Ultra-small cobalt nanoparticles from molecularly-defined Co-salen complexes for catalytic synthesis of amines

Article abstract of DOI:10.1039/c9sc04963k

We report the synthesis of in situ generated cobalt nanoparticles from molecularly defined complexes as efficient and selective catalysts for reductive amination reactions. In the presence of ammonia and hydrogen, cobalt-salen complexes such as cobalt(ii)-N,N′-bis(salicylidene)-1,2-phenylenediamine produce ultra-small (2-4 nm) cobalt-nanoparticles embedded in a carbon-nitrogen framework. The resulting materials constitute stable, reusable and magnetically separable catalysts, which enable the synthesis of linear and branched benzylic, heterocyclic and aliphatic primary amines from carbonyl compounds and ammonia. The isolated nanoparticles also represent excellent catalysts for the synthesis of primary, secondary as well as tertiary amines including biologically relevant N-methyl amines.

Full text of DOI:10.1039/c9sc04963k

Chemical
Science
View Article Online
View Journal
EDGE ARTICLE
Ultra-small cobalt nanoparticles from molecularly-
defined Co–salen complexes for catalytic synthesis
of amines†
Cite this: DOI: 10.1039/c9sc04963k
All publication charges for this article
have been paid for by the Royal Society
of Chemistry
Thirusangumurugan Senthamarai,a Vishwas G. Chandrashekhar,a
b
a
b
Manoj B. Gawande,
Paul C. J. Kamer,
Narayana V. Kalevaru,
Rajenahally V. Jagadeesh
Radek Zboril,
ˇ
a
a
a
*
*
and Matthias Beller
We report the synthesis of in situ generated cobalt nanoparticles from molecularly defined complexes as
efficient and selective catalysts for reductive amination reactions. In the presence of ammonia and
hydrogen, cobalt–salen complexes such as cobalt(II)–N,N0-bis(salicylidene)-1,2-phenylenediamine
produce ultra-small (2–4 nm) cobalt-nanoparticles embedded in a carbon–nitrogen framework. The
resulting materials constitute stable, reusable and magnetically separable catalysts, which enable the
synthesis of linear and branched benzylic, heterocyclic and aliphatic primary amines from carbonyl
compounds and ammonia. The isolated nanoparticles also represent excellent catalysts for the synthesis
of primary, secondary as well as tertiary amines including biologically relevant N-methyl amines.
Received 2nd October 2019
Accepted 3rd February 2020
DOI: 10.1039/c9sc04963k
rsc.li/chemical-science
activity and selectivity for the preparation of functionalized
amines,1d–i nitriles,1k,4c carboxylic acid derivatives,1f,k and cyclo-
aliphatic compounds.1j Although this preparation represents
Introduction
In recent years, 3d metal-based nanoparticles (NPs) emerged as
promising catalysts for the synthesis of functionalized and
complex organic molecules for advanced applications in life
and material sciences.1 Traditionally, such syntheses are per-
formed using homogeneous organometallic complexes,2 which
are oen sensitive and more difficult to recycle compared to
heterogeneous materials.1,2 For the preparation of stable but at
the same time active and selective NPs, the use of suitable
precursors and optimal methods is crucial.1 Commonly, nano-
particles are prepared by chemical reduction processes, calci-
nation or pyrolysis in the presence of suitable supports and
metal precursors. The resulting materials are applied particu-
larly in industrially-relevant bench mark reactions of less
functionalized molecules.3 However, in recent years there is an
increasing interest to use such catalysts for advanced organic
synthesis, specically for the preparation of life science prod-
ucts.1 In this respect, the preparation of specic NPs by
immobilization and pyrolysis of organometallic complexes or
metal organic frameworks (MOFs) on heterogeneous supports
attracted also attention.1,4 These supported NPs show high
a highly useful tool to produce novel nano-structured catalysts
on lab-scale, the upscaling can be difficult and requires
specialized equipment.1,4 Thus, the use of alternative, more
convenient methods is highly desired. One possibility is the
practical in situ generation of active heterogeneous NPs.5 Based
on this idea, herein we report a straightforward approach for
the generation of cobalt-based NPs in situ from molecularly-
dened metal complexes and their application in reductive
amination reactions using ammonia and molecular hydrogen
(Fig. 1).
The resulting amines represent privileged molecules widely
used in chemistry, medicine, biology, and material science.6 For
their synthesis, catalytic reductive amination of carbonyl
compounds using molecular hydrogen is widely applied as cost-
a
¨
¨
Leibniz-Institut fur Katalyse e. V. an der Universitat Rostock, Albert-Einstein-Str. 29a,
18059 Rostock, Germany. E-mail: jagadeesh.rajenahally@catalysis.de; matthias.
beller@catalysis.de
bRegional Centre of Advanced Technologies and Materials, Department of Physical
ˇ
´
˚
Chemistry, Faculty of Science, Palacky University, Olomouc, Slechtitelu 27,
Olomouc, 78371, Czech Republic
† Electronic supplementary information (ESI) available. See DOI:
10.1039/c9sc04963k
Fig. 1 In situ generation of Co-NPs for reductive aminations.
This journal is © The Royal Society of Chemistry 2020
Chem. Sci.

View Article Online
Chemical Science
Edge Article
effective methodology in both academic research and indus- “real” active species for the reductive amination reaction. To
try.1e,h,7–10 Among reductive aminations, the synthesis of primary conrm this, we performed a standard mercury test in the pres-
amines, which can be easily functionalized to high value ence of complex I and aer addition of 15 mg Hg the reductive
products, continues to be especially important.1e,h,j,8–10 amination reaction did not occur. Hot ltration of NPs and
Regarding catalysts for this reaction, precious metals-based testing the ltrate for the reaction showed that Co-NPs did not go
ones are known to a large extent.8,9 However, in recent years into solution as soluble particles. Studying the course of the
Co-1e and Ni1h,i-based nanocatalysts have successfully been benchmark reaction at different intervals of time showed a pro-
developed in addition to RANEY® nickel.10
longed catalyst preformation time and only aer 10 h 4-bromo-
benzylamine started to form (Fig. S3†). Apparently, complex I
generated nanoparticles slowly, which then catalyze the desired
amination process. For comparison, we also prepared cobalt
nanoparticles separately by mixing complex I, ammonia and
hydrogen (see S7a†). Aer isolation, they were tested under
similar conditions and exhibited comparable activity and selec-
tivity to that of in situ generated ones. Due to their physical
properties, the Co NPs could be magnetically separated and were
conveniently re-used up to three times (Fig. 2). However, aer the
third cycle we observed a signicant decrease in activity and
selectivity. In addition, the stability of the catalyst system was also
conrmed by recycling the NPs aer reduced reaction time
(Fig. S4†). Next, we compared the reactivity of these active NPs
with related supported NPs. However, addition of carbon or silica
support to the reaction led to completely inactive materials
(Fig. 2). On the other hand, materials prepared by immobilization
of complex I on carbon or silica and subsequent pyrolysis
produced catalysts with moderate activity (Fig. 2; 40–50% yield of
2). In addition, specic cobalt nanoparticles have been prepared
by using chemical reduction of cobalt salts11 and tested for their
activities. However, none of these cobalt nanoparticles formed
the desired product, 4-bromobenzylamine (Table S1,† entries 5–
6). All these results reveal the superiority of the simply in situ
generated Co NPs (Fig. 3).
Results and discussion
In situ generation of Co-NPs and their activities
Following our concept, we initially investigated the reaction of
cobalt salen complexes to obtain NPs. For example, using the
cobalt–N,N0-bis(salicylidene)-1,2-phenylenediamine (complex I)
in water–THF as solvent in the presence of ammonia and
ꢀ
molecular hydrogen at 120 C a black precipitate of Co NPs is
formed, which can be magnetically separated (Fig. 1 and S3†). To
explore their reactivity, preliminary catalytic experiments were
carried out for the reductive amination of 4-bromobenzaldehyde
1 to 4-bromobenzylamine 2 in presence of ammonia and
molecular hydrogen (Fig. 2). Indeed, using a mixture of cobalt(II)
acetate and N,N0-bis(salicylidene)-1,2-phenylenediamine (L1) led
to the formation of 15% of 2. In contrast, testing simple cobalt(II)
acetate under the same conditions produced no desired product.
Remarkably, the dened complex Co-L1 (complex I) exhibited
excellent activity as well as selectivity in the bench mark reaction
(98% of 4-bromobenzylamine). In addition, other molecularly-
dened Co–salen complexes have also been tested (Fig. S1†)
and complexes II–IV showed good activity (85–90% yield), while
complex V resulted in lower product yield (50%). In all cases of
active complexes, the reaction mixtures turned black aer some
hours. Hence, we assumed the in situ formed cobalt-NPs are the
Characterization of in situ Co-NPs
To understand the reactivity and to know the structural features
of the most active cobalt nanoparticles, we performed detailed
Fig. 2 Reductive amination of 4-bromobenzaldehyde: Activity of
a
cobalt catalystsa. Reaction conditions: 0.5 mmol 4-bromobenzalde-
hyde, 6 mol% Co-complex, (Co NPs), 5 bar NH3, 45 bar H2, 2.5 mL Fig. 3 Reductive amination of carbonyl compounds in presence of
H2O–THF (1.5 : 1), 120 ꢀC, 24 h, GC yields using n-hexadecane as NH3 and H2 using different Co NPs produced from cobalt–salen
standard.
complex.
Chem. Sci.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Edge Article
Chemical Science
characterizations using transmission electron microscopy C]O, and O]C–O type bonds with individual atomic% of
(TEM), energy dispersive X-ray spectroscopy (EDX), X-ray 64.09, 19.30, 8.59, 3.20 and 4.82 respectively, showing the
diffraction (XRD), and X-ray photoelectron spectroscopy (XPS). graphitic nature of the carbon material (Fig. 5a). The presence
TEM analysis of cobalt-particles at different magnication of a specic N1s peak at 399.7 conrms pyrrolic nitrogen
showed sheets and at some places thread bundles like (Fig. 5b). The three peak components at 529.5, 531.4, and
morphology where cobalt nanoparticles are embedded in 532.9 eV in O1s spectra originate from the presence of Co(OH)2
carbon and nitrogen framework (Fig. 4). Further detailed (6.99%), C]O (73.09%), and O–C on the surface of cobalt
morphological investigations were performed by HRTEM-STEM (19.92%). This reveals partial oxidation at the surface of the
analysis. A close inspection of HRTEM images at 20 nm, optimal material (Fig. 5c). In agreement, the two main compo-
revealed the presence of ultra-small (range 2–4 nm) cobalt nent peaks having binding energy at 780.7 eV (50.81%) and
nanoparticles (Fig. 4 and S6†) supported on graphitic carbon. 782.5 eV (22.54%) conrm the presence of Co2+ (Co(OH)2)
The HAADF-elemental mapping displayed a homogeneous (Fig. 5d).12a Three small peaks having binding energies at 778.09
distribution of the cobalt nanoparticles (Fig. 4). In case of the (2.45%), 781.09 (0.59%) and 783.09 (0.35%) eV indicate the
recycled catalyst, we observed that these particles were still presence of metallic cobalt.12b
intact and there are no noticeable changes in the morphology
The HR-XPS of reused catalysts revealed that there is no
(Fig. S8†).
shiing of binding energy in Co 2p3/2 peak but the ratio of
XRD patterns of in situ generated and reused Co- metallic cobalt vs. cobalt hydroxide was slightly changed
nanoparticles do not show variations on the phase composi- (Fig. S11d†). On the other hand, no perceptible change in the
tion (Fig. S9†). Two allotropes of metallic cobalt have been binding energies of C1s and N1s was discerned except for the
identied, one with face centered cubic arrangement (Co-fcc, slight shiing of the pyrrolic nitrogen peak from 399.7 eV to
ꢀ
space group Fm3m, PDF card 01-089-7093), and the other one 400.2 eV, thus reiterating no apparent alteration in the chemical
with hexagonal closed packing (Co-hcp, space group P63/mmc, nature of the carbon shell of the catalyst (Fig. S11a and b†).
PDF card 01-089-7373). Elemental analysis of the bulk material
It is interesting to note that these cobalt-particles exhibit
showed 96.8 wt% of Co, 0.15 wt% of C and only 0.5 wt% of N. ferromagnetic behaviour with distinct values of coercivity eld
Complementary, XPS analysis displayed the presence of larger and remanent magnetization (Fig. S12†). Ferromagnetic
amounts of C, N, and O (C ¼ 34.4, N ¼ 1.2, O ¼ 47.24 and Co ¼ behaviour at room temperature is due to the stronger effect of
16.8 at%) on the surface (Fig. S10†). The high resolution XP the magnetic dipole interaction compared with thermal uc-
spectra of NPs in C1s region can be deconvoluted into ve peak tuations. We do not observe any blocking temperature sug-
components with binding energies of 284.6, 285.6, 286.5, 288.5 gesting the size of nanoparticles above 10–15 nm (e.g. the
and 289.4 eV corresponding to C–C sp2, C–C sp3, C–O/C–N, and system is not superparamagnetic at room temperature).
Fig. 4 TEM images of in situ generated Co-NPs from complex I. (a and b) HRTEM images of cobalt catalyst, (C) magnified STEM image, (d–g)
elemental mapping images where C, N, O and Co are in blue, yellow, green and red colours.
This journal is © The Royal Society of Chemistry 2020
Chem. Sci.

View Article Online
Chemical Science
Edge Article
Fig. 5 HR-XPS spectra of in situ generated Co NPs.
to 92% yield. In addition, separately prepared Co NPs from
complex I, gave similar yields of amines to those obtained by in
situ generated nanoparticles.
Synthesis of linear primary amines
Then, we tested the general applicability of our in situ generated
nanoparticles for the synthesis of primary amines. As shown in
Schemes 1 and 2, a variety of structurally diverse and func-
tionalized benzylic, heterocyclic and aliphatic linear and
branched primary amines can be prepared in good to excellent
yields. Simple and substituted aldehydes underwent smooth
reaction to give primary benzylic amines in up to 92% yield
(Scheme 1, products 3–7). For example, uoro-, chloro-, and
bromo-substituted benzaldehydes produced corresponding
amines without signicant dehalogenations in 86–92% yields
(Scheme 1, entries 8–12). Different functionalized benzylic
amines containing methoxy, triuoromethoxy, dimethylamino,
and ester groups as well as C–C double bonds were synthesized
in up to 95% yield (Scheme 1, products 14–23). In addition to
benzylic amines, primary aliphatic ones were also prepared
under similar conditions (Scheme 1, products 25–27). Interest-
ingly, the natural product perillaldehyde was successfully ami-
nated to produce the corresponding amine in 87% yield
(product 27).
Synthesis of secondary and tertiary amines
Apart from primary amines synthesis, we explored the appli-
cability of Co-NPs for the synthesis of secondary and tertiary
amines. Interestingly, testing complex I which generates the
active NPs vide supra for the reaction of benzaldehyde and
aniline at 120 ꢀC in presence of molecular hydrogen (40 bar) led
to the formation of imine (N-benzylideneaniline) as the sole
product. Under these conditions no nanoparticles could be
isolated aer the reaction.
Apparently, the presence of both ammonia and hydrogen are
required for the generation of the active NPs! Indeed, using iso-
lated Co NPs, which were prepared from complex I, ammonia and
hydrogen, led to excellent activity and selectivity for the synthesis
of secondary and tertiary amines including N-methyl amines
(Scheme 3). As representative examples different benzaldehydes
were reacted with substituted anilines and the corresponding N-
benzylanilines were obtained in 87–98% yields (Scheme 1; prod-
ucts 41–45). Similarly, reactions of different benzaldehydes with
benzylic and aliphatic amines produced selectively the corre-
Synthesis of branched primary amines
Next, we tested the reductive amination of ketones (Scheme 2), sponding secondary and tertiary amines (Scheme 3; products 46–
which is more challenging compared to aldehydes.
Nevertheless, at higher temperature (130 C) nine aromatic went reductive amination and gave the corresponding secondary
and six aliphatic branched primary amines were prepared in up amines (Scheme 3, products 56–57). Finally, N,N0-dimethylamines
55). In addition, aliphatic aldehydes and 4-uoroaniline under-
ꢀ
Chem. Sci.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Edge Article
Chemical Science
Scheme 1 In situ generated Co-nanoparticles catalyzed synthesis of
linear primary amines from aldehydesa. aReaction conditions:
0.5 mmol aldehyde, 6 mol% complex I (22 mg), 5–7 bar NH3, 45 bar H2,
2.5 mL H2O–THF (1.5 : 1), 120 ꢀC, 24h, isolated yields. b Same as ‘a’ at
c
d
130 ꢀC. Same as ‘a’ in 2.5 mL H2O. same as ‘a’ using prepared and
isolated Co-NPs from complex I (2 mg; 6.5 mol% Co).
Scheme 3 Synthesis of secondary, tertiary and N-methyl amines using
Co-nanoparticles prepared from complex Ia. aReaction conditions:
0.6 mmol aldehyde, 0.5 mmol amine, 2 mg Co-NPS (6.5 mol% Co), 45
bar H2, 2.5 mL H2O, 130 ꢀC, 20 h, isolated yields. bSame as ‘a’ using
1 mL aq. N,N0-dimethylamine instead of amine.
carbonyl compounds in 5–10 g scale (Scheme 4). As expected, all
the tested reactions could be successfully upscaled and the
yields (92–96%) of the corresponding primary amines were
comparable to that of small scale (0.5 mmol) reactions.
Experimental
General considerations
All substrates were obtained commercially from various chem-
ical companies and their purity has been checked before use.
Cobalt(II) acetate tetrahydrate (cat no. 208396–50G), salicy-
laldehyde, phenylenediamine and other ligand precursors were
purchased from Sigma-Aldrich. Silica (silicon(IV) oxide,
Scheme 2 Synthesis of branched primary amines from ketones using
in situ generated Co-nanoparticlesa. aReaction conditions: 0.5 mmol
ketone, 6 mol% complex I (22 mg) 5–7 bar NH3, 45 bar H2, 2.5 mL H2O,
b
130 ꢀC, 24 h, isolated yields. Same as ‘a’ in H2O–THF (1.5 : 1 ratio).
cUsing prepared and isolated Co-NPs from complex I (2 mg; 6.5 mol%
Co).
were also prepared from three different aldehydes and aqueous
N,N0-dimethyl amine (Scheme 3, products 58–60).
Scheme 4 Gram-scale synthesis of selected primary amines using in
a
situ generated Co-NPs
.
aReaction conditions: 5–10 g of carbonyl
Reaction upscaling
compound, weight of complex I corresponds to 6 mol%, 5–7 bar NH3,
45 bar H2, 75–150 mL H2O–THF (1 : 1), 120 ꢀC, 24h, isolated yields.
Same as ‘a’ at 130 ꢀC in 75 mL H2O.
In order to demonstrate the synthetic utility of this novel
reductive amination protocol, we performed the amination of 5
b
This journal is © The Royal Society of Chemistry 2020
Chem. Sci.

View Article Online
Chemical Science
Edge Article
amorphous fumed, S.A. 300–350 m2 gꢁ1) was obtained from Alfa 0.25 mm. 1H, 13C, NMR data were recorded on a Bruker ARX 300
Aesar. Carbon powder, VULCAN® XC72R with Code XVC72R and Bruker ARX 400 spectrometers using DMSO-d6 and CDCl3
and CAS No. 1333-86-4 was obtained from Cabot Corporation solvents.
Prod. The pyrolysis experiments were carried out in a Nytech-
Qex oven.
All catalytic reactions were carried out in 300 mL and 100 mL
autoclaves (PARR Instrument Company). In order to avoid
X-ray diffraction patterns were recorded with an Empyrean unspecic reactions, catalytic reactions were carried out either
(PANalytical, The Netherlands) diffractometer in the Bragg– in glass vials, which were placed inside the autoclave, or glass/
Brentano geometry, Co-Ka radiation (40 kV, 30 mA, l ¼ 0.1789 Teon vessel tted autoclaves.
nm) equipped with a PIXcel3D detector (1D mode) and
programmable divergence and diffracted beam anti-scatter
slits. The measurement range was 2q: 5–105ꢀ, with a step size
of 0.026ꢀ. The identication of crystalline phases was per-
formed using the High Score Plus soware (PANalytical) that
includes the PDF-4+ database.
TEM images were obtained using a HRTEM TITAN 60-300
with X-FEG type emission gun, operating at 80 kV. This micro-
scope is equipped with a Cs image corrector and a STEM high-
angle annular dark-eld detector (HAADF). The point resolution
is 0.06 nm in TEM mode. The elemental mappings were ob-
tained by STEM-Energy Dispersive X-ray Spectroscopy (EDS)
with an acquisition time of 20 min. For HRTEM analysis, the
powder samples were dispersed in ethanol and ultrasonicated
for 5 min. One drop of this solution was placed on a copper grid
with holey carbon lm.
Preparation of Co–salen complexes (see scheme S1†)
(a) Preparation of salen ligand (L1). Salicylaldehyde (4 mmol;
in 15 mL ethanol), 1,2-phenylenediamine (2 mmol; in 10 mL
ethanol) were separately dissolved in ethanol. Then, the etha-
nolic solution of 1,2-phenylenediamine was slowly added to
salicylaldehyde solution. The resulting reaction mixture was
reuxed at 80 ꢀC for 8 h to obtain a solid compound. The
reaction mixture was cooled to room temperature and the
product was isolated by ltration. Then, the solid product was
washed with 30 mL of cold ethanol twice and dried in vacuo to
get corresponding salen ligand (L1) in 98% yield. Other salen
ligands were prepared by using similar method.
(b) Preparation of Co–salen complex (complex I).13 1 g of
Co(OAc)2$4H2O, (4 mmol; in 15 mL ethanol) and 1.28 g of N,N0-
bis(salicylidene)-1,2-phenylenediamine (ligand L1) (4 mmol; in
20 mL ethanol) were separately dissolved in ethanol. The cobalt
acetate solution was slowly added to the solution of ligand. The
XPS surface investigation has been performed on the PHI
5000 VersaProbe II XPS system (Physical Electronics) with
monochromatic Al-Ka source (15 kV, 50 W) and photon energy
of 1486.7 eV. Dual beam charge compensation was used for all
measurements. All the spectra were measured in the vacuum of
ꢀ
resulting red suspension was reuxed for 18 h at 80 C to give
a
reddish-brown solid compound. The obtained solid
compound was ltered. Then the obtained solid compound was
washed with 10 mL of cold ethanol and dried in vacuo to get
corresponding cobalt salen complex I in 94% yield. The same
procedure was applied to prepare other cobalt salen complexes
using different salen ligands.
1.3 ꢂ 10ꢁ7 Pa and at room temperature of 21 C. The analyzed
ꢀ
area on each sample was a spot of 200 mm in diameter. The
survey spectra were measured with pass energy of 187.850 eV
and electronvolt step of 0.8 eV while for the high resolution
spectra was used pass energy of 23.500 eV and electronvolt step
of 0.2 eV. The spectra were evaluated with the MultiPak (Ulvac –
PHI, Inc.) soware. All binding energy (BE) values were refer-
enced to the carbon peak C 1s at 284.80 eV.
Procedure for reductive amination
(a) Procedure for the synthesis of primary amines. The
Magnetic properties of cobalt-nanoparticles were analyzed magnetic stirring bar, 0.5 mmol of the carbonyl compound
using a Quantum Design Physical Properties Measurement (aldehyde or ketone) and 22 mg complex I (in case of in situ
System (PPMS Dynacool system) with the vibrating sample generated Co NPs) or 2 mg of prepared and isolated Co NPs were
magnetometer (VSM) option. The experimental data were cor- transferred to an 8 mL glass vial. Then, 2 mL of solvent (water or
rected for the diamagnetism and signal of the sample holder. THF/H2O (1.5 : 1)) was added and the vial was tted with
The temperature dependence of the magnetization was recor- septum, cap and needle. The reaction vials (8 vials with different
ded in a sweep mode of 1 K minꢁ1 in the zero-eld-cooled (ZFC) substrates at a time) were placed into a 300 mL autoclave. The
and eld-cooled (FC) measuring regimes. To get the ZFC autoclave was ushed with hydrogen twice at 40 bar pressure
magnetization curve, the sample was rstly cooled down from and then it was pressurized with 5–7 bar ammonia and 45 bar
300 to 5 K in a presence of zero magnetic eld and the hydrogen. The autoclave was placed into an aluminium block
ꢀ
ꢀ
measurement was carried out on warming from 5 to 300 K preheated at 135 C in case of aldehydes and 145 C in case of
under the external magnetic eld (1000 Oe). In the case of the ketones and the reactions were stirred for the required time.
FC magnetization measurements, the sample was cooled from During the reaction, the inside temperature of the autoclave was
300 to 5 K in an external magnetic eld (1000 Oe) and the measured to be 120 ꢀC in case of aldehydes and 130 ꢀC in case of
measurement was carried out on warming from 5 to 300 K at the ketones and this temperature was used as the reaction
same value of the external magnetic eld (1000 Oe). Hysteresis temperature. Aer completion of the reactions, the autoclave
loops were measured at room temperature (300 K) and at low was cooled to room temperature. The remaining ammonia and
temperature (5 K).
hydrogen were discharged and the vials containing reaction
GC conversion and yields were determined by GC-FID, products were removed from the autoclave. The reaction
HP6890 with FID detector, column HP530 m ꢂ 250 mm ꢂ mixtures containing the products were ltered off and washed
Chem. Sci.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Edge Article
Chemical Science
thoroughly with ethanol. The reaction products were analyzed ammonia gas and 45 bar hydrogen. The autoclave was placed
by GC-MS. The crude product was puried by column chro- into the heating system and the reaction was allowed to prog-
matography using ethyl acetate and n-heptane as the eluent. ress at 120 ꢀC (temperature inside the autoclave) by stirring for
The corresponding primary amines were converted to their 24 h. Aer completion of the reaction, the autoclave was cooled
respective hydrochloride salt and characterized by NMR and and the remaining ammonia and hydrogen pressure was dis-
GC-MS analysis. For converting into hydrochloride salt of charged. To the reaction products, 200 mL n-hexadecane as
amine, 0.3–0.5 mL 7 M HCl in dioxane or 1.5 M HCl in methanol standard was added. The catalyst was then separated by
was added to the dioxane solution of respective amine and centrifugation and the supernatant containing the reaction
stirred at room temperature for 4–5 h. Then, the solvent was products was subjected to GC analysis for determining
removed and the resulted hydrochloride salt of amine was dried conversion and yield. The separated catalyst was then washed
under high vacuum. The yields were determined by GC for the with ethanol, dried under vacuum and used without further
selected amines: aer completing the reaction, n-hexadecane purication or reactivation for the next run.
(100 mL) as standard was added to the reaction vials and the
reaction products were diluted with ethyl acetate followed by
ltration using plug of silica and then analyzed by GC.
Gram-scale reactions
The Teon or glass tted 300 mL autoclave was charged with
a magnetic stirring bar and 5–10 g of carbonyl compound
(aldehyde or ketone) and complex I (weight of complex I
corresponds 6 mol%). Then, 75–150 mL of solvent (THF/H2O
(1.5 : 1) in case of aldehyde and H2O in case of ketones) was
added and the autoclave was ushed with hydrogen twice at 40
bar pressure. Aerwards, it was pressurized with 5–7 bar
ammonia and 45 bar hydrogen. The autoclave was placed into
(b) Procedure for the synthesis of secondary and tertiary
amines. The magnetic stir bar, 0.5 mmol of amine and
0.6 mmol aldehyde were transferred to an 8 mL glass vial. Then
2 mg of Co NPs and 2 mL of water as solvent were added. The
vial was tted with septum, cap and needle. The reaction vials (8
vials with different substrates at a time) were placed into
a 300 mL autoclave. The autoclave was ushed with hydrogen
twice at 40 bar pressure and then it was pressurized 45 bar
hydrogen. The autoclave was placed into an aluminium block
preheated at 145 ꢀC and the reactions were stirred for the
required time. During the reaction, the inside temperature of
ꢀ
an aluminium block preheated at 135 C in case of aldehydes
and 145 ꢀC in case of ketones and the reactions were stirred for
the required time. During the reaction, the inside temperature
of the autoclave was measured to be 120 ꢀC in case of aldehydes
and 130 ꢀC in case of ketones and this temperature was used as
the reaction temperature. Aer completion of the reaction, the
autoclave was cooled to room temperature and the remaining
ammonia and hydrogen were discharged. The reaction mixtures
containing the products were ltered off and washed thor-
oughly with ethanol. The reaction products were analysed by
GC-MS and the crude primary amine product was puried by
column chromatography using ethyl acetate and n-heptane as
the eluent.
ꢀ
the autoclave was measured to be 130 C and this temperature
was used as the reaction temperature. Aer completion of the
reactions, the autoclave was cooled to room temperature. The
remaining hydrogen was discharged and the vials containing
the reaction products were removed from the autoclave. The
reaction mixtures containing the products were ltered off and
washed thoroughly with ethanol. The reaction products were
analyzed by GC-MS. The crude product was puried by column
chromatography using ethyl acetate and n-heptane as the
eluent. The corresponding amines were characterized by NMR
and GC-MS analysis.
Preparation of cobalt-nanoparticles
(a) Preparation of Co NPs from complex I. A magnetic stir bar
and 1.0 g of cobalt–salen complex (complex I) were transferred
to a glass tted 100 mL autoclave. Then, 20 mL THF–water
(1 : 1) was added. The autoclave was ushed with 40 bar
hydrogen and then it was pressurized with 5–7 bar ammonia
and 45 bar hydrogen. The autoclave was placed into the heating
system and the reaction was allowed to progress at 120 ꢀC
(temperature inside the autoclave) by stirring for 24 h. Aer 24 h
of reaction time, the autoclave was removed from the heating
system and cooled to room temperature. The remaining
ammonia and hydrogen pressure was discharged. The cobalt
nanoparticles formed were separated from solution by using the
magnetic stir bar. Then, the nanoparticles were separated from
the magnetic stir bar and washed with water and ethanol.
Isolation of in situ generated cobalt nanoparticles
Aer the completion of the reductive amination reaction of
carbonyl compound in presence of ammonia and hydrogen as
described in Section S3a,† the in situ generated cobalt nano-
particles from the solution containing products were separated
using the magnetic stir bar. Then, they were separated from the
magnetic stir bar and washed with water and ethanol. Finally
the recycled Co NPs were dried under vacuum and stored in
a glass vial.
Elemental analysis (wt%): Co ¼ 96.8% C ¼ 0.15% and N ¼
0.5%.
Recycling of in situ generated cobalt-nanoparticles
A magnetic stirring bar and 5 mmol 4-bromobenzaldehyde were Finally obtained nanoparticles were dried under vacuum and
transferred to a glass tted 100 mL autoclave and then 20 mL stored in a glass vial.
THF–water (1.5 : 1) was added. Subsequently, 20 mg isolated in
(b) Preparation of carbon and silica supported Co-nano-
situ generated Co NPs were added. The autoclave was ushed particles. In a 50 mL round bottomed ask, cobalt salen
with 40 bar hydrogen and then it was pressurized with 5–7 bar complex (316.8 mg) and 25 mL of ethanol were reuxed at 80 ꢀC
This journal is © The Royal Society of Chemistry 2020
Chem. Sci.

View Article Online
Chemical Science
Edge Article
for 15 minutes. To this, 700 mg of Vulcan XC 72R carbon analytical team of the Leibniz-Institute for Catalysis, Rostock for
powder or SiO2 was added and then the whole reaction mixture their excellent service. We gratefully thank Ondrej Tomanec,
was reuxed at 80 ꢀC for 4–5 h. The reaction mixture was cooled and Martin Petr (both from Palacky University) for HRTEM
to room temperature and the ethanol was removed in vacuum. elemental mapping and XPS data respectively. The authors
The solid sample obtained was dried in high vacuum, aer gratefully acknowledge the support by the Operational Pro-
which it was grinded to a ne powder. Then, the grinded gramme Research, Development and Education – European
powder was pyrolyzed at 800 ꢀC for 2 hours under an argon Regional Development Fund, projects no. CZ.02.1.01/0.0/0.0/
atmosphere and cooled to room temperature.
16_019/0000754 and CZ.02.1.01/0.0/0.0/15_003/0000416 of the
(c) Preparation of other cobalt nanoparticles reported in Ministry of Education, Youth and Sports of the Czech Republic.
literature.11 Method-I: 1.0 g of cobalt acetate and 1.5 mL of oleic
acid were mixed in 40 mL of diphenyl ether (DPE) and the
reaction mixture was heated to 200 ꢀC under N2 atmosphere.
Notes and references
Then, 1.0 mL of TOP (trioctylphosphine) was added and the
mixture was again heated to 250 ꢀC. Subsequently, the reducing
solvent such as 4.0 g of 1,2-dodecanediol dissolved in 10 mL
DPE at 80 ꢀC, was injected into the reaction mixture. Then
whole reaction mixture was held at 250 ꢀC for 30 min until the
completion of the reduction. The reaction products were cooled
to room temperature and ethanol was added to precipitate
nanoparticles. The formed cobalt nanoparticles were separated
by centrifugation and were nally dried and stored in glass vial.
Method-II: 2 mmol of cobalt acetate and 0.4 mmol of oleic
acid were mixed in 1 mL ethanol. Then, 3 mmol of NaBH4
dissolved in 1 mL of ethanol, was slowly added to the above
mixture at room temperature under stirred condition. The
whole mixture was stirred at room temperature for 4 h. The
formed cobalt nanoparticles were separated by centrifugation
and nally dried and stored in glass vial.
1 (a) D. Wang and D. Astruc, Chem. Soc. Rev., 2017, 46, 816–
854; (b) L. Liu and A. Corma, Chem. Rev., 2018, 8, 4981–
5079; (c) X. Cui, X. Dai, Y. Deng and F. Shi, Chem.–Eur. J.,
2013, 19, 3665–3675; (d) R. V. Jagadeesh, A.-E. Surkus,
H. Junge, M.-M. Pohl, J. Radnik, J. Rabeah, H. Huan,
¨
¨
V. Schunemann, A. Bruckner and M. Beller, Science, 2013,
342, 1073–1076; (e) R. V. Jagadeesh, K. Murugesan,
A. S. Alshammari, H. Neumann, M.-M. Pohl, J. Radnik and
M. Beller, Science, 2017, 358, 326–332; (f) L. He, F. Weniger,
H. Neumann and M. Beller, Angew. Chem., Int. Ed., 2016,
55, 12582–12594; (g) T. Schwob and R. Kempe, Angew.
Chem., Int. Ed., 2016, 55, 15175–15179; (h) G. Hahn,
P. Kunnas, N. de Jonge and R. Kempe, Nat. Catal., 2018, 2,
71–77; (i) K. Murugesan, M. Beller and R. V. Jagadeesh,
Angew. Chem., Int. Ed., 2019, 58, 5064–5068; (j)
K. Murugesan, T. Senthamarai, A. S. Alshammari,
R. M. Altamimi, C. Kreyenschulte, M.-M. Pohl, H. Lund,
R. V. Jagadeesh and M. Beller, ACS Catal., 2019, 9, 8581–
8591; (k) R. V. Jagadeesh, T. Stemmler, A.-E. Surkus,
M. Bauer, M.-M. Pohl, J. Radnik, K. Junge, H. Junge,
Conclusions
In conclusion, we demonstrated that the in situ formation of
cobalt nanoparticles from molecularly dened precursors is
straightforward and convenient. Such approach can be used as
a versatile tool to prepare selective and active heterogeneous
catalysts. In our specic case, Co NPs are formed from cobalt–
salen complexes (e.g. cobalt(II)–N,N0-bis(salicylidene)-1,2-
phenylenediamine) in the presence of ammonia and
hydrogen. Thereby, well-dened ultra-small metallic cobalt and
cobalt hydroxide nanoparticles embedded in a cobalt–nitrogen
framework are formed. The resulting NPs are stable in the
presence of air and water and allow for the preparation of
various functionalized and structurally diverse linear and
branched benzylic, heterocyclic and aliphatic primary amines
as well as secondary and tertiary amines. Moreover, they can be
easily magnetically separated enabling easy catalyst recycling
and product purication.
¨
A. Bruckner and M. Beller, Nat. Protoc., 2015, 10, 916–926;
(l) T. Schwob, P. Kunnas, N. de Jonge, C. Papp,
¨
H. P. Steinruck and R. Kempe, Sci. Adv., 2019, 5, eaav3680;
(m) T. Schwob, M. Ade and R. Kempe, ChemSusChem, 2019,
12, 3013–3017.
2 (a) G. W. Parshall and S. D. Ittel, Homogeneous Catalysis: The
Applications and Chemistry of Catalysis by Soluble Transition
Metal Complexes, Weily, 1992; (b) P. W. N. M. van Leeuwen
and J. C. Chadwick, Homogeneous Catalysts: Activity –
Stability – Deactivation, Wiley-VCH, 2011; (c) B. Cornils,
W. A. Herrmann, M. Beller and R. Paciello, Applied
Homogeneous Catalysis with Organometallic Compounds,
Wiley-VCH, 2017; (d) B. A. Averill, J. A. Moulijn, R. A. van
Santen and P. W. N. M. van Leeuwen, Catalysis: An
integrated approach, Elsevier, 1997.
3 (a) L. Filipponi and D. Sutherland, Nanotechnologies:
Principles, Applications, Implications and Hands-on Activities,
European Commission, European Union, 2012; (b)
M. B. Gawande, S. P. Branco and R. S. Varma, Chem. Soc.
Rev., 2013, 42, 3371–3393; (c) M. B. Gawande, A. Goswami,
T. Asefa, H. Guo, A. V. Biradar, D. L. Peng, R. Zboril and
R. S. Varma, Chem. Soc. Rev., 2015, 44, 7540–7590; (d)
P. Munnik, P. E. De Jongh and K. P. De Jong, Chem. Rev.,
2015, 115, 6687–6718; (e) M. Sankar, N. Dimitratos,
P. J. Miedziak, P. P. Wells, C. J. Kielye and G. J. Hutchings,
Conflicts of interest
There are no conicts to declare.
Acknowledgements
We gratefully acknowledge the European Research Council (EU
project 670986-NoNaCat), and the State of Mecklenburg-
Vorpommern for nancial and general support. We thank the
Chem. Sci.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Edge Article
Chemical Science
Chem. Soc. Rev., 2012, 41, 8099–8139; (f) F. Tao, Metal
Nanoparticles for Catalysis: Advances and Applications, Royal
Society of Chemistry, 2014; (g) E. M. van Schrojenstein
Lantman, T. Deckert-Gaudig, A. J. G. Mank, V. Deckert and
B. M. Weckhuysen, Nat. Nanotechnol., 2012, 7, 583–586; (h)
J. J. H. B. Sattler, J. Ruiz-Martinez, E. Santillan-Jimenez and
B. M. Weckhuysen, Chem. Rev., 2014, 114, 10613–10653; (i)
A. Balanta, C. Godard and C. Claver, Chem. Soc. Rev., 2011,
40, 4973–4985.
4 (a) S. Dang, Q.-L. Zhu and Q. Xu, Nat. Rev. Mater., 2017, 3,
17075; (b) J. Tang and Y. Yamauchi, Nat. Chem., 2016, 8,
638–639; (c) K. Shen, X. Chen, J. Chen and Y. Li, ACS
Catal., 2016, 6, 5887–5903.
metal; (f) K. N. Gusak, Z. V. Ignatovich and E. V. Koroleva,
Russ. Chem. Rev., 2015, 84, 288–309.
8 (a) Y. Nakamura, K. Kon, A. S. Touchy, K.-I. Shimizu and
W. Ueda, ChemCatChem, 2015, 7, 921–924; (b) G. Liang,
A. Wang, L. Li, G. Xu, N. Yan and T. Zhang, Angew. Chem.,
Int. Ed., 2017, 56, 3050–3054; (c) T. Komanoya,
T. Kinemura, Y. Kita, Y. K. Kamata and M. Hara, J. Am.
Chem. Soc., 2017, 139, 11493–11499; (d) M. Chatterjee,
T. Ishizakaa and H. Kawanami, Green Chem., 2016, 18,
487–496.
9 (a) T. Gross, A. M. Seayad, M. Ahmad and M. Beller, Org. Lett.,
2002, 4, 2055–2058; (b) T. Riermeier, K.-J. Haack,
¨
U. Dingerdissen, A. Borner, V. Tararov and R. Kadyrov, US
5 (a) N. Yan, Y. Yuan and P. J. Dyson, Dalton Trans., 2013, 42,
13294–13304; (b) Da. Cantillo, M. Baghbanzadeh and
C. O. Kappe, Angew. Chem., Int. Ed., 2012, 51, 10190–10193;
(c) J. Holz, C. Pfeffer, H. Zuo, D. Beierlein, G. Richter,
E. Klemm and R. Peters, Angew. Chem., Int. Ed., 2019, 58,
10330–10334; (d) L. Zadoina, K. Soulantica, S. Ferrere,
B. Lonetti, M. Respaud, A. F. Mingotaud, A. Falqui,
A. Genovese, B. Chaudret and M. Mauzac, J. Mater. Chem.,
2011, 21, 6988–6994.
6 (a) A. Ricci, Amino group chemistry: From synthesis to the life
sciences, Wiley-VCH, 2008; (b) Top 200 drugs production,
National Science Foundation, J Chem. Educ., 2010, 87,
1348; (c) S. A. Lawrence, Amines: synthesis, properties and
Pat., 6884887B1, 2005; (c) J. Gallardo-Donaire, M. Ernst,
O. Trapp and T. Schaub, Adv. Synth. Catal., 2016, 358, 358–
363; (d) J. Gallardo-Donaire, M. H. Wysocki, M. Ernst,
F. Rominger, O. Trapp, A. Stephen, L. Hashmi, A. Schaefer,
P. Comba and T. Schaub, J. Am. Chem. Soc., 2018, 140, 355–
361; (e) S. Ogo, K. Uehara, T. Abura and S. Fukuzumi, J.
Am. Chem. Soc., 2014, 126, 3020–3021; (f) R. Kadyrov and
T. H. Riermeier, Angew. Chem., Int. Ed., 2003, 42, 5472–
5474; (g) T. Senthamarai, K. Murugesan, J. Schneidewind,
N. V. Kalevaru, W. Baumann, H. Neumann, P. C. J. Kamer,
M. Beller and R. V. Jagadeesh, Nat. Commun., 2018, 9,
4123; (h) X. Tan, S. Gao, W. Zeng, S. Xin, Q. Yin and
X. Zhang, J. Am. Chem. Soc., 2018, 140, 2024–2027.
applications, Cambridge University Press, 2004; (d) F. Shi 10 (a) Z. Wang, “Mignonac reaction” in comprehensive organic
and X. Cui, Catalytic amination for N-alkyl amine synthesis,
Academic Press, 2018; (e) W. R. Meindl, E. V. Angerer,
H. Schoenenberger and G. Ruckdeschel, Med. Chem., 1984,
name reactions and reagents, Wiley, 2010; (b) https://
erowid.org/archive/rhodium/chemistry/
reductive.amination.html, 2004.
27, 1111–1118; (f) V. Froidevaux, C. Negrell, S. Caillol, 11 A. S. Zola, R. U. Ribeiro, J. M. C. Bueno, D. Zanchet and
J.-P. Pascault and B. Boutevin, Chem. Rev., 2016, 116, P. A. Arroyo, J. Exp. Nanosci., 2014, 9, 398–405.
14181–14224; (g) T. Yan, B. L. Feringa and K. Barta, Nat. 12 (a) J. Yang, H. Liu, W. N. Martens and R. L. Frost, J. Phys.
Commun., 2014, 5, 5602.
Chem., 2010, 114, 111–119; (b) C. D. Wagner, L. E. Davis,
J. F. Moulder and G. E. Mullenberg, Handbook of X-ray
Photoelectron Spectroscopy, Minnesota: Perkin-Elmer
Corporation, 1978.
7 (a) S. Gomez, J. A. Peters and T. Maschmeyer, Adv. Synth.
Catal., 2002, 344, 1037–1057; (b) H. Alinezhad, H. Yavari
and F. Salehian, Curr. Org. Chem., 2015, 19, 1021–1049; (c)
T. C. Nugenta and M. El-Shazlya, Adv. Synth. Catal., 2010, 13 (a) A. A. Khandar, B. Shaabani, F. Belaj and A. Bakhtiari,
352, 753–819; (d) K. Natte, H. Neumann, R. V. Jagadeesh
and M. Beller, Nat. Commun., 2017, 8, 1344; (e) https://
reagentguides.com/reagent-guides/reductive-amination/list-
of-reagents/hydrogen-metal-catalysts-precious-and-base-
Inorg. Chim. Acta, 2007, 360, 3255–3264; (b) A. van Den
Bergen, K. S. Murray and B. O. West, J. Organomet. Chem.,
1971, 33, 89–96.
This journal is © The Royal Society of Chemistry 2020
Chem. Sci.