Robust Aluminum and Iron Phosphinate Metal-Organic Frameworks for Efficient Removal of Bisphenol A

Article abstract of DOI:10.1021/acs.inorgchem.0c00201

Porous metal-organic frameworks (MOFs) have excellent characteristics for the adsorptive removal of environmental pollutants. Herein, we introduce a new series of highly stable MOFs constructed using Fe³⁺ and Al³⁺ metal ions and bisp

Full text of DOI:10.1021/acs.inorgchem.0c00201

pubs.acs.org/IC
Article
Robust Aluminum and Iron Phosphinate Metal−Organic
Frameworks for Efficient Removal of Bisphenol A
̌
̌
́
́
̌
̌
̊
̌
Daniel Buzek, Sona Ondrusova, Jan Hynek, Petr Kovar, Kamil Lang, Jan Rohlícek, and Jan Demel*
Cite This: https://dx.doi.org/10.1021/acs.inorgchem.0c00201
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: Porous metal−organic frameworks (MOFs) have
excellent characteristics for the adsorptive removal of environmental
pollutants. Herein, we introduce a new series of highly stable MOFs
constructed using Fe3+ and Al3+ metal ions and bisphosphinate
linkers. The isoreticular design leads to ICR-2, ICR-6, and ICR-7
MOFs with a honeycomb arrangement of linear pores, surface areas
up to 1360 m2 g−1, and high solvothermal stabilities. In most cases,
their sorption capacity is retained even after 24 h of reflux in water.
The choice of the linkers allows for fine-tuning of the pore sizes and
the chemical nature of the pores. This feature can be utilized for the
optimization of host−guest interactions between molecules and the
pore walls. Water pollution by various endocrine disrupting chemicals
has been considered a global threat to public health. In this work, we prove that the chemical stability and hydrophobic nature of the
synthesized series of MOFs result in the remarkable sorption properties of these materials for endocrine disruptor bisphenol A.
chemical compositions.8−10 The high surface area of
INTRODUCTION
Endocrine disrupting chemicals are gaining increased attention
among emerging pollutants due to their influence on the
endocrine system, such as mimicking or blocking natural
■
MOFs11,12 and the possibility to introduce functional
groups13,14 makes them highly promising materials for many
applications,15 including as sorbents of pollutants.16−18 In this
context, utilization in an aqueous environment represents a
challenging condition for MOFs due to hydrolytic cleavage of
the coordination bonds that constitute the MOF backbone.
Thus, the number of proven water-stable MOFs is limited so
far.19 Reported water-stable MOFs are often composed of
azolate linkers and M2+, e.g., ZIF-8,20 or carboxylate or
phosphonate linkers combined with M3+ and M4+ metal ions as
in the case of MIL-125 and UPG-1, respectively.21,22
hormones and causing the over- or under production of
specific hormones.1 This diverse group of pollutants is utilized
in a broad spectrum of human-made products, such as
herbicides (DDT and Propanil), antimicrobial agents (Triclo-
san), detergents, toiletries, cosmetics (parabens and phenols),
pharmaceuticals (diethylstilbestrol), and plastics (phthalates,
bisphenol A, and brominated flame retardants).1,2 The most
studied endocrine disruptor is bisphenol A (2,2-bis(4-
hydroxyphenyl)propane, BPA). Over a million tons of BPA
are produced per year, mainly for the production of plastics
such as polycarbonate and epoxy resins. It was found that BPA
can leach from these products (e.g., beverage containers and
packages, baby bottles, and dental sealants), migrate into the
environment, and enter the food chain.2−4 Therefore, methods
for the fast and effective removal of BPA from wastewater or
landfill leachate are of great importance. Commonly used
sorbents, such as activated charcoal, zeolites, or clays, have a
low affinity toward BPA or a low sorption capacity.5,6 For these
reasons, new materials have to be developed in order to stop
the spread of endocrine disruptors in the environment.
Recently, we demonstrated that the combination of Fe3+
ions with the bisphosphinate linker H2PBP(Me) (Figure 1)
leads to a new MOF with a honeycomb structure named Fe-
23
̌
̌
ICR-2 (ICR stands for Inorganic Chemistry Rez).
Importantly, Fe-ICR-2 is endowed with a higher hydrothermal
stability in comparison to the carboxylate-based analogue Fe-
MIL-53.24 The stability is related to the stronger coordination
bonds of the phosphinates to hard metals, such as Fe3+, in
comparison with the carboxylate groups. Notably, the methyl
group bound at the phosphorus atom is pointing into the
Received: January 20, 2020
Metal−organic frameworks (MOFs) belong to the fast
growing area of organic−inorganic coordination polymers.
MOFs combine metal nodes (secondary building units, SBUs)
and polytopic organic ligands (linkers).7 The number of
possible SBUs and organic linkers gives rise to thousands of
new structures with varying topologies, pore sizes, and
https://dx.doi.org/10.1021/acs.inorgchem.0c00201
© XXXX American Chemical Society
Inorg. Chem. XXXX, XXX, XXX−XXX
A

Inorganic Chemistry
pubs.acs.org/IC
Article
Figure 1. Honeycomb patterns of the 1D pores of the Fe-ICR MOFs running along the c-axis (top), where the pore limiting diameter (nm)
calculated by Poreblazer is indicated in the middle of the pore; 1D columns of the octahedrally coordinated iron atoms bridged by the phosphinate
acid groups (middle); coding of the ICR MOFs (bottom left); PXRD of the Fe-ICR MOFs (bottom middle); and adsorption isotherms of nitrogen
for the Fe-ICR MOFs (bottom right). Color coding is as follows: octahedrally coordinated iron atoms (blue), phosphinate tetrahedra (magenta), O
(red), C (gray), and H (white).
volume of the Fe-ICR-2 pore. Substitution with a bulkier
phenyl group leads to a decrease in the pore size (Fe-ICR-4),23
suggesting that this strategy can allow for fine-tuning of the
pore hydrophobicity.
