Metal-organic frameworks (MOFs) composed of (triptycenedicarboxylato)zinc

Article abstract of DOI:10.1002/ejic.200701335

Two novel 2D and 3D coordination polymers of 9,10-trip-tycenedicarboxylic acid and zinc nitrate, which form under solvothermal reaction conditions, are described. As determined by single-crystal X-ray structure analysis, their frameworks are assembled from dinuclear zinc coordination units which are interlinked by triptycenedicarboxylato (TDC) struts. In the 2D-MOF reported here, the extended network of van der Waals interactions between triptycene units as well as coordinated solvent molecules is responsible for the dense assembly of layers with a 4⁴ net topology into the final crystal structure. The framework of the 3D-MOF is composed from [Zn₂(TDC) _1.5]∞ layers with a 6₃ topology, the layers being interlinked by triptycenedicarboxylato pillars (bnn-net). This MOF possesses hexagonal channels which are filled with guest molecules. Wiley-VCH Verlag GmbH & Co. KGaA, 2008.

Full text of DOI:10.1002/ejic.200701335

FULL PAPER
DOI: 10.1002/ejic.200701335
Metal-Organic Frameworks (MOFs) Composed of (Triptycenedicarboxylato)zinc
Sergei Vagin,[a,b] Anna Ott,[a,b] Hans-Christoph Weiss,[c] Alexander Karbach,[c]
Dirk Volkmer,[b] and Bernhard Rieger*[a,b]
Keywords: Coordination polymers / Metal-organic frameworks / Microporous compounds
Two novel 2D and 3D coordination polymers of 9,10-trip-
tycenedicarboxylic acid and zinc nitrate, which form under
solvothermal reaction conditions, are described. As deter-
mined by single-crystal X-ray structure analysis, their frame-
works are assembled from dinuclear zinc coordination units
which are interlinked by triptycenedicarboxylato (TDC)
struts. In the 2D-MOF reported here, the extended network
of van der Waals interactions between triptycene units as
well as coordinated solvent molecules is responsible for the
dense assembly of layers with a 44 net topology into the final
crystal structure. The framework of the 3D-MOF is composed
from [Zn2(TDC)1.5]ϱ
layers with a 63 topology, the layers be-
ing interlinked by triptycenedicarboxylato pillars (bnn-net).
This MOF possesses hexagonal channels which are filled
with guest molecules.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2008)
the final MOF topology adopted upon self-assembly are
not always unambiguously predictable and controllable. In
general, they are very much dependent on the structures
and properties of organic ligands as well as on the condi-
tions of synthesis and the Zn-containing frameworks are
no exception. The simultaneous formation of coordination
polymers with distinct structures from the same compo-
nents (supramolecular isomers) is often the case.[7] Using
the same ligand, such as terephthalic acid, might lead to
different MOF topologies upon coordinating to Zn ions
(starting, for example, from Zn(NO3)2·4H2O, under distinc-
tive reaction conditions). Thus, topologically different
MOFs comprising tetranuclear {Zn4O} units (MOF-5),[8a]
trinuclear {Zn3}[8b] as well as dinuclear paddlewheel {Zn2}
units[8c] have been prepared with this dicarboxylic acid. The
partial metal exchange of Zn by divalent ions of, for exam-
ple, Ni, Co or Cd may also lead to completely rearranged
framework structures instead of the expected architec-
tures.[9] On the other hand, a wide range of MOFs based
on tetranuclear {Zn4O} coordination units (CUs) could be
obtained simply by adopting the common synthetic pro-
cedure for MOF-5 (IRMOF-1) but, instead, utilising struc-
turally different ditopic aromatic carboxylato groups
linkers.[10a] In this respect, comprehensive insights into the
fundamental principles driving the network assembly in the
early stages of MOF synthesis would be of immense theo-
retical and practical interest.
Introduction
In recent years, the interest in employing 3D metal-or-
ganic frameworks (MOFs) as micro-porous materials has
grown immensely due to the high number of potential ap-
plications in materials science and industrial technologies.
Among their useful properties, nonlinear optical activity,[1]
sensing,[2] selective sorption and selective catalytic ac-
tivity,[3] storage of gases, in particular H2,[4] etc. can be men-
tioned. The versatility of the attractive features manifested
by this class of materials is mostly due to the wide possibil-
ities for altering and fine-tuning their structures which can
be achieved by matching the organic polydentate linker (L)
and/or metal-containing secondary building block (SBU).[5]
Up to now, most investigations on MOFs were focused
on applications based around gas storage or selective sorp-
tion. Regarding hydrogen sorption, the strategies proposed
to increase hydrogen uptake were aimed both at modifica-
tion of the organic linker structures, e.g. by extending their
aromatic moieties,[6a] as well as at development of new
MOF topologies and optimisation of the pore size and
geometry.[6b–6d]
The vast empirical knowledge in the field of metal-or-
ganic frameworks shows that the constitution of SBUs and
[a] WACKER-Lehrstuhl für Makromolekulare Chemie, Tech-
nische Universität München,
Lichtenbergstraße 4, 85747 Garching, Germany
Fax: +49-89-289-13562,
Here we report two novel MOFs utilising triptycened-
icarboxylic acid (H2TDC) and zinc which can be prepared
under the reaction conditions that normally lead to a pcu-
net[10b] based on {Zn4O} units as, for example, in MOF-5.
However, instead of the anticipated framework, two novel
coordination polymers were obtained which are constructed
E-mail: rieger@tum.de
[b] Institut für Anorganische Chemie II, Universität Ulm,
Albert-Einstein-Allee 11, 89081 Ulm, Germany
[c] Bayer Industry Services GmbH & Co OHG,
Building Q 18, 51368 Leverkusen, Germany
Supporting information for this article is available on the
WWW under http://www.eurjic.org or from the author.
Eur. J. Inorg. Chem. 2008, 2601–2609
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2601

S. Vagin, A. Ott, H.-C. Weiss, A. Karbach, D. Volkmer, B. Rieger
FULL PAPER
from paddlewheel SBUs with three or four blades, respec-
tively.[10c] To the best of our knowledge, no coordination
compounds of H2TDC have been reported up to now in the
Cambridge Structural Database and the new coordination
networks described here are the first examples of MOFs
with this bulky dicarboxylato ligand.
Results and Discussion
The intention of utilising a triptycene dicarboxylato
group instead of the more common terephthalato ligand
can be rationalised as the attempt to create the additional
contact surface for the adsorption of small guest molecules
in modified MOFs. For example, increasing void diameters
or volumes in MOF structures does not automatically give
rise to a positive impact on hydrogen absorption and a less
porous network such as an interpenetrating network struc-
ture might in fact be desired for the more effective hydrogen
physisorption at low temperatures and pressures.[4b,6] Ac-
cording to another report, a fourfold interpenetrated
{Zn4O} MOF constructed from sterically demanding aro-
matic linkers possesses a six times lower pore volume than
MOF-5 although its hydrogen storage capacity (48 bar,
room temperature) is comparable to that of MOF-5 under
the same conditions.[11] Encouraged by these studies, we
considered 9,10-triptycenedicarboxylic acid as a suitable
bis(bidentate) ligand to design new MOFs for the sorption
of small molecules, e.g. H2. Indeed, we anticipated that
three benzene rings arranged at 120° angles to one another
could increase the interaction energy of an H2 molecule
with the aromatic π system of the ligand by means of a
partial “sandwiching” effect, i.e. simultaneous interaction
of H2 with two benzene rings of the three V-shaped moie-
ties, as observed, for example for I2 and Ne.[12]
Scheme 1. Synthesis of 9,10-triptycenedicarboxylic acid (H2TDC):
1. [CH2O]n, HCl(g), dioxane; 2. NaOAc, HOAc; 3. anthranilic acid,
isopentyl nitrite, dimethoxyethane, maleic anhydride, KOH(aq); 4
KOH, MeOH, H2O; 5. CrO3, H2SO4(aq), acetone.
tive abundances are strongly dependent on the synthetic
conditions, e.g. reaction temperature and time, as well as on
the reactant ratios. Small polygonal platelets (e.g. tetrago-
nal, hexagonal or octagonal) were observed both by light
and scanning electron microscopy (SEM). Relatively large
irregularly shaped crystals such as intergrown columns,
blocks or thick spikes were also formed, sometimes as the
major products, which are completely covered by micro-
platelets (see Figure S1 in the Supporting Information). Ad-
ditionally, regular hexagonal columns and morphologically
related crystals were revealed in some synthetic batches.
9,10-Triptycenedicarboxylic acid (H2TDC), as shown in
Scheme 1, was synthesised according to a route proposed
in a patent.[13] The overall yield of the product based on
anthracene was ca. 30% but this could still be optimised
thereby facilitating a large scale preparation of H2TDC.
The compound was then allowed to react with Zn(NO3)2·
4H2O or Zn(NO3)2·6H2O under solvothermal conditions in
diethylformamide (DEF) to give polycrystalline precipi-
tates.
Figure 1. Polarised light optical micrograph showing different types
of crystals formed during the synthesis of (triptycenedicarboxyl-
ato)zinc MOFs in DEF solvent.
The IR spectra of the resultant solids clearly indicated
a complete conversion of H2TDC into the corresponding
dicarboxylato derivative (TDC). The C=O stretching vi-
The single-crystal X-ray diffraction (SXRD) analysis of a
bration of the carboxylic acid groups at 1705 cm–1 in hexagonal plate (Figure 1) picked from one of the synthetic
H2TDC underwent a low-frequency shift to approximately batches allowed us to formulate its structure as a 2D MOF
1640 cm–1 and the absorption in the region from 2500 to (TDC-MOF-1) with the composition [Zn(TDC)(DEF)].
3000 cm–1 ascribed to the O–H stretching mode in the hy- This 2D coordination polymer is formed by slightly dis-
drogen bonded carboxylic acid groups disappeared. Ad- torted paddlewheel secondary building units (see parts a
ditionally, new bands typical for aliphatic C–H stretching and b of Figure 2). The distance of 2.93 Å between two zinc
modes in the IR spectra of the vacuum- and temperature- ions in the CUs of TDC-MOF-1 is typical for zinc pad-
dried materials were observed at 2875–2990 cm–1, indicating dlewheel complexes bridged by four carboxylic groups in a
the presence of DEF in the isolated MOF compound.
syn-syn mode.[14a] The compound crystallises in the mono-
Crystals of different sizes and morphologies (see Fig- clinic space group C2/c. The planar tetragonal structural
ure 1) were found in the precipitated materials. Their rela- fragment spanned between points ABCD of TDC-MOF-1
2602
www.eurjic.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2008, 2601–2609

Frameworks Composed of (Triptycenedicarboxylato)zinc
consisting of four di-zinc paddlewheel units linked by four
TDCs {(Zn2)4(µ-TDC)4} (see Figure 2, c) forms planar 44-
tiling layers parallel to the ab plane of the crystal lattice.
This structural element has a geometry of a rhombus which
results in a close contact between adjacent triptycene units
with an angle of approximately 84.2° at the A and C verti-
ces. Such a deviation from a regular square geometry in
the {(Zn2)4(µ-TDC)4} layers might be due to the gain in
interaction energy resulting from the “gearing” (CH–π and
π–π) interaction in two pairs of triptycene three-blade ro-
tors at the A and C vertices with the shortest C···C distance
between the interacting blades amounting to 3.27 Å. No
specific interactions were found between the 2D layers in
the crystal lattice. The layers seem to assemble only by me-
ans of van der Waals interactions between the DEF residues
of neighbouring 2D sheets or between the DEF “tails” of
one sheet and the triptycene V-shaped “clefts” of another
neighbouring sheet (see part d of Figure 1, the marked dis-
tances are in the range 3.4–3.8 Å). A very weak C–H···O
hydrogen bond can, however, also be postulated as existing
between the methyl groups of DEF in one layer and the
carboxylato groups in another layer with an H···O distance
of 2.5–2.8 Å.
As mentioned above, the crystal platelets described here
are not the only type of crystals formed in DEF at 110–
120 °C. Immediate analysis of the resultant materials by
powder X-ray diffraction (PXRD) and SEM allows us to
suggest the simultaneous formation of several MOF phases
at different ratios strongly depending on the conditions of
synthesis (see Figures S1–S3 in the Supporting Infor-
mation). PXRD patterns of the materials, however, changed
upon drying or workup, probably due to the loss of oc-
cluded solvent molecules. Finally, the synthesised products
become converted into TDC-MOF-1, for example upon
drying at elevated temperature in vacuo, as their PXRD
patterns show a rather good agreement with that calculated
from the single crystal analysis data, although the broaden-
ing of the experimental reflection peaks might indicate
some loss of product crystallinity upon fast drying (Figure
S2 in the Supporting Information). The shapes of the large
Figure 2. Structural features of TDC-MOF-1 (hydrogen atoms are
sometimes omitted for clarity): four-bladed dinuclear zinc pad-
dlewheel CU (a) and its schematic representation by a distorted
rectangle (b); 2D layer of interlinked paddlewheel units viewed at
normal direction to the ab plane (c); stacking of 2D layers viewed
along the [1,1,0] direction (d) and space-filling representation of
the rhombic structural element ABCD indicated in (c) which dem-
onstrates the dense arrangement of triptycene units within each 2-
D layer (e).
crystals as well as of the small platelets remain unchanged certain period into TDC-MOF-1 as well. Similar processes
after drying, though their exfoliation could be observed in other MOFs, e.g. self-assembled from terephthalic acid
with SEM. Exfoliation was also observed for the synthe- and zinc, are already known.[14a,14c]
sised plate-like crystals but it is not clear whether this comes
The statements above are also supported by thermogravi-
from drying out under the vacuum conditions of SEM (10 metric analysis. Although no occluded solvent molecules
to 40 Pa) or whether this is a defect introduced during crys- placed at crystallographically defined positions were found
tal growth (compare Figures S1 and S3 in the Supporting in the crystal structure of TDC-MOF-1, TGA data of air-
Information).
dried bulk material indicates the presence of up to 10% of
The elemental analysis of the dried bulk product fits sat- noncoordinated water and DEF in various ratios depending
isfactorily with the [Zn(TDC)(DEF)] composition. It is rea- on the synthetic batch and this might explain the irrepro-
sonable to suppose that the major phases in the synthesised ducibility and low accuracy of the elemental analysis for
material are composed of the stable 2D [Zn(TDC)(DEF)]ϱ the synthesised product. The loss of occluded molecules can
layers which are separated from each other by occluded sol- be observed as two rather sharp steps appearing in the TGA
vent molecules in different manners (e.g. with different off- curve at temperatures ranging from 90 to 110 °C and from
sets and distances depending on the phase) and which build 170 to 190 °C. TDC-MOF-1 dried at 150 °C in vacuo does
TDC-MOF-1 upon solvent removal. Additionally, the lay- not show any weight loss up to 270 °C, at which tempera-
ered structures based on trinuclear zinc units can also be ture the coordinated DEF is volatilised in two subsequent
supposed in the samples and this could transform over a steps (see Figure S7 in the Supporting Information and
Eur. J. Inorg. Chem. 2008, 2601–2609
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
2603

S. Vagin, A. Ott, H.-C. Weiss, A. Karbach, D. Volkmer, B. Rieger
FULL PAPER
¯
Exp. Sect.). Approximately 13.5% weight loss may be ob- space group P3 (no. 147) and in the hexagonal crystal sys-
served during the first step (up to 360 °C) which is some- tem with the space group P6/m (no. 175) but the final struc-
what higher than calculated for the split of one DEF mole- tural refinement was accomplished in P6/m. The X-ray
cule per dinuclear zinc coordination unit (10%). Complete structure analysis shows that the crystals are composed
removal of coordinated DEF molecules is achieved above from dinuclear zinc CUs symmetrically linked by three
400 °C, followed by the decomposition of coordinated TDC TDC units (“three-bladed” paddlewheels[10c]) to form lay-
units. Investigation of the porosity of dried TDC-MOF-1 ers. Additionally, each paddlewheel unit has two axially co-
by recording BET isotherms shows a very low “apparent ordinated TDCs which crosslink the layers and extend the
surface” (ca. 5 m2 gϪ1) which could be expected in light of structure into three dimensions. This framework can be
its dense packing arrangement.
thus considered as a pillared 63 net or as a distorted 5-
The general yield of bulk TDC-MOF-1 in the synthesis connected bnn-net.[10b] However, two different coordination
at 120 °C drops upon increasing the Zn(NO3)2 to H2TDC modes of axial TDCs can be distinguished (Figure 3). The
ratio (see Table 1). An additional prolongation of the reac- unit cell of TDC-MOF-2 contains eight paddlewheel clus-
tion time leads to competitive formation of crystals having ters, two of which have apical TDC ligands coordinated in
the shape of regular hexagonal columns, these being the a slightly distorted bidentate (chelating) mode with disorder
compound TDC-MOF-2 (see Figure 1). After having fil- of the carboxylato group about the vertical threefold sym-
tered off the plate-like crystals and blocks formed within metry axis (see part c of Figure 3, type-II clusters). The car-
the first 24 h, hexagonal columns continuously formed from bon atom C48 of the apical carboxylato group and both
the remaining reaction mixture as the sole product contain- zinc ions in such a cluster are positioned on a common axis.
ing TDC. However, there was some contamination con- However, hard restraints had to be applied for modelling
sisting of a yellowish often amorphous powder with a very this cluster in order to obtain reasonable Zn2–O7Ј and
high zinc content according to EDX measurements (see Zn2–O7 distances (1.98 and 2.03 Å, respectively), probably
Figures S4 and S5 in the Supporting Information). This as a consequence of the disorder and high diffuse electron
impurity could be zinc oxide/hydroxide or (formato)zinc density. Apical TDCs in the remaining six dinuclear units
and the relative amount of it was especially high with an (type I clusters) coordinate by the carboxylic groups in a
increasing amount of zinc salt relative to H2TDC and this is monodentate syn mode (the Zn1–O5 and Zn1–O6 distances
quite understandable. The largest hexagonal columns were are 1.94 and 2.67 Å, respectively, and the Zn1–Zn1–C43 an-
formed when the ratio of zinc salt to H2TDC was the high- gle is 164.5°, Figure 3, a) which leads to the tetracoordi-
est but they quickly turned opaque and became fragile al- nated zinc ions.
ready during synthesis. TDC-MOF-2 was also formed
The (triptycenedicarboxylato)zinc framework in TDC-
nearly solely when the solutions of reacting components MOF-2 possesses two different types of linear channels
were separately preheated to 120 °C before being mixed. with different shapes and sizes (see parts e and f in Figure 3)
The resultant hexagonal columns, however, feature abun- and a total potential solvent accessible volume (PSAV) of
dant macroscopic defects due to the high rate of crystal 51% calculated with PLATON.[15] Twelve DEF molecules
growth (see Figure S6a in the Supporting Information). per layer segment could be located in the “pocket” of larger
Contrary to what was mentioned above, an increase in the (regular hexagonal, PSAV = 20%) channels and reasonably
Zn to TDC ratio leads to finer crystals in this case due to resolved by SXRD analysis. Each DEF molecule interacts
the slower formation. Nevertheless, SXRD analysis of the with a neighbouring triptycene unit of the 2D layer in such
hexagonal columns from different batches indicated their a way that one methyl group of DEF fits into the triptycene
identity.
cleft with distances of 3.57 Å and 3.71 Å to the correspond-
The structure of TDC-MOF-2 could be reasonably ing benzo-rings as shown in Figure 4. The methylene group
solved and refined in the trigonal crystal system with the of another ethyl substituent in the DEF molecule appears
Table 1. Formation of MOFs in DEF (20 mL) using different ratios of Zn(NO3)2 to H2TDC at 120 °C.
Mass of
Mass of
Ratio of
reagents
Duration of
reaction / h
Mass of
precipitate formed / mg
Zn(NO3)2·6H2O / mg H2TDC / mg
Type of product
297
344
344
344
344
344
344
1:1
3:1
4:1
5:1
6:1
9:1
0–24
24–72
0–24
24–72
0–24
24–72
0–24
24–72
0–24
24–72
0–24
384
6
TDC-MOF-1
TDC-MOF-1 + -2
TDC-MOF-1
TDC-MOF-2 + impurities
TDC-MOF-1
TDC-MOF-2 + impurities
TDC-MOF-1 + -2 (as junctions)
TDC-MOF-2 + impurities
–
TDC-MOF-2 + impurities
TDC-MOF-1 (traces) + impurity
TDC-MOF-2 + impurities
892
281
237
161
430
71
706
–
776
69
1190
1487
1785
2677
24–72
1030
2604
www.eurjic.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2008, 2601–2609

Frameworks Composed of (Triptycenedicarboxylato)zinc
located on a six-fold symmetry axis which was resolved as
a disordered and slightly distorted octahedral hexaaqua-
zinc(II) complex (one half per elementary cell). The ob-
served Zn–O distances in this complex are 1.96 Å and the
O–Zn–O adjacent angles are 85° and 95°. Each coordinated
water molecule interacts with two DEF molecules by hydro-
gen bonding (OAqua–ODEF distances are 2.65 Å and 2.85 Å,
respectively). The observed structural parameters such as
bonds lengths and angles in this hydrogen-bonded network
are in the range of those typically observed for hexaaqua-
zinc complexes.[14d,14e] The only exception is that the Zn–O
bond length found here is somewhat shorter than the re-
ported values [the most common Zn–O distances in
2+
Zn(H2O)6 complexes are 2.04–2.10 Å] but this could be
a problem with accuracy due to the disorder and lowered
occupancy.
Other guest molecules, e.g. in smaller channels with the
flattened hexagonal cross-section and PSAV of 10%, are
highly disordered and the diffuse electron density was cor-
rected by applying the SQUEEZE routine in the PLATON
software package.[15]
One can notice that each cluster in TDC-MOF-2 is con-
structed from two zinc ions and five carboxylic groups,
among which three are basal and two are apical. It is unam-
biguous that the basal groups are negatively charged (de-
protonated) carboxylato moieties, whereas the nature of the
axial groups is not clear. The charge neutrality of the frame-
work requires that either the pillaring dicarboxylato groups
are monoprotonated with the protons disordered over the
carboxylato units or that the other crystallographically un-
resolved positively charged guest species, e.g. diethylammo-
nium, are present in the framework. Three-bladed zinc pad-
dlewheel clusters with axially coordinated nondeprotonated
carboxylic groups have, to the best of our knowledge, not
been reported up to now[10c] whereas a charge compensa-
tion by means of, for example, the alkylammonium cation
in pentacarboxylato-di-zinc CU is known.[14f] Our attempts
to resolve the structure and nature of the residual guest spe-
cies in TDC-MOF-2 before SQUEEZE procedures were,
nevertheless, unsuccessful. On the other hand, no clear
statements about the nature of pillaring TDCs can be drawn
by comparison of the structural parameters of the coordi-
nation units reported here with those reported in literature
data. First, no dinuclear carboxylatozinc CUs, which are
structurally and symmetry equivalent to those found in
TDC-MOF-2, have been yet described and second, the
analysis of data reported on zinc paddlewheel CUs reveals
a relatively high flexibility of such structures in terms of
geometric parameters. Taking into account all the above,
the most plausible stoichiometry for TDC-MOF-2 based
on SXRD can, in our opinion, be expressed as [Zn2-
Figure 3. Structures of paddlewheel CUs in TDC-MOF-2 (a: type
I cluster; c: type II cluster), their representation with polyhedra (b,
d) and assembly into a layer (e, perspective view normal to the ab
plane, axial TDC ligands are not shown; f, space-filling representa-
tion with axial TDCs). Solvent molecules are omitted for clarity.
Distances shown in (f) are between the van der Waals surfaces of
the carbon atoms.
to be in the proximity of the triptycene cleft which inter-
links the layers (the shortest CH2–CAr distance is 3.63 Å).
The carbonyl groups of the DEF molecules are turned
towards the centre of the channel and form a hydrogen-
bonding network (see Figure 4) with another guest species
(TDC)1.5+x(HTDC)1–x·[Zn(H2O)6]x/2·{solv}y], where x ≈ 1͞
8
and {solv} represents DEF and any other guests.
Unfortunately, analyses of the TDC-MOF-2 material by
TGA, EA and IR spectroscopy does not give enough infor-
mation to explicitly support the suggested composition, as
will be shown below. The TGA-MS data for TDC-MOF-2
measured for the process taking place under argon show
Figure 4. Structural fragment of TDC-MOF-2 showing the interac-
tions between the guest species in the hexagonal channel: a) view
along the ab plane of the unit cell; b) view along the c axis. Hydro-
gen atoms and the disorder of the hexaaquazinc complex are not
shown for clarity.
Eur. J. Inorg. Chem. 2008, 2601–2609
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
2605

S. Vagin, A. Ott, H.-C. Weiss, A. Karbach, D. Volkmer, B. Rieger
Preferential formation of zinc paddlewheel units from di-
FULL PAPER
the loss of included solvent molecules in the temperature
range 50–250 °C in one broad step, corresponding to an carboxylic acids and zinc nitrate in DMF under solvother-
approximately 24% weight loss. The slow weight decrease mal synthetic conditions has already been reported.[14a,14b]
occurs further as well, followed by a sharp step at ca. 350– On the contrary, a series of isoreticular MOFs based on
360 °C (6.8%) accompanied by a strong increment in MS Zn4O clusters was prepared in DEF by applying similar re-
detection of ions with m/z = 17 a.m.u. The decarboxylation action conditions.[10] Nevertheless, contamination of MOF-
of TDC becomes noticeable at 550 °C (see Figure S10 in 5 by other MOF phases was also observed and this ap-
the Supporting Information). TGA-MS measurements on peared to be strongly sensitive to the preparation condi-
the process in a stream of synthetic air indicate the loss of tions.[17] The reasons for the unexpectedly dominating Zn2
solvent molecules in nearly the same temperature range but paddlewheel assembly by H2TDC instead of Zn4O clusters
the corresponding weight loss is much higher and occurs in under the conditions of MOF-5 synthesis are not entirely
two steps (31% at 50–115 °C and 8% at 115–250 °C) which clear and many factors governing this process can be sug-
might be due to the differences in the handling of the gested. On the one hand, the bulkiness of triptycene frag-
probes. Additionally, the amount of inorganic remains after ments together with their specific symmetry might lead to
burning-out at 400–500 °C is dependant on the batch. This high steric demands and disturb the 3D self-assembly by
is due to the impurities with a high Zn content e.g. ZnO means of tetranuclear Zn4O carboxylato SBUs, although
which are nearly always present in the probes with hexago- this seems to contradict the preliminary modelling and cal-
nal crystals. TGA (synthetic air) of TDC-MOF-2 kept in culations.[16b] On the other hand, a more efficient spatial
CHCl3 for several days and subsequently dried in vacuo at arrangement (“gearing”) of triptycene fragments in a lay-
120 °C does not show any weight loss up to 200 °C. At ered structure such as TDC-MOF-1 can lead to an ad-
higher temperature a slow decrease of sample mass occurs, ditional energy gain due to the van der Waals (π-π and CH-
followed by burning-out and resulting in 15.3% of residual π) interactions. The aliphatic nature of the carboxylato
ZnO ashes. This value is the lowest obtained so far from groups in H2TDC in contrast to aromatic –COOH in
different TDC-MOF-2 probes.
terephthalic and other acids forming MOFs with Zn4O co-
The PXRD patterns also differ noticeably in the inten- ordination units might also play a role in controlling the
sities of peaks from batch to batch, as well as upon storage, MOF topology, although some Zn4O-based structures are
and this is most probably to do with the amount, location also known for aliphatic carboxylato[18a,18b] and carbamato
and nature of the guest species remaining in the framework groups.[18c,18d] The mechanism leading to the competitive
(see Figure S6 in the Supporting Information). Moreover, formation of a hexagonal framework TDC-MOF-2 could
reflections in the PXRD patterns of TDC-MOF-2 disap- not be reasonably established up to now, though our experi-
pear upon drying, indicating the low stability of such a co- ments point to the kinetically controlled formation of four-
ordination network. No conversion into TDC-MOF-1 bladed paddlewheel Zn-TDC units, whereas the three-
seems to take place in this case. The shape of the hexagonal bladed units seem to be preferred thermodynamically.
crystals, however, remains unchanged upon drying accord-
An interesting feature found from the structural analysis
ing to SEM studies (see, for example, Figure S3 in the Sup- of H2TDC-based MOFs is the affinity of DEF hydrocarbon
porting Information). We believe that a collapse of the residues for the V-shaped clefts of the triptycene moieties
pores in the framework takes place upon guest removal. composing the frameworks. This observation can be utilised
This would also explain the fact that despite high PSAV, in developing new MOFs for gas uptake with enhanced
the crystals of TDC-MOF-2 do not soak up chloroform or host-guest interactions. Additionally, the observed tendency
noticeably exchange the guest molecules against CHCl3 and of triptycene-dicarboxylic acid to form layered paddlewheel
thus remain swimming on the surface of this solvent, in structures with zinc ions can be employed in constructing
contrast to other porous MOFs.[16a]
new porous MOFs using a “pillaring strategy”.[14b,19] Based
IR spectra of dried TDC-MOF-2 (vac., 120 °C) reveal on the pillaring approach, several zinc paddlewheel based
the presence of aliphatic species in the structure, although a MOFs have already been reported, among which that con-
comparison of spectra of the dried and initially synthesised structed from terephthalic acid and DABCO shows H2 ab-
probes indicates that some amount of DEF is removed sorption comparable with MOF-5.[14b,20] Several other di-
upon drying (see Figure S9 in the Supporting Information). carboxylato ligands and diamines give analogous assem-
The residual aliphatic species could be either the coordi- blies, though often with interpenetration.[14b,21] The latter
nated DEF or diethylamine (DEA). The latter might be effect will be, however, unattainable if Zn2(TDC)2 layers
either coordinated to zinc or form the ion pairs with car- were pillared because of the overlapping triptycene moie-
boxylic groups of TDC. The elemental analysis of the dried ties. A possible advantage of such pillared structures created
material is unsatisfactory for carbon, however it fits better by H2TDC could be a strongly increased “contact surface”
for coordinated DEF instead of DEA in the structure with of the 2D layers capable of effectively interacting with
the composition [Zn2(TDC)1.5(HTDC)·{ZnO}1/16·{DEF}2]. guests. Additionally, the layers are impermeable to most
Accounting for the excessive ZnO as an impurity gives an small molecules (see Figure 2, e) thus giving the anisotropy
excellent fit of elemental analysis and TGA-data for the in permeability to such crystals which might become at-
composition [Zn16(TDC)12(HTDC)8·{DEF}17·{ZnO}3] (see tractive for advanced applications. The work in this field
Exp. Sect.).
continues in our laboratory.
2606
www.eurjic.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2008, 2601–2609

Frameworks Composed of (Triptycenedicarboxylato)zinc
Table 2. Summary of crystal and structure refinement data for
TDC-MOF-1 and -2.
Conclusions
Triptycenedicarboxylic acid (H2TDC) was applied here
for the first time to construct new coordination polymers
with zinc ions. Although the solvothermal synthesis in DEF
generally resulted, after the respective work-up, in the non-
porous 2D MOF based on zinc TDC paddlewheel coordi-
nation units, an additional crystalline phase such as small
hexagonal columns was also obtained under certain reac-
tion conditions. According to SXRD analysis, it appeared
to be a new 3D MOF based on zinc TDC “three-bladed
paddlewheels” which possesses relatively big channels filled
with guest molecules and might be potentially interesting
for studying the guest exchange, removal and binding pro-
cesses. Additionally, 2D MOFs can be modified using a
“pillaring” strategy into porous 3D systems in which the
triptycene fragments are expected to enhance the van der
Waals interactions between guest molecules and the frame-
work, acting as “clefts” for small molecules.
TDC-MOF-1
27H23NO5Zn
506.83
monoclinic
C2/c
16.0157(1)
14.4722(1)
21.6977(2)
90.00
TDC-MOF-2
Empirical formula
Formula weight
Crystal system
Space group
a / Å
b / Å
c / Å
α / °
C
C41.67H31NO7.75Zn1.38
759.56
hexagonal
P6/m
33.9418(7)
33.9418(7)
14.2185(6)
90.00
β / °
γ / °
108.804(1)
90.00
4760.72(6)
1.414
90.00
120.00
14185.8(7)
1.067
12
Vcalcd. / Å3
ρcalcd. / gcm–3
Z
8
2θmax
65.70
1.752
66.04
1.264
µ / mm–1
(Tmin/Tmax
)
0.90195/1.00000
13678
3974
0.84776/1.00000
69984
8562
Reflections measured
Reflections independent
Restrains/parameters
R, wR [IϾ2σ(I)]
0/309
4/505
0.0247, 0.0673
0.0279, 0.0689
0.378/–0.238
0.0633, 0.2055
0.0705, 0.2175
1.941/–1.159
R, wR (all data)
(ρmax/ρmin) / eÅ–3
Experimental Section
General: All chemicals and solvents were purchased from commer-
cial sources and were used without additional purification, unless
otherwise mentioned.
Synthesis
Instrumentation: NMR: Bruker AMX400 (400 MHz for 1H and
100 MHz for 13C). IR: Bruker FTIR IFS-113V; IFS-55. Elemental
Analysis: Elementar Vario EL. TGA: Mettler Toledo TGA/SDTA
851 (N2, 50 mLmin–1; 10 °Cmin–1); Netzsch STA 409 PC Luxx
coupled with Aeolos QMS 403 C (Ar or synthetic air, 60 mLmin–1;
10 °Cmin–1). BET: Quantachrome Autosorb-1. PXRD: PANalyt-
ical X’Pert PRO MPD system; STOE STADI P system with an IP-
PSD detector. SEM: Hitachi TM-1000 tabletop microscope; Zeiss
DSM 962 electron microscope with EDX equipment from EDAX
Boston.
9,10-Triptycenedicarboxylic acid (H2TDC): this was prepared from
anthracene according to the synthetic pathway proposed earlier.[13]
Chloromethylation of anthracene and synthesis of intermediate
9,10-acetoxymethylanthracene were carried out as described else-
where.[23,24] C22H14O4 (342.4): calcd. C 77.18, H 4.12; found C
77.07, H 4.20. 1H NMR (400 MHz, [D6]DMSO): δ, ppm = 7.12 (m
AAЈBBЈ, 6 H), 7.88 (m AAЈBBЈ, 6 H). 13C NMR (100.6 MHz,
[D6]DMSO): δ = 61.9 (s), 124.7 (s), 126.4 (s), 144.8 (s), 171.9
(s) ppm. FTIR (KBr): ν = 3060 (–2520 v. br. medium-intensity peak
˜
with multiple weak maxima), 1705 (vs), 1598 (w), 1453 (s), 1405
(m), 1317 (m), 1270 (s), 1215 (w), 1175 (w), 1144 (vw), 1048 (w),
1007 (w), 931 (m), 855 (vw), 772 (w), 754 (s), 684 (m), 642 (m), 606
(w), 524 (w), 480 (m), 440 (w) cm–1. TGA (N2 stream): weight loss
at 360–410 °C (–97%).
Crystal Structure Determination: The crystal structure determi-
nation was carried out on a Xcalibur diffractometer (Oxford Dif-
fraction, Ruby CCD) using Cu-Kα (λ = 1.54178 Å) radiation, osmic
mirrors at T = 100 K, with full-sphere data collection omega and
phi scans. The CrysAlis programs were used for data collection and
reduction. A semiempirical absorption correction using spherical
harmonics (SCALE3 ABSPACK algorithm) was applied. Crystal
structure solutions were achieved by direct methods and the refine-
ments were carried out on |F2| against all reflections after merging
using the programs implemented in SHELXTL Version 6.10.[22] All
non-hydrogen atoms were refined including anisotropic displace-
ment parameters.
One of the COO– groups in the structure TDC-MOF-2 was refined
as a disordered group. The channels in the structure are partly oc-
cupied by solvent molecules. Since most of them are highly disor-
dered with no possibility of refining them reasonably, they were
removed using the SQUEEZE routine implemented in the software
package PLATON.[15] Only one guest molecule of diethylform-
amide could be reasonably refined in the structure.
TDC-MOF-1 and -2: In a typical procedure, warm (ca. 50 °C) solu-
tions of zinc(II) nitrate tetrahydrate (0.78 g, 3 mmol) in DEF
(10 mL) and triptycenedicarboxylic acid (0.344 g, 1 mmol) in DEF
(10 mL) were mixed together in a pressure-resistant glass tube, se-
aled and placed in a block with a programmable thermostat (Bar-
key Company) preheated to 50 °C. The temperature was gradually
raised to 110–120 °C (at 10–20 °Ch–1) and kept constant for 15–
24 h. The hot reaction mixture was decanted and the resultant crys-
tals covered with fresh solvent. After filtering, the product was
dried under vacuum at room or elevated temperature to give TDC-
MOF-1.
The decanted reaction solution was placed in a pressure-resistant
glass tube, sealed and heated further at 120 °C for several days to
yield crystals of TDC-MOF-2 contaminated with an unidentified
by-product.
CCDC-664462 and -664463 contain the supplementary crystallo-
graphic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
Alternatively, preheated to 120 °C, the reactants (same quantities
as in a typical procedure above) were mixed in a pressure-resistant
glass tube, sealed and maintained at 120 °C for 8–15 h. An increase
of the Zn/H2TDC ratio resulted in a longer reaction time and im-
For summary of crystal structure analysis data on TDC-MOF-1 proved the appearance of TDC-MOF-2 crystals, though the
and -2 see Table 2.
amount of Zn-containing impurities also increased.
Eur. J. Inorg. Chem. 2008, 2601–2609
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
2607

S. Vagin, A. Ott, H.-C. Weiss, A. Karbach, D. Volkmer, B. Rieger
FULL PAPER
C.-D. Wu, A. Hu, L. Zhang, W. Lin, J. Am. Chem. Soc. 2005,
127, 8940–8941.
a) J. L. C. Rowsell, O. M. Yaghi, J. Am. Chem. Soc. 2006, 128,
1304–1315; b) J. L. C. Rowsell, O. M. Yaghi, Angew. Chem.
2005, 117, 4748–4758; Angew. Chem. Int. Ed. 2005, 44, 4670–
4679; c) A. J. Fletcher, K. M. Thomas, M. J. Rosseinsky, J. So-
lid State Chem. 2005, 178, 2491–2510.
M. Eddaoudi, D. B. Moler, H. Li, B. Chen, T. M. Reineke, M.
O’Keeffe, O. M. Yaghi, Acc. Chem. Res. 2001, 34, 319–330.
a) J. L. C. Rowsell, A. R. Millward, K. S. Park, O. M. Yaghi, J.
Am. Chem. Soc. 2004, 126, 5666–5667; b) J. L. C. Rowsell, J.
Eckert, O. M. Yaghi, J. Am. Chem. Soc. 2005, 127, 14904–
14910; c) M. Hirscher, B. Panella, Scripta Materialia 2007, 56,
809–812; d) H. Frost, T. Düren, R. Q. Snurr, J. Phys. Chem. B
2006, 110, 9565–9570.
a) A. L. Grzesiak, F. J. Uribe, N. W. Ockwig, O. M. Yaghi,
A. M. Matzger, Angew. Chem. 2006, 118, 2615–2618; Angew.
Chem. Int. Ed. 2006, 45, 2553–2556; b) N. L. Toh, M. Nagara-
thinam, J. J. Vittal, Angew. Chem. 2005, 117, 2277–2281; An-
gew. Chem. Int. Ed. 2005, 44, 2237–2241; c) P. D. C. Dietzel,
R. Blom, H. Fjellvag, Dalton Trans. 2006, 586–593; d) M. Ed-
gar, R. Mitchell, A. M. Z. Slawin, P. Lightfoot, P. A. Wright,
Chem. Eur. J. 2001, 7, 5168–5175.
a) H. Li, M. Eddaoudi, M. O’Keeffe, O. M. Yaghi, Nature
1999, 402, 276–279; b) H. Li, C. E. Davis, T. L. Groy, D. G.
Kelley, O. M. Yaghi, J. Am. Chem. Soc. 1998, 120, 2186–2187;
c) H. Li, M. Eddaoudi, T. L. Groy, O. M. Yaghi, J. Am. Chem.
Soc. 1998, 120, 8571–8572.
Y. Wang, B. Bredenkötter, B. Rieger, D. Volkmer, Dalton Trans.
2007, 689–696.
a) M. Eddaoudi, J. Kim, N. Rosi, D. Vodak, J. Wachter, M.
O’Keeffe, O. M. Yaghi, Science 2002, 295, 469–472; b) N. W.
Ockwig, O. Delgado-Friedrichs, M. O’Keeffe, O. M. Yaghi,
Acc. Chem. Res. 2005, 38, 176–182; c) S. I. Vagin, A. K. Ott,
B. Rieger, Chem. Ing. Technol. 2007, 79, 767–780.
TDC-MOF-1, as-synthesised: [Zn2(TDC)2(DEF)2]·xDEF·yH2O: no
reproducible TGA and elemental analysis. For example, for x = 2
and y = 2: calcd. C 61.40, H 5.80, N 4.47; found C 61.64, H 5.87,
N 4.27 (probe A). For x = 1 and y = 1: calcd. C 62.55, H 5.25, N
3.71; found C 62.54, H 5.22, N 3.48 (probe B). For x = 0.5 and y
= 1: calcd. C 62.70, H 4.98, N 3.24; found C 62.78, H 5.08, N 3.10
(probe C). TGA (N2, batch 1): weight loss at 90–110 °C (–9.5%;
H2O; y ≈ 6), 170–190 °C (–0.7%; DEF; x ≈ 0.08), 270–360 °C
(–13.2%), 370–490 °C (–13.1%), 510–620 °C (–22.8%). TGA (N2,
batch 2): weight loss at 90–110 °C (–4.2%; H2O; y ≈ 2.6), 170–
190 °C (–4.9%; DEF; x ≈ 0.54), 270–360 °C (–14.0%), 370–490 °C
(–12.8%), 510–620 °C (–23.1%).
[4]
[5]
[6]
TDC-MOF-1, dried at 150 °C/vacuum: For [Zn2(TDC)2(DEF)2]:
calcd. C 63.98, H 4.57, N 2.76; found C 63.11, H 4.61, N 2.66.
TGA (N2): weight loss at 270–360 °C (–13.6%), 370–490 °C
[7]
[8]
(–15.4%), 510–620 °C (–27.4%). FTIR (KBr): ν = 3054 (vw), 2974
˜
(w), 2934 (w), 2875 (vw), 1665 (sh), 1634 (vs), 1443 (m), 1399 (s),
1361 (sh), 1296 (w), 1264 (w), 1213 (w), 1182 (vw), 1104 (w), 1086
(vw), 1046 (w), 1015 (w), 941 (w), 899 (vw), 814 (m), 759 (w), 745
(m), 712 (m), 691 (w), 646 (w), 626 (m), 494 (vw), 469 (m) cm–1.
TDC-MOF-2, as-synthesised: [Zn2(TDC)1.5+x(HTDC)1–x·[Zn-
(H2O)6]x/2·{solv}y]: no reproducible TGA and elemental analysis.
TGA (Ar): weight loss at 50–260 °C (–25%), 260–340 °C (–4%),
340–360 °C (–7%), 360–500 °C (–4%), 500–700 °C (–29%). TGA
(synthetic air): weight loss at 50–115 °C (–31%), 115–260 °C (–8%),
[9]
260–400 °C (–6%), 400–550 °C (–40%). FTIR (KBr): ν = 3425 (w),
˜
3062 (w), 2976 (m), 2939 (w), 2875 (vw), 1636 (vs), 1444 (vs), 1400
(vs), 1298 (s), 1265 (m), 1215 (m), 1186 (w), 1164 (w), 1143 (vw),
1116 (w), 1105 (w), 1047 (m), 1016 (w), 943 (w), 902 (vw), 825
(m), 811 (m), 752 (s), 713 (w), 692 (m), 646 (s), 627 (s), 492 (m),
(br) cm–1.
[10]
[11]
[12]
B. Kesanli, Y. Cui, M. R. Smith, E. W. Bittner, B. C. Bockrath,
W. Lin, Angew. Chem. 2005, 117, 74–77; Angew. Chem. Int. Ed.
2005, 44, 72–75.
a) N. Kulevsky, K. Pierce, Spectrochim. Acta Part A: Mol. and
Biomol. Spectrosc. 1993, 49, 417–423; b) A. Furlan, S.
Leutwyler, M. J. Riley, J. Chem. Phys. 1994, 100, 840–855.
B. H. Klanderman, J. W. H. Faber, patent, FR 1520625, 1968,
9 pp.
a) H. F. Clausen, R. D. Poulsen, A. D. Bond, M.-A. S. Chevall-
ier, B. B. Iversen, J. Solid State Chem. 2005, 178, 3342–3351;
b) H. Chun, D. N. Dybtsev, H. Kim, K. Kim, Chem. Eur. J.
2005, 11, 3521–3529; c) M. Edgar, R. Mitchell, A. M. Z. Sla-
win, P. Lightfoot, P. A. Wright, Chem. Eur. J. 2001, 7, 5168–
5175; d) P. Knuuttila, Polyhedron 1984, 3, 303–305; e) J. W.
Steed, B. J. McCool, P. C. Junk, J. Chem. Soc., Dalton Trans.
1998, 3417–3423; f) J. Klunker, M. Biedermann, W. Schafer, H.
Hartung, Z. Anorg. Allg. Chem. 1998, 624, 1503–1508.
a) P. van der Sluis, A. L. Spek, Acta Crystallogr., Sect. A 1990,
46, 194–201; b) A. L. Spek, J. Appl. Crystallogr. 2003, 36, 7–
13.
TDC-MOF-2, dried at 120 °C/vacuum: For [Zn16(TDC)12(HTDC)8·
{DEF}17·{ZnO}3]: calcd. C 64.18, H 4.46, N 2.42; ZnO ashes 15.7;
found C 64.24, H 4.53, N 2.46; ZnO ashes 15.3. TGA (synthetic
air): weight loss at 200–400 °C (–16%), 400–550 °C (–69%). FTIR
(KBr): ν = 3421 (w), 3058 (w), 2979 (w), 2937 (vw), 1622 (vs), 1444
˜
[13]
[14]
(vs), 1408 (vs), 1296 (s), 1211 (w), 1186 (w), 1163 (w), 1144 (vw),
1086 (vw), 1045 (m), 1016 (w), 941 (w), 902 (vw), 814 (s), 748 (s),
715 (m), 692 (s), 646 (s), 627 (s), 484 (m), (br) cm–1.
Supporting Information (see also the footnote on the first page of
this article): including IR spectra, PXRD patterns, TGA plots, pho-
tos and SEM pictures of compounds as well as ORTEP plots of
asymmetrical units of TDC-MOF-1 and -2.
[15]
[16]
Acknowledgments
a) A. Ott, S. Vagin, B. Rieger, unpublished results. b) R.
Schmid, S. Amirjalayer, personal communications on the un-
published results.
We would like to thank Dr. W. Seidl (Lehrstuhl für Bauchemie,
Technische Universität München) for helpful discussions and for
help in acquiring of TGA-MS data.
[17]
[18]
A. L. Grzesiak, F. J. Uribe, N. W. Ockwig, O. M. Yaghi, A. J.
Matzger, Angew. Chem. Int. Ed. 2006, 45, 2553–2556.
a) W. Clegg, D. R. Harbron, C. D. Homan, P. A. Hunt, I. R.
Little, B. P. Straughan, Inorg. Chim. Acta 1991, 186, 51–60; b)
H. Koyama, Y. Saito, Bull. Chem. Soc. Jpn. 1954, 27, 112–
114; c) A. Belforte, F. Calderazzo, U. Englert, J. Straehle, Inorg.
Chem. 1991, 30, 3778–3781; d) C. S. McCowan, T. L. Groy,
M. T. Caudle, Inorg. Chem. 2002, 41, 1120–1127.
a) S. Kitagawa, M. Kondo, Bull. Chem. Soc. Jpn. 1998, 71,
1739–1753; b) G. Ferey, Chem. Mater. 2001, 13, 3084–3098.
D. N. Dybtsev, H. Chun, K. Kim, Angew. Chem. 2004, 116,
5143–5146; Angew. Chem. Int. Ed. 2004, 43, 5033–5036.
[1] O. R. Evans, R.-G. Xiong, Zh. Wang, G. K. Wong, W. Lin,
Angew. Chem. 1999, 111, 557–559; Angew. Chem. Int. Ed. 1999,
38, 536–538.
[2] T. M. Reineke, M. Eddaoudi, M. Fehr, D. Kelley, O. M. Yaghi,
J. Am. Chem. Soc. 1999, 121, 1651–1657.
[3] a) M. Fujita, Y. J. Kwon, S. Washizu, K. Ogura, J. Am. Chem.
Soc. 1994, 116, 1151–1152; b) J. S. Seo, D. Whang, H. Lee, S. I.
Jun, J. Oh, Y. J. Jeon, K. Kim, Nature 2000, 404, 982–986; c)
[19]
[20]
2608
www.eurjic.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2008, 2601–2609

Frameworks Composed of (Triptycenedicarboxylato)zinc
[21] B.-Q. Ma, K. L. Mulfort, J. T. Hupp, Inorg. Chem. 2005, 44,
4912–4914.
[22] G. M. Sheldrick, Universtität Göttingen (Germany), 2000.
[23] M. W. Miller, R. W. Amidon, P. O. Tawney, J. Am. Chem. Soc.
1955, 77, 2845–2848.
[24] T. Nakaya, T. Tomomoto, M. Imoto, Bull. Chem. Soc. Jpn.
1966, 39, 1551–1556.
Received: December 12, 2007
Published Online: April 23, 2008
Eur. J. Inorg. Chem. 2008, 2601–2609
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
2609