A recoverable Pd nanocatalyst for selective semi-hydrogenation of alkynes: Hydrogenation of benzyl-propargylamines as a challenging model

Article abstract of DOI:10.1039/c4gc00669k

We describe a recyclable heterogeneous palladium nanocatalyst for the selective hydrogenation of alkynes to alkenes. The catalyst was prepared through the decomposition of the organometallic precursor Pd₂(dba)₃ over a magnetic support, obtaining well-dispersed Pd nanoparticles that formed exclusively on the support surface, with average diameter of 3.5 ± 0.8 nm. The catalytic activity was investigated in the hydrogenation reactions of alkenes and alkynes, and the chemo- and stereoselectivity were evaluated in the hydrogenation of benzyl-propargylamines. The catalyst is highly selective in performing semi-hydrogenation reactions under mild conditions and short reaction times, with good overall yields. Furthermore, it can be easily recovered and recycled, with no leaching of palladium detected, and activities and selectivity retained over multiple reaction cycles. This journal is

Full text of DOI:10.1039/c4gc00669k

Green Chemistry
View Article Online
View Journal | View Issue
PAPER
A recoverable Pd nanocatalyst for selective
semi-hydrogenation of alkynes: hydrogenation of
benzyl-propargylamines as a challenging model†
Cite this: Green Chem., 2014, 16,
566
4
a
a
b,c
d
Paula M. Uberman, Natalia J. S. Costa, Karine Philippot, Rafaela C. Carmona,
d
a
Alcindo A. Dos Santos and Liane M. Rossi*
We describe a recyclable heterogeneous palladium nanocatalyst for the selective hydrogenation of
alkynes to alkenes. The catalyst was prepared through the decomposition of the organometallic precursor
2 3
Pd (dba) over a magnetic support, obtaining well-dispersed Pd nanoparticles that formed exclusively on
the support surface, with average diameter of 3.5 ± 0.8 nm. The catalytic activity was investigated in the
hydrogenation reactions of alkenes and alkynes, and the chemo- and stereoselectivity were evaluated in
the hydrogenation of benzyl-propargylamines. The catalyst is highly selective in performing semi-hydro-
genation reactions under mild conditions and short reaction times, with good overall yields. Furthermore,
it can be easily recovered and recycled, with no leaching of palladium detected, and activities and selecti-
vity retained over multiple reaction cycles.
Received 12th April 2014,
Accepted 20th July 2014
DOI: 10.1039/c4gc00669k
www.rsc.org/greenchem
be synthesized using several approaches, such as the addition
of aryl- and alkylacetylene derivatives to imines or enam-
Introduction
7
,8
9
,10
11
Hydrogenation reactions are considered one of the most valu- ines,
cross-dehydrogenative-coupling (CDC),
or the
1
2
able synthetic transformations in organic chemistry. In par- decarboxylation-coupling of secondary amino acids.
ticular, the selective semi-hydrogenation of alkynes to obtain addition to these methods, the A coupling reaction, a multi-
alkenes is a challenging task in organic synthesis, and it is an component reaction of an aldehyde, an amine, and an alkyne,
important reaction in the chemical industry. Selective palla- has received attention for its practicality. It is possible to
dium catalysts are usually promoted by a second metal, for obtain, in a single step, propargylamines substituted with
In
3
1
13
example the Lindlar catalyst comprised of Pd/CaCO
3
partially several functional groups with good yield. Propargylamines are
poisoned with lead. Catalyst deactivation and the presence of important intermediates in the synthesis of alkaloids, amino
1
4,15
toxic lead are the main drawbacks for the utilization of this acids, and nitrogen heterocycles.
Several steps of hydro-
catalyst system in true green processes. In addition, there is a genation could be necessary for the transformation of
growing demand for transformations of highly functionalized propargylamines into the target molecules, and usually only
7
,10,16,17
molecules to prepare complex molecules in fine chemical pro- the fully hydrogenated product is obtained.
When
duction. In this sense, the synthesis and modification of benzylamines are used, the hydrogenation process becomes
propargylamines have gained popularity in recent years due to challenging, as it could lead to a debenzylated product, leaving
1
0,18
their applications in the synthesis of natural products and the amino group unprotected.
Therefore, the development
2
–6
pharmaceutical compounds. Propargylamine derivatives can of a highly active and selective catalyst to perform alkyne
hydrogenation, as well as the more demanding hydrogenation
of propargylamines, is an area of current interest.
Pd-catalyzed reactions of pharmaceutically active ingredi-
ents may have problems associated with the retention of Pd
a
Laboratory of Nanomaterials & Catalysis, Institute of Chemistry, Universidade de
São Paulo, Av. Professor Lineu Prestes, 748, São Paulo 05508-000 SP, Brazil.
E-mail: lrossi@iq.usp.br
b
(and metal promoters) in the final product. There is a strict
CNRS, LCC (Laboratoire de Chimie de Coordination), 205, route de Narbonne,
1
9
limit for the levels of heavy metals in a drug substance. The
preparation of catalysts, especially lead-free Pd catalysts, in
which the leaching of metals can be minimized is highly rele-
vant to the development of green chemical processes, provid-
ing significant advantages in the safe production of organic
compounds. Functionalized magnetic nanoparticles (MNPs)
F-31077 Toulouse cedex 4, France
c
Université de Toulouse, UPS, INPT, LCC, F-31077 Toulouse, France
d
Laboratory of Organocatalysis and Organic Synthesis, Institute of Chemistry,
Universidade de São Paulo, Av. Professor Lineu Prestes, 748, São Paulo 05508-000
SP, Brazil
†
Electronic supplementary information (ESI) available: Characterization data
and NMR spectroscopy. See DOI: 10.1039/c4gc00669k
4566 | Green Chem., 2014, 16, 4566–4574
This journal is © The Royal Society of Chemistry 2014

View Article Online
Green Chemistry
Paper
2
0
have emerged as options to facilitate catalyst recovery.
to synthesize a magnetically recoverable Pd catalyst through
They offer the advantage of being magnetically separable, the direct decomposition of the organometallic precursor
thereby eliminating the requirement of catalyst filtra- [Pd (dba) ] with particular interest in selective hydrogenation
2
3
tion after completion of the reaction and avoiding the exposure of propargylamines.
of the catalyst to air during the recycling, which are
important and desired characteristics in the field of organic
synthesis.
