Single-crystal-to-single-crystal metalation of a metal-organic framework: A route toward structurally well-defined catalysts

Article abstract of DOI:10.1021/acs.inorgchem.5b00096

Metal-organic frameworks featuring ligands with open chelating groups are versatile platforms for the preparation of a diverse set of heterogeneous catalysts through postsynthetic metalation. The crystalline nature of these materials allows them to be characterized via X-ray diffraction, which provides valuable insight into the structure of the metal sites that facilitate catalysis. A highly porous and thermally robust zirconium-based metal-organic framework, Zr₆O₄(OH)₄(bpydc)₆ (bpydc^2- = 2,2′-bipyridne-5,5′-dicarboxylate), bears open bipyridine sites that readily react with a variety of solution- and gas-phase metal sources to form the corresponding metalated frameworks. Remarkably, Zr₆O₄(OH)₄(bpydc)₆ undergoes a single-crystal-to-single-crystal transformation upon metalation that involves a change in space group from Fm3m to Pa3. This structural transformation leads to an ordering of the metalated linkers within the framework, allowing structural characterization of the resulting metal complexes. Furthermore, Zr₆O₄(OH)₄(bpydc)₆ yields an active heterogeneous catalyst for arene C-H borylation when metalated with [Ir(COD)₂]BF₄ (COD = 1,5-cyclooctadiene). These results highlight the unique potential of metal-organic frameworks as a class of heterogeneous catalysts that allow unparalleled structural characterization and control over their active sites.

Full text of DOI:10.1021/acs.inorgchem.5b00096

Article
pubs.acs.org/IC
Single-Crystal-to-Single-Crystal Metalation of a Metal−Organic
Framework: A Route toward Structurally Well-Defined Catalysts
Miguel I. Gonzalez,† Eric D. Bloch,† Jarad A. Mason,† Simon J. Teat,‡ and Jeffrey R. Long*,†,§
†Department of Chemistry, University of California, Berkeley, California 94720, United States
‡Advanced Light Source, Lawrence Berkeley National Laboratory, Berkeley, California 94720, United States
§Materials Sciences Division, Lawrence Berkeley National Laboratory, Berkeley, California 94720, United States
S
* Supporting Information
ABSTRACT: Metal−organic frameworks featuring ligands
with open chelating groups are versatile platforms for the
preparation of a diverse set of heterogeneous catalysts through
postsynthetic metalation. The crystalline nature of these
materials allows them to be characterized via X-ray diffraction,
which provides valuable insight into the structure of the metal
sites that facilitate catalysis. A highly porous and thermally
robust zirconium-based metal−organic framework,
Zr6O4(OH)4(bpydc)6 (bpydc2− = 2,2′-bipyridne-5,5′-dicarbox-
ylate), bears open bipyridine sites that readily react with a
variety of solution- and gas-phase metal sources to form the
corresponding metalated frameworks. Remarkably,
Zr6O4(OH)4(bpydc)6 undergoes a single-crystal-to-single-
crystal transformation upon metalation that involves a change in space group from Fm3m to Pa3. This structural transformation
̅
̅
leads to an ordering of the metalated linkers within the framework, allowing structural characterization of the resulting metal
complexes. Furthermore, Zr6O4(OH)4(bpydc)6 yields an active heterogeneous catalyst for arene C−H borylation when
metalated with [Ir(COD)2]BF4 (COD = 1,5-cyclooctadiene). These results highlight the unique potential of metal−organic
frameworks as a class of heterogeneous catalysts that allow unparalleled structural characterization and control over their active
sites.
Metal−organic frameworks exhibit the exceptional capability
of adopting crystalline structures with both high porosity4 and
specific pore environments,5,6 resulting in their extensive
evaluation for applications in gas storage and gas separa-
tion.7−10 As readily functionalized porous structures, these
materials are uniquely suited to combine the structural control
available to homogeneous molecular catalysts with the inherent
ease of separation and reuse of heterogeneous catalysts. Similar
to molecular catalysts, metal−organic frameworks can be easily
tailored to give optimal catalytic activity and selectivity.11−13
Moreover, active site isolation within a solid framework can
impart stability to reactive species by preventing intermolecular
decomposition pathways, allowing their characterization and
use for productive chemistry.14−18 Finally, the ability to
determine the structures of these frameworks through X-ray
diffraction enables facile correlation of structure to catalytic
activity.14,19 Altogether, these distinct advantages open up a
promising avenue toward the design of robust catalysts with
isolated and well-defined active sites.
INTRODUCTION
■
The synthesis of catalysts with highly tunable and well-defined
active sites remains an enduring goal in the field of
heterogeneous catalysis, as the development of new catalysts
has been largely limited by the structural ambiguity and
nonuniformity in typical heterogeneous catalysts.1,2 Although
molecular catalysts exhibit tremendous advantages in the ability
to design and characterize their active sites, approximately 80%
of catalytic processes still adopt heterogeneous systems.3 This
strong industrial preference for heterogeneous catalysts arises
from their inherent stability and ease of recovery, which allows
more efficient catalyst separation and reuse. Significant
advances in preparing and characterizing heterogeneous
catalysts with distinct site-isolated active species have been
achieved by grafting molecular catalysts onto solid supports
coupled with characterization via a comprehensive array of
spectroscopic techniques.1,2 Molecular systems, however, still
maintain a considerably higher degree of structural control and
are typically more amenable to characterization than heteroge-
neous materials. In particular, the ability to systematically
change the ligand environment of the catalytically active metal
site and determine molecular structure by X-ray diffraction
remains unrealized in these supported catalysts.
Much of the research on metal−organic framework catalysis
has focused on reactions facilitated by the metal nodes.14,20−23
Received: January 13, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.inorgchem.5b00096
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 1. Structure of bpydc2− linkers (a) and the octahedral Zr6 inorganic nodes (b) in 1. A portion of the crystal structure of 1 (c) as determined
by analysis of single-crystal X-ray diffraction data. The bipyridine nitrogen atoms were found to be disordered over two positions in the crystal
structure, but are represented here in specific positions. Yellow, red, blue, and gray spheres represent Zr, O, N, and C atoms, respectively; H atoms
are omitted for clarity.
In comparison to molecular catalysts, however, achieving
tunability in these systems has been challenging, as only a
few metal and linker combinations lead to frameworks with
metal sites that are accessible for catalysis. This limits the range
of metal sites and coordination environments available to
evaluate as potential catalytically active sites. To overcome
these limitations, alternative strategies have been developed to
allow greater opportunity for catalyst design, such as the
encapsulation of catalytically active metal species24−26 and the
use of functionalized linkers to build catalysts into the
framework.13,17,27−30 Among these methods, postsynthetic
metalation has emerged as a versatile way to produce materials
with isolated and well-defined catalytically active sites.31−37
In analogy to how a single ligand family is used to make a
diverse set of molecular catalysts, we envisioned using a metal−
organic framework with bridging linkers featuring accessible
chelating sites as a platform to synthesize catalysts for a variety
of reactions. Through judicious choice of the metal source used
for postsynthetic metalation, one could potentially prepare a
wide range of catalytically active species based not only on the
identity of the metal cation but also on the coordinated
ancillary ligands.
The synthesis of metal−organic frameworks with open
chelating sites requires a strategy that prevents metalation of
the chelating site during framework synthesis. This has
previously been achieved by using a linker with hard
carboxylate donors and a softer chelating moiety, allowing
selective metal coordination based on their hard/soft acid
properties.19,38 The hard carboxylate groups preferentially bind
to hard oxophilic metal ions to form a porous framework, while
the chelating sites remain available for postsynthetic metalation.
