Structural and zeolitic features of a series of heterometallic supramolecular porous architectures based on tetrahedral {M(C2O 4)4}4- primary building units

Article abstract of DOI:10.1039/b503964a

The utilization of tetrahedral pre-formed coordination compounds {M(C ₂O₄)₄}^4- (M = Zr^IV, U^IV; C₂O₄^2- = oxalate) permitted the efficient construction of rare examples of heteronuclear supramolecular nano-porous architectures. A series of metal-organic coordination frameworks prepared by association of these building units with either Mn²⁺, Cd²⁺, or Mg²⁺ have been structurally characterized and are described. Their 3-D chemical scaffold is based on the primary tetrahedral building unit but their pore sizes and topologies could be varied through the M²⁺ metal ion involved in the assembling process, and the anionic tetrahedral moiety. These structures display channels with apertures up to 12 A × 8 A which are emptied of solvates at mild temperatures without affecting the chemical scaffold, the integrity of which is maintained up to 250-300°C. A certain degree of flexibility of the coordination polymers upon guest release is suggested by the temperature dependence of the powder X-ray patterns and N₂ sorption experiments, but reversible and selective sorption of small molecules has been observed substantiating that these open-frameworks behave like sponges. The Royal Society of Chemistry 2005.

Full text of DOI:10.1039/b503964a

View Article Online / Journal Homepage / Table of Contents for this issue
F U L L P A P E R
Structural and zeolitic features of a series of heterometallic
supramolecular porous architectures based on tetrahedral
4−
{M(C2O4)4} primary building units†
Inhar Imaz,a Georges Bravica and Jean-Pascal Sutter*a,b
a Institut de Chimie de la Matie`re Condense´e de Bordeaux-CNRS, Universite´ Bordeaux 1,
87 Avenue du Dr. Schweitzer, F-33608, Pessac, France
b Laboratoire de Chimie de Coordination du CNRS, Universite´ Toulouse 3, 205 Route de
Narbonne, F-31077, Toulouse, France. E-mail: sutter@lcc-toulouse.fr
Received 18th March 2005, Accepted 20th June 2005
First published as an Advance Article on the web 15th July 2005
4−
2−
The utilization of tetrahedral pre-formed coordination compounds {M(C2O4)4} (M = ZrIV, UIV; C2O4 = oxalate)
permitted the efficient construction of rare examples of heteronuclear supramolecular nano-porous architectures. A
series of metal–organic coordination frameworks prepared by association of these building units with either Mn2+,
Cd2+, or Mg2+ have been structurally characterized and are described. Their 3-D chemical scaffold is based on the
primary tetrahedral building unit but their pore sizes and topologies could be varied through the M2+ metal ion
involved in the assembling process, and the anionic tetrahedral moiety. These structures display channels with
˚
˚
apertures up to 12 A × 8 A which are emptied of solvates at mild temperatures without affecting the chemical
scaffold, the integrity of which is maintained up to 250–300 ◦C. A certain degree of flexibility of the coordination
polymers upon guest release is suggested by the temperature dependence of the powder X-ray patterns and N2
sorption experiments, but reversible and selective sorption of small molecules has been observed substantiating that
these open-frameworks behave like sponges.
based materials like bimetallic magnets, for instance,12–14 it
Introduction
has seldom been envisaged for the construction of porous
A contemporary aspect of supramolecular chemistry aims at de-
frameworks.15–22 We have considered a tetrafunctional synthon,
signing organizations with workable chemical, physical and/or
structural properties according to the new perspectives these
molecule-based materials open in material science. A promising
approach to such functional materials involves coordination
polymers. Given the diversity of geometries and electronic
properties exhibited by the metal ions and the limitless number of
organic ligands that can be envisaged, it is in principle possible to
achieve supramolecular architectures with specifically tailored
properties.1,2 This approach is currently intensively investigated
for the preparation of micro-porous materials.3–7 Compared to
traditional zeolites, coordination polymers offer the possibility
of designing open frameworks exhibiting high porosities and
large specific areas but also customized pore sizes, topologies,
and chemical compositions. Several prototypical example of
such porous architectures have been reported and were shown
to be effective functional materials for catalysis,8 gas sorption
and storage,9 or separation.10
The widely used route to form an open metal–organic
framework consists of the direct assembly of a metal ion
with a bridging ligand. The resulting network relies on the
formation (during the association process) of the so-called
secondary building units, or nodes, formed by the connection
of the coordinating groups of the ligands to the metal ions.
These nodes will determine the net topology and dimensionality
of the coordination polymer.11 An alternative route for the
construction of coordination polymers involves a pre-formed
coordination compound able to assemble with a complementary
module, typically a metal ion. Whereas this latter approach was
shown to be very efficient for the preparation of several molecule-
4−
{M(C2O4)4} , with tetrahedrally arranged oxalate linkers as a
primary building unit.23 Two evident benefits of this approach
are i. to permit the pre-design and ideally-fix a node of the
network by means of the linking capability and geometry of
the molecular building block, and ii. to provide access to
heterometallic open-frameworks through the assembling of this
primary building unit with a second metal ion.
Herein we report that robust heterometallic 3D-coordination
polymers with open frameworks are formed by association
4−
of tetrahedral building units {M(C2O4)4} (M = Zr4+, U4+)
with M2+ metal ions. A series of supramolecular nano-porous
architectures (SPA) constructed with either Mn2+, Cd2+ or Mg2+
are described. They exhibit open frameworks which are easily
emptied of solvates without collapsing, and become accessible
for the sorption of small molecules.
Results and discussion
Synthesis and structural features
(a) [K2Mn{U(C2O4)4}·9H2O], SPA-1, and [K2Cd{U(C2O4)4}·
5H2O]·4H2O, SPA-2. The reaction of K4U(C2O4)4 with
either Mn2+ or Cd2+ in H2O led respectively, to green [K2Mn-
{U(C2O4)4}·9H2O], SPA-1, and pink [K2Cd{U(C2O4)4}·5H2O];
4H2O, SPA-2. Since their X-ray structure analyses revealed that
the two compounds exhibit the same network organization, we
will describe here only the structure for the Cd derivative; SPA-1
is isostructural and has already been briefly communicated.24
Well-shaped single crystals of [K2Cd{U(C2O4)4}·5H2O]·4H2O,
SPA-2, suitable for X-ray diffraction analysis were grown by
slow inter-diffusion of the reagent solutions. The crystal data
for the compound are given in Table 1, an ORTEP plot with the
atom numbering scheme, bond lengths and angles are provided
as ESI.†
† Electronic supplementary information (ESI) available: Thermo-
gravimetric analysis for SPA-1, SPA-2, and EtOH@SPA-3, powder
X-ray diffractogram at different temperatures for SPA-1 and SPA-3;
positional coordinates, isotropic and anisotropic thermal parameters,
ORTEP plot, bond lengths, and angles for all the reported structures.
See http://dx.doi.org/10.1039/b503964a
As a result of the reaction, each {U(C2O4)4}4− building block
is linked to four CdII ions by means of its oxalate ligands, as
T h i s j o u r n a l i s
T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 5
D a l t o n T r a n s . , 2 0 0 5 , 2 6 8 1 – 2 6 8 7
2 6 8 1
©

View Article Online
Table 1 Crystal data for SPA-2, SPA-3, and SPA-4
SPA-2
SPA-3
SPA-4a
Empirical formula
Formula mass
Crystal system
Space group
C8H18O25K2CdU
942.8
Monoclinic
P21
11.473(1)
8.968(1)
11.524(1)
97.43(2)
1175.8(2)
2
C14H22O39K2Mg2U2
1417.7
Monoclinic
C2/m
20.437(3)
12.841(4)
14.614(4)
94.52(3)
3823.2(6)
4
C8H10O21K2MnZr
666.51
Monoclinic
P21/n
14.699(2)
12.630(2)
11.819(1)
106.76(6)
2100.8(8)
4
˚
a/A
˚
b/A
˚
c/A
b/◦
3
˚
V/A
Z
qcalc./g cm−3
F(000)
2.663
888
2.424
2568
2.107
1316
Crystal dimensions/mm
T/K
0.08 × 0.07 × 0.06
0.34 × 0.30 × 0.20
0.21 × 0.16 × 0.07
298
298
298
l(MoKa)/ mm−1
h rang◦e/◦
8.249
2.36–26.37
52.7
8.836
4.13–36.32
72.6
1.594
1.00–34.97
72.6
2hmax
/
Tot. no. of data collected
No. of obsd data [I ≥ 2r(I)]
R1, wR2 [I ≥ 2r(I)]
Goodness-of-fit on F2
No. of variables
8688
4662
0.0160, 0.0449
0.0171, 0.0642
1.207
22286
6805
0.0482, 0.1362
0.0658, 0.1510
1.047
18238
6192 [I ≥ 3r(I)]
0.077, 0.242
0.077, 0.242
1.093
R1, wR2 (all dat)
335
276
317
a The H2O molecules located in the channels are highly disordered (see Experimental section) and could not all be positioned; moreover, the
corresponding H atoms have not been included in the structure refinement but these have been considered for the formula, molecular mass, density,
adsorption coefficients and F(000).
two K+ cations. These are located in the chair-shaped hexagons
generated by the interconnected tetrahedral units (Fig. 1b). Each
K+ is positioned in the plane of the oxalate ligands and interacts
with four O atoms of two {M(C2O4)} moieties. The K atoms
thus complete the chemical scaffold forming the walls which
delimit the channels running through the structure.
SPA-2 contains nine H2O molecules. Five act as ligands, one
for the UIV center and two for each K+ ion, and the remaining
four are located in the channels. These water molecules are
easily released (vide infra) revealing an open framework. The
main channels running through the structure develop along the
˚
˚
b axis. Their apparent square opening of 6 A × 6 A results
from the staggered superposition of the 24-atom rings formed
by three interconnected {U–oxalate–Cd} fragments (Fig. 1).
The effective porous architecture of SPA-2 is clearly visible
in Fig. 2 where the van der Waals volume of the chemical
Fig. 1 View of the structure of the isostructural [K2Mn{U(C2O4)4}·
9H2O],24 SPA-1, and [K2Cd{U(C2O4)4}·5H2O]·4H2O SPA-2 (U4+ in
blue, M2+ (Mn or Cd) in green, K+ in orange, the H2O molecules
are not shown). a) Detail of the {U(C2O4)4-M} assembly; b) detail
of the 24-atom ring superposition framing the channel aperture and
location of the K+ cations; c) view of the porous architecture down
the b axis showing the channels running through the structure.
˚
Selected bond lengths (A): SPA-1: U–O(oxalate), 2.375(4)–2.456(4);
U–O(water), 2.542(5); Mn–O, 2.283(4); U–Mn, 6.102(3)–6.193(3).
SPA-2: U–O(oxalate), 2.384(4)–2.406(4); U–O(water), 2.554(5); Cd–O,
2.355(4)–2.452(4); U–Cd, 6.166(4)–6.252(4). K–O(oxalate), 2,797(5)–
2.824(5), K–O(water), 2.856(4)–2.828(4). Further data can be found in
the ESI.†
shown in Fig. 1. The coordination sphere of the Cd ions as
well consists of four Cd←{oxalate-U} linkages. Hence, every
metal ion is connected four times by oxalate bridges to the
other metal ion forming a three-dimensional framework with
a diamond-like topology. The charge neutrality is provided by
Fig. 2 View of the framework of SPA-2 taking into account the van
der Waals volume of the atoms (generated by Cerius 2). The blue surface
corresponds to the walls of the porous volume. Left: projection in the ac
plane. Right projection in the ab plane.
2 6 8 2
D a l t o n T r a n s . , 2 0 0 5 , 2 6 8 1 – 2 6 8 7

View Article Online
Fig. 3 View of the structure of [K2Mg2{U2(C2O4)7}·2H2O]·9H2O, SPA-3 (U4+ blue, K+ orange, Mg2+ green, the H2O molecules are not depicted). Top:
Detail of the {U(C2O4)} ladder organization and ladder interconnections framing the cavities. Bottom: View of the framework revealing the channels
˚
running through the structure along the a axis (left), b axis (center) and c axis (right). Selected bond lengths (A): U–O(oxalate), 2.372(5)–2.553(4);
Mg–O(oxalate), 2.799(9)–2.831(9); Mg–O(H2O), 2.736(9); K–O(oxalate), 2.715(6)–2.912(7); U–U, 6.449(1)–6.392(1); further data can be found in
the ESI.†
scaffold has been taken into account; the opening diameter of
˚
the channels is of ca. 3 A. The volume accessible for a small
3
molecule (H2O) determined with Platon25 is 347 A per unit cell
˚
3
˚
volume (1176 A ) which represents 30% of void per unit volume
for SPA-2. The same analysis has been performed for SPA-1
revealing a potential porosity of 29% per unit volume (unit cell
3
3
˚
˚
volume: 1151 A , potential free volume: 332 A ).
(b) [K2Mg2{U2(C2O4)7}·2H2O]·9H2O,
SPA-3. When
K4U(C2O4)4 was reacted with Mg2+ in H2O, green single crystals
of SPA-3 with formula [K2Mg2{U2(C2O4)7}·2H2O]·9H2O
were obtained. The X-ray structure analysis revealed a 3D-
coordination polymer with an association scheme of the
building units different to that observed with MnII and CdII.23
The crystal data for the compound are given in Table 1, an
ORTEP plot with the atom numbering scheme, the bond
lengths, and angles are provided as ESI.†
4−
For SPA-3, the {U(C2O4)4} units assemble by the mean of
{U–oxalate–U} linkages in a ladder type arrangement running
along the b axis (Fig. 3) where each metal center displays five
oxalate ligands in its coordination sphere but has lost formally
half an oxalate. The 3-D framework results from the ladder
interconnections realized by the coordination of the external
oxalate ligands to the K and Mg ions. Views of the resulting
scaffold revealing the porous nature of the structure are depicted
in Fig. 3.
Fig. 4 View of the framework of SPA-3 along the b axis taking into
account the van der Waals volume of the atoms (generated by Cerius 2).
The blue surface corresponds to the walls of the porous volume.
(Reproduced from ref. 23.)
Channels are running through the structure in the three
˚
˚
˚
directions of space with apertures of 6 A × 6 A (along a), 12 A ×
˚
˚
˚
8 A (along b), and 5.5 A × 5.5 A (along c). Approximately 32%
of the volume of the crystal remains available for small molecules
(H2O). 23,25 Taking into account the van der Waals volume of the
atoms, the main channels along the b axis can be described as a
with MnII ions in H2O [K2Mn{Zr(C2O4)4}·8H2O], SPA-4, was
obtained. Its composition was found to be similar to that of
SPA-1 but with Zr instead of U; however, X-ray structure
analysis for SPA-4 revealed a different network organization.
The crystal data for SPA-4 are given in Table 1, an ORTEP plot
with the atom numbering scheme, the bond lengths, and angles
are provided as ESI.†
˚
diamond-shaped section with sides of ca. 5 A and diagonals of
9 × 4 A (Fig. 4).
˚
[K2Mn{Zr(C2O4)4}·8H2O], SPA-4. We considered the pos-
sibility of replacing the U-based building unit by the related
4−
{Zr(C2O4)4} complex. When the latter unit was reacted
D a l t o n T r a n s . , 2 0 0 5 , 2 6 8 1 – 2 6 8 7
2 6 8 3

View Article Online
The most striking feature is found in the association of the
{Zr(C2O4)4} units with the MnII ions. Indeed, the anticipated
chelating interaction of the external oxygen atoms of the oxalate
ligands is not observed. Instead a single oxygen atom is involved
in the linkage of an oxalate to a Mn ion, in much the same
fashion as for the coordination to the K cations. A view of the
interactions of the K and Mn ions with a {Zr(C2O4)4} unit
is shown in Fig. 5. Both the Mn and K ions are linked
to up to four Zr units developing thus a 3D coordination
polymer. The resulting framework can be seen as compact layers
interconnected by bridging oxalate ligands. This scaffold frames
channels running in one direction parallel to the layers (Fig. 5),
their effective square shaped aperture is 5.5 A × 5.5 A. The
solvent accessible space in the crystal corresponds to 23% of the
total volume,25 which hosts eight H2O molecules; four of them
act as ligands for the Mn and K ions, the remaining four are
solvates.
For SPA-1–4, the coordination network is based on the ability
of the external atoms of the oxalates from the building unit to
act as ligands for the other metal ions. The resulting structural
features are, however, dependent on the engaged modules. The
effect of the M2+ metal ions on the realized assemblage can
be related to the different metal–ligand affinities and to the
coordination sphere for this ion, like for instance Mn2+ and
Mg2+ in SPA-1 and SPA-3. Conversely, ions with closely related
characteristics will lead to isostructural networks, i.e. Mn2+
and Cd2+ in SPA-1 and SPA-2. The origin of the differences
4−
found when changing from {U(C2O4)4} to its Zr counterpart,
4−
{Zr(C2O4)4} , is more subtle. As shown by SPA-1 and SPA-4,
the networks developed by these building units with Mn2+ are
˚
˚
different revealing different association schemes. Most likely,
IV
˚
the origin is to be found in the ionic radii (0.80 A for Zr and
IV
˚
0.97 A for U ) which govern the volume and flexibility of the
4−
4−
{M(C2O4)4} moieties. The geometrical data for {U(C2O4)4}
and {Zr(C2O4)4}4− obtained from the structures described above
reveal a distance between the central metal ion and the external
˚
oxygen atoms of the oxalate ligands ca. 0.2 A longer for the
U unit than for its Zr counter part (4.421–4.492 A versus 4.228–
4.269 A). Moreover, the coordination sphere of the larger U ion
˚
˚
is more flexible and may accommodate up to 10 atoms as shown
by SPA-3. These features are already expressed in the solid state
4−
organizations of the {M(C2O4)4} building units alone which
consist of discrete complexes for the Zr moieties26,27 whereas the
U units associate through oxalate bridges into 1-D coordination
polymers.28
Thermogravimetric analyses (TGA) revealed that all H2O
molecules contained in the frameworks are released under mild
conditions. The behavior for SPA-3 (Fig. 6) is a typical example.
First H2O leaves the framework which leads to a mass decrease
of 12% corresponding to the loss of 9 H2O. This release is
observed in the temperature range 30–120 ◦C and occurs in
three steps suggesting different binding energies even for the
lattice water. Above 120 ◦C and up to 250 ◦C the sample
mass remains unchanged, in this temperature domain SPA-3 is
fully desolvated. Finally, for higher temperatures the compound
decomposes. The characteristic temperatures for the end of
solvent release and stability for each SPA are given in Table 2;
the related TGA data are provided as ESI.† It can be noticed
Fig. 5 View of the structure of [K2Mn{Zr(C2O4)4}·8H2O], SPA-4, with
Zr4+ in blue, K+ in orange, and Mn2+ in green, the H2O molecules are
not depicted. Top: Detail of the linkages between the Mn2+ and K+ ions
with the building unit {Zr(C2O4)4}4−. Bottom: View of the 3D frame-
work revealing the channels running through the structure. Selected
˚
bond lengths (A): Zr–O(oxalate), 2.163(3)–2.239(4); Mn–O(oxalate),
2.741(5)–2.972(4); K–O(oxalate), 2.664(4)–2.901(5); further data can be
found in the ESI.†
The structural information gathered for this series of porous
architectures shows that the utilization of the pre-formed tetra-
4−
hedral coordination compound {M(C2O4)4} spontaneously
leads to 3D coordination polymers when associated with a
Mn2+, Mg2+ or Cd2+ metal ion. Attempts were made with
other M2+ ions (i.e. Co, Ni, Cu, Zn, Pd) but they did not
lead to crystalline solids suitable for structural characterization.
Nevertheless, infrared spectroscopy and chemical analyses for
these solids strongly suggest the formation of homometallic
{MII(C2O4)}n coordination polymers.
Fig. 6 Themogravimetric analysis for SPA-3.
Table 2 Temperatures of dehydration ending and limit of thermal stability for SPA-1–4 deduced from TGA dataa
Dehydration ending T/◦C
Upper thermal stability T/◦C
[K2Mn{U(C2O4)4}·9H2O], SPA-1
[K2Cd{U(C2O4)4}·9H2O], SPA-2
[K2Mg2{U2(C2O4)7}·11H2O], SPA-3
[K2Mn{Zr(C2O4)4}·8H2O], SPA-4
120
140
140
140
280
320
250
b
a Measured under a N2 stream, heating rate 1 ◦C min−1 b Not determined.
.
2 6 8 4
D a l t o n T r a n s . , 2 0 0 5 , 2 6 8 1 – 2 6 8 7

View Article Online
that for SPA-3 9 H2O are observed by TG whereas 11 are found
by X-ray structure analysis. For all compounds reported here a
decreased number of solvate molecules is found for the isolated
solids both by TGA and chemical analysis as compared to their
crystal structure data. This observation is in line with the fact
that solvent release occurs at mild temperatures.
of the frameworks discussed above which compacts upon guest
release and thus reduces the space freed by the solvates. It is
worth noting here that the lower limit for N2 sorption is a pore
˚
diameter of ca. 3.5 A, therefore this adsorbate may not be best
suited for the evaluation of microporosity in these compounds.
Despite the apparent small porosity, these frameworks exhibit
reversible sorption properties. We have pointed out above
that upon exposure to air, the evacuated frameworks recover
their initial crystallographic phase, obviously they adsorb H2O
molecules. The H2O uptake has been clearly established and
quantified for SPA-1 and SPA-3 by recording the mass increase
of a dehydrated powder (Fig. 8). Both the frameworks re-adsorb
ca. 5 H2O per formula unit. SPA-3 was also found to adsorb
small alcohol molecules (MeOH, EtOH) after dehydration
and immersion in the corresponding anhydrous alcohol. In
Fig. 9 is depicted the TGA of MeOH@SPA-3 showing that
Further evidence for the persistence and stability of these
frameworks upon solvent release was obtained from the X-ray
powder diffractograms (XRPD) recorded at different tempera-
tures. In Fig. 7 are depicted the XRPD for SPA-2◦(from bottom
◦
◦
to top) as synthesized at 25 C, at 60 C, at 120 C, just after
◦
cooling back to 25 C, and at the same temperature after 24 h
in air. The diffractograms recorded at 60 and 120 ◦C show
broader and weaker signals, revealing a loss of crystallinity
upon solvent removal. Finally, when the evacuated sample
was maintained in air for 24 h the initial X-ray pattern was
recovered, indicating the re-adsorption of H2O. The frameworks
of SPA-1 and SPA-3 exhibit a related behavior (see ESI†). The
dehydration–rehydration process for SPA-2 can also be followed
by its color change from pink to green and back to pink. These
supramolecular architectures thus belong to the third generation
compounds of solvated solids, i.e. the sample recovers its initial
crystalline phase after the desorption/adsorption process of its
guest molecules.6,29
Fig. 7 X-Ray powder diffractograms recorded for compound SPA-2.
From bottom to top: as synthesized at 25 ◦C, at 60 ◦C, at 120 ◦C, at 25 ◦C
after H2O release, after H2O re-adsorption at 25 ◦C.
The evolutions of the XRPD patterns, especially the shifts of
some of the diffraction peaks indicate that the crystal packing is
modified by the H2O release. However, the recovery of the initial
diffractogram unambiguously confirms that the integrity of the
framework, i.e. the metal–oxalate network connectivity, is not
affected by solvent release. The evolution of the diffractogramm
patterns during the desorption thus reveals a deformation of the
network which can be related to a certain flexibility of the 3-D
coordination polymer.30–33 The use of the rigid oxalate ligands
was anticipated to avoid the collapse of the framework but the
metal–ligand linkages remain flexible. Likely, the scaffold has a
tendency to compensate the volume freed by the released solvent
molecules. As a result, the effective porosity of the dehydrated
solid might be reduced with respect to the potential porosity
exhibited by the solvated structure.
Fig. 8 Mass increase upon H2O uptake for SPA-1 (top) and SPA-3
(bottom) in ambient conditions.
In order to evaluate the surface area developed by the
frameworks, adsorption isotherms of N2 at 77 K were measured
for SPA-1 and SPA-3. For both compounds N2 adsorption was
observed for low relative pressures, therefore data were collected
in the 0.05 to 0.2 P/P0 range. From these data, Langmuir (and
BET) surface areas were derived to be 25.4 (12.2) m2 g−1 and 34
(17.2) m2 g−1, respectively for SPA-1 and SPA-3. These results
suggest a very limited effective porosity for the desolvated frame-
works which is in strong contrast with the potential porosity
exhibited by the structures of the solvated architectures. The
origin of the low effective porosity may be found in the flexibility
Fig. 9 Themogravimetric analysis for MeOH@SPA-3. The sample was
heated from 20 to 115 ◦C and maintained at this temperature until the
mass remained invariant before cooling it back to room temperature.