In this work, we applied the reticular design and extended
the bisphosphinate linkers. We synthesized and delineated the
properties of a series of phosphinate MOFs denoted ICR-2,
ICR-4, ICR-6, and ICR-7 as both the Fe3+ and Al3+ versions
(Figure 1). All prepared MOFs possess the honeycomb
arrangement of linear pores, with sizes varying from 3 to 24
Å. The hydrophobic nature of the pores leads to the high
sorption capacity for BPA.
was confirmed by elemental analyses and FTIR spectra (Table
S1 and Figures S1−S7).
The thermal stability of the ICR MOFs in air was
investigated by thermogravimetric analyses in conjunction
with differential thermal analyses and mass spectroscopy
(TGA/DTA/MS) (Figures S8−S14). The TGA curves
indicate that all of the prepared ICR MOFs are endowed
with high thermal stabilities and contain negligible amounts of
solvent or water molecules inside the pores. The least stable is
Al-ICR-4, which starts to decompose at 350 °C, whereas Al-
ICR-7 is the most thermally stable ICR MOF with a
decomposition temperature of 550 °C.
The crystal structure of Fe-ICR-7 was obtained from powder
X-ray diffraction (PXRD) data (Table S2). The indexing was
performed in the DICVOL06 program,25 and the crystal
structure models were found ab initio using the FOX software
(Supporting Information).26 The final Rietveld fit confirms the
proposed structure (Table S2 and Figure S16). The low
crystallinity of Fe-ICR-6 did not allow for indexing of the
PXRD pattern. Nevertheless, the isoreticular design of the ICR
MOFs and the solved structures for Fe-ICR-2 and Fe-ICR-423
enabled the creation of a structural model, followed by
geometry optimizations using the PCFF force field and the
Rietveld refinement in the Materials Studio software (Figure
S18).27 The PXRD patterns of all of the Fe-ICR MOFs are
compared in Figure 1, bottom middle.
RESULTS AND DISCUSSION
■
The solvothermal reaction of the bisphosphinate linkers
(Figure 1, bottom left) with FeCl3·6H2O or AlCl3·6H2O in
EtOH at 250 °C yielded the crystalline ICR MOFs. The only
exception was the reaction of H2BBP(Me) with FeCl3·6H2O,
which led to the nonporous layered material Fe-ICR-5. In this
case, the synthetic conditions were optimized and the porous
Fe-ICR-6 was obtained in DMF at 120 °C after a three-day
reaction. However, Fe-ICR-6 is of inferior crystallinity due to
the lower temperature that was used compared to the other
ICR MOFs (see below). The formula of all of the synthesized
MOFs is M2(linker)3 (M = Fe3+ or Al3+, linkers are specified in
Figure 1), whereas the formula of the layered Fe-ICR-5 is
Fe(BBP(Me)). The composition of all of the prepared MOFs
B
https://dx.doi.org/10.1021/acs.inorgchem.0c00201
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
pubs.acs.org/IC
Article
Table 1. Specific Surface Areas, Pore Diameters, Calculated Pore Limiting Diameters, and Accessible Surface Areas for All of
the Synthesized MOFs
specific surface area pore diameter
pore volume
pore limiting
diameter (nm)
accessible surface area pore accessible volume
d e
a
b
c
sample
linker
(m2 g−1
)
(nm)
(cm3 g−1
)
(m2 g−1
)
(cm3 g−1
)
f
Fe-ICR-2
Fe-ICR-4
Fe-ICR-6
Fe-ICR-7
Al-ICR-2
Al-ICR-4
Al-ICR-6
Al-ICR-7
H2PBP(Me)
H2PBP(Ph)
H2BBP(Me)
H2BBP(Ph)
H2PBP(Me)
H2PBP(Ph)
H2BBP(Me)
H2BBP(Ph)
906
165
1134
1125
933
190
1362
1030
0.71
n.a.
2.39
2.16
0.74
n.a.
0.39
0.044
1.32
0.79
0.44
0.055
1.59
1.52
0.90
0.29
1.75
1.32
0.90
0.29
1.75
1.31
850
0
1562
1097
876
0
0.48
n.a.
1.00
0.64
0.48
n.a.
f
g
g
f
f
g
2.34
2.07
1660
1088
1.06
0.63
g
a
b
c
The pore diameter was obtained by the MP plot for ICR-2 and ICR-4, otherwise by the NLDFT method. Total pore volume. The pore limiting
d
e
diameter was calculated by the Poreblazer software. The accessible surface area was calculated by the Poreblazer software. The accessible pore
volume was calculated by the Poreblazer software. The specific surface area was calculated by the t-plot method. BET specific surface area.
f
g
a
Table 2. Specific Surface Areas of As-Prepared and Treated ICR MOFs
b
reflux
rt
activated
H2O
sample
as prepared
H2O
EtOH
toluene
H2O
EtOH
toluene
Fe-ICR-2
Fe-ICR-6
Fe-ICR-7
Al-ICR-2
Al-ICR-6
Al-ICR-7
906
1134
1125
933
1362
1030
738
172
896
851
171
701
360
1077
1012
806
1444
1210
790
1081
1061
878
1654
1076
917
1195
1065
887
1087
1099
952
945
1122
915
1126
1164
921
1174
1056
908
1212
1121
969
1092
1077
836
747
908
a
Specific surface areas are in m2 g−1. The t-plot method was used for Fe-ICR-2 and Al-ICR-2; otherwise, BET specific surface areas are given.
b
Activation from water.
Motivated by the successful syntheses of the Fe-ICR MOFs,
in a trigonal bipyramid formation, and every two neighboring
bipyramids are edge-sharing. In the chain, the pairs of edge-
shared bipyramids are connected through their vertices by four
phosphinate tetrahedra. The chains form bilayers that are held
together only by weak nonbonding interactions. This structural
arrangement is isoreticular to Fe-ICR-3.23 A comparison of
both structures is given in Figure S19. Since Fe-ICR-5 is
nonporous, this material was not further investigated.
The permanent porosity of the activated ICR MOFs was
probed by measuring the N2 adsorption isotherms at 77 K
(Figure 1, bottom right; Figure S43). All adsorption isotherms
display a steep N2 uptake at low P/P0 ratios, which is typical
for microporous materials. More specifically, ICR-4 contains
ultramicropores, whereas the pore diameters of ICR-6 and
ICR-7 are at the borderline between micropores and
mesopores (Table 1 and Figures S21−S24).
In order to better understand the porous structures of the
ICR MOFs, the MOFs were computationally analyzed using
the Poreblazer software29,30 for an N2 molecule 3.314 Å in
diameter (Table 1). The obtained parameters for Fe-ICR-2
and Al-ICR-2 are in agreement with the experimental values.
The calculated pore limiting diameter (PLD) of Fe-ICR-4 and
Al-ICR-4 is 2.9 Å; therefore, the accessible surface area cannot
be calculated. Nevertheless, the pores are still accessible to N2,
as evidenced by the corresponding adsorption isotherms
(Figure 1, bottom right; Figure S43). The BET specific
surface areas of Fe-ICR-7 and Al-ICR-7 fit well with the
calculated values, whereas in the case of Fe-ICR-6 and Al-ICR-
6 the BET specific surface areas are considerably lower,
probably due to a lower crystallinity or pore blocking. The
PLDs of Fe-ICR-6 and Al-ICR-6 and Fe-ICR-7 and Al-ICR-7
are considerably smaller than the pore diameters obtained by
the NLDFT method from adsorption isotherms. This differ-
we investigated the structural arrangements of the aluminum-
based ICR MOFs (Al-ICR MOFs). In the case of Al-ICR-4,
the quality of the PXRD pattern allowed for solving the crystal
structure ab initio using the Superflip package28 with the
histogram matching option (Figure S17). Detailed analysis of
the corresponding PXRD pattern confirms nearly the identical
crystal structure of Al-ICR-4 compared to Fe-ICR-423 (Table
S2). In general, the analyses of the PXRD patterns for the Al-
ICR MOFs revealed that they form identical structural motifs
for each linker compared to the Fe-ICR MOFs (Figure S20).
As illustrated for the Fe-ICR MOFs (Figure 1, middle),23
the secondary building units (SBUs) of the Fe-ICR MOFs and
the Al-ICR MOFs (not applicable for Fe-ICR-5, see below for
details) are composed of octahedrally coordinated metal atoms
bound together through OPO bridges that form one-
dimensional (1D) infinity columns. These columns are
connected via phenylene or biphenylene bridges that form
the three-dimensional (3D) honeycomb framework. The Fe
and Al versions of ICR-6 and ICR-7 are isoreticular structures
to Fe-ICR-2, with increased pore sizes due to the incorporated
biphenylene spacer. The crystal structures of Fe-ICR-4 and Al-
ICR-4 are constructed similarly to the structure of Fe-ICR-2,
i.e., the honeycomb arrangement is composed of 1D infinity
columns tied together by OPO bridges. However, the
phenylene groups connecting the 1D columns are not parallel
to each other but instead are crossed and rotated in
neighboring layers (Figure 1, left).
The crystal structure of Fe-ICR-5 was also solved from
PXRD data in this work (Table S2 and Figure S15). Its
structure is layered and composed of 1D infinity columns of
iron atoms coordinated by OPO bridges (Figure S19). In
this case, the oxygen atoms are coordinated to the Fe3+ centers
C
https://dx.doi.org/10.1021/acs.inorgchem.0c00201
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
pubs.acs.org/IC
Article
ence can be caused by the roughness of the pore walls
decreasing the smallest opening of the pores or the
hydrophobic nature of the pores differing significantly from
the chemical nature used by the kernel in the NLDFT method.
The synthesized ICR MOFs are expected to be chemically
stable under harsh conditions, as previously described for Fe-
ICR-2.23 After treatment of the ICR MOFs in water, EtOH,
and toluene at rt or under reflux, the PXRD patterns of most of
the ICR MOFs remained unchanged, suggesting the
preservation of the crystallinity and the original structure
(Figures S25−S42). Only Fe-ICR-6, Al-ICR-6, and Al-ICR-7
recrystallized or lost crystallinity in the boiling water.
Generally, the longer the linker, the lower is the stability of
the MOFs. For example, UiO-66, which was made of the
terephthalate linker, was stable in water and under a humid
atmosphere; however, both UiO-67 and UiO-68, which were
made of biphenyl-4,4′-dicarboxylate and p-terphenyl-4,4″-
dicarboxylate linkers, respectively, decomposed when exposed
to humid air.31 This behavior is not the case for the presented
ICR MOFs.
ICR MOFs possess pores large enough to accommodate the
BPA molecules. The kinetic curves and adsorption isotherms
of the Al-ICR MOFs were compared with those of conven-
tional activated charcoal (abbreviated as AC, Sigma-Aldrich)
measured under identical conditions. Prior to all measure-
ments, the adsorbents were activated under vacuum overnight
at 80 °C.
The adsorption rate is an important factor for practical
application in environmental remediation. Figure 2 depicts the
We also analyzed the effects of these treatments on the
specific surface area of the MOFs (Table 2 and Figures S44−
S61), except for Fe-ICR-4 and Al-ICR-4. In these two cases,
small molecules such as water block the pores and were not
removed even during activation (150 °C, vacuum, 24 h) when
keeping the PXRD pattern intact. Fe-ICR-2, Al-ICR-2, Fe-ICR-
6, Fe-ICR-7, and Al-ICR-7 retained their porosity in the tested
solvents at rt, in boiling toluene, and, with the exception of Fe-
ICR-2, in boiling EtOH. Interestingly, Al-ICR-6 behaved
differently. The specific surface area decreased after the solvent
treatments at rt, while the treatments in boiling EtOH or
toluene resulted in an increase in the specific surface area; this
was probably due to the formation of structural defects. In
general, boiling water represents one of the most challenging
conditions for MOFs. In this respect, Fe-ICR-2, Al-ICR-2, and
Fe-ICR-7 preserved the majority of their porosity. Clearly,
both Fe-ICR-7 and Al-ICR-7 are more solvothermally stable
than the corresponding ICR-6 MOFs. This behavior can be
rationalized by the hydrophobicity of the phenyl groups that
point into the pore accessible volume and effectively shield the
coordination bonds of the linkers.32,33
Figure 2. Kinetics of BPA adsorption by the Al-ICR MOFs compared
with that of AC. Reaction conditions are as follows: an initial BPA
concentration of 50 mg L−1 and 10 mg of the adsorbent dispersed in
50 mL of BPA solution at 25 1 °C. The experimental points are
obtained from triplicate experiments (see Figure S62 for error bars).
recorded kinetic curves. The kinetic parameters, including the
correlation factors obtained by nonlinear fitting to the pseudo-
second order kinetic model, are summarized in Table 3 and
Table 3. Pseudo-Second Order Kinetic Constants and
Langmuir Isotherm Constants Obtained by Nonlinear
a
Fitting to the Experimental Data
kinetic constants
Langmuir constants
We also investigated the stability of the ICR MOFs in regard
to activation from water, i.e., under conditions where wet
MOFs are dried in air without exchanging water for another
solvent before drying. Some water-stable Zr-MOFs, such as
PCN-222 or NU-1000, lose porosity during the activation
process from water.34 In contrast, the phosphinate ICR MOFs,
except for Al-ICR-6, display low variabilities in their surface
areas, indicating the exceptional stability of the porous
structure. The presented experimental results confirm that
the ICR family of MOFs represents robust materials that are
well-suited for applications in an aqueous environment.
Adsorption of Bisphenol A. The robustness, pore size
variability, and hydrophobic nature of the pores prompted us
to investigate the sorption properties of ICR MOFs in regard
to hydrophobic pollutants. For these experiments, we selected
the Al-ICR MOFs, as the pore volume of the Al-ICRs is greater
than that of the Fe analogues, and bisphenol A (BPA), a
pollutant from the family of endocrine disruptors that
represents a significant threat in the food chain. The
adsorption properties of the Al-ICR MOFs were analyzed
using high-performance liquid chromatography (HPLC). With
the exception of Al-ICR-4 (PLD of 2.9 Å), all of the other Al-
qm
(mg g−1
k2
Qm
sample
AC
Al-ICR-2
Al-ICR-4
Al-ICR-6
Al-ICR-7
)
(g mg−1 min−1
)
(mg g−1
)
KL (L mg−1
0.81 0.04
9.61 0.51
0.64 0.58
0.15 0.01
0.62 0.07
)
183
220
n.a.
194
234
2
7
0.022 0.001
0.052 0.002
n.a.
0.010 0.002
0.017 0.002
221
222
22
326
307
4
3
1
8
5
4
1
a
All data points were measured in triplicate experiments: qm is the
amount of BPA adsorbed at equilibrium, k2 is the pseudo-second
order kinetic rate constant, Qm is the Langmuir maximum sorption
capacity, and KL is the Langmuir constant.
Table S3, and the corresponding fits are presented in Figure
S62. Interestingly, the sorption equilibrium for AC, Al-ICR-2,
and Al-ICR-7 was nearly completed within 15 min. By
contrast, Al-ICR-6 behaved differently. The sorption kinetics
indicates two consecutive processes, where a fast initial step is
followed by a slow process so that the equilibrium is not
reached with the time frame of the sorption experiment (i.e.,
360 min). This behavior can be attributed to the slow
rearrangement of BPA molecules inside the pores, indicated by
D
https://dx.doi.org/10.1021/acs.inorgchem.0c00201
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
pubs.acs.org/IC
Article
molecular modeling. These results show that BPA can be
arranged in two positions in the pores of Al-ICR-6 (for details,
see below).
Adsorption isotherms of BPA for the Al-ICR MOFs and AC
(Figure 3) were obtained using initial BPA concentrations
removed BPA at low initial concentrations; however, due to
their high pore volumes, the Qm values are greater than that of
Al-ICR-2. This observation correlates well with the values of
the KL constants of the Langmuir isotherms (Table 3), which
are the measures of the adsorbent−adsorbate affinity. Thus,
the high value for Al-ICR-2 (KL = 9.61) indicates a high affinity
of BPA for Al-ICR-2. In contrast, the KL values for Al-ICR-6
and Al-ICR-7 are more than one order of magnitude lower
(0.15 and 0.62, respectively), indicating that the affinity of BPA
for the ICR MOFs with larger pores is significantly lower
compared to that of Al-ICR-2.