Metal−organic frameworks derived from Zr6O4(OH)4(bpdc)6
(bpdc2− = biphenyl-4,4′-dicarboxylate) or UiO-6739 represent
ideal systems for this strategy due to their inherent thermal and
chemical stability and the strong preference of ZrIV for hard
oxygen donors. We anticipated that Zr6O4(OH)4(bpydc)6
(bpydc2− = and 2,2′-bipyridine-5,5′-dicarboxylate), the bipyr-
idine analog of Zr6O4(OH)4(bpdc)6, could be synthesized and
employed as a platform for building an array of heterogeneous
catalysts through postsynthetic metalation (Figure 1). Indeed,
while this Article was in preparation, recent reports have
demonstrated the synthesis of this compound and its
application in gas separation,40 H2S removal,41 and cataly-
sis.32,33
In addition to being a versatile route toward new
heterogeneous catalysts, the structural determination of the
resulting metalated frameworks can provide insight that is
critical to both understanding the mechanism of catalysis and
precise tuning of catalyst structure to obtain ideal activity and
selectivity. This would demonstrate the ability to characterize
and tailor a heterogeneous catalyst to a degree previously only
possible for molecular systems. In practice, however, it can be
very challenging to determine the structures of the metal−
linker complexes formed from postsynthetic metalation,
because the high symmetry of most metal−organic frameworks
leads to disorder of these sites.18,30,34,42 Recent work, however,
has shown that structurally ordered metal−linker complexes
can be achieved if a metal−organic framework has sufficiently
low symmetry.19
Herein, we report the synthesis of Zr6O4(OH)4(bpydc)6 (1)
and its subsequent metalation with solution- and gas-phase
metal precursors. We demonstrate the unprecedented single-
crystal-to-single-crystal metalation of 1 and characterization of
the resulting metalated frameworks, 1·(CuCl2)5.8, 1·(CuCl)6.8,
1·(CoCl2)5.5, 1·(FeBr2)6.1, and 1·(Cr(CO)4)5.6, by single-crystal
X-ray diffraction. Remarkably, metalation of the framework
leads to a structural transition to a lower symmetry space group,
allowing the structural characterization of the metal-bipyridine
complexes within the pores. Furthermore, 1 shows catalytic
activity in arene C−H borylation upon metalation with
[Ir(COD)2]BF4. These results highlight the unique potential
of metal−organic frameworks as heterogeneous catalysts with
well-defined site-isolated active sites that can be precisely
characterized structurally.
EXPERIMENTAL SECTION
■
Materials and Methods. All manipulations were performed under
an N2 atmosphere in a VAC Atmospheres glovebox or using standard
Schlenk techniques. Acetonitrile, benzene, toluene, hexanes, diethyl
ether, and 1,2-difluorobenzene were deoxygenated by purging with
argon for 1 h and dried using a commercial solvent purification system
designed by JC Meyer Solvent Systems. The compounds 1,3-
dimethylbenzene and 1,3-dimethoxybenezne were purchased from
Sigma-Aldrich, dried over Na/benzophenone, and degassed via three
B
DOI: 10.1021/acs.inorgchem.5b00096
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
successive freeze−pump−thaw cycles. Bis(pinacolato)diboron was
purchased from Strem and purified by recrystallization from hexanes
followed by sublimation at 80 °C and 60 mTorr. The compounds 2,2′-
bipyridine-5,5′-dicarboxylic acid (H2bpydc) and dimethyl [2,2′-
bipyridine]-5,5′-dicarboxylate (dmbpydc) were synthesized using a
previously published procedure.29 The compounds ZrCl4, 5,5′-
dimethyl-2,2′-bipyridine, 1,3,5-trimethoxybenzene, triphenyl-phos-
phine (PPh3), and tricyclohexylphosphine (PCy3) were purchased
from Sigma-Aldrich and used as received. The compounds [Ir-
(COD)2]BF4, FeBr2, CoCl2, CuCl, and CuCl2 were purchased from
Strem and used as received. All other chemicals were purchased from
commercial vendors and used as received unless otherwise noted.
Thermogravimetric analyses were carried out with a TA Instru-
ments TGA Q5000 operating at a ramp rate of 1 °C/min under a 25
mL/min N2 flow. Air-sensitive samples were prepared as a slurry in
hexanes to temporarily protect the sample from oxygen and moisture
and quickly loaded into the instrument. Samples were then heated to
100 °C at a rate of 1 °C/min and held at that temperature for 1 h to
evaporate the hexanes prior to analysis. Infrared spectra were collected
on a PerkinElmer Avatar Spectrum 400 FTIR spectrophotometer
equipped with a Pike attenuated total reflectance accessory (ATR)
accessory. UV−vis spectra were recorded on a Cary 5000
spectrophotometer equipped with a reflectance sphere for diffuse
reflectance spectra. Scanning electron microscopy (SEM) samples
were prepared by dispersing crystals in dichloromethane and drop
casting onto a silicon chip. In order to dissipate charge, the samples
were sputter coated with ∼3 nm of Au (Denton Vacuum, LLC).
Crystals were imaged at 5 keV/12 μA by field emission SEM (JEOL
FSM6430). Elemental analyses were obtained from the Microanalytical
Laboratory of the University of California, Berkeley, using a
PerkinElmer 2400 series II combustion analyzer. NMR spectra were
acquired on Bruker AVB-400 and AVQ-400 instruments at the
University of California, Berkeley, NMR facility. All chemical shifts are
given in relation to residual solvent peaks or tetramethylsilane. NMR
yields were determined by 1H NMR spectroscopy with 1,3,5-
trimethoxybenzene as an internal standard.
Characterization of the crystals was performed using single-crystal X-
ray diffraction.
General Procedure for Metalation of 1 with Solution-Phase
Metal Precursors. Microcrystalline 1 (50−500 mg), the desired
metal source (0.10−1.0 equiv per bpydc), and acetonitrile (3 mL)
were mixed in a 20 mL Teflon-capped vial. The resulting mixture was
then heated for a week on a hot plate at 80 °C to afford a colored
powder. After cooling to room temperature, the solvent was decanted,
and the powder was soaked three times in fresh acetonitrile for 24 h at
80 °C in order to remove any unreacted metal source. The solvent was
then removed under reduced pressure at 80 °C to give the
corresponding metalated framework. Metal insertion reduces the
stability of the framework to moisture. Thus, metalation and
subsequent manipulations were carried out under an inert atomo-
sphere.
Single crystals of 1 in THF were placed in a 4 mL Teflon-capped
vial. Most of the solvent was pipetted out, followed by addition of
excess (10−20 mg) metal source in acetonitrile. The mixture was
allowed to react for 7 days at 80 °C, resulting in a color change of the
crystals. Most of the solution was removed by pipet, and the crystals
were subsequently soaked three times in fresh acetonitrile at 80 °C for
24 h, and were then used for single-crystal X-ray diffraction
experiments.
Metalation of 1 with Cr(CO)6. Microcrystalline powder of
Zr6O4(OH)4(bpydc)6 (10−50 mg) in a 4 mL vial was placed in a
20 mL vial containing excess Cr(CO)6. The 20 mL vial was sealed with
a Teflon cap and heated for 7 days at 80 °C. The resulting dark green
solid was washed three times with benzene at 80 °C to remove any
unreacted Cr(CO)6, and was then heated at 80 °C under reduced
pressure for 24 h to give 1·(Cr(CO)4)5.6.
Single crystals of Zr6O4(OH)4(bpydc)6 in THF or benzene were
placed in a 4 mL Teflon-capped vial. Most of the solvent was pipetted
out, and the 4 mL vial was placed within a 20 mL vial containing
excess Cr(CO)6. The 20 mL vial was sealed with a Teflon cap and
heated for 7 days at 80 °C. The resulting dark green crystals were
either used directly or soaked in acetonitrile then used for single-
crystal X-ray diffraction experiments.
Zr6O4(OH)4(bpydc)6 (1). H2bpydc (3.09 g, 12.5 mmol) and
benzoic acid (125 g, 1000 mmol), and N,N-dimethylformamide
(DMF; 1 L) from a newly opened bottle, were placed into a three-neck
2-L round-bottom flask equipped with a Schlenk adapter, glass
stoppers, and a magnetic stir bar, and the resulting mixture was purged
with dry nitrogen for 30 min. Solid ZrCl4 (2.96 g, 12.5 mmol) was
then added, and the mixture was purged with dry nitrogen for an
additional 30 min. Deionized water (410 μL, 22.8 mmol) was then
added, and the mixture was heated with magnetic stirring for 5 days at
120 °C under a nitrogen atmosphere. After allowing the mixture to
cool to room temperature, the solvent was decanted, and the resulting
white microcrystalline powder was washed by soaking three times in
500 mL aliquots fresh DMF for 24 h at 120 °C, followed by solvent
exchange with tetrahydrofuran (THF) via Soxhlet extraction for 3
days. The THF solvated powder was filtered under dry N2, followed by
heating at 120 °C under dynamic vacuum for 24 h to give fully
desolvated Zr6O4(OH)4(bpydc)6. Yield: 3.70 g, 82%. Anal. Calcd for
C72H36N12O32Zr6: C, 40.55; H, 1.98; N, 7.88. Found: C, 41.72; H,
1.98; N, 7.52. IR (solid-ATR): 3205 (br, w), 1653 (m), 1591 (s), 1536
(m), 1403 (s), 1167 (w), 1014 (w), 839 (2), 802 (w), 770 (s), 720
(w), 653 (s), 455 (s).
