Exploration of chemically cross-linked metal-organic frameworks

Article abstract of DOI:10.1021/ic500951b

A series of geometrically constrained, cross-linked benzene dicarboxylic acid (bdc) derivatives have been synthesized and incorporated into the canonical isoreticular metal-organic framework (IRMOF) lattice. Only certain cross-links, which allow for the proper relative orientation of the bdc subunits, form the desired IRMOF. Design criteria from these cross-linked ligands allowed for the rational design of two oligomeric ligands composed of three bdc monomers tethered together. These oligomeric ligands were also readily incorporated into an IRMOF lattice with a high degree of crystallinity and porosity, providing a new dimension to rational ligand design for metal-organic frameworks.

Full text of DOI:10.1021/ic500951b

Article
pubs.acs.org/IC
Exploration of Chemically Cross-Linked Metal−Organic Frameworks
C. A. Allen and S. M. Cohen*
Department of Chemistry and Biochemistry, University of California, San Diego, 9500 Gilman Dr. La Jolla, San Diego, California
92093-0358, United States
S
* Supporting Information
ABSTRACT: A series of geometrically constrained, cross-linked
benzene dicarboxylic acid (bdc) derivatives have been synthesized and
incorporated into the canonical isoreticular metal−organic framework
(IRMOF) lattice. Only certain cross-links, which allow for the proper
relative orientation of the bdc subunits, form the desired IRMOF.
Design criteria from these cross-linked ligands allowed for the rational
design of two oligomeric ligands composed of three bdc monomers
tethered together. These oligomeric ligands were also readily
incorporated into an IRMOF lattice with a high degree of crystallinity
and porosity, providing a new dimension to rational ligand design for
metal−organic frameworks.
the framework, but the flexibility of the linkers resulted in a
random orientation of the cross-linking alkyl groups. It was
proposed that at least four distinct cross-link orientations were
possible for some of these alkyl bridges (Figure S1, Supporting
Information).17 Herein we report a substantial extension of
these cross-linked building blocks. The ligands reported herein
are geometrically preorganized to reliably form the prototypical
IRMOF-1 lattice (Figure 1) with a single linker orientation.
More importantly, these cross-linked ligands allowed for the
rational design of oligomeric ligands that also produced the
canonical IRMOF-1 structure. These studies provide a
complementary, bottom-up approach that may lead to the
synthesis of polymer-MOF hybrid materials.
INTRODUCTION
■
Since the discovery of metal−organic frameworks (MOFs)1
many efforts have been made to optimize our ability to design
and control all facets of these materials including creating more
complex architectures.2,3 MOFs (sometimes referred to as
porous coordination polymers, PCPs) are known for being
versatile platforms for designing materials that arrange in
predicted periodic structures upon crystallization. The tuna-
bility of these hybrid materials has been a large factor in their
proposed adaptation for a multitude of applications.4,5 For any
given application, the desired characteristics of a material can be
engineered by careful choice of the initial building blocks6 and
then fine-tuned with other synthetic methods, such as
postsynthetic modification (PSM).7−12 In a particularly striking
example by Sada and co-workers,13 a crystalline MOF was
transformed into a polymer gel by PSM cross-linking, followed
by removal of the metal component. The level of chemical
cross-linking achieved was sufficient that upon removal of the
structural metal ions from the secondary building units (SBUs),
the overall morphology of the crystal was preserved in the
polymer gel monolith. This spectacular study bridges “hard”
crystalline materials with “soft” polymeric materials allowing for
the creation of a new area of materials research that harbors
characteristics of both classes of compounds.14−16
EXPERIMENTAL SECTION
■
General. Starting materials and solvents were purchased and used
without further purification from commercial suppliers (Sigma-Aldrich,
Alfa Aesar, EMD, TCI, Cambridge Isotope Laboratories, Inc., and
others). Chromatography was performed using a CombiFlash Rf 200
automated system from TeledyneISCO (Lincoln, USA). 1H NMR
spectra were collected on a Varian Mercury spectrometer running at
400 MHz. Mass spectrometry (MS) was performed at the Molecular
Mass Spectrometry Facility (MMSF) in the Department of Chemistry
& Biochemistry at the University of California, San Diego. Dimethyl 2-
hydroxyterephthalate (1, Scheme 1) and dimethyl 2,3-dihydroxyter-
ephthalate (7) were synthesized according to literature procedures.18
Compound 2. Compound 1 (1.6 g, 7.7 mmol) was dissolved in
N,N-dimethylformamide (DMF) (15 mL). 1,5-Dibromopentane (0.5
mL, 3.67 mmol) and K2CO3 (2.0 g, 14.7 mmol) were added to the
solution, and the mixture was stirred at 80 °C overnight. After the
mixture was cooled to room temperature, the K2CO3 was removed by
filtration, and excess water was added to precipitate out a beige solid
Recently, we described an alternative, presynthetic approach
to cross-linked MOFs. IRMOFs were prepared from cross-
linked ligands,17 derived from 2-amino-1,4-benzenedicarboxylic
acid (NH2-bdc). These cross-linked ligands were synthesized
by tethering two bdc monomer building blocks with an alkane
linker. By direct solvothermal synthesis, these ligands could
generate isoreticular metal−organic framework (IRMOF)
analogues. Experimental and computational data concluded
that the NH2-bdc tethers were structurally incorporated into
Received: April 25, 2014
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ic500951b | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 1. Schematic representation of cross-linked ligands and resulting IRMOFs (photographs shown). IRMOFs were only formed with L1 and L3,
where the bdc units can arrange in a specific relative orientation with respect to one another.
