Catalyst@Metal Hybrids in a One-Pot Multistep Opposing Oxidation and Reduction Reaction Sequence

Article abstract of DOI:10.1002/cctc.201601386

We report the feasibility of the catalysis of contrasting reactions, oxidation and reduction, in a multistep process that also involves a condensation reaction all performed with hybrids of catalytic organometallic complexes entrapped within metals ([complex]@metal). Two routes were explored for the multistep process. The first route used a combination of an oxidizing Ir complex entrapped in Ag ([Ir]@Ag) and a reducing Rh complex entrapped in Ag ([Rh]@Ag). The second route used a single composite material, namely, [Ir]@Pd, in which [Ir] acted as the oxidizing catalyst, and Pd was the reducing catalyst. Both routes were efficient for the heterogeneous catalytic oxidation and reduction reaction sequence. Towards the goal of a one-pot multistep oxidation and reduction process, a detailed study of the catalytic alcohol oxidation properties of the heterogeneous [Ir]@Ag and [Ir]@Pd was performed. As the arsenal of available hydrogenation catalysts exceeds by far that of oxidations, this part of the study is, we believe, of general interest in itself, as a new alternative to heterogeneous oxidative catalysis.

Full text of DOI:10.1002/cctc.201601386

Accepted Article
Title: "Catalyst@metal hybrids in a one-pot multistep opposing
oxidation and reduction reactions-sequence"
Authors: Leora Shapiro and David Avnir
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemCatChem 10.1002/cctc.201601386
Link to VoR: http://dx.doi.org/10.1002/cctc.201601386
A Journal of
www.chemcatchem.org

10.1002/cctc.201601386
ChemCatChem
FULL PAPER
Catalyst@metal hybrids in a one-pot multistep opposing
oxidation and reduction reactions-sequence
Leora Shapiro and David Avnir*
Abstract: We report the feasibility of catalysis of contrasting
reactions - oxidation and reductions – in a multistep process,
involving also a condensation reaction, all carried out with hybrids of
catalytic organometallic complexes entrapped within metals
([complex]@metal). Two routes were explored for that multistep
process: The first route used a combination of an oxidizing iridium
complex entrapped in silver, [Ir]@Ag, and a reducing rhodium
complex entrapped in silver [Rh]@Ag. The second route used a
single composite material, namely [Ir]@Pd, where [Ir] acted as the
oxidative catalyst, and the Pd as the reducing catalyst. Both routes
proved to be efficient for heterogeneous catalytic oxidation and
reduction reactions sequence. Towards the goal of a one-pot
multistep oxidation and reductions process, a detailed study of the
catalytic alcohols oxidation properties of the heterogeneous [Ir]@Ag
and [Ir]@Pd was carried out. As the arsenal of available
hydrogenations catalysts exceeds by far that of oxidations, this part
of the study is, we believe, of general interest in itself, as a new
alternative to heterogeneous oxidative catalysis.
hydrogenations of styrene and diphenyl acetylene[4]; a Co(II)-
(salen)@Ag for electrochemical asymmetric carboxylation[5];
catalytic polymers, for example the superacid Nafion@Ag used
for a pinacol-pinacolone rearrangement[6]; inorganic acids such
as H3[P(Mo3O10)4]@Ag and H3[P(W3O10)4]@Ag, used for
Friedel–Crafts (FC) adamantylation of toluene and anisole[7];
alkaloids, as in the case of (-)-cinchonidine@Pd or (-)-
cinchonine@Ag for the partial enantioselective hydrogenation of
ketones[5,8]
; and even biological catalysts, such as acid
phosphatase@Au for the hydrolysis of p-nitrophenyl phosphate[9].
Of particular interest is the third pathway, namely the entrapping
of a catalytic entity within a catalytic metal. Work in which the
entrapped catalyst was assisted by the cage for performing one
reaction include: ReOD@Ag (ReOD= ReO-(DPBS)2Cl3;
DPBS=3-(diphenylphosphino)benzenesulfonic acid sodium salt),
which was used for the epoxidation of alkens with the aid of
persulfate[10]; and pyridine entrapped within palladium, which
was pressed into a coin shape and used electrochemically for
the reduction of CO2 to methanol[11]
.
Alternatively, an
independent separate individual reaction may be catalyzed by
each component, as displayed with Nafion@Pd[12], in which the
Nafion catalyzed a dehydrogenation step and the palladium
catalyzed hydrogenation or disproportion on the same moiety.
Finally, a set of parallel reactions can be performed, the
products of which can then be combined to a single product, as
was shown by the dual catalytic activity of hydroformylation and
Introduction
Entrapping molecules within metals is a novel addition to the
arsenal of methods for preparation of new heterogeneous
catalysts, denoted molecular-dopant@metal[1]. Three main
routes for catalysis may be followed with these new metal
composites: First, the entrapping metal itself is a catalyst, the
properties of which are modified by the dopant[2]. For example,
silver doped with entrapped Congo Red has demonstrated
superior catalysis parameters of the oxidation of methanol to
formaldehyde, compared to pure silver[3]. The improvements
were attributed largely to the reduction in the size of the
elementary metal crystallites and to the maintenance of that size
during reaction. A second pathway utilizes the metal as a
heterogenizing matrix of molecular catalysts, much like the well-
known immobilization in ceramic or polymeric matrixes. This
pathway is possible, because the molecules of the catalyst are
embedded within cages made of aggregated metallic crystals,
and are accessible for reaction. It was found that the metallic
heterogenizing matrix provides efficient protection of the dopant,
and may alter also the distribution of products. Accounts of
catalytic activity performed by the entrapped species include a
wide variety of entrapped species: Organo-metallic complexes,
such as [RhCl(COD)TPPS]@Ag (COD = 1,5 cyclooctadiene,
hydrogenation with [RhCl(TPPS)3]@Pd [13]
.
While catalysis of hydrogenations, hydrogen shifts and other
reducing processes is well developed, oxidation catalysis offers
a smaller library of catalytic processes, and is considered, by
and large, more difficult to perform[14]. In the context of doped
metals, methanol oxidation and epoxidations by doped silver
were mentioned above. As a major theme of our work are
cascade reactions which do not require separation between
steps; and as the specific challenge we faced in this report has
been to have in one series both catalytic oxidation and reduction
steps – classical opposing processes - suitable doped metal
oxidation catalysts had to be developed. Towards this goal we
report here the entrapment of an oxidative dibenzobarrelene-
based PCSP3P iridium pincer catalyst[15] (denoted [Ir], see figure
1), in either silver or in palladium, resulting in [Ir]@Ag and
[Ir]@Pd. The mechanism of These two composite catalysts were
first studied on the oxidative dehydrogenation of alcohols and
once the oxidation activity was established, it followed into a
multi-stepped oxidation and reduction tandem reaction
sequence, where the hydrogenation of nitrobenzene to aniline
and of an imine to an amine, were performed. The latter was
affected by either applying [Rh]@Ag[4] (see figure 1 for [Rh]) or
by the Pd component of [Ir]@Pd. In the case of doped silver,
each complex was entrapped separately, so that parallel
opposing reactions can take place. We recall that silver is
usually an inert metal at the temperature range employed in this
study; it may catalyze oxidation using oxygen, but at very high
temperatures of at least 200 C with a yield which increases with
TPPS
=
{Ph2P(C6H4SO3Na)}, denoted [Rh]@Ag) for
[a]
Institute of Chemistry and the Center for Nanoscience and
Nanotechnology, the Hebrew University of Jerusalem, Jerusalem
91904, Israel
E-mail: david.avnir@mail.huji.ac.il
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601386
ChemCatChem
FULL PAPER
the temperature. For the second path, both functions – oxidation
and reduction – are catalyzed by the same hybrid material: The
oxidizing iridium complex was entrapped within palladium, which
itself is an excellent catalyst for the hydrogenation process.
We begin with the SEM and the wavelength-dispersive X-
ray spectroscopy (WDS) elemental analysis of the three
catalysts (Ir@Ag in Figure 2, Ir@Pd in Figure 3, and Rh@Ag in
Figure 4). We recall that in WDS, the detector counts only one
wavelength at a time, as opposed to EDX which produces a full
spectrum. This allows calibration of the probe for each element
separately, resulting in an extremely high degree of sensitivity
with a small error in the evaluation, and it is therefore useful
even for the lighter elements and for differentiation between
adjacent elements. Considering a 100 nm diameter beam, the
excitation volume is approximately 1 μm3. In the images, the
relative weight of each element is calculated from the raw counts
in each point of the sample and compared to the element’s
standard in the detector. Sections (a) in Figures 2-4 shows the
SEM images of the WDS-analyzed areas (see Fig. S1,
Supplementary Material for HR-SEM images; sections (b)
represent the WDS of central atom in the complex (Ir in Figures
2 and 3, Rh in Figure 4), entrapped within the metal; sections (c)
represent the surrounding metal, either silver or palladium; and
sections (d) show the ratio between the central metal atom of
the entrapped complex and the surrounding metal in the
samples. The micrographs were sampled in multiple spots in the
depicted area. The lettered points in the figures indicate where
measurement of the level of each metal atom in the composite
was actually determined. Each point marked by a letter is an
average measurement of nine points closely surrounding the
proximity of the marker. All of the spots, even those which
showed elevated levels of iridium or rhodium, showed the
presence of the metal in which the complex was caged.
Figure 1. The iridium pincer catalyst and the rhodium complex
catalyst used in this report.
Results and Discussion
Characterization of the catalytic materials:
The catalytic materials, [Ir]@Ag, [Ir]@Pd, were prepared by
reducing the particular metal cation in the presence of the
complex to be entrapped, modifying previously reported
procedures[4, 8, 13, 16], and [Rh]@Ag was prepared as described in
Ref 4. The catalytic activities of these new composites are
dictated not only by the specific catalytic dopants and by the
catalytic properties of Pd, but also by materials properties.