Single Crystals of Zr6O4(OH)4(bpydc)6 (1). H2bpydc (247 mg,
1.00 mmol), benzoic acid (9.89 g, 80.0 mmol), and anhydrous DMF
(80 mL) were placed into a 100 mL Teflon-capped jar and sonicated
for 1 min. Solid ZrCl4 (466 mg, 2.00 mmol) was then added, and the
mixture was sonicated for 1 min. Deionized water (128 μL) was added,
and the mixture was placed in an oven that was preheated to 120 °C,
and the vessel was kept at that temperature for 5 days, yielding
colorless octahedron-shaped single crystals on the walls of the jar. The
crystals were soaked three times in fresh DMF for 24 h at 120 °C,
followed by soaking three times in dry deoxygenated THF for 24 h at
70 °C in a Schlenk flask under N2 atmosphere. The crystals were then
kept in THF and stored in a glovebox under N2 atmosphere.
Zr6O4(OH)4(bpdc)6 (2). This material was synthesized by
substituting H2bpydc with H2bpdc in the synthetic procedure for 1
above. Anal. Calcd for C84H52O32Zr6: C, 47.58; H, 2.47; N, 0. Found:
C, 47.85; H, 2.11; N, 0.
[Ir(dmbpydc)(COD)]BF4 (3). In a 20 mL vial, dmbpydc (22.0 mg,
0.0808 mmol) was added to a solution of [Ir(COD)2]BF4 (40.2 mg,
0.0811 mmol) in acetonitrile (3 mL). The initially yellow solution
immediately turned dark green upon addition of dmbpydc. The vial
was sealed with a Teflon cap and allowed to stir at room temperature
for 12 h. The resulting dark green solution was then concentrated
under reduced pressure. Layering or slow diffusion of diethyl ether
over the concentrated solution afforded dark green crystals, which
were suitable for single-crystal X-ray analysis. After carefully decanting
the solvent, the crystals were washed with diethyl ether and dried
under reduced pressure to give [Ir(dmbpydc)(COD)]BF4. Yield: 38.2
mg, 72%. Anal. Calcd for C22H24BF4IrN2O4: C, 40.07; H, 3.67; N,
4.25. Found: C, 39.62; H, 3.40; N, 4.72. 1H NMR (400 MHz,
CD3CN) δ 8.778 (dd, 2H), 8.703 (d, 2H), 8.537 (d, 2H), 4.406 (m,
4H), 3.971 (s, 6H), 2.405 (m, 4H), 2.035 (m, 4H). IR (solid-ATR):
3121 (w), 3067 (w), 3000 (w), 2958 (w), 2894 (w), 2841 (w), 1721
(m), 1611 (2), 1427 (w), 1296 (m), 1276 (m), 1191 (w), 1128 (m),
1027 (br, m), 867 (w), 760 (m), 523 (w), 484 (w).
RESULTS AND DISCUSSION
■
Synthesis of Zr6O4(OH)4(bpydc)6. Reaction of ZrCl4 with
H2bpydc in the presence of 80 equiv of benzoic acid and 1.8
equiv of water in DMF at 120 °C for 5 days affords 1 as a white
microcrystalline solid. Powder X-ray diffraction data show that
1 is isostructural to Zr6O4(OH)4(bpdc)6 (Supporting Informa-
tion Figure S8). Thermogravimetric analysis data collected for
the as-synthesized material (Supporting Information Figure
S32) shows a 41% mass loss from 40 to 120 °C, which is
C
DOI: 10.1021/acs.inorgchem.5b00096
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 1. Selected Properties of the Metal−Organic Frameworks Presented in this Work
a
b
c
compd
color
white
metal loading (%)
SALang (m2/g)
SABET (m2/g)
1
2
2772
2805
1253
835
2730
2625
1101
701
white
1·(CuCl2)5.8
1·(CuCl)6.8
1·(CoCl2)5.5
1·(FeBr2)6.1
1·(Cr(CO)4)5.6
green
96(1)
light brown
pale blue
red-violet
dark green
114(3)
92(1)
1282
1073
1065
1204
971
101.1(7)
93(2)
934
a
b
Metal loading was determined by ICP-OES based on the molar ratio of the inserted metal relative to Zr in the framework. SALang = Langmuir
c
surface area. SABET = Brunauer−Emmett−Teller (BET) surface area.
Figure 2. Portion of the crystal structure of 1·(CuCl2)5.8 (a) as determined by analysis of single-crystal X-ray diffraction data. Structures of the
bipyridine metal complexes in 1·(CuCl2)5.8 (b), 1·(CuCl)6.8 (c), 1·(CoCl2)5.5 (d), 1·(FeBr2)6.1 (e), and 1·(Cr(CO)4)5.6 (f), as determined from
single-crystal X-ray diffraction data. Yellow, green, brown, purple, light orange, dark green, light green, dark red, red, blue, and gray spheres represent
ZrIV, CuII, CuI, CoII, FeII, Cr0, Cl, Br, O, N, and C atoms, respectively; H atoms and acetonitrile molecules in the pores are omitted for clarity. Note
that 53% of the Cu-bipyridine complexes in 1·(CuCl)6.8 were found to have [CuCl2]− as a counteranion instead of Cl− (Supporting Information
Figure S1), while the bipyridine complexes in 1·(CoCl2)5.5 and 1·(FeBr2)6.1 exist in different coordination geometries due to different degrees of
solvation (Supporting Information Figures S2−S3). The coordinated solvent molecules in 1·(CoCl2)5.5 and 1·(FeBr2)6.1 could not be modeled due
to disorder and weak scattering of the solvent molecules in comparison to the halide ligands.
associated with framework desolvation. No further mass loss is
observed until the framework decomposes at temperatures
above 450 °C, which suggests that 1 has comparable thermal
stability to Zr6O4(OH)4(bpdc)6.39,43 The fully desolvated
material exhibits Langmuir and BET surface areas of 2772
and 2730 m2/g, respectively, which are close to 3000 m2/g, the
reported Langmuir surface area of Zr6O4(OH)4(bpdc)6.39 SEM
images of the bulk powder sample of 1 (Supporting
Information Figure S39) reveal that the material forms well-
faceted octahedron-shaped crystals roughly 0.5−1 μm on an
edge.
Benzoic acid is critical to the synthesis of highly crystalline 1.
Attempts to synthesize it in the absence of benzoic acid result
in powders with poor crystallinity. Powder X-ray diffraction
analysis reveals that increasing the amount of benzoic acid (0−
80 equiv per linker) leads to more crystalline material
(Supporting Information Figure S10), which is in agreement
with results reported in a similar study for Zr6O4-
(OH)4(bpdc)6.43 This effect is thought to arise from
competition between benzoic acid and the dicarboxylic acid
linkers during framework assembly, making framework
formation more reversible.43 In addition, having more benzoic
acid present in the reaction mixture also requires longer
reaction times. As an example, precipitate formation can be
observed within a few hours if no benzoic acid is added, while it
takes 2−3 days for the solid product to be observed in the
presence of 80 equiv of benzoic acid per linker. Attempts to
synthesize the material with the addition of 100 or greater
equivalents of benzoic acid per linker yielded no observable
precipitate over 5 days.
The synthesis of X-ray quality single crystals of 1 required
more careful control of the reaction conditions. In particular,
both the amount of water added and the metal-to-ligand ratio
had significant effects on the size and quality of the crystals.
Though highly crystalline and porous 1 can be reproducibly
synthesized using commercially available DMF, reproducible
formation of single crystals requires the addition of a precise
amount of water to anhydrous DMF obtained from a solvent
purification system. Reaction conditions that use anhydrous
DMF alone fail to yield any precipitate, which is consistent with
the requirement of water to form the μ3-O and μ3-OH groups
of the Zr6 octahedral nodes.43 Interestingly, the addition of 2
equiv of ZrCl4 per linker yields isolated and regularly shaped
octahedral crystals. In contrast, conditions that employ
stoichiometric amounts of ZrCl4 per linker produce crystals
that are intergrown.
X-ray analysis of an as-synthesized single crystal of 1 further
confirmed that the framework is isostructural to Zr6O4-
(OH)4(bpdc)6 (Figure 1). The calculated powder diffraction
pattern obtained from the crystal structure agrees well with the
D
DOI: 10.1021/acs.inorgchem.5b00096
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 3. Comparison of the crystal structures of 1 (a) and 1·(CuCl2)5.8 (b) viewed along the [100] direction; unit cell edges are shown as black
lines. The bare framework exhibits mirror symmetry along the body diagonals of its unit cell and face-centering translations that relate the zirconium
clusters at the ends of each linker. In the metalated structure, the mirror symmetry is removed by the ordering of the linkers, while the face-centering
is lost because the clusters at the ends of each linker are tilted in opposite directions.
experimental powder pattern, indicating that the crystal
structure is representative of the bulk material (Supporting
Information Figure S9). As expected, the bipyridine units show
crystal quality after metalation. Analysis of the crystals by
single-crystal X-ray diffraction revealed a lowering of the
framework symmetry from Fm3m to Pa3. Remarkably, the
̅
̅
disorder over two positions due to the Fm3m symmetry of the
metalated bipyridine linkers were crystallographically ordered,
allowing structural determination of the metal complexes
formed within the framework (Figure 2).
̅
structure. While this Article was in preparation, similar results
were reported from X-ray analysis of crystals of 1 grown under
slightly different conditions.40 Refinement of the linker
occupancy in the structure revealed that 86% of the ligand
was present, corresponding to the absence of roughly 1/6 of
the linkers. When the ligand is not present, the vacancies on the
cluster have been suggested to be occupied with water or
hydroxide,44−46 which can be identified as weak electron
density peaks 2.18(4) Å away from Zr.
Closer inspection of the structures shows that each
bipyridine unit forms an arch to chelate the metal, presumably
facilitating better orbital overlap with the metal (Figure 3). The
zirconium clusters at the ends of each linker rotate slightly to
accommodate the arching, while linkers around each cluster
orient accordingly to conform to the direction of the cluster
rotation. These distortions from the unmetalated structure
collectively result in an ordering of the linker−metal complexes
Metalation of Zr6O4(OH)4(bpydc)6. Transition metal
halides were first considered for the metalation of 1, as these
were expected to form solvated metal species that would easily
fit through the triangular windows of the framework, which
have an incircle of ∼8 Å in diameter.39 Soaking microcrystalline
powders of 1 in acetonitrile solutions of metal halide salts
(typically 1 equiv per bipyridine linker) at 80 °C for 5 days
results in a color change (Table 1) of the powders over the
course of the reaction. The powders remain colored after three
acetonitrile washes at 80 °C, suggesting successful metalation of
the framework. The powder X-ray diffraction patterns of the
metalated frameworks 1·(CuCl2)5.8, 1·(CuCl)6.8, 1·(CoCl2)5.5,
and 1·(FeBr2)6.1 show additional peaks compared to that of 1,
indicating a possible change in space group upon metalation.
Metal to zirconium ratios determined by ICP-OES analysis
(Table 1) indicate a close to stoichiometric reaction of the
bipyridine sites. Thermogravimetric analyses of the resulting
powders generally showed slight decreases in thermal stability
of the framework upon metalation (Supporting Information
Figures S33−S38). This is likely due to weakened carboxylate−
Zr bonds that result from less electron density being available
to the linker carboxylate groups, as well as to strain induced by
arching of the linker upon coordination of the linker bipyrdine
moiety to a metal center, as discussed below.
as they go from a site that has three mirror planes in Fm3m to a
̅
site with no symmetry in Pa3.
̅
Although there are several examples of the structural
characterization of postsynthetically metalated metal−organic
frameworks, most structures fail to resolve the complete ligand
environment around the metal centers due to extensive
disorder of the metalated linkers.30,34,42 Recent work has
emphasized that full characterization of metal-linkers complexes
formed via postsynthetic metalation requires low symmetry at
the location of the open chelating site in the crystal, which can
be achieved by starting with a metal−organic framework with
low crystal symmetry.19 The crystallographic ordering of the
linkers in 1 after postsynthetic metalation demonstrates that the
structural characterization of the metalated linkers in high-
symmetry metal−organic frameworks can be achieved if the
framework has a pathway to lower symmetry during metalation.
To the best of our knowledge, this is the first reported example
of this phenomenon in a metal−organic framework.
X-ray analysis of 1·(CuCl2)5.8 revealed significant ordering of
both the framework and the acetonitrile molecules within the
pores, which enabled reliable measurement of interatomic
distances and angles for the (bpy)CuCl2 units (Figure 2b).
Interestingly, the CuII centers do not have acetonitrile bound,
despite the presence of additional acetonitrile in the pores. The
Cu−N distances of 1.993(4) and 2.007(4) Å and Cu−Cl
distances of 2.2101(15) and 2.2126(16) Å in the complex were
found to be very close to reported bond distances for the
Metal insertion in single crystals of 1 was carried out under
the same conditions used for the bulk microcrystalline samples,
with the exception of using excess metal precursor. Similar to
the powders, the crystals change color with no visible loss in
E
DOI: 10.1021/acs.inorgchem.5b00096
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
analogous molecular complex, Cu(tbbpy)Cl2 (tbbpy = 4,4′-
bis(tert-butyl)-2,2′-bipyridine).47 The CuII centers exhibit a
distorted square planar geometry, with the Cl−Cu−Cl plane
tilted 33° away from the N−Cu−N plane, which is likely a
result of steric repulsion between the Cl− ligands and
neighboring bipyridine H atoms. This distortion from planarity
is common in molecular (bpy)CuCl2 complexes and has been
shown to vary with crystal packing.47 While it was not possible
to collect a single-crystal structure of activated 1·(CuCl2)5.8, a
structure was obtained from high-resolution powder X-ray
diffraction data (Supporting Information Figure S16). In the
activated structure, the CuII centers are in a similar coordination
geometry, with the Cl−Cu−Cl plane tilted 45° away from the
N−Cu−N plane.
Structural determination of the bipyridine metal complexes
was more challenging for 1·(CuCl)6.8, 1·(CoCl2)5.5, and 1·
(FeBr2)6.1, due to the disorder attributed to rotation of the
linkers and variation in the degree of solvation of the metal
complexes. The linkers in 1·(CuCl)6.8 show rotational disorder
over two positions tilted 27° away from each other. Although
the CuI complexes in 1·(CuCl)6.8 were modeled as distorted
trigonal planar complexes with Cl− as a coordinated counter-
anion (Figure 2c), residual electon density close to the CuI
center suggests that 53% of the complexes actually have linear
[CuCl2]− units as a bound counteranion (Supporting
Information Figure S1). Molecular complexes that are
analogous to the (bpy)CuCl structural model exist.48,49 The
closest analogs to the (bpy)Cu(CuCl2) structural model
contain phenanthroline- or bipyridine-CuI units bridged by
[CuCl2]− anions,50,51 which cannot form in the framework due
to the isolation of the bipyridine sites.
Similar to the structure of 1·(CuCl)6.8, the linkers in 1·
(CoCl2)5.5 displayed rotational disorder over two positions,
tilted 14° away from each other. The bipyridine−CoCl2
adducts were clearly located and appear to be square planar
in one of the disordered positions (Figure 2d) and distorted
square planar in the other (Supporting Information Figure S2).
Residual electron density close to the CoII centers, however,
indicates that solvent is likely bound to CoII and that the
complexes are actually in octahedral and trigonal bipyramidal
geometries (Supporting Information Figure S2), which are
more consistent with published structures.52,53 The square
planar sites can be assigned to be octahedral with the bipyridine
and Cl− ligands coplanar and solvent on the axial sites, while
the distorted square planar sites can be assigned as partially
trigonal bipyramidal with the N and Cl on the axial sites and N,
Cl, and solvent on the equatorial sites. Unfortunately, the
disorder and weak scattering from the bound solvent molecules
prevent a more accurate assignment of the CoII complex
geometries.
X-ray diffraction data for 1·(FeBr2)6.1 collected on single
crystals that were washed with acetonitrile alone gave structures
with poorly resolved FeBr2 moieties. The disorder in the
structures can be attributed to the FeII complexes being in
multiple states of solvation. Consequently, crystals were further
washed with benzene at 80 °C in an attempt to fully desolvate
the FeII centers. X-ray analysis of one of these crystals gives a
much more ordered structure, with the FeII centers
predominantly in a pseudotetrahedral geometry (Figure 2e),
which is reasonable for FeBr2 complexes with nitrogen-based
chelating ligands.54 Additional electron density peaks found
near to the FeII centers suggests that a fraction of the complexes
have the Br− ligands closer to the Fe-bipyridine plane. Similar
to 1·(CoCl2)5.5, solvent may still be bound to on a fraction of
the FeII sites, forcing them into what appears to be a
pseodooctahedral geometry with the bipyridine and Br− ligands
on the equatorial sites and solvent on the axial sites
(Supporting Information Figure S3). The geometry of these
FeII sites, however, cannot be accurately assigned due to the
disorder and the weak scattering from the coordinated solvent
molecules on the FeII sites in comparison to the Br− ligands.
Given the permanent porosity of 1, we explored the
possibility of using framework to chelate metals from reagents
introduced in the gas phase. Heating a microcrystalline sample
of 1 in a sealed vial with excess Cr(CO)6 at 80 °C over 7 days
results in gradual color change of the framework from white to
dark green over the course of the reaction, consistent with the
metalation of the bipyridine sites. Analysis of the resulting
framework by infrared spectroscopy shows CO stretches at
2012, 1897, 1870, and 1845 cm−1 (Supporting Information
Figure S43), consistent with the formation of (bpy)Cr(CO)4
complexes within the framework.55 Remarkably, single crystals
of 1 can also be metalated and analyzed by X-ray diffraction,
leading to structural confirmation that the complexes formed
are indeed (bpy)Cr(CO)4 (Figure 2f). These results emphasize
the unique ability to perform and characterize gas phase
reactions in metal−organic frameworks due to their inherent
porosity and crystallinity.
The structures obtained for single crystals of 1·(Cr(CO)4)6
from data collected directly after metalation show rotational
disorder of the (bpy)Cr(CO)4 units within the framework.
Interestingly, soaking crystals of 1·(Cr(CO)4)6 in acetonitrile
results in ordering of the linkers, conceivably due to restriction
of the rotational freedom of the linkers as the pores are filled
with solvent. The Cr−N distances of 2.102(4) and 2.105(4) Å,
equatorial Cr−C distances of 1.842(7) and 1.842(7) Å, and
equatorial C−O distances of 1.173(8) and 1.167(9) Å are all in
good agreement with analogous complexes in the literature.56
In contrast, the axial Cr−C distances of 1.77(1) and 1.78(2) Å
and C−O distances of 1.29(2) and 1.24(2) Å are comparatively
shorter, which may be an artifact of slight disorder of the
(bpy)Cr(CO)4 complexes along the direction of these bonds.
Gas Adsorption. Low-pressure N2 adsorption data
(Supporting Information Figure S17) collected at 77 K for
metalated frameworks reveal a considerable decrease in surface
area upon metalation of 1 (Table 1). These results are
consistent with loss of accessible pore surface due to the metal-
linker complexes occupying additional space in the pores. In
line with this, frameworks with bulkier metal complexes exhibit
lower surface areas, with the exception of 1·(CuCl)6.8. The low
surface area of 1·(CuCl)6.8 relative to the other metal halide
metalated frameworks may be attributed to some of the Cu
sites in 1·(CuCl)6.8 possessing the much larger [CuCl2]− as a
counteranion instead of the Cl−.
To determine the accessibility of the metal sites in 1·
(CuCl2)5.8, 1·(CuCl)6.8, 1·(CoCl2)5.5, 1·(FeBr2)6.1, and 1·
(Cr(CO)4)5.6, low-pressure H2 adsorption measurements
were performed at 77 K (Figure 4). All metalated frameworks
except 1·(CuCl)6.8 displayed improved gravimetric H2 uptake
over the bare framework at low pressures, despite their lower
surface areas and increased formula weights. The isotherms,
however, did not show any steep uptake at very low pressures
characteristic of strong interaction with open metal sites.57
Instead, the increased H2 capacity likely arises from the
introduction of additional weakly polarizing sites and the
formation of tighter binding pockets within the pores of the
F
DOI: 10.1021/acs.inorgchem.5b00096
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
molecular CuI−CO complexes (Supporting Information Figure
S42).58 The CO stretching frequency in the CO adduct of 1·
(CuCl)6.8 is shifted 50 cm−1 lower relative to free CO,
indicating modest π backbonding from CuI to CO.
Substoichiometric uptake of CO per CuI suggests that CO
binds to only one type of CuI center among the three
determined in the single-crystal structure (Supporting
Information Figure S1). In contrast, the lack of any strong
interaction between CO and 1·(CuCl2)5.8, 1·(CoCl2)5.5, or 1·
(FeBr2)6.1 suggests that the metal centers in these materials
have insufficient π backbonding capability to form stable
adducts with CO, which is attributed to the higher formal
oxidation state of the metal centers in the bipyridine-MII
complexes.
Arene C−H Borylation. Iridium-catalyzed C−H borylation
has proven to be a practical and efficient way of functionalizing
inert feedstock chemicals to make valuable products that are
widely used in fine chemicals synthesis.59 Among the many
catalysts studied for this reaction, iridium complexes supported
by chelating N-donor ligands have been shown to be very
capable systems for the C−H borylation of arenes with either
4,4,4′,4′,5,5,5′,5′-octamethyl-2,2′-bi-1,3,2-dioxaborolane
(B2pin2) or 4,4,5,5-tetramethyl-1,3,2-dioxaborolane
(HBpin).60−63 Inspired by the considerable amount of work
on homogeneous systems, recent efforts have been directed
toward making heterogeneous analogs of these catalysts.33,64,65
Given that IrI precursors in combination with 2,2′-bipyrdine
ligands lead to highly active homogeneous systems, we
investigated whether 1 would show similar activity when
metalated with iridium.
Considering the size of the pore apertures in 1, [Ir(COD)2]-
BF4 (COD = 1,5-cyclooctadiene) was selected as an metalation
agent, as it forms a relatively compact complex, [Ir(COD)-
(MeCN)2]+, in acetonitrile.66 Metalation of a microcrystalline
powder of 1 with [Ir(COD)2]BF4 following the procedure used
for the metal halides indeed yields 1·(Ir(COD)BF4)4.7 as an
olive green powder. ICP-OES analysis of the powder indicates
78% Ir loading of the framework, which is much lower than the
loading determined for the frameworks metalated with first-row
transition metal halides. X-ray analysis of the metalated single
crystals revealed no change in space group and less than 10%
occupancy of Ir at the bipyridine sites. Consequently, the
ancillary ligands on Ir could not be identified. Disparity
between the Ir loading in the powder and single-crystal samples
implies that the extent of metalation at the framework surface is
greater than that of the interior. Compared to ∼60 μm single
crystals, the ∼0.5−1 μm crystals in the powder sample have a
much larger crystal surface area to volume ratio, which would
result in higher Ir loading in the powder. The Ir source may be
blocking the pore windows upon metalation, obstructing the
diffusion of additional Ir complexes to sites within the
framework. This may be due to the larger size of the Ir
precursor compared to the first row transition metal halides. In
addition, Ir forms much more stable complexes with bipyrdine,
making it less likely to reversibly dissociate from the bipyridine
sites and travel deeper into the pore structure.
Figure 4. Low-pressure H2 adsorption isotherms for 1, 1·(CuCl2)5.8, 1·
(CuCl)6.8, 1·(CoCl2)5.5, 1·(FeBr2)6.1, and 1·(Cr(CO)4)5.6 at 77 K.
framework. Similar results have been reported for CO2
adsorption in metalated samples of another bipyridine-
containing framework, Al(OH)(bpydc).38
Although there is ample space in the coordination sphere of
the metal-linker complexes in 1·(CuCl2)5.8 and 1·(CuCl)6.8,
these metal centers only show weak interaction with additional
ligands, preferring lower coordination as a result of their filled d
shells. While the CoII and FeII centers in 1·(CoCl2)5.5 and 1·
(FeBr2)6.1 should be able to achieve higher coordination, these
complexes have the propensity to form tetrahedral complexes
after desolvation. Weakly coordinating gases such as H2 cannot
bind to these metal centers, most likely because the resulting
binding energy would not compensate for the reorganization
energy required for conversion to geometries that support a
higher coordination number.
Carbon monoxide adsorption experiments were performed
on 1·(CuCl2)5.8, 1·(CuCl)6.8, 1·(CoCl2)5.5, and 1·(FeBr2)6.1 at
298 K to probe the interaction of the metal sites with a more
coordinating gas (Figure 5). Surprisingly, only 1·(CuCl)6.8
showed significant uptake of CO, reaching a level of
approximately 0.4 per copper center at 0.2 bar. Here,
coordination of CO to the CuI centers is confirmed by the
observation of a peak at 2093 cm−1 in the infrared spectrum,
which agrees with reported CO stretching frequencies in similar
To provide insight into the structure of the Ir centers formed
in 1·(Ir(COD)BF4)4.7, CO adsorption measurements were
performed. These indicate substantial irreversible binding of
CO with uptake of roughly two CO molecules per Ir center by
0.001 bar CO. This presumably occurs through the substitution
of CO for COD on the Ir center to form 1·(Ir(CO)2BF4)4.7.
After CO adsorption, the resulting yellow material features two
Figure 5. Low-pressure CO adsorption isotherms for 1, 1·(CuCl2)5.8,
1·(CuCl)6.8, 1·(CoCl2)5.5, 1·(FeBr2)6.1 and 1·(Ir(COD)BF4)4.8 at 298
K.
