Direct self-condensation of bio-alcohols in the aqueous phase

Article abstract of DOI:10.1039/c4gc00510d

Bio-alcohols (e.g. ethanol, butanol) are primarily obtained as diluted aqueous solutions from biomass fermentation, and thus the subsequent isolation is a very costly process. So the direct transformation of bio-alcohols in water will have great advantages. This study describes the development of catalysts used for the self-condensation of bio-alcohols in water (that mimic the primary fermentation solutions). Efficient iridium catalysts have been developed rationally from homogeneous to heterogeneous, and the immobilized catalysts could be reused without any loss of activity, which is very important for the development of practical processes. The expected self-condensation could be realized with 80-90% selectivity in water and air. Such a protocol might be used for producing butanol from ethanol solution directly, which is an improved higher-alcohol biofuel. Other useful chemicals, such as 2-ethylhexanol, could also be obtained from renewable resources through this condensation reaction. This journal is the Partner Organisations 2014.

Full text of DOI:10.1039/c4gc00510d

Green Chemistry
View Article Online
View Journal | View Issue
PAPER
Direct self-condensation of bio-alcohols in the
aqueous phase†
Cite this: Green Chem., 2014, 16,
971
3
a
b
a
a
a
Guoqiang Xu,* Tijs Lammens, Qiang Liu, Xicheng Wang, Linlin Dong,
b
c
a
a
Aldo Caiazzo, Nadim Ashraf, Jing Guan and Xindong Mu*
Bio-alcohols (e.g. ethanol, butanol) are primarily obtained as diluted aqueous solutions from biomass
fermentation, and thus the subsequent isolation is a very costly process. So the direct transformation of
bio-alcohols in water will have great advantages. This study describes the development of catalysts used
for the self-condensation of bio-alcohols in water (that mimic the primary fermentation solutions).
Efficient iridium catalysts have been developed rationally from homogeneous to heterogeneous, and the
immobilized catalysts could be reused without any loss of activity, which is very important for the devel-
opment of practical processes. The expected self-condensation could be realized with 80–90% selec-
tivity in water and air. Such a protocol might be used for producing butanol from ethanol solution
directly, which is an improved higher-alcohol biofuel. Other useful chemicals, such as 2-ethylhexanol,
could also be obtained from renewable resources through this condensation reaction.
Received 24th March 2014,
Accepted 13th June 2014
DOI: 10.1039/c4gc00510d
www.rsc.org/greenchem
4
raw materials for cosmetics, emulsifying agents, etc. Conden-
sation of bio-alcohols with fewer carbon atoms has also drawn
Introduction
Bio-alcohols, such as ethanol and butanol, can be used as sus- much attention from both industrial and academic commu-
tainable biofuels and future feedstocks for the chemical indus- nities, which might be used in manufacturing improved
1
try. Currently, they are produced in increasing amounts from higher-alcohol biofuels or other useful chemicals from
1
renewable biomass resources. However, the primarily obtained renewable resources. Based on a general hydrogen auto-trans-
crude bio-alcohols after fermentation are diluted aqueous fer mechanism (Scheme S1, ESI†), the alcohol condensation
solutions, and thus the subsequent purification is a very costly process is composed of dehydrogenation, condensation and
5
process. The polar nature of ethanol and butanol makes them hydrogenation steps, but can be carried out in one pot.
2
difficult to be separated from water. As in the production of Although other reaction pathways such as bimolecular conden-
st
1c
the 1 -generation fuel ethanol, usually more than half of the sation are possible under specific conditions, theoretically
energy is consumed in the distillation and dehydration steps. water is the sole byproduct. So, a water tolerant catalyst is
Therefore, the direct upgrading of aqueous ethanol or butanol intrinsically desired considering its inevitable formation.
solutions corresponding to those obtained after fermentation
will have great advantages.
On the other hand, the use of water as a reaction medium
has received significant attention due to its many advantages
3
Alcohol condensation, which is also named the Guerbet from economic, environmental, and safety standpoints.
