Formation of C-C bonds for the production of bio-alkanes under mild conditions

Article abstract of DOI:10.1039/c4gc00501e

It is of crucial importance to form C-C bonds between biomass-derived compounds for the production of bio-alkanes from biomass. In this study, it was found that C-C bonds can be formed between angelica lactones, key intermediates derived from biomass, through free radical reactions under mild conditions without using a noble catalyst or solvent, which gave elongated carbon chains of di/trimers with 10 or 15 carbons, with complete conversion and 100% selectivity. The di/trimers produced serve as a novel feedstock for the carbon backbones of bio-alkanes. Hydrogenation of the di/trimers produced C6-C13 hydrocarbons suitable for use as transportation fuels. This journal is the Partner Organisations 2014.

Full text of DOI:10.1039/c4gc00501e

Green Chemistry
View Article Online
View Journal | View Issue
PAPER
Formation of C–C bonds for the production of
bio-alkanes under mild conditions†
Cite this: Green Chem., 2014, 16,
589
3
a
a
a
a
a
Jiayu Xin, Suojiang Zhang,* Dongxia Yan, Olubunmi Ayodele, Xingmei Lu and
b
Jianji Wang
It is of crucial importance to form C–C bonds between biomass-derived compounds for the production
of bio-alkanes from biomass. In this study, it was found that C–C bonds can be formed between angelica
lactones, key intermediates derived from biomass, through free radical reactions under mild conditions
without using a noble catalyst or solvent, which gave elongated carbon chains of di/trimers with 10 or 15
carbons, with complete conversion and 100% selectivity. The di/trimers produced serve as a novel feed-
stock for the carbon backbones of bio-alkanes. Hydrogenation of the di/trimers produced C6–C13
hydrocarbons suitable for use as transportation fuels.
Received 21st March 2014,
Accepted 24th April 2014
DOI: 10.1039/c4gc00501e
www.rsc.org/greenchem
Biomass is regarded as a possible renewable resource for fuels and aldol condensation have been found to be effective for the
and chemicals. Current approaches for the conversion of purpose. However, ketonization has to be carried out at 623 K
1
2
biomass feedstocks into liquid fuels involve their gasification and 5.7 MPa with Pd or Pt as the catalysts, while aldol con-
to syngas followed by Fischer–Tropsch synthesis, thermal densation needs a ketone as a reactant and low yields have
1
3,13
liquefaction and fermentation. These approaches have one or always existed in this reaction.
more drawbacks, including the poor fuel properties of the pro- In this respect, it is desirable to develop an efficient process
ducts, low conversion efficiencies, low heating values, long that can convert widespread biomass without extensive pre-
treatment times or high processing costs. Recently, acid hydro- treatment to liquid alkanes, similar to those that have been
1
4–16
lysis to give sugars followed by aqueous-phase processing used for many years from petroleum resources,
and to
2
–7
(APP) to yield liquid fuel has attracted much interest.
The design an effective C–C bond formation reaction between
APP process can convert biomass-derived sugars into liquid biomass-derived compounds under mild conditions. Herein,
fuels with an energy recovery as high as 96% and is believed to we introduce an integrated process that converts commonly
8
17
be a promising route for industry-scale technology. However, available biomass into gasoline range alkanes. The strategies
the process needs sugars as the raw materials and the pro- of this process are to form elongated-carbon-chain oxygenated
duction of sugars is costly, and sugars tend to produce dehy- intermediates from functional chemicals produced from
dration products like hydroxylmethylfurfural or humins in the biomass, and then to remove the oxygens of the intermediates
presence of water and an acid catalyst, thus decreasing the by hydrodeoxygenation to produce gasoline range liquid
9
–11
yield of the target products.
alkanes/aromatics. As shown in Scheme 1, levulinic acid
Furthermore, typical gasoline and diesel fuels are hydro- derived from biomass is dehydrated to angelica lactones (ALs),
carbons 6 to 13 carbons in length, while the major monomeric then the ALs undergo a C–C bond formation reaction to
18
building blocks of biomass are carbohydrates which have 5 or produce C10 or C15 lactones, and these lactones are finally
carbons. In order to meet the required fuel specifications, deoxygenated to produce C6–C15 branched-chain alkanes as
6
new C–C bonds have to be constructed between sugar-, cellu- liquid fuels. In this process, a free radical reaction was
lose- or hemicellulose-derived furan compounds. Ketonization employed under mild reaction conditions in the C–C bond
formation step to give a high yield. Neither water nor any other
solvents are used in the process. Therefore, the reaction avoids
the shortcoming of energy consumption during the removal of
a
Beijing Key Laboratory of Ionic Liquids Clean Process, State Key Laboratory of
Multiphase Complex System, Institute of Process Engineering, Chinese Academy of
excessive water/solvents or heating up of the reaction mixture,
commonly employed in APP. Furthermore, AL, which has a
double bond and lactone structure produced by the dehy-
dration of levulinic acid catalyzed by a solid acid catalyst, can
be used as a new platform chemical derived from commonly
Sciences, Zhongguancun, Beiertiao, Haidian, Beijing 100190, China.
E-mail: sjzhang@ipe.ac.cn; Fax: (+86) 10-8262-7080
b
Key Laboratory of Green Chemical Media and Reactions, Ministry of Education,
School of Chemistry and Environmental Science, Henan Normal University, Xinxiang
453007, Henan, China
†
Electronic supplementary information (ESI) available: Supporting figures and
1
9–21
tables. See DOI: 10.1039/c4gc00501e
available biomass.
This journal is © The Royal Society of Chemistry 2014
Green Chem., 2014, 16, 3589–3595 | 3589

View Article Online
Paper
Green Chemistry
flask was connected to a vacuum pump set at 0.01 MPa. The
produced angelica lactones passed through a glass column
with cooling water reflux and were collected by a collector at
the end of the column. Products from the dehydration reaction
were washed 3–4 times with distilled water to remove the
remaining acid catalyst and levulinic acid. The composition of
the final products was analyzed by a Waters Acquity UPLC
system equipped with an Acquity UPLC HSS-T3 column
(
100 mm × 2.1 mm, 1.8 μm) and UV-vis detector at a wave-
length of 215 nm, yields of the ALs were determined by an
external method. The carrier solvent was 50% deionized water
and 50% acetonitrile at a flow rate of 0.1 ml min− and the
oven temperature was set to 313 K.
1
C–C bond formation
Catalysts used for the di/trimerization of angelica lactone were
crushed in an agate mortar and dried in an oven at 373 K over-
night. 100 g angelica lactone was loaded in a 500 ml flask and
Scheme 1 Reaction pathways for C–C bond formation between angel- heated to the designated temperature by an oil bath. The reac-
ica lactones (ALs) and the conversion of biomass into bio-alkanes.
tion temperature of the reactant was measured by a mercury
thermometer. When the temperature reached the designated
temperature, crushed catalyst was put into the flask. After each
reaction, the elemental compositions of the obtained products
Experimental
Catalyst preparation
were analyzed by a Vario EL cube element analyzer (Elementa
Co., Germany). The molecular weight distributions of the final
products were analyzed by gel permeation chromatography
HCl, H SO , H PO , K CO , Na CO , NaOH, H-ZSM5 and SiO
2
2
4
3
4
2
3
2
3
(60 mesh) were purchased from J&K Scientific Ltd, Beijing.
(
GPC) with an Agilent LC-1200 system under the following con-
1
0% Pt/C and NiCrFe (2 : 2 : 1) catalysts were prepared in ways
−
1
2
2,23
ditions: column, Shodex GF 321-HQ; flow rate, 1.0 ml min ;
similar to those reported before.
carrier solvent, acetonitrile; detector, refractive index detector
RID); and temperature, 313 K. The composition of the pro-
(
Hydrolysis
ducts was analyzed by an ultra performance liquid chromato-
graphy-mass spectrometer (UPLC-MS, micrOTOF-QII) and the
yields were calculated using the area normalization method.
The effluent from the LC column was directed into an electro-
spray ionization (ESI) probe. The mass spectrometer con-
ditions were optimized to obtain maximal sensitivity. The
source temperature and the desolvation temperature were
maintained at 423 K and 623 K, respectively. The probe voltage
All materials used were dried in an oven at 393 K overnight
before hydrolysis. Specified amounts of the materials for
hydrolysis, distilled water and acid solution were added to a
5
ml high pressure reactor made by Inconel-625 and sealed.
The reactor was quickly immersed in an oil bath preheated to
the designated temperature and kept there for a set time inter-
val. After the time interval, the reactor was moved to a water
bath to quench the reaction. A lignin-enriched solid was
removed by passing the contents through a 0.2 μm nylon
membrane filter. Levulinic acid and formic acid were analyzed
by a Waters Acquity UPLC system equipped with an Acquity
UPLC HSS-T3 column and UV-vis detector at wavelengths of
(
capillary voltage), cone voltage and extractor voltage were
fixed at 1.5 kV, 65 V and 2 V, respectively. Nitrogen was used as
−1
the source of the desolvation gas (650 l h ) and drying gas
−1
(
4 l min ). P57 was confirmed in selected ion recording (SIR)
+
mode. [M + Na] = 901.6 ion for P57 was selected as the detect-
ing ion. Mass spectra were obtained using a dwell time of 0.1 s
in SIR and a scan rate of 5000 Da s . H NMR spectra were
2
15 and 268 nm. The carrier solvent was 97% KH PO aqueous
2 4
−
1
solution (0.01 mol l and pH value adjusted to be 2.7 by
H PO ) and 3% acetonitrile at a flow rate of 0.1 ml min and
−1 1
−
1
3
4
recorded using a JNM-ECS400 NMR spectrometer operating at
the oven temperature was set to 313 K. The yields of levulinic
acid and formic acid were calculated by an external method
and reported as a percentage of the theoretical yield from the
C H O unit in cellulose according to Fig. S1.†
13
3
3 2
00 MHz in (CD ) SO. C NMR spectra were recorded on the
same instrument at an operating frequency of 75 MHz.
Electron spin resonance (ESR) measurements were carried
out on a JEOL JES FA200 ESR spectrometer at 298 K. The
typical spectrometer parameters include: microwave frequency,
6
12 5
Dehydration
9.4 GHz; microwave power, 4 mW; center field, 3374 G; sweep
1
00 g levulinic acid was loaded in a 500 ml distillation flask width, 8 mT; sweep time, 4 min; and time constant, 30 ms;
and heated to the designated temperature by an oil bath, and modulation frequency 100 kHz; and modulation amplitude,
then the flask was charged with the catalyst. The outlet of the 0.1 G.
3590 | Green Chem., 2014, 16, 3589–3595
This journal is © The Royal Society of Chemistry 2014

View Article Online
Green Chemistry
Paper
Hydrodeoxygenation
Table 1 Conversion of levulinic acid into ALs by employing 3 wt%
3 4 3 4 2 2
H PO , H PO /SiO , H-ZSM5/SiO as catalysts. The reactions were
The dimers and trimers of angelica lactones were washed with
distilled water to remove the remaining catalyst. After the
carried out in a 500 ml distillation flask and heated to the designated
temperature by an oil bath, and then the flask was charged with the
removal of the catalyst, the products were hydrogenated in a catalyst. The outlet of the flask was connected to a vacuum pump set at
.01 MPa. The produced ALs passed through a glass column with
0
batch type reactor using a 10% Pd/C catalyst at 523, 573 and
48 K for 8–60 h. The amount of catalyst, reaction temperature
and time for hydrodeoxygenation reactions are listed in
cooling water reflux and were collected by a collector at the end of the
column
6
Table S3.† Gas products were analyzed by gas chromatography
Yield [mol%]
(
GC) on a Shimadzu GC2014 instrument equipped with a TCD
detector and an HP-PLOT Q capillary column (Agilent, 30 m ×
.53 mm, 40.00 μm) to quantify low molecular weight alkanes,
Entry
Catalyst
T [K]
α-AL
β-AL
Total
0
1
2
3
4
5
6
7
8
9
H
3
H
3
H
3
H
3
H
3
H
2
PO
PO
PO
PO
PO
4
4
4
4
4
433
433
433
448
458
433
433
423
423
443
403
423
19
44
32
49
0
0
3
50
5
2
64
1
4
6
8
9
0
0
1
5
3
23
49
40
58
0
0
4
62
8
4
97
2
CO and CO2 by an external method. Qualitative analysis of
the liquid products was performed by an Agilent 6890
GC equipped with an HP-5 capillary column (Agilent, 30 m ×
0
.53 mm, 40.00 μm) and an Agilent 5973 mass selective detec-
SO
4
None/SiO
PO /SiO
2
tor (MSD, Agilent Technologies). The liquid products were
quantitatively analyzed by the same GC used for the gas
products, with an HP-5 capillary column (Agilent, 30 m × 10
0
H
3
4
2
H-ZSM5/SiO
H-ZSM5/SiO
H-ZSM5/SiO
2
2
2
2
33
1
1
1
1
2
.53 mm, 40.00 μm) and the yields were calculated using the
2 4 2
H SO /SiO
area normalization method.
In order to remove the produced ALs immediately by evap-
oration, we added SiO (60 mesh) to promote the generation of
AL bubbles in the reactant phase. As expected, a higher yield
was achieved at a lower temperature. H PO as the homo-
2
Results and discussion
Preparation of levulinic acid
3
4
We studied the process with levulinic acid as the raw material. geneous liquid catalyst may provide good contact with the reac-
Levulinic acid can be produced via the hydrolysis of sacchar- tant, but its boiling point is also low, and it may evaporate
ides, polysaccharides, cellulose, starch, etc., catalyzed by an under negative pressure and condense together with the pro-
1
9,26
acid at 100–200 °C with up to 73% yield (mol basis) without ducts. We further added H-ZSM5 as a solid catalyst
to
8
comprehensive pre-treatment. The production of levulinic replace H
PO
for the reaction. However, levulinic acid was
4
also evaporated without reacting at 423 K and 0.01 MPa, with
3
2
4,25
acid from various sources has been extensively studied,
thus some results obtained by using cellulose, cotton, fructose, the employment of fine particles of H-ZSM5. We carried out
glucose, cornstalk and starch powder are summarized and pro- the same reaction at 403 K and consequently found that the
vided in the ESI (Table S2†).
conversion of levulinic acid was greatly increased and the yield
was also increased to 97 mol%. We also noticed that the reac-
tion produced two isomers of AL, including the α and β con-
figurations, and that more β-AL was produced at higher
temperatures.
Dehydration of levulinic acid
The dehydration of levulinic acid results in the production of
α-AL, which is further isomerized to produce β-AL in the pres-
ence of an acid catalyst at a high temperature. In order to
obtain a high yield of AL and suppress side reactions, AL has
Formation of C–C bond
to be removed from the reactants once it is produced. The One important issue for the production of alkanes with mole-
boiling point of AL is about 438–443 K, while the boiling point cular weights appropriate for gasoline from biomass is to form
of levulinic acid is 518 K. The difference in boiling points C–C bonds between molecules, to meet the number of carbon
facilitates the separation of levulinic acid and AL by vacuum atoms needed for fuel quality requirements. Building a C–C
distillation. We carried out the dehydration reaction of levuli- linkage directly from two organic compounds has emerged as
nic acid by employing 3 wt% H PO as a catalyst under an attractive and challenging goal in catalysis. In the past
3
4
0
.01 MPa and the results are shown in Table 1. Up to 58% several decades, transition metal-catalyzed C–C bond for-
yield was obtained in the case of using H PO as a catalyst at mation has allowed chemists to assemble complex molecular
48 K. Such a low yield possibly existed because the produced frameworks.
AL could not be removed from the reactants immediately. tion between molecules can be achieved by an aldol conden-
3
4
2
7–31
7,32
4
Besides, Dumesic et al.
found the connec-
3
3–35
A temperature higher than 458 K led to no reaction, due to the sation reaction between HMF
or furfural and acetone,
introduced a method to increase the
4
was also studied and it was found that the reactants number of carbon atoms by hydroxyalkylation and alkylation.
3
6–38
rapid evaporation of levulinic acid. The catalytic activity of while Corma et al.
SO
H
2
quickly became dark due to the over-dehydration effect of
In this study, we found that C–C bonds were rapidly formed
H SO at such a temperature.
between AL molecules, without using expensive noble catalysts,
2
4
This journal is © The Royal Society of Chemistry 2014
Green Chem., 2014, 16, 3589–3595 | 3591

View Article Online
Paper
Green Chemistry
Table 2 AL conversion, isomerization and yield to dimers, trimers and
tetramers catalyzed by K
various temperatures
2 3 2 3 3 2
CO , Na CO , NaHCO , Ca(OH) and NaOH at
Yield [wt%]
Entry
Catalyst/T [K]/time [min]
α-AL
β-AL
Di
Tri
Tetr
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
K
K
K
K
K
K
K
K
Na
Na
Na
Na
NaHCO
NaHCO
NaHCO
Ca(OH)
Ca(OH)
2
2
2
2
2
2
2
2
CO
CO
CO
CO
CO
CO
CO
CO
3
3
3
3
3
3
3
3
/353/0
/353/1
/353/3
/353/5
/353/10
/353/30
/393/30
/433/30
90
0
0
0
0
0
0
0
18
6
10
0
0
0
0
0
0
0
82
94
96
0
61
89
0
33
90
49
86
99
0
0
0
0
0
0
2
2
2
2
5
5
0
0
0
4
0
0
3
0
0
0
0
0
65
64
64
65
64
63
0
0
0
71
0
0
73
0
0
0
0
33
34
34
33
31
32
0
0
5
25
0
0
24
0
0
0
0
2
CO
2
CO
2
CO
2
CO
3
/353/30
3
/353/180
3
/413/30
3
/433/1
0
1
2
3
4
5
6
7
8
9
0
4
0
Fig. 1 ESR spectrum of AL in CH
2 3
CN solution with 3 wt% K CO at
3
room temperature.
3
/353/30
3
/353/180
3
/413/1
39
11
0
67
10
51
14
1
2
/353/30
/413/30
2
NaOH/353/10
NaOH/353/720
NaOH/413/30
0
0
3
9
under mild conditions. It is observed that both α- and
β-ALs can be converted to a dimer (5-methyl-5-(2-methyl-5-oxo-
tetrahydrofuran-3-yl)furan-2(5H)-one) or trimer (2,2′-dimethyl-
3
5
-(2-methyl-5-oxo-2,5-dihydrofuran-2-yl)tetrahydro-2,3′-bifuran-
,5′(2H,2′H)-dione) through C–C bond formation in the pres-
ence of medium strength alkalis like K CO or Na CO as cata-
2
3
2
3
lysts. Firstly, C–C bond formation between the ALs was carried
out at 353 K under atmospheric pressure by employing 3%
K CO as the catalyst. The obtained products at various temp-
2
3
eratures are shown in Table 2. We found that the reaction can
be completed in a very short time (3.5 min) and a longer reac-
tion time did not significantly affect the composition of the
products. At the end of the reaction, a brown, viscous and
transparent liquid at room temperature was obtained. The
HPLC analysis showed that the ALs were completely converted
to dimers and trimers, as well as trace amounts of tetramers.
The reaction was exothermic, and resulted in a further increase
in temperature, and an induction period was observed before
the reaction. When the temperatures employed in the C–C
Scheme 2 Proposed free radical mechanism for the C–C bond for-
mation between ALs.
bond formation between the ALs were varied between best of our knowledge, the mechanism of the C–C bond for-
53–423 K, the induction periods and obtained dimers and mation between ALs has not been reported. From the ESR
3
trimers varied with the reaction temperatures. These results spectrum, by determining the g factors, three typical radicals
can be rationalized, assuming that the C–C bond formation can be labelled. g Factors equal to 2.31 and 2.30 are labelled as
reaction undergoes a free radical pathway.
radical 3 (Scheme 2), g factors equal to 2.46 and 2.42 are
•
−
Although we lack a clear picture of the initiation and propa- labelled as CO
and g factors equal to 1.00 are labelled as
3
gation of the C–C bond formation between ALs in the presence radical 4. On the basis of the observations described above, we
of a carbonate, we are able to offer some insights into the proposed the mechanism shown in Scheme 2. 1 is firstly iso-
mechanism. The ESR spectrum of the AL–acetonitrile solution merized to β-AL 2 at the appropriate temperature in the pres-
2
−
(
1
3 wt%) with K
2
CO
3
as a catalyst (3 wt%) and trapped with ence of K
2
CO
3
. CO
3
, from the ionization of K
2
CO
3
, gives an
•
−
00 mM DMPO is shown in Fig. 1. Considering the ESR spec- electron to 2 to form the radical anion CO
3
(ref. 40 and 41)
trum and induction period before the C–C bond formation and radical anion 3. 3 undergoes a homolytic cleavage to form
•
−
•−
reaction, it is evident that it is a free radical reaction. To the CO
3
and the highly reactive radical anion 4. The CO
3
3592 | Green Chem., 2014, 16, 3589–3595
This journal is © The Royal Society of Chemistry 2014

View Article Online
Green Chemistry
Paper
2 4
radical further reacts to produce CO and CO
2−. The produced
CO was observed in the experiment and confirmed by GC-MS.
2
Anion radical 4 rapidly reacts with 1 to give the dimer anion
radical 5, which may react with 1 to form the higher molecular
weight radical 7. Another possibility involving the reaction of 5
is single electron transfer to produce the target product 6.
Similarly, the trimer of AL is also produced in the same
manner.
The effects of different catalysts such as NaOH, Ca(OH)
2
,
Na
2
CO , CaCO and NaHCO on the dimerization/trimeriza-
3
3
3
tion were also investigated. The reaction catalyzed by 3 wt%
NaOH showed a significant difference from that catalyzed by
K
2
CO
while NaOH as a catalyst produced isomerized products of AL
in its α and β forms (Fig. S6†) and Ca(OH) showed similar be-
3 2 3
. K CO as a catalyst produced dimers and trimers of AL,
Fig. 2 Product distribution of the organic phase produced in the
hydrogenation reaction of the di/trimerization products of the ALs. (A)
Hydrodeoxygenation of dimers and trimers produced from ALs at 573 K
2
havior to that of the NaOH catalyst. These results suggest
that the basicity of the catalyst played an important role in
and 5 MPa H
2
with 10% Pd/C catalyst for 12 h. (B) Hydrodeoxygenation
with
of dimers and trimers produced from ALs at 623 K and 5 MPa H
2
this reaction. In order to illustrate the effect of the reaction 10% Pd/C catalyst for 18 h. (C) Hydrodeoxygenation of dimers and
trimers produced from ALs at 648 K and 5 MPa H
lyst for 60 h. (D) Dimer and trimer products of the ALs were pre-hydro-
genated at 648 K and 5 MPa H with a NiCrFe catalyst for 12 h, and then
the liquid products were hydrodeoxygenated under the same conditions
with a newly added catalyst.
2
with 10% Pd/C cata-
time on the di/trimerization of AL, the changes to α-AL heated
at 353 K for various reaction times without a catalyst were
studied. As seen from Fig. S6,† there was almost no change
within 1 h but more prolonged reaction times led to the pro-
duction of some unknown products, possibly due to thermal
decomposition.
2
Hydrogenation of di/trimers of AL
The produced dimers and trimers were hydrodeoxygenated in
a 100 ml batch-type reactor at 573–648 K with 10% Pd/C or
NiCrFe micro particles as catalysts. Hydrogen was consumed
during the reaction and extra hydrogen was supplied to main-
tain the overall pressure. The obtained products separated into
two phases when the products were settled and cooled at room
temperature. The upper phase was the organic phase, contain-
ing liquid alkanes/aromatics and partly hydrodeoxygenated
4
2
products, and the lower phase was the aqueous phase
mainly composed of water). We first carried out the reaction
(
with 5 g of products from the di/trimerization reaction from Fig. 3 Product distribution of the hydrogenation reaction of the
AL with H at a pressure of 5 MPa at 623 K in a batch-type di/trimerization products of the ALs in a fixed-bed reactor. Reaction
2
2
conditions: 0.5 g 10% Pd/C catalyst, 573 K, 5 MPa H and a flow rate of
reactor and the results are shown in Fig. 2A (selected values
from Table S3†). Alkanes containing more than 5 carbons con-
stitute 49% of the products and oxygenated compounds consti-
tute 51% of products, and were caused by the incomplete
−1
2
ml h .
4
3
47,48
hydrodeoxygenation reaction. We further increased the reac- removal.
The result also showed that Pd/C is effective for
tion temperature and time. As shown in Fig. 2B–2D, under oxygen removal. Other factors that affect the distribution of
suitable conditions, the obtained products were mainly the molecular weights of the alkanes are the temperature and
1
9,49
C6–C12 alkanes and/or aromatics with molecular weights time,
appropriate for gasoline.
a higher temperature and longer reaction time result
Ideally, the liquid organic pro- in higher alkane yields.
ducts from the hydrogenation of the dimers and trimers from
Since the catalyst lifetime is an important factor in the
4
4,45
AL should be C10 (from dimer) and C15 (from trimer) alkanes evaluation of the process, we carried out a continuous reaction
and accompanied by H O as a by-product. However, most of for 100 h with 10% Pd/C as the catalyst. After hydrogenation,
2
the products were found to be C8–C10 under the optimum the organic phase contained 90–97% hydrocarbons. There
conditions. The gaseous products produced during the hydro- were also 3–5% oxygenated compounds detected in the pro-
genation were CO/CO , methane and n-butane as shown in ducts, caused by incomplete deoxygenation under these con-
2
Fig. S8.† This indicates that C–C bond cleavage also took place ditions (Fig. 3).
4
6
during hydrogenation. The production of CO
2
in the final
The integrated system for the conversion of biomass into
products leads to the lower consumption of H2 for oxygen alkanes consists of four reactors in series. Raw materials,
This journal is © The Royal Society of Chemistry 2014
Green Chem., 2014, 16, 3589–3595 | 3593

View Article Online
Paper
Green Chemistry
liquid acid catalysts and water are fed into the first reactor to
carry out the hydrolysis reaction. An aqueous solution contain-
Conclusions
ing levulinic acid, formic acid and acid catalysts is filtered to The major monomeric building blocks of biomass are
remove the solid residue that mainly consists of lignin and carbohydrates typically containing 5 or 6 carbons. Petroleum-
humins. After its separation from formic acid, the catalyst and derived gasoline and diesel fuels are hydrocarbons having 6 to
water, levulinic acid is subjected to the second reactor for the 15 carbons. Therefore, to produce gasoline and diesel fuel
dehydration reaction under vacuum conditions by employing from biomass, there must be a C–C bond formation between
H-ZSM5 as the catalyst. The obtained products from the the biomass-derived molecules. In this study, it was found that
vacuum dehydration reaction are collected in a sealed con- C–C bonds can be formed between angelica lactones through
tainer, in which the products separate into two phases; the free radical reactions under mild conditions without using
upper (lighter) phase contains water and the lower (heavier)
a noble catalyst or solvent, which gave the elongated
phase contains AL. The lower phase (AL) is transferred to the carbon chain products with 10 or 15 carbons with complete
third reactor by employing carbonate as a catalyst to produce conversion and 100% selectivity. After hydrogenation, C6–C15
the AL dimers and trimers. Finally, the AL dimers and trimers hydrocarbons can be obtained for transportation fuels.
are washed with hot distilled water to remove the catalyst, and
then fed into the hydrodeoxygenation reactor to produce bio-
alkanes for gasoline. It is worth noting that part of the H for
2
Acknowledgements
hydrogenation can be obtained from the decomposition of
5
0–54
formic acid produced in the hydrolysis step,
which This work was supported by the National Basic Research
reduces the H consumption and makes economic use of the Program of China (973 Program no. 2013CB733506) and
2
by-product.
National Natural Science Foundation of China (no. 21210006,
The research shows that it is possible to produce alkanes 21006112, 21276260). We thank Prof. Z. Conrad Zhang for
in the range of C6–C13, most of them are C8 and helpful discussion and Dr Chunyan Shi for her help in experi-
C9 alkanes, which are of the carbon numbers suitable for the ment and analysis.
requirements of gasoline, from cellulose/hemicelluloses and
sugar-containing biomass. The raw materials used for this
process can be widely available biomass such as wood,
saw dust, rice straw, corn stalk, bagasse, waste paper and so
Notes and references
on. Under the optimal reaction conditions, 73 wt% carbon in
cellulose/sugar is converted into levulinic acid, the produced
levulinic acid can be upgraded to AL with a carbon yield
of 97 mol% and this can be further converted into dimers
and trimers with a carbon yield of 100 wt%. After hydro-
genation, 92 wt% dimers/trimers of AL are converted into
alkanes and remain in the liquid phase. According to
the above-mentioned yields, it is possible to convert nearly
1 P. Alvira, E. Tomas-Pejo, M. Ballesteros and M. J. Negro,
Bioresour. Technol., 2010, 101, 4851–4861.
2 E. L. Kunkes, D. A. Simonetti, R. M. West, J. C. Serrano-
Ruiz, C. A. Gartner and J. A. Dumesic, Science, 2008, 322,
417–421.
3 G. W. Huber, J. N. Chheda, C. J. Barrett and J. A. Dumesic,
Science, 2005, 308, 1446–1450.
4 J. Q. Bond, D. M. Alonso, R. M. West and J. A. Dumesic,
Langmuir, 2010, 26, 16291–16298.
5 J. C. Serrano-Ruiz, D. J. Braden, R. M. West and
J. A. Dumesic, Appl. Catal., B, 2010, 100, 184–189.
6 E. I. Gürbüz, J. M. R. Gallo, D. M. Alonso, S. G. Wettstein,
W. Y. Lim and J. A. Dumesic, Angew. Chem., Int. Ed., 2013,
52, 1270–1274.
6
5% of the carbon of cellulose or sugars into alkanes and
aromatics. In other words, the process can produce about
2 kg alkanes from 100 kg cellulose or sugars, or about 250 kg
3
wood. This amount is similar to that of ethanol from 100 kg
cellulose. Accordingly, the process consumes 5 mol H (1 CvC
2
bond and 4 oxygens) to produce 1 mol of decane from
cellulose. Besides, if H
sumption can be reduced. However, the obtained bio-alkanes
are oxygen-free, which results in a calorific value that is
2
is produced from formic acid, its con-
7 J. N. Chheda, G. W. Huber and J. A. Dumesic, Angew.
Chem., Int. Ed., 2007, 46, 7164–7183.
8 S. M. Sen, C. A. Henao, D. J. Braden, J. A. Dumesic and
C. T. Maravelias, Chem. Eng. Sci., 2012, 67, 57–67.
9 J. Song, H. Fan, J. Ma and B. Han, Green Chem., 2013, 15,
2619–2635.
1
.8 times higher than that of ethanol. A striking breakthrough
is that our process starts from the biomass directly rather than
from sugars. Moreover, the formation of the C–C bond in
di/trimerization could take place under mild conditions 10 H. Guo, Y. Lian, L. Yan, X. Qi and R. L. Smith, Green Chem.,
353 K) with 100% selectivity through free radical reactions, 2013, 15, 2167–2174.
while such reactions usually lead to a diverse spectrum of pro- 11 H. Kobayashi and A. Fukuoka, Green Chem., 2013, 15,
ducts. The general conclusion from this study is that alkanes/ 1740–1763.
aromatics can be formed by converting biomass into platform 12 E. I. Gurbuz, E. L. Kunkes and J. A. Dumesic, Green Chem.,
chemicals; increasing the number of carbons and finally 2009, 12, 223–227.
hydrogenating the products to remove the oxygen by 13 J. Carlos Serrano-Ruiz and J. A. Dumesic, Energy Environ.
hydrogenation. Sci., 2011, 4, 83–99.
(
3594 | Green Chem., 2014, 16, 3589–3595
This journal is © The Royal Society of Chemistry 2014

View Article Online
Green Chemistry
Paper
1
1
1
1
4 M. J. Climent, A. Corma and S. Iborra, Green Chem., 2014, 34 X. Qi, H. Guo, L. Li and R. L. Smith, ChemSusChem, 2012,
6, 516–547. 5, 2215–2220.
5 J. Han, S. M. Sen, D. M. Alonso, J. A. Dumesic and 35 H. Guo, Y. Lian, L. Yan, X. Qi and R. L. Smith, Green Chem.,
C. T. Maravelias, Green Chem., 2014, 16, 653–661. 2013, 15, 2167–2174.
6 G. Li, N. Li, J. Yang, L. Li, A. Wang, X. Wang, Y. Cong and 36 A. Corma, G. W. Huber, L. Sauvanaud and P. O’Connor,
T. Zhang, Green Chem., 2014, 16, 594–599. J. Catal., 2008, 257, 163–171.
7 J. Xin and S. Zhang, Integrated catalytic process for the 37 A. Corma, O. de la Torre, M. Renz and N. Villandier, Angew.
1
bio-alkane production from waste biomass, Busan, Korea,
012.
8 R. Lukes and K. Syhora, Collect. Czech. Chem. Commun.,
954, 19, 1205–1214.
9 A. M. Ruppert, K. Weinberg and R. Palkovits, Angew. Chem.,
Int. Ed., 2012, 0, 2564–2601.
Chem., Int. Ed., 2011, 50, 2375–2378.
38 G. W. Huber and A. Corma, Angew. Chem., Int. Ed., 2007,
46, 7184–7201.
39 M. Mascal, S. Dutta and I. Gandarias, Angew. Chem., Int.
Ed., 2014, 53, 1854–1857.
2
1
1
2
1
40 S. Goldstein and G. Czapski, J. Am. Chem. Soc., 1998, 120,
3458–3463.
0 P. N. R. Vennestrøm, C. M. Osmundsen, C. H. Christensen
and E. Taarning, Angew. Chem., Int. Ed., 2011, 50, 10502– 41 G. Wu, Y. Katsumura, Y. Muroya, M. Lin and T. Morioka,
1
0509.
J. Phys. Chem. A, 2002, 106, 2430–2437.
42 R. W. Gosselink, D. R. Stellwagen and J. H. Bitter, Angew.
Chem., Int. Ed., 2013, 52, 5089–5092.
2
2
1 R. Schlögl, Angew. Chem., Int. Ed., 2011, 50, 6424–6426.
2 E. C. G. Rufino and P. Olivi, Int. J. Hydrogen Energy, 2010,
3
5, 13298–13308.
3 T. Kan, H. Wang, H. He, C. Li and S. Zhang, Fuel, 2011, 90,
404–3409.
43 M. Shiramizu and F. D. Toste, Angew. Chem., Int. Ed., 2012,
51, 8082–8086.
44 J. Lee, Y. T. Kim and G. W. Huber, Green Chem., 2014, 16,
708–718.
45 D. D. Laskar, M. P. Tucker, X. Chen, G. L. Helms and
B. Yang, Green Chem., 2014, 16, 897–910.
2
2
2
3
4 S. Rivas, M. J. González-Muñoz, C. Vila, V. Santos and
J. C. Parajó, Ind. Eng. Chem. Res., 2013, 52, 3951–3957.
5 R. Saladino, T. Pagliaccia, D. S. Argyropoulos and
C. Crestini, in Materials, Chemicals, and Energy from Forest 46 S. K. Hanson, R. Wu and L. A. P. Silks, Angew. Chem., Int.
Biomass, American Chemical Society, 2007, vol. 954, ch. 17,
pp. 262–279.
6 S. J. Reitmeier, O. C. Gobin, A. Jentys and J. A. Lercher,
Ed., 2012, 51, 3410–3413.
47 C. V. Loricera, P. Castano, A. Infantes-Molina, I. Hita,
A. Gutierrez, J. M. Arandes, J. L. G. Fierro and B. Pawelec,
Green Chem., 2012, 14, 2759–2770.
2
2
2
2
3
3
3
3
Angew. Chem., Int. Ed., 2009, 48, 533–538.
7 A. Heckmann and C. Lambert, Angew. Chem., Int. Ed., 2011, 48 T. Dijkmans, S. P. Pyl, M.-F. Reyniers, R. Abhari, K. M. Van
1, 326–392. Geem and G. B. Marin, Green Chem., 2013, 15, 3064–3076.
8 C. C. C. Johansson Seechurn, M. O. Kitching, T. J. Colacot 49 C. Zhao and J. A. Lercher, Angew. Chem., Int. Ed., 2012, 51,
and V. Snieckus, Angew. Chem., Int. Ed., 2012, 51, 5062–5085. 5935–5940.
9 J. Yamaguchi, A. D. Yamaguchi and K. Itami, Angew. Chem., 50 L. Deng, J. Li, D.-M. Lai, Y. Fu and Q.-X. Guo, Angew.
Int. Ed., 2012, 51, 8960–9009. Chem., Int. Ed., 2009, 48, 6529–6532.
0 M. J. Monreal and P. L. Diaconescu, J. Am. Chem. Soc., 51 Q.-Y. Bi, X.-L. Du, Y.-M. Liu, Y. Cao, H.-Y. He and K.-N. Fan,
010, 132, 7676–7683. J. Am. Chem. Soc., 2012, 134, 8926–8933.
1 M. Espinal Viguri, M. A. Huertos, J. Pérez, L. Riera and 52 T. C. Johnson, D. J. Morris and M. Wills, Chem. Soc. Rev.,
I. Ara, J. Am. Chem. Soc., 2012, 134, 20326–20329. 2010, 39, 81–88.
2 J. N. Chheda, Y. Roman-Leshkov and J. A. Dumesic, Green 53 X. Zhou, Y. Huang, W. Xing, C. Liu, J. Liao and T. Lu,
Chem., 2007, 9, 342–350. Chem. Commun., 2008, 3540–3542.
3 X. Qi, M. Watanabe, T. M. Aida and R. L. Smith, Chem- 54 T. Thananatthanachon and T. B. Rauchfuss, Angew. Chem.,
SusChem, 2009, 2, 944–946. Int. Ed., 2010, 49, 6616–6618.
5
2
This journal is © The Royal Society of Chemistry 2014
Green Chem., 2014, 16, 3589–3595 | 3595