Network topology and cavity confinement-controlled diastereoselectivity in cyclopropanation reactions catalyzed by porphyrin-based MOFs

Article abstract of DOI:10.1039/c9cy00893d

In this work, we show that the stereoselectivity of a reaction can be controlled by directing groups of substrates, by network topology and by local cavity confinement of metal-organic framework (MOF) catalysts. We applied the porphyrin-based PCN-224(Rh), which contains no stereocenters in the cyclopropanation reaction using ethyl diazoacetate (EDA) as carbene source. When styrene and other non-coordinating olefins are used as substrates, high activity, but no diastereoselectivity is observed. Interestingly, conversion of 4-amino- and 4-hydroxystyrene substrates occurs with high diastereomeric ratios (dr) of up to 23 : 1 (trans : cis). We attribute this to local pore confinement effects as a result of substrate coordination to neighboring Rh-centers, which position the olefin with respect to the active site, causing a break of local symmetry of the coordinated substrate. The effect of local pore confinement was improved by using PCN-222(Rh) as catalyst, which is a structural analog of PCN-224(Rh) with characteristic Kagomé topology featuring shorter Rh-Rh distances. A remarkable dr of 42 : 1 (trans : cis) was observed for 4-aminostyrene. In this case, the length of the substrate corresponds to the average distance between two neighboring Rh centers within the pores of PCN-222(Rh), which drastically boosts the diastereoselectivity. This work showcases how diastereomeric control can be achieved by favorable substrate-catalyst interactions and thoughtful adjustment of confined reaction space using porphyrin-based MOFs, in which stereocenters are inherently absent.

Full text of DOI:10.1039/c9cy00893d

Catalysis
Science &
Technology
View Article Online
View Journal | View Issue
PAPER
Network topology and cavity confinement-
controlled diastereoselectivity in
cyclopropanation reactions catalyzed by
porphyrin-based MOFs†
Cite this: Catal. Sci. Technol., 2019,
9, 6452
Konstantin Epp,a Bart Bueken, b Benjamin J. Hofmann,c Mirza Cokoja,a
a
Karina Hemmer,a Dirk De Vos b and Roland A. Fischer
*
In this work, we show that the stereoselectivity of a reaction can be controlled by directing groups of
substrates, by network topology and by local cavity confinement of metal–organic framework (MOF)
catalysts. We applied the porphyrin-based PCN-224IJRh), which contains no stereocenters in the
cyclopropanation reaction using ethyl diazoacetate (EDA) as carbene source. When styrene and other non-
coordinating olefins are used as substrates, high activity, but no diastereoselectivity is observed.
Interestingly, conversion of 4-amino- and 4-hydroxystyrene substrates occurs with high diastereomeric
ratios (dr) of up to 23: 1 (trans : cis). We attribute this to local pore confinement effects as a result of
substrate coordination to neighboring Rh-centers, which position the olefin with respect to the active site,
causing a break of local symmetry of the coordinated substrate. The effect of local pore confinement was
improved by using PCN-222IJRh) as catalyst, which is a structural analog of PCN-224IJRh) with characteristic
Kagomé topology featuring shorter Rh–Rh distances. A remarkable dr of 42: 1 (trans : cis) was observed for
4-aminostyrene. In this case, the length of the substrate corresponds to the average distance between two
neighboring Rh centers within the pores of PCN-222IJRh), which drastically boosts the diastereoselectivity.
This work showcases how diastereomeric control can be achieved by favorable substrate–catalyst
interactions and thoughtful adjustment of confined reaction space using porphyrin-based MOFs, in which
stereocenters are inherently absent.
Received 9th May 2019,
Accepted 8th October 2019
DOI: 10.1039/c9cy00893d
rsc.li/catalysis
BINAP,3 or chiral multidentate carboxylates, as is the case
with the Sharpless epoxidation catalyst.4 The design of
1 Introduction
Stereoselective catalysis is
a
key technology due to its
heterogeneous stereoselective catalysts is, however, more
intricate. In heterogeneous catalysts, the stereoselectivity
usually originates from a molecular catalyst immobilized on a
solid support. The high tuneability of metal–organic
frameworks (MOFs) as heterogeneous catalysts allows much
versatility in tailoring and manipulating their structural and
chemical properties. In the context of stereoselective catalysis,
MOF design approaches involve either the heterogenization of
stereoselective molecular catalyst to the MOF linker,5 or using
linkers6 that themselves act as stereoselective ligands to
catalytically active metals.7,8
While the classic ‘catalyst-controlled’ concept for both
homogeneous and heterogenized molecular catalysts requires
well-defined stereocenters, very few reports describe the
stereoselectivity of solid catalyst@host materials, in which
neither the catalyst nor the host contain such well-defined
stereocenters. In this case, the induction of stereoselectivity
is derived from pore confinement effects of the host. While
this type of stereocontrol is known for soluble, defined
enormous economic relevance towards the production of
pharmaceuticals, agrochemicals, fungicides, pheromones,
flavors and fragrances.1 In stereoselective organometallic
catalysis, one of the most exploited design principles is
‘catalyst control’, i.e. the coordination of catalytically active
metal centers by chiral ligands to drive stereoselectivity,2 such
as diamines or diphosphines, for instance in the well-known
a Chair of Inorganic and Metal-Organic Chemistry, Catalysis Research Center and
Department of Chemistry, Technical University of Munich, Ernst-Otto-Fischer-
Straße 1, D-85748 Garching bei München, Germany.
