Solvent-free synthesis of C9 and C10 branched alkanes with furfural and 3-pentanone from lignocellulose

Article abstract of DOI:10.1016/j.catcom.2014.11.003

Jet fuel range branched alkanes were first synthesized under solvent-free conditions by the aldol condensation of furfural and 3-pentanone from lignocellulose followed by the one-step hydrodeoxygenation (HDO). Among the investigated solid base catalysts,

Full text of DOI:10.1016/j.catcom.2014.11.003

Catalysis Communications 59 (2015) 229–232
Contents lists available at ScienceDirect
Catalysis Communications
journal homepage: www.elsevier.com/locate/catcom
Short Communication
Solvent-free synthesis of C9 and C10 branched alkanes with furfural and
3-pentanone from lignocellulose
Fang Chen a,b, Ning Li a, , Shanshan Li a,b, Jinfan Yang a,b, Fei Liu a, Wentao Wang a, Aiqin Wang a, Yu Cong a,
⁎
a,⁎⁎
Xiaodong Wang a, Tao Zhang
a
State Key Laboratory of Catalysis, Dalian Institute of Chemical Physics, CAS, Dalian 116023, China
University of Chinese Academy of Sciences, Beijing 100049, China
b
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 26 June 2014
Received in revised form 28 October 2014
Accepted 4 November 2014
Available online 9 November 2014
Jet fuel range branched alkanes were first synthesized under solvent-free conditions by the aldol condensation of
furfural and 3-pentanone from lignocellulose followed by the one-step hydrodeoxygenation (HDO). Among the
investigated solid base catalysts, CaO and KF/Al2O3 demonstrated the highest activity for the aldol condensation
reaction of furfural and 3-pentanone. The aldol condensation product of furfural and 3-pentanone (liquid at room
temperature) was directly hydrodeoxygenated to 4-methyl-nonane and 4-methyl-octane. These alkanes have
low freezing points (174.1 K and 159.8 K) and can be blended into jet fuel without hydroisomerization.
Among the investigated HDO catalysts, the Ni-Cu/SiO2 exhibited the best performance.
Keywords:
Aldol condensation
Hydrodeoxygenation
Lignocellulose
Jet fuel
© 2014 Elsevier B.V. All rights reserved.
Branched alkane
1. Introduction
consumption. (2) Straight alkanes with lower octane numbers and
higher freezing points are produced from the HDO step. Both characters
As a solution to the current energy and environmental problems, the
catalytic conversion of renewable biomass to fuels and chemicals has
drawn tremendous attention [1–3]. Jet fuel is one of the often used liq-
uid fuels, which are mainly produced from petroleum nowadays. Ligno-
cellulose is the main component of agriculture waste and forest
residues. In recent years, the synthesis of jet fuel range alkanes with
the lignocellulose-derived platform chemicals becomes an active re-
search area [4–7].
Furfural is an important chemical which has been produced in large
scale by the catalytic hydrolysis-dehydration of hemicellulose [8]. In the
recent works of Dumesic and Huber [4,9–11], it was found that C8 and
C13 oxygenates can be obtained by the aldol condensation of furfural
and acetone. After the low-temperature hydrogenation and HDO of
these oxygenates, C8-C13 alkanes can be obtained (see Scheme 1a).
Pioneered by their works, a lot of research was also carried out by
other groups about the same reaction system [12,13]. These works
open a new era in the production of lignocellulosic biofuel. However,
there are also two limitations for the furfural–acetone route: (1) the
aldol condensation products of furfural and acetone are solid at room
temperature. Therefore, organic solvents were indispensable to facili-
tate the mass transfer, which will increase the cost and energy
are unfavorable for aviation fuel.
3-Pentanone is the ketonization product of propanoic acid that can
be obtained from the dehydration/hydrogenation of lactic acid [14], an
important lignocellulosic platform compound [15]. To our knowledge,
there is no report about the synthesis of diesel or jet fuel range alkanes
with 3-pentanone. In this work, we reported, for the first time, the high-
ly efficient synthesis of C9 and C10 branched alkanes by the aldol con-
densation of 3-pentanone and furfural followed by the one-step HDO
process under solvent-free conditions (Scheme 1b). These branched al-
kanes have low freezing points and can be blended into jet fuel without
hydroisomerization.
2. Experimental
2.1. Catalyst preparation
2.1.1. Solid base catalysts
MgO and CaO were purchased from Kermel Reagent Company. Be-
fore being used in reaction, these samples were calcined in nitrogen
flow at 873 K for 4 h.
MgO-ZrO2 and MgO-La2O3 mixed oxide, MgAl hydrotalcite (MgAl-
HT) and LiAl hydrotalcite (LiAl-HT) were prepared by the methods in-
troduced in our previous works [16,17].
⁎
Corresponding author. Tel.: +86 411 84379738; fax: +86 411 84685940.
KF/Al2O3 was prepared according to literature [18] by the impregna-
tion of 5.8 g γ-Al2O3 with a solution of KF (5 g KF⋅2H2O in 16 g water).
⁎⁎ Corresponding author. Tel.: +86 411 94379015; fax: +86 411 84691570.
E-mail addresses: lining@dicp.ac.cn (N. Li), taozhang@dicp.ac.cn (T. Zhang).
http://dx.doi.org/10.1016/j.catcom.2014.11.003
1566-7367/© 2014 Elsevier B.V. All rights reserved.

230
F. Chen et al. / Catalysis Communications 59 (2015) 229–232
Previous work
a)
Our work
b)
Scheme 1. (a) Previous route versus (b) our new protocol for the synthesis of jet fuel range branched alkanes.
The mixture was stirred at 343 K for 1 h, dried at 393 K for 4 h and cal-
cined in N2 flow at 873 K for 4 h.
through the back-pressure regulator and were analyzed online by an
Agilent 7890A GC. The liquid products were withdrawn periodically
from the gas-liquid separator and analyzed by another Agilent 7890A
GC.
2.1.2. Hydrodeoxygenation (HDO) catalysts
The mono-metallic M/SiO2 (M = Fe, Co, Ni, Cu, Pt) catalysts were
prepared by the incipient wetness impregnation of SiO2 (Qingdao
Ocean Chemical Ltd.) with the aqueous solutions of Fe(NO3)3 ⋅ 9H2O,
Co(NO3)2 ⋅ 6H2O, Ni(NO3)3 ⋅ 6H2O, Cu(NO3)2 ⋅ 3H2O and H2PtCl6.6H2O,
respectively. For all these catalysts, the theoretical contents of metal in
the M/SiO2 catalysts were fixed as 5 wt.%. After impregnation, the sam-
ples were kept at room temperature for 4 h, dried at 333 K overnight
and calcined in air at 773 K for 4 h.
The bimetallic Ni-Cu/SiO2 catalysts were prepared by the co-
impregnation of SiO2 with the solution of Ni(NO3)3 ⋅ 6H2O and
Cu(NO3)2 ⋅ 3H2O. The total theoretical metal contents of Ni and Cu in
the catalysts were controlled as 5 wt.%. To facilitate the comprehension,
the weight percentages of Ni and Cu were indicated in the names of Ni-
Cu/SiO2 catalysts. For example, the Ni4Cu1/SiO2 catalyst denotes the Ni-
Cu/SiO2 catalyst containing 4 wt.% Ni and 1 wt.% Cu.
3. Results and discussion
3.1. Aldol condensation
The solvent-free aldol condensation of furfural and 3-pentanone was
carried out over a series of solid bases. From the analysis of HPLC and
NMR (see Figs. S1 and S2 in supporting information), 1-(furan-2-yl)-
2-methylpent-1-en-3-one (i.e., 1A in Scheme 1) was identified as the
main product. No 1,5-di(furan-2-yl)-2,4-dimethylpenta-1,4-dien-3-
one (generated by the aldol condensation of one 3-pentanone molecule
with two furfural molecules) was detected in the product even at lower
ketone-to-furfural ratio of 0.5. This phenomenon was different with
what has been observed in the aldol condensation of furfural with ace-
tone [4,9] and can be explained by the lower reactivity of α-carbon
atoms in a 3-pentanone molecule (caused by the electron and steric ef-
fects of the methyl groups which are connected with the two α-carbon
atoms).
2.2. Activity test
The aldol condensation of furfural and 3-pentanone was performed
in batch reactor. Typically, 2.60 g furfural, 4.67 g 3-pentanone and
0.4 g solid base catalyst were used for each test. Before the reaction,
the reactor was purged with argon. The mixture was stirred at 443 K
for 8 h then cooled down to room temperature. The liquid products
were diluted, filtered and analyzed by an Agilent 1260 HPLC equipped
with a refractive index detector.
The HDO reaction was carried out in a 316 L stainless steel tubular
reactor described in our previous works [16,17]. Prior to reaction, the
catalysts (1.8 g) were reduced in situ by H2 flow at 773 K for 2 h. After
cooling down the reactor to 623 K and increasing the system pressure
to 6 MPa, the aldol condensation product purified by vacuum distillation
was pumped into the reactor at 0.04 mL min−1 with hydrogen at a flow
rate of 120 mL min−1. The products from the outlet of the reactor be-
come two phases in a gas-liquid separator. The gaseous products passed
Among the solid base catalysts investigated in this work, CaO and
KF/Al2O3 exhibited the highest activity for the aldol condensation of fur-
fural and 3-pentanone (see Fig. 1). Over them, about 60% 1A yields were
achieved after reacting at 443 K for 8 h. As we know, CaO is the main
component of quick lime. Compared with KF/Al2O3, CaO has lower
price, wider availability and lower toxicity. Therefore, we believe that
CaO is a promising catalyst in future application. Under the same reac-
tion conditions, 1A yield up to 40% could also be obtained over CaO cat-
alyst at the initial 3-pentanone/furfural molar ratio of 1:1. This result
further proved that the reaction can be conducted under solvent-free
conditions.
The 1A as obtained exists as liquid at room temperature, which
makes it possible to be directly used in HDO process. In practical appli-
cation, this is very advantageous because (1) the use of organic solvent
can be avoided, which will reduce the cost and energy consumption

F. Chen et al. / Catalysis Communications 59 (2015) 229–232
231
(a)
Conversion of furfural
Yield of 1A
80
60
40
20
0
(b)
Fig. 1. Conversions of furfural and yields of 1A over different solid base catalysts. Reaction
conditions: 2.60 g (27 mmol) furfural, 4.67 g (54 mmol) 3-pentanone and 0.4 g catalyst,
443 K, 8 h.
from the evaporation of solvent, and (2) the low-temperature hydroge-
nation step can be merged with HDO step, which will also improve the
energy and facility efficiency.
Analogously, we investigated the aldol condensation of furfural with
5-nonanone (the ketonization product of valeric acid which can be ob-
tained from the dehydration/hydrogenation of levulinic acid [10,19])
over CaO catalyst. From the analysis of HPLC, NMR and GC-MS (see
Figs. S3, S4 and S5 in supporting information), 4-(furan-2-ylmethylene)
nonan-5-one (i.e., 1B in Scheme 1), which also exists as liquid state at
room temperature, was identified as the main product. The liquid state
of 1A and 1B can be explained by their branched structures which are de-
termined by the positions of carbonyl group in 3-pentanone and 5-
nonanone molecules. Compared with 3-pentanone, 5-nonanone is less
reactive (see Fig. 2), which can be explained by the greater steric effect
of ethyl group than that of methyl group.
Fig. 3. (a) Carbon yields of different alkanes and (b) carbon percentages of 4-methyl-nonane
in jet fuel range alkanes over the M/SiO2 (M = Fe, Cu, Co, Ni, Pt) catalysts. Reaction condi-
tions: 623 K, 6 MPa, 1.8 g catalyst, liquid feedstock 1A flow rate 0.04 mL min−1, hydrogen
flow rate 120 mL min−1
.
SiO2 demonstrated the highest selectivity to 4-methyl-nonane (see
Fig. 3b). The higher HDO activity of Ni/SiO2 can be rationalized by
the higher efficiency of Ni for hydrogenation or hydrogenolysis,
while the great difference of Ni/SiO2 and Cu/SiO2 in the selectivity
to 4-methyl-nonane can be explained by their activities in decar-
bonylation [20]. According to Scheme S1 in supporting information, al-
dehyde group can be generated from the hydrolysis reaction of 1A
with the water produced during the HDO process. This aldehyde
group can be removed through decarbonylation over metal catalysts.
It is interesting that high HDO activity and better selectivity to 4-
methyl-nonane can be obtained simultaneously over bimetallic Ni-
Cu/SiO2 catalysts (see Fig. 4). The best result was achieved over
Ni1Cu4/SiO2 catalyst. Over this catalyst, larger carbon percentage of
4-methyl-nonane in diesel range alkanes and slightly higher carbon
yield to jet fuel range alkanes were obtained than those over Ni/SiO2,
which can be interpreted by the restraining of decarbonylation over
Ni/SiO2 in the presence of Cu. Analogous to what has been observed
by Resasco's group over Pd-Cu catalyst [21], the presence of Cu may
decrease the stability of side on η2-(C-O) aldehyde species over Ni cat-
alyst and inhibit the formation of acyl intermediate. Consequently, the
Ni-Cu/SiO2 catalysts are less active for decarbonylation. Compared
with the Pd-Fe/C catalyst used in our previous work [16], the Ni-Cu/
SiO2 catalyst has lower cost, wider availability, and lower HDO temper-
ature (623 K vs. 643 K). These characters are advantageous in real
application.
3.2. Hydrodeoxygenation (HDO)
The purified 1A was directly hydrodeoxygenated over a series of
catalysts. According to Fig. 3a, Ni/SiO2 exhibited the best HDO activity
among the M/SiO2 (M = Fe, Co, Ni, Cu, Pt) catalysts. Over it, high 1A
conversion (100%) and carbon yield to alkanes (95%) were achieved
at 623 K and 6 MPa H2. The sequence for the carbon yields to the jet-
fuel-range alkanes over monometallic catalysts is Ni/SiO2 ≈ Pt/SiO2 N
Co/SiO2 N Cu/SiO2 N Fe/SiO2. According to the GC-MS analysis, the
jet-fuel-range alkanes obtained in this work were composed of 4-
methyl-nonane and 4-methyl-octane. Both of them have low freezing
points (174.1 K and 159.8 K) and can be directly used as jet fuel with-
out isomerization. Among the investigated M/SiO2 catalysts, the Ni/
SiO2 exhibited the highest selectivity to 4-methyl-octane, while Cu/
Conversion of furfural
Yield of 1A or 1B
80
60
40
20
0
4. Conclusions
Furfural & 3-Pentanone
Furfural & 5-Nonanone
The solvent-free aldol condensation of furfural and 3-pentanone
followed by hydrodeoxygenation (HDO) is a promising route for the di-
rect synthesis of jet fuel range branched alkanes. Compared with the
Fig. 2. Conversions of furfural and yields of 1A or 1B over CaO catalyst. Reaction conditions:
2.60 g (27 mmol) furfural, 4.67 g (54 mmol) 3-pentanone (or 1.83 g (20 mmol) furfural,
5.43 g (40 mmol) 5-nonanone), 0.4 g CaO catalyst, 443 K, 8 h.

232
F. Chen et al. / Catalysis Communications 59 (2015) 229–232
Acknowledgments
(a)
This work was supported by the National Natural Science Founda-
tion of China (grant nos. 21106143, 21277140 and 21202163), 100-
talent project of Dalian Institute of Chemical Physics.
Appendix A. Supplementary data
Supplementary data to this article can be found online at http://dx.
doi.org/10.1016/j.catcom.2014.11.003.
References
[1] G.W. Huber, S. Iborra, A. Corma, Chem. Rev. 106 (2006) 4044–4098.
[2] A. Corma, S. Iborra, A. Velty, Chem. Rev. 107 (2007) 2411–2502.
[3] M. Besson, P. Gallezot, C. Pinel, Chem. Rev. 114 (2013) 1827–1870.
[4] G.W. Huber, J.N. Chheda, C.J. Barrett, J.A. Dumesic, Science 308 (2005) 1446–1450.
[5] J.Q. Bond, D.M. Alonso, D. Wang, R.M. West, J.A. Dumesic, Science 327 (2010)
1110–1114.
(b)
[6] A. Corma, O. de la Torre, M. Renz, N. Villandier, Angew. Chem. Int. Ed. 50 (2011)
2375–2378.
[7] J.Q. Bond, A.A. Upadhye, H. Olcay, G.A. Tompsett, J. Jae, R. Xing, D.M. Alonso, D.
Wang, T.Y. Zhang, R. Kumar, A. Foster, S.M. Sen, C.T. Maravelias, R. Malina, S.R.H.
Barrett, R. Lobo, C.E. Wyman, J.A. Dumesic, G.W. Huber, Energy Environ. Sci. 7
(2014) 1500–1523.
[8] J.-P. Lange, E. van der Heide, J. van Buijtenen, R. Price, ChemSusChem 5 (2012)
150–166.
[9] R. Xing, A.V. Subrahmanyam, H. Olcay, W. Qi, G.P. van Walsum, H. Pendse, G.W.
Huber, Green Chem. 12 (2010) 1933–1946.
[10] R.M. West, Z.Y. Liu, M. Peter, J.A. Dumesic, ChemSusChem 1 (2008) 417–424.
[11] H. Olcay, A.V. Subrahmanyam, R. Xing, J. Lajoie, J.A. Dumesic, G.W. Huber, Energy En-
viron. Sci. 6 (2013) 205–216.
[12] W. Xu, Q. Xia, Y. Zhang, Y. Guo, Y. Wang, G. Lu, ChemSusChem 4 (2011) 1758–1761.
[13] A.D. Sutton, F.D. Waldie, R.L. Wu, M. Schlaf, L.A. Silks, J.C. Gordon, Nat. Chem. 5
(2013) 428–432.
[14] J.C. Serrano-Ruiz, J.A. Dumesic, ChemSusChem 2 (2009) 581–586.
[15] Y.L. Wang, W.P. Deng, B.J. Wang, Q.H. Zhang, X.Y. Wan, Z.C. Tang, Y. Wang, C. Zhu,
Z.X. Cao, G.C. Wang, H.L. Wan, Nat. Commun. 4 (2013) 2141.
[16] J. Yang, N. Li, G. Li, W. Wang, A. Wang, X. Wang, Y. Cong, T. Zhang, ChemSusChem 6
(2013) 1149–1152.
Fig. 4. (a) Carbon yields of different alkanes and (b) carbon percentages of 4-methyl-
nonane in jet fuel range alkanes over the Ni/SiO2, Ni-Cu/SiO2 and Cu/SiO2 catalysts. Reac-
tion conditions: 623 K, 6 MPa, 1.8 g catalyst, liquid feedstock 1A flow rate 0.04 mL min−1
,
hydrogen flow rate 120 mL min−1
.
[17] J. Yang, N. Li, G. Li, W. Wang, A. Wang, X. Wang, Y. Cong, T. Zhang, Chem. Commun.
50 (2014) 2572–2574.
[18] C. Murugan, H.C. Bajaj, R.V. Jasra, Catal. Lett. 137 (2010) 224–231.
[19] J.C. Serrano-Ruiz, A. Pineda, A.M. Balu, R. Luque, J.M. Campelo, A.A. Romero, J.M.
Ramos-Fernandez, Catal. Today 195 (2012) 162–168.
[20] S. Sitthisa, D.E. Resasco, Catal. Lett. 141 (2011) 784–791.
[21] S. Sitthisa, T. Pham, T. Prasomsri, T. Sooknoi, R.G. Mallinson, D.E. Resasco, J. Catal. 280
(2011) 17–27.
literature route, the new protocol has many advantages such as solvent-
free, fewer steps and direct preparation of jet fuel range branched al-
kanes without isomerization.