Direct C-C coupling of bio-ethanol into 2,3-butanediol by photochemical and photocatalytic oxidation with hydrogen peroxide

Article abstract of DOI:10.1039/c6gc00883f

Theoretically, selective C-H manipulation in ethanol can result in a direct C-C coupling synthesis of 2,3-butanediol (2,3-BDO). However, this process is typically extremely difficult to achieve because of the high complexity of the involved chemical bonds. In this work, we determine that hydroxide radicals generated from the photolysis of H₂O₂ can selectively attack the α-hydrogen atom in ethanol aqueous solutions and crack the C-H bond to produce hydroxyethyl radicals, which subsequently undergo C-C coupling to form 2,3-BDO. This selective C-H breakage is determined by the reaction rate, which is primarily controlled by the local H₂O₂ concentration at a given irradiation intensity. At a moderate reaction rate of ethanol (37 mmol h^-1), the 2,3-BDO selectivity reaching as high as 91% can be obtained. The introduction of a catalyst can further increase ethanol conversion and enhance the 2,3-BDO formation rate by controlling the reaction rate. This result provides an environment-friendly approach to convert bio-ethanol directly to 2,3-BDO and to manipulate a single bond selectively in complex bonding situations.

Full text of DOI:10.1039/c6gc00883f

Green Chemistry
View Article Online
View Journal
PAPER
Direct C–C coupling of bio-ethanol into
2,3-butanediol by photochemical and
Cite this: DOI: 10.1039/c6gc00883f
photocatalytic oxidation with hydrogen peroxide†
Na Li,a Wenjun Yan,a Pengju Yang,a,b Hongxia Zhang,a,c Zhijian Wang,a
Jianfeng Zheng,a Suping Jiaa and Zhenping Zhu*a
Theoretically, selective C–H manipulation in ethanol can result in a direct C–C coupling synthesis of 2,3-
butanediol (2,3-BDO). However, this process is typically extremely difficult to achieve because of the high
complexity of the involved chemical bonds. In this work, we determine that hydroxide radicals generated
from the photolysis of H2O2 can selectively attack the α-hydrogen atom in ethanol aqueous solutions and
crack the C–H bond to produce hydroxyethyl radicals, which subsequently undergo C–C coupling to
form 2,3-BDO. This selective C–H breakage is determined by the reaction rate, which is primarily con-
trolled by the local H2O2 concentration at a given irradiation intensity. At a moderate reaction rate of
ethanol (37 mmol h−1), the 2,3-BDO selectivity reaching as high as 91% can be obtained. The introduction
of a catalyst can further increase ethanol conversion and enhance the 2,3-BDO formation rate by control-
ling the reaction rate. This result provides an environment-friendly approach to convert bio-ethanol
directly to 2,3-BDO and to manipulate a single bond selectively in complex bonding situations.
Received 29th March 2016,
Accepted 14th June 2016
DOI: 10.1039/c6gc00883f
www.rsc.org/greenchem
duction rapidly increased and the total annual capacity
reached 100 billion L in 2013.6 The transformation of bio-
Introduction
In the modern green chemical industry, selectively manipulat- ethanol into valuable chemicals or high calorific value fuels is
ing chemical bonds is a primary approach in organic chem- of great interest in different countries owing to its availability
istry and catalysis science that is becoming increasingly as a feedstock. However, some limitations need to be overcome
important.1 Among them, the practical and selective for the effective application of bio-ethanol. For example, bio-
functionalization of traditionally inert aliphatic C–H bonds ethanol obtained by fermentation of biomasses has large
has been developed as a powerful strategy to form new chemi- amounts of water (12% ethanol) and its dehydration is very
cal bonds.2,3 However, one of the major challenges facing the complex requiring the consumption of large amounts of
continued advance of this field is the ability to manage a energy with a higher cost due to ethanol–water azeotropic solu-
single bond controllably in a compound containing many tion.7,8 From the economic standpoint, development of the
such bonds and an array of functional groups.4 With the water phase reaction for ethanol conversion is very profitable.
