Novel Chiral PNNP Ligands with a Pyrrolidine Backbone – Application in the Fe-Catalyzed Asymmetric Transfer Hydrogenation of Ketones

Article abstract of DOI:10.1002/ejic.201900778

The PNNP ligand (R,R)-{PPh₂(2-C₆H₄)CH=N(pyrrolidine-NBn)-}₂ 2 was prepared by condensation between 2-diphenylphosphino benzaldehyde and the pyrrolidine-substituted diamine 1. Reduction with NaBH₄ in MeOH afforded (R,R)-{PPh₂(2-C₆H₄)CH–NH(pyrrolidine-NBn)-}₂ 3. The corresponding iron(II) complexes [FeCl₂(2)] (4), [Fe(CH₃CN)₂(2)](BF₄)₂ (5) and [Fe(CH₃CN)₂(2)](PF₆)₂ (6) were synthesized and fully characterized by NMR, ESI-HRMS, and EA. DFT calculations for complexes [FeCl₂(2)] (4) and [Fe(CH₃CN)₂(2)](BF₄)₂ (5) were carried out in order to investigate the stability of related structures. The PNNP ligands 2 and 3 in combination with Fe₃(CO)₁₂ as iron source were tested as catalysts in the asymmetric transfer hydrogenation of a variety of ketones with conversions higher than 95 % and enantioselectivities up to 97 %.

Full text of DOI:10.1002/ejic.201900778

DOI: 10.1002/ejic.201900778
Full Paper
Asymmetric Transfer Hydrogenation
Novel Chiral PNNP Ligands with a Pyrrolidine Backbone –
Application in the Fe-Catalyzed Asymmetric Transfer
Hydrogenation of Ketones
Elisabet Mercadé,[a] Ennio Zangrando,[b] Anna Clotet,[a] Carmen Claver,*[a,c] and
Cyril Godard*[a]
Abstract: The PNNP ligand (R,R)-{PPh2(2-C6H4)CH=N(pyrrol- ized by NMR, ESI-HRMS, and EA. DFT calculations for complexes
idine-NBn)-}2 2 was prepared by condensation between 2-di- [FeCl2(2)] (4) and [Fe(CH3CN)2(2)](BF4)2 (5) were carried out in
phenylphosphino benzaldehyde and the pyrrolidine-substi- order to investigate the stability of related structures. The PNNP
tuted diamine 1. Reduction with NaBH4 in MeOH afforded (R,R)- ligands 2 and 3 in combination with Fe3(CO)12 as iron source
{PPh2(2-C6H4)CH–NH(pyrrolidine-NBn)-}2 3. The corresponding were tested as catalysts in the asymmetric transfer hydrogen-
iron(II) complexes [FeCl2(2)] (4), [Fe(CH3CN)2(2)](BF4)2 (5) and ation of a variety of ketones with conversions higher than 95 %
[Fe(CH3CN)2(2)](PF6)2 (6) were synthesized and fully character- and enantioselectivities up to 97 %.
Introduction
Homogeneous catalyst progress has traditionally focused on
the second- and third-row transition metals in combination
with simple ligands.[1] In general, these catalysts provide high
turnover frequencies (TOF) and product selectivities which are
difficult to achieve by the less active first-row transition met-
als.[2] However, the limited availability of precious metals as well
as their high price and toxicity diminishes their attractiveness
and make desirable the search of more sustainable and eco-
nomically friendly alternatives.[3]
Chiral tetradentate ligands offer a great advantage in cataly-
sis due to their conformational and configurational rigidity aris-
ing from their chelation.[4,5] The first achiral tetradentate ligand
Scheme 1. Selected PNNP and macrocyclic N2P2 ligands.
containing a N2P2 donor scaffold was described by Rauchfuss
and co-workers in 1980.[6] Later, Pignolet and co-workers re-
ported the chiral version using the 2,2′-dimethyl-6,6′-diamino-
In 2010, Beller and co-workers reported the use of a chiral
PNNP ligand with [Et3NH][HFe3(CO)11] as iron source for the
asymmetric transfer hydrogenation of ketimines with enantio-
selectivities up to 98 %.[9] The group of Morris and co-workers
also developed a series of well-defined chiral Fe(II)–PNNP com-
plexes which provided high activities and enantioselectivities
for the asymmetric transfer hydrogenation of C=O[10] and
C=N[11] bonds.[12]
Modification in the diamine[13] backbone and the phos-
phine[14] framework lead to fine tuning of the properties of
these catalysts (Scheme 2). The complex [Fe(CH3CN)(CO)(A)]-
(BF4)2, also indicated as first generation, reduces acetophenone
with 63 % enantioselectivity and a TOF of 1600 h–1.[10] Later, it
was demonstrated that this type of complexes form metallic
nanoparticles under catalytic conditions.[15] Complexes
[Fe(Br)(CO)(B/C)](BPh4), which constitute the second generation,
hydrogenate acetophenone with ee's up to 90 % with TOF's of
30000 and 26000 h–1, respectively.[14b] The study of the mecha-
biphenyl as the chiral framework and its coordination to Rh(I)
and Ir(I).[5] However, the most important breakthrough was car-
ried out by Noyori, Gao and Ikariya in 1996, effectively applying
Ru–PNNP complexes in the asymmetric transfer hydrogenation
of ketones.[7] Recently, Mezzetti and co-workers prepared chiral
macrocyclic N2P2 ligands containing two stereogenic phos-
phorus donor atoms and their related Fe(II) complexes
(Scheme 1).[8]
[a] Departament de Química Física i Inorgànica, Universitat Rovira i Virgili,
Marcel.li Domingo s/n, 43007, Tarragona, Spain
E-mail: Cyril.godard@urv.cat
[b] Department of Chemical and Pharmaceutical Science, University of Trieste,
Via Giorgieri 1, Trieste, Italy
[c] Centre Tecnològic de la Química,
Marcel.li Domingo s/n, 43007, Tarragona, Spain
Supporting information and ORCID(s) from the author(s) for this article are
available on the WWW under https://doi.org/10.1002/ejic.201900778.
Eur. J. Inorg. Chem. 0000, 0–0
1
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
nism provided evidence for the reduction of one of the imine NaBH4 in MeOH afforded ligand 3 in quantitative yield
groups of the starting PNNP ligand during catalysis, giving a (Scheme 3). These new ligands were fully characterized by mul-
highly active catalyst containing an amido and ene-amido func- tinuclear NMR spectroscopy, ESI-HRMS and Elemental Analysis.
tionalities.[16] The proposed structure of the amido (ene-amido)
catalyst was further supported by reaction with acid providing
a coordinated ligand that contained an amine and imine func-
tions.[16] In view of the excellent results obtained with this set
of iron(II) complexes, the group of Morris and co-workers devel-
oped a third generation of iron(II) catalyst containing an imine
and an amine function.[17a] These new set of catalysts was ap-
plied in the asymmetric transfer hydrogenation of ketones and
imines with enantioselectivities up to 98 %. TOFs as high as
200 s–1 were obtained with these new P–N–NH–P iron(II) sys-
tems. It is worth mentioning that the third generation of iron(II)
catalysts provided greater TOF values than ruthenium- and os-
mium-based (R,S)-Josiphos systems and the ruthenium-based
P–NH–NH–P complexes, which provided a TOF of 89 and 92 s–1,
respectively.[18,19] Later, the same group showed in the asym-
metric transfer hydrogenation of acetophenone, the enantio-
meric excess ranged from 94 % (R) to 95 % (S) depending on
the nature of PR2 and whether the precatalyst contained an
imine or amine donor.[17b] It should be pointed out that for all
generations, the presence of CO in the precatalyst structure was
a prerequisite in order to obtain active catalysts.[10a]
Scheme 3. Synthesis of ligands 2 and 3. i) Benzylamine, xylene, reflux,
24 h. ii) LiAlH4, THF, reflux, 48 h. iii) HN3, PPh3, DIAD, toluene, r.t., overnight.
iv) Pd/C, 1atm H2, EtOH, overnight.
1
In the H NMR spectra of ligand 2, the formation of the li-
gand was confirmed by the detection of a doublet signal at
8.46 ppm corresponding to the CH=N moiety. In the case of
ligand 3, the disappearance of this signal and the detection of
a new signal at 3.84 ppm confirmed the reduction of the imine
fragment (CH2-NH), and thus the formation of ligand 3. In the
31P{1H} NMR spectrum, a singlet at –12.14 ppm appeared for
ligand 2 while that of 3 was observed at –16.21 ppm.
Scheme 2. Chiral PNNP reported by Morris and co-workers in the ATH of
ketones.
Synthesis of the Bis(chloride) Complex [FeCl2(2)] (4)
The pyrrolidine-based PNNP ligand reacts with FeCl2 in CH2Cl2
In view of these results, it was thought that the introduction
of a pyrrolidine moiety in PNNP ligands could potentially allow
the anchoring of the ligand through the functionalization of
the NH group. Here, we report a preliminary study including
the synthesis of the novel PNNP ligands containing a chiral
pyrrolidine backbone, their coordination chemistry with Fe(II)
and their application as catalysts in the asymmetric transfer
hydrogenation of ketones.
at room temperature to give the new paramagnetic neutral
complex [FeCl2(2)] (4) whose formation is indicated by a color
change from yellow to pink-red (Scheme 4).
Results and Discussion
Synthesis of the PNNP Ligands 2 and 3
The chiral PNNP ligand 2 was synthesized by condensation of
diamine 1, which is accessible from
steps, and the commercially available 2-diphenylphosphino
Scheme 4. Synthesis of [FeCl2(2)] (4).
L
-tartaric acid[20] in four
1
The H NMR spectrum showed paramagnetically shifted res-
benzaldehyde in 84 % yield. Reduction of the diimine with onances in the range –30 ppm to 110 ppm. Magnetic measure-
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
2
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
ments (Gouy's method) revealed a magnetic effective moment
μeff = 5.0. In the 31P{1H} NMR spectrum at room temperature, a
singlet signal was observed at –0.65 ppm, which chemical shift
is in the range of the uncoordinated phosphines. For this rea-
son, a low temperature NMR study was performed (Figure 1).
Figure 2. ORTEP drawing of complex [FeCl2(2)] (4) (ellipsoids at 50 % probabil-
ity). Solvent molecule omitted for clarity and only partial labeling scheme is
illustrated.
the ideal tbp geometry are evidenced in particular by the N(2)–
Fe–P(1) and N(1)–Fe–Cl(1) bond angles of 148.56 and 138.9°,
respectively. Similar geometries have been detected in other
structurally characterized complexes of type [(PNN)FeX2].[21] The
Fe–N(1) and Fe–N(2) bond lengths are equivalent [2.212(5) Å],
while the Fe–Cl distances are significantly different of
2.3719(17) vs. 2.2572(18) Å. Finally the Fe–P(1) bond length of
2.4729(15) Å is in agreement with values detected in other pre-
viously reported complexes of type [(PNN)FeCl2].[21] Selected
bond lengths and distances of complex [FeCl2(2)] (4) are listed
in Table 1.
Figure 1. 31P{1H} NMR spectrum of [FeCl2(2)] (4) at distinct temperatures.
(a) at r.t. (b) at –20 °C. (c) at –50 °C. (d) at –60 °C. (e) at –80 °C.
Table 1. Computed and experimental (Exp) selected bond lengths and non-
bonding distances (Å) and angles (deg) for tbp-N1N2 structure of [FeCl2(2H)]
and [FeCl2(2)] complexes.
Upon lowering the temperature to –20 °C, a new signal was
detected at 26.7 ppm (Figure 1b). When the temperature was
further lowered from to –80 °C (Figure 1c–e), the signal origi-
nally detected at –0.65 ppm became broader and was displaced
to higher chemical shifts whereas the intensity of the signal at
26.7 ppm became sharper and was displaced to lower chemical
shifts. These observations indicated that at room temperature,
complex 4 is involved in a fluxional process resulting in the
detection of an average signal at –0.65 ppm while at low tem-
perature, only one of the phosphine moieties of the ligand is
coordinated to the Fe center.
Crystals suitable for X-ray diffraction were obtained by va-
pour diffusion from CH2Cl2 in pentane at room temperature.
The X-ray diffraction analysis revealed the same coordination
mode of the ligand to the iron center in the solid state. Figure 2
shows the ORTEP view of complex [(PNN)FeCl2] (4).
FeCl2(2H)
d [Å]
FeCl2(2)
d [Å]
Exp.
d [Å]
Fe–N(1)
Fe–N(2)
Fe–Cl(1)
Fe–Cl(2)
Fe–P(1)
Fe–N(3)
Fe–P(2)
2.223
2.333
2.359
2.361
2.556
4.793
7.820
2.156
2.324
2.386
2.280
2.442
4.834
7.874
2.212(5)
2.212(5)
2.3719(17)
2.2572(18)
2.4729(15)
4.828(1)
8.306(1)
θ [°]
θ [°]
θ [°]
N(1)–Fe–N(2)
N(1)–Fe–Cl(1)
N(1)–Fe–Cl(2)
N(2)–Fe–Cl(1)
N(2)–Fe–Cl(2)
N(1)–Fe–P(1)
N(2)–Fe–P(1)
Cl(1)–Fe–Cl(2)
77.5
76.4
75.58(17)
138.89(15)
110.86(15)
92.81(13)
104.22(9)
79.94(13)
148.56(11)
110.21(7)
130.8
102.0
92.9
96.7
82.3
145.2
100.1
88.0
105.0
81.0
157.8
127.1
142.7
114.0
In the solid state the coordination geometry of the iron atom
in the chiral PNNP-iron complex 4 is best described as a
strongly distorted trigonal bipyramidal (tbp) where N2 and P1
In order to further investigate the stability of related struc-
ligand donors occupy the apical positions, while N1 and chlor- tures for complex 4, DFT calculations were carried out on the
ine atoms are located in the equatorial plane. Deviations from following coordination geometries: (a) a tbp structure (tbp-
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
3
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
N1N2) with one imine and the pyrrolidine nitrogen coordinated
In agreement with the experimental results, the most stable
to the iron center, (b) a symmetric tbp structure (tbp-N1N3) structure was the quintuplet tbp-N1N2 complex. Quintuplet
with both imines coordinated to the iron center, (c) an octahe- (2S + 1 = 5) is also the lowest energy state for the other struc-
dral (oct-N1N3) and (d) tetrahedral (td-N1N3) species (Figure 3). tures. The tbp-N1N3, oct-N1N3 and td-N1N3 structures fall in
Calculations were carried out using B3LYP functional, dichloro- a narrow energy range (5–7 kcal mol–1) (Table 2). When the
methane as solvent and for each geometry, various multiplici- phenyl substituents at the phosphorus atoms were included in
ties were considered. To reduce computational cost, the phenyl the computed structure for the singlet and quintuplet states
substituents on phosphorus atoms were first replaced with (Table 1 and Table 2), the quintuplet state of tbp-N1N2 struc-
hydrogen atoms (2H).
ture remained the most stable isomer while the oct-like (oct-
Figure 3. Atom labels for [FeCl2(2H)] (a) tbp-N1N2, (b) tpb-N1N3, (c) oct-N1N3, and (d) td-N1N3 structures. Hydrogen atoms are omitted for clarity. Phenyl
groups attached to P are not displayed.
