Catalytic properties of pristine and defect-engineered Zr-MOF-808 metal organic frameworks

Article abstract of DOI:10.1039/c8cy00742j

Various defect-engineered Zr-trimesate MOF-808 compounds (DE-MOF-808) have been prepared by mixing the tricarboxylate ligands with dicarboxylate ligands; viz. isophthalate, pyridine-3,5-dicarboxylate, 5-hydroxy-isophthalate, or 5-amino-isophthalate. The resulting mixed-ligand compounds, MOF-808-X (X = IP, Pydc, OH or NH₂) were all found to be highly crystalline and isostructural to the unmodified MOF-808. Pristine MOF-808 showed better catalytic performance than a UiO-66 reference compound for the Meerwein-Ponndorf-Verley (MPV) reduction of carbonyl compounds. This was attributed to a higher availability of coordinatively unsaturated Zr⁴⁺ sites (cus) in MOF-808 upon removal of formate ions. Meanwhile, cus in UiO-66 are only located at defect sites and are thus much less abundant. Further improvement of the catalytic activity of defect-engineered MOF-808-IP and MOF-808-Pydc was observed, which may be related with the occurrence of less crowded Zr⁴⁺ sites in DE-MOF-808. The wider pore structure of MOF-808 with respect to UiO-66 compounds translates into a sharp improvement of the activity for the MPV reduction of bulky substrates, as shown for estrone reduction to estradiol. Interestingly, MOF-808 produces a notable diastereoselectivity towards the elusive 17-α-hydroxy estradiol.

Full text of DOI:10.1039/c8cy00742j

View Article Online
View Journal
Catalysis
Sc ience &
Technology
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: H. Mautschke, F.
Drache, I. Senkovska, S. Kaskel and F. X. X. Llabrés i Xamena, Catal. Sci. Technol., 2018, DOI:
1
0.1039/C8CY00742J.
Volume 6 Number 1 7 January 2016 Pages 1–308
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Catalysis
Science &
Technology
www.rsc.org/catalysis
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
ISSN 2044-4753
PAPER
Qingzhu Zhang et al.
Catalytic mechanism of C–F bond cleavage: insights from QM/MM
analysis of fluoroacetate dehalogenase
rsc.li/catalysis

Page 1 of 16
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C8CY00742J
Catalytic properties of pristine and defect-engineered Zr-MOF-808
Metal Organic Frameworks
a
b
b
b
H. –H. Mautschke, F. Drache, I. Senkovska, S. Kaskel and F. X. Llabrés i
a*
Xamena
a
Instituto de Tecnología Química UPVꢀCSIC, Universitat Politècnica de València,
Consejo Superior de Investigaciones Científicas, Avda. de los Naranjos s/n, 46022
Valencia, Spain
b
Department of Inorganic Chemistry, Dresden University of Technology, Bergstraße
6
*
6, 01062 Dresden, Germany
Corresponding author:
eꢀmail: fllabres@itq.upv.es
Fax: (+34) 963877809
Abstract
Various defectꢀengineered Zrꢀtrimesate MOFꢀ808 compounds (DEꢀMOFꢀ808) have
been prepared by mixing the tricarboxylate ligands with dicarboxylate ligands; viz.
isophthalate, pyridineꢀ3,5ꢀdicarboxylate, 5ꢀhydroxyꢀisophthalate, or 5ꢀaminoꢀ
isophthalate. The resulting mixedꢀligand compounds, MOFꢀ808ꢀX (X = IP, Pydc, OH or
NH ) were all found to be highly crystalline and isostructural to the unmodified MOFꢀ
2
808. Pristine MOFꢀ808 showed better catalytic performance than a UiOꢀ66 reference
compound for the MeerweinꢀPonndorfꢀVerley (MPV) reduction of carbonyl
compounds. This was attributed to a higher availability of coordinatively unsaturated
4
+
Zr sites (cus) in MOFꢀ808 upon removal of formate ions. Meanwhile, cus in UiOꢀ66
are only located at defect sites and are thus much less abundant. Further improvement of
the catalytic activity of defectꢀengineered MOFꢀ808ꢀIP and MOFꢀ808ꢀPydc was
4+
observed, which may be related with the occurrence of less crowded Zr sites in DEꢀ
MOFꢀ808. The wider pore structure of MOFꢀ808 with respect to UiOꢀ66 compounds
translate into a sharp improvement of the activity for the MPV reduction of bulky
substrates, as shown for estrone reduction to estradiol. Interestingly, MOFꢀ808 produces
a notable diastereoselectivity towards the elusive 17ꢀαꢀhydroxy estradiol.
Keywords: Metal Organic Frameworks; heterogeneous catalysis; MOFꢀ808; ZrꢀMOFs,
Diastereoselective MeerweinꢀPondorfꢀVerley; estrone reduction.

Catalysis Science & Technology
Page 2 of 16
View Article Online
DOI: 10.1039/C8CY00742J
Introduction
Metal Organic Frameworks (MOFs) are crystalline hybrid materials formed by metal
ions (or oxoclusters) connected to polytopic organic molecules through coordination
bonds forming extended monoꢀ, biꢀ or tridimensional networks defining a strictly
1
regular system of pores and cavities of molecular dimensions. The huge variability of
chemical compositions and pore architectures, the large surface areas and pore volumes
attainable with these materials, and the possibility to introduce new functionalities in
2
, 3
preformed MOFs through postꢀsynthesis methods,
all make MOFs excellent
candidates for a number of technological applications, including heterogeneous
4
ꢀ7
8
catalysis. In this sense, the Zrꢀcontaining terephthalate known as UiOꢀ66 and its
9
ꢀ13
derivatives have attracted a great deal of attention in recent years.
These compounds
are formed by hexameric [Zr (OH) ] octahedral oxoclusters connected by 12
6
O
4
4
terephthalate linkers into a face centered cubic packing. The high coordination number
of the inorganic building units endows the material with a remarkable thermal, chemical
1
4
and mechanical stability, which is seldom found in many known MOF compounds.
Together with this high stability, an attractive characteristic of UiOꢀ66ꢀtype compounds
is the possibility to prepare a battery of derivatives by introducing additional functional
15ꢀ17
groups in the terephathalate linkers
and to prepare expanded versions of UiOꢀ66 by
replacing terephthalate by longer linear dicarboxylate linkers (viz., UiOꢀ67, UiOꢀ68 and
18
4+
related compounds ). However, the coordination sphere of Zr ions in UiOꢀ66 is
completely blocked by the carboxylate linkers, thus lacking coordination vacancies
available for substrate binding. Therefore, UiOꢀ66 materials rely on the formation of
linker defects (either intrinsic or intentional) to display the reactivity typical of solid
9
, 19
acid catalysts.
In this way, it is well know that a direct correlation exist between the
9
number of missing linker defects and the catalytic activity of UiOꢀ66. A second
limitation of UiOꢀ66 is that it features a relatively narrow pore system, accessible
8
through windows of about 6 Å, which largely delimits the substrates that can reach the
active sites located at the internal surface, and thus, the reactivity scope of this material.
Motivated by the excellent catalytic properties displayed by UiOꢀ66ꢀtype materials,
and in an attempt to overcome the two limitations mentioned above (viz., lack of
coordination vacancies and narrow pore system), we have addressed our efforts to other
Zrꢀcontaining compounds. Among several possible candidates, we have found that Zr
2
0
trimesate known as MOFꢀ808 can be an interesting alternative with a high potential
for catalytic applications. This compound has similar Zr ꢀoxoaggregates than those
6

Page 3 of 16
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C8CY00742J
present in UiOꢀ66, forming a tridimensional network with cavities of 18.4 Å and
apertures of 14 Å (see Fig. 1). In the MOFꢀ808 structure, each cluster is connected by 6
trimesate linkers, while the other coordination positions of Zr ions are saturated by
bridging formate molecules. These formate molecules can be removed by simple
solvent washing or mild thermal treatment, thus leaving two coordination vacancies on
each metal site.
6
Fig 1. Structure of MOFꢀ808, showing the Zr oxoaggregates and evidencing the large
adamantaneꢀlike cavities. C, black; O, red; Zr, blue polyhedra. H atoms are omitted for
clarity. Yellow and orange balls indicate the space in the framework. Reproduced with
20
permission from Ref. . Copyright 2014 American Chemical Society.
Besides MOFꢀ808, other wide pore compounds containing similar Zr
6
oxoaggregates
1
8
21
22
with open metal sites might be considered as well, such as NUꢀ1000, PCNꢀ222,
2
3
24
25
PCNꢀ700, DUTꢀ51, or the PIZOFs. However, a clear drawback of these
compounds is that they usually require the use of nonꢀcommercial and/or expensive
organic ligands, which introduce additional synthesis steps and rise up the final price,
thus hampering the large scale synthesis of these MOFs. This is in sharp contrast with
readily available and cheap trimesic acid used in the synthesis of MOFꢀ808.
In order to increase further the availability of the Zr sites in MOFꢀ808, we have
prepared defectꢀengineered MOFꢀ808 materials (DEꢀMOF) by a mixedꢀligand approach
(see Scheme 1). Thus, we have combined the tritopic trimesic acid ligands (1) with a

Catalysis Science & Technology
Page 4 of 16
View Article Online
DOI: 10.1039/C8CY00742J
small amount (ca. 10%) of a ditopic ligand, such as isophtalic acid (2), pyridineꢀ3,5ꢀ
dicarboxylic acid (3), or 5ꢀhydroxyꢀ (4) and 5ꢀaminoꢀisophthalic acid (5) to prepare a
series of mixedꢀligand MOFs. A similar strategy has been shown to increase the number
26
27, 28
of open metal sites in related Cu and Ru
trimesate compounds, which translated
into a considerable improvement of the catalytic properties of the DEꢀMOF with respect
to the pristine compound.
Scheme 1. Trimesate (1) is combined with various ditopic ligands (2ꢀ5) to prepare
defectꢀengineered MOFꢀ808, DEꢀMOFs.
Herein, we have investigated the catalytic properties of both, pristine and DEꢀMOFꢀ
808 compounds, and the results have been compared with those obtained over UiOꢀ66
as a reference compound. As a test reaction, we have considered the Meerweinꢀ
PonndorfꢀVerley reduction of carbonyl compounds using an alcohol as reducing agent.
Indeed, de Vos and coꢀworkers have recently shown that MOFꢀ808 can efficiently
29
catalyse this reaction. The benefits of MOFꢀ808 having a wider pore system and a
4
+
higher concentration of coordinatively unsaturated Zr sites (cus), as compared with
UiOꢀ66, are evidenced by studying the reduction of ketones of different size.
Experimental section
Synthesis of pristine and DE-MOFs
MOF-808: Pristine zirconium MOFꢀ808 material was prepared with slightly
20
modifications from an earlier reported procedure by Furukawa et. al. Briefly, a
solution was prepared containing 242,5 mg of ZrOCl *8H 0 (0,75 mmol), 105 mg of
trimesic acid (0,5 mmol) and 22,5 mL of a DMF/HCO H 1:1 (v/v) mixture. The
solution was transferred into a Teflon lined autoclave and heated inside an oven at 130
2
2
2

Page 5 of 16
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C8CY00742J
°C for 48 h. After cooling down to room temperature, the material was recovered by
centrifugation and washed for 3 days with DMF (changing the solvent 2 times per day)
and for another 3 days with EtOH (changing the solvent 2 times per day). After
removing the solvent by centrifugation, the solid was dried in air. Xꢀray diffraction
(PhillipsX’Pert, Cu Kα radiation) was used to confirm the expected structure type and
high crystallinity of the material.
DE-MOF-808: DEꢀMOFꢀ808 materials were prepared by replacing 10 mol% of
trimesic acid by equimolar amounts of defective linker: a) isophthalic acid (8,3 mg, 0,05
mmol) for MOFꢀ808ꢀIP, b) 3,5ꢀpyridinedicarboxylic acid (8,4 mg, 0,05 mmol) for
MOFꢀ808ꢀPydc, c) 5ꢀhydroxyisophthalic acid (9,1 mg, 0,05 mmol) for MOFꢀ808ꢀOH
and d) 5ꢀaminoisophthalic acid (9,1 mg, 0,05 mmol) for MOFꢀ808ꢀNH
synthesis steps were kept the same as for the pristine MOFꢀ808 material.
UiO-66: The reference zirconium terephthalate UiOꢀ66 material was prepared
2
. The remaining
1
6
according to the reported procedure. Briefly, 750 mg of ZrCl
terephthalic acid were dissolved in 90 mL of DMF (Zr:ligand:DMF molar ratio of
:1:220) and the solution was kept in a closed round bottom flask at 80ºC in an oil
4
and 740 mg of
1
heating bath for 12 h without stirring, followed by another 24 h at 100ºC. The resulting
material was recovered by filtration and washed thoroughly with fresh DMF. Then the
solid was washed three times by soaking in dichloromethane for 3 h. Finally, the solid
was recovered by filtration and dried under vacuum. The amount of missing linker
defects in this UiOꢀ66 sample was estimated to be ca. 7% from the corresponding TGA
30
curve, following the method proposed by Valenzano et. al.
DUT-67: ZrCl (1.38 g, 6 mmol) was dissolved in 150 mL of N,N
dimethylformamide/ ꢀmethylꢀ2ꢀpyrrolidone mixture (DMF/NMP, 1:1) by
4
ꢀ
N
ultrasonication for 10 minutes. Afterwards, 2,5ꢀthiophenedicarboxylic acid (0.66 g, 4
mmol) was added and the mixture was sonicated again for 5 minutes. Then formic acid
(26.8 mL, 120 eq) was added and the resulting mixture was divided in to 3 parts and
distributed between three Schott flasks (500 mL each) and placed into an oven for 72 h
at 85 °C. The white precipitate was filtered off and washed several times with DMF.
Results and discussion
Synthesis and characterization of pristine and DE-MOF-808
Pristine MOFꢀ808 was prepared following the procedure reported by Furukawa et
20
al
.
Slightly modified synthesis were used to prepare the mixedꢀligand, defect

Catalysis Science & Technology
Page 6 of 16
View Article Online
DOI: 10.1039/C8CY00742J
engineered MOFs, by replacing 10% of the trimesate ligands by the defectꢀinducing
ditopic ligands, as described in detail in the Experimental section. These compounds
will be hereafter referred to as MOFꢀ808ꢀX; where X = IP (isophthalate), Pydc
(
pyridineꢀ3,5ꢀdicarboxylate), OH (5ꢀhydroxyꢀisophthalate), or NH
2
(5ꢀaminoꢀ
isophthalate).
According to the powder Xꢀray diffraction patterns shown in Fig. 2, all the materials
were found to be isostructural and highly crystalline. All the diffraction peaks can be
1
assigned to the expected structure for MOFꢀ808 (space group
F
dꢀ3
m
, spn topology). H
NMR of the digested solids (see Fig.S1 in the ESI) confirmed the incorporation of the
ditopic ligands. Characteristic peaks of the ligands were used to determine the amount
of dicarboxylate ligands incorporated in each case, as summarized in Table S1 (ESI).
Incorporation of the defective ligands was found to be very low in all cases (ca. ~3ꢀ7%).
It was not possible to increase these values even when the amount of defective ligand
introduced in the starting reaction mixture was increased from 10% to 20%. As a
consequence, no significant differences were observed in the FTIR or TGA curves of
pristine and DEꢀMOF compounds. Meanwhile, introduction of ditopic ligands in MOFꢀ
808 produced only minor changes on the textural properties of MOFꢀ808, as
summarized in Table S2 (ESI). For instance, SBET and pore volume varied from 1345
2
ꢀ1
3
ꢀ1
2
ꢀ1
3 ꢀ1
m g and 0.60 cm g for pristine MOFꢀ808, to 1592 m g and 0.74 cm g for the most
porous sample, MOFꢀ808ꢀPydc.
1
0
20
30
40
50
Diffraction angle, 2θ

Page 7 of 16
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C8CY00742J
Fig. 2. Xꢀray diffraction patterns (Cu K
MOFꢀ808ꢀIP, MOFꢀ808ꢀPydc, MOFꢀ808ꢀOH and MOFꢀ808ꢀNH
α
radiation) of (from bottom to top): MOFꢀ808,
2
.
Catalytic properties of pristine and DE-MOF-808
Meerwein-Ponndorf-Verley reduction of cyclohexanone. The catalytic activity of
pristine and DEꢀMOFs for the MPV reduction of carbonyl compounds was first
evaluated using cyclohexanone as a model compound and isopropanol as both reducing
agent and solvent. The results were then compared with a reference UiOꢀ66 material
containing ca. 7% of linker defects measured under the same reaction conditions. Fig. 3
and Table 1 summarize the results obtained.
100
3
80
60
40
20
0
2
1
4
0
,0
0,5
1,0
1,5
2,0
t (h)
Fig. 3. Conversion of cyclohexanone over: MOFꢀ808 (curve 1), MOFꢀ808ꢀIP (curve 2),
MOFꢀ808ꢀPydc (curve 3), and a reference UiOꢀ66 containing ca. 7% missing linker
defects (curve 4).
As it can be seen in Fig. 3, conversion of cyclohexanone is almost complete after 2 h
using the pristine MOFꢀ808 catalyst. The reaction was found to be fully selective, being
cyclohexanol the only product detected in all cases. MOFꢀ808 was found to be much
more active than the reference UiOꢀ66 compound. Thus for instance, cyclohexanone
conversion after 1 h was 69% and 12% over MOFꢀ808 and UiOꢀ66, respectively
(compare entries 1 and 3 in Table 1). In our opinion, this difference in activity shall not
be attributed to eventual diffusion problems in the case of UiOꢀ66, since cyclohexanone
is small enough to enter the pores of UiOꢀ66 (see Fig. S2 in ESI). Indeed, full

Catalysis Science & Technology
Page 8 of 16
View Article Online
DOI: 10.1039/C8CY00742J
cyclohexanone conversion is slowly attained after ca. 24 h over UiOꢀ66 (Fig. S3 in
ESI). Moreover, SEM revealed that both UiOꢀ66 and MOFꢀ808 have similar particle
size (see Fig. S4 in ESI), so that no significant differences are expected in the amount of
sites exposed at the external surface of the crystallites. Rather, the difference in
reactivity between MOFꢀ808 and UiOꢀ66 is most likely due to the higher concentration
of Zr coordinatively unsaturated metal sites in the former material, created upon
removal of the formate anions during catalyst activation. As we have mentioned above,
UiOꢀ66 lacks such open metal sites, and the only coordination positions available are
those associated with missing linkers. Note that, according to the model suggested in
9
ref. for missing linker defects, an UiOꢀ66 sample with 7% of missing linkers defects
contains only 14% of the total Zr ions exposed and available for the reaction (each
missing linker molecule creates two open metal sites, one in each of two neighbour
oxoaggregates). In contrast, all Zr ions are accessible in MOFꢀ808 after removal of the
formate ligands.
To complete our study, and in order to lend further support to our conclusion that the
observed differences in catalytic activity are related directly with the relative amount of
available active sites, we have included a third Zrꢀcontaining MOF in the present study,
3
1
6
DUTꢀ67. In this compound, the Zr clusters are connected to 8 carboxylate ligands,
thus lying between MOFꢀ808 (6ꢀconnected) and UiOꢀ66 (12ꢀconnected). As expected,
the rate of cyclohexanone reduction is also intermediate between that observed for
MOFꢀ808 and UiOꢀ66 (see Fig. S3 in ESI).
a
Table 1. MPV reduction of cylohexanone over various ZrꢀMOFs.
b
Entry
Catalyst
Time (h)
Conv (mol%)
1
2
3
4
5
6
UiOꢀ66
1
24
1
12
>99
69
MOFꢀ808
2
97
MOFꢀ808ꢀIP
1
91
2
>99

