Synthesis, structural aspects and catalytic performance of a tetrahedral cobalt phosphonate framework

Article abstract of DOI:10.1002/ejic.201300639

A new phosphonate-based metal organic framework has been synthesized. The structure determined by X-ray crystallography reveals that it has an interesting 8-ring channel 3D structure having the Co₂-dimers as the secondary building unit. Topological analysis shows that it has 10-c uninodal 3D net structure with the Schlaefli symbol 3¹⁵.4²².5 ⁸. The magnetic measurement reveals the tetrahedral coordinated Co^II ion. The compound is highly active towards epoxidation in the presence of air and IBA (isobutyraldehyde). It catalyses almost all types of olefinic substrates with equal efficiency. The compound is highly stable thermally (370 °C) and under the reaction conditions as well. After the reaction it can be recovered quite easily and can be used for further reaction without any loss of activity over several cycles. A new tetrahedral cobalt phosphonate has been synthesized hydrothermally. It shows an interesting 8-ring channel structure. It catalyzes olefin epoxidation reaction in presence of O₂/IBA in heterogeneous medium. Copyright

Full text of DOI:10.1002/ejic.201300639

FULL PAPER
DOI:10.1002/ejic.201300639
Synthesis, Structural Aspects and Catalytic Performance
of a Tetrahedral Cobalt Phosphonate Framework
Rupam Sen,*[a] Debraj Saha,[b] Dasarath Mal,[a] Paula Brandão,[a]
Guillaume Rogez,[c] and Zhi Lin*[a]
Keywords: Cobalt / Phosphonate / Heterogeneous catalysis / Epoxidation / Molecular oxygen activation
A new phosphonate-based metal organic framework has
been synthesized. The structure determined by X-ray crystal-
lography reveals that it has an interesting 8-ring channel 3D
structure having the Co2-dimers as the secondary building
unit. Topological analysis shows that it has 10-c uninodal 3D
net structure with the Schläfli symbol 315.422.58. The mag-
netic measurement reveals the tetrahedral coordinated CoII
ion. The compound is highly active towards epoxidation in
the presence of air and IBA (isobutyraldehyde). It catalyses
almost all types of olefinic substrates with equal efficiency.
The compound is highly stable thermally (370 °C) and under
the reaction conditions as well. After the reaction it can be
recovered quite easily and can be used for further reaction
without any loss of activity over several cycles.
methyl group. In this study, we used a flexible ligand, 3-
pyridylmethylphosphonic acid, and produced a new 8-ring
tetrahedral cobalt phosphonate which is catalytically active
for aerial olefin epoxidation.
Introduction
The last decade witnessed a vast exploration of the syn-
thesis and applications of MOFs. Several approaches have
been taken to achieve the goal of making new and highly
applicable MOFs in the field of gas adsorption and storage,
and catalysis and magnetic material synthesis.[1–3] In this
regard, metal phosphonates now attract immense attention
for their multifunctional applications in different fields in-
cluding gas sensing and storage,[4] catalysis[5] and magnet-
ism.[6] Much effort has been invested to produce zero- to
multidimensional phosphonate frameworks either by using
alkyl phosphonates or rigid aromatic functionalized phos-
phonates.[7] Most of the studies have involved varying the
alkyl group or the attached aromatic moiety of the phos-
phonate. Paul and coworkers successfully developed porous
phosphonates with cobalt and nickel using large spacer
phosphonates.[8a,8b] In the literature there are notable exam-
ple of using flexible phosphonate ligands as highlighted by
Richards et al.[8c] They successfully developed and demon-
strated the interesting variability of the structural dimen-
sionality by using flexible phosphonate ligands such as 2-
pyridylmethylphosphonic acid. They successfully isolated
monomers, dimers and polymers by introducing only a
Molecular oxygen activation is a highly important pro-
cess in biological systems. For several decades, lots of effort
has been made to produce a suitable and simple method
to activate dioxygen according to the present day demand,
particularly from an economical and environmental point
of view.[9] In biological systems there are some complicated
processes catalyzed by enzymes (e.g. cytochrome P450)
which activate molecular oxygen and some versatile systems
based on metalloporphyrins,[10] phthalocyanin[11] and Schiff
base complexes[12] have been developed but the systems re-
main limited to homogeneous reaction conditions. Recently,
a few approaches have been made to heterogenize the sys-
tem in various catalytic reactions including oxidation of C–
H bonds or epoxidation. In most of the cases people used
the polymer supported homogeneous complexes for the ac-
tivation.[13] Other efforts include using zeolites as a sup-
port,[14] using POM based cluster compounds,[15] mineral
based materials[16] (e.g. hydrotalcite), nanoparticle clus-
ters[17] etc. and, most remarkably, by building up porphyrin-
based MOF systems.[18] More convincingly, and in contrast
to these systems, we demonstrate herein that instead of in-
tercalating metal ions into zeolites or using POMs, polymer
supported complexes, phosphonate based MOFs can be
used as heterogeneous catalysts in olefin epoxidation reac-
tions in the presence of air. In this report we introduce a
new phosphonate-based 8-ring channel MOF {Co(3-
pmp)}n 1 [3-pmp = 3-pyridylmethylphosphonic acid] which
is highly efficient at activating molecular oxygen in the pres-
ence of IBA and catalyzes epoxidation in heterogeneous
media. Remarkably, it can be used for several cycles with
[a] Department of Chemistry, CICECO, University of Aveiro,
3810-193 Aveiro, Portugal
E-mail: zlin@ua.pt; rupamsen1@yahoo.com
Homepage: http://www.ciceco.ua.pt/
[b] Department of Chemistry, Jadavpur University,
Kolkata 700032, Kolkata, India
[c] Institut de Physique et Chimie des Matériaux de Strasbourg,
UMR 7504 CNRS – Université de Strasbourg,
23 rue du Loess, BP43, 67034 Strasbourg Cedex, France
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejic.201300639.
Eur. J. Inorg. Chem. 2013, 5020–5026
5020
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
good selectivity without any loss of activity. This work
therefore provides a valuable addition to the series of
MOFs activating molecular oxygen in heterogeneous me-
dia. Additionally, variable temperature magnetic studies re-
veal that this framework compound shows a spin-flop tran-
sition at around 1.8 K.
Results and Discussion
X-ray Structure of Compound 1
Compound 1 crystallizes in the space group P21/c with
Z = 4. In the structure, the central metal ions are in a dis-
torted tetrahedral geometry (see ORTEP diagram, Figure
S1). The four corners are occupied by three oxygen atoms
from the phosphonate anion (O1, O2 and O3) and one ni-
trogen atom from the pyridine ligand (N1). The secondary
building unit is the Co2 dimer, formed by the two phos-
phonate anions with the two adjacent Co1 atoms being
linked together by two phosphonate groups to generate a
paddle-wheel [Co2(OPO)2] subunit with a Co···Co separa-
tion of 4.083 Å (Figure 1). The two axial sites of such a
paddle-wheel subunit are occupied by two different pyridyl
ligands with a Co–N bond length of 2.0517(14) Å (Table 1).
In the structure, the four Co1 tetrahedral centers are con-
nected through four phosphonate anions forming an 8-ring
channel along the crystallographic a axis and in turn it
forms a 3D framework (Figure 2). To further understand
the structure of 1, topological analysis by reducing the mul-
tidimensional structure to a simple node-and-linker net was
performed. The topological analysis by TOPOS[19] revealed
that the structure is a 10-c uninodal net with the Schläfli
symbol 315.422.58(TD10 = 3269) and the vertex symbol,
3.3.3.3.3.3.3.3.3.3.3.3.3.3.3.4.4.4.4.4.4.4.4.5.5.5.5.5.5.*.*.*.*.
*.*.*.*.*.*.*.*.*.*.*.*. (Figure 3).
Figure 2. The structural view of compound 1 along [100]. Yellow
and blue tetrahedra represent P and Co, respectively.
Figure 3. Simplified uninodal net structure of compound 1.
Magnetic Measurement
Magnetic measurements were used to determine the na-
ture of the interaction in the compound. A variable tem-
perature magnetic study of compound 1 reveals that the
Curie constant (C = 2.77 emuKmol–1) is in good agreement
with the expected value for a CoII ion in tetrahedral sym-
metry. The χT product decreases regularly upon cooling,
being indicative of overall antiferromagnetic behavior. The
susceptibility curve indicates the occurrence of antiferro-
magnetic ordering below 3.5 K (Figure S2). The magnetiza-
tion vs. field curve shows a spin-flop transition of around
1.4 T at 1.8 K. (Figure S3).
Nitrogen Adsorption Study
A nitrogen sorption measurement was undertaken to ver-
ify the porous nature of compound 1 as well as to calculate
the surface area of the frameworks (Figure S4). The BET
(Brunauer–Emmett–Teller) surface area of 1 was calculated
from the N2 adsorption isotherms. The adsorption of N2
follows a type II isotherm with a surface area (SBET) equal
to 59 m2 g–1. The calculated pore diameter is found 3.754 Å
and the pore volume 0.079 gcm-3. From the adsorption iso-
therm we observed that the compound is not porous in na-
ture. This result corroborates the information obtained
from X-ray crystallographic analysis.
