Enantioselective Hydrogenation of Ketones using Different Metal Complexes with a Chiral PNP Pincer Ligand

Article abstract of DOI:10.1002/adsc.201801511

The synthesis of different metal pincer complexes coordinating to the chiral PNP ligand bis(2-((2R,5R)-2,5-dimethyl-phospholanoethyl))amine is described in detail. The characterized complexes with Mn, Fe, Re and Ru as metal centers showed good activities regarding the reduction of several prochiral ketones. Comparing these catalysts, the non-noble metal complexes produced best selectivities not only for aromatic substrates, but also for different kinds of aliphatic ones leading to enantioselectivities up to 99% ee. Theoretical investigations elucidated the mechanism and rationalized the selectivity. (Figure presented.).

Full text of DOI:10.1002/adsc.201801511

Title: Enantioselective Hydrogenation of Ketones using Different Metal
Complexes with a Chiral PNP Pincer Ligand
Authors: Marcel Garbe, Zhihong Wei, Bianca Tannert, Anke
Spannenberg, Haijun Jiao, Stephan Bachmann, Michelangelo
Scalone, Kathrin Junge, and matthias Beller
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Adv. Synth. Catal. 10.1002/adsc.201801511
Link to VoR: http://dx.doi.org/10.1002/adsc.201801511

10.1002/adsc.201801511
Advanced Synthesis & Catalysis
UPDATE
DOI: 10.1002/adsc.201((will be filled in by the editorial staff))
Enantioselective Hydrogenation of Ketones using Different
Metal Complexes with a Chiral PNP Pincer Ligand
Marcel Garbe,a Zhihong Wei,a Bianca Tannert,a Anke Spannenberg,a Haijun Jiao,a
Stephan Bachmann,b Michelangelo Scalone,b Kathrin Junge,a and Matthias Bellera*
a
Marcel Garbe, Zhihong Wei, Bianca Tannert, Dr. Kathrin Junge, Dr. Anke Spannenberg, Dr. Haijun Jiao, Prof. Dr.
Matthias Beller*, Leibniz-Institut für Katalyse e. V. an der Universität Rostock, Albert-Einstein Straße 29a, Rostock,
18059, Germany; Matthias.Beller@catalysis.de
Dr. Stephan Bachmann, Dr. Michelangelo Scalone, F. Hoffmann-La Roche Ltd, Process Chemistry and Catalysis,
b
Grenzacherstrasse 124, CH-4070 Basel, Switzerland
Received: ((will be filled in by the editorial staff))
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/adsc.201######.((Please
delete if not appropriate))
Prize in 2002.[5] To date, one of the most prominent
Abstract. The synthesis of different metal pincer
complexes coordinating to the chiral PNP ligand bis(2- and powerful catalyst systems for asymmetric
((2R,5R)-2,5-dimethyl-phospholanoethyl))amine
is
reductions is the BINAP-derived ruthenium complex
synthesized by Noyori and co-workers.[6] In part,
other precious metals such as Rh[7], Pd[8], Ir[9] and
Os[10] also led to standard protocols for asymmetric
reduction of ketones.
Since the turn of the millennium an increasing
interest aroused in homogenous catalysis using non-
noble metal complexes based on Cu[11], Co[12], Ni[13]
and Fe[14] combined with different mono-, di-, tri-, or
tetradentate ligands. Apart from being less toxic,
more abundant, inexpensive and environmental-
friendly, these metals might offer different selectivity
control and bio-inspired reactivity. Nevertheless, it
should be recognized that for most asymmetric
catalysis, the price of the chiral ligand determines the
overall catalyst costs.
A thorough study on this topic was reported by the
group of Morris, who used predominantly chiral
PNNP ligands for the (transfer) hydrogenation of
ketones.[15] In 2014, they published the first chiral
iron PNP complex for asymmetric hydrogenation
(Figure 1).[16] Complex 1 showed a high activity for
aromatic as well as for aliphatic prochiral ketones but
only moderate enantioselectivities up to 46% ee were
achieved for the latter ones. The catalyst needs to be
activated by four equivalents of LiAlH4 and base
during the reaction. With the improved Fe pincer
complex 2 the pre-activation step could be avoided
and aromatic prochiral ketones are reduced with ee
values up to 96% ee.[17]
Further pincer ligand backbones were synthesized
for iron and manganese complexes by the groups of
Morris (3), Mezzetti (4) and Kirchner (5).[18] Whereas
3 and 4 were only studied for the asymmetric
reduction of aromatic ketones, 5 was also able to
reduce aliphatic substrates with enantioselectivities
up to 74% ee. In addition, during the preparation of
this manuscript Morris and co-workers described
described in detail. The characterized complexes with Mn,
Fe, Re and Ru as metal centers showed good activities
regarding the reduction of several prochiral ketones.
