Catalytic Production of Branched Small Alkanes from Biohydrocarbons

Article abstract of DOI:10.1002/cssc.201500375

Squalane, C30 algae-derived branched hydrocarbon, was successfully converted to smaller hydrocarbons without skeletal isomerization and aromatization over ruthenium on ceria (Ru/CeO₂). The internal CH₂-CH₂ bonds located between branches are preferably dissociated to give branched alkanes with very simple distribution as compared with conventional methods using metal-acid bifunctional catalysts.

Full text of DOI:10.1002/cssc.201500375

DOI: 10.1002/cssc.201500375
Communications
Catalytic Production of Branched Small Alkanes from
Biohydrocarbons
Shin-ichi Oya,[a] Daisuke Kanno,[a] Hideo Watanabe,[b] Masazumi Tamura,[a]
Yoshinao Nakagawa,*[a] and Keiichi Tomishige*[a]
Squalane, C30 algae-derived branched hydrocarbon, was suc-
cessfully converted to smaller hydrocarbons without skeletal
isomerization and aromatization over ruthenium on ceria (Ru/
CeO2). The internal CH2ÀCH2 bonds located between branches
are preferably dissociated to give branched alkanes with very
simple distribution as compared with conventional methods
using metal-acid bifunctional catalysts.
located between branches are preferably dissociated to give
branched alkanes with very simple distributions of isomers.
First, we used n-hexadecane as a model substrate, to identi-
fy promising catalysts (Figure 1). Among carbon-supported cat-
alysts (Figure 1a–d), Ru/C showed the highest activity. The ac-
Liquid alkanes, which are important components of fuels and
chemicals, are supplied by petroleum refining. Considering the
diminishing reserves of crude oil, biomass, as renewable organ-
ic carbon resources, is expected to be a promising substitute.[1]
The production of liquid alkanes has been attempted from
lignocellulose-derived substrates such as levulinic acid, furanic
compounds, and cellulose.[2–8] Some plants or microorganisms
produce pure (bio)hydrocarbons, such as terpenes.[9,10] One ex-
ample is squalene (2,6,10,14,18,22-hexaen-2,6,10,15,19,23-hex-
amethyltetracosane), high amounts of which have been report-
ed to accumulate in Aurantiochytrium microalgae strains.[11,12]
Typically, biohydrocarbons, in particular algae-derived ones, are
large molecules with many branches. While some amount of
squalene (derived from sharks) has been used in cosmetics,[11]
biohydrocarbons need to be refined into smaller molecules for
most other uses such as biofuel. Conventional methods for re-
fining large hydrocarbons typically use solid acids in combina-
tion with noble-metal catalysts, and many side reactions can
occur such as isomerization or coke formation.[13,14] Although
isomerization is beneficial for fuel production from linear-
alkane-based feedstock such as petroleum, isomerization is un-
desirable in the case of branched algal hydrocarbons and only
complicates the reaction mixture.
Figure 1. n-Hexadecane hydrogenolysis over various catalysts. Conditions: n-
hexadecane, 2.26 g; catalyst, 10–100 mg (5 wt% metal except Ir); H2, 6 MPa;
513 K. Gray and white bars represent linear and branched alkanes, respec-
tively. S: Selectivity based on carbon, C. N.: carbon number.
tivity of Ru/C was higher than that of Ir/SiO2 (Figure 1e), which
has been reported to be selective in hydrogenolysis of internal
CÀC bonds.[15–18] Ruthenium catalysts on other supports (Ru/
CeO2, Ru/SiO2; Figure 1 f and g) were also tested, and the reac-
tion times and catalyst amounts were adjusted to similar low
levels of conversion, in order to compare selectivity. The order
of activity was Ru/SiO2 (TOF 1.8”102 hÀ1)>Ru/C (79 hÀ1)>Ru/
CeO2 (39 hÀ1)>Ir/SiO2 (12 hÀ1)>Rh/C (2.8 hÀ1)@Pd/C, Pt/C
(<0.01 hÀ1). The products were mainly n-alkanes for all of the
tested active catalysts, and branched alkanes were hardly ob-
served. This result is in contrast to those achieved over solid-
acid catalysts, where branched alkanes are the main products
through isomerization.[14,19–25] In the cases of Rh/C, Ru/SiO2, and
Ru/C, more methane and n-pentadecane were formed than
C2–C14 alkanes, suggesting that these catalysts preferably
cleave terminal CÀC bonds before internal ones. In the litera-
ture, similar selectivity trends to Rh/C, Ru/SiO2, and Ru/C cata-
lysts have been reported for most metal-catalyzed CÀC hydro-
genolysis.[15,26,27] However, these trends are not desirable be-
cause the branches of algal hydrocarbons are methyl groups,
Herein, we show that a Ru/CeO2 catalyst can produce small
alkanes from biohydrocarbons by regioselective CÀC hydroge-
nolysis, without isomerization and coke formation. By using
this catalyst and molecular hydrogen, internal CH2ÀCH2 bonds
[a] S.-i. Oya, D. Kanno, Dr. M. Tamura, Dr. Y. Nakagawa, Prof. Dr. K. Tomishige
Department of Applied Chemistry, School of Engineering
Tohoku University
6-6-07 Aoba, Aramaki, Aoba-ku, Sendai, 980-8579 (Japan)
E-mail: yoshinao@erec.che.tohoku.ac.jp
tomi@erec.che.tohoku.ac.jp
[b] Dr. H. Watanabe
Faculty of Life and Environmental Sciences
University of Tsukuba
1-1-1 Tennodai, Tsukuba, Ibaraki, 305-8572 (Japan)
Supporting Information for this article is available on the WWW under
http://dx.doi.org/10.1002/cssc.201500375.
ChemSusChem 2015, 8, 2472 – 2475
2472
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communications
and cleavage of the terminal CÀC bond removes the branch.
On the other hand, on Ru/CeO2 as well as Ir/SiO2 the dissocia-
tion of terminal CÀC bonds was not more preferable than
those of other CÀC bonds.
We applied Ru/CeO2 to the hydrogenolysis of squalane
(2,6,10,15,19,23-hexamethyltetracosane), which can be easily
obtained by hydrogenation of squalene. Figure 2 shows GC
Figure 3. Squalane hydrogenolysis over Ru/CeO2. (a) Selectivity patterns.
Conditions: squalane, 4.23 g (10 mmol); Ru/CeO2, 50 mg; H2, 6 MPa; 513 K;
6 h. Gray bars represent the dominating isomer of products with each
carbon number, and white bars represent other isomers. S.: Selectivity based
on carbon, C.N.: carbon number. (b) Proposed CÀC dissociation positions.
the branches were selectively dissociated. The six branches in
the squalane were maintained in the main products. On the
other hand, Ru/SiO2 gave products with carbon numbers
below C30 (Supporting Information, Figure S3a). The reactions
at tertiary carbon, such as those giving C6–C8, C11–C13, and
C17-19 products, decrease the number of branches. Hydroge-
nolysis of squalane over Ir/SiO2 was also tested (Figure S3b).
The selectivity trend was intermediate between Ru/SiO2 and
Ru/CeO2. However, the activity (TOF 2.7 hÀ1) was much lower
than those of ruthenium catalysts (Ru/CeO2: 19 hÀ1; Ru/SiO2:
1.0”102 hÀ1). This result agrees with literature data, reporting
that iridium catalysts show very low activity towards the hy-
drogenolysis of branched alkanes.[18]
Figure 2. GC charts of liquid samples (diluted with mesitylene) of squalane
hydrogenolysis. Conditions: squalane 4.23 g (10 mmol); catalyst, (a) 5.5 wt%
Pt/H-USY 20 mg, (b) 5 wt% Ru/CeO2 100 mg; H2, 6 MPa; 513 K; (a) 3 h,
(b) 6 h. I and i.s. represent intensity and internal standard material (n-octaco-
sane), respectively.
Figure 4 shows the time course of the hydrogenolysis of
squalane over Ru/CeO2. The product molecules became smaller
at longer reaction times. The ratio of branched alkanes to n-al-
charts of liquid samples from the reaction of squalane over Pt/
H-USY and Ru/CeO2. Pt/H-USY is a metal-acid bifunctional cata-
lyst and was used as an example of a petroleum hydrocracking
catalyst. The reaction over Pt/H-USY gave a very complex mix-
ture of products, with>200 kinds of molecules (Figure 2a). In
literature studies of squalane hydrocracking, we noted that
products are also very complex and usually analyzed roughly
by boiling point.[24,28,29] On the other hand, the number of GC
peaks on Ru/CeO2 was much smaller than Pt/H-USY (Fig-
ure 2b).
Figure 3 shows the product distribution over Ru/CeO2. The
main products have carbon numbers of 9, 10, 14, 15, 16, 20,
21, 25, and 26. In particular, each structure with carbon
number of 4, 5, 6, 9, 10, and 15 was verified by GC-MS analysis
using standard samples (Supporting Information, Figure S1). All
the structures of these products were substructures of squa-
lane. The amounts of products whose formations involve iso-
merization was always very small. The absence of unsaturated
Figure 4. Reaction progress of squalane hydrogenolysis over Ru/CeO2. Con-
ditions: squalane, 4.23 g (10 mmol); Ru/CeO2, 50 or 100 mg; H2, 6 MPa;
513 K; 3–48 h. S: Selectivity based on carbon; C: conversion. Wcat: weight of
catalyst.
1
or cyclic products was confirmed by H NMR (Supporting Infor-
mation, Figure S2) and GC-MS(CI). The structures of the main
products indicate that the CH2ÀCH2 bonds located between
ChemSusChem 2015, 8, 2472 – 2475
www.chemsuschem.org
2473
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communications
kanes in ꢀC4 components was always >10. When the conver-
sion of squalane reached almost 100% (catalyst 0.1 g, time
18 h), the main products were tribranched C14–16 and di-
branched C9–10. The former group can be used as compo-
nents of diesel fuel or jet fuel.[30,31] A good yield of the C9–10
fraction was obtained after slightly further reaction. Further re-
action gave gasoline-range (C5–C8) products. All products in
this reaction were stable saturated hydrocarbons and have
very low melting points because of the branches.[30,32] These
properties are ideal for transportation fuels. The reaction prog-
ress of the hydrogenolysis of squalane over Ru/SiO2 is shown
in Figure S4 (Supporting Information). The selectivities to C14–
16 and C9–10 on Ru/SiO2 were not higher than 32%, and they
were lower than the maximum values on Ru/CeO2 (ca. 40%;
Figure 4). These trends reflect the wider product distribution
on Ru/SiO2 compared to Ru/CeO2.
ruthenium precursors supported on CeO2. On the other hand,
larger ruthenium particles were formed on other supports (<
1/5 dispersion relative to Ru/CeO2; Supporting Information,
Table S2).
The unique selectivity of the Ru/CeO2 catalyst might be due
to the high dispersion of ruthenium metal. In fact, when Ru/
CeO2 was heated at 773 K before catalytic use to increase the
particle size (ca. 2.5 nm from XRD; coordination number of Ru–
Ru=10.0 according to Ru K-edge EXAFS), the activity and se-
lectivity became similar to Ru/SiO2 (Supporting Information,
Figure S9). Clarifying the mechanism that induces the selectivi-
ty requires further investigation. Preparing further small ruthe-
nium particles might give even higher regioselectivity, which is
also one target of our further studies.
Experimental Section
We re-used the Ru/CeO2 catalyst in squalane hydrogenolysis,
and almost the same result was obtained as for the fresh cata-
lyst even after 4 runs (conversion: 72%!78%!72%!75%!
