Mixed (Fe2+ and Cu2+) double metal hexacyanocobaltates as solid catalyst for the aerobic oxidation of oximes to carbonyl compounds

Article abstract of DOI:10.1016/j.jcat.2013.12.011

Mixed Iron and copper hexacyanocolbatate was found to be a suitable heterogeneous and recoverable catalyst for the aerobic oxidation of oximes to the corresponding ketone. The reaction can be conveniently carried out in water-ethanol 1-1 mixture as solvent. The tinme-conversion plots shows the presence of an induction period that do not correspond to the leaching of metal ions or to the damage of the crystal structure of the material. The proposed reaction mechanism is based on the cooperation of the Lewis acidity of iron with the ability of copper to interact with oxygen. Given the remarkable stability of metal hexacyanocobaltates and the large diversity of metals that can contain, our reports opens the way for the general use of these affordable and accessible solids as heterogeneous catalysts.

Full text of DOI:10.1016/j.jcat.2013.12.011

Journal of Catalysis 311 (2014) 386–392
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Mixed (Fe2+ and Cu2+) double metal hexacyanocobaltates as solid
catalyst for the aerobic oxidation of oximes to carbonyl compounds
b,
Alma García-Ortiz a,b, Abdessamad Grirrane b, Edilso Reguera a, , Hermenegildo García
⇑
⇑
a Centro de Investigación en Ciencia Aplicada y Tecnología Avanzada-Unidad Legaria, Instituto Politécnico Nacional, México, D.F., Mexico
b Instituto Universitario de Tecnología Química CSIC-UPV, Universidad Politécnica de Valencia, Av. De los Naranjos s/n, 46022 Valencia, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
Mixed Iron and copper hexacyanocolbatate was found to be a suitable heterogeneous and recoverable
catalyst for the aerobic oxidation of oximes to the corresponding ketone. The reaction can be conve-
niently carried out in water-ethanol 1-1 mixture as solvent. The tinme-conversion plots shows the pres-
ence of an induction period that do not correspond to the leaching of metal ions or to the damage of the
crystal structure of the material. The proposed reaction mechanism is based on the cooperation of the
Lewis acidity of iron with the ability of copper to interact with oxygen. Given the remarkable stability
of metal hexacyanocobaltates and the large diversity of metals that can contain, our reports opens the
way for the general use of these affordable and accessible solids as heterogeneous catalysts.
Ó 2014 Elsevier Inc. All rights reserved.
Received 5 December 2013
Accepted 24 December 2013
Available online 31 January 2014
Keywords:
Catalytic aerobic oxidation
Exacyanocobaltates as solid catalysts
Oxidation of oximes
1. Introduction
For this reason double metal cyanides are used as heterogeneous
catalysis for the industrial ring epoxide aperture with alcohols
Double metal cyanides have a crystal structure constituted by
octahedral building units in which a divalent, trivalent or tetrava-
lent transition metal is octahedrally coordinated by the C atom to
six cyanide ligands. These hexacyanometallate building blocks are
spatially arranged and held in the lattice by a second divalent tran-
sition metal cation that also coordinates octahedrally through the
N atoms of the cyanide ligands of the building units [1]. Fig. 1 illus-
trates the porous structure of double metal cyanides and the cubic
arrangement of their lattice where the larger pores can be formed
by systematic vacancies for the building block. The crystal struc-
tures of double metal hexacyanide usually belong to the Fm3m
and Pm3m space group where the metals linked to the N end of
CN ligands are randomly distributed in their structural position
to form solid solution [1a,1b,2].
Double metal cyanides are obtained through a simple synthetic
procedure consisting in the precipitation of solutions of hexacy-
anometallate by addition of an aqueous solution of the appropriate
mixture of divalent metals [2b,3]. Besides affordable starting mate-
rials, double metal cyanides show considerably versatility in the
inner and outer metals, even allowing the presence of a third
divalent metal cation in various proportions [3]. In addition, these
solids can be prepared in multigram scale in scalable and repro-
ducible manner. The main structural characteristic of double metal
cyanides is that they have high content in two or more transition
metals together with a high thermal and chemical stability [2c].
leading to polyetherols [4]. However, double metal cyanides offer
many other possibilities that still have to be realized [5]. Recently
de Vos and coworkers have reported the use of these materials as
solid catalyst for hydroaminacion of alkenes and alkynes [1c,6]. It
was found that this reaction occurs exclusively on the external sur-
face on the catalyst particles, since the pore opening of the cubic
lattice (about 0.5 nm in Fig. 1) does not allow the internal diffusion
of reagents. Nevertheless, high catalytic activities comparable or
higher than those achieved by using other conventional solid cata-
lyst were obtained [7].
Continuing with the exploitation of the opportunities that dou-
ble metal cyanides offer as solid catalysts and considering that
transition metals have excellent activity for oxidation reactions,
in the present contribution we describe the use of metal hexacya-
nocobaltates for the aerobic oxidation of oximes to the correspond-
ing carbonyl compounds. One topic with high industrial relevance
in green chemistry is the use of molecular oxygen in combination
with a catalyst to produce oxidation of organic substrates [7,8].
Classically, stoichiometric amounts of transition metal oxidation
reagents have been used to perform oxidation of oximes and other
organic compounds and although this type of reaction does not
require catalysts, they generated a considerable amount of, fre-
quently, toxic wastes. In contrast, oxidation with oxygen is charac-
terized by high activation barriers and poor selectivity toward as
single product. These two limitations can be overcome by using
catalysts that should decrease the activation energies and direct
the process toward a single product. Catalytic oxidative deoxima-
tion is a well-established method to obtain the corresponding
⇑
Corresponding authors. Fax: +34 95 378 7809.
E-mail address: hgarcia@qim.upv.es (H. García).
0021-9517/$ - see front matter Ó 2014 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.jcat.2013.12.011

A. García-Ortiz et al. / Journal of Catalysis 311 (2014) 386–392
387
Fig. 1. Crystal structure of double metal hexacyanocobaltate showing the lattice
and the dimensions of cubic pores; the larger pores arise from vacancies in the
framework.
ketone [9]. In the present case, we will show that mixed-metal, Cu-
containing metal hexacyanocobaltates are stable and reusable sol-
ids promoting the oxidation of oximes by molecular oxygen.
Fig. 2. Typical morphology of double metal cyanide microcrystals. The obtained
powder is formed by small cubic crystals of about 1 lm of size.
2. Results and discussion
For the present study, we have prepared a series of twelve metal
hexacyanocobaltates that were obtained by precipitation of a solu-
tion of hexacyanocobaltate and aqueous solution of a divalent me-
tal (Mn2+, Co2+, Ni2+ and Cu2+) as well as mixtures of two of these
divalent metals. The selection of these materials was made on
purpose in order to have present in the solid hexacyanocobaltate
ions that can provide the structural integrity and other two transi-
tion metal ions that can interact with oxygen and introduce Lewis
acidity to bind hydroxylamine functionalities of the oximes. These
two transition metals may also participate in other elementary
steps occurring during oxidations such as decomposition of hydro-
peroxide intermediate and radical trapping. In the selection of the
double metal hexacyanocobaltate, we have used previous informa-
tion in the field of metal organic frameworks as catalysts that have
shown that mixtures of cobalt and copper imidazolate MOFs are
active and selective to perform the aerobic oxidation of tetrahydro-
naphthalene to the corresponding ol/one mixtures containing a
minimal amount of the correspond hydroperoxide [10]. This infor-
mation has shown that the efficiencies of catalyst containing two
metals, such copper and cobalt, are the best performing materials.
The list of the materials prepared and the most relevant physico-
Fig. 3. HR-TEM image of Cu3[Co(CN)6]2.
chemical properties and infrared vibration wavenumber for m(CN)
and d (CoCN) [11] are summarized in Table 1 (see Figs. S1–S11 in
the Supporting Information for details). Figs. S12–S23 and S25–
S36 in the Supporting Information provide a collection of XRD
and thermogravimetric curves, respectively. In addition, represen-
tative SEM and HR-TEM images of double metal cyanides are
shown in Figs. 2 and 3 in which the high crystallinity of these
materials are observed. Fig. 4, on the other hand, presents an
example of CO2 adsorption isotherm indicating that double metal
cyanides exhibit significant micro-/nanoporosity [1c].
Table 1
Physicochemical properties and infrared vibration wavenumber for the twelve
materials prepared.
Structural
BET surface
area (m2 gÀ1
Pore volume
IR bands m (CN)/
formula (Xa)
)
(cm3 gÀ1
)
d (CoCN) (cmÀ1
)
In the initial stage of our work, we screened the catalytic activ-
ity of the series of metal hexacyanocobaltates for the oxidation of
cyclohexanone oxime and oxygen at low pressure in a mixture of
ethanol/water as solvent. The results showing the oxime conver-
sion and cyclohexanone yield at final reaction times are presented
in Table 2. Reaction times were screened from 1 to 30.5 h. From
this preliminary screening, we can conclude that, in agreement
with our initial hypothesis, those cobaltates containing Cu were
the most efficient ones and, therefore, they were selected for addi-
tional studies.
Cu3[Co(CN)6]2 (5)
773
787
808
782
834
795
769
781
803
811
797
783
0.32
0.34
0.36
0.33
0.36
0.39
0.37
0.35
0.39
0.38
0.36
0.40
2181/475
2171/465
2191/465
2181/475
2181/454
2171/450
2173/462
2170/454
2171/454
2181/454
2181/465
2171/465
MnCu2[Co(CN)6]2 (8)
Ni1.3Cu1.7[Co(CN)6]2 (8)
FeCu2[Co(CN)6]2 (7.5)
Ni3[Co(CN)6]2 (7)
Co3[Co(CN)6]2 (8)
Mn1.2Fe1.8[Co(CN)6] (8)
Fe1.4Ni1.6[Co(CN)6]2 (7.5)
Mn3[Co(CN)6]2 (8)
Co1.5Ni1.5(Co(CN)6)2 (8)
Co1.4Cu1.6(Co(CN)6) (8)
Fe1.4Co1.6(Co(CN)6)2 (8)
Considering that the organic reaction takes place on the
external surface due to the small pore size of double metal
a
X: molecules of water present in the formula calculated with TG curves.

388
A. García-Ortiz et al. / Journal of Catalysis 311 (2014) 386–392
Conversion of acetophenone oxime into acetophenone can also
be carried out both with air and with pure oxygen. The reaction
rate estimated using air was as 0.48 mmol hÀ1. This reaction rate
is essentially the same as that found for the same system in pure
oxygen, indicating that the reaction is under their present condi-
tions of zero order with respect to the oxygen pressure. However,
at long reaction time, we noted that the final conversion of aceto-
phenone oxime under air is lower (about 80%, 25 h) than for the
reaction carried out under oxygen suggesting that the catalyst
undergoes some deactivation when air is used.
In order to gain understanding on the origin of the induction
period observed for some Cu-containing hexacyanocobaltate cata-
lysts, we performed some additional experimental work. In the
first control, the solvent was contacted with de catalyst under
the experimental reaction conditions, but in the absence of aceto-
phenone oxime and the liquid phase analyzed by ICP-OES for the
presence of dissolved metal ions, while simultaneously the solid
was submitted to ICP-OES analysis and X-ray diffraction. The re-
sults show that the solid catalyst maintains its crystal structure
and chemical composition under these conditions and very low
amounts of Cu or Co were found in the liquid phase (below 1% of
metal content present in the solid catalyst). These measurements
rule out the possibility that the induction period could be due to
the leaching of metal ions from the solid to the solution or to the
partial dissolution of the solid.
Fig. 4. CO2 adsorption isotherms for the mixed series Co3ÀxNix[Co(CN)6]2. Inset: the
region of low coverage (low pressures) for the same isotherms. No significant
modification for the pore volume is observed when solid solutions involving
different metals are formed.
Table 2
Results of the cyclohexanone oxime oxidation by molecular oxygen in the presence of
double metal hexacyanocobaltates as catalysts. Reaction conditions: Cyclohexanone
oxime: 0.5 mmol, catalyst/substrate ratio: 5 mol%, solvent: 2 ml EtOH/H2O 1:1,
temperature: 100 °C, oxygen pressure: 5 bar.
In a second experiment, the reaction was initiated under the
typical reaction conditions but in the absence of acetophenone
oxime and the substrate was added after 90 min and the mixture
surveyed for conversion and selectivity. It was observed that under
these conditions, an induction period similar to that shown in
Fig. 5 takes place. Therefore, the induction period observed in the
time-conversion plot for Ni1.3Cu1.7(Co(CN)6)2 and Co1.4Cu1.6
(Co(CN)6)2 catalysts should reflect probably changes on the surface
of the crystal, probably by coordination of the alcohol or by the
interaction with molecular oxygen forming peroxo species. This
surface conditioning would not take place or would be much faster
in the case of FeCu2[Co(CN)6]2 catalyst and for this reason this cat-
alyst was selected for further studies.
Catalyst
Time (h)
Conv. (%)
Yield (%)
Cu3[Co(CN)6]2
1
2
1
2
30.5
30.5
23
23
18
3
94
55
100
96
20
39
38
17
20
5
82
42
82a
77a
28
37
30
36
13
7
MnCu2[Co(CN)6]2
Ni1.3Cu1.7[Co(CN)6]2
FeCu2[Co(CN)6]2
Ni3[Co(CN)6]2
Co3[Co(CN)6]2
Mn1.2Fe1.8[Co(CN)6]2
Fe1.4Ni1.6[Co(CN)6]2
Mn3[Co(CN)6]2
Co1.5Ni1.5(Co(CN)6)2
Co1.4Cu1.6(Co(CN)6)2
Fe1.4Co1.6(Co(CN)6)2
3
23
100
33
100
36
a
Formation of 1,1-diethoxycyclohexanone was observed.
Fig. 6 shows the time-conversion plot for the reaction of aceto-
phenone oxime in the presence of FeCu2[Co(CN)6]2 catalyst. The
plot shows a pseudo first-order kinetics for the disappearance for
the acetophenone oxime and that acetophenone is the only and
stable product accounting completely for the disappearance of
the starting material.
hexacyanometallate [1c,2c] and that as a general observation in
heterogeneous catalysis, the solvent should play a role in the pro-
cess by reversible adsorption on the metal sites competing with
the substrate modulating the reaction rate, several solvents were
tested as reaction medium. In fact, as can be seen in Table 3 using
Cu3[Co(CN)6]2 as catalyst, cyclohexanone oxime conversion and
ketone selectivity were lower when the reaction is carried out in
toluene or in pure ethanol than in mixtures ethanol/water.
Reaction times from 1 to 22 h were screened. It was found that a
convenient reaction media is a mixture of ethanol/water in appro-
priate proportions 1:1.
After the initial screening, further work was pursued using
acetophenone oxime as substrate, the mixture of ethanol–water
1:1 as a solvent and a series of hexacyanocobaltates at 5 mol% as
catalysts. Also in these tests, almost complete conversions with
high selectivity toward acetophenone were observed for Cu-con-
taining hexacyanocobaltates. The results are presented in Table 4.
Although hexacyanocolbatates exhibit high conversion at final
reaction times, the temporal profile of the reaction was remarkably
different depending on the catalyst. Thus, for the FeCu2[Co(CN)6]2
catalyst, no induction period was observed, while, in contrast, for
Ni1.3Cu1.7[Co(CN)6]2 and Co1.4Cu1.6[Co(CN)6]2 the reaction in the
first hour occurred in a very low extent, indicating the existence
of an induction period for the process. Fig. 5 shows the temporal
profiles for acetophenone formation catalyzed by the set of Cu-
containing hexacyanocobaltates.
Fig. 7 shows, on the other hand, the kinetics of two consecutive
reuses of the same FeCu2[Co(CN)6]2 catalyst with an accumulated
9 h use of the solid. After one run, the solid catalyst was recovered
by filtration, washed with pure ethanol and reused for a consecu-
tive reaction without further treatment. As can be seen in Fig. 7,
the temporal profiles of the fresh and reused catalyst differ at short
reaction times, where an induction period of about 15 min was ob-
served. However, the initial reaction rate after the induction period
and final conversion of the fresh and first and second reuses are al-
most coincident, indicating the stability and recyclability of the
FeCu2[Co(CN)6]2 as catalyst. Since the appearance of the induction
period could be a sign of deactivation, the catalyst was surveyed
for crystallinity by XRD after its third consecutive use. The same
diffraction pattern as the fresh material was recorded indicating
the stability of the FeCu2[Co(CN)6]2 material under the present
reaction conditions (see Fig. S24 in the Supporting Information).
Further understanding of the origin of the induction period is
needed to rationalize its appearance upon reuse (vide supra for
the data on the induction period).
Besides cyclohexanone and acetophenone oximes, the scope of
the FeCu2[Co(CN)6]2 catalyst activity was further expanded by
using benzophenone and carvone oximes as substrates. It should

A. García-Ortiz et al. / Journal of Catalysis 311 (2014) 386–392
389
Table 3
Influence of the solvent on the performance of Cu3[Co(CN)6]2 as catalyst. Reaction conditions: Cyclohexanone oxime: 0.5 mmol, solvent: 2 ml, temperature: 100 °C, oxygen
pressure: 5 bar.
HO
N
O
O2 / solvent
catalyst
Entry
Catalyst (mol%)
Solvent time (h)
Conv. (%)
79
Yield (%)
49
1
2
3
4c
5
6
7
8
9
10
10
5
Toluene
22
EtOH
48
94
46
74
82
76
70
48
31a
82
35
59
63a
62
55
43
5
EtOH/H2O 1:1b
1
5
H2O/EtOH 1:1d
21
5
H2O/EtOH 1:1
6
EtOH/H2O 1.5:0.5
2
H2O/EtOH 1:1
16
EtOH/H2O 1:1
23
5
3
3
1
EtOH/H2O 1:1
28
a
b
c
Formation of 1,1-dietoxycyclohexanone was observed.
Ethanol was added before that water.
Air at 5 bar.
d
Water was added before Ethanol.
Table 4
Catalytic activity for the conversion of acetophenone oxime to acetophenone. Reaction conditions: Acetophenone oxime: 0.5 mmol, catalyst/substrate ratio: 5 mol%, solvent:
EtOH/H2O 1:1; 2 ml, temperature: 100 °C, oxygen pressure: 5 bar. Selectivity to acetophenone <99% in all cases.
OH
N
O
O2 / solvent
Catalyst
Catalyst
Time (h)
Conv. (%)
Yield (%)
Cu3[Co(CN)6]2
3
3
3
3
25
11
97
100
96
8
97
100
92
Ni1.3Cu1.7[Co(CN)6]2
FeCu2[Co(CN)6]2
Co1.4Cu1.6(Co(CN)6)2
FeCu2[Co(CN)6]2
80
74
90
60
30
90
60
30
0
0
0
1
2
3
0
1
2
3
Time (h)
Time (h)
Fig. 5. Time-yield plots for the oxidation of acetophenone oxime to acetophenone
catalyzed by FeCu2[Co(CN)6]2 (Â), Co1.4Cu1.6[Co(CN)6]2 (N), Ni1.3Cu1.7[Co(CN)6]2 (ꢀ),
Cu3[Co(CN)6]2 (j), FeCu2[Co(CN)6]2 (*). Reaction conditions: Acetophenone oxime:
0.5 mmol, catalyst/substrate ratio: 5 mol%, solvent: 2 ml of EtOH/H2O 1:1, temper-
ature: 100 °C, oxygen pressure: 5 bar.
Fig. 6. Time-percentage plot for the oxidation of acetophenone oxime by molecular
oxygen using FeCu2[Co(CN)6]2 as catalyst. Conversion of acetophenone oxime (ꢀ);
Yield of acetophenone (j). Reaction conditions: Acetophenone oxime: 0.5 mmol,
catalyst/substrate ratio: 5 mol%, solvent: 2 ml of EtOH/H2O 1:1; temperature:
100 °C, oxygen pressure: 5 bar.
be noted that carvone is used in the fragrance industry and one of
the commercial synthetic routes is based on deoximation of
carvone oxime. In spite of the larger molecular dimension of these
substrates also almost a full conversion to benzophenone and carv-
one were achieved indicating that FeCu2[Co(CN)6]2 catalyst is also
efficient for the oxidation of bulk aromatic oximes. This reactivity

390
A. García-Ortiz et al. / Journal of Catalysis 311 (2014) 386–392
quenchers (hydroquinone and ascorbic acid), carboxylic acids
90
60
30
0
(ascorbic and benzoic acid), and carboxylate anion (NaAcO).
Fig. 8 shows time-conversion plots for the acetophenone oxime
oxidation in the presence of these additives. Thus, it was found that
the hydroquinone leads to the appearance of an induction period
that can be interpreted as the time required to produce the oxida-
tion of hydroquinone. According to this, the reaction mechanism is
likely to involve some radical species that will be quenched by
hydroquinone. The presence of carboxylic acids, on the other hand,
increases the reaction rate, suggesting that protonation of aceto-
phenone oxime or a possible intermediate accelerates the reaction.
On the contrary, addition of acetate (1%) leads to a complete
quenching of the reaction that can be interpreted as arising from
the strong adsorption to the acetate on the Lewis acid sites of the
catalyst.
0
1
2
3
Time (h)
Fig. 7. Time-yield plot for the consecutive runs using the same FeCu2[Co(CN)6]2
sample as catalyst; (a) first run: ꢀ; (b) second run: j; (c) third run N. Reaction
conditions: Acetophenone oxime: 0.25 mmol, catalyst/substrate ratio: 5 mol%,
solvent: 1 ml of EtOH/H2O 1:1; temperature: 100 °C, oxygen pressure: 5 bar.
We also considered that possibility that the mechanism in-
volves partially acid-catalyzed hydrolysis of oxime, particularly
taking into account that oxime conversion and ketone selectivity
in the presence of water is higher than in pure ethanol and that
acids increase the reaction rate. To address this issue, we per-
formed controls using HCl (1 mol%) or H2SO4 (5 mol%) in ethanol/
water mixtures under the same conditions as those indicated in Ta-
ble 2. It was observed that using HCl or H2SO4 as Brönsted acid cat-
alysts cyclohexanone conversion at 1 h reaction time was 8% and
24%, with ketone yields of 7% and 21%, respectively. These values
contrast with the 94% oxime conversion and 82% cyclohexanone
yield achieved under the same conditions with Cu3[Co(CN)6]2
and indicate that the relative contribution of acid hydrolysis to
deoximation should be relatively minor.
90
60
30
0
0
0,25
0,5
0,75
1
Time (h)
Considering the relative acidity of the Fe2+ with respect to Cu2+
,
Fig. 8. Time-yield plots for the oxidation of acetophenone oxime in the presence of
hydroquinone 10 mol% (Â), ascorbic acid 10 mol% (ꢀ), benzoic acid 5 mol% (*),
NaAcO 1 mol% (N) and without any additive (j). Reaction conditions: Acetophe-
none oxime: 0.25 mmol, catalyst/substrate ratio: 5 mol%, solvent: EtOH/H2O 1:1;
1 ml, temperature: 100 °C, oxygen pressure: 5 bar. Amount of additive: hydroqui-
none, 2.8 mg; ascorbic acid, 4.5 mg; benzoic acid, 3 mg; NaAcO, 0.4 mg.
the Fe2+ ion will be the most likely binding site for interaction with
the oxime. Accordingly, Scheme 1 presents a mechanistic proposal
to rationalize the catalytic activity of double metal hexacyanoco-
baltates as solid catalysts in the aerobic oxidation of oximes.
According to this proposal, the oxime will bind to the most
acidic metal ion Fe2+, while Cu2+ will interact with oxygen and
the combination of these two metal sites adsorbing the oxime
(Fe2+) and the oxidizing reagent (Cu2+) will result in the formation
of the carbonyl compound by the oxidative cleavage of C@N. It
could be that the corresponding oxazyridine could be the reaction
intermediate. The role of the internal hexacyanocobaltate units in
this mechanism will be just structural, maintaining the crystallin-
ity of the material. In agreement with our proposal, other double
of benzophenone and carvone oxime strongly suggests that the
reaction take places exclusively on the external surface of the
hexacyanocobaltates, since both oximes are absolutely excluded
from intracrystalline diffusion.
Information about the reaction mechanism using FeCu2
[Co(CN)6]2 as catalyst was obtained by performing acetophenone
oxime oxidation in the presence of small percentages of radical
Scheme 1. Proposed mechanism for the aerobic oxidation of oximes to carboxylic compound in the presence of mixed double metal cyanides.

A. García-Ortiz et al. / Journal of Catalysis 311 (2014) 386–392
391
metal hexacyanocobaltates less efficient to interact with the oxime
will exhibit lower catalytic activity and the solid containing exclu-
sively copper ions will also be almost inactive due to the low affin-
ity to bind the oxime. However, on the other hand, the ability of
Cu2+ ions to bind molecular O2 will be reflected in the need for this
transition metal to have a high active catalyst.
involved metals (Mn2+, Fe2+, Co2+, Ni2+ y Cu2+). The resulting pre-
cipitates were aged for 24 h within the mother liquor, followed
by its separation by centrifugation. The solid fraction was repeat-
edly washed with distilled water to remove all the accompanying
ions and then air-dried until constant weight.
The crystal structure, thermal stability, and information on the
electronic structure for the materials under consideration are
available from previous studies [2c,13] and XRD powder patterns
and TG curves of the studied samples are provided in the Support-
ing Information Figs. S12–S36.
3. Conclusions
The present study constitutes one of the few examples illustrat-
ing the remarkable catalytic activity of (mixed) double metal hexa-
cyanocobaltates to promote oxidations with molecular oxygen.
Due to the narrow pore size, only the external surface of the solid
crystallites will be catalytically relevant. The potential of these so-
lid hexacyanocobaltates in heterogeneous catalysis is demon-
strated by the proposed reaction mechanism that is based on the
cooperation of two different metal sites to interact one preferen-
tially with the oxime and the other preferentially with molecular
oxygen. Also, we propose that the presence of functional groups
on the substrate able to bind with the external double metal hex-
acyanides ions such as OH groups is a necessary prerequisite for a
substrate to undergo a catalytic reaction in the presence of these
solids. Considering the easy synthesis of (mixed) double metal
hexacyanocobaltates and the large variety of transition metals that
can be incorporated in a wide range of proportions as well as the
interest in the developing oxidations using molecular oxygen, our
report opens the way for the general use of this type of materials
in aerobic oxidations of appropriate organic compounds.
4.3. Catalytic experiments
Catalytic experiments were performed in reinforced glass semi
continuous reactors equipped with temperature and pressure con-
trollers. For each reaction, a 2-ml mixture of reactants and solvent
was placed into the reactor (3 ml capacity) together with appropri-
ate amount of catalyst.
The reaction conditions for acid (HCl or H2SO4)-catalyzed
hydrolysis of cyclohexanone oxime (0..5 mmol) employed a cata-
lyst (HCl or H2SO4)/substrate ratio: 1 or 5 mol%, respectively, in
2 ml EtOH/H2O 1:1 as solvent at 100 °C under 5 bar oxygen
pressure.
In the cases in which no water was used as a solvent, conversion
and yields were estimated using dodecane as internal standard.
After sealing the reactor, air was purged by filling the reactor with
oxygen (5 bar) and pumping out three times before final pressuri-
zation of the reactor with O2 or air at 5 bar. The reactor was deeply
introduced into a stirring and temperature controller that was pre-
heated at the reaction temperature. During the experiment, the
pressure was maintained constant and the reaction mixture was
magnetically stirred at 1200 rpm. Aliquots were taken from the
reactor at different reaction times. Once the catalyst particles were
removed from the solution by centrifuging at 8000 rpm the sample
were analyzed by GC. The products were identified by GC–MS and
also by comparing their retention time with that of commercial
pure samples when available. When mixtures of water–ethanol
were used as a solvent, quantification was performed adding a
known weight of dodecane as external standard to aliquots of
the reaction mixtures diluted in ethanol. Only experiments with
mass balances further than 95% were considered.
4. Experimental
4.1. Materials and apparatus
The nature of the hexacyanocobaltates (III) was corroborated by
their IR spectra. The atomic metal ratio (T:Co) was estimated by
energy dispersive X-ray spectroscopy (EDS) analyses and the ex-
pected formula unit, TAx T3BÀx [Co(CN)6]2 Â H2O was established.
The hydration degree (x) was calculated from thermogravimetric
(TG) curves.
The structural characterization was carried out from X-ray dif-
fraction (XRD) data (see Supporting Informations for the patterns
Figs. S12–24).
Acknowledgments
IR spectra were recorded in an FT-IR spectrophotometer Nicolet
710. XRD powder patterns were obtained with Cu Ka radiation in
an HZG-4 diffractometer (from Carl Zeiss) and their evaluation car-
ried out using Dicvol program [12].
TG curves were analyzed in a METTLER TOLEDO TGA/SDTA851
and the specific surface area and porosity of the materials under
study was in a Micromeritics ASAP 2010 by N2 at 77 K after
exhaustive outgassing until constant weight.
Financial support by the Spanish ministry of Economy and
Competitiveness (Severo Ochoa and CTQ2012-32315) and General-
idad Valenciana (Prometeo 2012-014) is grateful acknowledged.
Alma Garcia thanks Mexican CONACYT for financial support for
her stay at Valencia. E.R. thanks the support provided by CONACyT
through the Projects 129048 and 174247.
Elemental analyses of metals in the liquid phase were per-
formed by inductively coupled plasma-optical emission spectros-
copy (ICP-OES) in a Varian 715-ES, after solid dissolution in
HNO3/HCl/HF aqueous solution. Gas chromatography (GC) was
performed using He as carrier gas with a Varian 3900 apparatus
equipped with an TRB-5MS column (5% phenyl, 95% poly-
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.jcat.2013.12.011.
References
methylsiloxane, 30 m, 0.25 mm  0.25
lm, Teknokroma). GC/MS
analyses were performed on an Agilent spectrometer equipped
with the same column as the GC and operated under the same
conditions.
[1] (a) G.W. Beall, D.F. Mullica, W.O. Milligan, Inorg. Chem. 19 (1980) 2876;
(b) D.F. Mullica, W.O. Milligan, G.W. Beall, W.L. Reeves, Acta Crystallogr., Sect.
B. B34(1978)3558.;
(c) A. Peeters, P. Valvekens, R. Ameloot, G. Sankar, C.E.A. Kirschhock, D. De Vos,
ACS Catal. 3 (2013) 597.
[2] (a) M.R. Hartman, V.K. Peterson, Y. Liu, S.S. Kaye, J.R. Long, Chem. Mater. 18
(2006) 3221;
4.2. Catalyst preparation
(b) S.S. Kaye, J.R. Long, J. Am. Chem. Soc. 127 (2005) 6506;
(c) J. Roque, E. Reguera, J. Balmaseda, J. Rodriguez-Hernandez, L. Reguera, C.L.F.
del, Microporous Mesoporous Mater. 103 (2007) 57.
The materials under study were prepared mixing aqueous
solutions (0.05 M) of K3[Co(CN)6] and of sulfate (0.063 M) of the

392
A. García-Ortiz et al. / Journal of Catalysis 311 (2014) 386–392
[3] C.P. Krap, J. Balmaseda, C.L.F. del, B. Zamora, E. Reguera, Energy Fuels 24 (2010)
[7] A. Grirrane, A. Corma, H. Garcia, J. Catal. 268 (2009) 350.
[8] A. Grirrane, A. Corma, H. Garcia, Science 322 (2008) 1661 (Washington, DC,
U.S.).
581.
[4] Yu, Guangzhou Huaxue 29 (2004) 47.
[5] (a) I. Kim, M.J. Yi, S.H. Byun, D.W. Park, B.U. Kim, C.S. Ha, Macromol. Symp. 224
(2005) 181;
[9] (a) Huaxue Jinzhan 24 (2012) 361.;
(b) G. Zhang, X. Wen, Y. Wang, W. Mo, C. Ding, J. Org. Chem. 76 (2011) 4665.
[10] I. Luz, i.X.F.X. Llabres, A. Corma, J. Catal. 276 (2010) 134.
[11] (a) K. Nakamoto, Infrared and Raman Spectra of Inorganic and Coordination
Compounds, John-Wiley & Sons, New York, Chichester, Brisbane, Toronto,
Singapore, 1986;
(b) X.-K. Sun, X.-H. Zhang, S. Chen, B.-Y. Du, Q. Wang, Z.-Q. Fan, G.-R. Qi,
Polymer 51 (2010) 5719;
(c) I. Kim, M.J. Yi, S.H. Pyun, D.W. Park, C.-S. Ha, Stud. Surf. Sci. Catal. 153
(2004) 239;
(d) J.L. Garcia, E.J. Jang, H. Alper, J. Appl. Polym. Sci. 86 (2002) 1553–1557;
(e) I.P. Singh, S. Singh, R. Patel, B. Mistry, M. Mehta, P.O. Otieno,
WO2010020053A1 (2010), Solid heterogeneous catalysts for the
esterification or transesterification of oils to produce biodiesel, 2010.;
(f) R.-J. Wei, X.-H. Zhang, B.-Y. Du, X.-K. Sun, Z.-Q. Fan, G.-R. Qi,
Macromolecules 46(2013)3693 (Washington, DC, U.S.).
[6] A. Peeters, P. Valvekens, F. Vermoortele, R. Ameloot, C. Kirschhock, D. De Vos,
Chem. Commun. 47 (2011) 4114 (Cambridge, U.K.).
(b) A.G. Sharpe, The Chemistry of Cyano Complexes of the Transition Metals,
Academic Press, New York, 1976.
[12] D. Louer, R. Vargas, J. Appl. Crystallogr. 15 (1982) 542.
[13] (a) J. Jimenez-Gallegos, J. Rodriguez-Hernandez, H. Yee-Madeira, E. Reguera, J.
Phys. Chem. C 114 (2010) 5043;
(b) E. Lima, J. Balmaseda, E. Reguera, Langmuir 23 (2007) 5752.