Documente Academic
Documente Profesional
Documente Cultură
Northwestern University.
3866
of the struts13 and (b) are largely shielded from direct interaction
with H2, exerting their effects instead by facilitating favorable
displacement of catenated frameworks. With these problems in
mind, we sought an alternative approach that would avoid catenation
and anchor ions far from carboxylates or nodes.
Catenation was addressed by using the recently developed octaoxygen ligand, L1.17 In contrast to most other carboxylate ligands
used in pillared-paddlewheel structures, L1 often yields noncatenated structures. As Figure 1 shows, the combination of L1 with
a Zn(II) source and the diol-containing strut, L2, gives, after 2 days
of heating, colorless block crystals of 1, which we have termed
DO-MOF. Single-crystal X-ray measurements confirmed that 1
consists of only a single network containing large cavities with
readily accessible alcohol functionalities. Application of the
SQUEEZE routine in PLATON revealed a remarkable 76% solventaccessible void volume.18 Characterization by thermal gravimetric
COMMUNICATIONS
Table 1. Summary of Adsorption Properties of 1 and 1-M
material
M/Zn2a
BET surface
area (m2/g)
pore volume
(cm3/g)c
H2 uptake (wt %)
at 1 atm, 77 K
Qst (kJ/mol)
at 0-1 atm
solvent/Me
1 (DO-MOF)
1-Li0.20
1-Li2.62d
1-Mg0.86
1-Mg2.02
H+
Li+
Li+
Mg2+
Mg2+
2b
0.20 ( 0.01
2.62 ( 0.05
0.86 ( 0.02
2.02 ( 0.02
810
840
270
820
510
0.35
0.46
0.20
0.40
0.29
1.23
1.32
0.77
1.16
1.01
6.3-4.7
6.3-6.6
5.6-0.5
6.2-6.9
7.3-5.0
n/a
0.40f
0.13f
0.08g
1.10h
a
Determined via inductively coupled plasma analysis of dissolved samples. b From the crystal structure. c Measured at P/Po 0.95. d Sample showed
substantial loss of crystallinity, suggesting degradation. e Total number of residual solvent molecules per M atom, as determined by 1H NMR analysis of
the evacuated material (see the SI). f DMF + THF. g DMF + THF + (methanol or methoxide). h 0.19 DMF + 0.04 THF + 0.88 (methanol or
methoxide).
3867
COMMUNICATIONS
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
(13)
(14)
(15)
(16)
(17)
(18)
(19)
(20)
(1) (a) Ferey, G. Chem. Soc. ReV. 2008, 37, 191214. (b) Collins, D. J.; Zhou,
H.-C. J. Mater. Chem. 2007, 17, 31543160. (c) Lin, X.; Jia, J.; Hubberstey,
(22)
(23)
(24)
(25)
References
3868
(21)
JA809954R
S1
S2
Synthesis of 1.
Single crystals of 1 were obtained upon heating Zn(NO3)26H2O (50 mg, 0.17 mmol), L1 (100
mg, 0.18 mmol), and L2 (25 mg, 0.12 mmol) in 25 ml DMF plus one drop of concentrated HCl.
This solution was divided equally between five 4 dram screw cap vials and heated to 80oC for 3
days at which time clear colorless block crystals had formed. Large scale materials were
obtained by simply scaling up the single crystal preparation. A sample preparation follows:
Zn2(NO3)6H2O (100 mg, 0.34 mmol), L1 (200 mg, 0.36 mmol), L2 (50 mg, 0.24 mmol), and 50
ml DMF were added to a 50 ml erlenmeyer flask and sonicated to disperse. After about 10
minutes, 2 drops concentrated HCl was added to the suspension and sonicated for an additional 5
minutes upon which time the suspension dissolved to give a clear pale yellow solution. The
solution was then divided equally among ten 4 dram screw cap vials and placed in an 80oC oven
for three days. The clear colorless crystals were removed from the flask, isolated via filtration,
washed with DMF, and allowed to dry in air. Zn2(L1)(L2) (1) recovered: ~30 mg, 5% yield
based on zinc. Anal. calcd. for 12H2O, C46H34N2O12Zn2: C, 58.93; H, 3.63; N, 2.99. Found: C,
58.78; H, 3.10; N, 1.93. Phase purity was verified by PXRD and TGA (Figure S1).