75%; selectivities: Supporting Information, Figure S5). Thermal
gravimetric (TG) analysis of the used catalyst showed that the
amount of coke formed was small (3 wt% loss from the cata-
lyst after 5 uses; 0.02%-C of used squalane). The amount of
ruthenium leaching, measured with ICP–OES, was below the
detection limit (<0.5% of total ruthenium).
Carbon-supported noble-metal catalysts (Pt/C, Pd/C, Rh/C and Ru/
C; 5 wt%) were purchased from Wako Pure Chemical Industries.
Ru/SiO2 and Ru/CeO2 (Ru: 5 wt%) were prepared by impregnating
SiO2 (BET 535 m2 gÀ1) and CeO2 (calcined at 873 K; 86 m2 gÀ1) with
Ru(NO)(NO3)3Àx(OH)x in HNO3aq. Pt/H-USY (Pt: 5.5 wt%) was pre-
pared by impregnating H-USY (Si/Al=6.3) with H2PtCl6aq. Ir/SiO2
(Ir: 4 wt%) was prepared by impregnating SiO2 with H2IrCl6 aq. All
the prepared catalysts were dried at 383 K for 12 h. After drying,
Ru/SiO2, Ru/CeO2, and Pt/H-USY were heated in flowing N2 at 573 K
for 1 h. Ir/SiO2 was calcined at 773 K for 3 h. Catalytic reaction was
carried out in a 190 mL autoclave. 3.5 MPa H2 was filled at r.t., and
the pressure became 6 MPa after ~60 min heating to 513 K. The re-
action time was counted after the heating. After appropriate reac-
tion time, the reactor was rapidly cooled. Products in both gas and
liquid phases were quantified with FID-GC, and qualified with GC-
MS in EI and CI modes. TOF (turnover frequency; [hÀ1]) was calcu-
lated by (increase of the total number of hydrocarbon molecules
[mol])/(total number of noble metal atom [mol])/(reaction time [h]).
The Ru/CeO2 catalyst was characterized by various tech-
niques. The profile of temperature-programmed reduction
with H2 had a signal at 323–493 K and the area corresponded
to the reduction of RuO2 to ruthenium metal (Supporting Infor-
mation, Figure S6 and Table S1). The ruthenium species in Ru/
CeO2 was reduced to Ru0 during the reaction. The X-ray diffrac-
tion (XRD) pattern of Ru/CeO2 did not show peaks for rutheni-
um metal (Supporting Information, Figure S7). The amount of
H2 adsorped onto Ru/CeO2 was 0.9H atoms per total rutheni-
um atoms (Supporting Information, Table S2). The ruthenium
K-edge EXAFS spectrum (Supporting Information, Figure S8)
could be fitted by one Ru–Ru shell with a coordination
number of ~5, which is much smaller than that of the ideal
hcp structure of ruthenium metal (12) and that of Ru/SiO2
(10.5) (Table 1). These results indicate the very small size
(<1.5 nm) of the ruthenium metal particles on Ru/CeO2. In the
literature, several reduced Ru/CeO2 catalysts have been pre-
pared for other reactions than CÀC hydrogenolysis, such as CO
conversions[33–35] and ammonia synthesis.[36] According to the
reported characterization results, similar highly dispersed
ruthenium particles (ꢁ2 nm) were formed by reduction of
Acknowledgements
This work was supported by the project of Next-generation Ener-
gies for Tohoku Recovery. We thank Prof. M. M. Watanabe and
Dr. M. Yoshida for providing the photos of Aurantiochytrium spe-
cies for table of content. We also thank Shimadzu Corporation
for GC-MS analysis.
Keywords: biomass
·
hydrogenolysis
·
metal oxides
·
microalgae · ruthenium
[1] a) G. W. Huber, S. Iborra, A. Corma,
Chem. Rev. 2006, 106, 4044–4098;
b) J. N. Chheda, G. W. Huber, J. A.
Dumesic, Angew. Chem. Int. Ed.
2007, 46, 7164–7183; Angew.
Chem. 2007, 119, 7298–7318; c) R.
Rinaldi, F. Schüth, Energy Environ.
Sci. 2009, 2, 610–626; d) A.-L. Mar-
shall, P. J. Alaimo, Chem. Eur. J.
2010, 16, 4970–4980; e) H. Ko-
bayashi, H. Ohta, A. Fukuoka, Catal.
Sci. Technol. 2012, 2, 869–883;
f) M. Besson, P. Gallezot, C. Pinel,
Chem. Rev. 2014, 114, 1827–1870.
Table 1. Curve fitting results of Ru K-edge EXAFS after reduction.
Catalyst
Shells
CN[a]
R/0.1 nm[b]
s/0.1 nm[c]
DE0 [eV][d]
Rf [%][e]
Ru/CeO2
Ru/CeO2 after reaction
Ru/SiO2
Ru–Ru
Ru–Ru
Ru–Ru
Ru–Ru
4.7 (Æ0.6)
2.61 (Æ0.01)
2.60 (Æ0.01)
2.66 (Æ0.00)
2.68
0.097 (Æ0.01)
0.083 (Æ0.01)
0.080 (Æ0.03)
0.06
À1.10 (Æ1.65)
À8.35 (Æ1.61)
À0.05 (Æ0.74)
0
1.4
1.3
0.21
–
5.2 (Æ0.6)
10.5 (Æ0.5)
12
Ru powder
[a] Coordination number. [b] Bond distance. [c] Debye–Waller factor. [d] Difference in the origin of photoelec-
tron energy between the reference and the sample. [e] Residual factor. Fourier filtering range: 0.169–0.288 nm.
ChemSusChem 2015, 8, 2472 – 2475
www.chemsuschem.org
2474
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communications
[2] a) R. Palkovits, Angew. Chem. Int. Ed. 2010, 49, 4336–4338; Angew.
Chem. 2010, 122, 4434–4436; b) M. J. Climent, A. Corma, S. Iborra,
Green Chem. 2014, 16, 516–547; c) Y. Nakagawa, S. Liu, M. Tamura, K.
Tomishige, ChemSusChem 2015, 8, 1114–1132.
[3] G. W. Huber, J. N. Chheda, C. J. Barrett, J. A. Dumesic, Science 2005, 308,
1446–1450.
[4] R. Xing, A. V. Subrahmanyam, H. Olcay, W. Qi, G. P. van Walsum, H.
Pendse, G. W. Huber, Green Chem. 2010, 12, 1933–1946.
[5] A. D. Sutton, F. D. Waldie, R. Wu, M. Schlaf, L. A. Silks, J. C. Gordon, Nat.
Chem. 2013, 5, 428–432.
[20] M. Roussel, J.-L. Lemberton, M. Guisnet, T. Cseri, E. Benazzi, J. Catal.
2003, 218, 427–437.
[21] A. Ishihara, K. Inui, T. Hashimoto, H. Nasu, J. Catal. 2012, 295, 81–90.
[22] A. Corma, A. Martinez, S. Pergher, S. Peratello, C. Perego, G. Bellusi,
Appl. Catal. A 1997, 152, 107–125.
[23] M. Busto, J. M. Grau, C. R. Vera, Appl. Catal. A 2010, 387, 35–44.
[24] M. Mitsios, D. Guillaume, P. Galtier, D. Schweich, Ind. Eng. Chem. Res.
2009, 48, 3284–3292.
[25] R. de Haan, G. Joorst, E. Mokoena, C. P. Nicolaides, Appl. Catal. A 2007,
327, 247–254.
[6] S. Liu, M. Tamura, Y. Nakagawa, K. Tomishige, ACS Sustainable Chem.
Eng. 2014, 2, 1819–1827.
[26] F. Regali, L. F. Liotta, A. M. Venezia, M. Boutonnet, S. Järås, Appl. Catal. A
2014, 469, 328–339.
[7] J. J. Bozell, G. R. Petersen, Green Chem. 2010, 12, 539–554.
[8] B. Op de Beeck, M. Dusselier, J. Geboers, J. Holsbeek, E. MorrØ, S.
Oswald, L. Giebeler, B. F. Sels, Energy Environ. Sci. 2015, 8, 230–240.
[9] P. Metzger, C. Largeau, Appl. Microbiol. Biotechnol. 2005, 66, 486–496.
[10] T. M. Mata, A. A. Martins, N. S. Caetano, Renewable Sustainable Energy
Rev. 2010, 14, 217–232.
[27] C. J. Machiels, R. B. Anderson, J. Catal. 1979, 58, 268–275.
ˇ
´
[28] K. P. de Jong, J. Zecevic, H. Friedrich, P. E. de Jongh, M. Bulut, S. van
Donk, R. Kenmogne, A. Finiels, V. Hulea, F. Fajula, Angew. Chem. Int. Ed.
2010, 49, 10074–10078; Angew. Chem. 2010, 122, 10272–10276.
[29] J. Francis, E. Guillon, N. Bats, C. Pichon, A. Corma, L. J. Simon, Appl.
Catal. A 2011, 409–410, 140–147.
[11] E. Naziri, F. Mantzouridou, M. Z. Tsimidou, Lipid Technol. 2011, 23, 270–
273.
[30] T. G. Smagala, E. Christensen, K. M. Christison, R. E. Mohler, E. Gjersing,
R. L. McCormick, Energy Fuels 2013, 27, 237–246.
[12] K. Kaya, A. Nakazawa, H. Matsuura, D. Honda, I. Inouye, M. M. Watanabe,
Biosci. Biotechnol. Biochem. 2011, 75, 2246–2248.
[31] T. C. R. Brennan, C. D. Turner, J. O. Krçmer, L. K. Nielsen, Biotechnol.
Bioeng. 2012, 109, 2513–2522.
[13] M. S. Rana, V. Sµmano, J. Ancheyta, J. A. I. Diaz, Fuel 2007, 86, 1216–
1231.
[14] J. Weitkamp, ChemCatChem 2012, 4, 292–306.
[15] A. MajestØ, S. Balcon, M. GuØrin, C. Kappenstein, Z. Paµl, J. Catal. 1999,
187, 486–492.
[16] a) P. T. Do, W. E. Alvarez, D. E. Resasco, J. Catal. 2006, 238, 477–488;
b) M. Santikunaporn, W. E. Alvarez, D. E. Resasco, Appl. Catal. A 2007,
325, 175–187.
[32] Diesel Fuels Technical Reviews, Chevron Corporation. Available from
http://www.chevron.com/productsservices/fuels/. (Accessed May 2015).
[33] L. A. Bruce, M. Hoang, A. E. Hughes, T. W. Turney, J. Catal. 1998, 178,
84–93.
[34] A. Satsuma, M. Yanagihara, J. Ohyama, K. Shimizu, Catal. Today 2013,
201, 62–67.
[35] X. Chen, J. J. Delgado, J. M. Gatica, S. Zerrad, J. M. Cies, S. Bernal, J.
Catal. 2013, 299, 272–283.
[17] A. Haas, S. Rabl, M. Ferrari, V. Calemma, J. Weitkamp, Appl. Catal. A
2012, 425–426, 97–109.
[36] Y. Izumi, Y. Iwata, K. Aika, J. Phys. Chem. 1996, 100, 9421–9428.
[18] a) D. W. Flaherty, E. Iglesia, J. Am. Chem. Soc. 2013, 135, 18586–18599;
b) D. W. Flaherty, D. D. Hibbitts, E. Iglesia, J. Am. Chem. Soc. 2014, 136,
9664–9676.
Received: March 13, 2015
[19] I. E. Maxwell, Catal. Today 1987, 1, 385–413.
Published online on June 11, 2015
ChemSusChem 2015, 8, 2472 – 2475
www.chemsuschem.org
2475
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim