Effect of the Metal within Regioisomeric Paddle-Wheel-Type Metal–Organic Frameworks

Article abstract of DOI:10.1002/chem.201903210

The effect of metal on the degree of flexibility upon evacuation of metal–organic frameworks (MOFs) has been revealed with positional control of the organic functionalities. Although Co-, Cu-, and Zn-based DMOFs (DMOF = DABCO MOF, DABCO = 1,4-diazabicyclo

Full text of DOI:10.1002/chem.201903210

A Journal of
Accepted Article
Title: Effect of the metal within regioisomeric paddle-wheel-type metal-
organic frameworks
Authors: Hyeonbin Ha, Youngik Kim, Dopil Kim, Jihyun Lee, Yoodae
Song, Suyeon Kim, Myung Hwan Park, Youngjo Kim,
Hyungjun Kim, Minyoung Yoon, and Min Kim
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201903210
Link to VoR: http://dx.doi.org/10.1002/chem.201903210
Supported by

10.1002/chem.201903210
Chemistry - A European Journal
FULL PAPER
Effect of the metal within regioisomeric paddle-wheel-type metal-
organic frameworks
Hyeonbin Ha,[a]† Youngik Kim, [a]† Dopil Kim, [a]† Jihyun Lee,[b] Yoodae Song,[b] Suyeon Kim,[a] Myung
Hwan Park,[c] Youngjo Kim,[a] Hyungjun Kim,*[d] Minyoung Yoon,* [b,e] and Min Kim*[a]
Abstract: The effect of metal on the degree of flexibility upon
evacuation of metal-organic frameworks (MOFs) has been revealed
with positional control of the organic functionalities. While Co-, Cu-,
and Zn-based DMOFs with ortho-ligands (2,3-NH2Cl) have
frameworks that are inflexible upon evacuation, MOFs with para-
ligands (2,5-NH2Cl) showed different N2 uptake amounts after
evacuation by metal exchange. Since the structural analysis were
not fully sufficiently different to explain the drastic changes in N2
adsorption after evacuation, quantum chemical simulation was
explored. A new index (η) was defined to quantify the regularity
around the metal based on differences in the oxygen-metal-oxygen
angles. Within 2,5-NH2Cl, the η value becomes larger as the metal
are varied from Co to Zn. A large η value means that the structures
around the metal center are less ordered. These results can be used
to explain flexibility changes upon evacuation by altering the metal
cation in this regioisomeric system.
the organic linker. These hybrid materials have been intensely
studied for a variety of applications, such as gas storage,
molecular separation, and catalysis, due to their unique porous
characteristics with a variety of functional groups.[1–4] Since
MOFs are constructed by infinite coordination bonds between
metal ions and chelating ligands, they are also called porous
coordination polymers (PCPs). These coordination bonds could
give MOFs some structural flexibility, and the framework and/or
the pore size could be changed by this flexible nature. External
stimuli, such as the presence of guest molecules, molecular
adsorption, and solvent evacuation, could allow structural
changes in flexible MOF systems.[5,6]
The flexible structures of MOFs are mainly dependent on the
type of secondary building unit (SBU) and coordinated ligands
present. For example, zinc-based IRMOFs (isoreticular MOFs),
which consists of Zn4O SBU and benzene-1,4-dicarboxylate
(BDC) ligands, displayed a very rigid structure. However, zinc-
paddle wheel-type DMOFs (DABCO MOF, DABCO = 1,4-
diazabicylco[2.2.2]octane), which consist of Zn2 paddle-wheel
SBU, BDC, and DABCO ligands, showed a flexible structure
upon external stimuli.[5,6] In addition, aluminum-based MOFs also
showed flexibility differences based on wine-rack type
structures.[7,8] While MIL-53(Al)-NH2 (MIL = Materials from
Institute of Lavoisier) showed flexible structures depending on
external stimuli, and another aluminum-based MOF, MIL-
101(Al)-NH2, is known for its rigidity.[5]
Introduction
Metal-organic frameworks (MOFs) are three-dimensional,
porous, organic-inorganic hybrid materials that consist of
inorganic nodes and multivalent organic struts. The size of the
pore and pore environment can be tuned by varying the
combination of metal cluster (or ion) and functional groups on
Although the flexibility of MOFs naturally arises from the type of
coordination bonds between the metal nodes and strut, the
functional groups (i.e., chemical tag) on the organic linkers also
have significant effects on the flexibility changes of MOFs.