The stability of the MOFs in an aqueous medium is an
important issue affecting their applicability. For this reason, we
also characterized the Al-ICR MOFs by PXRD and N2
adsorption isotherms after the sorption of BPA and
regeneration, done by washing with water and acetone (Figure
S64 and Table S4). These characteristics are in line with the
results presented above in regard to the treatment of the Al-
ICR MOFs with water at rt, confirming that the Al-ICR MOFs
are stable during the sorption experiments. In addition, the
adsorption process is reversible, i.e., BPA can be washed out
from the pores of the Al-ICR MOFs with acetone (Supporting
Information).
Summing up, the Al-ICR MOFs are endowed with greater
adsorption capacities than zeolites, graphene, imprinted
polymers, montmorillonite, and other materials.5 In recent
years, several MOFs were successfully tested as adsorbents of
BPA. The Qm values for typical carboxylate-based MOFs (such
as Fe-MIL-100 and Cr-MIL-101) do not exceed 260 mg g−1.35
Only Al-MIL-53 displays a similar maximum sorption capacity
(325 mg g−1) to Al-ICR-6.36
Figure 3. Adsorption isotherms of BPA expressed as the dependence
of the adsorbed amount of BPA (qe) on the BPA equilibrium
concentration (Ce). The experimental points were obtained from
triplicate experiments, and the solid lines are the corresponding
nonlinear fits to the Langmuir adsorption model. For details, see the
Supporting Information. Conditions are as follows: an initial BPA
concentration between 10 and 120 mg L−1 and 10 mg of the
adsorbent dispersed in 50 mL of BPA solution at 25 1 °C.
Molecular Modeling. We used molecular modeling in
order to analyze the interactions of BPA inside the MOF pores.
As described above, the pores are a hexagonal shape with the
phenyl or methyl substituents bonded to P atoms aiming at the
center of the pore. These substituents and the pore diameter
can influence the interactions and arrangement of guest
molecules as well as the sorption capacity.
The interaction energies between BPA and the Al-ICR
MOFs for relevant BPA amounts adsorbed in the pores are
summarized in Table 4. At low concentrations of BPA in the
framework (one BPA molecule per pore in the supercell, i.e., qe
≈ 10 mg g−1), the interaction energies decrease in the order
Al-ICR-2 > Al-ICR-6 > Al-ICR-7. Snapshots of the BPA
arrangements are given in Figure 4a−c. The interaction
energies in the Al-ICR-2 pores are nearly flat up to the
loading of approximately 200 mg g−1. This behavior is in good
agreement with the observed high affinity of BPA for Al-ICR-2,
indicated by the nearly quantitative adsorption of BPA at these
concentrations and a high KL value.
from 10 to 120 mg L−1 after 24 h of stirring at a constant
temperature (25 1 °C). The experimental data were fitted
using the Langmuir, Freundlich, and Langmuir−Freundlich
adsorption isotherm models (Supporting Information). Best
fits were obtained using the Langmuir model using the
parameters summarized in Table 3. The results indicate that
the sorption capacity for BPA increases in the order of Al-ICR-
4 < AC ≈ Al-ICR-2 < Al-ICR-7 < Al-ICR-6. The highest
adsorption capacity (Qm) was found for Al-ICR-6 (326 mg
g−1), which is approximately 50% greater than the Qm of AC
(221 mg g−1). On the other hand, the adsorption capacity of
Al-ICR-2 is comparable to that of AC, and Al-ICR-4 adsorbs
very little at the external surface due to it narrow pores (PLD
of 2.9 Å).
Interestingly, the course of the adsorption isotherm for Al-
ICR-2 is different from the isotherms of the other adsorbents.
BPA was completely adsorbed from the dispersions with initial
concentrations up to 30 mg L−1, and Al-ICR-2 became fully
saturated at the initial BPA concentration of 50 mg L−1. On the
other hand, Al-ICR-6, Al-ICR-7, and AC only partially
Interestingly, there are two positions for BPA in the Al-ICR-
6 pore (Figure S65), with an interaction energy difference of
a
Table 4. Interaction Energies between BPA and Al-ICR MOFs per BPA Molecule
qe (mg g−1
)
sample
10
100
200
300
n.a.
−16.4 0.2
−15.3 0.5
Al-ICR-2
Al-ICR-6
Al-ICR-7
−22.2 0.4
−20.3 0.4
−18.2 0.4
−21.3 0.3
−17.7 0.2
−16.3 0.3
−21.2 0.3
−17.6 0.2
−16.3 0.2
b
a
b
qe is the adsorbed amount of BPA per gram of the Al-ICR MOF. Interaction energies are given in kcal mol−1
weighted average over two BPA positions, as shown in Figure S65.
.
The interaction energy is the
E
https://dx.doi.org/10.1021/acs.inorgchem.0c00201
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
pubs.acs.org/IC
Article
Preparation of Fe-ICR-6. A Wheaton vial was charged with 37.2
mg of H2BBP(Me) (0.12 mmol) and overlaid with 25 mL of
dimethylformamide (DMF). After 10 min of sonication, 16.2 mg of
FeCl3·6H2O (0.06 mmmol) in 5 mL of DMF was added. The vial was
heated in a preheated oven (Memmert UF30 plus) at 120 °C for 72 h.
The resulting white powder was centrifuged (11,000 rpm, 5 min,
Hettich Rotina 380 R) and washed as described for Fe-ICR-2.
Stability of the ICR MOFs. Twenty milligrams of the MOF was
suspended in 10 mL of H2O, EtOH, or toluene, and the suspension
was shaken for 24 h at rt or refluxed for 24 h. After that, the solid
material was collected by centrifugation and washed twice with water
(only in the case of the stability tests in water) or EtOH (only in the
case of stability tests in EtOH and toluene) and twice with acetone.
The resulting powders were air dried at rt.
Adsorption of BPA. The adsorption experiments were performed
in sealed 100 mL reagent SIMAX glass bottles in a temperature-
controlled room with a constant temperature of 25 1 °C and BPA
concentrations between 10 and 120 mg L−1. The bottles were charged
with 10 mg of the Al-ICR MOF or activated charcoal (AC, DARCO,
100 mesh particle size, powder, Sigma-Aldrich) and 10 mL of water,
followed by 5 min of sonication. Then, 40 mL of a BPA solution was
added. The mixture was stirred for 24 h at 25 °C, and then 1 mL of
sample was taken and filtered through a PTFE microfilter (0.2 μm,
Whatman). The remaining concentration of BPA was analyzed using
HPLC-DAD.
Figure 4. BPA molecules in the 1D pores of (a) Al-ICR-2, (b) ICR-6,
and (c) ICR-7 with a view along the c-axis, qe ≈ 10 mg g−1. (d)
Arrangement of the BPA molecules in the 1D pore of Al-ICR-6 with a
view perpendicular to the c-axis, qe ≈ 30 mg g−1.
2.9 kcal mol−1 (Supporting Information). The existence of two
binding sites and the relocation of the BPA molecule between
these two sites during the simulation can be the reason for the
measured slow adsorption kinetics (Figure 2). At higher BPA
concentrations (up to qe ≈ 300 mg g−1), the interaction energy
decreases due to the BPA-BPA stacking interactions making
the system quite disordered (Figure 4d and Figure S66). The
average interaction energies for Al-ICR-6 and Al-ICR-7
decrease with the increasing loading of BPA, in agreement
with the low KL values found for these materials (Table 3).
ASSOCIATED CONTENT
■
sı
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.inorgchem.0c00201.
Detailed experimental procedures, FTIR spectra, PXRD
patterns, Rietveld fits, TGA curves, details for the
adsorption of BPA, and details of the molecular
modeling (PDF)
CONCLUSIONS
■
We synthesized new ICR-6 and ICR-7 MOFs that were
isoreticular with Fe-ICR-2 and Fe-ICR-4 MOFs described
earlier.23 We also showed that Al3+ cations can be successfully
used for the construction of ICR MOFs. ICR MOFs have high
thermal and solvothermal stability. Due to the hydrophobic
character of the ICR pore walls, ICR MOFs effectively adsorb
BPA with greater sorption capacities than the majority of
already-investigated adsorbents.
Summing up, this work extends the area of phosphinic acid-
based MOFs. The isoreticular design is applicable, and the
wide variety of water-stable MOFs can be prepared using
various substituents at the phosphorus atom. We envision that
the number of phosphinic acid-based MOFs will steeply
increase in the coming years.37
General ICR MOF crystal structure (XYZ)
Accession Codes
CCDC 1919607−1919608 and 1980135 contain the supple-
mentary crystallographic data for this paper. These data can be
obtained free of charge via www.ccdc.cam.ac.uk/data_request/
cif, or by emailing data_request@ccdc.cam.ac.uk, or by
contacting The Cambridge Crystallographic Data Centre, 12
Union Road, Cambridge CB2 1EZ, UK; fax: +44 1223 336033.
AUTHOR INFORMATION
Corresponding Author
■
Jan Demel − Institute of Inorganic Chemistry of the Czech
EXPERIMENTAL SECTION
̌
■
̌
Academy of Sciences, 250 68 Husinec-Rez, Czech Republic;
Preparation of Fe-ICR-2, Fe-ICR-4, Fe-ICR-7, and Al-ICR
MOFs. A Teflon lined autoclave (Berghof DAB-2) was charged with
0.08 mmol of linker and 0.04 mmol of FeCl3·6H2O (10.8 mg) or
AlCl3·6H2O (9.7 mg) and overlaid with 5 mL of absolute EtOH. The
sealed autoclave was heated in a preheated heating mantle (Berghof
BTC-3000) at 250 °C for 24 h. The resulting white powder was
centrifuged (11,000 rpm, 5 min, Hettich Rotina 380 R), washed five
times with EtOH (the third time, the powder was left in EtOH for 2
h), three times with water (the second time, the powder was left in
water overnight), and three times with acetone (the third time, the
powder was left in acetone for 1.5 hours), and activated at 80 °C for 5
h under vacuum.
Preparation of Fe-ICR-5. A Teflon lined autoclave (Berghof
DAB-2) was charged with 0.08 mmol of H2BBP(Me) and 0.04 mmol
of FeCl3·6H2O (10.8 mg) and overlaid with 5 mL of absolute EtOH.
The sealed autoclave was heated in a preheated heating mantle
(Berghof BTC-3000) at 250 °C for 24 h. The resulting white powder
was centrifuged (11,000 rpm, 5 min, Hettich Rotina 380 R), washed
five times with acetone, and dried in air.
orcid.org/0000-0001-7796-6338; Email: demel@iic.cas.cz
Authors
Daniel Bu
Academy of Sciences, 250 68 Husinec-Rez, Czech Republic;
̊
̌
zek − Institute of Inorganic Chemistry of the Czech
̌
̌
̌
Faculty of Environment, Jan Evangelista Purkyne University,
400 96, Czech Republic; orcid.org/0000-0002-7387-9461
̌
̌
́
Sona Ondrusova − Institute of Inorganic Chemistry of the Czech
̌
̌
Academy of Sciences, 250 68 Husinec-Rez, Czech Republic
Jan Hynek − Institute of Inorganic Chemistry of the Czech
Academy of Sciences, 250 68 Husinec-Rez, Czech Republic;
̌
̌
orcid.org/0000-0003-1883-9464
́
̌
Petr Kovar − Charles University, Faculty of Mathematics and
Physics, 121 16 Praha 2, Czech Republic
Kamil Lang − Institute of Inorganic Chemistry of the Czech
̌
̌
Academy of Sciences, 250 68 Husinec-Rez, Czech Republic;
orcid.org/0000-0002-4151-8805
F
https://dx.doi.org/10.1021/acs.inorgchem.0c00201
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
pubs.acs.org/IC
Article
(11) Koh, K.; Wong-Foy, A. G.; Matzger, A. J. A Porous
Coordination Copolymer with over 5000 m2/g BET Surface Area.
J. Am. Chem. Soc. 2009, 131, 4184−4185.
(12) Farha, O. K.; Eryazici, I.; Jeong, N. C.; Hauser, B. G.; Wilmer,
C. E.; Sarjeant, A. A.; Snurr, R. Q.; Nguyen, S. T.; Yazaydın, A. O.;
Hupp, J. T. Metal-Organic Framework Materials with Ultrahigh
Surface Areas: Is the Sky the Limit? J. Am. Chem. Soc. 2012, 134,
15016−15021.
̌
Jan Rohlícek − Institute of Physics of the Czech Academy of
Sciences, 182 21 Praha, Czech Republic; orcid.org/0000-
0001-6913-2667
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.inorgchem.0c00201
Notes
(13) Ali Akbar Razavi, S.; Morsali, A. Linker Functionalized Metal-
Organic Frameworks. Coord. Chem. Rev. 2019, 399, 213023.
(14) Cohen, S. M. Postsynthetic Methods for the Functionalization
of Metal-Organic Frameworks. Chem. Rev. 2012, 112, 970−1000.
(15) Pettinari, C.; Marchetti, F.; Mosca, N.; Tosi, G.; Drozdov, A.
Application of Metal-Organic Frameworks. Polym. Int. 2017, 66, 731−
744.
(16) Drout, R. J.; Robison, L.; Chen, Z.; Islamoglu, T.; Farha, O. K.
Zirconium Metal-Organic Frameworks for Organic Pollutant
Adsoprtion. Trends in Chem. 2019, 1, 304−317.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by the Czech Science Foundation
(no. 20-04408S) and partial project DPK/2018/14 from the
Technology Agency of the Czech Republic (GAMA COMNID
TG02010049). X-ray diffractometers were supported by the
Operational Program Research, Development, and Education
financed by the European Structural and Investment Funds
and the Ministry of Education, Youth, and Sports (no.
SOLID21 CZ.02.1.01/0.0/0.0/16_019/0000760). Computa-
tional resources were supplied by the project “e-Infrastruktura
CZ” (e-INFRA LM2018140) provided within the program
Projects of Large Research, Development, and Innovations
Infrastructures. The authors acknowledge the assistance
provided by the Research Infrastructure NanoEnviCz,
supported by the Ministry of Education, Youth, and Sports
of the Czech Republic under Project no. LM2018124.
(17) Bobbitt, N. S.; Mendonca, M. L.; Howarth, A. J.; Islamoglu, T.;
Hupp, J. T.; Farha, O. K.; Snurr, R. Q. Metal-Organic Frameworks for
the Removal of Toxic Industrial Chemicals and Chemical Warfare
Agents. Chem. Soc. Rev. 2017, 46, 3357−3385.
(18) Feng, M.; Zhang, P.; Zhou, H.-C.; Sharma, V. K. Water-Stable
Metal-Organic Frameworks for Aqueous Removal of Heavy Metals
and Radionuclides: A Review. Chemosphere 2018, 209, 783−800.
̊
̌
(19) Buzek, D.; Demel, J.; Lang, K. Zirconium Metal-Organic
Framework UiO-66: Stability in an Aqueous Environment and Its
Relevance for Organophosphate Degradation. Inorg. Chem. 2018, 57,
14290−14297.
(20) Luzuriaga, M. A.; Benjamin, C. E.; Gaertner, M. W.; Lee, H.;
Herbert, F. C.; Mallick, S.; Gassensmith, J. J. ZIF-8 degrades in cell
media, serum, and somebut not allcommon laboratory buffers.
Supramol. Chem. 2019, 31, 485−490.
REFERENCES
■
́
(1) Nowak, K.; Jabłonska, E.; Ratajczak-Wrona, W. Immunomodu-
latory Effects of Synthetic Endocrine Disrupting Chemicals on the
Development and Functions of Human Immune Cells. Environ. Int.
2019, 125, 350−364.
(21) Dan-Hardi, M.; Serre, C.; Frot, T.; Rozes, L.; Maurin, G.;
́
Sanchez, C.; Ferey, G. A new photoactive crystalline highly porous
titanium(IV) dicarboxylate. J. Am. Chem. Soc. 2009, 131, 10857−
́
(2) Galloway, T. S.; Lee, B. P.; Buric, I.; Steele, A. M.; BPA Schools
10859.
Study Consortium; Kocur, A. L.; Pandeth, A. G.; Harries, L. W.
Plastics Additives and Human Health: A Case Study of Bisphenol A
(BPA). In Plastics and the Environment; Hester, R. E., Harrison, R. M.,
Eds.; Royal Society of Chemistry: Cambridge, U.K., 2019; pp 131−
155.
(3) Michałowicz, J. Bisphenol A − Source, toxicity and
biotranformation. Environ. Toxicol. Pharmacol. 2014, 37, 738−758.
(4) Nam, S. H.; Seo, Y. M.; Kim, M. G. Bisphenol A Migration From
Polycarbonate Babe Bottle with Repeated Use. Chemosphere 2010, 79,
949−952.
(5) Bhatnagar, A.; Anastopoulos, I. Adsorptive Removal of Bisphenol
A (BPA) from Aqueous Solution: A Review. Chemosphere 2017, 168,
885−902.
(6) Wang, F.; Lu, X.; Peng, W.; Deng, Y.; Zhang, T.; Hu, Y.; Li, X. Y.
