From coordinatively weak ability of constituents to very stable alkaline-earth sulfonate metal-organic frameworks

Article abstract of DOI:10.1021/cg200078j

Four new compounds based on alkaline-earth elements ions and anthraquinone-2,6-disulfonate (2,6-AQDS) ligand were obtained and characterized: AEPF-2 is the first metal-organic framework (MOF) Mg-based disulfonate, in which the ligand directly coordinates

Full text of DOI:10.1021/cg200078j

ARTICLE
pubs.acs.org/crystal
From Coordinatively Weak Ability of Constituents to Very Stable
Alkaline-Earth Sulfonate MetalꢀOrganic Frameworks
ꢀ
Ana E. Platero-Prats, Marta Iglesias, Natalia Snejko, Angeles Monge,* and E. Gutiꢀerrez-Puebla
Instituto de Ciencia de Materiales de Madrid, ICMM-CSIC, Madrid, Spain
S Supporting Information
b
ABSTRACT: Four new compounds based on alkaline-earth
elements ions and anthraquinone-2,6-disulfonate (2,6-AQDS)
ligand were obtained and characterized: AEPF-2 is the first
metalꢀorganic framework (MOF) Mg-based disulfonate, in
which the ligand directly coordinates with Mg2þ ions to build a
two-dimensional net; AEPF-3, AEPF-4, and AEPF-5 for Ca-
(II), Sr(II), and Ba(II) ions, respectively, have different three-
dimensional structural types and two new topologies. The
catalytic behavior of these alkaline-earth materials in alkenes
hydrogenation and ketones hydrosilylation reactions makes
them cheap and environmentally friendly promising catalysts:
in all cases, 100% selectivity was achieved toward ethylbenzene,
in contrast with the behavior previously reported with calcium and strontium homogeneous catalysts. The high thermal stability (up
to 500 °C) is to be emphasized for all these materials.
’ INTRODUCTION
of arene-sulfonate groups11ꢀ13 as linkers in the search of multi-
functional materials, we report here four novel two- and three-
dimensional (2D and 3D) alkaline-earth polymeric frameworks
(AEPF) belonging to four different structural types, in which the
metal centers are coordinated to anthraquinone-2,6-disulfonate
(2,6-AQDS) ligands. The structure and properties of the four
resulting compounds are discussed, and the results of the catalytic
activity of these materials in alkenes hydrogenation and ketones
hydrosilylation reactions under mild conditions are also reported.
The immense majority of coordination polymers concerns
transition metals, and, more recently, rare earth cations; however,
s-blocks metals have received relatively less attention,1 probably
because as it happened with lanthanide elements, they are usually
regarded as unsuitable metal centers to form such polymers. Two
reasons prevent the use of s-blocks in the metalꢀorganic frame-
works (MOFs) design: their unpredictable coordination num-
bers and geometries as no ligand field stabilization effects govern
their coordinative bonding, and their tendency to form solvated
metal centers, and thus, the formation of the typical alternating
[M(H2O)6] L organicꢀinorganic ionic layers.
’ EXPERIMENTAL SECTION
General Information. All reagents were purchased (AR grade)
from Aldrich and used without further purification. IR spectra were
recorded from KBr pellets in the range 4000ꢀ400 cmꢀ1 on a Perkin-
Elmer spectrometer. Thermal gravimetric analyses (TGA) were per-
formed using a MettlerꢀToledo apparatus, between 25 and 1000 °C in
air (flow rate 50 mL minꢀ1, heating rate 10 °C minꢀ1).
Synthesis. The compounds were synthesized under hydrothermal
conditions, by heating the reactant mixture at 170 °C for 72 h in a
Teflon-lined stainless steel autoclave. After that time, the autoclave was
cooled to room temperature; the product was filtered and washed with
distilled water and acetone. Typically, 0.24 mmol of 2,6-AQDSNa2 and
0.24 mmol of alkaline-earth acetate were dissolved in a mixture of 5 mL
of water (278.8 mmol) and 5 mL of 1-butanol (54.6 mmol). In all cases,
the products were isolated as pure phases: yellow crystals in all cases
were obtained (AEPF-2 (65% yield), AEPF-3 (70% yield), AEPF-4
On the other hand, in the broad domain of MOFs research,
networks based on sulfonate ligation are much less studied,
despite the fact that interesting materials with layered structures
have been reported.2 Because of the weak coordination strength
of the sulfonate functionality, many of them do not displace water
from the primary coordination sphere. In the case of the Group 2
metals, only the larger alkaline-earth elements coordinate to both
water molecules and sulfonate groups.3 Although some research
groups have published MOF-type alkaline earth-based carboxy-
lates,3ꢀ5 not many networks based on sulfonate coordination
have been reported.1 The supramolecular chemistry of the sulfo-
nate group in extended solids was reviewed by C^ote and Shimizu,6
and that of the metal arenesulfonates, by Cai.7 Structures of
compounds of Group 1 and Group 2 metal ions with 1,5-naph-
thalenedisulfonate (1,5-NDS) are also found;2 recently, a new
review by Shimizu et al. shows that for sulfonates, hard metal ions
remain highly hydrated and typically results in zero- or one-
dimensional (0D or 1D) structures.8ꢀ10 In our ongoing studies
Received: December 17, 2010
Revised:
February 14, 2011
Published: March 14, 2011
r
2011 American Chemical Society
1750
dx.doi.org/10.1021/cg200078j Cryst. Growth Des. 2011, 11, 1750–1758
|

Crystal Growth & Design
ARTICLE
Table 1. Crystallographic Data
AEPF-2
AEPF-3
AEPF-4
AEPF-5
formula
C
14H10MgS2O10
C14H6CaS2O8
406.41
C14H8SrS2O9
471.96
C14H6BaS2O8
503.66
formula weight
temperature (K)
wavelength (Å)
crystal system
space group
426.67
296(2)
296(2)
296(2)
296(2)
0.71073
triclinic
0.71073
monoclinic
P2(1)/c
14.863(3)
5.5805(11)
9.3076(18)
90.00
0.71073
triclinic
0.71073
orthorhombic
Pnma
P1
P1
ꢀ
a/Å
4.8682(5)
5.7700(6)
13.4652(15)
94.500(2)
90.517(2)
95.403(2)
375.34(7)
1
5.4254(13)
10.219(2)
14.537(3)
103.584(4)
99.853(4)
95.790(4)
763.4(3)
2
29.1296(11)
6.3329(3)
7.8329(3)
90.00
ꢀ
b/Å
ꢀ
c/Å
R/°
β/°
γ/°
volume /Å3
105.968(3)
90.00
90.00
90.00
742.2(3)
2
1444.97(10)
4
Z
calc density/g cmꢀ3
1.888
1.818
2.053
2.315
μ/mmꢀ1
0.459
0.749
3.857
3.083
dimensions (mm)
0.15 ꢁ 0.10 ꢁ 0.04
ꢀ5 < h < 5
ꢀ6 < k < 6
ꢀ15 < l < 15
218
0.20 ꢁ 0.10 ꢁ 0.05
ꢀ19 < h < 19
ꢀ7 < k < 7
ꢀ12 < l < 12
412
0.20 ꢁ 0.12 ꢁ 0.08
ꢀ6 < h < 6
ꢀ11 < k < 12
ꢀ17 < l < 16
468
0.20 ꢁ 0.12 ꢁ 0.08
ꢀ37 < h < 38
ꢀ8 < k < 8
ꢀ23 < l < 23
968
limiting indices h
k
l
F(000)
reflections collected/unique with I > 2σ(I)
2682/1292
132
6281/1796
115
5095/2565
243
11404/1838
145
refined parameters
goodness-of-fit on F2
1.093
1.058
1.015
1.204
R1
0.0295
0.0499
0.0614
0.0281
wR2
0.0749
0.1376
0.1049
0.0544
(62% yield), AEPF-5 (65% yield)). The purity of the products was
tested by comparing the experimental and simulated powder X-ray
diffraction (PXRD) patterns (see Supporting Information, Section S5)
and through the elemental CHNS analyses. Calculated for AEPF-2: C,
39.41; H, 2.35; S, 15.03. Found: C, 38.69; H, 2.46; S, 14.49. Calculated
for AEPF-3: C, 41.37; H, 1.48; S, 15.78. Found: C, 41.43; H, 1.64; S,
15.02. Calculated for AEPF-4: C, 35.63; H, 1.70; S, 13.56. Found: C,
35.49; H, 1.73; S, 13.25. Calculated for AEPF-5: C, 33.38; H, 1.19; S,
12.71. Found: C, 33.38; H, 1.23; S, 12.25.
Powder X-ray Diffraction. PXRD measurements were perfor-
med with a Bruker D8 diffractometer in the θꢀθ mode using nickel-
filtered Cu KR1,2 (λ = 1.5418 Å) radiation. The best counting statistics
were achieved by using a scanning step of 0.02° taken between 5 and
30° Bragg angles with an exposure time of 0.5 s per step.
Structure Determination by Single Crystal X-ray Diffrac-
tion. A summary of the main crystal and refinement data for the four
compounds is given in Table 1. Data for single crystals of AEPF-2,
AEPF-3, AEPF-4, and AEPF-5 were collected in a Bruker SMART
CCD diffractometer equipped with a normal focus, 2.4 kW sealed tube
X-ray source (MoKR radiation = 0.71073 Å). Data were collected over a
hemisphere of the reciprocal space by a combination of three sets of
exposures. Each exposure of 20 s covered 0.3° in φ. Unit cell dimensions
were determined by a least-squares fit of 60 reflections with I > 2σ(I).
Main crystallographic data are given in Table 1. The structures were
solved by direct methods. The final cycles of refinement were carried out
by full-matrix least-squares analyses with anisotropic thermal parameters
for all non-hydrogen atoms. All hydrogen atoms were placed in geomet-
rically calculated positions and subsequently refined using a riding model
with Uiso(H) = 1.2Ueq(C), except in the case of coordinated water
molecules, for which hydrogen atoms were located from difference
Fourier maps. Calculations were carried out with SMART software for
data collection and data reduction and SHELXTL.14 CCDC reference
numbers 804863, 804864, 804865, 804866 contain the supplementary
crystallographic data for this paper. These data can be obtained free of
charge from the Cambridge Crystallographic Data Center.
Topological Analyses. Topological analyses were performed using
TOPOS software.15 For AEPF-3 and AEPF-4, whose nets have an equili-
brium placement with no vertex collisions, it was possible to idealize them
in their maximum symmetry embedding using Systre software.16,17
Catalytic Activity Experiments. For both Lewis-acidic catalytic
reactions, time-over-frequency (TOF) values were calculated from first
reaction run data points in the batch experiments (after 15 reaction minutes).
