NNP-Type Pincer Imidazolylphosphine Ruthenium Complexes: Efficient Base-Free Hydrogenation of Aromatic and Aliphatic Nitriles under Mild Conditions

Article abstract of DOI:10.1002/chem.201504709

A series of seven novel N^ImN^HP-type pincer imidazolylphosphine ruthenium complexes has been synthesized and fully characterized. The use of hydrogenation of benzonitrile as a benchmark test identified [RuHCl(CO)(N^ImN^HP^tBu)] as the most active catalyst. With its stable Ru-BH₄ analogue, in which chloride is replaced by BH₄, a broad range of (hetero)aromatic and aliphatic nitriles, including industrially interesting adiponitrile, has been hydrogenated under mild and base-free conditions.

Full text of DOI:10.1002/chem.201504709

DOI: 10.1002/chem.201504709
Full Paper
&
Synthesis Design
NNP-Type Pincer Imidazolylphosphine Ruthenium Complexes:
Efficient Base-Free Hydrogenation of Aromatic and Aliphatic
Nitriles under Mild Conditions
Rosa Adam+,[a] Elisabetta Alberico+,[a, b] Wolfgang Baumann,[a] Hans-Joachim Drexler,[a]
Ralf Jackstell,[a] Henrik Junge,[a] and Matthias Beller*[a]
Abstract: A series of seven novel NImNHP-type pincer imid-
azolylphosphine ruthenium complexes has been synthesized
and fully characterized. The use of hydrogenation of benzo-
nitrile as a benchmark test identified [RuHCl(CO)(NImNHPtBu)]
as the most active catalyst. With its stable RuÀBH4 analogue,
in which chloride is replaced by BH4, a broad range of (het-
ero)aromatic and aliphatic nitriles, including industrially in-
teresting adiponitrile, has been hydrogenated under mild
and base-free conditions.
Introduction
Pincer ligands and their complexes are of high interest to or-
ganometallic and organic chemists, since they are widely and
successfully applied in catalysis.[1] Their tridentate and, in the
majority of cases, planar coordination to the metal center con-
fers improved stability to the resulting catalysts, which are
therefore also suitable for reactions that might require forcing
conditions. Clearly, the design of pincer ligands serves to mod-
ulate the steric and electronic properties of the corresponding
metal center. In addition, the ligand often participates actively
in several elementary steps of the catalytic cycle through re-
versible bond activation (bifunctional catalysis).[2] This paved
the way to alternative reactivities, and thus, several redox
transformations could be achieved under milder conditions.[3]
The majority of today’s pincer ligands, which allow for bifunc-
tional catalysis, are built upon
a
2,6-lutidine skeleton[4a]
(Scheme 1a) or possesses a central aliphatic NH group[4b]
(Scheme 1b). Structural diversity in these systems is then
achieved by introducing different donor sets on the pincer
arms. Several versatile catalysts have been prepared, in which
pincer ligands of each class are symmetrically substituted by
[a] Dr. R. Adam,+ Dr. E. Alberico,+ Dr. W. Baumann, Dr. H.-J. Drexler,
Dr. R. Jackstell, Dr. H. Junge, Prof. Dr. M. Beller
Leibniz-Institut für Katalyse e.V. Albert Einstein Str. 29a
18059 Rostock (Germany)
E-mail: Matthias.Beller@catalysis.de
[b] Dr. E. Alberico+
Istituto di Chimica Biomolecolare, Consiglio Nazionale delle Ricerche
Tr. La Crucca 3, 07100 Sassari (Italy)
[+] These authors contributed equally to this work.
Supporting information and the ORCID for the author of this article can be
found under http://dx.doi.org/10.1002/chem.201504709. It contains experi-
mental procedures for the synthesis of ligands 1a–2c and complexes 3–6
and 8–10, their characterization and NMR spectra, crystallographic infor-
mation, additional tables, and characterization of isolated amines.
Scheme 1. Representative examples of pincer ruthenium complexes, the li-
gands of which are built upon a 2,6-lutedine skeleton (a) or possess a central
aliphatic NH group (b). For both classes, the reversible activation of dihydro-
gen is enabled by metal–ligand cooperation. Py=pyridine, Im=imidazole.
Chem. Eur. J. 2016, 22, 4991 – 5002
4991
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
alkyl- or aryl-substituted phosphino groups (PNPyP, PNHP): yet,
replacement of a phosphino group with either a dialkylamino
(NNPyP) or pyridino group (NPyNPyP, NPyNHP) has been shown to
improve catalyst activity in ester hydrogenation,[5] amide hy-
drogenolysis to alcohol and amine,[6] dehydrogenations of sec-
ondary alcohols to ketones,[7] dehydrogenative couplings of
primary alcohols to afford esters,[5a,8] acylation of secondary al-
cohols with esters,[9] and dehydrogenative couplings of alco-
hols and amines to form amides.[10]
first-row metals,[21] in this area ruthenium still remains the
benchmark metal to probe the ability of new ligands to pro-
vide efficient catalysts.[1a,b] In a preliminary report, we showed
the synthesis of one member of this new complex family and
its catalytic behavior in base-free hydrogenation of amides to
alcohols and amines.[22] Herein, we describe the general syn-
thesis of a number of ruthenium complexes with the new li-
gands and their application to the hydrogenation of nitriles.
In the last case, the authors ascribed the enhanced activity
of the ruthenium–NNPyPtBu catalyst to the potentially hemilabile
nature[11] of the ligated amino group. By providing a temporary
vacant coordination site at the metal, dehydrogenation of the
coordinated alcohol (and of the hemiaminal formed from the
reaction of the resulting aldehyde with amine) is facilitated.
However, DFT calculations have questioned this partial dissoci-
ation of the pincer ligand and shown that, under the applied
experimental conditions, dehydrogenation of the alcohol (and
hemiaminal) through a bifunctional double hydrogen transfer
in the metal outer coordination sphere is likewise feasible and
indeed energetically favored.[12] Even more interesting is the
observation that, when the RuÀPtBuNPyPtBu catalyst is used in-
stead, the dehydrogenative coupling of primary alcohols and
primary amines affords imines rather than amides.[13] Imines
arise from dehydration of the intermediate hemiaminal, which,
for the RuÀPtBuNPyPtBu-mediated process, is, in general, kinetical-
ly more favorable than the metal-mediated dehydrogenation
of the same intermediate to the corresponding amide.[14] The
switch in product selectivity is ruled by the diverse steric and
electronic properties of the side donor in the pincer ligand,
either a phosphino or an amino group, rather than the ability
of the latter to reversibly dissociate from the metal around the
catalytic cycle.[14] Furthermore, recent work by Langer and Xu
addressed the effect of varying the nature of the heterocyclic
group in RuÀXNHP-type catalysts: replacing the side arm pyri-
dino group with weaker coordinating heterocycles, such as
a furanyl or thiophenyl group, negatively affects the activity of
the corresponding catalyst in the acceptorless dehydrogena-
tion of alcohols and hydrogenation of ketones.[15]
Results and Discussion
Synthesis of complexes
The NImNHP pincer ligands, 1a–2c (Scheme 2), were easily as-
sembled in a one-pot, two-step procedure starting from com-
mercially available 1-methyl-2-imidazolecarboxaldehyde and
the respective aminophosphine. After condensation, the result-
ing imine was reduced with sodium borohydride. Relevant
ligand NMR spectroscopy data are reported in Table 1.
Scheme 2. General synthetic procedure to access the novel NImNHP pincer li-
gands 1a–2c. The chemical shifts of the highlighted atoms are reported in
Table 1. The synthesis of 2c has been already described, see ref. [22].