Sorption Behavior of Bisphenol A and Triclosan by Graphene:
Comparison with Activated Carbon. ACS Omega 2017, 2, 5378−
5384.
(7) Eddaoudi, M.; Kim, J.; Rosi, N.; Vodak, D.; Wachter, J.;
O’Keeffe, M.; Yaghi, O. M. Systematic Design of Pore Size and
Functionality in Isoreticular MOFs and their Application in Methane
Storage. Science 2002, 295, 469−472.
(8) Guillerm, V.; Kim, D.; Eubank, J. F.; Luebke, R.; Liu, X.; Adil, K.;
Lah, M. S.; Eddaoudi, M. A Supermolecular Building Approach for
the Design and Construction of Metal-Organic Frameworks. Chem.
Soc. Rev. 2014, 43, 6141−6172.
(9) Furukawa, H.; Cordova, K. E.; O’Keeffe, M.; Yaghi, O. M. The
Chemistry and Applications of Metal-Organic Frameworks. Science
2013, 341, 1230444.
(22) Taddei, M.; Costantino, F.; Marmottini, F.; Comotti, A.;
Sozzani, P.; Vivani, R. The first route to highly stable crystalline
microporous zirconium phosphonate metal−organic frameworks.
Chem. Commun. 2014, 50, 14831−14834.
́
̌
(23) Hynek, J.; Brazda, P.; Rohlícek, J.; Londesborough, M. G. S.;
Demel, J. Phosphinic Acid Based Linkers: Building Blocks in Metal−
Organic Framework Chemistry. Angew. Chem., Int. Ed. 2018, 57,
5016−5019.
̀
(24) Serre, C.; Millange, F.; Thouvenot, C.; Nogues, M.; Marsolier,
̈
́
G.; Louer, D.; Ferey, G. Very Large Breathing Effect in the First
Nanoporous Chromium(III)-Based Solids: MIL-53 or CrIII(OH)·
{O2C-C6H4- CO2}·{HO2C-C6H4-CO2H}x·H2Oy. J. Am. Chem. Soc.
2002, 124, 13519−13526.
(25) Boultif, A.; Louer, D. Powder pattern indexing with the
dichotomy method. J. Appl. Cryst. 2004, 37, 724−731.
̌
́
̌
̌
́
(26) Cerny, R.; Favre-Nicolin, V.; Rohlícek, J.; Husak, M. FOX,
Current State and Possibilities. Crystals 2017, 7, 1−10.
(27) Materials Studio Modeling Environment, Release 4.3 Documenta-
tion; Accelrys Software Inc.: San Diego, CA, 2003.
(28) Palatinus, L.; Chapuis, G. Superflip - a Computer Program for
the Solution of Crystal Structures by Charge Flipping in Arbitrary
Dimensions. J. Appl. Crystallogr. 2007, 40, 786−790.
(29) Sarkisov, L.; Harrison, A. Computational Structure Character-
isation Tools in Application to Ordered and Disordered Porous
Materials. Mol. Simul. 2011, 37, 1248−1257.
(30) Sarkisov, L.; Harrison, A. Computational structure character-
isation tools in application to ordered and disordered porous
materials. Molecular Simulation 2011, 37 (15), 1248−1257.
(31) Lawrence, M. C.; Schneider, C.; Katz, M. J. Determining the
Structural Stability of UiO-67 with Respect to Time: a Solid-State
NMR Investigation. Chem. Commun. 2016, 52, 4971−4974.
(10) Kim, D.; Liu, X.; Lah, M. S. Topology Analysis of Metal-
Organic Frameworks Based on Metal-Organic Polyhedra as
Secondary or Tertiary Building Units. Inorg. Chem. Front. 2015, 2,
336−360.
G
https://dx.doi.org/10.1021/acs.inorgchem.0c00201
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
pubs.acs.org/IC
Article
(32) Bosch, M.; Zhang, M.; Zhou, H.-C. Increasing the Stability of
Metal-Organic Frameworks. Adv. Chem. 2014, 2014, 182327.
(33) Burtch, N. C.; Jasuja, H.; Walton, K. S. Water Stability and
Adsorption in Metal-Organic Frameworks. Chem. Rev. 2014, 114,
10575−10612.
(34) Mondloch, J. E.; Katz, M. J.; Planas, N.; Semrouni, D.;
Gagliardi, L.; Hupp, J. T.; Farha, O. K. Are Zr6-based MOFs Water
Stable? Linker Hydrolysis vs. Capillary-Force-Driven Channel
Collapse. Chem. Commun. 2014, 50, 8944−8946.
(35) Qin, F.-X.; Jia, S.-Y.; Liu, Y.; Li, H.-Y.; Wu, S.-H. Adsorption
Removal of Bisphenol A from Aqueous Solution Using Metal-Organic
Frameworks. Desalin. Water Treat. 2015, 54, 93−102.
(36) Zhou, M.; Wu, Y.-N.; Qiao, J.; Zhang, J.; McDonald, A.; Li, G.;
Li, F. The Removal of Bisphenol A from Aqueous Solution by MIL-
53(Al) and Mesostructured MIL-53(Al). J. Colloid Interface Sci. 2013,
405, 157−163.
(37) Nakatsuka, S.; Watanabe, Y.; Kamakura, Y.; Horike, S.; Tanaka,
D.; Hatakeyama, T. Solvent-Vapor-Induced Reversible Single-Crystal-
to-Single-Crystal Transformation of a Triphosphaazatriangulene-
Based Metal−Organic Framework. Angew. Chem., Int. Ed. 2020, 59,
1435.
H
https://dx.doi.org/10.1021/acs.inorgchem.0c00201
Inorg. Chem. XXXX, XXX, XXX−XXX