Fischer and coworkers studied the functionalization of flexible
MOFs in great detail. BDC ligands with a series of different
functionalities/alkyl chain lengths were prepared and used to
construct to DMOFs. The flexibility of the DMOFs was controlled
by the alkyl chain length and the degree of saturation in the
chemical tags.[9,10] Recently, Cohen’s group and our group have
revealed that not only direct interframework interactions between
two linkers but also the electronic environment of the ligands are
significantly correlated with the flexibility of the MOFs.[11–13] We
prepared a series of bifunctionalized BDC ligands with two
chemical handles along with regioisomeric differences. All
symmetrical and unsymmetrical combinations of NH2, NO2, OMe,
and Cl functionalities at the ortho- and para-positions were
synthesized and used to construct DMOFs. Among the 10
combinations, only diamino-functionalized BDCs could not
successfully be incorporated into DMOFs due to their low
thermal stability and solubility under solvothermal conditions.[13]
Of the remaining 9 combinations, only para-substituted amino-
halo, amino-methoxy, and dimethoxy-functionalized DMOFs (i.e.,
DMOF-2,5-NH2X, DMOF-2,5-NH2OMe, and DMOF-2,5-(OMe)2)
[a]
[b]
H. Ha, Y. Kim, D. Kim. S. Kim, Prof. Dr. Y. Kim, Prof. Dr. M. Kim
Department of Chemistry and BK21Plus Research Team
Chungbuk National University
1 Chungdae-ro, Seowon-gu, Cheongju, 28644, Republic of Korea
E-mail: minkim@chungbuk.ac.kr
J. Lee, Y. Song, Prof. M. Yoon
Department of Nanochemistry
Gachon University
1342 Seongnamdae-ro, Sujeong-gu, Seongnam, 13120, Republic of
Korea
E-mail: myyoon@gachon.ac.kr
Prof. Dr. M. H. Park
Department of Chemistry Education
Chungbuk National University
1 Chungdae-ro, Seowon-gu, Cheongju, 28644, Republic of Korea
Prof. Dr. H. Kim
Department of Chemistry
Incheon National University
119 Academy-ro, Yeonsu-gu, Incheon 22012, Republic of Korea
E-mail: kim.hyungjun@inu.ac.kr
Prof. Dr. M. Yoon
[c]
[d]
[e]
Department of Chemistry
Kyungpook National University
80 Daehak-ro, buk-gu, Daegu, 41566, Republic of Korea
E-mail: myyoon@knu.ac.kr
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/chem.201903210
Chemistry - A European Journal
FULL PAPER
resulted in flexible frameworks upon evacuation. These three
ligands have more electron-rich benzene rings relative to those
in the other ligands (e.g., NH2-NO2, NO2-OMe, OMe-Cl, NO2-Cl,
NO2-NO2, and Cl-Cl) based on their Hammett constants.[12,13]
Herein, we focused on the effect of the metal on the magnitude
of the flexibility changes in the regioisomeric DMOF systems.
Three DMOFs of cobalt, copper, and zinc with ortho- and para-
NH2-Cl-functionalities will be compared based on their gas
adsorption differences after evacuation. There are several
reports about the relationships between flexibility (as well as
breathing behavior and gate-opening) and the metal cation on
MOFs.[14–19] In particular, Kaskel and coworkers prepared a
autoclave. Both DMOF(Co)-2,3-NH2Cl and DMOF(Co)-2,5-
NH2Cl were also successfully synthesized and were obtained as
purple crystals (Scheme 1). However, the nickel-based DMOFs,
DMOF(Ni)-2,3-NH2Cl and DMOF(Ni)-2,5-NH2Cl, were not
obtained after a thorough screening of synthetic conditions.
Since there is limited literature on functional group-containing Ni-
MOFs from BDC-type ligands, we explored the synthesis of
mono-functionalized
DMOF(Ni)
species
rather
than
regioisomeric, bifunctional species. Interestingly, we could
obtain DMOF(Ni)-1-Cl, but not DMOF(Ni)-1-NH2, under
solvothermal conditions (Figure S1, see the Supporting
Information for details). To date, there are very limited reports
about Ni(II)-based MOFs with NH2-functionalized ligands.[22,23]
We hypothesize that the coordination between nickel(II) and
BDC-NH2 prevents the formation of DMOFs under solvothermal
conditions.
series of DUT-8 MOFs (DUT
= Dresden University of
Technology) with Ni, Cu, Co, and Zn and studied the magnitude
of the changes in flexibility with different metal nodes. DUT-8
species are also DABCO-based pillared MOFs with a slightly
tilted 2,6-naphthalenedicarboxylic acid). In their results, Ni and
Co produces flexible and reversible frameworks, Zn-based DUT-
8 formed flexible-irreversible frameworks, and Cu-based DUT-8
formed inflexible structures.[14] Very recently, Fischer’s research
team studied the effect of the metal on methoxyethoxy-
functionalized DMOFs. Both the cobalt and nickel derivatives
showed structural swelling and switching, and the copper-based
MOFs and zinc-based MOFs showed only structural
switching.[19] Both Kaskel’s DUT-8 studies and Fischers’s DMOF
studies clearly indicate that the metal cation could impact the
flexibility of DMOFs; thus, in this study, we want to focus the
relationship between the regioisomer of the ligand and the
flexibility of MOFs with a metal salt.
Results and Discussion
Synthesis of the Ligands and MOFs
Both BDC-2,3-NH2Cl (1) and BDC-2,5-NH2Cl (2) ligands were
prepared by following reported procedures.[11] Regioisomeric
mixtures of methyl esters 1 and 2 were synthesized by the
chlorination of dimethyl-2-amino terephthalate (BDCE-NH2), and
esters 1 and 2 were separated by column chromatography. The
hydrolysis of each ester with potassium hydroxide gave desired
ligand 1 or 2 as the carboxylic acid (Scheme S1). Using these
regioisomeric ligands, the syntheses of DMOFs with different
metal salts were attempted. Zinc-based DMOFs was first
reported by K. Kim et al. in 2003.[20] Walton and coworkers
successfully prepared cobalt-, nickel-, copper-, and zinc-based
DMOF-1 without organic functional groups and studied their
stability to humidity.[21] Based on their synthetic conditions with
different metal salts, we attempted to synthesize DMOF(Co),
DMOF(Ni), DMOF(Cu), and DMOF(Zn) with two regioisomeric
ligands, BDC-2,3-NH2Cl and BDC-2,5-NH2Cl. First, for zinc, both
Scheme 1. Synthesis of regioisomeric DMOFs with different metal salts
(colored circle = M2 secondary building unit).
Finally, we studied the synthesis of copper-based DMOFs.
Interestingly, both DMOF(Cu)-2,3-NH2Cl and DMOF(Cu)-2,5-
NH2Cl were obtained as powdery microcrystals (Scheme 1).
Although the morphology varied with the metal salt (Co, Cu, and
Zn), the DMOF structures were perfectly constructed with the
two regioisomeric ligands, as evidenced by power X-ray
1
diffraction (PXRD, Figure 1). H NMR measurements after acid
digestion confirmed that each ligand was completely retained in
the frameworks (Figure S2) and infrared (IR) spectra clearly
confirmed that there were almost no trapped, remaining BDC-
NH2Cl and DABCO ligands after washing process of DMOFs
(by DMF and chloroform, Figures S3-S5). Thermogravimetric
analysis (TGA) showed that all six regioisomeric DMOFs with
different metal salts offered good thermal stability. Generally,
DMOF(Cu)s showed decomposition temperatures >250 °C, and
DMOF(Zn)-2,3-NH2Cl
and
DMOF(Zn)-2,5-NH2Cl
were
successfully obtained as colorless crystals, as reported (Scheme
1).[11] Second, in the case of cobalt, cobalt(II) nitrate was used
as the metal source instead of Zn(NO3)2. The solvothermal
syntheses of the DMOF(Co)s had to be performed in an
This article is protected by copyright. All rights reserved.

10.1002/chem.201903210
Chemistry - A European Journal
FULL PAPER
DMOF(Co)s/DMOF(Zn)s are thermally stable up to 300 °C
(Figure S6). Around 36-55 weight% losses were observed by
ligand portion before 500 oC (Figure S7-S9).
Figure 1. PXRD patterns of regioisomeric DMOFs with different metals.
N2 Adsorption Experiments of Regioisomeric DMOFs
The N2 adsorption differences among six different regioisomeric
DMOFs after evacuation were studied at 77 K (Figures 2 and
S10). In previous studies, flexibility changes upon evacuation
were observed by comparing the amounts of N2 adsorbed with
NH2-Cl, NH2-Br, NH2-OMe and OMe-OMe substituents at
different positions. In other words, DMOF(Zn)-2,5-NH2Cl, which
has narrow pores after evacuation, is almost nonporous to N2,
meaning that it has a flexible-irreversible structure. However,
DMOF(Zn)-2,3-NH2Cl, which is not collapsed after evacuation,
shows high N2 uptake, indicating the structure is inflexible to
evacuation. The N2 uptake difference is approximately 350
m2/g.[11] Once again, the as-synthesized DMOF(Zn)-2,3-NH2Cl
and DMOF(Zn)-2,5-NH2Cl have almost identical structures (i.e.,
very similar PXRD patterns). Interestingly, in the case of cobalt-
based DMOFs, there are no significant differences in the N2
uptakes of DMOF(Co)-2,3-NH2Cl and DMOF(Co)-2,5-NH2Cl
(Figure 2 and Table 1) upon evacuation. The difference in N2
uptake is only <50 m2/g. The type of isotherm and uptake
amount follow patterns typical of DMOFs. This result strongly
indicates that the flexibility of cobalt-based DMOFs was not
altered by the regioisomer of the BDC-NH2Cl ligand (1 or 2).
Table 1. N2 uptake amount by DMOFs with regioisomeric ligands (at a
relative pressure of P/P0 = 0.95)
DMOF
DMOF(Co)
354 cm3/g
DMOF(Cu)
473 cm3/g
DMOF(Zn)[9]
363 cm3/g
BDC-2,3-NH2Cl
Figure 2. N2 full isotherm at 77 K of regioisomeric DMOFs with different
metals.
BDC-2,5-NH2Cl
313 cm3/g
306 cm3/g
5 cm3/g
This article is protected by copyright. All rights reserved.

10.1002/chem.201903210
Chemistry - A European Journal
FULL PAPER
DABCO pillar ligand of DMOF(Zn)-2,3-NH2Cl was slightly longer
than the M-N bond in DMOF(Zn)-2,5-NH2Cl. However, the large
In the case of copper-based DMOFs, moderate differences in N2
uptake between DMOF(Cu)-2,3-NH2Cl and DMOF(Cu)-2,5-
NH2Cl were observed (~170 m2/g). Nevertheless, DMOF(Cu)-
2,3-NH2Cl showed a regular DMOF-1 N2 full isotherm pattern,
meaning it is inflexible. The N2 uptake by DMOF(Cu)-2,5-NH2Cl
is lower than that of DMOF(Cu)-2,3-NH2Cl; however, this higher
N2 uptake was in contrast with the almost nonporous DMOF(Zn)-
2,5-NH2Cl. In 2015, Verpoort et al. reported four different highly
porous DMOFs based on BDC-Ni, Co, Cu, and Zn. Their N2
uptake amounts at saturation were approximately 400~500
cm3/g.[24] Therefore, the low N2 uptakes of DMOF(Cu)-2,5-NH2Cl
and DMOF(Zn)-2,5-NH2Cl are caused by the functional groups
N2
adsorption
differences
between
and/or
DMOF(Zn)-2,3-
DMOF(Co)-2,5-
NH2Cl/DMOF(Zn)-2,5-NH2Cl
NH2Cl/DMOF(Zn)-2,5-NH2Cl are not fully explained by the small
structural differences in their solvated crystals. Moreover, the
states of the samples are not matched since the N2 adsorption
experiments were conducted on activated (i.e., evacuated and
fully dried) samples, and the crystal structures were obtained
from the solvated species.
