Inexpensive and rapid hydrogenation catalyst from CuSO4/CoCl2 — Chemoselective reduction of alkenes and alkynes in the presence of benzyl protecting groups

Article abstract of DOI:10.1016/j.tetlet.2018.02.026

The simple reduction of a number of alkenes and alkynes was performed with a typical reaction time of 20 min using a copper-cobalt catalytic system. The reduction did not cleave benzyl protecting groups which are usually vulnerable to catalytic hydrogenation reactions. The catalyst can be prepared in situ by reduction of the inexpensive precursor salts CuSO₄ and CoCl₂ with NaBH₄. Sodium borohydride was also used as an easily handled hydrogen source for the catalytic reductions. No pressure, heating or inert atmosphere is required and purification/catalyst removal is achieved using extraction procedures, making this approach simple and efficient.

Full text of DOI:10.1016/j.tetlet.2018.02.026

Accepted Manuscript
Inexpensive and rapid hydrogenation catalyst from CuSO /CoCl — chemose-
4
2
lective reduction of alkenes and alkynes in the presence of benzyl protecting
groups
Mario Ficker, Søren Wedel Svenningsen, Thomas Larribeau, Jørn B.
Christensen
PII:
S0040-4039(18)30196-5
https://doi.org/10.1016/j.tetlet.2018.02.026
TETL 49711
DOI:
Reference:
To appear in:
Tetrahedron Letters
Received Date:
Revised Date:
Accepted Date:
14 December 2017
6 February 2018
9 February 2018
Please cite this article as: Ficker, M., Svenningsen, S.W., Larribeau, T., Christensen, J.B., Inexpensive and rapid
hydrogenation catalyst from CuSO /CoCl — chemoselective reduction of alkenes and alkynes in the presence of
4
2
benzyl protecting groups, Tetrahedron Letters (2018), doi: https://doi.org/10.1016/j.tetlet.2018.02.026
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Inexpensive and rapid hydrogenation catalyst from CuSO /CoCl —
4
2
chemoselective reduction of alkenes and alkynes in the presence of benzyl
protecting groups
†
†
Mario Ficker, Søren Wedel Svenningsen, Thomas Larribeau, Jørn B. Christensen*
Department of Chemistry, University of Copenhagen, Thorvaldsensvej 40, Frederiksberg, DK-
1
871 Denmark
*corresponding author, e-mail: jbc@chem.ku.dk
†
Authors contributed equally
Graphical Abstract:

Abstract
The simple reduction of a number of alkenes and alkynes was performed with a typical reaction time
of 20 minutes using a copper-cobalt catalytic system. The reduction did not cleave benzyl protecting
groups which are usually vulnerable to catalytic hydrogenation reactions. The catalyst can be
prepared in situ by reduction of the inexpensive precursor salts CuSO and CoCl with NaBH . Sodium
4
2
4
borohydride was also used as an easily handled hydrogen source for the catalytic reductions. No
pressure, heating or inert atmosphere is required and purification/catalyst removal is achieved using
extraction procedures, making this approach simple and efficient.
Key words:
Catalytic hydrogenation, alternative hydrogen sources, benzyl protecting group

