Selective Implantation of Diamines for Cooperative Catalysis in Isoreticular Heterometallic Titanium–Organic Frameworks

Article abstract of DOI:10.1002/anie.202100176

We introduce the first example of isoreticular titanium–organic frameworks, MUV-10 and MUV-12, to show how the different affinity of hard Ti(IV) and soft Ca(II) metal sites can be used to direct selective grafting of amines. This enables the combination of Lewis acid titanium centers and available -NH₂ sites in two sizeable pores for cooperative cycloaddition of CO₂ to epoxides at room temperature and atmospheric pressure. The selective grafting of molecules to heterometallic clusters adds up to the pool of methodologies available for controlling the positioning and distribution of chemical functions in precise positions of the framework required for definitive control of pore chemistry.

Full text of DOI:10.1002/anie.202100176

Angewandte
Research Articles
Chemie
How to cite: Angew. Chem. Int. Ed. 2021, 60, 11868–11873
International Edition:
German Edition:
doi.org/10.1002/anie.202100176
doi.org/10.1002/ange.202100176
Metal–Organic Frameworks
Hot Paper
Selective Implantation of Diamines for Cooperative Catalysis in
Isoreticular Heterometallic Titanium–Organic Frameworks
Elena López-Maya, Natalia M. Padial, Javier Castells-Gil, Carolina R. Ganivet, Ana Rubio-
Gaspar, Francisco G. Cirujano, Neyvis Almora-Barrios, Sergio Tatay, Sergio Navalón, and
Carlos Martí-Gastaldo*
Abstract: We introduce the first example of isoreticular
titanium–organic frameworks, MUV-10 and MUV-12, to show
how the different affinity of hard Ti(IV) and soft Ca(II) metal
sites can be used to direct selective grafting of amines. This
enables the combination of Lewis acid titanium centers and
available -NH2 sites in two sizeable pores for cooperative
cycloaddition of CO2 to epoxides at room temperature and
atmospheric pressure. The selective grafting of molecules to
heterometallic clusters adds up to the pool of methodologies
available for controlling the positioning and distribution of
chemical functions in precise positions of the framework
required for definitive control of pore chemistry.
Introduction
Figure 1. a) Different strategies to combine cooperative active sites in
MOFs. b) Creation of two types of open metal sites (hard and soft)
generated by thermal treatment.
Metal–Organic Frameworks (MOFs) have arisen as
promising materials for cooperative catalysis due to the
combination of high surface areas,[1] dense number of open
metal sites (OMS),[2] and crystalline structures for control-
lable arrangements of active sites and additional catalytic
units.[3] Cooperative catalysis relies on the acceleration of the
reaction rate by simultaneous activation of two or more
reactants by different catalytic sites. These processes are
these shortcomings and produce cooperative catalysts that
feature a tailorable distribution of active sites by direct
implantation in precise positions of the framework.
a common feature in enzymes, that have naturally evolved to Results and Discussion
make use of the active center and the surrounding environ-
ment for tuning their catalytic activity.[4,5] Cooperativity in
MOFs can be achieved from the interplay of open metal sites
with organic co-catalysts, that can be incorporated to the solid
either by using functionalized linkers during the synthesis[6–8]
or loaded post-synthetically as functional guests (Fig-
ure 1a).[9,10] However, these strategies sometime offer poor
control over the distribution of active sites in the solid or
weak host-guest interactions detrimental to activity or
recyclability. Chemical design can be used to circumvent
Here, we have chosen a titanium–organic framework to
demonstrate this concept. Despite MOFs often suffer from
poor chemical stability that limits their use as heterogeneous
catalysts the incorporation of hard, oxophilic Ti or Zr(IV)
ions to their structure provides stronger bonds for exceptional
stability in water.[11] Although the challenging solution
chemistry of titanium has limited the isolation of crystalline
frameworks, the number of Ti-MOFs reported has increased
significantly in the last years.[12–16] We recently reported the
heterometallic MUV-10 framework [Ti(IV)3Ca(II)3(m3-O)3-
(btc)4(H2O)6] (btc = benzene-1,3,5-tricarboxylate).[14] This
material combines excellent thermal and chemical stabilities
with a surface area close to 1000 m2 gÀ1, all relevant properties
from a catalytic standpoint. In addition, it features tetranu-
clear heterometallic Ti2Ca2 clusters with axially coordinated
water molecules that enable straightforward generation of
two types of OMS by thermal treatment.
[*] Dr. E. López-Maya, Dr. N. M. Padial, Dr. J. Castells-Gil,
Dr. C. R. Ganivet, A. Rubio-Gaspar, Dr. F. G. Cirujano,
Dr. N. Almora-Barrios, Dr. S. Tatay, Dr. C. Martí-Gastaldo
Functional Inorganic Materials team, Instituto de Ciencia Molecular,
Universidad de Valencia
Catedrµtico Jose Beltrµn-2, 46980 Paterna (Spain)
E-mail: carlos.marti@uv.es
Dr. S. Navalón
We demonstrate how the different chemistry in the
affinity of unsaturated hard Ti(IV) and soft Ca(II) Lewis
acid sites to coordinate amines can be used for selective
implantation of basic Brçnsted sites to the last (Figure 1b).
Compared to other routes based on encapsulation or im-
Departamento de Química, Universitat Polit›cnica de Val›ncia
C/Camino de Vera, s/n, 46022 Valencia (Spain)
Supporting information and the ORCID identification number(s) for
the author(s) of this article can be found under:
https://doi.org/10.1002/anie.202100176.
11868
ꢀ 2021 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 11868 –11873

Angewandte
Research Articles
Chemie
pregnation, the generation of the co-catalyst is here con-
of DAP loading was evaluated with thermogravimetric
trolled by the chemistry of the metal-oxo cluster and is thus
compatible with the isoreticular expansion of the framework.
To exemplify this advantage, we extend this method to MUV-
12, a new material built from the interlinking of the same
Ti2Ca2 with an extended tricarboxylic linker, that permits
obtaining enlarged pores to generate a cooperative catalyst
for quantitative cycloaddition of CO2 to epoxides at room
temperature and atmospheric pressure that prevents the use
of an additional molecular co-catalyst.
MUV-10 was synthesized according to the reported
method.[14] Phase purity was confirmed by chemical analysis
and powder X-Ray diffraction (PXRD). We used water
adsorption isotherms and FT-IR spectra of CO adsorption to
confirm the creation of OMS by thermal treatment in vacuum
(Figure S1 and S2). MUV-10 displays sodalite-type cages of
1.2 nm in diameter intersected in the 3D network by channels
with 0.5 nm pore windows (Figure 2a), only compatible with
the grafting of small diamines. Accordingly, we reacted
desolvated MUV-10 crystals with 1,3-diaminopropane
(DAP) in anhydrous toluene for 12 hours at room temper-
ature under inert atmosphere to produce MUV-10-DAP. This
method is very similar to that used for appending diamines to
the metal centres (Mg, Mn, Zn) in MOF-74.[17–19] The PXRD
of MUV-10-DAP confirms that the appending of DAP is
respectful with the structure of the framework (Figure 2b).
The incorporation of the diamine in the solid is consistent
analysis (TGA) and proton nuclear magnetic resonance
1H NMR of the digested solid. Both suggest the grafting of
approximately 13 wt% of DAP for the unit formula [Ti-
(IV)3Ca(II)3(m3-O)3(btc)4(DAP)3], consistent with the CHN
analysis. Compared to homometallic Mg2(dobpdc) that re-
sults in complete metal grafting for M:DAP ratios of 1:1,[18]
our experimental 2:1 ratio suggests that only 50% of the
metal sites are grafted. This agrees well with a selective
implantation to the soft Ca(II) sites in the heterometallic
Ti2Ca2 cluster, more prone to interact with the primary amine
of DAP than the hard oxophilic Ti(IV) metals. We next
analyzed the changes in the porosity of the solid with N2 and
CO2 adsorption experiments. Compared to MUV-10, the
surface area of MUV-10-DAP is reduced to close the 50% for
a BET of 520 m2 gÀ1 (Figure 2c). This drop coincides with the
reduction in free pore volume calculated with Materials
Studio for a structural model of MUV-10 with DAP molecules
appended only to Ca(II) sites (Table S1).
CO2 isotherms also reveal an important decrease in gas
uptake of close to 75% compared to MUV-10, but display
a higher uptake in the low pressure region (P < 100 torr)
indicative of a stronger interaction with CO2 (Figure S7).[20,21]
This is confirmed by the isosteric heat of CO2 adsorption (Qst)
of diamine appended MUV-10-DAP (61.3 kJmolÀ1) that
almost triplicates the 27.7 kJmolÀ1 of the pristine framework
(Figure 2d).
À
with the presence of characteristic C H and N < C- < H
This experimental information suggested MUV-10-DAP
might behave as a bifunctional heterogeneous catalyst from
the combination of Lewis acid and available -NH2 sites.
amine vibrations at 2864 and 3280 respectively cmÀ-1 in the
FT-IR spectrum (Supplementary Section S2.2.4). The extent
Figure 2. a) Structure of MUV-10 featuring sodalite-type octahedral cages of 1.2 nm in diameter (lavender sphere). b) PXRD patterns of MUV-10
and MUV-10-DAP after reaction with the diamine. c) N2 isotherms showing the drop of surface area as result of the incorporation of DAP (MUV-
10-DAP), which also induces a drastic increase in the d) CO2 isosteric heat of adsorption. e) The simulated structure of MUV-10-DAP reveals
a reduction in the pore windows and internal cavity (red sphere) of close to 50% that limits substrate uptake.
Angew. Chem. Int. Ed. 2021, 60, 11868 –11873
ꢀ 2021 Wiley-VCH GmbH
www.angewandte.org 11869

Angewandte
Research Articles
Chemie
Before testing its ability to activate the cycloaddition of CO2
to epoxides in absence of a molecular co-catalyst, we deemed
necessary to generate iodide ions that are known to activate
ring opening.[22] MUV-10-DAP crystals were suspended in dry
CH3CN and reacted with methyl iodide (CH3I) at 808C for
48 h under N2 atmosphere according to a reported method.[22]
We evaluated the incorporation of iodine by ICP analysis
(Supplementary Section S2.1). Unfortunately, our results
discarded a successful generation of iodine anions. This is
likely due to the strict limitations imposed by the grafting of
DAP to the pore openings and empty space available in
MUV-10, which is quite small in origin (Figure 2e and
Table S1).
This limitation encouraged us to synthesize an isoreticular
framework that combined superior porosity and heterome-
tallic clusters amenable to selective DAP grafting. The
reticulation of organic linkers of the same geometry and
connectivity but variable size permits controlling the metrics
and functionality of a targeted framework topology, that is in
turn determined by the structural information encoded in the
inorganic node or secondary building unit (SBU).[23] Reticular
chemistry has been incredibly fruitful in the design of multiple
MOFs[24] but the synthetic difficulties inherent to the chemis-
try of titanium in solution have neglected the discovery of
persistent SBUs for the synthesis of isoreticular Ti-MOFs
thus far. We synthesized 3,5-tris(4-carboxyphenyl)benzene
(H3btb) according to the reported protocol (Figure 3a,
Supplementary Section S3.1). Reaction of H3btb with calcium
chloride and titanium(IV) isopropoxide at 1208C in N,N-
dimethylformamide (DMF) by using acetic acid as modulator
yields colourless, octahedral crystals of ca. 100 mm in size
(Figure 3b). Single-crystal XRD measurements revealed that
MUV-12 [Ti(IV)3Ca(II)3(m3-O)3(btb)4(H2O)6] is a doubly in-
terpenetrated framework isoreticular to MUV-10. It crystal-
¯
lizes in the cubic space group Im3 (a = 26.4010 Š) and,
despite interpenetration, doubles the accessible void space
with pore diameters of 2.4 nm (Figure 3c). Besides over-
coming the spatial limitations intrinsic to MUV-10, we argued
framework interpenetration might be beneficial from a cata-
lytic standpoint due to a higher density of active sites and
enhanced robustness.[25]
We synthesized a fresh batch of MUV-12 and confirmed
phase purity by PXRD and CHN (Supplementary Sec-
tion S3), followed by DAP grafting and thorough washing
according to the same method used for MUV-10. It is worth
noticing that all materials used in this work were subjected to
a thermal treatment to prevent the presence of uncoordinated
diamine in the pore as confirmed by X-Ray Photoelectron
Spectroscopy (XPS, Supplementary Figure S11). TGA and
1H NMR suggest the appending of approximately 7.8 wt% of
DAP to the MUV-12 solid for a M:DAP ratio of 2:1. This
value is equivalent to that obtained for MUV-10 and further
supports the selective implantation to Ca(II) sites for the
formation of [Ti(IV)3Ca(II)3(m3-O)3(btb)4(DAP)3]. This unit
formula was also corroborated by CHN analysis of the
crystals. It is worth noting that MUV-12 does not suffer from
the limitations to the diffusion and grafting of the diamine
imposed by MUV-10. As result, it could incorporate more
easily a stochiometric M:DAP ratio unless a marked differ-
ence in the affinity of both metals for the diamine coordina-
tion exists. As supported by MUV-12-DAP PXRD data, DAP
grafting has a negligible impact on the structure of MUV-12
Figure 3. a) Structure of H3btb linker. b) SEM images of MUV-12 crystals. c) Structure of MUV-12 featuring an interpenetrated structure and
sodalite-type octahedral cages of ca. 2.2 nm in diameter (lavender sphere). d) PXRD patterns of MUV-12, MUV-12-DAP and MUV-12-DAP(CH3I).
e) XPS of the N 1s region in MUV-12-DAP f) N2 isotherms showing the drop of surface area as result of the incorporation of DAP (MUV-12-DAP)
as well as the incorporation of iodide ions [MUV-12-DAP(CH3I)] g) CO2 isosteric heat of adsorption of MUV-12, MUV-12-DAP and MUV-12-
DAP(CH3I). h) Simulated pore structure of MUV-12-DAP reveals a reduction in the pore windows and internal cavity (red sphere) of close to 50%
that limits substrate uptake.
11870 www.angewandte.org
ꢀ 2021 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 11868 –11873

Angewandte
Research Articles
Chemie
(Figure 3d). We used nitrogen 1s XPS to confirm the
of MUV-12-DAP(CH3I), that suggest the formation of
appending of the diamine to the framework. Figure 3e shows
the N1s high-resolution XPS spectrum of MUV-12-DAP. The
N1s signal confirms the presence of nitrogen and reveals two
different chemical environments in roughly 1:1 ratio. The
contribution centered at 400.0 and 402.3 eV can be attributed
to free and coordinated to the metal amine groups, respec-
tively.[26,27] The effect of DAP grafting on the porosity of
MUV-12 is quite similar to that observed for MUV-10. As
shown in Figure 3 f, the experimental BET surface area of
2100 m2 gÀ1 in MUV-12 drops to around 50% in MUV-12-
DAP (980 m2 gÀ1). The analysis of CO2 isotherms at 273, 283
and 293 K reveals that MUV-12-DAP outperforms the non-
aminated framework at low-P regimes, indicative of enhanced
adsorption (Supplementary Figure S18). This is consistent
with the drastic increase in the isosteric heat of CO2
adsorption observed after appending of DAP (Figure 3g).
MUV-12-DAP Qst of 84.6 kJmolÀ1, reaches close to four times
the value of pristine MUV-12 (24.3 kJmolÀ1).
a tetramethyl ammonium salt by alkylation of the grafted
DAP molecules (Supplementary Section S3.2.4). The excess
of alkylating agent used seems to promote a Hofmann
elimination reaction for the formation of a quaternary
ammonium salt, [(CH3)4N+IÀ], and the corresponding alkene.
The generation of an ammonium salt after alkylation does
not lead to a high decrease in BET surface area value
(850 m2 gÀ1) compared to MUV-12-DAP (Figure 3 f). This is
likely mitigated by the sublimation of iodine during the
activation of the framework prior to gas adsorption, as
confirmed by the EDX analysis of the material after the
isotherm that reveals a complete loss of iodine (Supplemen-
tary Figure S19e). By contrast, the treatment with CH3I
reduced substantially the CO2 uptake at 208C (40 cm3 gÀ1).
The weaker interaction with CO2 denoted by the absence of
steep increases in the gas uptake at low-P is consistent with
the transformation of pendant -NH2 groups into quaternary
ammonium cations (SupplementaryFigure S18). This is fur-
ther corroborated by the decrease of the heat of adsorption at
zero coverage to 28.4 kJmolÀ1 (Figure 3g), slightly above
pristine MUV-12.
The simulated structural model of MUV-12-DAP reveals
3
a free volume of 2176 Š with accessible pore diameters of
close to 20 Š and pore windows of 7 Š (Figure 3h). Com-
pared to the spatial limitations intrinsic to MUV-10-DAP, this
value shall enable generation of nucleophilic iodide anions in
the pore more easily. Accordingly, we reacted MUV-12-DAP
crystals with CH3I by following the same procedure described
above for its isoreticular counterpart. The resulting material
MUV-12-DAP(CH3I) was analyzed with ICP (Supplementary
Section S3.1) and energy dispersive X-ray (EDX) to confirm
the generation of iodine ions. In addition, mapping images
show the homogeneous distribution of iodine across the
crystal and scanning electronic microscopy (SEM) images
verify that the size and morphology of the crystals is
maintained in MUV-12-DAP(CH3I) discarding significant
damage (Supplementary Figure S19). Moreover, the PXRD
of MUV-12-DAP(CH3I) confirms that the framework can
survive multistep post-synthetic modification (PSM) with
negligible changes to the original diffraction pattern (Fig-
ures 3d and S10). Nonetheless, CHN analysis reveal that this
PSM induces a loss of close to 60% of the nitrogen originally
present in MUV-12-DAP. Our NMR analysis after digestion
To illustrate the value of diamine grafting for in situ
generation of nucleophilic halides in the pores of the frame-
work, we tested the catalytic activity of the MUV-10 and
MUV-12 families in the formation of cyclic carbonates by the
cycloaddition of CO2 to epoxides. There are previous reports
on the use of MOFs to catalyze this reaction but they
generally require high pressures, high temperatures or the use
of a co-catalyst.[22,28–30] This renders a complicated scenario
that is not ideal from an economical and environmental
standpoint. The design of more efficient catalysts capable of
CO2 conversions at room temperature and atmospheric
pressure without a co-catalyst is highly desirable. To demon-
strate the efficiency of our material, we tested the cyclo-
addition of CO2 to form propylene carbonate at room
temperature and 1 atm of pressure (Figure 4a). The conver-
sion with pristine MUV-10 and MUV-12 is negligible in
absence of a co-catalyst, whereas addition of tetrabutylam-
monium bromide (TBAB) leads conversions slightly above
40% in both materials. We do not observe significant
Figure 4. a) Cyclic carbonate conversion though cycloaddition of CO2 into propylene oxide using MUV-10, MUV-12, MUV-10-DAP and MUV-12-
DAP in presence (8% mol) and in absence of co-catalyst and MUV12-DAP(CH3I) only in absence of co-catalyst. b) Recycled experiments of
cycloaddition of CO2 with propylene oxide catalyzed by MUV-12-DAP(CH3I). c) PXRD of MUV-12 simulated, MUV-12-DAP(CH3I), MUV-12-
DAP(CH3I) after the first cycle and MUV-12-DAP(CH3I) after the fourth cycle of catalysis.
Angew. Chem. Int. Ed. 2021, 60, 11868 –11873
ꢀ 2021 Wiley-VCH GmbH
www.angewandte.org 11871

Angewandte
Research Articles
Chemie
differences in conversion that could be attributed to the
superior porosity of the extended MUV-12 framework. More
clear differences can be observed for the changes in catalytic
activity after the appending of DAP. Just like their non-
aminated counterparts, MUV-10-DAP and MUV-12-DAP are
inactive in absence of TBAB but display comparatively higher
conversions after the addition of the co-catalyst. This boost is
particularly significant for MUV-12-DAP, that reaches close
to 90% in conversion compared to the 60% of MUV-10-DAP.
We argue these changes between isostructural frameworks
reveal the positive effect of -NH2 Brçnsted sites to activate
CO2 for the formation of the corresponding cyclic carbon-
ate.[31] This effect is more acute for MUV-12 possibly from its
superior porosity area and interpenetrated structure. This
combination results in a superior CO2 uptake at 208C
(46 cm3 gÀ1) and a higher density of appended DAP molecules
with the associated increase in Qst for more efficient CO2
activation. Finally, MUV-12-DAP(CH3I) displays the highest
efficiency of the series and is the only material that reaches
complete conversion in absence of the molecular co-catalyst.
It should be noted that for a control physical mixture of
MUV-12, DAP and CH3I the conversion obtained was only
40%, thus suggesting that the implantation of DAP plays
a key role for the subsequent generation of iodide ions inside
the pore in MUV-12-DAP(CH3I) making unnecessary the use
of an additional co-catalyst to complete the reaction. MUV-
12-DAP(CH3I) behaves as a cooperative catalyst from the
combination of Lewis acid Ti(IV) and nucleophilic iodide
anions, both cooperating in the pores of the framework for
synergetic interaction in a confined space. The established
mechanism for this reaction involves a Lewis acid site to
activate the epoxide towards the nucleophilic attack by the
halide (Scheme S1). Whereas the density of Lewis sites is
fixed by the heterometallic Ti2Ca2 cluster, our method permits
precipitation of the polymer at the surface of the crystals,
our method enables direct implantation of the cooperating
sites in the confined space by using the chemistry of the SBU.
To investigate if the CO2 transformation was ascribed
exclusively to in-pore reactivity, we also tested the same
reaction under equivalent conditions with other substrates of
increasing sizes (Supplementary Table S6). We observe a pro-
gressive decrease with the size of the substrate down to 30%
in conversion for 1,2-epoxybutane, 7% for butyl glycidyl
ether and below 3% for bigger substrates, suggesting in-pore
reactivity, although some other electronic effects linked to the
structure of the epoxides cannot by fully discarded. Also
important, the absence of the need for any additional
component simplifies the recovery of the catalyst after the
reaction. Figure 4b shows the cyclability of MUV-12-DAP-
(CH3I), which maintains the conversion of propylene carbon-
ate above 80% upon four consecutive cycles. The profile of
the activity upon cycling correlates very nicely with the
leaching of iodine to the reaction medium determined by ICP
analysis. This small loss in activity is associated to the leaching
of near 10% of the iodine originally present in the crystal
after the first test, followed by minimum concentration
changes for the following cycles. (Supplementary Section S4).
The stability of the catalyst was also confirmed by SEM and
EDX, that show negligible effects on the size or morphology
of the crystals and homogeneous distribution of Ca, Ti and I in
all cases. This is consistent with the PXRD of the material
after each test (Figure 4c), that rules out degradation and
only display changes in the relative intensity of the (002)
diffraction peaks, possibly as result of changes in the
occupancy of the pores of MUV-12-DAP(CH3I) upon cy-
cling.[32]
producing MUV-12-DAP(CH3I) catalysts with variable Conclusion
amounts of nucleophile to evaluate the effect of these iodide
anions on the overall performance. By changing the temper-
ature and reaction time used for the synthesis of MUV-12-
DAP(CH3I) (Table S4), we synthesized materials with differ-
ent amounts of iodine. The catalytic tests confirm that
epoxide conversion in our case is controlled by the nucleo-
phile, in good agreement with the poor efficiencies encoun-
tered for MUV-10-DAP and MUV-12-DAP in absence of
TBAB. This was further corroborated by control experiments
based on the use of [(CH3)4N+IÀ], its physical mixture with
MUV-12 or attempts to impregnate it in the pores of MUV-
12. All catalytic tests show negligible conversion, underpin-
ning the importance of integrating Lewis acid Ti(IV) and
nucleophilic iodide anions in the internal cavities of the
framework for cooperative catalysis.
To put our results in a general context, we have
summarized the catalytic activity reported for other MOFs
used for the transformation of epoxides into cyclic carbonates
at standard pressure (Supplementary Table S5). MUV-12-
DAP(CH3I) is the only one that displays quantitative
conversion at room temperature, just comparable to the
activity displayed by composites of mesoporous MIL-101 or
PCN-333 prepared by infiltration with an ionic polymer.[10]
Compared to this approach that might induce the co-
We have introduced here the first family of isoreticular
titanium frameworks, MUV-10 and MUV-12, to demonstrate
the ability of heterometallic Ti2Ca2 clusters to direct the
selective implantation of diamines to the soft Ca(II) sites in
the inorganic node. This results in the combination of Lewis
acid Ti(IV) centers and available -NH2 sites. Chemical
modification of MUV-12-DAP with CH3I enables quantita-
tive cycloaddition of CO2 to propylene oxide at room
temperature and atmospheric pressure for complete conver-
sion in absence of additional co-catalysts and with excellent
cyclability. This selective grafting of molecules to heterome-
tallic clusters adds up to the pool of methodologies available
for controlling the positioning and distribution of functional
moieties in MOFs. We are confident the combined use of the
empty space offered by these porous frameworks, alongside
the different chemistries intrinsic to metal clusters and
organic linkers, offers a promising playground for producing
chemically complex environments by implantation of multi-
ple components for unprecedented opportunities in syner-
gistic phenomena.[33]
11872 www.angewandte.org
ꢀ 2021 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 11868 –11873

Angewandte
Research Articles
Chemie
A. E. Platero-Prats, S. Tatay, et al., J. Am. Chem. Soc. 2020, 142,
6638 –6648.
Acknowledgements
[13] J. Castells-Gil, N. M. Padial, N. Almora-Barrios, I. Da Silva, D.
Mateo, J. Albero, H. García, C. Martí-Gastaldo, Chem. Sci. 2019,
10, 4313 –4321.
[14] J. Castells-Gil, N. M. Padial, N. Almora-Barrios, J. Albero, A. R.
Ruiz-Salvador, J. Gonzµlez-Platas, H. García, C. Martí-Gastal-
do, Angew. Chem. Int. Ed. 2018, 57, 8453 –8457; Angew. Chem.
2018, 130, 8589 –8593.
[15] L. Zou, D. Feng, T. F. Liu, Y. P. Chen, S. Yuan, K. Wang, X.
Wang, S. Fordham, H. C. Zhou, Chem. Sci. 2016, 7, 1063 –1069.
[16] H. L. Nguyen, F. Gµndara, H. Furukawa, T. L. H. Doan, K. E.
Cordova, O. M. Yaghi, J. Am. Chem. Soc. 2016, 138, 4330 –4333.
[17] C. M. McGuirk, R. L. Siegelman, W. S. Drisdell, T. Runcevski,
P. J. Milner, J. Oktawiec, L. F. Wan, G. M. Su, H. Z. H. Jiang,
D. A. Reed, et al., Nat. Commun. 2018, 9, 5133.
[18] P. J. Milner, R. L. Siegelman, A. C. Forse, M. I. Gonzalez, T.
Runcevski, J. D. Martell, J. A. Reimer, J. R. Long, J. Am. Chem.
Soc. 2017, 139, 13541 –13553.
This work was supported by the EU (ERC Stg Chem-fs-MOF
714122) and Spanish Government (CTQ2017-83486-P,
RTI2018-098568-A-I00
& CEX2019-000919-M). E.L.-M.
and S.T. thanks the Spanish Government for their Juan de
la Cierva Fellowship (FJCI-2017-32956) and Ramon y Cajal
contract (RYC-2016-1981), respectively. A.R.G. acknowledg-
es funding from Generalitat Valencia (ACIF/2020/090)and
Fondo Social Europeo. N.M.P. thanks the European Union
for a Marie Skłodowska-Curie Global Fellowship (H2020-
MSCA-IF-2016-GF-749359-EnanSET) and to the “2020 Post-
doctoral Junior Leader–Retaining Fellowship, la Caixa Foun-
dation (ID 100010434 and fellowship code LCF/BQ/PR20/
11770014)”. F.G.C. acknowledges the support of a fellowship
from “la Caixa” Foundation (ID 100010434 and the fellow-
ship code LCF/BQ/PI19/11690011). S.N. thanks financial
support by the Ministerio de Ciencia, Innovación y Universi-
dades (RTI 2018-099482-A-I00 project), Fundación Ramón
Areces (XVIII Concurso Nacional para la Adjudicación de
Ayudas a la Investigación en Ciencias de la Vida y de la
Materia, 2016) and Agencia Valenciana de la Innovación
(AVI-GVA, Carboagua project, INNEST/2020/111).
[19] E. J. Kim, R. L. Siegelman, H. Z. H. Jiang, A. C. Forse, J. H. Lee,
J. D. Martell, P. J. Milner, J. M. Falkowski, J. B. Neaton, J. A.
Reimer, et al., Science 2020, 369, 392 –396.
[20] C. Montoro, E. García, S. Calero, M. A. PØrez-Fernµndez, A. L.
López, E. Barea, J. A. R. Navarro, J. Mater. Chem. 2012, 22,
10155 –10158.
[21] A. Demessence, D. M. DꢀAlessandro, M. L. Foo, J. R. Long, J.
Am. Chem. Soc. 2009, 131, 8784 –8786.
[22] J. Liang, R. P. Chen, X. Y. Wang, T. T. Liu, X. S. Wang, Y. B.
Huang, R. Cao, Chem. Sci. 2017, 8, 1570 –1575.
Conflict of interest
[23] M. J. Kalmutzki, N. Hanikel, O. M. Yaghi, Sci. Adv. 2018, 4,
eaat9180.
[24] M. Eddaoudi, J. Kim, N. Rosi, D. Vodak, J. Wachter, M.
OꢀKeeffe, O. M. Yaghi, Science 2002, 295, 469 –472.
[25] Z. Lin, Z. M. Zhang, Y. S. Chen, W. Lin, Angew. Chem. Int. Ed.
2016, 55, 13739 –13743; Angew. Chem. 2016, 128, 13943 –13947.
[26] W. Cao, Z. Wang, Q. Zeng, C. Shen, Appl. Surf. Sci. 2016, 389,
404 –410.
The authors declare no conflict of interest.
Keywords: carbon dioxide conversion ·cooperative catalysis ·
heterometallic cluster ·isoreticular chemistry ·titanium–
organic frameworks
[27] J. Ederer, P. Janos, P. Ecorchard, J. Tolasz, V. Stengl, H. Benes,
M. Perchacz, O. Pop-Georgievski, RSC Adv. 2017, 7, 12464 –
12473.
[28] J. Zhu, P. M. Usov, W. Xu, P. J. Celis-Salazar, S. Lin, M. C.
Kessinger, C. Landaverde-Alvarado, M. Cai, A. M. May, C.
Slebodnick, et al., J. Am. Chem. Soc. 2018, 140, 993 –1003.
[29] H. He, Q. Sun, W. Gao, J. A. Perman, F. Sun, G. Zhu, B. Aguila,
K. Forrest, B. Space, S. Ma, Angew. Chem. Int. Ed. 2018, 57,
4657 –4662; Angew. Chem. 2018, 130, 4747 –4752.
[30] C. Huang, J. Dong, W. Sun, Z. Xue, J. Ma, L. Zheng, C. Liu, X.
Li, K. Zhou, X. Qiao, et al., Nat. Commun. 2019, 10, 2779, 1 –10.
[31] Q. Han, B. Qi, W. Ren, C. He, J. Niu, C. Duan, Nat. Commun.
2015, 6, 10007, 1 –8.
[32] H. S. Cho, J. Yang, X. Gong, Y. B. Zhang, K. Momma, B. M.
Weckhuysen, H. Deng, J. K. Kang, O. M. Yaghi, O. Terasaki, Nat.
Chem. 2019, 11, 562 –570.
[1] Z. Chen, P. Li, R. Anderson, X. Wang, X. Zhang, L. Robison,
L. R. Redfern, S. Moribe, T. Islamoglu, D. A. Gómez-Gualdrón,
et al., Science 2020, 368, 297 –303.
[2] Y. S. Wei, M. Zhang, R. Zou, Q. Xu, Chem. Rev. 2020, 120,
12089 –12174.
[3] Y. B. Huang, J. Liang, X. S. Wang, R. Cao, Chem. Soc. Rev. 2017,
46, 126 –157.
[4] M. Wodrich, X. Hu, Nat. Rev. Chem. 2018, 2, 0099, 1 –7.
[5] B. Le Bailly, Nat. Rev. Chem. 2018, 2, 385 –397.
[6] P. Z. Li, X. J. Wang, J. Liu, J. S. Lim, R. Zou, Y. Zhao, J. Am.
Chem. Soc. 2016, 138, 2142 –2145.
[7] C. Li, H. Tang, Y. Fang, Z. Xiao, K. Wang, X. Wu, H. Niu, C. Zhu,
H. C. Zhou, Inorg. Chem. 2018, 57, 13912 –13919.
[8] B. Li, K. Leng, Y. Zhang, J. J. Dynes, J. Wang, Y. Hu, D. Ma, Z.
Shi, L. Zhu, D. Zhang, et al., J. Am. Chem. Soc. 2015, 137, 4243 –
4248.
[9] X. Li, B. Zhang, L. Tang, T. W. Goh, S. Qi, A. Volkov, Y. Pei, Z.
Qi, C. K. Tsung, L. Stanley, et al., Angew. Chem. Int. Ed. 2017,
56, 16371 –16375; Angew. Chem. 2017, 129, 16589 –16593.
[10] B. Aguila, Q. Sun, X. Wang, E. OꢀRourke, A. M. Al-Enizi, A.
Nafady, S. Ma, Angew. Chem. Int. Ed. 2018, 57, 10107 –10111;
Angew. Chem. 2018, 130, 10264 –10268.
[33] J. Castells-Gil, N. M. Padial, N. Almora-Barrios, R. Gil-San-
Millµn, M. Romero-ngel, V. Torres, I. da Silva, B. C. J. Vieira,
J. C. Waerenborgh, J. Jagiello, J. A. R. Navarro, S. Tatay, C.
Martí-Gastaldo, Chem 2020, 6, 3118 –3131.
[11] A. J. Howarth, Y. Liu, P. Li, Z. Li, T. C. Wang, J. T. Hupp, O. K.
Farha, Nat. Rev. Mater. 2016, 1, 15018.
[12] N. M. Padial, B. Lerma-Berlanga, N. Almora-Barrios, J. Castells-
Gil, I. da Silva, M. de la Mata, S. I. Molina, J. Hernµndez-Saz,
Manuscript received: January 5, 2021
Revised manuscript received: February 24, 2021
Accepted manuscript online: February 25, 2021
Version of record online: March 26, 2021
Angew. Chem. Int. Ed. 2021, 60, 11868 –11873
ꢀ 2021 Wiley-VCH GmbH
www.angewandte.org 11873