E-mail: roland.fischer@tum.de
b Centre for Surface Chemistry and Catalysis, Department of Microbial and
Molecular Systems (M2S), KU Leuven, Celestijnenlaan 200F p.o. box 2461, 3001
Leuven, Belgium
c Molecular Catalysis, Catalysis Research Center and Department of Chemistry,
Technical University of Munich, Ernst-Otto-Fischer-Straße 1, D-85748 Garching bei
München, Germany
† Electronic supplementary information (ESI) available. See DOI: 10.1039/
c9cy00893d
6452 | Catal. Sci. Technol., 2019, 9, 6452–6459
This journal is © The Royal Society of Chemistry 2019

View Article Online
Catalysis Science & Technology
Paper
macromolecular cage compounds and others,9,10 only a
handful examples using solid/polymeric porous materials
such as zeolites,11,12 mesoporous silica13,14 or MOFs15–18 are
known. ‘Confinement’ or ‘confinement effects’ are rather
broad terms and not explicitly specified in literature. Hence,
it is not exactly clear how the confinement of a host defines
the stereochemical outcome of a reaction, since the local
substrate arrangement through its interaction within the host
are a multi-component problem.19 In many homogeneously
catalysed reactions, the stereo- or regioselectivity can be
influenced by interactions of ‘directing’ functional groups of
the substrate with the catalyst, e.g. via hydrogen-, covalent- or
coordinative bonds, or Coulomb or Lewis acid–base
interactions, which induce a preferential conformation or
orientation of the substrate at the catalyst.20 Therefore, it is
reasonable to assume that directing effects of functional
groups attached to substrates may also play a crucial role in
MOF-based catalysis by coordination to suitable sites located
within the pores. A thorough understanding of the origin of
stereocontrol in these cases is of great interest, as the
insights gained could enable further control over
confinement effects in MOF-based catalysts.
PCN-222IJRh), constructed from the same building blocks, i.e.
Zr6 inorganic nodes and MP linkers, but with different
topologies influence the stereoselectivity of
a catalytic
reaction. The utilization of unaltered building units,
especially the MP linkers as the catalytically active sites,
allows direct comparison of the catalytic results. Hence, the
difference in stereoselectivity can be attributed to the
different confinement effects of altered pore sizes and
geometries.
Since MPs are used in the cyclopropanation of olefins, we
choose this as a reaction. Cyclopropane motifs are not only
found in natural products; they are also important synthetic
targets in pharmaceuticals like antibiotics, in perfume
compounds and in biomimetic insect repellents.40,41 The
specific reactivity of the strained three-membered ring system
and its ability to induce conformational constraints on
otherwise flexible acyclic chains are key features of this
motif.41–43 The transition-metal catalyzed cyclopropanation of
double bonds using diazo compounds as a carbene source is
well studied and has been recognized as a useful design
route for substituted cyclopropanes. Since there is a vast
number of reports using Rh(II, III) catalysts in the CP
reaction,25–27,44,45 we decided to start with the RhIJIII)-
As we have been working with metalloporphyrin (MP)
MOFs as catalysts for the cycloaddition of carbon dioxide to
propylene oxide to form propylene carbonate,21 we extended
their application as heterogeneous catalysts to (dia)
stereoselective reactions. Molecular MPs have been widely
studied in literature, because of their ability to act as ligands
for various metals (M = Fe(II, III),22,23 CoIJII),24 RhIJIII),25–27
RuIJII),28 IrIJIII),29 OsIJIII)30 and others). Thus, it is not surprising
that their chemistry and catalytic properties are rich.31,32 To
be used in stereoselective catalysis, MPs are often
functionalized with bulky26,33 and/or chiral groups34–36 at the
meso-positions of the porphyrin backbone in order to induce
stereochemical or even chiral information and most
importantly, to engineer a pocket-like environment around
the catalytically active MP center. On the one hand catalytic
metalated
porphyrin,
namely
[5,10,15,20-tetrakisIJ4-
methoxycarbonylphenyl)porphyrinato]-RhIJIII) chloride (RhIJT-
CPPCO2Me)Cl) as a homogeneous reference system in the
cyclopropanation of styrene with ethyl diazoacetate (EDA) to
the corresponding cyclopropanation products (trans and cis).
Consecutively, we compared these results with those obtained
on stable heterogeneous MP-based PCN-224IJRh) and PCN-
222IJRh) catalysts with regard to activity and dr.
2 Experimental section
2.1 Materials and methods
All chemicals were purchased from commercial sources and
used without any purification. Benzoic acid, benzonitrile,
activities may increase to TOFs
>
100 000 h−1 as
cyclooctene,
diazoacetate
4-nitrostyrene,
(EDA), 4-hydroxystyrene,
4-tert-butylstyrene,
2-, 3-
ethyl
and
demonstrated by Gallo et al. using ‘totem’-shaped FeIJIII)-
porphyrins in the cyclopropanation of styrene,33 while on the
other hand synthetic yields of these sophisticated MPs are
extremely low (often <1%) and purification is necessary to
isolate the compounds. Moreover, deactivation via
dimerization may hinder their application in catalysis.
Motivated by the remarkable catalytic properties of
molecular MPs and the challenges in their synthesis, stability
and recyclability, we were interested to investigate how their
heterogeneous counterparts, namely, MP-MOFs, would
perform as catalysts. Heterogenization is achieved when MPs
are built into the structure of the framework as organic
linkers, as in the Zr-MOFs PCN-222,37 PCN-22438 and MOF-
525.39 Consequently, MP-MOFs can be seen as rigid, self-
supported heterogeneous catalysts with well-defined, spatially
distributed MP-sites, which are catalytically active and not
susceptible to dimerization. Thus, here we investigate for the
first time how different MP-MOFs phases, PCN-224IJRh) and
4-aminostyrene were purchased from Sigma Aldrich, RhCl3
from Precious Metals Online, methyl 4-formylbenzoate from
Merck, pyrrole, styrene, 1-octene, 4-chlorostyrene, 3,4-
dihydro-2H-pyran and N,N-diethylformamide (DEF) from TCI,
propionic acid, and N,N-dimethylformamide (DMF), N,N-
dimethyl-4-aminostyrene from abcr and ZrCl4 from Alfa Aesar.