Page 9 of 16
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C8CY00742J
7
MOFꢀ808ꢀPydc
1
2
93
8
>99
a
i
Reaction conditions: 10 mg cyclohexanone (0.1 mmol), PrOH (0.5 mL, ca. 6.5 eq)
b
and ZrꢀMOF (5 mg, ca. 14 mol% Zr), 80ºC.
Conversion, determined by GC.
Selectivity to cyclohexanol was >99% in all cases.
Interestingly, the use of DEꢀMOFs, MOFꢀ808ꢀIP and MOFꢀ808ꢀPydc, containing
defectꢀinducing linkers, brings about a marked increase of the catalytic activity
compared with the pristine MOFꢀ808 (see Fig. 3). Thus, the yield of cyclohexanol
obtained after 1 h increased from 69% (MOFꢀ808, entry 3) to 91% and 93% (MOFꢀ808ꢀ
IP and MOFꢀ808ꢀPydc, entries 5 and 7, respectively). FTIR spectroscopy of adsorbed
CO at 77 K (see Fig S5 in ESI) revealed that both, the concentration and relative acid
4
+
strength of the Zr open metal sites are virtually identical in pristine MOFꢀ808 and
MOFꢀ808ꢀPydc compounds. Therefore, we attribute the observed increment of the
reactivity of DEꢀMOFs with respect to the pristine material to the creation of additional
coordination vacancies on the already exposed Zr sites due to the replacement of tritopic
trimesate ligands by the ditopic ligands, thus leaving less sterically crowded Zr active
sites (see Fig. S6 in ESI). In this sense, it is important to stress that the generally
accepted mechanism for the MPV reduction of ketones over Lewis acid catalysts
assumes the simultaneous adsorption of the alcohol and the ketone on the active site and
the direct hydrogen transfer from the alcohol to the ketone, involving a sixꢀmembered
32
cyclic transition state, as shown in Scheme 2. Therefore, it seems reasonable to
hypothesize that less shielded Zr acid sites will form such a transition state more easily.
Further computational studies are currently underway to evaluate this hypothesis in
more detail.
Scheme 2. Proposed transition state for the MPV reduction of cyclohexanone with
isopropanol over Zrꢀcontaining MOFꢀ808 compounds.
In the case of MOFꢀ808ꢀOH and MOFꢀ808ꢀNH
observed with respect to pristine MOFꢀ808 (see Fig. S7 in ESI). This probably indicates
that the presence of –OH and –NH groups in position 5 of the ligands introduces
2
, no significant improvement was
2

Catalysis Science & Technology
Page 10 of 16
View Article Online
DOI: 10.1039/C8CY00742J
further steric hindrance around the Zr sites, similar to what happens in pristine MOFꢀ
08, thereby precluding any improvement of the catalytic activity with respect to the
8
parent material.
Finally, it is important to point out that all the materials were found to be stable
under the reactions conditions used, and no significant differences were observed in the
corresponding XRDs of the materials after reaction (see Fig. S8 in ESI). Consequently,
no significant decrease of the catalytic activity was observed upon reuse of the material
for at least 4 consecutive cycles.
Meerwein-Ponndorf-Verley reduction of estrone (E1). As we have mentioned in the
Introduction, MOFꢀ808 has a wider pore system with respect to UiOꢀ66, with pore
entries of about 14 Å and 6 Å, respectively. From the catalytic point of view, this is a
clear advantage when dealing with the conversion of bulky substrates (with a size
comprised between 6 and 14 Å), which will only react at the external surface of UiOꢀ66
but will freely diffuse inside the pores of MOFꢀ808, thus reaching all the active sites.
In order to evaluate the benefits of a wideꢀpore catalyst such as MOFꢀ808 as
compared with UiOꢀ66, we have studied the MPV reduction of a bulky ketone: estrone
(
E1, approximate dimensions: 11.2 Å x 6.2 Å x 4.2 Å, as extracted from the optimized
3
3
structure using MOPAC2016, see Fig. S9 in ESI). A summary of the results obtained
with various ZrꢀMOF catalysts is shown in Table 2.
a
Table 2. MPV reduction of estrone over various ZrꢀMOFs.
b
c
Entry
Catalyst
UiOꢀ66
Time (h)
Conv (mol%)
d.r.
1
2
3
4
5
48
4
2
n.d.
MOFꢀ808
24
87
61
91
24
4
40:60
38:62
MOFꢀ808ꢀPydc
24

