Selective catalytic hydrogenation of nitriles to primary amines using iron pincer complexes

Article abstract of DOI:10.1039/c6cy00834h

The selective catalytic hydrogenation of nitriles to primary amines with the well-defined Fe(PNP^Cy) pincer complex 2 is reported. This iron pincer catalyst shows superior catalytic activity and selectivity in the reduction of various nitriles including industrially relevant adipodinitrile in high yields under relatively mild conditions.

Full text of DOI:10.1039/c6cy00834h

Catalysis
Science &
Technology
View Article Online
View Journal
COMMUNICATION
Selective catalytic hydrogenation of nitriles to
primary amines using iron pincer complexes†
Cite this: DOI: 10.1039/c6cy00834h
S. Lange,a S. Elangovan,a C. Cordes,a A. Spannenberg,a H. Jiao,a H. Junge,a
Received 18th April 2016,
Accepted 19th May 2016
b
b
a
a
a
*
S. Bachmann, M. Scalone, C. Topf, K. Junge and M. Beller
DOI: 10.1039/c6cy00834h
www.rsc.org/catalysis
The selective catalytic hydrogenation of nitriles to primary amines
with the well-defined FeIJPNPCy) pincer complex 2 is reported. This
iron pincer catalyst shows superior catalytic activity and selectivity
in the reduction of various nitriles including industrially relevant
adipodinitrile in high yields under relatively mild conditions.
application of non-noble metal pincer complexes in homoge-
neous reactions is studied very intensively by the catalytic
community.11,12 Regarding the reduction of nitriles, until
now, only two other examples of non-noble metal catalysts
have been reported by the group of Milstein.12f,i Compared to
iron catalyst 1, the Co PNN as well as the Fe PNP pincer com-
plexes seem to be less active, requiring addition of base, lon-
ger reaction time (16–48 h) and higher temperatures (135–140
°C). Thus, there is room for improvement. Herein, we evalu-
ate the influence of alkyl substituents at the phosphorus
binding site on the catalytic performance of the complexes in
the hydrogenation of nitriles.‡ Therefore, we synthesized the
new iron pincer complexes 2 and 3, bearing different substitu-
ents on the phosphine moiety (Scheme 1).13
Primary amines constitute important building blocks in the
agrochemical and pharmaceutical industries as well as for the
manufacture of dyes, pigments, polymers and for gas
treatment.1 Compared to classic organic reactions, such as
alkylation of ammonia and reductive amination, the
reduction of nitriles represents an alternative way to form
primary amines. Usually, stoichiometric amounts of metal
hydride reagents are applied for nitrile reduction with high
efficiency.2 Due to the huge quantities of waste metal salts,
these procedures are connected with economic and ecologic
limitations. Therefore, catalytic transformation of nitriles
with molecular hydrogen offers an atom-economic and envi-
ronmentally benign alternative for the synthesis of amines.3
Although heterogeneous metal catalysts are known for the re-
duction of nitriles, often they suffer from low to moderate se-
lectivity and limited functional group tolerance under the re-
quired drastic conditions. Instead, homogeneous catalysts for
the hydrogenation of nitriles are generally based on precious
metals such as Ru,4 Ir,5 Rh,6 Re7 and Mo.8 Recently, we
developed the first homogeneous iron-based catalyst 1 for the
selective reduction of a large number of aromatic, aliphatic
and heterocyclic nitriles to the corresponding primary amines
requiring no addition of base.9 This iron-pincer complex 1
has been developed parallel in our group and in Schneider's
group, showing extraordinary catalytic activities in various hy-
drogenation and dehydrogenation reactions.10 Presently, the
The synthesis of FeIJPNPEt) complex
3
has been
described,10i while we show here the preparation of FeIJPNPCy
)
2: first, the dibromo complex [(PNPCy)FeIJCO)Br2] 4 is pro-
duced by the treatment of FeBr2·2THF with the PNP ligand,
bis(2-dicyclohexyl
phosphinoethyl)amine
under
an
a Leibniz-Institut für Katalyse e.V. an der Universität Rostock, Albert-Einstein-
Straße 29a, 18059 Rostock, Germany. E-mail: matthias.beller@catalysis.de
b F. Hoffmann-La Roche AG, Process Research & Development, CoE Catalysis,
4070 Basel, Switzerland
† Electronic supplementary information (ESI) available. See DOI: 10.1039/
c6cy00834h
Scheme 1 Structure of different Fe(PNP) catalysts 1–3 and synthesis
of complex 2.
This journal is © The Royal Society of Chemistry 2016
Catal. Sci. Technol.

View Article Online
Communication
Catalysis Science & Technology
atmosphere of CO.14 Complex 4 is isolated as a dark blue
solid compound in 96% yield, showing a singlet at δ = 61.2
ppm in the 31P NMR spectra. In the IR spectra of 4, a strong
CO stretching band at 1943 cm−1 and an N–H vibration at
3117 cm−1 are observed. The hydridoborato complex 2 is pre-
pared in 76% yield by the reaction of complex 4 with an ex-
cess of NaBH4. The bright yellow complex 2 was characterized
by different spectroscopic methods (NMR, IR, MS). In the 31P
NMR spectra of the dihydride complex 2, a mixture of two
isomers was detected at δ = 91.5 ppm (major isomer) and δ =
92.7 ppm (minor isomer).15 Additionally, the 1H NMR spec-
trum shows a sharp triplet at δ = −19.6 ppm (major isomer)
for the hydride ligand, whereas the BH4 ligand resonates as a
broad signal at δ = −2.9 ppm. In the IR spectrum, bands at
1905 cm−1 indicate the coordination of CO to the metal cen-
tre. Furthermore, crystals suitable for X-ray diffraction analy-
sis were obtained from a THF/n-heptane mixture. In the mo-
lecular structure of complex 2 shown in Fig. 1, the central
iron atom is placed in a distorted octahedral coordination ge-
ometry.16 The HBH3 ligand and the N–H bond are arranged
in the cis position. Moreover, the CO ligand and the N atom
as well as the hydride ligand and the HBH3 ligand are located
trans to each other.
The reaction of benzonitrile to benzyl amine was chosen
as a benchmark system to study the influence of structural
variation of the phosphine moiety on the activity of this type
of catalysts. In preliminary experiments, different amounts of
catalyst loadings (0.25–1 mol%) of 1–3 were tested at 30 bar
H2, 3 h, 70 °C (Scheme 2). With 1 mol% catalyst concentra-
tion, all three complexes produced benzyl amine in high
yields, while the catalytic performance of FeIJPNPEt) complex 3
completely dropped down with half the amount of catalyst.
This finding is in contrast to the results for the iron-catalysed
ester reduction, where the FeIJPNPEt) complex 3 is the most
active one of these three catalysts.10i We were pleased that
complex 2 afforded 90% yield of benzyl amine using 0.5
mol% catalyst concentration, 30 bar H2 at 70 °C. Regarding
the influence of the alkyl substituents at the phosphorus
binding site on the catalytic efficiency of the three Fe pincer
complexes 1–3, the following sequence was found: Cy = iPr > Et.
Scheme 2 Yield of benzyl amine using different catalyst loadings of
1–3. Reaction conditions: 1.0 mmol of benzonitrile, x mol% 1–3, 2 mL
of iPrOH, 30 bar H2, 3 h, 70 °C.
A short optimization of the reaction parameters (time, sol-
vent, temperature) applying the new active complex 2 is sum-
marized in Table 1. The temperature and the solvent play a
significant role for the catalyst efficiency. Good to excellent
yields of benzyl amine are produced with 0.5–1 mol% of 2 at
70 °C in 3 h (Table 1, entries 1 and 2). The reaction runs fast,
yielding complete conversion and 86% of amine after 2 h.
Interestingly, at temperatures below 70 °C there is still con-
version of the starting material, but most of the detected
product is the corresponding N-benzyl-1-phenylmethanimine.
When toluene was used as solvent, traces of benzaldehyde
were observed.
Next, we studied the scope of the nitrile reduction with
the FeIJPNPCy) pincer complex 2 to illustrate the generality of
this iron-catalyzed procedure. Therefore, various aromatic, al-
iphatic and heterocyclic nitriles (Table 2 and Scheme 3) were
hydrogenated to their corresponding amines in high yields
under mild conditions (30 bar H2, 70–100 °C, 1 mol% 2).
Here, in general nitriles with electron withdrawing substitu-
ents such as halides or esters (Table 2, entries 2, 4 and 7)
were hydrogenated in good yields. Only in the case of 3,5-
difluorobenzonitrile no further increase of the yield was
reached with 5 mol% catalyst loading (Table 2, entry 4). More
interestingly, the hydrogenation of nitriles selectively took
place even in the presence of an ester group producing an ex-
cellent yield (95%). 4-Phenyl substituted benzonitrile provided
Table 1 Optimization of the reaction conditions for the hydrogenation
of benzonitrilea
Catalyst
[mol%]
Temp.
[°C]
Time
[h]
Conv.b
[%]
Yieldb
[%]
Entry
1
2
3
4
0.5
1
1
1
1
70
70
70
55
40
70
3
3
2
3
3
3
99
99
99
92
44
17
90
89
86
—
—
—
5
6c
1
Fig. 1 Molecular structure of complex 2 in the crystal. Only one of the
two molecules of the asymmetric unit is depicted. Displacement
ellipsoids are drawn at the 30% probability level. Hydrogen atoms
except those on Fe, N and B are omitted for clarity.
a
Conditions: 1.0 mmol of benzonitrile, 1 mol% 2, 2 mL of iPrOH,
b
c
30 bar H2, 2–3 h, 40–70 °C. Determined by GC analysis. 2 mL of
toluene.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Catalysis Science & Technology
Communication
Table 2 Hydrogenation of various (hetero)aromatic nitrilesa
Entry
1c
Nitrile
Amine
Yieldb [%]
95
2
3
84
41d
44g
Scheme 3 Yield of different alkyl amines using 2. Reaction conditions:
1.0 mmol of nitrile, 1 mol% 2, 2 mL of iPrOH, 30 bar H2, 3 h, 70 °C; *
100 °C, ca. 8% of CC hydrogenation.
4d
88
lyst was demonstrated for the transformation of aliphatic ni-
triles and adipodinitrile to the corresponding amines
(Scheme 3). Here, cyclic as well as linear nitriles are smoothly
reduced in good yields under mild conditions (70 °C, 30 bar,
3 h) forming exclusively primary amines. In the case of
cinnamyl nitrile, a higher temperature (100 °C) was needed
in order to obtain full conversion. However, under these con-
ditions small amounts of the C–C double bond were reduced.
It should be pointed out that the synthesis of the industrially
important hexane-1,6-diamine was performed with excellent
selectivity and quantitative yield.
5
93
6d
63f
7e
95
8
9
70f
Finally, in order to understand the different reactivities of
catalysts 1A–3A (1–3 without BH3), B3PW91 density func-
tional theory computations were carried out (see the ESI†). In
our first study of the hydrogenation of nitrile to amine by
using catalyst 1A, an outer-sphere mechanism was identified
based on experimental and computational analyses.9 Here,
the most important feature in this reaction is the reversible
process between the hydrogenation and dehydrogenation of
catalyst 1A and its amido intermediate 1B. In contrast to the
first step from nitrile to imine, the second step from imine to
amine is barrier less, while we compared only the difference in
the first step for 1A–3A (Scheme 4). For hydrogenation of aceto-
nitrile, the free energy barrier is 17.55, 20.05 and 17.74 kcal
mol−1 for catalysts 1A–3A, respectively. The reaction is slightly ex-
ergonic for catalysts 1A (−0.96 kcal mol−1) and 3A (−0.92 kcal
mol−1), but endergonic for catalyst 2A (2.57 kcal mol−1).
Hence, for acetonitrile as an aliphatic nitrile, catalysts 1A and
3A have similar barriers, which are also close to those of cata-
lyst regeneration; therefore catalysts 1A and 3A should have
approximately the same reactivity. In contrast, catalyst 2A has
a higher barrier and should have lower reactivity, so that the
regeneration of 2A should determine the reaction rate. For
benzonitrile as an aromatic nitrile, the free energy barrier is
15.35, 17.80 and 14.65 kcal mol−1 for catalysts 1A–3A, respec-
tively; the reaction is exergonic by 4.10, 0.56 and 4.05 kcal
mol−1, respectively. The computation for both cases shows
that catalysts 1A and 3A should have similar reactivity, while
the reactivity of catalyst 2A should be lower. However, for the
catalytic experiments a different sequence of reactivity was
observed: Cy (2) = iPr (1) > Et (3). Comparing theory and
94
81f
10
11d
a
41
84g
90
Conditions: 1.0 mmol of nitrile, 1 mol% 2, 2 mL of iPrOH, 30 bar
b
c
d
H2, 3 h, 70 °C. Determined by GC analysis. 0.5 mol% 2. 30 bar
H2, 100 °C, 3 h. 1 mol% 3, 30 bar H2, 100 °C, 3 h. Isolated yields
of ammonium salt. 5 mol% 2, 30 bar H2, 100 °C, 3 h.
e
f
g
93% of the corresponding amine. Unfortunately, easily reduc-
ible ketone was not tolerated (Table 2, entry 6).
Regarding the influence of different substitutions in
aminobenzonitrile, a higher yield was obtained for ortho-
substituted derivatives (94%) compared to para-substituted
ones (72%) (Table 2, entries 8 and 9). In addition, nitrogen-
and sulfur-containing heteroaromatic nitriles (Table 2, en-
tries 10 and 11) were hydrogenated to their corresponding
amines with moderate to good yields up to 90%. No reaction
was observed in the case of nitro- and hydroxyl-substituted
benzonitriles, while 2-formylbenzonitrile was reduced to the
corresponding alcohol.
Furthermore, the flexibility of FeIJPNPCy) pincer complex 2
as a suitable and selective homogeneous hydrogenation cata-
This journal is © The Royal Society of Chemistry 2016
Catal. Sci. Technol.

View Article Online
Communication
Catalysis Science & Technology
4 (a) R. A. Grey, G. P. Pez, A. Wallo and J. Corsi, J. Chem. Soc.,
Chem. Commun., 1980, 783; (b) R. A. Grey, G. P. Pez and A.
Wallo, J. Am. Chem. Soc., 1981, 103, 7536; (c) T. Li, I.
Bergner, F. N. Haque, M. Zimmer-De Iuliis, D. Song and
R. H. Morris, Organometallics, 2007, 26, 5940; (d) S.
Enthaler, D. Addis, K. Junge, G. Erre and M. Beller, Chem. –
Eur. J., 2008, 14, 9491; (e) S. Enthaler, K. Junge, D. Addis, G.
Erre and M. Beller, ChemSusChem, 2008, 1, 1006; ( f ) D.
Addis, S. Enthaler, K. Junge, B. Wendt and M. Beller,
Tetrahedron Lett., 2009, 50, 3654; (g) R. Reguillo, M. Grellier,
N. Vautravers, L. Vendier and S. Sabo-Etienne, J. Am. Chem.
Soc., 2010, 132, 7854; (h) C. Gunanathan, M. Hölscher and
W. Leitner, Eur. J. Inorg. Chem., 2011, 3381; (i) S.
Werkmeister, K. Junge, B. Wendt, A. Spannenberg, H. Jiao,
C. Bornschein and M. Beller, Chem. – Eur. J., 2014, 20, 427;
( j) J.-H. Choi and M. H. G. Prechtl, ChemCatChem, 2015, 7,
1023; (k) J. Neumann, C. Bornschein, H. Jiao, K. Junge and
M. Beller, Eur. J. Org. Chem., 2015, 5944.
Scheme
regeneration.
4 Mechanism for nitrile hydrogenation and catalyst
experiment reveals that the optimal conditions should be
able to overcome the barrier of catalyst 2A effectively. Obvi-
ously, the observed differences in reactivity among catalysts
1A–3A are stronger influenced by the lower stability of cata-
lyst 3A then by kinetic effects. This assumption is supported
by the vanished reactivity at low catalyst loading (Scheme 2).
In conclusion, the synthesis and application of the new
iron PNP pincer complex 2 is presented. This catalyst al-
lows for the selective hydrogenation of aromatic, aliphatic
and heterocyclic nitriles including adipodinitrile to pri-
mary amines under mild conditions requiring no additive
(base).
5 K. Junge, B. Wendt, H. Jiao and M. Beller, ChemCatChem,
2014, 6, 2810.
6 T. Yoshida, T. Okano and S. Otsuka, J. Chem. Soc., Chem.
Commun., 1979, 870.
7 K. Rajesh, B. Dudle, O. Blacque and H. Berke, Adv. Synth.
Catal., 2011, 353, 1479.
8 S. Chakraborty and H. Berke, ACS Catal., 2014, 4, 2191.
9 C. Bornschein, S. Werkmeister, B. Wendt, H. Jiao, E.
Alberico, W. Baumann, H. Junge, K. Junge and M. Beller,
Nat. Commun., 2014, 5, 4111.
Acknowledgements
We thank B. Wendt, Dr. C. Fischer, S. Schareina and S.
Buchholz (all at the LIKAT) for their excellent support. This
work was supported by F. Hoffmann-La Roche Ltd.
10 (a) E. Alberico, P. Sponholz, C. Cordes, M. Nielsen, H.-J.
Drexler, H. Junge and M. Beller, Angew. Chem., Int. Ed.,
2013, 52, 14162; (b) I. Koehne, T. J. Schmeier, E. A. Bielinski,
C. J. Pan, P. O. Lagaditis, W. H. Bernskoetter, M. K. Takase,
C. Würtele, N. Hazari and S. Schneider, Inorg. Chem.,
2014, 53, 6066; (c) S. Werkmeister, K. Junge, B. Wendt, E.
Alberico, H. Jiao, W. Baumann, H. Junge, F. Gallou and M.
Beller, Angew. Chem., Int. Ed., 2014, 53, 8722; (d) S.
Chakraborty, H. Dai, P. Bhattacharya, N. T. Fairweather,
M. S. Gibson, J. A. Krause and H. Guan, J. Am. Chem. Soc.,
2014, 136, 7896; (e) S. Chakraborty, W. W. Brennessel and
W. D. Jones, J. Am. Chem. Soc., 2014, 136, 8564; ( f ) E. A.
Bielinski, P. O. Lagaditis, Y. Zhang, B. Q. Mercado, C.
Würtele, W. H. Bernskoetter, N. Hazari and S. Schneider,
J. Am. Chem. Soc., 2014, 136, 10234; (g) M. Pena-Lopez, H.
Neumann and M. Beller, ChemCatChem, 2015, 7, 865; (h) R.
Xu, S. Chakraborty, S. M. Bellows, H. Yuan, T. R. Cundari
and W. D. Jones, ACS Catal., 2016, 6, 2127; (i) S. Elangovan,
B. Wendt, C. Topf, S. Bachmann, M. Scalone, A.
Spannenberg, H. Jiao, W. Baumann, K. Junge and M. Beller,
Adv. Synth. Catal., 2016, 358, 820.
Notes and references
‡ Catalytic procedure: catalytic transformations were performed in a 300 mL au-
toclave equipped with an internal aluminum plate to include seven uniform re-
action glass vials (4 mL) sealed with cap, septum and needle. The autoclave was
placed into an aluminum block as the heating system to perform the reactions.
General procedure: in a reaction vial (4 mL), iron complex 2 (0.01 mmol) was
mixed with 2 mL of iso-propanol (degassed). After a short time of stirring, the
nitrile (1 mmol) was added under argon. The vial was placed into the alloy plate
in the autoclave. The apparatus was purged three times with hydrogen, pressur-
ized to 30 bar H2, and stirred for 3 h at 70 or 100 °C. Afterwards, the autoclave
was cooled to room temperature, depressurized, hexadecane was added as an
internal standard and the reaction mixture was analysed by GC. Isolated HCl
salts were analysed by NMR, GC-MS and HRMS.
1 Amines: Synthesis, Properties and Application, ed. S. A.
Lawrence, Cambridge University Press, Cambridge, 2004.
2 (a) J. Seyden-Penne, Reduction by Alumino- and Borohydrides
in Organic Synthesis, Wiley, New York, 1997; (b) P. G.
Anderson and I. J. Munslow, Modern Reduction Methods,
Wiley-VCH, Weinheim, 2008.
3 For recent reviews highlighting the reduction of nitriles, see:
(a) S. Werkmeister, K. Junge and M. Beller, Org. Process Res.
Dev., 2014, 18, 289; (b) D. B. Bagal and B. M. Bhanage, Adv.
Synth. Catal., 2015, 357, 883; (c) D. S. Mérel, M. L. T. Do, S.
Gaillard, P. Dupau and J.-L. Renaud, Coord. Chem. Rev.,
2015, 288, 50.
11 For recent reviews on pincer-catalysts, see: (a) T. Zell and D.
Milstein, Acc. Chem. Res., 2015, 48, 1979; (b) S. Werkmeister,
J. Neumann, K. Junge and M. Beller, Chem. – Eur. J.,
2015, 21, 12226.
12 (a) G. Zhang, B. L. Scott and S. K. Hanson, Angew. Chem.,
Int. Ed., 2013, 51, 12102; (b) P. O. Lagalitis, P. E. Sue, J. F.
Sonnenberg, K. Y. Wan, A. J. Lough and R. H. Morris, J. Am.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Catalysis Science & Technology
Communication
Chem. Soc., 2014, 136, 1367; (c) J. F. Sonnenberg, A. J. Lough
and R. H. Morris, Organometallics, 2014, 33, 6452; (d) T. Zell,
Y. Ben-David and D. Milstein, Catal. Sci. Technol., 2015, 5,
822; (e) S. Rösler, J. Obenauf and R. Kempe, J. Am. Chem.
Soc., 2015, 137, 79998; ( f ) A. Mukherjee, D. Srimani, S.
Chakraborty, Y. Ben-David and D. Milstein, J. Am. Chem.
Soc., 2015, 137, 8888; (g) D. Srimani, A. Mukherjee, A. F. G.
Goldberg, G. Leitus, Y. Diskin-Posner, L. J. W. Shimon, Y.
Ben-David and D. Milstein, Angew. Chem., Int. Ed., 2015, 54,
12357; (h) S. Rösler, M. Ertl, T. Irrgang and R. Kempe,
Angew. Chem., Int. Ed., 2015, 54, 15046; (i) S. Chakraborty, G.
Leitus and D. Milstein, Chem. Commun., 2016, 52, 1812; ( j)
B. Butschke, M. Feller, Y. Diskin-Posner and D. Milstein,
Catal. Sci. Technol., 2016, DOI: 10.1039/c5cy02036k; (k) J. A.
Gang, S. Chakraborty, Y. Ben-David and D. Milstein, Chem.
Commun., 2016, 52, 5285.
13 F. Schenk, M. Assmann, M. Balmer, K. Harms and R.
Langer,
Organometallics,
2016,
DOI:
10.1021/acs.
organomet.6b00251.
14 For a similar compound [(PNPCy)Fe(CO)Cl2], see ref. 10b.
15 Additionally, a second signal at −20.44 ppm appeared in the
hydride region of the 1H NMR spectrum. This result supports
the assumption that in solution two isomers of the borane
adduct in a ratio of 2.4 : 1 are present. After keeping the NMR
tube for several hours at room temperature, a complete shift
of the equilibrium to the major isomer is observed.
16 CCDC 1450524 data can be obtained free of charge from
www.ccdc.cam.ac.uk/data_request/cif.
This journal is © The Royal Society of Chemistry 2016
Catal. Sci. Technol.