The Rh linker was synthesized with slight changes to the
published procedures.46 The Rh-porphyrin ester RhIJT-
CPPCO2Me)Cl was prepared by mixing TCPPCO2Me (168.2
mg, 0.2 mmol) and RhCl3 (104.2 mg, 0.8 mmol) in
benzonitrile (5 mL) under stirring at 220 °C for 2 h.
Benzonitrile was removed under reduced pressure and the
residue was dried at 80 °C overnight. The resulting
RhIJTCPPCO2Me)Cl was hydrolyzed (120 mg) by refluxing a
mixture of THF (6 mL) and MeOH (6 mL), to which a solution
of KOH (100 mg) in H2O (6 mL) was added within 5 h. After
the reaction mixture was cooled down to room temperature,
This journal is © The Royal Society of Chemistry 2019
Catal. Sci. Technol., 2019, 9, 6452–6459 | 6453

View Article Online
Paper
Catalysis Science & Technology
the organic solvents were removed by rotary evaporation.
Additional water was added to the residual product mixture
and acidified with 1 M HCl until no further precipitation of
RhIJTCPPCO2H)Cl was observed. The red product was washed
with water (3 × 50 mL) and dried in vacuum overnight.
solution for additional 24 h to ensure sufficient solvent
exchange. The powder was separated by centrifugation and
activated at 120 °C for 24 h in dynamic vacuum. Anal. calcd.
(%) for PCN-224: C, 42.01; H, 2.45; N, 4.08%. Found: C,
44.51; H, 2.74; N, 3.77%.
Synthesis of PCN-224IJRh). In a Pyrex vial ZrCl4 (30 mg),
RhIJTCPPCO2H)Cl (10 mg) and benzoic acid (400 mg) were
ultrasonically dissolved in DMF (2 mL). The mixture was
heated in an oven at 120 °C for 24 h. After cooling down to
room temperature, orange-red crystals were collected by
filtration, washed with DMF (3 × 20 mL) and acetone (3 × 20
mL), and transferred in acetone solution for 24 h. The
powder was separated by centrifugation and activated at 120
°C for 24 h in dynamic vacuum. Anal. calcd. (%) for PCN-
224IJRh): C, 39.49; H, 2.16; N, 3.84%; found: C, 40.20; H, 3.01;
N, 3.92%.
Synthesis of PCN-222. In a Pyrex vial ZrCl4 (70 mg), H2-
TCPP (50 mg), and benzoic acid (2.7 g) were ultrasonically
dissolved in DMF (8 mL). The mixture was heated in an oven
at 120 °C for 2 d. After cooling down to room temperature,
violet powder was collected by filtration and activated with
1.5 mL 8 M HCl in DMF at 120 °C for 12 h to remove residual
modulator, followed by washing with DMF (3 × 20 mL) and
acetone (3 × 20 mL) and transferred into acetone solution for
24 h. The powder was separated by centrifugation and
activated at 120 °C for 24 h in dynamic vacuum. Anal. calcd.
(%) for PCN-222: C, 48.27; H, 2.70; N, 4.69%. Found: C,
46.36; H, 2.68; N, 5.40%.
Synthesis of PCN-222IJRh). In a Pyrex vial ZrCl4 (70 mg),
Rh(TCPP)Cl (50 mg) and benzoic acid (2.7 g) were
ultrasonically dissolved in DMF (8 mL). The mixture was
heated in an oven at 120 °C for 2 d. After cooling down to
room temperature, the resulting dark red powder was
collected by filtration, washed with DMF (6 × 20 mL) and
acetone (6 × 20 mL) and transferred into acetone solution for
48 h. During that time, acetone was exchanged every 12
hours. The powder was separated by centrifugation and
activated at 120 °C for 24 h in dynamic vacuum. Anal. calcd
(%) for PCN-222IJRh): C, 44.94; H, 2.36; N, 4.28%; found: C,
46.90; H, 2.66; N, 4.41%.
2.2 Powder X-ray diffraction (PXRD)
Powder X-ray diffraction measurements were performed
using Bragg–Brentano geometry in a PANalytical Empyrean
diffractometer equipped with
a PANalytical PIXcel 1D
detector. X-ray Cu Kα radiation (λ1 = 1.5406 Å, λ2 = 1.5444 Å,
I2/I1 = 0.5) was used for the measurements. Kβ radiation was
removed with a Ni filter. Voltage and intensity were 45 kV
and 40 mA, respectively. The measurement range was from
5.0° to 70.0° (2θ) with a step size of 0.040° (2θ) and an
acquisition time of 35 seconds per step. The measurement
was performed at 298 K and the sample was rotated on a
reflection-transmission spinner during the measurement (0.5
rps). High-throughput powder X-ray diffraction patterns were
recorded
on
a
STOE
COMBI
P
diffractometer
(monochromated Cu Kα1 radiation, λ = 1.54060 Å) equipped
with an IP-PSD detector in transmission geometry.
2.3 N2-Physisorption (Brunauer–Emmett–Teller, BET)
Nitrogen physisorption measurements were performed on a
Micromeritics 3Flex surface analyzer and carried out at 77 K.
Prior to the measurements, the samples (∼100 mg) were
activated for 12 h at 393 K in vacuum. The specific surface
area of the materials was calculated using the multipoint
BET method (Brunauer–Emmett–Teller) applied to the
isotherm adsorption branch, while taking into account the
Rouquerol consistency criteria.47
2.4 NMR spectra
1H and 13C NMR spectra were recorded on a Bruker AMX-300
spectrometer at 300 MHz and 75 MHz, respectively (16 and
1024 scans, respectively).
2.5 UV-VIS
UV-VIS spectroscopy was performed on an Agilent Cary 60
UV-VIS instrument.
2.8 Synthesis and characterisation
PCN-224IJRh) was obtained by solvothermal synthesis. A
metalated Rh-porphyrin linker46 was synthesized by
saponification of the RhIJTCPPCO2Me)Cl ester. Metalation was
confirmed via UV-VIS spectroscopy (see the ESI,† Fig. S1). The
2.6 Energy dispersive spectroscopy (EDS)
Measurements were performed on
a Hitachi TM-1000
Tabletop Microscope with an acceleration voltage of 15 kV
and an acquisition time of 200 seconds.
obtained
linker
[5,10,15,20-tetrakisIJ4-carboxyphenyl)-
porphyrinato]RhIJIII) chloride, ZrCl4 and benzoic acid as
modulator were mixed in DMF and heated for 1 day at 120
°C.38 Accordingly, PCN-222IJRh) was synthesized utilizing
lower Zr : linker ratios and extended reaction times.38
Experimental Zr : Rh molar ratios of the MP-MOFs were
determined by energy dispersive spectroscopy (EDS; see ESI,†
Fig. S2 and S3 and Table S1) and are in good agreement with
the calculated ones derived from the idealized sum formulas.
2.7 Material synthesis
Synthesis of PCN-224. In a Pyrex vial ZrCl4 (30 mg), H2-
TCPP (10 mg) and benzoic acid (400 mg) were ultrasonically
dissolved in DMF (2 mL). The mixture was heated in an oven
at 120 °C for 24 h. After cooling down to room temperature,
purple crystals were collected by filtration, washed with DMF
(3 × 20 mL) and acetone (3 × 20 mL), and placed in acetone
6454 | Catal. Sci. Technol., 2019, 9, 6452–6459
This journal is © The Royal Society of Chemistry 2019

View Article Online
Catalysis Science & Technology
Paper
Fig. 3 N2-Physisorption measurement of PCN-222IJRh) measured at
77 K.
Fig. 1 Experimental (exp.) powder X-ray diffraction pattern of PCN-
224(Rh, Fe) compared to non-metalated exp. PCN-224 and the
simulated (sim.) pattern of PCN-224IJNi) calculated from
a model
obtained from single crystal data.
literature Sa of non-metalated PCN-222 = 2223 m2 g−1),37
showing an expected additional step in the isotherm
compared to PCN-222 which could be attributed to the large
hexagonally shaped 3D mesoporous channels (see ESI,† Fig.
S6–S9 for pore size distributions).
PCN-224IJRh) was obtained as phase pure micro-crystalline
powder, which was confirmed by powder X-ray diffraction
(PXRD, Fig. 1) (see ESI,† Fig. S4 for the PXRD of PCN-
222IJRh)). The comparison to the patterns of parent non-
metalated MOFs and to the simulated powder patterns reveal
that all MP-MOFs are isostructural to each other. A type I
isotherm and a Brunauer–Emmett–Teller (BET) surface area
(Sa) of 1400 m2 g−1 were found for PCN-224IJRh) (Fig. 2). This
value is lower than, but still in the same order of magnitude
as for non-metalated PCN-224 (Sa = 2147 m2 g−1). This might
indicate slightly incomplete activation prior to the
measurement, or possibly the presence of additional Rh-ions,
which would increase the molar weight but hardly affect the
pore volume. Pore size distributions (PSD), BET values and
pore volumes are summarized in ESI,† Table S2.
2.9 Catalytic tests
As a test reaction, we selected the cyclopropanation (CP) of
olefins (see Scheme 1) using ethyl diazoacetate (EDA) as
carbene source. The products were analyzed by 1H and 13C
NMR spectroscopy (see ESI,† Fig. S10 and S11).
First, we tested RhIJTCPPCO2Me)Cl as a homogeneous
reference system. In the catalytic reaction 135 μL EDA (1.1
mmol) was diluted in 2 mL CH2Cl2 and added manually via a
syringe to a solution of 75 μL styrene (0.8 mmol) in 3 mL
CH2Cl2 within 20 min. Further manual introduction of EDA
did not lead to full conversion. The activity at room
temperature (r.t.) under stirring in CH2Cl2 is moderate (yield
= 20% after 22 h; see Table 1). Expectedly, there is no
diastereoselectivity (1 : 1), since this porphyrin ligand is
nearly planar and lacks directing bulky groups attached at
the meso-position of the porphyrin backbone.
The synthesized micro- and mesoporous PCN-222IJRh)
sample exhibited a surface area of 1912 m2 g−1 (Fig. 3;
Secondly, we used the heterogeneous counterpart PCN-
224IJRh) as catalyst under the same reaction conditions as
above and found a similar yield (29% after 22 h). The
Scheme
1 Cyclopropanation of styrene using ethyl diazoacetate
(EDA) to the corresponding diastereomeric cyclopropanation products
(trans, cis and their corresponding enantiomers (not shown)) under Rh-
MOF catalysis.
Fig.
2 N2-Physisorption measurements of PCN-224(Rh, no metal)
measured at 77 K.
This journal is © The Royal Society of Chemistry 2019
Catal. Sci. Technol., 2019, 9, 6452–6459 | 6455

View Article Online
Paper
Catalysis Science & Technology
Table 1 Catalytic results for the cyclopropanation of olefins and ethyl
diazoacetate (EDA) using MP-MOFs as catalyst
which makes it relevant in terms of sustainability, showed a
moderate yield of 20%.
Interestingly, a high dr of 23 : 1 (trans : cis) was observed
with 4-aminostyrene (4-AS) as substrate. Presumably, the
amino functionality acts as an anchoring ligand to the Rh-
porphyrin linkers and/or via hydrogen bonding to the Zr-oxo-
clusters, as it is indicated by infrared spectroscopy (see the
ESI,† Fig. S13 and S14). Such a coordination mode is likely to
align the substrate in a certain position, and sterically
orienting the olefin moiety to the Rh-carbene in a favored
fashion. In this sense, the reaction can be considered
catalyst-controlled, because of the well-defined and rigid
RhIJZr)–Rh distances favoring certain olefin orientations. On
the other hand, the reaction can be viewed as substrate-
controlled, whereby the coordination ability of functional
groups on the substrates drives the stereoselectivity. In this
example, the linkers and/or nodes carrying functional groups
may act as binding sites facilitating substrate–catalyst
interactions for possible enrichment fixation, and activation
of substrates, providing special transfer pathways for
No.
Olefin
Catalysta
Yieldc [%]
drd
1
2
3
4
5
6
7
8
Styrene
Rh-TCCPOMe
Rh-TCCPOMe
Rh-TCCPOMe
Rh-TCCPOMe
Rh-TCCPOMe
Rh-TCCPOMe
Rh-TCCPOMe
PCN-224IJRh)
PCN-224IJRh)
PCN-224IJRh)
PCN-224IJRh)
PCN-224IJRh)
PCN-224IJRh)
PCN-224IJRh)
PCN-224IJRh)
PCN-224IJRh)
PCN-224IJRh)
PCN-224IJRh)
PCN-224IJRh)
PCN-224IJRh)
PCN-224IJRh)
PCN-222IJRh)
PCN-222IJRh)
PCN-222IJRh)
PCN-222IJRh)
20
22
5
1 : 1
1 : 1
1 : 1
1 : 1
2 : 1
1 : 1
1 : 1
1 : 1
1 : 1
1 : 1
1 : 1
1 : 1
1 : 1
21 : 1
23 : 1
5 : 1
14 : 1
2 : 1
1 : 1
1 : 1
1 : 1
42 : 1
10 : 1
15 : 1
1 : 1
4-Methoxystyrene
4-Hydroxystyrene
4-Aminostyrene (4-AS)
3-Aminostyrene (3-AS)
2-Aminostyrene (2-AS)
N,N-Dimethyl-4-AS
4-Methylstyrene
4-Methoxystyrene
4-Nitrostyrene
28
30
10
33
58
52
68
52
71
29
4
35
36
12
31
22
49
20
29
23
12
16
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
4-Chlorostyrene
Styrene
Styreneb
4-Hydroxystyrene
4-Aminostyrene (4-AS)
3-Aminostyrene (3-AS)
2-Aminostyrene (2-AS)
N,N-Dimethyl-4-AS
1-Octene
Cyclooctene
3,4-Dihydro-2H-pyran
4-Aminostyrene (4-AS)
3-Aminostyrene (3-AS)
2-Aminostyrene (2-AS)
N,N-Dimethyl-4-AS
stereoselective
chemical
transformation.48
Hence,
coordination of the substrate caused by the pore confinement
results in its symmetry breaking, with the carbene transfer
proceeding no longer statistically but favorably towards one
diastereomer over the other. The high diastereoselectivity
thus could possibly follow from stabilization of an alternative
transition state, compared to molecular Rh-porphyrins in
solution, which naturally lack any pore confinement effect or
substrate-controlled preferential steric orientation in the
transition state of the carbene transfer. The abovementioned
stabilization is also known from cage-catalyzed reactions
leading to unusual reactivity/selectivity compared to their
bulk counterparts.10,49,50
All reactions were carried out at room temperature and stirring
(rpm = 500) operated under air after 6 h reaction time. Yields are
obtained applying an excess of substrate: 135 μL (1.1 mmol) EDA
diluted in 3 mL CH2Cl2 was added via a motorized syringe pump
(rate
=
0.5 mL h−1
)
to
a
suspension of catalyst in
2
mL
a
b
substrate. Catalyst loading: 0.0033 mmol, 0.4 mol% Rh. Manual
addition of 135 μL EDA (1.1 mmol) in 2 mL CH2Cl2 to a solution
of 75 μL styrene in 2 mL CH2Cl2 within 20 min. GC-yield; based
on EDA as the limiting reagent expect entry 13, whereas styrene is
the limiting reagent.
products.
c
d
Diastereomeric ratio (trans : cis) of the
Further, we investigated the conversion of 4-hydroxystyrene
and we found a similarly high dr of 21 : 1 (trans : cis), which
supports the idea of the directing group effect of the
substrate. However, the yield of the corresponding
cyclopropane was low due to side reactions with propylene
glycol, which is present in the solution of 4-hydroxystyrene
(10 wt%), acting as a stabilizer. In order to probe our
hypothesis that the observed diastereoselectivity is controlled
by coordinating substituents attached to the styrene core, we
selectivity towards the cyclopropanation products was
decreased due to the uncontrolled dimerization of EDA
(found by GC-MS and 1H and 13C NMR analysis); EDA is
therefore no longer available for the carbene transfer to the
olefin. Therefore, the reaction was performed using
a
motorized pump and excess of olefin conditions; hereby, 135
μL (1.1 mmol) EDA diluted in 3 mL CH2Cl2 was added at a
rate of 0.5 mL h−1 to a suspension of catalyst in 2 mL olefin
solution. As a result, a conversion of 71% styrene and a dr of
1 : 1 (trans : cis) was observed; formation of the coupling side
product diethyl fumarate was drastically reduced, revealing
the crucial dependence of controlled introduction of EDA on
the yield of the reaction.
Furthermore, we determined the scope of the reaction, as
summarized in Table 1. Products of the substituted 4-methyl-,
4-methoxy-, 4-nitro-, 4-chlorostyrenes were obtained in
moderate to good yields up to 68% at r.t. after 6 h under
stirring under air applying excess substrate. Cyclic and
terminal olefins exhibited yields of 49% (cyclooctene) and
22% (1-octene). In addition, 3,4-dihydro-2H-pyran, a sugar-
based glycal, which can be obtained from biological feedstock
further
investigated
N,N-dimethyl-4-aminostyrene
and
4-methoxystyrene. In both substrates, the coordination ability
is reduced by sterical hindrance and the absence of possible
H-bonding interactions. As expected, in neither case we did
observe diastereoselective transformations, while the product
yield for N,N-dimethyl-4-aminostyrene was consistent with that
of 4-aminostyrene (4-methoxystyrene does not contain a
stabilizer, hence a direct comparison of yields cannot be
made). Moreover, the other regio isomers 2- and
3-aminostyrene (2-AS, 3-AS) were also tested to study how
other positions of the amino-group would affect the
diastereoselectivity of the reaction as compared to
4-aminostyrene (4-AS). Interestingly, both regio-isomers 2-AS
and 3-AS show a reduced dr. For PCN-224IJRh) the dr of 5 : 1
6456 | Catal. Sci. Technol., 2019, 9, 6452–6459
This journal is © The Royal Society of Chemistry 2019

View Article Online
Catalysis Science & Technology
Paper
for 3-AS is significantly lower than the dr of 23 : 1 for 4-AS.
The same trend is observed for PCN-222IJRh). We attribute this
effect to a less favorable alignment within the pores as
compared to 4-AS. 2-Aminostyrene (2-AS) shows a reduced
yield from 29% to 12% and a less drastic drop to a dr of 14 : 1.
This reduced yield could be explained by the close proximity
of the amino group to the olefin functionality. This could
affect the carbene transfer which would preferentially occur
over one side due to the steric demand of the neighbored
amino group or because of an electronic effect caused by the
ortho-positioning of the NH2 group. Thus, the dr does not
simply depend on the substitution pattern of the amino group
at the styrene core, obviously the details of secondary
PCN-222IJRh) provides smaller Rh–Rh distances of 9.7 Å
within its smaller trigonal micropores (Fig. 4, right).
4-Aminostyrene has a dimension of 8.1 Å and when we
consider a bond length of around 2 Å for Rh-carbene,51
a
cumulative length of ∼10 Å is obtained. Hence, the summed
distance of substrate and carbene matches quite well with the
Rh–Rh distance of 9.7 Å in PCN-222IJRh). Indeed, reaction of
EDA and 4-aminostyrene proceeds with a moderate yield of
29%, but with an exceptional dr of 42: 1 (trans : cis). The small
trigonal pore is built from three distinct Rh-centers with
identical Rh–Rh distances. Thus, in our view, this specific
micro environment confines the reaction space and makes it
more likely that these two species will react in very selective
manner, leading to strongly enhanced diastereoselectivity (see
the ESI,† Fig. S12 for qualitative graphic illustrations
highlighting the possible orientations of 4-aminostyrene inside
a cavity of PCN-222IJRh)). Since the Zr–Rh distances in both
MOFs are identical and the major differences in the Kagomé
structure of PCN-222IJRh) are the shorter Rh–Rh distances, we
conclude that the role of H-bonding interactions of amino or
hydroxy groups with the Zr-oxo clusters are not that important.
However, we cannot strictly exclude nor quantify the
contribution of H-bonding interactions for substrate
alignment. When the regio-isomers 2- and 3-aminostyrene were
tested for comparison, less pronounced dr values were found
for both catalysts (Table 1). This indicates the importance of
appropriate size-matching and alignment upon secondary
interactions such as amino group coordination within the
pores of the MOF-catalysts. The combination of 4-AS and PCN-
222IJRh) appears as an optimum within our series of test
reactions. In a conceptually similar study Zhang et al. applied
Zn-BCTA MOF (BCTA = bisij4-(5-carboxy-2-thienyl)phenyl]IJ4-
carboxyphenyl)amine) as a photocatalyst in the sulfonylation–
cyclization of activated alkenes.52 Interpenetration of the MOF
allowed closer substrate-redox center contacts, resulting in
superior efficiency and more importantly, in higher
diastereoselectivities compared to the homogeneous
counterpart.52 These results are in line with the observed
diastereoselectivities in our work, which also critically depend
on confinement effects, controlling the alignment of the
substrate with the catalyst's active site and neighboring groups.
substrate–catalyst
interactions
are
rather
complex.
Nevertheless, the data (Table 1) altogether support our
assumption that the diasteroselectivity depend on the
positioning of the substrate influenced by coordinatively
active functional side groups at the substrate.
In contrast to other substrate-directed reactions which
often proceed via an intramolecular pathway, the ascribed
reaction is intermolecular, since coordination and catalytic
reactions do proceed at two distinct centers. When we
consider the coordination of the amino group to a Rh-
porphyrin, the Rh-porphyrin may have two important
different functions: (1) coordination acceptor and pre-
alignment of non-innocent substrate and (2) catalytically
active center. This interplay of two neighbouring Rh-centers
may critically influence the diastereoselectivity of the
reaction.
2.10 Topology-dependent diastereoselectivity
To investigate the interplay of two neighboring Rh-
porphyrins as a function of the Rh–Rh distances, we studied
PCN-222IJRh), which consists out of the same building units
as PCN-224IJRh) but differs in pore geometry and topology
exhibiting small trigonal and large hexagonal pores. In
contrast to a Rh–Rh distance of 13.6 Å found in PCN-224IJRh),
3 Conclusions
We introduced an example of a catalytic process wherein the
diastereoselectivity crucially depends on specific local
confinement effects, which can be adjusted by the careful
choice of an appropriate MOF system. For the first time,
heterogeneous porphyrin-based metal–organic frameworks
PCN-222IJRh) and PCN-224IJRh) catalysts were applied in the
diastereoselective cyclopropanation of styrene and styrene
derivatives with ethyl diazoacetate. Styrene and substituted
styrenes were converted with high catalytic activity.
Interestingly, styrenes carrying coordinating amino and
hydroxy groups show a high diastereomeric ratio (dr) of up to
23 : 1 (trans : cis) under PCN-224IJRh) catalyst, which was
Fig.
4 Schematic representation of PCN-224IJRh) (left) exhibiting
squared micropores and PCN-222IJRh) revealing its small trigonal
micropores and large hexagonal mesopores (right). Black and bold
arrows indicate Rh–Rh distances, while black arrows represent Zr–Rh
distances, respectively. Red arrows highlight the most feasible
(shortest possible) Rh–Rh distances, which would favor the assisted
binding of a coordinating substrate molecule.
This journal is © The Royal Society of Chemistry 2019
Catal. Sci. Technol., 2019, 9, 6452–6459 | 6457

View Article Online
Paper
Catalysis Science & Technology
attributed to their coordination to Rh centres caused by pore
confinement effects derived from favourable substrate–
catalyst interactions. The diastereoselectivity is further
improved to a dr of 42 : 1 (trans : cis) by selecting PCN-222IJRh)
catalyst, a structural analogue of PCN-224 with Kagomé
topology, featuring closer Rh–Rh distances. Moreover,
reaction of the regio-isomers 2- and 3-aminostyrene reveal a
significant drop of the diastereoselectivity compared to
4-aminostyrene, demonstrating that the exact positioning of
the anchored amino groups may be critical to control the
diastereomeric ratio. These results demonstrate that the
diastereoselectivity of the cyclopropanation reaction can be
controlled by (1) the choice and positioning of functional
groups attached to substrates and (2) by structural and
topological differences of the two MOFs used in this study,
i.e. the distances between two rhodium centres. The insights
that secondary substrate–catalyst interactions, namely,
substrate driven coordination effects within the well-defined
pores of topology altered MOFs may be interesting for the
development of novel MOFs catalysts for stereoselective
transformations. The results showcase possible structure–
property relationships and more importantly, stress the
impact of confined space as a critical reaction parameter to
obtain stereoselective reaction products.
9 S. H. A. M. Leenders, R. Gramage-Doria, B. de Bruin and
J. N. H. Reek, Chem. Soc. Rev., 2015, 44, 433–448.
10 A. M. Castilla, W. J. Ramsay and J. R. Nitschke, Acc. Chem.
Res., 2014, 47, 2063–2073.
11 J. C. Jansen, E. J. Creyghton, S. L. Njo, H. van Koningsveld
and H. van Bekkum, Catal. Today, 1997, 38, 205–212.
12 A. Corma, M. E. Domine, L. Nemeth and S. Valencia, J. Am.
Chem. Soc., 2002, 124, 3194–3195.
13 C. Li, H. Zhang, D. Jiang and Q. Yang, Chem. Commun.,
2007, 547–558.
14 S. H. Kim, C. K. Shin, J. H. Seok, C. Y. Lee and G. J. Kim,
Stud. Surf. Sci. Catal., 2007, 165, 745–748.
15 A. Corma, M. Iglesias, F. X. L. i. Xamena and F. Sánchez,
Chem. – Eur. J., 2010, 16, 9789–9795.
16 F. G. Cirujano, F. X. Llabrés i Xamena and A. Corma, Dalton
Trans., 2012, 41, 4249–4254.
17 V. L. Rechac, F. G. Cirujano, A. Corma and F. X. Llabrés i
Xamena, Eur. J. Inorg. Chem., 2016, 4512–4516.
18 H. H. Mautschke, F. Drache, I. Senkovska, S. Kaskel and
F. X. Llabrés
i Xamena, Catal. Sci. Technol., 2018, 8,
3610–3616.
19 R. E. Morris and X. Bu, Nat. Chem., 2010, 2, 353–361.
20 T. Sawano and H. Yamamoto, J. Org. Chem., 2018, 83,
4889–4904.
21 K. Epp, A. L. Semrau, M. Cokoja and R. A. Fischer,
ChemCatChem, 2018, 10, 3506–3512.
22 C. G. Hamaker, G. A. Mirafzal and L. K. Woo,
Organometallics, 2001, 20, 5171–5176.
Conflicts of interest
There are no conflicts to declare.
23 P. Tagliatesta and A. Pastorini, J. Mol. Catal. A: Chem.,
2003, 198, 57–61.
Acknowledgements
24 Y. Chen, J. V. Ruppel and X. P. Zhang, J. Am. Chem. Soc.,
2007, 129, 12074–12075.
25 H. J. Callot and C. Piechocki, Tetrahedron Lett., 1980, 21,
3489–3492.
26 S. O'Malley and T. Kodadek, Organometallics, 1992, 11,
2299–2302.
27 A. G. M. Barrett, D. C. Braddock, I. Lenoir and H. Tone,
J. Org. Chem., 2001, 66, 8260–8263.
28 M. Frauenkron and A. Berkessel, Tetrahedron Lett., 1997, 38,
7175–7176.
29 J.-C. Wang, Y. Zhang, Z.-J. Xu, V. K.-Y. Lo and C.-M. Che,
ACS Catal., 2013, 3, 1144–1148.
30 D. A. Smith, D. N. Reynolds and L. K. Woo, J. Am. Chem.
Soc., 1993, 115, 2511–2513.
31 Z. Gross, N. Galili and L. Simkhovich, Tetrahedron Lett.,
1999, 40, 1571–1574.
The authors thank Werner Heinz for his assistance, fruitful
discussions and EDS measurements. BB is grateful to the
Research Foundation – Flanders (FWO) for funding (post-
doctoral grant). DDV thanks FWO for project funding, and
KU Leuven (Metusalem grant). This work was funded by the
European Union‘s Horizon 2020 research and innovation
program under the Marie Sklodowska-Curie grant agreement
no. 641887 (project acronym: DEFNET).
Notes and references
1 R. Noyori, Angew. Chem., Int. Ed., 2002, 41, 2008–2022.
2 S. F. Kirsch, Angew. Chem., Int. Ed., 2009, 48, 2450–2451.
3 A. Miyashita, A. Yasuda, H. Takaya, K. Toriumi, T. Ito, T.
Souchi and R. Noyori, J. Am. Chem. Soc., 1980, 102,
7932–7934.
4 T. Katsuki and K. B. Sharpless, J. Am. Chem. Soc., 1980, 102,
5974–5976.
32 D. Intrieri, A. Caselli and E. Gallo, Eur. J. Inorg. Chem.,
2011, 5071–5081.
5 K. Manna, T. Zhang and W. Lin, J. Am. Chem. Soc.,
2014, 136, 6566–6569.
6 J. S. Seo, D. Whang, H. Lee, S. I. Jun, J. Oh, Y. J. Jeon and K.
Kim, Nature, 2000, 404, 982–986.
7 A. Gheorghe, M. A. Tepaske and S. Tanase, Inorg. Chem.
Front., 2018, 5, 1512–1523.
8 M. Yoon, R. Srirambalaji and K. Kim, Chem. Rev., 2012, 112,
1196–1231.
33 D. M. Carminati, D. Intrieri, A. Caselli, S. L. Gac, B. Boitrel,
L. Toma, L. Legnani and E. Gallo, Chem. – Eur. J., 2016, 22,
13599–13612.
34 L. Huang, Y. Chen, G.-Y. Gao and X. P. Zhang, J. Org. Chem.,
2003, 68, 8179–8184.
35 S. Fantauzzi, E. Gallo, E. Rose, N. Raoul, A. Caselli, S. Issa,
F. Ragaini and S. Cenini, Organometallics, 2008, 27,
6143–6151.
6458 | Catal. Sci. Technol., 2019, 9, 6452–6459
This journal is © The Royal Society of Chemistry 2019

View Article Online
Catalysis Science & Technology
Paper
36 S. Zhu, J. V. Ruppel, H. Lu, L. Wojtas and X. P. Zhang, J. Am.
Chem. Soc., 2008, 130, 5042–5043.
45 M. P. Doyle, W. R. Winchester, J. A. A. Hoorn, V. Lynch,
S. H. Simonsen and R. Ghosh, J. Am. Chem. Soc., 1993, 115,
9968–9978.
46 J. Liu, Y.-Z. Fan, X. Li, Z. Wei, Y.-W. Xu, L. Zhang and C.-Y.
Su, Appl. Catal., B, 2018, 231, 173–181.
47 F. Rouquerol, J. Rouquerol and K. S. W. Sing, Adsorption by
Powders and Porous Solids: Principles, Methodology and
Applications, San Diego, USA, 1999.
48 X. Jing, C. He, L. Zhao and C. Duan, Acc. Chem. Res.,
2019, 52, 100–109.
37 D. Feng, Z.-Y. Gu, J.-R. Li, H.-L. Jiang, Z. Wei and H.-C.
Zhou, Angew. Chem., Int. Ed., 2012, 51, 10307–10310.
38 D. Feng, W.-C. Chung, Z. Wei, Z.-Y. Gu, H.-L. Jiang, Y.-P.
Chen, D. J. Darensbourg and H.-C. Zhou, J. Am. Chem. Soc.,
2013, 135, 17105–17110.
39 W. Morris, B. Volosskiy, S. Demir, F. Gándara, P. L. McGrier,
H. Furukawa, D. Cascio, J. F. Stoddart and O. M. Yaghi,
Inorg. Chem., 2012, 51, 6443–6445.
40 H. Lebel, J.-F. Marcoux, C. Molinaro and A. B. Charette,
Chem. Rev., 2003, 103, 977–1050.
49 J. D. Bass, S. L. Anderson and A. Katz, Angew. Chem., Int.
Ed., 2003, 42, 5219–5222.
41 C. Ebner and E. M. Carreira, Chem. Rev., 2017, 117, 11651–11679.
42 G. Maas, Chem. Soc. Rev., 2004, 33, 183–190.
43 A. de Meijere, Methods of Organic Chemistry, Carbocyclic
50 J. L. Bolliger, in Effects of Nanoconfinement on Catalysis, ed.
R. Poli, Springer International Publishing, Cham, 2017, DOI:
10.1007/978-3-319-50207-6_2, pp. 17–48.
Three-Membered
Ring
Compounds
(Houben-Weyl):
51 L. R. Collins, M. van Gastel, F. Neese and A. Fürstner, J. Am.
Chem. Soc., 2018, 140, 13042–13055.
52 T. Zhang, Y. Shi, S. Zhang, C. Jia, C. He and C. Duan, New J.
Chem., 2018, 42, 18448–18457.
Cyclopropanes, Transformations, Stuttgart, Germany, 1997.
44 H. J. Callot, F. Metz and C. Piechocki, Tetrahedron, 1982, 38,
2365–2369.
This journal is © The Royal Society of Chemistry 2019
Catal. Sci. Technol., 2019, 9, 6452–6459 | 6459