Heterometallic coordination polymers assembled from trigonal trinuclear Fe2Ni-pivalate blocks and polypyridine spacers: Topological diversity, sorption, and catalytic properties

Article abstract of DOI:10.1021/ic503061z

Linkage of the trigonal complex [Fe₂NiO(Piv)₆] (where Piv^- = pivalate) by a series of polypyridine ligands, namely, tris(4-pyridyl)triazine (L²), 2,6-bis(3-pyridyl)-4-(4-pyridyl)pyridine (L³), N-(bis-2,2-(4-pyridyloxymethyl)-3-(4-pyridyloxy)propyl))pyridone-4 (L⁴), and 4-(N,N-diethylamino)phenyl-bis-2,6-(4-pyridyl)pyridine (L⁵) resulted in the formation of novel coordination polymers [Fe₂NiO(Piv)₆(L²)]_n (2), [Fe₂NiO(Piv)₆(L³)]_n (3), [Fe₂NiO(Piv)₆(L⁴)]_n·nHPiv (4), and [{Fe₂NiO(Piv)₆}₄{L⁵}₆]_n·3nDEF (5, where DEF is N,N-diethylformamide), which were crystallographically characterized. The topological analysis of 3, 4, and 5 disclosed the 3,3,4,4-connected 2D (3, 4) or 3,4,4-connected 1D (5) underlying networks which, upon further simplification, gave rise to the uninodal 3-connected nets with the respective fes (3, 4) or SP 1-periodic net (4,4)(0,2) (5) topologies, driven by the cluster [Fe₂Ni(?14;₃-O)(?14;-Piv)₆] nodes and the polypyridine ?14;₃-L^3,4 or ?14;₂-L⁵ blocks. The obtained topologies were compared with those identified in other closely related derivatives [Fe₂NiO(Piv)₆(L¹)]_n (1) and {Fe₂NiO(Piv)₆}₈{L⁶}₁₂ (6), where L¹ and L⁶ are tris(4-pyridyl)pyridine and 4-(N,N-dimethylamino)phenyl-bis-2,6-(4-pyridyl)pyridine, respectively. It was shown that a key structure-driven role in defining the dimensionality and topology of the resulting coordination network is played by the type of polypyridine spacer. Compounds 2 and 3 possess a porous structure, as confirmed by the N₂ and H₂ sorption data at 78 K. Methanol and ethanol sorption by 2 was also studied indicating that the pores filled by these substrates did not induce any structural rearrangement of this sorbent. Additionally, porous coordination polymer 2 was also applied as a heterogeneous catalyst for the condensation of salicylaldehyde or 9-anthracenecarbaldehyde with malononitrile. The best activity of 2 was observed in the case of salicylaldehyde substrate, resulting in up to 88% conversion into 2-imino-2H-chromen-3-carbonitrile.

Full text of DOI:10.1021/ic503061z

Article
pubs.acs.org/IC
Heterometallic Coordination Polymers Assembled from Trigonal
Trinuclear Fe2Ni-Pivalate Blocks and Polypyridine Spacers:
Topological Diversity, Sorption, and Catalytic Properties
Svetlana A. Sotnik,† Ruslan A. Polunin,† Mikhail A. Kiskin,*,‡ Alexander M. Kirillov,§
Victoria N. Dorofeeva,†,∥ Konstantin S. Gavrilenko,∥,⊥ Igor L. Eremenko,‡ Vladimir M. Novotortsev,‡
and Sergey V. Kolotilov*,†
†L. V. Pisarzhevskii Institute of Physical Chemistry of the National Academy of Sciences of the Ukraine, Prospekt Nauki 31, Kiev
03028, Ukraine
‡N. S. Kurnakov Institute of General and Inorganic Chemistry, Russian Academy of Sciences, Leninsky Prosp. 31, 119991 GSP-1
Moscow, Russian Federation
§Centro de Química Estrutural, Complexo I, Instituto Superior Tec
Portugal
́
nico, Universidade de Lisboa, Av. Rovisco Pais, 1049-001 Lisbon,
∥Research-And-Education ChemBioCenter, National Taras Shevchenko University of Kyiv, Chervonotkackaya str., 61, Kiev 03022,
Ukraine
⊥Enamine Ltd., A. Matrosova str. 23, Kiev 01103, Ukraine
S
* Supporting Information
ABSTRACT: Linkage of the trigonal complex [Fe2NiO(Piv)6] (where Piv− =
pivalate) by a series of polypyridine ligands, namely, tris(4-pyridyl)triazine (L2),
2,6-bis(3-pyridyl)-4-(4-pyridyl)pyridine (L3), N-(bis-2,2-(4-pyridyloxymethyl)-3-
(4-pyridyloxy)propyl))pyridone-4 (L4), and 4-(N,N-diethylamino)phenyl-bis-
2,6-(4-pyridyl)pyridine (L5) resulted in the formation of novel coordination
polymers [Fe2NiO(Piv)6(L2)]n (2), [Fe2NiO(Piv)6(L3)]n (3), [Fe2NiO-
(Piv)6(L4)]n·nHPiv (4), and [{Fe2NiO(Piv)6}4{L5}6]n·3nDEF (5, where DEF
is N,N-diethylformamide), which were crystallographically characterized. The
topological analysis of 3, 4, and 5 disclosed the 3,3,4,4-connected 2D (3, 4) or
3,4,4-connected 1D (5) underlying networks which, upon further simplification,
gave rise to the uninodal 3-connected nets with the respective fes (3, 4) or SP 1-
periodic net (4,4)(0,2) (5) topologies, driven by the cluster [Fe2Ni(μ3-O)(μ-
Piv)6] nodes and the polypyridine μ3-L3,4 or μ2-L5 blocks. The obtained
topologies were compared with those identified in other closely related derivatives [Fe2NiO(Piv)6(L1)]n (1) and
{Fe2NiO(Piv)6}8{L6}12 (6), where L1 and L6 are tris(4-pyridyl)pyridine and 4-(N,N-dimethylamino)phenyl-bis-2,6-(4-
pyridyl)pyridine, respectively. It was shown that a key structure-driven role in defining the dimensionality and topology of
the resulting coordination network is played by the type of polypyridine spacer. Compounds 2 and 3 possess a porous structure,
as confirmed by the N2 and H2 sorption data at 78 K. Methanol and ethanol sorption by 2 was also studied indicating that the
pores filled by these substrates did not induce any structural rearrangement of this sorbent. Additionally, porous coordination
polymer 2 was also applied as a heterogeneous catalyst for the condensation of salicylaldehyde or 9-anthracenecarbaldehyde with
malononitrile. The best activity of 2 was observed in the case of salicylaldehyde substrate, resulting in up to 88% conversion into
2-imino-2H-chromen-3-carbonitrile.
magnetic, catalytic, luminescent, and other properties of PCPs
are determined by the composition and structure of metal-
containing units. Despite numerous methods having been
developed for the synthesis of PCPs, in the overwhelming
majority of cases it is impossible to rationally predict the
structure of the resulting compounds, since the same reagents
can often lead to completely different products, depending on
INTRODUCTION
■
Porous coordination polymers (PCPs) are considered as a basis
for the creation of functional materials for sorption and
separation processes,1 catalysis,2 luminescent,3 or magnetic
applications.4 Due to the unique possibility of PCPs to combine
several properties within one structure, such compounds have a
recognized potential to become multifunctional materials.1c,5
Sorption properties of PCPs are determined by their porosity
that depends on the crystal structures of such compounds and
the arrangement of structural elements. On the other hand,
Received: December 30, 2014
© XXXX American Chemical Society
A
DOI: 10.1021/ic503061z
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 1. Polypyridine building blocks (nitrogen atoms that coordinate to metal ions are shown in blue).
solvent, stoichiometry, reaction temperature, or time.6 Poly-
nuclear “building blocks” in the frameworks of PCPs are usually
formed by the self-assembly of metal ions and organic ligands in
the reaction mixture,7 or via the decomposition of high
nuclearity derivatives.8 However, the number of examples of
presynthesized polynuclear complexes being used as building
blocks is much lower.9−13 Some success in the directed
synthesis of PCPs possessing the desired structures can only be
expected if a very similar system has already been studied.14 In
particular, the series of similar PCPs was synthesized using the
strategy of reticular synthesis.15 In contrast, the use of
polynuclear complexes as building blocks (without undergoing
rearrangement during reaction) can lead to the assembly of
porous lattices, the structure of which is predetermined to a
certain extent by the topology of such blocks and bridging
ligands.
In recent years, we reported that the combination of
heterometallic trigonal blocks [Fe2NiO(Piv)6] (hereinafter
abbreviated as [Fe2Ni]) and several polypyridine ligands
resulted in coordination polymers or discrete complexes, the
topologies of which depend on the symmetry of [Fe2Ni] blocks
and the type of polypyridine spacers.9−11 The [Fe2NiO(Piv)6]
can be considered as a pseudotrigonal building block, since
both the FeIII and NiII atoms show the same coordination
environments and thus could not be distinguished by the X-ray
diffraction analysis. The reaction of [Fe2Ni] with linear
bidentate ligands such as 4,4′-bipyridine (bipy) or trans-bis-
(4-pyridyl)ethylene (dpe) produced 2D coordination polymers
[Fe2NiO(Piv)6(bipy)1.5]n or [Fe2NiO(Piv)6(dpe)1.5]n, respec-
tively, which possess honeycomb structures.10,11 Each honey-
comb in a 2D layer of these compounds is constructed from six
[Fe2Ni] moieties linked by six polypyridine ligands, resulting in
the interpenetrated layers. The interpenetration is possible due
to the sufficiently large size of the cavities in honeycombs. The
combination of the same [Fe2Ni] building block with a
pseudotrigonal planar tridentate 2,4,6-tris(4-pyridyl)pyridine
ligand (L1, Figure 1) also gave a 2D coordination polymer
[Fe2NiO(Piv)6(L1)]n·3nDMSO·0.84nH2O (1) with a honey-
comb structure.9 In contrast to [Fe2NiO(Piv)6(bipy)1.5]n or
[Fe2NiO(Piv)6(dpe)1.5]n, each honeycomb in [Fe2NiO-
(Piv)6(L1)]n is built from three [Fe2Ni] and three μ3-L1 blocks,
and the 2D layers are not interpenetrated presumably due to
the small cavity size. The linkage of [Fe2Ni] units by bidentate
angular bipyridines resulted in the formation of cyclic
structures, namely, a 1D chain with {Fe2NiO(Piv)6}2(cis-
dpe)2 cycles in the case of cis-bis(4-pyridyl)ethylene (cis-dpe),
or a discrete 0D “nanocube” {Fe2NiO(Piv)6}8{L6}12 in the case
of bis-2,6-(4-pyridyl)-4-(4-N,N-dimethylaminophenyl)pyridine
(L6, Figure 1).9,11 Each of these structures also corresponds to
one of the possibly expected topological types.
As a continuation and possible merging of these studies, the
present work aimed at investigating the influence of the type of
polypyridine ligand on structure and topology of the resulting
coordination polymers based on the pseudotrigonal [Fe2NiO-
(Piv)6] blocks. Three analogues of L1 were used, namely, (i)
tris(4-pyridyl)triazine L2, containing a “central” triazine ring
instead of pyridine, (ii) a derivative L3 that bears two 3-pyridine
groups instead of two 4-pyridine rings in L1 and L2, and (iii) a
tripyridine derivative L4 that possesses a nonplanar structure in
contrast to L1 and L2 (Figure 1). As expected, a coordination
polymer with L2, [Fe2NiO(Piv)6(L2)]n·1.25nH2O (2), is
isostructural to previously reported compound 1. In contrast,
the replacement of L1 by L3 or L4 resulted in the generation of
two novel 2D coordination polymers [Fe2NiO(Piv)6(L3)]n·
1.5nDMSO·3nH2O (3) and [Fe2NiO(Piv)6(L4)]n·nHPiv (4),
respectively, the topologies of which are different from that of
[Fe2NiO(Piv)6(L1)]n (1). In addition, a similar reaction of an
analogue of L6 that comprises a diethylamino group (L5, Figure
1) was examined, giving rise to a novel double chain 1D
polymer [{Fe2NiO(Piv)6}4{L5}6]n·3nDEF (5, where DEF is
N,N-diethylformamide) instead of the discrete 0D “nanocube”
{Fe2NiO(Piv)6}8{L6}12·30DMF (6).
In the present work, we report the synthesis and character-
ization of 2, 3, 4, and 5, as well as their structural features and
topological analysis. For comparative purposes, the latter has
also been performed for 1 and 6. Since solvent has a significant
influence on ordering the crystal lattice of 3 (as evidenced by
powder XRD) and its sorption characteristics, two different
samples were studied: 3 and 3′ that were synthesized in DMSO
or DMF, respectively. For characterization of porous structures
B
DOI: 10.1021/ic503061z
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
spectrometry measurements as described above, which confirmed that
the condensation reactions do not occur without a catalyst.
of compounds 2 and 3, N2 and H2 sorption was measured.
Sorption of methanol, ethanol, salicylaldehyde, and 9-
anthracenecarbaldehyde by 2 was also investigated. Moreover,
catalytic activity of 2 in the condensation of aldehydes with
malononitrile was evaluated.
Synthesis of L3. A modified procedure published by Smith et al.19
was used. 3-Acetylpyridine (10 g, 82.5 mmol) was added to a
suspension of crushed NaOH (3.3 g, 82.5 mmol) in polyethylenegly-
col PEG-400 (100 mL) and stirred at 0 °C for 10 min. 4-
Pyridinecarboxaldehyde (4.42 g, 41.2 mmol) was then added, and
the suspension was left standing at 0 °C for 2 h. Every 30 min the
mixture was manually stirred with a spatula. After 2 h, NH4OAc (20 g,
excess) was added, and the suspension was heated at 110 °C for 10 h.
During this time, the color of the mixture changed from cherry-red to
auburn, and a precipitate of the product formed. Water (200 mL) was
then added, and the precipitate was isolated by filtration and washed
with water (100 mL) and cold acetone (20 mL). Recrystallization from
CHCl3 gave white solid, yield 9 g (70%). Anal. Found/Calcd for
C20H14N4, %: C, 77.3/77.4; H, 4.55/4.55; N, 18.2/18.1.
EXPERIMENTAL SECTION
■
General Methods. Commercially available reagents and solvents
(Aldrich, Merck) were used as received. The starting materials
[Fe2NiO(Piv)6(HPiv)3],11 tetrakis(O-tosyl)pentaerythritol,16 and
2,4,6-tris-(4-pyridyl)triazine (L2)17 were prepared as previously
reported. C,H,N-elemental analyses were carried out using a Carlo
Erba 1106 analyzer. 1H NMR spectra were recorded on a Varian Unity
Plus 400 spectrometer at 400.4 MHz. Chemical shifts are reported in
ppm. Powder X-ray diffraction experiments were run on a Bruker D8
instrument with Cu radiation (λ = 1.540 56 Å). UV−vis spectra were
measured on a Specord 210 spectrophotometer.
Synthesis of L4. 4-Hydroxypyridine (0.95 g, 10 mmol), tetrakis(O-
tosyl)pentaerythritol (1.88 g, 2.5 mmol), and K2CO3 (5 g) were
heated under reflux in DMSO (50 mL) for 8 h. Then, the reaction
mixture was cooled down to room temperature and poured in 250 mL
of water. The obtained precipitate was filtered off, recrystallized from
chloroform, and dried in vacuum over P2O5. Yield 0.10 g (10%). Anal.
Found/Calcd for C25H24N4O4, %: C, 67.8/67.6; H, 5.55/5.44; N,
12.5/12.6.
Measurements of N2 and H2 sorption were studied on a
Sorptomatic-1990 instrument by volumetric method at 78 K, while
sorption of methanol and ethanol was investigated gravimetrically at
293 K, as previously described.12 In order to remove solvent and other
compounds that could be captured in the voids, the samples of 2, 3,
and 3′ were dried in vacuum and hold in acetone for one week
(acetone was changed several times) prior to the measurements. The
samples were then finally dried in vacuum (10−3 Torr) at 150 °C.
Sorption measurements of salicylaldehyde (Sal) or 9-anthracene-
carbaldehyde (Aca) by 2 were performed at 373 1 K in hermetic
vials as follows: weighed samples of 2 (2 × 10−3 mmol) were mixed
with toluene solutions (2 mL) of substrate (c(Sal) was from 0.005 to
0.055 M, c(Aca) = 0.05 M), and the concentration of aldehyde was
measured at certain time moments by spectrophotometry at 330 and
405 nm for salicylaldehyde and 9-anthracenecarbaldehyde, respec-
tively. In sorption and catalytic experiments, effective concentration of
2 in suspension was determined as the ratio of a quantity of suspended
2 (in moles) to the volume of solution.
Synthesis of L5. The synthetic procedure was similar to that for L3,
with the difference that 4-acetylpyridine (10 g, 82.5 mmol) was used
instead of 3-acetylpyridine, and p-diethylaminobenzaldehyde (7.3 g,
41.2 mmol) was applied instead of 4-pyridinecarboxaldehyde. For the
isolation of the product, water (150 mL) was added, and the
precipitate was removed by filtration and washed with H2O (100 mL)
and cold acetone (40 mL). Recrystallization from acetone gave yellow
solid, yield 7.0 g (45%). Anal. Found/Calcd for C24H24N4, %: C, 78.3/
78.2; H, 6.55/6.57; N, 15.3/15.2.
Synthesis of Compound 2. A solution of [Fe2NiO(Piv)6(HPiv)3]
(0.198 g, 0.18 mmol) in DMF (15 mL) was slowly added to a DMF
solution (60 mL) containing L2 (0.066 g, 0.21 mmol) and pivalic acid
(0.81 g, 7.93 mmol), and the reaction mixture was heated at 120 °C
for 1 h resulting in the formation of yellowish-green microcrystals.
These were collected by filtration, held in acetone for 3 days, and then
filtered off and dried in vacuum. Yield 0.190 g (95%). Anal. Found/
Calcd for Fe2NiC48H68.5N6O14.25, [Fe2NiO(Piv)6(L2)]n·1.25nH2O, %:
C, 50.9/51.1; H, 5.92/6.12; N, 7.46/7.45. Water molecules were
captured from air prior to elemental analysis.
Synthesis of Compound 3. Synthesis was carried out using the
same procedure as for 2, with the difference that L3 (0.065 g, 0.21
mmol) was used instead of L2. Two alternative solvents were used:
DMSO or DMF. Yield of 3 in the reaction with DMSO as solvent was
0.22 g (95%). Anal. Found/Calcd for C53H83N4O17.5S1.5Fe2Ni,
[Fe2NiO(Piv)6(L3)]n·1.5nDMSO·3nH2O, %: C 49.9/49.9; H 6.21/
6.56; N 4.58/4.40. Water molecules were captured from air prior to
elemental analysis. Yield of 3′ in the reaction with DMF as solvent was
0.190 g (95%). Anal. Found/Calcd for C50H70N4O14Fe2Ni, [Fe2NiO-
(Piv)6(L3)]n·nH2O, %: C, 53.2/53.6; H, 5.99/6.29; N, 4.89/5.00.
Water molecule was captured from air prior to elemental analysis.
Synthesis of Compound 4. A solution of L4 (0.022 g, 0.05
mmol) in DMF (10 mL) was added to a DMF/MeCN (40 mL/20
mL) solution containing [Fe2NiO(Piv)6(HPiv)3] (0.055 g, 0.05
mmol) and HPiv (0.05 g, 0.49 mmol). Greenish-brown crystals
formed in 3 days and were then collected and dried in air. Yield 0.03 g
(50%). Anal. Found/Calcd for C60H88N4O19Fe2Ni, [Fe2NiO-
(Piv)6(L4)]n·nHPiv, %: C, 53.6/53.8; H, 6.80/6.62; N, 4.10/4.18.
Synthesis of Compound 5. A solution of [Fe2NiO(Piv)6(HPiv)3]
(0.030 g, 0.027 mmol) in acetonitrile (6 mL) was carefully layered
over an N,N-diethylformamide (DEF) solution (2 mL) containing L5
(0.019 g, 0.05 mmol) and HPiv (0.06 g, 0.58 mmol). Dark greenish-
brown crystals were formed in 10 days and then were collected and
dried in air. Yield 0.03 g (50%). Anal. Found/Calcd for
C279H393N27O55Fe8Ni4, [{Fe2NiO(Piv)6}4(L2)6]n·3nDEF, %: C, 59.2/
58.9; H, 7.00/6.97; N, 6.45/6.65.
In catalytic experiments, typical reaction mixtures were prepared as
follows: aldehyde (0.2 mmol), malononitrile (0.2 mmol), and 2 (2.2 ×
10−3 mmol) were continuously stirred in toluene (2 mL) at 100 °C
during certain time period (2−40 h). Concentrations of products were
18
1
measured by H NMR. Separate portions of the reaction mixtures
1
were used for determination of conversion versus time. For H NMR
analysis the following steps occurred: stirring of the reaction mixture
was stopped, catalyst was filtered off and washed with hot
dichloromethane several times, solvent was evaporated, and the
obtained residues were quantitatively dissolved in CDCl3 or DMSO-d6
for reaction mixtures when using salicylaldehyde or 9-anthracene-
carbaldehyde as substrates, respectively. In order to determine
conversion of salicylaldehyde, 1,2-dimethoxyethane (0.02 mmol) was
added to the probes as standard. Signal with δ = 11.02 ppm was used
for integration. In the case of 9-anthracenecarbaldehyde, the standard
was not added, and the yields of 2-(9-anthrylmethylene)malononitrile
were determined as ω = 100% × I(product)/[I(product} + I(Aca)],
where I(product) is integral intensity of signal with δ = 9.65 ppm, and
I(Aca) is integral intensity of signal with δ = 11.48 ppm. Typical NMR
spectra are shown on Figure S1 (Supporting Information).
Control tests for condensations of aldehydes and malononitrile
were also performed in the presence of L2 (2.2 × 10−3 mmol) over 40
h under typical reaction conditions as mentioned above. Concen-
trations of aldehydes were measured by spectrophotometry. For
spectrophotometric measurements, heating and stirring of reaction
mixtures was stopped, and aliquots (0.2 mL) of the clear reaction
solutions (without particles of L2) were diluted by toluene (2 mL);
spectra of resulting solutions were measured (λ = 320−400 nm in the
case of reaction mixtures with Sal, λ = 320−600 nm in the case of
reaction mixtures with Aca).
Blank experiments in the absence of any catalyst were also
performed by mixing 0.2 mmol of aldehyde with 0.2 mmol of
malononitrile in 2 mL of toluene. Reaction mixtures were stirred at
100 °C over 40 h. Concentrations of aldehydes were determined by
C
DOI: 10.1021/ic503061z
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 1. Crystal Data and Structure Refinement Details for 2−5
2
3
3
empirical formula
C48H66Fe2N6NiO13
1105.48
C50H68Fe2N4NiO13
1103.49
160(2)
C50H68Fe2N4NiO13
1103.49
230(2)
fw (g mol−1
)
temp (K)
150(2)
wavelength (Å)
cryst size, mm3
cryst syst
Space group
a (Å)
0.710 73
0.710 73
0.2 × 0.1 × 0.1
monoclinic
P21/c
0.710 73
0.1 × 0.1 × 0.1
monoclinic
P21/c
0.15 × 0.1 × 0.1
trigonal
P3 /c
̅
1
16.667(5)
16.667(5)
14.966(4)
90
13.1615(11)
16.9299(14)
30.922(3)
92.319(2)
6884.5(10)
4
13.164(4)
17.071(5)
31.150(9)
92.110(6)
6996(4)
4
b (Å)
c (Å)
β (deg)
V (Å3)
3600(2)
2
Z
calcd density (g cm−3
)
1.020
1.065
1.048
abs coeff (mm−1
)
0.705
0.736
0.724
F(000)
1160
2320
2320
θ range for data collection (deg)
refns collected
reflns unique
R(int)
1.41−23.31
19 891
1751
1.32−25.69
58 968
1.31−25.69
57 839
13 092
13 291
0.1502
121
0.1362
0.2014
params
631
631
GOF on F2
1.206
0.834
0.817
a
R1 ([I > 2σ(I)]
0.0752
0.2301
0.0622
0.0774
b
wR2 [I > 2σ(I)]
0.1419
0.1805
4
5
empirical formula
C60H87Fe2N4NiO19
1338.75
296(2)
C135H180Fe4N12Ni2O26
fw (g mol−1
)
2727.73
296(2)
0.71073
0.20 × 0.15 × 0.1
monoclinic
P21
temp (K)
wavelength (Å)
cryst size, mm3
cryst syst
space group
a (Å)
0.71073
0.12 × 0.08 × 0.01
monoclinic
P21/n
11.917(6)
34.750(18)
23.067(12)
98.817(8)
9439(9)
4
12.514(2)
60.998(9)
12.873(2)
98.645(3)
9714(3)
2
b (Å)
c (Å)
β (deg)
V (Å3)
Z
calcd density (g cm−3
)
0.942
0.933
abs coeff (mm−1
)
0.550
0.532
F(000)
2828
2884
θ range for data collection (deg)
reflns collected
reflns unique
1.07−26.41
75 894
0.67−25.69
42 413
19 292
29 160
R(int)
0.1734
0.1245
params
765
1237
GOF on F2
0.722
0.841
a
R1 ([I > 2σ(I)]
0.0763
0.1013
b
wR2 [I > 2σ(I)]
0.1798
0.2270
a
b
2
2
R1 = ∑||Fo| − |Fc||/∑|Fo|. wR2 = {∑[w(Fo − Fc2)2]/∑[w(Fo )2]}1/2
.
X-ray Crystal Structure Determinations. The X-ray data sets for
cell measurements and powder XRD). X-ray data for 3 were collected
at 160 and 230 K. The structures were solved using SHELXS-9721 and
refined with SHELXL-9721 by full-matrix least-squares on F2. The H
atoms were treated by a riding model. Disordered solvent molecules in
compounds 3 (at 230 K), 4, and 5 could not be localized, and thus, the
corresponding electronic density was corrected by SQUEEZE.22 Table
1 contains crystal data. CCDC deposition numbers are 1040164−
1040168 for 2, 3 (160 K), 3 (230 K), 4, and 5, respectively. The
channel diameters in 1, 2, and 3 were estimated from the Mercury
diagram for a probe molecule with r = 1.4 Å, taking into account that
the compounds 2, 3, 4, and 5 were collected on a Bruker APEX II
diffractometer equipped with a CCD camera and a graphite
monochromated Mo Kα radiation source (λ = 0.710 73 Å).20 The
X-ray quality crystals of compounds 4 and 5 were taken from the
reaction mixtures, while single crystals of 3 were grown by diffusion of
diethyl ether into solution of starting compound in DMSO. In the case
of 2, X-ray data were collected using an air-dried sample in order to
check the stability of the crystal structure upon desolvation
(isostructural nature of 1 and 2 was confirmed by single-crystal unit
D
DOI: 10.1021/ic503061z
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 2. Scheme illustrating the formation of 1−6 and their structural fragments. All H atoms, t-Bu groups, and noncoordinated molecules (if any)
are omitted for clarity. In 3, planes marked by yellow or green belong to sets; planes within each set are parallel, but planes from different sets are
inclined ca. 85° to each other. In 4, yellow and green indicate different metallocycles. Structural fragments of 1 and 6 are drawn using published
data.9
the image of the channel corresponds to the centers of spheres that
touch channel “wall”, and a true channel is 2r wider.
[Fe2Ni] unit (i.e., N atoms of central pyridine and dialkyamino
groups, and all O atoms of L4 were not counted) was equal to a
number of “free” coordination sites in [Fe2NiO(Piv)6], thus
allowing a comparison of the observed and expected structures.
Two samples were prepared by the reaction of [Fe2NiO-
(Piv)6(HPiv)3] with L2 in DMSO or DMF (3 and 3′,
respectively), which possess the same ratio of [Fe2Ni]
trinuclear block and L2, but differ by solvent composition.
Despite being rather similar, these samples exhibit significantly
distinct sorption properties, which can be explained by different
crystallinity of these compounds, vide infra.
Crystal Structures. Compound 2. Compound 2 is
isostructural to a previously characterized derivative 1,9 as
confirmed by powder X-ray data (see Figure S3, Supporting
Information) and unit cell parameters measured for a single
crystal. In order to check the stability of crystal lattice of 2 upon
desolvation, the X-ray data were collected using a crystal dried
in air for several days. It was found that a single crystal of 2
preserves crystallinity under these conditions, and the crystal
structure was successfully solved. Unit cell parameters a and b
for 2 were 0.1 Å lower compared to those of 1 (16.667(5) Å in
RESULTS AND DISCUSSION
■
Synthesis. Synthesis of the L3 and L5 ligands was performed
similarly to reported procedures.19 The preparation of L4
involved a reaction of tetrakis(O-tosyl)pentaerythritol with 4-
hydroxypyridine, similar to reported procedures.23 However,
instead of the expected derivative with four 4-pyridine groups, a
different compound was obtained that bears one N-substituted
4-pyridone fragment (Scheme S2, Supporting Information).
This group formed as a result of N-alkylation of 4-
hydroxypyridine instead of O-alkylation.
The coordination polymers ([Fe2NiO(Piv)6(L2)]n·
1.25nH2O) 2, ([Fe2NiO(Piv)6(L3)]n·1.5nDMSO·3nH2O) 3,
([Fe2NiO(Piv)6(L4)]n·nHPiv) 4, and ([{Fe2NiO-
(Piv)6}4{L5}6]n·3nDEF) 5 were synthesized via treatment of
[Fe2NiO(Piv)6(HPiv)3] with the corresponding polypyridine
spacers, with adoption of the protocols reported for related
compounds.9−11 In all cases the total number of donor atoms in
polypyridine ligands that can be bound by the metal ions in the
E
DOI: 10.1021/ic503061z
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
2 vs 16.775(2) Å in 1 at the same temperature), while the
values of c were closer (14.966(4) Å vs 14.973(2) Å,
respectively). We suppose that these differences are not caused
by solvent content in voids, but by the replacement of a central
pyridine ring in 1 into a triazine one in 2. Since the L1 and L2
structures have a local C3 axis, the C−C and C−N bonds in
central rings were averaged in the structural model. These
average values were equal to 1.341(6) Å in 2 and 1.364(6) Å in
1, and can be responsible for unit cell contraction, even despite
slightly longer M−N bonds in 2 compared to 1 (2.192(7) vs
2.175(7) Å, respectively). Other structural features, including
porosity characteristics, are almost the same for 1 and 2. Crystal
lattice of 2 features similar channels with the cavity diameter of
∼10.6 Å, window diameter of ∼6.4 Å, and pore volume of
∼0.24 cm3/g (estimated from X-ray data by Platon24 for a
probe molecule with r = 1.4 Å).25
N bonds (where M are metal atoms of different trinuclear
blocks, Figure 3). The metal atoms in each [Fe2Ni]2(L3)2 cyclic
motif lie in one plane, and the angle between the mean planes
of the adjacent motifs (yellow and green in Figure 2) is 84.14°
at 160 K (83.20° at 230 K). Four above-mentioned
[Fe2Ni]2(L3)2 cyclic fragments form a larger ring-containing
void (vide infra).
Due to the fact that the arrangement of 2D layers in 3 can be
considered as a result of translation along the a axis without a
shift, the voids of each 2D layer are arranged in channels along
the a axis (Figure 4). Such channels can be presented as a
consequence of cavities (diameter ∼13.8 Å) separated by
windows (diameter ∼8.1 Å). Solvent-accessible volume of voids
in 3, estimated by PLATON, is equal to 27% at 160 K (28% at
230 K) for a probe molecule with r = 1.4 Å, corresponding to
the pore volume of ∼0.25 cm3/g. As can be seen from
structural data, the channels in 3 are bigger than those in 1
because they are located inside a larger {[Fe2Ni]2(L3)2}4 “cycle”
compared to [Fe2Ni]3(L1)3 in 1. However, the total solvent-
accessible volumes in 1 and 3 are close, which is consistent with
more “thick” walls between the channels in 3.
Compound 3. The crystal structure of 3 was solved using
the data collected at two temperatures, 160 and 230 K. Though
main features of the structure at these temperatures are similar,
only low-temperature modification will be described in detail
and significant differences will be noted.
The crystal structure of 3 features the corrugated 2D layers
(Figure 2 and Supporting Information Figures S6, S7).
Similarly to 1 and 2, each L3 moiety is bound to three
[Fe2Ni] units. The L3 spacers are almost planar (the largest
deviation of carbon atoms from the mean plane of L3 is
0.217(2) Å at 160 K or 0.209 Å at 230 K), which is essential for
the formation of a network with a predetermined topology
(Figure 3). Two adjacent [Fe2Ni] blocks are linked by two L3
Compound 4. The crystal lattice of the coordination
polymer 4 is built from 2D layers (Figure 2). Each L4 spacer
acts as a tridentate ligand and links three [Fe2Ni] units, while
each [Fe2Ni] block is bound by three L4 moieties (Figure S4,
Supporting Information). Every L4 ligand participates in the
formation of two different metallocycles, which involve two and
four [Fe2Ni] units, respectively (marked by green and yellow in
Figure 2). All metal ions within the metallocycle containing two
[Fe2Ni] units lie in one plane, while the neighboring
metallocycles of this type are not coplanar. Thus, the bent
character of 2D layers is consistent with a nonplanar structure
of L4, in contrast to L1 and its perfectly planar derivative 1.9
The adjacent 2D layers are shifted with a distance of 5.244 Å,
which is estimated as the separation between the parallel mean
planes that go through metal ions in the [Fe2Ni] units from the
adjacent layers. The pyridone rings of L4 from the neighboring
layers are coplanar and located at 3.396 Å from each other (the
distance between the mean planes of such rings, Supporting
Information Figure S5), indicating some stacking interactions
between them.
Compounds 5 and 6. The compound 5 is a 1D
coordination polymer with a double chain structure, wherein
two {[Fe2Ni](L5)}n chain motifs are held together via
additional linkage of L5 between each [Fe2Ni] unit (Figure
2). Alternatively, 5 can be considered as a series of 24-
membered units [Fe2Ni]4(L5)4 interconnected by the L5
linkers. As expected,9 the [Fe2Ni]4(L5)4 fragments are not
planar. Previously, we reported a related compound
[Fe2Ni]8(L4)12 (6) with a cube-like structure.9 The replacement
of dimethylamino group in L6 by a diethylamino moiety in L5
led to the formation of a 1D chain structure in 5 instead of
“folding” into a 0D cube (Figure 2). Cube-like molecules of 6
were polycatenated in the solid state.9 It can be assumed that
such polycatenation was important to stabilize the crystal lattice
and decrease ΔG of compound 6 formation. In such a case, a
possible reason for the difference between the topologies of 5
and 6 can consist of a larger size of the diethylamino group over
the dimethylamino group in L5 and L6, respectively. In 6, there
is a short contact between the H atoms of dimethylamino
group of L6 of one cube and the H atoms of tert-butyl group of
[Fe2NiO(Piv)6] of adjacent cube (Figure S8, Supporting
Information). Hence, there is probably no enough space for
Figure 3. Metallocycles [Fe2Ni]x(L)y formed as a result of junction of
trinuclear pivalate with various polypyridines in different combinations
in 1−3, 5, 6, and related compound derived from cis-(4-pyridyl)-
ethylene (cis-dpe).11
ligands in such way that the centrosymmetric cyclic
[Fe2Ni]2(L3)2 motifs can be distinguished (highlighted by
yellow and green in Figure 2 and separately shown in Figure 3).
Each [Fe2Ni]2(L3)2 motif has a structure resulting from the
combination of a trigonal metal-containing building block and
rigid ligand, which predetermine the 60° angle between the M−
F
DOI: 10.1021/ic503061z
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 4. Visualization of voids in the crystal structure of 3 (160 K). Mercury diagram drawn for a probe molecule with r = 1.4 Å. Projections along
the a and b axes.
the ethyl group in a hypothetic polycatenated cubic structure,
which leads to the formation of a chain. The adjacent 2D layers
in 5 are arranged in such a way that no apparent voids are
formed (Supporting Information Figure S9).
Topological Diversity of 1−6. To get further insight into the
structures of 2D (1−4), 1D (5), and 0D (6) networks, we have
performed their topological analysis.26 Following the concept of
the underlying net,27,28 all the μ-Piv and μ-L1−6 moieties in 1−6
have been reduced to their centroids resulting in simplified
metal−organic networks.
Thus, an underlying net of 2 is composed of the 4-connected
Fe and Ni nodes (topologically equivalent), 3-connected μ3-L2
and μ3-O nodes, as well as 2-connected μ2-Piv linkers (Figure
5a and Figure S10, Supporting Information). This trinodal
3,3,4-connected net possesses a rare 3,3,4L40 topology26,29
expressed by the point symbol of (33.92.10)3(33)(93), with the
(33.92.10), (33), and (93) indices corresponding to the Fe/Ni,
μ3-O, and μ3-L2 nodes, respectively. This network can be
Figure 5. Topological representations of the simplified uninodal 3-
connected networks in 2 (a), 3 (b), 5 (c), and 6 (d) after contracting
the [Fe2Ni(μ3-O)(μ-Piv)6] units to 3-connected cluster nodes (second
simplification; for initial simplification, see Supporting Information
Figure S10). (a) 2D net with the hcb [Shubnikov hexagonal plane net
reduced further to a uninodal 3-connected net with the hcb
[Shubnikov hexagonal plane net (6,3)] topology and the point
symbol of (63) (Figure 5a), after treating the heterometallic
(6,3)] topology and the point symbol of (63) (view along the c axis).
(b) 2D net with the fes [Shubnikov plane net] topology and the point
symbol of (4.82). (c) 1D chain with the SP 1-periodic net (4,4)(0,2)
topology and the point symbol of (42.6). (d) 0D → 1D catenated net
with the 3M8-1 topology and the point symbol of (43). Networks
shown in parts b, c, and d are along the a axis. Further details:
[Fe2Ni(μ3-O)(μ-Piv)6] cluster node is shown in the center [t-Bu
groups of μ-Piv are omitted for clarity; Fe (green), Ni (cyan), O (red),
C (gray)]; (a−d) centroids of [Fe2Ni(μ3-O)(μ-Piv)6] cluster nodes
after simplification (green), centroids of μ3-L1,3 nodes or μ2-L5,6 linkers
(blue).
[Fe2Ni(μ3-O)(μ-Piv)6] blocks as the 3-connected cluster
nodes. The compound 1 is topologically similar to 2.
From the topological viewpoint, the underlying 2D networks
of 3 and 4 are similar. These are assembled from the 4-
connected Fe and Ni nodes, 3-connected μ3-O and μ3-L3,4
nodes, and 2-connected μ-Piv linkers (for 3, see Figure 5b and
Supporting Information Figure S10). Their topological
analysis26 discloses the complex tetranodal 3,3,4,4-connected
networks with a unique topology described by the point symbol
of (33.122.13)(33.6.72)2(33)(6.122), wherein the (33.122.13),
(33.6.72), (33), and (6.122) notations correspond to the Ni, Fe,
μ3-O, and μ3-L3,4 nodes, respectively. Further simplification of
the obtained networks, namely, by considering the hetero-
metallic [Fe2Ni(μ3-O)(μ-Piv)6] blocks as the 3-connected
cluster nodes, furnishes the uninodal 3-connected nets (for 3,
see Figure 5b) with the fes [Shubnikov plane net] topology and
the point symbol of (4.82).26,30,31 Although various compounds
with the fes topology have been described,32 3 and 4 reveal the
first example of the present topological type that is assembled
G
DOI: 10.1021/ic503061z
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
from a μ3-oxo trimetallic hexacarboxylate cluster [M3(μ3-O)(μ-
RCOO)6] as a secondary building unit (SBU).26,30,33
nonactivated samples (Supporting Information Figures S11
and S12). Under the same conditions, the crystallinity of 3
slightly deteriorated, and reflections on the PXRD pattern
became wider and less intense.
After simplification procedure, the obtained trinodal 3,3,4-
connected double chain of 5 (Figure 5d and Supporting
Information Figure S10) can be topologically described by the
point symbol of (33.8.92)2(33.82.9)(33) with the (33.8.92),
(33.82.9), and (33) notations corresponding to the 4-connected
Fe1/Fe3/Ni1/Ni2, 4-connected Fe2/Fe4, and 3-connected μ3-
O nodes, respectively. Further simplification of this topolog-
ically unique 1D network by treating the [Fe2Ni(μ3-O)(μ-
Piv)6] units as 3-connected cluster nodes and μ2-L5 moieties as
2-connected linkers furnishes a uninodal net (Figure 5d) with a
rare SP 1-periodic net (4,4)(0,2) topology26 defined by the
point symbol of (42.6).
The underlying structure of 6 is composed of the discrete 0D
clusters that are catenated into a 1D array. The topological
analysis reveals a binodal 3,4-connected net (Figure 5c and
Supporting Information Figure S10) that features an
unreported topology26,28,30,33 described by the point symbol
of (33.82.9)3(33), wherein the (33.82.9) and (33) indices are
those of the 4-connected Fe and Ni nodes (topologically
equivalent) and 3-connected μ3-O nodes, respectively. The
[Fe2Ni(μ3-O)(μ-Piv)6] blocks can be contracted to the 3-
connected cluster nodes which, along with the 2-connected μ2-
L6 linkers, give rise to a further simplified uninodal net (Figure
5c) with the 3M8-1 topology and the point symbol of (43). The
present type of topology is very rare and has been reported only
in a single case.34
Since crystallographic studies showed that only compounds 2
and 3 possess voids in crystal lattices, sorption properties of 4
and 5 were not studied (sorption properties of 1 and 6 were
9
previously reported ). Bearing in mind a significant difference
between powder XRD patterns for 3 and 3′ ([Fe2NiO-
(Piv)6L3]n, sorption properties of both samples were inves-
tigated.
Desolvated compounds 2, 3, and 3′ exhibit permanent
porosity, which has been confirmed by sorption of N2 and H2 at
78 K (Figure 6a,b). Nitrogen adsorption isotherms for
compounds 2 and 3′ are typical for microporous sorbents.35
A sharp growth of N2 adsorption up to 150 and 40 cm3 g−1 at a
Thermal Stability and Sorption Properties of 2, 3, and 3′.
Partial or complete desolvation of 2 has almost no influence on
its crystal structure, as confirmed by comparison of powder
XRD patterns (Supporting Information Figure S11) and
successful X-ray structure determination of the desolvated
crystal. Only some redistribution of relative intensities of
reflections was observed in the experimental patterns compared
to calculated ones. All expected reflections were found on
experimental patterns.
In contrast, the powder XRD pattern of air-dried sample of 3,
prepared in DMSO (the same solvent as used for single-crystal
growth), contains wide reflections which are evidence for poor
crystallinity of the bulk sample of this compound or small
particle size (Supporting Information Figure S12). While
positions of these reflections correspond to the expected ones
(with some redistribution of relative intensities, as in the case of
2), some expected reflections are not found. In particular, all
low-angle reflections (at 2θ = 5.6°, 5.9°, and 6.7°, which
correspond to 002, 011, and 100 planes, respectively) are
absent, and there is also no reflection at 2θ = 18.7 ° (12−5
plane). Reflections expected between 2θ = 8.5° and 9.0°
(planes 110, 003, 10−2, 11−1, 102, and 111) merge in the
experimental powder XRD pattern of 3.
As mentioned in the Experimental Section, the reaction
between the same reagents in DMF instead of DMSO led to
the formation of sample 3′. Surprisingly, powder XRD pattern
of 3′ contained more distinct and narrow reflections, which fit
much better to the expected powder XRD pattern (Supporting
Information Figure S12). The dominating majority of expected
reflections was found (redistribution of relative intensities
occurred).
□
○
Figure 6. (a) Isotherms of N2 ( ) and H2 ( ) sorption for 2; (b)
Heating of samples 2 and 3′ in 10−3 Torr vacuum, performed
for activation prior to sorption measurements, did not lead to
significant structural changes, as evidenced by comparison of
their powder XRD patterns with X-ray diffraction for
□
○
△
isotherms of N2 ( ) and H2 ( ) sorption for 3, and N2 ( ) and H2
◇
(
) sorption for 3′; (c) Saito−Foley distribution of micropores with
diameter (D) derived from N2 adsorption data for 1−3. dV/dr for 1
was calculated on the basis of published data.9
H
DOI: 10.1021/ic503061z
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
very low pressure (3 Torr) for 2 and 3′, respectively, is caused
by micropores filling. The second sharp growth of N2
adsorption isotherms at P close to PS can be explained by
interparticle condensation. The N2 sorption is completely
reversible for both these compounds. At the same time, the
shape of nitrogen sorption isotherm of 3 indicates the presence
of not only regular micropores, but also a significant
contribution of irregular mesopores. Adsorption isotherm for
3 continuously grows after micropores filling at PPS−1 between
3 and 750 Torr (Figure 6b). In addition, there is a small
hysteresis of N2 adsorption−desorption in 3, which is absent in
2 and 3′. The difference between the N2 sorption for
compounds 2 and 3′ on one side versus 3 on the another
side is consistent with their powder XRD patterns (vide supra).
In fact, compounds 2 and 3′ possess high crystallinity, while 3 is
significantly disordered.
To get additional information about accessibility of pores of
2 to various substrates, sorption of methanol and ethanol was
measured at 293 K from the gaseous phase, and sorption of
salicylaldehyde and 9-anthracenecarbaldehyde was studied from
toluene solution at 373 K as well. Both isotherms of alcohol
sorption reach saturation at V of ∼0.3 cm3 g−1 (Figure 7) that is
For 2, the micropore volume estimated using the Dubinin−
Radushkevich approach (VDR = 0.29 cm3 g−1, Table 2) is
Table 2. Parameters of N2 and H2 Sorption Isotherms of 1−3
□
○
Figure 7. Isotherms of methanol ( ) and ethanol ( ) sorption by 2 at
293 K. Arrows indicate direction of isotherm at absorption and
desorption.
a
b
compd
SBET, m2 g−1
VDR, cm3 g−1 u(H2), % by weight
ref
1
730
730
140
205
0.28
0.29
0.03
0.08
0.91
0.93
0.20
0.25
9
2
this work
this work
this work
3
very close to VDR calculated from N2 adsorption (Table 2),
providing evidence that 2 possesses “rigid” micropores, which
do not change on interaction with guest molecules in contrast
to previously reported cases.11,12,38,39 This observation is
consistent with powder XRD data, which indicate that the
crystal lattice of 2 is stable upon desolvation (vide supra).
Pressure decrease to zero leads to complete desorption of
methanol without significant hysteresis, providing evidence that
MeOH is not bound by covalent bonds. In fact, coordination to
metal ion is impossible, because there are no free sites available
in 2.
Isotherm of ethanol absorption reaches saturation at lower
PPS−1, compared to methanol absorption (Figure 7). This can
be explained by a higher affinity of more hydrophobic ethanol
(compared to methanol) to hydrophobic tert-butyl groups of
[Fe2NiO(Piv)6] units. These results are consistent with the
difference, observed previously between methanol and ethanol
absorption by pivalate-based coordination polymers [Fe2MO-
(Piv)6(L)1.5]n (M = CoII or NiII, L = 4,4′-bipyridine or trans-
1,2-bis(4-pyridyl)ethylene).11,12 Though the absorption of
alcohol by [Fe2MO(Piv)6(L)1.5]n probably led to expansion of
the crystal lattice due to rearrangement of 2D layers (and
alcohol absorption could not be considered as filling of rigid
micropores), ethanol sorption capacity was higher than that of
methanol.11,12
Catalytic Activity of 2 in Aldehyde Condensation with
Malononitrile. Compound 2 can be considered as a potential
heterogeneous catalyst for condensation reactions that can be
catalyzed by N atoms of triazine rings in pores, similarly to
reported cases.40−42 Hence, condensation of salicylaldehyde or
9-anthracenecarbaldehyde with malononitrile was studied in the
present work. For analysis of catalytic results (vide infra),
sorption of salicylaldehyde from toluene solution at T = 373 K
was investigated. In contrast to sorption of N2 or H2, isotherm
of salicylaldehyde sorption by 2 was not typical for pores filling
in microporous sorbent (Figure 8a). Noticeable sorption begins
at equilibrium salicylaldehyde concentration c(Sal) ≈ 7 × 10−3
M, and then the adsorption isotherm continuously rises,
reaching the sorption capacity of 1.2 mol salicylaldehyde per
3′
a
All compounds were desolvated in vacuum prior to sorption
b
measurements. SBET, VDR, and u(H2) stand for BET surface area,
Dubinin−Radushkevich micropore volume, and the highest exper-
imentally achieved hydrogen capacity at 78 K, respectively.
consistent with the value calculated from crystallographic data
(0.24 cm3 g−1). In contrast, VDR is significantly lower for both 3
and 3′ (0.03 and 0.08 cm3 g−1, respectively, Table 2) than the
value expected from the X-ray structure (0.25 cm3 g−1). This
difference can be explained by partial blocking of micropores or
crystal structure disorder (more significant for 3, less significant
for 3′). Sorption properties of 2 resemble those of its
isostructural analogue 1 (Table 2).9
The diameter of micropores (D) was estimated from N2
adsorption isotherms by the Saito−Foley model using potential
function derived for N2 on zeolite at 77.3 K as an
approximation.36 Such an approach gave the D values of
∼10.0 and ∼10.8 Å (maxima of dV/dr curves, Figure 6c) for 1
and 2, respectively. These values are consistent with crystallo-
graphic data (vide supra). A similar analysis performed for 3 and
3′ showed maxima of dV/dr at D of ∼13.6 and ∼14.3 Å,
respectively, which fit well with the value expected from the X-
ray structure of 3. Notably, the dV/dr curve for 3 has a second
poorly defined maximum at D ∼18.9 Å, which can be
considered as a numerical estimation of effective average
diameter of “additional” irregular pores present in the sample.
Surprisingly, hydrogen sorption capacity values for 3 and 3′ are
very close (Figure 6b and Table 2). This can be rationalized
assuming that channels in the crystal lattice of 3 are partially
blocked and not accessible for N2 (r = 3.64 Å),37 but are
accessible to H2 (r = 2.8 Å). Notably, close values of H2
sorption for 3 and 3′ can be considered as an argument,
confirming that (i) there is no impurity of nonporous phase in
these samples, and (ii) possible presence of such impurity
cannot be responsible for the difference between N2 sorption
isotherms for these compounds.
I
DOI: 10.1021/ic503061z
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 8. (a) Isotherm of salicylaldehyde sorption by 2 from toluene solution and (b) time dependence of salicylaldehyde sorption by 2 from
□
○
toluene solution for c(Sal)0 = 0.039 M ( ) or c(Sal)0 = 0.055 M ( ). T = 373 K.
Figure 9. Condensation reactions of aldehydes with malononitrile catalyzed by 2.
Fe2Ni unit at c(Sal) = 5.3 × 10−2 M. Estimation of
salicylaldehyde sorption capacity, assuming that pores are filled
by liquid salicylaldehyde at room temperature, gives the value
corresponding to 3.1 molecules of salicylaldehyde per Fe2Ni
formula unit, indicating that the experimental data are
consistent with partial filling of pores of 2 by salicylaldehyde.
Notably, in contrast to salicylaldehyde sorption by 2, isotherms
of this aldehyde sorption by other MOFs under similar
conditions were quite similar to those expected for type I, and
the values of saturation sorption capacity were consistent with
those expected for pore filling by liquid aldehyde.43,44
In addition, kinetics of salicylaldehyde sorption by 2 was
studied (Figure 8b). The obtained data cannot be satisfactorily
fit by a kinetic equation of the first order45 or a mixed-sorption
model,45 thus indicating that sorption rate is not controlled by
salicylaldehyde diffusion in pores. Application of the second-
order equation46 allowed us to fit the experimental data
(Supporting Information Figure S13), giving k2 = 0.38
mol(Fe2Ni) mol(Sal)−1 min−1 (average value for two different
concentrations, details are presented in Supporting Information
Figure S13). This value is significantly lower in comparison to
sorption rate constants determined previously with the same
pseudo-second-order model for salicylaldehyde sorption in CuII
benzenetricarboxylate [Cu3(btc)2]n, HKUST-1 (k2 = 5.7
mol(Cu2) mol(Sal)−1 min−1 at 373 K),44 or FeIII benzenedi-
carboxylate [Fe2(OH)0.3(H2O)1.7(btc)4/3]Cl1.7 (k2 = 4.3 mol-
(Fe2) mol(Sal)−1 min−1 at 373 K).43 Besides, it is also much
lower than the rate constant of xylenol orange adsorption in
MIL-101(Cr) (3−90 mol(Cr2) mol(XO)−1 min−1, depending
on dye concentration, at T between 298 and 318 K).47 Hence,
the salicylaldehyde adsorption by 2 is much slower compared to
that for similar systems, which can be explained by a smaller
pore size or stronger competition between salicylaldehyde and
solvent (toluene). Nevertheless, adsorption data do not
contradict to the fact that pores of 2 are accessible to
salicylaldehyde. In contrast, compound 2 does not adsorb 9-
anthracenecarbaldehyde in similar conditions.
As mentioned above, catalytic activity of 2 was investigated in
a condensation reaction between aldehydes and malononitrile.
This process is typically catalyzed by coordination polymers
containing basic40 or acidic48 sites, and in the case of 2 the
noncoordinated N-atoms of triazine ring can be active catalytic
sites. Condensation of aldehydes with malononitrile normally
gives the derivative of 1,1-dicyanoethylene, which in the case of
salicylaldehyde undergoes further intermolecular transforma-
tion to furnish 2-imino-2H-chromen-3-carbonitrile (Figure
9).49 These reactions do not proceed without a catalyst, as
confirmed by “blank experiments”.
Reactions of malononitrile with salicylaldehyde or 9-
anthracenecarbaldehyde in the presence of 2 gave expected
1
products; no byproducts could be detected by H NMR in the
reaction mixtures. It was also shown previously that selectivity
in this reaction is almost quantitative.48 Thus, the values of
aldehyde conversion correspond to the product yields (these
were determined from spectra with respect to concentration of
aldehyde, though equal initial concentrations of reagents were
used). In the presence of 2, the conversion of salicylaldehyde to
condensation product, 2-imino-2H-chromen-3-carbonitrile, rea-
ches 70% in 2 h. Then, the reaction rate decreases, and further
heating of the reaction mixture for 38 h leads to just 18%
increase of conversion (total conversion is 88%, Figure 10). In
contrast, total conversion of 9-anthracenecarbaldehyde after 15
h of reaction is about 20%, increasing to 30% after 40 h of
heating. Thus, the catalytic activity of 2 in the condensation of
salicylaldehyde is much higher compared to that when using 9-
anthracenecarbaldehyde as substrate. The conversions achieved
with catalyst 2 in condensation of salicylaldehyde with
malononitrile are comparable with those reported for a number
J
DOI: 10.1021/ic503061z
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
μ2-L5,6 blocks. The latter appear to play a crucial structure-
driven role in defining the dimensionality and topological type
of the resulting network.
In 2, the values of pore volume found from N2 sorption and
sorption capacity regarding methanol and ethanol were close to
those expected from the X-ray crystal structure. The isotherm
of N2 sorption by 2 resembled those of type I according to
IUPAC classification, typical for microporous sorbents. The
difference between sorption of methanol or ethanol by 2 was
only in the shape of the isotherm (which can be explained by
host−guest interaction energy), but not in the volume of pores
accessible to these alcohols. In 2, the volume of pores accessible
to N2, methanol, or ethanol was close to 0.3 cm3 g−1, indicating
that the pores could be completely filled by these substrates
without any size discrimination. These findings point out that
the crystal lattice of 2 is rigid and does not change upon
interaction with N2 and alcohols. Pores of 2 were accessible to
salicylaldehyde, but the rate of its adsorption is rather low if
compared to those of similar previously reported systems.
Nevertheless, catalytic activity of 2 in the condensation of
salicylaldehyde and malononitrile was significantly higher than
its activity in a similar reaction involving 9-anthracenecarbalde-
hyde and malononitrile substrates. In addition, the difference
between salicylaldehyde and 9-anthracenecarbaldehyde con-
versions catalyzed by 2 was higher than the corresponding
difference in the reactions catalyzed by L2. This finding can be
partially explained by the sieving effect (discrimination of
aldehyde molecules by size) rather than the higher reactivity of
salicylaldehyde. Pore volumes for 3 and 3′ found from N2
adsorption were much lower than the values expected from
crystallographic data, which can be explained by pore blocking.
Significant dependency of sorption characteristics of 3 on a
sample’s crystallinity (estimated from the width of reflections
on powder XRD patterns) was found. It can be concluded that
crystal lattice of 3 is less stable toward collapse than that of 2.
The design of novel metal−organic materials constructed
from the different heterometallic hexacarboxylate SBUs and
polypyridine type spacers and the investigation of their
functional properties are currently in progress.
Figure 10. Time dependency of the condensation product yields for
□
■
the reactions of salicylaldehyde ( or ) or 9-anthracenecarbaldehyde
2
○
(
●
or ) with malononitrile catalyzed by 2 (red) or L (blue).
c(aldehyde) = 0.1 M, c(malononitrile) = 0.1 M, c(2) = c(L2) = 1.1 ×
10−3 M (effective concentration), toluene, 100 °C.
of MOFs or polynuclear complexes in similar reactions. In
addition, the activity of 2 in terms of catalyst turnover number
(TON) exceeds the majority of reported catalysts48,50−54
(Table S1, Supporting Information).
Considering the possibility of triazine moieties to catalyze
such condensation reactions, we also performed control tests
with L2 as a catalyst instead of compound 2. As in the case of 2,
reactions of salicylaldehyde or 9-anthracenecarbaldehyde with
malononitrile in the presence of L2 gave the same products,
although in much lower yields under similar reaction conditions
(e.g., 27% for L2 vs 88% for 2 in the case of Sal substrate for the
same catalyst loading and identical concentrations of the
reagents, see also Figure 10). While the process involving 2 was
heterogeneous, catalytic reaction in the presence of L2 was
partially homogeneous, but without achieving complete
dissolution of L2. Thus, higher activity of 2 compared to L2
can be caused by better accessibility of the catalyst’s active sites.
Conversion of salicylaldehyde in the presence of L2 was 2−5
times higher compared to conversion of 9-anthracenecarbalde-
hyde, which can be explained by distinct reactivity of these
aldehydes. However, there was 20-fold difference between
conversions of these aldehydes at t = 2 h (or 10-fold at t = 5 h)
in the case of catalysis by 2. This value exceeds the difference in
conversions associated with reactivity of aldehydes, and it can
be explained by better accessibility of active sites of 2 to
salicylaldehyde versus bulkier substrate (9-anthracenecarbalde-
hyde). The fact that the reaction of 9-anthracenecarbaldehyde
occurs in the presence of 2 can provide evidence for the
presence of some accessible active sites, located probably on the
surface of microparticles of 2. Apparently, these sites are active
in catalytic condensations of both aldehydes, making the sieving
effect (if any) less explicit. It should be noted that the substrate
selectivity of 2 with respect to salicylaldehyde versus 9-
anthracanecarbaldehyde was significantly higher than that for
ASSOCIATED CONTENT
* Supporting Information
■
S
Additional characterization details, Figures S1−S11, Table S1
with comparison of catalytic activity, as well as CIF files for 2−
5. The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/ic503061z.
AUTHOR INFORMATION
Corresponding Authors
■
52
reported MOF [Zr6O4(OH)4(L7)6]n (Supporting Informa-
tion Table S1).
*E-mail: m_kiskin@mail.ru. Phone: +7(495)9522084. Fax:
+7(495)9541279.
CONCLUSIONS
■
*E-mail: svk001@mail.ru. Phone: +38(044)5256661. Fax:
+38(044)5256216.
The present study has extended the growing family of
coordination polymers built from the trimetallic hexacarbox-
ylate [M3(μ3-O)(μ-RCOO)6] SBUs, resulting in the synthesis
and characterization of the four new heterometallic Fe/Ni
coordination polymers 2, 3, 4, and 5. These, along with the
related derivatives 1 and 6, have been topologically classified,
revealing uninodal 3-connected underlying 2D, 1D, or 0D nets
with distinct topologies. After additional simplification, these
include the hcb (1, 2), fes (3, 4), SP 1-periodic net (4,4)(0,2)
(5), and 3M8-1 (6) topologies constructed from the cluster
[Fe2Ni(μ3-O)(μ-Piv)6] nodes and the polypyridine μ3-L1,2,3,4 or
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was partially supported by a joint grant of the
National Academy of Sciences of Ukraine and Russian
Foundation for Basic Research (No. 03-03-14), program of
the National Academy of Sciences “Hydrogen in alternative
energetics and novel technologies”. A.M.K. acknowledges the
K
DOI: 10.1021/ic503061z
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(10) Polunin, R. A.; Kolotilov, S. V.; Kiskin, M. A.; Cador, O.;
Mikhalyova, E. A.; Lytvynenko, A. S.; Golhen, S.; Ouahab, L.;
Ovcharenko, V. I.; Eremenko, I. L.; Novotortsev, V. M.; Pavlishchuk,
V. V. Eur. J. Inorg. Chem. 2010, 5055−5057.
Foundation for Science and Technology (FCT), Portugal
(PTDC/QUI-QUI/121526/2010, PEst-OE/QUI/UI0100/
2013). V.M.N. acknowledges the Russian Foundation for
Basic Research (No. 14-03-90423). M.A.K. and I.L.E. acknowl-
edge the Council on Grants of the President of the Russian
Federation (Grants NSh-4773.2014.3).
(11) Polunin, R. A.; Kiskin, M. A.; Cador, O.; Kolotilov, S. V. Inorg.
Chim. Acta 2012, 380, 201−210.
(12) Polunin, R. A.; Kolotilov, S. V.; Kiskin, M. A.; Cador, O.;
Golhen, S.; Shvets, O. V.; Ouahab, L.; Dobrokhotova, Z. V.;
Ovcharenko, V. I.; Eremenko, I. L.; Novotortsev, V. M.; Pavlishchuk,
V. V. Eur. J. Inorg. Chem. 2011, 4985−4992.
REFERENCES
■
(1) (a) Gu, Z.-Y.; Yang, C.-X.; Chang, N.; Yan, X.-P. Acc. Chem. Res.
2012, 45, 734−745. (b) Chen, B.; Xiang, S.; Qian, G. Acc. Chem. Res.
2010, 43, 1115−1124. (c) Qiu, S.; Zhu, G. Coord. Chem. Rev. 2009,
253, 2891−2911. (d) Zhao, Z.; Ma, X.; Kasik, A.; Li, Z.; Lin, Y. S. Ind.
Eng. Chem. Res. 2013, 52, 1102−1108. (e) Huang, K.; Liu, S.; Li, Q.;
Jin, W. Sep. Purif. Technol. 2013, 119, 94−101. (f) Han, S.; Wei, Y.;
Valente, C.; Lagzi, I.; Gassensmith, J. J.; Coskun, A.; Stoddart, J. F.;
Grzybowski, B. A. J. Am. Chem. Soc. 2010, 132, 16358−16361. (g) Lu,
H.; Zhu, S. Eur. J. Inorg. Chem. 2013, 1294−1300. (h) Li, J.-R.; Sculley,
J.; Zhou, H.-C. Chem. Rev. 2012, 112, 869−932. (i) Burtch, N. C.;
Jasuja, H.; Walton, K. S. Chem. Rev. 2014, 114, 10575−10612.
(2) (a) Maihom, T.; Wannakao, S.; Boekfa, B.; Limtrakul, J. J. Phys.
Chem. C 2013, 117, 17650−17658. (b) Wang, S.; Bromberg, L.;
Schreuder-Gibson, H.; Hatton, T. A. ACS Appl. Mater. Interfaces 2013,
5, 1269−1278. (c) Fang, Q.-R.; Yuan, D.-Q.; Sculley, J.; Li, J.-R.; Han,
Z.-B.; Zhou, H.-C. Inorg. Chem. 2010, 49, 11637−11642. (d) Kim, W.-
S.; Lee, K. Y.; Ryu, E.-H.; Gu, J-M.; Kim, Y.; Lee, S. J.; Huh, S. Eur. J.
Inorg. Chem. 2013, 4228−4233. (e) Sen, R.; Saha, D.; Mal, D.;
(13) (a) Schoedel, A.; Wojtas, L.; Kelley, S. P.; Rogers, R. D.;
Eddaoudi, M.; Zaworotko, M. J. Angew. Chem., Int. Ed. 2011, 50,
11421−11424. (b) Dulcevscaia, G. M.; Filippova, I. G.; Speldrich, M.;
van Leusen, J.; Kravtsov, V. Ch.; Baca, S. G.; Kogerler, P.; Liu, S.-X.;
̈
Decurtins, S. Inorg. Chem. 2012, 51, 5110−5117. (c) Pavlishchuk, A.
V.; Kolotilov, S. V.; Zeller, M.; Thompson, L. K.; Addison, A. W. Inorg.
Chem. 2014, 53, 1320−1330.
(14) (a) Deng, H.; Grunder, S.; Cordova, K. E.; Valente, C.;
Furukawa, H.; Hmadeh, M.; Gandara, F.; Whalley, A. C.; Liu, Z.;
́
Asahina, S.; Kazumori, H.; O’Keeffe, M.; Terasaki, O.; Stoddart, J. F.;
Yaghi, O. M. Science 2012, 336, 1018−1023. (b) Deng, H.; Doonan, C.
J.; Furukawa, H.; Ferreira, R. B.; Towne, J.; Knobler, C. B.; Wang, B.;
Yaghi, O. M. Science 2010, 327, 846−850.
(15) (a) O’Keeffe, M. Chem. Soc. Rev. 2009, 38, 1215−1217.
(b) O’Keeffe, M.; Peskov, M. A.; Ramsden, S. J.; Yaghi, O. M. Acc.
Chem. Res. 2008, 41, 1782−1789. (c) Dybtsev, D. N.; Yutkin, M. P.;
́
Peresypkina, E. V.; Virovets, A. V.; Serre, C.; Ferey, G.; Fedin, V. P.
Brandao, P.; Lin, Z. Eur. J. Inorg. Chem. 2013, 5103−5109. (f) Lee, J.;
̃
Inorg. Chem. 2007, 46, 6843−6845. (d) Furukawa, H.; Go, Y. B.; Ko,
N.; Park, Y. K.; Uribe-Romo, F. J.; Kim, J.; O’Keeffe, M.; Yaghi, O. M.
Inorg. Chem. 2011, 50, 9147−9152.
Farha, O. K.; Roberts, J.; Scheidt, K. A.; Nguyen, S. T.; Hupp, J. T.
Chem. Soc. Rev. 2009, 38, 1450−1459. (g) Zhao, M.; Ou, S.; Wu, C.-D.
Acc. Chem. Res. 2014, 47, 1199−1207. (h) Ou, S.; Wu, C.-D. Inorg.
Chem. Front. 2014, 1, 721−734.
(16) Shostakovskii, M. F.; Atavin, A. S.; Mirskova, A. N. Zh. Obshch.
Khim. 1965, 35, 804−807.
(17) Barrios, L. A.; Ribas, J.; Aromí, G. Inorg. Chem. 2007, 46, 7154−
7162.
(3) (a) Suresh, P.; Radhakrishnan, S.; Babu, C. N.; Sathyanarayan, A.;
Sampath, N.; Prabusankar, G. Dalton Trans. 2013, 42, 10838−10846.
(b) Lestari, W. W.; Lonnecke, P.; Sar
M.; Wickleder, C.; Handke, M.; Einicke, W.-D.; Glaserd, R.; Hey-
́
osi, M. B.; Streit, H. C.; Adlung,
̈
(18) (a) Prikhod’ko, A.; Pointillart, F.; Golhen, S.; Gavrilenko, K. S.;
Ouahab, L.; Kolotilov, S. V. New J. Chem. 2012, 36, 2070−2077.
(b) Pauli, G. F.; Jaki, B. U.; Lankin, D. C. J. Nat. Prod. 2007, 70, 589−
̈
Hawkins, E. CrystEngComm 2013, 15, 3874−3884. (c) Rao, X.; Song,
T.; Gao, J.; Cui, Y.; Yang, Y.; Wu, C.; Chen, B.; Qian, G. J. Am. Chem.
Soc. 2013, 135, 15559−15564. (d) Wanderley, M. M.; Wang, C.; Wu,
C.-D.; Lin, W. J. Am. Chem. Soc. 2012, 134, 9050−9053.
595. (c) Nunez, A.; Hickner, M. A. ACS Macro Lett. 2013, 2, 49−52.
̃
(19) Smith, C. B.; Raston, C. L.; Sobolev, A. N. Green Chem. 2005, 7,
650−654.
(4) (a) Wriedt, M.; Yakovenko, A. A.; Halder, G. J.; Prosvirin, A. V.;
Dunbar, K. R.; Zhou, H.-C. J. Am. Chem. Soc. 2013, 135, 4040−4050.
(b) Maurice, R.; Verma, P.; Zadrozny, J. M.; Luo, S.; Borycz, J.; Long,
J. R.; Truhlar, D. G.; Gagliardi, L. Inorg. Chem 2013, 52, 9379−9389.
(c) Mohapatra, S.; Rajeswaran, B.; Chakraborty, A.; Sundaresan, A.;
Maji, T. K. Chem. Mater. 2013, 25, 1673−1679.
(20) Bruker APEX2 Software Package; Bruker AXS: Madison, WI,
2005.
(21) Sheldrick, G. M. Acta Crystallogr. 2008, A64, 112−122.
(22) Sluis, P. v. d.; Spek, A. L. Acta Crystallogr., Sect. A 1990, 46,
194−201.
(23) Allampally, N. K.; Strassert, C. A.; Cola, L. D. Dalton Trans.
2012, 41, 13132−13137.
(24) Spek, A. L. Acta Crystallogr., Sect. A 1990, 46, c34.
(25) The values that characterize channel size in 1, published in ref 9,
did not take into account diameter of probe molecule, which should be
added to the Mercury image.
(26) (a) Blatov, V. A. IUCr Comp. Comm. Newsletter 2006, 7, 4.
(b) Blatov, V. A.; Shevchenko, A. P.; Proserpio, D. M. Cryst. Growth
Des. 2014, 14, 3576−3586.
(27) (a) Blatov, V. A.; Proserpio, D. M. In Modern Methods of Crystal
Structure Prediction; Oganov, A. R., Ed.; Wiley: New York, 2010; pp 1−
28. (b) Blatov, V. A.; O’Keeffe, M.; Proserpio, D. M. CrystEngComm
2010, 12, 44. (c) Alexandrov, E. V.; Blatov, V. A.; Kochetkova, A. V.;
Proserpio, D. M. CrystEngComm 2011, 13, 3947−3958.
(28) O’Keeffe, M.; Yaghi, O. M. Chem. Rev. 2012, 112, 675−702.
(29) For compounds with the 3,3,4L40 topology, see: (a) Zhang, Q.;
Chung, I.; Jang, J. I.; Ketterson, J. B.; Kanatzidis, M. G. Chem. Mater.
2009, 21, 12−14. (b) Lytvynenko, A. S.; Kolotilov, S. V.; Cador, O.;
Gavrilenko, K. S.; Golhen, S.; Ouahab, L.; Pavlishchuk, V. V. Dalton
Trans. 2009, 3503−3509. (c) Zheng, N.; Bu, X.; Feng, P. Chem.
Commun. 2005, 2805−2807. (d) Heller, M.; Sheldrick, W. S. Z. Anorg.
Allg. Chem. 2004, 630, 1191−1195.
(5) (a) Lan, A.; Li, K.; Wu, H.; Kong, L.; Nijem, N.; Olson, D. H.;
Emge, T. J.; Chabal, Y. J.; Langreth, D. C.; Hong, M.; Li, J. Inorg.
Chem. 2009, 48, 7165−7173. (b) Maspoch, D.; Ruiz-Molina, D.;
Veciana, J. Chem. Soc. Rev. 2007, 36, 770−818. (c) Kolotilov, S. V.;
Kiskin, M. A.; Eremenko, I. L.; Novotortsev, V. M. Curr. Inorg. Chem.
2013, 3, 144−160.
(6) (a) Robson, R. Dalton Trans. 2008, 5113−5131. (b) Scherb, C.;
Schodel, A.; Bein, T. Angew. Chem., Int. Ed. 2008, 47, 5777−5779.
̈
(c) Kim, H.; Das, S.; Kim, M. G.; Dybtsev, D. N.; Kim, Y.; Kim, K.
Inorg. Chem. 2011, 50, 3691−3696. (d) Rosi, N. L.; Kim, J.; Eddaoudi,
M.; Chen, B.; O’Keeffe, M.; Yaghi, O. M. J. Am. Chem. Soc. 2005, 127,
1504−1518.
(7) (a) Hong, K.; Bak, W.; Moon, D.; Chun, H. Cryst. Growth Des.
2013, 13, 4066−4070. (b) Vuong, G.-T.; Pham, M.-H.; Do, T.-O.
Dalton Trans. 2013, 42, 550−557. (c) Perry, J. J., IV; Perman, J. A.;
Zaworotko, M. J. Chem. Soc. Rev. 2009, 38, 1400−1417.
(8) Botezat, O.; van Leusen, J.; Kravtsov, V. Ch.; Filippova, I. G.;
Hauser, J.; Speldrich, M.; Hermann, R. P.; Kramer, K. W.; Liu, S.-X.;
̈
Decurtins, S.; Kogerler, P.; Baca, S. G. Cryst. Growth Des. 2014, 14,
4721−4728.
̈
(9) Dorofeeva, V. N.; Kolotilov, S. V.; Kiskin, M. A.; Polunin, R. A.;
Dobrokhotova, Z. V.; Cador, O.; Golhen, S.; Ouahab, L.; Eremenko, I.
L.; Novotortsev, V. M. Chem. Eur. J. 2012, 18, 5006−5012.
L
DOI: 10.1021/ic503061z
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(30) The Reticular Chemistry Structure Resource (RCSR) Database:
O’Keeffe, M.; Peskov, M. A.; Ramsden, S. J.; Yaghi, O. M. Acc. Chem.
Res. 2008, 30, 1782−1789.
(31) Mitina, T. G.; Blatov, V. A. Cryst. Growth Des. 2013, 13, 1655−
1664.
(32) For recent examples of compounds with the fes topology, see:
(a) Sun, J.-Y.; Wang, L.; Zhang, D.-J.; Li, D.; Cao, Y.; Zhang, L.-Y.;
Zeng, S.-L.; Pang, G.-S.; Fan, Y.; Xu, J.-N.; Song, T.-Y. CrystEngComm
2013, 15, 3402−3411. (b) Li, D.-S.; Zhang, P.; Zhao, J.; Fang, Z.-F.;
Du, M.; Zou, K.; Mu, Y.-Q. Cryst. Growth Des. 2012, 12, 1697−1702.
(c) Gong, Y.; Wu, T.; Li, J.; Li, J. Inorg. Chem. Commun. 2012, 19, 39−
42. (d) Wang, Y.-B.; Liu, D.-S.; Pan, T.-H.; Liang, Q.; Huang, X.-H.;
Wu, S.-T.; Huang, C.-C. CrystEngComm 2010, 12, 3886−3893.
(33) See the Cambridge Structural Database (CSD, version 5.34,
May 2013): Allen, F. H. Acta Crystallogr. 2002, B58, 380−388.
(34) Kahr, J.; Mowat, J. P. S.; Slawin, A. M. Z.; Morris, R. E.; Fairen-
Jimenez, D.; Wright, P. A. Chem. Commun. 2012, 48, 6690−6692.
(35) (a) Sing, K. S. W.; Everett, D. H.; Haul, R. A. W.; Moscou, L.;
Pierotti, R. A.; Rouquerol, J.; Siemieniewska, T. Pure Appl. Chem.
1985, 57, 603−619. (b) Gregg, S. J.; Sing, K. S. W. In Adsorption,
Surface Area and Porosity, 2nd ed.; Academic Press: London, 1982; p 4.
(36) Ross, S.; Olivier, J. P. On Physical Adsorption; Wiley and Sons:
New York, 1964.
(37) Breck, D. W. Zeolite Molecular Sieves; Wiley-Interscience: New
York, 1974.
(38) (a) Choi, S. B.; Furukawa, H.; Nam, H. J.; Jung, D.-Y.; Jhon, Y.
H.; Walton, A.; Book, D.; O’Keeffe, M.; Yaghi, O. M.; Kim, J. Angew.
Chem. 2012, 124, 8921−8925.
(39) (a) Sheveleva, A. M.; Kolokolov, D. I.; Gabrienko, A. A.;
Stepanov, A. G.; Gromilov, S. A.; Shundrina, I. K.; Sagdeev, R. Z.;
Fedin, M. V.; Bagryanskaya, E. G. J. Phys. Chem. Lett. 2014, 5, 20−24.
(b) Kozachuk, O.; Meilikhov, M.; Yusenko, K.; Schneemann, A.; Jee,
B.; Kuttatheyil, A. V.; Bertmer, M.; Sternemann, C.; Poppl, A.; Fischer,
̈
R. A. Eur. J. Inorg. Chem. 2013, 26, 4546−4557.
(40) Fang, Q.-R.; Yuan, D.-Q.; Sculley, J.; Li, J.-R.; Han, Z.-B.; Zhou,
H.-C. Inorg. Chem. 2010, 49, 11637−11642.
́
(41) Wolinska, E. Tetrahedron 2013, 69, 7269−7278.
(42) Gamez, P.; de Hoog, P.; Lutz, M.; Spek, A. L.; Reedijk, J. Inorg.
Chim. Acta 2003, 351, 319−325.
(43) Sotnik, S. A.; Kolotilov, S. V.; Kiskin, M. A.; Dobrokhotova, Zh.
V.; Gavrilenko, K. S.; Novotortsev, V. M.; Eremenko, I. L.; Imshennik,
V. K.; Maksimov, Yu. V.; Pavlishchuk, V. V. Russ. Chem. Bull. 2014, 4,
862−869.
(44) Sotnik, S. A.; Gavrilenko, K. S.; Lytvynenko, A. S.; Kolotilov, S.
V. Inorg. Chim. Acta 2015, 426, 119−125.
(45) Alosmanov, R. M. Moscow Univ. Chem. Bull. 2011, 52, 145−148.
(46) Wui, F.-C.; Tseng, R.-L.; Juang, R.-S. Water Res. 2001, 35, 613−
618.
(47) Chen, C.; Zhang, M.; Guan, Q.; Li, W. Chem. Eng. J. 2012, 183,
60−67.
(48) Opanasenko, M.; Dhakshinamoorthy, A.; Shamzhy, M.;
̌
́ ̌
ek, M.; Garcia, H.; Cejka, J. Catal. Sci. Technol.
Nachtigall, P.; Horac
2013, 3, 500−507.
(49) (a) Deb, M. L.; Bhuyan, P. J. Tetrahedron Lett. 2005, 46, 6453−
6456. (b) Karade, N. N.; Gampawar, S. V.; Shinde, S. V.; Jadhav, W. N.
Chin. J. Chem. 2007, 25, 1686−1689.
(50) Liu, Y.; Zhang, R.; He, C.; Dang, D.; Duan, C. Chem. Commun.
2010, 46, 746−748.
(51) Wu, P.; Wang, J.; Li, Y.; He, C.; Xie, Z.; Duan, C. Adv. Funct.
Mater. 2011, 21, 2788−2794.
(52) Yang, Y.; Yao, H.-F.; Xi, F.-G.; Gao, E.-Q. J. Mol. Catal. A: Chem.
2014, 390, 198−205.
(53) Murase, T.; Nishijima, Y.; Fujita, M. J. Am. Chem. Soc. 2012,
134, 162−164.
(54) Samanta, D.; Mukherjee, P. S. Chem. Commum. 2013, 49, 4307−
4309.
M
DOI: 10.1021/ic503061z
Inorg. Chem. XXXX, XXX, XXX−XXX