Noble metal nanoparticles with high specific catalytic
Results and discussion
activity are ubiquitous in modern synthetic organic chem-
2
1
istry, and hydrogenation reactions are not an exception.
Recently, a variety of magnetically recoverable metal nanocata-
lysts have been studied in the hydrogenation of alkynes with
The catalyst preparation was performed through the direct
decomposition of the organometallic precursor [Pd (dba) ]
with dihydrogen in the presence of an amino-modified silica-
coated magnetite nanoparticle (FFSiNH ). The [Pd (dba) ]
2 2 3
precursor decomposes easily under dihydrogen atmosphere
leading to the production of Pd nanoparticles free of contami-
2
3
2
2–25
variable results.
activity but low selectivity to the alkene product.
et al. prepared Pt nanoparticles supported on MNPs modified
Fe-based nanocatalysts exhibit good
3
4
2
3,24
Alper
2
5
with ionic liquids (ILs), and they exhibited a very high
activity and selectivity for (Z)-alkene in the hydrogenation of
diphenylacetylene under harsh reaction conditions (90 °C,
2
8
nants after the purification step.
In such a way, this
methodology allowed us to obtain well-dispersed Pd nanoparti-
cles formed exclusively on the support surface, with average
diameter of 3.5 ± 0.8 nm (Fig. 1). The organometallic precursor
was totally decomposed over the magnetic support, and no
metal remained in the final solution (palladium < 0.02 ppm).
The Pd loading determined by the ICP OES was 3.6 wt%. A
representative transmission electron microscopy (TEM) image
of the obtained nanomaterial (Fig. 1) also revealed that the
core–shell nanostructured catalyst support was unchanged
after palladium deposition.
The catalytic behavior of the supported Pd nanocatalyst was
first tested using the well-established hydrogenation of cyclo-
hexene as a model substrate for the hydrogenation reactions.
Thus, cyclohexene was placed in a Fischer–Porter glass in the
presence of 0.005 mol% catalyst, 6 bar of dihydrogen, and
heated at 75 °C. The reaction was monitored by dihydrogen
consumption versus time. Under these conditions, the turnover
2
6
2
2
00 psi H , and 16 hours). Conversely, Scott et al. synthesized
Pd nanoparticles supported on MNPs stabilized by tetraalkyl-
phosphonium ILs, but they showed a very low selectivity for
the alkene in the hydrogenation of 3-hexyn-1-ol. Under the
best conditions, the alkene was obtained with a 41% yield,
whereas 59% of the hexanol was produced. Despite this, Pd
nanoparticles supported on magnetic carriers still remain
largely unexplored in selective hydrogenations. Moreover, most
of the catalytic studies were performed with simple or less
functionalized alkynes.
Several methods have been established for the synthesis of
metallic nanoparticles with high control of the size and mor-
phology, including the use of physical techniques, such as
atom evaporation, and techniques based on salt reduction or
compound precipitation, as well as electrochemical synthesis,
the use of micelles as nanoreactors, and the organometallic
approach. The decomposition of metal(0) organometallic com-
pounds in the presence of a stabilizing agent in organic media
was reported by Chaudret and Philippot as an innovative strat-
−
1
−1
frequency (TOF) obtained was 40 000 molcyclohexene molPd
at initial rates. This value corresponds to an enhanced activity
compared with a palladium catalyst prepared from PdCl
immobilized in the same support and reduced by dihydro-
h
2
2
7,28
egy for synthesizing metal nanoparticles.
This methodo-
3
5
gen. However, it is difficult to precisely determine whether
this increase is attributed to the effect of particle size, size dis-
tribution, and metal dispersion or the surface state of the pal-
ladium nanoparticles. Indeed it is expected that the
organometallic approach leaves the metallic surface more
logy allowed the preparation of metal nanoparticles under very
mild conditions (room temperature, low gas pressure) with the
additional advantage of precise control of the surface of
the nanoparticles, in particular the absence of oxidation and
the number and nature of surface species (ligands). Based on
the advantages of using organometallic precursors, our
research group developed the preparation of supported nano-
catalysts through the direct decomposition of the organometal-
lic precursor over silica-coated MNP. Highly active and
2
9
30
magnetically recoverable Ni and Pd nanocatalysts were pre-
pared by decomposition of [Ni(cod) and Pd (dba) ]. We also
2
2
3
demonstrated that the co-decomposition of such precursors
can be used to produce bimetallic NiPd nanoparticles of
various compositions as highly active hydrogenation cata-
31
32,33
lysts. Later, Reiser et al.
netic separation (carbon-coated cobalt nanobeads) and the
decomposition of Pd (dba) to prepare magnetically recover-
also used the concept of mag-
2
3
Fig. 1 TEM image of the magnetic catalyst FFSiNH
(dba) ] over the magnetic support FFSiNH
2
Pd prepared from
able catalysts for alkene hydrogenation. Here we have chosen [Pd
2
3
2
.
This journal is © The Royal Society of Chemistry 2014
Green Chem., 2014, 16, 4566–4574 | 4567

View Article Online
Paper
Green Chemistry
a
clean and available for catalysis in comparison with the pre- Table 1 Catalytic hydrogenation of 3-butyn-1-ol (1) with FFSiH
2
Pd
cursor PdCl , in which the chlorides contaminate the surface
of the catalyst.
2
b
Time
(min)
Conv.
(%)
Select.
(%)
Initial
rate
c
d
Entry
Solvent
Ethanol
CH Cl
2 2
Toluene
TOF
Afterward, we tested the activity and selectivity of the Pd
catalyst in alkyne hydrogenation. In this case, 3-butyn-1-ol (1)
was used as a model substrate (Scheme 1). This simple alkyne
was chosen to evaluate and compare the catalyst selectivity for
partial and full-hydrogenation products 3-buten-1-ol (2) and
n-butanol (3).
Initially, we evaluated the effect of the solvent on the cata-
lytic activity. Cyclohexane, toluene, dichloromethane, and
ethanol were tested for the hydrogenation reaction in the
1
2
3
4
5
25
27
41
70
45
100
100
100
100
98
3: 100%
3: 100%
3: 100%
3: 100%
2: 82%
9.42
10.18
7.28
2410
2907
1848
736
Cyclohexane
Cyclohexane
2.58
3
: 18%
2: 68%
: 31%
2: 89%
: 11%
e
6
Cyclohexane
Cyclohexane
150
90
100
93
0.63
2.97
1.06
173
7270
287
3
f
7
3
g
8
Cyclohexane
Cyclohexane
480
270
100
94
3: 100%
2: 27%
3: 73%
2: 58%
g
presence of 0.2 mol% of Pd, at 27 °C, and 3 bar of H
2
. The
9
reaction progress was followed by H pressure measurements
2
g
10
Cyclohexane
120
54
(
Fig. 2). Ethanol would be the first choice as a green and in-
expensive solvent, but the reaction was too fast, and in
0 minutes the alkyne 1 was converted into the fully hydrogen-
3: 42%
a
Reaction conditions: 1.5 mmol alkyne, ∼10 mg catalyst (0.2 mol%),
2
b
c
5
2
mL solvent, 27 °C, 3 bar H . Quantified by GC analysis. Initial rate:
d e
ated n-butanol (3) (entry 1, Table 1). The reactions performed
in dichloromethane and toluene also produced only product 3
after 30 and 40 minutes, respectively (entries 2 and 3, Table 1).
The reaction in cyclohexane required longer reaction times for
completion (ca. 72 minutes), but it also produced only product
−1
−1
−1
2 2 2
mmol H h . TOF initial: mol H mol Pd h . H pressure at 1 bar.
g
f
The reaction was performed with 0.04 mol% Pd. Reaction performed
with Pd/C (0.2 mol%).
3
(entry 4, Table 1). The fully hydrogenated product 3 was 1 was nearly completed, and the major product was alkene 2,
obtained in all solvents, which is in agreement with the con- obtained in 82% selectivity (entry 5, Table 1).
sumption of two equivalents of dihydrogen (Fig. 2). After the
In the reaction performed at 1 bar H , at 100% conversion,
consumption of one equivalent of H , the conversion of alkyne the product composition was 68% alkene 2 and 31% of alkane
2
2
3
(entry 6, Table 1). Thus, under this condition, the catalyst
was more selective than the reaction at 3 bar (entry 4, Table 1);
however, 150 minutes were necessary to achieve complete con-
version. When a lower catalyst loading was used (0.04 mol%
Pd), 89% selectivity for the semi-hydrogenated product 2 was
obtained at 93% conversion (entry 7, Table 1). The catalyst
reached a high TOF value of 7270 h− at initial rates. In
addition, when the reaction was performed without a Pd
source, alkyne 1 was recovered without changes after 8 h of
reaction. The reaction performed under similar conditions
with the commercial Pd/C catalyst (3 wt%) took nearly 8 h to
convert alkyne 1 into product 3 (entry 8, Table 1). When this
reaction was stopped at about 4.5 h after the reaction began,
the conversion of alkyne 1 reached 94%, but the major
product was again the fully hydrogenated product 3, obtained
in 73% selectivity (entry 9, Table 1). At lower conversion, the
fully hydrogenated product 3 was also present (entry 10,
1
Scheme 1 Catalytic hydrogenation of 3-butyn-1-ol (1) with the
FFSiNH Pd nanocatalyst.
2
Table 1). Therefore, the FFSiNH
activity and selectivity to perform the semi-hydrogenation of
-butyn-1-ol than Pd/C under the studied conditions. The
optimum condition for high selectivity for the semi-hydro-
genated product is 3 bar of H , 0.04 mol% Pd and cyclohexane.
2
Pd catalyst exhibited higher
3
2
In an attempt to explore the hydrogenation mechanism, a tem-
poral evolution of the product composition was examined
(
Fig. 3). The hydrogenation reaction takes place exclusively on
the alkyne, maintaining a high selectivity (>90% selectivity),
and only after almost all of alkyne 1 was reduced the alkene
hydrogenation started. This behavior suggests that the active
sites of the catalyst are preferentially occupied by the alkyne,
thus retarding the hydrogenation of 3-buten-1-ol (2). Conse-
Fig. 2 Solvent evaluation in the catalytic hydrogenation of 3-butyn-1-
ol with FFSiNH Pd: ethanol (circle), dichloromethane (square), toluene
triangle), and cyclohexane (open circle).
2
(
4568 | Green Chem., 2014, 16, 4566–4574
This journal is © The Royal Society of Chemistry 2014

View Article Online
Green Chemistry
Paper
2
Scheme 3 Hydrogenation of propargylamine 4a with the FFSiNH Pd
catalyst.
Fig. 3 Temporal evaluation of the product composition in the catalytic
hydrogenation of 3-butyn-1-ol with FFSiNH
circle), and selectivity to alkene (triangle).
2
Pd. Alkene (square), alkane
(
quently, in the presence of a high concentration of alkyne 1,
the binding of 3-buten-1-ol (2) to the catalyst is prevented until
the concentration of alkyne 1 is significantly reduced to start
the hydrogenation of alkene 2. Alkynes are actually known to
coordinate more strongly with the catalyst surface than the
alkenes, providing high selectivity to the semi-hydrogenation
1
,36–39
reaction over modified Pd catalysts.
2
The control of the Fig. 4 Catalytic hydrogenation of propargylamine 4a with FFSiNH Pd
(0.20 mol% (circle), 0.04 mol% (triangle)) and Pd/C (0.04 mol% (square)).
hydrogenation selectivity is usually obtained by surface poison-
ing when adding surface modifiers (metal, carbon, sulfur,
but Pd can also be modified under reaction con-
ditions, not only by surface deposits, but also by sub-surface
species. As no extra stabilizer was added for the synthesis of
the Pd nanoparticles, the most plausible catalyst modifiers are
the –N donor ligands present in the amino-functionalized
supports.
3
5,40,41
etc.),
was followed by H2 consumption (Fig. 4). The reaction pro-
ceeded very fast when 0.2 mol% of the FFSiNH Pd was
4
2
2
2
employed, and one equivalent of H was consumed after 0.5 h.
When a longer reaction time was allowed, after 2 h the compo-
sition of the product was 82% alkene 5a and 18% fully hydro-
genated product 6. The product’s composition remained
essentially unchanged for a few hours, reaching an equili-
brium point in the hydrogenation curve, as shown in Fig. 4.
After 24 h, the product’s composition changed slightly: 75%
alkene 5a and 25% product 6. These results indicated that the
coordination with the catalyst surface was hindered more in
the case of alkene 5a, and the alkene was forced to leave the
surface; thus, the over-hydrogenation process was efficiently
suppressed. Only after 24 h of reaction at 40 °C was the fully
hydrogenated product 6a obtained in 68% yield. Under lower
catalyst loading conditions (0.04 mol% of Pd), propargylamine
Next, the chemo- and steroselectivity of the catalyst
FFSiNH
substrates, such as propargylamine 4a–c, were considered
Scheme 2). A series of propargylamine 4a–c was synthesized
2
Pd were evaluated. For this purpose, more challenging
(
3
through an A coupling reaction employing 1 mol% CuCl for
2 h in ethyl acetate under 105 °C in a glass pressure reactor.
The alkynes 4a–c were obtained in one step, with 64–83% iso-
1
4
3
lated yield (Scheme 2).
The propargylamine 4a (1 mmol), the catalyst FFSiNH Pd,
and cyclohexane (5 mL) were charged in a Fischer–Porter glass
reactor, and the hydrogenation reaction was performed at
bar H and 27 °C. Two products were observed, the alkene 5a
2
2
4a was hydrogenated in the corresponding alkene 5a with 88%
3
selectivity after 1.5 h (68% isolated yield). After 1 h of reaction,
the H2 consumption reached one equivalent; however, some
alkyne 4a still remained in the reaction mixture. Commercial
Pd/C (3 wt%) was also examined under similar conditions, but
it exhibited lower activity (Fig. 4). The reaction took more than
and the fully-hydrogenated product 6 (Scheme 3). The reaction
3
h to start. Therefore, the hydrogenation of propargylamine
4
a–c was performed with FFSiNH Pd, employing 0.04 mol%
2
Pd. The results are summarized in Table 2. Propargylamines
b and 4c showed similar reactivity to propargylamine 4a for
4
the hydrogenation reaction. After 2 h in the presence of 3 bar
hydrogen, alkenes 5b and 5c were obtained in 53% and 61%
3
Scheme 2 Synthesis of propargylamines using an A coupling reaction.
This journal is © The Royal Society of Chemistry 2014
Green Chem., 2014, 16, 4566–4574 | 4569

View Article Online
Paper
Green Chemistry
a
2
Table 2 Catalytic hydrogenation of propargylamine 4a–c with magnetic FFSiNH Pd
b
Entry
1
Substrate
Product
Conversion (selectivity)
Isolated yield (%)
68
100% (88%)
91% (97%)
2
3
53
61
c
n.d.
a
b
Reaction conditions: 1 mmol alkyne, ∼2 mg catalyst (0.04 mol%), 5 mL solvent, 27 °C, 2 h, and 3 bar hydrogen atmosphere. Determined by
c
CG analysis. n.d. = not determined.
isolated yields, respectively (entries 2 and 3, Table 2). In terms
of stereoselectivity, when the reactions were performed under
short reaction times, the main products detected by GC analy-
sis were the alkenes 5a–c. Under these conditions, no side
reactions, such as isomerization or debenzylation, were
observed. The stereochemistry of these products was studied
1
using H NMR analysis. The products 5b–c correspond to the
Z-isomers, and the configuration of 5a could not be indubita-
1
bly attributed to the Z-isomer, since the H-NMR signal corre-
sponds to a multiplet; however, it does not suggest the
4
4
presence of the E-isomer either.
The magnetic catalyst FFSiNH
2
Pd was highly active in per-
forming the alkyne hydrogenation under mild reaction con-
ditions and in short time reactions. With a careful selection of
reaction conditions, the reaction proceeded efficiently with a
high selectivity for partial hydrogenated products. At this
Fig. 5 Recyclability test of FFSiNH
-butyn-1-ol.
2
Pd in the catalytic hydrogenation of
3
point, it is important to mention that the catalyst FFSiNH
2
Pd
is easily separated from the reaction medium through the
application of an external magnet, and the recovered catalyst pounds with pharmacological properties, as the levels of Pd
could be easily tested in successive reactions. The catalyst must be kept very low (typically 2–20 ppm) for the compound
1
9
FFSiNH
1) employing 0.2 mol% of Pd, wherein the catalyst maintained
the activity and selectivity for 7 successive hydrogenations
Fig. 5). There is no evidence of catalyst deactivation, which
2
Pd was evaluated in the hydrogenation of 3-butyn-1-ol to be used for testing and ultimately for medicinal use.
(
(
Conclusions
means that further reaction cycles are possible. After each reac-
tion and magnetic separation cycle, the leaching of Pd into the We have shown that a palladium magnetic nanocatalyst can
organic phase was measured using ICP OES analysis. The Pd be prepared easily through the decomposition of the organo-
concentration was lower than 0.005 ppm (limit of detection) in metallic precursor [Pd (dba) ] directly over the support, result-
2
3
all samples. These results indicate that no significant leaching ing in a high yield and well-dispersed Pd nanoparticles that
of Pd took place, since it represented less than 0.06% of the form exclusively on the support surface. This methodology pro-
total Pd used in the reaction, and also reinforce that the mag- vides a highly active catalyst for hydrogenation reactions, but
netic separation using a neodymium-iron-boron permanent also a selective catalyst in performing the semi-hydrogenation
magnet (4000 G) is very efficient. In fact, a complete recovery of alkynes to alkenes without the addition of metal promoters.
of the catalyst is observed in the reactor wall where the magnet The Pd catalyst was also demonstrated to be chemo- and
is placed in few minutes. The magnetic support and its mag- stereoselective with more challenging substrates, such as
3
4
netic properties were published elsewhere. The low catalyst benzylated propargylamines. In addition, if compared with
loading and the absence of Pd leaching in the organic phase commercial Pd/C, the Pd nanocatalyst was more active, more
are important and valuable characteristics for a catalyst that selective and easily recovered from the reaction media through
could be employed in the synthesis and modification of com- the application of an external magnet. There was negligible
4570 | Green Chem., 2014, 16, 4566–4574
This journal is © The Royal Society of Chemistry 2014

View Article Online
Green Chemistry
Paper
leaching of Pd to the organic product and the Pd nanocatalyst Catalyst preparation
was reused for seven cycles, maintaining its activity and
The silica-coated magnetite support (FFSi) was prepared using
selectivity. We are currently pursuing the use of this catalyst in
other important catalytic reactions, such as selective
oxidations.
reverse microemulsion, and it was functionalized with amino
34
groups (FFSiNH ), as previously reported. The Pd catalyst
2
was prepared through the direct decomposition of the organo-
metallic precursor onto the support. Initially, 25 mL of a
−
1
toluene solution of [Pd (dba) ] (0.45 mol L ) was prepared in
2
3
a Schlenk tube under argon conditions. The solution was then
Experimental
General procedures
transferred under argon with a syringe to a Fischer–Porter
reactor containing 1 g of FFSiNH , which was previously dried
2
under vacuum conditions for 1 h. The mixture was submitted
to 3 bar of dihydrogen at 25 °C under magnetic stirring for
The synthesis of the nanocatalyst was carried out under an
argon atmosphere using vacuum-line techniques and a glove-
box. The catalytic reactions were performed with standard
Fischer–Porter glass reactors.
1
h. After the reaction time, the solid was magnetically recov-
ered, washed with toluene (4 × 6 mL), and dried under vacuum
conditions for 1 h. The catalyst, denoted as FFSiNH Pd, was
2
1
13
H and C NMR spectra were obtained on a Bruker spectro-
stored under air conditions.
1
13
meter ( H at 200 MHz and 500 MHz, and C at 50 MHz and
50 MHz). The spectra were recorded in chloroform-d (CDCl ),
The morphology, size, and dispersion of the supported Pd
nanoparticles were characterized by an electronic microscopy
technique using a JEOL JEM 2010 microscope, working at
1
3
and the chemical shifts (δ) are given in ppm using tetramethyl-
silane (TMS) as the internal standard. High resolution mass
spectra (HRMS) were recorded in a Micro TOF-MS Bruker Dal-
tonics ESI. IR spectra were recorded in an IRPrestige-21 Shi-
madzu FTIR spectrometer, using an ATR module. Selected
characteristic peaks are reported in cm . The determination
of the palladium content on solid and liquid samples was per-
formed on a Spectro Arcos ICP OES. The solid samples were
analyzed after digestion with nitric acid. The liquid samples
were collected after the magnetic separation step and sub-
jected to analysis without further purification. The only
additional step performed was the catalyst washing (3 × 10 mL
200 kV with a resolution of 2.35 Å. The samples were prepared
by the deposition of a drop of the suspension on a covered
holey copper grid, followed by drying under vacuum con-
ditions for over 15 h. The diameter of ca. 200 particles was
measured from enlarged micrographs using the Image Tool
software. Using the Origin software, size-distribution histo-
grams fitted to Gaussian function were used to determine the
mean diameter and the statistical size distributions of the
nanoparticles.
−
1
3
Synthesis of propargylamines by the A coupling
2 2
CH Cl ). The organic phases were combined, evaporated to
dryness and then digested with nitric acid. The final solution The following procedure is representative of all the A coupling
was transferred to a volumetric flask and filled to the mark reactions. In a 100 mL screw cap pressure glass reactor,
3
before analysis.
copper(I) chloride (2 mg, 0.02 mmol), dibenzylamine (197 mg,
For the catalytic tests, the ends of the hydrogenation reac- 1 mmol), and ethyl acetate (2 mL) were sequentially added
tions were checked carefully by determining the dihydrogen under an air atmosphere. The mixture changed from yellow to
consumption curve and by gas chromatography, using a Shi- bluish-green. Then, with vigorous stirring, butyraldehyde
madzu GC-2010 gas chromatograph equipped with an AOC 20i (94 mg, 1.3 mmol) and 3-butyn-1-ol (140 mg, 2 mmol) were
auto injector and a flame ionization detector (FID) with a 30 m added. The reaction mixture was heated at 105 °C for 12 h.
Rtx-wax column. Column chromatography separations of com- The progress of the reaction was monitored using thin layer
pounds 4a–c, 5a–c, and 6 were carried out using silica gel flash chromatography. The reaction mixture was filtered in a Celite
chromatography, employing as the eluent mixtures of hexane pad short column. The Celite residue was washed with ethyl
and ethyl acetate with increasing polarity.
4 4
acetate (3 × 5 mL). A 10% (v/v) NH OH in saturated NH Cl
Toluene was purchased from SDS and dried in a purifi- solution (10 mL) was added, and the mixture was vigorously
cation machine (MBraun MB SPS-800), and it was used stirred for 20 min. The organic phase was collected in an
immediately after being dried. All solvents were of analytical Erlenmeyer flask; the aqueous phase was washed twice (5 mL)
grade and used as received. 3-Butyn-1-ol, propargyl alcohol, with ethyl acetate and the organic phases were collected with
dibenzylamine, N-benzylmethylamine, butyraldehyde, anisole, the first one. After combination, the organic phase was dried
magnesium sulfate, ammonium chloride, and ammonium over MgSO
hydroxide were commercially available and used as received. pressure. The residue was purified using silica-gel flash
The tris(dibenzylideneacetone) dipalladium(0) [Pd (dba) ] was chromatography, obtaining alkyne 4a in 83% isolated yield.
purchased from Stream Chemicals. Pd 3 wt% on activated The same procedure was used with alkynes 4b–c.
carbon was supplied by Sigma-Aldrich. Cyclohexene was dried 5-(N,N-Dibenzylamino)-oct-3-yn-1-ol (4a). Pale yellow oil;
with MgSO and distilled under a nitrogen atmosphere. Silica yield: 83% yield. H NMR (200 MHz, CDCl
gel (40–63 μm, 230–400 mesh particle size, Fluka) was used for J 7.2 Hz, 3H), 1.27–1.81 (m, 4H), 2.62 (td, J 6.2 Hz, J 2.0 Hz,
4
, filtered, and concentrated under reduced
2
3
1
4
3
), δ: 0.83 (t,
flash chromatography.
2H), 3.32–3.43 (m, 3H), 3.80–3.88 (m, 4H), 7.29–7.45 (m, 10H);
Green Chem., 2014, 16, 4566–4574 | 4571
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Green Chemistry
1
3
C NMR (50 MHz, CDCl
2C), 61.5, 80.4, 81.1, 126.9 (2C), 128.2 (4C), 128.9 (4C), 139.9 10 °C min−
2C). νmax 3085, 3063, 3028, 2961, 2949, 2864, 2791, 1495,
453, 1369, 1123, 1044, 1027, 906, 733, 696. HRMS (ESI): calcd 3-butyn-1-ol (1, 0.1051 g), 10 mg of catalyst (3.4 × 10 mmol
3
), δ: 13.7, 19.7, 23.2, 36.2, 51.5, 54.8 GC parameters: T
i
= 50 °C, T
f
= 150 °C, and heating rate =
1
(
(
1
.
Alkyne hydrogenation. In a typical experiment, 1.5 mmol of
−
3
+
for C H NO [M − H ] 322.2171, found 322.2168.
Pd), and 5 mL of cyclohexane were placed in a Fischer–Porter
-(N,N-Dibenzylamino)-hept-2-yn-1-ol (4b). Pale yellow oil; glass reactor. Then, the reactor was filled with 3 bar H , and
), δ: 0.84 (t, J 7.2 Hz, 3H), the reaction was magnetically stirred for the necessary time at
.31–1.75 (m, 4H), 3.40–3.49 (m, 3H), 3.87 (d, J 13.8 Hz, 2H), 27 °C. The end of the hydrogenation reaction was checked
.44 (d, J 1.8 Hz, 2H), 7.30–7.46 (m, 10H); 13C NMR (50 MHz, carefully by determining when the dihydrogen consumption
): δ 13.7, 19.6, 35.9, 51.2, 51.4, 54.8 (2C), 83.1, 84.0, 126.9 stopped. The reactor was then depressurized and opened to
2C), 128.2 (4C), 128.8 (4C), 139.8 (2C). νmax 3085, 3063, 3029, the air. After each reaction, the catalyst was magnetically separ-
2
2
27
4
2
1
yield: 64%; H NMR (200 MHz, CDCl
1
4
CDCl
(
3
3
2
1
3
957, 2931, 2871, 2807, 1716, 1496, 1451, 1366, 1123, 1070, ated using a neodymium-iron-boron permanent magnet (4000 G).
+
025, 904, 744, 694. HRMS (ESI): calcd for C21
08.2014, found 308.2020.
H
25NO: [M − H ] The catalyst responds immediately to the applied magnetic
field and accumulates in the reactor wall where the magnet is
4
-(N-Benzyl-N-methylamino)-hept-2-yn-1-ol (4c). Pale yellow placed. The liquid phase was withdrawn with a syringe. The
1
oil; yield: 73%; H NMR (500 MHz, CDCl
3
), δ: 0.89 (t, J 7.5 Hz, catalyst was kept inside the reactor and was washed with
3
3
7
3
H), 1.42–1.67 (m, 4H), 2.14 (s broad, 1H), 2.19 (s, 3H), CH Cl (3 × 10 mL). All the organic phases were combined and
2
2
.39–3.46 (m, 2H), 3.65 (d, J 13 Hz, 1H), 4.34 (d, J 1.8 Hz, 2H), the products were analyzed using GC and GC-MS and quanti-
1
3
.23–7.62 (m, 5H); C NMR (150 MHz, CDCl
3
): δ 13.8, 19.7, fied by GC using pure samples of the products and anisole as
5.8, 37.7, 51.2, 55.2, 59.1, 83.5, 83.8, 127.0, 128.3 (2C), 129.0 the standard in the internal standard method. GC parameters:
−
1
(2C), 139.2. νmax 3085, 3066, 3029, 2957, 2931, 2867, 2796,
T
i
= 40 °C, T
f
= 170 °C, and heating rate = 10 °C min . Before
1
725, 1629, 1494, 1455, 1365, 1117, 1072, 1015, 904, 744, 694. the reuse of the catalyst, no further purification was made.
+
HRMS (ESI): calcd for C H NO: [M − H ] 232.1701, found
When the alkynes 4a–c were used, the same procedure was
1
5
21
2
32.1699.
employed. After the reaction was complete, the catalyst was
magnetically separated and washed with CH
the organic phases were combined and the products were ana-
The catalytic reactions were carried out in a Fischer–Porter lyzed using GC and GC-MS. GC parameters: T = 250 °C and
2 2
Cl (3 × 10 mL),
Catalytic tests
i
glass reactor connected to a hydrogen gas reservoir with a time = 25 min. The reaction mixtures were purified using
pressure regulator that was responsible for keeping a constant silica-gel flash chromatography, employing the eluent mixtures
pressure inside the reactor. The dihydrogen consumption was of hexane and ethyl acetate with increasing polarity.
measured with a signal transducer connected to the dihydro-
gen reservoir. The catalytic activity of each catalyst was deter- yield: 68%; H NMR (200 MHz, CDCl ), δ: 0.85 (t, J 7 Hz, 3H),
5-(N,N-Dibenzylamino)-oct-3-en-1-ol (5a). Pale yellow oil;
1
3
mined at the beginning of the reaction by the slope of the 1.26–1.41 (m, 2H), 1.70–1.79 (m, 2H), 1.99–2.08 (m, 2H),
linear curve obtained from the plot of TON (turnover number) 3.36–3.47 (m, 3H), 3.57 (t, J 6 Hz, 2H), 3.74 (d, J 13.6 Hz, 2H),
1
3
versus the time of the reaction or mmol of dihydrogen con- 5.53–5.72 (m, 2H), 7.16–7.41 (m, 10H); C NMR (50 MHz,
sumed versus time. CDCl ): δ 14.2, 19.8, 31.2, 34.3, 53.7 (2C), 54.4, 62.1, 126.8 (2C),
Cyclohexene hydrogenation. In a typical experiment, 2.9 mg 128.2 (4C), 128.7 (4C), 129.2, 131.5, 140.3 (2C). νmax 3085,
3
−
4
of catalyst (9.75 × 10 mmol Pd) were placed in a Fischer– 3061, 3021, 3003, 2954, 2927, 2867, 2799, 1742, 1601, 1494,
Porter glass reactor. The catalyst was first subjected to a 1 bar 1449, 1365, 1133, 1049, 1026, 904, 739, 722, 693. HRMS (ESI):
+
of H
cyclohexene (1.6 g, 19.5 mmol) was then added (under N2
atmosphere with a syringe), and the mixture, under magnetic yield: 53%; H NMR (200 MHz, CDCl
stirring, was subjected to 6 bar H , and it was heated at 75 °C 1.29–1.87 (m, 4H), 3.38–3.51 (m, 3H), 3.77 (d, J 13.6 Hz, 2H),
until the completion of the reaction. The end of the hydrogen- 3.92–3.97 (m, 2H), 5.50–5.61 (m, 1H), 5.89 (dtd, J 11 Hz, J 7 Hz,
2
and heated at 75 °C for 1 h. After that, fresh distilled calcd for C22
H
29NO [M − H ] 324.2327, found; 324.2327.
4-(N,N-Dibenzylamino)-hept-2-en-1-ol (5b). Pale yellow oil;
1
3
), δ: 0.91 (t, J 7 Hz, 3H),
2
1
3
3
ation reaction was checked carefully by determining the dihy- J 1 Hz, 1H), 7.18–7.40 (m, 10H); C NMR (50 MHz, CDCl ):
drogen consumption curve. Finally, to stop the reaction, the δ 14.1, 19.8, 33.4, 53.7 (2C), 54.7, 58.8, 126.9 (2C), 128.2 (4C),
reactor was cooled down, depressurized and opened to the air. 128.7 (4C), 131.8, 132.3, 139.8. νmax 3085, 3059, 3025, 2957,
After each reaction, the catalyst was magnetically separated 2912, 2867, 2845, 2799, 1731, 1607, 1494, 1449, 1370, 1201,
using a neodymium-iron-boron permanent magnet (4000 G). 1071, 1026, 970, 742, 727, 696. HRMS (ESI): calcd for
+
The catalyst responds immediately to the applied magnetic C H NO [M − H ] 310.2165, found 310.2169.
2
1
27
field and accumulates in the reactor wall where the magnet is
placed. The liquid phase was withdrawn with a syringe. The oil; yield: 61%; H NMR (500 MHz, CDCl
catalyst was kept inside the reactor and was washed with 3H), 1.21–1.39 (m, 3H), 1.69–1.73 (m, 1H), 2.16 (s, 3H),
CH Cl (3 × 10 mL). All the organic phases were combined and 3.42–3.44 (m, 1H), 3.55 (d, J 13 Hz, 1H), 3.62 (d, J 13 Hz, 1H),
4-(N-Benzyl-N-methylamino)-hep-2-en-1-ol (5c). Pale yellow
1
3
), δ: 0.94 (t, J 7 Hz,
2
2
the products were analyzed using GC and GC-MS and quanti- 4.09 (dd, J 14 Hz, J 5.5 Hz, 1H), 4.29 (ddd, J 14 Hz, J 5.5 Hz,
fied by GC using pure samples of the products as the standard. J 2 Hz, 1H), 5.54–5.58 (m, 1H), 5.96 (dtd, J 11.5 Hz, J 5.5 Hz,
4572 | Green Chem., 2014, 16, 4566–4574
This journal is © The Royal Society of Chemistry 2014

View Article Online
Green Chemistry
Paper
1
3
J 1.5 Hz, 1H), 7.21–7.32 (m, 5H); C NMR (150 MHz, CDCl
3
): 16 S. Turcaud, F. Berhal and J. Royer, J. Org. Chem., 2007, 72,
δ 14.2, 20.0, 30.5, 36.7, 57.6, 59.7, 59.9, 127.2, 128.4 (2C), 129.1
7893.
(2C), 133.5, 133.7, 138.6. νmax 3085, 3060, 3026, 3009, 2954, 17 N. Gommermann and P. Knochel, Tetrahedron, 2005, 61, 9.
2
1
C
927, 2869, 2791, 1740, 1494, 1454, 1371, 1315, 1254, 1208, 18 N. Gommermann and P. Knochel, Chem. Commun., 2005,
070, 1014, 942, 911, 803, 733, 697. HRMS (ESI): calcd for
4175.
+
15
H
23NO [M − H ] 234.1852, found 234.1847.
19 C. E. Garrett and K. Prasad, Adv. Synth. Catal., 2004, 346,
5
-(N,N-Dibenzylamino)-octan-1-ol (6). Pale yellow oil;
889.
1
H NMR (200 MHz, CDCl ), δ: 0.83 (t, J 7.2 Hz, 3H), 1.19–1.69 20 V. Polshettiwar, R. Luque, A. Fihri, H. Zhu, M. Bouhrara
3
(
m, 10H), 2.36–2.51 (m, 1H), 3.55–3.62 (m, 6H), 7.16–7.38 (m,
and J.-M. Basset, Chem. Rev., 2011, 111, 3036.
1
3
1
3
1
2
1
0H); C NMR (50 MHz, CDCl
3
): δ 14.3, 20.4, 23.1, 29.5, 31.8, 21 L. L. Chng, N. Erathodiyil and J. Y. Ying, Acc. Chem. Res.,
2.8, 53.3 (2C), 56.7, 63.0, 126.6 (2C), 128.0 (4C), 128.9 (4C),
40.78 (2C). νmax 3084, 3061, 3025, 2958, 2927, 2862, 2798, 22 V. Kelsen, B. Wendt, S. Werkmeister, K. Junge, M. Beller
736, 1745, 1603, 1496, 1451, 1372, 1361, 1256, 1244, 1205, and B. Chaudret, Chem. Commun., 2013, 49, 3416.
144, 1105, 1069, 1049, 1028, 972, 954, 905, 808, 743, 695. 23 C. Rangheard, C. de Julian Fernandez, P.-H. Phua,
2013, 46, 1825.
+
HRMS (ESI): calcd for C22
26.2470.
H
31NO [M − H ] 326.2478, found:
J. Hoorn, L. Lefort and J. G. de Vries, Dalton Trans., 2010,
39, 8464.
3
2
2
4 M. Stein, J. Wieland, P. Steurer, F. Tölle, R. Mülhaupt and
B. Breit, Adv. Synth. Catal., 2011, 353, 523.
5 R. Abu-Reziq, D. Wang, M. Post and H. Alper, Adv. Synth.
Catal., 2007, 349, 2145.
Acknowledgements
The authors are grateful for the financial support offered by 26 A. Banerjee, R. Theron and R. W. J. Scott, ChemSusChem,
FAPESP, CNPq, TWAS, CAPES, and the University of São Paulo 2012, 5, 109.
through the NAP-CatSinQ (Research Core in Catalysis and 27 K. Philippot and B. Chaudret, C. R. Chim., 2003, 6,
Chemical Synthesis).
1019.
2
2
8 B. Chaudret, C. R. Phys., 2005, 6, 117.
9 N. J. S. Costa, R. F. Jardim, S. H. Masunaga,
D. Zanchet, R. Landers and L. M. Rossi, ACS Catal.,
Notes and references
2012, 2, 925.
1
2
3
4
5
M. Crespo-Quesada, F. Cárdenas-Lizana, A.-L. Dessimoz 30 M. Guerrero, N. J. S. Costa, L. L. R. Vono, L. M. Rossi,
and L. Kiwi-Minsker, ACS Catal., 2012, 2, 1773.
I. Matsuda, J. Sakakibara and H. Nagashima, Tetrahedron
Lett., 1991, 32, 7431.
E. V. Gusevskaya and K. Philippot, J. Mater. Chem., 2013, 1,
1441.
31 N. J. S. Costa, M. Guerrero, V. Colliere, E. Teixeira-Neto,
R. Landers, K. Philippot and L. M. Rossi, ACS Catal., 2014,
4, 1735.
B. M. Trost and S. F. Chen, J. Am. Chem. Soc., 1986, 108,
6
053.
M. Konishi, H. Ohkuma, T. Tsuno, T. Oki, G. D. VanDuyne 32 Q. M. Kainz, R. Linhardt, R. N. Grass, G. Vilé, J. Pérez-
and J. Clardy, J. Am. Chem. Soc., 1990, 112, 3715.
M. Miura, M. Enna, K. Okuro and M. Nomura, J. Org.
Chem., 1995, 60, 4999.
Ramírez, W. J. Stark and O. Reiser, Adv. Funct. Mater., 2014,
24, 2020.
33 R. Linhardt, Q. M. Kainz, R. N. Grass, W. J. Stark and
O. Reiser, RSC Adv., 2014, 4, 8541.
6
7
B. Jiang and Y.-G. Si, Angew. Chem., Int. Ed., 2004, 43, 216.
G. Huang, Z. Yin and X. Zhang, Chem. – Eur. J., 2013, 19, 34 M. J. Jacinto, P. K. Kiyohara, S. H. Masunaga, R. F. Jardim
1992. and L. M. Rossi, Appl. Catal., A, 2008, 338, 52.
F. Colombo, M. Benaglia, S. Orlandi, F. Usuelli and 35 L. M. Rossi, I. M. Nangoi and N. J. S. Costa, Inorg. Chem.,
G. Celentano, J. Org. Chem., 2006, 71, 2064. 2009, 48, 4640.
C. Koradin, K. Polborn and P. Knochel, Angew. Chem., Int. 36 C. Na-Chiangmai, N. Tiengchad, P. Kittisakmontree,
1
8
9
Ed., 2002, 41, 2535.
0 C. Koradin, N. Gommermann, K. Polborn and P. Knochel,
Chem. – Eur. J., 2003, 9, 2797–2811.
O. Mekasuwandumrong, J. Powell and J. Panpranot, Catal.
Lett., 2011, 141, 1149.
37 P. Sautet and J.-F. Paul, Catal. Lett., 1991, 9, 245.
38 X. C. Guo and R. J. Madix, J. Catal., 1995, 155, 336.
1
1
1 Z. Li and C.-J. Li, J. Am. Chem. Soc., 2004, 126, 11810.
1
2 H.-P. Bi, Q. Teng, M. Guan, W.-W. Chen, Y.-M. Liang, 39 T. Ouchaib, J. Massardier and A. Renouprez, J. Catal., 1989,
X. Yao and C.-J. Li, J. Org. Chem., 2010, 75, 783. 119, 517.
3 V. A. Peshkov, O. P. Pereshivko and E. V. Van der Eycken, 40 N. Lopez and C. Vargas-Fuentes, Chem. Commun., 2012, 48,
Chem. Soc. Rev., 2012, 41, 3790. 1379.
4 E. M. Vieira, F. Haeffner, M. L. Snapper and A. H. Hoveyda, 41 A. Borodziński and G. C. Bond, Catal. Rev. Sci. Eng., 2008,
Angew. Chem., Int. Ed., 2012, 51, 6618. 50, 379.
5 E. M. Campi, W. Roy Jackson and Y. Nilsson, Tetrahedron 42 M. Armbrüster, M. Behrens, F. Cinquini, K. Föttinger,
1
1
1
Lett., 1991, 32, 1093.
Y. Grin, A. Haghofer, B. Klötzer, A. Knop-Gericke,
This journal is © The Royal Society of Chemistry 2014
Green Chem., 2014, 16, 4566–4574 | 4573

View Article Online
Paper
Green Chemistry
H. Lorenz, A. Ota, S. Penner, J. Prinz, C. Rameshan, 43 G. Z. Melgar, E. P. Wendler, A. A. dos Santos and
Z. Révay, D. Rosenthal, G. Rupprechter, P. Sautet, A. L. M. Porto, Tetrahedron: Asymmetry, 2010, 21, 2271.
R. Schlögl, L. Shao, L. Szentmiklósi, D. Teschner, D. Torres, 44 A. J. Cresswell, S. G. Davies, J. A. Lee, M. J. Morris,
R. Wagner, R. Widmer and G. Wowsnick, ChemCatChem,
012, 4, 1048.
P. M. Roberts and J. E. Thomson, J. Org. Chem., 2012, 77,
7262.
2
4574 | Green Chem., 2014, 16, 4566–4574
This journal is © The Royal Society of Chemistry 2014