1
that was isolated by vacuum filtration. Yield: 1.0 g (59%). H NMR
(400 MHz, CDCl3, 35 °C): δ 1.73 (m, 2H; CH2), 1.95 (m, 4H, CH2),
3.89 (s, 6H, CO2CH3), 3.93 (s, 6H, CO2CH3), 4.13 (t, 4H; CH2), 7.60
(d, 4H; ArH), 7.8 (d, 2H; ArH). ESI-MS(+): m/z 489.23 [M + H]+,
506.22 [M + NH4]+, 511.19 [M + Na]+.
Compound 8. Compound 7 (0.52 g, 2.3 mmol)18 was dissolved in
DMF (100 mL). Compound 6 (2.0 g, 5.1 mmol) and K2CO3 (1.3 g,
9.3 mmol) were added, and the mixture was stirred at 80 °C overnight.
After being cooled to room temperature, the K2CO3 was filtered off,
and the solvent removed under a vacuum to yield a beige solid. The
product was purified by trituration with acetone followed by vacuum
filtration to give a white solid. Yield: 1.0 g (44%). 1H NMR (400 MHz,
CDCl3, 35 °C): δ 3.88 (t, 16H; CO2CH3), 5.09 (s, 4H, CH2), 5.14 (s,
4H, CH2), 7.36 (s, 4H; ArH), 7.48 (s, 4H; ArH), 7.58 (s, 2H; ArH),
7.62 (s, 4H; ArH), 7.80 (d, 2H; ArH). ESI-MS(+): m/z 873.17 [M +
NH4]+, 874.18 [M + Na]+, 889.10 [M + K]+.
Compound L5. Compound 8 was dissolved in 1:1 (v:v) THF/4%
KOH(aq) solution (30 mL) and stirred overnight at room
temperature. The aqueous layer was collected and acidified to pH ≈
1 with 1 M HCl. A white precipitate formed that was collected by
vacuum filtration. Yield: 0.4 g (65%). 1H NMR (400 MHz, DMSO-d6,
35 °C): δ 5.05 (s, 4H; CH2), 5.12 (s, 4H, CH2), 7.33 (s, 4H; ArH),
7.43 (d, 2H; ArH), 7.48 (t, 4H; ArH), 7.50 (d, 2H; ArH), 7.62 (s, 2H;
ArH), 7.68 (d, 2H; ArH). ESI-MS(−): m/z 765.21 [M − H]−, 787.10
[M + Na-2H]−, 803.11 [M + K-2H]−, 809.10 [M + 2Na-3H]−.
Compound 9. Diethyl-2,5-dihydroxyterephthalate (0.44 g, 1.7
mmol) was dissolved in DMF (100 mL). Compound 6 (1.5 g, 3.8
mmol) and K2CO3 (0.96 g, 6.9 mmol) were added, and the mixture
was stirred at 80 °C overnight. After the mixture was cooled to room
temperature, the K2CO3 was filtered off, and the solvent removed
under a vacuum to reveal a beige solid. The product was purified by
trituration with acetone followed by vacuum filtration to give a white
solid. Yield: 1.1 g (64%). 1H NMR (400 MHz, CDCl3, 35 °C): δ 1.31
(t, 6H; CH2CH3), 3.9 (d, 12H; CO2CH3), 4.34 (q, 4H; CH2CH3),
5.16 (s, 4H, CH2), 5.24 (s, 4H, CH2), 7.43 (m, 2H; ArH), 7.49 (s, 6H;
ArH), 7.58 (s, 2H; ArH), 7.70 (d, 2H; ArH), 7.83 (s, 2H; ArH), 7.85
(d, 2H; ArH). ESI-MS(+): m/z 901.22 [M + Na]+.
Compound L6. Compound 9 was dissolved in 1:1 (v:v) THF/4%
KOH(aq) solution (30 mL) and stirred overnight at room
temperature. The aqueous layer was collected and acidified to pH ≈
1 with 1 M HCl. A white precipitate formed that was collected by
vacuum filtration. Yield: 1.6 g (86%). 1H NMR (400 MHz, DMSO-d6,
35 °C): δ 5.15 (s, 4H; CH2), 5.24 (s, 4H, CH2), 7.44 (m, 8H; ArH),
7.56 (m, 4H; ArH), 7.69 (m, 4H; ArH). ESI-MS(−): m/z 765.16 [M
− H]−, 787.09 [M + Na-2H]−, 803.09 [M + K-2H]−, 809.17 [M +
2Na-3H]−.
Preparation of IRMOF-1-L1, IRMOF-1-L3. Zn(NO3)2·6H2O (0.342
g, 1.15 mmol) and a cross-linked ligand (L1, L3, 0.2 mmol) were
dissolved in DMF (5 mL) in a scintillation vial (20 mL). The vial was
placed in a sand bath that was placed in a programmable oven. The
temperature was raised from room temperature to 100 °C at 2.5 °C/
min, held for 24 h, and cooled at 2.5 °C/min to room temperature.
Clear blocks (L1) or truncated cubes (L3) were formed (Figure 1).
Once the vial cooled to room temperature, the mother liquor was
decanted, and the crystals were washed with DMF (3 × 10 mL), rinsed
Compound L1. Compound 2 (0.7 g, 1.4 mmol) was dissolved in
1:1 v:v THF/4% KOH(aq) solution (30 mL) and stirred overnight at
room temperature. The aqueous layer was collected and acidified to
pH ≈ 1 with 1 M HCl. A white precipitate formed that was collected
by vacuum filtration. Yield (0.61 g, ∼99%). 1H NMR (400 MHz,
DMSO-d6, 35 °C): δ 1.61 (m, 2H; CH2), 1.77 (m, 4H, CH2), 4.08 (t,
4H; CH2), 7.52 (m, 4H; ArH), 7.65 (d, 2H; ArH). ESI-MS(−): m/z
431.40[M − H]−.
General Synthesis for Compounds L2−L4. Compound 1 (2.0
g, 9.5 mmol) was dissolved in DMF (100 mL). The appropriate
bis(bromomethyl)benzene isomer (1,2-; 1,3-; or 1,4-; 1.2 g, 4.5 mmol)
and K2CO3 (2.5 g, 18 mmol) were added, and the mixture was stirred
at 80 °C overnight. After the mixture was cooled to room temperature,
the K2CO3 was removed by filtration. Excess water was added to
precipitate out a white solid that was isolated by vacuum filtration and
washed with minimal acetone. The resulting intermediate (compounds
3, 4, or 5) was dissolved in 1:1 v:v THF/4% KOH(aq) solution (30
mL) and stirred overnight at room temperature. The aqueous layer
was collected and acidified to pH ≈ 1 with 1 M HCl. A white
precipitate formed that was collected by vacuum filtration. Character-
ization for compounds 4 and L3 is given below; similar data for L2 and
L4 and their associated intermediates can be found in the Supporting
Information.
Compound 4. Yield: 1.65 g (70%). 1H NMR (400 MHz, CDCl3, 35
°C): δ 3.92 (d, 12H; CO2CH3), 5.25 (s, 4H, CH2), 7.47 (m, 3H,
ArH), 7.60 (s, 1H; ArH), 7.66 (s, 2H; ArH), 7.68 (s, 2H; ArH), 7.85
(d, 2H; ArH). ESI-MS(+): m/z 540.12 [M + NH4]+, 545.23 [M +
Na]+.
Compound L3. Yield: 1.0 g (∼99%). 1H NMR (400 MHz, DMSO-
d6, 35 °C): δ 5.26 (s, 4H; CH2), 7.43 (m, 1H; ArH), 7.48 (m, 2H;
ArH), 7.57 (d, 2H; ArH), 7.59 (s, 1H; ArH), 7.70 (m, 4H; ArH). ESI-
MS(−): m/z 465.37 [M − H]−.
Compound 6. Compound 1 (2.9 g, 13.8 mmol) was dissolved in
MeCN (150 mL). 1,3-Bis(bromomethyl)benzene (7.3 g, 27.7 mmol)
and K2CO3 (2.3 g, 16.6 mmol) were added, and the mixture was
stirred at 40 °C overnight. After being cooled to room temperature,
the K2CO3 was removed by filtration, and the solvent removed under a
vacuum to reveal a beige solid. A SiO2 column using 5% ethyl acetate
(EtOac) in hexanes as eluent was used to remove excess dibromide,
and the eluent was changed to 100% CH2Cl2 to liberate the product as
1
a white solid upon removal of solvent. Yield: 4.0 g (74%). H NMR
(400 MHz, CDCl3, 35 °C): δ 3.94 (s, 6H; CO2CH3), 4.52 (s, 2H,
CH2), 5.22 (s, 2H, CH2), 7.36 (m, 2H; ArH), 7.43 (d, 1H; ArH), 7.56
(s, 1H; ArH), 7.66 (m, 2H; ArH), 7.86 (d, 1H; ArH). ESI-MS(+): m/z
392.79 [M + H]+, 409.70 [M + NH4]+.
B
dx.doi.org/10.1021/ic500951b | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
a
Scheme 1. Synthetic Scheme for Selected Compounds
a
Williamson ether chemistry is utilized to crosslink the bdc subunits, followed by hydrolysis to obtain ligands L1, L3, L5, and L6.
with CHCl3 (2 × 10 mL), and left to soak for 3 days with fresh CHCl3
added every 24 h. The crystals were stored in CHCl3 until needed.
Yield: IRMOF-1-L1 186 mg (>100% due to trapped solvent, FW
920.04 g/mol for Zn4OL11.5). IRMOF-1-L3 147 mg (76%, FW 971.07
g/mol for Zn4OL31.5). Under the same reaction conditions, L2 gave
no crystals or precipitate, and L4 gave only an amorphous powder. A
variety of other solvothermal conditions were also explored with L2
and L4 (Table S1, Supporting Information), but no crystalline
products were obtained.
Preparation of IRMOF-1-L5. Zn(NO3)2·6H2O (0.29 g, 0.97 mmol)
and L5 (0.063 g, 0.08 mmol) were dissolved in DMF (5 mL) in a
scintillation vial (20 mL). The vial was placed in a sand bath that was
placed in a programmable oven. The temperature was raised from
room temperature to 100 °C at 0.5 °C/min, held for 24 h and cooled
at 0.5 °C/min to room temperature. The product formed as
transparent yellow blocks. Yield: 22 mg (22%, FW 1038.13 g/mol
for Zn4OL5).
Preparation of IRMOF-1-L6. Zn(NO3)2·6H2O (0.257 g, 0.86
mmol) and L6 (0.063 g, 0.08 mmol) were dissolved in 5 mL of
DMF in a scintillation vial (20 mL). The vial was placed in a sand bath
that was placed in a programmable oven. The temperature was raised
from room temperature to 100 °C at 0.5 °C/min, held for 24 h, and
cooled at 0.5 °C/min to room temperature. The product formed as
clear blocks. Yield: 55 mg (53%, FW 1038.13 g/mol for Zn4OL6).
Digestion and Analysis by 1H NMR. 1H NMR spectra were
recorded on a Varian FT-NMR spectrometer (400 MHz).
Approximately 5 mg of IRMOF was immediately used after BET
analysis and digested with sonication in 500 μL of DMSO-d6 and 100
C
dx.doi.org/10.1021/ic500951b | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
μL of dilute DCl (23 μL of 35% DCl in D2O diluted with 1.0 mL of
1
DMSO-d6) prior to H NMR analysis.
Digestion and Analysis by ESI-MS. ESI-MS was performed using a
ThermoFinnigan LCQ-DECA mass spectrometer, and the data were
analyzed using the Xcalibur software suite. Samples for ESI-MS
analysis were prepared by diluting 10 μL of digested 1H NMR solution
in 1 mL of MeOH.
Thermal Analysis. Approximately 10−15 mg of IRMOF (dried
after gas sorption analysis) was used for TGA measurements. Samples
were analyzed under a stream of dinitrogen (10 mL/min) using a TA
Instrument Q600 SDT running from room temperature to 600 °C
with a scan rate of 5 °C/min.
PXRD Analysis. Approximately 15 mg of IRMOF (soaked in DMF)
was air-dried (∼10 min) before PXRD analysis. Powder X-ray
diffraction (PXRD) data were collected at ambient temperature on a
Bruker D8 Advance diffractometer using a LynxEye detector at 40 kV,
40 mA for Cu Kα (λ = 1.5418 Å), with a scan speed of 1 s/step, a step
size of 0.02 in 2θ, and a 2θ range of 5−40°.
BET Surface Area Analysis. Approximately 35−60 mg of IRMOF
(previously soaking in CHCl3) was evacuated on a vacuum line for ∼1
min at room temperature. The sample was then transferred to a
preweighed sample tube and degassed at 105−150 °C on an
Micromeritics ASAP 2020 Adsorption analyzer for a minimum of 12
h or until the outgas rate was <5 μmHg. The sample tube was
reweighed to obtain a consistent mass for the degassed sample. BET
surface area (m2/g) measurements were collected on three
independent samples, unless residual solvent was still present, of
each MOF at 77 K with dinitrogen on an Micromeritics ASAP 2020
Adsorption analyzer using the volumetric technique.
Figure 2. PXRD of IRMOF-1-L1, -L3, -L5, and -L6 after air-drying
from DMF. All of these MOFs show a high degree of uniformity and
crystallinity.
solvent in this MOF is a characteristic we have observed with
other cross-linked MOFs and may prove to be a means to
modulate guest sorption properties.
In an effort to control the orientation and improve the
organization of the cross-linking group, ligands with more rigid
tethers were designed and prepared. Ortho-, meta-, and para-
xylene dibromides were used to prepare three new ligands (L2,
L3, L4) in good yields (Scheme 1, Schemes S1−S2). Of these,
the m-xylene derivative creates a cross-link that is roughly
identical in length to L1 but with a more rigid structure (Figure
1).22
RESULTS AND DISCUSSION
■
The first target of this study was an alkyl linked HO-bdc (L1),
which was similar to our previously reported systems,17 but
utilizes an ether, rather than an amide linker. This target ligand
is an analogue to the shortest tether previously described17 and
was designed to confirm that small changes in the chemical
composition of the cross-link (i.e., ether vs amide) would be
tolerated during IRMOF formation. Using the well-established
Williamson ether synthesis, gram quantities of ligand L1 were
readily obtained (Scheme 1).19 Consistent with our earlier
findings, L1 produces the desired cross-linked IRMOF as
transparent blocks (IRMOF-1-L1) under standard solvother-
mal procedures for an IRMOF material.20 Extrapolation from a
prior computational analysis suggested that this short tether
must force the bdc monomers into an adjacent, orthogonal
conformation (isomer 1, Figure S1, Supporting Information)
based on length and geometric requirements. Powder X-ray
diffraction (PXRD) of IRMOF-1-L1 matches that of IRMOF-1
(Figure 2), and the connectivity and topology was
unambiguously confirmed by single-crystal X-ray diffraction
(XRD, Table S2). As expected, the alkyl linker was disordered
and could not be located but was determined to be present and
Using standard solvothermal synthetic conditions, only L3
produced the desired MOF (IRMOF-1-L3) as clear truncated
blocks (Figure 1). These crystals displayed the expected PXRD
pattern for an IRMOF (Figure 2). Despite numerous attempts,
under a variety of solvothermal conditions (Table S1), no
crystalline materials were obtained using L2 or L4. Under some
conditions, L2 and L4 produced white powders that were
amorphous as gauged by PXRD (data not shown). To obtain
the desired ligand orientation (isomer 1, Figure S1), the angle
between the linked bdc struts would appear to be too acute
with L2 and too obtuse with L4. Only L3 has the correct
geometry to facilitate formation of the IRMOF lattice.
Furthermore, L2 and L4 are too short to allow for the
formation of other cross-linked isomers (Figure S1).
IRMOF-1-L3 gave a BET surface area of 1619 m2/g after
1
activation at 105 °C under vacuum. The H NMR spectra of
1
intact by H NMR and mass spectral analysis of acid digested
digested samples of IRMOF-1-L3 showed a significant amount
of DMF was trapped in the material under these activation
conditions (Figure S3). However, the DMF could be removed
by heating the material at 150 °C overnight (under a vacuum)
IRMOF-1-L1 (Figure S2). After extensive rinsing with CHCl3
and heating to 150 °C under a vacuum, significant amounts of
DMF remained trapped in IRMOF-1-L1, similar to that
observed with other cross-linked MOFs.17 Thermogravimetric
analysis (TGA) shows a ∼5% weight loss at just under 200 °C,
indicative of trapped solvent (Figure 3), followed by
decomposition of the framework at ∼380 °C, which is typical
for IRMOFs. Activated IRMOF-1-L1 gave a Brunauer−
Emmett−Teller (BET) surface areas of 1711 m2/g and 1450
m2/g when activated at 150 and 105 °C, respectfully, consistent
with greater solvent trapping at lower temperature. In contrast,
typical, unlinked IRMOFs will release all solvent guests under
the aforementioned activation conditions.21 The trapped
giving a BET value of 2121
54 m2/g (Figure 4). TGA
analysis of IRMOF-1-L3 activated at 105 °C shows a 4% weight
loss from 50 to 100 °C but not when activated at 150 °C
indicative of complete activation (Figure 3). Activation at either
temperature results in a steep 10% weight loss at 350 °C
followed by framework degradation at 360 °C (Figure S3). We
attribute the 10% weight loss at 350 °C to degradation of the
m-xylene linker unit. Again, the trapping of solvent in IRMOF-
1-L3 at lower temperatures is suggestive of the potential gating
behavior of these materials.
D
dx.doi.org/10.1021/ic500951b | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 3. TGA traces of IRMOFs following activation at 105 °C (left). The right graph explicitly shows IRMOF-1-L3 after both activation at 105
and 150 °C. The distinct weight loss prior to 200 °C at the lower activation temperature is attributed to trapped DMF molecules. A higher activation
temperature shows the disappearance of this weight loss, corroborating our assignment.
Figure 5. Schematic representation of the design of L5 and L6 from
L3. These oligomeric ligands readily form IRMOF crystals under slow
ramping conditions (photographs shown).
Figure 4. Dinitrogen sorption isotherms of cross-linked IRMOFs
(performed at 77 K). Samples were activated at 105 °C (IRMOF-1-
2,3- and 2,5-substitutional isomers of one another, but both
L6) or 150 °C (all others). The overlap of IRMOF-1-L1 and L6 is
limit the bdc units to the isomer 1 orientation (Figure 3, Figure
attributed to the residual solvent trapped in L1 thus rendering similar
S1). These ligands were synthesized using similar procedures to
L3 as shown in Scheme 1. A Williamson ether synthesis using
an excess of 1,3-dibromoxylene provided an intermediate
competent for a second coupling in good yields (6, Scheme 1).
Coupling of the benzyl bromide 6 with one of two diol
intermediates, followed by hydrolysis, produced L5 and L6 in
reasonable yields (Scheme 1).
Our previous cross-linked ligands have required minimal, if
any, screening efforts to form IRMOFs. While ligands L5 and
L6 do form IRMOF under standard solvolthermal conditions, it
was found that improved crystal quality and uniformity was
achieved by heating with slower temperature ramping (0.5
instead of 2.5 °C/min) and the use of a larger excess of metal
ion. The need for slower ramping is consistent with a more
demanding “annealing” process that would be required with
these oligomeric structures.
sorption properties to the triple bdc linked IRMOF-1-L6.
XRD structure determination of IRMOF-1-L3 (Table S2)
verified the IRMOF topology but also revealed that parts of the
m-xylene cross-linker could be located in the difference map.
Location of substituents by XRD data is very unusual in the
highly symmetric IRMOF system23,24 and demonstrates that
the rigid cross-link is confined to a specific geometry inside the
MOF pores. There is some disorder of the cross-linker, but
electron density was located for the linker in the expected
isomer 1 conformation (Figures S1 and S4), confirming our
computational and design analysis. This suggests that by
designing ligands that restrict the orientation of the bdc
subunits, an increased level of organization and structural
predictability can be engineered into the MOF topology.
Successful incorporation of L3 into the IRMOF structure led
to the expansion and evolution of L3 into ligands L5 and L6.
These “oligomeric” type ligands connect three bdc monomers
via two m-xylene cross-linker units (Figure 5). L5 and L6 are
Upon combining a 1:12 ratio of L5 with Zn(NO3)2·6H2O in
DMF and heating to 100 °C for 24 h, yellow cubes were
obtained that were suitable for single-crystal XRD (Figure 5,
Table S3). IRMOF-1-L5 displayed the expected cubic topology,
E
dx.doi.org/10.1021/ic500951b | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
as demonstrated by both the XRD structure and PXRD
patterns (Figure 2). Unfortunately, the xylene linker could not
be located in the difference map; however, H NMR and mass
ACKNOWLEDGMENTS
■
We thank Dr. Min Kim (Chungbuk National University,
Korea) for helpful discussions, Dr Y. Su (U.C.S.D.) for
performing mass spectrometry experiments, and Dr. C.
Moore, Prof. A. Rheingold (U.C.S.D.), and Phuong V. Dau
for assistance with single-crystal XRD. This work was supported
by a grant from Department of Energy, Office of Basic Energy
Sciences, Division of Materials Sciences and Engineering under
Award No. DE-FG02-08ER46519. C.A.A. was supported by a
National Science Foundation Graduate Research Fellowship
under Award No. DGE1144086.
1
spectral analysis of digested samples confirmed that the ligand
remained intact (Figure S5). Activation of the MOF at 105 °C
(under vacuum) gave a low BET value of 650 m2/g and a TGA
trace indicative of trapped solvent. By heating at 150 °C (under
a vacuum) the BET value was raised to 1654
89 m2/g
(Figure 4). The increased temperature for activation was not
surprising as the xylene units occupy a larger percentage of the
available pore volume.
To explore how the positioning of the xylyl rings affected the
MOF growth and characteristics, L6 was used in the
preparation of a cross-linked IRMOF. Using a 1:10.7 L6 to
zinc ratio, under slow ramping conditions (0.5 °C/min), clear
blocks were obtained after 24 h at 100 °C (Figure 5). XRD
failed to locate the xylene electron density in the difference
map, but again the IRMOF topology was verified by XRD
(Table S3) and PXRD (Figure 2). Interestingly, heating at 105
°C was sufficient for complete activation of IRMOF-1-L6 as
corroborated by the high BET value of 1948 34 m2/g (Figure
4). TGA analysis confirmed no solvent loss at <150 °C
followed by MOF degradation at around 380 °C (Figure 3). 1H
NMR of the digested material also verified the intact ligand and
no residual solvent (Figure S6). Even though the overall
number of xylyl groups incorporated is the same as in IRMOF-
1-L5, the change in orientation of the groups manifests itself in
subtly different physical properties of the material.
REFERENCES
■
(1) Rowsell, J. L. C.; Yaghi, O. M. Microporous Mesoporous Mater.
2004, 73, 3−14.
(2) Tian, D.; Chen, Q.; Li, Y.; Zhang, Y.-H.; Chang, Z.; Bu, X.-H.
Angew. Chem., Int. Ed. 2013, 53, 837−841.
(3) Roy, X.; Hui, J. K. H.; Rabnawaz, M.; Liu, G.; MacLachlan, M. J.
Angew. Chem., Int. Ed. 2011, 50, 1597−1602.
(4) Hurd, J. A.; Vaidhyanathan, R.; Thangadurai, V.; Ratcliffe, C. I.;
Moudrakovski, I. L.; Shimizu, G. K. H. Nat. Chem. 2009, 1, 705−710.
̂
(5) Chen, Z.; Wang, G.; Xu, Z.; Li, H.; Dhotel, A.; Zeng, X. C.; Chen,
B.; Saiter, J.-M.; Tan, L. Adv. Mater. 2013, 25, 6106−6111.
(6) Li, D.-S.; Zhao, J.; Wu, Y.-P.; Liu, B.; Bai, L.; Zou, K.; Du, M.
Inorg. Chem. 2013, 52, 8091−8098.
(7) Mu, Y.-Q.; Zhu, B.-F.; Li, D.-S.; Guo, D.; Zhao, J.; Ma, L.-F. Inorg.
Chem. Commun. 2013, 33, 86−89.
(8) Medishetty, R.; Tandiana, R.; Koh, L. L.; Vittal, J. J. Chem.Eur.
J. 2014, 20, 1231−1236.
(9) Distefano, G.; Suzuki, H.; Tsujimoto, M.; Isoda, S.; Bracco, S.;
Comotti, A.; Sozzani, P.; Uemura, T.; Kitagawa, S. Nat. Chem. 2013, 5,
335−341.
CONCLUSIONS
■
In this study, we have synthesized a variety of new cross-linked
ligands that eludicate design criteria that are required for the
formation of IRMOF under solvothermal conditions. L1 and
L3 were capable of adhering to the restraints of a cubic lattice
because both cross-links, pentane and m-xylene, accommodate
a specific orientation of the two bdc units. However, L2 and L4
that employ o- and p-xylene linkers were not able to attain the
proper geometry, and this was corroborated by an inability to
isolate crystalline materials. The characteristics required for
successful incorporation of tethered bdc molecules were used
to rationally design extended, oligomeric ligands L5 and L6,
each possessing six carboxylic acid units per ligand. Amazingly,
these also both form IRMOFs. This discovery is a first step
toward an advanced level of design and tuning in MOF
materials. Expansion of this concept into other frameworks and
larger oligomeric and polymeric ligands is currently underway.
(10) Holten-Andersen, N.; Harrington, M. J.; Birkedal, H.; Lee, B. P.;
Messersmith, P. B.; Lee, K. Y. C.; Waite, J. H. Proc. Natl. Acad. Sci. U.
S. A. 2011, 108, 2651−2655.
(11) Ma, L.; Wu, C.-D.; Wanderley, M. M.; Lin, W. Angew. Chem.,
Int. Ed. 2010, 49, 8244−8248.
(12) Cohen, S. M. Chem. Rev. 2012, 112, 970−1000.
(13) Ishiwata, T.; Furukawa, Y.; Sugikawa, K. J. Am. Chem. Soc. 2013,
135, 5427−5432.
(14) Schubert, U. Chem. Soc. Rev. 2011, 40, 575−582.
(15) Furukawa, Y.; Ishiwata, T.; Sugikawa, K.; Kokado, K.; Sada, K.
Angew. Chem., Int. Ed. 2012, 51, 10566−10569.
(16) Appel, E. A.; Del Barrio, J.; Loh, X. J.; Scherman, O. A. Chem.
Soc. Rev. 2012, 41, 6195−6214.
(17) Allen, C. A.; Boissonnault, J. A.; Cirera, J.; Gulland, R.; Paesani,
F.; Cohen, S. M. Chem. Commun. 2013, 49, 3200−3202.
(18) Tanabe, K. K.; Allen, C. A.; Cohen, S. M. Angew. Chem., Int. Ed.
2010, 49, 9730−9733.
(19) Huh, S.; Jin, J.; Achard, M. F.; Hardouin, F. Liq. Cryst. 1998, 25,
285−293.
ASSOCIATED CONTENT
(20) Eddaoudi, M.; Kim, J.; Rosi, N.; Vodak, D.; Wachter, J.;
O’Keeffe, M.; Yaghi, O. M. Science 2002, 295, 469−472.
(21) Garibay, S. J.; Wang, Z. Q.; Tanabe, K. K.; Cohen, S. M. Inorg.
Chem. 2009, 48, 7341−7349.
■
S
* Supporting Information
Synthetic and experimental details, characterization of com-
pounds, and crystallographic tables. CIF files have been
deposited to the Cambridge Structural Database under
CCDC 990793-990796. This material is available free of
charge via the Internet at http://pubs.acs.org.
(22) Perry; Kravtsov, V. C.; McManus, G. J.; Zaworotko, M. J. J. Am.
Chem. Soc. 2007, 129, 10076−10077.
(23) Dau, P. V.; Polanco, L. R.; Cohen, S. M. Dalton Trans. 2013, 42,
4013−4018.
(24) Burrows, A. D.; Frost, C. G.; Mahon, M. F.; Richardson, C.
Angew. Chem., Int. Ed. 2008, 47, 8482−8486.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: scohen@ucsd.edu. Telephone: (858) 822-5596. Fax:
(858) 822-5598.
Notes
The authors declare no competing financial interest.
F
dx.doi.org/10.1021/ic500951b | Inorg. Chem. XXXX, XXX, XXX−XXX