Page 11 of 16
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C8CY00742J
a
i
Reaction conditions: 20 mg estrone (0.08 mmol), PrOH (1 mL, ca. 16 eq) and Zrꢀ
b
MOF (5 mg, ca. 18 mol% Zr), 80ºC. Conversion. Determined by GC. Estradiols were
the only product detected. α−E2:
NMR spectra of the reaction filtrates (for the exact procedure used, see Fig.S10 in ESI).
c
1
β
ꢀE2 diastereomeric ratio, calculated from the H
Under the experimental conditions used, UiOꢀ66 was found to be almost inactive for
the reduction of estrone (only 2% conversion after 48 h). This is most likely due to a
sizeꢀexclusion effect, which prevents the large estrone molecule to cross the ~6 Å
triangular pores to reach the active sites located (mostly) at the internal surface of UiOꢀ
6
6. On the contrary, almost full estrone conversion was attained over the wide pore
MOFꢀ808 materials after 24 h of reaction, with full selectivity to the expected estradiol
E2). As it was the case of cyclohexanone reduction, the introduction of defectiveꢀ
(
inducing pyridinedicarboxylate ligands was found to be beneficial for the catalytic
activity. Thus, the conversion obtained after 4 h of reaction was 24% and 61%,
respectively for MOFꢀ808 and MOFꢀ808ꢀPydc. However, the catalytic performance of
MOFꢀ808ꢀIP was very similar to that of the pristine MOFꢀ808: maximum yield of 78%
after 24 h. This lower activity with respect to MOFꢀ808ꢀPydc might be related to a
faster deactivation associated to product adsorption, although more experiments would
be necessary to reach a more satisfactory explanation.
In general, the synthesis of 17
corresponding 17ꢀoxo compound is challenging. Usually, the 17
almost exclusively when NaBH or other reducing agents are employed, due to the
αꢀhydroxy steroids (such as
α-E2) by reduction of the
β
isomer is obtained
4
steric hindrance imposed by the 18ꢀmethyl group on the approaching direction of the
reducing agent. Additional reactions are then required to produce the inversion of the
3
4
17
β
ꢀhydroxy steroids to the 17
involve multiple protectionꢀdeprotection steps. And even then, the yields and
selectivities to the 17 ꢀOH steroids are only moderate at best. For example, Han et al.
reported on the synthesis of a related 17 ꢀOH steroid (5 ꢀandrostaneꢀ3 ,17 ꢀdiol) from
the 17ꢀoxo compound by a 4ꢀsteps reaction involving reduction, tosylation, acetate
α compounds through Mitsunobu reactions, which
α
α
α
β
α
35
substitution and hydrolysis, with a final yield of the 17
al. reported later a modification of this procedure by adding three additional steps:
epoxidation with peracetic acid, reduction with NaBH and hydrolysis with NaOH. The
reported final yield of the 17 ꢀOH compound was 54%. Göndös and Orr followed a
different approach to 17 ꢀOH steroid compounds by using a chiral Rh complex as
α product as low as 4%. Ohta et
4
36
α
α

Catalysis Science & Technology
Page 12 of 16
View Article Online
DOI: 10.1039/C8CY00742J
stereoselective reducing agent, yielding the 17αꢀhydroxyꢀestroneꢀ3ꢀmethyl ether in
37
3
6%, together with the 17
β
ꢀcompound in 26% yield.
Given the elusive character of the 17αꢀhydroxy steroid derivatives, it is evident that
development of alternative preparative methods would be highly desirable. In this sense,
it would be very convenient to use easily recoverable (and cheap) solid catalysts, along
with the design of single step reaction processes and minimization of waste generation
and use of additional reagents.
Therefore, given the well known high chemoꢀ, regioꢀ and stereoselectivity attainable
32
with MPV reactions, we have evaluated the diastereomeric ratio between alphaꢀ and
betaꢀestradiol ( -E2 and -E2) obtained using both MOFꢀ808 and MOFꢀ808ꢀPydc as
α
β
catalysts. Interestingly, we found that both materials afforded a mixture of the two
alcohols in a ca. 40:60 α:β ratio, which translates into an overall yield of ca. 35% of the
α
-E2 isomer after 24 h of reaction time. This finding is most probably related with the
relative steric hindrance of the two transition states leading to ꢀE2 and ꢀE2 products
inside the restricted space of the MOF cavities, similar to what was already pointed out
α
β
38, 39
for MPV reactions taking place inside the pores of zeolites.
stereoselectity attained with MOFꢀ808 compounds is still far from optimal, it is worth
mentioning that 17 ꢀhydroxyꢀestradiol was cleanly obtained directly from estrone with
moderate yields in a single reaction step, using only isopropanol as the sole reagent and
avoiding any additional protection/deprotection steps. Isolation of pure and isomers
Although the
α
α
β
from the reaction filtrate can be readily achieved by TLC or column purification without
any further workout.
Conclusions
Herein we have reported on the synthesis of various defectꢀengineered MOFꢀ808
materials by mixing the tricarboxylate trimesate ligands of MOFꢀ808 with dicarboxylate
ligands; viz. isophthalate (IP), pyridineꢀ3,5ꢀdicarboxylate (Pydc), 5ꢀhydroxyꢀ
2
isophthalate (OH), or 5ꢀaminoꢀisophthalate (NH ). The resulting mixedꢀligand
compounds, MOFꢀ808ꢀX (X = IP, Pydc, OH or NH ) were all found to be highly
2
crystalline and isostructural to the unmodified MOFꢀ808.
The catalytic activity of pristine and mixedꢀligand MOFꢀ808 compounds have been
evaluated for the MeerweinꢀPonndorfꢀVerley (MPV) reduction of carbonyl compounds
using isopropanol as reducing agent, and the results were compared with a reference

Page 13 of 16
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C8CY00742J
UiOꢀ66 material containing ca. 7% of linker defects. When a small molecule such as
cyclohexanone was used as substrate, MOFꢀ808 was found to be more active than UiOꢀ
66. The higher activity of MOFꢀ808 with respect to UiOꢀ66 was attributed to the fact
4
+
that in MOFꢀ808, all Zr ions are accessible and can participate in the reaction (once
the bridging formate ions are removed). Meanwhile, the only accessible acid sites in
UiOꢀ66 are those associated with linker defects (which represent only 14% of the total
Zr ions of the solid). Partial replacement of trimesate linkers in MOFꢀ808 by ditopic
ligands results in a further improvement of the catalytic performance with respect to the
pristine MOFꢀ808. This is probably related to the creation of less hindered Zr open
metal sites when dicarboxylate ligands are used, which facilitate the formation of the 6ꢀ
membered cyclic transition state (see Scheme 2). The presence of wide pores in MOFꢀ
808 with respect to UiOꢀ66 translates into a significant improvement of the catalytic
activity for converting bulky substrates, as it is shown here for the reduction of estrone
to estradiol. Interestingly, the process produces reasonable amounts of the elusive 17
hydroxy estradiol in a single reaction step.
αꢀ
Conflicts of interest
There are no conflicts of interest to declare.
Acknowledgements
This project has received funding from the European Union’s Horizon 2020 research
and innovation programme under the Marie SklodowskaꢀCurie grant agreement No.
641887 (project acronym:DEFNET). Financial support from the Spanish Ministry of
Economy and Competitiveness (program Severo Ochoa SEV20120267), the Spanish
Ministry of Science and Innovation (project MAT2014ꢀ52085ꢀC2ꢀ1ꢀP), and the German
Research Foundation (project KA 1698/19ꢀ1) is also gratefully acknowledged. The
Microscopy Service of the Universitat Politècnica de València are gratefully
acknowledged for the SEM images.
References
1
2
.
.
J. L. C. Rowsell and O. M. Yaghi, Microporous Mesoporous Mater., 2004, 73, 3-14.
Z. Q. Wang and S. M. Cohen, Chem. Soc. Rev., 2009, 38, 1315-1329.

Catalysis Science & Technology
Page 14 of 16
View Article Online
DOI: 10.1039/C8CY00742J
3
4
.
.
A. D. Burrows, in Metal Organic Frameworks as Heterogeneous Catalysts, eds. F. X.
Llabrés i Xamena and J. Gascon, The Royal Society of Chemistry, Cambridge, 2013, ch.
3
, pp. 31-75.
S. M. J. Rogge, A. Bavykina, J. Hajek, H. Garcia, A. I. Olivos-Suarez, A. Sepúlveda-
Escribano, A. Vimont, G. Clet, P. Bazin, F. Kapteijn, M. Daturi, E. V. Ramos Fernandez, F.
X. Llabrés i Xamena, V. Van Speybroeck and J. Gascon, Chem. Soc. Rev., 2017, 46, 3134-
3
184.
J. Gascon, A. Corma, F. Kapteijn and F. X. Llabrés i Xamena, ACS Catal., 2014, 4, 361-
78.
5
.
3
6
7
.
.
A. Corma, H. Garcia and F. X. Llabrés i Xamena, Chem. Rev., 2010, 110, 4606-4655.
J. Y. Lee, O. K. Farha, J. Roberts, K. A. Scheidt, S. T. Nguyen and J. T. Hupp, Chem. Soc.
Rev., 2009, 38, 1450-1459.
8
.
J. H. Cavka, S. Jakobsen, U. Olsbye, N. Guillou, C. Lamberti, S. Bordiga and K. P. Lillerud,
J. Am. Chem. Soc., 2008, 130, 13850-13851.
9
1
.
0.
F. G. Cirujano, A. Corma and F. X. Llabrés i Xamena, Chem. Eng. Sci., 2015, 124, 52-60.
V. L. Rechac, F. G. Cirujano, A. Corma and F. X. Llabrés i Xamena, Eur. J. Inorg. Chem.,
2
016, 2016, 4512-4516.
1
1.
C. Rösler, S. Dissegna, V. L. Rechac, M. Kauer, P. Guo, S. Turner, K. Ollegott, H.
Kobayashi, J. Yamamoto, D. Peeters, Y. Wang, S. Matsumura, G. Van Tendeloo, H.
Kitagawa, M. Muhler, F. X. Llabrés i Xamena and R. A. Fischer, Chem. Eur. J, 2017, 23,
3
583-3594.
F. Vermoortele, R. Ameloot, A. Vimont, C. Serre and D. De Vos, Chem. Commun., 2011,
7, 1521-1523.
1
1
2.
3.
4
F. Vermoortele, M. Vandichel, B. Van de Voorde, R. Ameloot, M. Waroquier, V. Van
Speybroeck and D. E. de Vos, Angew. Chem., Int. Ed., 2012, 51, 4887-4890.
H. Wu, T. Yildirim and W. Zhou, J. Phys. Chem. Lett., 2013, 4, 925-930.
S. J. Garibay and S. M. Cohen, Chem. Commun., 2010, 46, 7700-7702.
M. Kandiah, M. H. Nilsen, S. Usseglio, S. Jakobsen, U. Olsbye, M. Tilset, C. Larabi, E. A.
Quadrelli, F. Bonino and K. P. Lillerud, Chem. Mater., 2010, 22, 6632-6640.
M. Kandiah, S. Usseglio, S. Svelle, U. Olsbye, K. P. Lillerud and M. Tilset, J. Mater.
Chem., 2010, 20, 9848-9851.
1
1
1
4.
5.
6.
1
1
1
7.
8.
9.
Y. Bai, Y. Dou, L. H. Xie, W. Rutledge, J. R. Li and H. C. Zhou, Chem. Soc. Rev., 2016, 45,
2
327-2367.
F. Vermoortele, B. Bueken, G. Le Bars, B. Van de Voorde, M. Vandichel, K. Houthoofd,
A. Vimont, M. Daturi, M. Waroquier, V. Van Speybroeck and D. E. de Vos, J. Am. Chem.
Soc., 2013, 135, 11465-11468.
2
2
0.
1.
H. Furukawa, F. Gandara, Y. B. Zhang, J. Jiang, W. L. Queen, M. R. Hudson and O. M.
Yaghi, J. Am. Chem. Soc., 2014, 136, 4369-4381.
J. E. B. Mondloch, W., D. Fairen-Jimenez, S. Kwon, E. J. DeMarco, M. H. Weston, A. A.
Sarjeant, S. T. Nguyen, P. C. Stair, R. Q. Snurr, O. K. Farha and J. T. Hupp, J. Am. Chem.
Soc., 2013, 135, 10294-10297.
2
2
2
2
2
2.
3.
4.
5.
6.
D. Feng, Z. Y. Gu, Y. P. Chen, J. Park, Z. Wei, Y. Sun, M. Bosch, S. Yuan and H. C. Zhou, J.
Am. Chem. Soc., 2014, 136, 17714-17717.
S. Yuan, W. Lu, Y.-P. Chen, Q. Zhang, T.-F. Liu, D. Feng, X. Wang, J. Qin and H.-C. Zhou,
J. Am. Chem. Soc., 2015, 137, 3177-3180.
V. Bon, V. Senkovskyy, I. Senkovska and S. Kaskel, Chem. Commun., 2012, 48, 8407-
8
409.
A. Schaate, P. Roy, T. Preuße, S. J. Lohmeier, A. Godt and P. Behrens, Chem. Eur. J,
011, 17, 9320-9325.
S. Marx, W. Kleist and A. Baiker, J. Catal., 2011, 281, 76-87.
2

Page 15 of 16
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C8CY00742J
2
2
7.
8.
O. Kozachuk, I. Luz, F. X. Llabrés i Xamena, H. Noei, M. Kauer, H. Bauke Albada, E. D.
Bloch, B. Marler, Y. Wang, M. Muhler and R. A. Fischer, Angew. Chem., Int. Ed., 2014,
5
3, 7058-7062.
W. Zhang, M. Kauer, O. Halbherr, K. Epp, P. Guo, M. I. Gonzalez, D. J. Xiao, C. Wiktor, F.
X. Llabrés i Xamena, C. Wöll, Y. Wang, M. Muhler and R. A. Fischer, Chem. Eur. J, 2016,
2
2, 14297-14307.
E. Plessers, G. Fu, C. Y. X. Tan, D. E. de Vos and M. B. J. Roeffaers, Catalysts, 2016, 6,
04.
L. Valenzano, B. Civalleri, S. Chavan, S. Bordiga, M. H. Nilsen, S. Jakobsen, K. P. Lillerud
and C. Lamberti, Chem. Mater., 2011, 23, 1700-1718.
2
3
3
3
3
9.
0.
1.
2.
3.
1
V. Bon, I. Senkovska, I. A. Baburin and S. Kaskel, Cryst. Growth Des., 2013, 13, 1231-
1
237.
C. F. de Graauw, J. A. Peters, H. Van Bekkum and J. Huskens, Synthesis, 1994, 10, 1007-
017.
1
MOPAC2016, HTTP://OpenMOPAC.net/, Version 17.048W, J.J.P. Stewart, Stewart
Computational Chemistry, Colorado Springs, CO, USA
3
3
3
3
3
3
4.
5.
6.
7.
8.
9.
J. A. Dodge and C. W. Lugar III, Bioorg. Med. Chem. Lett., 1996, 6, 1-2.
G.-D. Han, Z.-Y. Lin and C.-P. Yan, Zhongguo Yiyao Gongye Zazhi, 1994, 25, 490-491.
T. Ohta, H. Zhang, Y. Torihara, T. Hida and I. Furukawa, Steroids, 1998, 63, 630-632.
G. Göndös and J. C. Orr, J. Chem. Soc., Chem. Commun., 1982, 1238-1239.
J. C. van der Waal, K. Tan and H. Van Bekkum, Catal. Lett., 1996, 41, 63-67.
A. Corma, M. E. Domine, L. Nemeth and S. Valencia, J. Am. Chem. Soc., 2002, 124,
3
194-3195.

Catalysis Science & Technology
Page 16 of 16
View Article Online
Table of contents entry
DOI: 10.1039/C8CY00742J
Defect-engineered Zr-MOF-808 are superior catalysts for Meerwein-Ponndorf-Verley
reduction of (bulky) carbonyls