(i) Alkenes Hydrogenation. The catalytic properties of these
materials in the hydrogenation of styrene to form ethyl benzene (see
Supporting Information, Scheme S1) were examined under mild con-
ditions (toluene, 373 K, 5 atm H2) using 1 mol % amount of catalyst.
Reaction mixture was monitored by gas chromatography (GC) using a
Hewlett-Packard 5890 II, coupled with a mass detector, and a methylsi-
licone (ov-1701) and permethylcyclodextrine column. (ii) Ketones
Hydrosilylation. The catalytic properties of these materials in the
hydrosilylation of benzaldehyde with diphenylsilane (see Supporting
Information, Scheme S2) were examined using 10% mol amount of
catalyst and an optimal ratio of carbonyl/diphenylsilane (1:2) under
mild conditions (toluene, 363 K, nitrogen atmosphere). The reaction
mixture was analyzed by GC using a KONIK HRGC 40000B chroma-
tograph equipped with an FID detector and a KAP-120212 column.
Recycling Experiments. The best catalyst for each studied
reaction was chosen to perform recycling experiments. Thus, AEPF-2
was reused in three consecutive styrene hydrogenation cycles and AEPF-3,
in benzaldehyde hydrosilylation. Before reuse, the solid was separated from
1751
dx.doi.org/10.1021/cg200078j |Cryst. Growth Des. 2011, 11, 1750–1758

Crystal Growth & Design
ARTICLE
Figure 1. (a) Inorganic layer for each compound and its topological simplification. (b) Topological net simplification for each compound.
Figure 2. Polyhedral representation for (AEPF-2): (a) Layers perpendicular to the b direction, where MgO6 octahedra form chains along the a
direction; (b) representation of interlayer hydrogen bonds along the a direction; and (c) representation of interlayer hydrogen bonds along the c
direction.
the reaction medium by centrifugation, washed with toluene, and dried at
120 °C under a vacuum; then fresh substrates and solvent were added, in
three consecutive experiments. To verify the heterogeneous nature of the
catalytic reaction, the residual activity of the supernatant solution after
separation of the catalyst was studied. To rule out a potential leaching, the
liquid phase of the first run was separated from the catalyst. New fractions
of reagents were added to the clear filtrate, and the composition of the
homogeneous reaction was determined by GC. This mixture was used in a
standard catalytic experiment. After 3 h, the mixture composition was
checkedandnoreactionwasobservedfor eachexperiment, whichexcluded
the presence of active catalytic species in solution.
four compounds are given in Tables S1ꢀS4, Supporting Infor-
mation. Figure 1 summarizes the topological analyses perfor-
med. Details of topological analyses performed for the four
compounds are shown in Figures S5ꢀS8, Supporting Informa-
tion. The molecular formulas of the four compounds are [Mg-
(2,6-AQDS)(H2O)2] (AEPF-2), [Ca(2,6-AQDS)] (AEPF-3),
[Sr(2,6-AQDS)(H2O)] (AEPF-4), and [Ba(2,6-AQDS)]
(AEPF-5).
In [Mg (2,6-AQDS) (H2O)2] (AEPF-2), Mg2þ ion is hexa-
coordinated to four oxygen atoms coming from two (2,6-AQDS)
linkers, and to two oxygen atoms of two different water molecules
to give MgO6 octahedra, which are connected through the S atoms
forming chains along the a direction (Figure 2). Junction among
chains is made via the complete linker, which acts in a η2μ2-η2μ2
coordination mode giving rise to layers perpendicular to the b
direction and thus to a 2D structure. From the topological point
’ RESULTS AND DISCUSSION
Crystal Structure Description and Topological Analyses.
Details of the unit cells, data collection, and refinement for the
1752
dx.doi.org/10.1021/cg200078j |Cryst. Growth Des. 2011, 11, 1750–1758

Crystal Growth & Design
ARTICLE
Figure 3. Polyhedral representation for AEPF-3: (a) inorganic layer; (b, c) three-dimensional net.
of view, S atoms were considered as three-connected nodes
instead of analyzing the whole ligand as a unique node. We think
that this kind of simplification, which has been applied for the
four compounds, is chemically clearer than other possibilities.
Thus, by simplification of the layers, two types of nodes can be
found: four-connected (Mg) and three-connected (S) nodes.
This two-periodic net exhibits a 3,4L13 (point symbol (4.62)2-
(42.62.82)) topology (see Supporting Information, Figure S5b).
The layers are bonded through three types of hydrogen bonds
among the coordinated water molecules and the 2,6-AQDS
oxygen atoms (one strong: d: O1ꢀO3 = 2.748(2) Å, O3 H1B
=
=
3 3 3
Figure 4. Polyhedral representation for compound (AEPF-4) (a) inor-
ganic layer and (b) three-dimensional net.
1.81(2) Å; two weak: d: O1ꢀO2 = 3.060(2) Å, O2 H1A
3 3 3
2.40(4) Å; d: O3ꢀO1 = 3.355(3) Å, O3 H1A = 2.47(2) Å),
3 3 3
along the b direction. Taking into account these interactions, the
supramolecular structure would be 3D, exhibiting now the typical
inorganic layer described in this kind of compound. By simpli-
fication of these inorganic layers, two types of nodes can be
found: eight-connected (Mg) and four-connected (S) nodes,
giving rise to inorganic layers with point symbol (420.68)(46)2
(Figure 4a). Surprisingly, this layer type does not appear in
TOPOS nor in the RCSR databases. Joining of these layers
through the three-connected nodes gives rise to a three-periodic
net, which exhibits a 5,8T16 topology (point symbol (420.68)
(46.64)2) (Figure 1b).
joining of the inorganic kgd layers in AEPF-4 gives rise to a new
net with point symbol (43.63)2(46.67.82). In order to clarify this
point, we performed the idealization of both nets with Systre
software (see Experimental Section).16 Once both structures
were idealized in their maximum symmetry embedding, com-
parative topological studies were performed to understand the
observed structural differences in their nets. Our analyses showed
that the different topologies found can be explained in terms of
the different distortions present in the inorganic layers for both
compounds. Considering a six-member ring of the kgd net to
describe the inorganic layers (Figure 5a), we determined differ-
ent “conformations” (Figure 5b): for AEPF-3, a “chair” con-
formation; for AEPF-4, a “boat” conformation. This explains the
different packing modes of the kgd layers in the final struc-
tures (Figure 5c): A-A type packing (fsh net) in AEPF-3, and
A-B type packing (new topological type net with point symbol
(43.63)2(46.67.82)) in AEPF-4. Those differences are due to the
change in the bond directionalities because of the presence of a
coordinated water molecule in the Sr coordination polyhedron.
Hydrogen bonds among the coordinated water molecules and the
coordinated 2,6-AQDS oxygen atoms (d: O1ꢀO3 = 2.845(7) Å,
This is, to the best of our knowledge, the first MOF-type Mg-
based disulfonate, in which the ligand directly coordinates with
Mg2þ ion to build a 2D net.
In [Ca(2,6-AQDS)] (AEPF-3), Ca2þ ions also occupy the
centers of regular CaO6 octahedra, but in this case no water
molecules are coordinated, and each Ca atom is bonded to six
sulfonate oxygen atoms. The sulfonate linker acts in a hex-
atopic η3μ3-η3μ3 coordination mode (Figure 3). The inorganic
layer is formed by six-connected (Ca) and three-connected
(S) nodes, to give rise to a 2D net of the type kgd (Figure 1a).
Joining of these layers through the three-connected nodes
gives rise to an fsh type 3D net with point symbol (43.63)2
(46.66.83) (Figure 1b).
O3 H1A = 1.86(2) Å) and (O1ꢀO8 = 3.090(7) Å, O8 H1B
=
3 3 3
3 3 3
2.39 Å) (Figure 6) might also have some influence in the AEPF-4
net distortion.
[Sr(2,6-AQDS)(H2O)] (AEPF-4). The Sr coordination
polyhedron is a monocapped trigonal prism. The water molecule
is situated in one of the prism squared faces. In this compound,
the ligand also acts in a hexatopic η3μ3-η3μ3 coordination mode
(Figure 4). Since the water molecule does not participate in con-
nectivity, the inorganic layer is quite similar to that of AEPF-3. It
is again of the type kgd formed by six-connected (Sr) and three-
connected (S) nodes, but some subtle differences appear when
connecting them through the three-connected nodes to give rise
to the 3D net. In fact, contrary to AEPF-3 with a fsh net topology,
[Ba(2,6-AQDS)] (AEPF-5) exhibits a 3D net, with the
AQDS2ꢀ anions coordinated to the Ba atoms (Figure 7). Dif-
ferent from the former compounds and given the capability of the
Ba cation to have different coordination numbers and ways of
coordination, the alkaline-earth ion in AEPF-5 is octa-coordi-
nated to oxygen atoms, belonging to sulfonate groups, giving a
BaO8 triangulated dodecahedron. It is worth noticing that in this
compound the 2,6-AQDS ligand acts as a hepta-topic η3μ4-η3μ3
linker, since one sulfonate oxygen atom is shared by two Ba2þ
1753
dx.doi.org/10.1021/cg200078j |Cryst. Growth Des. 2011, 11, 1750–1758

Crystal Growth & Design
ARTICLE
Figure 5. The figure depicts a comparative topological analysis for both fsh ideal net (left) and for AEPF-3 (middle) and AEPF-4 (right) idealized nets.
(a) Six-member ring defined to describe the different kgd nets found in fsh net, AEPF-3 and AEPF-4; (b) joining of two six-member rings; (c) different
packing patterns: fsh net when “chair” conformation is found in the kgd layers of AEPF-3 (left, middle); the new topological type net of AEPF-4 with
“boat” conformation (right).
Figure 6. Hydrogen bonds in of AEPF-4 between coordinated water
molecules, and the coordinated 2,6-AQDS oxygen atoms.
cations. The inorganic layer is formed by seven-connected (Ba),
four-connected (S), and three-connected (S) nodes, to give rise
to inorganic layers with point symbol (411.610)(43)(45.6)
(Figure 1). Joining of these layers through the four- and three-
Figure 7. Polyhedral representation for AEPF-5: (a) (Left) BaO8 trian-
gulated dodecahedron and inorganic layer; (b, c) three-dimensional net.
connected nodes gives rise to a 3D net with point symbol
(411.610)(43.63)(45.65). For this structure, the topological types
of neither the inorganic layer nor the 3D net appear in TOPOS or
in the RCSR databases.
down Group 2, the alkaline-earth ions form fewer bonds with
water and more with SO3ꢀ groups, which can be rationalized in
terms of hardꢀsoft acidꢀbase principles. On the other hand, an
alternative explanation can be based on electronegativities. Thus,
moving down Group 2, as the metals become more ionic in
Role of the Alkaline-Earth Ion. As it has been previously
shown for this kind of alkaline-earth disulfonate, different
degrees of dimensionality are obtained depending on the metal
used.9,10 This fact can be explained as follows. First, moving
1754
dx.doi.org/10.1021/cg200078j |Cryst. Growth Des. 2011, 11, 1750–1758

Crystal Growth & Design
ARTICLE
character, the interactions with SO3ꢀ groups are favored over the
interactions with water molecules.
AEPF-4. Dehydration proceeds in one stage with a weight loss
of ∼4.0% at ∼150 °C consistent with the removal of the water
molecules present in the compound (calc. 3.8%). The second
step at ∼490 °C corresponds to the complete decomposition of
the anhydrous compound. The calculated weight loss for the
whole process (61.1%) is in good agreement with the experi-
mental value of 60.9%. PXRD pattern shows strontium sulfate
(celestine) as a main final product (PDF = 5-593).
Mg2þ ion (hexa-coordinated), which can be considered as a
hard acid ion, is normally present as a 0D hexaaquo species that
forms charge-assisted hydrogen bonds with sulfonate ligands.2ꢀ10
However, by using solvothermal conditions, we were able to
obtain the first MOF-type Mg based disulfonate, in which the
ligand directly coordinates with Mg2þ ions to build a 2D net
(AEPF-2).
AEPF-5. The complete decomposition of the compound
proceeds in one step at ∼518 °C. The calculated weight loss
for the whole process (53.7%) is in good agreement with the
experimental value of 52.9%. The PXRD pattern shows barium
sulfate (Barite) as a main final product (PDF = 24-1035).
Catalytic Activity Tests. Alkenes Hydrogenation. The
catalytic hydrogenation of unsaturated compounds represents
one of the earliest examples in heterogeneous as well as homo-
geneous catalysis.18 However, because of its high industrial
potential, this particular conversion is even currently intensively
investigated.19 Nowadays, a new class of hydrogenation catalysts
is required to develop cheap and environmentally friendly alter-
natives, because traditional hydrogenation catalysts are based on
precious metals (Pt, Pd, Rh, Ir, Re).20 In fact, non-transition
metal catalysts such as NaH, KH, MgH2, or LiAlH4 are indust-
rially used, although they need forcing conditions (150ꢀ225 °C,
60ꢀ100 bar H2) and give predominantly oligomeric or poly-
meric subproducts.21 In this way, the possibility to use alkaline-
earth elements based catalysts in hydrogenation processes has
been recently reviewed by Harder.22 Two years earlier, the good
performance of calcium and strontium homogeneous catalysts in
alkenes hydrogenation has been also shown.23
Continuing with our previous studies on the catalytic activity
of alkaline-earth MOFs, in which we showed the high catalytic
activity in hydrogenation processes under mild conditions of a
flexible calcium MOF,24,25 AEPF-2, AEPF-3, AEPF-4, and
AEPF-5 were tested as catalysts for the hydrogenation of alkenes
using styrene as substrate model under mild conditions (see
Experimental Section, Section S6). The obtained results are
summarized in Figure 8 and Table 2. The best results were
obtained with AEPF-2, for which the total and selective hydro-
genation of styrene to ethylbenzene was achieved after 4 h (TOF =
38.8 hꢀ1). It is worth highlighting that in all cases 100% selec-
tivity was achieved toward ethylbenzene, in contrast with the
behavior of previously reported calcium and strontium homo-
geneous catalysts.23
For Ca2þ (hexa-coordinated) and Sr2þ (hepta-coordinated)
ions, 3D nets with related topologies were obtained (AEPF-3
and AEPF-4 , respectively) (Figures 4 and 7). The increase in the
net dimensionality is explained on the basis of the decrease in the
charge/radius ratio. It is worth mentioning that both ions present
different coordination numbers: 6 for Ca2þ and 7 in the case of
Sr2þ, the latter additionally coordinating to one water molecule
decorating the net. Thus, in AEPF-4 the presence of coordinated
water molecules changes the bond directionalities in the inor-
ganic layer, giving rise to a new net with point symbol (43.63)2-
(46.67.82).
For Ba2þ (octa-coordinated) ion, a 3D net has been obtained
(AEPF-5) (Figures 6 and 7). The remarkable increase in the
polarizability and atomic radius for Ba2þ ion results in an increase
in its number of coordination. AEPF-5 exhibits a more compli-
cated net with point symbol (411.610)(43.63)(45.65), which
corresponds, as in the case of AEPF-4, to a new topological
type. On the other hand, the presence of a hepta,- tetra-, and
three-connected three-nodal inorganic layer in AEPF-5 forces
the aromatic rings to be much closer than in the other
compounds, giving rise to πꢀπ, and π
O stacking interac-
tions, along the (010) direction (distances of 3.15 and 3.76 Å,
respectively).
3 3 3
Infrared Spectroscopy Studies. IR spectra of AEPF-2,
AEPF-3, AEPF-4, and AEPF-5 show the presence of both υas
and υs of the SO3 groups found at 1240ꢀ1180 (υas) and
1100ꢀ1040 (υs) cmꢀ1, the aromatic CꢀH stretching mode
being situated in the area of 3100ꢀ2860 cmꢀ1 (see Support-
ing Information, Figures S9ꢀ10). For AEPF-2 and AEPF-4,
IR spectra show additional bands in the ꢀOH stretching mode
region (3750ꢀ3200 cmꢀ1), which correspond to the coordi-
nated water molecules present in these materials (see Sup-
porting Information, Figure S9).
Thermal Behavior. TGA curves for AEPF-2, AEPF-3,
AEPF-4, and AEPF-5 are shown in Supporting Information
(Figures S15ꢀS16). The exceptional thermal stability is to be
emphasized for all four MOF materials. Thus, TGA results for the
thermal decomposition for the four compounds reveal that their
frameworks are stable up to ∼500 °C in air; above this
temperature the frameworks decompose completely.
AEPF-2. Dehydration proceeds in one stage with a weight loss
of ∼8.5% at ∼180 °C consistent with the removal of the water
molecules present in the compound (calc. 8.4%). The second
step at ∼500 °C corresponds to the complete decomposition of
the anhydrous compound. The calculated weight loss for the
whole process (90.6%) is in good agreement with the experi-
mental value of 88.5%. The PXRD pattern shows magnesium
oxide as a main final product (PDF = 45-946).
AEPF-3. The complete decomposition of the compound
proceeds in one step at ∼500 °C. The calculated weight loss
for the whole process (66.5%) is in good agreement with the
experimental value of 67%. The PXRD pattern shows calcium
sulfate (anhydrite) as a main final product (PDF = 37-1496).
All catalysts employed in styrene hydrogenation were recov-
ered by centrifugation, washed with toluene, and then character-
ized by PXRD. Comparison of the PXRD patterns before and
after the catalytic reaction confirm the robustness of the four
compounds in the studied conditions (see Supporting Informa-
tion, Figures S19ꢀS22).
A mechanism for the alkaline-earth MOF-mediated hydro-
genation, based on the H2 heterolytic cleavage, could be analo-
gous to that proposed by Harder et al. for soluble calcium hydride
complexes,22,23 in which the catalytic activity is explained by the
formation of Ca-hydride intermediate species. Taking this into
account, the observed decrease in the catalytic activity in the
AEPF-2 > AEPF-3 > AEPF-4 > AEPF-5 direction can be
explained by the loss of Lewis acidity in the Mg(II) > Ca(II) >
Sr(II) > Ba(II) direction (Figure 8 and Table 2).
Recycling Experiments. Alkenes Hydrogenation. To in-
vestigate the lifetime and the stability of the better catalyst in
styrene hydrogenation (AEPF-2), a recycling experiment was
1755
dx.doi.org/10.1021/cg200078j |Cryst. Growth Des. 2011, 11, 1750–1758

Crystal Growth & Design
ARTICLE
Figure 8. (Left) Kinetic profiles of styrene hydrogenation with AEPF-2, AEPF-3, AEPF-4, and AEPF-5 as catalysts. (Right) Kinetic profiles in three
consecutive reaction cycles employing AEPF-2 as catalyst.
Table 2. Styrene Hydrogenation in Toluene, at 373 K, 5 atm
H2, Using 1 mol % Catalyst
Table 3. Benzaldehyde Hydrosilylation with Diphenylsilane
in Toluene, at 363 K, Using 10 mol % Catalyst
compound
active center
time (h)
conversion (%)
TOF (hꢀ1
)
compound
active center
time (h)
conversion (%)
TOF (hꢀ1
)
AEPF-2
AEPF-3
AEPF-4
AEPF-5
Mg(II)
Ca(II)
Sr(II)
4
5
5
5
100
100
68.2
64.7
38.8
12.3
14.3
36.5
AEPF-2
AEPF-3
AEPF-4
AEPF-5
Mg(II)
Ca(II)
Sr(II)
22
22
22
22
34.6
85.3
65.5
39.5
0.21
1.52
1.41
0.24
Ba(II)
Ba(II)
catalytic activity was clearly lower (silylated product yield lower
than ∼40% after 22 h).
performed employing it in three consecutive reaction cycles.
The observed activity is kept over the three cycles of reaction
(Figure 8 right). A blank run with magnesium acetate tetra-
hydrate replacing AEPF-2 was also carried out, in order to
compare both activity and robustness (Figure S23, Supporting
Information).
The catalyst employed in the three consecutive cycles of
reaction was recovered as explained before, and then character-
ized by PXRD. The comparison of the PXRD patterns of the
catalyst before and after the recycling allows us to confirm that
the compound is unaltered after the three cycles of reaction (see
Supporting Information, Figure S19).
Catalytic Activity Tests. Ketones Hydrosilylation. Hydro-
silylation reactions are considered as a highly atom-efficient
key transformation, which has great importance for the silicon
industry and organic synthesis, dendrimer, and polymer chemistry.26
In particular, catalytic hydrosilylation of ketones is a convenient
one-step procedure to prepare protected alcohols. The mechan-
ism is similar to that of the hydrogenation reaction, and generally
similar catalysts are employed for the two catalytic processes.
Recently, main group metal homogeneous catalysts have been
introduced for this reaction.22
All the catalysts employed in hydrosilylation of benzaldehyde
with diphenylsilane were recovered as explained before and then
characterized by PXRD. The comparison of the PXRD patterns
of the catalysts before and after the reaction allows us to confirm
the robustness of the four compounds in the studied conditions
(see Supporting Information, Figures S24ꢀS27).
The mechanism proposed for ketones hydrosilylation usually
assumes the intermediacy of a metal complex that contains a
hydride, a silyl ligand (R3Si), and the ketone substrate.22,23
Taking into account this type of mechanism, two different kinds
of requirements are needed: (i) Lewis acidity, that favors the
formation of hydride intermediate species and (ii) capability of
the catalytically active alkaline-earth center to increase its co-
ordination number and give rise to the metal complex inter-
mediate species. Thus, in the studied conditions, the observed
decrease in the catalytic activity in the AEPF-3 > AEPF-4 >
AEPF-5 direction can be explained with the loss of Lewis
acidity in the Ca(II) > Sr(II) > Ba(II) direction. For AEPF-2,
in which Mg(II) is the active center, the low activity is explained
in terms of the substrate steric impediment related to the Mg
ion size.
Continuing with our previous work,24 AEPF-2, AEPF-3,
AEPF-4, and AEPF-5 were tested as ketones hydrosilylation
catalysts in the hydrosilylation of benzaldehyde with diphenylsi-
lane (see Experimental Section). The obtained results are shown
in Table 3, Figure 9 left. The best result was obtained with
AEPF-3, for which the silylated product yield was 85.3% after 22
h (TOF = 1.52 hꢀ1). A similar behavior was obtained with
AEPF-4 (silylated product yield 65.5% after 22 h, TOF = 1.41
Recycling Experiments. Ketones Hydrosilylation. To in-
vestigate the lifetime and the stability of the better catalyst in
benzaldehyde hydrosilylation with diphenylsilane (AEPF-3), a
recycling experiment was performed employing it in three conse-
cutive reaction cycles. The observed activity is kept over the three
cycles of reaction, being slightly decreased in the third one
(Figure 9).
The catalyst employed in the three consecutive cycles of
reaction was recovered as explained before, washed with
h
ꢀ1). However, in the case of both AEPF-2 and AEPF-5, the
1756
dx.doi.org/10.1021/cg200078j |Cryst. Growth Des. 2011, 11, 1750–1758

Crystal Growth & Design
ARTICLE
Figure 9. (Left) Kinetic profiles for bezaldehyde hydrosilylation with diphenylsylane with AEPF-2, AEPF-3, AEPF-4, and AEPF-5 as catalysts. (Right)
Kinetic profiles in three consecutive reaction cycles employing AEPF-3 as catalyst.
toluene, and then characterized by PXRD. The comparison of
the PXRD patterns of the catalyst before and after the recyc-
ling allows us to confirm that the compound is unaltered after
the three cycles of reaction (see Supporting Information,
Figure S25).
’ AUTHOR INFORMATION
Corresponding Author
*E-mail: amonge@icmm.csic.es.
’ ACKNOWLEDGMENT
This work has been supported by the Spanish MCYT Project
Mat 2007-60822, Mat 2010-17571, CAM S2009/MAT-1756/
CAM, and Consolider-Ingenio CSD2006ꢀ2010. A.E.P.P. ac-
knowledges the JAE fellowship from CSIC and Fondo Social
Europeo from EU. The authors thank Prof. Davide M. Proserpio
and Dr. Felipe Gꢀandara for their help with the topological
analysis.
’ CONCLUSIONS
In summary, by using appropriate solvothermal conditions,
the four new compounds, based on coordinatively “weak”
alkaline-earth elements cations and anthraquinone-2,6-disulfo-
nate (2,6-AQDS) anions, were obtained and characterized. In the
novel first MOF-type Mg-based disulfonate, AEPF-2, the ligand
coordinates directly with hexa-coordinated Mg2þ ion to build a
layered 2D structure; these layers are bonded through hydrogen
bonds giving 3D supramolecular structure. In AEPF-3, AEPF-4,
and AEPF-5 for Ca2þ, Sr2þ, and Ba2þ ions respectively, the
structures possess different 3D structural types and represent two
new topologies (for AEPF-4 and AEPF-5).
The exceptional thermal stability (up to 500 °C) is to be
emphasized for all these MOF compounds. On the other hand,
and although these frameworks do not allow the substrate
accessibility to the interior, and thus the catalytic reactions take
place on the surface, the new compounds exhibit activity as
alkene hydrogenation and ketones hydrosilylation hetero-
geneous catalysts in mild conditions, AEPF-2 being the best
one for the hydrogenation of styrene. It is worth highlighting that
in all cases 100% selectivity was achieved toward ethylbenzene, in
contrast with the behavior previously reported for calcium and
strontium homogeneous catalysts. AEPF-3 is the most active
catalyst for benzaldehyde hydrosilylation with diphenylsilane.
’ REFERENCES
(1) C^otꢀe, A. C.; Shimizu, G. K. H. Chem.—Eur. J. 2003, 9, 5361–5370
and references therein.
(2) (a) Russell, V. A.; Etter, M. C.; Ward, M. D. J. Am. Chem. Soc.
2004, 116, 1941–1952. (b) Pivovar, A. M.; Holman, K. T.; Ward, M. D.
Chem. Mater. 2001, 13 (9), 3018–3031.
(3) Cai, J. W.; Chen, C. H.; Liao, C. Z.; Feng, X. L.; Chen, X. M. Acta
Crystallogr. 2001, B57, 520–530.
(4) (a) Dincꢁa, M.; Long, J. R. J. Am. Chem. Soc. 2005, 127 (26),
9376–9377. (b) Davies, R. P.; Less, R. J.; Lickiss, P. D.; White, A. J. P.
Dalton Trans. 2007, 24, 2528–2535. (c) Senkovska, I.; Fritsch, J.; Kaskel,
S. Eur. J. Inorg. Chem. 2007, 35, 5475–5479. (d) Dietzel, P. D. C.; Blom,
ꢂ
R.; Fjellvag, H. Eur. J. Inorg. Chem. 2008, 23, 3624–3632.
(5) (a) de Lill, D. T.; Bozzuto, D. J.; Cahill, C. L. Dalton Trans. 2005,
ꢂ
12, 2111–2115. (b) Birkedal Nielsen, R. K.; Kongshaug, K. O.; Fjellvag,
H. Solid State Sci. 2006, 8 (10), 1237–1242. (c) Volkringer, C.; Marrot,
J.; Fꢀerey, G.; Loiseau, T. Cryst. Growth Des. 2008, 8 (2), 685–689.
(6) C^ote, A. P.; Shimizu, G. K. H. Coord. Chem. Rev. 2003,
245, 49–64.
(7) Cai, J. Coord. Chem. Rev. 2004, 248, 1061–1083 and references
therein.
’ ASSOCIATED CONTENT
(8) Shimizu, G. K. H.; Vaidhyanathan, R.; Taylor, J. M. Chem. Soc.
Rev. 2009, 38, 1430–1449.
(9) C^otꢀe, A. P.; Shimizu, G. K. H. Chem.—Eur. J. 2003, 9 (21),
S
Supporting Information. Structural analyses by single
b
crystal X-ray diffraction, topological analyses, infrared spectros-
copy analyses, thermal gravimetric analyses, Pawley refinements,
and catalytic studies for AEPF-2, AEPF-3, AEPF-4, and AEPF-5.
This material is available free of charge via the Internet at http://
pubs.acs.org.
5361–5370.
(10) Kennedy, A. R.; Kirkhouse, J. B. A.; McCarney, K. M.;
Puissegur, O.; Smith, W. E.; Staunton, E.; Teat, S. J.; Cherryman,
J. C.; James, R. Chem.—Eur. J. 2004, 10 (18), 4606–4615.
1757
dx.doi.org/10.1021/cg200078j |Cryst. Growth Des. 2011, 11, 1750–1758

Crystal Growth & Design
ARTICLE
(11) Gꢀandara, F.; Fortes-Revilla, C.; Snejko, N.; Gutiꢀerrez-Puebla,
E.; Iglesias, M.; Monge, M. A. Inorg. Chem. 2007, 46, 3475–3484.
(12) Gꢀandara, F.; Perles, J.; Snejko, N.; Iglesias, M.; Gꢀomez-Lor, B.;
Gutiꢀerrez-Puebla, E.; M. Monge., M. A. Angew. Chem. Int. Ed. 2006,
45, 7998–8001.
(13) Gꢀandara, F.; Puebla, E. G.; Iglesias, M.; Proserpio, D. M.;
ꢀ
Snejko, N.; Monge, M. A. Chem. Mater. 2009, 21 (4), 655–661.
(14) Software for the SMART System V5.04 and SHELXTL V 5.1;
Bruker-Siemens Analytical X-ray Instrument Inc.: Madison, WI, 1998.
(15) Blatov, V. A. IUCr Comput. Comm. Newslett. 2006, 7, 4–38; see
also http://www.topos.ssu.samara.ru.
(16) DelgadoꢀFriedrichs, O.; O’Keeffe, M. Acta Crystallogr. 2003,
B59, 351–360.
(17) Ramsden, S. J.; Robins, V.; Hyde, S. T.; Hungerford, S.
EPINET: Euclidean Patterns in Non-Euclidean Tilings. The Australian
National University: Canberra, Australia, 2005ꢀ2009; http://epinet.
anu.edu.au/.
(18) (a) Sabatier, P. La Catalyse en Chimie Organique; Librairie Poly-
technique: Paris, 1913. (b) Sabatier, P.; Nye, M. J. Chem. World 2004, 1
(12), 46–49. (c) Wilkinson, G. Bull. Soc. Chim. Fr. 1968, 12, 5055–5058.
(19) de Vries, J. G.; Elsevier, C. J. The Handbook of Homogeneous
Hydrogenation; Wiley-VCH: Weinheim, 2007.
(20) (a) Ertl, G.; Knc-zinger, H.; Weitkamp, J. Handbook of Hetero-
geneous Catalysis; Wiley-VCH: Weinheim, 1997; Vol. 5. (b) Sheldon,
R. A.; Van Bekkum, H. Fine Chemicals through Heterogeneous Catalysis;
Wiley-VCH: Weinheim, 2001; Vol. 5.
(21) (a) Haenel, M. W.; Narangerel, J.; Richter, U.-B.; Rufiꢀnska, A.
Angew. Chem. In. Ed. 2006, 45 (7), 1061–1066. (b) Slaugh, L. H. Tetra-
hedron. 1966, 22 (6), 1741–1746.
(22) Harder, S. Chem. Rev. 2010, 110 (7), 3852–3876.
(23) Spielmann, J.; Buch, F.; Harder, S. Angew. Chem. In. Ed. 2008,
47 (49), 9434–9438.
(24) Platero-Prats, A. E.; de la Pe~na-O’Shea, V. A.; Iglesias, M.; Snejko,
ꢀ
N.; Monge, A.; Gutiꢀerrez-Puebla, E. ChemCatChem 2010, 2, 147–149.
(25) Platero-Prats, A. E.; de la Pen~a-O’Shea, V. A.; Snejko, N.; Monge,
A.; Gutiꢀerrez-Puebla, E. Chem.—Eur. J. 2010, 16 (38), 11632–11640.
ꢀ
(26) Ojima, I.; Li, Z.; Zhu, J. The Chemistry of Organosilicon Com-
pounds; Rappoport, Z., Apeloig, Y., Eds.; John Wiley: New York, 1998;
Vol 2, p 1687.
1758
dx.doi.org/10.1021/cg200078j |Cryst. Growth Des. 2011, 11, 1750–1758