Figure 1. Paddle wheel type structural moiety in compound 1.
Table 1. Selected bond lengths [Å] and bond angles [°] of 1.
Bond lengths
Bond Angles
Co1–O1
Co1–O2
Co1–O3
Co1–N1
P2–O1
1.9515(13)
1.9347(13)
1.9449(12)
2.0517(14)
1.5235(12)
1.5181(14)
O2–Co1–O1
O1–Co1–O3
O2–Co1–O3
O1–Co1–N1
O2–Co1–N1
O3–Co1–N1
110.61(5)
115.95(6)
117.29(6)
96.80(6)
111.51(6)
102.29(6)
Thermogravimetric (TG) Analysis
Thermogravimetric (TG) analysis between room temp.
and 800 °C shows high thermal stability for compound 1.
P2–O2
Eur. J. Inorg. Chem. 2013, 5020–5026
5021
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
[22]
Upon increasing the temperature there was no mass loss conditions by using H2O2 or tBuOOH (TBHP)[23–25] but
up to 370 °C but further increase in temperature results in few studies are on MOF catalysts for epoxidation reactions
degradation of the organic moiety with rapid mass loss and in the presence of air.[18,26] Metalloporphyrin encapsulated
decomposition completes at around 600 °C, producing only zeolite-like 3D MOFs can efficiently (a total yield of 91.5%)
the oxide compound (Figure S5). This result clearly indi- catalyze the oxidation of cyclohexane with TBHP to cyclo-
cates that the compound is highly stable and able to be used hexanol/cyclohexanone.[23] However, no selectivity was re-
for higher temperature catalytic reactions.
ported. The epoxidation of cyclohexene can be done by
using a cobalt containing pyrazolate MOF in the presence
of TBHP.[24a,24b] Ying et al. successfully converted cyclohex-
ene up to 62% by using a cobalt pyrazolate framework.[24a]
In a recent report Volkmer and his colleague carried out
Catalytic Activity
Compound 1 is highly efficient at catalyzing epoxidation another study of cyclohexene oxidation where the maxi-
reactions in air. In the literature, use of this type of phos- mum conversion achieved after 22 h was 27.5%.[24b] Li et
phonate for aerial olefin epoxidation is unprecedented. It al. reported a new cobalt based layered carboxylate frame-
catalyzes the epoxidation reaction quite smoothly in the work that catalyzes the reaction of several alkenes with high
presence of air although a sacrificial oxidant, isobutyral- efficiency in the presence of TBHP.[25] Previously, we re-
dehyde (IBA), was needed to carry out the reaction. The ported on layered metal carboxylates and the lanthanide
IBA/O2 system has been well studied with various catalysts based MOFs which are catalytically active in epoxidation
for aerobic oxidation reactions in the recent past.[18,20] The reactions of olefins with TBHP in heterogeneous condi-
catalytic activity can be summarized based on the results tions.[27,28] Besides these results, less effort has been devoted
accumulated in Table 2. Compound 1 catalyzes the reaction to produce the efficient MOF-based catalysts that are cap-
of all kinds of alkenes from terminal linear to internal able of activating molecular oxygen. Corma et al. reported
cyclic, the styrene system and also the stilbene system. The heterogeneous catalytic oxidation of tetralin with molecular
cyclic olefins cyclohexene and cyclooctene were converted oxygen using copper-based MOFs.[29a] A remarkable effort
in high yield (93 and 96%) to their corresponding products has been made by Jiang et al. where they introduced FeIII
with high selectivity. In the case of cyclooctene, the epoxide ions into the porphyrin based MOF system which catalyzes
was the only product while for cyclohexene, 2-cyclohexen- the reaction of different kinds of olefins in the presence of
1-ol was also produced owing to allylic C–H oxidation. Al- molecular oxygen.[18] They successfully converted all kind
though the terminal alkenes need higher activation energies of alkenes with good selectivity. The linear olefins 1-hexene
we got a fair conversion of 100 and 95% for 1-hexene and and 1-octene have been converted by up to 62% with selec-
1-octene, respectively, having the product selectivity of tivity of 99% with a TOF value of 23. In the case of cyclic
100% within 16 hours which clearly demonstrates the high olefins, cyclooctene was converted by up to 95% with 97%
efficiency of the catalyst. To study the conjugated olefins, epoxide selectivity but the TOF value again remains around
styrene and its methyl derivative were taken into account. 25. For styrene oxidation, the conversion was not impres-
Both the substrates were converted with 100% efficiency sive enough with a value of 55% with the epoxide selectivity
but the product selectivity remains within the range of 49– of 69% where other byproduct was also formed. In their
54% with concomitant production of the usual correspond- study, the more remarkable result was for trans-stilbene oxi-
ing by-product as benzaldehyde. Compound 1 exhibited dation where they found almost 99% conversion with selec-
good efficiency for trans-stilbene oxidation, after 16 h we tivity of 96%. Baiker et al. also successfully employed a
got 86% conversion with 100% product selectivity. A dif- phosphonate (STA-12)[26] for the trans-stilbene oxidation
ferent system, limonene was also subjected to oxidation. where they found 95% conversion with 89% epoxide selec-
The conversion was good with the product selectivity of tivity. It is worth noting that the present compound is a
75% along with some unidentified product. To carry out all good candidate besides those mentioned above.[18,26] This
the reactions, the presence of both IBA and air (O2) is es- new compound shows good catalytic activity, giving 100%
sential and in absence of either species, the reaction does selectivity for cyclooctene, 1-hexene, 1-octene and trans-
not proceed. To account for the possibility of autooxidation stilbene with 86–100% conversion and having a TOF value
of the sacrificial oxidant (IBA) as documented by Hermans ranging from 54–63 in acetonitrile. The process with 100%
et al.,[21] we carried out the reactions in the absence of the selectivity can bypass the purification step and is more
catalyst but at this low temperature we did not encounter economic. The conversion rates are remarkable for a variety
any conversion of the substrate. This result confirms the of different substrates. The possible reason could be based
necessity of the catalyst. It is also important to note that on the mechanistic pathway of the reactions, by which con-
the epoxidation reaction with 1-hexene and cyclooctene was version of the olefin to its corresponding epoxide product
also carried out in homogeneous media using Co(NO3)2 as depends only on the slow step of the reaction. If we investi-
the catalyst but we got very poor conversions of 39 and 34 gate the reaction mechanism of this type of reaction re-
with a selectivity of 33 and 26, respectively. These results ported to date, then it will help to find out a possible reason
also support the necessity of the framework catalyst (Table behind this good catalytic activity.[29b] This type of reaction
S1). Olefin epoxidation reactions catalyzed by various mechanism involves the generation of peroxyacid from O2
MOF based catalysts are well documented in heterogeneous and an aldehyde by means of a radical pathway. The reac-
Eur. J. Inorg. Chem. 2013, 5020–5026
5022
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
tion between metal compounds and peroxyacid then takes smaller butyl group.[29c] In view of the conjugative effect of
place to form a metal-peroxy adduct. The metal–peroxy ad- the benzene ring, styrene should show lower activity but it
ducts then release a peroxy radical thereby regenerating the was interesting to find that styrene gave styrene oxide in
metal environment. This is followed by a reaction between 49% yield with 100% conversion and, along with this, benz-
the peroxy radical and olefin to form a peroxy species that aldehyde was also detected. When the electron-donating
will further undergo migration of oxygen to form the epox- methyl group was used as a substituent in the para-position
ide product and acid as a by-product. If any step of the on styrene, similar conversions to the epoxide to those seen
reaction mechanism, except the formation of epoxide, is the for styrene were observed but with higher selectivity (54%).
slow step then that will control the conversion process. In the case of trans-stilbene, the reaction gave 100% epoxide
Moreover, there are no such strong electron-withdrawing or selectivity with 86% alkene conversion. The difference in
donating groups or any bulky groups involved in the sub- the conversion between styrene, 4-Me styrene and trans-
strates studied for catalytic reactions which could signifi- stilbene may be attributed to the electronic effects of the
cantly influence the epoxide formation electronically or ste- substituents.[25] In the case of Limonene, due to different
rically. However, the results of various olefins differed activity of the double bonds, 75% epoxidation occurred at
somewhat for various reasons. Epoxidation of endocyclic the cyclic double bond.
alkenes such as cylcooctene led to 96% conversion and
In order to understand the nature of the catalyst and to
100% epoxide selectivity because of its highly active double optimize the conditions, we carried out some control ex-
bond and the relative stability of cyclooctene oxide. For cy- periments by varying the temperature, solvent and catalyst
clohexene, only 68% epoxide selectivity was obtained, since amount. To search for a greener process, we carried out all
its active allylic site could also be attacked in the oxidation the reactions in the environmentally friendly solvent eth-
process. Linear aliphatic olefins such as 1-hexene and 1- anol. In ethanol we found good catalytic activity for all of
octene were converted into their corresponding 1,2-epoxy the substrates but the conversions are slightly low although
alkanes as the sole products. However, the conversion of 1- the selectivity is impressive (Table 3). The highest conver-
hexene was 100% where as for 1-octene it was 95%. It sion we got was 82% for 1-hexene and 4-Me styrene. In
shows that the catalytic activity decreases along with the comparison with the reactions carried out in acetonitrile,
chain length of the olefin. This may be due to the larger the selectivity for most of the substrates was found to be
hexyl group of 1-octene connected to a double bond which higher. Interestingly, the selectivity of styrene and 4-Me
sterically hinders it in approaching the active site with re- styrene epoxide is much higher and notable in ethanol than
spect to 1-hexene in which the double bond carries a in acetonitrile. These results clearly give us a new improve-
ment opportunity to study the reactions in greener solvents
Table 2. Epoxidation of olefins catalyzed by compounds 1 in aceto-
nitrile.[a]
Table 3. Epoxidation of olefins catalyzed by compounds 1 in ethan-
ol.[a]
[a] Reaction conditions: alkenes (10 mmol); isobutyraldehyde
(10 mmol); catalyst (2.0 mg); flow rate of air ca. 5 mLmin–1; sol-
vent (20 mL); temperature 60 °C. The results of the catalytic reac-
tion are given here after 16 h of reaction. [b] Turn over frequency
(TOF): mol converted per mol of active site per hour. [c] 2-
Cyclohexen-1-ol. [d] Benzaldehyde. [e] 4-methylbenzaldehyde. [f]
Limonene oxide. [g] Unidentified product.
[a] Reaction conditions were the same as given in footnote of
Table 1. [b] Turn over frequency (TOF): mol converted per mol of
active site per hour. [c] 2-Cyclohexen-1-ol. [d] Benzaldehyde. [e] 4-
methylbenzaldehyde. [f] Limonene oxide. [g] Unidentified product.
Eur. J. Inorg. Chem. 2013, 5020–5026
5023
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
as well. To understand the full catalytic behavior, we tested
all possible solvents with cyclooctene and styrene oxidation
and the optimum solvent is acetonitrile. The conversion se-
quence follows the order CH3CN Ͼ CH3COCH3 Ͼ EtOH
Ͼ MeOH Ͼ CHCl3 Ͼ CH2Cl2 but, for the product selectiv-
ity for styrene conversion, the order is reverse (Table 4).
Table 4. Influence of solvent on the catalyzed epoxidation reaction
of compound 1.[a]
Figure 4. Conversion (%) vs. time plot for comparison between re-
actions with a radical scavenger and a control reaction without
adding a radical scavenger. Reaction conditions were the same as
given in the footnote of Table 1. (1) Without adding a radical scav-
enger in acetonitrile (3) with a radical scavenger in acetonitrile (2)
Without adding a radical scavenger in ethanol (4) with a radical
scavenger in ethanol.
[a] Reaction conditions were the same as given in the footnote of
Table 2.
phase of the reaction mixture was collected by filtration at
This study unveils the necessity of optimum polarity of the reaction temperature after 6 h and the residual activity
the solvent needed for higher conversion of the substrates. of the supernatant solution after separation of the catalysts
To observe the temperature effect we monitored the reac- was studied. The leaching test was done as follows: the li-
tion conditions from room temperature to 80 °C, a good quid phase of a run reaction was separated from the solid
yield with good product selectivity was obtained at the catalyst and new reagents were then added to this liquid
moderate temperature of 60 °C. Higher temperature results phase, forming a fresh reaction mixture, and the composi-
in higher conversion but for further oxidation, the epoxide tion of this fresh reaction mixture was determined by GC.
goes to different oxidized products (for styrene, only 30% This homogeneous reaction mixture was treated as a stan-
epoxide was found) which clearly demonstrates the vital dard catalytic solution. After 6 h, we started redetermi-
role of the temperature on the reaction course (Table S2, nation of the composition of the reaction mixture at 3
for oxidation in ethanol see Table S3). The catalyst amount hourly intervals, no further conversion was observed which
also has some effect on the reaction conversion. The proves the absence of any active species in solution and also
amount of catalyst used was varied from 0.5 to 3 mg. It is clearly demonstrates that metal is not leached from the
interesting to note that with increasing the catalyst amount framework during epoxidation reactions (Table S6 and Fig-
the conversion also increased but reached almost maximum ure S8). Atomic absorption spectrometric analysis (sensitiv-
at 2 mg. Therefore, this weight can be fixed as the optimum ity level 0.001 ppm) was also carried out after collecting the
one for the reactions (Table S4 and Figure S6 for aceto- liquid phase of the reaction mixtures and it confirms the
nitrile and Table S5 and Figure S7 for ethanol, respectively). absence of metal ions in the liquid phase. The solid catalyst
Thus the overall reaction study shows that this compound was also characterized by powder X-ray and FTIR spec-
is superior and highly efficient besides the results published troscopy to check the stability after the reactions. After the
in recent studies.[18,26] To understand the reaction mecha- catalytic reactions are over, solid catalyst was recovered by
nism and whether it is going through the radical pathway, centrifugation and washed with fresh acetonitrile several
we investigated our system by introducing a radical scaven- times and dried in air in an oven. Comparison of X-ray
ger as reported earlier.[30] Cyclooctene epoxidation was diffraction patterns and FTIR spectroscopic peaks of the
studied in both the solvent (acetonitrile and ethanol) and fresh and recovered catalysts convincingly demonstrates
in the presence of a radical scavenger (e.g. hydroquinone). that the structural integrity of 1 remains unaltered after the
During the reactions, the scavenger was introduced after 3 h epoxidation reactions (see Supporting Information Figures
and after that point, the reactions do not proceed further S9 and S10). Notably, the recovered catalyst can be reused
(Figure 4). This result clearly indicates that the reaction in epoxidation reactions several times with no considerable
pathway follows the usual radical chain mechanism.
loss of activity (Table 5). The conversion profiles of the
The advantage of heterogeneous catalysts over homogen- epoxidation reaction of cyclooctene using fresh and reco-
eous ones is that they can be recovered from the reaction vered catalyst remained virtually the same up to 5th cycle
mixtures by simple filtration and recycling. To test metal (Figure S11). These results support the true heterogeneity
leaching of the solid catalyst during reaction, the liquid and admirable efficiency of compound 1.
Eur. J. Inorg. Chem. 2013, 5020–5026
5024
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
tered by suction and washed with copious amount of distilled water
and dried in air. Powder XRD indicates the formation of the pure
sample. Elemental analysis of 1 C6H6O3NPCo: calcd Co 25.62, C
31.30, H 2.60, N 6.08; found Co 25.92, C 30.52, H 2.71, N 6.45.
The FTIR spectrum of 1 shows two strong absorptions at 1002 and
1071 cm–1 which can be assigned to the terminal P=O stretching
and P–O vibrations, respectively. Bands at 1134 and 1162 cm–1 were
assigned to ν(P–O) vibrations. Notably, the same compound,
{Co(3-pmp)}n 1 was obtained in this process when other bases such
as LiOH, CsOH and NaN3 were used in place of imidazole.
Table 5. Catalytic efficacy of the recovered catalyst at different
cycles.[a]
Catalytic Reaction: Catalytic reactions were performed in a three-
necked flask (50 mL) equipped with a reflux condenser, magnetic
stirrer, and gas inlet. Then olefin (10 mmol), isobutyraldehyde
(10 mmol) and solvent (20 mL) were added to the flask and im-
mersed in an oil bath kept at 60 °C under vigorous stirring. Bub-
bling air at atmospheric pressure was flown through the reaction
mixture at a flow rate of ca. 5 mL/min. The mixture was then equil-
ibrated to desired temperature in an oil bath. The reaction was
started after addition of catalyst (2 mg) into the reaction mixture.
The products of the epoxidation reactions were collected at dif-
ferent time intervals and were identified and quantified by gas
chromatography.
[a] Reaction conditions were the same as given in the footnote of
Table 2. [b] Turn over frequency (TOF): mol converted per mol of
active site per hour.
Conclusions
In summary, a new cobalt phosphonate has been
hydrothemally synthesized and its structure determined by
single crystal X-ray diffraction. The CoII in this compound
is tetracoordinated with oxygen and nitrogen atoms. This
tetrahedral cobalt phosphonate can be used as selective cat-
alyst for the epoxidation of a variety of olefins using air
and IBA at a moderate temperature. It behaved as a hetero-
geneous catalyst, can easily be recovered after reaction by
filtration and washing, and be subsequently used without
significant deactivation of the catalyst due to either leaching
of active species or degradation of the structure. This study
shows a new vista for designing highly active MOFs for the
molecular oxygen activation.
X-Ray Crystallography: X-ray crystal structure determination: X-
ray diffraction data for 1 were collected on a Bruker SMART
APEX CCD X-ray diffractometer using graphite-monochromated
Mo-Kα radiation (λ = 0.71073 Å). The determination of integrated
intensities and cell refinement were performed with the SAINT[31]
software package using a narrow-frame integration algorithm. An
empirical absorption correction (SADABS)[32] was applied. The
structure was solved by direct methods and refined using full-ma-
trix least-squares techniques against F2 with anisotropic displace-
ment parameters for nonhydrogen atoms with the programs
SHELXS97 and SHELXL97.[33] The hydrogen atoms of the C–H
bonds were placed at geometrical positions and refined with Uiso
= 1.2Ueq of the atom to which they are attached. In the final differ-
ence Fourier maps there were no remarkable peaks except the ghost
peaks surrounding the metal centers. Details of the crystal data and
relevant refinement parameters for compound 1 are given in Table
S7.
Experimental Section
CCDC number 900738 contains the supplementary crystallo-
graphic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
Physical Measurements: FTIR spectra of samples suspended in
KBr pellets were measured on a Unicam Mattson Mod 7000 spec-
trophotometer equipped with a DTGS CsI detector. The cobalt
content of the sample was estimated on a Varian Techtron AA-
ABQ atomic absorption spectrometer. The powder X-ray diffrac-
tion (XRD) patterns of the samples were recorded with a Scintag
XDS-2000 diffractometer using Cu-Kα radiation. The products of
the catalytic reactions were identified and quantified by a Varian
CP-3800 gas chromatograph. Magnetic measurements were per-
formed using a Quantum Design Squid-VSM magnetometer. DC
measurements were performed under a static field of 0.01 T. Mag-
netization measurements at different fields at room temperature
confirm the absence of ferromagnetic impurities. Data were cor-
rected for the sample holder and diamagnetism was estimated from
PascalЈs constants.
Supporting Information (see footnote on the first page of this arti-
cle): ORTEP diagram, TG profile, PXRD, FTIR, more catalytic
results (Figures S1–S11, Tables S1–S7).
Acknowledgments
R. S. wishes to thank Fundação para a Ciência e a Tecnologia
(FCT) for a postdoctoral fellowship (SFRH/BPD/71798/2010). The
work was also financially supported by the Programa Operacional
Ciência e Inovação 2010 (POCI2010), the Fundo Social Europeu
(FSE) and the Fundo Europeu de Desenvolvimento Regional
(FEDER).
Synthesis of {Co(3-pmp)}n (1): The phase pure compound was syn-
thesized by hydrothermal treatment of cobalt nitrate hexahydrate
(0.291 g, 1 mmol) with 3-pyridylmethyl phosphonic acid (0.091 g,
0.5 mmol) in the presence of imidazole (0.068 g, 1 mmol) dissolved
in milliQ water (10 mL) in a 25 mL teflon-lined autoclave for three
days at 170 °C followed by slow cooling the autoclave. Dazzling
violet needle shaped crystals (yield 70% based on the metal ion)
were formed during hydrothermal treatment. The crystals were fil-
[1] S. Kitagawa, K. Uemura, Chem. Soc. Rev. 2005, 34, 109.
[2] M. B. Duriska, S. M. Neville, J. Lu, S. S. Iremonger, J. F. Boas,
C. J. Kepert, S. R. Batten, Angew. Chem. 2009, 121, 9081; An-
gew. Chem. Int. Ed. 2009, 48, 8919.
[3] A. M. Ako, V. Mereacre, R. Clérac, W. Wernsdorfer, I. J. Hew-
itt, C. E. Anson, A. K. Powell, Chem. Commun. 2009, 544.
Eur. J. Inorg. Chem. 2013, 5020–5026
5025
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
[4]
[19] V. A. Blatov, A. P. Shevchenko, V. N. Serezhkin, J. Appl. Crys-
tallogr. 2000, 33, 1193.
[20] T. Yamada, T. Takai, O. Rhode, T. Mukaiyama, Chem. Lett.
1991, 20, 1.
[21] U. Neuenschwander, F. Guignard, I. Hermans, ChemSusChem
2010, 3, 75.
[22] P. Roy, M. Nandi, M. Manassero, M. Ricc´o, M. Mazzani, A.
Bhaumik, P. Banerjee, Dalton Trans. 2009, 9543.
[23] M. H. Alkordi, Y. Liu, R. W. Larsen, J. F. Eubank, M. Edda-
oudi, J. Am. Chem. Soc. 2008, 130, 12639.
[24] a) Y. Lu, M. Tonigold, B. Bredenkötter, D. Volkmer, J. Hitz-
bleck, G. Langstein, Z. Anorg. Allg. Chem. 2008, 634, 2411; b)
M. Tonigold, Y. Lu, B. Bredenkötter, B. Rieger, S. Bahnmüller,
J. Hitzbleck, G. Langstein, D. Volkmer, Angew. Chem. 2009,
121, 7682; Angew. Chem. Int. Ed. 2009, 48, 7546.
[25] J. Zhang, A. V. Biradar, S. Pramanik, T. J. Emge, T. Asefa, J.
Li, Chem. Commun. 2012, 48, 6541.
S. R. Miller, G. M. Pearce, P. A. Wright, F. Bonino, S. Chavan,
S. Bordiga, I. Margiolaki, N. Guillou, G. Ferey, S. Bourrelly,
P. L. Llewellyn, J. Am. Chem. Soc. 2008, 130, 15967.
M. Pramanik, M. Nandi, H. Uyama, A. Bhaumik, Catal. Sci.
Technol. 2012, 2, 613.
T.-H. Yang, Y. Liao, L.-M. Zheng, R. E. Dinnebier, Y.-H. Sua,
J. Mac, Chem. Commun. 2009, 3023.
K. J. Gagnon, H. P. Perry, A. Clearfield, Chem. Rev. 2012, 112,
1034.
a) M. T. Wharmby, J. P. S. Mowat, S. P. Thompson, P. A.
Wright, J. Am. Chem. Soc. 2011, 133, 1266; b) S. R. Miller, S.
Chavan, G. M. Pearce, S. Bordiga, P. A. Wright, I. Margiolaki,
S. Bourrelly, F. Bonino, N. Guillou, P. L. Llewellyn, G. Férey,
J. Am. Chem. Soc. 2008, 130, 15967; c) A. Zavras, J. A. Fry,
C. M. Beavers, G. H. Talboa, A. F. Richards, CrystEngComm
2011, 13, 3551.
[5]
[6]
[7]
[8]
[26] M. J. Beier, W. Kleist, M. T. Wharmby, R. Kissner, B. Kim-
merle, P. A. Wright, J.-D. Grunwaldt, A. Baiker, Chem. Eur. J.
2012, 18, 887.
[27] R. Sen, S. Bhunia, D. Mal, S. Koner, Y. Miyashita, K.-I. Oka-
moto, Langmuir 2009, 25, 13667.
[28] R. Sen, D. K. Hazra, M. Mukherjee, S. Koner, Eur. J. Inorg.
Chem. 2011, 2826.
[29] a) F. X. L. i Xamena, O. Casanova, R. G. Tailleur, H. Garcia,
A. Corma, J. Catal. 2008, 255, 220; b) P. Buranaprasertsuk, Y.
Tangsakol, W. Chavasiri, Catal. Commun. 2007, 8, 310; c) R.
Sen, D. Saha, S. Koner, Catal. Lett. 2012, 142, 124.
[30] A. Zhang, L. Li, J. Li, Y. Zhang, S. Gao, Catal. Commun. 2011,
12, 1183.
[31] Bruker, APEX 2, SAINT, XPREP, Bruker AXS Inc., Madison,
Wisconsin, USA, 2007.
[32] Bruker, SADABS, Bruker AXS Inc., Madison, Wisconsin,
USA, 2001.
[9]
[10]
B. M. Trost, Science 1991, 254, 1471.
R. A. Sheldon, Metalloporphyrins in Catalytic Oxidations, CRC
Press, New York, 1994.
[11] V. B. Sharma, S. L. Jain, B. Sain, Tetrahedron Lett. 2003, 44,
383.
[12] S. Koya, Y. Nishioka, H. Mizoguchi, T. Uchida, T. Katsuki,
Angew. Chem. 2012, 124, 8368; Angew. Chem. Int. Ed. 2012,
51, 8243.
[13] R.-M. Wang, E.-X. Hao, G.-R. Shen, Y.-F. He, Z.-Q. Lei, J.
Appl. Polym. Sci. 2009, 111, 1999.
[14] T. Joseph, S. B. Halligudi, C. Satyanarayan, D. P. Sawant, S.
Gopinathan, J. Mol. Catal. A 2001, 168, 87.
[15] H. Lü, Y. Zhang, Z. Jiang, C. Li, Green Chem. 2010, 12, 1954.
[16] B. M. Choudary, M. L. Kantam, A. Rahman, C. V. Reddy,
K. K. Rao, Angew. Chem. 2001, 113, 785; Angew. Chem. Int.
Ed. 2001, 40, 763.
[17] A. P. Woodham, G. Meijer, A. Fielicke, Angew. Chem. 2012,
124, 4520; Angew. Chem. Int. Ed. 2012, 51, 4444.
[18] L. Chen, Y. Yang, Z. Guo, D. Jiang, Adv. Mater. 2011, 23,
3149.
[33] G. M. Sheldrick, SHELXS97 and SHELXL97, Programs for
Crystal Structure Solution and Refinement, University of
Göttingen, Germany, 1997.
Received: May 19, 2013
Published Online: July 30, 2013
Eur. J. Inorg. Chem. 2013, 5020–5026
5026
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim