Heterometallic Titanium-Organic Frameworks as Dual-Metal Catalysts for Synergistic Non-buffered Hydrolysis of Nerve Agent Simulants

Full text of DOI:10.1016/j.chempr.2020.09.002

ll
Article
Heterometallic Titanium-Organic Frameworks
as Dual-Metal Catalysts for Synergistic Non-
buffered Hydrolysis of Nerve Agent Simulants
Javier Castells-Gil, Natalia
-gnskip-->M. Padial, Neyvis
Almora-Barrios, ..., Jorge A.R.
Navarro, Sergio Tatay, Carlos
Mart´ı-Gastaldo
carlos.marti@uv.es
HIGHLIGHTS
Controlled distribution of metals
in heterometallic titanium-organic
frameworks
Dual-metal catalysis for
degradation of chemical warfare
agent simulants in water
Synergetic cooperation of Ti+4
and Fe+3 sites reminiscent of
purple acid phosphatase
A family of heterometallic titanium-organic frameworks is built from the
combination of Ti(IV) with M(II) metal ions. Among them, only MUV-101(Fe) is
capable of degrading the nerve agent simulant diisopropyl-fluorophosphate in
non-buffered aqueous media, thanks to a dual-metal synergetic mechanism
reminiscent of bimetallic enzymes. This work highlights the importance of
controlling metal distribution in heterometallic frameworks at an atomic level for
advancing the design of synergetic catalysts for unprecedented boosts in
performance.
Castells-Gil et al., Chem 6, 1–14
November 5, 2020 ª 2020 Elsevier Inc.
https://doi.org/10.1016/j.chempr.2020.09.002

Please cite this article in press as: Castells-Gil et al., Heterometallic Titanium-Organic Frameworks as Dual-Metal Catalysts for Synergistic Non-
buffered Hydrolysis of Nerve Agent Simulants, Chem (2020), https://doi.org/10.1016/j.chempr.2020.09.002
ll
Article
Heterometallic Titanium-Organic Frameworks
as Dual-Metal Catalysts for Synergistic
Non-buffered Hydrolysis of Nerve Agent Simulants
Javier Castells-Gil,1,6 Natalia M. Padial,1,6,7 Neyvis Almora-Barrios,1,6 Rodrigo Gil-San-Milla´n,2
1
2
3
4
4
´
Marıa Romero-Angel, Virginia Torres, Iva´n da Silva, Bruno C.J. Vieira, Joao C. Waerenborgh,
´
Jaciek Jagiello,5 Jorge A.R. Navarro,2 Sergio Tatay,1 and Carlos Martı-Gastaldo1,8,
*
´
SUMMARY
The Bigger Picture
Heterometallic metal-organic
frameworks (MOFs) can offer
Mixed-metal or heterometallic metal-organic frameworks (MOFs)
are gaining importance as a route to produce materials with
increasing chemical and functional complexities. We report a family
of heterometallic titanium frameworks, MUV-101(M), and use them
to exemplify the advantages of controlling metal distribution across
the framework in heterogeneous catalysis by exploring their activity
toward the degradation of a nerve agent simulant of Sarin gas.
MUV-101(Fe) is the only pristine MOF capable of catalytic degrada-
tion of diisopropyl-fluorophosphate (DIFP) in non-buffered aqueous
media. This activity cannot be explained only by the association of
two metals, but to their synergistic cooperation, to create a whole
that is more efficient than the simple sum of its parts. Our simula-
tions suggest a dual-metal mechanism reminiscent of bimetallic en-
zymes, where the combination of Ti(IV) Lewis acid and Fe(III)–OH
Bro¨ nsted base sites leads to a lower energy barrier for more effi-
cient degradation of DIFP in absence of a base.
important advantages over their
homometallic counterparts to
enable targeted modification of
their adsorption, structural
response, electronic structure, or
chemical reactivity. However,
controlling metal distribution in
these solids still remains a
challenge. The family of
mesoporous titanium-organic
frameworks, MUV-101(M),
displays heterometallic TiM2
nodes assembled from direct
reaction of Ti(IV) and M(II) salts.
We use the degradation of nerve
agent simulants to demonstrate
that only TiFe2 nodes are capable
of catalytic degradation in non-
buffered conditions. By using an
integrative experimental-
INTRODUCTION
Metal-organic frameworks (MOFs) have emerged as a versatile platform to access a
broad range of applications built upon their large structural and chemical diversity.1
The unlimited number of combinations in which inorganic secondary building units
(SBUs) can be linked to organic connectors by reticular design has been used to produce
more than 84,000 porous crystalline frameworks2 for promising advances in applications
as gas storage and separation,3,4 drug delivery,5 or catalysis,6,7 to cite a few. Among
these, the degradation of chemical warfare agents (CWAs) and their simulants8–10 has
gained an increasing importance since early reports demonstrating the high activity of
Zr-MOFs in the detoxification of nerve agents.6,11,12 The activity of these materials orig-
inates from the presence of Zr6 nodes that combine accessible Lewis acid Zr(IV) and
basic-nucleophilic O2À/OHÀ sites, capable of activating P–X (X = F, O, S) bonds.
computational approach, we
rationalize how the two metals
influence each other, in this case,
for a synergistic mechanism
reminiscent of bimetallic
enzymes. Our results highlight the
importance of controlling metal
distribution at an atomic level to
span the interest of heterometallic
MOFs to a broad scope of
However, most detoxification studies have been carried out in the presence of basic
buffers as N-ethylmorpholine, that behaves as a sacrificial base and a nucleophilic
co-catalyst. In absence of this buffer, the catalyst is typically poisoned as result of
the irreversible binding of the degradation products to the Zr(IV) active centers.13
This problem can be partially overcome by heterogenization of basic-nucleophilic
sites in the framework.12,14–19 However, further improvement of CWA degradation
remains limited by the intrinsic activity of the MOFs currently available.
cascade or tandem reactions.
Chem 6, 1–14, November 5, 2020 ª 2020 Elsevier Inc.
1