increasing demands for energy, concerns about anthropogeni- Moreover, ethanol has multifarious chemical bonds, including
cally caused global climate change, and depletion of fossil α-C–H, β-C–H, O–H, C–O, and C–C. These bonds provide
feedstock, more and more attempts have been focusing on ethanol with versatile functional properties, which has
alternative and renewable biomass sources for fuels and resulted in the current extensive ethanol industry that pro-
chemicals. Bio-ethanol, as an important biomass platform duces aldehydes, acids, ethers, esters, olefins, etc.9 However, it
molecule, has great potential as a renewable alternative to pet- is difficult to selectively cleave C–H bonds to form new C–C
roleum, with the advantage of being compatible with the bonds, which is important and challenging in increasing the
current infrastructure.5 In the past decade, bio-ethanol pro- conversion efficiency and purity. Wang et al. found that the
initial dissociation step of ethanol prefers to begin with its
O–H bond cleavage leading to CH3CH2O and H species rather
aState Key Laboratory of Coal Conversion, Institute of Coal Chemistry, Chinese
than the cleavage of other bonds.10 Davis et al. reported the
Academy of Sciences, Taiyuan, 030001 Shanxi, China. E-mail: zpzhu@sxicc.ac.cn
bGraduate University of Chinese Academy of Sciences, Beijing 100039, China
cInstitute of Application Chemistry, shanxi University, Taiyuan 030001, China
†Electronic supplementary information (ESI) available. See DOI:
10.1039/c6gc00883f
coupling reaction of ethanol into butanol over hydroxyapatite
and magnesia, and the products formed are ethane and acet-
aldehyde in addition to butanol.11 The direct transformation
of ethanol into ethyl acetate through acetaldehyde formation
This journal is © The Royal Society of Chemistry 2016
Green Chem.

View Article Online
Paper
Green Chemistry
with bimetallic Pd/Zn catalytic membrane reactors was water as the solvent to avoid the elimination of extensive water
reported. In this process, the main job was how to avoid the in bio-ethanol, which increases the availability and decreases
transformation of ethanol to CO and CH4.12 Thus, it is very the cost. For increasing the ethanol conversion rate, we investi-
difficult to effectively and selectively cleave C–H bonds of gated the influence of some catalysts on the C–C coupling reac-
ethanol because two or more bonds may be broken tion efficiency. From the experiment, we observe that the
simultaneously.
introduction of a catalyst can further accelerate ethanol conver-
2,3-BDO is a versatile platform chemical and a potential sion and enhance the formation rate of 2,3-BDO by controlling
aviation fuel with a high calorific value (27 198 J g−1), which the reaction rate. This result not only indicates the practical
can be used to produce butadiene (a monomer of synthetic possibility of directly coupling bio-ethanol into diols but also
rubber), methyl ethyl ketone (an excellent organic solvent for demonstrates how a single bond in a complex bonding circum-
resins and lacquers), acetoin, and diacetyl (a flavor enhancer) stance can be precisely manipulated.
through a dehydration or dehydrogenation reaction.13–15 The
increasing annual demand for 2,3-BDO has become a valuable
commercial opportunity.16 Generally, 2,3-BDO could be pro-
duced from carbohydrates (glucose, sucrose, glycerol etc.) by a
Results and discussion
bacterial fermentation pathway. So far, the generation rate of To validate the selectivity of H2O2 in attacking the chemical
2,3-BDO from glucose is higher than other carbon sources, bond of ethanol in aqueous solutions, the electron paramag-
which is not ideal for industrial production.17,18 Moreover, netic resonance (EPR) spin trapping technique was used. The
large-scale commercial production of 2,3-BDO using bacterial reaction of H2O2 with bio-ethanol under UV irradiation in situ
fermentation might be difficult in terms of safety regulations was investigated. The 5,5-dimethyl-l-pyrroline N-oxide (DMPO)
and the high cost of the feedstock.19–21 Thus, developing a spin trap was used in these experiments to detect radical inter-
green, clean and high efficiency process based on using low- mediates. As shown in Fig. 1A, a four-line EPR signal from an
cost and renewable resources as substrates is imperative.
apparent hydroxyl radical adduct with DMPO (DMPO/•OH,
In this work, we reported the one-step synthesis of 2,3-BDO aN = aH = 14.47 G), which was formed by the photolysis of
from bio-ethanol via the C–C coupling reaction at room temp- H2O2, could be observed. Furthermore, the other six-line EPR
erature. In this process, we found that hydroxyl radicals gener- signal was characteristic of DMPO/•CH(OH)CH3 with aN
=
ated from the photolysis of H2O2 under UV irradiation can 15.8 G and aH = 22.6 G, which was the result of the abstraction
•
selectively attack the α-hydrogen atom in ethanol aqueous of the α-hydrogen atom of ethanol by OH. Fig. 1 (curve b)
solutions and crack the α-C–H bond to produce hydroxyethyl shows the computer simulation of the experimental spectrum
radicals (•CH(OH)CH3), which subsequently undergo C–C at the third minute of illumination based on the coupling con-
•
coupling to form 2,3-BDO. At a moderate ethanol reaction rate stants given for the DMPO spin adducts of CH(OH)CH3 and
(37 mmol h−1), the 2,3-BDO selectivity can reach as high as •OH radicals. On the basis of the intensity ratio of these two
•
91%. Suzuki et al. also described the laser-induced selective adducts (9 : 1), we confirmed that CH(OH)CH3 was the major
synthesis of diols from methanol and ethanol in the presence intermediate radical for the reaction system. This result
of H2O2.22 When ethanol was irradiated, the products were suggests that the hydroxyl radical from the photolysis of H2O2
acetaldehyde and butanediol (including 2,3-, 1,3- and 1,4-butane- can precisely crack the α-C–H of ethanol aqueous solutions
diol). Acetaldehyde was the major product and the selectivity into the •CH(OH)CH3 radical.
was as high as 62%, which showed that the O–H bond scission
was the main in this process, consistent with the report of
Wang et al..10 Moreover, the selectivities of butanediol e.g. 2,3-,
1,3-, and 1,4-butanediol were 29%, 1%, and 2%, respectively.
For diols, they found that hydroxyl radicals competitively
attacked α-C–H and β-C–H of ethanol to form α-hydroxyethyl
radicals
(•CH(OH)CH3)
and
β-hydroxyethyl
radicals
(•CH2CH2OH), which subsequently dimerized to form 2,3- and
1,4-butanediol. However, the selectivity of diols was extremely
low at 29%, which demonstrates that this technology is unsui-
table for precisely manipulating single α-C–H bond breakage.
In comparison with the report of Suzuki et al., our advantages
were mainly manifested on the following two aspects: (1) selec-
tively manipulating α-C–H of ethanol: the hydroxyl radicals
can selectively break the α-C–H of ethanol to form a new C–C
bond, which can be demonstrated by EPR and the high selecti-
vity of 2,3-BDO (91%). Moreover, other butanediols (e.g. 1,4 or
1,3-butanediol) were not detected in our experiment. (2) Bio-
Fig. 1 EPR spectrum of the CH3CH2OH–H2O–H2O2–DMPO system at
the third minute of illumination. (a) Experimental spectrum; (b) spectrum
for computer simulation with two species: I, DMPO/OH; aN = aH = 14.47
ethanol instead of ethanol as the raw material: we choose G II, DMPO/CH(OH)CH3 with aN = 15.8 G and aH = 22.6 G.
Green Chem.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Green Chemistry
Paper
•
Moreover, we performed the photochemical C–C coupling large amount of OH during the early stage, which could not
reaction in a 200 mL aqueous solution under an argon atmo- only attack the α-C–H of ethanol to form •CH(OH)CH3 radicals,
sphere under UV irradiation at room temperature. The initial but could also further break down the other O–H, C–O, and
concentration of ethanol was 2.39 mol L−1, whereas that of C–C bonds of ethanol to form aldehydes, acids, or even CO2.
H2O2 was 0.25 mol L−1. 2,3-BDO was produced as a major With the increment of time, the H2O2 concentration and the
•
product, whereas acetaldehyde, acetic acid, carbon dioxide, decomposition rate decreased to produce moderate OH rad-
carbon monoxide and methane were the minor products.
icals, which preferentially manipulated the α-C–H bond of
The conversion of ethanol was 14.2%, and the selectivity of ethanol in aqueous solutions. Thus, the selectivity of 2,3-BDO
•
2,3-BDO was 84% after 5 h (Fig. 2). On the basis of these pre- improved eventually. When OH disappeared, the quantity of
liminary data, 2,3-BDO could be selectively produced through 2,3-BDO became approximately constant.
the C–C coupling of bio-ethanol. We examined the photolysis
From the preceding data, we conjectured that H2O2 concen-
of H2O2, the conversion of ethanol, and the selectivity of pro- tration might be one of the important factors that controlled
ducts as a function of reaction time. As shown in Fig. 2A, the the selectivity of 2,3-BDO. Thus, a series of comparative experi-
photolysis rate of H2O2 was extremely rapid in the beginning. ments was performed using a 2.4 mol L−1 ethanol aqueous
However, as the reaction time increased, the photolysis rate solution with different H2O2 initial concentrations under a 300
slowed down and the concentration of H2O2 decreased in the W high-pressure mercury vapor lamp for 5 h. Moreover, the
solution. The conversion of ethanol is basically consistent with utilization efficiency of H2O2 was also investigated through
the photolysis of H2O2. When the photolysis rate is fast, the comparative experiments, which can reflect how much H2O2
quantity of •OH is large, and the conversion of ethanol is high. was used for effective coupling of bio-ethanol to 2,3-BDO,
•
When H2O2 decomposed completely, the conversion of because the OH from the photolysis of H2O2 can not only
•
ethanol approached equilibrium for approximately 5 h. As crack the α-C–H bond to produce CH(OH)CH3 (eqn (1)–(3)),
shown in Fig. 2B, the selectivity of excessive oxidation pro- but can also cause a chain self-decomposition reaction with
ducts, namely, acetaldehyde and acetic acid, were high at the residual H2O2 to form O2 (eqn (4) and (5)).
initial reaction stage. By contrast, the selectivity of 2,3-BDO
•
H2O2 þ hv ! OH
ð1Þ
ð2Þ
ð3Þ
ð4Þ
ð5Þ
was low. As the reaction progressed, the selectivity of acet-
aldehyde and acetic acid evidently reduced, whereas the
selectivity of 2,3-BDO gradually increased and then leveled off.
The gas products, namely, CO, CO2, and CH4, did not evidently
change in the entire reaction process. This outcome could be
attributed to the high concentration of H2O2 that produced a
•OH þ CH3CH2OH ! CHðOHÞCH3 þ H2O
2•CHðOHÞCH3 ! CH3CHðOHÞCHðOHÞCH3
•OH þ H2O2 ! HO2• þ H2O
•
•OH þ HO•2 ! H2O þ O2
The H2O2 utilization efficiency was defined as the ratio of
the amount of H2O2 used to produce 2,3-BDO with the total
amount of H2O2 consumed in the reaction, as reported by
Wang et al.23 As the H2O2 dosage increased, the selectivity of
acetic acid improved from 3.9% to 44.3%, and the gas products
increased slightly from 1.3% to 11.8%; aldehyde did not change
evidently. However, the selectivity of 2,3-BDO decreased from
93.1% to 36.5% (Fig. 3A). Furthermore, the H2O2 utilization
efficiency was reduced from 85.5% to 19.6% with the increase
of H2O2 concentration (Fig. 3B). Assuming that extensive H2O2
was consumed during excessive oxidation or during its self-
decomposition the decrease of the utilization efficiency is
reasonable. These results are consistent with the conclusion in
Fig. 2. When the H2O2 quantity is low, the selectivity of 2,3-BDO
is high, and the H2O2 utilization efficiency is increased.
However, the conversion of ethanol is low. Thus, the means to
achieve high efficiency in both ethanol conversion and 2,3-BDO
formation must be determined.
From the preceding question, we conjectured if 2,3-BDO
formation and ethanol conversion could be improved by the
successive addition of H2O2. The successive addition method
may control the local concentration of H2O2 through the flow
rate to avoid the excessive oxidation of ethanol and the ineffec-
tive self-decomposition of H2O2. Thus, we compared the
Fig. 2 (A) The conversion of ethanol and the photolysis of H2O2 as a
function of the reaction time. (B) The selectivity of products as a func-
tion of the reaction time (conduction: ethanol 0.479 mol, H2O2 (30%)
0.05 mol, 200 ml; 300 W high pressure Hg lamp).
This journal is © The Royal Society of Chemistry 2016
Green Chem.

View Article Online
Paper
Green Chemistry
at the initial reaction stage and became nearly constant after
5 h. Compared with simultaneous addition, the successive
addition method of H2O2 improved the ethanol conversion
rate from 24.9 mmol h−1 to 37.2 mmol h−1, the selectivity of
2,3-BDO from 74.0% to 91.3%, and the utilization efficiency of
H2O2 from 46.3% to 84.9%. Moreover, for the successive
addition of H2O2, the selectivity of 2,3-BDO remained basically
constant at 91% during the entire irradiation time, and the
amounts of aldehyde and acetic acid were minimal (Fig. 4B).
The amount of 2,3-BDO formation and ethanol conversion
were significantly higher in the successive addition of H2O2
than in the simultaneous addition. These results verified that
the selective C–H manipulation of bio-ethanol could
be achieved successfully by controlling the local H2O2
concentration.
In order to further enhance the ethanol conversion rate, we
consider if it can be achieved via accelerating the decompo-
sition of H2O2. It was reported that photocatalysts can catalyze
H2O2 decomposition into •OH by accepting the conduction
band electrons, thus we investigated the influence of TiO2
photocatalysts on the ethanol coupling reaction (entries 2–5,
Table 1).24–26 We found that rutile–TiO2 can accelerate the
decomposition of H2O2 to increase the instantaneous concen-
Fig. 3 (A) Effect of the H2O2 concentration on the selectivity of pro-
ducts (B) H2O2 utilization efficiency as a function of concentration (con-
ditions: 298 K, 5 h, ethanol 0.479 mol, 200 ml; 300 W high pressure Hg
lamp).
•
tration of OH radicals, which improved the ethanol conver-
different addition methods of H2O2 through simultaneous and
successive addition. The initial concentration of ethanol was
4.8 mol L−1. As shown in Fig. 4A (curve a), the conversion of
ethanol increased linearly with the successive addition of
H2O2 at 2 mL h−1 for 5 h. During simultaneous addition
(Fig. 4A, curve b), the conversion of ethanol steeply increased
sion rate from 37.2 mmol h−1 to 40.6 mmol h−1. After the Pt
cocatalyst was loaded on rutile, the ethanol conversion rate
was further enhanced to 47.9 mmol h−1. When using the
highly active P25–TiO2 instead of rutile as the photocatalyst,
the ethanol conversion rate can reach the maximum value of
55.1 mmol h−1 and the conversion was 31.4%. However, speed-
ing up H2O2 decomposition leads to the decrease of 2,3-BDO
selectivity and increase of excessive oxide products because of
•
high OH radical instantaneous concentration (Table S1†). As
shown in Fig. S2,† the generation amount of by-products
enhanced obviously with the increase of the reaction rate. The
improvement of the ethanol conversion rate and maintaining
a constant selectivity of 2,3-BDO were contradictory, however,
in general, the formation rate of 2,3-BDO can be enhanced by
accelerating the reaction rate. For example, the formation rate
of 2,3-BDO increased from 15.3 mmol h−1 to 18.3 mmol h−1
using 0.2% Pt/rutile as the photocatalyst. Thus, controlling
and choosing a moderate reaction rate was very important to
improve the ethanol conversion rate with a high 2,3-BDO for-
mation rate.
Moreover, based on the decomposition of H2O2 catalyzed
by ceramic oxides (e.g. Al2O3 and SiO2),27 we investigate the
effect of some oxides (e.g. SiO2 nanopowders, SiO2 mesoporous
molecular sieves (SBA-15) and γ-Al2O3) on the bio-ethanol
coupling reaction (entries 6–8, Table 1). The results were
similar to the above rutile-based catalyst, both the ethanol con-
version rate and the 2,3-BDO formation rate were improved. In
the coupling reaction, γ-Al2O3 displayed the highest ethanol
conversion rate (55.4 mmol h−1) and the 2,3-BDO formation
rate (18.9 mmol h−1). Both the improvement of ethanol conver-
sion and the increase of 2,3-BDO formation rate can be rea-
lized successfully by the introduction of a catalyst. In the
Fig. 4 The effect of the H2O2 addition method on ethanol conversion
and 2,3-BDO selectivity (curve a: H2O2 was added by successive addition
at 2 ml h−1; curve b: H2O2 was added by simultaneous addition); (reac-
tion conditions: 298 K, ethanol 0.959 mol, H2O2 (30%) 0.1 mol, 200 ml;
300 W high pressure Hg lamp).
Green Chem.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Green Chemistry
Paper
Table 1 Experimental data for the photocatalytic C–C coupling of ethanola
Ethanol
conversion (%)
Conversion rateb
2,3-BDO
selectivity (%)
Formation ratec of
Entry
Catalyst
of ethanol (mmol h−1
)
2,3-BDO (mmol h−1
)
1
2
3
4
5
6
7
8
No catalyst
Rutile
0.2% Pt/rutile
1% Pt/rutile
1% Pt/P25
SiO2
19.4
22.4
24.9
26.8
31.4
25.0
27.9
32.1
37.2
40.6
44.9
47.9
55.1
42.4
49.6
55.4
91.3
86.1
81.4
61.9
36.5
75.5
72.5
68.6
15.3
17.5
18.3
14.9
10.1
16.0
18.0
18.9
SBA-15
γ-Al2O3
a Reaction conditions: 298 K, ethanol 0.959 mol, H2O2 (30%) 0.1 mol was added by successive addition at 2 ml h−1, 0.2 g of catalyst with 0.2% or
1% of Pt co-catalyst, 200 ml; a 300 W high pressure Hg lamp, 5 h of irradiation time. b The rate was calculated on the basis of the converted
ethanol. c The rate was calculated on the basis of the generated 2,3-BDO.
previous work, Yang et al. reported that sacrificial ethanol can shorter reaction time, higher 2,3-BDO selectivity and faster
be selectively oxidized to 2,3-BDO during the water splitting ethanol conversion rate. This approach is a rational strategy to
reaction using Pt/TiO2 as the photocatalyst, which discovered manipulate a single bond selectively in alcohols and to help
that the hydroxyl radicals of the TiO2 surface played an impor- explain C–C formation in natural photosynthesis.
tant role in the selective oxidation process. Through fluorine
substitution of OH groups of the Degussa P25 surface, the 2,3-
BDO selectivity enhanced from approximately 2.6% to approxi-
mately 65%.28 However, this process showed slightly low 2,3-
BDO selectivity and requires a long reaction time (24 h). In
Experimental section
All chemical reagents were analytical grade reagents and used
without further purification. SiO2 nanopowders (15 nm,
99.5%) and γ-Al2O3 (40–60 mesh) were purchased from
Aladdin. Particle surface areas were determined on a Quanta-
chrome Autosorb 1 analyzer using nitrogen adsorption and de-
sorption at equilibrium vapor pressure and the Brunauer–
Emmet–Teller (BET) method of surface area calculation.
Specific surface areas of SiO2 nanopowders, SiO2 mesoporous
molecular sieves (SBA-15) and γ-Al2O3 particles were 124.3,
•
this work, we utilized directly the OH radical from the photo-
lysis H2O2 to selectively oxidize bio-ethanol to 2,3-BDO, which
accelerated the ethanol conversion rate to shorten the reaction
time with a relatively high 2,3-BDO selectivity. Moreover, our
strategy displayed higher controllability for aliphatic C–H
breakage than that reported by Suzuki et al.. In a word, this
result does not only provide an environment-friendly approach
to convert bio-ethanol to 2,3-BDO but also provide a promising
channel for selectively manipulating inert aliphatic C–H
bonds.
712.0 and 209.6 m2 −1, respectively. The adsorption average
g
pore diameter of SiO2 nanopowders, SBA-15 and γ-Al2O3 par-
ticles were 11.2, 5.52 and 8.24 nm, respectively.
Rutile–TiO2 was synthesized by a calcination method with
commercial TiO2 powder in a muffle furnace in air at 800 °C
for 8 h. SBA-15 was synthesized according to the method
Conclusions
We determined that the α-C–H bond of bio-ethanol could be reported elsewhere.29
selectively attacked to achieve a direct C–C coupling synthesis
Photochemical experiments were conducted in a 250 mL
of 2,3-butanediol with hydroxide radicals from the photolysis inner irradiation-type Pyrex reactor. A 300 W high-pressure
of H2O2 at room temperature. This selective C–H breakage was mercury vapor lamp was used as the incandescent light source
determined by the reaction rate, which was primarily con- and was cooled by water to maintain the reactions at room
trolled by the local H2O2 concentration at a given irradiation temperature. All the photochemical tests were performed in a
intensity. The low local H2O2 concentration was favorable for 200 mL aqueous solution of reactants and were magnetically
selectively cleaving the α-C–H bond of alcohols. At a moderate stirred to achieve full dispersion, with pure Ar continuously
reaction rate of ethanol (37 mmol h−1), the 2,3-butanediol bubbling. Before the photochemical reaction, Ar was purged
selectivity could reach 91%. The introduction of a catalyst into the reactor for 30 min to remove the air thoroughly. The
can further increase the ethanol conversion rate from reaction course was monitored by periodically sampling the
37.2 mmol h−1 to 55.4 mmol h−1 and enhance the formation liquid and gas from a sampling valve. Liquid samples were
rate of 2,3-BDO from 15.3 mmol h−1 to 18.9 mmol h−1 via analyzed using
a gas chromatograph (Haixin; GC-950)
accelerating H2O2 decomposition. In comparison with the pre- equipped with a flame ionization detector (FID). Gaseous
vious synthetic method of 2,3-BDO, our strategy displays a samples were analyzed using a gas chromatograph (Fuli;
This journal is © The Royal Society of Chemistry 2016
Green Chem.

View Article Online
Paper
Green Chemistry
GC-9790) equipped with a thermal conductivity detector
(TCD), a FID. Quantities of the products and reactants were
calculated from the peak areas of the compounds using cali-
bration curves.
9 J. F. DeWilde, C. J. Czopinski and A. Bhan, ACS Catal.,
2014, 4, 4425–4433; C. Bianchini and P. K. Shen, Chem.
Rev., 2009, 109, 4183–4206; T. K. Phung and G. Busca,
Chem. Eng. J., 2015, 272, 92–101; T. K. Phung and G. Busca,
Catal. Commun., 2015, 68, 110–115.
All the photocatalytic tests were carried out using 0.2 g of
catalyst and 200 mL of 0.959 mol ethanol aqueous solution 10 R. G. Zhang, Z. X. Liu, L. X. Ling and B. J. Wang, Appl. Surf.
and H2O2 (30%) (0.1 mol) was added by using a peristaltic Sci., 2015, 353, 150–157.
pump at 2 ml h−1 over 5 h followed by an appropriate amount 11 S. Hanspal, Z. D. Young, H. Shou and R. J. Davis, ACS
(0.2% and 1% Pt) of chloroplatinic acid (H2PtCl6·6H2O). Before Catal., 2015, 5, 1737–1746.
the photocatalyst reaction, Ar was purged into the reactor for 12 A. B. Śanchez, N. Homs, S. Miachon, J. A. Dalmon,
30 min to remove the air thoroughly. The other reaction course
was similar to the above photochemical experiment.
J. G. Fierrodand and P. R. Piscina, Green Chem., 2011, 13,
2569–2575.
EPR spectra were recorded using a Bruker EMXplus-10/12. 13 M. Kopke, C. Mihalcea, F. Liew, J. H. Tizard, M. S. Ali,
The typical parameters were as follows: sweep width = 100 G,
EPR microwave power = 10 mW, modulation amplitude =
J. J. Conolly and S. D. Simpson, Appl. Environ. Microbiol.,
2011, 77, 5467–5475.
0.8 G, time constant = 40.96 ms, and sweep time = 167.77 s. 14 X. J. Ji, H. Huang and P. K. Ouyang, Biotechnol. Adv., 2011,
The values of the g-tensor were calculated using Bruker’s 29, 351–364.
WIN-EPR SimFonia 2.3 program, which allowed the control of 15 J. V. Haveren, E. L. Scott and J. Sanders, Biofuels, Bioprod.
the Bruker EPR spectrometer, data acquisition, automation Biorefin., 2008, 2, 41–57.
routines, tuning, and calibration programs on a PC with a 16 S. Hrivastav, J. Lee, H. Y. Kim and Y. R. Kim, J. Microbiol.
Windows operating system. The exact g-values of the key Biotechnol., 2013, 23, 885–896.
spectra were determined by comparing with the 2,2-diphenyl- 17 C. F. Huang, Y. F. Jiang, G. L. Guo and W. S. Hwang, Bio-
1-picrylhydrazyl (DPPH) standard.
resour. Technol., 2013, 135, 446–453.
18 T. W. Yang, Z. M. Rao, X. Zhang, M. J. Xu, Z. H. Xu and
S. T. Yang, Appl. Microbiol. Biotechnol., 2013, 97, 7651–7658.
19 S. Hrivastav, J. Lee, H. Y. Kim and Y. R. Kim, J. Microbiol.
Biotechnol., 2013, 23, 885–896.
Acknowledgements
This research was financially supported by the National 20 S. Kim, S. O. Seo, Y. C. Park, Y. S. Jin and J. H. Seo, J. Bio-
technol., 2014, 192, 376–382.
U1510108) and the Knowledge Innovation Project of the 21 X. X. Wang, H. Y. Hu, D. H. Liu and Y. Q. Song, New Bio-
Natural Science Foundation of China (21173250 and
Chinese Academy of Sciences (KGCX2-EW-311).
technol., 2016, 1, 16–22.
22 Y. Shimizu, S. Sugimoto, S. Kawanishi and N. Suzuki, Bull.
Chem. Soc. Jpn., 1991, 64, 3607–3612.
23 W. G. Li, Y. Wang and A. Irinic, Chem. Eng. J., 2014, 244,
1–8.
Notes and references
24 J. G. Jia, T. Ohno and M. Matsumura, Chem. Lett., 2000,
908–909.
1 Y. Y. Ma, Z. T. Du, J. X. Liu, F. Xia and J. Xu, Green Chem.,
2015, 17, 4968–4973.
25 B. L. Tang, G. H. Jiang, Z. Wei, X. Li, X. H. Wang,
T. T. Jiang, W. X. Chen and J. M. Wan, Acta Metall. Sin.,
2014, 27, 124–130.
26 C. F. Yu, S. Y. Dong, J. L. Feng, J. Y. Sun, L. M. Hu, Y. K. Li
and J. H. Sun, Environ. Sci. Pollut. Res., 2014, 21, 2837–
2845.
2 Y. Yan, P. Feng, Q. Z. Zheng, Y. F. Liang, J. F. Lu, Y. Cui and
N. Jiao, Angew. Chem., Int. Ed., 2013, 52, 5827–5831.
3 T. Patra, R. Watile, S. Agasti, T. Naveen and D. Maiti, Chem.
Commun., 2016, 52, 2027–2030.
4 E. M. Simmons and J. F. Hartwig, Nat. Lett., 2012, 483, 70–73.
5 J. R. Hansen, C. H. Christensen, J. Sehested, S. Helveg,
J. R. R. Nielsen and S. Dahl, Green Chem., 2007, 9, 1016–
1021.
27 A. Hiroki and J. A. LaVerne, J. Phys. Chem. B, 2005, 109,
3364–3370.
28 P. J. Yang, J. H. Zhao, B. Y. Cao, L. Li, Z. J. Wang,
X. X. Tian, S. P. Jia and Z. P. Zhu, ChemCatChem, 2015, 7,
2384–2390.
29 L. N. Zhang, H. Wang, Z. F. Qin, J. G. Wang and W. B. Fan,
RSC Adv., 2015, 5, 22838–22846.
6 D. P. Ho, H. H. Ngo and W. Guo, Bioresour. Technol., 2014,
169, 742–749.
7 M. Samei, T. Mohammadi and A. A. Asadi, Chem. Eng. Res.
Des., 2013, 91, 2703–2712.
8 M. R. Ladisch and K. Dyck, Science, 1997, 205, 898–900.
Green Chem.
This journal is © The Royal Society of Chemistry 2016