Introduction
The catalytic reduction of alkenes and alkynes with hydrogen and a transition metal catalyst is a
1,2
standard reaction in organic synthesis. These reactions typically require expensive metals from
groups 8 and 10, with platinum metals representing the most active, and a hydrogen source, such as
3- 7
H gas, hydrazine or formates. Specialized glassware and inert atmospheres are typically required,
2
and often these reactions are performed under pressure (H ) or using highly toxic and
2
8,9
environmentally harmful reagents (hydrazine, formates), which are not practical for a non-
industrial laboratory use.
Recently, we described the formation and catalytic activity of bimetallic copper-cobalt nanoparticles
10
that were stabilized by encapsulation in a dendrimer. This procedure gave nanoparticles (1-3 nm)
capable of reducing alkenes and alkynes with low catalyst loading and short reaction times using
NaBH as a solid hydrogen source; however due to cost of the dendrimer this system is expensive,
4
requires a nitrogen atmosphere and is difficult to scale-up. Herein, we describe the development of
an inexpensive and simple copper-cobalt catalyst which is prepared in situ, and utilises sodium
borohydride as a solid and easily manageable hydrogen source. The reaction is fast with an
operationally simple extraction work up which could become a viable alternative to Pd/C for
standard hydrogenation reactions.
Results and Discussion
A promising bi-metallic nano-catalytic system comprising of a 1:1 mixture of cobalt and
10
copper(Cu Co @G3-PAMAM-Pyr, 1 nm metal particle size) was previously described. These
6
6
nanoparticles were highly potent in catalytic reductions with low catalyst loading and linear kinetics.
However, the system was difficult to transform into an easy and cheap standard laboratory reaction.
In this study we investigated the ability of non-stabilized Cu-Co bimetallic particles to reduce alkenes
and alkynes without the need of a dendrimer as support.
Gratifyingly, it was found that a mixture of CuSO and CoCl salts dissolved in methanol formed a
4
2
metallic catalyst system in situ upon the addition of NaBH . These bimetallic particles were found to
4
promote the decomposition of NaBH to hydrogen and could be used as a solid hydrogen source. The
4
reduction of 10-undecen-1-ol was used as test substrate. An initial analysis of different ratios
between copper and cobalt revealed that the best ratio for catalytic hydrogenation was a catalyst
with the original composition of 91% Cu and 9% Co.

As a general experimental procedure, the precursor salts CuSO •6H O and CoCl •5H O (10:1) were
4
2
2
2
dissolved in methanol, and the unsaturated starting material added directly to the solution. The
reaction was initiated by the addition of NaBH to the solution, which reduced the copper(II) and
4
cobalt(II) salts to the copper-cobalt-boride catalyst as finely dispersed particles, similar to the
previously employed dendrimer encapsulated nanoparticles. However, the metal catalyst aggregated
to form larger particles during the reaction, due to the lack of a stabilizing polymer. The in situ
generated catalyst immediately started to decompose the excess NaBH to hydrogen, which was
4
transferred to the double or triple bond of the starting material. Additional portions of NaBH were
4
typically added over 20 minutes, which was sufficient to quantitatively reduce most double and triple
bonds. The operationally simple work-up procedure was achieved by extraction with water/CH Cl ,
2
2
where the catalyst and the boronic acid salts remained in the water phase, while the product was
extracted into the organic layer, which was dried over MgSO and evaporated to give isolated yields
4
of typically above 95%.

Scheme 1. Test substrates and products of the catalytic hydrogenation reaction using the copper-cobalt-boride
catalyst. Reaction conditions and conversions are listed in Table 1.

a
Table 1. Conditions for the catalytic hydrogenation of compounds 1-11a.
NaBH4
equiv.)
t
b
Substrate
Catalyst
Conversion
(
(min)
Isolated Yield
1
0
3
0
3
0
3
0
5
0
3
0
1
1
3
0
5
0
3
0
2
2
3
0
5
0
.0 mol% Cu
.1 mol% Co
.0 mol% Cu
.3 mol% Co
.0 mol% Cu
.3 mol% Co
.0 mol% Cu
.3 mol% Co
.0 mol% Cu
.5 mol% Co
.0 mol% Cu
.3 mol% Co
0 mol% Cu
.0 mol% Co
.0 mol% Cu
.3 mol% Co
.0 mol% Cu
.5 mol% Co
.0 mol% Cu
.3 mol% Co
5 mol% Cu
.5 mol% Co
.0 mol% Cu
.3 mol% Co
.0 mol% Cu
.5 mol% Co
1
2
3
4
5
6
7
8
9
1
1
1
1
a
1.5
2.0
2.0
2.0
4.0
2.0
1.5
2.0
4.0
2.0
6.0
2.0
4.0
10
10
20
20
20
20
10
20
20
20
25
20
20
quant.
quant.
quant.
quant.
quant.
quant.
91%
a
94%
65%
89%
96%
95%
a
a
a
a
c
26% 7b
e
a
-
56% 7c
a
quant.
93%
96%
97%
89%
66%
94%
a
quant.
quant.
0a
1a
2a
3a
d
89
88
quant. 13c
4
9% 14c
26% 14b
5% 14d
2
0
.0 mol% Cu
.2 mol% Co
e
1
1
4a
2.0
15
-
2
6
0
3
0
.0 mol% Cu
.6 mol% Co
.0 mol% Cu
.3 mol% Co
35% 15b
36% 15c
e
5a
6a
4.0
2.0
20
20
-
1
64%
93%
a
Reaction performed at ambient temperature and atmosphere. Methanol was used as the standard solvent. Extracted
b
1
c
yields and further reaction details can be found in the ESI. Conversion determined by H-NMR and GC-MS. 3% of mono
d
e
reduced compound as the 1,4 addition product. 2% was doubly reduced and 9% of the starting material remained.
Isolated as a product mixture containing starting material or intermediate products (see ESI for details and composition).
A systematic study of the catalyst potential was performed with the compounds shown in Scheme 1
and reaction conditions presented in Table 1. α,β-Unsaturated esters were readily reduced, leaving
the ester group intact. The lowest reaction time and catalyst loading was observed for dimethyl

itaconate 1a (1 mol% Cu, 0.1 mol% Co, 10 min.), where the conjugated double bond is less hindered
than in acrylates (2a and 3a), methyl cinnamate (4a) and cis,cis-dimethyl muconate (5a).
Reduction of a terminal double bond conjugated to an aromatic ring system (styrene 6a) was rapid
with low catalyst loading (3mol% Cu and 0.3 mol% Co), while internal double bonds (7a) were harder
to reduce (10 mol% Cu and 1 mol% Co), which could also be explained by the trans-trans structure of
the compound, which is difficult to reduce via the syn-addition of hydrogen. In all aromatic
compounds the conjugated ring system remained intact during the reaction. Typical ester protection
groups such as benzyl ester (2a) and tert-butyl ester (3a) were unaffected by the catalytic system.
The stability of the benzyl ester is notable, since this group is usually removed by hydrogenation
11-13
reactions;
this could open new pathways for reducing unsaturated carbon systems in the
presence of benzyl protecting groups. The chemoselective hydrogenation of olefins is usually
achieved by selective poisoning of Pd/C or other expensive rare metal catalysts, which is time
14-16
consuming due to catalyst preparation and long reaction times of the deactivated catalyst.
Terminal double bonds not conjugated to another π-system were readily reduced as demonstrated
with compounds 8a, 9a and 10a. The acetyl ester in compound 9a and the benzyl ether (10a), a
13,17
group potentially vulnerable to hydrogenation reactions,
remained intact during the reaction.
The preference for terminal double bonds was demonstrated with (+)-limonene (11a) where the
terminal alkene was selectively hydrogenated, leaving the more hindered double-bond virtually
untouched; only 2% of the fully reduced compound was observed by GC-MS. However, the catalyst
loading for 11a was almost ten times higher than for the other compounds (25 mol% Cu and 2.5
mol% Co). This suggests that hydrogenation under these conditions is sensitive to the substitution
pattern of the double-bond. The internal double bond in β-citronellol (12a) could be almost
quantitatively reduced; this could be due to the less sterically hindered nature of the double bond
compared to the aliphatic ring system in (+)-limonene where the double bond can hardly be accessed
by the catalyst.
Furthermore, alkynes could also be successfully reduced to the alkene and further to the
corresponding alkane. The reduction of compounds 13a was quantitative with a fully saturated
product, whereas compound 14a was chosen for a mechanism study with low catalyst loading (2
mol% Cu and 0.2 mol% Co) and short reaction time (15 min) in order to isolate intermediate products
(
vide infra, Fig. 3).
Amine containing compounds (15a and 16a) showed incomplete reaction (36% 15c, 64% 16b) under
the applied catalytic conditions. Furthermore, the copper salts from the catalyst seemed to form
stable complexes with the amines which interfered with NMR measurements due to the
paramagnetic character of Cu(II). Especially, nitrile containing compound 15a showed, besides the
desired reaction product 15c (36%), mono reduced species 15c (35%) and starting material 15a

(16%), and some unidentified products (12%), possibly due to cyclization reactions. Thus, the
application of this catalytic system to amine containing compounds may require additional treatment
to remove Cu(II), for instance reaction of the Cu(II) with a chelating agent. However, for typical
products (e.g. 1-6, 7-10 and 13) an operationally simple extraction procedure was sufficient to obtain
the desired product without further purification. The catalytic reaction was also demonstrated to be
up-scalable, as demonstrated by the 4.7 g reaction scale of β-citronellol 12a.
Figure 1. Plot showing reaction progress versus reaction time for (+)-limonene. The decrease in reaction
1
progress indicates Ostwald-ripening of the catalyst. Reaction progress was determined using H-NMR and GC-
MS. Reagents and conditions: 15 mol% Cu / 1.5 mol% Co in methanol, NaBH (4.3 equiv.).
4
Preliminary studies of the reaction progress indicated that the catalytic system is prone to Ostwald-
18
ripening (Fig. 1). Ostwald-ripening is the formation of larger clusters of metal-particles from smaller
metal-particles which decreases the available surface area of the catalyst. If the reaction occurs on
the surface of the metal this should lower the reaction rate, since the number of available binding
sites is lowered. (+)-Limonene (10a) was chosen because the reaction rate was sufficiently low to
allow several samples to be removed during the reaction. The reaction rate proceeded relatively
quickly in the first two minutes, converting 39% of the starting material to compound 10b. The initial
-
1
turn over frequency (TOF) for the first 2 minutes was calculated to be 78 h . After the first 2 minutes
the reaction followed zeroth order kinetics, which is expected for a catalyst with constant surface
-
1
area. Here, the TOF was calculated to be approximately 8 h , a ten-fold decrease in reaction rate.
This indicates that the catalyst starts with a large number of available binding sites, which drastically
decreases to a constant level within 2 minutes.
The observed difference in reaction times and catalyst loading between isolated- and carbonyl
conjugated double bonds was investigated by two mechanistic studies. The carbonyl conjugated

systems can be hydrogenated by at least two possible pathways: (i) syn addition of H to the double
2
bond, assisted by the catalyst and (ii) “Michael-type” attack of the hydride to the conjugated double
bond followed by proton abstraction from the solvent (Fig. 2). In this experiment the reaction was
performed in deuterated methanol under the standard conditions. In the “Michael-type” addition
the product would be 4c with deuterium in the β-position. If the addition is a syn type addition of HD,
which is produced when NaBH is decomposed by the catalyst in MeOD, the product will be a 1:1
4
1
mixture of 4c and 4d. Examination of the product H-NMR spectrum would then help to elucidate
1
whether one or both possible pathways are operating by comparing the H-integrals of the α- and β-
positions. While the “Michael-type” addition would lead to a 2:1 integral ratio between the signals of
1
the α- and β-hydrogens, syn addition would lead to a 1:1 integral ratio. The section of the H-NMR
spectrum showing the signals from the α,β-region is shown in Figure 2. The integrals displayed a
1
.89:1.19 ratio between the α- and β-positions, which does not correlate completely to either
pathway. The overserved results indicate that ¾ of the product is produced by the “Michael-type”
pathway, whereas ¼ is produced by the syn pathway.
Figure 2. Left: Two possible pathways for the reduction of compound 2a to 2b. (a) “Michael-type” addition
assisted by the catalyst forming product 2c. (b) syn addition of H-D assisted by the catalyst forming products 2c
1
and 2d in a 1:1 mixture. Right: The α,β-region of the product H-NMR spectrum from the reduction performed
in MeOD. Integrals are based on the methyl group, set to 3H. Reaction conditions: 6 mol% Cu / 0.6 mol% Co in
MeOD (6 mL), NaBH (2 equiv.), 20 min.
4

Figure 3. Monitored reaction process for the reduction of diphenyl acetylene with sodium
borohydride using 2 mol% Cu and 0.2 mol% Co.
Further investigation was performed on diphenyl acetylene. In a study by Magoon and Slaugh,several
19
alkynes were treated with lithium aluminium hydride in THF to afford the trans-isomers. The
authors explained this outcome by the anti-addition of aluminium hydride followed by substitution
of the aluminium with hydrogen. Formation of a cis-stilbene (14c) would suggest a syn mechanism,
where hydrogen is adsorbed onto the catalyst surface and then transferred to the double-bond. The
trans-stilbene (14b) would be formed if another mechanism was responsible for the observed
reduction. As depicted in Figure 3, the reduction proceeds readily to product 14d; the intermediate
stilbene products were quantifiable by NMR and GC-MS. The product ratio between cis- and trans-
stilbene was 3.5:1 was after five minutes reaction time. In this particular case it seems that the
reaction proceeds mainly through syn addition of H . The observed trans-stilbene can either be
2
explained by photo isomerization of cis-stilbene, to the thermodynamically more stable trans isomer
during work-up and characterization, or by a second pathway similar to the one proposed by Magoon
1
9
and Slaugh.
Conclusion
In summary, an inexpensive and easily manageable catalyst for the catalytic hydrogenation of
alkenes and alkynes was generated in situ from the inexpensive starting materials CuSO •6H O and
4
2
CoCl •5H O. The reduction did not cleave hydrogenation sensitive benzyl protecting groups. The
2
2
reaction can be performed in an open round bottom flask and, thus, does not require specialised

glassware or inert conditions. NaBH is used as solid hydrogen source, which is an advantage
4
compared to H which needs to be handled as a highly flammable gas. The reaction times were fast
2
(typically 20 min.) and a simple extraction procedure could be applied for workup and purification.
Including workup and preparation time, the isolated hydrogenated compounds can be obtained
within 1 h, which makes this route very useful synthetic procedure in everyday laboratory work.
References
1
2
3
4
5
6
7
8
9
1
.
Blaser HU, Malan C, Pugin B, Spindler F, Steiner H, Studer M, Adv. Synth. Catal. 2003; 345: 103-151.
Brieger G, Nestrick TJ, Chem. Rev. 1974; 74: 567-580.
.
.
Roessler F, CHIMIA 2003; 57: 791-798.
.
Hünig S, Müller H, Thier W, Angew. Chem. Int. Ed. 1965; 4: 271-280.
Furst A, Berlo R C, Hooton S, Chem. Rev. 1965; 65: 51-68.
.
.
Johnstone RAW, Wilby AH, Entwistle ID, Chem. Rev. 1985; 85: 129-170.
Rylander P. N. In Hydrogenation methods. Academic Pr: Chambridge, 1990; Vol. 1.
Witkin LB, Arch. Indust. Health 1956; 13: 34-6.
.
.
.
Thrasher JD, Kilburn KH, Arch Environ Health 2001; 56: 300-311.
Ficker M, Petersen JF, Gschneidtner T, Rasmussen AL, Purdy T, Hansen JS, Hansen TH, Husted S,
Moth Poulsen K, Olsson E, Christensen JB, Chem. Commun. 2015; 51: 9957-9960.
Felix AM, Heimer EP, Lambros TJ, Tzougraki C, Meienhofer J, J. Org. Chem. 1978; 43: 4194-4196.
Rao VS, Perlin AS, Carbohydr. Res. 1980; 83: 175-177.
0.
11.
1
1
1
1
1
1
1
1
2.
3.
4.
5.
6.
7.
8.
9.
ElAmin B, Anantharamaiah GM, Royer GP, Means GE, J. Org. Chem. 1979; 44: 3442-3444.
Sajiki H, Hattori K, Hirota K, J. Org. Chem. 1998; 63: 7990-7992.
Sajiki H, Hirota K, Tetrahedron 1998; 54: 13981-13996.
Hattori K, Sajiki H, Hirota K, Tetrahedron 2000; 56: 8433-8441.
Llàcer E, Romea P, Urpí F, Tetrahedron Lett. 2006; 47: 5815-5818.
Ostwald W. in Lehrbuch der allgemeinen Chemie. W. Engelmann: Leipzig, 1886; Vol. 2.
Magoon E, Slaugh L, Tetrahedron 1967; 23: 4509-4515.

Highlights




Cheap and efficient catalyst prepared in situ.
Short reaction time.
Easy work-up.
Does not cleave benzyl groups.