Manganese Pincer Complexes for the Base-Free, Acceptorless Dehydrogenative Coupling of Alcohols to Esters: Development, Scope, and Understanding

Article abstract of DOI:10.1021/acscatal.6b03554

Aliphatic PNP pincer-supported earth-abundant manganese(I) dicarbonyl complexes behave as effective catalysts for the acceptorless dehydrogenative coupling of a wide range of alcohols to esters under base-free conditions. The reaction proceeds under neat conditions, with modest catalyst loading and releasing only H₂ as byproduct. Mechanistic aspects were addressed by synthesizing key species related to the catalytic cycle (characterized by X-ray structure determination, multinuclear (¹H, ¹³C, ³¹P, ¹⁵N, ⁵⁵Mn) NMR, infrared spectroscopy, inter alia), by studying elementary steps connected to the postulated mechanism, and by resorting to DFT calculations. As in the case of related ruthenium and iron PNP catalysts, the dehydrogenation results from cycling between the amido and amino-hydride forms of the PNP-Mn(CO)₂ scaffold. For the dehydrogenation of alcohols into aldehydes, our results suggest that the highest energy barrier corresponds to the hydrogen release from the amino-hydride form, although its value is close to that of the outer-sphere dehydrogenation of the alcohol into aldehyde. This contrasts with the ruthenium and iron catalytic systems, where dehydrogenation of the substrate into aldehyde is less energy-demanding compared to hydrogen release from the cooperative metal-ligand framework.

Full text of DOI:10.1021/acscatal.6b03554

Subscriber access provided by University of Newcastle, Australia
Article
Manganese pincer complexes for the base-free, acceptorless dehydrogenative
coupling of alcohols to esters: development, scope and understanding.
duc hanh nguyen, Xavier Trivelli, Frederic Capet, Jean-Francois Paul, Franck Dumeignil, and Régis M. Gauvin
ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.6b03554 • Publication Date (Web): 30 Jan 2017
Downloaded from http://pubs.acs.org on February 1, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
ACS Catalysis is published by the American Chemical Society. 1155 Sixteenth Street
N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 14
ACS Catalysis
1
2
3
4
5
6
Manganese pincer complexes for the base-free, acceptorless
dehydrogenative coupling of alcohols to esters: development,
scope and understanding
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
b
a
a
a
Duc Hanh Nguyen , Xavier Trivelli , Frédéric Capet , Jean-François Paul , Franck Dumeignil ,
Régis M. Gauvin *
a,
a
:
Univ. Lille, CNRS, Centrale Lille, ENSCL, Univ. Artois, UMR 8181 - UCCS - Unité de Catalyse et Chimie du Solide,
F-59000 Lille, France.
b
:
Univ. Lille, CNRS, UMR 8576 - UGSF - Unité de Glycobiologie Structurale et Fonctionnelle, F-59000 Lille, France.
ABSTRACT: Aliphatic PNP pincer-supported Earth-abundant manganese(I) dicarbonyl complexes behave as effective
catalysts for the acceptorless dehydrogenative coupling of a wide range of alcohols to esters under base free conditions.
Reaction proceeds under neat conditions, with modest catalyst loading and releasing only H as by-product. Mechanistic
2
aspects were addressed by synthesizing key species related to the catalytic cycle (characterized by X-ray structure deter-
1
13
31
15
55
mination, multinuclear ( H, C, P, N, Mn) NMR, infrared spectroscopy, inter alia), by studying elementary steps con-
nected to the postulated mechanism, and by resorting to DFT calculations. As in the case of related ruthenium and iron
PNP catalysts, the dehydrogenation results from cycling between the amido and amino-hydride forms of the PNP-
Mn(CO) scaffold. For the dehydrogenation of alcohols into aldehydes, our results suggest that the highest energy barrier
2
corresponds to the hydrogen release from the amino-hydride form, although its value is close to that of the outer-sphere
dehydrogenation of the alcohol into aldehyde. This contrasts with the ruthenium and iron catalytic systems, where dehy-
drogenation of the substrate into aldehyde is less energy-demanding compared to hydrogen release from the cooperative
metal-ligand framework.
Keywords: Dehydrogenative Coupling • Manganese • Pincer Ligands • Metal-Ligand Cooperativity • Green Chemis-
try • Esters
INTRODUCTION
reaction appears as an environmentally friendly, oxidant-
free and atom-efficient catalytic oxidation process that
simultaneously produces esters and hydrogen. Accord-
Esters and their derivatives occupy a central position in
the chemistry of naturally occurring and synthetic organic
4
1
ingly, symmetric and mixed primary-secondary esters and
lactones as well as polyesters derivatives can be sustaina-
compounds. Consequently, methods for the synthesis
and conversion of these functional groups are of prime
importance for the fine and bulk chemical industry.
Common routes for ester preparation rely on the direct
reaction of carboxylic acids and acid derivatives with
hydroxylic compounds, including reactions of acids with
alcohols, alcoholysis of acyl halides, anhydrides, ketenes,
5
bly synthesized.
Scheme 1. Reaction network for ADC reaction of al-
cohols to esters.
1
nitriles and amides, and transesterifications. Transition-
metal
approaches
such
as
alkoxycarbonyla-
tion/carbonylation processes under CO/CO pressure also
2
produce ester compounds from laboratory- to industrial-
2
scale production. In addition, direct catalytic oxidative
esterifications of alcohols with molecular oxygen and of
3
aldehydes with alcohols have also received attention.
Despite all of these well-established methodologies, the
development of environmentally benign and cost-
effective procedures for the synthesis of esters continues
to attract considerable interest. In this respect, the prima-
ry alcohol acceptorless dehydrogenative coupling (ADC)
The proposed reaction pathway involves a dehydrogena-
tive coupling sequence (Scheme 1). The alcohol is primari-
ACS Paragon Plus Environment

ACS Catalysis
Page 2 of 14
ly dehydrogenated into an aldehyde that further reacts [Mn(Br)(CO) (N{CH CH (PR )} ] complexes for N-
2
2
2
2
2
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
with a second molecule of alcohol to give a hemiacetal
intermediate, which finally undergoes dehydrogenation
into an ester. All elementary steps are reversible. From a
thermodynamic viewpoint, alcohol dehydrogenation is
alkylation of amines with alcohols, and hydrogenation of
esters, that proceed with both high catalyst and base
loadings ([Mn] = 2-3 mol%, [t-BuOK] = 10-75 mol%).
4
,6
18
generally an uphill process: the release of H gas however
2
contributes to shift the equilibrium favorably under open
system conditions.
Figure 1. Manganese pincer complexes
Catalyst design plays an essential role for such a trans-
formation. Molecular systems involving conventional
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
“
innocent” ligands typically require both high catalyst
7
loading and high reaction temperatures. On the other
hand, Milstein and coworkers achieved multiple, major
breakthroughs in this field by introducing the ligand-
metal cooperativity concept: Through nowadays well-
established aromatization-dearomatization process of
heteroaromatic-based pincer complexes, significantly
higher catalytic efficiency is achieved under comparative-
ly milder conditions in the ADC of alcohols, as well as in
8
other related transformations. Several other related ru-
thenium and osmium pincer-type (with PNP, PNN, SNS…
cooperative ligands bearing N-H moiety) complexes, were
9,10
further developed to improve catalytic activity.
The quest for ever greener chemical processes has
sparked considerable interest in the replacement of the
highly active and selective well-established precious metal
catalysts by finding new earth-abundant metal-based
counterparts that usually have lesser health and environ-
In the present study, we focused on manganese complex-
es featuring aliphatic PNP ligands based on the amino-
bis(2-phosphinoethyl) framework. We demonstrate for
the first time that these systems are catalytically active
and selective in ADC reactions under base-free conditions
applied to both aliphatic and benzylic substrates. Mecha-
nistic aspects have also been addressed by combining
DFT calculations and isolation of organometallic species
related to the catalytic cycle.
19
1
1
mental impact. Thus, several nickel- and iron-based
catalysts for the ADC reaction of alcohols to esters were
developed but they generally suffer from lower activi-
12
ty/robustness, as well as from narrower substrate scope.
A further landmark was established when Milstein et al.
reported the first manganese-catalyzed dehydrogenative
coupling of alcohols and amines to form imines with
RESULTS AND DISCUSIONS
release of water and H , using a pyridine-based pincer
2
1
3
ligand (Figure 1). The reaction is effective for benzylic
alcohols while aliphatic counterparts are less efficiently
converted. This demonstrated efficiency of pincer com-
plexes of manganese, an Earth-abundant base metal, has
triggered tremendous interest within the field. Thus,
along the same line, Kirchner et al. reported (PNP)Mn(I)
complexes based on the 2,6-diaminopyridine scaffold that
is also able to catalyze efficiently the same reaction with
benzylic derivative alcohols but still suffers from low
The red dicarbonyl manganese(I) PNP amido complex 1
was straightforwardly prepared in good isolated yield
(67%) by reaction of the starting carbonyl manganese
compound [Mn (CO) ] with 2.0 molar equiv. of the ali-
2
10
phatic PNP pincer (iPr PC H ) NH ligand (L-iPr) in a
2
2
4 2
hexamethyldisiloxane/toluene mixture and under reflux
conditions followed by crystallization in n-pentane at low
20
temperature (Scheme 2). Our NMR and IR spectroscopic
data as well as X-ray structure for 1 are fully in agreement
14
15
15
1
yields with the aliphatic substrates. Shortly afterwards
and during our efforts to develop herein-described re-
with those very recently reported. Notably, the 2D N- H
HMBC NMR spectrum allows determining the downfield
15
sults, Boncella et al. reported [N{CH CH (PiPr )} ]-
chemical shift δ( N) of 231 ppm which is characteristic for
2
2
2
2
2
10
supported Mn(I) dicarbonyl complexes that promote the
the sp hybridization of the amido nitrogen atom. Fur-
thermore, as a complementary spectroscopic tool, we
15
decomposition of formic acid into CO and H2 Inde-
2
55
pendently, Beller et al. also reported the use of similar
resorted to Mn NMR, as valuable information may de-
complexes (formed in-situ, using basic conditions) for the
hydrogenation of nitriles, ketones, and aldehydes. The
duced from comparison of isotropic chemical shift of
various complexes. Mn is a quadrupolar nucleus, with a
16
55
drive towards manganese catalysts is further exemplified
through the report from Kempe et al., who disclosed tria-
zine-based PNP-supported Mn complexes that display
enhanced catalytically activity in hydrogenation of the
5/2 spin, a natural abundance of 100 % and an absolute
receptivity of 0.18, with very similar properties to the
27
commonly used Al nucleus (5/2, 100 %, 0.21, respective-
55
ly). The Mn chemical shift range spreads over 3000 ppm:
17
same substrates. Very recently, Beller et al. further ex-
-2000 to -3000 ppm for Mn(0) and -1000 to -1700 ppm for
21
55
tended
catalytic
applications
of
Mn(I).
Indeed,
Mn NMR spectrum for penta-
ACS Paragon Plus Environment

Page 3 of 14
ACS Catalysis
1
6
coordinated 1 displays a broad peak centered at -1635 ppm,
compound 2-Br, with the PNP framework in meridional
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
with a full width at half maximum (FWHM) of about 1850 Hz
(Table 1).
configuration and the two carbonyl ligands in mutually
cis position. The N-H and the BH moieties (the latter
4
being coordinated as a monohapto borohydride ligand
onto the metal center) are in syn configuration. The N1-
Mn1 bond length (Mn1-N1 = 2.1098(16) Å) is significantly
Scheme 2. Ligands and complexes involved in this
study.
15
longer with respect to that of 1 (Mn1-N1 = 1.882(1) Å ) as
2
expected from the nitrogen sp hybridization for the lat-
CO
PiPr2
ter.
H
0
.5 eq. Mn
2
(CO)10
N
N
iPr P
2
PiPr2
Mn
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
PhMe/O(SiMe
3
)
2
P
L-iPr
iPr
CO
2
1
i) HOTf
MnBr(CO)5
4
ii) NaBH ,
EtOH/PhMe
BH3
H
Br
Mn
CO
H
N
H
PiPr
CO
NaBH
EtOH/PhMe
4
PiPr2
N
2
Mn
P
iPr
P
CO
iPr
2
2
CO
2-BH
2-Br
4
On the other hand, a one-pot, two-step synthesis starting
from [Mn(Br)(CO) ] and L-iPr afforded the new dicar-
5
bonyl manganese (I) PNP amino complex 2-BH as an
4
orange-yellow microcrystalline solid in good yield (62%),
with the reported 2-Br as intermediate species which is
1
6
reacted with NaBH4 (Scheme 2). Alternatively, 2-BH4
was also prepared in overall 59 % isolated yield by reac-
tion of 1 with the trifluoromethanesulfonic acid (HOTf)
Figure 2. ORTEP view of solid-state structure of 2-BH . Ellip-
4
soids are given at the 50% probability level. iPr groups and
H-atoms (except for N-H and BH ) are omitted for clarity.
4
Selected bond distances (Å): Mn1-P2 = 2.2819(6), Mn1-P1 =
followed by the treatment with NaBH . The complex was
4
2
1
.2937(6), Mn1-N1 = 2.1098(16), Mn1-C17 = 1.768(2), Mn1-C18 =
.758(2), Mn1-H1b = 1.77(2), N1-H1 = 0.85(2), O1-C17 = 1.156(2),
O2-C18 = 1.164(2), B1-H1b = 1.18(2), B1-H1c = 1.06(3), B1-H1c =
fully characterized by IR and multinuclear NMR spectros-
copies and elemental analysis (see SI). This compound is
stable enough to be shortly exposed to air without notice-
able decomposition.
1
1
.10(3), B1-H1e = 1.10(3). Selected angles (deg): P2-Mn1-P1 =
66.59(2), P2-Mn1-H1b = 88.6(6), P1-Mn1-H1b = 88.6(6), N1-
Mn1-P2 = 83.37(5), N1-Mn1-P1 = 83.47(5), N1-Mn1-H1b =
9.1(6), C17-Mn1-P2 = 95.64(7), C17-Mn1-P1 = 97.37(7), C17-
Mn1-N1 = 177.14(8), C17-Mn1-H1b = 88.2(7), C18-Mn1-P2 =
3.31(7), C18-Mn1-P1 90.12(7), C18-Mn1-N1 93.59(8), C18-Mn1-
55
8
Table 1. Mn NMR data for selected complexes.
9
Complexes
Solvent
C D
δ (ppm)
-2270
-1145
FWHM (Hz)
18
C17 = 89.14(10), C18-Mn1-H1b = 176.9(6). O1-C17-Mn1 =
178.2(2), O2-C18-Mn1 = 177.30(19).
[Mn (CO) ]
2
10
6
6
[
MnBr(CO)5]
THF-d8
C D
760
The IR spectrum displays two strong bands at 1913 and
a
1
-
1
-1635
1850
7
8
1836 cm characteristic for accounting for terminal CO
1
elongation. The presence of
η -BH is evidenced by the
4
2
-Br
C D
-1456
-1390
-1450
2660
1660
ꢀ1
22
6
6
ν(B-H) bands at 2373, 2348 and 2039 cm . In addition,
-
1
1
NH stretching bond is observed at 3215 cm . On the H
NMR spectrum, the borohydride ligand resonates as a
very broad peak centered at -3.15 ppm (FWHM of about
2-BH4
CD Cl
2
2
b
2-OBn
C D
6000
6
6
3
00 Hz) which indicates fast hydride exchange on the
a
b
NMR temperature measurements: 300K, : 323K, : 280K.
1
1
NMR time scale at RT. Moreover, the H- H NOESY NMR
spectrum features a cross-peak between the NH and Mn-
HBH signals in agreement with the solid-state structure
3
,
Single crystals of 2-BH were subjected to X-ray diffrac-
4
where both mutually located in syn-position (See SI).
tion studies (Figure 2). The complex features a distorted
octahedral geometry highly similar to that of reported
3
1
1
P{ H} spectrum features a sharp singlet at 88.2 ppm
which is upfield-shifted with respect to 1 (113.1 ppm). The
ACS Paragon Plus Environment

ACS Catalysis
N resonance is located at 47 ppm, which is consistent
Page 4 of 14
1
5
3
10
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
with the sp hybridization for the amino nitrogen atom.
We next probed the efficiency of complexes 1 and 2-BH4
1
3
1
The C{ H} spectrum comprises two broad signals charac-
teristic of CO ligands centered at 231.1 and 229.2 ppm.
for the catalytic ADC reaction of hexan-1-ol under base
free conditions, in neat substrate and under an argon
stream (Table 2). First, the reaction was carried out at
1
3
55
Indeed, C and Mn resonance frequencies are very close
respectively 100.6 and 99.24 MHz at 9.4 T), leading to an
efficient scalar relaxation mechanism which broadens the
(
1
50°C and in the presence of 0.1 mol% of 1. After 24 h,
hexan-1-ol was moderately converted to the correspond-
ing ester, hexyl caproate, in 34% yield (Table 2, Entry 1).
Very low amount of n-hexanal (< 1.0%) was also obtained.
Performing the reaction for prolonged reaction times
1
3
signal from C carbonyl ligands directly bound to the Mn
21
center. Compound 2-BH gives rise to a singlet at -29.9
4
1
1
1
55
ppm on the B{ H} NMR spectrum. The Mn NMR spec-
trum for this hexa-coordinated complex shows a broad
peak centered at -1390 ppm (FWHM of about 1660 Hz)
downfield relative to that of penta-coordinated 1 (ꢁδ =
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
48h) does not improve yield, which indicates that the
catalyst is no longer active. Increasing gradually the cata-
lyst loading from 0.2 to 0.6 mol% resulted in the yield
improvement from 58 to 92% (Table 2, Entries 2-6). Selec-
tivity remained unchanged, namely higher than 99%. At
2
45 ppm, Table 1).
0
.6 mol% catalyst concentration, monitoring the yield
-1
Table 2. Catalytic optimization for ADC reaction of
hexan-1-ol.
with time allowed determining the TOF of about 20 h
0
(
Figure 3). After 24h, an excellent yield of 92% was
achieved (Table 2, Entry 6). Next, influence of reaction
temperature was studied. Decreasing the reaction tem-
perature from 150 to 130°C lead to decreased yield, from
9
2 to 65% (Table 2, Entries 6-9), respectively.
T
(°C)
Yield
(%)
Entry
[Mn] (% mol)
1 (0.1)
a
TON
340
290
230
183
In the view of developing easily accessible systems, the
use of in situ systems was also probed, first by selecting
reagents mixtures that would lead to formation of pre-
catalyst 1. The [Mn (CO) ]/L-iPr mixture (1:2 molar ratio)
catalyzes the coupling reaction to form the ester in 51%
yield (Table 2, Entry 10). In order to ascertain the for-
mation of the same catalytic species under such condi-
tions, the reaction of 1 with two molar equiv. of L-iPr in
toluene was independently performed under reflux condi-
tions. Indeed, only ca. 64% of 1 was observed after 24 h
along with unreacted PNP ligand according to P{ H}
NMR. In analogy with the report from Beller and cowork-
ers for catalytic hydrogenation,
rivative 2-Br was combined with excess tBuOK to gener-
ate 1 in the reaction mixture, which afforded the ester
with only 20% yield (Table 2, Entry 11). However, in this
case, formation of by-products occurred, presumably due
1
150
150
150
150
150
150
140
135
130
34
58
69
73
80
92
83
71
2
10
2
3
1 (0.2)
1 (0.3)
4
5
6
7
8
9
10
1 (0.4)
1 (0.5)
158
153
3
1
1
1 (0.6)
1
6,18
the bromo-amino de-
1 (0.6)
138
1 (0.6)
118
1 (0.6)
65
108
10
to base-promoted aldolization reactions. The bromo
derivative alone is not active (Table 2, Entry 12). Compa-
rable results were obtained with the [MnBr(CO) ]/L-iPr
[Mn
(CO)10]/L
iPr (0.6)
150
150
51
85
33
2
-
b
1
1
2-Br (0.6)
20
5
systems: only 12 % yield in the presence of excess alkox-
ide, and no conversion in its absence (Table 2, Entries 13
and 14, respectively).
12
2-Br (0.6)
150
150
0
0
[
MnBr(CO) ]/L-iPr
5
1
3
12
20
b
(0.6)
[MnBr(CO) ]/L-iPr
5
1
4
5
150
150
150
150
150
0
43
0
0
72
0
Figure 3. Yield vs. time for ADC reaction of hexan-1-ol
(0.6)
(
150°C, 0.6 mol% of 1).
1
[Mn (CO)10]/L-Cy (0.6)
2
[
Mn (CO) ]/L-tBu
2
10
16
(
0.6)
1
7
2-BH
4
(0.6)
84
17
140
28
1
8
3 (0.6)
Reaction conditions: Hexan-1-ol (1.5 mL), 24 h, argon stream.
a
1
b
:
Hexyl caproate yield determined by H NMR. : reaction in
the presence of tBuOK (2 mol%).
ACS Paragon Plus Environment

Page 5 of 14
ACS Catalysis
major product is the symmetric ester along with trace
1
2
3
4
5
6
7
8
9
amount of corresponding aldehyde (< 1% in most cases).
Long, linear aliphatic chain C -C alcohols were trans-
8
18
formed into the corresponding symmetric waxy esters in
good yield from 76 to 83% (Table 3, entries 4-7). n-
pentan-1-ol can also be efficiently converted to pentyl
valerate in 84% yield (Table 3, Entry 3). Reactions with
lighter alcohols such as propan-1-ol and butan-1-ol were
carried out at lower temperatures (110 and 130°C), thus
requiring longer reaction times (72 h) to reach satisfacto-
ry yield of 74 and 89%, respectively (Table 3, Entries 1 and
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
). However, ethanol, which features comparatively lower
Influence of substituent variation at the phosphorous
atoms of the PNP ligand was further studied. Combina-
tion of [Mn (CO) ] with HN(CH CH PCy ) (L-Cy) bear-
boiling point, afforded ethyl acetate with only 3% yield
under reflux conditions.
2
10
2
2
2 2
ing bulkier cyclohexyl substituents gave slightly lower
yield of 43% with respect to that of [Mn (CO) ]/L-iPr
Table 3. Catalytic ADC reaction of various alcohols.
2
10
system (51 %, Table 2, Entry 15). The [Mn (CO) ]/L-tBu
2
10
Yield
system (L-tBu = HN(CH CH PtBu ) ) bearing very bulky
2
2
2 2
Entry Substrate
T(°C) t(h)
TON
a
t-butyl groups is catalytically inactive (Table 2, Entry 16).
It is worthy of note that the less moisture/oxygen sensi-
(%)
74
1
110
72
72
24
24
123
148
140
138
tive pre-catalyst 2-BH (with respect to 1) also efficiently
4
promotes the ADC reaction, achieving 84% yield of hexyl
2
3
4
130
150
150
89
84
83
caproate (Table 1, Entry 17). This indicates that 2-BH is
4
readily converted into the active species 1 under base-free
reaction conditions, as previously reported with PNP-
1
0
ruthenium borohydride derivatives.
5
150
150
24
24
76
82
127
137
Scheme 3. Synthesis of [Mn(H)(CO) (MePNP)] 3.
2
Me
CO
CO
6
N
PiPr
2
i) MeOTf
N
PiPr
2
Mn
Mn
P
iPr2
ii) NaBH4,
MeOH/PhMe
P
iPr2
CO
CO
7
8
9
150
120
150
150
24
72
24
24
80
70
74
67
133
117
123
112
H
1
3
To further probe the influence of the amino proton on the PNP
framework, or more specifically, the importance of the metalꢀ
ligand cooperativity, complex 3, [Mn(H)(CO) (MePNP)]
2
(
MePNP = MeN(CH CH PiPr) ) was prepared by N-
2 2 2
10
methylation of 1 with methyl trifluoromethanesulfonate
and subsequent reaction with NaBH in MeOH/toluene
4
(
Scheme 3; see SI for full experimental details). This com-
11
120
150
150
72
24
24
0
0
pound is unable to perform catalysis via involvement of
the amido/amino-hydride pair, contrary to derivatives
featuring the parent L-iPr ligand (vide infra). Based on
NMR data, we propose that the methyl and hydride
groups are in mutually anti positions in 3, in contrast to
what is observed for the 2-H compound (vide infra). No
isomer was detected. This complex proved to be signifi-
12
13
37
95
62
158
cantly less active with respect to 1 or 2-BH : Only 17% of
4
1
4
5
150
150
24
24
88
82
147
137
the hexyl caproate was formed under similar conditions
(
Table 2, Entry 18). This observation is indicative of the
strong beneficial influence of the metal-ligand cooperativ-
ity in the case of the N-H derivatives.
1
With optimized reaction conditions in hand (0.6 mol% 1,
1
50°C, 24h), a substrate scope including aliphatic and
benzylic alcohols was studied (Table 3). In all cases, the
ACS Paragon Plus Environment

ACS Catalysis
Page 6 of 14
(
Table 4, Entry 1) instead of 83% (Table 3, Entry 4). MeOH
has an even lower detrimental influence (78% yield, Table
, Entry 2). On the contrary, the presence of butyric acid
2 mol%) strongly inhibits the catalytic reaction (Table 4,
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
16
17
150
24
66
12
110
4
(
150
150
24
24
20
Entry 3) that could be explained by the formation of car-
boxylate manganese [Mn(C H CO )(CO) (L-iPr)] complex
which is catalytically no longer active. In addition, as
3
7
2
2
1
5
b
18
95
158
23
experienced in the related iron chemistry, the presence
of a halide salt (KBr in the present case) has some nega-
tive impact on the conversion presumably due to partial
back-formation of catalytically inactive 2-Br or a bromide
derivative (Table 3, Entry 4). When considering the com-
paratively high ligand cost and the poorer performance
relative to benchmark Ru systems, the benefits from use
of manganese systems are not obvious. Nevertheless, this
calls for deeper investigations on such base-metal systems
in order to make these practically useful, which motivated
detailed studies described below.
Reaction conditions: Catalyst 1 (0.6 mol%), alcohol (12.04
mmol), argon stream. : Ester yield determined by H
NMR. : γ-butyrolactone was formed as the only product.
a
1
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
b
Catalytic efficiency is significantly dependent on the sub-
strate bulk, as demonstrated when studying branched
aliphatic alcohols. Indeed, 2-methylpropan-1-ol afforded
isobutyl isobutyrate in 70% yield (Table 3, Entry 8) that is
1
9% lower than that obtained with the linear isomer (n-
butanol). Similarly, both 3-methylbutan-1-ol and 2-
methylbutan-1-ol isomers afforded the corresponding
esters in 74 and 67% yields, respectively (Table 3, Entries
9
and 10). The more sterically crowded derivative
Table 4. Catalytic ADC reaction of alcohols in pres-
ence of different additives.
branched 2-ethyl hexan-1-ol proved to be less reactive, as
only 37% yield of corresponding ester was formed (Table
, Entry 12) while the very sterically crowed neopentyl
alcohol is not converted (Table 3, Entry 11). Under similar
conditions, benzyl alcohol was converted to benzyl ben-
zoate with a very good yield of 95% (Table 3, Entry 13). 4-
isopropyl and 3,5-dimethyl benzyl alcohols produced the
corresponding esters in slightly lower yield (88 and 82%,
respectively, Table 3, Entries 14 and 15). 4-methoxy benzyl
alcohol was transformed to the corresponding ester in
3
Addi-
tive
%mol)
En
try
Yield
(%)
Substrate
t(h)
TON
(
1
H O (5)
24
24
73
78
122
130
2
MeOH
2
(
5)
Butyric
acid (2)
6
6% yield while the 2-methoxybenzyl alcohol isomer is
3
24
24
6
10
less reactive: Only 12% of the corresponding ester was
obtained (Table 3, Entries 16 and 17). Interestingly, butan-
,4-diol was transformed to the γ-butyrolactone in excel-
4
KBr (2)
50
83
1
Reaction conditions: Catalyst 1 (0.6 mol%), Alcohol (12.04
a
1
lent yield of 95% (Table 3, Entry 18).
mmol), 150°C, Argon stream. : Ester yield determined by H
NMR.
In order to assess the tolerance of these catalytic systems
toward common impurities, we studied the influence of
different additives on the outcome of the reaction. At 5
mol% concentration, water has only a slight impact on
the ADC reactions of octan-1-ol with 73% yield reached
Scheme 4. Proposed mechanism for manganese-catalyzed ADC reaction.
ACS Paragon Plus Environment

Page 7 of 14
ACS Catalysis
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Mechanistic studies
ppm assigned to the two terminal CO ligands, while the
methylenic carbon [Mn]-OCH Ph resonates as a singlet at
As previously demonstrated, the ADC reaction of alcohols
to esters proceeds via a dehydrogenative coupling se-
quence: i) dehydrogenation of the alcohol substrate to the
corresponding aldehyde, ii) formation of a hemiacetal
from the aldehyde and an alcohol, and its dehydrogena-
2
5
5
74.0 ppm. Mn NMR spectrum for 2-OBn (recorded in
the presence of a large excess of benzyl alcohol, about ~12
equiv.) features a very broad peak centered at - 1450 ppm
1
1
(FWHM ≈ 6000 Hz, Table 1). The 2D H- H NOESY NMR
comprises a cross-peak between the NH proton and aro-
matic protons, thus indicating proximity between the NH
and benzyl moieties, in line with the syn-configuration
observed in the solid state that further supports a syn-
addition mechanism of hydroxy function on 1. Further-
more, in the presence of benzyl alcohol, observation of
positive cross-peaks between two pairs of protons
4,6
tion into the ester. Inspired by previous reported works
on ruthenium- and iron-catalyzed ADC and in line with
above-described catalytic observations,
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5b,6,12c,13,24
we pro-
pose a metal-ligand cooperative mechanism involving 1
and the corresponding amino-hydride 2-H as catalytical-
ly-relevant species, with formation of aldehyde as a pri-
mary product (detected in small quantities under our
catalytic conditions, vide supra) (Scheme 4). More pre-
cisely, this consists of outer-sphere dehydrogenation of
the alcohol into aldehyde (cycle A, Scheme 4) with the
amido species 1 acting as a hydrogen acceptor, thus form-
ing amino hydride species 2-H. The catalytic cycle is then
(NH/HOCH Ph and Mn-OCH Ph/HOCH Ph) demon-
2
2
2
strates that the free benzyl alcohol and the benzyloxy
moiety are involved in chemical exchange.
completed by H loss from this compound and regenera-
2
tion of the active species 1. A similar cycle converts the
hemiacetal (generated from the aldehyde and the alcohol
substrate) into the ester (cycle B, Scheme 4).
In order to gain detailed fundamental insight into the
mechanism, specific elementary steps were investigated.
Thus, addition of excess benzyl alcohol (3.0 molar equiv.)
to the red solution of the catalytic active species 1 in C D
at room temperature induced an immediate color change
to yellow, leading to formation of the alkoxide 2-OBn
6
6
(
Scheme 5). The O-H bond activation reaction is very
rapid (being completed before NMR spectrum acquisi-
tion) and presumably occurs by means of metal–ligand
cooperation, as the O-H bond formally adds over the Mn-
N amido moiety. Multinuclear NMR spectroscopy (see
below) unambiguously evidences the formation of the
benzyloxy manganese complex 2-OBn as the only organ-
ometallic species in solution (See SI). Crystallization from
pentane at ꢀ20°C afforded single crystals amenable to X-ray
diffraction studies.
In the solid-state, the complex features a distorted octa-
hedral geometry, similar to that of related complexes,
Figure 4. ORTEP view of the solid-state structure of 2-OBn.
Ellipsoids are given at the 50% probability level. iPr and H-
atoms (except for N-H) are omitted for clarity. Selected bond
distances (Å): Mn1-P1 = 2.2943(9), Mn1-P2 = 2.3030(9), Mn1-
O1 = 2.040(2), Mn1-N1 = 2.105(3), Mn1-C24 = 1.770(3), Mn1-
C25 = 1.753(4), N1-H1 = 0.85(4), O1-C17 = 1.390(3), O3-C25 =
1.181(4), O2-C24 = 1.173(4), Selected angles (deg): P1-Mn1-P2
= 163.55(4), O1 Mn1-P1 = 86.28(7), O1-Mn1-P2 = 83.76(7), O1-
Mn1-N1 = 76.13(10), N1-Mn1-P1 = 82.25(8), N1-Mn1-P2 =
1
6
such as 2-Br or 2-BH4 described above: The amino-
phosphine pincer ligand is coordinated in meridional
fashion, the two terminal carbonyl ligands are in mutually
cis configurations, one being trans to the amino moiety,
and the residual position is occupied by the alkoxide
ligand, which is arranged in syn configuration with the N-
H function (Figure 4). The IR spectrum displays two in-
tense, sharp characteristic bands for ν(CO) at 1878 and
-
1
-1
8
2.68(8), C24-Mn1-P1 = 97.03(11), C24-Mn1-P2 = 97.55(11),
C24-Mn1-O1 = 99.93(12), C24-Mn1-N1 = 176.03(14), C25-Mn1-
P1 93.71(11), C25-Mn1-P2 94.32(11), C25-Mn1-O1
71.85(13), C25-Mn1-N1 = 95.78(14), C25-Mn1-C24 = 88.16(16).
1
799 cm , along with a ν(N-H) signal at 3164 cm .
3
1
1
In solution, the P{ H} NMR spectrum of 2-OBn consists
=
=
=
1
of a singlet at 84.2 ppm. In the H NMR spectrum, the
1
OCH Ph and NH protons resonate as a singlet at 5.0 ppm
2
O2-C24-Mn1 = 177.2(3), O3-C25-Mn1 = 177.8(3).
3
and a broadened triplet centered at 3.43 ppm ( J
Hz), respectively. Consistently, the sp NH amino nitro-
gen resonates at 56.5 ppm, as deduced from the H- N
HSQC NMR spectrum. The C{ H} spectrum features two
characteristic broad signals centered at 231.7 and 226.8
= 11.8
H-H
3
1
15
Indeed, the reaction of 1 with alcohol is reversible: disso-
1
3
1
lution of 2-OBn in C D or treatment of 1 with 1.0 equiv.
6 6
ACS Paragon Plus Environment

ACS Catalysis
Page 8 of 14
of benzyl alcohol resulted in the presence of both 1 and 2- we propose that the isomer 2-OBn’ features NH and ben-
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
OBn in equilibrium. The 2-OBn/1 molar ratio decreases
zyloxy ligands in mutually anti-position (Scheme 5).
with increasing temperature, evolving from 5.1 at 275 K to
Heating a solution of 2-OBn in C D at 80°C for 14 h in a
closed system resulted in the formation of two new hy-
dride manganese complexes (~ 11% conversion) in a 1.4:1
6
6
0
.93 at 322 K. The corresponding thermodynamic values
for this exchange obtained through a van’t Hoff plot are
0
-1
0
-1
-1
ꢀ
H = 10.5 kcal.mol and ꢀS = 26.5 cal.K .mol (Scheme 5,
ratio that are attributed to the two isomers of formula
16
See SI). Thus, the amido species 1 reacts with the alcohol
substrate via an exothermic reaction, forming an alkoxide
species. Consistently, when under catalytic reaction con-
dition, that is at 110-150°C, a red color of the reaction mix-
ture was observed (as expected from the presence of 1 in
solution). On the other hand, a yellow-green solution was
obtained upon addition of 1 to alcoholic substrates at
room temperature (as expected from the formation of
yellow compound 2-OBn).
[
(
(Mn)(L-iPr)(H)(CO) ], 2-H
and 2-H’, respectively
2
1
Scheme 6). The H NMR spectrum exhibits two charac-
teristic triplets centered at -5.55 (JHP = 50.7 Hz) and -5.76
31
1
ppm (JHP = 51.3 Hz). Accordingly, on the P{ H} spectrum,
two new respective singlets are observed at 109.2 and 106.9
ppm. Benzyl benzoate was observed as the only organic
by-product, and no benzaldehyde was detected. The for-
mation of 2-H from 2-OBn can be rationalized by 1) gen-
eration of 1 and benzyl alcohol, 2) ADC of this alcohol
into the corresponding ester with formation of 1 and H2,
and 3) reaction of 1 with H2 affording 2-H.16 Addition of
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 5. Reaction of 1 with benzylic alcohol and the
1
.0 equiv. of benzaldehyde to 1 at room temperature does
0
0
1
15
31
van’t Hoff plot for ꢀH and ꢀS determination.
not result in any change on the H, N and P NMR spec-
tra, indicating that the coordination of the aldehyde to
the manganese center should be very weak. This contrast
with the reported formation of an alkoxide derivative
from reaction with benzaldehyde within the pyridine-
Ph
H
O
H
CO
Mn
CO
PiPr
PhCH OH
PiPr
PiPr2
N
2
2
N
2
N
Mn
Mn
CO
Ph
P
iPr
P
iPr
CO
P
iPr
2
CO
2
CO
2
O
13
based PNP system from Milstein and coworkers.
1
2-OBn
2-OBn'
Scheme 6. Formation of hydrides from 2-OBn.
Ph
H
O
H
H
H
CO
Mn
H
8
14 h
0°C
N
PiPr
2
N
PiPr
2
N
PiPr
2
O
Mn
CO
Mn
+
+
P
iPr
CO
P
iPr
CO
P
iPr
2
CO
Ph
O
Ph
2
2
CO
2-H
.5 %
2
-OBn
2-H'
6
4.5 %
As a comparison, formation of the related alkoxide
M(RO)(H)(CO)(PNHP)] complexes (M = Fe, Ru) from
[
the amido derivatives [M(H)(CO)(PNP)] has also been
described, although their relevance to catalytic cycles
9g,12c
remains a matter of discussion.
Based on DFT calcula-
tions, Jones and Schneider proposed that the alkoxide
iron species (formed from the corresponding amido and
the alcohol) is not part of the ADC dehydrogenative reac-
tion scheme, but is rather involved in the generation of
the hemiacetal intermediate species, with concomitant
As a side issue, when stored at room temperature for
several days at room temperature, the composition of a
solution of 2-OBn slowly evolves and a small amount of
isomer 2-OBn’ (~ 5%) is detected, which was character-
25 31
1
12c
ized in-situ by NMR spectroscopy (See SI). P{ H} signal
is shifted downfield at 92.6 ppm (84.2 ppm for 2-OBn).
re-generation of the amido species. On the other hand,
very recently, Beller proposed that under basic condi-
1
13
The methylenic fragment Mn-OCH features H and
C
2
tions,
the
deprotonated
form
of
formula
-
signals at 5.34 and 74.0 ppm, respectively. The NH moiety
[Ru(OR)(H)(CO)(PNP)] can be involved in the catalytic
3
26
give rise to signals at 4.38 ppm (t, J = 12.6 Hz) and 54.3
HH
cycle. Thus, under our neutral conditions, these alkoxide
derivatives would be considered as off-catalytic cycle
1
15
1
ppm, on the H and N spectrum, respectively. The H N-
H signal is thus significantly downfield shifted from that
of 2-OBn (3.43 ppm), which indicates a different type of
(
reservoir) species when considering dehydrogenation
steps. This may be detrimental to catalytic activity since
their formation induces depletion of amido active species.
Under typical catalytic conditions (150°C, 0.6% mol cata-
15
environment. On the other hand, the N signal is only
marginally affected, which is indicative of overall struc-
tural features similar to those of 2-OBn, (most particular-
ly with a CO ligand in the trans position). Furthermore,
lyst loading), the equilibrium constant K was estimated
eq
-1
to 2.44 mol.L : The initial [1]/[2-OBn] ratio is about ¼,
1
1
H- H NOESY shows no proximity between NH and aro-
matic protons. From these combined spectroscopic data,
which will increase overtime with the consumption of the
benzylic alcohol (i.e. decrease of its concentration in the
catalytic mixture). Thus, this substrate inhibition (reac-
ACS Paragon Plus Environment

Page 9 of 14
ACS Catalysis
tion of 1 with the alcohol generating a catalytically inac-
C18-Mn1-C17 = 90.87(10), C18-Mn1-H = 85.5(8), C17-Mn1-P2 =
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
99.31(7), C17-Mn1-P1 = 95.50(7), C17-Mn1-N1 = 97.26(9), C17-
Mn1-H = 175.1(8), O2-C18-Mn1 = 179.7(2), O1-C17-Mn1 =
172.6(2).
tive species) is not marginal. Furthermore, this may par-
tially account for lower reactivity observed at lower tem-
perature, as the proportion of 2-OBn increases under
such conditions (Table 2). As a further alternative, Gusev
proposed that for ruthenium-based pincer systems, such
alkoxide species would on the contrary be involved within
the ADC catalytic cycle. The reaction thus proceeds via a
Tishchenko reaction mechanism. However, under our
catalytic conditions (0.6 mol%, 150°C, 24h) and using
benzaldehyde as substrate, 2-OBn gave only 16% yield of
benzyl benzoate (to be compared with 95% yield when
starting from benzyl alcohol, Table 3, Entry 13), which
indicates that the Tishchenko type mechanism is not
involved as the main pathway in our case.
In the solid state, 2-H is stable and can be stored under
argon at low temperature for several weeks, while in solu-
tion at room temperature, 2-H very slowly loses dihydro-
gen to form 1, as evidenced by a color change from pale
yellow to red. In addition, the formation of the isomer 2-
H’ was also observed (Scheme 7). Therefore, 2-H was
characterized in solution by multinuclear NMR spectros-
6
d
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
5
1
copy at low temperature (260 K). Notably, the N and H
resonances for the N-H group are determined at 36.8 and
1
5
1
1
.25 ppm, respectively, from 2D N- H HSQC NMR. This is
The last step of the catalytic cycle consists of H release
3
2
consistent with the sp hybridization of the amino nitro-
from the amino-hydride 2-H and regeneration of the
active catalytic amido species 1. We prepared and isolated
the hydride manganese complex 2-H as an off-white mi-
crocrystalline solid at low temperature according to the
gen, as observed for 2-BH and 2-OBn. In a closed system
4
(
NMR tube) and in the absence of substrate, the H2-
release reaction is rapid at 120°C, leading to 80% conver-
sion of 2-H into a mixture of 1, 2-H and 2-H’ within 15min.
16
reported method with slight modification. Crystallization
1
31
1
Follow-up by H or P{ H} NMR of these species concen-
tration was performed at different temperatures (See
Scheme 7 for an example, and see SI for further exam-
ples).
from toluene/pentane mixture at ꢀ20°C afforded single crys-
tals suitable to X-ray diffraction studies. The molecular X-
ray structure reveals a distorted classical octahedral ge-
ometry that is similar to that of 2-BH and 2-OBn (Figure
4
5
). The amino-diphosphine tridentate ligand is coordinat-
Scheme 7. Monitoring of a 2-H solution evolution
over time at 318K.
ed in meridional fashion, the two terminal carbonyl lig-
ands are in mutually cis configurations, one being trans to
the amino moiety, and the residual position is occupied
by the hydride ligand, which is arranged in syn configura-
tion with the N-H function.
Figure 5. ORTEP view of solid-state structure of 2-H. Ellip-
soids are given at the 50% probability level. iPr groups and
H-atoms (except for N-H and Mn-H) are omitted for clarity.
Selected bond distances (Å): Mn1-P2 = 2.2375(7), Mn1-P1 =
Kinetic parameters for the dehydrogenation of 2-H to 1
could be determined by applying a first order kinetic
model. From the Eyring equation, the activation parame-
2.2357(7), Mn1-N1 = 2.1278(19), Mn1-C18 = 1.756(3), Mn1-C17 =
#
-1
ter were estimated to ꢁH of about 19.1 kcal.mol . The fact
that we observe a build-up of 2-H’ (which cannot undergo
1.779(2), Mn1-H = 1.59(2), N1-H1 = 0.86(3), O1-C17 = 1.183(3),
O2-C18 = 1.176(3). Selected angles (deg): P2-Mn1-H = 84.4(8),
P1 Mn1 P2 160.30(3), P1-Mn1-H = 81.7(8), N1-Mn1-P2 =
H syn-elimination) over time is an indication that the
2
81.99(5), N1-Mn1-P1 = 83.31(5), N1-Mn1-H = 86.5(8), C18-Mn1-
conversion of 2-H’ into 2-H is slow compared to the re-
verse reaction (2-H’ being the thermodynamic species).
P2 = 95.91(7), C18-Mn1-P1 = 96.79(7), C18-Mn1-N1 = 171.83(9),
ACS Paragon Plus Environment

ACS Catalysis
Page 10 of 14
On the other hand, treatment of 2-H with ca. 10 molar from 2-H). Likely, the reaction of 2-H with alcohol occurs
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
equiv. of benzylic alcohol at room temperature resulted in
a color change from faint yellow to yellow in few minutes.
Accordingly, NMR studies reveal the formation of 2-OBn
as the main product along with the presence of small
amount of the residual 2-H and its isomer 2-H’ (~10%, 3:1
with the so-called self-promotion-type mechanism, as
already suggested for isoelectronic [M(H)(CO)(PNP)]
1
2c, 24
complexes (M= Ru, Fe) based on DFT studies
: H loss
2
is promoted by the alcohol acting as a proton transfer
shuttle to form amido 1, which is followed by its rapid
reaction with excess alcohol to form 2-OBn. A similar
mechanistic pathway was proposed by Wang and co-
respective ratio) within minutes. Thus, H release oper-
2
ates with much higher rate than that in absence of alco-
hol which evidences the beneficial effect of the latter in
the last step of the catalytic cycle (that is, removal of H2
2
7
workers for a Fe-PNP complex related to 2-BH4.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 6. Computed lowest free-energy pathways for ADC reaction of ethanol with complex 1 at 298K. Free energies ΔG and
−1
enthalpies ΔH (in parenthesis) in kcal.mol computed at the B3PW91/SDD(Mn)/6-31G** level of DFT. Unassisted H2 release and
ethanol-bridged proton transfer pathways are presented as orange and blue lines, respectively.
The ADC reaction mechanism mediated by manganese
catalyst 1 was further examined using DFT calculations.
Ethanol was chosen as a model substrate. Relative free
energies (kcal.mol ) reported in the Figure 6 have been
-1
ACS Paragon Plus Environment

Page 11 of 14
ACS Catalysis
The influence of the presence of alcohol was probed on
obtained at the B3PW91/SDD(Mn)/6-31G** level of densi-
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ty functional theory (DFT) and are reported for a temper-
ature of 298 K. Calculations were focused on cycle A
the dehydrogenation step of 2-H. The first step is the
formation of an adduct 2-H-EtOH resulting from hydro-
gen bonding between the N-H group and the hydroxyl
group (N-H•••O-H) which is slightly endergonic (1.7
(
Scheme 4), as similar mechanistic pathways are expected
for the dehydrogenation of the intermediate hemiacetal
cycle B, Scheme 4).
-
1
(
kcal.mol ). The concerted displacement of the two N-H
and O-H protons leads to the formation of a manganese-
The beginning of catalytic cycle is an adduct 1-EtOH
generated through a N···H-O hydrogen bonding between
the five-coordinate amido complex 1 and EtOH. From this
point, a transfer process of the hydroxyl proton to the
ligand’s nitrogen takes place through transition state 1-
2
coordinated
η
-H2 molecule 1-H -EtOH. The activation
2
-
1
-1
free enthalpy is 17.7 kcal.mol , which is 3.8 kcal.mol
lower than in the direct mechanism. This is in line with
the involvement of the alcohol in this step, as already
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2c,24
proposed in ruthenium and iron systems.
This is also
EtOH-TS to produce an outer-sphere alkoxide complex,
-1
confirmed by our experimental findings, as we observed
1
-OEt-H in endergonic manner of 15.1 kcal.mol . The free-
-1
faster H release from 2-H in the presence of alcohol (vide
energy barrier for O-H bond cleavage is 14.0 kcal.mol . As
a reference point in the energy diagram, we selected the
resting state ethoxide complex 2-OEt, which is calculated
2
supra). The last step consists of H2 release from 1-H2-
EtOH with regeneration of 1-EtOH that proceeds in exer-
-
1
-1
gonic manner (12.8 kcal.mol ). The dissociation of the
hydrogen molecule requires an activation energy of only
to be 3.7 kcal.mol more stable than 1-EtOH. The next
step consists of a α-methylene hydride transfer of the
ethanolate moiety onto the metal center through transi-
tion state 1-OEt-H-TS that corresponds to the free energy
-
1
1
.3 kcal.mol . Indeed, the hydrogen release process from
the MLC framework is only slightly more energy-
demanding than alcohol dehydrogenation into aldehyde
-1
barrier of 2.8 kcal.mol . It leads to the formation of an
-
1
(
19.4 vs. 17.9 kcal.mol , respectively).
adduct complex, 2-H-CH CHO, with an acetaldehyde
3
moiety bridging between the N-H and Mn-H groups of
To further complement our mechanistic investigations,
the reaction of 1 with benzyl alcohol to form 2-OBn was
also computed. DFT calculations show that the reaction is
-1
complex 2-H via an exergonic process of a 7.1 kcal.mol .
The dissociation of acetaldehyde from 2-H-CH CHO
3
-1
-1
leading to formation of 2-H is a barrier-less, 5.1 kcal.mol
-11.0 kcal.mol downhill step, which is fully in agreement
downhill step.
with the experimental value determined above by in situ
-1
NMR studies on benzyl alcohol (-10.5 kcal.mol ). Since
The last step for aldehyde production cycle (A, Scheme 4)
experimentally, the reversible reaction reaches equilibri-
um very quickly even at low temperature, this probably
takes place with very low energy barrier. No direct path-
way was found leading from 2-OBn to 2-H.
is the release of H from 2-H to regenerate the catalytic
2
active species 1. Indeed, two pathways for the formation
of 1 were investigated: i) Direct proton transfer without
alcohol involvement and ii) proton transfer via substrate-
2
4
mediated proton shuttling. The more straightforward
route is the direct transfer of a proton from the ligand
nitrogen to the metal hydride through transition state 2-
H-TS to form a very unstable hexacoordinated intermedi-
CONCLUSIONS
In conclusion, we have demonstrated the first successful
example of use of isolated aliphatic PNP pincer-supported
earth abundant manganese(I) catalysts for acceptorless
dehydrogenative coupling of wide range of alcohols to
esters under base free conditions. The importance of the
cooperativity on the efficiency of the system was demon-
strated by the lesser performances of the system featuring
N-methylated PNP ligand. Mechanistic studies allowed to
unveil some fundamental aspects related to alcohol acti-
vation and to the postulated mechanism. Thus under
reaction conditions, species 1 is mainly present as its
masked analogue 2-OBn (which is more stable) through
reversible reaction with the alcohol substrate. For the
dehydrogenation of alcohols into aldehydes, our results
suggest that the highest energy barrier corresponds to the
hydrogen release from the amino-hydride form, although
its value is close to that of the outer-sphere dehydrogena-
tion of the alcohol into aldehyde. This contrasts with the
ruthenium and iron catalytic systems, where dehydro-
genation of the substrate into aldehyde is less energy-
demanding compared to hydrogen release from the coop-
erative metal-ligand framework. Even though, care should
be taken when comparing these Mn catalysts to the re-
ported Ru and Fe PNP systems, as even if they are
ate 1-H . The reaction was calculated to be endergonic of
2
-1
20.1 kcal.mol . Noteworthy, this direct concerted N-
H/[Mn]-H proton transfer pathway exhibits a free-energy
-1
barrier of 21.5 kcal.mol , which is close to that reported by
-1 16
≠
Beller (20.2 kcal.mol ). The activation enthalpy ΔH
-1
calculated of 21.2 kcal.mol is comparable to the experi-
mental value we determined above (19.1 kcal.mol ). Fur-
-1
thermore, free-energy barriers were found to be 29.4 and
-1
2
4.0 kcal.mol for isoelectronic ruthenium and iron sys-
24
tems, respectively. Consistently, the H release from the
2
isoelectronic [Ru(H) (CO)PNHP)] without assistance of
2
26
alcohol is very slow even under heating conditions. Fi-
nally, the very fast dissociation of H from 1-H through
2
2
-1
transition state 1-H -TS is a 21.1 kcal.mol downhill step
2
and requires a very weak free-energy barrier of only 0.1
-1
kcal.mol to regenerate catalytic species 1. Complex 1 is 1.0
-1
kcal.mol slightly more stable than 2-H which is con-
sistent with the experimental observation that H is slow-
2
ly released from 2-H to form 1 at room temperature. The
last step consists of the interaction of the five-coordinate
amido complex 1 with EtOH to regenerate the adduct 1-
EtOH throughout a N···H-O hydrogen bonding and via
-1
an endergonic process (5.8 kcal.mol ).
ACS Paragon Plus Environment

ACS Catalysis
Page 12 of 14
isolelectronic, 18 electron complexes, the position trans to
Corresponding Author
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
the hydride is occupied by a strongly π-accepting carbon-
yl in the case of manganese, and by a strongly σ-donating
hydride in the case of ruthenium and iron. This may be a
further lead for the design of more efficient manganese-
based catalytic systems.
* E-mail: regis.gauvin@ensc-lille.fr
ACKNOWLEDGMENT
This article is dedicated to Prof. David Milstein on the occa-
th
sion of his 70 birthday. This work was performed in part-
nership with the SAS PIVERT, within the frame of Institut
pour la Transition Energétique P.I.V.E.R.T. (www.institut-
pivert.com) selected as an Investment for the Future (“Inves-
tissements d’Avenir”). This work was supported, as part of
the Investments for the Future, by the French Government
under the reference ANR-001-01. CNRS, Chevreul Institute
(FR 2638), Ministère de l’Enseignement Supérieur et de la
Recherche, Région Nord – Pas de Calais and FEDER are
acknowledged for supporting and funding partially this work.
ASSOCIATED CONTENT
Supporting Information Available: Experimental meth-
ods, synthetic details, spectroscopic data, catalytic proce-
dures and DFT-calculated coordinates. This material is avail-
able free of charge via the Internet at http://pubs.acs.org.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
AUTHOR INFORMATION
REFERENCES
1) Synthesis of Carboxylic Acids, Esters and Their Derivatives (1991), John Wiley & Sons Ltd, Ogliaruso, M. A.; Wolfe,
J. F.
(2) a) Li, H.; Dong, K.; Jiao, H.; Neumann, H.; Jackstell, R.; Beller, M. Nat. Commun. 2016, 8, 1159-1166. b) Sargent, B.
T.; Alexanian, E. J. J. Am. Chem. Soc. 2016, 138, 7520-7523. b) Wu, L.; Liu, Q.; Fleischer, I.; Jackstell, R.; Beller, M. Nat.
Commun. 2014, 1-6. c) Jimenez Rodriguez, C.; Foster, D. F.; Eastham, G. R.; Cole-Hamilton, D. J. Chem. Commun.
2004, 1720–1721.
(3) a) Tana, T;. Guo, X-W.; Xiao, Q.; Huang, Y.; Sarina, S.; Christopher, P.; Jia, J.; Wu, H.; Zhu, H. Chem. Com-
mun.2016, 51, 11567--11570. b) Davis, S. E.; Ide M. S.; Davis, R. J. Green Chem. 2013, 15, 17. c) Wang, B.; Ran, W.; Sun,
W.; Wang, K. Ind. Eng. Chem. Res. 2012, 51, 3932–3938.
4) a) S. Werkmeister, J. Neumann, K. Junge, M. Beller, Chem. Eur. J. 2015, 21, 12226-12250. b) C. Gunanathan, D.
Milstein, Chem. Rev. 2014, 114, 12024-12087.
5) a) Cheng, J.; Zhu, M.; Wang, C.; Li, J.; Jiang, X.; Wei, Y.; Tang, W.; Xue, D.; Xiao, J. Chem. Sci., 2016, 7, 4428-4434.
b) Srimani, D.; Balaraman, E.; Gnanaprakasam, B.; Ben-David, Y.; Milstein, D. Adv. Synth. Catal. 2012, 354, 2403. c)
Gnanaprakasam, B.; Ben-David, Y.; Milstein, D. Adv. Synth. Catal. 2010, 352, 3169. d) Zhang, J.; Balaraman, E.; Leitus,
G.; Milstein, D. Organometallics, 2011, 30, 5716. e) Zhao, J.; Hartwig, J. F. Organometallics, 2005, 24, 2441. f)
Hunsicker, D. M.; Dauphinais, B. C.; Mc Ilrath, S. P.; Robertson, N. J.; Macromol. Rapid. Commun. 2012, 33, 232.
6) a) Vicent, C.; Gusev, D. G. ACS Catal. 2016, 6, 3301−3309. b) Gusev, D. G. ACS Catal. 2016, 6, 6967–6981.
7) a) Murahashi, S-I.; Ito, K.; Naota, T.; Maeda Y. Tetrahedron Lett. 1981, 22, 5327-5330. b) Murahashi, S-I.; Naota, T.;
Ito, Maeda Y.; Taki, H. J. Org. Chem. 1987, 52, 4319-4327. c) Shahane, S.; Fischmeister, C.; Bruneau, C. Catal. Sci.
Technol, 2012, 2, 1425–1428.
8) Zhang, J.; Leitus, G.; Ben-David, Y.; Milstein D. J. Am. Chem. Soc. 2005, 127, 10840-10841.
(
(
(
(
(
(
(
9) a) Kim, D.; Le, L.; Drance, M. J.; Jensen, K. H.; Bogdanovski, K.; Cervarich, T. N.; Barnard, M. G.; Pudalov, N. J.;
Knapp, S. M. M.; Chianese, A. R. Organometallics 2016, 35, 982−989. b) Spasyuk, D.; Vicent, C.; Gusev, D. G. J. Am.
Chem. Soc. 2015, 137, 3743−3746. c) Fogler, E.; Garg, J. A.; Hu, P.; Leitus, G.; Shimon, L. J. W.; Milstein, D. Chem. Eur.
J. 2014, 20, 15727-15731. d) Tseng, K-N; T.; Kampf, J. W.; Szymczak, N. K. Organometallics 2013, 32, 2046−2049. e)
Musa, S.; Fronton, S.; Vaccaro, L.; Gelman D. Organometallics 2013, 32, 3069−3073. f) Spasyuk, D.; Smith, S.; Gusev,
D. G. Angew. Chem. Int. Ed. 2012, 51, 2772-2775. g) Spasyuk, D.; Gusev, D. G. Organometallics 2012, 31, 5239-5242. h)
Nielsen,M.; Junge, H.; Kammer, A.; Beller, M. Angew. Chem. Int. Ed. 2012, 51, 5711 –571. i) Chen, C.; Zhang, Y.; Hong,
S. H. J. Org. Chem. 2011, 76, 10005−10010. k) del Pozo, C.; Iglesias,M.; Sanchez, F. Organometallics 2011, 30, 2180–
188. l) Bertoli, M.; Choualeb, A.; Lough, A. J.; Moore, B.; Spasyuk, D.; Gusev, D. G. Organometallics 2011, 30, 3479–
482. m) Nielsen, M.; Kammer, A.; Cozzula, D.; Junge, H; Gladiali, S.; Beller, M. Angew. Chem. Int. Ed. 2011, 50, 9593
9597.
10) a) Zhang, L.; Raffa, G.; Nguyen, D. H.; Swesi, Y.; Corbel-Demailly, L.; Capet, F.; Xavier Trivelli, X.; Desset, S.;
2
3
–
(
Paul, S.; Paul, J-P.; Fongarland, P.; Dumeignil, F.; Gauvin, R. M. J. Catal. 2016, 340, 331-343. b) Zhang, L.; Nguyen, D.
ACS Paragon Plus Environment

Page 13 of 14
ACS Catalysis
1
2
3
4
H.; Raffa, G.; Trivelli, X.; Capet, F.; Desset, S.; Paul, S.; Dumeignil, F.; Gauvin, R. M. ChemSusChem, 2016, 9, 1413–
423.
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
(
(
11) Chirik, P.; Morris R. Acc.Chem.Res. 2015, 48, 2495-2495.
12) a) Chakraborty, S.; Piszel, P. E.; Brennessel, W. W.; Jones, W. D. Organometallics 2015, 34, 5203−5206. b) Boni-
tatibus Jr., P. J.; Chakraborty, S.; Mark D. Doherty, M. D.; Siclovan, O.; Jones, W. D.; Soloveichik, G. L. PNAS 2015,
12, 1687–1692. c) Bielinski, E. A.; Förster, M.; Zhang, Y.; Bernskoetter, W. H.; Hazari, N.; Holthausen, M. C. ACS
1
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Catal. 2015, 5, 2404−2415. d) Chakraborty, S.; Lagaditis, P. O.; Förster, M.; Bielinski, E. A.; Hazari, N.; Holthausen, M.
C.; Jones, W. D.; Schneider, S. ACS Catal. 2014, 4, 3994−4003.
(
13) Mukherjee, A.; Nerush, A.; Leitus, G.; Shimon, L. J. W.; David, Y. B.; Jalapa, N. A. S.; Milstein D. J. Am. Chem.
Soc. 2016, 138, 4298–4301.
14) Mastalir, M.; Glatz, M.; Gorgas, N.; Stçger, B.; Pittenauer, E.; Allmaier, G.; Veiros, F. L.; Kirchner, K. Chem. Eur.
(
J. 2016, 22, 12316-12320.
(15) Tondreau, A. M.; Boncella, J. M. Organometallics, 2016, 35, 2049-2052.
(16) Elangovan, S.; Topf, C.; Fischer, S.; Jiao, H.; Spannenberg, A.; Baumann, W.; Ludwig, R.; Junge, K.; Beller, M. J.
Am. Chem. Soc. 2016, 138, 8809–8814.
(17) Kallmeier, F.; Irrgang, T.; Dietel, T.; Kempe, R. Angew. Chem. Int. Ed. 2016, 55, 1180duc6–11809
(18) a) Elangovan, S.; Neumann, J.; Sortais, J-B.; Kathrin Junge, K.; Darcel, C.; Beller, M. Nat. Chem. 2016, 7, 12641. b)
Elangovan, S.; Garbe, M.; Jiao H, Spannenberg, A.; Junge, K.; Beller, M. Angew. Chem. Int. Ed. 2016, 55, 15364–15368.
(
19) Kuriyama, W.; Matsumoto, T.; Ogata, O.; Ino, Y.; Aoki, K.; Tanaka, S.; Ishida, K.; Kobayashi, T.; Sayo, N.; Saito,
T. Org. Proc. Res. Dev. 2012, 16, 166-171.
20) Radosevich, A. T.; Melnick, J. G.; Stoian, S. A.; Bacciu, D.; Chen, C-H.; Foxman, B. M.; Ozerov, O. V.; Nocera, D.
G. Inorg. Chem. 2009, 48, 9214–9221.
21 a) Heuer, B.; Matthews, M. L.; Reid, G.; Ripley, M. J. Organomet. Chem. 2002, 655, 55-62. b) Rentsch, D.; Hany,
(
(
R.; von Philipsborn W. Magn. Reson. Chem. 1997, 35, 832-838. c) Rentsch, D.; Nill, L.; von Philipsborn, W.; Sidler, D.
R.; Rybczynski, P. J.; DeShong, P. Magn. Reson. Chem. 1998, 36, S54-S60. d) Benjamin S. L.; Levason W.; Reid G.
Organometallics, 2013, 32, 2760-2767. e) Levason W.; Orchard S. D.; Reid G. Organometallics 1999, 18, 1275-1280.
(
(
22) Marks, T. J.; Kolb, J. R. Chem. Rev., 1977, 77, 263-293.
23) Werkmeister, S.; Junge, K.; Bianca Wendt, Alberico, E.; Jiao, H.; Baumann, W.; Junge, H.; Gallou, F.; Beller, M.
Angew. Chem. Int. Ed. 2014, 53, 8722 –8726.
(
(
(
24) a) Yang, X. ACS Catal. 2013, 3, 2684−2688. b) Yang, X. ACS Catal. 2014, 4, 1129−1133.
25) We also observed the presence of traces amounts of L-iPr (see SI).
26) Alberico, E.; Lennox, A. J. J.; Vogt, L. K.; Jiao, H.; Baumann, W.; Drexler, H-J.; Nielsen, M.; Spannenberg, A.;
Checinski, M. P.; Junge, H.; Beller, M. J. Am. Chem. Soc. 2016, 138, 14890–14904.
27) a) H. Li, M. Wen, Z.-X. Wang, Inorg. Chem. 2012, 51, 5716-5727; b) S. Qu, Y. Dang, C. Song, M. Wen, K.-W.
Huang, Z.-X. Wang, J. Am. Chem. Soc. 2014, 136, 4974-4991.
(
TOC Graphic
13
ACS Paragon Plus Environment

ACS Catalysis
Page 14 of 14
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
14
ACS Paragon Plus Environment