S3
Crystal structure of 1.
Single crystals of 1 were mounted on a BRUKER APEX2 V2.1-0 diffractometer equipped
with a graphite-monochromated MoKa ( = 0.71073 ) radiation source in a cold nitrogen
stream. All crystallographic data were corrected for Lorentz and polarization effects (SAINT).
Data diffracted out to 0.84 Angstroms but fit poorly at low resolution suggesting limitations in
the proposed model. The structures were solved by direct methods and refined by the full-matrix
least-squares method on F2 with appropriate software implemented in the SHELXTL program
package. Most of the guest DMF solvent molecules within the pores are severely disordered,
which hindered satisfactory development of the model; therefore, the SQUEEZE routine in
PLATON2 was applied to remove the contributions of electron density from disordered solvent
molecules. The outputs from the SQUEEZE calculations are attached to the CIF file. All of the
non-hydrogen atoms were refined anisotropically. Heavy restraints (esd 0.001) on the bond
distances and angles were used on L2 to idealize the pyridyl ring and to stabilize the solution of
1. Rigid-bond restraints (esd 0.001) were imposed on displacement parameters for the DPG
ligand similar amplitudes (esd 0.001) were imposed on displacement parameters for the sites on
the ligand separated by less than van der Waals radii. All efforts to reduce symmetry to avoid
twofold and mirror disorders failed.
S4
128DMF
empirical formula a
C65H113N15O19Zn
formula weight
1474.07
colorless, tabular
orthorhombic
space group
Pmmm
a ()
11.3014 (11)
b ()
15.7553 (13)
c ()
15.7553(13)
a (deg)
90
b (deg)
90
g (deg)
90
V (3)
2805.3(4)
(calcd, g/cm3)
1.745
(mm-1)
0.541
goodness-of-fit on F2
0.793
Rb
0.1078
Rw
a
0.2575
The SQUEEZE routine in PLATON was employed to mask diffuse electron density in the cavities due to
disordered solvent (DMF) molecules. The outputs from the SQUEEZE routine gave 28 DMF solvent molecules per
unit cell; these have been added to the unit cell formula. bR(F) = (Fo Fc)/Fo. cRw(Fo2) = [[w(Fo2
Fc2)2]/wFo4]1/2.
S5
normalized intensity
as synthesized
simulation
10
15
20
25
30
35
two theta
1A
100
80
% mass
54.5%
60
40
20
0
100
200
300
400
500
o
temperature ( C)
1B
S6
600
700
Alkoxide formation.
In general, to perform the proton exchange and create the alkoxide frameworks, crystalline
samples of 1 were submerged in THF for 1 day to replace the pore-filling DMF. The THFexchanged samples were then stirred vigorously in dry THF/CH3CN with a metal source
overnight. The solid was then isolated by filtration, washed with copius amounts of THF to
remove any weakly physisorbed ions, dried briefly in air, and quickly transferred to an
adsorption sample tube. This procedure was tested on the L2 ligand alone and the complete
removal of the hydroxyl proton verified by 1H NMR, Figure S2.
Stoichiometric loading is defined here as the amount of alkoxide formation per L2 strut. The
crystal structure of 1 gives 1 L2 strut per Zn2 cluster, stoichiometric loading here is 2 Li(I) per 2
Zn(II). Specific preparation procedures follow:
Synthesis of 1-Li0.20: 1 (~10 mg) was added to a small scintillation vial, covered with 3 ml
CH3CN, and stirred gently with 2 ml LiOtBu solution (1.0M THF solution) overnight. The solid
was washed with THF several times and isolated via filtration. ICP analysis for Li/Zn returned
0.20 0.01 Li / Zn2 unit.
Synthesis of 1-Li2.62: 1 (13 mg) was added to a 100 ml flask, covered with 20 ml THF, and
stirred vigorously. Lithium-tert-butoxide (4 mg, 0.05 mmol) was added to the flask and stirred at
room temperature overnight. ICP analysis for Li/Zn returned 2.62 0.05 Li / Zn2 unit.
Synthesis of 1-Mg0.86: 1 (~30 mg) was added to a small conical vial, covered with 3 ml THF
and stirred vigorously to create a strong dispersion. Mg(OMe)2 (60 l, 6-10% solution in
MeOH) was added to the vial, which was then capped and allowed to stir overnight. ICP
analysis for Mg/Zn returned 0.86 0.02 Mg / Zn2 unit.
Synthesis of 1-Mg2.02: 1 (~30 mg) was added to a small conical vial, covered with 5 ml THF
and stirred vigorously to create a strong dispersion. Mg(OMe)2 (200 l, 6-10% solution in
MeOH) was added to the vial, which was then capped and allowed to stir overnight. ICP
analysis for Mg/Zn returned 2.02 0.02 Mg / Zn2 unit.
S7
Figure S2. 500 MHz 1H NMR in DMSO-d6 of L2 (bottom, black) and L2-2Li+ (top, blue).
Hydroxyl protons at 5.7 ppm are not present in spectrum of alkoxide-functionalized ligand.
S8
Figure S3. N2 adsorption isotherms of 1 and 1-Li (A) and 1 and 1-Mg (B). Closed symbols,
adsorption; open symbols, desorption.
400
1
1-Li0.20
350
1-Li2.62
450
300
250
200
150
100
50
0
0.0
0.2
0.4
0.6
0.8
1.0
P / Po
3A
300
400
200
100
1
1-Mg0.86
1-Mg2.02
0
0.0
0.2
0.4
0.6
P / Po
3B
S9
0.8
1.0
Figure S4. 77K and 87K H2 adsorption isotherms of 1 (A), 1-Li0.20 (B), 1-Li2.62 (C), 1-Mg0.86
(D), 1-Mg2.02 (E). Closed symbols, adsorption; open symbols, desorption.
1.4
1.2
1.0
wt %
0.8
0.6
0.4
77K
87K
0.2
0.0
0.0
0.2
0.4
0.6
0.8
1.0
P (atm)
4A
1.4
1.2
wt %
1.0
0.8
0.6
0.4
77K
87K
0.2
0.0
0.0
0.2
0.4
0.6
P (atm)
4B
S10
0.8
1.0
0.8
wt %
0.6
0.4
0.2
77K
87K
0.0
0.0
0.2
0.4
0.6
0.8
1.0
P (atm)
4C
1.2
1.0
wt %
0.8
0.6
0.4
0.2
77K
87K
0.0
0.0
0.2
0.4
0.6
P (atm)
4D
S11
0.8
1.0
1.0
wt %
0.8
0.6
0.4
0.2
77K
87K
0.0
0.0
0.2
0.4
0.6
P (atm)
4E
S12
0.8
1.0
Figure S5. Details of fitting and calculation of isosteric heat of adsorption for 1(A), 1-Li0.20 (B),
1-Li2.62 (C), 1-Mg0.86 (D), and 1-Mg2.02 (E).
Both hydrogen isotherms (77K, 87K) for each material were fit to a virial equation of the form
given in Equation 1.3 The heat of adsorption is then calculated from the fitting parameters using
Equation 2. The isotherms and fitting parameters are shown in Figure S5. Red lines are virial
equation fit to data.
ln p = ln N +
1
T
a N + b N
i
(1)
i=0
i=0
m
qst ( N ) = R ai N i
(2)
i =0
77K
87K
6
ln P (torr)
a0 = -756.02427 31.15921
a1 = 67.82439 13.37311
a2 = -19.05574 7.38794
a3 = 3.13151 1.54678
a4 = -0.2381 0.13837
a5 = 0.00673 0.00444
b0 = 11.06465 0.37528
-2
0
N, mg/g
5A
S13
10
12
14
77K
87K
ln P (torr)
a0 = -780.62964 67.40931
a1 = 27.0409 13.07371
a2 = -7.94879 3.10377
a3 = 0.76295 0.35668
a4 = -0.02531 0.01318
b0 = 10.91304 0.8191
b1 = 0.2849 0.12696
-2
0
10
12
14
N, mg/g
5B
77K
87K
ln P (torr)
a0 = -669.12954 80.23388
a1 = 353.43436 56.88392
a2 = -247.5793 65.36447
a3 = 103.6587 32.66379
a4 = -21.82458 7.72871
a5 = 2.23554 0.86588
a6 = -0.08875 0.03696
b0 = 10.37112 0.9703
b1 = -0.74008 0.23481
-2
0
N, mg/g
5C
S14
8
77K
87K
ln P (torr)
a0 = -772.48033 64.86774
a1 = 52.73039 21.37485
a2 = -28.33613 9.49021
a3 = 5.21312 2.11209
a4 = -0.43229 0.20039
a5 = 0.01318 0.00682
b0 = 11.24135 0.78551
b1 = 0.39617 0.16078
-2
0
10
12
N, mg/g
5D
77K
87K
ln P (torr)
a0 = -874.85671 23.15925
a1 = 44.33304 5.80591
a2 = -4.21387 1.77951
a3 = 0.58996 0.26291
a4 = -0.02792 0.01262
b0 = 12.33106 0.28064
b1 = -0.09992 0.04975
-2
0
N, mg/g
5E
S15
10
Figure S6. Isosteric H2 heat of adsorption of 1 and 1-Li (A) and 1 and 1-Mg (B).
8
7
Qst, kJ/mol
6
5
4
3
2
1
1-Li0.20
1-Li2.62
0
0
10
N, mg/g
6A
8
7
Qst, kJ/mol
6
5
4
3
2
1
1-Mg0.86
1-Mg2.02
0
0
N, mg/g
6B
S16
10
Figure S7. 1H NMR analysis of evacuated samples to quantify residual solvent content of 1-Li
and 1-Mg.
Following complete evacuation and N2 and H2 adsorption measurements, samples of 1, 1-Li and
1-Mg were dissolved in D2SO4 (96-98 wt% in D2O, 99.5 atom% D) in order to detect residual
solvent in the pores or coordinated to Li+ and Mg2+. For consistency, the spectra were referenced
to the first solvent peak of DMF ( = 3.01 ppm in H2O). The proton peaks for DMF (two -CH3
groups, 6H), THF (only one of two peaks, 4H), t-butanol (1-Li samples, 9H) and methanol (1Mg samples, 3H) were integrated and referenced to proton peaks for the dissolved MOF ligands
(specifically those of the -protons to the hydroxyl group of L2). Figure S7 shows the NMR
spectra for each sample. Table S2 outlines the calculations. See also reference 16 in the text for
previous use of this method.
1-Li0.20
1-Li2.62
1-Mg0.86
1-Mg2.02
9.0
8.5
8.0
7.5
7.0
6.5
6.0
5.5
5.0
4.5
S17
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.20
2.62
0.86
2.02
L2 proton
integration
20
20
20
20
M/Zn2
# L2
ligands
10
10
10
10
# DMF / M
DMF
integration
2.6
17.8
2.2
22.6
# THF / M
#
DMF
0.43
2.97
0.37
3.77
THF
integration
1.2
1.3
0.7
2.9
# R-OH / M
0.20
0.22
0.15
1-Li0.20
2.62
0.11
0.01
1-Li2.62
0.86
0.04
0.02
1-Mg0.86
2.02
0.19
0.04
1-Mg2.02
[a] R-OH (1-Li) = t-butanol, R-OH (1-Mg) = methanol.
0
0.00
0.02
0.88
S18
#
THF
0.30
0.33
0.18
0.73
total # of solvent
molecules / M
0.37
0.12
0.08
1.10
R-OH
integration[a]
0
0.6
0.4
53.1
#
R-OH
0
0.67
0.13
17.7
total residual
mass loss (%)
0.28
0.10
0.05
0.48
References
1) Farha, O. K.; Mulfort, K. L.; Hupp, J. T., Inorg. Chem. 2008, 47, 10223-10225.
2) Spek, A. L., J. Appl. Crystallogr. 2003, 36, 7-13.
3) Czepirski, L.; Jagiello, J., Chemical Engineering Science 1989, 44, 797-801.
S19