Investigation of porous ni-based metal-organic frameworks containing paddle-wheel type inorganic building units via high-throughput methods

Article abstract of DOI:10.1021/ic200381f

In the search of Ni based metal-organic frameworks (MOFs) containing paddle-wheel type building units, three chemical systems Ni²⁺/HnL/ base/solvent with HnL = H₃BTC (1,3,5-benzenetricarboxylic acid), H3BTB (4,40,400,-benzene-1,3,5-triyl-tris- (benzoic acid)), and H₂BDC (terephthalic acid) were investigated using high-throughput (HT) methods. In addition to the conventional heating, for the first time HT microwave assisted synthesis of MOFs was carried out. Six new compounds were discovered, and their fields of formation were established. In the first system, H₃BTC was employed and a comprehensive HT-screening of compositional and process parameters was conducted. The synthesis condition for the Ni paddle-wheel unit was determined and two compounds [Ni₃(BTC)₂(Me ₂NH)₃]·(DMF)₄(H₂O) ₄ (1a) and [Ni₆(BTC)₂(DMF)₆(HCOO) ₆] (1b) were discovered (Me₂NH = dimethylamine, DMF = dimethylformamide). In the second system, the use of the extended tritopic linker H₃BTB and the synthesis conditions for the paddle-wheel units led to the porous MOF, [Ni₃(BTB)₂(2-MeIm) _1.5(H₂O)_1.5]·(DMF)_9- (H ₂O)_6.5 (2), (2-MeIm = 2-methylimidazole). This compound shows a selective adsorption of H₂O and H₂ with a strong hysteresis. In the third system, H₂BDC was used, and the base (DABCO) was incorporated as a bridging ligand into all structures. Thus, two pillared layered porous MOFs [Ni₂(BDC)₂(DABCO)]·(DMF) ₄(H₂O)_1.5(3a) and [Ni₂(BDC) ₂(DABCO)]·(DMF)₄(H₂O)₄(3b) as well as a layered compound [Ni(BDC)(DABCO)]·(DMF)_1.5(H ₂O)₂ (3c) were isolated. The 3a and 3b polymorphs of the [Ni₂(BDC)₂(DABCO)] framework can be selectively synthesized. The combination of microwave assisted heating, low overall concentration, stirring of the reaction mixtures, and an excess of DABCO yields a highly crystalline pure phase of 3b. The fields of formation of all compounds were established, and scale-up was successfully performed for 1b, 2, 3a, 3b, and 3c. All compounds were structurally characterized. In addition to IR, elemental and TG analyses, gas and vapor sorption experiments were carried out.

Full text of DOI:10.1021/ic200381f

ARTICLE
pubs.acs.org/IC
Investigation of Porous Ni-Based MetalꢀOrganic Frameworks
Containing Paddle-Wheel Type Inorganic Building Units via
High-Throughput Methods
Palanikumar Maniam and Norbert Stock*
Institut f€ur Anorganische Chemie, Christian-Albrechts-Universit€at zu Kiel, Max-Eyth-Strasse 2, D-24118 Kiel, Germany
S Supporting Information
b
ABSTRACT: In the search of Ni based metalꢀorganic frameworks (MOFs) containing
paddle-wheel type building units, three chemical systems Ni2þ/HnL/base/solvent with HnL
= H3BTC (1,3,5-benzenetricarboxylic acid), H3BTB (4,40,400,-benzene-1,3,5-triyl-tris-
(benzoic acid)), and H2BDC (terephthalic acid) were investigated using high-throughput
(HT) methods. In addition to the conventional heating, for the first time HT microwave
assisted synthesis of MOFs was carried out. Six new compounds were discovered, and their
fields of formation were established. In the first system, H3BTC was employed and a
comprehensive HT-screening of compositional and process parameters was conducted. The
synthesis condition for the Ni paddle-wheel unit was determined and two compounds
[Ni3(BTC)2(Me2NH)3] (DMF)4(H2O)4 (1a) and [Ni6(BTC)2(DMF)6(HCOO)6] (1b)
3
were discovered (Me2NH = dimethylamine, DMF = dimethylformamide). In the second
system, the use of the extended tritopic linker H3BTB and the synthesis conditions for the
paddle-wheel units led to the porous MOF, [Ni3(BTB)2(2-MeIm)1.5(H2O)1.5] (DMF)9-
3
(H2O)6.5 (2), (2-MeIm = 2-methylimidazole). This compound shows a selective adsorption of H2O and H2 with a strong hysteresis.
In the third system, H2BDC was used, and the base (DABCO) was incorporated as a bridging ligand into all structures. Thus, two
pillared layered porous MOFs [Ni2(BDC)2(DABCO)] (DMF)4(H2O)1.5 (3a) and [Ni2(BDC)2(DABCO)] (DMF)4(H2O)4
3
3
(3b) as well as a layered compound [Ni(BDC)(DABCO)] (DMF)1.5(H2O)2 (3c) were isolated. The 3a and 3b polymorphs of the
3
[Ni2(BDC)2(DABCO)] framework can be selectively synthesized. The combination of microwave assisted heating, low overall
concentration, stirring of the reaction mixtures, and an excess of DABCO yields a highly crystalline pure phase of 3b. The fields of
formation of all compounds were established, and scale-up was successfully performed for 1b, 2, 3a, 3b, and 3c. All compounds were
structurally characterized. In addition to IR, elemental and TG analyses, gas and vapor sorption experiments were carried out.
’ INTRODUCTION
type inorganic building units.15 The chemical and thermal
stability as well as the presence of accessible Lewis acid sites
are some of the reasons why HKUST-1 has been studied by many
groups. A search in the CSD database shows that most of the
compounds containing paddle-wheel units are based on Cu and
Zn, but examples with other ions such as Mo, Fe, and Cr were
also reported.16 Only few nickel carboxylates with paddle-wheel
units have been described. While most of these nickel com-
pounds exhibit no porosity,17ꢀ20 only two porous structures
were recently reported.21,22 The combination of porosity and the
presence of coordinatively unsaturated Ni2þ sites is also of
special interest because of catalytic properties23,24 and the strong
H2 binding affinity.25,26 The accessibility of the metal sites in the
these compounds depends very much on the nature and size of
Ni-based inorganic units. Simple dimeric clusters such paddle-
wheel units provide much less steric hindrance and thus, could
enhance hostꢀguest interactions. For these reasons, we started a
systematic investigation to explore the feasibility of synthesizing
High-throughput (HT) methodologies which are regarded as a
core component in the screening of new drugs in the pharmaceu-
tical fieldarealsoveryhelpfulinthe area ofsolidstatechemistry.1ꢀ3
By implementing this alternative method in solvothermal syntheses,
many groups have discovered new compounds such as zeolites,
metal-arsenates, -phosphonates, -carboxylates and -imidazolates.4ꢀ8
In contrast to conventional methods, HT methods allow the rapid
and systematic investigation of large synthesis fields.3,9 This enables
the efficient discovery of new compounds, the fast optimization of
synthesis conditions, and because of the large amount of data, it
allows the extraction of reaction trends. HT methods have been
successfully applied in the investigation of porous metalꢀorganic
frameworks (MOFs), which are of interest because of their potential
application in fields such as gas storage, gas separation, catalysis, and
drug delivery.10,11 Using HT methods, compounds such as MIL-101-
NDC,12 the amino-functionalized MOFs MIL-53, MIL-88, and
MIL-10113,14 and imidazolate based compounds such as ZIF-8 have
been discovered.8
One of the most intensively investigated MOF to-date is
HKUST-1 [Cu3(BTC)2(H2O)3] which contains paddle-wheel
Received: February 23, 2011
Published: May 03, 2011
r
2011 American Chemical Society
5085
dx.doi.org/10.1021/ic200381f Inorg. Chem. 2011, 50, 5085–5097
|

Inorganic Chemistry
ARTICLE
porous Ni2þ based MOFs with paddle-wheel units. Starting from
the well established linkers 1,3,5-benzenetricarboxylic acid (H3BTC),
4,40,400,-benzene-1,3,5-triyl-tris(benzoic acid) (H3BTB) and ter-
ephthalic acid (H2BDC), we investigated the three chemical
systems Ni2þ/HnL/base/solvent with HnL = H3BTC, H3BTB,
and H2BDC. Here we report the fields of formation of six
compounds, four containing Ni paddle-wheel units, in these
three systems, and their structural elucidation, as well as their
detailed characterization.
[Ni3(BTC)2(Me2NH)3] (DMF)4(H2O)4 (1a). In a HT experiment
1a was obtained by the following procedure (molar ratio Ni-
3
(NO3)2 6H2O/H3BTC/2-MeIm = 2:1.5:1). DMF solutions of Ni-
3
3
(NO3)2 6H2O (0.03 mmol, 30 μL), H3BTC (0.015 mmol, 60 μL),
2-MeIm (0.01 mmol, 10 μL), and DMF (100 μL) were pipettedinto a 300
μL reactor. The mixture was heated at 170 °C for 48 h in a conventional
oven. The synthesis of compound 1a could not be scaled-up in glass
reactors or larger Teflon reactors. Therefore, a larger amount of the pure
phase product was collected from a separate HT experiment containing 24
identical reaction mixtures with the molar ratio Ni(NO3)2 6H2O/
3
H3BTC/2-MeIm = 2:1.5:1. The product (69 mg, 35% based on H3BTC)
containing dark green octahedral crystals (SEM micrograph Supporting
Information, Figure S1) was washed with fresh DMF and identified by
XRPD measurements. The measured and simulated XRPD patterns
(Supporting Information, Figure S4) compare well. Elemental analysis
’ EXPERIMENTAL SECTION
Material and Methods. All reagents are of analytical grade (Sigma-
Aldrich and ABCR) and were used without any further purification.
H3BTB was provided by BASF SE. High-throughput X-ray powder
diffraction (XRPD) experiments were carried out in transmission mode
using a STOE Stadi P combi high-throughput X-ray powder diffract-
ometer equipped with an image-plate position-sensitive detector
(IPPSD). MIR spectra were recorded with an ATI Matheson Genesis
spectrometer in the spectral range 4000ꢀ400 cmꢀ1 using the KBr disk
method. Thermogravimetric analyses were carried out in air (75 mL
minꢀ1, 30ꢀ800 °C, 4 °C minꢀ1) or nitrogen (75 mLminꢀ1, 30ꢀ800 °C,
4 °C minꢀ1) atmosphere using a NETZSCH STA 409 CD analyzer.
CHN analyses were performed with a Eurovektor EuroEA Elemental
Analyzer. The SEM images and semiquantitative elemental analyses were
performed with a Phillips ESEM XL 30 and JEOL JSM-6500F hot
cathode scanning electron microscope equipped with an energy disper-
sive X-ray (EDX) EDAX analyzer forelementalanalysis. N2, CO2, Ar, and
H2O sorption experiments were performed using a Belsorp-max appa-
ratus (BEL JAPAN INC.). Low pressure volumetric hydrogen sorption
measurements were conducted with a Quantachrome Autosorb 1-C
instrument at 77 K; H2 as well as He gases of 99.999% purity were used.
For the activation prior tosorptionmeasurement, allsampleswere heated
overnight at 150 °C in vacuum (10ꢀ3 mbar).
High-Throughput Experiments (Conventional Heating).
The reaction system Ni2þ/HnL/base/solvent was investigated using
various molar ratios Ni2þ:HnL:base (discovery and focused arrays)
employing high-throughput methods. The process and compositional
parameters arelisted intheSupporting Information, TablesS1ꢀS15. The
HT experiments via conventional heating were performed under sol-
vothermal conditions in a custom-made stainless steel high-throughput
reactor system containing 48 PTFE inserts each with a maximum volume
of 300 μL.27 For most of the HT-experiments, standard solutions of
starting materials were made by dissolving all solid reagents in the solvent
under magnetic stirring (Supporting Information, Tables S1ꢀS15).
Respective volumes of each reactant were manually dosed into the PTFE
inserts. In some of the HT-experiments, solid reactants with low
solubility were weighed in directly into the Teflon inserts. The exact
amounts of starting materials are given in the Supporting Information,
Table S1ꢀS15. The evaluations of the HT experiments are based on
XRPD measurements.
of [Ni3(BTC)2(Me2NH)3] (DMF)4(H2O)4 (1a), M = 1090.02 g
3
molꢀ1: found C 39.6, H 5.1, N 9.3; calcd C 39.6, H 5.8, N 9.0.
[Ni6(BTC)2(HCOO)6(DMF)6] (1b). In a HT experiment, 1b was
obtained by the following procedure (molar ratio Ni(ClO4)2 6H2O/
3
H3BTC/2-MeIm = 3:1:1). DMF solutions of Ni(ClO4)2 6H2O (0.03
3
mmol, 60 μL), H3BTC (0.01 mmol, 40 μL), 2-MeIm (0.01 mmol, 20
μL), and DMF (80 μL) were pipetted into a 300 μL reactor. The mixture
was heated at 150 °C for 48 h in a conventional oven. The scale-up of 1b
was accomplished in DURAN glass tubes with Vmax = 4 mL by employing
Ni(ClO4)2 6H2O (0.3 mmol, 109.7 mg), H3BTC (0.1 mmol, 21.0 mg),
3
2-MeIm (0.1 mmol, 8.2 mg) and 2 mL of DMF. The product (49.4 mg,
67% based on H3BTC) containing green hexagonal plates (SEM
micrograph Supporting Information, Figure S1) was washed with fresh
DMF and identified by XRPD measurements. The measured and
simulated XRPD patterns (Supporting Information, Figure S4) compare
well. Elemental analysis of [Ni6(BTC)2(HCOO)6(DMF)6] (1b), M =
1475.05 g molꢀ1: found C 34.5, H 4.0, N 5.8; calcd C 34.2, H 3.7, N 5.7.
[Ni3(BTB)2(2-MeIm)1.5(H2O)1.5] (DMF)9(H2O)6.5 (2). In a HT
3
experiment under conventional heating 2 was obtained by the following
procedure (molar ratio Ni(NO3)2 6H2O/H3BTB/2-MeIm = 1.5:3:3.5).
3
DMF solutions DMF of Ni(NO3)2 6H2O (0.015 mmol, 15 μL), H3BTB
3
(0.03 mmol, 105 μL), 2-MeIm (0.035 mmol, 35 μL), and DMF (45 μL)
were pipetted into a 300 μL reactor. The mixture was heated at 170 °C
for 48 h. The scale-up of 2 was accomplished in a batch Teflon-lined
steel autoclave with Vmax = 20 mL by employing Ni(NO3)2 6H2O
3
(1.5 mmol, 436 mg), H3BTB (3 mmol, 1314 mg), 2-MeIm (3.5 mmol,
287 mg), and 20 mL of DMF. The product (769 mg, 78% based on
Ni(NO3)2 6H2O) containing dark brown blocks (SEM micrograph
3
Supporting Information, Figure S1) was washed with fresh DMF and
identified by XRPD measurements. The measured and simulated XRPD
patterns (Supporting Information, Figure S5) compare well. Elemental
analysis of [Ni3(BTB)2(2-MeIm)1.5(H2O)1.5] (DMF)9(H2O)6.5 (2),
3
M = 1971.83 g molꢀ1: found C 52.6, H 5.4, N 8.8; calcd C 52.9, H 6.0,
N 8.5.
[Ni2(BDC)2(DABCO)] (DMF)4(H2O)1.5 (3a). In a HT experi-
3
ment under conventional heating, 3a was obtained by the following
procedure (molar ratio Ni(NO3)2 6H2O/H2BDC/DABCO = 2:2:1).
3
DMF solutions of Ni(NO3)2 6H2O (0.02 mmol, 20 μL), H2BDC (0.02
High-Throughput Experiments (Microwave Heating). The
system Ni2þ/H2BDC/DABCO/DMF was investigated under micro-
wave assisted heating. All reactions were performed using Anton Paar
Synthos 3000 multimode microwave unit employing a 4 ꢁ 24MG5 rotor.
Sealed borosilicate glass vials with reaction mixtures were used, which
were inserted into the silicon carbide plate (6 ꢁ 4 matrix). The
temperature of the plates is monitored using an IR sensor located at
the bottom of the microwave unit. Samples requiring stirring were
agitated with 10 mm magnetic stir bars during synthesis. The exact
amounts of starting materials and reaction conditions are given in the
Supporting Information, Tables S16ꢀS19. The evaluations of the HT
experiments are based on XRPD measurements.
3
mmol, 60 μL), DABCO (0.01 mmol, 10 μL), and DMF (110 μL) were
pipetted into a 300 μL reactor. The mixture was heated at 110 °C for
48 h. The scale-up of 3a was accomplished in a batch Teflon-lined steel
autoclave with Vmax = 20 mL by employing Ni(NO3)2 6H2O (2 mmol,
3
580 mg), H2BDC (2 mmol, 332 mg), DABCO (1 mmol, 112 mg) and
20 mL of DMF. The product (614 mg, 70% based on H2BDC)
containing green microcrystalline powder (SEM micrograph Supporting
Information, Figure S1) was washed with fresh DMF and identified by
XRPD measurements (Supporting Information, Figure S6). Elemental
analysis of [Ni2(BDC)2(DABCO)] (DMF)4(H2O)1.5 (3a), M =
3
877.12 g molꢀ1: found C 46.2, H 5.4, N 9.3; calcd C 46.5, H 5.8, N 9.6.
5086
dx.doi.org/10.1021/ic200381f |Inorg. Chem. 2011, 50, 5085–5097

Inorganic Chemistry
ARTICLE
Table 1. Crystallographic Data for [Ni3(BTC)2(Me2NH)3] (DMF)4(H2O)4 (1a), [Ni6(BTC)2(DMF)6(HCOO)6] (1b), and
3
[Ni3(BTB)2(2-MeIm)1.5(H2O)1.5] (DMF)9(H2O)6.5 (2)
3
1aa
1b
2aa
chemical formula
formula weight (g/mol)
crystal system
space group
a (Å)
Ni3C36H63N7O20
1090.02
cubic
Ni6C42H54N6O30
1475.05
trigonal
P3 (No. 147)
13.8230(4)
13.8230(4)
7.9625(2)
90
Ni3C87H118N12O29
1971.83
cubic
Fm3m (No. 225)
26.5941(7)
26.5941(7)
26.5941(7)
90
Im3 (No. 204)
26.816(3)
26.816(3)
26.816(3)
90
b (Å)
c (Å)
R (deg)
β (deg)
90
90
90
γ (deg)
90
120
90
V (Å3)
18808.6(9)
16
1317.60(6)
1
19283(6)
8
Z
λ (Å)
Dcalcd (g cmꢀ3
0.71013
1.025
0.71013
1.859
0.71013
0.831
)
μ (mmꢀ1
)
1.23
2.20
0.62
temp (K)
293
120
293
R1 (on Fo2, I > 2σ(I))
wR2 (on Fo2, I > 2σ(I))
0.068
0.040
0.076
0.180
0.070
0.239
a The SQUEEZE routine of the program PLATON was used to eliminate the contribution of disordered solvents molecules.29
[Ni2(BDC)2(DABCO)] (DMF)4(H2O)4 (3b). In a HT-MW experi-
ment 3b was obtained by the following procedure (molar ratio Ni-
SQUEEZE routine (PLATON).29 The single crystal of 1b was measured
using a BrukerꢀNonius APEX II CCD diffractometer equipped with a
BrukerꢀNonius FR591 rotating anode Mo-KR radiation source. Ab-
sorption correction was done using the SADABS software.30 The single
crystal structures were solved by direct methods and refined using the
program package SHELXTL.31 The results of single crystal structure
determinations are summarized in Table 1. The structure of compound
3c was solved using the following software programs: EXPO2009,32
Material Studio,33 and FOX.34 The full Rietveld refinement of com-
pounds 3a, 3b, and 3c were carried out with TOPAS.35 Results and final
Rietveld refinement plots are given in Table 2 and Supporting Informa-
tion, Figures S7ꢀS9. Comparison of NiꢀO and NiꢀN bond lengths of
the paddle-wheel based Ni(II) compounds are summarized in the
Supporting Information, Table S20.
3
(NO3)2 6H2O/H2BDC/DABCO
= 1:1:4). DMF solutions of
3
Ni(NO3)2 6H2O (0.1 mmol, 200 μL), H2BDC (0.1 mmol, 300 μL),
3
DABCO (0.4 mmol, 800 μL), and DMF (700 μL) were pipetted into a
4 mL glass vial. The mixture was heated under stirring in the microwave
unit at 110 °C for 2 h (set power value =400 W). Larger amounts of 3b
were obtained by using a separate HT-MW experiment containing six
identical reaction mixtures with the molar ratio Ni(NO3)2 6H2O/
3
H2BDC/DABCO = 1:1:4. The product (163 mg, 59% based on H2BDC)
containing green microcrystalline powder (SEM micrograph Supporting
Information, Figure S1) was washed with fresh DMF and identified by
XRPD measurements (Supporting Information, Figure S6). Elemental
analysis of [Ni2(BDC)2(DABCO)] (DMF)4(H2O)4 (3b), M = 922.23 g
3
molꢀ1: found C 43.8, H 5.7, N 9.2; calcd C 44.3, H 6.1, N 9.1.
[Ni(BDC)(DABCO)] (DMF)1.5(H2O)2 (3c). In a HT-MW experi-
ment 3c was obtained by the following procedure (molar ratio Ni-
’ RESULTS AND DISCUSSION
3
The systematic investigation of a chemical system comprises a
large number of parameters. Many chemical parameters such as
molar ratios of starting materials, the overall concentration,
additives (e.g., a base), the solvent, or the metal salt employed
have a strong influence on the product formation and the phase
purity.36,37 Furthermore, process parameters such as the heating
program, the reaction time, and the reaction temperature have to
be taken into account.38 To identify the reaction conditions
leading to Ni-based paddle-wheel units, all these parameters were
investigated. Starting from so-called discovery arrays, large parts
of the parameter fields are scanned. Once a new compound is
discovered, focused arrays are set up which focus on a smaller
part or the parameter space and thus allow us to establish the
optimized synthesis conditions.27 In the following sections the
results of our HT- investigations of the three respective systems
are described and discussed.
(NO3)2 6H2O/H2BDC/DABCO
= 3:1:4). DMF solutions of
3
Ni(NO3)2 6H2O (0.3 mmol, 600 μL), H2BDC (0.1 mmol, 300 μL),
3
DABCO (0.4 mmol, 800 μL), and DMF (300 μL) were pipetted into a
4 mL glass vial. The mixture was heated in the microwave unit at 110 °C
for 2 h (set power value = 400 W). Larger amounts of 3c were obtained
using a separate HT-MW experiment containing six identical reaction
mixtures with the molar ratio Ni(NO3)2 6H2O/H2BDC/DABCO =
3
3:1:4. The product (153 mg, 53% based on H2BDC) containing pale
green microcrystalline powder (SEM micrograph Supporting Informa-
tion, Figure S1) was washed with fresh DMF and identified by XRPD
measurements (Supporting Information, Figure S6). Elemental analysis
of [Ni(BDC)(DABCO)] (DMF)1.5(H2O)2 (3c), M = 480.65 g molꢀ1
:
3
found C 46.2, H 6.0, N 10.4; calcd C 46.2, H 6.4, N 10.2.
X-ray Crystallography. Suitable crystals of the compounds were
carefully selected from the HT experiments using a polarizing micro-
scope. Single-crystal X-ray diffraction for 1a and 2 were performed with a
STOE IPDS-1 diffractometer equipped with a fine-focus sealed tube
(Mo-KR radiation, λ = 71.073 pm). For data reduction and absorption
correction the programs XRED and XSHAPE were used.28 For 1a and 2,
contributions from disordered solvent molecules were removed by the
HT-Investigation of the System Ni2þ/H3BTC/Base/Sol-
vent. The first and most extensive screening of the reaction
parameters was carried out in the first system Ni2þ/H3BTC/
base/solvent (Figure 1). At first the influence of different bases,
5087
dx.doi.org/10.1021/ic200381f |Inorg. Chem. 2011, 50, 5085–5097

Inorganic Chemistry
Table 2. Crystallographic Data for [Ni2(BDC)2(DABCO)] (DMF)4(H2O)1.5 (3a), [Ni2(BDC)2(DABCO)] (DMF)4(H2O)4
ARTICLE
3
3
(3b), and [Ni(BDC)(DABCO)] (DMF)1.5(H2O)2 (3c)
3
3a
3b
3c
chemical formula
formula weight (g/mol)
crystal system
Ni2C34H51N6O13.5
877.12, 557.75a
tetragonal
1.360, 0.861a
I4/mcm (No. 140)
15.1496(7)
15.1496(7)
18.6597(8)
90
Ni2C34H56N6O16
922.23, 557.75a
trigonal
1.215, 0.735a
P3m1 (No. 164)
21.5930(3)
21.5930(3)
9.3672(1)
90
NiC18.5H30.5N3.5O7.5
480.65, 334.98a
monoclinic
1.762, 1.228a
P2/m (No. 10)
10.6682(7)
7.0125(4)
6.0822(4)
90
calculated density (g cmꢀ3
)
3
space group
a (Å)
b (Å)
c (Å)
R (deg)
β (deg)
90
90
95.269(5)
90
γ (deg)
90
120
V (Å3)
4282.6(4)
4
3782.4(1)
3
453.1(1)
1
Z
λ (Å)
1.5406
1.5406
1.5406
2θ range (deg)
2.0 to 80.0
13
2.0 to 80.0
11
2.0 to 80.0
17
no. of non H-atoms
GoF
1.158
2.930
3.570
no. structural parameters
39
40
79
Rwp
Rp
0.044
0.054
0.139
0.033
0.039
0.097
RBragg
0.009
0.032
0.055
a The values are calculated without guest molecules.
imidazole (Im), 2-methylimidazole (2-MeIm), 1,4-diazabicyclo-
[2.2.2]octane (DABCO), and 4,4-bipyridine (4,4-Bipy) in the
optimization of 1a (Figure 1, Supporting Information, Table S3).
On the basis of the results of the XRPD measurements,
only pure, dry DMF led to the most crystalline and pure phase
product 1a. Nevertheless, small amounts of H2O introduced
system NiCl2 6H2O/H3BTC/base/DMF, was investigated
3
varying the molar ratio Ni2þ/H3BTC. At 150 °C, only 2-MeIm
led to a new, well crystalline phase which was identified as
through the use of Ni(NO3)2 6H2O lead to the partial hydro-
3
[Ni3(BTC)2(Me2NH)3] (DMF)4(H2O)4 (1a) and is isostruc-
lysis of DMF and the formation of dimethylamine (Me2NH)
and formic acid. While formate ions can act as anionic
ligands, dimethylamine can act as a monodentate ligand or in
the protonated form as a counterion to balance the framework
charge.39,40 In 1a and 1b dimethylamine and formate ions are
observed, respectively.
3
tural to HKUST-1 (Supporting Information, Table S1).15 In
contrast to the copper-analogue HKUST-1 which can be ob-
tained in various solvents without any additives via solvothermal,
electrochemical, microwave, or sonothermal methods (Supporting
Information, Table S21), the presence of the base 2-MeIm is
essential for the formation of 1a. The role of 2-MeIm can not be
explained up to now. It is not a structure directing agent since no
2-MeIm could be found in the final product 1a.
The reaction temperature also plays a crucial role in solvother-
mal syntheses, especially with respect to the crystallinity and the
yield of the phase.38,41 Therefore, the reaction system Ni2þ
/
To findthe optimal Ni2þ source for the synthesis, differentsalts,
H3BTC/2-MeIm in DMF was investigated at five different
temperatures between 140 and 180 °C using 48 different starting
compositions, each (Figure 2, Supporting Information, Tables
S4ꢀS8). The products were analyzed per XRPD measurements,
and the results are plotted in ternary crystallization diagrams
(Figure 2). In addition to 1a and 1b, a new crystalline phase was
detected (1c) at temperatures between 150 and 170 °C, which
rapidly decomposed upon exposure to air. This phase is only
found at the edge of the crystallization diagram in the area of high
2-MeIm and low Ni2þ concentrations. 1a and 1c crystallize as
dark green octahedral and rod-shaped crystals, respectively, while
1b forms light green hexagonal crystals (Supporting Information,
Figure S1). Figure 2 shows that 1a preferentially forms at
temperatures of 160ꢀ180 °C, but only at 170 °C crystals with
a well-defined shape (octahedral morphology) were obtained
(Supporting Information, Figure S1). Pure phase 1a with high
crystallinity was observed at the molar ratio Ni(NO3)2/H3BTC/
2-MeIm = 2:1.5:1 at 170 °C. Large amounts of 1a could only be
obtained in our 48-er reactor system. Title compound 1b is the
that is, NiCl2 6H2O, NiSO4 6H2O, Ni(ClO4)2 6H2O, Ni-
3
3
3
(CH3COO)2 4H2O, and Ni(NO3)2 6H2O, were screened in
3
3
combination with H3BTC, 2-MeIm, and DMF at a reaction
temperature of 150 °C (Figure 1, Supporting Information, Table
S2). On the basis of the results of the XRPD measurements,
NiSO4 6H2O and Ni(CH3COO)2 4H2O produce only X-ray
3
3
amorphous products while Ni(ClO4)2 6H2O led exclusively to
3
the new dense compound [Ni6(BTC)2(DMF)6(HCOO)6] (1b).
Ni(NO3)2 6H2O yields most often a mixture of both phases 1a
3
and 1b depending on the molar ratio of the reactants (Supporting
Information, Table S2). Although the use of NiCl2 6H2O re-
3
sulted always in the formation of 1a, EDX measurements showed
the presence of large amounts of Clꢀ ions in the as synthesized
product (Ni/Cl = 1:4ꢀ5). Removal of the Clꢀ ions was investi-
gated but always led to the structural decomposition. Therefore,
Ni(NO3)2 6H2O was used for all further investigations.
3
The use of various solvents well-known for the synthesis of
MOFs and mixtures thereof were screened for the synthesis
5088
dx.doi.org/10.1021/ic200381f |Inorg. Chem. 2011, 50, 5085–5097

Inorganic Chemistry
ARTICLE
based porous MOFs with tetragonal framework structures were
reported in recent years.21,43ꢀ45 On the basis of the Cu-com-
pound, Seki et al. proposed the structure model in 2001.44 The
tetragonal structure was later refined from single crystal XRD data
of the isoreticular Zn-based compound.45,46 A Zn-based poly-
morph of this MOF with a trigonal framework (Kagomꢀe net
topology47) was first synthesized by Chun et al. in a ternary
solvent system.48 The polymorph formation, tetragonal phase
versus trigonal phase, can be controlled using different reaction
times or using anion-templating as observed in mechanochemical
synthesis.49,50 While the Zn-based compounds have been studied
in detail, almost no data is available on the Ni-based systems.21,22
Rather than focusing only on the solvent system or the reaction
time, we decided to study the polymorph formation by system-
atically investigating the compositional parameters as well as
the reaction temperature. By employing the HT-reactor system,
a discovery array was setup with 48 different Ni(NO3)2/H2BDC/
DABCO compositions using DMF as the sole solvent (Supporting
Information, Figure S3 and Supporting Information, Tables
S11ꢀS14). Four temperatures (90°, 110°, 130°, and 150 °C)
were chosen, and the reactors were subjected to conventional
heating for 48 h. Both polymorphs, the tetragonal [Ni2(BDC)2-
(DABCO)] (DMF)4(H2O)1.5 (3a) and trigonal [Ni2(BDC)2-
3
(DABCO)] (DMF)4(H2O)4 (3b) phases were observed, and
3
the new structure [Ni(BDC)(DABCO)] (DMF)1.5(H2O)2
3
Figure 1. Summary of the HT screening of nickel(II) salts, bases, and
solvents which provided the optimal reactants (shaded box) for the
subsequent reaction temperature screening in HT-investigation of the
system Ni2þ/H3BTC/base/solvent. The definitions of the abbrevia-
tions used are following: Im = imidazole, 2-MeIm = 2-methylimidazole,
DABCO = 1,4-diazabicyclo[2.2.2]octane, 4,4-Bipy =4,4-bipyridine,
DMA = dimethylacetamide, DMF = dimethylformamide, DEF =
diethylformamide, H3BTC = 1,3,5-benzenetricarboxylic acid. Details
of the experiments are given in Supporting Information, Tables S1ꢀS3.
(3c) was discovered.
Title compound 3a is the prevalent phase in the upper part of the
crystallization diagram within the molar ratio range Ni(NO3)2/
H2BDC/DABCO = 1ꢀ5:2ꢀ6:1ꢀ4, whereas phase 3c dominates
the lower part in the molar ratio range Ni(NO3)2/H2BDC/DABCO
= 1ꢀ4:1:3ꢀ6. A scale-up synthesis of 3a could be performed
successfully using a 20 mL Teflon batch reactor with the optimal
molar ratio Ni(NO3)2/H2BDC/DABCO = 2:2:1 at 110 °C for 48 h.
In the overlapping section of the 3a and 3c formation fields, title
compound 3b was observed. It always cocrystallizes as the minor
phase with 3a or 3c. Comparing thenumberofoccurrences of 3b at
different temperatures, this compound was predominantly formed
at 110 °C (Supporting Information, Figure S3). Since in the
corresponding Zn-based system, a short reaction time leads to
the formation of the trigonal phase,49 we also investigated the
influence of the reaction time. A decrease to 1 h helps to increase
the yield of 3b, but the cocrystallization of 3a could not be
avoided (Figure 4, Supporting Information, Table S15). The
results of HT-investigations via conventional heating at different
reaction times have generally shown that both 3a and 3b are
rapidly formed in relatively short time. In addition, only micro-
crystalline products of 3a and 3b have been obtained in all
experiments. These factors indicate that the nucleation rate of
both phases is high which usually leads to many small crystals.51
To further investigate the polymorph formation, high-through-
put experiments using microwave-assisted heating (HT-MW)
with magnetic stirring were conducted (Figure 5). Microwave
heating, which has been mostly utilized in the field of organic
chemistry, is reported to have many advantages such as short
reaction times, higher product yields, and increased product
purity.52,53 Recent reports have also shown that porous materi-
als can be synthesized with microwave heating and
most importantly, phase selectivity has been demonstrated,
for example, for silicoaluminiumphosphates.54,55 Therefore, the
following studies were carried out: (a) influence of the molar
ratios Ni2þ/H2BDC/DABCO (Figure 5A), (b) role of stirring
during the reaction (Figure 5B), (c) influence of the overall
dominant phase at lower temperatures, and scale-up was easily
accomplished.
HT-Investigation of Ni(NO3)2/H3BTB/2-MeIm System. Once
the synthesis condition of the Ni paddle-wheel unit for 1a was
established, H3BTC was replaced with the extended linker H3BTB
and the optimal reactants and solvent, that is, nickel(II)-nitrate
salt, 2-methylimidazole, and DMF, wereemployed. Usinganinitial
reaction temperature of 150 °C and a reaction time of 72 h, a
discovery array containing 48 different molar ratios of Ni(NO3)2/
H3BTB/2-MeIm wereinvestigated. As shown inthe crystallization
diagram (Figure 3), only one crystalline phase is observed. Large
dark brown cubic crystals of the paddle-wheel based structure
[Ni3(BTB)2(2-MeIm)1.5(H2O)1.5] (DMF)9(H2O)6.5 (2) were
3
obtained (Supporting Information, Figure S2). Upon exposure to
air the crystals change their color to green in minutes without the
loss of crystallinity. The title compound is only formed in a small,
well-defined area of the crystallization diagram (Figure 3). A
focused array was set up to narrow down the compositional
parameters, and the reaction temperature was varied between
130 and 170 °C. Well crystalline product of 2 is obtained at the
molar ratio range Ni(NO3)2/H3BTC/2-MeIm = 1.25ꢀ1.5:1.25ꢀ
3.8:3ꢀ5.5 at 170 °C. Employing these conditions, the synthesis
scale-up was carried out without any problems. In comparison to
analogous Cu-based MOF-14 which is formed under mild synth-
esis conditions,42 compound 2 prefers relatively high reaction
temperatures and 2-methylimidazole as an additive.
HT-Investigation of Ni2þ/H2BDC/DABCO System. Series of
[(M2þ)2(BDC)2(DABCO)] (M = Ni, Co, Cu, Zn) paddle-wheel
5089
dx.doi.org/10.1021/ic200381f |Inorg. Chem. 2011, 50, 5085–5097

Inorganic Chemistry
ARTICLE
Figure 2. Ternary crystallization diagrams showing the results of HT reaction temperature screening between 140° and 180 °C. Compound 1b
dominates the formation fields at lower temperatures. 1a forms preferentially at higher temperatures, mostly in the center of the diagram. 1c crystallizes
in a small formation field with high 2-MeIm amounts at 150ꢀ170 °C. Overlapping symbols stand for mixed phases. Results are based on XRPD
measurements. Molar ratios of reactants are normalized to 1.00. Details of the experiments are given in Supporting Information, Tables S4ꢀS8.
concentration (Figure 5C), and (d) influence of the Ni salt
counterion (Figure 5D). These experiments were carried out
repeatedly (2ꢀ3 times each) to confirm the trends observed.
On the basis of the previous results (Figure 4), molar ratios
Ni(NO3)2/H2BDC = 1:1, 1.5:1, and 2:1 were chosen for the HT-
MW experiment and the amount of DABCO was varied
(Figure 5A, Supporting Information, Table S16). In contrast to
the experiments under conventional heating (no stirring), pure
phase 3b could be obtained byapplying thesamereactiontimeand
temperature. Employing a molar ratio of Ni(NO3)2/H2BDC/
DABCO/DMF = 1:1:4:260 and microwave heating at 110 °C for
1 h under stirring yields a highly crystalline single-phase product of
3b. Another HT-MW experimentwas performed to cross-examine
the stirring effect on the selective crystallization of 3a and 3b
(Figure 5B, Supporting Information, Table S17). Interestingly, in
this concentration regime pure phase of 3b only forms under
stirring and using large molar excess of DABCO, while the
nonstirred reaction mixtures and smaller amounts of DABCO
lead to the formation of 3a. The effect of stirring on the phase
selectivity is a relatively undeveloped field, and so far, it has been
reported in the area of selective asymmetric synthesis56,57 and
synthesis of zeolites, for example, for zeolite Y polymorphs.58
The overall concentration has also a large influence on the
product formation. In a HT-MW experiment, the Ni(NO3)2/
H2BDC/x DABCO/y DMF (x = 0.5 and 4) reaction mixtures
with y ranging from 65 to 5200 were investigated (Figure 5C,
Supporting Information, Table S18, previously used concentra-
tions in experiment 5B are marked by an asterisk). The results
show that under highly diluted conditions and independent of
stirring or base amount, only phase 3b forms. Thus, the overall
concentration exerts a greater influence on the product forma-
tion than the base amount. Under higher overall concentrations,
the influence of stirring and base amount is observed. Phase 3a
crystallizes preferentially at concentrated conditions and low
3
3
5090
dx.doi.org/10.1021/ic200381f |Inorg. Chem. 2011, 50, 5085–5097

Inorganic Chemistry
ARTICLE
Figure 3. Crystallization diagram displaying the results of HT-investigations of the system Ni2þ/H3BTB/2-MeIm/DMF. Reactions were conducted at
150 °C for 72 h in a conventional oven. Left: discovery array showing the approximate formation field of 2. Right: focused array confirming the specific
formation field of 2. Results are based on XRPD measurements. Molar ratio values are normalized to 1.00. Details of the HT-experiment are given in
Supporting Information, Tables S9ꢀS10.
procedure, the added anions were incorporated into the structure
and lead to the phase selectivity. The addition of nitrate salts led
to the formation of the tetragonal framework, while sulfate salts
resulted in the formation of the trigonal framewꢀork structure.
2ꢀ
Therefore, we also investigated the role of NO3 and SO4
anions of the Ni2þ salt on the product formation (Figure 5D,
Supporting Information, Table S19). In the solvothermal HT-
MW experiment, the NO3 salt leads to better crystalline
ꢀ
products of 3a and 3b. Pure phase 3b could not be obtained
2ꢀ
starting from the SO4 salt. Thus the anion-templating effect
cannot be transferred to the Ni-system with solvothermal reac-
tion conditions.
Crystal Structure Description. The coordination of carbox-
ylate ions to metal ions can be categorized in three types
depending on the CꢀO M angle.62 Paddle-wheel units are
3 3 3
Figure 4. Crystallization diagram displaying the results of HT-investi-
gations of the system Ni(NO3)2/H2BDC/DABCO. Reactions were
conducted at 110 °C for 1 h in a conventional oven. Formation of 3b was
observed at molar ratios Ni(NO3)2/H2BDC = 1:1, 1.5:1, and 2:1
(dashed lines). These parameters were used in the HT-MW setup for
further investigations (Figure 5). Details of the HT-experiment are given
in Supporting Information, Table S15.
defined as two Mnþ ions which are bridged by four (COO)ꢀ ions
in syn-syn mode (Figure 6). In this work, four porous nickel(II)
based MOFs with paddle-wheel building units and two dense
non paddle-wheel containing compounds were synthesized.
Crystal structures of compounds 1a, 1b, and 2 were determined
via single crystal X-ray diffraction, and Rietveld refinement was
applied for the microcrystalline compounds 3a, 3b, and 3c.
Suitable starting structures were only available for 3a, 3b, and
for 3c, the structure solution was accomplished using in-house
XRPD data.
amounts of base and phase 3c forms exclusively at high overall
concentration, an excess of DABCO and under stirring. Similar
observations were also reported in the hydrothermal synthesis of
Al/Cr-MIL-101 and Al/Cr-MIL-53, in which the former prefers
dilute and more basic conditions whereas Al/Cr-MIL-53 requires
concentrated and acidic conditions.60,61 Recently Friscic et al.
have demonstrated the anion-templating effect of nitrate and
sulfate ions in the liquid assisted mechanochemical synthesis of
the analogue Zn-based polymorphs.50 Because of their synthesis
[Ni3(BTC)2(Me2NH)3] (DMF)4(H2O)4 (1a). Title compound
3
1a has a 3D framework built of Ni(II) paddle-wheel units and
tritopic BTC linker molecules. The paddle-wheel unit consists of
four carboxylate fragments in the equatorial positions and two
dimethylamine molecules in the axial positions (Figure 7,i). The
axial ligands were formed in situ through the partial hydrolysis of
5091
dx.doi.org/10.1021/ic200381f |Inorg. Chem. 2011, 50, 5085–5097

Inorganic Chemistry
ARTICLE
Figure 5. Results of HT-MW investigations of the system Ni(NO3)2/
H2BDC/DABCO/DMF. Under stirring, pure phase and highly crystal-
line 3b was synthesized using large amounts of DABCO (A). In B, the
effect of stirring on the product formation was studied with the optimal
reactant molar ratios for 3a (2:2:1:130) and 3b (1:1:4:260). In C, the
effect of the increasing overall concentration is shown for unstirred and
stirred reaction mixtures, respectively [1 Ni(NO3)2/1 H2BDC/x DAB-
CO/y DMF, with x = 0.5, 4 and y = 65ꢀ5200]. The influence of sulfate
and nitrate ions on the 3a/3b formation is shown in diagram D.
[(H2DABCO)(BDC)] is the salt formed from protonated DABCO
and benzenedicarboxylate ions.59
Figure 7. Paddle-wheel building units of (i) 1a with two axial dimethy-
lamine ligands, (ii) 2 with one axial water and one axial 2-methylimida-
zole ligand, (iii) 3a/3b with two axial 1,4-diazabicyclo[2.2.2]octane
ligands. Section of the crystal structures of 1a (iv), view along [100], ∼9
and ∼5 Å pores, 2 with two interwoven nets (v), view along [100], 16 Å
cavities, activated 3a (vi), view along [001], 7 Å along c axis and 4 Å pores
along the a and b axes, 3b (vii), view along [001] 14 Å and 4 Å pores
along the c axis and 3.5 Å pore windows along a and b axes. Square
pyramidal NiO5/NiO4N units are displayed as green polyhedra.
with Fm3m symmetry. The structure of 1a has two cages with
inner pore diameters of ∼9 and ∼5 Å, respectively (Figure 7,iv).
To the best of our knowledge, isostructural compounds have
only been reported with copper, iron, molybdenum, and
chromium.15,64ꢀ66 The pores are occupied by DMF and H2O
molecules, which can be removed by thermal activation and
renders the pores accessible to gas molecules.
[Ni3(BTB)2(2-MeIm)1.5(H2O)1.5] (DMF)9(H2O)6.5 (2). Com-
3
pound 2 is built of Ni(II) paddle-wheel units which are com-
posed of four carboxylate groups in equatorial positions and one
water molecule as well as one 2-methylimidazole (2-MeIm)
ligand in the axial positions. Similar to 1a, the paddle-wheel
units are connected to BTB linker molecules to construct an
extended 3D (3,4)-connected pto-a network.42,63 Because of
freely rotating phenyl rings around the central phenyl ring of
BTB linker (35° torsion angle), the symmetry of 2 is reduced to
Im3.67 In analogy to the isostructural Cu(II)-based compound
(MOF-14),42 title compound 2 has a 2-fold interwoven network
with a minimum displacement of 3 Å between the subframe-
works (Figure 7,v). Although large cavities of approximately
16 Å are observed, the pore opening is restricted to a window of
3 ꢁ 4 Å (without the axial water ligands). The cavities are
Figure 6. Three types of carboxylate ion coordination modes to metal
ions. Syn-bonding: 0ꢀ80°, direct bonding: 80ꢀ90°, and anti-bonding:
ꢀ90ꢀ0°. Thus, a paddle-wheel unit has a syn-syn bridging mode.62
DMF. Thus, each paddle-wheel unit contains two five-coordi-
nated Ni2þ ions with distorted square pyramidal polyhedra.
These units are connected to four carboxylate groups of the
BTC linker to form a 3D (3,4)-connected tbo-a63 framework
5092
dx.doi.org/10.1021/ic200381f |Inorg. Chem. 2011, 50, 5085–5097

Inorganic Chemistry
ARTICLE
occupied by DMF and water molecules. Upon thermal activation
the guest molecules are removed. The crystallinity of sample
increases drastically, and a structural transformation takes place
(Supporting Information, Figure S5). The new reflections ob-
served in the XRPD pattern of activated 2 can be explained by
using a larger isomorphic unit cell (2-ht) and same space group
undergo a solvent-mediated transformation to form the more
thermodynamically stable 3a.70 Microwaveassistedheatingisknown
to increase the nucleation rate in crystallization processes.71,72
Therefore, the results can be interpreted as follows: high solvent
content together with stirring and large amounts of DABCO,
microwave assisted heating leads exclusively to nuclei of 3b.
Because of the high nucleation rate, the reactant concentrations
drop below the critical nucleation concentration necessary for
the formation of 3a and thereby, only crystals of 3b are formed.
Under these reaction conditions, a solvent mediated transforma-
tion is not observed since no crystal seeds of 3a are present.
These findings highlight the importance of microwave assisted
heating compared to conventional methods in discovering new
phases, investigating selective synthesis of polymorphs and less
stable structures with larger pore sizes. In addition, this also
supports the fact that the new metalꢀorganic structures can be
still obtained with well established common organic linkers such
as terephthalic or trimesic acid by systematically trying out new
synthesis methods. Often metastable products of a reaction
system are overlooked because of inadequate understanding of
the kinetic and thermodynamic parameters.
Crystal structure of non paddle-wheel based compounds (1b,
3c): Compound [Ni6(BTC)2(DMF)6(HCOO)6] (1b) crystal-
lizes in a dense layered kgd-a structure with hexanuclear Ni6O32
clusters similar to the isostructural Mn2þ compound.39,63 Each
Ni2þ ion is six-coordinated by oxygen atoms from one DMF
molecule, three formate ions, and two BTC linkers to form
distorted NiO6 polyhedra. Six corner-sharing polyhedra form
hexanuclear cores which are linked by BTC linkers to construct a
2D layered network along the ab plane (Supporting Information,
Figure S11). The BTC linker molecules are all coordinated to the
Ni ions in a syn-mode, while the sterically less hindered formate
groups act as bridging ligands connecting three Ni2þ ions. Thus
coordination in syn- and anti-mode is observed.
Im3 with cell parameters 3 (a, b, c).68
3
[Ni2(BDC)2(DABCO] (DMF)4(H2O)1.5 (3a). Rietveld refine-
3
ment of the compound 3a was performed starting from the
structure model of the isostructural Zn-based compound as it
always crystallizes as a microcrystalline powder (Supporting
Information, Figure S1, Table 2, Supporting Information, Figure
S7).45 As seen in Figure 7,iii, the paddle-wheel unit is comprised
of four carboxylate groups and the DABCO ligands are located in
the axial positions. A 2D layer with a (4,4) sql63 grid is formed by
the connection of the paddle-wheel units with BDC linkers in the
ab plane. The layers are interconnected to form a pillared layered
3D structure via DABCO ligands along [001]. Two pore sizes of
∼7 (along c axis) and ∼4 Å (along a and b axes) are observed in
the microporous structure of activated 3a (Figure 7,vi). The as-
synthesized compound crystallizes in the space group I4/mcm
because of bent BDC linker molecules. Upon activation the
symmetry changes to P4/mmm since the BDC linkers are now
planar. This flexibility is caused by the presence and the type of
guest molecules residing in the pores.45
[Ni2(BDC)2(DABCO] (DMF)4(H2O)4 (3b). Title compounds
3
3b and 3a exhibit polymorphic framework structures, containing
different amounts of H2O guest molecules. Rietveld refinement of
the compound 3b was performed starting from the structure
model of the isostructural Zn-based compound as it always
crystallizes as a microcrystalline powder (Supporting Informa-
tion, Figure S1, Table 2, Supporting Information, Figure S8).48
The structure consists of (3,4)-connected kgm63 layers build by
the paddle-wheel units and linked by the BDC molecules. These
layers are interconnected by the DABCO molecules along the c
axis, and a pillared 3D network is obtained. Along the a and b axes
∼3.5 Å rectangular windows and along the c axis two types of
pores with trigonal (4.5 Å) and hexagonal (14 Å) shapes are
observed (Figure 7,vii). The pores are occupied by the DMF and
water molecules. These can be removed by thermal activation, but
under ambient conditions the structure decomposes gradually.
Whereas Zn2þ ions have a very flexible coordination environ-
ment, which can easily be distorted, Ni2þ ions exhibit preferen-
tially quadratic planar and quadratic pyramidal coordination
geometries. Because of the change from a (4,4) square grid in
3a to a (3,4) Kagomꢀe layer in 3b (Supporting Information,
Figure S10), ring strain is observed in the latter which could cause
its structural instability. The information can also be used in the
attempt to explain the formation of 3a and 3b under the different
reaction conditions. Under conventional heating 3b is preferen-
tially formed at short reactions times, but we were never able to
obtain a pure phase product. Only the use of microwave assisted
heating as well as the optimal reaction conditions of low overall
reaction mixture concentration, stirring, and excess base amounts
led to single phase 3b. By applying Ostwald’s rule of stages,69 the
structure 3b seems to be the kinetically preferred metastable
product. This is also supported by the fact that 3b has a lower
framework density compared to 3a (Fframework = 0.735/0.861
g cmꢀ3 for 3b and 3a respectively). Thus, during the product
formation nucleation of 3b should occur first. Under conven-
tional heating, nuclei of both polymorphs are formed that grow
to mixtures of 3a and 3b. Since crystals of 3a are present, 3b can
The structure of [Ni(BDC)(DABCO)] (DMF)1.5(H2O)2
3
(3c) was solved from XRPD data. Title compound 3c crystallizes
in a layered (4,4) rectangular-grid topology containing NiO4N2
polyhedra.73 Direct bonding of two carboxylate groups to the
Ni2þ ion is observed (Figure 6), which leads to the formation of
chains of alternating Ni2þ and BDC2ꢀ ions along the a axis
(Figure 8). These chains are connected along the b axis by the
DABCO molecules to form layers. DMF and water molecules are
located between these layers.
Sorption Measurements. Nitrogen sorption measurements
were performed at 77 K on the title compounds with paddle-wheel
building units (1a, 2, 3a, 3b) to examine their porosity (Figure 9).
The samples were first activated at 150 °C under vacuum for 12 h
to remove the guest molecules in the pores. The experimental
apparent specific surface area (BrunauerꢀEmmettꢀTeller (BET)
method) and micropore volume of 1a were determined to be
920 m2/g and 0.35 cm3/g respectively (Table 3). These values are
smaller than expected probably because of partial decom-
position of the structure or the incomplete activation. Different
synthesis methods and improved activation procedures could yield
higher porosity values. Isostructural copper-based HKUST-1 has
been reported with surface areas and micropore volumes of
692ꢀ1820 m2/g and 0.33ꢀ0.74 cm3/g, respectively (Supporting
Information, Table S21).
Compound 2 does not exhibit any porosity for nitrogen as
anticipated because of the restricted pore window size of 3 ꢁ 4 Å
(without the axial water ligands). On the other hand, the
isostructural copper-based MOF-14 shows surface area of
5093
dx.doi.org/10.1021/ic200381f |Inorg. Chem. 2011, 50, 5085–5097

Inorganic Chemistry
ARTICLE
Table 3. Specific Surface Areas and Micropore Volumes of
the Paddle-Wheel Based Ni(II) Compoundsa
micropore
volume/cm3 gꢀ1
specific surface area
SBET (obs.)/m2 gꢀ1
obs. at
calcd
(PLATON)29
compound
p/p0 = 0.5
1a
2
920
0
0.35
0
0.83
0.68
0.66
0.84
3a
3b
1807
1231
0.66
0.53
a Theoretical micropore volumes are calculated without the axial ligands.
Figure 8. Sections of the crystal structure of the layered compound 3c.
Alternating BDC linkers and direct-bonded Ni2þ ions form chains along
the a axis (bottom). Top left diagram showing DABCO molecules
connecting the chains along the b axis. The hydrogen bonding (broken
lines) between DMF and H2O molecules are shown in the top right
diagram in the ab plane.
Figure 10. Increasing adsorption of molecules with decreasing kinetic
diameters in compound 2. Measurements using H2, Ar, and N2 were
conducted at 77 K; H2O and CO2 at 298 K. The uptake depends
strongly on the kinetic diameter of the guest molecules employed. The
inset graph shows the hysteresis occurring during the desorption of H2O
and H2 (solid symbols: adsorption, hollow: desorption).
volume of the polymorph 3b with trigonal frameworks with the
theoretical value indicates the decomposition of the structure
which was confirmed via XRPD measurement. The comparison
of values obtained with isostructural compounds with other
transition metal ions are given in Supporting Information,
Table S21.
Chemical and Thermal Stability. High thermal stability and
inertness to air as well as moisture are some of the essential
requirements for the application of porous MOFs. For these
reasons, thermogravimetric (TG) analyses and XRPD measure-
ments of samples exposed to air as well as solvent-exchanged
samples were performed. The weight losses observed in the TG
experiments deviate in some cases from the results of the
elemental analyses. Slightly lower values in TG experiments are
found since the content of the pores and cages can vary because
of the gas flow at the beginning of the experiments. TG analysis
Figure 9. Nitrogen sorption isotherms measured at 77 K for the four
paddle-wheel based Ni2þ compounds 1, 2, 3a, and 3b. All compounds
with the exception of 2 exhibit microporous type I behavior. Solid
symbols denote adsorption, hollow symbols denote desorption.
1502 m2/g and a micropore volume of 0.53 cm3/g (Supporting
Information, Table S21).42 Nevertheless, sorption experiments
of 2 using other adsorbate molecules with smaller kinetic
diameters74 have shown increasing adsorption in the following
order: N2 < Ar < H2 (77 K) and CO2 < H2O (298 K) (Figure 10).
Substantial hysteresis was observed in the H2 and the H2O
sorption measurement. While for water as the adsorbent this is
often observed, hysteresis in H2 sorption measurements is rather
rare. Interestingly, almost no H2 molecules were liberated during
the desorption process indicating a kinetic trapping of the
adsorbates in the pores (Figure 10 inset graph).75 The highest
specific surface area was achieved with compound 3a, and the
experimental micropore volume is in good agreement with
calculated values (Table 3, Figure 9). This proves the robustness
and high stability of 3a. In comparison to the lower micropore
of 1a in nitrogen, [Ni3(BTC)2(Me2NH)3] (DMF)4(H2O)4,
3
revealed a gradual weight loss up to 320 °C which corresponds
to the removal of four H2O and three DMF molecules (obs.
26.9%, calcd 26.7%). The next significant weight loss up to
600 °C yielded elemental nickel (Supporting Information, Figure
S12). The temperature-dependent XRPD measurements in a
capillary (Supporting Information, Figure S12) confirm the
5094
dx.doi.org/10.1021/ic200381f |Inorg. Chem. 2011, 50, 5085–5097

Inorganic Chemistry
ARTICLE
thermal stability of 1a up to 300 °C. Upon exposure to ambient
conditions 1a decomposes rapidly at room temperature to form a
green X-ray amorphous product (Supporting Information, Fig-
ure S2). Solvent-exchange experiments with diverse organic
solvents such ethanol, dichloromethane, and acetonitrile have
shown a rapid disintegration of the crystal structure which
indicates the instability of compound 1a. The TG curve of
[Ni6(BTC)2(HCOO)6(DMF)6] 1b shows three weight losses.
The first weight loss (14.5%, calcd 14.9%) up to 250 °C could be
due to the loss of three coordinated DMF molecules. Between
250 and 350 °C a loss of 24.9% (calcd 24.0%) is observed which is
in accordance with the removal of the other three coordinated
DMF molecules and three formate ions (Supporting Informa-
tion, Figure S12). The solid residue after the third weight loss was
determined to be NiO.
aliphatic ν(CꢀH) bands are observed in the 2950ꢀ2850 cmꢀ1
For all compounds the aromatic CꢀH stretching bands could not
be seen because of overlap with the broad OꢀH bands. Two sharp
aromatic γ(CꢀH) bands of 1,3,5-BTC linker are observed at 770
( 5 cmꢀ1 and 725 ( 5 cmꢀ1 in 1a, and 1b.76 Three sharp aromatic
γ(CꢀH) deformation bands of BTB linker of 2 were seen at 857,
observed at 810 ( 5 cmꢀ1 and 750 ( 5 cmꢀ1 are characteristic and
correspond to the γ(CꢀH) and the δ(CꢀC) vibration of the 1,4-
BDC linker, respectively.78 The bands at 1090 ( 5 and 1055 (
5 cmꢀ1 could be due to the νas(CꢀN) and νs(CꢀN) vibrations of
DABCO.79 In all spectra, strong bands at 1670ꢀ1600 cmꢀ1 and
1400ꢀ1350 cmꢀ1 are observed which can be assigned to the
νas(CꢀO) and νs(CꢀO) vibrations of the ꢀCOOꢀ groups.
Around 1680 cmꢀ1 the band of the ν(CdO) vibration of DMF
molecules is detected.
.
783, and 705 cm .
ꢀ1 77 For compounds 3a, 3b, and 3c, the bands
Compound 2, [Ni3(BTB)2(2-MeIm)1.5(H2O)1.5] (DMF)9-
3
(H2O)6.5, with the BTB linker is stable in air after the thermal
activation, and its crystallinity even increases (Supporting Infor-
mation, Figure S5). Unlike compound 1a, the crystal structure of
2 remains intact even after 72 h of stirring in pure dichloro-
methane at room temperature. Upon heating during TG analysis,
compound 2 loses a considerable amount of guest molecules until
the first plateau at 300 °C is reached (Supporting Information,
Figure S12). The observed weight loss of 33.4% corresponds to
eight H2O and seven DMF molecules (calcd 33.2%). Above
350 °C the organic units decompose and NiO is formed. The
XRPD measurement of the sample treated at 250 °C confirms the
stability of the framework and the structural transformation to the
’ CONCLUSION
With the aid of high-throughput (HT) methods we have
identified the fields of formation of six new compounds. In
addition to four porous Ni based MOFs with paddle-wheel
building units, two compounds with dense layered structures
(1b, 3c) were also discovered. All compounds were fully char-
acterized. In the system containing H3BTC as starting material,
a comprehensive screening of compositional and process param-
eters allowed to establish the reaction conditions for the Ni
paddle-wheel unit. Application of these reaction conditions and
replacing H3BTC with the extended tritopic linker molecule
H3BTB, the paddle-wheel containing compound 2 was discov-
ered. Changing the tritopic to a ditopic linker molecule (H2BDC)
and adding DABCO as the base, the formation of two pillared
layered structures with the composition [Ni2(BDC)2(DABCO)]
was accomplished. The synthesis condition for each polymorph
was scrutinized by employing HT-methods under conventional
and microwave assisted heating. Rapid nucleation via microwave
assisted heating coupled with the systematic investigation has
been proven valuable for selectively obtaining the MOF 3b. On
the basis of the results of this study, the synthesis of other Ni
containing MOFs with paddle-wheel units should be feasible,
which have been described previously mainly with Cu2þ and
Zn2þ ions.
isomorphic subgroup with an enlarged unit cell of 3 (a, b, c).
3
During the thermal decomposition of 3a, [Ni2(BDC)2
(DABCO)] (DMF)4(H2O)1.5 (Supporting Information, Figure
3
S12), allguest molecules areliberated fromthe pores up to 290 °C
in a two step process (obs. 36.6%, calcd 36.4%). The substantial
weight loss from 290 to 800 °C can be attributed to the
decomposition of two BDC and one DABCO linker molecules
(obs. 48.6%, calcd 50.2%). The residual powder was determined
to be NiO by XRPD. Compound 3a is also stable against moisture
since after treatment with water, the crystallinity was preserved. 3a
has also been found to be tolerant againstsolvents such as ethanol,
dichloromethane, and acetonitrile in solvent exchange experi-
ments at room temperature. TG analysis of the compound 3b,
[Ni2(BDC)2(DABCO)] (DMF)4(H2O)4, which has a poly-
3
morphic framework, two weight loss steps because of the removal
of the guest molecules are observed (Supporting Information,
Figure S12). The first weight loss between 40ꢀ120 °C corre-
sponds to the removal of four H2O molecules (obs. 7.4%, calcd
7.8%), while the second weight loss (120ꢀ300 °C) can be
attributed to the removal of the four DMF molecules (obs.
29.6%, calcd 31.7%). Further heating up to 800 °C led to NiO.
Unlike 3a, 3b with the trigonal framework is moisture sensitive
and gradual decomposition upon storage at ambient conditions is
observed. This gradual decomposition of the 3b crystal structure
was also observed after solvent exchange experiments. The TG
’ ASSOCIATED CONTENT
S
Supporting Information. Exact amounts used for the
b
high-throughput synthesis, SEM and optical images, Rietveld
refinement results, XRPD patterns, TGA analyses and IR spectra
are available. This material is available free of charge via the
Internet at http://pubs.acs.org. The Cambridge Crystallographic
Data Center (CCDC) 802889ꢀ802894 contains the supple-
mentary crystallographic data for this paper. These data can be
obtained free of charge via the Internet at ww.ccdc.cam.ac.uk/
conts/retrieving.html (or from the CCDC, 12 Union Road,
Cambridge CB2 1EZ, U.K; fax, þ44 1223 36033; e-mail,
deposit@ccdc.ac.uk).
analysis of 3c, [Ni(BDC)(DABCO)] (DMF)1.5(H2O)2, shows
3
that the guest molecules are removed in two steps up to 350 °C
(obs. 29.2%, calcd 30.3%). Above 350 °C the oxidation of the
organic linker molecules takes place, and NiO is formed.
IR Spectroscopy. All compounds were characterized by IR
spectroscopy (Supporting Information, Figure S13, S14). Their IR
spectra exhibit broad bands in the region between 3600 and
3100 cmꢀ1, which are due to ν(OꢀH) vibrations of the uncoor-
dinated water molecules involved in weak hydrogen bonds. Weak
’ AUTHOR INFORMATION
Corresponding Author
*E-mail: stock@ac.uni-kiel.de. Phone: (þ49)431-880-1675. Fax:
(þ49)431-880-1775.
5095
dx.doi.org/10.1021/ic200381f |Inorg. Chem. 2011, 50, 5085–5097

Inorganic Chemistry
ARTICLE
’ ACKNOWLEDGMENT
(26) Lamberti, C.; Zecchina, A.; Groppo, E.; Bordiga, S. Chem. Soc.
Rev. 2010, 39, 4951–5001.
The authors thank Inke Jess (Christian-Albrechts-Universit€at
zu Kiel) and Dr. Alexandra Lieb (Otto-von-Guericke-Universit€at
Magdeburg) for the acquisition of single crystal data, Adam
Wutkowski (Christian-Albrechts-Universit€at zu Kiel) for TG
measurements, Steffen Schmidt (Ludwig-Maximilians-Universit€at
M€unchen) for high-resolution SEM images, Prof. Michael Fr€oba,
Ute Sazama, and Sandra Maracke (Universit€at Hamburg) for
H2-sorption measurements. The work has been supported by the
State of Schleswig-Holstein and the Deutsche Forschungsge-
meinschaft (DFG) through the priority program SPP 1362 “Porous
MetalꢀOrganic Frameworks” under Grant STO 643/5-1.
(27) Maniam, P.; N€ather, C.; Stock, N. Eur. J. Inorg. Chem.
2010, 3866–3874.
(28) XRED version 1.19, XSHAPE version 1.06; Stoe & Cie GmbH:
Darmstadt, Germany, 1999.
(29) Spek, A. L. PLATON; a multipurpose crystallographic tool;
Utrecht University: Utrecht, The Netherlands, 2001.
(30) Sheldrick, G. M. SADABS version; Bruker AXS Inc., Madison,
WI, 2007.
(31) Sheldrick, G. M. Acta Crystallogr. 2008, A64, 112–122.
(32) Altomare, A.; Camalli, M.; Cuocci, C.; Giacovazzo, C.; Moliterni,
A.; Rizzi, R. J. Appl. Crystallogr. 2009, 42, 1197–1202.
(33) Materials Studio, v5.0; Accelrys Software Inc.: San Diego, CA,
2009.
(34) Favre-Nicolin, V.; Cerny, R. J. Appl. Crystallogr. 2002, 35,
734–743.
(35) Coelho, A. A. TOPAS-Academic; Program for indexing, structure
resolution and Rietveld refinement on powder data, V4.1; Bruker AXS Inc.,
Madison, WI, 2007.
(36) Bauer, S.; Bein, T.; Stock, N. Inorg. Chem. 2005, 44, 5882–5889.
(37) Biemmi, E.; Christian, S.; Stock, N.; Bein, T. Microporous
Mesoporous Mater. 2009, 117, 111–117.
(38) Sonnauer, A.; Stock, N. Eur. J. Inorg. Chem. 2008, 5038–5045.
(39) Chen, J.; Ohba, M.; Kitagawa, S. Chem. Lett. 2006, 35, 526–527.
(40) Biemmi, E.; Bein, T.; Stock, N. Solid State Sci. 2006, 8, 363–370.
(41) Forster, P. M.; Burbank, A. R.; Livage, C.; Fꢀerey, G.; Cheetham,
A. K. Chem. Commun. 2004, 368–369.
(42) Chen, B.; Eddaoudi, M.; Hyde, S. T.; O’Keeffe, M.; Yaghi, O. M.
Science 2001, 291, 1021–1023.
(43) Takei, T.; Ii, T.; Kawashima, J.; Ohmura, T.; Ichikawa, M.;
Hosoe, M.; Shinya, Y.; Kanoya, I.; Mori, W. Chem. Lett. 2007, 36,
1136–1137.
(44) Seki, K.; Takamizawa, S.; Mori, W. Chem. Lett. 2001, 30,
332–333.
(45) Dybtsev, D. N.; Chun, H.; Kim, K. Angew. Chem., Int. Ed. 2004,
43, 5033–5036. Angew. Chem. 2004, 116, 5143–5146.
(46) Chun, H.; Dybtsev, D. N.; Kim, H.; Kim, K. Chem.—Eur. J.
2005, 11, 3521–3529.
(47) Syozi, I. Prog. Theor. Phys. 1951, 6, 306–308.
(48) Chun, H.; Moon, J. Inorg. Chem. 2007, 46, 4371–4373.
(49) Kondo, M.; Takashima, Y.; Seo, J.; Kitagawa, S.; Furukawa, S.
CrystEngComm 2010, 12, 2350–2353.
(50) Friscic, T.; Reid, D. G.; Halasz, I.; Stein, R. S.; Dinnebier, R. E.;
Duer, M. J. Angew. Chem., Int. Ed. 2010, 49, 712–715. Angew. Chem.
2010, 122, 724–727.
’ REFERENCES
(1) Koinuma, H.; Takeuchi, I. Nat. Mater. 2004, 3, 429–438.
(2) Maier, W. F.; St€owe, K.; Sieg, S. Angew. Chem., Int. Ed. 2007,
46, 6016. Angew. Chem. 2007, 119, 6122–6179.
(3) Stock, N. Microporous Mesoporous Mater. 2010, 129, 287–295.
(4) Stock, N.; Bein, T. Solid State Sci. 2003, 5, 1207–1210.
(5) Akporiaye, D. E.; Dahl, I. M.; Karlsson, A.; Wendelbo, R. Angew.
Chem., Int. Ed. 1998, 37, 609–611. Angew. Chem. 1998, 110, 629–631.
(6) Stock, N.; Bein, T. Angew. Chem., Int. Ed. 2004, 43, 749–752.
Angew. Chem. 2004, 116, 767–770.
(7) Forster, P. M.; Burbank, A. R.; O’Sullivan, M. C.; Guillou, N.;
Livage, C.; Fꢀerey, G.; Stock, N.; Cheetham, A. K. Solid State Sci. 2005, 7,
1549–1555.
(8) Banerjee, R.; Phan, A.; Wang, B.; Knobler, C.; Furukawa, H.;
O’Keeffe, M.; Yaghi, O. M. Science 2008, 319, 939–943.
(9) Bauer, S.; Stock, N. Chem. Unserer Zeit 2007, 41, 390–398.
(10) Mueller, U.; Schubert, M.; Teich, F.; Puetter, H.; Schierle-
Arndt, K.; Pastrꢀe, J. J. Mater. Chem. 2006, 16, 626–636.
(11) Horcajada, P.; Serre, C.; Vallet-Regi, M.; Sebban, M.; Taulelle,
F.; Fꢀerey, G. Angew. Chem., Int. Ed. 2006, 45, 5974–5978. Angew. Chem.
2006, 118, 6120–6124.
(12) Sonnauer, A.; Hoffmann, F.; Fr€oba, M.; Kienle, L.; Duppel, V.;
Thommes, M.; Serre, C.; Fꢀerey, G.; Stock, N. Angew. Chem., Int. Ed.
2009, 48, 3791–3794. Angew. Chem. 2009, 121, 3849–3852.
(13) Bauer, S.; Serre, C.; Devic, T.; Horcajada, P.; Marrot, J.; Fꢀerey,
G.; Stock, N. Inorg. Chem. 2008, 47, 7568–7576.
(14) Ahnfeldt, T.; Gunzelmann, D.; Loiseau, T.; Hirsemann, D.;
Senker, J.; Ferey, G.; Stock, N. Inorg. Chem. 2009, 48, 3057–3064.
(15) Chui, S. S.-Y.; Lo, S. M.-F.; Charmant, J. P. H.; Guy Orpen, A.;
Williams, I. D. Science 1999, 283, 1148–1150.
(16) Allen, F. H. Acta Crystallogr. 2002, B58, 380ꢀ388. Databank
search in CSD Conquest version 5.31 with Feb 2010 update.
(17) Lee, S. W.; Kim, H. J.; Lee, Y. K.; Park, K.; Son, J.-H.; Kwon, Y.-
U. Inorg. Chim. Acta 2003, 353, 151–158.
(51) Dunitz, J. D.; Bernstein, J. Acc. Chem. Res. 1995, 28, 193–200.
(52) Herrero, M. A.; Kremsner, J. M.; Kappe, C. O. J. Org. Chem.
2008, 73, 36–47.
(53) Sonnauer, A.; Stock, N. J. Solid State Chem. 2008, 181,
3065–3070.
(18) Yuen, T.; Danilovic, D.; Li, K.; Li, J. J. Appl. Phys. 2008,
103, 07B725.
(19) Shi, Q.; Sun, Y.; Sheng, L.; Ma, K.; Cai, X.; Liu, D. Inorg. Chim.
Acta 2009, 362, 4167–4173.
(54) Jhung, S. H.; Chang, J.-S.; Hwang, J. S.; Park, S.-E. Microporous
Mesoporous Mater. 2003, 64, 33–39.
(55) Jhung, S. H.; Lee, J. H.; Forster, P. M.; Fꢀerey, G.; Cheetham,
A. K.; Chang, J.-S. Chem.—Eur. J. 2006, 12, 7899–7905.
(56) Kondepudi, D. K.; Kaufman, R. J.; Singh, N. Science 1990, 250,
975–976.
(20) Zhu, L.; Chen, X.; Zhao, Q.; Li, Z.; Zhang, X.; Sun, B. Z. Anorg.
Allg. Chem. 2010, 636, 1441–1443.
(21) Arstad, B.; Fjellvag, H.; Kongshaug, K. O.; Swang, O.; Blom, R.
ꢂ
Adsorption 2008, 14, 755–762.
(57) Mcbride, J. M.; Carter, R. L. Angew. Chem., Int. Ed. Engl. 1991,
30, 293–295. Angew. Chem. 1991, 103, 298–300.
(58) Hanif, N.; Anderson, M. W.; Alfredsson, V.; Terasaki, O. Phys.
Chem. Chem. Phys. 2000, 2, 3349–3357.
(59) Yang, E.; Song, X.-C.; Zhu, J.-W. Acta Crystallogr., Sect. E 2008,
64, o1764.
(60) Ahnfeldt, T.; Guillou, N.; Gunzelmann, D.; Margiolaki, I.;
Loiseau, T.; Fꢁerey, G.; Senker, J.; Stock, N. Angew. Chem., Int. Ed.
2009, 48, 5163–5166. Angew. Chem. 2009, 121, 5265–5268.
(61) Khan, N. A.; Jun, J. W.; Jhung, S. H. Eur. J. Inorg. Chem.
2010, 1043–1048.
(22) Klein, N.; Herzog, C.; Sabo, M.; Senkovska, I.; Getzschmann, J.;
Paasch, S.; Lohe, M. R.; Brunner, E.; Kaskel, S. Phys. Chem. Chem. Phys.
2010, 12, 11778–11784.
(23) Guillou, N.; Gao, Q.; Forster, P. M.; Chang, J.-S.; Noguꢁes, M.;
Park, S.-E.; Fꢁerey, G.; Cheetham, A. K. Angew. Chem., Int. Ed. 2001,
40, 2831–2834. Angew. Chem. 2001, 113, 2913–2916.
(24) Zou, R.-Q.; Sakurai, H.; Xu, Q. Angew. Chem., Int. Ed. 2006,
45, 2542–2546. Angew. Chem. 2006, 118, 2604–2608.
(25) Forster, P. M.; Eckert, J.; Heiken, B. D.; Parise, J. B.; Yoon,
J. W.; Jhung, S. H.; Chang, J.-S.; Cheetham, A. K. J. Am. Chem. Soc.
2006, 128, 16846–16850.
5096
dx.doi.org/10.1021/ic200381f |Inorg. Chem. 2011, 50, 5085–5097

Inorganic Chemistry
ARTICLE
(62) Carrell, C. J.; Carrell, H. L.; Erlebacher, J.; Glusker, J. P. J. Am.
Chem. Soc. 1988, 110, 8651–8656.
(63) O’Keeffe, M.; Peskov, M. A.; Ramsden, S. J.; Yaghi, O. M. Acc.
Chem. Res. 2008, 41, 1782–1789.
(64) Xie, L.; Liu, S.; Gao, C.; Cao, R.; Cao, J.; Sun, C.; Su, Z. Inorg.
Chem. 2007, 46, 7782–7788.
(65) Kramer, M.; Schwarz, U.; Kaskel, S. J. Mater. Chem. 2006, 16,
2245–2248.
(66) Murray, L. J.; Dinca, M.; Yano, J.; Chavan, S.; Bordiga, S.;
Brown, C. M.; Long, J. R. J. Am. Chem. Soc. 2010, 132, 7856–7857.
(67) Ma, S.; Sun, D.; Ambrogio, M.; Fillinger, J. A.; Parkin, S.; Zhou,
H.-C. J. Am. Chem. Soc. 2007, 129, 1858–1859.
(68) Wondratschek, H.; M€uller, U. International Tables for Crystal-
lography; Kluwer Academic Publishers: Dordrecht, The Netherlands,
2006, Vol. A1, p 689.
(69) Ostwald, W. Z. Phys. Chem. 1897, 22, 289–330.
(70) Davey, R. J.; Blagden, N.; Potts, G. D.; Docherty, R. J. Am.
Chem. Soc. 1997, 119, 1767–1772.
(71) Gharibeh, M.; Tompsett, G. A.; Conner, W. C.; Yngvesson,
K. S. ChemPhysChem 2008, 9, 2580–2591.
(72) Khan, N. A.; Haque, E.; Hwa Jhung, S. Phys. Chem. Chem. Phys.
2010, 12, 2625–2631.
(73) Natarajan, S.; Mahata, P. Chem. Soc. Rev. 2009, 38, 2304–2318.
(74) Li, J.-R.; Kuppler, R. J.; Zhou, H.-C. Chem. Soc. Rev. 2009, 38,
1477–1504.
(75) Zhao, X.; Xiao, B.; Fletcher, A. J.; Thomas, K. M.; Bradshaw, D.;
Rosseinsky, M. J. Science 2004, 306, 1012–1015.
(76) Green, J. H. S.; Harrison, D. J.; Kynaston, W. Spectrochim. Acta,
Part A 1971, 27, 793–806.
(77) Jutand, A.; Negri, S. Eur. J. Inorg. Chem. 1998, 9, 1811–1821.
(78) Arenas, J. F.; Marcos, J. I. Spectrochim. Acta, Part A 1980,
36, 1075–1081.
(79) McDivitt, J. R.; Humphrey, G. L. Spectrochim. Acta, Part A 1974,
30, 1021–1033.
5097
dx.doi.org/10.1021/ic200381f |Inorg. Chem. 2011, 50, 5085–5097