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
4
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
N1N3) structure lied 4.7 kcal mol–1 above in energy and in this attributed to an oxidation of the Fe during the course of the
case, the singlet state was preferred. With these parameters, the crystallization since no sign of paramagnetism was observed in
energy of the quintuplet state of tbp-N1N3 structure compared the NMR spectra of the complex 6. The molecular structure of
to that of the tbp-N1N2 increases by 13 kcal mol–1. Interest- this complex is displayed in Figure 4, while the coordination
ingly, only a singlet structure very high in energy converged for bond lengths and angles are reported in Table 3.
the td-like structure, indicating that this type of coordination
geometry can be discarded.
Table 2. Computed relative energies [kcal mol–1] of complexes [FeCl2(2H)] and
[FeCl2(2)] with different multiplicities (2S + 1).
2S + 1 tbp-N1N2
oct-N1N3
tbp-N1N3 td-N1N3
[FeCl2(2H)]
[FeCl2(2)]
1
3
5
7
29.7
14.0
0.0
14.8
19.3
7.1
33.3
22.8
7.2
48.1
38.4
5.1
42.5
56.0
53.2
57.0
1
5
19.8
0.0
4.7
7.2
28.5
13.0
44.1
–
Synthesis of the bis(acetonitrile) complexes [Fe(CH3CN)2(2)]-
(BF4)2 (5) and [Fe(CH3CN)2(2)](PF6)2 (6). The pyrrolidine-base
PNNP ligand 2 reacts with Fe(BF4)2·6H2O in CH3CN to form the
diamagnetic dicationic complex 5 (81 % yield) whose formation
is indicated by a color change from yellow to orange (Scheme 5).
In the presence of NaPF6, [Fe(CH3CN)2(2)](BF4)2 (5) reacts in
CH2Cl2 at room temperature to provide the diamagnetic dicat-
ionic complex [Fe(CH3CN)2(2)](PF6)2 (6) in 90 % yield (Scheme 5).
The 31P{1H} NMR spectra of both complexes contained two mu-
tually coupled doublet signals at ca. 51 and 52 (2JP-P′ = 42–44 Hz),
indicating the presence of two slightly inequivalent phosphine
moieties in the complex. These chemical shifts were similar to
those observed by Morris and co-workers for the iron(II) PNNP
complexes [Fe(MeCN)2(A)] (BF4)2 (51.8), although in that case, a
singlet signal was reported.[10c]
Figure 4. ORTEP drawing of the complex [Fe(CH3CN)2(2)](PF6)3 (ellipsoid prob-
ability at 50 %). Solvent molecules and anions are omitted and only a partial
labeling scheme is shown for clarity.
Table 3. Selected bond lengths [Å] and angles [deg] for complex
[Fe(CH3CN)2(2)](PF6)3.
Bond lengths
Fe–N(1)
Fe–N(2)
Fe–N(4)
2.042(4)
2.039(4)
1.905(6)
Fe–N(5)
Fe–P(1)
Fe–P(2)
1.923(5)
2.2573(14)
2.2800(13)
Bond angles
N(1)–Fe–N(2)
N(4)–Fe–N(5)
P(1)–Fe–P(2)
N(1)–Fe–N(4)
N(1)–Fe–N(5)
N(2)–Fe–N(4)
N(2)–Fe–N(5)
N(1)–Fe–P(1)
83.96(17)
N(1)–Fe–P(2)
N(2)–Fe–P(1)
N(2)–Fe–P(2)
N(4)–Fe–P(1)
N(4)–Fe–P(2)
N(5)–Fe–P(1)
N(5)–Fe–P(2)
172.88(14)
173.3(2)
97.51(5)
87.9(2)
86.5(2)
86.6(2)
172.32(12)
89.76(11)
91.12(16)
88.35(14)
92.52(16)
96.73(15)
89.1(2)
88.63(13)
In this complex, the iron center exhibits a distorted octahedral
geometry where equatorial positions are occupied by the P2N2
donor set of the ligand 2 and linearly coordinated CH3CN mole-
cules are at axial positions. The Fe–N(imine) bond lengths are
comparable within their esd's [2.042(4) and 2.039(4) Å], as well
Scheme 5. Formation of [Fe(CH3CN)2(2)](BF4)2 (5) and [Fe(CH3CN)2(2)](PF6)2
(6).
Crystals suitable for X-ray diffraction were obtained by vapour as the Fe–NCCH3 distances, of 1.905(6) and 1.923(5) Å. On the
diffusion of pentane into a CH2Cl2 solution containing complex other hand, the values of the Fe–P bond lengths are significantly
6. To our surprise, the X-ray structural analysis revealed a Fe(III) different [2.2573(14) and 2.2800(13) Å]. This Fe(III) complex crys-
complex, [Fe(CH3CN)2(2)](PF6)3, where the ligand 2 is coordinated tallizes with a disordered lattice molecule of pentane beside the
to the iron center in a tetradentate manner. This fact is tentatively PF6 anions.
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
5
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Attempted Synthesis of the Acetonitrilecarbonyl Complex
[Fe(CH3CN)(CO)(2)](BF4)2 (7)
ter [Et3NH][HFe3(CO)11] are used. In view of the results obtained
with the iron carbonyl clusters, other iron(0) carbonyl precursors
were tested (entries 6 and 7). Nevertheless, when Fe2(CO)9 or
Fe(CO)5 were used, no conversion was obtained. When Fe(CO)5
was used with a ligand/Fe ratio of 1:3 (entry 8) or a combination
of Fe(CO)5/Fe2(CO)9 (1:1) (entry 9), no relevant conversion was
achieved either. It should be mentioned that Gao and co-workers
reported similar observation in the asymmetric transfer hyd-
rogenation of ketones catalyzed by PNHNHP ligands.[22]
The reaction was carried out using the standard procedure de-
scribed for this transformation.[10a] The complex [Fe(CH3CN)2(2)]-
(BF4)2 (5) was reacted in acetone under CO atmosphere (1atm)
at room temperature for 16 h. However, the resulting 31P{1H}
NMR spectrum only revealed the presence of the starting ligand
2. Various strategies were tested to obtain complex 7, however,
all attempts proved unsuccessful (Scheme 6).
Table 4. Asymmetric transfer hydrogenation of isobutyrophenone catalyzed
by Fe/2. Screening of iron precursors.[a]
Entry
Iron precursor
Conv. [%][b]
ee [%][b]
Scheme 6. Attempted synthesis of complex [Fe(CH3CN)(CO)(2)](BF4)2 (7).
1
2
3
4
5
6
7
8
9
Fe(BF4)2·6H2O
0
97
0
–
96
–
Fe3(CO)12
Fe3(CO)12
We performed DFT calculations, using B3LYP functional and
acetone as solvent, for different structures and multiplicities of
the complexes [Fe(CH3CN)2(2H)]2+ and [Fe(CH3CN)(CO)(2H)]2+. The
most stable structure is the low spin (singlet) oct structure for
both complexes, in agreement with experimental results for
bis(acetonitrile) complex. The quintuplet-tbpN1N2 structure only
lies 5.8 kcal mol–1 higher in energy for [Fe(CH3CN)2(2H)]2+ but is
[c]
–
0
–
[Et3NH][HFe3(CO)11
Fe2(CO)9
]
98
< 5
0
8
< 5
97
85
–
50
10
[d]
Fe(CO)5
Fe(CO)5
[e]
Fe2(CO)9/Fe(CO)5
20.0 kcal mol–1 higher in the case of [Fe(CH3CN)2(CO)(2H)]2+
.
[a] General conditions: Fe/2/KOH = 1:1:8; 0.0068 mmol of Fe, 0.0068 mmol of
2, 0.054 mmol of KOH, 0.136 mmol of isobutyrophenone, 3 mL of iPrOH,
time = 16 h, temperature = 70 °C. [b] Conversion and enantiomeric excess
was calculated by GC using a ChirasilDex CB column. [c] No ligand used. [d]
Ligand/Fe = 1:1. [e] Ligand/Fe = 1:3.
Other computed structures of [Fe(CH3CN)2(2H)]2+ calculated
were between 9 and 19 kcal mol–1 higher in energy than the
singlet oct structure (see supporting information for details).
Finally, the computed free energy for the substitution reaction
(Scheme 4) is of –3.2 kcal mol–1. Consequently, from a thermo-
The nature and the amount of base in the catalytic reaction
dynamic point of view, [Fe(CH3CN)(CO)(2H)]2+ could be obtained. using the iron carbonyl cluster Fe3(CO)12 as iron source were ex-
It was therefore concluded that the impossibility to obtain this amined (Table 5). The best results were obtained when KOH was
complex from 5 must arise from kinetic issues. Unraveling kinetic used as base, with 97 % conversion and 96 % of enantioselec-
aspects is out of the scope of the present work.
tivity (entry 1). The use of NaOH considerably lowered the con-
version whereas the enantioselectivity was maintained (entry 2).
When the basicity was increased, lower conversions were also
observed although the enantiomeric ratio remained high (entry
3). However, when NaOiPr was used, no activity was observed
(entry 4). Decreasing the amount of KOH from a ratio of 1:8 to
1:1 had a strong impact on the activity while the enantioselec-
tivity slightly decreased (entry 7–9).
The effect of the temperature and catalyst loading was also
examined obtaining an optimal temperature of 70 °C and 5 mol-
% of catalyst loading (see supporting information for details).
Monitoring of the performance of the catalytic systems con-
Asymmetric transfer hydrogenation (ATH) of ketones. The
bis(acetonitrile) complexes [Fe(CH3CN)2(2)](BF4)2 (5) and
[Fe(CH3CN)2(2)](PF6)2 (6) were tested as catalyst precursors in the
asymmetric transfer hydrogenation of ketones in isopropanol.
However, both complexes were inactive in this transformation, in
agreement with the reports by Morris and co-workers.[14a] These
authors indeed reported that a carbonyl ligand is necessary to
make these types of iron complexes active.
The use of in situ catalytic systems using the PNNP ligand 2
with various iron precursors was therefore investigated for the
asymmetric transfer hydrogenation of isobutyrophenone
(Table 4). As expected, using Fe(BF4)2·6H2O (entry 1) as precursor, taining the PNNP ligands 2 and 3 using 5 mol-% of catalyst
no conversion was observed. However, the use of the iron carb- loading, Fe3(CO)12 as iron source, at 70 °C of temperature and
onyl cluster Fe3(CO)12 as metal source achieved high conversion 16 h of reaction time is depicted in Scheme 7. It should be
and excellent enantioselectivity for this transformation (entry 2). pointed out that both systems exhibited similar enantioselectivi-
It is important to note that for blank experiments, namely in the ties (≈ 97 %) while displayed very distinct activities. Comparing
absence of ligand (entry 3) or of iron source (entry 4), no reaction the initial turnover frequencies (TOF h–1) of both systems at ca.
was observed. When the iron hydride carbonyl cluster complex 15 % of conversion, the Fe3(CO)12/2 catalytic system is three
[Et3NH][HFe3(CO)11] was tested, isobutyrophenone was trans- times more active (TOF: 4.8 h–1) than its diamino analogue
formed with 98 % conversion and 97 % ee (entry 5). This feature {Fe3(CO)12/3} (TOF: 1.5 h–1). After 500 minutes, the Fe3(CO)12/2
might suggest that a similar active catalyst is formed when the system provided a conversion of 95 % whereas the Fe3(CO)12/3
iron carbonyl cluster Fe3(CO)12 or the iron hydride carbonyl clus- system only exhibited ca. 40 % conversion.
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
6
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Table 5. Asymmetric transfer hydrogenation of isobutyrophenone catalyzed
by Fe3(CO)12/2. Optimization of the base.[a]
Entry
Base
L/Base
Conv.[%][b]
ee [%][b]
1
2
3
4
5
6
7
8
9
KOH
NaOH
KOtBu
NaOiPr
–
KOH
KOH
KOH
KOH
1:8
1:8
1:8
1:8
–
1:10
1:4
1:2
1:1
97
46
70
0
96
91
86
–
0
0
98
38
32
25
97
89
86
83
[a] General conditions: Fe3(CO)12/2/KOH= 1:1:8; 0.0068 mmol of Fe3(CO)12
,
0.0068 mmol of 2, 0.054 mmol of KOH, 0.136 mmol of isobutyrophenone,
3 mL of iPrOH, time = 16 h, temperature = 70 °C. [b] Conversion and enantio-
meric excess was calculated by GC using a ChirasilDex CB column.
Figure 5. Scope and limitations of the asymmetric transfer hydrogenation of
a variety of ketones catalyzed by the Fe3(CO)12/2 system. General conditions:
Fe3(CO)12/2/KOH= 1:1:8; 0.0068 mmol of Fe3(CO)12, 0.0068 mmol of 2,
0.054 mmol of KOH, 0.136 mmol of Ketone3 mL of iPrOH, time = 16 h, tem-
perature = 70 °C. Conversion and enantiomeric excess were calculated by GC
using a ChirasilDex CB column.
phenyl ketones and ketones containing annulated-rings was car-
ried out. The reduction of cyclobutyl phenyl ketone provided
high conversion and enantioselectivities. However, when more
flexible rings were located in the α-position of the ketone, such
as cyclopentyl and cyclohexyl phenyl ketone, lower conversions
were observed although the ee's remained unchanged or even
increased in the case of cyclohexyl phenyl ketone up to 97 %.
Finally, when the 3,4-dihydronaphthalen-1(2H)-one was em-
ployed as substrate, low conversion was obtained and the enan-
tioselection was moderate.
Scheme 7. Catalytic performance of Fe3(CO)12/2 vs. Fe3(CO)12/3.
Asymmetric Transfer Hydrogenation (ATH) of Ketones:
Substrate Scope
Mechanistic Observations
Based on the reaction conditions, the scope and limitation of
the Fe3(CO)12/2 catalytic system were explored in the asymmetric Previous studies conducted by the groups of Morris[15,23] and
transfer hydrogenation of a variety of ketones. The results are Gao[24] with similar Fe-N2P2 and Fe3(CO)12/(NH)4P2 (Scheme 8)
summarized in Figure 5.
catalytic systems, have proven to be heterogeneous rather than
homogeneous, involving iron nanoparticles as the active catalytic
species. Various poisoning agents such as phosphines, sulfides,
or mercury have been used in order to differentiate an homoge-
neous process from a heterogeneous one.[25] In this regard, we
have chosen Hg(0) to carry out the poisoning experiment, a
poison usually used to distinguish homogeneous catalysis from
catalysis by metal particles.
The hydrogenation of acetophenone provided high conver-
sion towards the corresponding alcohol together with a modest
enantioselectivity (35 % ee). Upon increasing the steric bulk at
the aliphatic position of the ketone, an increase of the enantio-
meric excess was observed (Me < Et < iPr = tBu) with enantio-
selectivities up to 96 % in the case of isobutyrophenone. Follow-
ing the same trend, isovalerophenone exhibited high conversion
(94 %) and enantioselectivities in a range between isobutyro-
First of all, the reaction mixture is cloudy throughout the en-
phenone and tert-butyl phenyl ketone (73 % ee). It has to be tire ATH process, as previously seen by other groups.[24] In our
mentioned that replacing the phenyl group of the acetophenone case, the use of 3eq of Hg was sufficient to stop the ATH com-
by the naphthalene ring gave similar results in terms of conver- pletely. This result is in agreement with the previous work re-
sion as well as enantiomeric excess. Then, the reduction of cyclic ported by Gao and co-workers with similar in situ systems.[24]
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
7
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
atmosphere. CH3CN, iPrOH and MeOH were dried, distilled and de-
gassed using standard procedure.[26] Deuterated solvents were de-
gassed via three freeze-pump-thaw cycles and kept in the glovebox
over 4Å molecular sieves. Reagents were purchased from Aldrich, Alfa
Aesar and Strem. 1H, 13C, 31P, 11B and 19F NMR spectra were recorded
on a Varian Mercury VX 400 (400 MHz, 100.6 MHz, 162 MHz,
1
128.5 MHz and 376.8 MHz respectively). Chemical shift values for H
and 13C were referred to internal SiMe4 (0.0 ppm) and for 31P was
referred to H3PO4 (85 % solution in D2O, 0 ppm). 2D correlation spec-
tra (gCOSY, gHSQC and gHMBC) were visualized using VNMR pro-
gram (Varian®). Enantiomeric ratios were determined by GC analysis.
Gas chromatography analyses were carried out in a Hewlett-Packard
HP 6890 gas chromatograph, using the CP Chirasil-Dex CB chiral col-
umn. ESI-HMRS and EA were performed at the Serveis Tècnics de
Recerca from the Universitat de Girona (Spain).
Scheme 8. Previous heterogeneous Fe-N2P2 and Fe3(CO)12/(NH)4P2 catalytic
systems reported for the ATH and AH of ketones respectively.
Synthesis of (2): 2-(diphenylphosphino)benzaldehyde (0.155g,
0.53 mmol) was added to 4 mL of dry and degassed toluene pro-
vided with molecular sieves. Then, (3R,4R)-3,4-diamino-1-benzyl-
pyrrolidine (1) (49.7 mg, 0.26 mmol) was added to the mixture. This
solution was allowed to stir overnight at 90 °C. The reaction mixture
was filtered under nitrogen atmosphere and collected by vacuum
filtration. Recrystallization with cool MeOH in a glovebox afforded a
yellow solid in 84 % yield. 1H-NMR (CD2Cl2, 400 MHz, δ in ppm): 8.45
(d, 2H, J = 4Hz, HC=N), 7.80 (m, 2H, Ar-), 7.38 (m, 2H, Ar-), 7.33 (d,
4H, J = 8Hz, Ar-), 7.29–7.18 (m, 23H, Ar-), 6.85 (m, 2H, Ar-), 3.59 (d,
1H, J = 4Hz, -CH2-Ph, -CH2-), 3.55 (d, 1H, J = 4Hz, -CH2-Ph, -CH2-),
3.48 (quint, 2H, J = 4Hz, -CH-), 2.70 (dd, 2H, J = 4 & 8Hz, -CH2-), 2.46
(dd, 2H, J = 4 & 8Hz, -CH2-). 13C{1H}-NMR (CD2Cl2, 100.6 MHz, δ in
ppm): 159.4 (d, JC-P = 16Hz, HC=N), 134.2, 134.0, 133.8, 133.5, 129.9,
128.8, 128.7, 128.5, 128.4, 128.3, 128.1, 126.8, overlapped signals, 75.8
(-CH-), 60.4 (-CH2-), 60.2 (-CH2Ph). 31P{1H}-NMR (CD2Cl2, 161MHz, δ in
ppm): –12.14. ESI-HRMS: Calculated for C49H43N3P2: Exact: (M: 735.29;
[M + H]+: 736.29); Experimental ([M + H]+: 736.2997). Elemental Anal-
ysis: Calculated for C49H43N3P2: C, 79.98; H, 5.89; N, 5.71; found C,
79.09; H, 5.73; N, 5.37.
Consequently, it is suggested that the ATH catalyzed by
Fe3(CO)12/2 is probably heterogeneous, having 2-modified iron
particles acting as active catalyst.
Conclusions
The novel PNNP ligands 2 and 3 containing a pyrrolidine scaffold
have been synthesized and fully characterized by NMR, ESI-HRMS
and EA. The incorporation of the pyrrolidine backbone could po-
tentially allow the anchoring of the ligand through the function-
alization of the NH group. The reaction between ligand 2 and
iron(II) precursors enables the synthesis of monomeric complexes
stable and paramagnetic in the case of complex [FeCl2(2)] (4) or
diamagnetic in the case of [Fe(CH3CN)2(2)] (BF4)2 (5) and
[Fe(CH3CN)2(2)] (PF6)2 (6). The obtained iron(II) complexes were
characterized by NMR, ESI-HRMS, EA and by X-ray diffraction for
4 and 6. DFT calculations were carried out for complex [FeCl2(2)]
(4) in order to investigate the stability of different coordination
isomers, obtaining the quintuplet tbp-N1N2 as the most stable
complex. Similarly, theoretical computation for [Fe(CH3CN)2(2)]
(BF4)2 (5) indicated the low spin (singlet) oct-N1N3 complex as
the most stable species. The impossibility to obtain complex
[Fe(CH3CN)(CO)(2)] (BF4)2 (7) was attributed to kinetic arguments.
The PNNP ligands 2 and 3 were used in combination with
Fe3(CO)12 in the asymmetric transfer hydrogenation of a variety
of ketones and provided low to high conversions (21–97 %) and
enantioselectivities (32–97 %) depending of the substrate. Upon
increasing the steric bulk at the aliphatic position of the ketone,
higher enantioselectivities were achieved, up to 97 % when iso-
butyrophenone was used as substrate. From a mechanistic point
of view, it is suggested that the Fe3(CO)12/2 system is probably
heterogeneous, in nature. The introduction of the pyrrolidine
thus did not affect substantially the asymmetric induction, when
compared with the results previously reported using related li-
gands.
Synthesis of (3): (2) (90 mg, 0.12 mmol) was dissolved in 5 mL of
dry methanol. Then, NaBH4 (28 mg, 0.73 mmol) was added under
nitrogen atmosphere. The reaction was refluxed with stirring under
nitrogen for 24 h. The solution was cooled to room temperature and
deoxygenated H2O (3 mL) was added to destroy the excess of NaBH4.
The mixture solution was extracted with deoxygenated CH2Cl2 (3 mL
× 3). The organic layer was dried with anhydrous MgSO4 and evapo-
ration under vacuum afforded the product in 95 % yield. 1H-NMR
(CD2Cl2, 400 MHz, δ in ppm): 7.45–7.42 (m, 3H, Ar-), 7.32–7.19 (m,
26H, Ar-), 7.16–712 (m, 2H, Ar-), 6.87–6.84 (m, 2H, Ar-), 3.84 (qd, 4H,
J = 12 & 4Hz, CH2-NH, -CH2-), 3.43 (s, 2H, -CH2-Ph, -CH2-), 2.80 (quint,
2H, J = 4Hz, -CH-), 2.55 (dd, 2H, J = 4 & 12Hz, -CH2-), 2.11 (dd, 2H,
J = 4 & 12Hz, -CH2-). 13C{1H}-NMR (CD2Cl2, 100.6 MHz, δ in ppm):
2
145.0 (d, JC-P = 24.1Hz, Ar-, -C-), 139.4 (CH2Ph, Ar-, -C-), 137.2 (d,
1
1JC-P = 11.1Hz, PPh2, Ar-, -C-), 137.1 (d, JC-P = 11.1Hz, PPh2, Ar-, -C-),
1
135.9 (d, JC-P = 11.1Hz, Ar-, -C-), 133.9 (d, J = 2Hz, Ar-, -CH-), 133.8
(d, J = 2Hz, Ar-, -CH-), 133.5 (Ar-, -CH-), 129.3 (d, J = 5Hz, Ar-, -CH-),
128.9 (Ar-, -CH-), 128.8 (Ar-, -CH-), 128.7 (d, J = 3Hz, Ar-, -CH-), 128.6
(d, J = 3Hz, Ar-, -CH-), 128.5 (d, J = 3Hz, Ar-, -CH-), 128.1 (Ar-, -CH-),
127.1 (Ar-, -CH-), 126.7 (Ar-, -CH-), overlapped signals, 64.3 (CH-NH,
3
-CH-), 60.2 (CH2-Ph, -CH2), 59.7 (CH2-CH-NH, -CH2), 50.8 (d, JC-P
=
21.1Hz, CH2-NH, -CH2-). 31P{1H}NMR (CD2Cl2, 161MHz, δ in ppm):
–16.21. ESI-HRMS: Calculated for C49H45N3P2: Exact: (M: 739.32; [M +
H]+: 740.32); Experimental ([M + H]+: 740.37).
Experimental Section
General Considerations: All preparations and manipulations were
carried out under an oxygen-free nitrogen atmosphere using conven-
tional Schlenk techniques or inside a glovebox. Solvents were puri-
fied by the system Braun MB SPS-800 and stored under nitrogen
Synthesis of [Fe(Cl)2(2)] (4): Iron dichloride anhydrous (12 mg, ex-
cess) dissolved in 1 mL of CH2Cl2 was added to a solution of (R,R)-
{PPh2(2-C6H4)CH=N(pyrrolidine-NBn)-}2 (2) (50 mg, 0.068 mmol) in
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
8
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
3 mL of CH2Cl2. The mixture was allowed to stir at room temperature
for 48 h to which a change of color was appreciated from pale yellow
to pink-red. The mixture was filtered via cannula under nitrogen at-
mosphere and ether was added to the pink-red solution to force the
product precipitation. The product was isolated as a pink-red solid in
70 % of yield. Crystals suitable for X-ray diffraction were ground by
Procedure for the Asymmetric Transfer Hydrogenation of Iso-
butyrophenone in Presence of Mercury: Fe3(CO)12 (3.4 mg,
0.0068 mmol), (2) (5 mg, 0.0068 mmol) were placed in a tube
equipped with a Teflon-coated magnetic stirring bar. Isopropyl alco-
hol (3 mL) was then added and the mixture was stirred at 70 °C for
30 min. KOH (3.05 mg, 0.054 mmol) was then added, and the mixture
was continually stirred for another 15 min. Isobutyrophenone
(20.4 μL, 0.136 mmol) was then introduced and the mixture was
stirred at 70 °C for 5 h. An aliquot was taken under N2 and analyzed
1
vapor diffusion of pentane into CH2Cl2 solution of the complex. H-
NMR (CD2Cl2, 400 MHz, δ in ppm): 105.08, 85.55, 72.60, 60.50, 16.48,
15.43, 12.66, 10.33, 9.94, 8.04, 7.56, 7.32, 7.22, 6.87, 6.72, 6.54, 6.17,
6.02, –1.01, –7.17, –30.37. 31P{1H}-NMR at r.t. (CD2Cl2, 161MHz, δ in by GC using the ChirasilDex CB chiral column. At the same time, Hg
ppm): –0.694. 31P{1H}-NMR at –80 °C (CD2Cl2, 161MHz, δ in ppm):
23.83 & 5.32 (bs). ESI-HRMS: Calculated for C49H43FeN3P2Cl2: Exact:
(M: 861.1659, M – Cl: 826.1970); Experimental (M – Cl: 826.1986).
Elemental Analysis: Calculated for C49H43FeN3P2Cl2·0.5CH2Cl2: C,
65.69; H, 4.90; N, 4.64; found C, 65.89; H, 4.66; N, 4.16.
was added to the reaction and left to stir at 70 °C for 10 h. After
cooling down to room temperature, a sample was carefully taken
from the top of the reaction mixture, the rest of which was quenched
with sulfur and collected in a special bottle.
Computational Details: All calculations were performed using Gaus-
sian 09.[27] For [FeLL′(2H)] complexes, B3LYP density functional was
used and the basis set was SDD for iron atom (including relativistic
core potentials) and 6-31++G(d,p) for Cl, P, N, O, C and H atoms. We
used B97D density functional (which includes dispersion) for [FeCl2(2)]
complex calculations and all atoms were treated with 6-31++G(d,p) all
electron basis set. Solvent effects were included using the Polarizable
Continuum Model (PCM) with the integral equation formalism variant
(IEFPCM). Solvent was dichloromethane for bis(chloride) complex and
acetone for acetonitrile complexes, CO and CH3CN free molecules.
Calculated structures were generated by J mol.[28]
Synthesis of [Fe(MeCN)2(2)](BF4)2 (5): (2) (0.135 mmol) in 4 mL of
MeCN was added to a solution of Fe(BF4)2·6H2O (0.133 mmol) in
1 mL of MeCN. After stirring for 3 h, the red solution was evaporated
to dryness. Then, 1 mL of CH2Cl2 was added to re-dissolve the deep
red powder. Ether (8 mL) was added to the solution and a precipitate
started to form. The solid was filtered under nitrogen via cannula
and washed repeatedly with ether to afford an orange solid in 81 %
yield. 1H-NMR (CD2Cl2, 400 MHz, δ in ppm): 8.82–8.77 (m, 2H, CH=
CN), 8.06–7.96 (m, 3H, Ar-), 7.77–7.21 (m, 21H, Ar-), 6.99 (t, 1H, J =
8Hz, Ar-), 6.88 (t, 1H, J = 8Hz, Ar-), 6.79–6.68 (m, 5H, Ar-), 6.43 (t, 2H,
J = 8Hz Ar-), 5.32 (s, 2H, CH2-Ph, -CH2-), 4.98–4.92 (m, 1H, -CH-), 4.80–
4.71 (m, 2H, -CH2-), 4.55–4.50 (m, 1H, -CH2-), 4.19–4.07 (m, 2H, -CH2-),
1.87 (s, 3H, CH3CN, -CH3), 1.76 (s, 3H, CH3CN, -CH3). 31P{1H}-NMR
X-ray Crystallography: X-ray diffraction data for compounds 4 and
6 were collected at the X-ray diffraction beamline of synchrotron
Elettra (Trieste) at 100(2) K, with λ = 0.7000 and 0.8000 Å, respec-
tively. Cell refinement, indexing, and scaling of the data sets were
carried out using the CCP4 software suite.[29] Both the structures
were solved by direct methods and refined by a full-matrix least-
squares procedure based on F2 using SHELXL-97.[30]
(CD2Cl2, 161MHz, δ in ppm): 52.32 (d, JP-P = 42Hz) and 51.65 (JP-P
=
42Hz). 11B{1H}-NMR (CD2Cl2, 128.51MHz, δ in ppm): –1.03. 19F{1H}-
NMR (CD2Cl2, 376.8MHz, δ in ppm): –150.05. ESI-HRMS: Calculated for
C53H49B2F8FeN5P2: Exact: [M: 1047.2871, M – (CH3CN)2(BF4)2/Z:
395.6136]; Experimental [M – (CH3CN)2(BF4)2:/Z: 395.6146]. Elemental
Analysis: Calculated for C53H49B2F8FeN5P2·3CH2Cl2: C, 51.65; H, 4.26;
N, 5.38; found C, 51.62; H, 4.49; N, 5.61.
The ΔF map revealed a molecule of CH2Cl2 (refined at half occu-
pancy) in 4, while a disordered penthane molecule and a residual
refined as water (H atoms not fixed) were detected in the unit cell
of 6. All non-hydrogen atoms were refined anisotropically. The
hydrogen atoms were placed at calculated positions and treated
using the appropriate riding models. Crystal data and details of
refinement are given in Supporting information.
Synthesis of [Fe(MeCN)2(2)](PF6)2 (6): NaPF6 (16 mg, 0.095 mmol)
was added to a solution of 5 (50 mg, 0.047 mmol) in dichloro-
methane (4 mL). After stirring for 1 h the resulting orange mixture
was filtered through a small pad of Celite and the solvents evapo-
1
rated to dryness to give 6 as an orange solid in 90 % yield. H-NMR
(CD2Cl2, 400 MHz, δ in ppm): 8.80 (br s, 1H, CH=CN), 8.71 (br s, 1H,
CH=CN), 8.04–7.97 (m, 2H, Ar-), 7.78–7.20 (m, 22H, Ar-), 7.02 (t, 1H,
J = 8Hz, Ar-), 6.87 (t, 1H, J = 8Hz, Ar-), 6.80–6.71 (m, 5H, Ar-), 6.36 (t,
2H, J = 8Hz, Ar-), 5.21–5.14 (m, 1H, -CH-), 4.82–4.76 (m, 4H, -CH2-),
4.55–4.50 (m, 1H, -CH-), 4.13–4.08 (m, 2H, -CH2-), 1.86 (s, 3H, CH3CN,
-CH3), 1.75 (s, 3H, CH3CN, -CH3). 31P{1H}-NMR (CD2Cl2, 161MHz, δ in
ppm): 51.98 (d, JP-P = 43.7Hz), 51.13(d, JP-P = 43.7Hz), –144.3 (sept,
CCDC 1941308 (for 4), and 1941306 (for 6) contain the supplemen-
tary crystallographic data for this paper. These data can be obtained
free of charge from The Cambridge Crystallographic Data Centre.
Supporting Information (see footnote on the first page of this
article): Associated details, product characterization data, crystallo-
graphic data for 4 and 6, and 1H, 13C, 31P, 11B and 19F spectra are
available.
J
P-F = 712.8Hz, PF6). 19F{1H}-NMR (CD2Cl2, 376.8MHz, δ in ppm): –71.5
(d, 6F, JF-P = 712.8Hz, PF6). ESI-HRMS: Calculated for C53H49F12FeN5P2:
Exact: [M: 1163.2096, M – (CH3CN)2(PF6)2/Z: 395.6136]; Experimental
[M – (CH3CN)2(PF6)2/Z: 395.6130]. Elemental Analysis: Calculated for
C53H49F12FeN5P2·3CH2Cl2: C, 47.42; H, 3.91; N, 4.94; found C, 47.32;
H, 3.21; N, 4.81.
Acknowledgments
General Procedure for the Asymmetric Transfer Hydrogenation
of Ketones: In a typical experiment, inside the glovebox Fe3(CO)12
(3.4 mg, 0.0068 mmol), (2) (5 mg, 0.0068 mmol) were placed in a
tube equipped with a Teflon-coated magnetic stirring bar. Isopropyl
alcohol (3 mL) was then added and the mixture was stirred at 70 °C
for 30 min. An appropriate amount of KOH was then added, and the
mixture was continually stirred for another 15 min. Ketone was then
introduced and the mixture was stirred at 70 °C for the required
reaction time. Conversion and enantiomeric ratios were calculated by
GC using the ChirasilDex CB chiral column.
The authors are grateful to the Spanish Ministerio de Economía
y Competitividad and the Fondo Europeo de Desarrollo Re-
gional FEDER (PhD grant BES-2011-043510 to E. M., CTQ2016-
75016-R, AEI/FEDER,UE), and the Generalitat de Catalunya
(2017SGR) for financial support.
Keywords: Homogeneous catalysis · N,P ligands · Ligand
design · Asymmetric catalysis
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
9
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[1] T. Ikariya, A. J. Blacker, Acc. Chem. Res. 2007, 40, 1300–1308.
[2] a) H. Doucet, T. Ohkuma, K. Murata, T. Yokozawa, M. Kozawa, E. Katayama,
A. F. England, T. Ikariya, R. Noyori, Angew. Chem. Int. Ed. 1998, 37, 1703–
1707; Angew. Chem. 1998, 110, 1792; b) K. Junge, K. Schroder, M. Beller,
Chem. Commun. 2011, 47, 4849–4859; c) R. M. Bullock, Science 2013,
342, 1054–1055; d) P. E. Sues, K. Z. Demmans, R. H. Morris, Dalton Trans.
2014, 43, 7650–7667; e) R. M. Bullock, Molybdenum and Tungsten Cata-
lysts for Hydrogenation, Hydrosilylation and Hydrolysis, in Catalysis with-
out Precious Metals (Wiley-VCH Verlag GmbH & Co. 2010).
[3] a) S. Enthaler, K. Junge, M. Beller, Angew. Chem. Int. Ed. 2008, 47, 3317–
3321; Angew. Chem. 2008, 120, 3363; b) G. Bauer, K. A. Kirchner, Angew.
Chem. Int. Ed. 2011, 50, 5798–5800; Angew. Chem. 2011, 123, 5918.
[4] For a review on structural aspects of tri- and tetradentate P- and P,N
donor ligands, see: a) J. C. Hierso, R. Amardeil, E. Bentabet, R. Broussier,
B. Gautheron, P. Meunier, P. Kalck, Coord. Chem. Rev. 2003, 236, 143–206;
b) Y.-Y. Li, Z.-R. Dong, J.-X. Gao, Curr. Top. Catal. 2007, 48, 2587–2598.
[5] T. L. Marxen, B. J. Johnson, P. V. Nilsson, L. H. Pignolet, Inorg. Chem. 1984,
23, 4663–4670.
[15]
[16]
[17]
J. F. Sonnenberg, N. Coombs, P. A. Dube, R. H. Morris, J. Am. Chem. Soc.
2012, 134, 5893–5899.
A. A. Mikhailine, M. I. Maishan, A. J. Lough, R. H. Morris, J. Am. Chem. Soc.
2012, 134, 12266–12280.
a) W. Zuo, A. J. Lough, Y. F. Li, R. H. Morris, Science 2013, 342, 1080–1083;
b) K. Z. Demmans, C. S. G. Seo, A. J. Lough, R. H. Morris, Chem. Sci. 2017,
8, 6531–6541.
V. Rautenstrauch, X. Hoang-Cong, R. Churlaud, K. Abdur-Rashid, R. H.
Morris, Chem. Eur. J. 2003, 9, 4954–4967.
W. Baratta, F. Benedetti, Z. A. Del, L. Fanfoni, F. Felluga, S. Magnolia, E.
Putignano, P. Rigo, Organometallics 2010, 29, 3563–3570.
J. Skarzewski, A. Gupta, Tetrahedron: Asymmetry 1997, 8, 1861–1867.
a) J. Zhang, M. Gandelman, D. Herrman, G. Leitus, L. J. W. Shimon, Y.
Ben-David, D. Milstein, Inorg. Chim. Acta 2006, 359, 1955–1960; b) B. M.
Schmiege, M. J. Carney, B. L. Small, D. L. Gerlach, J. A. Halfen, Dalton
Trans. 2007, 2547–2562; c) T. Zell, R. Langer, M. A. Iron, L. Konstantinov-
ski, L. J. W. Shimon, Y. Diskin-Posner, G. Leitus, E. Balaraman, Y. Ben-David,
D. Milstein, Inorg. Chem. 2013, 52, 9636–9649.
J.-S. Chen, L.-L. Chen, Y. Xing, G. Chen, W.-Y. Shen, Z.-R. Dong, Y.-Y. Li, J.-
X. Gao, Acta Chim. Sin. 2004, 62, 1745–1750.
J. F. Sonnenberg, R. H. Morris, Catal. Sci. Technol. 2014, 4, 3426–3438.
Y. Li, S. Yu, X. Wu, J. Xiao, W. Shen, Z. Dong, J. Gao, J. Am. Chem. Soc.
2014, 136, 4031–4039.
[18]
[19]
[20]
[21]
[6] J. C. Jeffery, P. A. Tucker, T. B. Rauchfuss, Inorg. Chem. 1980, 19, 3306–
3316.
[22]
[7] a) J. X. Gao, T. Ikariya, R. Noyori, Organometallics 1996, 15, 1087–1089;
b) R. Noyori, S. Hashiguchi, Acc. Chem. Res. 1997, 30, 97–102; c) Y.-Y. Li,
S.-L. Yu, W.-Y. Shen, J.-X. Gao, Acc. Chem. Res. 2015, 48, 2587–2598.
[8] a) R. Bigler, E. Otth, A. Mezzetti, Organometallics 2014, 33, 4086–4099; b)
R. Bigler, A. Mezzetti, Org. Lett. 2014, 16, 6460–6463; c) R. Bigler, R. Huber,
A. Mezzetti, Angew. Chem. Int. Ed. 2015, 54, 5171–5174; Angew. Chem.
2015, 127, 5260; d) L. De Luca, A. Mezzetti, Angew. Chem. Int. Ed. 2017,
56, 11949–11953; Angew. Chem. 2017, 129, 12111.
[9] S. Zhou, S. Fleischer, K. Junge, S. Das, D. Addis, M. Beller, Angew. Chem.
Int. Ed. 2010, 49, 8121–8125; Angew. Chem. 2010, 122, 8298.
[10] a) C. Sui-Seng, F. Freutel, A. J. Lough, R. H. Morris, Angew. Chem. Int. Ed.
2008, 47, 940–943; Angew. Chem. 2008, 120, 954; b) N. Meyer, A. J.
Lough, R. H. Morris, Chem. Eur. J. 2009, 15, 5605–5610; c) C. Sui-Seng,
F. N. Haque, A. Hadzovic, A. M. Pütz, V. Reuss, N. Meyer, A. J. Lough,
M. Zimmer-De Iuliis, R. H. Morris, Inorg. Chem. 2009, 48, 735–743; d) A.
Mikhailine, A. J. Lough, R. H. Morris, J. Am. Chem. Soc. 2009, 131, 1394–
1395.
[23]
[24]
[25]
[26]
J. A. Widegren, R. Finke, J. Mol. Catal. A 2003, 198, 317–341.
D. D. Perrin, W. L. F. Armarego, Purification of Laboratory Chemicals (3rd
ed., Pergamon Press, Oxford, 1989).
[27]
M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R.
Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Na-
katsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G.
Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hase-
gawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A.
Montgomery Jr., J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Broth-
ers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghava-
chari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega,
J. M. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J.
Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi,
C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski,
G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, Ö.
Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, D. J. Fox, Gaussian 09,
Revision D.01, Gaussian, Inc., Wallingford CT, 2009.
[11] A. A. Mikhailine, M. I. Maishan, R. H. Morris, Org. Lett. 2012, 14, 4638–
4641.
[12] For selected examples see: a) P. O. Lagaditis, A. J. Lough, R. H. Morris, J.
Am. Chem. Soc. 2011, 133, 9662–9665; b) D. E. Prokopchuk, J. F. Sonnen-
berg, N. Meyer, M. Zimmer-De Iuliis, A. J. Lough, R. H. Morris, Organome-
tallics 2012, 31, 3056–3064; c) D. E. Prokopchuk, R. H. Morris, Organome-
tallics 2012, 31, 7375–7385.
[13] a) A. A. Mikhailine, E. Kim, C. Dingels, A. J. Lough, R. H. Morris, Inorg.
Chem. 2008, 47, 6587–6589; b) A. A. Mikhailine, R. H. Morris, Inorg. Chem.
2010, 49, 11039–11044.
[28]
[29]
[30]
J mol: an open-source Java viewer for chemical structures in 3D. http://
www.jmol.org/.
Collaborative Computational Project, Number 4, Acta Crystallogr., Sect. D
2011, 67, 235–242.
G. M. Sheldrick, Acta Crystallogr., Sect. A 2008, 64, 112–122.
[14] a) P. O. Lagaditis, A. L. Lough, R. H. Morris, Inorg. Chem. 2010, 49, 10057–
10066; b) P. E. Sues, A. J. Lough, R. H. Morris, Organometallics 2011, 30,
4418–4431.
Received: July 18, 2019
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
10
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Asymmetric Transfer
Hydrogenation
E. Mercadé, E. Zangrando, A. Clotet,
C. Claver,* C. Godard* ................... 1–11
Novel Chiral PNNP Ligands with a
Pyrrolidine Backbone – Application
in the Fe-Catalyzed Asymmetric
Transfer Hydrogenation of Ketones
PNNP ligands containing a pyrrolidine asymmetric transfer hydrogenation of
backbone were prepared and their co- ketones, the compounds afforded
ordination chemistry to iron was inves- conversions higher than 95 % and
tigated through experimental and DFT enantioselectivities up to 97 % were
studies. When used in combination achieved.
with Fe3(CO)12 as iron source in the
DOI: 10.1002/ejic.201900778
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
11
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim