Manganese PNP-pincer catalyzed isomerization of allylic/homo-allylic alcohols to ketones-activity, selectivity, efficiency

Article abstract of DOI:10.1039/c9cy01502g

We report the first manganese catalyzed isomerization of allylic alcohols to produce the corresponding carbonyl compounds. The ligand plays a decisive role in the efficiency of this reaction. Very high conversions could be obtained using a solvent-free reaction system. A detailed DFT study reveals a self-dehydrogenation/hydrogenation reaction mechanism which was verified by the isolation of the α,β-unsaturated ketone as intermediate and a deuterium labeling experiment. It also provided a rationale for the observed selectivity and the higher efficiency of phenyl over isopropyl substitution.

Full text of DOI:10.1039/c9cy01502g

Catalysis
Science &
Technology
View Article Online
View Journal | View Issue
PAPER
Manganese PNP-pincer catalyzed isomerization of
allylic/homo-allylic alcohols to ketones – activity,
selectivity, efficiency†
Cite this: Catal. Sci. Technol., 2019,
9, 6327
Tian Xia, ‡ Brian Spiegelberg, ‡ Zhihong Wei, ‡ Haijun Jiao,
Sergey Tin,
*
Sandra Hinze
and Johannes G. de Vries
We report the first manganese catalyzed isomerization of allylic alcohols to produce the corresponding
carbonyl compounds. The ligand plays a decisive role in the efficiency of this reaction. Very high
conversions could be obtained using a solvent-free reaction system. A detailed DFT study reveals a self-
dehydrogenation/hydrogenation reaction mechanism which was verified by the isolation of the α,β-
unsaturated ketone as intermediate and a deuterium labeling experiment. It also provided a rationale for
the observed selectivity and the higher efficiency of phenyl over isopropyl substitution.
Received 27th July 2019,
Accepted 10th October 2019
DOI: 10.1039/c9cy01502g
rsc.li/catalysis
dehydrogenative alkylation reactions.11 Although excellent
results in various chemical processes have been obtained by
1. Introduction
Atom economy has been widely used in synthesis and
catalysis as an important metric, since Trost proposed this
concept in 1995.1 Alkene isomerization is one of the prime
examples of atom economy and green chemistry because all
atoms from the substrate are used to form the product
without generation of any waste.2 One of the most interesting
and widely investigated isomerization reactions is the
transformation of readily available allylic alcohols into
ketones and aldehydes which are the basic components of
many chemicals in both laboratory and industry.3 During the
last half century, numerous all-around and highly active
homogeneous allylic alcohol isomerization catalysts,
containing precious transition metals like Ir, Ru, Pd and Rh
have been widely investigated.3 However, noble metals are
expensive and there are severe limitations on their use in the
production of pharmaceuticals, in particular in the last step
of a synthetic sequence. Consequently, much attention has
been focused on their replacement with first-row transition
metals which are earth-abundant, environmentally friendly
and often less toxic. Recently, reports have appeared on the
sole use of base as catalyst for the isomerization of allylic
alcohols to the corresponding ketones.4
using
manganese
pincer
complexes
as
catalysts,
isomerization reactions of allylic alcohols to produce ketones
are still unknown so far. We have previously reported the
isomerization of allylic alcohols using iron and cobalt PNP
pincer complexes.12 Herein, we report the first manganese-
catalyzed isomerization of allylic/homo-allylic alcohols to
ketones.
First, we decided to screen
a number of different
manganese PNP pincer complexes. The first PNP pincer
ligated manganese complex was reported by Nocera and
Ozerov in 2009. This complex was prepared by a reaction of
the PNP ligand with Mn2IJCO)10 or MnIJCO)5Br to afford
(PNP)MnIJCO)3.13 This method has been utilized as a model
reaction in the syntheses of many other PNP manganese
complexes. Subsequently, Boncella and co-worker synthesized
and characterized
a range of PNP pincer manganese
Homogeneous manganese catalysts were found to be
effective catalysts for hydrogenation,5 dehydrogenation,6
transfer
hydrogenation,7
alkyne
semi-hydrogenation,8
CH-activation reactions,9 hydrosilylation reactions10 and
Leibniz-Institut für Katalyse e.V. an der Universität Rostock, Albert-Einstein-
Strasse 29a, 18059 Rostock, Germany. E-mail: johannes.devries@catalysis.de
† Electronic supplementary information (ESI) available. See DOI: 10.1039/
c9cy01502g
Fig. 1 PNP–Mn complexes A–F tested in the isomerization of allylic
‡ These authors contributed equally to this work.
alcohols.
This journal is © The Royal Society of Chemistry 2019
Catal. Sci. Technol., 2019, 9, 6327–6334 | 6327

View Article Online
Paper
Catalysis Science & Technology
complexes of which complex A (Fig. 1) is an example.14 In
order to investigate the role of substitution on the efficiency
of the isomerization we tested the complexes B, C and D
hitherto known for hydrogenation, N-alkylation and
dehydrogenative coupling.5a,11a,j Beyond the PNHP ligand
sphere we were interested in testing complexes E and F due
to their different structural motif and catalytic performance
we found that the best results were obtained using D as
catalyst, with quantitative conversion at 120 °C in only two
hours (entry 13). Relatively poor results were obtained with
the Milstein catalyst (E) (entries 14–16) and the Kempe
catalyst (F) (entries 17–19). Next, we optimized the reaction
conditions of the reaction catalyzed by complex D. Use of 1,4-
dioxane, which is a good solvent for ester hydrogenation
catalyzed by complex D, resulted in the highest amount
(98%) of ketone (Table 2, entry 2). This catalyst showed the
same selectivity in both toluene and cyclohexane (entries 1
and 3). Compared to other solvents, the reaction in heptane
resulted in poor selectivity under the same reaction
conditions (entry 4). Under neat conditions, full conversion
of the substrate was observed with 97% yield of the desired
product (entry 5). Decreasing catalyst and base loadings to 1
mol% resulted in quantitative conversion and yield of the
desired product at 120 °C after 1 hour in 1,4-dioxane (entry
6). Surprisingly, this catalytic reaction also showed good
conversion and yield even at room temperature after both 1
hour and 16 hours (entry 7). Furthermore, the conversion
and yield dropped when the catalyst and tBuOK loadings
were decreased to 0.1 mol% in 1,4-dioxane (entry 8). Catalyst
D (0.1 mol%) showed high activities in both cyclohexane and
toluene (entries 9–11). It is worth mentioning that the
isomerization reaction proceeds in neat oct-1-en-3-ol and
yields 94% of the desired 3-octanone (entry 12). The catalytic
activity of compound D was also tested under the same
reaction conditions at room temperature (entry 13). However,
this led to a dramatic decline of its activity (6% yield after 16
h). On the other hand, we were interested in finding the
maximum efficiency of catalyst D, for which reason we
in
dehydrogenative
coupling
and
hydrogenation
reactions.5b,15
2. Results and discussion
2.1 Screening and optimization of the isomerization
Our initial experiments were performed at 80 or 120 °C in
different solvents using oct-1-en-3-ol as substrate in the
presence of a catalytic amount of manganese compounds A–F
(Table 1, full screening see ESI†). Only 3% of ketone was
observed by using toluene as solvent in 20 hours at 80 °C
t
with compound A and BuOK (Table 1, entry 1). Then toluene
was replaced by THF and a higher conversion and yield
(17%) was obtained (entry 2). By using different solvents in
the presence of complex B which presumably is the active
catalyst that forms from A in the presence of base, we found
heptane was a somewhat better solvent than THF and
toluene (entries 3–5). Furthermore, higher conversions and
yields were observed by using 5 mol% of compound B in
both 1 and 9 hours at 120 °C (entries 6–7). Interestingly, 69%
of the desired product was obtained by using neat condition
after 9 hours (entry 8). Complex C was used as catalyst in
heptane or under neat condition at 120 °C, which resulted in
good conversions and yields (entries 9–12). To our delight,
Table 1 Catalyst screening in oct-1-en-3-ol isomerizationa
Entry
Cat. (%)
tBuOK (%)
Solvent
Time (h)
Conv.b [%]
Yield 2b [%]
Yield 3b [%]
1
2
3
4
A (1)
A (1)
B (1)
B (1)
B (1)
B (5)
B (5)
B (5)
C (5)
C (5)
C (5)
C (5)
D (5)
E (5)
E (5)
E (0.5)
F (5)
F (5)
F (0.5)
1
1
0
0
0
0
0
0
5
5
5
5
5
5
5
0.5
5
Toluene
THF
Toluene
THF
Heptane
Heptane
Heptane
—
Heptane
Heptane
—
20
20
16
16
16
1
9
9
2
16
2
3
17
8
0
24
30
90
70
45
3
17
8
0
24
30
88
69
44
88
51
90
93
0
0
0
0
0
0
2
1
1
5
6c
7c
8c
9c
10c
11c
12c
13c
14c
15c
16c
17c
18c
19c
98
52
92
10
1
2
—
16
2
Toluene
Cyclohexane
Heptane
100
27, 43
17, 22
1, 1
1, 7
3, 9
22, 27
7
2, 6
2, 6
2, 6
2, 6
2, 6
2, 6
24, 39
15, 18
1, 1
0, 3
2, 3
18, 21
3, 4
2, 4
0, 0
0, 4
1, 6
4, 6
d
—
Cyclohexane
Heptane
5
0.5
d
—
a
b
d
1.0 mmol of substrate. Determined by GC with dodecane as internal standard. c Reaction temperature is 120 °C. 10.0 mmol of substrate.
6328 | Catal. Sci. Technol., 2019, 9, 6327–6334
This journal is © The Royal Society of Chemistry 2019

View Article Online
Catalysis Science & Technology
Paper
Table 2 Solvent and catalyst loading screening in the isomerization of oct-1-en-3-ol to 3-octanonea
Entry
Cat. (%)
tBuOK (%)
Solvent
Time (h)
Conv.b [%]
Yield 2b [%]
Yield 3b [%]
1
2
3
4
5
6
5
5
5
5
5
1
1
0.1
0.1
0.1
0.1
0.1
0.1
0.01
5
5
5
5
5
1
1
0.1
0.1
0.1
0.1
0.1
0.1
0.01
Toluene
1,4-Dioxane
Cyclohexane
Heptane
No
1,4-Dioxane
1,4-Dioxane
1,4-Dioxane
Cyclohexane
Toluene
Toluene
No
2
2
2
2
100
100
100
100
100
100
88, 100
70
93
98
93
87
97
4
1
7
10
3
0
2
1
1, 16
1
1
1
1
1
16
16
100
76, 83
68
7c
8d
9d
10d
11e
12f
13c
14g
10, 15
2
2
1
0
0
0
0
97
100
85
94
6
95
99
85
94
6
3
No
No
3
a
b
c
d
1.0 mmol of substrate. Determined by GC with dodecane as internal standard. Room temperature. 10.0 mmol of substrate and 10 mL of
solvent. e 10.0 mmol of substrate and 10 mL of solvent, open system. f 10.0 mmol of substrate. g 100.0 mmol of substrate.
reduced the catalyst loading to 0.01 mol% while increasing at
the same time the amount of substrate (100 mmol) (entry
14). The latter resulted in a turnover number (TON) of 300
and was obtained in the absence of any solvent. Nevertheless,
the use of solvent is still preferred due to the slightly higher
turnover frequency (TOF) which can be achieved (1000 h−1,
entry 10) compared to the neat conditions (TOF: 940 h−1,
entry 12).
carbonyl compound in 95% yield (entry 8). Interestingly,
good to excellent yields of the corresponding carbonyl
derivatives were also obtained by using homo-allylic alcohols
(entries 9–11) which have a carbon number of exact one
between the alcohol and the double bond functionality. In
comparison, when the carbon number is equal to two (entry
12), no isomerization reaction took place, which nicely
confirms that the isomerisation does not take place via a
metal-catalysed double bond isomerization. In the case of the
homo-allylic alcohols we assume that the double bond is
shifted by base catalysis at the unsaturated ketone stage.
Aliphatic allylic alcohols could also be converted to the
ketones in good yield. 1-Cyclohexyl-but-2-en-1-ol (entry 13),
which was used as a mixture of cis- and trans-isomers, was
converted to 1-cyclohexylbutan-1-one in a high isolated yield.
Linear allylic alcohols were isomerized to the saturated
ketones with excellent yields using this catalytic system (entry
14 and 15). Notably, when we used 10 mmol (1.5 gram scale)
of oct-1-en-3-ol as substrate in 10 mL of toluene, the
purification procedure of 3-octanone required only filtering
off the catalyst and salts and removal of the solvent.
2.2 Substrate scope
With the optimized reaction conditions in hand, we
subsequently tested the scope and limitations of this catalytic
system which included various aromatic and aliphatic allylic
alcohols (Table 3). Excellent yields of the ketones were
obtained in the isomerization of aromatic allylic alcohols and
homo-allylic alcohols. Both electron-rich and electron-poor
substituents on the aromatic ring were well-tolerated (entries
1–11). Isomerization of 1-phenylprop-2-en-1-ol which was
selected as the parent substrate resulted in a 91% yield of the
corresponding ketone (entry 1). A methyl substituent on the
para-position of the aromatic ring had no influence on the
yield of the desired product (entry 2). A slight decrease of the
catalytic activity was observed when the methyl substituent
was replaced by a methoxy group in the para position of the
aromatic ring (entry 3). A bromo-susbtrituent was also well-
tolerated and isomerization of the para-bromo compound led
to formation of the ketone in 87% yield (entry 4). An excellent
yield (89%) was obtained when the reaction was carried out
on the naphthyl derivative (entry 5). Substrates with
heteroaromatic rings, e.g. furanyl- and thienyl-, showed high
isomerization activity in terms of the yield of the desired
ketones (entries 6 and 7). The substrate with an internal
double bond was also converted into the corresponding
2.3 Mechanism – theory and experiment
We also carried out a computational study to understand the
kinetic and thermodynamic parameters of the reaction. All
computational details are given in the ESI.†
Both previous experimental and computational studies
showed that the bromo–amine complex [MnIJBr)IJCO)2HN-
IJCH2CH2PR2)] (Fig. 1, A and D) is the pre-catalyst and needs
to be activated by base to form the amido catalyst
[MnIJCO)2NIJCH2CH2PR2)]. Isomerization from 1 to 2 via the
proposed outer-sphere mechanism in Scheme
1
was
computed. According to the proposed mechanism, the first
This journal is © The Royal Society of Chemistry 2019
Catal. Sci. Technol., 2019, 9, 6327–6334 | 6329

View Article Online
Paper
Catalysis Science & Technology
Table 3 Substrate scope of manganese-catalyzed allylic/homo-allylic alcohol isomerizationa
Entry
1
Allylic/homo-allylic alcohols
Ketones
Yieldb
91
2
3
90
74
4
5
87
89
6
7
8
91
96
95
9
93
93
73
10
11
12
13
0
93
14
15
94
91
a
1.0 mmol of substrate, 1 mol% of catalyst D, 1 mol% of tBuOK, 1 mol of toluene. b Isolated yield.
step is the dehydrogenation of allyl alcohol 1 to α,β-
unsaturated ketone 4 by using complex 1M-iPr/Ph.
In our previous study of the ketone hydrogenation by
using 2M,16 which is the reverse reaction of the
dehydrogenation reaction, we found a two-step process, in
which the first step is the hydride transfer from M–H to the
carbon center of ketone, followed by an ion-pair intermediate
and the second step is the proton transfer from N–H to the
oxygen center of ketone; and the hydride transfer is the rate-
determining step.
On the basis of these results, we also tried to locate a
revised two-step process for the dehydrogenation reaction.
6330 | Catal. Sci. Technol., 2019, 9, 6327–6334
This journal is © The Royal Society of Chemistry 2019

View Article Online
Catalysis Science & Technology
Paper
2a, which can tautomerize into 2. For the 1,2-addition, we
found a stepwise mechanism; i.e.; the first step is the hydride
transfer of Mn–H to the terminal C1 carbon via the transfer
state TS4–2IJ1,2-H−) resulting in the formation of the ion-pair
intermediate intermIJ1,2). The second step is the proton
transfer of N–H to the internal C2 carbon via the transition
state of TS4–2IJ1,2-H+). The free energy barrier of the first step
is 76.2 and 80.0 kJ mol−1 for 2M-iPr and 2M-Ph, respectively.
The formation of intermIJ1,2) is endergonic by 45.3 and 55.4
kJ mol−1 for 2M-iPr and 2M-Ph, respectively.
Scheme 1 Proposed mechanism (R = C5H11, E = PIJiPr)2 or PPh2).
The second step has free energy barrier of 52.1 and 58.0 kJ
mol−1 for 2M-iPr and 2M-Ph, respectively. Totally, the first
step is rate-determining; and this hydrogenation is exergonic
by 94.7 and 90.0 kJ mol−1 for 2M-iPr and 2M-Ph, respectively.
For the 1,4-addition, we also found a stepwise mechanism;
i.e.; the first step is the hydride transfer of Mn–H to the
terminal C1 carbon via the transfer state TS4–2IJ1,4-H−)
resulting in the formation of the ion-pair intermediate
intermIJ1,4). The second step is the proton transfer of N–H to
the terminal O4 via the transition state of TS4–2IJ1,4-H+). The
free energy barrier of the first step is 77.7 and 72.4 kJ mol−1
for 2M-iPr and 2M-Ph, respectively. The formation of
intermIJ1,4) is endergonic by 40.2 and 47.2 kJ mol−1 for 2M-
iPr and 2M-Ph, respectively. The second step has a free
energy barrier of 40.9 and 45.9 kJ mol−1 for 2M-iPr and 2M-
However, only one transition state corresponding to the
hydride transfer [TS1–4IJ3,4-H−)] was located for the
dehydrogenation of allyl alcohol 1; and neither the expected
transition state of the proton transfer nor the ion-pair
intermediate could be located. It shows that the
dehydrogenation of allyl alcohol
1 follows a concerted
mechanism. Actually, similar results have been found for the
reaction catalyzed by using the corresponding Fe–PNP12 as
well as Ru–PNP17 complexes. It is found that the hydride
transfer has a Gibbs free energy barrier of 128.6 and 117.6 kJ
mol−1 for 1M-iPr and 1M-Ph, respectively; and the reaction is
endergonic by 4.0 kJ mol−1 for 1M-iPr and somewhat
exergonic by 0.7 kJ mol−1 for 1M-Ph; indicating a quite
thermodynamically balanced process.
Ph, respectively. The tautomerization from 2a to
2 is
The next step is the CC bond hydrogenation of the α,β-
unsaturated ketone 4 by using complex 2M-iPr/Ph via either
1,2-addition directly to 2 or via 1,4-addition to form the enol
exergonic by 46.9 kJ mol−1. The first step is rate-determining.
Comparison in Scheme 2 shows that the 1,4-route is more
favored kinetically than the 1,2-route for 2M-Ph (80.0 vs. 72.4
Scheme 2 Gibbs free energy profile for the isomerization of 1 to 2 (R1 = C5H11, E = PIJiPr)2 or PIJPh)2).
This journal is © The Royal Society of Chemistry 2019
Catal. Sci. Technol., 2019, 9, 6327–6334 | 6331

View Article Online
Paper
Catalysis Science & Technology
kJ mol−1) by 7.6 kJ mol−1, whereas the 1,4- and 1,2-routes
have close barriers (77.7 and 76.2 kJ mol−1, respectively) for
2M-iPr.
Compared with the dehydrogenation of 1 to 4, the
hydrogenation of 4 to 2 is more favored kinetically by 52.4
(128.6 kJ mol−1 for 1M-iPr vs. 76.2 kJ mol−1 for 2M-iPr) and
45.2 kJ mol−1 (117.6 kJ mol−1 for 1M-Ph vs. 72.4 kJ mol−1 for
2M-Ph), and thermodynamically by 94.7 (−90.7 vs. 4.0 kJ
mol−1) and 90.0 kJ mol−1 (−90.7 vs. –0.7 kJ mol−1) for 1M-iPr
and 1M-Ph. Therefore, once 4 formed, it can be easily
converted to 2. The detailed course of the dehydrogenation
step of alcohols by a manganese PNP pincer complex was
previously described by Gauvin and coworkers.6b In
agreement with their findings, an immediate color change
(red to yellow) was noticed when an allylic alcohol was added.
This observation is associated to the formation of an alkoxide
species which, when heated further becomes the amino
hydride species 2M.
Fig. 2 Gibbs free energy profile of H2 elimination from 2M. (E = PIJiPr)2
or PIJPh)2).
whole potential energy surface. Since the barrier of H2
elimination is lower than that of 1 to 4 dehydrogenation, it is
necessary to run the reaction in a closed system in order to
maintain the stability of the catalyst 2M-iPr/Ph. This explains
clearly why the reaction needs to be carried out in a pressure
tube. Nevertheless, when the reaction on 1 was performed in
an open system, only a slightly lower yield (87%) of 2 was
obtained (Table 2, entry 11).
In addition, we computed the direct hydrogenation of 1 to
3 by using complexes 2M-iPr/Ph after the dehydrogenation of
one equivalent 1 into 4. For 1 hydrogenation to 3, this
reaction proceeds via
a
one-step mechanism mainly
corresponding to the hydride transfer and is exergonic by
101.2 and 96.5 kJ mol−1 for 2M-iPr and 2M-Ph, respectively.
The barrier is 133.0 and 133.4 kJ mol−1 for 2M-iPr and 2M-
Ph, respectively. This indicates that the isolated CC double
bond hydrogenation of 1 is less competitive kinetically than
the conjugated CC double bond hydrogenation of 4 by 56.8
(133.0 vs.76.2 kJ mol−1) for 2M-iPr and 61.0 kJ mol−1 (133.4
vs. 72.4 kJ mol−1) for 2M-Ph.
It is noted that the free energy barrier of 1M-Ph is lower
than 1M-iPr by 11.0 kJ mol−1 in the process from 1 to 4, and
by 10.0 kJ mol−1 in the 1,4-route from 4 to 2. This shows the
more favored substitution effect of phenyl over isopropyl.
Therefore, we employed the proposed activation strain model18
(ASM, Fig. 2) to investigate the substitution effect. According
to the ASM scheme, we dissected the electronic activation
energy of the transition states of rate-determining step
Furthermore, we computed the consecutive hydrogenation
of
2
to
3
by using complexes 2M-iPr/Ph after the
(TS1–4IJ3,4-H−), ΔE‡) into the geometrical strain energy (ΔEstrain
)
dehydrogenation of another one equivalent 1 into 4. The
hydrogenation of 2 to 3 by using complex 2M-iPr/Ph is found
to have a one-step process. The computed barrier is 141.8
and 122.1 kJ mol−1 for 2M-iPr and 2M-Ph, respectively; and
the reaction is exergonic by 10.4 and 5.8 kJ mol−1 for 2M-iPr
and 2M-Ph, respectively.
All these reveal that 2 is the principal and preferred
product. The hydrogenation of the isolated CC bond from
1 to 3 and the isolated CO bond from 2 to 3 by using
complex 2M-iPr/Ph are not competitive kinetically. This
agrees perfectly with the results that the ketone is the major
product.
and interaction energy (ΔEint), where the electronic activation
energy ΔE‡ is the electronic energy difference between the
optimized TS and the sum of substrate and catalyst in their
optimized structures; and the geometrical strain energy ΔEs‡train
is the electronic energy difference between the sum of the
structurally deformed substrate and catalyst individually taken
from the optimized TS and the sum of reactant and catalyst in
their optimized structures. Accordingly, the difference between
ΔE‡strain and ΔE‡ is the interaction energy between substrate
and catalyst in the TS. In addition, ΔE‡strain can be divided into
To investigate the stability of the catalysts (Fig. 2), we
computed the dehydrogenation or hydrogen elimination from
complex 2M-iPr/Ph to complex 1M-iPr/Ph (2M = 1M + H2),
which has a Gibbs free energy barrier of 89.5 and 86.4 kJ
mol−1 for 2M-iPr and 2M-Ph, respectively; is slightly exergonic
by 1.3 kJ mol−1 for 2M-iPr, while slightly endergonic by 3.3 kJ
mol−1 for 2M-Ph.
Comparisons show that the barrier of H2 elimination from
2M-iPr/Ph to complex 1M-iPr/Ph is higher than that of the
hydrogenation of
4
to
2
and lower than that of the
dehydrogenation of
dehydrogenation of 1 to 4 is the rate-determining step on the
1
to 4. This reveals that the
Fig. 3 ASM analysis of TS1–4IJ3,4-H−) of reaction [1M + 1 = 2M + 4].
6332 | Catal. Sci. Technol., 2019, 9, 6327–6334
This journal is © The Royal Society of Chemistry 2019

View Article Online
Catalysis Science & Technology
Paper
To supply further proof for the proposed mechanism we
have performed a deuterium labeling experiment. Applying
the same conditions as used for the substrate scope, 1-D-
labelled 1-phenylprop-2-en-1-ol (Fig. 4) was isomerized
efficiently to the corresponding ketone. From this reaction
the saturated ketone was recovered exclusively labeled in the
beta-position. This result is in perfect agreement with our
proposed dehydrogenation-1,4- reduction pathway.
Scheme 3 Slow isomerization of cyclohexenol allows isolation of the
intermediate 15-I. Conditions: 2.0 mmol, 1 mol% D, 1 mol% tBuOK, 2
mL toluene, 120 °C, 3 h.
3. Conclusions
the strain energy of substrate (ΔE‡strain/sub) and catalyst (ΔEs‡train/
cat) accordingly (ΔE‡strain = ΔEs‡train/sub + ΔEs‡train/cat). All these
data based on reactions [1M + 1 = 2M + 4] are shown in Fig. 3.
As shown in Fig. 3, for the hydrogenation of 1 to 4, the
energy barrier of TS1–4IJ3,4-H−)-Ph is lower in energy than that
of the TS1–4IJ3,4-H−)-iPr (ΔΔE‡ = ΔE‡(iPr) − ΔE‡(Ph)) by 13.8 kJ
mol−1. Further analyses into the difference of geometrical
strain energy (ΔΔE‡strain = ΔEs‡train(iPr) − ΔEs‡train(Ph)) and
interaction energy (ΔΔE‡int = ΔEi‡nt(iPr) − ΔEi‡nt(Ph)) reveal that
the ΔΔE‡ is dominated by ΔΔEs‡train. In addition, the ΔΔE‡strain/
In summary, we have shown for the first time that
manganese complexes are efficient catalysts for the
isomerization of allylic alcohols to the corresponding ketones
with catalyst amounts as low as 0.1 mol% in the presence of
base. Notably, the efficiency of the catalysts could be
increased by changing the substituent on the phosphorus
atom from isopropyl to phenyl. Good to excellent yields could
be achieved in both aromatic and aliphatic allylic or homo-
allylic alcohols.
A
two-step self-hydrogen borrowing
dehydrogenation–hydrogenation
mechanism involving
sub contributes much stronger than the ΔΔEs‡train/cat
.
isomerization is proposed and verified by DFT computations.
The calculated results in verifying the selectivity and
revealing the higher activity of the phenyl over isopropyl
substitution are in perfect agreement with experiments. The
high activity of phenyl over isopropyl substitution is due to
the weaker structural strain of the substrate interacting with
the phenyl substitution than isopropyl substitution, and this
is responsible for the lower barrier for phenyl substitution.
These results revealed that the isopropyl substituted Mn
PNP catalyst causes larger structural distortion for the
substrate in the transition state than phenyl substituted Mn
PNP catalyst, and explains why the phenyl substituted
catalysts (1M/2M-Ph) shows higher activity than the isopropyl
substituted catalysts (1M/2M-iPr).
These computational results highlight
a consecutive
dehydrogenation–hydrogenation sequence. According to this,
cyclohexenol (15) was chosen as a substrate in order to find
cyclohexenon (15-I) as an intermediate which requires
typically more time for alkene hydrogenation in comparison
to the before mentioned substrates.12 Hence, after 3 h at 120
°C (Scheme 3) a mixture of starting material (15), the
intermediate (15-I) and the desired product cylcohexanon (15-
P) was obtained. The overall conversion of cyclohexenol is
quite low (58%) whereby 18% were found to be the
intermediate 15-I. This discovery not only supports the
evidence of a dehydrogenation–hydrogenation mechanism
but also the fact that the dehydrogenation is the rate-
determining step.
Experimental section
Synthesis of catalyst D
A suspension of HCl·PNP(Ph) ligand (600 mg, 1.26 mmol) and
NEt3 (0.2 mL, 1.45 mmol) in THF (20 mL) was stirred for 15
minutes at room temperature. The liquid part of the resulting
suspension was added dropwise through cannula filtration to
a suspension of MnIJCO)5Br (275 mg, 1.0 mmol) in THF (40
mL) (be aware of CO evolution). After one hour, the mixture
was heated to 100 °C and stirred for 24 hours (a precipitate
was formed during the course of the reaction). After this time,
the mixture was allowed to cool to room temperature and the
suspension was filtered. The resulting solid was washed with
heptane (3 × 20 mL) and dried in vacuo to afford the title
compound (388 mg, 61%). Single crystals suitable for X-ray
diffraction measurement were obtained from slow diffusion
of heptane into the solution of the title compound in toluene.