Table 2. M-O bond distances of DMOF(Co)-2,5-NH2Cl, DMOF(Zn)-2,3-
NH2Cl, and DMOF(Zn)-2,5-NH2Cl in the solvated crystal structures
and
regioisomerism-derived
flexibility-irreversibility
upon
evacuation before N2 adsorption. Once again, these six
regioisomeric DMOFs showed almost identical PXRD patterns in
their as-synthesized states, and the differences in their N2
uptake amounts varied between regioisomers.
average
d (Å)
(M-O)
d (Å)
(M-O1)
d (Å)
(M-O2)
d (Å)
(M-O3)
d (Å)
(M-O4)
DMOFs
DMOF(Co)
-2,5-NH2Cl
2.0319
1.9437
2.0216
1.9437
2.0488
2.0609
2.0552
2.0609
2.0394
2.0023
Structural Analysis of the DMOFs
DMOF(Zn)
-2,3-NH2Cl
(solvated)
The functionalization studies provided experimental evidence of
the relationship between electron density and the flexibility of the
MOF. More electron-rich ligands could allow flexibility changes
by regioisomerism (BDC-NH2-X, BDC-NH2-OMe, and BDC-
(OMe)2).[11–13] In the present study, we have the same functional
groups on the ligands but different metals; thus, the coordination
bonds between the SBUs and ligands should be carefully
analyzed. The single-crystal structures of solvated DMOF(Zn)-
2,3-NH2Cl and DMOF(Zn)-2,5-NH2Cl have been reported
previously,[11] and the structure of solvated DMOF(Co)-2,5-
NH2Cl and desolvated (dried) DMOF(Zn)-2,3-NH2Cl was solved
for the first time in this work (Figures S11, S12 and Tables S1,
S2). The diffraction data of DMOF(Co)-2,3-NH2Cl was not
suitable for determining its crystal structure. Unlike DMOF(Co)-
2,3-NH2Cl, DMOF(Zn)-2,5-NH2Cl gives nice diffraction pattern,
but it fails to determine the unit cell using the diffraction data.
Despite good diffraction peaks in PXRD, it was impossible to
even index the unit cell of the framework. The overall structures
of DMOF(Co)-2,5-NH2Cl (solvated), DMOF(Zn)-2,3-NH2Cl
(solvated), DMOF(Zn)-2,3-NH2Cl (desolvated) and DMOF(Zn)-
2,5-NH2Cl (solvated) from their crystallographic data are
consistent with the reported structures of DMOFs, as originally
evidenced by their PXRD patterns (Figures 1, S11, and S12).
Then, their single-crystal structures were carefully analyzed with
a focus on the metal cluster geometries and bond lengths
(Tables 2 and S3). First, we compared the metal-oxygen bond
(M-O bond) distances (dM-O) of the DMOFs. Interestingly,
DMOF(Zn)-2,3-NH2Cl has the shortest average M-O bond length,
whereas the M-O bond lengths in DMOF(Co)-2,5-NH2Cl and
DMOF(Zn)-2,5-NH2Cl are similar. The small difference between
the M-O bond lengths of DMOF(Co)-2,5-NH2Cl and DMOF(Zn)-
2,5-NH2Cl may be derived from the ionic radius of Co being
smaller than that of Zn. The shorter bond length (0.042 Å , 2.1%)
in DMOF(Zn)-2,3-NH2Cl relative to those of DMOF(Co)-2,5-
NH2Cl and DMOF(Zn)-2,5-NH2Cl suggests slightly stronger M-O
bonding. In addition, the metal-nitrogen bond length in the
DMOF(Zn)
-2,3-NH2Cl
(desolvate
d)
2.0190
2.0364
2.0190
2.0447
2.0190
2.0671
2.0190
2.0289
2.0190
2.0443
DMOF(Zn)
-2,5-NH2Cl
(solvated)
DMOF(Zn)
-2,3-NH2Cl
(desolvate
d)
Not
determin
ed
Not
determin
ed
Not
determin
ed
Not
determin
ed
Not
determin
ed
The flexible-irreversible structural changes in DMOF(Zn)-2,5-
NH2Cl, unlike the other DMOFs (DMOF(Co)s, DMOF(Cu)s, and
DMOF(Zn)-2,3-NH2Cl), were observed based on the PXRD
pattern shift after evacuation (Figure S13). Other DMOFs
showed no significant shifts in their PXRD data. Since the PXRD
data could provide structural information, we attempted to collect
PXRD data with better resolution and signal-to-noise ratio
(Figures S14-S22). We tried to determine the unit cell
parameters of the DMOFs (Table S4). The both DMOFs
possessing cobalt ions have similar unit cell parameters
(DMOF(Co)-2,3-NH2Cl and DMOF(Co)-2,5-NH2Cl), which
supports their similar N2 uptakes (Figures S14-S16). However,
we found a significant increase in the background noise in the
PXRD pattern for DMOF(Cu)-2,5-NH2Cl, which suggests partial
decomposition of the framework. Because of the partial
decomposition, DMOF(Cu)-2,5-NH2Cl has a smaller N2 sorption
amount compared to that of DMOF(Cu)-2,3-NH2Cl (Figures S17-
S19). The unit cell parameters of DMOF(Zn)-2,3-NH2Cl were
also determined, and they were similar to the values calculated
by single-crystal structure analysis. When the PXRD patterns of
DMOF(Zn)-2,3-NH2Cl and DMOF(Zn)-2,5-NH2Cl were compared
(Figures S20-S22), we found that the major peaks
This article is protected by copyright. All rights reserved.

10.1002/chem.201903210
Chemistry - A European Journal
FULL PAPER
corresponding to the (n,0,0) planes shifted to higher 2 theta
values, which may be due to the shrinking of the framework (unit
cell parameter reduction). In addition, many peaks in the PXRD
pattern of DMOF(Zn)-2,5-NH2Cl disappeared, but the peaks
corresponding to the (n,0,0) planes did not. Although it is difficult
to say that the structure of DMOF(Zn)-2,5-NH2Cl was destroyed,
it is clear that the DMOF(Zn) structures differ depending on the
position of the functional groups on the BDC ligand. Because of
the significant disappearances and decreases in diffraction peak
intensity, it was impossible to determine the unit cell from the
PXRD unit cell indexing process of DMOF(Zn)-2,5-NH2Cl. We
can estimate the unit cell parameters from the peak position of
the MOFs and DMOF(Zn)-2,5-NH2Cl has much smaller 2theta
values in PXRD compared to that of DMOF(Zn)-2,3-NH2Cl,
which means the smaller unit cell parameter of the framework. In
addition, peak intensities of the framework have been drastically
changed, which also indicate local structure change in the
framework. Although it was unable to determine the unit cell
parameter of DMOF(Zn)-2,5-NH2Cl, the significant difference of
the PXRD pattern, implying an unusual change in the MOF
structure.
Figure 3. The predicted ground state structures of the model systems having
Co, Cu, and Zn metal centers. The atoms directly connected to the metal ions
are described with ball-and-stick models, and other atoms are described with
simple line models. The hydrogen atoms have been omitted for clarity. Color
scheme: carbon-cyan, nitrogen-blue, oxygen-red, chlorine-brown, cobalt-
magenta, copper-green, and zinc-dark gray.
Quantum Chemical Simulation Approaches
The ground state structures of the model systems having
substituents at the 2,3- (ortho-, top) and 2,5-positions (para-,
bottom) predicted by the quantum chemical simulations are
displayed in Figure 3. (The top and side views are available in
the Supporting Information, Figures S23-S25.) All six DMOF
models in Figure 3 have five-coordinate metal centers, and their
overall shapes are square pyramidal. Notably, the quantum
chemical simulations located the minimum-energy structures of
four-coordinate nickel ions (three oxygen atoms of carboxyl
groups and one nitrogen of a DABCO) rather than five-
coordinate metal centers like the others. This result provides one
piece of evidence as to why we have not been able to
synthesize DMOF(Ni)s (the optimized DMOF(Ni) structures are
available in the Supporting Information, Figure S26). Inspired by
the geometry index used in structural chemistry and
crystallography,[26] we defined a new index (η) to quantify the
regularity around the metal ions. This index measures the
difference between the two largest oxygen-metal-oxygen angles
(∠OMOs, M: metal, and O: oxygen) in the highlighted region of
Figure 3. There are two metal centers in the DMOF model;
therefore, the numerical value of η is given as the average of
two measurements. A zero value of η indicates that the structure
around the metal atom is highly ordered regardless of its point
group symmetry. In other words, both perfect square pyramidal
geometries (i.e., the two ∠OMOs are the same, and the values
are 180 ) and distorted square pyramidal geometries (the two ∠
OMOs are the same but are <180 ) could have η values equal to
0. A large η value means that the two ∠OMOs are significantly
different. Therefore, the structures around the metal center are
less ordered and are less sensitive to subtle geometric changes.
This results in a number of different feasible local minimum
structures; therefore, compounds with low symmetry would be
more flexible than highly symmetric compounds.
Next, we investigated the coordination environment differences
of the simplified metal-ligand complexes using quantum
chemical simulation approaches. Since the size of a MOF is
nearly infinite, its structure is too large to be illustrated quantum
chemically with current computational capabilities. Therefore,
simplified model structures, M2(BDC)(DABCO) complexes in
which two divalent metal ions were held by four functionalized
carboxylate ligands and capped with two DABCO units, were
studied (e.g., Co2(BDC-2,3-NH2Cl)4(DABCO)2/Co2(BDC-2,5-
NH2Cl)4(DABCO)2,
2,5-NH2Cl)4(DABCO)2,
NH2Cl)4(DABCO)2/Zn2(BDC-2,5-NH2Cl)4(DABCO)2).
Cu2(BDC-2,3-NH2Cl)4(DABCO)2/Cu2(BDC-
and Zn2(BDC-2,3-
The
electronic mapping on the simplified repeating units is universal
methodology on quantum chemical simulation approaches for
polymeric material. The main interest of the current study is the
flexibility upon evacuation, which is expected to strongly
correlate with the structural environment around the metal
dications. This model is expected to reproduce the original
MOF’s electronic and geometric features since it incorporates all
the organic fragments directly bound to the metal dications.
There are several possible conformers due to the relative
orientation of the amine and chlorine substituents on the
benzenes with respect to those on nearby benzenes. In this
study, we focused on the statistically most likely combination–
two amine groups (chlorine atoms) on the same side and the
other two amine groups (chlorine atoms) on the opposite side.
The minimum-energy structures were located by density
functional theory (DFT) employing the ωB97X-D functional. The
6-31G* basis sets were used for the light atoms (hydrogen,
carbon, oxygen, and chlorine), and LANL2DZ ECP and the
corresponding basis sets were selected for describing the metal
atoms. All quantum chemical simulations were performed with
Q-Chem 5.1.[25]
Table 3 lists the numerical values of ∠OMO of the six DMOF
models studied herein. The DMOF structures with cobalt as the
This article is protected by copyright. All rights reserved.

10.1002/chem.201903210
Chemistry - A European Journal
FULL PAPER
central metal atoms are nearly perfect square pyramids, and the
two ∠OMO values are nearly the same (approximately 180 )
regardless of the positions of the substitutions. The DMOF(Cu)s
feature more distorted square pyramidal geometries than are
found in the cobalt derivatives. The largest deviation from
square pyramidal geometry was observed for the DMOF(Zn)s.
The values of ∠OMO were significantly lower (150~160°). While
DMOF(Zn)-2,3-NH2Cl is well-ordered (i.e., the two ∠OMOs are
nearly the same, and they give an η value of only 0.5), the
coordination environment of DMOF(Zn)-2,5-NH2Cl is quite
asymmetric. Generally, DMOF(M)-2,3-NH2Cl yields structures
that are relatively more symmetric and ordered than those of
DMOF(M)-2,5-NH2Cl, which could explain the observed rigidity
(inflexibility) of DMOF(M)-2,3-NH2Cl (M = Co, Cu, and Zn).
Within the DMOF(M)-2,5-NH2Cl series, the η values become
larger as the metal atoms are varied from Co to Zn. As a small η
value implies a well-ordered structure, these results explain the
inflexibility of DMOF(Co)-2,5-NH2Cl, lower flexibility of
DMOF(Cu)-2,5-NH2Cl, and flexibility of DMOF(Zn)-2,5-NH2Cl
upon evacuation. We found that the trend in the O-M-O angles
does not exactly match the X-ray crystallographic data of
solvated DMOF(Co)-2,5-NH2Cl, DMOF(Zn)-2,3-NH2Cl, and
DMOF(Zn)-2,5-NH2Cl (Table 2 and Table 3). The discrepancy
may be derived from the difference of the solvent system and
the unit-framework.
the changes in flexibility with two regioisomeric ligands (BDC-
2,3-NH2Cl and BDC-2,5-NH2Cl). Although regioisomeric
functionalized DMOFs with nickel salts were not obtained, we
could clearly observe the trends and relationships between the
metal salts and changes in flexibility. All six regioisomeric
DMOFs have similar frameworks their as-synthesized forms, as
evidenced by their PXRD patterns; however, different
regioisomers of the ligands resulted in different magnitudes of
changes in their N2 uptakes. In particular, the N2 uptake
differences are very small in cobalt-based DMOFs, moderate in
copper-based DMOFs, and large in zinc-based DMOFs.
Although all DMOF(M)-2,3-NH2Cl derivatives showed high
porosity after evacuation and followed the typical N2 full isotherm
of DMOFs at 77K, DMOF(M)-2,5-NH2Cl derivatives displayed
very different N2 uptake behaviors. High porosity for DMOF(Co)-
2,5-NH2Cl, moderate porosity for DMOF(Cu)-2,5-NH2Cl, and
poor porosity for DMOF(Zn)-2,5-NH2Cl after evacuation were
observed. This result means that DMOF(Zn)-2,5-NH2Cl has a
flexible-irreversible structure upon evacuation, but the N2 uptake
of DMOF(Cu)-2,5-NH2Cl is reduced, and DMOF(Co)-2,5-NH2Cl
is inflexible. All three DMOF(M)-2,3-NH2Cl species were, of
course, inflexible upon evacuation (M = Co, Cu, and Zn).
While the PXRD patterns of all six as-synthesized DMOFs
matched and there were no significant differences in the X-ray
crystallographic data of solvated DMOF(Co)-2,5-NH2Cl,
DMOF(Zn)-2,3-NH2Cl,
and
DMOF(Zn)-2,5-NH2Cl,
only
DMOF(Zn)-2,5-NH2Cl (i.e., flexible-irreversible structures)
showed significant changes in their PXRD patterns after
evacuation. In addition, quantum chemical simulation
approaches helped clarify our findings. A new index (η) was
defined to quantify the regularity of the structural environment
around the metal ion based on differences in the oxygen-metal-
oxygen bond angles ( ∠ OMOs). Within the BDC-2,5-NH2Cl
series, the η value becomes larger as the metal atoms are
varied from Co to Zn. As a small η value implies a well-ordered
structure, these results justify the inflexibility of DMOF(Co)-2,5-
NH2Cl, the limited flexibility of DMOF(Cu)-2,5-NH2Cl, and the
flexibility of DMOF(Zn)-2,5-NH2Cl upon evacuation. Finally, in
the case of DMOF(Ni)s, the nickel unit cells with BDC-2,3-NH2Cl
and BDC-2,5-NH2Cl were not formed in the quantum chemical
simulation because the minimum energy was identified with a
four-coordinate nickel species. This result is consistent with our
experimental results; our thorough testing of synthetic conditions
for DMOF(Ni)-2,3/2,5-NH2Cl were not successful. This study
suggests that the metal cation, coordination environment of the
metal-ligand, and the position of functional groups in ligand are
important for MOF synthesis and the resulting properties,
including flexibility.
Table 3. Numerical values of ∠OMO () and index η () for DMOF(M)-2,3-
NH2Cl and DMOF(M)-2,5-NH2Cl. (M=Co, Cu, and Zn)
MOFs
Parameters
Co
Cu
Zn
176.5
176.0
176.6
175.8
0.6
173.8
164.1
171.2
163.2
8.9
157.0
156.1
153.7
153.6
0.5
1st metal center
∠OMO ()
DMOF(M)-2,3-
NH2Cl
2nd metal center
∠OMO ()
η
176.8
176.1
175.7
175.5
0.5
171.5
164.0
174.2
163.5
9.1
160.9
150.0
160.6
149.6
11.0
1st metal center
∠OMO ()
DMOF(M)-2,5-
NH2Cl
2nd metal center
∠OMO ()
η
Experimental Section
Conclusions
DMOF(Co)-2,3-NH2Cl: BDC-2,3-NH2Cl (1, 107 mg, 0.5 mmol) and
Co(NO3)2·6H2O (146 mg, 0.5 mmol) were dissolved in 12.5 mL of DMF.
The pillar ligand (DABCO, 56 mg, 0.5 mmol) was added to this mixture. A
purple gel formed upon the addition of DABCO. The gel was filtered
through a fritted disc of fine porosity. The solution was then transferred to
an autoclave and heated at a rate of 2.5 °C/min from room temperature
Four new, bifunctional (NH2-Cl) DMOFs have been successfully
synthesized using cobalt and copper salts. A total of six series of
DMOFs (including two previously reported Zn-DMOFs) were
compared to study the effect of the metal on the magnitude of
This article is protected by copyright. All rights reserved.

10.1002/chem.201903210
Chemistry - A European Journal
FULL PAPER
to 120 °C. The temperature was then held for 48 h, and the material was
cooled to room temperature at a rate of 2.5 °C/min. The resulting crystals
were then washed three times with 5 mL of DMF. The solvent was then
exchanged with chloroform (5 mL) over three days, replacing the old
chloroform with fresh chloroform every 24 h.
Keywords: metal-organic frameworks • coordination polymers •
metal cation effect • flexibility • regioselectivity
[1]
[2]
[3]
[4]
[5]
H.-C. Zhou, J. R. Long, O. M. Yaghi, Chem. Rev. 2012, 112, 673–
674.
DMOF(Co)-2,5-NH2Cl: BDC-2,5-NH2Cl (2, 107 mg, 0.5 mmol) was used
instead of BDC-2,3-NH2Cl (1) with the same protocol as was used for the
synthesis of DMOF(Co)-2,3-NH2Cl.
L. J. Murray, M. Dincă, J. R. Long, Chem. Soc. Rev. 2009, 38,
1294–1314.
J.-R. Li, R. J. Kuppler, H.-C. Zhou, Chem. Soc. Rev. 2009, 38,
1477–1504.
DMOF(Cu)-2,3-NH2Cl: BDC-2,3-NH2Cl (1, 107 mg, 0.5 mmol) and
Cu(NO3)2·3H2O (121 mg, 0.5 mmol) were dissolved in 12.5 mL of DMF.
The pillar ligand (DABCO, 56 mg, 0.5 mmol) was added to this mixture. A
green gel formed upon the addition of DABCO. The mixture was then
transferred to a scintillation vial and heated at a rate of 2.5 °C/min from
room temperature to 120 °C. The temperature was then held for 24 h,
and then the material was cooled to room temperature at a rate of
2.5 °C/min. The resulting crystals were then washed three times with 5
mL of DMF. The solvent was then exchanged with chloroform (5 mL)
over three days, replacing the old chloroform with fresh chloroform every
24 h.
J. Lee, O. K. Farha, J. Roberts, K. A. Scheidt, S. T. Nguyen, J. T.
Hupp, Chem. Soc. Rev. 2009, 38, 1450–1459.
A. Schneemann, V. Bon, I. Schwedler, I. Senkovska, S. Kaskel, R.
A. Fischer, Chem. Soc. Rev. 2014, 43, 6062–6096.
A. Clearfield, Dalton Trans. 2016, 45, 4100–4112.
T. Loiseau, C. Volkringer, M. Haouas, F. Taulelle, G. Férey, C. R.
Chim. 2015, 18, 1350–1369.
[6]
[7]
[8]
[9]
G. Férey, C. Serre, Chem. Soc. Rev. 2009, 38, 1380–1399.
S. Henke, D. C. Florian Wieland, M. Meilikhov, M. Paulus, C.
Sternemann, K. Yusenko, R. A. Fischer, CrystEngComm 2011, 13,
6399–6404.
DMOF(Cu)-2,5-NH2Cl: BDC-2,5-NH2Cl (2, 107 mg, 0.5 mmol) was used
instead of BDC-2,3-NH2Cl (1) with the same protocol as was used for the
synthesis of DMOF(Cu)-2,3-NH2Cl.
[10]
[11]
[12]
[13]
[14]
S. Henke, A. Schneemann, A. Wütscher, R. A. Fischer, J. Am.
Chem. Soc. 2012, 134, 9464–9474.
M. Kim, J. A. Boissonnault, P. V. Dau, S. M. Cohen, Angew. Chem.
Int. Ed. 2011, 50, 12193–12196.
X-ray Crystal Structure Analysis: A single crystal of the MOF was coated
with Paratone-N oil, and the diffraction data were measured by
synchrotron radiation ( = 0.69999 Å) on an ADSC Quantum-210
detector at 2D SMC with a silicon (111) double-crystal monochromator
(DCM) at the Pohang Accelerator Laboratory (PAL; Pohang, Korea).
ADSC Q210 ADX software[27] was used for data collection (detector
distance of 63 mm, omega scan;  = 1°, exposure time of 10 s per
frame), and HKL3000sm (Ver. 703r)[28] was used for cell refinement,
reduction, and absorption corrections. The crystal structure was solved
by the direct method using SHLEXT (2014/4)[28] for MOFs and refined by
full-matrix least-squares refinement using the SHELXL (2014/7)
computer program. For all the crystals, the contributions from disordered
solvent molecules were removed by using the SQUEEZE routine
(PLATON),[29] and the outputs from the SQUEEZE calculations are
attached to each CIF file. The positions of all non-H atoms were refined
with anisotropic displacement factors. All H atoms were placed using a
riding model, and their positions were constrained relative to their parent
atoms using the appropriate HFIX command in SHELXL-2014. The X-ray
diffraction data for the MOFs were collected using synchrotron (PAL).
The supplementary crystallographic data for this study are found in the
CCDC-1909253 and CCDC-1919446 for DMOF(Co)-2,5-NH2Cl and
DMOF(Zn)-2,3-NH2Cl, respectively. The data can be obtained free of
charge from www.ccdc.cam.ac.uk/conts/retrieving.html (or from the
CCDC, 12 Union Road, Cambridge CB2 1EZ, UK; fax: +44 1223 336033;
e-mail: deposit@ccdc.cam.ac.uk).
H. Hahm, K. Yoo, H. Ha, M. Kim, Inorg. Chem. 2016, 55, 7576–
7581.
H. Ha, H. Hahm, D. G. Jwa, K. Yoo, M. H. Park, M. Yoon, Y. Kim, M.
Kim, CrystEngComm 2017, 19, 5361–5368.
N. Klein, H. C. Hoffmann, A. Cadiau, J. Getzschmann, M. R. Lohe,
S. Paasch, T. Heydenreich, K. Adil, I. Senkovska, E. Brunner, et al.,
J. Mater. Chem. 2012, 22, 10303–10312.
[15]
[16]
[17]
F. Nouar, T. Devic, H. Chevreau, N. Guillou, E. Gibson, G. Clet, M.
Daturi, A. Vimont, J. M. Grenèche, M. I. Breeze, et al., Chem.
Commun. 2012, 48, 10237–10239.
J. Bergmann, K. Stein, M. Kobalz, M. Handke, M. Lange, J. Möllmer,
F. Heinke, O. Oeckler, R. Gläser, R. Staudt, et al., Microporous
Mesoporous Mater. 2015, 216, 56–63.
K. Kobalz, M. Kobalz, J. Möllmer, U. Junghans, M. Lange, J.
Bergmann, S. Dietrich, M. Wecks, R. Gläser, H. Krautscheid, Inorg.
Chem. 2016, 55, 6938–6948.
[18]
[19]
Y.-C. Ou, Y.-Y. Song, M.-M. Hao, J.-Z. Wu, J.-Z. Wu, Y.-C. Ou, Y.-Y.
Song, H.-M. Du, M.-M. Hao, J.-Z. Wu, Crystals 2017, 7, 311.
A. Schneemann, P. Vervoorts, I. Hante, M. Tu, S. Wannapaiboon, C.
Sternemann, M. Paulus, D. C. F. Wieland, S. Henke, R. A. Fischer,
Chem. Mater. 2018, 30, 1667–1676.
[20]
[21]
[22]
[23]
D. N. Dybtsev, H. Chun, K. Kim, Angew. Chem. Int. Ed. 2004, 43,
5033–5036.
Acknowledgements
H. Jasuja, Y. Jiao, N. C. Burtch, Y. Huang, K. S. Walton, Langmuir
2014, 30, 14300–14307.
This research was supported by the Basic Science Research
Program (2019R1A2C4070584) and the Science Research
Center (2016R1A5A1009405) through the National Research
Foundation of Korea (NRF) funded by the Ministry of Science
and ICT.
D. Sarma, K. V. Ramanujachary, S. E. Lofland, T. Magdaleno, S.
Natarajan, Inorg. Chem. 2009, 48, 11660–11676.
R. Vismara, G. Tuci, N. Mosca, K. V. Domasevitch, C. Di Nicola, C.
Pettinari, G. Giambastiani, S. Galli, A. Rossin, Inorg. Chem. Front.
2019, 6, 533–545.
This article is protected by copyright. All rights reserved.

10.1002/chem.201903210
Chemistry - A European Journal
FULL PAPER
[24]
[25]
S. Chaemchuen, K. Zhou, N. A. Kabir, Y. Chen, X. Ke, G. Van
[26]
A. W. Addison, T. N. Rao, J. Reedijk, J. van Rijn, G. C. Verschoor, J.
Chem. Soc., Dalton Trans. 1984, 0, 1349–1356.
Tendeloo, F. Verpoort, Microporous Mesoporous Mater. 2015, 201,
277–285.
[27]
[28]
Z. Otwinowski, W. Minor, Methods Enzymol. 1997, 276, 307–326.
G. M. Sheldrick, IUCr, Acta Crystallogr. Sect. A Found. Crystallogr.
2008, 64, 112–122.
Y. Shao, Z. Gan, E. Epifanovsky, A. T. B. Gilbert, M. Wormit, J.
Kussmann, A. W. Lange, A. Behn, J. Deng, X. Feng, et al., Mol.
Phys. 2015, 113, 184–215.
[29]
A. L. Spek, Acta Crystallogr. Sect. C Struct. Chem. 2015, 71, 9–18.
This article is protected by copyright. All rights reserved.

10.1002/chem.201903210
Chemistry - A European Journal
FULL PAPER
Entry for the Table of Contents
FULL PAPER
The effect of metal on the degree of
flexibility upon evacuation of paddle-
wheel-type metal-organic frameworks
(MOFs) has been revealed with
positional control of the organic
functionalities. The difference in the
amount of N2 taken by species with
ortho- and para-substituents
Hyeonbin Ha, Youngik Kim, Dopil Kim,
Jihyun Lee, Yoodae Song, Suyeon Kim,
Myung Hwan Park, Youngjo Kim,
Hyungjun Kim*, Minyoung Yoon*, Min
Kim*
Page No. – Page No.
Effect of the metal within
regioisomeric paddle-wheel type
metal-organic frameworks
increased in the order of cobalt,
copper, and zinc.
This article is protected by copyright. All rights reserved.