Please cite this article in press as: Castells-Gil et al., Heterometallic Titanium-Organic Frameworks as Dual-Metal Catalysts for Synergistic Non-
buffered Hydrolysis of Nerve Agent Simulants, Chem (2020), https://doi.org/10.1016/j.chempr.2020.09.002
ll
Article
Figure 1. MUV-101 Heterometallic Titanium Organic Frameworks
(A and B) Structure of the family of mesoporous materials MUV-101 (A) assembled from the
interlinking of heterometallic titanium clusters (B) with the formula [TiM2(m3-O)(O2CR)6X3] (M = Mg,
Fe, Co, Ni; X = H2O, OHÀ, O2À) and trimesate linkers.
In this regard, replacement of Zr(IV) with Ti(IV) might be beneficial due to its higher
natural abundance, stronger Lewis acidity, and strength of Ti–O bonds.20 Despite
these promising features, the use of titanium-organic frameworks for the degrada-
tion of nerve agents has been only studied from a theoretical standpoint.21 We argue
that this lack of experimental information is due to the synthetic challenges intrinsic
to the chemistry of titanium in solution. Compared with Zr(IV), this narrows the syn-
thetic conditions required for targeting specific SBUs, thus restricting the assembly
of targeted architectures.22 We have recently reported the synthesis of MIL-
100(Ti).23 This mesoporous Ti-MOF is based on [TiIV3(m3-O)(O2CR)6] metal-oxo clus-
ters, also present in other frameworks based on trivalent metals as the archetypical
MIL-100 family of Cr, Al, and Fe(III),24 or other Ti(III) frameworks as COK-6925 or MIL-
101(Ti).26 Previous works point out the versatility of the SBUs in accommodating
different metals with variable charge for a persistent structure director.27–31 We
have demonstrated how the addition of a second metal for heterometallic titanium
frameworks has proven an effective way to modify their photocatalytic activity by
1Functional Inorganic Materials Team, Instituto
de Ciencia Molecular (ICMol), Universitat de
Vale` ncia, Paterna 46980, Vale` ncia, Spain
tuning of the band gap without compromising stability.32 Following this strategy,
the combination of Ti(IV) Lewis acid sites with other metal transition ions might result
in synergetic cooperation for more efficient degradation of CWAs. Just like purple
acid phosphatase (PAP) metalloenzymes that combine Fe3+ and M2+ in the active
centers,33 heterometallic Ti-MOFs might also enable dual-metal catalysis for supe-
rior activity compared with homometallic Zr(IV) analogs.
2Departamento de Quımica Inorga´nica,
´
Universidad de Granada, Avenida Fuentenueva
S/N, Granada 18071, Spain
3ISIS Facility, Rutherford Appleton Laboratory,
Chilton, Didcot, Oxfordshire OX11 0QX, UK
4Centro de Cieˆ ncias e Tecnologias Nucleares,
Instituto Superior Te´ cnico, Universidade de
Lisboa, Bobadela LRS 2695-066, Portugal
We demonstrate this concept for a new family of heterometallic titanium MUV-
101(M) frameworks (M = Mg, Fe, Co, and Ni) isostructural to the archetypal MIL-
100 family (Figure 1).24 These mesoporous materials can be prepared from direct re-
action of the molecular precursors to ensure good control over the distribution of the
metals across the structure. Compared to post-synthetic doping of preformed ma-
terials, that can result in partial or non-homogeneous metal substitution,34 we argue
that homogeneity is fundamental to rationalize the effect of the heteroatom over ac-
tivity. Our results show that pristine MUV-101(Fe) displays excellent catalytic activity
for the degradation of the Sarin gas simulant DIFP (diisopropyl fluorophosphate) in
water and non-buffered conditions. This distinctive behavior, that is not accessible
5Micromeritics Instrument Corporation, 4356
Communications Drive, Norcross, GA, USA
6These authors contributed equally
7Present address: Department of Chemistry,
Scripps Research, 10550 North Torrey Pines
Road, La Jolla, CA 92037, USA
8Lead Contact
*Correspondence: carlos.marti@uv.es
https://doi.org/10.1016/j.chempr.2020.09.002
2
Chem 6, 1–14, November 5, 2020

Please cite this article in press as: Castells-Gil et al., Heterometallic Titanium-Organic Frameworks as Dual-Metal Catalysts for Synergistic Non-
buffered Hydrolysis of Nerve Agent Simulants, Chem (2020), https://doi.org/10.1016/j.chempr.2020.09.002
ll
Article
Figure 2. Characterization of MUV-101 Heterometallic Titanium-Organic Frameworks
(A) Comparison of the PXRD of MUV-101(Mg, Fe, Co and Ni) solids with the pattern simulated from the structure refined for MUV-101(Fe).
˚
(B–D) Rietveld refinement of MUV-101(Fe) (l = 0.442655 A) (B), SEM micrograph (Scale bar: 2 mm) (C), and TEM-EDX mapping images of as-synthesized
MUV-101(Fe) crystals (D) (scale bar: 100 nm).
(E) N2 adsorption-desorption isotherms of MUV-101 solids at 77 K.
(F) Calculated isosteric heat of adsorption of CO2.
to its homometallic counterparts or other heterometallic combinations, results from
the synergetic cooperation of Ti(IV) and Fe(III) sites in close proximity for a cooper-
ative mechanism that mimics bimetallic phosphatases.
RESULTS AND DISCUSSION
Heterometallic MUV-101 Titanium-Organic Frameworks
MUV-101 heterometallic [TiM2(m3-O)(L)2X3] (M = Mg, Fe, Co, Ni; L = benzene-1,3,5-
tricarboxylate; X = H2O, OHÀ, O2À) materials were synthesized by using similar con-
ditions to those reported for the heterometallic titanium-framework MUV-10.32 In a
typical experiment, titanium(IV) isopropoxide was reacted with benzene-1,3,5-tricar-
boxylic acid and the corresponding chloride metal salts, in a mixture of N,N-dime-
thylformamide (DMF) and acetic acid (AcOH) at 120ꢀC (See Section S2 for experi-
mental details). After 48 h, the resulting microcrystalline materials were separated
by centrifugation, washed with copious amounts of DMF and methanol (MeOH),
and allowed to dry under reduced pressure. Compared with our previous method
based on metal exchange reactions with MUV-10,35 this one-step synthesis can be
easily scaled-up to larger volume vessels to produce close to 1 g of material per re-
action batch. Also, the use of higher temperatures enables the formation of pure
MUV-101(Fe) phases not accessible at 65ꢀC.
Phase purity of the solids was confirmed with powder X-ray diffraction (PXRD), ther-
mogravimetric analysis (TGA), scanning electron microscopy (SEM), and inductively
coupled plasma mass spectrometry (ICP-MS). LeBail refinement of the PXRDs
converged in a cubic Fd-3m space group with excellent residual values in all cases
to confirm the formation of pure crystallographic phases isostructural to MIL-100
Chem 6, 1–14, November 5, 2020
3

Please cite this article in press as: Castells-Gil et al., Heterometallic Titanium-Organic Frameworks as Dual-Metal Catalysts for Synergistic Non-
buffered Hydrolysis of Nerve Agent Simulants, Chem (2020), https://doi.org/10.1016/j.chempr.2020.09.002
ll
Article
(Figures 2A, S1, and S2; Table S3).24 Rietveld refinement was performed on MUV-
101(Fe) as a representative example of this family of heterometallic solids (Fig-
ure 2B; Table S2). TGA ruled out the formation of contaminant oxide phases, based
on the good agreement between the experimental and calculated weight percent-
ages of residue that results from thermal decomposition of the solids in air.
Compared with the homometallic MIL-100(Ti),23 the substitution of Ti(IV) with softer
M(II) metals reduces the temperature at which the heterometallic phases decom-
pose in synthetic air, from 450ꢀC down to a minimum of 350ꢀC (Figure S4; Table
S4). As for the microscopic structure, SEM revealed all solids to be composed of
submicrometric particles with octahedral morphologies and a homogeneous size
dispersion of ca. 1 mm (Figures 2C, S5, and S6). Energy dispersive X-ray spectros-
copy (EDX) single-point mapping measurements reveal average ratios close to 1:2
(Ti:M), consistent with the formation of heterometallic [TiM2(m3-O)(O2CR)6X3] clus-
ters. Agreement with the proposed formula was also corroborated with ICP analysis
(Table S1). The homogeneous distribution of both metals throughout the solid was
used to discard metal clustering (Figures 2D and S7–S11). To confirm the formation
of heterometallic solids rather than segregated homometallic phases, we also ran
control experiments by individual reaction of the linker, with each metal precursor
under the same conditions used for the synthesis of MUV-101. Whereas reactions
with Mg, Co, or Fe(II) led to clear solutions and no solid could be isolated, reaction
with Ti(IV) or Ni(II) produced an amorphous solid or a different crystalline phase,
respectively (Figure S3). These experiments suggest that the simultaneous pres-
ence of Ti(IV) and M(II) metals in the reaction medium is necessary to induce the
assembly of the heterometallic TiM2 metal-oxo clusters required for the formation
of the framework. The heterometallic nature of the local structure of the cluster was
confirmed with synchrotron X-ray total scattering and pair distribution function
(PDF) analysis of MIL-100(Fe), MIL-100(Ti), and MUV-101(Fe). Compared to the ho-
mometallic frameworks that display a single peak in the M–M region, MUV-101(Fe)
shows the presence of two contributions to that intermetallic region (Figure S12).
Further analysis of the M–O, M–C, and M–M contributions from the differential
PDFs confirms that the M–M distances in heterometallic MUV-101(Fe) are clearly
different from those displayed by homometallic MIL-100 phases (Figure S13; Table
S5). These changes are consistent with the incorporation of Ti and Fe into a heter-
ometallic TiFe2 metal-oxo cluster, ruling out the coexistence of segregated homo-
metallic phases or clusters that would simply correspond to the superposition of
the signals corresponding to the individual Ti–Ti and Fe–Fe homometallic compo-
nents. Nonetheless, we note that further techniques, as neutron powder diffraction
and neutron total scattering experiments might be necessary to unambiguously
assess the metallic distribution along the seven crystallographically independent
metal sites in MUV-101(M) materials.
Permanent porosity of the MUV-101(M) family was analyzed with N2 isotherms at 77
K (Figure 2E). All solids show a reversible type-I isotherm with no hysteresis and two
additional uptakes at P/P0 = 0.04 and 0.12, due to the filling of the two type of mes-
opores characteristic of the mesoporous structure of MIL-100. The experimental Bru-
nauer-Emmet-Teller (BET) surface areas for all solids oscillate between 1,840 and
2,200 m2$gÀ1 (Figures 2E and S14–S17). These values are in good agreement with
those described for other MIL-100 phases36 and exceed the 1,320 m2$gÀ1 reported
for homometallic MIL-100(Ti). The pore size distribution (PSD) was calculated by
non-linear density functional theory (NLDFT) methods and reveals two types of mes-
˚
opores of ca. 17–21 and 22–26 A, consistent with the crystallographic structure
refined for MUV-101(Fe). CO2 adsorption isotherms collected between 273 and
293 K show clear differences in the adsorption profile that can be attributed to the
4
Chem 6, 1–14, November 5, 2020

Please cite this article in press as: Castells-Gil et al., Heterometallic Titanium-Organic Frameworks as Dual-Metal Catalysts for Synergistic Non-
buffered Hydrolysis of Nerve Agent Simulants, Chem (2020), https://doi.org/10.1016/j.chempr.2020.09.002
ll
Article
incorporation of different divalent metals to the framework (Figures S18A–S18D).
Changes are likely due to the different interaction of CO2 molecules with the vacant
M sites generated during the activation of the solids by thermal heating in vacuum.
The isosteric heats of adsorption vary according to the sequence Mg z Ni > Co > Fe
(Figure 2F). Similar trends have been observed for MOF-74 upon replacement of Mg
with Ni or Co(II), highlighting the impact that metal substitution can have on tuning
the selectivity of gas adsorption.37 We also estimated the impact of incorporating
Ti(IV) sites on the number of open metal sites accessible by thermal activation, by us-
ing water adsorption isotherms according to a recently reported protocol (Figures
S19 and S20; Table S7).38 Compared to homometallic MIL-100 materials, MUV-
101(Fe) displays a smaller number of open metal sites per formula unit (1.68). This
agrees well with the proposed formula as the excess of charge that results from
the incorporation of highly charged Ti(IV) units to the cluster must be counterbal-
anced by negatively charged species (OHÀ, O2À); thus reducing the number of water
molecules per cluster available.
One of the main limitations of MOFs for broad range application is their limited
chemical stability, in particular to water. This might be an issue for the degradation
of nerve agents, as these experiments generally require buffered aqueous solu-
tions or the formation of acid molecules as the reaction product. Accordingly,
we evaluated the hydrolytic stability of MUV-101 solids under acid and basic con-
ditions (See Section S5 for details). Except for MUV-101(Mg), which showed very
poor stability leading to complete degradation of the solid even in pure water
(Figure S21), the PXRD profiles of the rest of materials remained intact after soak-
ing during 24 h in water solutions at pH between 2 and 12 (Figures S22–S24). At
first, this suggested a good hydrolytical stability, but we also collected N2 iso-
therms of the solids after the treatment to confirm this point. From a porosity
standpoint, only MUV-101(Fe) retains the original properties and can be consid-
ered stable toward the attack of water. MUV-101(Ni) and MUV-101(Co) suffer
from a partial loss in the BET value that reaches a maximum of close to 60% for
the heterometallic Ni(II) phase in basic conditions (Figures S25–S27; Tables S9–
S11). These changes in hydrolytical stability linked to M(II) replacement were
further confirmed by ICP and SEM analysis (Figures S28–S31; Table S12). This
shows a higher concentration of M(II) in solution for MUV-101(Co, Ni), also associ-
ated with a reduction in particle size and morphology, which could compromise
the long-term stability of these materials in water given the presence of labile
M(II)–O bonds. We argued the higher stability of MUV-101(Fe) was probably due
to the presence of stronger Fe(III)–O coordination bonds, less likely to undergo
water hydrolysis. This was confirmed with Mossbauer spectroscopy measurements
¨
of MUV-101(Fe) (Figure 3; Table S8), which revealed the complete transformation
of Fe(II) into Fe(III) in the final material. It is worth noting that all our attempts to
synthesize heterometallic MUV-101(Fe) from Fe(III) salts under analogous reaction
conditions were unsuccessful. This suggests the inability of heterometallic TiM2
SBUs to incorporate metals with higher oxidation states directly from solution.
Just like for the case of Fe-MOF-74, the gradual oxidation of Fe(II) sites after incor-
poration to the framework is possibly more respectful with the structure formed
originally.39
Catalytic Activity for the Detoxification of Chemical Warfare Simulants in Non-
buffered Conditions
Phosphonate-based nerve agents can act as inhibitors of the acetylcholinesterase
enzyme (AChE), present in the central nervous system, by causing a continuous stim-
ulation of the nerve fiber for asphyxiation and death. This family of molecules
Chem 6, 1–14, November 5, 2020
5

Please cite this article in press as: Castells-Gil et al., Heterometallic Titanium-Organic Frameworks as Dual-Metal Catalysts for Synergistic Non-
buffered Hydrolysis of Nerve Agent Simulants, Chem (2020), https://doi.org/10.1016/j.chempr.2020.09.002
ll
Article
¨
Figure 3. Mossbauer Spectra of MUV-101(Fe) at 295 and 4 K Confirm the Presence of Fe(III) in the
Framework
generally display a stereogenic P(V) center with one P=O bond and one alkyl (–R),
one fluoride (–F) and/or –XR residues (X= O, S, N). Their degradation involves the
hydrolysis of P–F or P–XR bonds to form non-toxic phosphate or alkylphosphonic
acids (Figure 4A).40 Some metalloenzymes, such as phosphotriesterases, are
capable of hydrolyzing phosphate ester bonds through cooperative catalysis
between the two metals that form the active site. We hypothesized that the
combination of a Ti(IV) Lewis acid site with another metal ion in the heterometallic
[TiM2(m3-O)(O2C)6X3] cluster of MUV-101 might mimic the activity of the metalloen-
zyme PAP that displays bimetallic Fe(III)–M(II) active sites. To prove our hypothesis,
we tested the activity of the MUV-101 family for the degradation of DIFP, a simulant
of the Sarin nerve agent. Initial experiments were carried out in non-buffered
aqueous solutions with an equimolar MOF:DIFP ratio (See Section S6 for experi-
mental details). Figures 4B and 4C shows the hydrolysis reaction profiles of
MOF:DIFP 1:1 mixtures in the presence of heterometallic MUV-101 materials, homo-
metallic MIL-100(Ti), MIL-100(Fe) and their physical mixture in a 1:2 ratio respectful
with the composition of the heterometallic cluster in MUV-101(Fe). Noteworthy,
MUV-101(Fe) is the only one that degrades 100% of the agent within 490 min in
non-buffered aqueous solution at room temperature with a half-life time of
165 min for the formation of diisopropylphosphate (DIP) as detected by 1H and
31P-NMR (Figures S40–S43). The half-life time can be reduced to 8 min when using
an excess of catalyst (Figures S32 and S33; Table S14). It is worth noting that
MUV-101(Fe) shows a multi-step degradation profile that combines a rapid increase
up to 25 min, associated to the uptake of DIFP by the substrate, followed by a
sigmoidal profile reminiscent of allosteric enzymes. This shape suggests a coopera-
tive process, that in our case, might correspond to the fixation of the substrate at
Lewis acid Ti(IV) sites followed by its degradation supported by Fe(III)–OH Bronsted
¨
basic sites. This is consistent with the analysis of the reaction mixture during the cat-
alytic tests, which confirms that degradation is negligible until 100 min (Figure S35).
Control experiments, ran in absence of MUV-101(Fe), also confirm the inability of
DIFP to degrade in the same experimental conditions after 48 h. The catalytic nature
of DIFP detoxification with MUV-101(Fe) was demonstrated by varying the
MOF:DIFP ratio from 1:1 to 1:10. DIFP is fully degraded in all cases but the rate of
transformation slows down with the concentration of the substrate down to a
maximum half-life of 802 min for 1:10 ratio (Figure 4D). The heterogeneous nature
6
Chem 6, 1–14, November 5, 2020

Please cite this article in press as: Castells-Gil et al., Heterometallic Titanium-Organic Frameworks as Dual-Metal Catalysts for Synergistic Non-
buffered Hydrolysis of Nerve Agent Simulants, Chem (2020), https://doi.org/10.1016/j.chempr.2020.09.002
ll
Article
Figure 4. Detoxification of Nerve Agents with Heterometallic MUV-101 Frameworks
(A) Scheme of the hydrolytic degradation of DIFP.
(B) Hydrolysis profiles of DIFP of MUV-101(Mg, Fe, Co, and Ni) the family in water.
(C) Comparison of the activity of MUV-101(Fe) with homometallic MIL-100(Ti and Fe) and their
physical mixture.
(D) Variation of the reaction kinetics with the MUV-101(Fe):DIFP ratio.
(E) Cyclability of MUV-101(Fe) for four consecutive catalytic tests for a fixed 1:1 ratio.
of the catalytic process was also confirmed by filtrating the catalyst after 2 h, which
resulted in an immediate drop of activity (Figure S36).
Compared to MUV-101(Fe), MIL-100(Ti), MIL-100(Fe), MUV-101(Mg, Co, Ni), MUV-
10(Mn),32 and UiO-66(Zr),41 all display a poor performance or become poisoned
by DIFP degradation products13 in non-buffered aqueous catalytic conditions (Fig-
ures S37 and S38). To find out if the poor activity of the other titanium homometallic
and heterometallic frameworks might be due to their partial degradation in the re-
action conditions, we examined their stability after 24 h by using PXRD. Our results
confirmed the stability trends in water. MUV-101(Mg) is not stable in the reaction
conditions but the rest of the solids did not show any sign of structural degradation
and remained highly crystalline after the catalytic tests (Figure S50), suggesting that
the poorer activity of MUV-101(Co, Ni) must have a different origin. Extraction
of the reaction mixture with dichloromethane after 24 h of reaction shows no pres-
ence of unreacted DIFP in the case of MUV-101(Fe), whereas for homometallic
MIL-100(Ti), MIL-100(Fe) and their physical mixture allowed to recover close to
100%, 91.7%, and 97.3% of the nerve agent added at the beginning of the tests (Fig-
ure S37B). This suggests that adsorption of the guest DIFP and not degradation
dominates the changes in DIFP concentration over time in the homometallic MIL-
100 solids. To put our results in a more general context, we have also compared
our results with the performance reported for the most representative MOFs used
Chem 6, 1–14, November 5, 2020
7

Please cite this article in press as: Castells-Gil et al., Heterometallic Titanium-Organic Frameworks as Dual-Metal Catalysts for Synergistic Non-
buffered Hydrolysis of Nerve Agent Simulants, Chem (2020), https://doi.org/10.1016/j.chempr.2020.09.002
ll
Article
in the degradation of CWAs or their simulants in different experimental conditions
(Tables S15 and S16). MUV-101(Fe) is the only pristine MOF capable of catalytic
degradation of DIFP with no signs of catalyst poisoning in non-buffered aqueous
media without the presence of a basic-nucleophilic co-catalyst (i.e., N-ethylmorpho-
line, dimethylaminoepyridine, or metal-alkoxide). This is a remarkable result in com-
parison with the state-of-the-art Zr-MOF materials currently in use.19 To confirm this
point, we tested MUV-101(Fe) and UiO-66 under equivalent conditions to identify
the product of the reaction and quantify the poisoning of the catalyst. Analysis of
the 1H NMR and 31P-NMR spectra of the supernatant and the solids after digestion
in NaOD/D2O (2M) confirms that the product of the reaction is exclusively DIP in
both cases. Compared to UiO-66 that adsorbs close to 25% of DIP, the poisoning
of MUV-101(Fe) is minor and drops down to 5% (See Section S6). Also important,
MUV-101(Fe) can endure three consecutive reuses without substantial reduction of
its catalytic activity (Figure 4E). Similar activities have been reported for UiO-66-
0.25NH2@LiOtBu16 and NU-1,000@Mg(OMe)2_1:419, but these materials require a
pre-treatment with strong basic agents and suffer from poorer cyclability.
Overall, our findings suggest that the catalytic activity of pristine MUV-101(Fe)
cannot be explained by the association of these two metals but to their
synergistic cooperation to create a whole that is more efficient than the simple
sum of its parts. Like bimetallic PAP enzymes and other biological systems, this
heterometallic titanium framework seems to display the activity expected for a
dual-metal catalyst, in which the two different metals, Ti(IV) and Fe(III), act simulta-
neously on two different substrates, DIFP and water, to accelerate the hydrolysis
reaction.
Dual-Metal Synergistic Degradation Mechanism in MUV-101(Fe)
To guide this study, we used the mechanism proposed in the literature for bimetallic
PAP as a reference.33,42 The active center of these enzymes is based on binuclear
(HO)–FeIII–(m-OH)–MII–(H2O) units. Hydrolysis of phosphate esters in PAP undergoes
by activation of the P=O at M(II) acting as a Lewis acid site, whereas the neighboring
Fe(III) –OH centers act as a Bronsted base, activating the hydrolysis of water to
¨
generate nucleophilic OH-anions that will ultimately attack the P(V) atom. We
used computational modeling to rationalize our experimental results on the basis
of this mechanism to understand why MUV-101(Fe) outperforms other isostructural
homo- and heterometallic systems in the degradation of DIFP (See Section S7 for
details).
As depicted in Figure 5A, this would involve the activation of DIFP and water at
neighboring metal sites in the framework. The steric hindrance and lack of accessible
coordination sites revealed that hydrolysis was not possible for one single cluster,
thus pointing to a cooperative mechanism (Figure S51). Hence, we modeled the
interaction of the DIFP molecules between two adjacent SBUs in the mesoporous
cage of MUV-101(Fe). By using density functional theory (DFT), we first investigated
the activation of DIFP by water displacement at the metal axial positions in hetero-
metallic MUV-101(Fe, Co) and homometallic MIL-100(Fe, Ti) (Figure 5B, stage 0). For
simplicity, we used MUV-101(Co) as representative of the poor performance of the
MUV-101(Mg, Ni, Co) phases. In a first step, DIFP molecules were allowed to interact
directly with the metal atom upon release of the axially coordinated water
molecule (Figure 5B, stages 0–2). This competitive stage is exothermic in all
cases and results in DIFP binding to the axial position of Ti(IV) through the P=O
group, with adsorption energies (Eads) ranging from À53.1 for MIL-100(Ti) to
À33.8 kJ molÀ1 for MUV-101(Co) (Figures 6 and S52; Table S17). In all cases,
8
Chem 6, 1–14, November 5, 2020

Please cite this article in press as: Castells-Gil et al., Heterometallic Titanium-Organic Frameworks as Dual-Metal Catalysts for Synergistic Non-
buffered Hydrolysis of Nerve Agent Simulants, Chem (2020), https://doi.org/10.1016/j.chempr.2020.09.002
ll
Article
Figure 5. Proposed Mechanism for Dual-Metal Catalytic Detoxification in Heterometallic MUV-
101(Fe)
(A) Scheme illustrating the cooperative activation of DIFP and water at neighboring metal sites in
the framework.
(B) Proposed reaction mechanism for the dual-metal synergetic degradation of DIFP with MUV-
101(Fe) involving Ti(IV) (green) and Fe(III) (orange) sites. Step 0: Initial state of reactants and
catalyst. Transition State 1(TS1): Transition state of water displacement by DIFP molecules. Step 1:
Adsorption of DIFP on the active site of Ti(IV) ions. Transition State 2 (TS2): Transition state of the
nucleophilic attack on the P(V) center by activated water molecules. Step 2: Release of HF.
(C) Reaction energy profile for homometallic MIL-100(Ti and Fe) (dashed lines) and heterometallic
MUV-101(Fe and Co) (solid lines).
coordination of DIFP in Ti(IV) metal sites is clearly preferred over Co(II) or Fe(III), likely
due to the stronger Lewis acidity of titanium. Fixation of DIFP in the axial position of
Fe(III) sites in MIL-100(Fe) is much less favorable, À4.82 kJ molÀ1, suggesting that
the low performance of this material for the degradation of DIFP might be linked
to an ineffective activation of the P=O bond. This value is significantly higher for
MUV-101(Fe) via Fe, À11.58 kJ molÀ1, but still less favorable than fixation to Ti(IV),
which is more likely to be dominant (Figure 6). We next looked into the hydrolysis
reaction by assuming that the nucleophilic attack of the activated Ti(IV)–O=P(V)
bond would involve a OHÀ anion generated by M(II/III/IV)–X (X = H2O, OHÀ, O2À
)
sites in close proximity. For MUV-101(Fe), this would correspond to the transition
state (TS) represented in Figure 5B (see also Figure S53). TS2 is a concerted step
that involves the dissociation of a water molecule to generate nucleophilic OHÀ
by interaction with a Fe(III)–OH site acting as a Bronsted base that will attack the P
¨
center in the DIFP molecule bond to the neighboring Ti(IV) acid site for final hydro-
lysis of the P–F bond. The basicity of the iron sites can be regenerated by the fluoride
anions produced during the hydrolysis reaction or by proton exchanged with the sol-
vent (See Video S1 for an overview of the degradation mechanism). Figure 5C
summarizes the calculated activation energy barriers of TS2 for homometallic MIL-
100(Ti) (200.7 kJ molÀ1), MIL-100(Fe) (70.4 kJ molÀ1), and heterometallic MUV-
101(Co) (288.5 kJ molÀ1), and MUV-101(Fe) (59.8 kJ molÀ1) (Table S18). This confirms
that DIFP hydrolysis in MUV-101(Fe) is the most favorable of all.
Chem 6, 1–14, November 5, 2020
9

Please cite this article in press as: Castells-Gil et al., Heterometallic Titanium-Organic Frameworks as Dual-Metal Catalysts for Synergistic Non-
buffered Hydrolysis of Nerve Agent Simulants, Chem (2020), https://doi.org/10.1016/j.chempr.2020.09.002
ll
Article
Figure 6. Adsorption Energy of Reactants and Products by Water Displacement at the Axial
Positions of Ti(IV) (Green), Fe(III) (Orange), Co(II) (Magenta), or Zr(IV) (Blue) in the Homo- and
Heterometallic Clusters of MIL-100(Ti), MIL-100(Fe), and MUV-101(Fe) via Ti and via Fe, MUV-
101(Co) via Ti and via Co, and UiO-66(Zr)
To put these numbers into context, we also compared the activation energies calculated
for this cooperative process with those reported for the accepted mechanism of Zr(IV)-
MOFs (Table S19).43 In the case of Zr(IV), the water displacement by the nerve agent is
the rate determining step (RDS) with a minimum activation energy of 91.8 kJ molÀ1 for
MOF-808.44 In our case, this process only limits the reactivity of MIL-100(Fe), whereas all
MOFs containing Ti(IV) undertake water displacement quite easily, likely boosted by the
acidity of this metal. In our case, the RDS is the nucleophilic attack by a dissociated water
molecule, that is much more favorable for the MUV-101(Fe) from the increased strength
of Fe-OH sites as a Bronsted base. Our results suggest that the inactivity of MIL-100(Ti)
¨
and MUV-101(Co), together with the poor performance reported for MIP-177,21 origi-
nate from the absence of this basic site. Just like for PAP enzymes, the synergetic coop-
eration of Ti(IV) and Fe(III) centers in this dual-metal catalyst leads to a much lower en-
ergy barrier (59.8 kJ molÀ1) for more efficient degradation of DIFP in absence of a base.
We also used this model to evaluate the affinity of the products of the hydrolysis to bind
the metal active sites, which might lead to the poisoning of the catalyst for concomitant
drop of the catalytic activity. To avoid this scenario, the adsorption energy of the reac-
tants must be greater than the adsorption energy of the products derived from partial
hydrolysis of DIFP molecules. Our calculations suggest that the DIFP molecules are
much strongly adsorbed than the products (DIP and HF) if the adsorption proceeds
via Ti(IV) sites, which can be, in turn, desorbed more easily in the presence of water
and reactant molecules (see Figure 6; Table S20). Adsorption of DIP is slightly favorable
for the case of MUV-101(Fe), which might result in partial competition with the activation
of DIFP. These predictions are consistent with our inhibition tests (Figure S39). Whereas
the activity of MUV-101(Fe) is not altered significatively in the presence of an excess of
fluoride, the addition of an equimolar amount of DIP causes a drop in the reaction rate
that can be recovered back by washing the solid in water.
Effect of the Heteroatom on the Acidity of the Framework
Our computational study suggests that the differences in activity for DIFP degrada-
tion are associated to the changes in basicity of the M–X sites in the homo- and
10
Chem 6, 1–14, November 5, 2020

Please cite this article in press as: Castells-Gil et al., Heterometallic Titanium-Organic Frameworks as Dual-Metal Catalysts for Synergistic Non-
buffered Hydrolysis of Nerve Agent Simulants, Chem (2020), https://doi.org/10.1016/j.chempr.2020.09.002
ll
Article
Figure 7. Effect of the Heteroatom over the Acidity of the Frameworks
Density of acid centers and pKa distribution obtained by fitting the proton binding curve extracted
from the titration curves in water. From top to bottom: MIL-100(Ti), MIL100(Fe), MUV-101(Fe), and
MUV-101(Co).
heterometallic clusters of the MOFs studied. This is similar to the changes in the hy-
drolytic activity of phosphotriesterase enzymes (PTEs) ascribed to the ability of
different metals to modify the pKa of the bounded water or hydroxide molecules.33
We argued that the acidity of the different MIL-100 solids would be controlled by
the Lewis acidity of the different metal ions and the changes in the axially coordi-
nated capping linkers (X = H2O, OHÀ, O2À) to maintain charge balance.45 The
experimental acidity of the solids was evaluated by potentiometric acid-base titra-
tions (See Section S7 for experimental details). The density of acid centers and pKa
distribution was determined for each MOF by fitting the proton binding curve ex-
tracted from the titration curves by using the SAEIUS numerical procedure.46,47
The four solids showed different titration curves and changes in the distribution
of acid centers for different pKa values (Figures 7 and S55–S58). All of them
show a predominant peak centered at pKa0 z 3.1, that can be attributed to the
protonation of all the oxo, hydroxo axial groups at very low pH. Noticeable differ-
ences can be observed at higher pKa values, that can be linked to variations in the
metal and capping linkers. MIL-100(Ti) shows an almost flat pKa distribution above
4.0, consistent with a highly acidic environment. In turn, MIL-100(Fe) and MUV-
101(Fe) showed a more basic character with a more marked and narrower pKa1
contribution with an average pKa1 centered around 7.9 that we ascribe to the
loss of Fe(III)–OH protons. The titration curve of MUV-101(Co) is more complex
due to the presence of Co–H2O species in the cluster. Compared to Ti(IV) and
Fe(III), Co(II) is a weaker Lewis acid, but axial H2O molecules can easily lose a pro-
ton to render Co–OH–. This results in two contributions at pKa1 = 5.78 (Co–OH2)
and pKa2 = 9.54 (Co–OH–). Considering pKa1 as the dominant acid-base equilibria
for the reaction conditions (non-buffered water), the higher basicity of MUV-
101(Fe) combined with more favorable adsorption of DIFP is consistent with the
proposed origin for its catalytic activity.
Chem 6, 1–14, November 5, 2020
11

Please cite this article in press as: Castells-Gil et al., Heterometallic Titanium-Organic Frameworks as Dual-Metal Catalysts for Synergistic Non-
buffered Hydrolysis of Nerve Agent Simulants, Chem (2020), https://doi.org/10.1016/j.chempr.2020.09.002
ll
Article
Concluding Remarks
The combination of Ti(IV) with other transition metals can be an efficient tool to pro-
duce materials with tunable function, provided good control of the distribution of
the metals across the framework. This route can combine the advantageous proper-
ties of the frameworks produced from this naturally abundant metal: excellent chem-
ical stability, low toxicity or photoactivity, with synergetic cooperation for improved
catalytic performance. We have illustrated this concept for a new family of titanium
heterometallic frameworks MUV-101(Mg, Fe, Co, and Ni). Compared with other
homo- and heterometallic MOFs, MUV-101(Fe) is very efficient in degrading DIFP
in aqueous medium and non-buffered conditions. The activity of the pristine mate-
rial does not rely on pre-conditioning with basic buffers or metal-alkoxides, thus
simplifying its potential integration in protective clothing or gas masks.
We use an integrative experimental-computational approach to clarify the origin of
the distinctive catalytic performance that arises from this specific combination of
metals and is not accessible to the isostructural homometallic analogs. Our simula-
tions suggest that the activity of MUV-101(Fe) is due to synergetic cooperation of
Ti(IV) Lewis acid and Fe(III)–OH Bronsted base sites for a cooperative mechanism
¨
that mimics bimetallic PAP enzymes. To the best of our knowledge, this is the first
example of a dual-metal TS in heterometallic MOFs that enables clear understand-
ing of the individual roles played by the metals combined and their mutual cooper-
ation. We are confident our results represent an excellent platform to guide the
design of other heterometallic frameworks and span the increasing interest in this
family of MOFs48,49 to a broad scope of cascade or tandem reactions in which syn-
ergetic catalysis might yield unprecedented boosts in performance.
EXPERIMENTAL PROCEDURES
Resource Availability
Lead Contact
Further information and requests for resources and materials should be directed to
and will be fulfilled by the Lead Contact, Carlos Mart´ı-Gastaldo (carlos.marti@uv.es).
Materials Availability
All materials generated in this study are available from the Lead Contact with a
completed Materials Transfer Agreement.
Data and Code Availability
The X-ray crystallographic data for the structure of MUV-101(Fe) reported in this pa-
per have been deposited in the Cambridge Crystallographic Data Centre (CCDC)
with the code 1960226 and can be obtained free of charge via www.ccdc.cam.ac.
uk/data_request/cif.
Detailed experimental procedures are provided in the Supplemental Information.
SUPPLEMENTAL INFORMATION
Supplemental Information can be found online at https://doi.org/10.1016/j.chempr.
2020.09.002.
ACKNOWLEDGMENTS
This work was supported by the EU (ERC Stg Chem-fs-MOF 714122) and Spanish
government (CTQ2017-83486-P, RTI2018-098568-A-I00, and CEX2019-000919-
M). S.T., M.R.-A., and J.C.-G. thank the Spanish government for a Ramo´ n y Cajal
12
Chem 6, 1–14, November 5, 2020

Please cite this article in press as: Castells-Gil et al., Heterometallic Titanium-Organic Frameworks as Dual-Metal Catalysts for Synergistic Non-
buffered Hydrolysis of Nerve Agent Simulants, Chem (2020), https://doi.org/10.1016/j.chempr.2020.09.002
ll
Article
Fellowship (RYC-2016-19817) and FPI Scholarships (CTQ2017-83486-P and
CTQ2014-59209-P). N.M.P. thanks the EU for a Marie Skłodowska-Curie Global
Fellowship (H2020-MSCA-IF-2016-GF-749359-EnanSET). J.A.R.N. and R.G.-S.-M.
are grateful to Spanish MINECO (CTQ2017-84692-R) and EU Feder Funding. We
acknowledge ALBA Synchrotron (proposal 2017092396) for provision of synchrotron
radiation facilities, and we would like to thank Catalin Popescu and Alicia Manjo´ n for
assistance in using the beamline BL04-MSPD. We also thank BSC-RES for computa-
tional resources (QS-2020-2-0024) and Celia Castillo-Blas and Ana Platero-Prats for
their help with the PDF analysis.
AUTHOR CONTRIBUTIONS
C.M.-G. was responsible for the overall conception, direction, and supervision of the
project. J.C.-G. and N.M.P. performed most of the experimental work and data anal-
ysis. M.R.-A. assisted with the experimental work. N.A.-B. and S.T. were responsible
for the computational work. N.M.P., R.G.-S.-M., V.T., and J.A.R.N. carried out the
catalytic tests. J.C.-G. and I.d.S. carried out the structural analysis. J.C.W. and
B.C.J.V. carried out the Mossbauer data collection and analysis. J.J. performed
¨
the data analysis of the potentiometric titration experiments. All authors discussed
the results and contributed to the writing of the manuscript.
DECLARATION OF INTERESTS
The authors declare that Universidad de Vale` ncia has applied for a patent on the ma-
terials discussed herein, on which J.C.-G., N.M.P., N.A.B., R.G.-S.-M., S.T., J.A.R.N.,
and C.M.-G. are included as inventors.
Received: March 14, 2020
Revised: April 13, 2020
Accepted: September 3, 2020
Published: September 29, 2020
REFERENCES
1. Furukawa, H., Cordova, K.E., O’Keeffe, M., and
Yaghi, O.M. (2013). The chemistry and
applications of metal-organic frameworks.
Science 341, 1230444.
6. Mondloch, J.E., Katz, M.J., Isley, W.C., Ghosh,
P., Liao, P., Bury, W., Wagner, G.W., Hall, M.G.,
Decoste, J.B., Peterson, G.W., et al. (2015).
Destruction of chemical warfare agents using
metal-organic frameworks. Nat. Mater. 14,
512–516.
11. Lopez-Maya, E., Montoro, C., Rodrıguez-
˜ ´
Albelo, L.M., Aznar Cervantes, S.D., Lozano-
Pe´ rez, A.A., Cenı´s, J.L., et al. (2015). Textile/
metal-organic-framework composites as self-
detoxifying filters for chemical-warfare agents.
Angew. Chem. Int. Ed. 54, 6790–6794.
2. Matito-Martos, I., Moghadam, P.Z., Li, A.,
Colombo, V., Navarro, J.A.R., Calero, S., and
Fairen-Jimenez, D. (2018). Discovery of an
optimal porous crystalline material for the
capture of chemical warfare agents. Chem.
Mater. 30, 4571–4579.
12. Moon, S.Y., Liu, Y., Hupp, J.T., and Farha, O.K.
(2015). Instantaneous hydrolysis of nerve-agent
simulants with a six-connected zirconium-
based metal-organic framework. Angew.
Chem. Int. Ed. 54, 6795–6799.
7. Zhao, M., Yuan, K., Wang, Y., Li, G., Guo, J., Gu,
L., Hu, W., Zhao, H., and Tang, Z. (2016). Metal-
organic frameworks as selectivity regulators for
hydrogenation reactions. Nature 539, 76–80.
3. Mason, J.A., Oktawiec, J., Taylor, M.K.,
Hudson, M.R., Rodriguez, J., Bachman, J.E.,
Gonzalez, M.I., Cervellino, A., Guagliardi, A.,
Brown, C.M., et al. (2015). Methane storage in
flexible metal-organic frameworks with intrinsic
thermal management. Nature 527, 357–361.
8. Liu, Y., Howarth, A.J., Vermeulen, N.A., Moon,
S.Y., Hupp, J.T., and Farha, O.K. (2017).
Catalytic degradation of chemical warfare
agents and their simulants by metal-organic
frameworks. Coord. Chem. Rev. 346, 101–111.
13. Plonka, A.M., Wang, Q., Gordon, W.O.,
Balboa, A., Troya, D., Guo, W., Sharp, C.H.,
Senanayake, S.D., Morris, J.R., Hill, C.L., and
Frenkel, A.I. (2017). In situ probes of capture
and decomposition of chemical warfare agent
simulants by Zr-based metal organic
9. Bobbitt, N.S., Mendonca, M.L., Howarth, A.J.,
Islamoglu, T., Hupp, J.T., Farha, O.K., and
Snurr, R.Q. (2017). Metal-organic frameworks
for the removal of toxic industrial chemicals
and chemical warfare agents. Chem. Soc. Rev.
46, 3357–3385.
4. Cadiau, A., Adil, K., Bhatt, P.M., Belmabkhout,
Y., and Eddaoudi, M. (2016). A metal-organic
framework-based splitter for separating
propylene from propane. Science 353,
137–140.
frameworks. J. Am. Chem. Soc. 139, 599–602.
14. Chen, Z., Wang, X., Noh, H., Ayoub, G.,
Peterson, G.W., Buru, C.T., et al. (2019).
Scalable, room temperature, and water-based
synthesis of functionalized zirconium-based
metal-organic frameworks for toxic chemical
removal. Cryst. Eng. Comm. 21, 2409–2415.
5. Horcajada, P., Chalati, T., Serre, C., Gillet,
B., Sebrie, C., Baati, T., Eubank, J.F.,
Heurtaux, D., Clayette, P., Kreuz, C., et al.
(2010). Porous metal-organic-framework
nanoscale carriers as a potential platform
for drug delivery and imaging. Nat. Mater.
9, 172–178.
10. Barea, E., Maldonado, C.R., and Navarro, J.A.R.
(2018). MOFs for the capture and degradation
of chemical warfare agents. In Metal-Organic
Frameworks: Applications in Separations and
Catalysis, H. Garcı´a and S. Navalo´ n, eds.
(Wiley-VCH Verlag), pp. 199–221.
15. Katz, M.J., Moon, S.Y., Mondloch, J.E., Beyzavi,
M.H., Stephenson, C.J., Hupp, J.T., and Farha,
O.K. (2015). Exploiting parameter space in
MOFs: A 20-fold enhancement of phosphate-
Chem 6, 1–14, November 5, 2020
13

Please cite this article in press as: Castells-Gil et al., Heterometallic Titanium-Organic Frameworks as Dual-Metal Catalysts for Synergistic Non-
buffered Hydrolysis of Nerve Agent Simulants, Chem (2020), https://doi.org/10.1016/j.chempr.2020.09.002
ll
Article
ester hydrolysis with UiO-66-NH2. Chem. Sci. 6,
2286–2291.
a titanium(III) metal-organic framework. Inorg.
Chem. 54, 10096–10104.
38. Hall, J.N., and Bollini, P. (2020). Quantification
of open-metal sites in metal-organic
frameworks using irreversible water
16. Gil-San-Millan, R., Lo´ pez-Maya, E., Hall, M., M.
Padial, N., Peterson, G.W., DeCoste, J.B., et al.
(2017). Chemical warfare agents detoxification
properties of zirconium metal-organic
27. Wongsakulphasatch, S., Nouar, F., Rodriguez,
J., Scott, L., Le Guillouzer, C., Devic, T., et al.
(2015). Direct accessibility of mixed-metal (iii/ii)
acid sites through the rational synthesis of
porous metal carboxylates. Chem. Commun.
51, 10194–10197.
adsorption. Langmuir 36, 1345–1356.
39. Rubio-Gime´ nez, V., Waerenborgh, J.C.,
Clemente-Juan, J.M., and Martı-Gastaldo, C.
´
frameworks by synergistic incorporation of
nucleophilic and basic sites. ACS Appl. Mater.
Interfaces 9, 23967–23973.
(2017). Spontaneous magnetization in
heterometallic NiFe-MOF-74 microporous
magnets by controlled iron doping. Chem.
Mater. 29, 6181–6185.
28. Zhai, Q.G., Bu, X., Mao, C., Zhao, X., Daemen,
L., Cheng, Y., Ramirez-Cuesta, A.J., and Feng,
P. (2016). An ultra-tunable platform for
molecular engineering of high-performance
crystalline porous materials. Nat. Commun. 7,
13645.
17. De Koning, M.C., Van Grol, M., and Breijaert, T.
(2017). Degradation of paraoxon and the
chemical warfare agents VX, tabun, and Soman
by the metal-organic frameworks UiO-66-NH2,
MOF-808, NU-1000, and PCN-777. Inorg.
Chem. 56, 11804–11809.
40. Jang, Y.J., Kim, K., Tsay, O.G., Atwood, D.A.,
and Churchill, D.G. (2015). Update 1 of:
destruction and detection of chemical warfare
agents. Chem. Rev. 115, PR1–PR76.
29. Wang, Y., Zhao, X., Yang, H., Bu, X., Wang, Y.,
Jia, X., et al. (2019). A tale of two trimers from
two different worlds: a COF-inspired synthetic
strategy for pore-space partitioning of MOFs.
Angew. Chem. Int. Ed. 58, 6316–6320.
41. Katz, M.J., Mondloch, J.E., Totten, R.K., Park,
J.K., Nguyen, S.T., Farha, O.K., et al. (2014).
Simple and compelling biomimetic metal-
organic framework catalyst for the degradation
of nerve agent simulants. Angew. Chem. Int.
Ed. 53, 497–501.
18. Kalaj, M., Palomba, J.M., Bentz, K.C., and
Cohen, S.M. (2019). Multiple functional groups
in UiO-66 improve chemical warfare agent
simulant degradation. Chem. Commun. 55,
5367–5370.
30. Gime´ nez-Marque´ s, M., Santiago-Portillo, A.,
19. Gil-San-Millan, R., Lo´ pez-Maya, E., Platero-
Prats, A.E., Torres-Pe´ rez, V., Delgado, P.,
Augustyniak, A.W., Kim, M.K., Lee, H.W., Ryu,
S.G., and Navarro, J.A.R. (2019). Magnesium
exchanged zirconium metal–organic
frameworks with improved detoxification
properties of nerve agents. J. Am. Chem. Soc.
141, 11801–11805.
´
Navalo´ n, S., Alvaro, M., Briois, V., Nouar, F.,
42. Erxleben, A. (2019). Mechanistic studies of
homo- and heterodinuclear zinc
Garcia, H., and Serre, C. (2019). Exploring the
catalytic performance of a series of bimetallic
MIL-100(Fe, Ni) MOFs. J. Mater. Chem. A 7,
20285–20292.
phosphoesterase mimics: what has been
learned? Front. Chem. 7, 82.
43. Momeni, M.R., and Cramer, C.J. (2019).
Computational screening of roles of defects
and metal substitution on reactivity of different
single- vs double-node metal-organic
frameworks for Sarin decomposition. J. Phys.
Chem. C 123, 15157–15165.
31. Yang, H., Wang, Y.Y., Krishna, R., Jia, X., Wang,
Y.Y., Hong, A.N., Dang, C., Castillo, H.E., Bu, X.,
and Feng, P. (2020). Pore-space-partition-
enabled exceptional ethane uptake and
ethane-selective ethane-ethylene separation.
J. Am. Chem. Soc. 142, 2222–2227.
20. M. Padial, N., Castells-Gil, J., Almora-Barrios,
N., Romero-Angel, M., Da Silva, I., Barawi, M.,
et al. (2019). Hydroxamate titanium-organic
frameworks and the effect of siderophore-type
linkers over their photocatalytic activity. J. Am.
Chem. Soc. 141, 13124–13133.
44. Rosen, A.S., Notestein, J.M., and Snurr, R.Q.
(2019). Identifying promising metal–organic
frameworks for heterogeneous catalysis via
high-throughput periodic density functional
theory. J. Comput. Chem. 40, 1305–1318.
32. Castells-Gil, J., M. Padial, N., Almora-Barrios,
N., Albero, J., Ruiz-Salvador, A.R., Gonza´lez-
Platas, J., et al. (2018). Chemical engineering of
photoactivity in heterometallic titanium–
organic frameworks by metal doping. Angew.
Chem. Int. Ed. 57, 8453–8457.
21. Vieira Soares, C., Maurin, G., and Leita˜o, A.A.
(2019). Computational exploration of the
catalytic degradation of Sarin and its simulants
by a titanium metal–organic framework.
J. Phys. Chem. C 123, 19077–19086.
45. Klet, R.C., Liu, Y., Wang, T.C., Hupp, J.T., and
Farha, O.K. (2016). Evaluation of Brønsted
acidity and proton topology in Zr- and Hf-
based metal-organic frameworks using
potentiometric acid-base titration. J. Mater.
Chem. A 4, 1479–1485.
33. Kimura, E. (2000). Dimetallic hydrolases and
their models. Curr. Opin. Chem. Biol. 4,
207–213.
22. Assi, H., Mouchaham, G., Steunou, N., Devic,
T., and Serre, C. (2017). Titanium coordination
compounds: from discrete metal complexes to
metal-organic frameworks. Chem. Soc. Rev. 46,
3431–3452.
34. Zou, L., Feng, D., Liu, T.F., Chen, Y.P., Yuan, S.,
Wang, K., Wang, X., Fordham, S., and Zhou,
H.C. (2016). A versatile synthetic route for the
preparation of titanium metal-organic
46. Jagiełło, J. (1994). Stable numerical solution of
the adsorption integral equation using splines.
Langmuir 10, 2778–2785.
23. Castells-Gil, J., M. Padial, N., Almora-Barrios,
N., Da Silva, I., Mateo, D., Albero, J., et al.
(2019). De novo synthesis of mesoporous
photoactive titanium(iv)-organic frameworks
with MIL-100 topology. Chem. Sci. 10, 4313–
4321.
frameworks. Chem. Sci. 7, 1063–1069.
47. Jagiełło, J., Bandosz, T.J., Putyera, K., and
Schwarz, J.A. (1995). Determination of proton
affinity distributions for chemical systems in
aqueous environments using a stable
numerical solution of the adsorption integral
equation. J. Colloid Interface Sci. 172,
341–346.
35. M. Padial, N., Lerma-Berlanga, B., Almora-
Barrios, N., Castells-Gil, J., Da Silva, I., De La
Mata, M., et al. (2020). Heterometallic titanium-
organic frameworks by metal-induced dynamic
topological transformations. J. Am. Chem. Soc.
142, 6638–6648.
24. Fe´ rey, G., Serre, C., Mellot-Draznieks, C.,
Millange, F., Surble´ , S., Dutour, J., et al. (2004).
A hybrid solid with giant pores prepared by a
combination of targeted chemistry, simulation,
and powder diffraction. Angew. Chem. Int. Ed.
43, 6296–6301.
36. Llewellyn, P.L., Bourrelly, S., Serre, C., Vimont,
A., Daturi, M., Hamon, L., De Weireld, G.,
Chang, J.S., Hong, D.Y., Kyu Hwang, Y.K., et al.
(2008). High uptakes of CO2 and CH4 in
mesoporous metal-organic frameworks MIL-
100 and MIL-101. Langmuir 24, 7245–7250.
48. Abednatanzi, S., Gohari Derakhshandeh, P.,
Depauw, H., Coudert, F.X., Vrielinck, H., Van
Der Voort, P., and Leus, K. (2019). Mixed-metal
metal-organic frameworks. Chem. Soc. Rev. 48,
2535–2565.
25. Bueken, B., Vermoortele, F., Vanpoucke,
D.E.P., Reinsch, H., Tsou, C.C., Valvekens, P.,
et al. (2015). A flexible photoactive titanium
metal-organic framework based on a [TiIV3(m3-
O)(O)2(COO)6] cluster. Angew. Chem. Int. Ed.
54, 13912–13917.
37. Jiao, Y., Morelock, C.R., Burtch, N.C.,
Mounfield, W.P., Hungerford, J.T., and Walton,
K.S. (2015). Tuning the kinetic water stability
and adsorption interactions of Mg-MOF-74 by
partial substitution with co or Ni. Ind. Eng.
Chem. Res. 54, 12408–12414.
49. Masoomi, M.Y., Morsali, A.,
Dhakshinamoorthy, A., and Garcia, H. (2019).
Mixed-metal MOFs: unique opportunities in
metal–organic framework (MOF) functionality
and design. Angew. Chem. Int. Ed. 58, 15188–
15205.
26. Mason, J.A., Darago, L.E., Lukens, W.W., and
Long, J.R. (2015). Synthesis and O2 reactivity of
14
Chem 6, 1–14, November 5, 2020