Manganese-Catalyzed Upgrading of Ethanol into 1-Butanol

Article abstract of DOI:10.1021/jacs.7b05939

Biomass-derived ethanol is an important renewable feedstock. Its conversion into high-quality biofuels is a promising route to replace fossil resources. Herein, an efficient manganese-catalyzed Guerbet-type condensation reaction of ethanol to form 1-butanol was explored. This is the first example of upgrading ethanol into higher alcohols using a homogeneous non-noble-metal catalyst. This process proceeded selectively in the presence of a well-defined manganese pincer complex at the parts per million (ppm) level. The developed reaction represents a sustainable synthesis of 1-butanol with excellent turnover number (>110 000) and turnover frequency (>3000 h^-1). Moreover, mechanistic studies including control experiments, NMR spectroscopy, and X-ray crystallography identified the essential role of the "N-H moiety" of the manganese catalysts and the major reaction intermediates related to the catalytic cycle.

Full text of DOI:10.1021/jacs.7b05939

Article
pubs.acs.org/JACS
Manganese-Catalyzed Upgrading of Ethanol into 1‑Butanol
Shaomin Fu,‡ Zhihui Shao,‡ Yujie Wang, and Qiang Liu*
Center of Basic Molecular Science (CBMS), Department of Chemistry, Tsinghua University, Beijing 100084, China
S
* Supporting Information
ABSTRACT: Biomass-derived ethanol is an important renew-
able feedstock. Its conversion into high-quality biofuels is a
promising route to replace fossil resources. Herein, an efficient
manganese-catalyzed Guerbet-type condensation reaction of
ethanol to form 1-butanol was explored. This is the first
example of upgrading ethanol into higher alcohols using a
homogeneous non-noble-metal catalyst. This process proceeded
selectively in the presence of a well-defined manganese pincer
complex at the parts per million (ppm) level. The developed
reaction represents a sustainable synthesis of 1-butanol with
excellent turnover number (>110 000) and turnover frequency
(>3000 h−1). Moreover, mechanistic studies including control
experiments, NMR spectroscopy, and X-ray crystallography
identified the essential role of the “N−H moiety” of the manganese catalysts and the major reaction intermediates related to the
catalytic cycle.
Scheme 1. Guerbet Reaction Process
INTRODUCTION
■
Biofuels produced from renewable biomass are considered to
be greener alternatives to conventional gasoline.1 Currently,
(bio)ethanol produced from fermentation of crops is the largest
liquid biofuel contributor.2 However, ethanol has several
disadvantages, including water solubility, corrosivity, and low
energy density, which limit its use as a blend additive to
gasoline in high ratio in unmodified gasoline engines. These
problems associated with ethanol can be alleviated by using
butanol (or other longer-chain alcohols), because butanol is
immiscible with water, noncorrosive, and more energy dense
than ethanol.3 Although butanol is a desirable advanced biofuel,
large-scale synthesis of butanol from biofeedstocks remains
challenging. The best reported method to date, the so-called
ABE fermentation process,4 generates a mixture of acetone,
butanol, and ethanol in low conversion.5
Upgrading ethanol into butanol could be a promising
alternative strategy to produce an improved higher alcohol
biofuel from renewable biomass. In this respect, the Guerbet
reaction is an ideal choice to generate a longer-chain alcohol
from ethanol via alcohol condensation with water as the only
byproduct.6 Based on a general “hydrogen-borrowing”
mechanism,7 this reaction involves three steps: dehydrogen-
ation of an alcohol into aldehyde, aldol condensation, and
hydrogenation of the unsaturated aldehyde to a higher alcohol
product (Scheme 1). Unfortunately, ethanol has proven to be a
challenging substrate for this reaction because of the
thermodynamically unfavored dehydrogenation and poorly
selective aldol condensation of highly reactive acetaldehyde.
Therefore, only a handful of catalysts that realize the catalytic
Guerbet reaction of ethanol have been developed. Heteroge-
neous catalytic systems generally require harsh reaction
conditions and/or suffer from low selectivity.8 In contrast,
well-defined homogeneous catalysts based on precious metals
(Ru9 and Ir10) show higher reactivity and selectivity for this
transformation under milder conditions (Figure 1).
Substitution of precious metals with economical and
ecologically friendly earth-abundant metal catalysts is highly
desirable in terms of sustainability. However, the use of
homogeneous non-noble metal catalysts to accomplish the
condensation of ethanol into higher alcohols has not been
reported to the best of our knowledge. Since last year, a series
of novel manganese pincer catalysts has been successfully
developed for both hydrogenation11 and dehydrogenation12
reactions. Encouraged by these significant achievements, herein
we report the first Mn-catalyzed Guerbet reaction of ethanol.
Notably, this robust manganese pincer catalyst achieves the
highest turnover number (TON up to 114 120) and turnover
frequency (TOF 3078 h−1) among all Guerbet-type processing
of ethanol with 92% selectivity for 1-butanol production.
Received: June 8, 2017
© XXXX American Chemical Society
A
DOI: 10.1021/jacs.7b05939
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
However, it was difficult to isolate a pure manganese carbonyl
complex from a mixture of di- and tricarbonyl Cy-substituted
complexes [Mn]-2. Fortunately, the dicarbonyl manganese
complex [Mn]-2a formed selectively when THF was used as
the solvent at 70 °C. Use of a bulkier tBu-substituted PNP
ligand resulted in the five-coordinate dicarbonyl manganese
complex [Mn]-3 with square-pyramidal geometry. In contrast,
six-coordinate facial complexes [Mn]-4 and [Mn]-5 were
generated using less sterically hindered NNP pincer ligands. A
complex of MnCl2 with the iPrPNP ligand [Mn]-6 was also
prepared.