G
DOI: 10.1021/acs.inorgchem.5b00096
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
new peaks in the infrared spectrum at 2088 and 2026 cm−1
(Supporting Information Figure S41), which can be assigned to
the asymmetric and symmetric CO stretching modes of the
(bpy)Ir(CO)2 complex. These frequencies agree with those
reported for similar cationic IrI dicarbonyl complexes.67,68
The compound [Ir(dmbpydc)(COD)]BF4 (3) was also
synthesized as a molecular analog and characterized by single-
crystal X-ray diffraction. The structure of 3 (see Figure 6)
Table 2. Control Experiments for Benzene C−H Borylation
with HBpin
a
catalyst
amount (mol %)
yield (%)
b
1·(Ir(COD)(BF4))0.6
1
0.065 (Ir)
0.68 (Zr)
0.18 (Ir)
0.84 (Zr)
0.065 (Ir)
0.069 (Ir)
100
0
[Ir(COD)2]BF4
0.9
0.03
4
2 + 0.6[Ir(COD)2]BF4
1·(Ir(COD)(BF4))0.6 + 0.6PPh3
1·(Ir(COD)(BF4))0.6 + 0.6PCy3
4
a
1
Yields were determined by H NMR using 1,3,5-trimethoxybenzene
b
as an internal standard. Determined as an average of three
replications.
Table 3. Control Experiments for Benzene C−H Borylation
with B2pin2
Figure 6. Crystal structure of 3, the molecular analog of 1·
(Ir(COD)BF4)4.7 as determined by single-crystal X-ray diffraction.
Dark blue, yellow green, olive green, red, blue, and gray spheres
represent IrI, F, B, O, N, and C atoms, respectively; H atoms are
omitted for clarity.
a
catalyst
amount (mol %)
yield (%)
b
1·(Ir(COD)(BF4))0.6
1
0.21 (Ir)
2.4 (Zr)
0.71 (Ir)
2.3 (Zr)
0.21 (Ir)
0.20 (Ir)
94
0
[Ir(COD)2]BF4
0.9
0
2 + 0.6[Ir(COD)2]BF4
1·(Ir(COD)(BF4))0.6 + 0.6PPh3
1·(Ir(COD)(BF4))0.6 + 0.6PCy3
0.05
0
supports the initial description of the Ir complexes in 1·
a
Yields were determined by 1H NMR spectroscopy using 1,3,5-
trimethoxybenzene as an internal standard. Determined as an average
of three replications.
(Ir(COD)BF4)4.7, with dmbpydc and COD bound to Ir in a
−
b
square planar geometry and BF4 as a noncoordinating anion.
The UV−vis spectrum of 1·(Ir(COD)BF4)4.7 (Supporting
Information Figure S44) reveals a shoulder at 466 nm and a
peak at 594 nm, which likely correspond to the metal-to-ligand
charge transfer bands of the Ir-bpy units in the framework.
These correspond well to peaks at 466 and 602 nm in the UV−
vis spectrum of 3. Furthermore, infrared spectra of 1·
(Ir(COD)BF4)4.7 (Supporting Information Figure S41) exhibits
peaks between 2922 and 2841 cm−1, which are assigned to the
aliphatic ν(C−H) of the coordinated COD moieties, and a
broad peak at 1056 cm−1, which is assigned to ν(B−F) of the
of magnitude of that reported for the best molecular systems.61
Note that supporting ligands that are stronger electron donors
are better at stabilizing the proposed IrV intermediate that
forms upon C−H activation of the arene,59,61,63 which is often
the rate-determining step of the reaction. Thus, the bipyridine-
Ir units in 1·(Ir(COD)BF4)0.6 are expected to be less active in
comparison to the best molecular Ir catalysts with more
electron-rich supporting ligands, such as 4,4′-di-tert-butylbipyr-
idine or 3,4,7,8-tetramethylphenanthroline.61,63
−
BF4 anion. Both features can also be observed in the same
regions in the infrared spectrum of 3. These results suggest that
the structure of 3 is representative of the Ir-bipyridine
complexes in 1·(Ir(COD)BF4)4.7.
Several control experiments were performed to establish that
catalysis is indeed facilitated by the Ir-bipyridine complexes in
framework (Tables 2 and 3). As predicted, [Ir(COD)2]BF4 and
1 both display limited or no catalytic activity. The metal−
organic framework Zr6O4(OH)4(bpdc)6 was synthesized and
treated with [Ir(COD)2]BF4 following the same procedure
used for 1. The Ir-treated biphenyl framework did not show
significant catalytic activity, excluding the possibility of catalysis
by adventitious Ir species that may form upon interaction with
the framework. A hot filtration experiment was performed to
determine if any catalytic activity was due to soluble Ir species
that may have formed during the reaction. No further increase
in turnover number was observed after 1·(Ir(COD)BF4)0.6 was
filtered from the reaction mixture (Supporting Information
Figure S46), which further supports that catalysis is facilitated
by the Ir-bipyridine complexes bound within the material.
To determine if catalysis occurs within the pores or only on
the surface of the metal−organic framework crystallites,
reactions were carried out in the presence of bulky coordinating
groups, which can selectively poison the surface Ir sites due to
Borylation reactions were generally conducted over 24 h at
80 °C in neat arene using either B2pin2 or HBpin as a boron
reagent. To ensure that the Ir sites in the framework are
accessible and to achieve catalyst loadings below 1 mol % Ir,
only 10% of the bipyridine sites were metalated in the samples
employed for catalysis. The Ir-metalated framework, 1·(Ir-
(COD)BF4)0.6, proved to be a highly active catalyst for the C−
H borylation of benzene to form 4,4,5,5-tetramethyl-2-phenyl-
1,3,2-dioxaborolane (PhBpin) in good yield at a very low
catalyst loading (Tables 2 and 3). Comparison of the powder
X-ray diffraction patterns of the material before and after
catalysis reveals no loss in crystallinity, suggesting that the
material is stable under the conditions used for catalysis.
Following conversion of benzene to PhBpin by 1H NMR
spectroscopy, the average turnover frequency of the catalyst
was determined to be 860 mol PhBpin mol Ir−1 h−1 at 80 °C
(Supporting Information Figure S45), which is within an order
H
DOI: 10.1021/acs.inorgchem.5b00096
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
their large size (Table 3).66 Addition of 1 equiv of either PPh3
or PCy3 per Ir to the reaction mixture led to drastically reduced
yields, suggesting that catalysis occurs primarily at the surface of
the material. This is consistent with Ir binding to only surface
bipyrdine sites and not to those within the pores.
Table 5. Selected Substrate Scope for C−H Borylation with
a
HBpin
Catalyst cycling experiments indicated no significant loss in
activity over five cycles when HBpin was employed, although
activity decreased substantially after the third and fourth cycles
when B2pin2 was used (Table 4). The much lower catalyst
Table 4. Catalyst Cycling Experiments for Benzene C−H
Borylation with Zr6O4(OH)4(bpydc)6(Ir(COD)BF4)0.6
a
yield (%)
boron reagent
1
2
3
4
5
b
HBpin
96
94
100
91
100
90
100
80
96
54
c
B2pin2
a
1
Yields were determined by H NMR using 1,3,5-trimethoxybenzene
b
as an internal standard. Reaction conditions: 1.3 mmol HBpin, 3.0
mL benzene, 0.10 mol % Ir, 80 °C, 24 h. Reaction conditions: 0.65
c
mmol B2pin2, 3.0 mL benzene, 0.20 mol % Ir, 80 °C, 24 h.
a
stability with B2pin2 may be attributed to impurities such as
trace moisture in the B2pin2 that can poison the catalytic sites.
It has previously been reported that the maximum turnover
number of the catalyst is highly dependent on the purity of the
B2pin2 used.60,61 If the catalyst were inherently unstable to the
reaction conditions, decreased activity after each cycle should
have also been observed when HBpin was used as a boron
reagent.
Reaction conditions: 2.0 mmol HBpin, 3 mL arene or arene mixture,
b
0.065 mol % Ir, 80 °C, 24 h. Product ratios and conversions were
determined by 1H NMR using 1,3,5-trimethoxybenzene as an internal
c
standard. Determined as an average of three replications.
reported results and the results in this work may be due to
differences in particle size, surface morphology, and metal
loading of the material, which indicates that these properties
may have unexpected and profound effects on catalysis and,
thus, should be more carefully studied.
The activity of 1·(Ir(COD)(BF4))0.6 in the borylation of
several substituted arenes reveals that the catalyst exhibits size
selectivity. Reaction with toluene, or 1,2-difluorobenzene,
results in nearly quantitative borylation, whereas substantially
lower activity is observed using tert-butylbenzene, o-xylene, and
m-xylene as substrates (Table 5). Assuming that the catalytic
sites are predominantly on the surface, as suggested by surface
poisoning experiments and low Ir loading after metalation, the
unexpected size selectivity suggests that the local steric
environment of the catalytic sites on the framework surface
can prevent access to the Ir centers or destabilize ideal
transition state conformations when bulkier substrates are used.
A substrate competition experiment between benzene and m-
xylene in 1:1 mixture showed that the catalyst has 95%
selectivity for benzene, further suggesting that the lower yields
observed for bulkier arenes are due to size or shape selectivity
and not catalyst instability or poisoning from trace impurities in
these solvents. Although size selectivity is not necessarily ideal
for this specific reaction, which has applications that require the
ability to functionalize a broad scope of substrates, it is
intriguing to observe that size selectivity still occurs even if the
catalytic sites are likely on the surface of the particles. This
implies that size selectivity in metal−organic frameworks does
not have to arise solely from restriction of substrate access to
catalytic sites by the pore apertures, but can also be influenced
by the local steric environment around the catalytic sites, as
dictated by the unique pore structure of a specific framework.
While this Article was in preparation, a recent paper reported
the similar application of Zr6O4(OH)4(bpydc)6 using [Ir-
(COD)(OMe)]2 as an Ir source and B2pin2 as the boron
source.33 Interestingly, the catalyst described in the report
maintained activity after 20 cycles using B2pin2 and displayed
no pronounced size selectivity. The disparity between the
CONCLUSIONS AND OUTLOOK
■
The foregoing results demonstrate that Zr6O4(OH)4(bpydc)6, a
metal−organic framework featuring open 2,2′-bipyridine sites,
can readily be metalated by a host of solution- and gas-phase
metal reagents. Moreover, single-crystal-to-single-crystal metal-
ation of the framework results in the ordering of the metal-
linker complexes, enabling structural characterization by single-
crystal X-ray diffraction. This remarkable structural ordering
arises from a transition from Fm3m to Pa3 symmetry induced
̅
̅
by the collective distortion of the framework linkers upon metal
chelation. These findings emphasize that structural determi-
nation of the metal complexes in postsynthetically metalated
metal−organic frameworks can be achieved if a high-symmetry
framework can transition to a lower symmetry upon metal
insertion. Furthermore, the [Ir(COD)2]BF4-metalated frame-
work is a highly active and recyclable catalyst for arene C−H
borylation.
Ongoing efforts are directed toward finding other frame-
works that display similar structural transitions upon metal-
ation, as well as to developing methods to limit disorder due to
linker rotation and variable solvation of the metal. In addition,
the in situ structural observation of reactions at the metal sites
in these frameworks by X-ray diffraction is being pursued.
Finally, efforts are underway to demonstrate that postsynthetic
metalation of metal−organic frameworks can be used to design
catalysts with unparalleled reactivity and selectivity through
judicious choice of metal node, chelating linker, and metal
precursor. Altogether, these results exemplify the distinct
advantages that metal−organic frameworks hold as highly
tunable, well-defined platforms for catalysis and the exceptional
I
DOI: 10.1021/acs.inorgchem.5b00096
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(11) Corma, A.; García, H.; Llabres
2010, 110, 4606−4655.
́
, i; Xamena, F. X. Chem. Rev.
ability to structurally characterize these materials through X-ray
diffraction.
(12) Wang, C.; Zheng, M.; Lin, W. J. Phys. Chem. Lett. 2011, 2,
1701−1709.
ASSOCIATED CONTENT
* Supporting Information
■
(13) Ma, L.; Abney, C.; Lin, W. Chem. Soc. Rev. 2009, 38, 1248.
(14) Xiao, D. J.; Bloch, E. D.; Mason, J. A.; Queen, W. L.; Hudson,
M. R.; Planas, N.; Borycz, J.; Dzubak, A. L.; Verma, P.; Lee, K.;
S
Additional experimental procedures and structural, powder X-
ray diffraction, gas adsorption, thermogravimetric analysis, and
catalysis data. Crystallographic data in CIF format. This
material is available free of charge via the Internet at http://
pubs.acs.org.
̀
Bonino, F.; Crocella, V.; Yano, J.; Bordiga, S.; Truhlar, D. G.;
Gagliardi, L.; Brown, C. M.; Long, J. R. Nat. Chem. 2014, 6, 590−595.
(15) Pullen, S.; Fei, H.; Orthaber, A.; Cohen, S. M.; Ott, S. J. Am.
Chem. Soc. 2013, 135, 16997−17003.
(16) Fei, H.; Shin, J.; Meng, Y. S.; Adelhardt, M.; Sutter, J.; Meyer,
K.; Cohen, S. M. J. Am. Chem. Soc. 2014, 136, 4965−4973.
(17) Feng, D.; Gu, Z.-Y.; Li, J.-R.; Jiang, H.-L.; Wei, Z.; Zhou, H.-C.
Angew. Chem., Int. Ed. 2012, 51, 10307−10310.
AUTHOR INFORMATION
Corresponding Author
*E-mail: jrlong@berkeley.edu.
■
(18) Anderson, J. S.; Gallagher, A. T.; Mason, J. A.; Harris, T. D. J.
Am. Chem. Soc. 2014, 136, 16489−16492.
(19) Bloch, W. M.; Burgun, A.; Coghlan, C. J.; Lee, R.; Coote, M. L.;
Doonan, C. J.; Sumby, C. J. Nat. Chem. 2014, 6, 906−912.
(20) Dhakshinamoorthy, A.; Alvaro, M.; Garcia, H. Catal. Sci.
Technol. 2011, 1, 856.
Present Address
For E.D.B.: Department of Chemistry and Chemical Biology,
Harvard University, 12 Oxford Street, Cambridge, Massachu-
setts 02138, United States.
Notes
(21) Dhakshinamoorthy, A.; Alvaro, M.; Garcia, H. Chem. Commun.
The authors declare no competing financial interest.
2012, 48, 11275.
̌
(22) Dhakshinamoorthy, A.; Opanasenko, M.; Cejka, J.; Garcia, H.
Catal. Sci. Technol. 2013, 3, 2509.
ACKNOWLEDGMENTS
■
̌
(23) Horike, S.; Dinca, M.; Tamaki, K.; Long, J. R. J. Am. Chem. Soc.
This work was supported by the Laboratory Directed Research
and Development Program of Lawrence Berkeley National
Laboratory under U.S. Department of Energy Contract DE-
AC02-05CH11231. The Advanced Light Source is supported
by the Director, Office of Science, Office of Basic Energy
Sciences, of the U.S. Department of Energy under Contract
DE-AC02-05CH11231. We thank the 11-BM staff at the
Advanced Photon Source at Argonne National Laboratory for
assisting with powder X-ray diffraction experiments. Use of the
Advanced Photon Source at Argonne National Laboratory was
supported by the U.S. Department of Energy, Office of Science,
Office of Basic Energy Sciences, under Contract DE-AC02-
06CH11357. We thank Gerald K. Branch and Arkema for the
fellowship support of E.D.B and Chevron for fellowship
support of J.A.M. We are grateful for the assistance of Antonio
DiPasquale and Kevin J. Gagnon with single-crystal X-ray
diffraction experiments and Jonathan Bachman with SEM
measurements. We also thank Kohei Takahashi, Matthew
Larsen, and Carl Liskey for fruitful discussions.
2008, 130, 5854−5855.
(24) Juan-Alcaniz, J.; Gascon, J.; Kapteijn, F. J. Mater. Chem. 2012,
̃
22, 10102.
(25) Genna, D. T.; Wong-Foy, A. G.; Matzger, A. J.; Sanford, M. S. J.
Am. Chem. Soc. 2013, 135, 10586−10589.
(26) Meilikhov, M.; Yusenko, K.; Esken, D.; Turner, S.; Van
Tendeloo, G.; Fischer, R. A. Eur. J. Inorg. Chem. 2010, 2010, 3701−
3714.
(27) Li, L.; Matsuda, R.; Tanaka, I.; Sato, H.; Kanoo, P.; Jeon, H. J.;
Foo, M. L.; Wakamiya, A.; Murata, Y.; Kitagawa, S. J. Am. Chem. Soc.
2014, 136, 7543−7546.
(28) Hasegawa, S.; Horike, S.; Matsuda, R.; Furukawa, S.; Mochizuki,
K.; Kinoshita, Y.; Kitagawa, S. J. Am. Chem. Soc. 2007, 129, 2607−
2614.
(29) Wang, C.; Xie, Z.; deKrafft, K. E.; Lin, W. J. Am. Chem. Soc.
2011, 133, 13445−13454.
(30) Wang, C.; deKrafft, K. E.; Lin, W. J. Am. Chem. Soc. 2012, 134,
7211−7214.
(31) Dau, P. V.; Cohen, S. M. Chem. Commun. 2013, 49, 6128.
(32) Fei, H.; Cohen, S. M. Chem. Commun. 2014, 50, 4810−4812.