General procedure for the isomerization of allylic alcohols
An oven dried 10 mL pressure tube with a stirring bar was
t
charged with complex D (0.01 mmol, 1 mol%), 0.01 mmol -
BuOK and 1 mL toluene sequentially. Then 1 mmol of
substrate was added immediately to the pressure tube. The
solution was stirred for 1 hour at 120 °C. GC yields were
determined with n-dodecane as internal standard. Isolated
yields were obtained by using silica gel chromatography
Fig. 4 Deuterium labeling experiment. Conditions: 1.0 mmol, 1 mol%
D, 1 mol% BuOK, 1 mL toluene, 120 °C, 1 h.
t
This journal is © The Royal Society of Chemistry 2019
Catal. Sci. Technol., 2019, 9, 6327–6334 | 6333

View Article Online
Paper
Catalysis Science & Technology
(cyclohexane: ethyl acetate = 50 : 1) after rotary evaporation. For
the neat reaction, base was added together with complex D.
F. Capet, J.-F. Paul, F. Dumeignil and R. M. Gauvin, ACS
Catal., 2017, 7, 2022; (c) M. Andérez-Fernández, L. K. Vogt,
S. Fischer, W. Zhou, H. J. Jiao, M. Garbe, S. Elangovan, K.
Junge, H. Junge, R. Ludwig and M. Beller, Angew. Chem., Int.
Ed., 2017, 56, 559.
Conflicts of interest
The authors declare that there is no conflict of interest.
7 (a) A. Bruneau-Voisine, D. Wang, V. Dorcet, T. Roisnel, C.
Darcel and J. B. Sortais, Org. Lett., 2017, 19, 3656; (b) M.
Perez, S. Elangovan, A. Spannenberg, K. Junge and M. Beller,
ChemSusChem, 2017, 10, 83; (c) A. Zirakzadeh, S. de Aguiar,
B. Stoger, M. Widhalm and K. Kirchner, ChemCatChem,
2017, 9, 1744; (d) K. Z. Demmans, M. E. Olson and R. H.
Morris, Organometallics, 2018, 37, 4608; (e) O. Martínez-
Ferraté, C. Werlé, G. Francìo and W. Leitner, ChemCatChem,
2018, 10, 4514; ( f ) D. Wang, A. Bruneau-Voisine and J. B.
Sortais, Catal. Commun., 2018, 105, 31.
Acknowledgements
Tian Xia is grateful for the financial support of Chinese
Scholarship Council (CSC). This work was also supported by
the state of Mecklenburg-Vorpommern and the Leibniz
Association (Leibniz Competition, SAW-2016-LIKAT-1523). We
are also thankful to Marcel Garbe (Leibniz-Institute for
Catalysis) for the provision of some chemicals.
8 (a) A. Brzozowska, L. M. Azofra, V. Zubar, I. Atodiresei, L.
Cavallo, M. Rueping and O. El-Sepelgy, ACS Catal., 2018, 8,
4103; (b) Y. P. Zhou, Z. B. Mo, M. P. Luecke and M. Driess,
Chem. – Eur. J., 2018, 24, 4780.
Notes and references
1 B. M. Trost, Angew. Chem., Int. Ed. Engl., 1995, 34, 259.
2 (a) M. Hassam, A. Taher, G. E. Arnott, I. R. Green and W. A.
van Otterlo, Chem. Rev., 2015, 115, 5462; (b) A. Vasseur, J.
Bruffaerts and I. Marek, Nat. Chem., 2016, 8, 209.
9 (a) W. Liu and L. Ackermann, ACS Catal., 2016, 6, 3743; (b)
Y. Hu, B. Zhou and C. Wang, Acc. Chem. Res., 2018, 51, 816.
10 X. Yang and C. Wang, Chem. – Asian J., 2018, 13, 2307.
11 (a) S. Elangovan, J. Neumann, J. B. Sortais, K. Junge, C.
Darcel and M. Beller, Nat. Commun., 2016, 7, 12641; (b) M.
Peña-López, P. Piehl, S. Elangovan, H. Neumann and M.
Beller, Angew. Chem., Int. Ed., 2016, 55, 14967; (c) M. K.
Barman, A. Jana and B. Maji, Adv. Synth. Catal., 2018, 360,
3233; (d) S. Chakraborty, P. Daw, Y. Ben David and D.
Milstein, ACS Catal., 2018, 8, 10300; (e) A. Jana, C. B. Reddy
and B. Maji, ACS Catal., 2018, 8, 9226; ( f ) Y. K. Jang, T.
Krückel, M. Rueping and O. El-Sepelgy, Org. Lett., 2018, 20,
7779; (g) V. G. Landge, A. Mondal, V. Kumar, A.
Nandakumar and E. Balaraman, Org. Biomol. Chem.,
2018, 16, 8175; (h) T. T. Liu, L. D. Wang, K. K. Wu and Z. K.
Yu, ACS Catal., 2018, 8, 7201; (i) G. Y. Zhang, T. Irrgang, T.
Dietel, F. Kallmeier and R. Kempe, Angew. Chem., Int. Ed.,
2018, 57, 9131; ( j) J. C. Borghs, Y. Lebedev, M. Rueping and
O. El-Sepelgy, Org. Lett., 2019, 21, 70.
12 (a) T. Xia, Z. Wei, B. Spiegelberg, H. Jiao, S. Hinze and J. G.
de Vries, Chem. – Eur. J., 2018, 24, 4043; (b) B. Spiegelberg,
A. Dell'Acqua, T. Xia, A. Spannenberg, S. Tin, S. Hinze and
J. G. de Vries, Chem. – Eur. J., 2019, 25, 7820.
13 A. T. Radosevich, J. G. Melnick, S. A. Stoian, D. Bacciu, C.-H.
Chen, B. M. Foxman, O. V. Ozerov and D. G. Nocera, Inorg.
Chem., 2009, 48, 9214.
14 A. M. Tondreau and J. M. Boncella, Polyhedron, 2016, 116, 96.
15 A. Mukherjee, A. Nerush, G. Leitus, L. J. W. Shimon, Y. Ben
David, N. A. E. Jalapa and D. Milstein, J. Am. Chem. Soc.,
2016, 138, 4298.
16 S. Elangovan, C. Topf, S. Fischer, H. Jiao, A. Spannenberg,
W. Baumann, R. Ludwig, K. Junge and M. Beller, J. Am.
Chem. Soc., 2016, 138, 8809.
3 (a) R. C. van der Drift, E. Bouwman and E. Drent,
J. Organomet. Chem., 2002, 650, 1; (b) R. Uma, C. Crévisy and
R. Grée, Chem. Rev., 2003, 103, 27; (c) V. Cadierno, P.
Crochet and J. Gimeno, Synlett, 2008, 1105; (d) L. Mantilli
and C. Mazet, Chem. Lett., 2011, 40, 341; (e) N. Ahlsten, A.
Bartoszewicz and B. Martín-Matute, Dalton Trans., 2012, 41,
1660; ( f ) P. Lorenzo-Luis, A. Romerosa and M. Serrano-Ruiz,
ACS Catal., 2012, 2, 1079; (g) D. Cahard, S. Gaillard and J.-L.
Renaud, Tetrahedron Lett., 2015, 56, 6159; (h) H. Li and C.
Mazet, Acc. Chem. Res., 2016, 49, 1232.
4 (a) H.-X. Zheng, Z.-F. Xiao, C.-Z. Yao, Q.-Q. Li, X.-S. Ning,
Y.-B. Kang and Y. Tang, Org. Lett., 2015, 17, 6102; (b) S.
Martinez-Erro, A. Sanz-Marco, A. Bermejo Gómez, A.
Vázquez-Romero, M. S. G. Ahlquist and B. Martín-Matute,
J. Am. Chem. Soc., 2016, 138, 13408; (c) H.-X. Zheng, C.-Z.
Yao, J.-P. Qu and Y.-B. Kang, Org. Chem. Front., 2018, 5, 1213.
5 (a) S. Elangovan, C. Topf, S. Fischer, H. J. Jiao, A.
Spannenberg, W. Baumann, R. Ludwig, K. Junge and M.
Beller, J. Am. Chem. Soc., 2016, 138, 8809; (b) F.
Kallmeier, T. Irrgang, T. Dietel and R. Kempe, Angew.
Chem., Int. Ed., 2016, 55, 11806; (c) A. Bruneau-Voisine,
D. Wang, T. Roisnel, C. Darcel and J. B. Sortais, Catal.
Commun., 2017, 92, 1; (d) N. A. Espinosa-Jalapa, A.
Nerush, L. J. W. Shimon, G. Leitus, L. Avram, Y. Ben-
David and D. Milstein, Chem. – Eur. J., 2017, 23, 5934;
(e) M. Garbe, K. Junge, S. Walker, Z. H. Wei, H. J. Jiao,
A. Spannenberg, S. Bachmann, M. Scalone and M. Beller,
Angew. Chem., Int. Ed., 2017, 56, 11237; ( f ) H. Y. Ge,
X. Y. Chen and X. Z. Yang, Chem. – Eur. J., 2017, 23,
8850; (g) K. S. Rawat and B. Pathak, Catal. Sci. Technol.,
2017, 7, 3234; (h) M. Glatz, B. Stöger, D. Himmelbauer,
L. F. Veiros and K. Kirchner, ACS Catal., 2018, 8, 4009.
6 (a) R. He, Z.-T. Huang, Q.-Y. Zheng and C. Wang, Angew.
Chem., Int. Ed., 2014, 53, 4950; (b) D. H. Nguyen, X. Trivelli,
17 R. A. Farrar-Tobar, Z. Wei, H. Jiao, S. Hinze and J. G. de
Vries, Chem. – Eur. J., 2018, 24, 2725.
18 I. Fernández and F. M. Bickelhaupt, Chem. Soc. Rev.,
2014, 43, 4953.
6334 | Catal. Sci. Technol., 2019, 9, 6327–6334
This journal is © The Royal Society of Chemistry 2019