The X-ray structures of [Mn]-1b, [Mn]-3, and [Mn]-4 were
obtained (Figure 2), which unambiguously confirmed the
Figure 2. Molecular structures of Mn complexes 1b, 3, and 4 (thermal
ellipsoids set at 50% probability; hydrogen atoms are omitted for
clarity).
Figure 1. Homogeneous catalysts that upgrade ethanol to 1-butanol.
structures of these manganese complexes. These results indicate
that the tricarbonyl manganese complexes form more readily in
the presence of less bulky ligands. In contrast, sterically
hindered ligands tend to generate dicarbonyl manganese
complexes.
RESULTS AND DISCUSSION
■
Catalysts Development. In line with our interest in the
development of pincer-type earth-abundant metal catalysts,13
this investigation commenced with preparing the well-defined
PNP and NNP pincer Mn catalysts. We synthesized these Mn
complexes by reacting Mn(CO)5Br with the corresponding
pincer ligand in toluene at 100 °C, which is similar to the
reported procedure (Scheme 2).11a Interestingly, usage of
The catalytic activities of these manganese pincer complexes
were then investigated in the Guerbet reaction of ethanol
(Table 1). Initially, we reacted 0.01 mol% Mn catalyst, 6 mol%
EtONa, and 100 mmol EtOH at 160 °C for 24 h. Apart from
the higher alcohols generated as the liquid fraction (including
C4 and C6 aliphatic alcohols and C10 and C12 araliphatic
alcohols15), NaOAc was also formed in these reactions as a
byproduct (see the Supporting Information for details). NaOAc
is probably produced by the reaction of ethanol with water and
base, liberating H2.16 Besides, it could also undergo Cannizzaro
or Tishchenko-type reaction pathways.9b Notably, other
possible dehydrogenative condensation products of ethanol,
either ethyl acetate or ethyl acetal,17 were not detected in any of
these reactions. It is worth mentioning that the TON,
selectivity, yield, and conversion of these Mn-catalyzed ethanol
condensation reactions were calculated in the same manner as
in previous reports (shown in Figure 1) to allow direct
comparison. Specifically, [Mn]-1 gave 9.7% conversion (TON
of 972) of ethanol, resulting in 6.8% yield of 1-butanol with
83% selectivity (entry 1). The di- and tricarbonyl Mn(I)
complexes [Mn]-1a and [Mn]-1b as well as CyPNP-Mn(I)
complexes [Mn]-2 and [Mn]-2a gave similar results (entries 2,
3, 6, and 7). The comparable catalytic activities of di- and
tricarbonyl Mn(I) complexes can be attributed to the
conversion of the tricarbonyl complex to dicarbonyl species
at high temperature. There was no reaction using Mn(I)
complexes supported by the bulkier tBuPNP ligand or less
sterically hindered NNP ligands ([Mn]-3 and [Mn]-4),
whereas [Mn]-5 displayed low reactivity for this transformation
(entries 8−10). In addition, both the Mn(II) catalyst [Mn]-6
and commercially available Mn(CO)5Br gave no higher-alcohol
products either, revealing the importance of the supporting
Scheme 2. Synthesis of Pincer Manganese Catalysts
different ligands led to the formation of various coordination
structures. A mixture of di- and tricarbonyl manganese
complexes ([Mn]-1 or [Mn]-2) was produced using
(R2PC2H4)2NH (R = iPr or Cy) as the ligands.14 [Mn]-1a
and [Mn]-1b could be successfully separated because of their
distinct solubilities in toluene. Additionally, the meridional
complex [Mn]-1b possessed an octahedral geometry, which
could be completely converted to the dicarbonyl complex
[Mn]-1a by heating at higher temperature (120 °C) in toluene.
B
DOI: 10.1021/jacs.7b05939
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
a
Table 1. Manganese-Catalyzed Guerbet Reaction of
Ethanol
Table 2. Optimization of the Reaction Conditions
a
conv, %
(yield, %)
entry base (mol%) time, h
selectivity, %
TON
selectivity,
b
c
d
entry
[Mn]
[Mn]-1
T, °C conv, % (yield, %)
%
TON
1
2
3
4
5
6
7
8
9
EtONa (6)
tBuONa (6)
NaOH (6)
KOH (6)
24
24
24
24
24
24
48
96
48
96
96
96
24
48
72
12
72
168
168
9.7 (6.8)
8.5 (6.1)
8.1 (5.8)
9.7 (6.4)
1.7 (1.7)
7.6 (4.8)
12.1 (8.0)
12.7 (8.4)
14.7 (8.2)
19.5 (14.1)
20.1 (14.5)
28.6 (22.7)
5.4 (4.5)
8.9 (6.5)
6.9 (5.9)
3.6 (2.7)
7.1 (6.1)
8.9 (7.7)
11.2 (9.8)
83
84
84
81
100
79
81
80
73
83
82
87
90
84
90
972
848
1
160
160
160
160
160
160
160
160
160
160
160
160
160
140
180
9.7 (6.8)
9.7 (6.7)
10.0 (6.8)
12.6 (8.0)
12.2 (7.8)
9.7 (6.9)
9.1 (6.3)
0 (0)
83
83
82
79
79
83
82
0
972
966
1002
1259
1224
969
909
0
810
2
[Mn]-1a
[Mn]-1b
[Mn]-1c
[Mn]-1d
[Mn]-2
[Mn]-2a
[Mn]-3
[Mn]-4
[Mn]-5
[Mn]-6
Mn(CO)5Br
none
965
3
NaOAc (6)
EtONa (6)
EtONa (6)
EtONa (6)
EtONa (6)
EtONa (12)
EtONa (12)
EtONa (12)
EtONa (6)
EtONa (6)
EtONa (6)
EtONa (6)
EtONa (6)
EtONa (6)
EtONa (12)
170
4
b
c
763
5
1207
1274
1473
1950
1031
286
6
7
8
10
11
12
13
14
15
16
17
18
19
9
0 (0)
0
0
d
e
f
e
10
11
12
13
14
15
2.2 (1.3)
0 (0)
66
0
221
0
1073
18160
42240
37040
72616
90968
114120
0 (0)
0
0
g
h
i
0 (0)
0
0
[Mn]-1
[Mn]-1
2.0 (1.6)
12.6 (8.3)
86
80
203
1255
j
79
i
a
90
91
92
Reaction conditions: EtOH (6 mL, 100 mmol), [Mn] (0.01 mmol),
i
and base (6 mmol) were reacted under Ar in a 25 mL autoclave for 24
k
b
h. Total conversion of ethanol to the higher alcohol products, with
the yield of n-butanol in parentheses (determined by GC). Total
c
a
Reaction conditions: In a 25 mL autoclave, EtOH (6 mL, 100 mmol),
selectivity for n-butanol was determined by mmol of nBuOH per
mmol of all the higher alcohol products. Turnover number (TON)
0.01 mol% [Mn]-1 (0.01 mmol), and base (6−12 mmol) were reacted
d
under Ar at 160 °C for 12−168 h. Conversion, yield, selectivity, and
b
based on the amount of EtOH (in mmol) converted into higher
TON were determined in the same way as in Table 1. H2O (10
e
c
d
e
alcohol products per mmol of Mn. Crotyl alcohol was also produced
mmol) was added. 180 °C. 0.02 mol% [Mn]-1 (0.02 mmol). 0.1
f
with 21% selectivity.
mol% [Mn]-1 (0.1 mmol). EtOH (12 mL, 200 mmol), EtONa (12
mmol). In a 60 mL autoclave, using EtOH (30 mL, 510 mmol),
[Mn]-1 (0.0025 mmol), and EtONa (30 mmol). In a 60 mL
g
h
carbonyl and pincer ligands (entries 11 and 12). There was no
reaction in the absence of Mn catalyst (entry 13). The effect of
reaction temperature was also investigated (entries 14 and 15).
A lower temperature (140 °C) gave decreased conversion and
yield, while higher TON and conversion were obtained at
increased temperature (180 °C).
autoclave, using EtOH (45 mL, 765 mmol), [Mn]-1 (0.00125 mmol),
and EtONa (46 mmol). In a 100 mL autoclave, using EtOH (75 mL,
i
1275 mmol), [Mn]-1 (0.00125 mmol), and EtONa (76 mmol).
j
k
Crotyl alcohol was also produced with 11% selectivity. In a 100 mL
autoclave, using EtOH (75 mL, 1275 mmol), [Mn]-1 (0.00125
mmol), and EtONa (150 mmol).
Reaction Conditions Optimization. We then examined
the performance of the most effective and easily prepared
catalyst [Mn]-1 in detail. Different reaction parameters were
screened to improve the efficiency of this transformation to a
more practical level (Table 2). First, the influence of base was
investigated. Strong bases, such as EtONa, NaOH, and KOH,
all afforded good TON and selectivity for 1-BuOH production
(entries 1−4). Such bases could be gradually converted to
acetate salts during the reaction process as stated earlier.
Moreover, the weak base NaOAc only gave a TON of 170,
albeit with excellent selectivity (entry 5). These results are in
agreement with the decrease of reaction rate during the
reaction process using a strong base (entries 1, 7, 8). In terms
of using alcohols as a fuel additive, C4+ longer-chain alcohols
are even more desirable than 1-butanol because of their higher
energy density. As a result, the TON of the catalyst would be
more important than the C4 selectivity for the upgrade of
(bio)ethanol to biofuels. An increase of the base loading of
EtONa from 6 to 12 mol% led to higher conversion (entries 1,
10−12). In addition, water can be generated with higher
alcohols in the catalytic Guerbet reaction. Therefore, we tested
the influence of water on this reaction process. The reaction
was less efficient when 10 mmol of water was added to the
reaction system (entry 6). Interestingly, an increase of the
selectivity for 1-BuOH to 90% was observed when 12 mL of
ethanol was used instead of 6 mL under reaction conditions
that were otherwise the same (entry 13). This is likely a result
of the higher pressure of the H2 generated in situ because of the
smaller headspace in the autoclave at higher ethanol volume.
Indeed, H2 gas was detected by GC in the headspace of the
pressure vessel after reaction.18 A higher H2 pressure would
accelerate the hydrogenation of the reaction intermediate
crotonaldehyde, thus preventing its condensation to C4+
higher oligomers under basic conditions. The same trend of
selectivity was also observed for the reactions performed in a 60
mL autoclave (entries 14 and 15). Notably, a further increase of
ethanol amount and decrease of catalyst loading gave a
dramatic improvement of the TON of the Mn catalyst
(Table 2, entries 16−19). Remarkably, decreasing the load of
[Mn]-1 to 0.0001 mol% (8 ppm) resulted in a record TON of
114120 with 92% C4 selectivity after 7 days (Table 2, entry
19). The TOF of Mn catalyst could reach up to 3078 h−1 for
the first 12 h (Table 2, entry 16).
Mechanistic Studies. To exclude the possibility that small
quantities of Pt-group metals in the reagents could promote
this transformation, ICP-MS analysis of the catalyst [Mn]-1
and NaOEt was performed. The amounts of trace precious
metal contaminants were all below 1 ppm (Table S8).
Therefore, it was not possible for these contaminants to be
C
DOI: 10.1021/jacs.7b05939
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
effective catalysts for this reaction because of their low
concentration. Furthermore, no inhibition effect was found
when 73 equiv of mercury or 0.5−1 equiv of PMe3 with respect
to Mn catalyst was added to the reaction system (Table S3).
This suggests that the Mn catalyst is likely to be homogeneous
under these reaction conditions.
To investigate the possible involvement of the N−H moiety
on the PNP pincer catalysts in this reaction, the N-methyl-
substituted manganese complexes [Mn]-7 and [Mn]-8 were
synthesized, and their catalytic activity was measured (Scheme
3). The crystal structure of dicarbonyl complex [Mn]-7 is
aldol reaction of crotonaldehyde 2 (Scheme 4, eq 2). [Mn]-1
could not inhibit the conversion of 2 into higher oligomers
either. These results are different from the results for a
ruthenium catalyst reported previously, which could prevent
the aldol reaction of C4 aldehydes, thus leading to high
selectivity for 1-BuOH production.9a As a result, we speculate
that efficient hydrogenation of the reaction intermediate
crotonaldehyde 2 could be the origin of the high selectivity
of this Mn-catalyzed Guerbet reaction of ethanol.
Consequently, the hydrogenation of acetaldehyde 1 and
crotonaldehyde 2 under 10 bar H2 pressure in THF was carried
out using [Mn]-1 and the corresponding N-methylated
complex [Mn]-7 as the catalysts. C4 alcohols (1-BuOH and
crotyl alcohol) were produced with 64% selectivity along with
other higher alcohols in [Mn]-1-catalyzed hydrogenation of 1
(Scheme 5, eq 1). This result demonstrates that acetaldehyde 1
Scheme 3. Catalytic Upgrading of Ethanol Using N-Me-
Substituted Manganese Catalysts
Scheme 5. Mn-Catalyzed Hydrogenation of Acetaldehyde 1
and Crotonaldehyde 2
shown in Scheme 3, and has a meridional configuration with
octahedral geometry. Much lower reactivity and selectivity were
observed for the upgrade of ethanol to 1-butanol using [Mn]-7
and [Mn]-8 as the catalysts compared with the results using
[Mn]-1a and [Mn]-2a under the same reaction conditions
(Table 1, entries 2 and 7). This indicates the strong beneficial
influence of the cooperative N−H functionality of Mn catalysts
on this transformation.19
To further understand the origin of the high selectivity, the
base-catalyzed aldol condensation step that governs the
selectivity of the Guerbet reaction was studied (Scheme 4).
and H2 can be the reaction intermediates for this Mn-catalyzed
ethanol upgrading reaction. Ethanol was also generated in this
reaction, which indicates that the Mn-catalyzed dehydrogen-
ation of ethanol is a reversible process. The [Mn]-1-catalyzed
hydrogenation of crotonaldehyde 2 proceeded smoothly as well
and gave 72% selectivity for C4 alcohols (Scheme 5, eq 2).
Conversely, [Mn]-7 gave much lower C4 selectivity for the
hydrogenation of both aldehydes 1 and 2. This implies that the
metal−ligand cooperative fashion of the pincer Mn catalysts
bearing N−H moieties is essential for the hydrogenation of the
key reaction intermediate crotonaldehyde 2.
Scheme 4. Aldol Condensation Reactions of Acetaldehyde 1
and Crotonaldehyde 2
To gain more mechanistic insights into this reaction, we
investigated the reactivity of the catalyst [Mn]-1a with EtONa
and EtOH via stoichiometric reactions (Scheme 6). The
complex [Mn]-1a could be converted into Mn(I) PNP amido
complex [Mn]-1c in the presence of 3 equiv of EtONa with
69% isolated yield. The reaction of [Mn]-1c with 8 equiv of
EtOH led to a rapid formation of ethoxide complex [Mn]-1d
with full conversion detected by 31P NMR spectroscopy
(Figure S17). Interestingly, [Mn]-1c could be regenerated
from ethoxide [Mn]-1d by removing the excess ethanol under
vacuum or Ar flow. In addition, the catalyst precursor [Mn]-1a
could also be directly converted into [Mn]-1d in an ethanolic
solution of EtONa, which is closely related to the catalytic
reaction process (Figure S19). To our delight, complex [Mn]-
1d was crystallized from hexane at −20 °C, affording single
crystals suitable for X-ray diffraction analysis (Figure 3). In the
We first investigated the EtONa-catalyzed aldol reaction of
acetaldehyde 1 and compared the results with those of the
reaction in the presence of the catalyst [Mn]-1. It was found
that EtONa-catalyzed condensation of 1 only produced 5%
crotonaldehyde 2 after 4 h at 160 °C (Scheme 4, eq 1). The
rest of the products were C4+ higher oligomers. Addition of the
Mn catalyst into the aldol reaction of 1 did not afford higher
selectivity. A similar trend of reactivity was observed for the
D
DOI: 10.1021/jacs.7b05939
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Scheme 6. Stoichiometric Reactions of [Mn]-1a with EtONa
and EtOH
Scheme 7. Reactivity of [Mn]-1d with Different Bases in
EtOH
two Mn−H complexes were consistent with those reported.12h
Meanwhile, hydroxide and acetate manganese complexes
[Mn]-1f and [Mn]-1g were also produced under these reaction
conditions. 13C NMR analysis of this reaction mixture revealed
the formation of NaOAc (Figure S23). Using NaOH as the
base instead of NaOEt led to similar results (Figure S25). In
contrast, when [Mn]-1d was reacted with the weak base
NaOAc, which can be generated in the catalytic reactions,
acetate complex [Mn]-1g was afforded in full conversion at
room temperature (Figure S31). There is no reaction after
heating [Mn]-1g in ethanol at 110 °C for 10 h as indicated by
31P NMR spectroscopy (Figure S32). This complex could be
also prepared independently from the reaction of amido
complex [Mn]-1c and HOAc at room temperature (Scheme 8),
which was fully characterized by 31P, 1H, and 13C NMR
spectroscopies. Crystallization of [Mn]-1g from hexane at −20
°C afforded single crystals. The X-ray structure revealed [Mn]-
1g possessed octahedral geometry with a hydrogen bond
between the N−H moiety and carbonyl group of the acetate
anion (Figure 3). In addition, [Mn]-1g is more stable than
[Mn]-1d, which could not be converted to amido complex
[Mn]-1c under vacuum. Notably, heating [Mn]-1d in toluene
without base at 110 °C did not lead to the formation of Mn−H
complexes [Mn]-1e or [Mn]-1e′ (Figure S27). These results
indicate that manganese hydride species could be generated by
the C−H cleavage of ethoxide complex [Mn]-1d with the
assistance of a strong base, such as NaOEt and NaOH.
Additionally, amido complex [Mn]-1c was not observed in any
of these reactions, which excludes the possibility of the
reversible formation of [Mn]-1c from ethoxide species [Mn]-
1d in ethanolic solution. As a result, it is clear that the reactivity
of the ethoxide complex [Mn]-1d is quite different from that of
the benzyloxide manganese complex reported by Gauvin and
co-workers.12h The latter complex afforded manganese hydrides
at 80 °C in the absence of base and was in equilibrium with
amido complex [Mn]-1c.
Figure 3. Single-crystal structures of [Mn]-1c, [Mn]-1d, [Mn]-1f, and
[Mn]-1g (thermal ellipsoids set at 50% probability; most hydrogen
atoms are omitted for clarity).
solid state, this complex was stabilized by N···H and O···H
hydrogen bonds with one molecule of co-crystallized ethanol.
This crystal structure indicates that ethanol could function as a
proton shuttle for the deprotonation of the N−H moiety on
the ligand by the ethoxy group on the Mn center. Moreover,
the X-ray structure of [Mn]-1c was obtained as well (Figure 3).
Both Mn complexes [Mn]-1c and [Mn]-1d were catalytically
active in the upgrading of ethanol (Table 1, entries 4 and 5),
which means they could be reaction intermediates in this
transformation.
The next major step of the catalytic cycle consists of the
formation of manganese hydride species from ethoxide complex
[Mn]-1d and release of acetaldehyde 1 (Scheme 7). Under
reaction conditions similar to the catalytic process, the cis- and
trans-isomers of Mn−H complexs [Mn]-1e and [Mn]-1e′ were
generated upon heating a solution of [Mn]-1d and EtONa in
EtOH at 110 °C for 8 h in a closed system (Figures S20−S24).
The observed 31P{1H} and featured 1H chemical shifts for these
The interconversion of manganese complexes [Mn]-1c,
[Mn]-1d, [Mn]-1e, [Mn]-1e′, [Mn]-1f, and [Mn]-1g, which
are closely related to the catalytic process, was studied (Scheme
8). Treatment of [Mn]-1d with NaOH (10 equiv) in ethanol at
room temperature resulted in the formation of a new complex
(38% conversion) that exhibited a broad 31P{1H}NMR signal at
E
DOI: 10.1021/jacs.7b05939
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
reaction conditions because of the formation of water, NaOH,
and NaOAc. [Mn]-1f could be easily converted back to [Mn]-
1d in ethanol, these species are therefore in equilibrium.
However, another off-cycle species [Mn]-1g could only
regenerate [Mn]-1c and [Mn]-1d in the presence of strong
base (NaOEt or NaOH), both of which are transformed
gradually into NaOAc during the catalytic reaction. Therefore,
the accumulation of the off-cycle species [Mn]-1g would be
expected as the amount of NaOAc is increased, which could be
one reason for the decreasing reactivity during the reaction
process.
Scheme 8. Interconversion of Manganese Complexes
Furthermore, we studied the reactivity of the above-
mentioned four manganese complexes with H2. Specifically,
amido complex [Mn]-1c could react with H2 (10 bar) at room
temperature to generate cis-hydride complex [Mn]-1e in full
conversion (Figure S45). Heating a toluene solution of [Mn]-
1e at 30 °C for 4 h leaded to the formation of a mixture of
[Mn]-1e, [Mn]-1e′, and [Mn]-1c in a ratio of 1:0.21:0.36
(Figure S46). Ethoxide complex [Mn]-1d could be also
hydrogenated at room temperature with 10 bar H2 to give
trans-hydride complex [Mn]-1e′ as the major product ([Mn]-
1e′: [Mn]-1e = 11:1) as shown in Figure S48. In contrast, the
hydrogenation of hydroxide complex [Mn]-1f under the same
condition mainly produce cis-hydride species [Mn]-1e in 17:1
cis/trans selectivity (Figure S51). Moreover, [Mn]-1e were
stable in the presence of EtOH or water, which could not be
transformed to [Mn]-1d or [Mn]-1f under these reaction
conditions (Figures S49 and 52). Interestingly, acetate complex
[Mn]-1g could not react with H2 even at more harsh
conditions (50 bar H2 and 100 °C) to afford hydride complex
[Mn]-1e or [Mn]-1e′ (Figure S53). It again indicates the low
reactivity of [Mn]-1g under catalytic reaction conditions.
Additionally, [Mn]-1g could be generated from [Mn]-1e or
[Mn]-1e′ in 100% conversion by reacting with acetic acid at
room temperature (Figures S54 and S55). At last, it is worth
mentioning that [Mn]-1e′ could not convert to [Mn]-1e upon
heating in toluene at 60 °C (Figure S58). However, [Mn]-1e′
could react with crotonaldehyde 2 to produce a mixture of
[Mn]-1d, [Mn]-1f, and [Mn]-1g at room temperature in full
conversion after 3 h (Figure S60). This reaction mixture could
be hydrogenated under 50 bar H2 at 65 °C to generate [Mn]-
1e and [Mn]-1e′ in 1:2.5 cis/trans selectivity (Figure S61).
These results indicated that [Mn]-1e′ could also be trans-
formed to [Mn]-1e under catalytic reaction conditions.
Based on the experimental findings, a catalytic cycle for the
Guerbet reaction of ethanol was proposed (Scheme 9). Initially,
the catalytically active species [Mn]-1d was generated from
precatalyst [Mn]-1a in the presence of EtONa in ethanol.
Complex [Mn]-1d further transformed into manganese hydride
species [Mn]-1e with the assistance of strong base (NaOH or
NaOEt) and released acetaldehyde. We proposed that this step
involved “inner-sphere” C−H bond cleavage via C−H
coordination of ethoxide to the Mn metal center. A similar
reaction process for the corresponding methoxide Ru-PNP
complex was demonstrated by Beller and co-workers very
recently.20 Under basic conditions, the aldol condensation of
acetaldehyde led to the formation of the important reaction
intermediate crotonaldehyde. It was then reduced by hydride
complex [Mn]-1e to afford the final product 1-BuOH along
with the generation of amido complex [Mn]-1c. The
alcoholysis of the latter complex regenerated ethoxide complex
[Mn]-1d and completed the catalytic cycle. Meanwhile, [Mn]-
1d was in equilibrium with hydroxide complex [Mn]-1f under
86.6 ppm (Figure S29). The same new species was also
obtained with 55% conversion for the reaction of [Mn]-1d with
H2O (25 equiv) (Figure S30). We assigned this new complex as
hydroxide manganese complex [Mn]-1f, which could also be
generated in full conversion from the reaction of amido
complex [Mn]-1c and H2O (Figure S33). [Mn]-1f turned out
to be unstable, and removal of the solvent under vacuum
regenerated [Mn]-1c (Figure S35). We attempted to crystallize
complex [Mn]-1f from hexane in the presence of a small
amount of degassed water at −20 °C for several times.
Surprisingly, a CO2 adduct of [Mn]-1f was always crystallized.
It was assumed that this adduct formed because of the strong
nucleophilicity of this hydroxide manganese complex toward
trace amount of CO2 in the solvent. The X-ray crystal structure
of this ion-pair complex 2[Mn]-1f·CO2 is shown in Figure 3. In
addition, when [Mn]-1f was treated with ethanol at room
temperature, it was converted to ethoxide complex [Mn]-1d
with 28% conversion after 0.5 h (Figure S36). [Mn]-1f could
be also converted to [Mn]-1g by reaction with NaOAc at room
temperature (Figure S38). Moreover, the reaction of acetate
complex [Mn]-1g and NaOEt in ethanol resulted in the
formation of [Mn]-1d with over 90% conversion after 30 min
(Figure S42). In contrast, there is no reaction between [Mn]-
1g and ethanol even at 110 °C (Figure S41). [Mn]-1g reacted
with NaOH (10 equiv) at 70 °C in THF to give [Mn]-1c with
complete conversion (Figure S43). In fact, [Mn]-1g could also
be directly transformed into [Mn]-1d and [Mn]-1f in 3.6:1
ratio by reaction with NaOH (10 equiv) in ethanol at room
temperature (Figure S44). It is worth pointing out that the off-
cycle species [Mn]-1f and 1g could be generated from the
catalytically active species [Mn]-1d and 1c under the catalytic
F
DOI: 10.1021/jacs.7b05939
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
ORCID
Scheme 9. Proposed Catalytic Cycle for the Guerbet
Reaction of Ethanol
Qiang Liu: 0000-0002-8402-029X
Author Contributions
‡S.F. and Z.S. contributed equally to this work.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We are grateful for the financial support of the National
Program for Thousand Young Talents of China and 111
project. We also thank Mr. Zi-Ang Nan (iChEM) and Dr. Xin
He (Tsinghua University) for their kind help with the single-
crystal X-ray diffraction analysis. Prof. Peng Kang and Miss
Fang-Wei Liu from TIPCCAS are also appreciated for their
assistance with the gas phase GC analysis.
REFERENCES
■
(1) (a) Ragauskas, A. J.; Williams, C. K.; Davison, B. H.; Britovsek,
G.; Cairney, J.; Eckert, C. A.; Frederick, W. J.; Hallett, J. P.; Leak, D. J.;
Liotta, C. L.; Mielenz, J. R.; Murphy, R.; Templer, R.; Tschaplinski, T.
Science 2006, 311, 484. (b) Sreekumar, S.; Balakrishnan, M.; Goulas,
K.; Gunbas, G.; Gokhale, A. A.; Louie, L.; Grippo, A.; Scown, C. D.;
Bell, A. T.; Toste, F. D. ChemSusChem 2015, 8, 2609. (c) Wu, L.;
Moteki, T.; Gokhale, A. A.; Flaherty, D. W.; Toste, F. D. Chem 2016,
1, 32.
(2) Biello, D. Can Ethanol from Corn Be Made Sustainable?. Scientific
American, Feb 20, 2013; http://www.scientificamerican.com/article/
can-corn-ethanol-be-made-sustainable/.
(3) (a) Durre, P. Biotechnol. J. 2007, 2, 1525. (b) Harvey, B. G.;
̈
the catalytic reaction conditions. In addition, [Mn]-1d could
convert to another off-cycle species acetate complex [Mn]-1g
by reacting with NaOAc generated during the reaction process.
However, [Mn]-1g could only regenerate the catalytically active
species [Mn]-1c and [Mn]-1d with the assistance of strong
base (NaOH or NaOEt), which are both consumed during the
reaction process.21
Meylemans, H. A. J. Chem. Technol. Biotechnol. 2011, 86, 2.
(4) Anbarasan, P.; Baer, Z. C.; Sreekumar, S.; Gross, E.; Binder, J. B.;
Blanch, H. W.; Clark, D. S.; Toste, F. D. Nature 2012, 491, 235.
(5) Jin, C.; Yao, M.; Liu, H.; Lee, C.-f. F.; Ji, J. Renewable Sustainable
Energy Rev. 2011, 15, 4080.
(6) (a) Guerbet, M. C. R. Acad. Sci. Paris 1899, 128, 1002.
(b) Guerbet, M. C. R. Acad. Sci. Paris 1909, 149, 129. (c) Veibel, S.;
Nielsen, J. I. Tetrahedron 1967, 23, 1723. (d) O’Lenick, A. J., Jr. J.
Surfactants Deterg. 2001, 4, 311. (e) Goulas, K. A.; Sreekumar, S.;
Song, Y.; Kharidehal, P.; Gunbas, G.; Dietrich, P. J.; Johnson, G. R.;
Wang, Y. C.; Grippo, A. M.; Grabow, L. C.; Gokhale, A. A.; Toste, F.
D. J. Am. Chem. Soc. 2016, 138, 6805.
(7) (a) Obora, Y. ACS Catal. 2014, 4, 3972. (b) Yang, Q.; Wang, Q.;
Yu, Z. Chem. Soc. Rev. 2015, 44, 2305. (c) Freitag, F.; Irrgang, T.;
Kempe, R. Chem. - Eur. J. 2017, DOI: 10.1002/chem.201701211.
(8) (a) Tsuchida, T.; Kubo, J.; Yoshioka, T.; Sakuma, S.; Takeguchi,
T.; Ueda, W. J. Catal. 2008, 259, 183. (b) Riittonen, T.; Toukoniitty,
E.; Madnani, D. K.; Leino, A.-R.; Kordas, K.; Szabo, M.; Sapi, A.; Arve,
CONCLUSIONS
■
In conclusion, we developed a Mn-catalyzed process to upgrade
ethanol into 1-BuOH with high selectivity. It demonstrated an
active catalytic system for ethanol upgrading via the Guerbet
reaction, reaching the highest TON (114 120) and TOF (3078
h−1) achieved to date. Moreover, a mechanistic study was also
implemented to identify the essential role of the “N−H moiety”
on the manganese catalyst and several major reaction
intermediates related to the catalytic cycle. The observed
remarkable catalyst performance represents a considerable step
toward a sustainable and efficient synthesis of biofuels with high
energy density from bio-derived feedstocks.
K.; Warna, J.; Mikkola, J.-P. Catalysts 2012, 2, 68. (c) Ogo, S.; Onda,
̈
̊
A.; Iwasa, Y.; Hara, K.; Fukuoka, A.; Yanagisawa, K. J. Catal. 2012, 296,
24. (d) Xu, G.; Lammens, T.; Liu, Q.; Wang, X.; Dong, L.; Caiazzo, A.;
Ashraf, N.; Guan, J.; Mu, X. Green Chem. 2014, 16, 3971. (e) Galadima,
A.; Muraza, O. Ind. Eng. Chem. Res. 2015, 54, 7181. (f) Earley, J. H.;
Bourne, R. A.; Watson, M. J.; Poliakoff, M. Green Chem. 2015, 17,
3018. (g) Moteki, T.; Flaherty, D. W. ACS Catal. 2016, 6, 4170.
(h) Taborga Claure, M.; Lee, L.-C.; Goh, J. W.; Gelbaum, L. T.;
Agrawal, P. K.; Jones, C. W. J. Mol. Catal. A: Chem. 2016, 423, 224.
(9) (a) Dowson, G. R. M.; Haddow, M. F.; Lee, J.; Wingad, R. L.;
Wass, D. F. Angew. Chem., Int. Ed. 2013, 52, 9005. (b) Wingad, R. L.;
Gates, P. J.; Street, S. T. G.; Wass, D. F. ACS Catal. 2015, 5, 5822.
(c) Tseng, K.-N. T.; Lin, S.; Kampf, J. W.; Szymczak, N. K. Chem.
Commun. 2016, 52, 2901. (d) Xie, Y.; Ben-David, Y.; Shimon, L. J. W.;
Milstein, D. J. Am. Chem. Soc. 2016, 138, 9077.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/jacs.7b05939.
■
S
Crystallographic data for [Mn]-1b, [Mn]-1c, [Mn]-1d,
[Mn]-1f, [Mn]-1g, 3, 4, 5, and 7 (ZIP)
Experimental details, characterization data, and spectra,
including Figures S1−S61, Schemes S1−S4, and Tables
S1−S8 (PDF)
(10) (a) Koda, K.; Matsu-ura, T.; Obora, Y.; Ishii, Y. Chem. Lett.
2009, 38, 838. (b) Chakraborty, S.; Piszel, P. E.; Hayes, C. E.; Baker,
R. T.; Jones, W. D. J. Am. Chem. Soc. 2015, 137, 14264.
(11) (a) Elangovan, S.; Topf, C.; Fischer, S.; Jiao, H.; Spannenberg,
A.; Baumann, W.; Ludwig, R.; Junge, K.; Beller, M. J. Am. Chem. Soc.
AUTHOR INFORMATION
Corresponding Author
*qiang_liu@mail.tsinghua.edu.cn
■
G
DOI: 10.1021/jacs.7b05939
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
2016, 138, 8809. (b) Elangovan, S.; Garbe, M.; Jiao, H.; Spannenberg,
A.; Junge, K.; Beller, M. Angew. Chem., Int. Ed. 2016, 55, 15364.
(c) Kallmeier, F.; Irrgang, T.; Dietel, T.; Kempe, R. Angew. Chem., Int.
Ed. 2016, 55, 11806. (d) Espinosa-Jalapa, N. A.; Nerush, A.; Shimon,
L. J. W.; Leitus, G.; Avram, L.; Ben-David, Y.; Milstein, D. Chem. - Eur.
J. 2017, 23, 5934.
(12) (a) Mukherjee, A.; Nerush, A.; Leitus, G.; Shimon, L. J. W.; Ben
David, Y.; Espinosa Jalapa, N. A.; Milstein, D. J. Am. Chem. Soc. 2016,
138, 4298. (b) Mastalir, M.; Glatz, M.; Gorgas, N.; Stoger, B.;
̈
Pittenauer, E.; Allmaier, G.; Veiros, L. F.; Kirchner, K. Chem. - Eur. J.
2016, 22, 12316. (c) Mastalir, M.; Glatz, M.; Pittenauer, E.; Allmaier,
G.; Kirchner, K. J. Am. Chem. Soc. 2016, 138, 15543. (d) Tondreau, A.
M.; Boncella, J. M. Organometallics 2016, 35, 2049. (e) Elangovan, S.;
Neumann, J.; Sortais, J.-B.; Junge, K.; Darcel, C.; Beller, M. Nat.
Commun. 2016, 7, 12641. (f) Pena-Lop
́
ez, M.; Piehl, P.; Elangovan, S.;
Neumann, H.; Beller, M. Angew. Chem., Int. Ed. 2016, 55, 14967.
(g) Anderez-Fernandez, M.; Vogt, L. K.; Fischer, S.; Zhou, W.; Jiao,
̃
́
́
H.; Garbe, M.; Elangovan, S.; Junge, K.; Junge, H.; Ludwig, R.; Beller,
M. Angew. Chem., Int. Ed. 2017, 56, 559. (h) Nguyen, D. H.; Trivelli,
X.; Capet, F.; Paul, J.-F.; Dumeignil, F.; Gauvin, R. M. ACS Catal.
2017, 7, 2022. (i) Deibl, N.; Kempe, R. Angew. Chem., Int. Ed. 2017,
56, 1663. (j) Bruneau-Voisine, A.; Wang, D.; Dorcet, V.; Roisnel, T.;
Darcel, C.; Sortais, J.-B. J. Catal. 2017, 347, 57. (k) Chakraborty, S.;
Gellrich, U.; Diskin-Posner, Y.; Leitus, G.; Avram, L.; Milstein, D.
Angew. Chem., Int. Ed. 2017, 56, 4229. (l) Deibl, N.; Kempe, R. Angew.
Chem., Int. Ed. 2017, 56, 1663. (m) Kallmeier, F.; Dudziec, B.; Irrgang,
T.; Kempe, R. Angew. Chem., Int. Ed. 2017, 56, 7261.
(13) (a) Shao, Z.; Fu, S.; Wei, M.; Zhou, S.; Liu, Q. Angew. Chem.,
Int. Ed. 2016, 55, 14653. (b) Fu, S.; Chen, N.-Y.; Liu, X.; Shao, Z.;
Luo, S.-P.; Liu, Q. J. Am. Chem. Soc. 2016, 138, 8588.
(14) In previous work (ref 10a), only dicarbonyl Mn complexes were
isolated using (R2PC2H4)2NH (R = iPr or Cy) as the ligands.
(15) Moteki, T.; Rowley, A. T.; Flaherty, D. W. ACS Catal. 2016, 6,
7278.
(16) Balaraman, E.; Khaskin, E.; Leitus, G.; Milstein, D. Nat. Chem.
2013, 5, 122.
(17) (a) Zhang, J.; Leitus, G.; Ben-David, Y.; Milstein, D. J. Am.
Chem. Soc. 2005, 127, 10840. (b) Gunanathan, C.; Shimon, L. J. W.;
Milstein, D. J. Am. Chem. Soc. 2009, 131, 3146.
(18) After reaction, the gas phase in the headspace of the pressure
vessel contains 81% H2, 11% CH4, 5% CO, 2% CO2, and 1% C2H4.
(19) Dub, P. A.; Scott, B. L.; Gordon, J. C. J. Am. Chem. Soc. 2017,
139, 1245.
(20) 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.
(21) A further disscussion of the base effect for the dehydrogenation
of ethanol is presented in the Supporting Information (Schemes S3
and S4).
H
DOI: 10.1021/jacs.7b05939
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX