Selective production of 1,3-butadiene using glucose fermentation liquor

Article abstract of DOI:10.1039/c4gc00485j

The production of 2,3-butanediol from glucose fermentation products and its subsequent esterification and conversion to 1,3-butadiene is reported. The addition of formic acid and acetic acid (C1-C2 acids) to the esterification reaction mixture resulted in yields of the diesters of 70% and 85%, without the loss of C1-C2 acids during the reaction. In the pyrolysis step, a highly selective (94% and 82% for formic acid 2-formyloxy-1-methyl-propyl ester and acetic acid 2-acetoxy-1-methyl-propyl ester, respectively) C-O cleavage to 1,3-butadiene over diesters was achieved without a catalyst. In the case of acetic acid, 100% was recovered, whereas in the case of formic acid, only 20% was recovered. Based on these results, it can be concluded that using glucose fermentation liquor as the starting material with external addition of acetic acid (2,3-butanediol:formic acid:acetic acid = 1:0.5:2.5), 70% yield of 1,3-butadiene can be achieved where the loss of formic acid is compensated by the acid in the starting material. This journal is the Partner Organisations 2014.

Full text of DOI:10.1039/c4gc00485j

Green Chemistry
View Article Online
View Journal | View Issue
PAPER
Selective production of 1,3-butadiene using
glucose fermentation liquor†
Cite this: Green Chem., 2014, 16,
3501
Jayeon Baek,‡a Tae Yong Kim,‡a Wooyoung Kim,b Hee Jong Leeb and
Jongheop Yi*a
The production of 2,3-butanediol from glucose fermentation products and its subsequent esterification
and conversion to 1,3-butadiene is reported. The addition of formic acid and acetic acid (C1–C2 acids) to
the esterification reaction mixture resulted in yields of the diesters of 70% and 85%, without the loss of
C1–C2 acids during the reaction. In the pyrolysis step, a highly selective (94% and 82% for formic acid
2-formyloxy-1-methyl-propyl ester and acetic acid 2-acetoxy-1-methyl-propyl ester, respectively) C–O
cleavage to 1,3-butadiene over diesters was achieved without a catalyst. In the case of acetic acid, 100%
was recovered, whereas in the case of formic acid, only 20% was recovered. Based on these results, it can
be concluded that using glucose fermentation liquor as the starting material with external addition of
acetic acid (2,3-butanediol : formic acid : acetic acid = 1 : 0.5 : 2.5), 70% yield of 1,3-butadiene can be
achieved where the loss of formic acid is compensated by the acid in the starting material.
Received 20th March 2014,
Accepted 22nd April 2014
DOI: 10.1039/c4gc00485j
www.rsc.org/greenchem
used in the production of synthetic rubber, plasticizers,
and fuel additives. Therefore, the development of a suitable
Introduction
The worldwide lack of petroleum has stimulated a search for technology for the conversion of 2,3-butanediol into other
alternative sources of clean energy. The use of biomass for this value-added products such as 1,3-butadiene would be highly
purpose, turning ‘consumptive use’ into ‘recyclable use’, is an desirable.4,5 1,3-Butadiene is currently produced by the steam
ideal goal for the future. The use of feedstock derived from cracking of paraffinic hydrocarbons, the catalytic dehydrogena-
biomass, e.g. sugars, vegetable oil, lignocellulosic biomass, tion of n-butane and n-butene, and the oxidative dehydrogena-
and algae, emits the greenhouse gas, carbon dioxide, but it tion of n-butene, all of which are based on a petrochemical
can be reused in the photosynthetic process and could result process.6
in the creation of a carbon-neutral and sustainable society.1
Herein, we report a process for the production of 1,3-buta-
Glucose, a monosaccharide produced by plants, could be diene from the mixture of 2,3-butanediol and C1–C2 acids
efficiently utilized after a fermentation process as the carbon which are representative products in glucose fermentation.
source for microorganisms.2 Glucose can be converted into Importantly, this process can replace some petrochemical-
2,3-butanediol by fermentation in which the major byproducts based processes. We proved that esterification would occur
are formic acid and acetic acid (C1–C2 acids), although the between the 2,3-butanediol and C1–C2 acids to form diesters
proportion of these byproducts can vary depending on the (products 4 and 7). In addition, the C–O cleavage in the
conditions of the fermentation.3 The use of 2,3-butanediol diester (products 4 and 7) sequentially occurs during the pyro-
is expected to steadily increase because its derivatives can be lysis process, which results in the selective formation of 1,3-
butadiene (Scheme 1). A similar procedure was reported in
1945 by S. Marshak and co-workers, however the experimental
details concerning it are not available.7
aWorld Class University (WCU) Program of Chemical Convergence for Energy &
To the best of our knowledge, the present work is the first
Environment (C2E2), Institute of Chemical Processes, School of Chemical and
report of esterification followed by pyrolysis for the formation
Biological Engineering, Seoul National University, Seoul 151-742, Republic of Korea.
of 1,3-butadiene by reacting 2,3-butanediol with products pro-
duced by the fermentation of glucose. This process would
result in diminished cost and energy which are consumed in
the pre-separation of reactants including 2,3-butanediol and
C1–C2 acids. In addition, 1,3-butadiene (gas phase) can be
easily isolated from the glucose fermentation products.
E-mail: jyi@snu.ac.kr
bChemical & Polymer Lab. R&D Center, GS-Caltex Corporation, Daejeon 305-380,
Republic of Korea
†Electronic supplementary information (ESI) available: 1H NMR & 13C NMR
spectra, photographs of esterification products, and GC chromatograms are
provided. See DOI: 10.1039/c4gc00485j
‡These authors contributed equally to this work.
This journal is © The Royal Society of Chemistry 2014
Green Chem., 2014, 16, 3501–3507 | 3501

View Article Online
Paper
Green Chemistry
C1–C2 acids is 1 : 2, as shown in Scheme 1. However, taking
the relatively low boiling point of formic acid (100.8 °C) and
acetic acid (118 °C) into account, excess C1–C2 acids were used
in the esterification; 2,3-butanediol (1) : formic acid (2) = 1 : 4
and 2,3-butanediol (1) : acetic acid (5) = 1 : 2.5 molar ratio. To
force the equilibrium to the right, a Dean–Stark trap was used
to remove water from the reaction mixture.
Esterification of 2,3-butanediol with formic acid
The esterification between 2,3-butanediol (1) and formic acid
(2) occurs sequentially, in which the formic acid 2-hydroxy-1-
methyl-propyl ester (3) is formed first and another 2,3-butane-
diol (1) reacts with the product 3 to produce formic acid 2-for-
myloxy-1-methyl-propyl ester (4) (Scheme 2). This reaction
process can be confirmed by time-on-stream reaction exper-
iments (Fig. 1) as the continuous decrease in formic acid
2-hydroxy-1-methyl-propyl ester (3) is correlated with the gradual
increase of 2,3-butanediol (1) conversion and formic acid
Scheme 1 Conversion of biomass-derived 2,3-butanediol and car-
boxylic acid to 1,3-butadiene.
Results and discussion
The composition of glucose fermentation liquor fermented
using Klebsiella oxytoca KCTC12133BP (Korean Collection for
Type Cultures, Daejeon, Korea) is presented in Table 1. The
microbial fermentation experiment and the analysis were con-
ducted by GS-Caltex Corporation. As can be seen from Table 1,
the glucose fermentation liquor is comprised of 2,3-butanediol
as the main product, followed by formic acid, acetic acid, lactic
acid, succinic acid (C1–C4 acids), and acetoin, along with
some inorganic salts. The molar ratio of 2,3-butanediol,
formic acid, and acetic acid is typically 1 : 0.5 : 0.1 in the liquor
where the amount of 2,3-butanediol is dominant. It should be
noted that the isomer ratio between meso and racemic 2,3-
butanediol used for the esterification coincide with that
of the glucose fermentation product (meso : racemic = 9 : 1,
Fig. S1–S7, ESI†).
Scheme 2 Esterification of 2,3-butanediol (1) with formic acid (2) using
sulfuric acid.
The esterification process was evaluated as the first step
using sulfuric acid as a catalyst. After the reaction, the sulfuric
acid can be recovered in the course of the extraction procedure.
The stoichiometric molar ratio between 2,3-butanediol and
Table 1 Composition of microbial fermented liquora
(Unit: g kg−1
)
2,3-
Butanediol
Formic Acetic
acid acid
Lactic Succinic
Acetoin acid
acid
Ethanol
0.00
480.13
122.64 13.91 5.08
34.10 36.93
(Unit: ppm (w))
Cl−
PO4
SO4
Na+
n.a.
NH4
n.a.
K+
Mg2+
n.a.
Ca2+
n.a.
3−
2−
+
Fig. 1 Time-on-stream of esterification of 2,3-butanediol with formic
acid. Reaction conditions: 1 (0.22 mol), 2 (0.88 mol) and 0.2 mL of
H2SO4. The temperature of the reaction mixture gradually increased
during the course of the run from about 110 °C to 130 °C. The conver-
sion of 2,3-butanediol and selectivity was determined by GC-FID.
2850
70 962
48 760
n.a.
a See the reference for the detailed information concerning
the microbial fermentation of glucose using Klebsiella oxytoca
KCTC12133BP (Korean Collection for Type Cultures, Daejeon, Korea).8
3502 | Green Chem., 2014, 16, 3501–3507
This journal is © The Royal Society of Chemistry 2014

View Article Online
Green Chemistry
Paper
2-formyloxy-1-methyl-propyl ester (4) formation until 7 h of
reaction time. The desired product formic acid 2-formyloxy-
1-methyl-propyl ester (4) for 1,3-butadiene was obtained with 70%
selectivity as well as formic acid 2-hydroxy-1-methyl-propyl
ester (3) with 17% selectivity as an intermediate in 5 h at
130 °C (Fig. 1). The calculated m/z for the formic acid 2-formyl-
oxy-1-methyl-propyl ester (4) was 147.0657 and measured as
147.0659 and the structure was additionally confirmed by 1H
NMR and 13C NMR (Fig. S8 and S9, ESI†). As the reaction pro-
ceeded, the color of the reaction mixture changed from light
yellow to light brown (Fig. S12, ESI†). This color change can be
attributed to the increased production of formic acid 2-formyl-
oxy-1-methyl-propyl ester (4). While the composition of formic
acid 2-hydroxy-1-methyl-propyl ester (3) and unknown pro-
ducts changed during the course of the reaction of 5 h to 7 h,
the yield of the targeted formic acid 2-formyloxy-1-methyl-
propyl ester (4) was maintained at a constant level of 70%,
indicating that the formation and degradation rates of formic
acid 2-formyloxy-1-methyl-propyl ester (4) are equal. After 7 h
of reaction time, however, the yield of other compounds orig-
inating from 2-formyloxy-1-methyl-propyl ester (4) rapidly
increased as the decreased amount of the product 4 was trans-
formed into other products. Hence, the optimum time for the
esterification between 2,3-butanediol and formic acid was 5 h
at 130 °C. After the reaction, the loss of excess formic acid was
calculated to be 2%. For the subsequent experimental pyrolysis
process, we used an esterification mixture that was produced
after a reaction time of 5 h.
Fig. 2 Time-on-stream of esterification of 2,3-butanediol with acetic
acid. Reaction conditions: 1 (0.22 mol), 5 (0.55 mol) and 0.2 mL of
H2SO4. The temperature of the reaction mixture gradually increased
during the course of the run from about 110 °C to 140 °C. The conver-
sion of 2,3-butanediol and selectivity was determined by GC-FID.
enhanced at the expense of acetic acid 2-hydroxy-1-methyl-
propyl ester (6) as the reaction proceeded. After 10 h, the reac-
tion mixture turned dark brown which can be attributed to the
production of acetic acid 2-acetoxy-1-methyl-propyl ester (7)
(Fig. S13, ESI†). Unlike the esterification of 2,3-butanediol with
formic acid, the formed acetic acid 2-acetoxy-1-methyl-propyl
ester (7) did not degrade into other products. The esterification
of 2,3-butanediol with acetic acid required an extended reac-
tion time to achieve the maximum production of acetic acid
2-acetoxy-1-methyl-propyl ester (7) compared to the esterification
of 2,3-butanediol with formic acid; this can be attributed to
the steric effect produced by the longer alkyl chain.9 It should
be noted, however, that the selectivity at the equilibrium state
for acetic acid 2-acetoxy-1-methyl-propyl ester (7) (selectivity:
85%) is higher than formic acid 2-formyloxy-1-methyl-propyl
ester (4) (selectivity: 70%), indicating that the selectivity in the
reaction is influenced more by thermodynamics than the
steric hindrance caused by the length of the alkyl chain. In
this experiment, no loss of acetic acid was observed. The esteri-
fication mixture produced after a reaction time of 10 h was
used in the subsequent pyrolysis step.