reaction, is an important method to produce a dimeric However, despite the various efficient catalytic systems in
alcohol. The process can be conducted on a scale suitable for water that have been developed to date, execution of the de-
commercial application when the starting-alcohols contain 6 hydration reaction in water is still one of the most challenging
6
or more carbon atoms and the products are mainly used as research topics. The alcohol condensation, including butanol
and ethanol, is usually carried out in organic solvent, pure
alcohol or in the diluted gas phase. The addition of water to
the reactant mixture is usually detrimental to the catalysts. For
a
CAS Key Laboratory of Bio-based Materials, Qingdao Institute of Bioenergy and
Bioprocess Technology, Chinese Academy of Sciences, Qingdao, Shandong 266101,
example, Carlini and coworkers reported the application of
PR China. E-mail: xugq@qibebt.ac.cn, muxd@qibebt.ac.cn;
[
Pd(dppe)Cl
the self-condensation of butanol in the pure butanol phase at
80–280 °C. The product 2-ethylhexanol could be obtained in
2
] or Pd/C in combination with BuONa to catalyze
Fax: (+86) 532-80662724
b
Shell Global Solutions International B.V., Grasweg 31, 1031HW Amsterdam,
The Netherlands
1
c
Shell International Explorations and Production, Westhollow Technology Center,
about 60% yield with good selectivity. But the progressive
hydrolysis of BuONa to the inactive NaOH would destroy the
catalytic systems. In their series of related work, solid bases of
3333 Highway 6 South, Houston, TX 77082-3101, USA
†Electronic supplementary information (ESI) available. See DOI:
10.1039/c4gc00510d
This journal is © The Royal Society of Chemistry 2014
Green Chem., 2014, 16, 3971–3977 | 3971

View Article Online
Paper
Green Chemistry
Mg/Al mixed oxides derived from hydrotalcite-type (HT) pre- the fast catalyst evaluation, and then the selected catalytic
cursors were developed to replace the soluble sodium alcoho- systems could be applied to ethanol condensation after some
late. Although such catalysts were tolerant of the co-produced modifications.
water, generally the reactions were still carried out in dry
alcohol mixtures. Solid bases, such as hydroxyapatites (HAP)
7
Catalyst development from butanol condensation
and HT derivatives, could also be used alone to catalyze the
Initial exploration. Both microwave irradiation and conven-
ethanol or butanol condensation. However, the reaction temp- tional heating were applied, and their performance was gener-
erature was over 300 °C and the pure alcohol reactant should ally similar. The reaction rate was faster under microwave
8
be vaporized and diluted with inert gas first. Ishii and co- irradiation, which was normal because the energy could be
workers reported the use of iridium catalysts (e.g. [Ir(acac)- delivered directly to the reacting partners and usually
1
1
(
cod)], [Cp*IrCl ] ) to promote the Guerbet reaction of primary higher reaction rates were achieved. So microwave irradiation
2 2
alcohols. Condensation of ethanol could be obtained in about was applied for fast catalyst screening, and then a more
0% conversion and over 80% selectivity towards higher conventional heating method was used for further catalyst
alcohol products. Similarly, the reactions were performed in a optimization.
4
pure alcohol or p-xylene solvent, and additional alkene addi-
Bifunctional catalysts comprised of transition metal precur-
tives (e.g. 1,7-octadiene) were necessary, which might function sors (Ru, Rh, Cu, Ir, Pd, Ag, etc.) and bases which would
as hydrogen acceptors or as weakly coordinated ligands to promote the expected dehydrogenation/hydrogenation and
9
stabilize the Ir intermediates. Other metal catalysts of condensation steps were rationally screened (Table S1, ESI†).
ruthenium and rhodium were also employed, and sometimes Our initial results suggested that the condensation of butanol
1
0
the selectivity could be more than 90%. However, because of could be performed in organic solvents smoothly under basic
the detrimental effect of water, generally all reactions were conditions. Transition metal precursors, such as CuBr and
carried out in organic solvent or pure alcohol. So, new catalysts RhCl , could effectively catalyze the reaction. But the addition
3
are necessary considering the alcohol condensation in an of a small amount of H O would totally destroy the catalysts,
2
aqueous solvent. Here we are pleased to report our explora- and condensation would not proceed (Table S1, entries 3–8,
tions in the catalyst development for butanol and ethanol self- ESI†). Such an incompatibility with water was similar to the lit-
condensations in water.
erature reports and clearly evidenced the challenge in realizing
the alcohol condensation in the aqueous phase. Although
some combinations of Rh, Ru and other metal precursors gen-
erated trace amounts of products in the aqueous phase, we did
not obtain an acceptable transformation using these catalysts.
Results and discussion
According to the previous reports, the condensation of ethanol But to our delight, we found that the combination of IrCl and
3
might generate butanol and other C and C products, which 1,10-phenanthroline (Phen) exhibited an obvious activity for
6
8
come from further condensation steps, while the products butanol condensation in water (Table 1, entry 1). So the related
from butanol condensation are relatively simple (Scheme 1). iridium catalysts were carefully studied by variations of
In addition, ethanol is considered to be a more sluggish iridium precursors, bases, catalyst loading, reaction time,
7
–9
reactant. So initially the condensation of butanol is used for temperature, etc.
The use of other organic solvents such as hexane was
intended to promote the transformation by phase separation,
but experimental results suggested that it was unnecessary.
The condensation performed well in pure water. An interesting
phenomenon was that large amounts of IrCl and phenanthro-
3
line mixtures were insoluble in water initially, but when the
base was added, the aqueous phase turned clear and looked
like homogeneous. Phenanthroline might function as a biden-
tate ligand or a stabilizer, and possibly active Ir intermediates
were formed. However, the catalyst agglomerated and black
precipitates were formed during the reaction, even at 125 °C.
3
TEM analyses also suggested that the IrCl -phenanthroline
catalyst dispersed well before the reaction, but aggregated evi-
dently during the transformation (Fig. S1–S2, ESI†). Similar
1
2
phenomenon was reported by Cadierno and coworkers. They
investigated palladium and bipyridine catalysts in the reaction
of arylboronic acids with unsaturated carbonyl compounds in
aqueous medium. Under strongly basic conditions, the catalyst
solubility increased and aqueous nano-dispersed solutions
could be prepared, but extensive aggregation was observed
Scheme 1 Main products in butanol and ethanol condensation.
3972 | Green Chem., 2014, 16, 3971–3977
This journal is © The Royal Society of Chemistry 2014

View Article Online
Green Chemistry
Paper
a
Table 1 Catalyst development for butanol condensation in water
With regard to butanol condensation in diluted aqueous
solution, nearly 60% yield and 90% selectivity towards the
desired 2-ethylhexanol could be obtained within 24 h (Table 1,
entry 10), which clearly evidenced the feasibility of our original
Condition
solvent-T [°C]-t [h]
Yield (Sele.)
of 1a [%]
Entry
Catalyst [M]–base
7,9
1
2
3
4
5
6
7
8
9
1
1
IrCl
IrCl
IrCl
IrCl
IrCl
IrCl
IrCl
IrCl
IrCl
IrCl
IrCl
3
3
3
3
3
3
3
3
3
3
3
/Phen–LiOH
/Phen–LiOH
/Phen–KOH
/Phen–KOH
/Phen–KOH
/Phen–KOH
/Phen–KOH
/Phen–KOH/NaOAc
/Phen–KOH/NaOAc
/Phen–KOH/NaOAc
/Phen–KOH
Hexane/H
Hexane/H
Hexane/H
2
2
2
O-120-2
O-140-2
O-140-2
6
12
14
6
design. Such results are comparable to literature reports,
while the direct transformation in water is a significant advan-
tage. 2-Ethylhexanol is an important product which is usually
used as a precursor to plasticizers. It is now produced indust-
rially not from butanol but from butanal derived from
b
H
2
H
2
H
2
H
2
H
2
H
2
H
2
H
2
O-125 -40
b
O-160 -6
45 (82)
55 (81)
31 (78)
48 (87)
45 (81)
59 (88)
27 (72)
b
O-160 -24
c
b
O-160 -6
b
3
propene. One major drawback of the present IrCl -phenanthro-
O-160 -6
b
O-170 -6
line catalyst is that it is unstable and agglomerates during the
reaction. Such aggregation makes the catalyst difficult to be
recycled, and in addition, further improvement of activity is
possible if the aggregation could be avoided. Therefore
it would be advantageous to find a way to stabilize and reuse
d
b
0
1
O-160 -24
e
b
O-160 -6
a
Reaction conditions: 1 g butanol, 0.2 mol% metal chloride hydrates
abbreviated as metal chlorides), 0.05 g Phen if used, 1 equiv. of base
for mixed bases, 0.8 equiv. of NaOAc was added), 15 mL solvent (for
O is 1 : 2 the precious metal catalyst.
except for entries 4 and 5). Microwave irradiation was used unless
(
(
mixed solvents, the volume ratio of organic solvent to H
2
Ligand development. The modification of ligands from
steric and electronic aspects was usually employed to obtain a
more suitable catalyst. The N–N multi-coordinated structure in
phenanthroline was speculated to be critical in creating an
active catalyst, so more ligands with this key function were
investigated in order to improve the catalyst stability and
b
otherwise stated. The reactions were heated using an oil bath and
c
d
carried out under air. 0.4 mol% IrCl
3
·3H
2
O was used. 2 equiv. of
e
KOH were used. The reaction was carried out under N
2
.
during the reaction. Such aggregates of nanoparticles were still activity (Scheme 2). All reactions were performed under similar
active to catalyze the condensation, as with the increasing conditions as specified in Table S2 in ESI.†
reaction time more products formed, but it was not surprising
that their activity would be reduced inevitably (Table 1,
entries 5 and 6).
Reaction temperature exerted an important influence on
the transformation. The conversion was improved evidently
at 160 °C, but higher temperatures were ineffective (Table 1,
entry 9). Increasing the iridium loading exhibited no improve-
ment (Table 1, entry 7), possibly because of the catalyst aggre-
gation. A high ligand to iridium ratio would suppress the
activity at 140 °C, but little difference was found at 155–160 °C.
Common inorganic bases, such as KOH, could promote the
condensation effectively. More expensive and water sensitive
t
bases (e.g. KO Bu, NaOEt) could be avoided. About 1.0 equi-
valent of the base relative to the amount of butanol was
needed to obtain a good conversion (Table S2, ESI†). NaOAc
could not promote the condensation alone, but the combi-
nation of NaOAc with KOH would increase the reaction rate,
1
3
especially at low temperatures. However, little difference was
found at 160 °C, except that the selectivity was a little
improved (Table 1, entries 5 and 8). The reaction could be
performed under air, and more interestingly, the reaction rate
was even faster than that under inert conditions (Table 1,
1
4
entry 11). Very small amounts of condensation products with
unsaturated CvC and CvO bonds were formed under both
the nitrogen and air atmosphere (less than 1%). Another
unwanted byproduct which could usually be observed was
1
c
butyric acid. The oxidation of alcohol to acid had been investi-
gated before and transition metal catalysts such as Pt could
1
5
promote the transformation. Sometimes ethanol was also
formed, which should come from the retro-aldol reaction.
However, these byproducts could be suppressed to less than
Scheme 2 Modification of ligands (yields of 2-ethylhexanol in
1
% after optimization (Fig. S5, ESI†).
brackets).
This journal is © The Royal Society of Chemistry 2014
Green Chem., 2014, 16, 3971–3977 | 3973

View Article Online
Paper
Green Chemistry
Very small amounts of products were generated using IrCl
3
3 3
Different iridium precursors such as [Ir(acac) ] and [Ir(OAc) ]
alone or in combination with other phosphine ligands. Such exhibited similar activity. After extraction, the water soluble
results clearly indicated the crucial role played by a suitable catalyst could be reused three times without an obvious
ligand. Methyl substitutes in phenanthroline near the N-atoms decrease in activity (Table 2, entries 4–8). The base solubilized
would sharply decrease the catalyst activity (L1 in Scheme 2), in aqueous solution could also be reused. It could be antici-
which implied that the sterically crowded coordination struc- pated from the reaction mechanism, which suggested that the
ture would block the active center. Substitutes away from the base would not be consumed during the transformation.
N-atoms did not decrease the activity obviously; however, this However, more precipitates formed later. TEM analyses also
change did not generate a more active catalyst (L2 in suggested that the water soluble catalyst dispersed well during
Scheme 2). Although there is one more N-coordination site in the reaction, but some aggregation was observed after the fifth
L5 and L6 compared to phenanthroline, the catalyst formed run (Fig. S3–S4, ESI†).
also exhibited very low activity, and iridium black formed.
It is possible to find more suitable water soluble ligands or
The imine ligands were not active possibly because of the reaction conditions after further optimization. For example,
similar steric reason and low solubility in water (L8 and L9 in higher molar ratio of ligand to metal exhibited better results,
Scheme 2). With regard to electronic modification of the and the expected selectivity could be up to 91% (Table 2, entry
ligands, the electron donating hydroxide group decreased the 11). However, it should be noted that the water soluble catalyst
activity evidently (L4 in Scheme 2), while the electron with- possibly will not be the best choice for the condensation of
drawing sulfo-group increased the activity (L3 and L10 in butanol or ethanol in water from a practical viewpoint.
Scheme 2). Possibly a higher electron density in the iridium Because bio-ethanol or bio-butanol is primarily accessed as a
center would make the alcohol dehydrogenation step difficult. diluted aqueous solution from biomass fermentation, our
The aliphatic amine ligand was ineffective (L7 in Scheme 2) design is to carry out the condensation directly in water. The
because of the lack of the π-back-bonding between iridium catalyst solubilized in the aqueous phase is difficult to be iso-
and aromatic N–N ligands, which would ensure tighter lated from water, so practically its reuse is still difficult. Hetero-
1
3
coordination.
geneous catalysts might be better considering the ultimate
The water soluble ligand L10 (bathophenanthroline disulfo- practical application.
nic acid disodium salt) exhibited very interesting character-
Catalyst immobilization. Based on the above explorations, Ir
istics. The catalyst remained in the aqueous phase and is the most suitable candidate to be immobilized and its pro-
dispersed well after the reaction, and the product yield was perties should be modified in the immobilized catalyst. Ir/C
considerable. Reaction temperature was also important (Table 2, (5 wt% iridium content) was not a very active catalyst when
entry 9), but the temperature could be about 10 °C lower than used separately, and only 6% yield was obtained (Table 3,
the IrCl
catalyst was stable and there was no black precipitate. But at some N-containing ligands, the transformation could be pro-
60 °C, small precipitates formed. At 170 °C, more precipitates moted and the activity was comparable to the IrCl -phen-
formed and the activity was destroyed (Table 2, entry 2). anthroline catalyst. Phenanthroline and its water soluble
3
-phenanthroline catalyst. At 150 °C, the water soluble entry 1). However, when Ir/C was used in combination with
1
3
a
a
Table 2 Butanol condensation using the water-soluble ligand L10
Table 3 Butanol condensation using heterogeneous catalysts
Catalyst [Ir]-ligand
(molar ratio)
Condition
T [°C]-t [h]
Conv.