Comparing these catalysts, the non-noble metal complexes
produced best selectivities not only for aromatic substrates,
but also for different kinds of aliphatic ones leading to
enantioselectivities up to 99% ee. Theoretical
investigations elucidated the mechanism and rationalized
the selectivity.
Keywords: Asymmetric Catalysis; Chiral Pincer;
Asymmetric Hydrogenation; Prochiral Ketones; Chiral
Alcohols
Introduction
Developing chiral organometallic complexes
continues to be an important and challenging field in
asymmetric catalysis.[1] In this area, enantioselective
hydrogenations and transfer hydrogenations are
commonly applied for the synthesis of enantiopure
alcohols from prochiral ketones leading to building
blocks for pharmaceuticals, agrochemicals, as well as
flavours and fragrances.[2]
The development of catalysts for asymmetric
hydrogenations originated in 1968 when Knowles
and Horner published their first systems
independently from each other.[3] These catalysts
were based on rhodium as a metal center coordinated
to chiral phosphine ligands. In the following decades
several chiral mono- and bidentate phosphine ligands
in the presence of noble metals were established and
notable advancements in asymmetric hydrogenations
as well as transfer hydrogenations have been
achieved by academia and industry.[4] Based on the
seminal work of Noyori and the original contributions
of Knowles, this area was honoured with the Nobel
1
This article is protected by copyright. All rights reserved.

10.1002/adsc.201801511
Advanced Synthesis & Catalysis
asymmetric transfer hydrogenation of ketones with
well defined PNN Mn(I) complexes in which the
ligands were previously reported and derived from (S,
S)-DPEN.[19] Only one of the complexes showed
moderate activities with enantioselectivities up to
53% ee for aromatic substrates while the mechanism
of transfer hydrogenation was described in more
detail.
the isomers 9a and 9b, whereas 9a could be isolated
by crystallisation. The structure of 9a was confirmed
by X-ray crystallography (see SI).
Figure 2. Chiral metal pincer complexes with bis(2-
((2R,5R)-2,5-dimethyl-phospholanoethyl))amine as ligand.
For the synthesis of the iron complex 12, FeBr2 was
reacted with the corresponding ligand. Next, the
resulting paramagnetic complex 10 was converted by
insertion of CO to 11, a suitable pre-catalyst for
catalysis.
Figure 1. Selected chiral Fe and Mn pincer complexes
used for the asymmetric reduction of prochiral ketones.
Scheme 1. Synthesis of chiral iron pincer complexes 11
and 12.
Recently, our group reported the synthesis of the
chiral
bis(2-((2R,5R)-2,5-dimethyl-phospholano-
ethyl))amine ligand[20,21] and the corresponding
manganese based complex 7 for the hydrogenation of
ketones.[21] While only low selectivities were
observed for the reduction of acetophenone and its
derivatives some aliphatic ketones especially
cyclohexyl methyl ketone gave enantiomeric excesses
up to 84% ee.
To avoid additional activation steps by base the
corresponding borohydride complex 12 was prepared.
NMR characterization shows two isomers where the
hydride is orientated cis or trans to the corresponding
NH proton.
After synthesizing these complexes, they were
tested regarding their activity and selectivity in two
Based on this work, herein we compare in detail
the behaviour of different metal pincer complexes
benchmark
hydrogenations
of
aromatic
(acetophenone) and aliphatic (cyclohexyl methyl
ketone) ketones. In preliminary experiments the
previously optimized conditions of the manganese
pincer catalyst system 7 were determined: 1 mol%
cat., 5 mol% base, 30 °C, 30 bar, 4 h, 1,4-dioxane for
acetophenone and 1 mol% cat., 5 mol% base, 40 °C,
30 bar, 4 h, t-amyl alcohol for cyclohexyl methyl
ketone.
coordinating
to
bis(2-((2R,5R)-2,5-dimethyl-
phospholano-ethyl))amine. Interestingly, non-noble
Mn and Fe complexes are superior compared to
related Ru and Re catalysts.
Initially, we synthesized the chiral rhenium
complex 8 as well as the corresponding ruthenium
(9a and 9b) and iron (12) derivatives, which are all
reported here for the first time. Analoguouse to the
preparation of the manganese complex 7,
[ReBr(CO)5] was reacted with bis(2-((2R,5R)-2,5-
dimethylphospholano-ethyl))amine in toluene at
100°C for twenty hours. After crystallization, 8 was
obtained in 69% yield as a grey powder. Furthermore,
refluxing carbonylchlorohydridotris(triphenylphos-
phine)ruthenium(II) with the chiral pincer ligand in
toluene for three hours gave a mixture of two
different Ru-species which were detected by
31P NMR. We assumed that these species belong to
As shown in Table 1, entry 1, the manganese
complex 7 led to full conversion for both substrates
and a high enantioselectivity regarding cyclohexyl
methyl ketone with 83% ee.[21] All new catalysts
produced complete conversion of acetophenone, too;
although for the rhenium-derived complex 8 a higher
temperature was necessary (Table 1, entry 3). As
expected, applying the Ru complex 9a or a mixture of
9a and 9b with base, the same activity was observed
indicating that the same active species is formed (SI).
Unfortunately, the low enantioselectivity using
2
This article is protected by copyright. All rights reserved.

10.1002/adsc.201801511
Advanced Synthesis & Catalysis
acetophenone could not be improved in the presence
prochiral ketones. Therefore the reaction parameters
for the asymmetric reduction of aromatic (30 bar,
30 °C, 3 h, EtOH) and aliphatic ketones (30 bar,
30 °C, 6 h - see SI) were optimized and compared to
the results of Mn pincer complex 7.
of complexes 8, 9a/b, 11 or 12.
Table 1. Comparison of the different catalysts for the
reduction of acetophenone and cyclohexyl methyl ketone.
Entr
y
cat.
conv.[a] ee in % conv.[b] ee in %
1
>99
0
18 (S)[21]
-
>99
0
83 (R)[21]
-
7
8
2
3[c]
>99
>99
>99
n.d.
>99
6 (S)
rac.
>99
41
60 (R)
62 (R)
62 (R)
47 (R)[e]
47 (R)[e]
8
4
11
12
9a
9a/b
5[d]
6
rac.
n.d.
6 (S)
>99
>99
>99
7
Reaction conditions: [a] 1 mmol acetophenone, 1 mol%
cat., 5 mol% KOtBu, 30 °C, 30 bar, 4 h, 1,4-dioxane
(2 mL). Conversion was determined by GC by using
hexadecane as internal standard. [b] 1 mmol cyclohexyl
methyl ketone, 1 mol% cat., 5 mol% KOtBu, 40 °C, 30 bar,
4 h, t-amyl alcohol (2 mL). Conversion was determined by
GC by using hexadecane as internal standard. [c] 100 °C,
1,4-dioxane (2 mL) for acetophenone and t-amyl alcohol
(2 mL) for cyclohexyl methyl ketone. [d] No base is
needed. [e] In EtOH a higher selectivity was achieved by
showing the same activity.
Scheme 2. Conversion vs. time diagram for the
comparison of 7, 9a/b and 12. Reaction conditions:
1 mmol substrate, 1 mol% cat., 5 mol% KOtBu (only for 7
and 9a/b), 40 °C, 30 bar, t-amyl alcohol (2 mL).
Conversion was determined by GC by using hexadecane as
internal standard.
As shown in Table 2, the decrease of the ring size
in the cyclic part of the ketone (Table 2, entries 1-3)
does not have a significant influence on the activity
of complex 12, but the enantioselectivity dropped
down using cyclopropyl methyl ketone (13a) instead
of cyclopentyl- or cyclohexyl methyl ketone (13b and
13c). Furthermore, the steric effect of the aliphatic
chain in cyclohexyl alkyl ketone was elucidated. In
case of cyclohexyl isopropyl or trifluoromethyl
ketone (13d-e), a decrease of the activity as well as
selectivity for both complexes was detected.
In general a better catalytic performance was
obtained for the aliphatic substrate cyclohexyl methyl
ketone. Nevertheless, to obtain full conversion
increased temperature is required using 8. In this case
only a moderate enantiomeric excess of 60% ee was
observed. Nearly the same selectivity was achieved
using the two different iron species 11 and 12
(Table 1, entries 4 and 5), but complex 12 showed a
higher activity regarding the reduction of cyclohexyl
methyl ketone. Surprisingly, a low enantiomeric
excess (47% ee) was observed for the ruthenium
complexes 9a/b (Table 1, entries 6 and 7).
Remarkably, this is one of the few cases where noble
metal complexes are less selective compared to their
base metal congeners.[22]
Next, the activity of the different complexes was
investigated in a conversion time diagram for the
reduction of cyclohexyl methyl ketone. In general,
the Re complex 8 was less selective vide supra; hence
only the Mn, Fe and Ru complexes 7, 9a/b, and 12
were studied in more detail (Scheme 2). Under these
conditions, the noble-metal complexes 9a/b are most
reactive giving complete conversion already after half
an hour. With complexes 7 and 12 four to five hours
were needed, while in case of 12 a slow activation
period was observed. For all complexes the
enantioselectivity was unaffected over the time.
4-Acetyltetrahydropyran (13f) showed only
a
moderate value of 46% ee for the iron complex (12),
while the enantioselectivity reached 79% ee for
complex 7. In contrast to this 1-cyclohexenyl methyl
ketone (13g) is reduced with nearly the same
enantiomeric excess for both complexes (51% ee and
52% ee, respectively). Thereby, complex 7 is less
selective regarding the C=C double bonds, producing
a
significant
amount
25%).
of
side
Whereas
product
using
(cyclohexylethanol:
cyclohexenone (13h) only small amounts of side
product could be detected and good to moderate
enantioselectivity of 62% ee (complex 7) and 46% ee
(complex 12) were determined. By using different
analogues of cyclohexenone (13i-13j), the
enantiomeric excess could not be increased. Notably,
the bulky substrate adamantyl methyl ketone (13k)
was hydrogenated with excellent enantioselectivity of
>99% ee with both complexes whereas the iron
complex 12 shows still a higher activity.
Due to its good activity and enantioselectivity the
iron pincer complex 12 was investigated for the
hydrogenation of different aromatic and aliphatic
3
This article is protected by copyright. All rights reserved.

10.1002/adsc.201801511
Advanced Synthesis & Catalysis
Table 2. Hydrogenation of cyclic aliphatic ketones using
complexes 7 and 12.[a]
– 14b) was realized with low enantioselectivities of
25% ee and 32% ee. A variation of the bulkiness of
the substituents on the keto group does not cause an
improvement of the enantioselectivity (see 14a-c).
The best result was obtained for the reduction of
3-methyl-2-butanone with complex 7 giving an
enantiomeric excess of 68% ee by using 7 as catalyst
(14a).
Catalyst 7[a]
Catalyst 12[b]
prochiral
ketone
conv.
[%]
ee
[%]
conv.
[%]
Entry
1
ee [%]
96[21]
70
>99
31
13a
13b
2
61[21] 74(R)
>99
60(R)
>99
3
>99[21]
83(R)
39
64(R)
32
[83%][c]
13c
13d
13e
13f
78[d]
5
54
[51%][c]
Figure 3. Reduction of non-cyclic aliphatic ketones by 7
(upper results) and 12 (lower results). General conditions
for 7: 1 mmol substrate, 2 mol% cat., 5 mol% KOtBu,
30 bar H2, 5 h 80 °C, t-amyl alcohol. General conditions
for 12: 1 mmol substrate, 1 mol% cat, 30 bar H2, 6 h, 30 °C,
EtOH (2 mL). Conversion was determined by GC using
hexadecane as internal standard. [a] 60 °C. [b] 5 mol%
NaOtBu, 50 °C, EtOH – 13 % of side product: 1-phenyl-3-
butanol. [c] Isolated yield is given.
6
23
57
>99
>99
16
96[21]
7
79
46
[74%][c]
>99
(25%)
97
(3%)
8[e]
9[e]
10
51
52
13g
13h
13i
As mentioned above, the asymmetric reduction of
acetophenone and different substituted aromatic
derivatives gave only low enantioselectivities (Figure
4 - 15a-f). The highest enantiomeric excess was
observed for the reduction of 1-phenyl-1-butanol
(19% ee) using complex 7. In most cases ketones
were reduced to racemic mixtures in the presence of
12. Only, for cyclohexyl phenyl ketone a chiral
induction of 26% ee was detected. Surprisingly, with
benzalacetophenone an enantioselectivity of 67% ee
was achieved using the chiral Mn based catalyst 7.
Here, slightly elevated temperature (50 °C) was
required which decreased the chemoselectivity
forming 36% of 1,3-diphenylpropan-1-ol. When the
Fe pincer complex 12 is applied for
benzalacetophenone only 6% of the side product was
detected and the chiral alcohol was obtained in 56%
ee.
To our delight, for 1-benzosuberol, α-tetralone and
α-indanone good ee values between 74% ee and
84% ee were detected with the manganese pincer
complex 7. By using complex 12 slightly smaller
enantioselectivities of 64% ee (15j) and 63% ee (15k)
were achieved. Reducing 1-benzosuberol with 12 the
selectivity dropped down significantly to 45% ee. We
also tested some heterocyclic aromatic ketones
(15l-15n) which were reduced by both catalytic
systems with high activities. Again the selectivity of
the manganese pincer complex 7 with ee values up to
94[21]
(6%)
98
(8%)
62
46
>99[21]
64[21]
28
51
99
55
96
11
12
61(S)
52(S)
[83%][c,f]
13j
>99
(R)
>99
(R)
65
13k
[a] General conditions for 7: 1 mmol substrate, 1 mol%
cat., 5 mol% KOtBu, 30 bar H2, 4 h 40 °C, tert-amyl
alcohol (2 mL). [b] General conditions for 12: 1 mmol
substrate, 1 mol% cat, 30 bar H2, 6 h, 30 °C, EtOH (2 mL).
Conversion was determined by GC using hexadecane as
internal standard. [c] isolated yield. [d] 60 °C – no conv. at
30 °C. [e] Conversion to the fully hydrogenated product is
given in brackets. [f] ~5% of impurities.
Furthermore, a number of non-cyclic aliphatic
ketones were applied for the asymmetric reduction
with chiral Mn and Fe pincer complexes 7 and 12.
The reduction of 3,3-dimethyl-butan-2-one (Figure 3
4
This article is protected by copyright. All rights reserved.

10.1002/adsc.201801511
Advanced Synthesis & Catalysis
69% ee for 2-furyl methyl ketone (15m) was slightly
better than for the iron catalyst system.
Scheme 3. Proposed mechanism for ketone hydrogenation
(E for chiral phospholane ligand in R,R configuration; X =
CO for M = Mn and Re; X = H for M = Fe and Ru).
Now, the corresponding Fe, Ru and Re chiral
pincer complexes were included in our computation
at the B3PW91 level of theory and compared with the
experimental results (Figure 5). Therefore, the
activated amido (1M) and amine (2M) complexes
were used as catalysts as well as cyclopentyl methyl
ketone, cyclohexyl methyl ketone and α-tetralone as
substrates. The computed results are listed in Table 3
(for more details see SI).
Figure 4. Asymmetric reduction of aromatic prochiral
ketones by 7 (upper results) and 12 (lower results). General
conditions for 7: 1mmol substrate, 1 mol% cat., 5 mol%
KOtBu, 30 bar H2, 4 h 30 °C, 1,4-dioxane (2 mL). General
conditions for 12: 1 mmol substrate, 1 mol% cat, 30 bar H2,
3 h, 30 °C, EtOH (2 mL). Conversion was determined by
GC using hexadecane as internal standard. [a] 2 mol% cat.,
3 h, 50 °C, EtOH (2 mL). [b] 6 h. [c] 2 mol% cat., 50 °C, 1
h, iPrOH (2 mL), 36% 1,3-diphenylpropan-1-ol (10% ee).
[d] 6% 1,3-diphenylpropan-1-ol. [e] Isolated yield is given.
Figure 5. Active catalysts 1M and 2M used for the
calculated hydrogenation reactions.
In case of 2Mn as catalyst a concerted but
asynchronous transition state was calculated for all
three substrates (cyclohexyl methyl ketone,
cyclopentyl methyl ketone and α-tetralone), where
the transfer of Mn-H and N-H to the ketone group
takes place concertedly.[21] Such
a
one-step
In our previous work,[21] an outer-sphere
mechanism (Scheme 3) was proposed for the
enantioselective hydrogenation of the Mn pincer
complexes 7 based on B3PW91 density functional
theory computations. Here, the asymmetric origin
comes from the approach of the prochiral ketone
group perpendicular to the N-H and M-H groups in
the cis position with either the re or si enantioface.
mechanism is also found for the reactions using 1Fe
and 1Ru as catalysts. However, for 1Re a two-step
mechanism is found, where the first step is the Re-H
transfer to the prochiral carbon center, followed by an
intermediate, and the second step is the N-H transfer
for the oxygen atom of the carbonyl group. Because
the transition state for Re-H transfer is higher in
energy than that of the N-H transfer, it is the rate-
determining step. This finding is in conclusion with
other two-step hydrogenation reactions.[23]
5
This article is protected by copyright. All rights reserved.

10.1002/adsc.201801511
Advanced Synthesis & Catalysis
Table 3. Comparison of the determined (er/Expt) and
computed (er/DFT) enantiomeric ratio on the basis of the
difference (bold and underlined value in kcal/mol; T = 313
K) of free energy (ΔG‡/TS, kcal/mol) between R and S
transition state configurations
1.1 kcal/mol for cyclopentyl and cyclohexyl methyl
ketone, however, a much larger deviation is found for
α-tetralone (3.1 kcal/mol).
In addition to the agreement in the enantio-
selectivity and enantiomeric ratios between theory
and experiment; we also found general agreement in
the stability between theory and experiment. Thus,
the computed free energy barriers of all
hydrogenation reactions are higher than the barriers
of the hydrogenation elimination of the catalyst [2M
= 1M + H2], indicating that high H2 pressure is
needed to maintain the stability of the catalysts. This
finding is supported by the high H2 pressure in the
experiment (30 bar). In case that the barrier of the
hydrogenation is lower than that of H2 elimination, no
high H2 pressure should be needed.
Mn (7)[21]
ΔG‡/TS
er/DFT 97/3 (R/S; –2.2)
31.3/33.5 (R/S)
31.4/33.5 (R/S) 33.5/35.6 (S/R)
97/3 (R; –2.1)
91/9 (R, –1.4)
Fe (12)
97/3 (S, –2.1)
er/Expt
87/13 (R, –1.1)
90/10 (S; –1.4)
ΔG‡/TS
er/DFT
er/Expt
25.5/27.5 (R/S)
97/3 (R; –2.0)
80/20 (R; –0.9)
24.3/26.1 (R/S) 26.6/27.9 (S/R)
95/5 (R; –1.8) 90/10 (S, –1.3)
According to the computed barriers, the catalytic
activity should follow the decrease order of Ru > Fe
> Re > Mn. Experimentally, Ru-based catalysts are
most active; however, Mn-based catalysts are more
active than the Fe-based catalysts (Scheme 2).
81/19 (R; –0.9) 80/20 (S; –0.9)
Ru (9)
ΔG‡/TS
er/DFT
er/Expt
21.0/22.7 (R/S)
95/5 (R; –1.7)
72/28 (R; –0.6)
20.5/22.3 (R/S) 22.4/23.8 (S/R)
Since reactions are usually carried out in solution
and the large substitutes might have strong van der
Waals interactions, one will always question the
accuracy and reliability of only gas phase computed
results despite of excellent agreement between theory
and experiment in stability and reactivity as well as
selectivity. Therefore, we computed the Fe-based
reaction (1Fe + H2 = 2Fe) considering van der Waals
dispersion correction and solvation effects (1,4-
dioxane) (B3PW91-SCRF-D3). It is found that
including dispersion and solvation correction the
barrier (15.05 vs. 17.18 kcal/mol) is lowered and
makes the reaction more exergonic (-6.69
vs. -2.92 kcal/mol). The rather large reaction free
energy does not support the expected equilibrium.
These effects have been observed computationally on
other pincer systems.[25]
95/5 (R; –1.8)
91/9 (S; –1.4)
73/27 (R; –0.6) 78/22 (S; –0.8)
Re (8)
ΔG‡/TS
er/DFT
er/Expt
29.3/31.1 (R/S)
95/5 (R; –1.8)
80/20 (R, –0.9)
29.5/31.5 (R/S) 29.9/33.6 (S/R)
96/4 (R; –2.0)
99/1 (S; –3.7)
80/20 (R, –0.9) 71/29 (S, –0.6)
Firstly, we compared the stability and reactivity
with respect to the exchange reaction between 1M
and hydrogen to form 2M (1M + H2 = 2M). It is
found that this exchange reaction is slightly exergonic
by 0.82, 2.92, 2.83 and 2.66 kcal/mol, respectively,
for 1Mn, 1Fe, 1Ru and 1Re. The free energy barrier
is 19.46, 17.18, 18.48 and 21.74 kcal/mol for 1Mn,
1Fe, 1Ru and 1Re, respectively. The free energy
barrier for the back reaction is 20.28, 20.10, 21.31
and 24.41 kcal/mol for 1Mn, 1Fe and 1Ru, and 1Re.
Such energetic properties are similar with those of the
non-chiral counterparts.[23]
A comparison of the enantiomeric ratios and the
energy differences of the R and S transition states for
the hydrogenation of cyclopentyl methyl ketone,
cyclohexyl methyl ketone and α-tetralone shows a
good agreement of the computed enantioselectivity
for all three substrates and all four catalysts with the
experiment (Table 3). Thus, the R isomer is more
selective for cyclopentyl and cyclohexyl methyl
ketone, while the S isomer is more selective for α-
tetralone.
Although the computed enantiomeric ratios are
larger than the experimentally determined data at the
first glance, the energetic difference to differentiate
the enantiomeric ratios does not vary too much. For
the Mn-based system, for example, the computed
energy difference is 2.1-2.2 kcal/mol; while the
estimated energy difference from the experimentally
determined enantiomeric ratios is 1.1-1.4 kcal/mol,
and the largest deviation is 1.1 kcal/mol. The same
results are found for the Fe- and Ru-based systems.
For the Re-based system, the deviation is 0.9 and
In addition, we computed the hydrogenation and
the enantiomeric ratio of α-tetralone as an example
for the consideration of the van der Waals dispersion
correction and solvation effect. To our surprise, the
computed barrier is rather low (9.72 and
7.37 kcal/mol for the R and S isomer, respectively);
much lower than the gas phase values (27.9 and
26.6 kcal/mol) as well as the barrier of the back
reaction (2Fe = 1Fe + H2) in both gas phase and in
solution (20.10 and 21.74 kcal/mol, respectively).
These results do not agree with the determined
reaction parameters. Here, the high H2 pressure
(30 bar) implies
a
higher barrier for the
hydrogenation than for the H2 elimination from the
catalyst and a stabilizing effect on the catalyst. In
case the barrier of the hydrogenation is lower than
that of H2 elimination, no selectivity should be
observed. Despite the fact of the lower hydrogenation
barriers, the selectivity is not altered by solvation and
dispersion correction. On the basis of these results,
one should take care about the choice of the method;
and benchmark calculations are needed to scale the
computed results.[21-23]
6
This article is protected by copyright. All rights reserved.

10.1002/adsc.201801511
Advanced Synthesis & Catalysis
Conclusions
[12] a) H. Pellissier, H. Clavier, Chem. Rev. 2014, 114(5),
2775-2823; b) D. Zhang, E.-Z. Thus, Z.-W. Lin, Z.-B.
Wei, Y.-Y. Li, J.-X. Gao, Asian J. Org. Chem. 2016,
5, 1323-1326.
[13] a) Y. Hamada, Y. Koseki, T. Fujii, T. Maeda, T.
Hibino, K. Makino, Chem. Commun. 2008, 6206-
6208.; b) R. Noyori, M. Tokunaga, M. Kitamura, Bull.
Chem. Soc. Jpn. 1995, 68, 36-55.
In conclusion, we have synthesized different chiral
metal pincer complexes using bis(2-((2R,5R)-2,5-
dimethylphos-pholanoethyl))amine as ligand as well
as Ru, Re, Mn and Fe as metal centers. These
complexes showed high activities as well as
selectivities especially regarding aliphatic ketones.
Among all the studied complexes, surprisingly the
manganese pincer complex performed best.
[14] a) M. Darwish, M. Wills, Catal. Sci. Technol., 2012,
2, 243-255; b) K. Gopalaiah, Chem. Rev. 2013, 113
(5), 3248-3296.
[15] C. Sui-Seng, F. Freutel, A. J. Lough, R. H. Morris,
Angew. Chem. Int. Ed. 2008, 47, 940-943; Angew.
Chem. 2008, 120, 954-957; b) N. Meyer, A. J. Lough,
R. H. Morris, Chem. - Eur. J. 2009, 15, 5605-5610;
c) A. Mikhailine, A. J. Lough, R. H. Morris, J. Am.
Chem. Soc. 2009, 131, 1394-1395. d) R. H. Morris,
Chem. Soc. Rev. 2009, 38, 2282-2291.
[16] a) P. O. Lagaditis, P. E. Sues, J. F. Sonnenberg, K. Y.
Wan, A. J. Lough, R. H. Morris, J. Am. Chem. Soc.
2014, 136, 1367-1380; b) J. F. Sonnenberg, A. J.
Lough, R. H. Morris, Organometallics 2014, 33,
6452-6465; c) J. F. Sonnenberg, K. Y. Wan, P. E.
Sues, R. H. Morris, ACS Catal. 2016, 316-326.
[17] S. A. M. Smith, P. O. Lagaditis, A. Lupke, A. J.
Lough, R. H. Morris, Chem. - Eur. J. 2017, 23, 7212-
7216.
[18] a) A. Zirakzadeh, K. Kirchner, A. Roller, B. Stöger,
M. Widhalm, R. H. Morris, Organometallics 2016,
35, 3781-3787; b) R. Huber, A. Passera, A. Mezzetti,
Organometallics 2018, 37, 396-405; c) A. Zirakzadeh,
S. R. M. M. de Aguiar, B. Stoger, M. Widhalm, K.
Kirchner, ChemCatChem 2017, 9, 1744-1748; d) M.
B. Widegren, G. J. Harkness, A. M. Z. Slawin, D. B.
Cordes, M. L. Clarke, Angew. Chem. Int. Ed. 2017,
56, 5825-5828; Angew. Chem. 2017, 129, 5919-5922.
[19] K. Z. Demmans, M. E. Olson, R. H. Morris,
Organometallics 2018, 37, 4608-4618.
[20] a) M. J. Burk, J. E. Feaster, R. L. Harlow,
Tetrahedron: Asymmetry 1991, 2, 569–592; b) M. J.
Burk (Du Pont), US19910691954, 1993; c) K. Abdur-
Rashid, US20040985058, 2005.
[21] M. Garbe, K. Junge, S. Walker, Z. Wei, H. Jiao, A.
Spannenberg, S. Bachmann, M. Scalone, M. Beller,
Angew. Chem. Int. Ed. 2017, 56, 11237-11241;
Angew. Chem. 2017, 129, 11389-11393.
[22] S. Monfette, Z. R. Turner, S. P. Semproni, P. J.
Chirik, J. Am. Chem. Soc. 2012, 134, 4561–4564.
[23] a) Z. Wei, A. de Aguirre, K. Junge, M. Beller, H. Jiao,
Catal. Sci. Technol. 2018, 8, 3649-3665; b) Z. Wei, A.
de Aguirre, K. Junge, M. Beller, H. Jiao, Eur. J.
Inorg. Chem. 2018, 4643-4657.
Acknowledgements
This work was supported by F. Hoffmann-La Roche AG. We
thank Dr. C. Fischer, S. Buchholz, S. Schareina, K. Fiedler (all
LIKAT) for their excellent analytical support. We thank Prof. Dr.
A. Börner and Dr. J. Holz for a gift of (R,R)-2,5-dimethyl-1-
(trimethylsilyl)phospholane. Zhihong Wei is grateful for the
financial support of the Leibniz Foundation (Leibniz Competition,
SAW-2016-LIKAT-1).
[1] a) H.-U. Blaser, Rend. Fis. Acc. Lincei 2007, 18, 281-
302; b) B. M. Trost, Proc. Natl. Acad. Sci. U. S. A.
2004, 101, 5348-5355.
[2] a) H.-U. Blaser, F. Spindler, M. Studer, App. Cat., A
2001, 221, 119-143; b) H.-U. Blaser, Chem. Commun.
2003, 293-296.
[3] W. S. Knowles, M. J. Sabacky, Chem. Comm. 1968,
1445-1446.
[4] a) H.-U. Blaser, E. Schmidt, Asymmetric Catalysis on
industrial Scale: Challenges, Approaches and
Solutions, Wiley-VCH. Weinheim, 2004; b) A.
Börner, Phosphorus Ligands in Asymmetric
Catalysis: Synthesis and Applications, Wiley-VCH,
Weinheim, 2008; c) W. Tang, X. Zhang, Chem. Rev.
2003, 103, 3029-3069.
[5] a) R. Noyori, Angew. Chem. Int. Ed. 2002, 41, 2008-
2022; Angew. Chem. 2002, 114, 2108-2123; b) W. S.
Knowles, Angew. Chem. Int. Ed. 2002, 41, 1998;
Angew. Chem. 2002, 114, 2096-2107.
[6] a) R. Noyori, T. Ohkuma, M. Kitamura, J. Am. Chem.
Soc. 1987, 109, 5856-5858; b) T. Ohkuma, H. Ooka,
S. Hashiguchi, T. Ikariya, R. Noyori, J. Am. Chem.
Soc. 1995, 117, 2675-2676; c) T. Ohkuma, H. Ooka,
T. Ikariya, R. Noyori, J. Am. Chem. Soc. 1995, 117,
10417-10418; d) R. Malacea, R. Poli, E. Manoury,
Coord. Chem. Rev. 2010, 254, 729-752.
[7] P. A. Avans, J. Tsuji, Modern Rhodium-Catalyzed
Organic Reactions Wiley-VCH, Weinheim, 2005, pp
22-35.
[8] Q. A. Chen, Z. S. Ye, Y. Duan, Y. G. Zhou, Chem.
Soc. Rev. 2013, 42, 497-511.
[24] Z. Wei, H. Jiao, Adv. Inorg. Chem. 2019, 73, (Eds. R.
[9] R. Malacea, R. Poli, E. Manoury, Coord. Chem. Rev.
2010, 254, 729-752.
[10] W. Baratta, C. Barbato, S. Magnolia, K. Siega, P.
Rigo, Chem. - Eur. J. 2010, 16, 3201-3206.
[11] a) H. Shimizu, D. Igarashi, W. Kuriyama, Y. Yusa, N.
Sayo, T. Saito, Org. Lett. 2007, 9, 1655-1657; b) K.
Mashima, T. Ohshima, H. Shimizu, T. Nagano, N.
Sayo, T. Saito, Synlett 2009, 3143-3146; c) K. Junge,
B. Wendt, D. Addis, S. Zhou, S. Das, S. Fleischer, M.
Beller, Chem .- Eur. J. 2011, 17, 101-105.
van
Eldik;
R.
Puchta),
https://doi.org/10.1016/bs.adioch.2018.10.002.
[25] a) H. Jiao, K. Junge, E. A. Alberico, M. Beller, J.
Comput. Chem. 2016, 37, 168-176; b) E. Alberico, A.
J. J. Lennox, L. K. Vogt, H. Jiao, W. Baumann, H.-J.
Drexler, M. Nielsen, A. Spannenberg, M. P.
Checinski, H. Junge, M. Beller, J. Am. Chem. Soc.
2016, 138, 14890-14904.
7
This article is protected by copyright. All rights reserved.

10.1002/adsc.201801511
Advanced Synthesis & Catalysis
UPDATE
Enantioselective Hydrogenation of Ketones using
Different Metal Complexes with a Chiral PNP
Pincer Ligand
Adv. Synth. Catal. Year, Volume, Page – Page
Marcel Garbe, Zhihong Wei, Bianca Tannert, Anke
Spannenberg, Haijun Jiao, Stephan Bachmann,
Michelangelo Scalone, Kathrin Junge, and Matthias
Beller*
8
This article is protected by copyright. All rights reserved.