Ligand-elaboration as a strategy for engendering structural diversity in porous metal-organic framework compounds

Article abstract of DOI:10.1039/b714160b

A series of 4,4′-ethynylenedibenzoic acids were synthesized and used in the construction of Zn-based, mixed-ligand metal-organic frameworks; through variation of functionality in the 3- and 3′-positions of these linkers, a collection of MOFs with differing connectivities and varying levels of interpenetration was obtained. The Royal Society of Chemistry.

Full text of DOI:10.1039/b714160b

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Ligand-elaboration as a strategy for engendering structural diversity
in porous metal–organic framework compoundsw
Tendai Gadzikwa, Bi-Shun Zeng, Joseph T. Hupp* and SonBinh T. Nguyen*
Received (in Berkeley, CA, USA) 13th September 2007, Accepted 24th April 2008
First published as an Advance Article on the web 30th June 2008
DOI: 10.1039/b714160b
A series of 4,40-ethynylenedibenzoic acids were synthesized and
used in the construction of Zn-based, mixed-ligand metal–organic
frameworks; through variation of functionality in the 3- and 30-
positions of these linkers, a collection of MOFs with differing
connectivities and varying levels of interpenetration was obtained.
which would eventually be employed in the synthesis of Zn
paddlewheel MOFs.15–17 Taking guidance from previous work,5,15
we focused on the synthesis of anisotropic MOFs from bis(aryl
carboxylic acid)ethyne and BIPY classes of ligands. MOFs ob-
tained from dipyridyl and dicarboxylic acids incorporate fewer
metal centers in the clusters that comprise the MOF vertices and
are generally smaller than the analogous Zn4O clusters found in
cubic isoreticular MOFs. Together with the relatively long lengths
of the BIPY and bis(aryl carboxylic acid)ethyne struts, the use of
slightly smaller clusters should facilitate the interpenetration of up
to several independent networks.
The chemistry of metal–organic frameworks (MOFs) has
matured to a stage where, together with the exploratory
synthesis and characterization of new materials, there is a
growing drive towards designing and synthesizing new materi-
als specifically to study the influence of chemical functionality,
catenation, and coordination environment on various MOF
applications, e.g., gas uptake, separations, and catalysis.1–4
Investigations of this kind become more precise with an
increased ability to hold constant as many variables as possible,
while systematically changing only the property under exam-
ination.5–7 The level of network interpenetration in MOFs has
been suggested to play a very important role in many sorption
and separation applications, due to its direct impact on pore
sizes as well as internal surface area and guest-accessible
volume.8,9 As network interpenetration can be strongly influ-
enced by interactions between the struts in neighboring net-
works, varying the sterics, chemical functionality and van der
Waals surface areas of the ligands might allow one to manip-
ulate catenation. Surprisingly, this has not yet been demon-
strated to date, although small changes to the substituents on a
parent ligand have been shown to affect topology10,11 and
chirality.12 In this communication, we report a strategy for
diversifying MOFs in which minor changes in the organic
ligands lead to significant changes in both the connectivities
and the levels of catenation in the resulting structures.
We found the one-pot synthesis of bisarylethynes via
Sonogashira coupling to be invaluable here, as it is reliable,
high-yielding, and tolerant to a wide range of functional
groups.18 This last property is particularly useful, as it allows
one to efficiently create different functional environments
within MOFs possessing similar struts. Additionally, and most
notably, we are able to investigate how small changes in
ligands, at points remote from their coordinating groups, lead
to significant changes in the topologies and levels of inter-
penetration of the resulting MOFs. To this end, a series of
4,40-(1,2-ethynediyl)bis(benzoic acid)s (1a–d, Fig. 1) was
synthesized and applied to the construction of anisotropic
MOFs A–D. These were formed in good yields upon heating
dimethylformamide (DMF) solutions of Zn(NO3)2ꢀ6H2O, the
appropriate diacid, and 4,40-dipyridyl (BIPY) in sealed vials.
We were surprised to obtain the four-fold interpenetrating
diamond-like structure A (Fig. 2) from 1a, as several two-fold
interpenetrating Zn paddle-wheel MOFs15 have been synthe-
sized in our labs under the same conditions. Despite having
only a single Zn atom complexing two 1a and two BIPY
ligands at the vertex, A is remarkably stable. Thermogravi-
metric analysis (TGA) of a bulk sample of A showed that the
structure can be desolvated and resolvated reversibly. Addi-
tionally, A exhibited the same powder X-ray diffraction
(PXRD) pattern before and after evacuation (see ESIw).
Ligand 1b, possessing methyl groups in the 3,30 position of the
dibenzoic acid moiety, gave a three-fold interpenetrating, paddle-
wheel MOF structure (B, Fig. 3), which, in our experience, is rare
We are interested in the use of anisotropic MOF structures,
consisting of both dicarboxylate and dipyridyl ligands, for cata-
lysis,13 separations, and gas storage;6,14 specifically, to observe
how the chemical pore environment affects MOF performance in
these applications. Judging that carboxylates are better suited for
modification than pyridines, we set out to find a simple and
efficient route to chemically diverse, symmetric dicarboxylic acids
Department of Chemistry and the Institute for Catalysis in Energy
Processes, Northwestern University, 2145 Sheridan Road, Evanston,
IL 60208-3113, USA. E-mail: j-hupp@northwestern.edu;
stn@northwestern.edu; Fax: +1 874-467-1425; +1 874-491-7713;
Tel: +1 847-491-3504; +1 847-467-3347
w Electronic supplementary information (ESI) available: Synthetic
protocols and characterization data for compounds 1a–d; TGA and
PXRD data for A–D; N2 and H2 isotherms for C; Ar isotherms for A
and B; CIF file for A, B and D and [P4P + RAW] data files for C.
CCDC 661009–661012. See DOI: 10.1039/b714160b
Fig. 1 Ligands used in synthesis of Zn-based mixed-ligand MOFs.
ꢁc
This journal is The Royal Society of Chemistry 2008
3672 | Chem. Commun., 2008, 3672–3674

View Article Online
pillared-paddlewheel framework (Fig. 3, C. See Notes and
references section for unit cell data).z
Hypothesizing that differences in steric demand for the
bisarylethyne struts were responsible for the differences in
interpenetration, we synthesized ligand 1d, with 3,30-Br moi-
eties that are similar in size to the methyl groups of 1b,19 and
combined it with BIPY in MOF synthesis. The resulting
structure D was indeed three-fold interpenetrating (Fig. 3), like
B. This result supports the notion that direct influence over the
catenation structure of MOFs can be obtained by synthetic
manipulation of the sterics of the organic building blocks.
We investigated and compared the gas adsorption capabil-
ities of A, B, and D (Fig. 4; the isotherms of compound C were
not included as the material is unstable to evacuation (see ESI
for XRD and TGA dataw)). Surprisingly, A adsorbs neither N2
nor H2 at 77 K, despite the finding that it can be evacuated and
resolvated reversibly with DMF at room temperature (Fig. 4).
However, it does adsorb CO2 at 273 K, giving a Langmuir
Fig. 2 Views of A: (a) Single vertex in the network. The yellow
polyhedron represents the zinc ion complexing two 1a and two BIPY
ligands. Carbon: gray; oxygen: red; nitrogen: blue. (b) A schematic
representation of the single network unit in A. (c) A schematic
representation of the four-fold network interpenetration in A. (d) A
view down the crystallographic b-axis of A, showing open channels.
for pillared paddlewheel compounds. (Indeed, in our labs we have
encountered only one other example.) Despite the relatively high
degree of interpenetration, B is characterized by a substantial void
volume (34%, calculated by PLATON). Upon evacuation and
resolvation, it only loses a modest amount of its capacity to take
up DMF, as evidenced by TGA experiments (see ESIw). Although
this may be due to partial collapse of the pores, PXRD indicates
that the resolvated sample retains its crystallinity (see ESIw).
When the steric bulk of the 3,30-disubstituted bisarylethyne
strut was increased by replacing the methyl groups with hydro-
xymethyls, as in 1c, we obtained single crystals of material C.
While the current single-crystal diffraction data were inade-
quate for modeling all the disorder in the ligands, we are
confident that this structure is a two-fold interpenetrating,
surface area of 200 m2 ꢂ1. Most likely, the larger kinetic
g
energy of the CO2 molecules at higher temperature, together
with the greater framework flexibility at 273 K, allows the
sorbate to enter pores that are too narrow and rigid at 77 K.20
Interestingly, and in contrast to A, B readily took up both
N2 and H2, despite indications that its structure may have
partially collapsed when desolvated (its TGA data show
B20% loss of DMF uptake capacity, see ESIw). Analysis of
the N2 uptake for B based on a type-I Langmuir isotherm
yielded a Brunauer–Emmett–Teller (BET) surface area of
650 m2
g
ꢂ1. Analysis of the CO2 isotherm of B gave a
Langmuir surface area of 400 m2 gꢂ1 (data were insufficient
Fig. 3 Paddlewheel MOF structures B, C, and D. Left column: Chemical structures of the corresponding ligands 1b, 1c, and 1d. Center column:
Schematic representations of the single network units in each MOF. Right column: Views down BIPY axes showing network interpenetration and
framework pores for each MOF. Only the metal-cluster corners for structure C are shown as the single-crystal X-ray diffraction data were too weak
to establish the location of the ligand atoms with absolute confidence.
ꢁc
This journal is The Royal Society of Chemistry 2008
Chem. Commun., 2008, 3672–3674 | 3673

View Article Online
We acknowledge the Institute of Catalysis for Energy Processes
at Northwestern University, AFOSR, NSF, and the Basic Energy
Sciences Program, Office of Science, US Department of Energy
(Grant No. DE-FG02-01ER15244) for support of this work.
Notes and references
z Crystal data. Compound A: C32H30N4O6Zn1, M = 631.97, monoclinic,
C2/c, a = 28.848(7), b = 6.1751(15), c = 17.144(4) A, b = 98.838(4)1,
U = 3017.8(13) A3, Z = 4, Dc = 1.391 g cmꢂ3, m = 0.850 mmꢂ1, F(000)
= 1312, GOF = 1.111. R1 and wR2 are 0.072 and 0.195, respectively for
197 parameters and 2702 reflections [I
4 2s(I)]. Compound B:
C69H48N3O12Zn13, M = 1307.21, monoclinic, C2, a = 21.792(2), b =
27.979(3), c = 14.0334(15) A, b = 93.579(2)1, U = 8539.9(16) A3, Z = 4,
Dc = 1.017 g cmꢂ3, m = 0.88 mmꢂ1, F(000) = 2676, GOF = 0.89. R1
and wR2 are 0.051 and 0.116, respectively for 783 parameters and 11 758
reflections [I 4 2s(I)]. Compound C: C46H32N2O12Zn2, M = 935.52,
monoclinic, C2, a = 21.909(16), b = 27.458(16), c = 14.013(9) A, b =
93.183(17)1, U = 8417(10) A3. Compound D: C126H60Br12N6O24Zn6, M
= 3392.94, monoclinic, P2(1), a = 14.0828(9), b = 28.2518(18), c =
21.4518(14) A, b = 93.5600(10)1, U = 8518.4(9) A3, Z = 2, Dc = 1.323 g
cmꢂ3, m = 3.70 mmꢂ1, F(000) = 3300, GOF = 0.91. R1 and wR2 are
0.066 and 0.199, respectively for 1565 parameters and 22 525 reflections
[I 4 2s(I)]. The data were collected on a SMART CCD 1000 with MoKa
Fig. 4 Top left: TGA curves for A. Remaining panels, clockwise from
top right: N2, CO2, and H2 adsorption (open), and desorption (closed)
isotherms for A (&), B (J), and D (n).
´
radiation (ı) 0.710 73 A at 120 K. The structures were solved by direct
methods and refined by a full matrix least-squares technique based on F2
using the SHELXL 97 program. Modeling of solvent molecules for all but
A proved impossible due to severe disorder, thus the SQUEEZE sub-
routine in the PLATON software package was applied to mask the
electron density in the cavities. CCDC 661009–661012.
for BET fitting). The difference in adsorption capabilities of
the two materials can potentially be attributed to the smaller
pore mouth of A, B7 A vs. B12 A for B. Indeed, when Ar, a
probe with a similar kinetic diameter to N2 (cross-sectional
area for Ar is 14.2 A2 vs. B16.2 A2 for N2), was used in the gas
sorption experiments, the results were similar to those for the
N2 uptake, with only B indicating adsorption within the pores.
Structure B also adsorbed H2, with an uptake of B1 wt% at 1
atm and 77 K.
1 C.-J. Li, S. Hu, W. Li, C.-K. Lam, Y.-Z. Zheng and M.-L. Tong,
Eur. J. Inorg. Chem., 2006, 1931–1935.
2 J. L. C. Rowsell and O. M. Yaghi, J. Am. Chem. Soc., 2006, 128,
1304–1315.
3 R. Vaidhyanathan, D. Bradshaw, J.-N. Rebilly, J. P. Barrio, J. A.
Gould, N. G. Berry and M. J. Rosseinsky, Angew. Chem., Int. Ed.,
2006, 45, 6495–6499.
4 A. G. Wong-Foy, A. J. Matzger and O. M. Yaghi, J. Am. Chem.
Soc., 2006, 128, 3494–3495.
5 M. Eddaoudi, J. Kim, N. Rosi, D. Vodak, J. Wachter, M. O’Keeffe
and M. Yaghi Omar, Science, 2002, 295, 469–472.
6 B. Chen, S. Ma, F. Zapata, E. B. Lobkovsky and J. Yang, Inorg.
Chem., 2006, 45, 5718–5720.
7 X. Lin, J. Jia, X. Zhao, K. M. Thomas, A. J. Blake, G. S. Walker, N.
R. Champness, P. Hubberstey and M. Schroeder, Angew. Chem., Int.
Ed., 2006, 45, 7358–7364.
8 B. Kesanli, Y. Cui, R. Smith Milton, W. Bittner Edward, C. Bockrath
Bradley and W. Lin, Angew. Chem., Int. Ed., 2004, 44, 72–75.
9 J. L. C. Rowsell and O. M. Yaghi, Angew. Chem., Int. Ed., 2005,
44, 4670–4679.
10 X.-C. Huang, Y.-Y. Lin, J.-P. Zhang and X.-M. Chen, Angew.
Chem., Int. Ed., 2006, 45, 1557–1559.
11 A. L. Grzesiak, F. J. Uribe, N. W. Ockwig, O. M. Yaghi and A. J.
Matzger, Angew. Chem., Int. Ed., 2006, 45, 2553–2556.
12 D. Bradshaw, T. J. Prior, E. J. Cussen, J. B. Claridge and M. J.
Rosseinsky, J. Am. Chem. Soc., 2004, 126, 6106–6114.
13 S.-H. Cho, B. Ma, S. T. Nguyen, J. T. Hupp and T. E. Albrecht-
Schmitt, Chem. Commun., 2006, 2563–2565.
14 H. Chun, D. N. Dybtsev, H. Kim and K. Kim, Chem.–Eur. J.,
2005, 11, 3521–3529.
15 B.-Q. Ma, K. L. Mulfort and J. T. Hupp, Inorg. Chem., 2005, 44,
4912–4914.
16 D. N. Dybtsev, H. Chun and K. Kim, Angew. Chem., Int. Ed.,
2004, 43, 5033–5036.
17 S. I. Vagin, A. K. Ott and B. Rieger, Chem.-Ing.-Tech., 2007, 79,
767–780.
18 M. J. Mio, L. C. Kopel, J. B. Braun, T. L. Gadzikwa, K. L. Hull,
R. G. Brisbois, C. J. Markworth and P. A. Grieco, Org. Lett.,
2002, 4, 3199–3202.
19 L. Pauling, The Nature of the Chemical Bond, Cornell University
Press, New York, 3rd edn, 1960.
As expected, based on similar ligand sterics and degree of
catenation to B, D also adsorbed all three gases, with H2 uptake
of B0.3 wt% at 1 atm and 77 K. The BET surface area based on
N2 adsorption was 230 m2 ꢂ1, while CO2 gave a Langmuir
g
surface area of 280 m2 gꢂ1. That only B and D, among the four
materials reported herein, can take up all three gases suggests a
‘‘Goldilocks’’ principle for designing MOFs with good general
gas-sorption capabilities: the pores of such MOFs must be large
enough to allow gas molecules to access the MOF interior, but
not so large as to be non-robust.
In conclusion, we have demonstrated that substituents in
ligand struts can be used as structure-influencing tools in the
synthesis of MOFs with different levels of catenation. That small
changes in the side groups of the ethynylene dibenzoate struts,
from H to Me and CH2OH, can lead to three different struc-
tures, all with large pores and different gas adsorption behaviors,
hints at the vast potential for manipulating solid structures via
organic synthesis. In addition, because similarly sized Me and Br
substituents afford similarly interpenetrated MOFs, it may be
possible that ligand sterics can be relied on to induce specific
topologies with different chemical environments.
While the results described herein pertain to a specific set of
ligands and metal ion, there exists a tantalizing possibility that a
wide range of different MOF structures can be accessed with very
nearly the same ligand motif. By modifying the substituents of the
organic struts, a diverse collection of MOFs with differing con-
nectivities and varying levels of interpenetration should be possi-
ble, allowing one to unravel the effect of chemical functionality
and structural differences on the material properties of MOFs.
20 D. Cazorla-Amoros, J. Alcaniz-Monge and A. Linares-Solano,
Langmuir, 1996, 12, 2820–2824.
ꢁc
This journal is The Royal Society of Chemistry 2008
3674 | Chem. Commun., 2008, 3672–3674