D a l t o n T r a n s . , 2 0 0 5 , 2 6 8 1 – 2 6 8 7
2 6 8 5

View Article Online
MeOH is gradually released upon heating. For EtOH@SPA-3
a related but slightly different release versus T behavior has
been obtained (see ESI†). The XRPD showed these compounds
to be rather amorphous. The sorption capacity per gram was
roughly estimated from the TG data to be in the order of 1.2
and 0.93 mmol of alcohol, respectively MeOH and EtOH. These
observations suggest that these coordination polymers behave
as open solids provided that a chemical affinity exists between
the potential guest molecules and the frameworks. In the
reported examples when the external molecules have interaction
capabilities (coordination, H-bonding), as is the case for H2O
or ROH, the network behaves like a sponge and soaks up these
molecules but it stays sealed to molecules like N2 which have no
interactions with the scaffold.
(0.27 g; 75%). The single-phase composition of the powder was
checked by XRPD. Larger crystals were obtained when EtOH
was allowed to mix by vapor diffusion with the H2O solution.
After 3 days green crystals of SPA-1 (0.22 g; 62%) were collected
and manually separated from the precipitated white manganese
oxalate and light-green uranium oxalate. Anal. (%) Calcd. for
C8O16K2MnU·7H2O: C, 11.31; H, 1.66; K, 9.20; Mn, 6.47; U,
28.02 Found: C, 12.16; H, 1.62; K, 9.32; Mn, 6.58; U, 28.24. IR
(KBr, cm−1): 3467 (s), 1654 (s), 1455 (m), 1440 (m), 1303 (m),
907 (w), 800 (m), 485 (m).
[K2Cd{U(C2O4)}·5H2O]·4H2O, SPA-2.
A
solution of
K4U(C2O4)4 was prepared from U(C2O4)2·6H2O (0.2 g,
0.4 mmol) and K2(C2O4)·H2O (0.15 g, 0.8 mmol) in H2O
(12 mL) at 80 ◦C. A solution of Cd(NO3)2·4H2O (0.12 g,
0.4 mmol) in H2O (5 mL) was dropwise added to this hot
solution with slow stirring which yielded immediately a pink
microcrystalline precipitate of SPA-2 (0.215 g; 68%). The single-
phase composition of the powder was checked by XRPD. Larger
single crystals of SPA-2 were obtained within 3 weeks by slow
interdiffusion of the two aqueous solutions in an H-shaped tube
(total volume 60 mL). These were washed with EtOH and dried
in air (160 mg; 50%). Anal. (%) Calcd. for C8H18O25CdK2U: C,
10.19; H, 1.92; K, 8.29; Cd, 11.92; Found: C, 10.35; H, 1.92;
K, 7.83; Cd, 11.47. IR (KBr, cm−1): 3482 (s), 1654 (s), 1458 (m),
1438 (m), 1303 (m), 907 (w), 800 (m), 486 (m).
Conclusion
The utilization of the tetrahedral pre-formed coordination
4−
compounds {M(C2O4)4} permitted the efficient construction
of rare examples of heteronuclear supramolecular nanoporous
architectures. Their 3D chemical scaffold is based on the primary
tetrahedral building unit but their pore sizes and topologies
could be varied through the M2+ metal ion involved in the
assembling process. It was found that the relative flexibility of
the anionic tetrahedral moiety, in relation with its MIV central
ion (U versus Zr), may also influence the assembling scheme
of the framework. These results further demonstrate that the
use of a coordination compound as primary building unit is
certainly a valuable and versatile approach for the construction
of robust open frameworks. Moreover, this approach appears
perfectly suited for the design of open frameworks with various
compositions of metal ions.
[K2Mg2{U2(C2O4)7}·2H2O]·9H2O, SPA-3. K4U(C2O4)4 was
synthesized in situ from U(C2O4)2·6H2O (0.2 g, 0.4 mmol) and
K2(C2O4)·H2O (0.15 g, 0.8 mmol) in H2O (12 mL) at 80 ◦C. To
the hot green solution, a solution of Mg(ClO4)2·6H2O (0.12 g,
0.4 mmol) in H2O (2 mL) was added. The reaction mixture
was allowed to cool, and ethanol was added by vapor diffusion.
After 7 days green crystals (152 mg; 27%) were obtained that
were manually separated from the precipitated white magnesium
oxalate and light-green uranium oxalate. Anal. (%) Calcd. for
C14H22O39K2Mg2U2: C, 11.87; H, 1.56; Mg, 3.43; U, 33.59
A remarkable feature of the reported compounds is that
whereas their effective porosity is strongly reduced upon dehy-
dration they are still able to re-adsorb small molecules, and thus
behave as porous solids. This reversible sorption process comes
with a breathing framework which opens to accommodate guest
molecules and compacts upon guest release. The chemical inter-
actions between the guest molecules and the framework seems to
be the key to these sorption properties. These observations point
to the fact that the sorption potentiality of such flexible networks
can not just be evaluated by the effective porosity exhibited by
the evacuated solid.
The solvent release not only renders the pores accessible to
other molecules but also liberates coordination sites on the
metal ions. These may be regarded as Lewis acid centers and,
consequently, these supramolecular porous architectures could
be envisaged as catalysts. From this point of view, access to
frameworks with given active centers (metal ions) becomes
important and we currently develop our approach in this respect.
Found: C, 11.81; H, 1.36; Mg, 3.11; U, 34.01. IR (KBr, cm−1)
:
3397 (s), 1667 (s), 1633 (s), 1443 (m), 1372 (m), 1323 (m), 830
(m), 490 (m).
[K2Mn{Zr(C2O4)4}·8H2O], SPA-4. An aqueous solution
(10 mL) of K4Zr(C2O4)4·5H2O (0.345 g, 0.5 mmol) was pre-
pared and warmed to 80 ◦C. To the hot colorless solu-
tion Mn(ClO4)2·6H2O (0.18 g, 0.5 mmol) in H2O (3 mL)
was added. The reaction mixture was allowed to cool, and
ethanol was added by vapor diffusion. After 1 night colorless
crystals (290 mg, 41%) were obtained. Anal. (%) Calcd. for
C8H16O24K2MnZr: C, 13.34; H, 2.24; K,10.85; Mn, 7.62; Zr,
12.66 Found: C, 13.11; H, 2.04; K, 11.11; Mn, 7.41; Zr, 12.37 IR
(KBr, cm−1): 3587 (s), 3491 (s), 1663 (s), 1457 (s), 1372 (m), 1337
(w), 1301 (m), 1122 (w), 917 (m), 808 (s).
Experimental
Crystallographic studies
The starting compounds [K4U(C2O4)4]28 and [K4Zr(C2O4)4]34
have been prepared according to described procedures.
UO2(NO3)2, ZrOCl2, K2C2O4 were purchased from usual com-
mercial sources. All reagents were used as received.
CAUTION: Perchlorate salts are known to be potentially
hazardous and must be used with care and always in small
quantities. The 238U is an alpha emitter, the chemicals containing
this metal ion must be handled with the required attention.
Single crystal diffraction data were collected on a Nonius Kappa
˚
CCD diffractometer at 298 K (k(MoKa) = 0.71069 A). The
structures were solved by direct methods and refined by full-
matrix least-squares on F2 values using SHELXL-97.35 Crystal
data are given in Table 1. For SPA-4 the guest water molecules
inside the channels were highly disordered in such a manner
that they could not be readily resolved. No attempt was made to
place the H atoms. The exact number of water molecules (eight)
was deduced from chemical analysis.
Synthesis
CCDC reference numbers 246086, 266741 and 275039.
See http://dx.doi.org/10.1039/b503964a for crystallographic
data in CIF or other electronic format.
The X-ray powder diffractograms for SPA-1 and SPA-3
were collected using a conventional (h–2h) Phillips X-Pert
diffractometer, with kCuKa, coupled to an Anton Parr oven. All
data were collected in 5◦ < 2h < 50◦ range, with 0.02 steps and 10
seconds of exposure. The compounds were heated at 1 ◦C min−1.
[K2Mn{U(C2O4)}]·9H2O, SPA-1. K4U(C2O4)4 was synthe-
sized in situ from U(C2O4)2·6H2O (0.2 g, 0.4 mmol) and
K2(C2O4)·H2O (0.15 g, 0.8 mmol) in H2O (12 mL) at 80 ◦C. To
the hot green solution, a solution of Mn(ClO4)2·6H2O (0.15 g,
0.4 mmol) in H2O (2 mL) was added. The reaction mixture
was allowed to cool, and ethanol was slowly added without
stirring yielding SPA-1 as a micro-crystalline light-green solid
2 6 8 6
D a l t o n T r a n s . , 2 0 0 5 , 2 6 8 1 – 2 6 8 7

View Article Online
The X-ray powder diffractograms for SPA-2 were collected with
a (h–2h) Panalytical X-Pert pro MPD diffractometer equipped
with a X-celerator detector, using kCoKa1 and kCoKa2 with an Fe
filter, coupled to an Anton Parr oven. Data were collected in 5◦ <
2h < 100◦ range with 0.02 steps and 10 seconds of expositure.
The compound was heated at 1 ◦C min−1.
14 S. Tanase, F. Tuna, P. Guionneau, T. Maris, G. Rombaut, C.
Mathonie`re, M. Andruh, O. Kahn and J.-P. Sutter, Inorg. Chem.,
2003, 42, 1625–1631.
15 P. Gravereau, E. Garnier and A. Hardy, Acta Crystallogr., Sect. B,
1979, 35, 2843–2848.
16 L. G. Beauvais, M. P. Shores and J. R. Long, J. Am. Chem. Soc.,
2000, 122, 2763–2772.
17 R. Kitaura, G. Onoyama, H. Sakamoto, R. Matsuda, S. Noro and
S. Kitagawa, Angew. Chem. Int. Ed., 2004, 43, 2684–2687.
18 M. Shmilovits, Y. Diskin-Posner, M. Vinodu and I. Goldberg, Cryst.
Growth Des., 2003, 3, 855–863.
19 D. W. Smithenry, S. R. Wilson and K. S. Suslick, Inorg. Chem., 2003,
42, 7719–7721.
20 F. Tuna, S. Golhen, L. Ouahab and J.-P. Sutter, C. R. Chim., 2003,
6, 377–383.
21 P. A. W. C. Dean, D. Dance, I. Russel and V. Scudder, M., Inorg.
Chem., 2004, 43, 443–449.
22 S. R. Halper and S. M. Cohen, Inorg. Chem., 2005, 44, 486–488.
23 I. Imaz, G. Bravic and J.-P. Sutter, Chem. Commun., 2005, 993–995.
24 K. P. Mortl, J.-P. Sutter, S. Golhen, L. Ouahab and O. Kahn, Inorg.
Chem., 2000, 39, 1626–1627.
Acknowledgements
This work was supported by the French Ministry for Education
and Research by a grant to I.I. and by the CNRS. The authors
gratefully acknowledge Mrs L. Raison (Electron probe X-ray
analysis), Dr S. Pechev and Mr E. Lebraud (XRPD), Mr
D. Denux and Mr P. Dagault (TGA) and Dr R. Pailler (N2
sorption) for technical assistance.
References
1 B. Moulton and M. J. Zaworotko, Chem. Rev., 2001, 101, 1629–58.
2 C. Janiak, Dalton Trans., 2003, 2781–804.
3 B. F. Hoskins and R. Robson, J. Am. Chem. Soc., 1990, 112, 1546–54.
4 O. M. Yaghi, M. O’Keeffe, N. W. Ockwig, H. K. Chae, M. Eddaoudi
and J. Kim, Nature, 2003, 423, 705–14.
5 C. N. R. Rao, S. Natarajan and R. Vaidhyanathan, Angew. Chem.
Int. Ed., 2004, 43, 1466–1496.
6 S. Kitagawa, R. Kitaura and S.-i. Noro, Angew. Chem. Int. Ed., 2004,
43, 2334–2375.
7 A. K. Cheetham, G. Fe´rey and T. Loiseau, Angew. Chem. Int. Ed.,
1999, 38, 3268–3292.
8 K. Endo, T. Koike, T. Sawaki, O. Hayashida, H. Masuda and Y.
Aoyama, J. Am. Chem. Soc., 1997, 119, 4117–4122.
9 N. L. Rosi, J. Eckert, M. Eddaoudi, D. T. Vodak, J. Kim, M. O’Keeffe
and O. M. Yaghi, Science, 2003, 300, 1127–1129.
10 D. Bradshaw, T. J. Prior, E. J. Cussen, J. B. Claridge and M. J.
Rosseinsky, J. Am. Chem. Soc., 2004, 126, 6106–6114.
11 R. Robson, J. Chem. Soc., Dalton Trans., 2000, 3735–3744.
12 O. Kahn, Acc. Chem. Res., 2000, 33, 647–657.
13 C. Mathonie`re, J.-P. Sutter and J. V. Yakhmi, in Bimetallic magnets:
Present and perspectives, ed. J. S. Miller and M. Drillon, Wiley-VCH,
Weinheim, 2002.
25 A. L. Spek, PLATON, A Multipurpose Crystallographic Tool,
Utrecht, The Netherlands, 2001.
26 G. L. Glen, J. V. Silverton and J. L. Hoard, Inorg. Chem., 1963, 2,
250–256.
27 B. Kojic-Prodic, Z. Ruzic-Toros and M. Sljukic, Acta Crystallogr.,
Sect. B, 1978, 34, 2002–2002.
28 M. C. Favas, D. L. Kepert, J. M. Patrick and A. H. White, J. Chem.
Soc., Dalton Trans., 1983, 571–581.
29 S. Kitagawa and M. Kondo, Bull. Chem. Soc. Jpn., 1998, 71, 1739–
1753.
30 L. C. Tabares, J. A. R. Navarro and J. M. Salas, J. Am. Chem. Soc.,
2001, 123, 383–387.
31 C. Serre, F. Millange, C. Thouvenot, M. Nogue`s, G. Marsollier, D.
Loue¨r and G. Fe´rey, J. Am. Chem. Soc., 2002, 124, 13519–13526.
32 T. Loiseau, C. Serre, C. Huguenard, G. Fink, F. Taulelle, M. Henry,
T. Bataille and G. Fe´rey, Chem. Eur. J., 2004, 10, 1373–1382.
33 M.-H. Zeng, X.-L. Feng and X.-M. Chen, Dalton Trans., 2004, 2217–
2223.
34 F. A. Johnson and E. M. Larsen, Inorg. Synth., 1965, Coll. Vol. VIII,
40–44.
35 G. M. Sheldrick, in Program for the refinement of crystal structures,
University of Go¨ttingen, Germany, 1997.
D a l t o n T r a n s . , 2 0 0 5 , 2 6 8 1 – 2 6 8 7
2 6 8 7