[%]
Yield (Sele.)
of 1a [%]
Conv. Yield (Sele.)
Entry
Entry [Ir] Catalyst
[%]
of 1a [%]
1
2
3
4
IrCl
IrCl
[Ir(acac)
Reuse entry 3 for 2
Reuse entry 3 for 2
Reuse entry 3 for 3
Reuse entry 3 for 4
Reuse entry 3 for 5
[Ir(acac)
[Ir(OAc)
[Ir(OAc)
Reuse entry 11 for 2
Reuse entry 11 for 3
3
–L10 (1 : 3)
–L10 (1 : 3)
150-12
170-12
150-12
150-24
150-24
150-24
150-24
150-24
140-24
150-12
150-12
150-24
150-24
45
40
55
51
49
51
36
21
21
57
54
55
50
35 (78)
20
40 (73)
34
35
38
26
12
16 (76)
42 (74)
49 (91)
46 (84)
42
1
2
3
4
5
6
7
8
9
10
11
12
13
Ir/C
Ir/C–Phen
Ir/C–Phen
Reuse of Ir/C in entry 2 – Phen
16
36
25
25
53
32
6
31 (86)
19
12
42 (80)
7
11
37 (88)
36 (86)
36 (88)
37 (86)
36 (88)
36 (86)
3
b
3
]–L10 (1 : 3)
nd
nd
rd
th
b
5
IrCl
3
(0.2 mol%)–Phen-AC
6
7
8
9
1
1
1
1
Reuse entry 5
[Ir(acac)
[Ir(acac)
3
]–Phen-AC pyrolyzed (1 wt% Ir) 19
]–Phen-AC pyrolyzed (5 wt% Ir) 42
]–Phen-AC pyrolyzed (5 wt% Ir) 42
th
3
3
]–L10 (1 : 3)
[Ir(OAc)
3
nd
0
1
2
3
3
]–L10 (1 : 3)
]–L10 (1 : 8)
Reuse entry 9 for 2
Reuse entry 9 for 3
Reuse entry 9 for 4
Reuse entry 9 for 5
41
43
41
42
rd
th
th
3
nd
rd
a
a
Reaction conditions: 1 g butanol, 0.01 g iridium precursors, 1.0
Reaction conditions: 1 g butanol, 0.15 g supported catalyst, 0.05 g
O, 160 °C,
equiv. of KOH and 0.8 equiv. of NaOAc, 18 mL H
2
O. The reactions were phenanthroline hydrate if used, 1.0 equiv. of KOH, 18 mL H
2
carried out under air and heated using an oil bath. The aqueous 16 h. The catalysts were filtered, rinsed with ethanol and water, and
solution containing the iridium catalyst and base was reused after then dried in an oven for the recycling tests. Preparation details for the
b
b
extraction with ether and hexane. Fresh 0.3 equiv. of KOH and NaOAc pyrolyzed catalysts are described in ESI. The reaction was carried out
were added.
at 125 °C.
3974 | Green Chem., 2014, 16, 3971–3977
This journal is © The Royal Society of Chemistry 2014

View Article Online
Green Chemistry
Paper
counterpart L10 were still the most active ligands (Table 3, enriched carbon surface formed in this protocol, which would
entry 2, and Table S2 in ESI†). Upon the reuse of Ir/C, its make the interaction between iridium and carbon supports
1
8
activity decreased (Table 3, entry 4). The iridium concentration tighter. Five consecutive experiments were performed in
−
5
−1
was determined to be about 5 × 10 g mL in aqueous solu- order to investigate the catalyst stability and reusability.
tion after the reaction, which implied that about 10% iridium Notably, the pyrolyzed catalyst was highly stable and could be
in Ir/C leached into the aqueous solvent. Consequently the reused without any loss of activity, which was important
color of the aqueous solvent changed to brownish-yellow for the development of practical processes (Table 3, entries
(
Fig. S6, ESI†). Ir/C might act as the iridium source to afford 10–13).
1
6
the real active intermediate. Another interesting aspect was
the activity at low temperatures. Ir/C-phenanthroline was more
Application to ethanol condensation
active than IrCl -phenanthroline at 125 °C (Table 3, entry 3), Bio-ethanol is produced in large amounts from renewable
3
although the reaction rate was still low. This improvement resources currently. Condensation of ethanol in water could be
could be attributed to the avoidance of iridium aggregation. used in the production of improved higher-alcohol biofuels
Activated carbon was added to the IrCl -phenanthroline and other useful chemicals from the fermentation of ethanol
3
1
7
system for the immobilization of the active iridium catalyst.
solution directly. High energy-cost dehydration and distillation
Such a combination did not generate a negative impact steps might be eliminated because such products could be
1
,19
(Table 3, entry 5), but the reuse of this solid exhibited sharply less expensive to isolate from water.
Application of the cata-
decreased activity (Table 3, entry 6). The interaction between lysts developed from butanol condensation to ethanol conden-
the iridium catalyst and the carbon support was not tight sation exhibited similar tendency. It was found that the main
enough and possibly the active catalyst was lost during products in ethanol condensation were butanol, hexanol and
isolation.
octanol (n- and i-). The unsaturated counterparts were much
Recently, Beller and coworkers reported a novel method for less. Reaction temperature also influenced the transformation
the preparation of heterogeneous catalysts via immobilization evidently (Table 4, entry 3). Water soluble catalysts exhibited
and pyrolysis of molecularly defined complexes. The hetero- better activity than Ir-phenanthroline catalysts (Table 4, entries
genized carbon supported cobalt or iron–nitrogen catalysts were 1 and 2), and similarly the aqueous catalyst solution could be
successfully applied in the nitroarene reduction and alcohol reused without an obvious loss of activity (Table 4, entry 5).
1
8
selective oxidation. This protocol also proved to be applicable Interestingly, the distribution of products could be adjusted by
here. Pyrolysis of the in situ-generated Ir-phenanthroline com- the catalytic systems. [Ir(OAc) ] generated more deep conden-
3
plexes supported on carbon led to highly active catalysts. The sation products than IrCl3 (Table 4, entries 2 and 4). High
preparation of the catalyst material, such as iridium loading, molar ratio of ligand to Ir and the addition of NaOAc were
pyrolysis temperature, different iridium precursors and carbon slightly better for the formation of butanol (Table 4, entries
supports, would all influence the catalyst activity (Table S3, 6–8). Ir/C and the pyrolyzed iridium–nitrogen catalysts were
ESI†). After optimization, the transformation could be also applicable, and butanol could be formed with more than
achieved comparable to the homogeneous iridium catalyst 50% selectivity (Table 4, entries 10 and 11). Until now, the con-
(Table 3, entries 8 and 9). Moreover, the resulting aqueous densation products butanol, hexanol and octanol could be
solvent was colorless and there was no iridium leaching obtained in more than 40% one-way overall yield and 80%
according to AAS determination (Fig. S6, ESI†). The nitrogen- total selectivity.
a
Table 4 Ethanol condensation in the aqueous phase
Yield [%]
Entry
Catalyst [Ir]–base
Condition T [°C]-t [h]
Conv. [%]
1b
2b
3b
4b
5b
Overall sele. [%]
1
2
3
4
5
6
7
IrCl
IrCl
IrCl
3
3
3
–Phen (1 : 10, molar ratio)–KOH/NaOAc
–L10 (1 : 3)–KOH
–L10 (1 : 3)–KOH/NaOAc
160-16
150-16
115-40
150-16
150-16
150-16
150-16
150-16
160-24
160-16
160-16
45
43
10
45
40
48
47
52
41
43
45
17
18
5
3
4
1
7
7
7
6
7
5
7
7
1
2
0.5
4
3
4
3
3
2
1
1
0.5
3
3
2
2
3
3
1
1
0.5
2
2
2
2
3
2
51
60
75
78
78
77
77
81
78
76
77
[Ir(OAc)
Reuse entry 4 for 2
[Ir(OAc)
[Ir(OAc)
[Ir(OAc)
Ir/C-Phen–KOH
[Ir(OAc) ]–Phen-AC pyrolyzed (5 wt% Ir)–KOH
Reuse entry 10 for 2
3
]–L10 (1 : 3)–KOH
19
16
22
23
26
20
23
25
nd
3
3
3
]–L10 (1 : 3)–KOH/NaOAc
]–L10 (1 : 8)–KOH
]–L10 (1 : 10)–KOH/NaOAc
b
8
9
10
1
3
1
1
1
1
0.5
0.5
nd
1
a
Reaction conditions: 1 g ethanol, 0.01 g iridium precursors or 0.2 g supported catalysts, 1.0 equiv. of KOH and 1 equiv. of NaOAc if used, 18 mL
O. All reactions were carried out under air and heated in an oil bath. The reuse and preparation of the pyrolyzed catalyst were the same as
H
2
b
3
described in Tables 2 and 3. 0.02 g [Ir(OAc) ] and 1.5 equiv. of KOH were used.
This journal is © The Royal Society of Chemistry 2014
Green Chem., 2014, 16, 3971–3977 | 3975

View Article Online
Paper
Green Chemistry
geneous, and the immobilized catalysts are evidenced to be
stable and reusable. The reaction could be performed under
air conditions and the use of other additives such as H or
2
Experimental
Chemicals and equipment
Commercially available chemicals including transition metal alkene could be avoided. Considering the ultimate practical
precursors, carbon supported catalysts, inorganic bases, application, the catalytic activity and selectivity should be
organic reagents and GC standards were purchased from improved further, especially for ethanol condensation. In
SCRC, Alfa Aesar or TCI and used as received. Commercially addition, nearly stoichiometric amount of KOH solution is not
unavailable ligands were prepared according to the literature eco-friendly, which is also a common concern in the industrial
2
0
in ESI.† The concentration of ethanol or butanol usually processes catalyzed by bases. Such challenges are being
varied from 5% to 15% to mimic the primary fermentation addressed in our lab now.
solution.
Microwave irradiation experiments were performed on an
Anton Paar Synthos 3000. Experiments under conventional
thermal treatment were carried out in an autoclave and heated
Acknowledgements
using an oil bath or heating plate with magnetic stirring. GC Financial support from Shell Global Solutions under
analyses were carried out on a VARIAN 450-GC equipped with contract PT27918 is greatly acknowledged. We also thank
a DB-FFAP capillary column (30 m × 0.32 mm × 0.25 µm). the Natural Science Foundation of China (no. 21273260)
GC-MS analyses were carried out on an Agilent 7890A GC with and the Shandong Provincial Natural Science Foundation for
a 5975C MS-unit. TEM (transmission electron microscopy) Distinguished Young Scholar, China (no. JQ201305) for
images were obtained using a field emission H-7600 electron financial support.
microscope at 120 kV. AAS (atomic absorption spectroscopy)
was measured on an INESA 4510F.
Notes and references
Quantitative GC analyses
Generally all reactions in this research were carried out in
diluted aqueous solutions. Because the organic products
exhibit low solubility in water, it is necessary to collect the pro-
ducts by extraction. In addition, butanol is partly soluble in
water and ethanol is miscible with water, so that the analyses
of the aqueous phase are also necessary for the quantitative
determination. The internal standard method was used for the
analyses of the organic phase. Because the organic internal
standards were insoluble in water, a modified external
standard method was applied for the analyses of the aqueous
phase. The known concentrations of alcohol reactants and
products were analyzed using this method, and the obtained
results suggested its feasibility.
1 (a) J. Goldemberg, Science, 2007, 315, 808; (b) C. Angelici,
B. M. Weckhuysen and P. C. A. Bruijnincx, ChemSusChem,
2013, 6, 1595; (c) J. T. Kozlowski and R. J. Davis, ACS Catal.,
2013, 3, 1588.
2 (a) O. J. Sanchez and C. A. Cardona, Bioresour. Technol.,
2008, 99, 5270; (b) X. Gao, X. Zou, F. Zhang, S. Zhang,
H. Ma, N. Zhao and G. Zhu, Chem. Commun., 2013, 49,
8839.
3 (a) C. H. Christensen, B. Jørgensen, J. R. Hansen,
K. Egeblad, R. Madsen, S. K. Klitgaard, S. M. Hansen,
M. R. Hansen, H. C. Andersen and A. Riisager, Angew.
Chem., Int. Ed., 2006, 45, 4648; (b) J. C. S. Ruiz, R. Luque
and A. S. Escribano, Chem. Soc. Rev., 2011, 40, 5266.
4
A. J. O’Lenick, J. Surfactants Deterg., 2001, 4, 311.
5 G. E. Dobereiner and R. H. Crabtree, Chem. Rev., 2010, 110,
81.
C. J. Li, Chem. Rev., 2005, 105, 3095.
General reaction procedure
6
Take the butanol condensation using the IrCl
3
-phenanthroline
6
catalyst for example: butanol (1 g, 13.5 mmol), IrCl ·3H O
(
3
2
0.01 g, 2.8 × 10− mmol, 0.2 mol%) and phenanthroline
2
7 (a) C. Carlini, A. Macinai, A. M. R. Galletti and G. Sbrana,
J. Mol. Catal. A: Chem., 2004, 212, 65; (b) C. Carlini,
C. Flego, M. Marchionna, M. Noviello, A. M. R. Galletti,
G. Sbrana, F. Basile and A. Vaccari, J. Mol. Catal. A: Chem.,
hydrate (0.05 g, 0.25 mmol) were mixed in H O (9 mL). Then a
2
KOH solution (0.76 g, 13.5 mmol in 9 mL H O) was added.
2
The mixture was heated using microwave irradiation or an oil
bath without exclusion of air. After heating for a specified reac-
2004, 220, 215; (c) C. Carlini, M. Marchionna, M. Noviello,
A. M. R. Galletti, G. Sbrana, F. Basile and A. Vaccari, J. Mol.
Catal. A: Chem., 2005, 232, 13.
tion time, Et O was added for extraction. Both the organic and
2
aqueous phases were analyzed by GC. Products were identified
using pure chemicals and a GC-MS.
8
(a) T. Tsuchida, S. Sakuma, T. Takeguchi and W. Ueda,
Ind. Eng. Chem. Res., 2006, 45, 8634; (b) T. Tsuchida,
T. Yoshioka, S. Sakuma, T. Takeguchi and W. Ueda, Ind.
Eng. Chem. Res., 2008, 47, 1443; (c) M. León, E. Díaz and
S. Ordónez, Catal. Today, 2011, 164, 436.
Conclusions
Direct self-condensation of bio-alcohols including butanol and
ethanol in water has been successfully realized. Novel iridium
catalysts are rationally developed from homogeneous to hetero-
9 (a) T. Matsu-ura, S. Sakaguchi, Y. Obora and Y. Ishii, J. Org.
Chem., 2006, 71, 8306; (b) K. Koda, T. Matsuura, Y. Obora
and Y. Ishii, Chem. Lett., 2009, 38, 838.
3976 | Green Chem., 2014, 16, 3971–3977
This journal is © The Royal Society of Chemistry 2014

View Article Online
Green Chemistry
Paper
1
0 (a) D. F. Wass and G. R. M. Dowson, WO Pat, 004 572A1, 16 L. Yin and J. Liebscher, Chem. Rev., 2007, 107, 133.
2
0
012; (b) Y. Tanaka and M. Utsunomiya, US Pat, 17 N. Kania, N. Gokulakrishnan, B. Léger, S. Fourmentin,
298 613A1, 2010. E. Monflier and A. Ponchel, J. Catal., 2011, 278, 208.
1
1 (a) A. J. A. Watson, A. C. Maxwell and J. M. J. Williams, 18 (a) R. V. Jagadeesh, G. Wienhofer, F. A. Westerhaus,
J. Org. Chem., 2011, 76, 2328; (b) W. Zhang, X. Dong and
W. Zhao, Org. Lett., 2011, 13, 5386; (c) G. Xu, Q. Li, J. Feng,
Q. Liu, Z. Zhang, X. Wang, X. Zhang and X. Mu,
ChemSusChem, 2014, 7, 105.
A. E. Surkus, M. M. Pohl, H. Junge, K. Junge and M. Beller,
Chem. Commun., 2011, 47, 10972; (b) R. V. Jagadeesh,
H. Junge, M. M. Pohl, J. Radnik, A. Brückner and M. Beller,
J. Am. Chem. Soc., 2013, 135, 10776; (c) F. A. Westerhaus,
R. V. Jagadeesh, G. Wienhofer, M. M. Pohl, J. Radnik,
A. E. Surkus, J. Rabeah, K. Junge, H. Junge, M. Nielsen,
A. Bruckner and M. Beller, Nat. Chem., 2013, 5, 537;
(d) R. V. Jagadeesh, A. E. Surkus, H. Junge, M. M. Pohl,
J. Radnik, J. Rabeah, H. Huan, V. Schünemann,
A. Brückner and M. Beller, Science, 2013, 342, 1073.
1
2 E. T. Mendivil, J. Diez and V. Cadierno, Catal. Sci. Technol.,
2
011, 1, 1605.
1
3 (a) G. J. Brink, I. W. C. E. Arends and R. A. Sheldon,
Science, 2000, 287, 1636; (b) G. J. Brink, I. W. C. E. Arends
and R. A. Sheldon, Adv. Synth. Catal., 2002, 344, 355.
1
1
4 (a) S. Liao, K. Yu, Q. Li, H. Tian, Z. Zhang, X. Yu and Q. Xu,
Org. Biomol. Chem., 2012, 10, 2973; (b) Q. Xu and Q. Li, 19 (a) M. Lapuerta, R. Garcia-Contreras, J. Campos-Fernandez
Chin. J. Org. Chem., 2013, 33, 18.
and M. P. Dorado, Energy Fuels, 2010, 24, 4497;
(b) X. Zhang, Z. Liu, X. Xu, H. Yue, G. Tian and S. Feng,
ACS Sustainable Chem. Eng., 2013, 1, 1493.
5 (a) T. Wang, H. Shou, Y. Kou and H. Liu, Green Chem.,
2
009, 11, 562; (b) S. E. Davis, B. N. Zope and R. J. Davis,
Green Chem., 2012, 14, 143.
20 G. Busca, Chem. Rev., 2010, 110, 2217.
This journal is © The Royal Society of Chemistry 2014
Green Chem., 2014, 16, 3971–3977 | 3977