Esterification of 2,3-butanediol with acetic acid
We further investigated the esterification of 2,3-butanediol (1)
with acetic acid (5) (Scheme 3). The esterification resulted in
the formation of the corresponding acetic acid 2-hydroxy-1-
methyl-propyl ester (6) and acetic acid 2-acetoxy-1-methyl-
propyl ester (7); the chemical structures of which were con-
firmed from 1H-NMR and 13C NMR data (Fig. S10 and S11,
ESI†). Also, the m/z estimated for the acetic acid 2-acetoxy-1-
methyl-propyl ester (7) was 175.0970 and the measured value
was 175.0971. An evaluation of esterification using 2,3-butane-
diol with acetic acid time-on-stream (Fig. 2) indicates that the
optimum time required for the reaction is 10 h where the
selectivity of acetic acid 2-acetoxy-1-methyl-propyl ester (7) is
Pyrolysis of products 4 and 7
Considering the proposed mechanism shown in Scheme 1, the
pyrolysis of formic acid 2-formyloxy-1-methyl-propyl ester (4)
and acetic acid 2-acetoxy-1-methyl-propyl ester (7) would
sequentially occur following the esterification reaction to
produce 1,3-butadiene. In this step, the pyrolysis of formic
acid 2-formyloxy-1-methyl-propyl ester (4) and acetic acid
2-acetoxy-1-methyl-propyl ester (7) would proceed in the absence
of a catalyst at a relatively high temperature.
Scheme 3 Esterification of 2,3-butanediol (1) with acetic acid (5) using
sulfuric acid.
This journal is © The Royal Society of Chemistry 2014
Green Chem., 2014, 16, 3501–3507 | 3503

View Article Online
Paper
Green Chemistry
Fig. 3 Electrostatic potential map for (a) formic acid 2-hydroxy-1-
methyl-propyl ester (3) and (b) formic acid 2-formyloxy-1-methyl-
propyl ester (4).
Electrostatic potential maps for the formic acid 2-hydroxy-1-
methyl-propyl ester (3) and formic acid 2-formyloxy-1-methyl-
propyl ester (4) show that the electron densities are highly loca-
lized on ester groups, which can function as a good leaving
group because the anion produced in the reaction is stable
(Fig. 3, we only discuss products 3 and 4 as a similar expla-
nation can be valid for products 6 and 7). In the case of
product 3, since the –OH in 2,3-butanediol is a much poorer
leaving group, it is typically aided by protonation; the electro-
philic ester functional group leaves first via cleavage of the C–O
bond to give a secondary carbocation.10 Consistent with the
E1 mechanism in organic chemistry, the ester anion would
attack the β-hydrogen in the carbocation intermediate with the
regeneration of formic acid and the production of a carbon–
carbon double bond in the carbocation intermediate.11 This
step is repeated with another ester functional group in product
4 with 1,3-butadiene being produced as a final product. When
the elimination of the ester functional group occurs in the
mono-ester (products 3 and 6), 3-buten-2-ol and methyl ethyl
ketone are readily produced. The formation of the carbocation
intermediate crosses the high energy barrier which is the rate-
determining step in the reaction, and, because of this, a high
temperature is required for the pyrolysis of formic acid 2-formyl-
oxy-1-methyl-propyl ester (4) and acetic acid 2-acetoxy-1-
methyl-propyl ester (7).11
The reaction temperature for the pyrolysis was optimized in
the range of 400–600 °C (Fig. 4a). Prior to pyrolysis, the yield of
1,3-butadiene was low due to the slightly soluble nature of 1,3-
butadiene in the methanol.12 The constant yield of 1,3-buta-
diene could be obtained after 5 h. When the reaction was con-
ducted at 400 °C, the conversion of acetic acid 2-acetoxy-1-
methyl-propyl ester (7) was lowered. An increase in the reaction
temperature to 600 °C, however, promoted the decomposition
of the products, resulting in the formation of C1–
C3 hydrocarbons (methane, ethylene, and propylene). Based
on the above results, the reaction temperature for the pyrolysis
was set at 500 °C for the conversion of all the esterification
products with high selectivity for 1,3-butadiene. In the pyro-
Fig. 4 (a) Pyrolysis results for product 7 at 400 °C, 500 °C, and 600 °C;
(b) Pyrolysis result of products 4 and 7 after 5 h at 500 °C in the absence
of a catalyst. Reaction conditions: 0.3 mL h−1 of feed flow rate, 30 mL
min−1 of N2, LHSV = 0.028 h−1. The final liquid products were trapped in
methanol. aYield (%) = produced 1,3-butadiene (mol) compared to the
introduced products 4 (mol) and 7 (mol).
Fig. 3. It is noteworthy that coke formation after the pyrolysis
step is negligible (0.5% for the pyrolysis of product 7). In
addition, an analysis of the output gas phase stream shows
that highly pure 1,3-butadiene is produced. The high purity of
1,3-butadiene (the peak at 51 min in the GC chromatogram)
above 93% for both the pyrolysis of formic acid 2-formyloxy-1-
methyl-propyl ester (4) and acetic acid 2-acetoxy-1-methyl-
propyl ester (7) was analyzed by on-line GC-FID (Fig. S14 and
S15, ESI†). The small peak in the GC chromatogram at 27 min
corresponds to trans-2-butene which is the main side reaction
product in the gas phase along with the minor decomposition
to C1–C3 hydrocarbons (methane, ethylene, and propylene).
Building upon these results, this approach would be poten-
tially useful for extracting pure 1,3-butadiene in the form of a
gas phase from the glucose-derived fermentation products.
Sustainable production of 1,3-butadiene from glucose
fermentation liquor
lysis of formic acid 2-formyloxy-1-methyl-propyl ester (4), the The attractive point of this process is that the original C1–C2
selectivity was about 94% whereas it was 82% for acetic acid acids in the final liquid products can be recovered. If a recy-
2-acetoxy-1-methyl-propyl ester (7) (Fig. 4b). The high selectivity cling process were possible, the C1–C2 acids produced in the
for 1,3-butadiene in the pyrolysis can be attributed to the selec- reaction could be transferred to the esterification process,
tive elimination of the ester functional group as explained in which would allow the C1–C2 acids to accumulate where exces-
3504 | Green Chem., 2014, 16, 3501–3507
This journal is © The Royal Society of Chemistry 2014

View Article Online
Green Chemistry
Paper
sive amounts are needed to adjust the stoichiometric molar acids after pyrolysis was also quantified. Interestingly, the
ratio. Given the fact that the C1–C2 acids are collected at the recovery of acetic acid was 100% whereas the formic acid
final stage, they could be trapped in methanol and the results shows a much lower recovery rate (20%) which could be
were as expected. In the GC-MS chromatogram of the trapped related to the low stability of formic acid at high tempera-
liquid phase, formic acid and acetic acid were regenerated tures.13 Some unknown products were present in the resulting
with methyl ethyl ketone (Fig. 5). The recovery rate of C1–C2 liquid phase after the pyrolysis of the esterification product
using 2,3-butanediol with acetic acid. This is in agreement
with the observed decreased selectivity for 1,3-butadiene in the
output gas phase stream under these conditions (Fig. 4b).
Based on these results, we attempted to apply the model
mixture to the esterification followed by pyrolysis reaction. The
model mixture proposed is comprised of 2,3-butanediol :
formic acid : acetic acid = 1 : 0.5 : 2.5 molar ratio, considering
the recovery rate of C1–C2 acids in the two sequential reactions
and low boiling point of C1–C2 acids. This model mixture pro-
portion allows the process to be recycled if an additional
2.4 mol of acetic acid is added to the glucose fermentation
liquor (2,3-butanediol : formic acid : acetic acid = 1 : 0.5 : 0.1).
The reaction temperature was increased to 140 °C in the light
of the dominant composition of acetic acid in the model
mixture. In Table 2, conversion of 2,3-butanediol reached
100% after 3 h which can be attributed to enough amounts of
C1–C2 acids for the esterification reaction step. In accordance
with the previous esterification results, the composition of
mono-esters (products 3 and 6) decreased as the reaction pro-
ceeded by self-destruction to the diesters (products 4 and 7).
The formed products 4 and 7 are also confirmed. Interestingly,
an entirely new peak in the GC-FID chromatogram was gene-
rated and it is confirmed as acetic acid 2-formyloxy-1-methyl-
propyl ester (8) where each formic acid and acetic acid
comprises two ester functional groups to one 2,3-butanediol
molecule (1H NMR and 13C NMR, Fig. S16 and S17, ESI†). The
product 8 also has the capability of producing 1,3-butadiene
with a high selectivity of 92% (data not shown here). It should
be noted that the yield of diesters (products 4, 7, and 8)
remained essentially unchanged after a reaction time of 10 h.
Thus, the optimum reaction time is considered to be 10 h for
Fig. 5 GC-MS chromatogram of the liquid products obtained after the
pyrolysis of (a) products 4 and (b) 7 (500 °C, 5 h, the final liquid products
the following pyrolysis. Concerning the origin of 1,3-buta-
were trapped in methanol).
Table 2 Esterification of the model mixture of glucose fermentation liquora
Selectivity in organic phaseb (%)
Entry
Temperature (°C)
Time (h)
Conversionb (%)
3
4
6
7
8
Others
1
2
3
4
140
140
140
140
3
7
10
12
99
100
100
100
3
2
1
1
2
1
1
1
17
11
7
54
60
66
68
18
18
16
15
6
8
9
8
7
a Reaction conditions: 1 (0.22 mol), 2 (0.11 mol), 5 (0.55 mol) and 0.2 mL of H2SO4. The temperature of the reaction mixture gradually increased
during the course of the run from about 110 °C to 140 °C. b The conversion and selectivity was determined by GC-FID.
This journal is © The Royal Society of Chemistry 2014
Green Chem., 2014, 16, 3501–3507 | 3505

View Article Online
Paper
Green Chemistry
fitted with a Dean–Stark trap and a condenser. The solution
was stirred and the reaction temperature was gradually
increased from 110 °C to 130 °C and 110 °C to 140 °C for the
esterification of 2,3-butanediol with formic acid and acetic
acid, respectively (direct increasing of the reaction temperature
results in a higher yield of methyl ethyl ketone and a loss of
C1–C2 acids used for the esterification). The resulting mixture
was then extracted with dichloromethane and the organic layer
was dried over anhydrous esterification. The resulting mixture
was then extracted with dichloromethane and the organic layer
was dried over anhydrous magnesium sulfate, filtered, and
flash evaporated. The liquid products were analyzed by GC
using a Younglin YL 6100 gas chromatograph equipped with a
FID. The amount of the loss of C1–C2 acids during the esterifi-
cation was analyzed by HPLC using a Younglin YL 9100.
Table 3 1,3-Butadiene production from the model mixture of glucose
fermentation liquora
Temperature (°C)
500
Time (h)
5
Conversion (%)
100
Yieldb (%)
70
a Reaction conditions: the feed (Table 2, entry 3) flow rate was set at
0.3 mL h−1 with 30 mL min−1 of N2 gas. LHSV = 0.028 h−1 b Yield (%)
.
= produced 1,3-butadiene relative to the initial 2,3-butanediol.
diene, one can predict that the yield of 1,3-butadiene from the
glucose fermentation is maximized at 83%. Finally, 70% of
1,3-butadiene was obtained via the pyrolysis reaction after 5 h
(Table 3). This is probably the highest yield ever reported.14
Along with the obtained high yield of 1,3-butadiene from
glucose fermentation liquor, the recycling of recovered C1–C2
acids after pyrolysis as reactants has considerable merit from
the environmental and economic point of view.
General procedure for the pyrolysis process
The pyrolysis in the absence of a catalyst was carried out in a
flow-type quartz reactor, which was then placed in an electric
furnace. The temperature was monitored by means of a K-type
thermocouple controlled by a PID controller in the range of
400–600 °C. The feed was preheated to 250 °C in a stream of
dry N2 (30 cm3 min−1). The feed consisted of the organic
phase and the flow rate was set at 0.3 mL h−1. The liquid
hourly space velocity (LHSV) was calculated from the volu-
metric feed rate and the volume of the reactor. The liquid
phase after pyrolysis was trapped in 30 mL of methanol cooled
at −15 °C and analyzed using an Agilent 7890 gas chromato-
graph coupled with a model 5975 mass spectrometer. The
output gas phase stream was analyzed using an online gas
chromatograph Donam DS 6200. For the quantification of the
loss of C1–C2 acids during the pyrolysis, D. I. water was used
for trapping the produced C1–C2 acids and this was analyzed
by HPLC using a Younglin YL 9100. The coke formation was
analysed using a CHNS-932 elemental analyser with SiO2
(Degussa, Aerosil 200) as an inert material.
Conclusions
An integrated route consisting of esterification followed by
pyrolysis for the conversion of glucose fermentation products
into 1,3-butadiene is demonstrated. This biomass-based
process appears to be quite efficient, sustainable, and reusable
compared with typical petrochemical processes. The catalytic
esterification of 2,3-butanediol with the excess C1–C2 acids
that are produced from glucose fermentation was achieved,
with 70% and 85% selectivity for the diesters (products 4 and
7, respectively) in an organic phase at 100% conversion of 2,3-
butanediol. The subsequent elimination of the ester functional
group in the pyrolysis process enables C–O bond cleavage to
occur, via the E1 mechanism resulting in the formation of 1,3-
butadiene and the regeneration of the C1–C2 acids. Further-
Characterization of products
The new compounds were characterized by H NMR, 13C NMR
1
and high resolution mass spectrometry (HRMS). 1H and 13C
NMR spectra were recorded on a Bruker Avance 500 MHz NMR
spectrometer. Chemical shifts (δ) are quoted in parts per
more, this combined process proceeded when
a model
mixture of glucose fermentation products was used. The
present methodology provides an example of producing a
diene product (1,3-butadiene) from a diol compound (2,3-butane-
diol) from biomass products via fermentation. Finally, as
the controlled C–O cleavage is a very general issue in the con-
version of biomass-derived substrates, the procedure reported
herein would be expected to have broad applicability for the
production of dienes from the corresponding diol compounds.
1
million relative to acetone-d6 2.05 ppm for H and 29.92 ppm
for 13C as the internal standards. HRMS was conducted with a
JEOL JMS-700 instrument and accurate masses were reported
for the molecular ion (M+).
Formic acid 2-hydroxy-1-methyl-propyl ester (3). 1H NMR
(500 MHz, acetone-d6): δ = 8.116 (d, J = 11.5 Hz, 1H),
4.854–4.807 (m, 1H), 3.843 (d, J = 31 Hz, 1H), 1.267–1.190 (m,
3H), 1.116 (d, J = 6.5 Hz, 3H); 13C NMR (125 MHz, acetone-d6):
161.78, 161.39, 75.02, 69.38, 19.01, 18.87, 15.48, 15.23.
Formic acid 2-formyloxy-1-methyl-propyl ester (4). 1H NMR
(500 MHz, acetone-d6): δ = 8.116 (d, J = 11.5 Hz, 2H),
Experimental
General procedure for the esterification process
A mixture of 2,3-butanediol with formic acid or acetic acid and 5.124–5.059 (m, 2H), 1.267–1.190 (m, 6H); 13C NMR (125 MHz,
0.2 mL of sulfuric acid was added into a round-bottomed flask acetone-d6): 161.78, 161.39, 71.62, 71.57, 16.39, 16.09. HRMS
3506 | Green Chem., 2014, 16, 3501–3507
This journal is © The Royal Society of Chemistry 2014

View Article Online
Green Chemistry
Paper
(CI): [M + H]+ (C6H11O4) m/z calculated: 147.0657; Found
147.0659.
Rev., 2007, 107, 2411; (f) P. Gallezot, ChemSusChem, 2008,
1, 734.
Acetic acid 2-hydroxy-1-methyl-propyl ester (6). 1H NMR
(500 MHz, acetone-d6): δ = 4.727–4.678 (m, 1H), 3.758–3.696
(m, 1H), 1.997–1.954 (m, 4H), 1.192–1.133 (m, 3H), 1.093 (dd,
J1 = 6.5 Hz J2 = 4 Hz, 3H); 13C NMR (125 MHz, acetone-d6):
170.63, 170.47, 170.44, 74.99, 74.90, 69.53, 69.42, 21.23, 21.05,
20.98, 19.17, 18.95, 15.36, 15.26.
2 (a) C. Y. Ng, M. Y. Jung, J. Lee and M. K. Oh, Microbiol. Cell
Fact., 2012, 11, 68; (b) M.-J. Syu, Appl. Microbiol. Biotechnol.,
2001, 55, 10; (c) J. G. Zeikus, M. K. Jain and P. Elankovan,
Microbiol. Cell Fact., 1999, 51, 545.
3 (a) L. Johansen, K. Bryn and F. C. Størmer, J. Bacteriol.,
1975, 123, 1124; (b) A. Orjuela, A. Kolah, C. T. Lira and
D. J. Miller, Ind. Eng. Chem. Res., 2011, 50, 9209.
4 (a) X. J. Ji, H. Huang, J. Du, J. G. Zhu, L. J. Ren, S. Li and
Z. K. Nie, Bioresour. Technol., 2009, 100, 5214; (b) L. Ge,
X. Wu, J. Chen and J. Wu, J. Biomater. Nanobiotechnol.,
2011, 2, 335; (c) M. Anvari and M. R. S. Motlagh, J. Biomed.
Biotechnol., 2010, 2011; (d) M. Anvari and G. Khayati, Asian
J. Chem., 2009, 21, 2131; (e) S. H. Shin, S. Kim, J. Y. Kim,
S. Lee, Y. Um, M. K. Oh, Y. R. Kim, J. Lee and K. S. Yang,
J. Bacteriol., 2012, 194, 2736. The high price of 1,3-butadiene
can be obtained from Duncan Seddon & Associates Pty. Ltd
via http://www.duncanseddon.com/docs/pdf/the-high-price-
of-butadiene.pdf.
5 (a) N. L. Morrow, Environ. Health Perspect., 1990, 86, 7;
(b) A. Singh, A. Chavda, S. Nandula, R. V. Jasra and
M. Maiti, Ind. Eng. Chem. Res., 2012, 51, 11066; (c) European
Commission. 2007. European Union Risk Assessment
Report. 1,3-butadiene. CAS no: 106-66-0. EINECS no: 203-
450-8. This report can be obtained free of charge via http://
echa.europa.eu/documents/10162/1f512549-5bf8-49a8-ba51-
1cf67dc07b72.
Acetic acid 2-acetoxy-1-methyl-propyl ester (7). 1H NMR
(500 MHz, acetone-d6): δ = 4.957–4.900 (m, 2H), 1.997–1.954
(m, 6H), 1.192–1.133 (m, 6H); 13C NMR (125 MHz, acetone-d6):
170.63, 170.47, 170.44, 71.85, 71.75, 21.23, 21.05, 20.98, 16.20,
15.91. HRMS (CI): [M + H]+ (C8H15O4) m/z calculated: 175.0970;
found 175.0971.
Acetic acid 2-formyloxy-1-methyl-propyl ester (8). 1H NMR
(500 MHz, acetone-d6): δ = 8.158–8.134 (m, 1H), 5.079–5.053
(m, 1H), 4.979–4.881 (m, 1H), 2.021–2.000 (m, 3H), 1.236–1.087
(m, 6H); 13C NMR (125 MHz, acetone-d6): 170.38, 161.62,
161.42, 76.62, 76.51, 76.25, 76.16, 76.12, 76.00, 75.96, 73.62,
73.57, 73.44, 73.05, 72.91, 72.56, 21.20, 21.05, 20.99, 16.21,
15.29.
Electrostatic potential map of products
The electrostatic potential map was obtained using the Gauss-
View 5.0 program based on the DFT (density functional theory)
calculation results. All structures were fully optimized at the
DFT level using the B3LYP hybrid functional and the 6-311
basis set with the Gaussian 03W program package. The
reported optimized structure of (R,S)-2,3-butanediol was used
to construct the initial structure of the formic acid 2-hydroxy-
1-methyl-propyl ester (3) and the formic acid 2-formyloxy-
1-methyl-propyl ester (4).15
6 W. C. White, Chem. Biol. Interact., 2007, 166, 10.
7 (a) N. Shlechter, D. F. Othmer and S. Marshak, Ind. Eng.
Chem., 1945, 37, 900; (b) N. Shlechter, D. F. Othmer and
R. Brand, Ind. Eng. Chem., 1945, 37, 905.
8 J. M. Park, H. Song, H. J. Lee and D. Seung, Microbiol. Cell
Fact., 2013, 12, 20.
Acknowledgements
9 (a) K. M. Parida and S. Mallick, J. Mol. Catal. A: Chem.,
2007, 275, 77; (b) K. Saravanan, B. Tyagi and H. C. Bajaj,
Catal. Sci. Technol., 2012, 2, 2512; (c) K. Mantri, K. Komura
and Y. Sugi, Green Chem., 2005, 7, 677; (d) S. S. Dash and
K. M. Parida, J. Mol. Catal. A: Chem., 2007, 266, 88.
10 (a) W. Zhang, D. Yu, X. Jia and H. Huang, Green Chem.,
2012, 14, 3441; (b) A. V. Tran and R. P. Chambers, Bio-
technol. Bioeng., 1987, 29, 343.
This work was financially supported by the grant from the GS
Caltex Corporation. This work was also supported by the
Supercomputing Center/Korea Institute of Science and Tech-
nology Information with supercomputing resources including
technical support (KSC-2013-C1-025). And this work was also
supported by the National Research Foundation of Korea
(NRF) grant funded by the Korea government (MEST) (no.
2013R1A2A2A01067164).
11 W. H. Brown, C. S. Foote and B. L. Iverson, Organic
Chemistry, Brooks/Cole Publishing Inc., California, 4th
edn, 2005.
12 Y. Miyano and K. Fukuchi, Fluid Phase Equilib., 2004, 226,
183.
Notes and references
1 (a) J. R. Regalbuto, Science, 2009, 325, 822; 13 J. Yu and P. E. Savage, Ind. Eng. Chem. Res., 1998, 37, 2.
(b) G. M. Whitesides and G. W. Crabtree, Science, 2007, 315, 14 A. N. Bourns and R. V. V. Nicholls, Can. J. Res., 1947,
796; (c) F. W. Lichtenthaler and S. Peters, C. R. Chim., 2004,
25b, 80.
7, 65; (d) J. N. Chheda, G. W. Huber and J. A. Dumesic, 15 A. J. L. Jesus, M. T. S. Rosado, I. Reva, R. Fausto,
Angew. Chem., 2007, 119, 7298, (Angew. Chem., Int. Ed.,
2007, 46, 7164); (e) A. Corma, S. Iborra and A. Velty, Chem.
M. E. Eusébio and J. S. Redinha, J. Phys. Chem. A, 2006,
110, 4169.
This journal is © The Royal Society of Chemistry 2014
Green Chem., 2014, 16, 3501–3507 | 3507