Table 1. Selected NMR spectroscopy data of pincer ligands 1a–2c.[a]
Ligand
P d (31P)[ppm]
CHIm d (1H)[ppm]
CHIm d (1H)[ppm]
Following our previous work on the synthesis of bidentate
imidazolyl phosphines and their successful application in the
ruthenium-catalyzed hydrogenation of carboxylic acid deriva-
tives,[16] we considered adding to the diversity of NNHP-type
pincer ligands by introducing a 1-methylimidazolyl group.[17] 1-
Methylimidazole, with a pKaH of 7.0, has a basicity that is inter-
mediate between that of the side-arm nitrogen donors so far
employed in NNP pincer ligands, such as a pyridine (pyridine
pKaH =5.2), or an amine moiety NR2 (fully saturated aliphatic
amines have pKaH values mostly within the range 9–11).[18] Its
coordinating properties are expected to be likewise intermedi-
ate.[19] Because of the commercial availability of the corre-
sponding building blocks and the ease of synthesis, ligands
with different substituents at phosphorus and either a two- or
three-carbon chain connecting the central aliphatic nitrogen
and phosphorus donor have been prepared.[20]
1a
1b
2a
2b
2c[b]
À20.40
À1.17
À15.80
4.06
6.83
6.83
6.87
6.83
6.87
6.82
6.82
6.83
6.82
6.86
28.71
[a] Spectra were recorded in CD2Cl2 at 298 K. [b] Data of 2c, the synthesis
of which has been already described (ref. [22]), are reported herein for
comparison.
To prepare the corresponding ruthenium complexes, the li-
gands were reacted with two different precursors: [RuHCl-
(CO)(PPh3)3] or [RuCl2(PPh3)3].
The type of complex arising from the reaction with [RuHCl-
(CO)(PPh3)3] turned out to depend on the length of the tether
between the central nitrogen atom and the phosphino group
in the corresponding pincer ligand, either a two- (1a, 1b) or
three-carbon chain (2a–c). When either ligand 1a or 1b was
Although increasing effort is being put into the design of
catalysts based on more abundant, cheaper, and less toxic
Chem. Eur. J. 2016, 22, 4991 – 5002
www.chemeurj.org
4992
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
reacted with [RuHCl(CO)(PPh3)3] in toluene at reflux for 3 h, the
corresponding cationic complexes 3 and 4 were isolated as
a mixture of cis and trans isomers, so defined based on the rel-
ative orientation of the ruthenium hydride and the hydrogen
on the aliphatic nitrogen (Scheme 3).[23,24]
the hydrogen on the aliphatic nitrogen (cis) and the hydride,
respectively (see the Supporting Information).
In solution, the binding of the imidazolyl sp2-N donor to
1
ruthenium was inferred by changes in the H chemical shifts of
the imidazolyl H4 and H5 resonances relative to those in the
free ligands. Whereas in the latter case they give rise to an AB
system with very close chemical shifts, in the complexes the
relative signals move apart, both to higher fields.
All complexes were fully characterized and their NMR spec-
troscopy data revealed similar structural features (Table 2).
The trans orientation of the two phosphorus donors and the
ensuing facial coordination of the pincer ligand is established
for all complexes based on the value of the J(P,P) coupling
constants (260–283 Hz) in the 31P NMR spectrum.[25] In the
1H NMR spectrum, the hydride appeared as a triplet due to the
almost equivalent coupling to the two adjacent cis-oriented
phosphino groups. For both 3 and 4, the structure of each
isomer, either cis or trans, was assigned based on diagnostic
NOE contacts between H4 of the imidazolyl group (trans) or
Crystals of cis-4 suitable for XRD were grown by slow diffu-
sion of diethyl ether into a concentrated solution of the two
isomers in dichloromethane (ratio of cis-4 and trans-4 in solu-
tion 96:4; Figure 1). The solid structure confirmed the ligand
facial coordination deduced from NMR spectroscopy data with
the two phosphorus donors arranged trans to each other and
to the metal center and CO ligand coordinated trans to the ali-
phatic nitrogen. For both complexes, the cis isomer appears to
Scheme 3. Synthesis of complexes 3 and 4: NOE contacts allowed assignment of the cis and trans structures of the two isomers (see the Supporting Informa-
tion).
Table 2. Selected NMR spectroscopy data of complexes 3–10.[a]
Complex
m
PR2 d (31P)
[ppm]
m PPh3 d (31P)
J(P,P)
[Hz]
m
t
RuH d (1H)
[ppm]
J(H,PR2)
[Hz]
J(H,PPh3)
[Hz]
CHIm5 d (1H)
[ppm]
CHIm4 d (1H)
[ppm]
[ppm]
cis-3
trans-3
d
d
62.30
54.21
d
d
42.77
44.41
272.4
283.5
À12.39
20.1
20.7
20.1
17.1
6.43
6.28
6.33
6.06
dd À12.27
cis-4
trans-4
d
d
80.19
72.72
d
d
46.61
44.93
260.3
269.2
t
t
À13.08
À12.60
19.6
18.8
19.6
18.8
6.68
6.59
6.70
6.42
maj-5
s
50.85
–
–
–
d
À15.04
23.9
–
6.89
7.16
maj-6
min-6
brs 65.12
brs 65.1206
–
–
—
–
–
–
d
d
À15.73
À15.80
19.5
21.4
–
–
6.90
6.90
7.17
7.14
cis-7[b]
trans-7[b]
s
s
78.11
74.19
–
–
–
–
–
–
d
d
À15.91
À16.25
25.2
23.0
–
–
6.90
6.89
7.15
7.15
maj-8
min-8
d
d
51.60
55.60
d
d
51.94
50.35
31.3
31.2
–
–
–
–
–
–
–
–
6.62
n.d.
6.10
n.d.
maj-9
min-9
d
d
55.38
63.63
d
d
52.63
50.25
29.0
29.0
–
–
–
–
–
–
–
–
6.59
6.40
6.12
6.14
maj-10
min-10
d
d
28.86
39.47
d
d
53.87
47.79
33.6
29.0
–
–
–
–
–
–
–
–
6.54
6.48
6.14
6.00
[a] Spectra were recorded in CD2Cl2 at 298 K. [b] The synthesis and characterization of 7 have been reported elsewhere, see ref. [22]. Selected NMR spec-
troscopy data are included here for comparison.
Chem. Eur. J. 2016, 22, 4991 – 5002
www.chemeurj.org
4993
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
tablishing the relative orientation of all coordinated ligands in
the various isomers, some important structural features of the
complexes could be established based on 1D and 2D NMR
spectra. For each complex, regardless of the substituents at
phosphorus or the isomer, only one signal, a singlet, was de-
tected in the 31P{1H} NMR spectrum, which indicated that PPh3
had been completely displaced from the ruthenium coordina-
tion sphere (Table 2). The hydride signal appears as a doublet
ranging from d=À15.04 ppm in maj-5 to d=À16.25 ppm in
trans-7.[26] For 6 and 7, the chemical shifts of the hydrides in
the two isomers are quite close, which suggests that the hy-
dride must be coordinated trans to either the same ligand or
a ligand of comparable trans influence.[27] In all cases, the hy-
dride is located cis to the phosphorus donor, as determined
from the value of J(H,P), which falls in the range of 19.4–
25.2 Hz (Table 2). For each complex, one strong broad band in
the IR attenuated total reflectance (ATR) spectrum corroborates
the presence of one coordinated CO, the stretching frequency
of which shifts from n˜ =1906 cmÀ1 in 5, to n˜ =1892 cmÀ1 in 6,
to n˜ =1890 cmÀ1 in 7; these values clearly reflect the more pro-
nounced back-donation in the complexes 6 and 7, which is
made possible by the presence of the more s-donating PiPr2
and PtBu2 ligands in 6 and 7, respectively, relative to PPh2 in 5.
For 7, the most abundant isomer could be assigned the cis
configuration.[28] Complexes 5–7 are not soluble or poorly solu-
ble in solvents such as 1,2-dichloroethane, toluene, THF, ace-
tone, MeOH, and CH3CN. However, they are perfectly soluble in
CD2Cl2. Hence, this solvent was used to record their NMR spec-
tra. Unfortunately, in this solvent, the hydride ligand of 5 and 6
is gradually substituted by chloride, as shown by the steady
disappearance of the hydride resonances with concomitant
growth of a signal at d=2.99 ppm (pent, J(H,D)=1.6 Hz),
which is assigned to CHD2Cl (see the Supporting Informa-
tion).[29] Likewise, in the 31P NMR spectra, the signals relative to
the ruthenium monohydride species are gradually replaced by
a signal due to the corresponding ruthenium dichloride spe-
cies Cl2-5 (31P{1H} NMR: d=41.27 ppm (s)) and Cl2-6
(31P{1H} NMR: d=46.03 ppm (s)). Although a kinetic investiga-
tion was not performed, the rate of nucleophilic substitution
of chloride in CD2Cl2 by the ruthenium hydride, as determined
from monitoring of the process by NMR spectroscopy, decreas-
es upon going from 5 to 6: the ruthenium dichloride complex
forms within hours once 5 or 6 are dissolved in dichlorome-
thane; H/Cl exchange does not take place in 7. For 5, yellow
crystals of Cl2-5, [RuCl2(CO)(PPh3){3-(diphenylphosphino)-N-[(1-
methyl-1H-imidazol-2-yl)methyl]-propylamine}], suitable for
XRD formed from a solution of 5 in CD2Cl2 inside the J-Young
NMR tube after a few days. The X-ray structure revealed that
Cl2-5 possessed a distorted octahedral geometry with the
pincer ligand coordinated in a meridional fashion and the CO
ligand bound to ruthenium trans to the central aliphatic nitro-
gen (Figure 2). The two chlorides occupy the apical positions
and are disposed trans to each other.
Figure 1. X-ray structure of cis-4 with thermal ellipsoids drawn at the 30%
probability level. Hydrogen atoms other than H1 and H3 are omitted for
clarity. H1 and H3 were refined based on electron density. Selected bond
lengths and angles are reported in Table 3.
be the thermodynamic product because its relative amount in-
creases upon standing in solution (see the Supporting Informa-
tion).
When ligands 2a–c were reacted with [RuHCl(CO)(PPh3)3] in
toluene at reflux for 3 h, the corresponding neutral complexes
5–7 were isolated as mixtures of isomers, the relative amounts
of which depended on the ligand (Scheme 4). In this case, the
ratio of isomers obtained following the chosen standard syn-
Scheme 4. Synthesis of complexes 5–7: for the ratio of isomers, see text and
the Supporting Information.
thetic procedure could not be altered by extending the reac-
tion time of the latter, neither by prolonged standing nor heat-
ing in solution. For 5, the most abundant isomer, maj-5, repre-
sents 88% of the total content; the rest is equally distributed
among three minor isomers, of which only the hydride signals
and the corresponding phosphorus signals could be assigned.
Complex 6 afforded two isomers in a ratio of 57:43 but, be-
cause of extensive overlapping, complete assignment of NMR
signals to either maj-6, the most abundant of the two isomers,
or min-6, the less abundant one, was not possible. For com-
plex 7, the configuration of the two isomers, in a ratio of
74:26, was confidently assigned.[22] Despite the difficulty in es-
Next, the synthesis of neutral ruthenium complexes with an
ancillary PPh3 in place of CO was considered, as a further possi-
bility to modulate the electronic and steric properties of the
metal center (Scheme 5). Thus, complexes 8–10 were obtained
Chem. Eur. J. 2016, 22, 4991 – 5002
www.chemeurj.org
4994
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
in good yields through ligand exchange reactions with the pre-
cursor [RuCl2(PPh3)3] to afford, in all cases, a mixture of two iso-
mers, in a ratio ranging from 93:7 to 80:20, as red–orange crys-
talline solids.[30] In all complexes, regardless of the isomer, the
J(P,P) coupling constant falls between 29 and 34 Hz, which in-
dicates that the phosphorus donor on the pincer ligand and
the ancillary triphenylphosphine are coordinated cis to each
other (Table 3).
For all complexes, it was possible to grow crystals suitable
for XRD, which revealed in every case a meridional arrange-
ment of the pincer ligand with PPh3 coordinated trans to the
central aliphatic nitrogen and the two chlorides bound to
ruthenium trans to each other (Figures 3–5). However, it was
not possible to correlate the solid-state structure with that of
either isomer in solution. Interestingly, by comparing the struc-
tures of complexes 3 and 8, or 4 and 9, it is clear that ligands
1a and 1b can adopt both the facial and meridional coordina-
tion modes.[31]
Figure 2. X-ray structure of Cl2-5 with thermal ellipsoids drawn at the 30%
probability level. Hydrogen atoms other than H3 are omitted for clarity. Se-
lected bond lengths and angles are reported in Table 3.
Nitrile hydrogenation
The potential of the new complexes as catalysts was tested in
the hydrogenation of nitriles.[32] This chemical transformation
was chosen because of its synthetic importance and the possi-
bility to directly compare the activity of the new catalysts with
that of analogous PNPÀRu ones.[33] The hydrogenation of ni-
triles allows for a direct synthesis of amines, which are valuable
organic compounds that have wide industrial applications as
solvents, additives, antifoam agents, corrosion inhibitors, deter-
gents, dyes, and bactericides.[34] Moreover, the amino group is
widespread in agrochemicals and pharmaceuticals.[34] On an in-
dustrial level, nitriles are reduced by using heterogeneous cata-
lysts,[35] mainly Raneyꢁ-Ni and -Co; however, these reactions re-
quire the presence of ammonia or ammonium salts to prevent
Scheme 5. Synthesis of complexes 8–10: each complex was obtained as
a mixture of two isomers: maj-8/min-8 91:9; maj-9/min-9 93:7; maj-10/min-
10 80:20. The structure of the trans-(Cl)(Cl)-isomer, as established by X-ray
analysis for all complexes, is drawn.
Table 3. Selected bond lengths [Š] and angles [8] for complexes syn-4, Cl2-5, syn-7, 8, 9, and 10.
cis-4
Cl2-5
cis-7[a]
8[b]
9[b]
10[b]
Ru1ÀCl1
–
–
2.4015(6)
2.3881(6)
–
2.1248(18)
2.208(2)
2.2882(6)
1.841(3)
–
1.125(3)
95.81(5)
75.91(7)
171.61(6)
–
2.5741(7)
–
2.4216(4)
2.4033(4)
–
2.1564(14)
2.1631(14)
2.2965(5)
–
2.3126(5)
–
82.88(4)
75.75(5)
158.33(4)
104.080(16)
–
–
2.4376(10)
2.4147(10)
–
2.1694(14)
2.1592(16)
2.3168(8)
–
2.3021(7)
–
83.56(4)
75.66(6)
158.36(4)
102.17(3)
–
–
2.4173(10)
2.4477(10)
–
2.1611(15)
2.2161(16)
2.3063(6)
–
2.3062(7)
–
88.70(5)
76.75(6)
165.02(4)
98.10(3)
–
–
Ru1ÀCl2
Ru1ÀH1
1.71(3)
2.196(2)
2.240(3)
2.3044(9)
1.811(4)
2.3825(8)
1.166(4)
81.79(7)
74.39(9)
99.26(7)
165.31(3)
93.29(11)
90.08(11)
96.30(7)
–
1.52(2)
2.1137(15)
2.2219(16)
2.2995(6)
1.809(2)
–
1.161(3)
94.14(4)
77.15(6)
169.28(4)
–
Ru1ÀN1Im
Ru1ÀN3
Ru1ÀP1
Ru1ÀCO1
Ru1ÀP2
CÀO1
P1-Ru1-N3
N3-Ru1-N1Im
P1-Ru1-N1Im
P1-Ru1-P2
P1-Ru1-CO1
CO1-Ru1-P2
N3-Ru1-P2
Cl1-Ru1-Cl2
N1Im-Ru1-CO1
92.29(8)
–
–
171.35(2)
96.06(9)
92.89(6)
–
–
–
–
–
–
169.439(14)
–
166.185(17)
–
165.371(17)
–
–
95.18(7)
[a] The X-ray structure has been reported elsewhere; see ref. [22]. Selected bond lengths and angles are included herein for comparison. [b] Data refer, for
each complex, to the isomer in which the two chloride ligands are coordinated to ruthenium trans to each other.
Chem. Eur. J. 2016, 22, 4991 – 5002
www.chemeurj.org
4995
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 3. X-ray structure of trans-Cl2-8 with thermal ellipsoids drawn at the
30% probability level. Hydrogen atoms other than H3 are omitted for clarity.
Selected bond lengths and angles are reported in Table 3.
Figure 5. X-ray structure of trans-Cl2-10 with thermal ellipsoids drawn at the
30% probability level. Hydrogen atoms other than H3 are omitted for clarity.
Selected bond lengths and angles are reported in Table 3.
Scheme 6. Catalytic hydrogenation of nitriles and possible side reactions.
imine (Scheme 6). This crucial intermediate can be further hy-
drogenated to afford the desired amine (Scheme 6, path A), or
can react with a molecule of the latter to provide a secondary
imine with release of ammonia (Scheme 6, path B). Hydrogena-
tion of the secondary imine leads to the secondary amine.
Classically, side-product formation has been mitigated by the
addition of base.
Figure 4. X-ray structure of trans-Cl2-9 with thermal ellipsoids drawn at the
30% probability level. Hydrogen atoms other than H3 are omitted for clarity.
Selected bond lengths and angles are reported in Table 3.
Several reports have appeared on the homogeneous hydro-
genation of nitriles promoted by ruthenium,[16a,38] rhodium,[39]
iridium,[40] molybdenum,[41] and rhenium[42] complexes. More re-
cently, successful systems based on the use of non-noble
metals, such as iron[43] and cobalt,[44] in combination with PNP
pincer ligands, have been also disclosed. With the exception of
the FeÀPNP complex[43b] and a few other RuÀpincer cata-
lysts,[33,38a,c] most of the reported systems require a basic addi-
tive to hamper the formation of secondary products, which
can be a drawback in terms of functional group tolerance.
Therefore, efforts are being devoted to the development of
the formation of byproducts (secondary and tertiary amines/
imines).[36] For small-scale preparations, nitriles are usually re-
duced by using metal hydrides or boranes;[37] a methodology
that suffers from low atom economy because it generates
(over)stoichiometric amounts of difficult to dispose waste by-
products. In this respect, the possibility of carrying out the
same transformation with molecular hydrogen is highly desir-
able. The key for the efficient synthesis of primary amines from
nitriles is control of the catalytic hydrogenation of the primary
Chem. Eur. J. 2016, 22, 4991 – 5002
www.chemeurj.org
4996
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
À
novel catalytic systems that allow the hydrogenation of nitriles
to be performed under mild conditions, without additives,
while providing a high selectivity in the desired amines.
catalyst BH4-7,[46] in which the Cl ligand was replaced by BH4
,
was used with no added base.[47]
Indeed, compound 11 was quantitatively hydrogenated over
3 h under base-free conditions with 1 mol% of BH4-7 at 708C
and 30 bar of hydrogen (Table 4, entry 1). To our delight, the
reaction proceeded smoothly, even at lower catalyst loadings
(Table 4, entries 2–5), reaching a TON of about 2000 at
0.05 mol%; this is the highest TON reported so far for this reac-
tion. When either the catalyst loading (Table 4, entry 6) or the
hydrogen pressure were reduced further (Table 4, entry 7), the
reaction became slower thus favoring the formation of the sec-
ondary imine in moderate yields.
To assess the catalytic activity of the novel complexes in ni-
trile hydrogenation, benzonitrile (11) was chosen as a standard
substrate (Figure 6 and Table S1 in the Supporting Informa-
tion). To begin with, reactions were carried out for 3 h in iPrOH
at 708C under 30 bar of hydrogen with 1 mol% of the catalyst
Quantitative yields of 12 were obtained, even at 508C, when
using 0.1 mol% BH4-7 (Table 4, entry 9). By comparing the re-
sults in entries 9 and 11 in Table 4 (reaction times of 3 and 1 h,
respectively, under otherwise identical conditions), it is evident
that in the initial stages of the reaction the main product is
the secondary imine, which, through the equilibria described
in Scheme 6, reverts back to the primary imine as the reaction
proceeds and is eventually hydrogenated to the primary
amine.[33b]
A few solvents, other than iPrOH, were screened (Table S3 in
the Supporting Information): complex BH4-7 was only active in
THF; however, the yield of 12 was quite low. The reaction did
not proceed in the absence of hydrogen (Table 4, entry 8). Sim-
ilar experiments were carried out to study the influence of cat-
alyst loading, pressure, and temperature on the reduction of
13 in the presence of BH4-7 (Table 5). As for 11, the screening
results show that for better selectivity the three parameters
have to be adjusted to secure a fast enough reaction to pre-
vent the intermediate imine from reacting with the product
amine (Table 5, entries 3 and 6). The best compromise was ob-
tained when using 0.5 mol% of catalyst at 708C under 15 bar
of H2, affording heptylamine (14) in excellent yield (98%;
Table 5, entry 4) over 3 h. Even in this case, no reaction took
place in the absence of hydrogen (Table 5, entry 5).
Figure 6. Comparison of the catalytic activity of complexes 3–10 in the hy-
drogenation of 11 to afford 12 in 30 bar hydrogen.
and a minimum amount of base (5 mol% of KOtBu). Under
these conditions, all catalysts afforded quantitative yields of
benzylamine (12), with the exception of 8 and 10, which gave
the product in moderate to low yield (Table S1, entries 1–8, in
the Supporting Information). To probe the influence of base,
the best performing catalysts were tested with the amount of
base strictly necessary to generate the catalytic active species
by their dehydrochlorination: one equivalent to ruthenium for
3–7; two for 9, under otherwise identical conditions.[45] Indeed
complexes 4, 5, 6, and 7 catalyzed the formation of 12 in ex-
cellent yields, whereas 3 and 9 showed a lower activity and se-
lectivity, which resulted in the preferential formation of the
secondary imine S1 (Table S1, entries 9–14, in the Supporting
Information). At a lower catalyst loading (0.5 mol%), only com-
plexes 5–7 were active enough to secure the quantitative and
selective formation of 12 over the 3 h of reaction time
(Table S1, entries 16–18, in the Supporting Information). Cata-
lyst 4 was less efficient and afforded exclusively the secondary
imine S1 (Table S1, entries 15, in the Supporting Information).
Finally, by decreasing the temperature to 508C, it was possible
to identify 7, among the structurally related catalysts 5–7, as
the most active catalyst for the reduction of 11 to 12, which
proceeded in 85% yield (Table S1, entries 19–20, in the Sup-
porting Information). Also, in a similar screening, complex 7
was the most active and selective catalyst for the hydrogena-
tion of the aliphatic nitrile 1-heptanenitrile (13) as well
(Table S2 in the Supporting Information).
Having assessed to which extent reaction conditions might
be mitigated without compromising yield and selectivity, the
general applicability of the new system was evaluated in the
hydrogenation of several aromatic and aliphatic nitriles. Table 6
summarizes the results obtained with aromatic and heteroaro-
matic nitriles. All experiments were carried out under 30 bar of
hydrogen. In most cases, excellent yields of primary amines
were obtained at 508C with 0.5 mol% of BH4-7. Both electron-
donating (Table 6, entries 3–8) and -withdrawing groups
(Table 6, entries 9–12) are tolerated, including ortho-substituted
derivatives.
Benzonitriles with either an ester or an amido group are se-
lectively hydrogenated, too. For methyl 4-cyanobenzoato, se-
lectivity was achieved by shortening the reaction time (Table 6,
entry 13). In case of N-(4-cyanophenyl)acetamide, reduction of
the sole cyano group was achieved at 708C over 3 h.[22] Gratify-
ingly, selected heteroaromatic nitriles were efficiently hydro-
genated to the corresponding amines (Table 6, entries 15–20).
Under mild reaction conditions, 3- and 4-pyridinecarbonitrile,
but not 2-pyridinecarbonitrile, afforded the corresponding pi-
colylamines in high yields (Table 6, entries 15 and 16). The 2-
Notably, catalyst 7 requires only one equivalent of base to
promote the reduction of 11. This means that base might be
needed only in the initial stage of the process to generate the
catalytic active species. Therefore, for subsequent experiments,
Chem. Eur. J. 2016, 22, 4991 – 5002
www.chemeurj.org
4997
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Table 4. Optimization of reaction conditions for the hydrogenation of 11 with ruthenium complex BH4-7.[a]
Entry
H2 [bar]
T [8C]
BH4-7 [mol%]
Conversion [%][b]
Yield 12 [%][b]
TON[c]
1
2
3
4
30
30
30
30
30
30
15
–
30
30
30
30
70
70
70
70
70
70
70
70
50
50
50
40
1
0.5
100
100
100
100
100
70
87
–
100
91
54
>99
>99
>99
>99
>99
–
5
–
>99
20
99
198
396
990
1980
–
100
–
990
400
–
0.25
0.1
0.05
0.025
0.05
0.5
0.1
0.05
0.1
5
6[d]
7[d]
8
9
10[d]
11[d,e]
12[d]
–
–
0.1
79
–
[a] Standard reaction conditions at 1–0.5 mol% of Ru: 11 (0.5 mmol, 51.6 mg), BH4-7 (1–0.5 mol%), dry iPrOH (2 mL) under H2 over 3 h; standard reaction
conditions at 0.25 mol% of Ru: 11 (1 mmol, 103.1 mg), BH4-7 (0.25 mol%), dry iPrOH (2 mL) under H2 over 3 h; standard reaction conditions at 0.1 mol%
of Ru: 11 (2 mmol, 206.2 mg), BH4-7 (0.1 mol%), dry iPrOH (2 mL) under H2 over 3 h; standard reaction conditions at 0.05 mol% of Ru: 11 (4 mmol,
412.5 mg), BH4-7 (0.025 mol%), dry iPrOH (4 mL) under H2 over 3 h; standard reaction conditions at 0.025 mol% of Ru: 11 (10 mmol, 1.0 g), BH4-7
(0.1 mol%), dry iPrOH (10 mL) under H2 over 3 h. [b] The conversion of 11 and yield of 12 were calculated by GC with hexadecane as an external standard.
[c] Turnover number (TON) was calculated by dividing the yield of product by mol% of catalyst. [d] Intermediate S1 formed in yields of 65 (entry 6), 75
(entry 7), 71 (entry 10), 37 (entry 11), and 61% (entry 12). [e] The reaction time was 1 h.
15 bar of hydrogen, linear, branched, and cyclic ni-
Table 5. Optimization of the reaction conditions for the hydrogenation of 13 with
ruthenium complex BH4-7.[a]
triles were successfully hydrogenated to the corre-
sponding primary alkyl amines over 3 h in high
yields, regardless of the alkyl chain length (Table 7,
entries 1–5) or steric bulk close to the cyano group
(Table 7, entries 6–9). Additionally, benzyl-substituted
nitriles and 3-phenylpropanenitrile also smoothly af-
forded the corresponding amines (Table 7, en-
tries 10–12). Remarkably, the industrially relevant
hexane-1,6-diamine, used for the production of
Nylon-6,6, was obtained in high yield by the hydro-
genation of adiponitrile at 708C (Table 7, entry 13).
Compared with previously known reduction of ni-
triles promoted by pincer ruthenium complexes
(Figure 7), the novel catalyst BH4-7 allowed reaction
conditions to be mitigated further.
Entry
H2 [bar]
T [8C]
BH4-7 [mol%]
Conv. [%][b]
Yield 14 [%][b]
1
2
3[c]
4
30
30
30
15
–
70
70
70
70
70
50
70
1
100
100
85
100
–
>99
>99
67
98
–
0.5
0.25
0.5
0.5
0.5
0.5
5
6[c]
7[d]
30
30
80
95
49
90
[a] Standard reaction conditions: 13 (0.5 mmol, 55.6 mg), BH4-7 (1-0.25 mol%), dry
iPrOH (2 mL) under H2 over 3 h. [b] The conversion of 13 and yield of 14 were calcu-
lated by GC with hexadecane as an external standard. [c] N-Heptylideneheptan-1-
amine was formed as a side product. [d] The reaction time was 1 h.
In the case of the RuÀPtBuNHPtBu catalyst 17, which
represents a pertinent example because it carries the
same tBu substituents at phosphorus as BH4-7, the
reported substrate scope is less broad.[33b] The system
and 3-(thiophen-2-yl)acetonitrile derivatives, as well as 2-
(furan-2-yl)acetonitrile and 1H-indole-5-carbonitrile, could be
also hydrogenated (Table 6, entries 17–20), although for these
substrates higher temperatures and/or higher catalyst loadings
were required. Finally, the hydrogenation of terephthalodini-
trile afforded the corresponding diamine in high yield (86%;
Table 6, entry 21).
Next, the scope of BH4-7 in aliphatic nitrile hydrogenation
was explored (Table 7). With 0.5 mol% catalyst, at 708C under
Figure 7. RuÀPNHP pincer complexes used for the hydrogenation of nitriles.
Chem. Eur. J. 2016, 22, 4991 – 5002
www.chemeurj.org
4998
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Table 6. Substrate scope for the hydrogenation of various aromatic nitriles with BH4-7.[a]
Entry Nitrile
Amine
T
Conversion
Yield 12
[%][b]
Entry Nitrile
Amine
T
Conversion
Yield 2
[%][b]
[8C] [%][b]
[8C] [%][b]
1
2
50
50
>99
>99
99[c]
98
11[d]
12
70 >99
99
130 >99
92
3
4
50
50
>99
>99
99
98
13[e]
50 >99
70 >99
85[c]
84[c]
14[d]
5
70
50
>99
>99
88
90
15
16
50 >99
50 >99
86
95
6[d]
7[d]
50
>99
99
17
70 >99
89
82
8
9
50
50
>99
>99
90[c]
92
18[d]
130 >99
19[f]
150 >99
150 >99
75
98
20[g]
10[d]
50
>99
89
21
50 >99
86[c]
[a] Standard reaction conditions: nitrile (0.5 mmol), BH4-7 (0.5 mol%), iPrOH (2 mL), 30 bar H2, 3 h. [b] The conversion and yields were calculated by GC
with hexadecane as an external standard. [c] Yield of product isolated as the ammonium salt. [d] 1 mol% of BH4-7. [e] 1 hour. [f] 1 mol% of BH4-7, KOtBu
10 mol%, 15 h. [g] 1 mol% of BH4-7, 15 h.
is remarkable because it operates at very low pressure, only
4 bar, which, in the case of 11, provides an almost quantitative
yield of the corresponding secondary imine in the first hour,
but longer reactions times are necessary to convert it into the
primary amine.[33b]
case of complex 17, high conversions were detected for the
hydrogenation of both 11 and 13. However, this catalyst was
not selective to the primary amine and the corresponding sec-
ondary imine was detected as a side product (Table 8, entries 3,
6, and 9).
A direct comparison of catalysts BH4-7, Ru-MachoTM-BH, and
17 in the reduction of both 11 and 13 under the same reaction
conditions was performed (Table 8). Ru-MachoTM-BH was less
active than BH4-7 for the hydrogenation of 11 affording low
conversions under the tested conditions (Table 8, entries 2 and
5), whereas for the hydrogenation of 13 the two catalysts
showed a similar performance at 708C (Table 8, entry 8). In the
Conclusions
A small family of novel pincer ruthenium complexes was pre-
pared and tested in the hydrogenation of nitriles. The novel
hybrid NNP imidazolylphosphino pincer ligands were easily as-
sembled from commercially available building blocks that al-
Chem. Eur. J. 2016, 22, 4991 – 5002
www.chemeurj.org
4999
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Table 7. Substrate scope for the hydrogenation of various aliphatic nitriles with BH4-7.[a]
Entry
Nitrile
Amine
Conversion [%][b]
Yield [%][b]
1
2
3
4
5
>99
>99
>99
>99
>99
98[d]
88
88
99
88[d]
6
>99
99
7
>99
76[d]
8
9
>99
>99
99
99
10
>99
>99
91
94
11[c]
12
13
>99
>99
99
87[d]
[a] Standard reaction conditions: nitrile (0.5 mmol), BH4-7 (0.5 mol%), iPrOH (2 mL), 15 bar H2, 3 h. [b] The conversion of RCN and yield of RCH2NH were cal-
culated by GC with hexadecane as an external standard. [c] 1 mol% of BH4-7. [d] Yield of product isolated as the ammonium salt.
flushed with argon three times. Dry isopropanol (2 mL) and 11
(0.5 mmol, 51.56 mL) were added under argon. The vial was set in
an alloy plate and introduced into a 300 mL autoclave (Parr instru-
ment) filled with argon. The autoclave was sealed, purged three
times (20 bar of H2), and pressurized with H2 (30 bar). The auto-
clave was placed into an aluminum block and heated to 508C for
3 h with magnetic stirring. After the desired reaction time, the au-
toclave was quickly cooled with an ice-water bath and the gas
carefully released. The reaction mixture was analyzed by GC-MS
and GC with n-hexadecane as an internal standard.
lowed for structural diversity. The complexes had different
steric and electronic properties, which manifested in their dif-
ferent performances as catalysts for the homogeneous hydro-
genation of nitriles. Indeed catalyst BH4-7, with bulky tBu sub-
stituents at phosphorus and a molecule of CO as an ancillary
ligand at ruthenium, stood out as an efficient catalyst for the
reduction of a range of (hetero)aromatic and aliphatic nitriles
in high yield and selectivity. No added base was necessary to
promote the reaction. Although the catalyst was able to hydro-
genate esters and amides,[22] through modulation of reaction
conditions, it was possible to selectively reduce the cyano
group in the presence of either an ester or amide functionality.
Experimental Section
Isolation of reaction products as HCl salts
General procedure for the hydrogenation of nitriles with
BH4-7
HCl (1m in MeOH, 1 mL) was added to the reaction mixture and
stirring was maintained for 30 min at room temperature. Then, the
reaction mixture was transferred to a 100 mL round-bottomed
flask containing diethyl ether (50 mL). The resulting precipitate was
filtered and washed with diethyl ether and ethyl acetate.
A 4 mL glass vial containing a stirrer bar was charged with BH4-7
(1.1 mg, 0.0025 mmol, 0.5 mol%). The vial, sealed with a septum
equipped with a syringe needle, was evacuated and subsequently
Chem. Eur. J. 2016, 22, 4991 – 5002
www.chemeurj.org
5000
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Table 8. Hydrogenation of 11 and 13 with BH4-7, Ru-MachoTM-BH, and 17 under standard reaction conditions.[a]
Entry
H2 [bar]
T [8C]
Catalyst ([mol%])
Nitrile
Conversion [%][b]
Yield [%][b]
1
2
3[c]
4
30
30
30
30
30
30
15
15
15
50
50
50
70
70
70
70
70
70
BH4-7 (0.5)
11
11
11
11
11
11
13
13
13
100
14
100
100
–
100
100
100
100
>99
13
59
>99
–
15
>99
85
15
Ru-MachoTM-BH (0.5)
17[d] (0.5)
BH4-7 (0.05)
5
Ru-MachoTM-BH (0.05)
17[d] (0.05)
6[e]
7
BH4-7 (0.5)
8
9[f]
Ru-MachoTM-BH (0.5)
17[d] (0.5)
[a] Standard reaction conditions for the hydrogenation of 11 at 0.5 mol% of Ru: 11 (0.5 mmol, 51. 6 mg), catalyst (0.5 mol%), dry iPrOH (2 mL) over 3 h;
standard reaction conditions for the hydrogenation of 11 at 0.05 mol% of Ru: 11 (4 mmol, 412.5 mg), catalyst (0.05 mol%), dry iPrOH (4 mL) over 3 h; stan-
dard reaction conditions for the hydrogenation of 13 at 0.5 mol% of Ru: 13 (0.5 mmol, 55.6 mg), catalyst (0.5 mol%), dry iPrOH (2 mL) over 3 h. [b] The
conversion of 11 and 13 and yields of the corresponding amines, 12 and 14, respectively, were calculated by GC with hexadecane as an external standard.
[c] Intermediate S1 formed in a yield of 41%. [d] Catalyst 17 was prepared in situ by dehydrochlorination of [RuHCl(CO){HN(CH2CH2P{C(CH3)3}2)2}] with one
equivalent of tBuOK. [RuHCl(CO){HN(CH2CH2P{C(CH3)3}2)2}] was prepared by following a published procedure.[48] [e] Intermediate S1 was formed in a yield
of 33% together with an unidentified polymer. [f] N-Heptylideneheptan-1-amine was formed as side product.
[8] a) D. Spasyuk, C. Vicent, D. G. Gusev, J. Am. Chem. Soc. 2015, 137, 3743–
Acknowledgements
3746; b) J. Zhang, G. Leitus, Y. Ben-David, D. Milstein, J. Am. Chem. Soc.
2005, 127, 10840–10841.
We would like to thank Dr. C. Fischer, S. Buchholz, S. Schareina,
K. Fiedler, and S. Rossmeisl (all at the Leibniz-Institut für Kata-
lyse e.V.) for excellent analytical and technical support. Finan-
cial support from the State of Mecklenburg–Vorpommern and
the BMBF is gratefully acknowledged. R.A. thanks the Ramon
Areces Foundation for a postdoctoral fellowship.
[9] B. Gnanaprakasam, Y. Ben-David, D. Milstein, Adv. Synth. Catal. 2010,
352, 3169–3173.
[10] C. Gunanathan, Y. Ben-David, D. Milstein, Science 2007, 317, 790–792.
[11] a) R. Lindner, B. van den Bosch, M. Lutz, J. N. H. Reek, J. I. van der Vlugt,
Organometallics 2011, 30, 499–510; b) M. Bassetti, Eur. J. Inorg. Chem.
2006, 4473–4482; c) P. Braunstein, F. Naud, Angew. Chem. Int. Ed. 2001,
40, 680–699; Angew. Chem. 2001, 113, 702–722.
[12] a) D. Cho, K. Chul Ko, J. Yong Lee, Organometallics 2013, 32, 4571–
4576; b) H. Li, X. Wang, F. Huang, G. Lu, J. Jiang, Z.-X. Wang, Organome-
tallics 2011, 30, 5233–5247; c) G. Zeng, S. Li, Inorg. Chem. 2011, 50,
10572–10580.
Keywords: homogeneous catalysis
ligands · ruthenium · synthesis design
· hydrogenation · N,P
[13] a) B. Gnanaprakasam, J. Zhang, D. Milstein, Angew. Chem. Int. Ed. 2010,
49, 1468–1471; Angew. Chem. 2010, 122, 1510–1513; b) for an analo-
gous switch of chemoselectivity in the dehydrogenative coupling of b-
amino alcohols, depending on the catalyst, either NNP-Ru or PNP-Ru,
see: B. Gnanaprakasam, E. Balaraman, Y. Ben-David, D. Milstein, Angew.
Chem. Int. Ed. 2011, 50, 12240–12244; Angew. Chem. 2011, 123, 12448–
12452.
[1] a) H. A. Younus, W. Su, N. Ahmad, S. Chen, F. Verpoort, Adv. Synth. Catal.
2015, 357, 283–330; b) C. Gunanathan, D. Milstein, Chem. Rev. 2014,
114, 12024–12087; c) Organometallic Pincer Chemistry (Eds.: G. van Ko-
ten, D. Milstein), Springer, Heidelberg, 2013; d) M. Albrecht, M. M. Lind-
ner, Dalton Trans. 2011, 40, 8733–8744; e) The Chemistry of Pincer Com-
pounds (Eds.: D. Morales-Morales, C. M. Jensen), Elsevier, Amsterdam,
2007.
[14] H. Li, X. Wang, M. Wen, Z.-X. Wang, Eur. J. Inorg. Chem. 2012, 5011–
5020.
[15] W. Xu, R. Langer, Dalton Trans. 2015, 44, 16785–16790.
[16] a) S. Werkmeister, K. Junge, B. Wendt, A. Spannenberg, H. Jiao, C. Born-
schein, M. Beller, Chem. Eur. J. 2014, 20, 4227–4231; b) K. Junge, B.
Wendt, F. A. Westerhaus, A. Spannenberg, H. Jiao, M. Beller, Chem. Eur.
J. 2012, 18, 9011–9018.
[17] For examples of pincer ligands containing one or more (benzo)imid-
azolyl unit, see: a) H. Chai, T. Liu, Q. Wang, Z. Yu, Organometallics 2015,
34, 5278–5284; b) K. Li, J.-L. Niu, M.-Z. Yang, Z. Li, L.-Y. Wu, X.-Qi Hao,
M.-P. Song, Organometallics 2015, 34, 1170–1176; c) W. Du, Q. Wang, L.
Wang, Z. Yu, Organometallics 2014, 33, 974–982; d) B. G. Hashiguchi,
K. J. H. Young, M. Yousufuddin, W. A. Goddard III, R. A. Periana, J. Am.
Chem. Soc. 2010, 132, 12542–12545; e) F. Zeng, Z. Yu, Organometallics
2009, 28, 1855–1862; f) F. Zeng, Z. Yu, Organometallics 2008, 27, 2898–
2901.
[2] R. Noyori, M. Yamakawa, S. Hashiguchi, J. Org. Chem. 2001, 66, 7931–
7944.
[3] a) J. R. Khusnutdinova, D. Milstein, Angew. Chem. Int. Ed. 2015, 54,
12236–12273; b) V. T. Annibale, D. Song, RSC Adv. 2013, 3, 11432–
11449; c) D. Gelman, S. Musa, ACS Catal. 2012, 2, 2456–2466; d) J. I. van
der Vlugt, J. N. H. Reek, Angew. Chem. Int. Ed. 2009, 48, 8832–8846;
Angew. Chem. 2009, 121, 8990–9004.
[4] a) C. Gunanathan, D. Milstein, Acc. Chem. Res. 2011, 44, 588–602; b) S.
Schneider, J. Meiners, B. Askevold, Eur. J. Inorg. Chem. 2012, 412–429.
[5] a) D. Spasyuk, S. Smith, D. G. Gusev, Angew. Chem. Int. Ed. 2012, 51,
2772–2775; Angew. Chem. 2012, 124, 2826–2829; b) J. Zhang, G.
Leitus, Y. Ben-David, D. Milstein, Angew. Chem. Int. Ed. 2006, 45, 1113–
1115; Angew. Chem. 2006, 118, 1131–1133.
[18] Heterocyclic Chemistry, 5th ed. (Eds.: J. A. Joule, K. Mills), Wiley, Hoboken,
2010.
[19] L. E. Kapinos, B. Song, H. Sigel, Inorg. Chim. Acta 1998, 280, 50–56.
[20] A related pincer imine ligand derived from the condensation of 1-
methyl-2-imidazolecarboxaldehyde and 2-(diphenylphosphino)ethyl-
[6] E. Balaraman, B. Gnanaprakasam, L. J. W. Shimon, D. Milstein, J. Am.
Chem. Soc. 2010, 132, 16756–16758.
[7] J. Zhang, E. Balaraman, G. Leitus, D. Milstein, Organometallics 2011, 30,
5716–5724.
Chem. Eur. J. 2016, 22, 4991 – 5002
www.chemeurj.org
5001
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
amine, and its NiII complex, have been reported. The latter has been ap-
plied as a catalyst in the oligomerization of ethylene: A. Boudier, P.-A. R.
Breuil, L. Magna, H. Olivier-Bourbigou, P. Braunstein, J. Organomet.
Chem. 2012, 718, 31–37.
[34] a) S. A. Lawerence, Amines: Synthesis, Properties and Applications, Cam-
bridge University Press, Cambridge, 2006; b) K. Eller, E. Henkes, E. R.
Rossbacher, H. Hoke, in Ullman’s Encyclopedia of Industrial Chemistry,
Wiley-VCH, Weinheim, 2000, Amines, Aliphatic. DOI: 10.1002/
14356007.a02_001.
[35] a) L. A. Oro, D. Carmona, J. M. Fraile, Metal Catalysis in Industrial Organic
Processes (Eds.: G. P. Chiusoli, P. M. Maitilis) pp. 79–113, RCS, London,
2006; b) L. Hegedu˝s, T. MµthØ, Appl. Catal. A 2006, 296, 209–215; c) P.
Kukula, M. Studer, H.-U. Blaser, Adv. Synth. Catal. 2004, 346, 1487–1493;
d) S. Nishimura, Handbook of Heterogeneous Catalytic Hydrogenation for
Organic Synthesis, Wiley, Hoboken, 2001, Ch. 7.
[21] a) ACS Select on First Row Transition Metal Complexes http://pubs.ac-
s.org/page/vi/2015/tm complexes; b) T. Zell, D. Milstein, Acc. Chem. Res.
2015, 48, 1979–1994; c) S. Chakraborty, P. Bhattacharya, H. Dai, H.
Guan, Acc. Chem. Res. 2015, 48, 1995–2003; d) J. I. van der Vlugt, Eur. J.
Inorg. Chem. 2012, 363–375; e) Catalysis without Precious Metals (Ed.:
R. M. Bullock), Wiley-VCH, Weinheim, 2010.
[22] J. R. Cabrero-Antonino, E. Alberico, H.-J. Drexler, W. Baumann, K. Junge,
H. Junge, M. Beller, ACS Catal. 2016, 6, 47–54.
[36] S. Gomez, J. A. Peters, T. Maschmeyer, Adv. Synth. Catal. 2002, 344,
1037–1057.
[37] J. Seyden-Penne, Reductions by the Alumino- and Borohydrides in Organ-
ic Synthesis, 2nd ed., Wiley, Hoboken, 1997.
[23] Likewise, a cationic complex was obtained when the pyridyl analogue
of ligand 1a was reacted with [RuHCl(CO)(PPh3)3], see ref. [15]. Indeed,
the use of [RuHCl(CO)(AsPh3)3], containing the more weakly coordinat-
ing AsPPh3 ligand, allowed Gusev et al. to completely displace the ancil-
lary ligands and obtain the neutral complex [RuHCl(CO)(NPyNHPiPr)], see
ref. [5a].
[24] For two recent examples of cationic pincer ruthenium complexes that
retain a PPh3 ligand and possess facial coordination of the pincer
ligand, see a) M. Hernµndez-Juµrez, M. Vaquero, E. lvarez, V. Salazar, A.
Suµrez, Dalton Trans. 2013, 42, 351–354; b) E. Fogler, E. Balaraman, Y.
Ben-David, G. Leitus, L. J. W. Shimon, D. Milstein, Organometallics 2011,
30, 3826–3833.
[25] a) P. S. Pregosin, NMR in Organometallic Chemistry, Wiley-VCH, Wein-
heim, 2012; b) O. Kühn, Phosphorus-31 NMR Spectroscopy, Springer, Hei-
delberg, 2008.
[26] In case of complex 5, the hydride signals of the three less abundant
isomers (in CD2Cl2) are increasingly shifted to lower fields than that of
the hydride signal of maj-5 at d=À14.81, À13.00, and À11.98 ppm;
see the Supporting Information.
[38] a) C. Gunanathan, M. Hçlscher, W. Leitner, Eur. J. Inorg. Chem. 2011,
3381–3386; b) X. Miao, J. Bidange, P. H. Dixneuf, C. Fischmeister, C. Bru-
neau, J.-L. Dubois, J.-L. Couturier, ChemCatChem 2012, 4, 1911–1916;
c) R. Reguillo, M. Grellier, N. Vautravers, L. Vendier, S. Sabo-Etienne, J.
Am. Chem. Soc. 2010, 132, 7854–7855; d) D. Addis, S. Enthaler, K.
Junge, B. Wendt, M. Beller, Tetrahedron Lett. 2009, 50, 3654–3656; e) S.
Enthaler, K. Junge, D. Addis, G. Erre, M. Beller, ChemSusChem 2008, 1,
1006–1010; f) S. Enthaler, D. Addis, K. Junge, G. Erre, M. Beller, Chem.
Eur. J. 2008, 14, 9491–9494; g) T. Li, I. Bergner, F. N. Haque, M. Zimmer-
De Iuliis, D. Song, R. H. Morris, Organometallics 2007, 26, 5940–5949;
h) S. Takemoto, H. Kawamura, Y. Yamada, T. Okada, A. Ono, E. Yoshikawa,
Y. Mizobe, M. Hidai, Organometallics 2002, 21, 3897–3904; i) R. P. Beatty,
R. A. Paciello, WO Patent WO/1996/23802–804, 1996; j) R. A. Grey, G. P.
Pez, A. Wallo, J. Am. Chem. Soc. 1981, 103, 7536–7542.
[39] T. Yoshida, T. Okano, S. Otsuka, J. Chem. Soc. Chem. Commun. 1979,
870–871.
[27] a) A. D. Buckingham, P. J. Stephens, J. Chem. Soc. 1964, 2747–2759;
b) T. G. Appleton, H. C. Clark, L. E. Manzer, Coord. Chem. Rev. 1973, 10,
335–422.
[28] X-ray structure analysis of complex syn-7 has been reported in ref. [22].
[29] T.-Z. Cheng, B. S. Brunschwig, R. M. Bullock, J. Am. Chem. Soc. 1998, 120,
13121–13137.
[40] C. Chin, B. Lee, Catal. Lett. 1992, 14, 135–140.
[41] S. Chakraborty, H. Berke, ACS Catal. 2014, 4, 2191–2194.
[42] K. Rajesh, B. Dudle, O. Blacque, H. Berke, Adv. Synth. Catal. 2011, 353,
1479–1484.
[43] a) S. Chakraborty, G. Leitus, D. Milstein, Chem. Commun. 2016, 52,
1812–1815; b) C. Bornschein, S. Werkmeister, B. Wendt, H. Jiao, E. Alber-
ico, W. Baumann, H. Junge, K. Junge, M. Beller, Nat. Commun. 2014, 5,
4111.
[44] A. Mukherjee, D. Srimani, S. Chakraborty, Y. Ben-David, D. Milstein, J.
Am. Chem. Soc. 2015, 137, 8888–8891.
[45] Complexes 3 and 4, in which ClÀ acts as a counterion and is not directly
coordinated to ruthenium, may not require any base to be activated.
The fact that these complexes are cationic may affect the energetics of
hydride tranfer to the substrate.
[30] For related pincer ruthenium complexes of the type Ru-NPyNHP, contain-
ing a pyridine moiety in place of an imidazolyl one, see: D. Spasyuk,
D. G. Gusev, Organometallics 2012, 31, 5239–5242; these complexes
have been successfully applied to the acceptorless dehydrogenative
coupling of ethanol and hydrogenation of esters and imines.
[31] For example, see: a) P. Li, B. de Bruin, J. N. H. Reek, W. I. Dzik, Organome-
tallics 2015, 34, 5009–5014; b) Y. Gloaguen, W. Jacobs, B. de Bruin, M.
Lutz, J. I. van der Vlugt, Inorg. Chem. 2013, 52, 1682–1684; c) S. Wan-
niarachchi, J. S. Hewage, S. V. Lindeman, J. R. Gardinier, Organometallics
2013, 32, 2885–2888; d) S. Wanniarachchi, B. J. Liddle, J. Toussaint, S. V.
Lindeman, B. Bennett, J. R. Gardinier, Dalton Trans. 2011, 40, 8776–
8787; e) E. Calimano, T. D. Tilley, Dalton Trans. 2010, 39, 9250–9263.
[32] a) D. B. Bagal, B. M. Bhanage, Adv. Synth. Catal. 2015, 357, 883–900;
b) D. S. MØrel, M. L. T. Do, S. Gaillard, P. Dupau, J.-L. Renaud, Coord.
Chem. Rev. 2015, 288, 50–68; c) S. Werkmeister, J. Neumann, K. Junge,
M. Beller, Chem. Eur. J. 2015, 21, 12226–12250; d) S. Werkmeister, K.
Junge, M. Beller, Org. Process Res. Dev. 2014, 18, 289–302.
[46] For the synthesis of BH4-7, see ref. [22].
[47] C. A. Sandoval, Y. Yamaguchi, T. Ohkuma, K. Kato, R. Noyori, Magn.
Reson. Chem. 2006, 44, 66–75.
[48] Z. Han, L. Rong, J. Wu, L. Zhang, Z. Wang, K. Ding, Angew. Chem. Int. Ed.
2012, 51, 13041–13045; Angew. Chem. 2012, 124, 13218–13222.
[33] a) J. Neumann, C. Bornschein, H. Jiao, K. Junge, M. Beller, Eur. J. Org.
Chem. 2015, 5944–5948; b) J.-H. Choi, M. H. G. Prechtl, ChemCatChem
2015, 7, 1023–1028.
Received: November 23, 2016
Published online on February 19, 2016
Chem. Eur. J. 2016, 22, 4991 – 5002
www.chemeurj.org
5002
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim