UTL) exceeds 2.5 and the UV irradiation density exceeds
30 mW cmÀ2 (Fig. 1A and Fig. S2, ESI†). The (200) peak of the
UTL phase at 6.21 was rarely changed after stabilization, indicat-
ing the successful stabilization of the zeolite framework without
degradation. The Si/Ge ratio of the stabilized Ti-UTL increased to
30.1 when the Ti content almost kept constant. The stabilized
Ti-UTL was extremely stable in the stability test, see Fig. 1B
and C. In contrast, the UTL structure is less stable with a
The surface area and porosity of the parent and stabilized
samples were analyzed by N2 adsorption, demonstrating that
the porosity of Ti-UTL remained intact after stabilization
(Fig. S11 and Table S1, ESI†). The UV-visible spectra of the
stabilized Ti-UTL showed an intensive absorption band cen-
tered around 215 nm (Fig. S12, ESI†), which resulted from the
tetrahedral-coordinated framework Ti species.22 The IR spectra
under vacuum conditions were collected and all three
Ti-UTL showed a typical band around 960 cmÀ1 attributed to
the stretching vibration of framework Si–O–Ti species (Fig. S8,
ESI†),23 while the same band was absent for Ti-free UTL, which
confirms that tetrahedrally coordinated Ti species were retained
after stabilization.
Fig. 1 PXRD patterns of (A) the parent sample and treated samples
obtained under UV conditions: (a) parent Ti-UTL and stabilization with a
TEOS/GeO2 molar ratio of (b) 1, (c) 2, (d) 2.5, (e) 5 and (f) 10. The PXRD
patterns of the parent and treated samples in stability tests upon (B) acid
treatment and (C) calcination.
Chem. Commun.
This journal is ©The Royal Society of Chemistry 2019

View Article Online
ChemComm
Communication
ratio were observed and the hyperfine coupling constants of the
nitrogen and b-proton were aN = aHb = 14.9 G, and g = 2.0020
(Fig. 2a). These numbers were regarded as typical evidence of
25
ꢀ
the DMPO– OH adduct according to the previous study and
matched well with the simulated EPR spectra. The six-line
spectrum with aN = 15.8 G, aHb = 24.5 G, and g = 2.0017, implies
the existence of alkane radicals (Fig. 2a), which were generated
by the ethanol from TEOS hydrolysis under UV irradiation
and then trapped by DMPO.26,27 Under H2O2 conditions, the
ꢀ
EPR data showed similar signals of both DMPO– OH and
ꢀ
DMPO– OCH2CH3 adducts (Fig. 2b). In contrast, no signal
was observed under dark conditions (Fig. 2c), proving that the
ꢀ
OH can assist the stabilization of the unstable zeolite structure
for the first time.
As discussed before, the germanosilicate stabilization is a
three-step process. In the three experimental conditions, nearly
75% of Ge atoms (located in d4rs) were easily removed by
hydrolysis in the first ten minutes during the treatment. How-
ever, the hydrolysis of TEOS and the reconstruction of the new
framework Si–O bonds were much slower compared to the
cleavage of the framework Ge–O bonds especially under the
room temperature and neutral conditions. Hence, the latter two
procedures had key roles in the stabilization process. For one
ꢀ
thing, the hydrolysis of TEOS can be promoted by the OH
radicals without additional catalyst.28 For another, the presence
ꢀ
of OH can remarkably accelerate the rate of Si–O–Si bond
remaking and polymerization of the silica species in the zeolite
nucleation process.18 Hence, we deduce that the OH promotes
Fig. 2 EPR spectra of the mixture containing DMPO under (a) UV condi-
tions, (b) H2O2 conditions, and (c) dark conditions. The EPR signals were
marked as follows: blue rectangles, DMPO–ꢀOH adduct; green circles,
DMPO–ꢀOCH2CH3 adduct. The red lines are simulated EPR spectra of the
DMPO–ꢀOH adduct.
ꢀ
the whole stabilization process, especially the last two steps, and
thus successfully modulates these multiple dynamic equilibria
under mild conditions.
ꢀ
To further confirm the role of OH radicals in the stabili-
zation process, controlled experiments were conducted by
ꢀ
To verify the presence of the OH radical, electron para- adding propanetriol as the radical scavenger during stabili-
magnetic resonance (EPR) spectroscopy was used to study the zation. Under both the UV and the H2O2 conditions, the (200)
stabilization process under all three conditions with 5,5-dimethyl-1- peak of the samples became broad with a shoulder peak at 6.91,
pyrroline-N-oxide (DMPO) as the free-radical trap.24 Under UV referring to partially collapsed structures (Fig. S13, ESI†). Therefore,
ꢀ
conditions, four splitting EPR signals with a 1 : 2 : 2 : 1 intensity the result proved that the introduction of OH by either UV
Table 1 Oxidative desulfurization of DBT and 4,6-DMDBT over various titanosilicatesa
Conv. (%)
Conv. (%)
Yieldg (%)
Si/Geb Si/Tib DBTc 4,6-DMDBTc TBHPDBT TBHP4.6-DMDBT TONe TONf DBTSO DBTSO2 DMDBTSO DMDBTSO2
d
d
Samples
Ti-UTL(UV)h
30.1
119.2 94.3 98.7
125.5 92.9 99.3
37.4 27.8 26.1
72.8
71.5
25.4
9.6
76.9
77.6
23.9
8.4
19.3
20.3
1.8
17.4 18.4
18.6 17.6
1.4 23.1
75.9
75.1
4.7
18.1
17.8
23.9
8.6
80.6
81.5
2.2
Ti-UTL(H2O2)h 24.9
TS-1
UTL
—
19.5
—
10.6
9.4
—
—
9.2
1.4
0.8
a
b
Reaction conditions: 5 mL 1000 mg mLÀ1 model fuels, 60 1C, 10 mg catalysts, 30 mL 65% TBHP. Measured by inductively coupled plasma (ICP).
c
d
e
After 2 hours of the catalytic reaction. Converision of TBHP in the catalytic reaction after 2 hours. Turnover number (moles of converted DBT/
f
g
mole of Ti sites). Turnover number (moles of converted DMDBT/mole of Ti sites). Product yield based on the DBT or DMDBT after 2 hours of the
catalytic reaction. The stabilized Ti-UTL with TEOS/GeO2 molar ratio of 2.5 under UV conditions or H2O2 conditions.
h
This journal is ©The Royal Society of Chemistry 2019
Chem. Commun.

View Article Online
Communication
ChemComm
irradiation or H2O2 plays a positive role in forming a more
This work was supported by the National Natural Science
stable Ge-rich sample, which was hard to stabilize under dark Foundation of China (No. 21621061, 21527803, 21471009,
conditions.
21871009) and the Ministry of Science and Technology of China
The oxidative desulfurization of dibenzothiophene (DBT) (No. 2016YFA0301004).
and dimethyldibenzothiophene (DMDBT) with tert-butyl hydro-
peroxide as an oxidant was used to characterize the catalytic
activity of the samples. A Ti-free UTL was synthesized following
the same route of Ti-UTL for the catalytic analysis and had a
negligible catalytic activity in oxidative desulfurization of DBT
(10.6%) and DMDBT (9.4%), implying that the residual Ge
atoms in the framework are inactive for catalytic activity
(Table 1). Although the stabilized Ti-UTL had relatively low
titanium contents, Si/Ti = 119.2 and 125.5, they showed a
superior catalytic behavior compared to the typical TS-1 catalyst
(Table 1 and Fig. S14, ESI†). This phenomenon is mainly
related to the multiple extra-large channel system of the UTL
structure. Owing to the different contents of Ti active sites in
the catalysts, the catalytic turnover number (TON) was calcu-
lated to compare their catalytic behavior. The stabilized Ti-UTL
afforded a TON of DBT of around 20 (Table 1), much higher
than that of TS-1 (1.8) (Table 1). Meanwhile, similar results were
observed in the oxidation of the bulky DMDBT (Table 1 and
Conflicts of interest
There are no conflicts to declare.
Notes and references
1 A. Corma, J. Catal., 2003, 216, 298–312.
2 M. E. Davis, Nature, 2002, 417, 813–821.
3 J. Jiang, J. Yu and A. Corma, Angew. Chem., Int. Ed., 2010, 49, 3120–3145.
4 J. L. Paillaud, B. Harbuzaru, J. Patarin and N. Bats, Science, 2004,
304, 990–992.
5 J. Sun, C. Bonneau, A. Cantin, A. Corma, M. J. Diaz-Cabanas,
M. Moliner, D. Zhang, M. Li and X. Zou, Nature, 2009, 458, 1154–1157.
6 J. Jiang, J. L. Jorda, M. J. Diaz-Cabanas, J. Yu and A. Corma, Angew.
Chem., Int. Ed., 2010, 49, 4986–4988.
´
˜
7 T. Blasco, A. Corma, M. J. Dıaz-Cabanas, F. Rey, J. A. Vidal-Moya and
C. M. Zicovich-Wilson, J. Phys. Chem. B, 2002, 106, 2634–2642.
8 A. Rojas and M. A. Camblor, Angew. Chem., Int. Ed., 2012, 51,
3854–3856.
9 F. Gao, M. Jaber, K. Bozhilov, A. Vicente, C. Fernandez and V. Valtchev,
J. Am. Chem. Soc., 2009, 131, 16580–16586.
Fig. S15, ESI†). The catalytic performance achieved on the 10 M. El-Roz, L. Lakiss, A. Vicente, K. N. Bozhilov, F. Thibault-Starzyk
and V. Valtchev, Chem. Sci., 2014, 5, 68–80.
11 H. Xu, J.-g. Jiang, B. Yang, L. Zhang, M. He and P. Wu, Angew. Chem.,
stabilized Ti-UTL is comparable to that of the Ti-UTL catalysts
reported before.29,30 The stabilized Ti-UTL showed a slightly
Int. Ed., 2013, 53, 1355–1359.
lower conversion of DBT than that of Ti-Beta and Ti-MWW. 12 L. Burel, N. Kasian and A. Tuel, Angew. Chem., Int. Ed., 2014, 53,
1360–1363.
However, for the large DMDBT substrate, the stabilized Ti-UTL
had a better performance than Ti-Beta (Table S2, ESI†). The
13 A. Corma, M. J. Diaz-Cabanas, F. Rey, S. Nicolopoulus and K. Boulahya,
Chem. Commun., 2004, 1356–1357.
crystal structure and the morphology of the stable Ti-UTL were 14 P. Eliasova, M. Opanasenko, P. S. Wheatley, M. Shamzhy, M. Mazur,
P. Nachtigall, W. J. Roth, R. E. Morris and J. Cejka, Chem. Soc. Rev.,
2015, 44, 7177–7206.
15 W. J. Roth, O. V. Shvets, M. Shamzhy, P. Chlubna, M. Kubu, P. Nachtigall
well preserved after six cycles in the oxidative desulfurization,
proved by PXRD patterns, SEM images, UV-vis spectra and N2
adsorption isotherms (Fig. S16–S18, ESI†). Although the con-
tent of Ti active sites in the sixth recycled catalyst decreased
slightly, the catalyst behavior still remained 93% of its initial
and J. Cejka, J. Am. Chem. Soc., 2011, 133, 6130–6133.
16 W. J. Roth, P. Nachtigall, R. E. Morris, P. S. Wheatley, V. R. Seymour,
S. E. Ashbrook, P. Chlubna, L. Grajciar, M. Polozij, A. Zukal, O. Shvets
and J. Cejka, Nat. Chem., 2013, 5, 628–633.
activity (Fig. S19 and Table S3, ESI†). The untreated Ti-UTL 17 L. Xu and J. Sun, Adv. Energy Mater., 2016, 6, 1600441.
18 G. Feng, P. Cheng, W. Yan, M. Boronat, X. Li, J. H. Su, J. Wang, Y. Li,
sample exhibited inferior catalytic performance in oxidative
A. Corma, R. Xu and J. Yu, Science, 2016, 351, 1188–1191.
desulfurization of DBT compared with the stabilized ones treated
19 N. Kasian, A. Tuel, E. Verheyen, C. E. A. Kirschhock, F. Taulelle and
by UV and H2O2 (Table S4, ESI†). Its structure deteriorated after
the catalytic reaction (Fig. S20, ESI†).
In summary, a facile post-synthesis strategy in obtaining a
stable Ti-UTL zeolite was achieved in room temperature and
J. A. Martens, Chem. Mater., 2014, 26, 5556–5565.
20 C. Blackwell, J. Phys. Chem., 1979, 83, 3257–3261.
21 A. J. M. de Man and R. A. van Santen, Zeolites, 1992, 12, 269–279.
¨
22 P. Ratnasamy, D. Srinivas and H. Knozinger, Advances in Catalysis,
Academic Press, 2004, vol. 48, pp. 1–169.
23 G. Ricchiardi, A. Damin, S. Bordiga, C. Lamberti, G. Spano, F. Rivetti
`
ꢀ
neutral conditions by introducing OH to accelerate the break-
and A. Zecchina, J. Am. Chem. Soc., 2001, 123, 11409–11419.
24 V. Roubaud, S. Sankarapandi, P. Kuppusamy, P. Tordo and J. L.
Zweier, Anal. Biochem., 1997, 247, 404–411.
25 F. A. Villamena, E. J. Locigno, A. Rockenbauer, C. M. Hadad and
J. L. Zweier, J. Phys. Chem. A, 2006, 110, 13253–13258.
26 P. Berczynski, I. Kruk, T. Piechowska, M. Ceylan-Unlusoy,
O. Bozdag-Dundar and H. Y. Aboul-Enein, Luminescence, 2013, 28,
900–904.
ing of Si–O–Si bonds from the added silica source and promote
the substitution of Si for framework Ge. The stabilized Ti-UTL
zeolite with extra-large pores showed an outstanding catalytic
activity for the oxidative desulfurization of bulky organic mole-
cules. This method opens up new ideas for promoting the
practical applications of germanosilicates with low thermal/
hydrothermal stability. A future study will make better use of
27 G. Feng, J. Wang, M. Boronat, Y. Li, J. H. Su, J. Huang, Y. Ma and
J. Yu, J. Am. Chem. Soc., 2018, 140, 4770–4773.
this method to stabilize more types of germanosilicates, such as 28 S. Okusaki and T. Ohishi, J. Non-Cryst. Solids, 2003, 319, 311–313.
ˇ
ˇ
29 J. Prech, R. E. Morris and J. Cejka, Catal. Sci. Technol., 2016, 6,
IWR, and CTH, which cannot withstand the high-temperature
combustion of organic structure directing agents or subsequent
exposure to a moist environment.
2775–2786.
30 X. Liu, L. Zhang, H. Xu, J. Jiang, M. Peng and P. Wu, Appl. Catal., A,
2018, 550, 11–19.
Chem. Commun.
This journal is ©The Royal Society of Chemistry 2019