(33) Manna, K.; Zhang, T.; Lin, W. J. Am. Chem. Soc. 2014, 136,
6566−6569.
(34) Falkowski, J. M.; Sawano, T.; Zhang, T.; Tsun, G.; Chen, Y.;
Lockard, J. V.; Lin, W. J. Am. Chem. Soc. 2014, 136, 5213−5216.
(35) Manna, K.; Zhang, T.; Carboni, M.; Abney, C. W.; Lin, W. J.
Am. Chem. Soc. 2014, 136, 13182−13185.
(36) Valvekens, P.; Bloch, E. D.; Long, J. R.; Ameloot, R.; De Vos, D.
E. Catal. Today 2014, 246, 55.
REFERENCES
■
(1) Serna, P.; Gates, B. C. Acc. Chem. Res. 2014, 47, 2612−2620.
(2) Wegener, S. L.; Marks, T. J.; Stair, P. C. Acc. Chem. Res. 2012, 45,
206−214.
(3) Hagen, J. Industrial Catalysis; John Wiley & Sons: New York,
2006.
(4) Furukawa, H.; Ko, N.; Go, Y. B.; Aratani, N.; Choi, S. B.; Choi,
E.; Yazaydin, A. O.; Snurr, R. Q.; O’Keeffe, M.; Kim, J.; Yaghi, O. M.
Science 2010, 329, 424−428.
(37) Evans, J. D.; Sumby, C. J.; Doonan, C. J. Chem. Soc. Rev. 2014,
43, 5933−5951.
(38) Bloch, E. D.; Britt, D.; Lee, C.; Doonan, C. J.; Uribe-Romo, F. J.;
Furukawa, H.; Long, J. R.; Yaghi, O. M. J. Am. Chem. Soc. 2010, 132,
14382−14384.
(5) Eddaoudi, M.; Kim, J.; Rosi, N.; Vodak, D.; Wachter, J.; O’Keeffe,
M.; Yaghi, O. M. Science 2002, 295, 469−472.
(6) Cohen, S. M. Chem. Rev. 2012, 112, 970−1000.
(39) Cavka, J. H.; Jakobsen, S.; Olsbye, U.; Guillou, N.; Lamberti, C.;
Bordiga, S.; Lillerud, K. P. J. Am. Chem. Soc. 2008, 130, 13850−13851.
(40) Li, L.; Tang, S.; Wang, C.; Lv, X.; Jiang, M.; Wu, H.; Zhao, X.
Chem. Commun. 2014, 50, 2304−2307.
̌
(7) Murray, L. J.; Dinca, M.; Long, J. R. Chem. Soc. Rev. 2009, 38,
1294−1314.
(8) Sumida, K.; Rogow, D. L.; Mason, J. A.; McDonald, T. M.; Bloch,
E. D.; Herm, Z. R.; Bae, T.-H.; Long, J. R. Chem. Rev. 2012, 112, 724−
(41) Nickerl, G.; Leistner, M.; Helten, S.; Bon, V.; Senkovska, I.;
Kaskel, S. Inorg. Chem. Front. 2014, 1, 325−330.
(42) Dau, P. V.; Kim, M.; Cohen, S. M. Chem. Sci. 2013, 4, 601.
(43) Schaate, A.; Roy, P.; Godt, A.; Lippke, J.; Waltz, F.; Wiebcke,
M.; Behrens, P. Chem.Eur. J. 2011, 17, 6643−6651.
781.
(9) Herm, Z. R.; Bloch, E. D.; Long, J. R. Chem. Mater. 2014, 26,
323−338.
(10) Mason, J. A.; Veenstra, M.; Long, J. R. Chem. Sci. 2013, 5, 32−
51.
J
DOI: 10.1021/acs.inorgchem.5b00096
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(44) Valenzano, L.; Civalleri, B.; Chavan, S.; Bordiga, S.; Nilsen, M.
H.; Jakobsen, S.; Lillerud, K. P.; Lamberti, C. Chem. Mater. 2011, 23,
1700−1718.
́
(45) Katz, M. J.; Brown, Z. J.; Colon, Y. J.; Siu, P. W.; Scheidt, K. A.;
Snurr, R. Q.; Hupp, J. T.; Farha, O. K. Chem. Commun. 2013, 49, 9449.
(46) Shearer, G. C.; Chavan, S.; Ethiraj, J.; Vitillo, J. G.; Svelle, S.;
Olsbye, U.; Lamberti, C.; Bordiga, S.; Lillerud, K. P. Chem. Mater.
2014, 26, 4068−4071.
(47) Awad, D. J.; Schilde, U.; Strauch, P. Inorg. Chim. Acta 2011, 365,
127−132.
(48) Healy, P. C.; Pakawatchai, C.; White, A. H. J. Chem. Soc., Dalton
Trans. 1985, 2531−2539.
(49) Hong, J.; Li, X.-Y.; Ng, S. W. Acta Crystallogr., Sect. E 2011, 67,
M1322−U1496.
(50) Xiao, L. N.; Wang, Y.; Pan, C. L.; Xu, J. N.; Wang, T. G.; Ding,
H. CrystEngComm 2011, 13, 4878.
(51) Cui, Y.; Chen, J.; Chen, G.; Ren, J.; Yu, W.; Qian, Y. Acta
Crystallogr., Sect. C 2001, 57, 349−351.
(52) Desroches, C.; Ohrstrom, L. Acta Crystallogr., Sect. C 2007, 63,
̈
m190−m192.
(53) Brewer, B.; Brooks, N. R.; Sykes, A. G. J. Chem. Crystallogr.
2003, 33, 663−668.
(54) Volbeda, J.; Meetsma, A.; Bouwkamp, M. W. Organometallics
2009, 28, 209−215.
(55) Vlcek, A.; Grevels, F. W.; Snoeck, T. L.; Stufkens, D. J. Inorg.
Chim. Acta 1998, 278, 83−90.
(56) Baxter, P. N. W.; Connor, J. A.; Wallis, J. D.; Povey, D. C.;
Powell, A. K. Polyhedron 1992, 11, 1771−1777.
(57) Zhou, W.; Wu, H.; Yildirim, T. J. Am. Chem. Soc. 2008, 130,
15268−15269.
(58) Ugozzoli, F.; Lanfredi, A. M. M.; Marsich, N.; Camus, A. Inorg.
Chim. Acta 1997, 256, 1−7.
(59) Mkhalid, I. A. I.; Barnard, J. H.; Marder, T. B.; Murphy, J. M.;
Hartwig, J. F. Chem. Rev. 2010, 110, 890−931.
(60) Ishiyama, T.; Takagi, J.; Ishida, K.; Miyaura, N.; Anastasi, N. R.;
Hartwig, J. F. J. Am. Chem. Soc. 2002, 124, 390−391.
(61) Boller, T. M.; Murphy, J. M.; Hapke, M.; Ishiyama, T.; Miyaura,
N.; Hartwig, J. F. J. Am. Chem. Soc. 2005, 127, 14263−14278.
(62) Tajuddin, H.; Harrisson, P.; Bitterlich, B.; Collings, J. C.; Sim,
N.; Batsanov, A. S.; Cheung, M. S.; Kawamorita, S.; Maxwell, A. C.;
Shukla, L.; Morris, J.; Lin, Z.; Marder, T. B.; Steel, P. G. Chem. Sci.
2012, 3, 3505.
(63) Larsen, M. A.; Hartwig, J. F. J. Am. Chem. Soc. 2014, 136, 4287−
4299.
(64) Waki, M.; Maegawa, Y.; Hara, K.; Goto, Y.; Shirai, S.; Yamada,
Y.; Mizoshita, N.; Tani, T.; Chun, W.-J.; Muratsugu, S.; Tada, M.;
Fukuoka, A.; Inagaki, S. J. Am. Chem. Soc. 2014, 136, 4003−4011.
(65) Wu, F.; Feng, Y.; Jones, C. W. ACS Catal. 2014, 1365−1375.
(66) Green, M.; Kuc, T. A.; Taylor, S. H. J. Chem. Soc., A 1971,
2334−2337.
(67) Rahaman, S. M. W.; Dinda, S.; Ghatak, T.; Bera, J. K.
Organometallics 2012, 31, 5533−5540.
(68) Harding, D. A. J.; Hope, E. G.; Singh, K.; Solan, G. A.
Polyhedron 2012, 33, 360−366.
NOTE ADDED AFTER ASAP PUBLICATION
■
This paper was published ASAP on February 26, 2015.
Additional Supporting Information was posted on the Web
on February 27, 2015.
K
DOI: 10.1021/acs.inorgchem.5b00096
Inorg. Chem. XXXX, XXX, XXX−XXX