Exploring the catalytic performance of a series of bimetallic MIL-100(Fe, Ni) MOFs

Article abstract of DOI:10.1039/c9ta01948k

A series of mixed-metal Fe^III/Ni^II metal-organic frameworks (MOFs) of the MIL-100 type containing different metal ratios have been synthesized de novo, following an approach that requires tuning of the Fe^III/Ni^II reactivity. The resulting heterometallic MIL-100(Fe, Ni) materials maintain thermal, chemical and structural stability with respect to the parent MIL-100(Fe) MOF as can be deduced from various techniques. The nature and the oxidation state of the accessible metal cations have been evaluated by in situ infrared spectroscopy and extended X-ray absorption fine structure measurements. The obtained mixed-metal MOFs and the parent material have been evaluated as heterogeneous catalysts in a model acid-catalyzed reaction, i.e., the Prins reaction. It is found that the catalytic activity improves by more than one order of magnitude upon incorporation of Ni^II, with a complete selectivity for the formation of nopol. This increase in the catalytic activity upon incorporation of Ni^II correlates with the enhancement in the Lewis acidity of the material as determined by CO adsorption. The heterometallic MOF can be recycled without observation of metal leaching, while maintaining the crystal structure under the reaction conditions.

Full text of DOI:10.1039/c9ta01948k

Journal of
Materials Chemistry A
View Article Online
View Journal
PAPER
Exploring the catalytic performance of a series of
bimetallic MIL-100(Fe, Ni) MOFs†
Cite this: DOI: 10.1039/c9ta01948k
Monica Gimenez-Marques, *ab Andrea Santiago-Portillo,
c
c
Sergio Navalon,
´
´
´
´
c
d
a
c
Mercedes A´ lvaro,
Valerie Briois,
*
Farid Nouar, Hermenegildo Garcia
*
´
a
and Christian Serre
III
II
A series of mixed-metal Fe /Ni metal–organic frameworks (MOFs) of the MIL-100 type containing different
III
II
metal ratios have been synthesized de novo, following an approach that requires tuning of the Fe /Ni
reactivity. The resulting heterometallic MIL-100(Fe, Ni) materials maintain thermal, chemical and structural
stability with respect to the parent MIL-100(Fe) MOF as can be deduced from various techniques. The
nature and the oxidation state of the accessible metal cations have been evaluated by in situ infrared
spectroscopy and extended X-ray absorption fine structure measurements. The obtained mixed-metal
MOFs and the parent material have been evaluated as heterogeneous catalysts in a model acid-catalyzed
reaction, i.e., the Prins reaction. It is found that the catalytic activity improves by more than one order of
Received 20th February 2019
Accepted 6th August 2019
II
magnitude upon incorporation of Ni , with a complete selectivity for the formation of nopol. This increase
II
in the catalytic activity upon incorporation of Ni correlates with the enhancement in the Lewis acidity of
DOI: 10.1039/c9ta01948k
the material as determined by CO adsorption. The heterometallic MOF can be recycled without
observation of metal leaching, while maintaining the crystal structure under the reaction conditions.
rsc.li/materials-a
the new multimetallic MOFs present unsatisfactory chemical
stability under catalytic conditions, an issue that has to be
Introduction
The development of multifunctional solid catalysts is currently of addressed before MOF catalysts can meet industrial applica-
foremost relevance in the eld of heterogeneous catalysis. The bility. Direct post-synthetic methods (PSMs) provide a conve-
reason is that it is expected that bi/multi-functionalization can be nient strategy for the preparation of mixed-metal MOFs, which
17,18
used to promote cascade or tandem reactions in one single step, has been widely explored recently.
This PSM approach has
thus noticeably decreasing processing costs. In this context, been successfully applied to partially replace in MOFs some of
metal–organic frameworks (MOFs) can be leading materials, given the cations by ion metathesis, or to directly coordinate metal
19–21
their chemically modulable nature and structural diversity, and complexes in the framework.
their already proved suitable performance as solid heterogeneous
catalysts for specic liquid phase chemical transformations.
A more recently developed strategy consists in the direct
synthesis of bi- or multi-metallic MOFs using a straightforward
1–5
Multifunctional MOFs presenting different ligands and/or cations method with pre-designed mixed-metal subunits as starting
6–9
have already been used in catalysis, some of these materials materials. This approach has been denoted as the secondary
1
0–14
22
being able to promote one-pot tandem reactions.
The real challenge in the synthesis of stable mixed-metal successfully used in the synthesis of porous pure cationic iron(III)
building unit (SBU) approach. These SBU methods have been
23
MOFs consists in the control of the metal stoichiometry dicarboxylate (MIL-88s) (MIL stands for Materials of Institut
particularly for cations with different oxidation states and Lavoisier) or zirconium(IV) dicarboxylate MOFs (UiO-66's) (UiO
24
15,16
therefore distinct chemical reactivities.
In addition, most of stands for Universitetet i Oslo) and have been recently applied for
the preparation of a mesoporous mixed-metal MIL-127 MOF or
25
soc-MOF, as well as for the preparation of different iron-
containing robust MOFs with a preserved Fe M O metal
cluster core (M ¼ Co , Ni , or Mg ).
considered the use of mixed-metal acetate building blocks, i.e. the
2 3
trimeric oxocentered bimetallic Fe /M acetate system [Fe M(m -
a
´
III
2
II
Institut des Mat ´e riaux Poreux de Paris, UMR 8004 CNRS, Ecole Normale Sup ´e rieure,
E´ cole Sup ´e rieure de Physique et de Chimie Industrielles de la ville de Paris, PSL
II
2+
2+
2+ 26,27
These latter examples
University, 75005 Paris, France
b
Instituto de Ciencia Molecular, Universidad de Valencia, Catedr a´ tico Jos ´e Beltr a´ n 2,
III
II
4
6980 Paterna, Spain. E-mail: monica.gimenez-marques@uv.es
2+
2+
2+
2+
28
c
O)(CH
COO)
] (M ¼ Co , Ni , Mn , Zn .) obtained from the
Departamento de Qu ´ı mica e Instituto de Tecnolog ´ı a Qu ´ı mica (ITQ-CSIC)-Universitat
Polit `e cnica de Val `e ncia, C/Camino de Vera, s/n, 46022, Valencia, Spain
3
6
29
pure iron(III) oxo acetate cluster. In this context, we have devoted
d
Synchrotron SOLEIL-UR1, 91192, Gif-Sur-Yvette, France
particular efforts to the preparation of the archetypal mesoporous
†
Electronic supplementary information (ESI) available. See DOI:
0.1039/c9ta01948k
mixed-metal MIL-100(Fe
2
M) material using the pre-dened
1
This journal is © The Royal Society of Chemistry 2019
J. Mater. Chem. A

View Article Online
Journal of Materials Chemistry A
Paper
II
3
mixed-metal acetate building blocks. The main reason to select aqueous solutions of HNO and Ni salt. Similarly, pure MIL-
III
0
MIL-100(Fe M) is the ability of its parent MIL-100(Fe) solid to 100(Fe ) was obtained using Fe as the only metal precursor. All
2
perform as a heterogeneous catalyst in a large variety of reactions, materials were obtained in stoichiometric metal-to-ligand
leading in many cases to large conversions and selectivity. The amounts, using water as the solvent and stirring for 24 h at
ꢀ
advantage in obtaining this particular MOF lies in its valuable relatively low temperatures (80 C), avoiding in this case the
33
III
features: (i) MIL-100(Fe) exhibits a very high surface area and pore addition of HF. Under these conditions, the Fe cations are
volume, allowing easy diffusion of substrates and products of very slowly released in the reaction media, permitting to some
molecular dimensions smaller than the pore windows; (ii) it can extent the incorporation into the framework of the less favoured
II
be synthesized on a large scale following green methodologies acidic Ni cation that is 100% available since the early stage of
III
II
and at low temperature; and (iii) its robust porous structure the reaction, in large Ni to Fe excess. Three different Fe /Ni
possesses a suitable chemical and thermal stability. Following the metal ratios were considered, obtaining the bimetallic MIL-
rational mixed-metal building block approach, we could obtain 100(Fe, Ni) (1–3) MOF materials whose analytical data are
MIL-100(Fe
2
Ni) with an appropriate crystallinity, although with summarized in Table 1.
Further attempts to obtain mixed-metal MIL-100 including
several drawbacks: (i) the formation of undesired iron oxides
during the synthesis, which considerably affect the properties of post-synthetic ion metathesis of MIL-100(Fe) by soaking the
II
the material with regard to its use as a catalyst; and (ii) a dramatic crystalline material in concentrated aqueous Ni salt solutions,
III
reduction of chemical stability with respect to the parent robust as well as direct mixture of the corresponding inorganic Fe
II
MIL-100(Fe). These inconveniences prevented the practical and Ni salts with trimesic acid, were carried out. In all cases,
application of the obtained MIL-100(Fe Ni) material and the single metal MIL-100(Fe) material was obtained regardless
2
II
prompted us to explore alternative synthetic routes leading to the of the M concentration and the immersion time. This failure to
incorporation of low Ni content in heterometallic MIL-100(Fe, Ni) obtain the mixed-metal MIL-100(Fe, Ni) under these conditions
III
materials. So far, to our knowledge, only mixed metal(III) MIL-100 is likely due to the preferential interaction of Fe hard cations
12
30
II
(
Sc combined with Al, Fe, or Cr) materials and MIL-100(Fe, Mn)
with carboxylates with respect to so cations (i.e. Ni ).
have been obtained.
The obtained heterometallic MIL-100(Fe, Ni) MOFs 1–3 and
Herein we present the direct synthesis of a series of mixed- the corresponding homometallic MIL-100(Fe) were character-
III
II
metal MIL-100(Fe , Ni ) MOFs containing different propor- ized by powder X-ray diffraction (PXRD), Fourier transform
II
tions of Ni . The method follows a synthetic route previously infrared spectroscopy (FT-IR), thermogravimetric analysis
31
developed by some of us, in which the preparation of a mixed- (TGA), inductively coupled plasma-atomic emission spectros-
III
III
2
metal MIL-53(Cr , Fe ) required the adjustment of the metal copy (ICP-AES), N sorption measurements, scanning electron
III
reactivity to permit the more inert Cr cation to enter into the microscopy (SEM) and extended X-ray absorption ne structure
framework. The mixed-metal MIL-100(Fe, Ni) MOFs obtained (EXAFS). As seen by PXRD, all powder diffraction patterns of the
here are compared with their parent MIL-100(Fe) materials as as-synthesized MIL-100(Fe) and mixed-metal MIL-100(Fe, Ni)
heterogeneous catalysts in the Prins reaction between b-pinene (1–3) samples match that of the simulated MIL-100(Fe) mate-
32
and paraformaldehyde.
rial, conrming the formation of an isostructural single-phase
MIL-100 material (Fig. 1).
The metal content of the mixed-metal MOFs was deter-
mined by ICP-OES analysis and revealed increasing amounts
of Ni in the range of 1 to 5%, respectively, for samples 1–3.
These metal compositions differ from the initial metal stoi-
Results and discussion
Synthesis and characterization of mixed-metal MIL-100(Fe,
Ni) MOFs
III
II
The synthesis of a series of heterometallic MIL-100(Fe , Ni ) chiometry used under the synthetic conditions (see the
(
1–3) MOFs with three different Fe/Ni ratios was carried out by Experimental section), evidencing the difficulty in incorpo-
directly mixing Fe with trimesic acid in the presence of rating a divalent cation into the metal nodes of MIL-100. This
0
Table 1 Summary of the Ni content, specific surface area and metal composition of the obtained heterometallic MOFs
%
TGA based
IIa
2
ꢁ1
%
Ni
a
BET (m g
)
Inorganic
Organic
ꢁ
1
MIL-100(Fe), M.W. 659.77 g mol
Fe O(NO )[(CO (C )]
MIL-100(Fe Ni), M.W. 597.77 g mol
(C )]
,
Theoretical
Experimental
Theoretical
—
1600
37.2
36.7
30.7
62.8
63.3
69.3
3
3
2
)
3
6
H
3
2
ꢁ1
2
,
Fe
1
2
NiO[(CO
2
)
3
6 3 2
H
Experimental
1.0 ꢂ 0.3
3.1 ꢂ 0.2
5.0 ꢂ 0.2
1545
1525
1570
33.5
33.7
34.1
66.5
66.3
65.9
2
3
a
ICP based. Standard deviation based on 3 independent syntheses. Normalisation of the samples was performed by considering 100% weight at
00 C corresponding to the dried samples.
ꢀ
2
J. Mater. Chem. A
This journal is © The Royal Society of Chemistry 2019

View Article Online
Paper
Journal of Materials Chemistry A
Fig. 2 (a) N
2
sorption isotherms for MIL-100(Fe) and heterometallic
Fig. 1 (a) Structure of the heterometallic MIL-100(Fe, Ni) and trimeric
heterometallic unit. Iron and nickel octahedra, and carbon and oxygen
atoms are in orange, blue, grey and red, respectively. (b) Powder X-ray
diffraction patterns of mixed-metal MIL-100(Fe, Ni) 1–3 materials as
compared with the bare MIL-100(Fe) material.
MIL-100(Fe, Ni) 1–3 materials at 77 K. (b) Chemisorption of CO in MIL-
00(Fe) and MIL-100(Fe, Ni) 3 at 303 K. Solids were activated at 423 K
1
for 12 h under vacuum. Chemisorption is presented as a relative value
of the total sorption isotherm.
the porosity (see the logarithmic scale of the pressure range in
Fig. S3†). The obtained accessible surface area values calculated
using the Brunauer–Emmett–Teller (BET) model are summa-
rized in Table 1. The minor differences encountered in BET
fact can be attributed to the structural instability generated
in the MIL-100(Fe) framework when the larger divalent Ni
cation is incorporated leading to a neutral trimeric oxo-
cluster unit (Fig. 1). In an effort to increase the percentage
of incorporated divalent Ni, a synthetic route based on the
SBU approach was followed using the preformed mixed-
II
2
ꢁ1
areas (1600–1521 m g ) are in the range of possible experi-
mental error in mass calculation. It is important to mention
that a substantial drop in the BET surface area from 1600 to
metal Fe
2 3 3 6
Ni(m -O) (CH COO) oxo-acetate trimer as the
2
ꢁ1
II
1
180 m g is detected upon further insertion of the Ni cation
2
4
metal source, as previously reported by some of us. MIL-
(10–15% metal substitution) (Fig. S7 and S8†). This decrease in
III
II
1
00(Fe , Ni ) containing 33% Ni was synthesized, leading to
surface area may be related to the presence of occluded metal
oxides, which is supported with a larger inorganic percentage
detected in thermogravimetric studies.
CO adsorption experiments were performed to evaluate the
effect of the nature and proportion of the different Lewis acid
sites present in the framework as compared to the single
cationic material.
the formation of an amorphous gel-like material with a much
2
ꢁ1
lower specic surface area (SBET ¼ 700 m
g
). Similar
results have been recently obtained in a series of acetate-
based MOFs using in this case the acetate of the trimeric
Cr (m -O) building block and biphenyl-4,4-dicarboxylic acid
3
3
3
4
(
H
2
-BPDC).
Thermogravimetric analysis performed on the solids
Heterometallic material 3 was selected for these CO
ꢀ
revealed comparable thermal stability up to 300 C with weight
losses of the mixed-metal MOFs in agreement with the expected
range: about 66% ligand loss for MIL-100(Fe, Ni) 1–3 (63% and
II
adsorption measurements considering its larger Ni content
(5%) and preserved porosity. Fig. 2 (right) depicts the CO
chemisorption isotherm for sample 3 compared to that of pure
MIL-100(Fe) aer activation at 423 K for 12 hours under
vacuum. It is important to note that under these conditions, it
has been shown previously that Fe metal sites remain in the +3
69% calc. for the Fe O and Fe NiO ratio, respectively) (see Table
3 2
1
and Fig. S1–S4†). FT-IR analysis of MOFs 1–3 also conrmed
the formation of the MIL-100 structure. All materials exhibited
the characteristic asymmetric and symmetric stretching bands
II
oxidation state, meaning that the presence of reduced Fe
species is negligible.
ꢁ
1
of the carboxylate anions at 1630–1576 cm
and 1450–
382 cm , respectively. The metal–oxygen bands of the M (m
O) group were identied at 624 cm in MIL-100(Fe) corre-
sponding to the Fe O trimeric unit, while the appearance of
22,35
A noticeably lower CO chemisorption
ꢁ
1
1
3
3
-
capacity is observed in this pressure range for the pure iron
solid, while an increase in the chemisorption capacity of
approximately 10% is measured for the mixed metal MIL-
ꢁ
1
3
ꢁ1
a new band at 569 cm in 1–3 attributed to the Fe NiO unit was
2
100(Fe, Ni) solid activated under the same conditions. This is
recorded for the mixed metal samples (see Fig. S5†). The
chemical stability of the obtained MIL-100(Fe, Ni) 1–3 materials
was evaluated in water as compared to the bare MIL-100(Fe)
material. Aer 24 h of stirring at room temperature, none of
the materials presented leaching of Ni or Fe metals (<1%) and
maintained their crystallinity (Fig. S6†). Upon water treatment
consistent with previous studies carried out by some of us on
24
MIL-127(Fe, Ni) materials. These results point out the effective
impact of mixed-metal cations on the Lewis acidity of the
material and, therefore, also presumably on its catalytic activity
as will be described below.
SEM and energy dispersive X-ray spectroscopy (EDS)
elemental mapping of heterometallic MIL-100(Fe, Ni) material
ꢀ
at 80 C a leaching of ca. 15–20% of the total Ni content was
observed for materials 1–3, while no Fe leaching was detected.
3
conrmed the mixed-metal composition revealed by ICP-OES
ꢀ
Similar studies conducted in MeCN at 80 C revealed negligible
but also endorsed the regular Fe and Ni distribution in the
material at submicrometric resolution, ruling out the presence
of large aggregates of independent Ni or Fe nanoparticles
Ni or Fe metal leaching (<1%).
The porosity of the solids was characterised by N
analysis. Fig. 2 shows the isotherms of MIL-100(Fe) and MIL-
00(Fe, Ni) 1–3 revealing characteristic type I isotherms ex-
2
sorption
(Fig. S9†). This elemental mapping is compatible with the
1
homogeneous spatial distribution at the micrometric scale of
pected for the MIL-100 microporous system with preservation of
II
Ni cations in the heterometallic MIL-100 material. Further
This journal is © The Royal Society of Chemistry 2019
J. Mater. Chem. A

View Article Online
Journal of Materials Chemistry A
Paper
evidence of the structural composition was provided by anal- Table 2 Summary of the main NO bands attributed to the different
III
II
M –NO and M –NO nitrosyls compounds formed
ysis of the EXAFS spectra of the materials at the ROCK beam-
line of the SOLEIL synchrotron radiation laboratory. At the Fe K
edge, iron is conrmed at the oxidation state +3. Least squares
3+
ꢁ1
2+
2+
Fe /NO n(NO) cm Ni /NO Fe /NO

ttings of EXAFS spectra recorded for sample MIL-100(Fe) MIL-100(Fe, Ni)423K (3) 1894
1868
1865
—
—
1805
1794
1816, 1796
(
(
without Ni) and heterometallic sample MIL-100(Fe, Ni) 3 MIL-100(Fe, Ni)503K (3) 1895
containing 5% Ni) are fully consistent with the MIL-100 re-
MIL-100(Fe)423K
MIL-100(Fe)503K
1895
1895
—
ported structure (see Table S1 and Fig. S10–S17†). At the Ni K
edge, Ni is conrmed at the oxidation state +2 with a rising
edge position at the same energy position as that of the NiO or
Ni(OH) references. This conclusion concerning the oxidation
state of the metals is similar regardless of the Ni loading of the
sample.
Least squares ttings of EXAFS spectra recorded for materials
MIL-100(Fe, Ni) 2 (3% Ni) and 3 (5% Ni) have been performed
with the substitution of one iron atom by Ni. Very similar
structural parameters were found for second neighbours (carbon)
and iron second neighbours compared to those for character-
izing the materials at the Fe K edge. The rst Ni–O coordination
shell is described with longer distances than that of Fe–O, which
is consistent with the Ni(2+) oxidation state. Debye Waller factors
are larger for Ni data than for Fe data likely due to the presence of
adsorbed nitrosyls on coordinatively unsaturated sites (CUSs).
The results are summarised in Table 2. NO adsorption experi-
ments on the heterometallic MIL-100(Fe, Ni) activated at
2
ꢀ
ꢀ
150 C, MIL-100(Fe, Ni)150 C, reveal the appearance of two
ꢁ
1
n(NO) bands centred at 1895 and 1868 cm . These bands are
3
+
2+
attributed to NO interacting with Fe and Ni CUSs, respec-
2
4,36
tively.
The intensity of these bands increases with the
amount of NO introduced, this effect being more pronounced
3
+
in the Fe nitrosyl band (see the ESI†). When the sample is
ꢀ
activated at higher temperature (MIL-100(Fe, Ni)230 C), a new
ꢁ
1
band centered at 1805 cm appears, which is assigned to NO
2
+
2+
interacting with Fe CUSs. These new Fe sites will be
3
+
III
II
generated by thermal reduction of Fe under the activation
III conditions, as has been established in related trimeric-based
local distortion upon substitution of Fe by Ni in the structure.
Therefore, it is established that Ni atoms likely replace Fe
positions in the trimeric units of the crystal and they are
distributed homogeneously throughout the network.
II
37
MOF materials. Similar NO adsorption studies performed
ꢀ
on the MIL-100(Fe) material activated at 150 C (MIL-100(Fe)
ꢀ
ꢁ1
1
50 C) show that there is rst a n(NO) band at 1901 cm
(
Fig. 3). Its intensity increases rapidly with the amount of NO
In situ IR spectroscopy. Analysis of the accessible metal sites
ꢁ
1
introduced while its wavenumber shis to 1895 cm
To gain further insights into the nature and the oxidation state
of the accessible cations in the framework, in situ IR analysis
was carried out on mixed-metal MOF MIL-100(Fe, Ni) 3,
comparing the results with those of the pure iron-based MIL-
(
1
Fig. S18†). Then, another very weak band is observed at
ꢁ1
793 cm at the NO equilibrium pressure. As indicated above,
3+
2+
these two bands are assigned to Fe and Fe nitrosyl species,
respectively. The infrared spectra of MIL-100(Fe) activated at
2
expected very intense couple of bands around 1800 cm
assigned to NO adsorption on Fe CUSs, in agreement with the
known formation of supplementary Fe metal sites upon
higher activation temperatures.
1
100(Fe) material. Adsorption of NO as a molecular probe was
ꢀ
ꢀ
30 C (MIL-100(Fe)230 C) and aer NO adsorption display the
monitored by in situ IR spectroscopy, given the well-known
n(NO) bands of adsorbed nitrosyls in the related MIL-127(Fe,
ꢁ1
,
2+
24
Ni) material. The infrared spectra were recorded at low
II
ꢀ
temperature (ꢁ125 C) aer NO adsorption on MIL-100(Fe, Ni)
38,39
Notably, the n(NO) band at
and MIL-100(Fe) activated under vacuum at two different
ꢁ1
868 cm has not been observed in MIL-100(Fe) as is expected
ꢀ
temperatures, 150 and 230 C (Fig. 3). The addition of small NO
for the pure iron based MOF.
doses on the activated materials causes the appearance of new
absorption bands in the spectra, which are assigned to the
This study not only corroborates the efficient incorporation
2+
of Ni centers into the heterometallic MIL-100(Fe, Ni) structure
based on the observation of free CUSs able to interact with NO,
but also establishes their accessible Lewis acid character.
III
Importantly, the characteristic partial reduction of Fe in MIL-
100 is also evidenced in the heterometallic MIL-100(Fe, Ni)
material upon activation at higher temperatures. Generation of
II
Fe is likely to be due to the coexistence of both types of trimeric
units, pure (Fe O) and mixed metal (Fe NiO) trimers, in the
3
2
structure of MIL-100(Fe, Ni).
Catalytic activity
Fig. 3 IR spectra of MIL-100(Fe, Ni) 3 (a) and MIL-100(Fe) (b) recorded
after introduction of an equilibrium pressure of NO (300 Pa) at low
temperature (150 K). Spectra after activation of the solid at 423 K are
represented with a black line while spectra after activation of the solid
at 503 K are represented with a dashed red line. The assignment of the
bands has been indicated in the drawings.
To determine the distinctive catalytic activity of the obtained
heterometallic materials MIL-100(Fe, Ni) 1–3 with respect to
that of their parent MIL-100(Fe) material, we tested them as
40
solid catalysts in the acid-catalyzed Prins reaction. In
J. Mater. Chem. A
This journal is © The Royal Society of Chemistry 2019

View Article Online
Paper
Journal of Materials Chemistry A
Fig. 4 (a) Time conversion plot of b-pinene, using as a catalyst either the mixed-metal MIL-100(Fe, Ni) 1–3 or homometallic MIL-100(Fe). (b) Hot
filtration test and (c) reusability of mixed-metal material MIL-100(Fe, Ni) 3 as a catalyst for the b-pinene conversion. Reaction conditions: catalyst
ꢀ
(
1
15 mg, 0.5 mol% total metal based), b-pinene (1 mmol) and paraformaldehyde (1 mmol) in MeCN at 80 C. Legend (a): MIL-100(Fe) (-), MIL-
st
nd
rd
th
th
00(Fe, Ni) 1 (B), 2 (C) and 3 (,); Legend (b): 1 use (,), 2 use (C), 3 use (,), 4 use (B) and 5 use (:). Inset: PXRD patterns of the fresh
nd
th
MIL-100(Fe, Ni) (black pattern) and after the 2 and 5 reuse (red and blue patterns, respectively).
particular, the model reaction studied was the synthesis of solid catalyst established the absence of iron or nickel leaching
nopol
paraformaldehyde.
via
Prins
condensation
of
b-pinene
and from the solid to the liquid phase.
30
The heterogeneity of the catalytic procedure was ascertained
Prior to the catalytic reaction, the samples MIL-100(Fe, Ni) 1–3 by Sheldon's hot ltration test. Fig. 4 shows a comparison of the
ꢀ
and MIL-100(Fe) were thermally activated at 150 C under vacuum temporal prole of b-pinene conversion in the presence of MIL-
overnight. In a typical procedure, the Prins condensation was 100(Fe, Ni) 3 and upon hot ltration of the solid catalyst aer 1
ꢀ
carried out with the MIL-100(Fe, Ni) at 80 C in MeCN, using hour of reaction. Noticeably, the reaction did not proceed aer
0
.5 mol% (total metal based) of solid (Fig. S8†). It is worth removal of the catalyst, demonstrating that no catalytically active
mentioning that further increase in the amount of catalyst (up to metal species are present in the ltrate and therefore estab-
.7 mol%) had no signicant effect on the reaction rate. Fig. 4a lishing the heterogeneity of the MOF. A productivity test using
0
shows the temporal evolution of b-pinene conversion to the cor- a large excess of b-pinene (10 mmol) over a minimal amount of
responding nopol promoted by the mixed-metal MOF and the mixed-metal MIL-100(Fe, Ni) MOF (0.001 mmol Fe + Ni metal)
analogous MIL-100(Fe). Aer 10 h of reaction, all materials show was performed. Under these extreme conditions, full conver-
a complete selectivity for the formation of nopol, and while the sions were achieved at long reactions times (see Fig. S20†),
4
use of homometallic MIL-100(Fe) affords <20% conversion, the reaching a turnover number of 1 ꢃ 10 . Importantly, a selectivity
use of heterometallic samples as catalysts affords between 70 and of 100% was maintained over the reaction. It should be pointed
100% conversion. Indeed, only the MIL-100(Fe, Ni) 3 material out that although the Prins reaction has been studied with other
resulted in a total conversion to nopol aer 10 h. Therefore, the MOFs and zeolites, the selectivity of nopol has been found to
activity test clearly shows that the presence of Ni signicantly decrease in the presence of adventitious Brønsted acidity that
increases the catalytic activity of MIL-100 as a Lewis acid, causes the undesirable isomerization and other side reactions of
increasing the initial reaction rate with the percentage of Ni in the norbornadiene. Also, although a combination of Amberlyst 15
range under study. These catalytic data are in accordance with the and montmorillonite has been reported as an efficient hetero-
41
increase of the Lewis acidity determined by CO titration upon geneous catalyst, the present mixed-metal MIL-100(Fe, Ni) is
2+
incorporation of Ni . A plausible explanation for the observed more versatile, still allowing for further optimization.
improved catalytic behaviour may be associated with the distor-
II
tion caused in the framework upon insertion of Ni Lewis acid
sites, as demonstrated by EXAFS experiments (see Fig. S10–S17†). Conclusions
It is believed that the formation of the mixed-metal trimers in the
framework may facilitate access to the Fe Lewis acid sites, thus
enhancing the catalytic activity of the catalyst.
The stability of the best MIL-100(Fe, Ni) 3 catalyst under the
reaction conditions was established by recovering and reusing
the material in ve consecutive runs, observing that the initial
reaction rate, nal conversion and complete selectivity to nopol
were maintained when the material was properly washed (uses
III
II
III
Heterometallic Fe /Ni MOFs of the MIL-100 type material with
different metal ratios have been synthesized, exhibiting the
thermal, chemical and structural stability of the parent MIL-
00(Fe) MOF. The incorporation of Ni centers in the hetero-
II
1
III
II
metallic MIL-100(Fe , Ni ) structure was based on analytical
data, EXAFS and in situ IR analysis, establishing both the
homogeneous distribution throughout the network and their
accessible Lewis acid character.
The mixed-metal MIL-100(Fe, Ni) MOF containing Ni and the
parent MIL-100(Fe) material have been studied as heteroge-
neous catalysts in a model acid-catalyzed reaction, i.e., the Prins
reaction. It was observed that the initial reaction rate increases
2, 4 and 5) (Fig. 4b). X-ray powder diffraction analysis of the
fresh and the reused materials (Fig. 4b inset) conrmed the
preservation of the crystallinity upon successive catalytic reac-
tions, while analysis of the liquid phase aer removal of the
This journal is © The Royal Society of Chemistry 2019
J. Mater. Chem. A

View Article Online
Journal of Materials Chemistry A
Paper
II
gradually with the percentage of Ni incorporated in the a gas-dosing facility. Transmission IR spectra were recorded in
ꢁ
1
ꢁ1
material. An improvement of the catalytic activity of more than the 500–5600 cm range, at 4 cm resolution. Aer activation,
ꢀ
one order of magnitude has been determined in the case of the the wafers were cooled to ꢁ125 C under dynamic vacuum fol-
mixed-metal MIL-100(Fe, Ni) catalyst 3 (5% Ni). This catalytic lowed by NO dosing at an increasing pressure (0.4–6 mbar). IR
II
behaviour, originating from the insertion of new Ni Lewis acid spectra were collected aer each dosage. All IR spectra that
sites, is maintained over successive cycles and correlates with correspond to the NO adsorption measurement were normal-
CO adsorption measurements. The improved catalytic activity ized to the weight of the wafer.
may have its origin in the distortion originating upon inclusion
II
of Ni ions in the framework, although local electronic modi-
Thermogravimetric analysis

cations are also expected to play a role. A future in-depth study
The thermal stability was evaluated by thermogravimetric
analysis (TGA). TG measurements were carried out with a Met-
to gain a more exhaustive understanding of this catalytic
improvement is envisioned.
It is proposed that the use of this simple and direct synthetic
approach may be applicable in the preparation of other mixed-
metal MOFs suitable as heterogeneous catalysts. Considering
the wide use of MOFs as heterogeneous catalysts, the present
tler Toledo TGA/DSC 1, STAR System apparatus under an O
2
ꢁ
1
ꢀ
ꢁ1
ꢀ

ow of 50 mL min , at a heating rate of 3 C min to 500 C.
Powder X-ray diffraction analysis
results showing the enhancement of the catalytic activity in Polycrystalline samples of MIL-100(Fe, Ni) were lightly ground
heterometallic MIL-100 open the way to prepare new genera- in an agate mortar and pestle and lled into 0.5 mm borosilicate
tions of even more active materials by introducing appropriate capillaries prior to being mounted and aligned on an Empyrean
amounts of two or more metals at the nodes.
PANalytical powder diffractometer using Cu Ka radiation (l ¼
˚
1
.54056 A). For each sample, three repeated measurements were
ꢀ
collected at room temperature (2q ¼ 2 ꢁ 20 ) and merged into
Methods
Materials
a single diffractogram.
Iron powder, 1,3,5-benzenetricarboxylic acid (95%) (trimesic Adsorption measurements
acid), HNO (65%) aqueous solution, paraformaldehyde, and b-
pinene were purchased from Sigma-Aldrich. Solvents were used
without further purication.
3
Adsorption/desorption isotherms for N were measured at 77 K
2
using a Tristar (Micromeritics). Samples were thoroughly
ꢀ
dehydrated by heating at 150 C for 5 h under vacuum.
Adsorption/desorption isotherms for CO were measured at 303
K using a Quantachrome Autosorb automated gas sorption
Synthesis of MIL-100(Fe) and mixed-metal MIL-100(Fe, Ni)
MOFs (1–3)
system (Quantachrome Instruments). Samples were dehydrated
ꢀ
The homometallic MIL-100(Fe) MOF was obtained by directly by heating at 150 C for 12 h under vacuum.
mixing 1.680 g of trimesic acid (8 mmol) and 0.670 g of Fe
2
powder (12 mmol) in 90 mL of distilled H O, followed by the
EXAFS experiments
addition of a 65% HNO
mmol). The whole mixture was then heated to 85 C for 24
hours. Aer cooling to room temperature, the solution was
3
aqueous solution (0.424 mL, 6
The Ni and Fe K edge XAS measurements were performed at the
ROCK beamline at the SOLEIL synchrotron radiation facility.
ꢀ
42
Pellets of the powdered samples diluted with boron nitride were
prepared with sample masses optimized in order to achieve an
edge jump of about 1 at the Fe K edge and ꢄ0.1 to 0.2 at the Ni K
edge. Data were collected at room temperature in transmission
mode using a Si(111) channel cut monochromator and ioniza-
tion chambers lled with nitrogen. Harmonic rejection was

ltered and the obtained powder washed with water. The solid
ꢀ
was then dispersed in 30 mL of EtOH and heated to 60 C for
1
2 hours to remove excess of trimesic acid. Aer cooling to
room temperature, the resulting compound was recovered by
ltration and washed with water. For the mixed-metal MIL-

III
II
1
00(Fe/Ni) MOFs 1–3 different molar M /M mixtures were
ensured using two mirrors with B
4
C stripes aligned to 2.8 mrad
II
used: 1 (11.88 and 0.12 mmol of Fe powder and Ni acetate,
respectively); 2 (11.4 and 0.6 mmol); 3 (10.8 and 1.2 mmol).
The amount of HNO aqueous solution added was adjusted in
3
with respect to the incident beam. XAS analysis was carried out
43
using Athena soware. Energy calibration was done at the
tabulated values using the rst derivative maximum of iron (E0
0
all cases to 0.5 mmol of Fe . MIL-100 is formulated as Fe3–
¼
7112 eV) and nickel (E
0
¼ 8333 eV) metallic foils located
Ni O(C H (CO ) ) (OH) $(H O)3–x$(solvent).
x
x
6
3
2 3 2
x
2
between the second and third ionization chambers and recor-
ded simultaneously with the samples. Aer linear pre-edge and
atomic background removal, XANES spectra were normalized in
In situ IR spectroscopic analysis
For in situ IR analysis, samples were pressed into self-supported the middle of the rst EXAFS oscillation. Fe K edge and Ni K
ꢁ
2
3
wafers (4–10 mg cm ) and treated in a vacuum (10–6 mbar) at edge EXAFS spectra were k -weighted before Fourier-
ꢀ ꢀ
50 C or 230 C for 12 h. NO-FTIR spectra were recorded by transformation using a Kaiser–Bessel window in the 3.3–11
1
ꢁ1
using a Nexus 8700 FTIR spectrophotometer, using an IR cell and 10 A˚ k range with dk ¼ 2. Distances and Debye–Waller
that allowed us to perform in situ treatments at a controlled factors of the different contributions characterizing the local
temperature and was connected to a high vacuum system with order of MIL-100(Fe) and mixed-metal MIL-100(Fe, Ni) MOFs
J. Mater. Chem. A
This journal is © The Royal Society of Chemistry 2019

View Article Online
Paper
2–3) were determined by least-squares renements using the
Journal of Materials Chemistry A
(
Acknowledgements
42
Artemis interface to IFeFFIT while the coordination numbers
of those contributions were xed at the values described in the This work was supported by the European Commission under
44
MIL-100(Fe) crystallographic structure.
the Marie Sklodowska-Curie agreement H2020-MSCA-IF-
The amplitude reduction factors were obtained by tting the 658224. Measurements at the ROCK synchrotron beamline of
2
EXAFS spectra of a-Fe
Fe K edge) and 1.23 (Ni K edge)) were kept xed for the simu- National Research Agency (ANR) as part of the “Investissements
lation of MIL-100(Fe) and mixed-metal MIL-100(Fe, Ni) MOFs d’Avenir” program (reference: ANR10-EQPX45). H. G. thanks
2–3) while the energy shi values which allow the match nancial support by the Spanish Ministry of Science and
2
O
3
and NiO and their values (S
0
¼ 0.68 SOLEIL were supported by a public grant overseen by the French
(
(
between theoretical and experimental k-scales were allowed to Innovation (Severo Ochoa and RTI2018-098237-CO21) and
vary within the limits 7127.7 ꢂ 3.0 eV and 8345.6 ꢂ 3.4 eV. The Generalitat Valenciana (Prometeo 2017/083). S. N. thanks
reliability of the ts was assessed by the minimization of the nancial support by the Fundaci ´o n Ram ´o n Areces (XVIII Con-
EXAFS reliability factor R
with respect to the data and the reduced c
metrics dened by the IXS Standards and Criteria Committee.
f
which measures the relative mist curso Nacional para la Adjudicaci ´o n de Ayudas a la Inves-
2
n
quality factor, two tigaci ´o n en Ciencias de la Vida y de la Materia, 2016) and
45
Ministerio de Ciencia, Innovaci ´o n y Universidades CTQ-2018
RTI2018-099482-A-I00 project.
Catalytic experiments
The required amount of catalyst (typically 10 mg, except for
Notes and references
productivity tests) was added to a round-bottom ask (25 mL).
ꢀ
Catalyst activation was performed at 150 C under vacuum
1
2
3
See the CrystEngComm 2017 themed issue "Metal–organic
framework catalysis".
L. Zhu, X.-Q. Liu, H.-L. Jiang and L.-B. Sun, Chem. Rev., 2017,
overnight. The reaction system was purged with molecular
oxygen using a balloon, the substrate was introduced (20 mmol)
and the system was sonicated for 5 min. Then, the reaction
117, 8129.
ꢀ
mixture was placed in a pre-heated bath at 80 C and magnet-
H. Garc ´ı a and S. Naval ´o n, Metal-Organic Frameworks:
Applications in Separations and Catalysis, Wiley, ISBN: 978-
ically stirred.
The reusability of MIL-100(Fe, Ni) was evaluated under the
general reaction conditions. At the end of the reaction the
catalyst was ltered through a Nylon membrane lter of 0.2
mm. Then, the catalyst was washed by suspending the solid in
3-527-80910-3, 2018.
4
S. M. J. Rogge, A. Bavykina, J. Hajek, H. Garcia, A. I. Olivos-
Suarez, A. Sep u´ lveda-Escribano, A. Vimont, G. Clet,
P. Bazin, F. Kapteijn, M. Daturi, E. V. Ramos-Fernandez,
F. X. Llabr ´e s i Xamena, V. Van Speybroeck and J. Gascon,
Chem. Soc. Rev., 2017, 46, 3134.
ꢀ
methanol at 60 C for 2 h. This washing procedure was repeated
ꢀ
two more times. Finally, the catalyst was dried at 100 C over-
night and used in subsequent catalytic cycles. Product analysis
was conducted as follows: previously ltered reaction aliquots
5
6
W. Tu, Y. Xu, S. Yin and R. Xu, Adv. Mater., 2018, 30, 1707582.
H. Depauw, I. Nevjesti ´c , G. Wang, K. Leus, F. Callens, E. De
Canck, K. De Buysser, H. Vrielinck and P. Van Der Voort, J.
Mater. Chem. A, 2017, 5, 24580.
(100 mL) were diluted in a MeCN solution (500 mL) containing
a known amount of nitrobenzene as the external standard.
Then, the reaction aliquots were analyzed by gas chromatog-
raphy using a ame ionization detector. Quantication was
carried out by using calibration curves of authentic samples.
7
8
9
C. Huang, R. Liu, W. Yang, Y. Li, J. Huang and H. Zhu, Inorg.
Chem. Front., 2018, 5, 1923.
A. Dhakshinamoorthy, A. M. Asiric and H. Garcia, Catal. Sci.
Technol., 2016, 6, 5238.
Leaching measurements
´
P. Silva, S. M. F. Vilela, J. P. C. Tome and F. A. Almeida Paz,
Chem. Soc. Rev., 2015, 44, 6774.
0 Y.-B. Huang, J. Liang, X.-S. Wang and R. Cao, Chem. Soc. Rev.,
Aer reaction completion, the reaction mixture was ltered
while still hot. Then, the organic phase was processed to extract
1
2017, 46, 126.
the metal ions by stirring in an aqueous nitric acid solution
ꢀ
11 L. M. Aguirre-D ´ı az, F. G ´a ndara, M. Iglesias, N. Snejko,
E. Guti ´e rrez-Puebla and M. A. Monge, J. Am. Chem. Soc.,
(
HNO , 3 M) at 80 C overnight. The metal content of the
3
aqueous phase was analyzed by chemical analysis using an ICP-
AES instrument.
2015, 137, 6132.
1
2 L. Mitchell, P. Williamson, B. Ehrlichov, A. E. Anderson,
V. R. Seymour, S. E. Ashbrook, N. Acerbi, L. M. Daniels,
R. I. Walton, M. L. Clarke and P. A. Wright, Chem.–Eur. J.,
Sheldon's hot ltration test
During this test, the catalyst was removed from the reaction by
2014, 20, 17185.

ltration while hot aer 60 min of reaction, and the ltrate was
1
1
3 A. Schejn, A. Aboulaich, L. Balan, V. Falk, J. Lalev ´e e,
G. Medjahdi, L. Aranda, K. Mozet and R. Schneider, Catal.
Sci. Technol., 2015, 5, 1829.
4 J. Park, J.-R. Li, Y.-P. Chen, J. Yu, A. A. Yakovenko,
Z. U. Wang, L.-B. Sun, P. B. Balbuena and H.-C. Zhou,
Chem. Commun., 2012, 48, 9995.
monitored for continued activity.
Conflicts of interest
There are no conicts to declare.
This journal is © The Royal Society of Chemistry 2019
J. Mater. Chem. A

View Article Online
Journal of Materials Chemistry A
Paper
1
5 C. Castillo-Blas, V. A. de la Pe n˜ a-O’Shea, I. Puente-Orench, 32 (a) M. Opanasenko, A. Dhakshinamoorthy, Y. K. Hwang,
J. Romero de Paz, R. S ´a ez-Puche, E. Guti ´e rrez-Puebla,
F. G ´a ndara and M. Monge, Sci. Adv., 2017, 3, e1700773.
6 L. M. Aguirre-D ´ı az, F. G ´a ndara, M. Iglesias, N. Snejko,
E. Guti ´e rrez-Puebla and M. Monge, J. Am. Chem. Soc., 2015,
J.-S. Chang, H. Garcia and J. Cejka, ChemSusChem, 2013, 6,
865; (b) G. G ´o mez-Pozuelo, C. Palomino Cabello,
M. Opanasenko, M. Hor ´a ˇc ek and J. Cejka, ChemPlusChem,
2017, 82, 152.
ˇ
ˇ
1
1
37, 6132.
33 J. Shi, S. Hei, H. Liu, Y. Fu, F. Zhang, Y. Zhong and W. Zhu,
Journal of Chemistry, 2013, 792827.
1
1
1
7 C. K. Brozek and M. Dincǎ, Chem. Soc. Rev., 2014, 43, 5456.
8 S. M. Cohen, Chem. Rev., 2012, 112, 970.
34 L. Peng, M. Asgari, P. Mieville, P. Schouwink, S. Bulut,
D. T. Sun, Z. Zhou, P. Pattison, W. van Beek and
W. L. Queen, ACS Appl. Mater. Interfaces, 2017, 9, 23957.
35 (a) J. F. Eubank, P. S. Wheatley, G. Lebars, A. C. Mckinlay,
H. Leclerc, P. Horcajada, M. Daturi, A. Vimont,
R. E. Morris and C. Serre, APL Mater., 2014, 2, 124112; (b)
9 M. Mon, J. Ferrando-Soria, M. Verdaguer, C. Train,
C. Paillard, B. Dkhil, C. Versace, R. Bruno, D. Armentano
and E. Pardo, J. Am. Chem. Soc., 2017, 139, 8098.
´
´
2
0 P. A. Szilagyi, P. Serra-Crespo, J. Dugulan, J. Gascon,
H. Geerlingsad and B. Dama, CrystEngComm, 2013, 15,
˚
10175.
C. Serre, P. Horcajada, G. FeALrey, A. Vimont, M. Daturi,
2
1 J. Castells-Gil, N. M. Padial, N. Almora-Barrios, J. Albero,
A. R. Ruiz-Salvador, J. Gonz ´a lez-Platas, H. Garc ´ı a and
C. Mart ´ı -Gastaldo, Angew. Chem., Int. Ed., 2018, 57, 8453.
2 C. Serre, F. Millange, S. Surbl ´e and G. F ´e rey, Angew. Chem.,
Int. Ed., 2004, 43, 6296.
3 S. Surbl ´e , C. Serre, C. Mellot-Draznieks, F. Millange and 37 J. W. Yoon, Y.-K. Seo, Y. K. Hwang, J.-S. Chang, H. Leclerc,
G. F ´e rey, Chem. Commun., 2006, 284.
4 V. Guillerm, S. Gross, C. Serre, T. Devic, M. Bauer and
G. F ´e rey, Chem. Commun., 2010, 46, 767.
P. L. Llewellyn, Y. K. Hwang, J.-W. Yoon and J. S. Chang,
Use of a porous crystalline hybrid solid as a nitrogen oxide
reduction catalyst and devices, 106660/FR PCT/FR2010/
000402, 28/05/2010.
2
2
2
2
36 K. I. Hadjiivanov, Catal. Rev.: Sci. Eng., 2000, 42, 71.
S. Wuttke, P. Bazin, A. Vimont, M. Daturi, E. Bloch,
P. L. Llewellyn, C. Serre, P. Horcajada, J.-M. Grenche,
A. E. Rodrigues and G. F ´e rey, Angew. Chem., Int. Ed., 2010,
49, 5949.
5 S. Wongsakulphasatch, F. Nouar, J. Rodriguez, L. Scott, C. Le
Guillouzer, T. Devic, P. Horcajada, J.-M. Gren `e che, 38 S. Wuttke, P. Bazin, A. Vimont, C. Serre, Y. Seo, Y. K. Hwang,
P. L. Llewellyn, A. Vimont, G. Clet, M. Daturi and C. Serre,
J. S. Chang, G. F. Rey and M. Daturi, Chemistry, 2012, 18,
Chem. Commun., 2015, 51, 10194.
11959.
26 D. Feng, K. Wang, Z. Wei, Y.-P. Chen, C. M. Simon, 39 J. F. Eubank, P. S. Wheatley, G. Lebars, A. C. McKinlay,
R. K. Arvapally, R. L. Martin, M. Bosch, T.-F. Liu,
S. Fordham, D. Yuan, M. A. Omary, M. Haranczyk, B. Smit
and H.-C. Zhou, Nat. Commun., 2014, 5, 5723.
7 X. -L. Wang, L. -Z. Dong, M. Qiao, Y. -J. Tang, J. Liu, Y. Li,
S. -L. Li, J. -X. Su and Y. -Q. Lan, Angew. Chem., Int. Ed.,
H. Leclerc, P. Horcajada, M. Daturi, A. Vimont,
R. E. Morris and C. Serre, APL Mater., 2014, 2, 124112.
40 Common Fragrance and Flavor Materials, Preparation,
Properties, and Uses, ed. H. Surburg and J. Panten, Wiley-
VCH, Weinheim, 2006, vol. 67.
2
2
2
018, 57, 9660.
41 J. S. Yadav, B. V. Subba Reddy, G. Mahesh Kumar and
C. V. S. R. Murthy, Tetrahedron Lett., 2001, 42, 89.
42 V. Briois, C. La Fontaine, S. Belin, L. Barthe, T. Moreno,
V. Pinty, A. Carcy, R. Girardot and E. Fonda, J. Phys.: Conf.
Ser., 2016, 712, 012149.
8 H. Vrubel, T. Hasegawa, E. Oliveira and F. S. Nunes, Inorg.
Chem. Commun., 2006, 9, 208.
29 A. Schoedel and M. J. Zaworotko, Chem. Sci., 2014, 5, 1269.
0 W. Zhang, Y. Shi, C. Li, Q. Zhao and X. Li, Catal. Lett., 2016,
3
1
46, 1956.
43 B. Ravel and M. Newville, J. Synchrotron Radiat., 2005, 12,
537.
3
1 F. Nouar, T. Devic, H. Chevreau, N. Guillou, E. Gibson,
G. Clet, M. Daturi, A. Vimont, J. M. Gren `e che, M. I. Breeze, 44 P. Horcajada, S. Surbl ´e , C. Serre, D.-Y. Hong, Y.-K. Seo,
R. I. Walton, P. L. Llewellyne and C. Serre, Chem.
Commun., 2012, 48, 10237.
J.-S. Chang, J.-M. Gren `e che and I. M. G. F ´e rey, Chem.
Commun., 2007, 2820.
45 http://ixs.iit.edu/subcommittee_reports/sc/err-rep.pdf.
J. Mater. Chem. A
This journal is © The Royal Society of Chemistry 2019