These include the metal crystallite size, a known parameter
which affects catalysis; the porosity which form by the
aggregation and which dictates the kinetics and the diffusional
aspects of these composites; the homogeneity of the
entrapment distribution and the resulting chemisorption
interactions between the organometallic catalyst which affect
each other’s electronic structure; and the protectability of the
metallic cage; the materials aspects are thus described next.
Figure 3: WDS mapping of [Ir]@Pd. Bar = 20 μm: (a) scanning
electron microscopy image.; (b) iridium; (c) palladium; (d) the Ir:Pd
ratio.
Figure 2. Wavelength-dispersive X-ray spectroscopy (WDS)
mapping of [Ir]@Ag. Bar = 20 μm: (a) scanning electron microscopy
image; (b) iridium; (c) silver; (d) the Ir:Ag ratio.
[Ir]@Pd
a
b
[Ir]@Ag
b
d
a
d
c
c
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601386
ChemCatChem
FULL PAPER
Figure 4. WDS mapping of [Rh]@Ag. Bar = 20 μm: (a) scanning
electron microscopy image; (b) represents rhodium; (c) represents
silver; (d) represents the ratio between Rh:Ag.
the desired reaction. This porosity also enables the reaction
products to leave the entrapped catalyst back into the reaction
solution.
Figure 5. XRD spectra of [Ir]@Ag powder (top), [Ir]@Pd (center) and
[Rh]@Ag powder (bottom).
a
b
[Rh]@Ag
d
c
Let us focus, for instance, on Figure 2, which collects the
results for [Ir]@Ag: The SEM image (Figure 2a) shows several
lumps which all contain silver and the iridium complex. Figure 2b
is the WDS of the iridium, while Figure 2c focuses on the
entrapping silver. Figure 2d shows the ratio between (b) and (c)
- note the different color-code sensitivities in b, c and d. The fact
that the ratio (Figure 2d) is relatively homogenous throughout
the sample is evident in the fact that the image is relatively dark,
showing few spots in which the iridium stands out with a high
ratio compared to the silver. It should be noted that the silver
content in the sample is much higher than the iridium and so
there are areas in which there is silver but no iridium complex.
In figure 3, it is visible that the iridium complex is scattered,
and spread out throughout the sample, matching the spreading
of palladium. In this sample, the powder seen is finer, covering
the surface more fully (Fig. 3a). This variance in image can be
explained by the much smaller elemental crystal size of the
palladium matrix compared to that of the silver matrices
explained later on (see XRD section). It appears that both the
palladium and the iridium are scattered throughout the surface
(Fig. 3b,c). The ratio between the iridium and palladium also
shows the intertwined relationship of the two. Two points on the
image indicate an above the average Ir:Pd ratio, A and D, in
which, still, the palladium is well present. Larger magnification
images using HR-SEM (Fig. 1S, Supplementary) show that the
samples are formed of small metallic nanocrystals, ranging in
size between 5 and 25 nm, aggregated into microns-scale
particles - here too the smaller crystal size of palladium
compared to the silver is visible (see also XRD analysis below).
These aggregates accommodate interstitial pores through which
the substrate can reach the entrapped complex which catalyzes
Figure 6. TGA of [Ir] and [Ir]@Ag (top) and first derivative (under air)
and [Ir] and [Ir]@Pd (middle) and first derivative (under N2). For
comparison, [Rh]@Ag TGA and first derivative are presented on the
bottom.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601386
ChemCatChem
FULL PAPER
100
90
80
70
60
50
40
0.0000
-0.0005
-0.0010
-0.0015
-0.0020
-0.0025
-0.0030
0.0002
0.0000
-0.0002
-0.0004
-0.0006
-0.0008
-0.0010
[Ir]@Ag
pure [Ir]
[Ir]@Ag
pure [Ir]
30
20
100
100
100
200
300
400
500
600
600
600
700
800
100
200
300
400
500
600
700
800
Ts C
Ts / C
100
90
80
70
60
50
0.0000
0.0000
-0.0001
-0.0002
-0.0003
-0.0004
-0.0005
-0.0006
-0.0002
-0.0004
-0.0006
-0.0008
-0.0010
-0.0012
-0.0014
[Ir]@Pd
pure [Ir]
[Ir]@Pd
pure [Ir]
100
200
300
400
500
600
700
800
200
300
400
500
700
800
Ts / C
Ts / C
100
90
80
70
60
50
40
30
0.001
0.0000
-0.0002
-0.0004
-0.0006
0.000
-0.001
-0.002
-0.003
-0.004
-0.005
-0.006
[Rh]@Ag
pure [Rh]
[Rh]@Ag
pure [Rh]
100
200
300
400
500
600
700
800
200
300
400
500
700
800
Ts / C
Ts / C
Finally, Fig. 4 represents the WDS map of [Rh]@Ag. The
aggregated nature of [Rh]@Ag is similar to [Ir]@Ag. There are
two spots on the image (A and B in Fig. 4b) in which there
seems to be a higher concentration of [Rh] compared to the rest
of the sample. Yet, the ratio is smaller than with [Ir]@Ag or
[Ir]@Pd, reflecting the fact that the synthetic ratio [Rh]:Ag
needed for the catalysis is smaller than [Ir]:Ag and [Ir]@Pd. In
the areas of high rhodium concentration silver is present as well
and the spots seem blended in on the image of the ratio (Fig.
4d). In prior work, this composite was shown to have the
complex entrapped within the silver through a phenomenon of
‘oozing out’ of the complex from the matrix due to heating by the
SEM electrons[4]. The specific data results of these spots
measurements are collected in Table S1, Supplementary
Material.
Further structural information is provided by x-ray diffraction
(XRD) analysis (Fig. 5). These measurements reveal that the
metal face-centered cubic crystalline structures of the silver and
the palladium remained intact upon entrapment[17]
.
The
mechanism of entrapment within metals, as detailed before,
involves the aggregation of the metallic elementary nanocrystals
around the entrapped species, therefore leaving the metal in its
natural crystalline form. According to the Scherrer equation
applied to the diffractograms, the elementary crystal size of the
palladium in [Ir]@Pd is 5.3 nm, silver in [Ir]@Ag is 23.6 nm, and
18.5 nm in [Rh]@Ag. These values are in agreement with
previous reports[13,16]. The smaller particle size of the palladium
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601386
ChemCatChem
FULL PAPER
compared to the silver is probably due to the fact that since its
reduction potential is more positive than that of silver, more
nucleation sites form in the initial stage of the metal reduction.
Characterization of the matrix utilizing thermogravimetric
analysis (TGA) focused by its nature on the organic components
of the hybrid matrices (Fig. 6). While silver does not oxidize in
the presence of oxygen upon heating, porous palladium reacts
at temperatures over 350 °C to form palladium oxide, PdO.
Therefore, while the silver containing samples were measured
under standard air conditions, samples with palladium were
measured under nitrogen atmosphere. For comparison purposes
the pure iridium complex was therefore analyzed both under air
(for comparison with [Ir]@Ag) and under nitrogen (for
comparison with [Ir]@Pd). The analysis of [Ir] under air shows
degradation of roughly 70 % of the weight which occurs
gradually from 200 °C and ends at 500 °C. In addition, there is a
minor weight loss peak above 500 °C, possibly correlated to the
burning of the phosphines in the complex. The weight loss of the
pure complex during the evaluation under air is concurrent to
calculation of the ratio between the iridium atom and the pincer
ligand that shows that nearly 30 % of the weight is contributed
by the iridium. The degradation profile of the entrapped species
is similar to that of the free complex, showing a steady loss of
mass from 200 up to 500 °C and minor weight loss before and
after those temperatures. The first derivative profiles are similar
in shape, but emphasize temperature shifts in the entrapped
case to lower values; such shifts have already been recorded
several times for doped metals[18]. The total weight loss of
[Ir]@Ag between 200-800 °C, i.e. after release of some well
entrapped residue solvent (water and methanol) molecules, was
16%, which is attributed to loss of the ligands of the complex.
Based on the weight loss of the free complex, calculation shows
that the entrapped complex consists of ~23 wt% of the
composite. The degradation profile of [Ir]@Pd shows that most
of the weight loss occurred between 200-500 °C, although the
measurement was done under nitrogen, therefore represents
mostly pyrolysis and not burning. In this case, the weight loss
started as early as 100 °C, due to release of locked-in water and
methanol. The weight loss in the free complex under nitrogen
was 46%. In the [Ir]@Pd composite, the weight loss after 200 °C
is roughly 8 %. This translates to roughly 22 wt% of the [Ir] in the
composite (a molar ratio of ~1:22 [Ir]:Pd). Again, the derivative
presentation emphasized the shift to lower temperatures,
indicating the catalytic contribution of the metal to the
decomposition. For comparison, the TGA behavior of [Rh] and
[Rh]@Ag were also tracked in this study. In brief, the TGA
analysis of [Rh] and [Rh]@Ag showed most of the mass loss up
to 550 °C. The [Rh] TGA first derivative showed a small
degradation peak near 200 °C, which may be attributed to
locked-in solvent. Another peak is visible at ~450 °C as well as a
major peak just above 500 °C. These are attributed to the loss of
the organic ligands which include solfunated triphenylphosphine
and cyclooctadiene. The profile of [Rh]@Ag indicates the same
peaks although the two later peaks are far smaller than in the
free complex.
The catalysis:
Towards our next goal - a multiple step, oxidation-reduction
process - there has been the interest in investigating the
oxidation process by itself, because oxidation catalysts under
mild conditions are by far less abundant compared to
hydrogenation catalysts. Thus, the catalytic tunable activity of
the composites was carried out first for the oxidation of alcohols,
namely the dehydrogenation of three secondary alcohols – 2-
octanol, 1-phenyl-2-propanol and dibenzosuberol - to ketones
(Scheme 1), and the oxidation of a primary alcohol, benzyl
alcohol, to benzaldehyde and further to benzylbenzoate
(Scheme 2). Two catalysts were tested (having in mind in their
construction the next goal of a multistep oxidation-reduction
process – see below): [Ir]@Ag and [Ir]@Pd. For the oxidation
step, both metals serve as heterogenizing matrices (although
each can serve also for catalysis in other processes). We note
that standard heterogenizing matrices in catalysis have
traditionally been ceramic supports and polymers; the use of a
metal for that purpose is only in its infancy[4-5, 7, 9a, 11, 13, 16, 18a, 19]
.
Scheme 1. The studied catalytic oxidations of secondary alcohols to
ketones.
Scheme 2. The catalytic oxidation of the primary benzyl alcohol to
benzaldehyde and to benzyl benzoate.
O
OH
([Ir]@Ag + [Rh]@Ag)
or [Ir]@Pd
([Ir]@Ag + [Rh]@Ag)
or [Ir]@Pd
([Ir]@Ag + [Rh]@Ag)
H2
NO2
NH2
or [Ir]@Pd
N
NH
H2
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601386
ChemCatChem
FULL PAPER
tested the possibility that the hydrogen released in the oxidation
step may reduce nitrobenzene. Thus, [Ir]@Pd and also the
combination [Ir]@Ag and [Rh]@Ag were heated with p-xylene
containing both the alcohol and nitrobenzene - no aniline formed
unless external hydrogen was added into the vessel. The
reductions and the condensation between the initial products are
quantitative, and the over-all yield for the full sequence is
dictated by the oxidation step – see 2-ocatnol, Table 2 - that is,
an almost full conversion to the final product in the case of
[Ir]@Pd (traces of azobenzene and of the 16 carbon β-
hydroxyketone aldol condensation product were detected as
well), and a somewhat smaller yield for the Ag catalysts. In the
case of [Ir]@Pd the condensation reaction took place
spontaneously, but in the case of silver this condensation
needed the addition of a small amount of hydrochloric acid after
the oxidation and reduction steps. Note that the sequence in
Scheme 3 requires secondary alcohols: The condensation
reaction of the intermediate aldehyde in the case of the primary
alcohol (Scheme 2) with aniline is slower than its conversion to
the ester, and the ester formation is irreversible: When benzyl
alcohol was oxidized in the presence of aniline, the ester still
formed as the major product.
The evaluation of the recyclability focused on the more sensitive
and difficult oxidation step (of dibenzosuberol). The test was
measured for three repetitions, each with fresh components
except for the catalyst which was centrifuged and placed back
into the reaction vessel. As seen in Table 3, the recyclability of
[Ir]@Pd is better than that of [Ir]@Ag for the first two cycles. The
catalysts decay however in the third cycle. In addition, the
reaction mixture was tested after reaction for residual activity: In
both cases of [Ir]@Ag and [Ir}@Pd, fresh dibenzosuberol was
added to the used reaction mixture after filtration, and no
reaction took place. This shows that the complex did not leach
out into the solution. It is therefore assumed that the loss of yield
in the 3rd cycle is due to destruction of the complex structure,
and not leaching. The interesting observation is that of the two
optional routes of the multistep process, the one in which both
the matrix and the dopant are catalytic, is a better choice.
Scheme 3. Four steps oxidation-reduction reaction sequence carried
out with either [Rh]@Ag and [Ir]@Ag or with [Ir]@Pd. The four steps
are (a) oxidation (top left), (b) reduction (bottom left), (c)
condensation (center) and (d) second reduction (right).
Table 1 collects the results for the oxidation of the
secondary alcohols. It is seen that both catalysts are effective,
with a better performance of [Ir]@Pd and higher yields for 2-
octanol and dibenzosuberol. The superior activity of [Ir]@Pd
may be attributed to the fact that palladium can reversibly store
hydrogen on its surface: as the iridium catalyst removes the
hydrogen molecule from the alcohol, the palladium can retain
the H2 molecule, therefore pulling the reaction to completion.
The primary alcohol, benzyl alcohol, tested for the
oxidation, resulted mainly in the ester (Scheme 2, Table 2). This
result agrees with previous reports on the activities of the free
catalyst[15] and the sol-gel entrapped catalyst[20]
.
This
esterification occurs via a Tischenko reaction between the
formed aldehyde and the starting alcohol, resulting in a
hemiacetal that is further oxidized to form the ester.
Table 1. Oxidation yields of the secondary alcohols (in %).
2-octanol
1-phenyl-2-
propanol
72
dibenzosuberol
catalyst
Ir@Pd
Ir@Ag
95
73
97
84
48
Table 2. Oxidation yields of benzyl alcohol (in % at 90% conversion).
benzyl
benzoate
62
benzaldehyde
catalyst
Ir@Pd
27
12
Ir@Ag
free Ir
78
complex
73
15
Table 3. Yields (%) in the recycling experiment involving dibenzosuberol
The multi-step oxidation-reduction process:
Catalyst
Cycle 1
[Ir]@Pd
97
[Ir]@Ag
84
Two approaches were tested regarding the design of the multi-
step oxidation-reduction process (Scheme 3): First, using a
combination of [Ir]@Ag – the oxidizing catalyst – and [Rh]@Ag –
the reducing catalyst - where each of the catalytic complexes
was separately entrapped within silver, which serves, in this
case, as a heterogenizing matrix. The second approach was to
use a single composite catalyst, [Ir]@Pd, which is constructed
from the oxidative catalyst entrapped within a reducing metal
catalyst. The oxidation step selected for this process is the
conversion of 2-octanol to the ketone, described above. The
reducing steps selected are the reduction of nitrobenzene to
aniline and the reduction of the imine, formed in the first reaction,
to an amine (Scheme 3). While the reduction of nitrobenzene
using palladium is a relatively straight forward reaction, [Rh]@Ag
was previously used for the reduction of unsaturated carbon-
carbon bonds[4]. In our hands, we report that this catalyst also
proved efficient for the two reductions with full conversion. We
Cycle 2
Cycle 3
94
36
29
8
Conclusion
To conclude, both the silver and palladium metal frames provide
stable heterogeneous environment for the iridium
a
organometallic oxidizing complex entrapped within. The
combinations of an oxidant iridium complex entrapped in silver
or in palladium, and a reductant rhodium complex entrapped in
silver were synthesized and catalytically investigated in a
multiple steps process of opposing oxidations and reductions. It
is our belief that many types of catalysts may benefit from this
method of heterogenization for synergism in reaction sequences.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601386
ChemCatChem
FULL PAPER
molecular sieves, 4 Å, to absorb water released during the condensation
reaction. It was then loaded with 100 psi H2, where it was held at 90 °C
for 24 hours, after which the H2 was released. Note that if the order of
steps is reversed, then the reaction is a three-step sequence, stopping at
the imine stage. The reversed order is the following: The catalysts along
with nitrobenzene and 2-octanol were first placed in a pressure vessel
under hydrogen pressure with p-xylene at 90 °C for 24 hours, and then
transferred to a round bottom flask and refluxed for an additional 24
hours with molecular sieves. Then HCl was added and allowed an
additional 24 hours to reach completion. The reaction components and
products, in all cases, were identified by NMR spectroscopy and gas
chromatography coupled to mass chromatography. NMR was compared,
Experimental Details
Chemicals: Dibenzosuberol and dibenzosuberone were purchased from
Alfa Aeser. AgNO3, zinc (20 mesh, ~840 μm) and p-xylene were
purchased from Sigma Aldrich. 2-octanol, 1-phenyl-2-propanol and
benzyl alcohol used in this work and their derivative ketones or
aldehydes, aniline, nitrobenzene, Cs2CO3 and PdCl2 were all purchased
from Acros Organics. Triply distilled water was obtained by a UV fitted
Millipore water purification system with 18.2 MΩ resistance. The [Ir]
catalyst was kindly synthesized and donated by Sanaa Musa, Orit Cohen
and Prof. D. Gelman (the Institute of Chemistry at the Hebrew University
of Jerusalem). The synthesis is described in Musa et al., 2011[15]
RhCl(COD)TPPS was synthesized as described in ref. [4] from chloro(1,5-
cyclooctadiene)rhodium(I) dimer and 3-(diphenylphosphino)-
.
when possible, to known spectra, or an NMR prediction tool was used[21]
.
Testing the catalyst recovery and leaching: 0.08 g of the dried desired
catalyst ([Ir]@Pd or [Ir]@Ag) was placed in a round-bottom flask along
with 0.25 g dibenzosuberol (1.110-3 mol), 0.1 g Cs2CO3 (used as a
catalytic base for the oxidation reaction with the iridium catalyst) and 4
mL p-xylene. The reaction mixture was refluxed for 24 hours after which
the supernatant was poured off, centrifuged at 50,000 rpm for 20 min and
evaporated for measurement. The remaining catalyst was collected back
into the reaction vessel for repeated use and was recycled three times.
For testing the product supernatant for oxidation activity, the liquid was
filtered through a polytetrafluoroethylene syringe filter (0.2 μm) before
reuse.
bezenesulfonic acid sodium salt (TPPS), both purchased from Strem
chemicals.
Synthesis of [Ir]@Ag and [Rh]@Ag: 60 mg (0.0694 mmol) of the iridium
complex, [Ir], was dispersed with 10 mL of methanol under argon for
several minutes, after which 0.24 g (1.4 mmol) of AgNO3 was added.
Then 15 mL of triple distilled water was added, followed by 0.065 g (1.0
mmol) of zinc granules, considering that each zinc atom supplies two
electrons. The slurry was stirred for 24 hours at room temperature, after
which it was filtered through a mixed cellulose ester filter (pore size 1.2
μm) and washed with triple distilled water three times (20 mL). The
powder was then placed in a desiccator for at least 24 hours prior to use.
Before initial use the catalyst was hand milled in an agate mortar and
pestle for several minutes until it appeared homogeneous. TGA analysis
indicates an [Ir]/Ag molar ratio of 1:24 in the final product, [Ir]@Ag.
Instrumentation: For electron probe microanalysis (EPMA), back-
scattered electron (BSE) and secondary electron (SE), images were
produced using a JEOL 8230 superprobe with EDS and four wavelength-
dispersive spectrometers (WDS) for microanalysis. Beam conditions
were set to voltage and current of 15 keV and 15 nA, respectively. All
phases were analyzed for Pd and Ag using silicate and oxide standards
(SPI 53 minerals). Data was processed with a ZAP correction procedure.
For SEM measurements, a MagellanT XHR SEM microscope was used,
operating at 5 kV, fitted with an Oxford X-Max EDS free detector.
Samples were prepared by placing the air-dried powder on an aluminum
stub suitable for SEM using carbon double sided tape.
Thermogravimetric analysis was performed by using a Mettler–Toledo
TGA/SDTA 851e from 50 to 800 °C at a heating rate of 10 °Cmin-1 by
flowing N2 or air. Gas chromatographs were measured by using an
Agilent 7890A GC, fitted with an HP-5 column. Solution analysis was
performed by introducing sample solutions by direct splitless injection (1
mL by volume). For the chromatography procedure, the oven
temperature was elevated to 80 °C and held for 2 min. Then it was
elevated 250 °C at 15 °C∙min-1, and held again for 3 min. Helium carrier
gas flow was 3 mL∙min-1. GC-MS (Hewlett Packard, GC 5890 Series II,
MS 5972 Series) was fitted with an Agilent HP-5 column. X-ray diffraction
(XRD) measurements were performed with a Philips automated powder
diffractometer (with a PW1830 generator, PW1710 control unit, PW1820
vertical goniometer, 40 kV, 30 mA, CuKα1 (1.5406 Å), stepscan mode
0.02 s-1). NMR spectra were recorded with Bruker DRX-400 in CDCl3.
[Rh]@Ag, (molar ratio of 1:139) was synthesized according to
procedure described in detail by Yosef et al[4]
a
.
Synthesis of [Ir]@Pd: 60 mg (0.0694 mmol) of [Ir] was dispersed with 10
mL of methanol under argon for several minutes, after which 0.25 g (1.4
mmol) PdCl2 was added. Then 15 mL of triple distilled water were added,
followed by 0.098 g (1.5 mmol) of zinc granules. The slurry was stirred
for 24 hours at room temperature after which it was filtered through a
mixed cellulose ester filter (pore size 1.2 μm) and was washed with triple
distilled water three times (20 mL). The powder was then placed in a
desiccator for at least 24 hours prior to use. Before initial use the catalyst
was hand milled in an agate mortar and pestle for several minutes until it
appeared homogeneous. TGA analysis indicates an [Ir]/Pd molar ratio of
1:28 in the final product, [Ir]@Pd.
The catalytic alcohol oxidation reactions: 0.1 gr of the dried catalyst,
[Ir]@Ag or [Ir]@Pd was placed in a round bottom flask to which 0.1 mL
(approximately 10-3 mol) of the alcohol and p-xylene (4 mL) was added
along with 0.1 g Cs2CO3 (used as a catalytic base for the oxidation
reaction with the iridium catalyst). The mixture was refluxed for 24 hours.
The catalytic reduction reaction: 0.1 of the dried catalyst, [Rh]@Ag or
[Ir]@Pd was placed in a pressure vessel. 0.1 mL of nitrobenzene
(9.8X10-4 mol) and p-xylene (4 mL) were added and then it was loaded
with 100 psi H2. The vessel was held at 90 °C for 24 hours, after which
the H2 was released.
Acknowledgements
We thank Prof. Matthias Driess for useful discussions. Special
thanks are due to Sanaa Musa, Orit Cohen and Prof. D. Gelman
(the Institute of Chemistry at the Hebrew University of
Jerusalem) for donating the iridium complex. Supported by the
Einstein Foundation, Berlin, by the Israel Science Foundation,
grant 703/12 and by the Deutsche Forschungsgemeinschaft
(DFG) through grant SCHO687/8-2.
The multi-step catalytic reactions: 0.1 mL nitrobenzene (9.8X10-4 mol)
and of 2-octanol (6.3X10-4 mol) were placed in a round bottom flask with
p-xylene (4 mL). Into the flask the dried catalyst (either 0.1 g [Ir]@Pd, or
0.1 g [Ir]@Ag and 0.1 g [Rh]@Ag) was added along with 0.1 g Cs2CO3 (a
co-catalyst for the oxidation reaction with the iridium catalyst). The
mixture was refluxed for 24 hours and the whole mixture transferred to a
pressure vessel along with
5 drops of HCl and several granules
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601386
ChemCatChem
FULL PAPER
Keywords: heterogeneous catalysis • mesoporous materials •
multiple catalysis • organic–inorganic hybrid composites •
supported catalysts•
[1]
[2]
D. Avnir, Accounts of Chemical Research 2014, 47, 579-592.
L. Chen, I. Polishchuk, E. Weber, A. N. Fitch, B. Pokroy, Crystal Growth
& Design 2016, 16, 2972-2978.
[3]
a) H. Behar-Levy, D. Avnir, Chemistry of Materials 2002, 14, 1736-
1741; b) Y. A. Yohann Aouat, Gad Marom, and David Avnir, Science of
Advanced Materials 2013, 5, 598-605; c) G. M. Y. Aouat, D. Avnir, V.
Gelman, G. E. Shter, and G. S. Grader, The Journal of Physical
Chemistry C 2013, 117, 22325-22330.
[4]
[5]
[6]
[7]
[8]
[9]
I. Yosef, R. Abu-Reziq, D. Avnir, Journal of the American Chemical
Society 2008, 130, 11880-11882.
H. Yang, D. Chi, Q. Sun, W. Sun, H. Wang, J. Lu, Chemical
Communications 2014, 50, 8868-8870.
H. Behar-Levy, D. Avnir, Advanced Functional Materials 2005, 15,
1141-1146.
S. Krackl, A. Company, Y. Aksu, D. Avnir, M. Driess, Chemcatchem
2011, 3, 227-232.
L. D. Pachon, I. Yosef, T. Z. Markus, R. Naaman, D. Avnir, G.
Rothenberg, Nature Chemistry 2009, 1, 160-164.
a) R. Ben-Knaz, D. Avnir, Biomaterials 2009, 30, 1263-1267; b) R.
Ben-Knaz, R. Pedahzur, D. Avnir, Rsc Advances 2013, 3, 8009-8015.
[10] A. Indra, M. Greiner, A. K. Gericke, R. Schlçgl, D. Avnir, M. Driess,
Chemcatchem 2014, 6, 1935–1939.
[11] H.-P. Yang, S. Qin, H. Wang, J.-X. Lu, Green Chemistry 2015, 17,
5144-5148.
[12] D. A. Leora Shapiro, Chemcatchem 2013, 5, 2195-2198.
[13] L. Shapiro, M. Driess, D. Avnir, Chemcatchem 2015, 7, 2033-2037.
[14]
a) R. Ciriminna, M. Pagliaro, Organic Process Research &
Development 2010, 14, 245-251; b) J. Hartung, M. Greb, Journal of
Organometallic Chemistry 2002, 661, 67-84; c) B. L. Ryland, S. S.
Stahl, Angewandte Chemie-International Edition 2014, 53, 8824-8838;
d) G. E. Dobereiner, R. H. Crabtree, Chemical Reviews 2010, 110, 681-
703.
[15] S. Musa, I. Shaposhnikov, S. Cohen, D. Gelman, Angewandte Chemie
International Edition 2011, 50, 3533-3537.
[16] L. Shapiro, D. Avnir, ChemCatChem 2013, 5, 2195-2198.
[17]
[18]
a) ICDD – Card Number 00–004–0783 for Silver, ; b) ICDD – Card
Number 00–005–0681 for Palladium,
a) H. Behar-Levy, D. Avnir, Advanced Functional Materials 2005, 15,
1141-1146; b) I. Yosef, D. Avnir, Chemistry of Materials 2006, 18,
5890-5896; c) H. Behar-Levy, G. E. Shter, G. S. Grader, D. Avnir,
Chemistry of Materials 2004, 16, 3197-3202; d) G. Nesher, G. Marom,
D. Avnir, Chemistry of Materials 2008, 20, 4425-4432.
[19] A. Indra, M. Greiner, A. K. Gericke, R. Schlögl, D. Avnir, M. Driess,
Chemcatchem 2014, 6, 1935-1939.
[20] K. Oded, S. Musa, D. Gelman, J. Blum, Catalysis Communications
2012, 20, 68-70.
[21] L.
Patiny,
J.
Wist,
NMR
Predict,
2015,
http://www.nmrdb.org/new_predictor/index.shtml?v=v2.15.1.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601386
ChemCatChem
FULL PAPER
Entry for the Table of Contents (Please choose one layout)
Layout 1:
FULL PAPER
Text for Table of Contents
Leora Shapiro and David Avnir*
Page No. – Page No.
Catalyst@metal hybrids in a one-pot
multistep opposing oxidation and
reduction reactions-sequence
((Insert TOC Graphic here: max.
width: 5.5 cm; max. height: 5.0 cm))
Layout 2:
FULL PAPER
Author(s), Corresponding Author(s)*
((Insert TOC Graphic here; max. width: 11.5 cm; max. height: 2.5 cm))
Page No. – Page No.
Title
Text for Table of Contents
This article is protected by copyright. All rights reserved.