Cerium oxide as a catalyst for the ketonization of aldehydes: Mechanistic insights and a convenient way to alkanes without the consumption of external hydrogen

Article abstract of DOI:10.1039/c6gc03511f

The ketonization of aldehydes joins two molecules, with n carbon atoms each, to a ketone with 2n - 1 carbon atoms. When employing cerium oxide as a catalyst with nano-sized crystals (<15 nm) the ketone can be obtained in almost 80% yield. In addition, other ketones are observed so that the total ketone selectivity reached almost 90%. Water is consumed during the reaction when the aldehyde is oxidized to the corresponding carboxylic acid, which is established as a reaction intermediate, co-producing hydrogen. Consequently, water has to be co-fed in the reaction to enhance the reaction rate and to improve the catalyst stability with time on stream. In contrast to zirconium oxide which possesses catalytic activity for the aldol condensation liberating water, with cerium oxide water is not abundant on the surface and the reaction kinetics show that the reaction rate depends on the concentration of the water in the gas-phase, in addition to the dependence on the gas-phase concentration of the aldehyde. The liberated hydrogen can be consumed in the hydrodeoxygenation of the ketone product. Doing so, when starting from heptanal, a biomass derived aldehyde, an alkane mixture was obtained with almost 90% diesel content. For the whole cascade reaction with five single steps no reagents are necessary and the only by-product is one molecule of innocuous carbon dioxide (related to two molecules of aldehyde). This shows that cerium oxide possesses a big potential to convert biomass derived aldehydes into biofuels in a very sustainable way.

Full text of DOI:10.1039/c6gc03511f

View Article Online
View Journal
Green
Chemistry
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: L. M. Orozco, M.
Renz and A. Corma, Green Chem., 2017, DOI: 10.1039/C6GC03511F.
This is an Accepted Manuscript, which has been through the
Volume 18 Number
7 7 April 2016 Pages 1821–2242
Royal Society of Chemistry peer review process and has been
accepted for publication.
Green
Chemistry
Accepted Manuscripts are published online shortly after
Cutting-edge research for a greener sustainable future
www.rsc.org/greenchem
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
ISSN 1463-9262
CRITICAL REVIEW
G. Chatel et al.
Heterogeneous catalytic oxidation for lignin valorization into valuable
chemicals: what results? What limitations? What trends?
rsc.li/green-chem

Please do not adjust margins
Page 1 of 16
Green Chemistry
View Article Online
DOI: 10.1039/C6GC03511F
Green Chemistry
ARTICLE
Cerium oxide as catalyst for the ketonization of aldehydes:
Mechanistic insights and a convenient way to alkanes without
consumption of external hydrogen
Received 00th January 20xx,
Accepted 00th January 20xx
Lina M. Orozco,a Michael Renz*a and Avelino Corma*a
DOI: 10.1039/x0xx00000x
The ketonization of aldehydes joins two molecules, with n carbon atoms each, to a ketone with 2n–1 carbon atoms. When
employing cerium oxide as catalyst with nano-sized crystals (<15 nm) the ketone can be obtained in almost 80% yield. In
addition, other ketones are observed so that the total ketone selectivity reached almost 90%. Water is consumed during the
reaction when the aldehyde is oxidized to the corresponding carboxylic acid, which is established as a reaction intermediate,
co-producing hydrogen. Consequently, water has to be co-fed in the reaction to enhance reaction rate and to improve
catalyst stability with time on stream. In contrast to zirconium oxide which possesses catalytic activity for the aldol
condensation liberating water, with cerium oxide water is not abundant on the surface and the reaction kinetics show that
the reaction rate depends on the concentration of the water in the gas-phase, in addition to the dependence on the gas-
phase concentration of the aldehyde. The liberated hydrogen can be consumed in the hydrodeoxygenation of the ketone
product. Doing so, when starting from heptanal, a biomass derived aldehyde, an alkane mixture was obtained with almost
90% diesel content. For the whole cascade reaction with five single steps no reagents are necessary and the only by-product
is one molecule of innocuous carbon dioxide (related to two molecules of aldehyde). This shows that cerium oxide possesses
a big potential to convert biomass derived aldehydes into biofuels in a very sustainable way.
www.rsc.org/
derivatives. Manifold carbon-carbon forming reactions are
known for these compounds as e.g. the Guerbet reaction,3,4 the
aldol condensation of carbonyl compounds,5,6 the Claisen
condensation of esters,7,8 the ketonic decarboxylation of
carboxylic acids,9 and many others. On the other hand, the first
option, i.e. initial hydrodeoxygenation, is mainly applied for fuel
production.
Introduction
The formation of carbon-carbon bonds is an interesting task in
chemistry and in biomass conversion. Non-edible biomass
consists mainly of lignocellulose which is a composite material
out of three biopolymers: cellulose, hemicellulose and lignin.
The former are built up by acetal linkages which can be
hydrolysed relatively easily so that the monomers are accessible
from this feedstock. These monomers consist of
monosaccharides with five or six carbon atoms. This is a small
number of carbon atoms for a number of chemical products and
fuels and, consequently, two or more molecules have to be
joined.
For the formation of larger molecules with a low oxygen content
from monosaccharides, two strategies can be applied: a
complete hydrodeoxygenation to oxygen-free olefins with
subsequent oligomerization1,2 or partial deoxygenation
followed by more selective transformations. The latter can be
achieved with molecules with simple oxygen functionalities,
such as alcohols, aldehydes, ketones or carboxylic acid
Among the reactions for oxygen-containing molecules, the
ketonization of aldehydes is a less classical pathway for the
formation of carbon-carbon bonds (eq. 1). Aldehydes are a class
of organic compounds abundant in biomass derived mixtures
such as bio-oils or similar.10–15 In this reaction, two molecules of
aldehyde are transformed into a symmetrical ketone and an
oxygenated C1 fragment, carbon monoxide or carbon dioxide.
Although the reaction is carried out at high temperature (> 400
°C), many demands of Sustainable Chemistry are met as e.g. the
use of a catalyst and the absence of a solvent or further
reactants which minimizes wastes. As an additional benefit of
the reaction, the energy, which in general is wasted when an
aldehyde is oxidized to its corresponding carboxylic acid
employing a sacrificial oxidant, can be recovered in form of
hydrogen during the ketonization reaction. Recently, we have
shown that the reaction proceeds in two steps when employing
zirconium oxide as catalyst.16 First the aldehyde is oxidized to
the corresponding carboxylic acid (eq. 2) which is then
transformed into the ketone via the classic ketonic
decarboxylation (eq. 3).
a.Instituto de Tecnología Química, Universitat Politècnica de Valencia-Consejo
Superior de Investigaciones Científicas (UPV - CSIC), Av. de los Naranjos s/n, E-
46022 Valencia, Spain. Tel.: +34 96 387 7800, Fax: +34 96 387 9444,
acorma@itq.upv.es, mrenz@itq.upv.es
Electronic Supplementary Information (ESI) available: further experimental and
analytical details, HRTEM images of nanocrystals of CeO2, details on the kinetic data,
the distribution of alkanes produced in the cascade reaction and the McLafferty
rearrangement
of
7-tridecanone
to
produce
2-octanone.
See
DOI: 10.1039/x0xx00000x
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 1
Please do not adjust margins

Please do not adjust margins
Green Chemistry
Page 2 of 16
ARTICLE
Green Chemistry
aldehyde in side-reaction pathways, but alsoVietwheArtic“lewOantliener
DOI: 10.1039/C6GC03511F
supply”, required for the aldehyde ketonization and provided by
aldol condensation, might be critical.
When the alkanes are the desired products the ketone product
can be completely hydrodeoxygenated by a hydrogenation-
dehydration-hydrogenation sequence which is carried out over
a supported metal in combination with a weakly acidic support.
When considering the whole sequence from the two aldehyde
molecules (eq. 4), this reaction cascade resembles olefinations
of aldehydes following the Wittig, Peterson or Tebbe
procedure19 with subsequent double bond hydrogenation and
with the particularity that the carbon chain length is shortened
by one carbon atom. As the addition of sophisticated reagents
is not required, the atom efficiency is much better on the
expense of the loss of one carbon atom as carbon dioxide.
Therefore, the combination of these two reaction sequences
will be studied here with the aim to confirm that this cascade
reaction provides a useful alternative for the direct conversion
of aldehydes into alkanes.
Interestingly, the oxidation proceeds consuming water which
provides the necessary oxygen atom and a proton. In a surface-
assisted reaction an aldehyde-hydrate species is formed and
this species is able to transfer a hydride species to the surface
(Scheme 1). Combination of the latter with a proton liberates
molecular hydrogen. The carboxylic acid can then react with a
second molecule following the established mechanism,17
forming the carbon-carbon bond between
a double-
deprotonated and a dehydroxylated species. This intermediate
releases carbon dioxide and forms the final ketone product.
This mechanism has been established for the reaction with
zirconium oxide and the question which has to be still addressed
is if this mechanism applies in general for all catalysts or if it
changes for other catalytic materials. Therefore, herein an
alternative material, namely cerium oxide, has been chosen to
see if the mechanism is the same for both oxides, and how
different surface properties of cerium oxide can influence
catalytic results. One fundamental difference between
zirconium oxide and cerium oxide is that the former is a non-
reducible material whereas in the latter the metal oxide
possesses redox-properties and the metal atoms can change
their oxidation state between plus three and plus four. In
addition, catalytic activity induced by surface properties is
different. For instance, zirconium oxide facilitates aldol
condensation (with water as side product) whereas cerium
oxide does so in a less pronounced way.18 As a consequence, a
better selectivity for the ketonization reaction over cerium
oxide might be expected due to a lower consumption of the
Experimental
General
Heptanal and heptanoic acid were purchased from Sigma-
Aldrich and heptanal distilled under reduced pressure before
being used. Water was employed in deionized form. Cerium
oxide (CeO2-6nm and CeO2-11nm) were received from Rhodia
or purchased from Aldrich (CeO2-43nm) as a powder. m-ZrO2
and Al2O3 were purchased from Chempur as pellets.
Chloroplatinic acid hydrate (H2PtCl6
· 6H2O), Sodium
borohydride (NaBH4) and 5% Palladium on activated charcoal
(5% Pd/C) were bought from Aldrich.
Scheme 1. Proposed mechanism for the ketonization of aldehydes. An aldehyde molecule coordinates to a surface hydroxyl group changing the geometry at the
carbonyl carbon atom from trigonal planar to tetrahedral. In the latter configuration it is proposed that a hydride species is shifted to the surface which then
combines with a proton and desorbs as molecular hydrogen. The remaining carboxylate can then react via the established mechanism, after a second deprotonation,
with a second carboxylic acid molecule. The latter is activated by dehydroxylation and the carbon-carbon bond is formed by a nucleophilic attack onto this carbonyl
carbon atom. The formed beta-ketoacid is decarboxylated and the products, ketone and carbon dioxide, are desorbed and the surface hydroxyl group restored by
adsorption of one molecule of water.16
2 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 3 of 16
Green Chemistry
Please do not adjust margins
Green Chemistry
ARTICLE
CeO2 synthesis with different particle size
and 20 Pa. Temperature programmed reduction (TPR) for ceria
catalysts was carried out on a conventional flow apparatus
View Article Online
DOI: 10.1039/C6GC03511F
Ce(NO3)3 · 6H2O was used as the cerium source to obtain CeO2
nanoparticles following a literature procedure.20 To get CeO2-
5nm and CeO2-7nm, Ce(NO3)3 · 6H2O (13.4 g) and NaOH (221.6
g) were dissolved in 76 and 540 mL of deionized water,
respectively. The two solutions were mixed and stirred for 30
min forming a milky slurry. Subsequently, the mixture was
placed in a 1-L glass bottle and sealed tightly. Finally, the bottle
was heated statically in an oven to 100 °C for 24 h. After the
(Autochem 2910, Micromeritics).
A 0.3-g sample was
pretreated in an O2 (2%)/He flow at 550 °C for 1 h and cooled in
a He flow to room temperature, followed by Ar purge. The
sample was then reduced in a H2 (10%)/Ar flow. Temperature
was increased from room temperature to 950 °C at a constant
heating rate of 10 K·min–1 and held for 5 min. Water produced
during the reduction was eliminated with a frozen n-propanol
trap and the amount of consumed H2 was monitored by a
thermal conductivity detector (TCD). The FTIR measurement on
the catalyst CeO2-11nm was performed with a Nicolet iS10
spectrometer from Thermo Fisher equipped with a DTGS
hydrothermal treatment,
a fresh white precipitate was
separated by filtration, washed several times with deionized
water until neutral pH, dried at 100 °C in air overnight and
calcined at 450 °C for 4 h. Yellow powders (5.3 g, and 5.4 g
respectively, 40% yield) were obtained after calcination. To get
CeO2-12nm, the same procedure was employed but the mixture
was prepared using different quantities of the reagent:
Ce(NO3)3 · 6H2O (3.34 g) and NaOH (55.4 g), which were
dissolved in 19 and 135 mL of deionized water, respectively. The
reaction was carried out in a teflon-lined stainless steel
autoclave following the procedure mentioned before (1.2 g,
36% yield). The latter procedure was also applied to get CeO2-
29nm and CeO2-277nm but the hydrothermal treatment was
done at 150 °C and 200 °C, respectively. In the case of the CeO2-
29nm synthesis, 0.062 g NaOH was used and 0.36 g of solid
obtained (11% yield). CeO2-277nm was produced in 29% yield
(0.96 g).
−
1
detector using a vacuum cell (4 cm , 32 scans). The catalyst was
employed as self-supported wafers of 1 cm diameter and 22 mg
weight. For the D2O adsorption experiments the catalyst was
treated for 1.5 h at 120 °C and 1 h at 450 °C under vacuum
(8.6·10−2 Pa). After activation, the sample was cooled down to
room temperature followed by D2O dosing (1.55 mL gas
volume) at increasing pressures (260–530 Pa) until surface
saturation, indicated by physisorption, and the IR spectrum
recorded after each dose. The active site strength was
measured by water evacuation at 8.6·10− Pa at room
2
temperature, 150, 350 and 450 °C, recording an IR spectrum
after each evacuation. For the thermal gravimetric (TG) analysis,
a Mettler Toledo TGA/SDTD 851e was utilized in a 50–900 °C
temperature range and a ramp of 10 K·min−1 under an air flow
The CeO2-11nm-400min was recovered from the ketonization
reactor after reaction in the absence of water for 400 min of
time on stream with a low final catalytic activity indicated by a
conversion of only 2%. CeO2-11nm-700min was employed co-
feeding water for 700 min of time on stream (77% final
conversion).
−
1
1
of 35 mL·min . H and 13C NMR experiments were performed
1
on a Bruker 300 / 54 mm Ultrashield (300 MHz for H and 75
1
MHz for 13C). H and 13C chemical shifts are reported in ppm
versus Si(CH3)4 (TMS).
Catalytic tests in a fixed-bed, continuous-flow reactor
2% Pt/Al2O3 synthesis
Transformations of heptanal and alternative reactions
(hexanoic acid-heptanal, hexanoic acid-heptanoic acid,
heptanoic acid-nonanoic acid and aldol condensation product
2-pentyl-2-nonenal) were carried out in a tubular stainless-steel
reactor. The ceria catalysts (1.0 g, pellets, 0.2 – 0.4 mm) were
diluted with silicon carbide (2.0 g), placed as a fixed bed in a
stainless-steel tube (0.4 cm internal diameter and 20 cm length)
and calcined at 450 °C for 2 h in air (100 mL·min–1). The reactor
was heated to 450 °C under ambient pressure. Heptanal (5 mL)
was reacted at 0.2 mL·min–1 under N2 flow of 198 mL·min–1 or
feeding heptanal-water molar ratio of 1 to 1 (0.026 mL·min–1
water, 163 mL·min–1 N2) with CeO2-6nm and CeO2-11nm. For
CeO2-5nm, CeO2-7nm and CeO2-12nm, heptanal was reacted
feeding 0.1 mL·min–1 with a N2 flow of 98 mL·min–1 or feeding
heptanal-water with a molar ratio of 1 to 1 (0.013 mL·min–1
water, 80 mL·min–1 N2). 2-pentyl-2-nonenal (0.147 mL·min–1)
and water (0.0847 mL·min–1) were fed separately into the
reactor with molar ratio of 1 : 8 together with a nitrogen flow of
29 mL·min–1 employing CeO2-6nm as catalyst. The reaction
mixtures mentioned before were passed through the reactor at
450 °C, with a flow rate of 0.147 mL·min─1 together with a N2
flow of 144 mL·min–1 for heptanoic acid-nonanoic acid (over m-
ZrO2) and, for the other mixtures, with a N2 flow of 122 mL·min–
1 and water of 0.02 mL·min–1 using CeO2-11nm.
The metal catalyst was prepared by incipient wetness
impregnation using an aqueous solution of a platinum precursor
(H2PtCl6 · 6H2O) and Al2O3 (pellets 0.4 – 0.8 mm). The solution
was prepared dissolving H2PtCl6 · 6H2O (0.081 g, 0.0002 mol Pt)
in water (1.6 mL) to impregnate the metal oxide support Al2O3
(1.5 g). The catalyst prepared was dried overnight in an oven at
100 °C and reduced in-situ.
Catalyst characterization
The as-prepared and commercial cerium oxide catalysts were
characterized by X-ray diffraction (XRD) to confirm crystallinity
of the active phases and to measure the average particle size.
Analyses were performed on
a PANalytical CUBIX-PRO
diffractometer equipped with a PW3050 goniometer (Cu Kα
radiation) provided with a variable divergence slit. High-
resolution transmission electron microscopy (HRTEM JEOL JEM-
2100F) was used to verify the particle size of the CeO2-5nm,
CeO2-6nm and CeO2-11nm catalysts. The Brunauer–Emmett–
Teller (BET) analysis was performed using a Micromeritics ASAP
2420 analyser with a nitrogen bath. Catalysts were out-gassed
at 200 °C in vacuum prior to the analysis until the static pressure
remained lower than 70 Pa. The ceria catalysts were examined
using N2 adsorption and the BET method was used to calculate
the surface area in the range of relative pressures between 1
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 3
Please do not adjust margins

Please do not adjust margins
Green Chemistry
Page 4 of 16
ARTICLE
Green Chemistry
The liquid product was condensed with an ice bath at the exit acid partial pressure in the temperature range of 390 to 450 °C
View Article Online
of the reactor and analyzed offline by GC with an Agilent 7890A employing 25 mg of CeO2-11nm. The actDivOaIt:i1o0n.10e3n9e/rCg6yGCfo0r35t1h1eF
apparatus, equipped with a HP-5 column (30 m, 0.32 mm, 0.25 heptanal reaction was determined in the presence of water
µm) and a FID detector, using n-dodecane (Aldrich) as an using a water-heptanal molar ratio of 1 : 1 with a heptanal flow
external standard. Gaseous products were trapped in a gas of 0.147 mL·min–1, a water flow of 0. 019 mL·min–1 and a N2 flow
burette and analyzed by GC on a Varian 450 with a “refinery gas of 118 mL·min–1, 1.5·104 Pa heptanal partial pressure in the
analyser” configuration with three channels. Hydrogen was temperature range from 390 to 450 °C employing 25 mg of
analyzed after separation on a 2 m molecular sieve 5 Å column CeO2-11nm.
by a thermal conductivity detector. Permanent gases such as CO Coupling of heptanal ketonization and 7-tridecanone
and CO2 were separated on a 2.5 m molecular sieve 13X column hydrodeoxygenation carried out in two catalytic beds
and quantified by a thermal conductivity detector. Low
The catalysts (CeO2-11-nm and Pt/Al2O3) were placed
molecular weight hydrocarbons were separated on a 50 m
separately as fixed beds in two connected stainless steel tubular
Plot/Al2O3 column and quantified with a flame ionization
reactors (length of 20 cm and i.d. of 0.4 cm for each one). CeO2-
11-nm (1.0 g, pellets 0.4 – 0.8 mm) and 2 % Pt/Al2O3 (1.5 g,
pellets 0.4 – 0.8 mm) were packed with 2.0 g and 1.0 g of silicon
carbide, respectively. The Pt-containing catalyst was reduced in-
situ in the fixed-bed continuous-flow reactor.
The catalytic reactions were carried out at a total hydrogen
detector.
For testing the catalyst stability in the presence and in the
absence of water, 65 g of heptanal were fed into the reactor (1
g CeO2-11nm) until complete catalyst deactivation. With a fresh
catalyst bed (1 g CeO2-11nm) 109 g of heptanal were fed into
the reactor with heptanal-water molar ratio 1 : 1, at 450 °C
pressure of 2·106 Pa controlled by a back-pressure regulator.
without intermediate catalyst calcinations. Liquid and gas
The first reactor packed with CeO2-11-nm was heated to 450 °C
phases were analyzed by GC as mentioned above.
and the second one filled with 2 % Pt/Al2O3 was heated to 300
Kinetic measurements of initial rates in a fixed-bed, continuous-
flow reactor.
−
1
°C. After catalyst reduction, heptanal (5 mL, 0.1 mL·min ) and
−
1
water (0.638 mL, 0.013 mL·min ) were fed separately into the
Kinetic measurements were carried out in a tubular stainless- reactor with a molar ratio of 1 : 1 together applying a hydrogen
steel reactor at 450 °C and ambient pressure using 0.2 – 0.4 mm flow of 220 mL·min–1 at a total pressure of 2·106 Pa. At the exit
pellets of CeO2-6nm. Adequate N2 and heptanal flow rates and of the reactor the liquid product was condensed at 0°C with an
catalyst particle size were selected to avoid heat and mass ice bath and analyzed off-line by GC y GC-MS. Gaseous products
transfer limitations. With the aim to determine the kinetic were trapped in a gas burette and analyzed by GC on a Varian
expression of the reaction, initial rates of formation of the 450 with a “refinery gas analysers” configuration with three
principal product (7-tridecanone) were calculated by keeping channels as mentioned above. A simulated distillation was
heptanal partial pressure, water partial pressures and total flow carried out on a Varian 3800-GC chromatograph according to
constant while the contact time (W/F) was varied by employing the ASTM-2887-D86 procedure, which is suitable for the
different amounts of catalyst. Heptanal conversions were not distillation of crude oil components boiling at temperatures
higher than 12 %.
lower than 400 °C.
Initial rates were calculated for different partial pressures
feeding 5 mL of heptanal with partial pressures of 1.5, 3.0, 4.5
and 6.0·104 Pa, using heptanal flows of 0.147, 0.293, 0.440 and
0.587 mL·min–1 and N2 flows of 118, 93, 68 and 42 mL·min–1
respectively. For each heptanal partial pressure 10, 25 and 40
mg of CeO2-6nm were used. Water partial pressure of 1.5·104
Pa was maintained constant using water flow of 0.019 mL·min–
Results and discussion
Characterization of the materials
In the present work several cerium oxide materials were
employed involving different particle sizes, pore diameters and
surface areas. CeO2-6nm, CeO2-11nm and CeO2-43nm were
industrial/commercial catalysts, and CeO2-5nm, CeO2-7nm,
CeO2-12nm, CeO2-29nm and CeO2-277nm were synthesized by
hydrothermal treatment at different temperatures. The
average particle size was estimated by X-ray powder diffraction
(XRD) by means of the Debye-Scherrer equation and the value
included into the sample name: CeO2-xxnm (Table 1). For
selected samples the crystal size was verified by high-resolution
transmission electron microscopy (HRTEM, Figure S1).
1
.
To test the initial rate effect of adding water to the reaction,
heptanal partial pressure of 1.5·104 Pa was maintained constant
using heptanal flow of 0.147 mL·min–1, while heptanal : water
molar ratio was varied from 1 : 0, 1 : 0.1, 1 : 0.3, 1 : 0.5, 1 : 1 to
1 : 2 and 1 : 3. Kinetic modelling to fit the experimental data
with the proposed rate expression was carried out using the
Origin program (Non-linear fit, Levenberg-Marquardt iteration
algorithm, coefficient of determination R2 = 0.998).
The activation energy for the heptanoic acid reaction was
determined in the absence of water using a heptanoic acid flow
of 0.147 and a N2 flow of 144 mL·min–1, and in the presence of
water using a water-heptanoic acid molar ratio of 1 : 1 with a
heptanoic acid flow of 0.147 mL·min–1, a water flow of 0. 019
mL·min–1 and a N2 flow of 118 mL·min–1, 1.5·104 Pa heptanoic
The BET surface areas of the samples, measured by nitrogen
−
adsorption, varied from 19 to 234 m2·g , but were not in line
with the particle size. The CeO2-6nm sample possessed a much
higher surface area than the synthesized ones, e.g. CeO2-5nm,
1
−
1
with values of 234 and 174 m2·g , respectively (Table 1, entries
1 and 2). Also the average pore size of the two industrial
samples, CeO2-6nm and CeO2-11nm, were different from the
4 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 5 of 16
Green Chemistry
Please do not adjust margins
Green Chemistry
ARTICLE
View Article Online
DOI: 10.1039/C6GC03511F
Table 1. Physical and textural properties of the cerium oxide materials.
a
b
c
SBET
Dp
Vp
Reducible sitesd
[wt% (°C)]
Crystal sizee
1
1
Entry
1
2
Catalyst
CeO2-5nm
CeO2-6nm
[m2 g–
174
234
138
114
110
33
]
[Å]
64
31
79
47
66
87
95
114
50
54
[cm3 g–
0.22
0.10
0.23
0.15
0.14
0.05
0.10
0.04
0.15
0.14
]
[nm]
5
6
n.d.
35 (465); 32 (747)
n.d.
17 (452); 17 (769)
n.d.
18 (477); 21 (712)
15 (387); 28 (750)
9 (392); 24 (713)
n.d.
3
CeO2-7nm
7
4
CeO2-11nm
11
12
29
43
277
11
12
5
CeO2-12nm
6
CeO2-29nm
7
CeO2-43nm
47
19
8
9
10
CeO2-277nm
CeO2-11nm-400minf
CeO2-11nm-700ming
106
92
n.d.
a
b
c
d
Specific surface area calculated by the BET method using N2 adsorption isotherm at −196 °C. Average BJH pore size. Average BJH pore volume. From
e
Temperature-Programmed Reduction (TPR) measurements, see also Figure 2. From X-ray measurements: the particle size determination was done using the
Debye-Scherrer equation for particles of spherical shape.44 f Catalyst used in the absence of water for 400 min; area measurement after calcination. g Catalyst used
in the presence of water for 700 min; area measurement after calcination.
synthesized materials with 31 and 47 Å, respectively (Table 1, out on the CeO2-11nm sample with deuterated water. The
entries 2 and 4). All other samples had much larger average sample was calcined at 450 °C during 60 min prior to the
pore diameters (from 64 to 114 Å; Table 1, entries 1, 3 and 5 to treatment with D2O and IR spectra were recorded at room
8). The XRD patterns of all CeO2 samples corresponded to the temperature before and after introducing repeatedly small
cubic phase of fluorite, in which Ce has a cubic eight-fold determined amounts of D2O gas into the cell. Finally, pressure
coordination to oxygen21 (Figure 1).
was equilibrated at 227 Pa with D2O vapour, excess water was
The redox properties were all similar and TPR curves of selected removed again by applying high vacuum at room temperature
samples are shown in Figure 2. Reduction occurred in two and then desorbed successively at increasing temperature (150,
temperature ranges: a low temperature region (300–550 °C) 350 and 450 °C).
with a maximum at around 470 °C attributed to the surface Selected indicative spectra are shown in Figure 3. Spectrum (a)
reduction of the Ce4+ ions, and a high temperature region (600– was obtained after the initial calcination not showing any bands
900 °C) with a maximum at around 750 °C, probably related to in the region from 2800 to 2550 cm−1 corresponding to surface
the reduction of the bulk oxide.18 With decreasing BET surface deuteroxyl groups. After saturation of the surface with
area, the area of the peak in the higher temperature region deuterated water and applying vacuum at room temperature,
(indicating bulk reduction) becomes bigger with respect to the one main band is observed in spectrum (b) at 2700 cm−1 which
−
1
one in the lower temperature region (Table 1) which is in is maintained after desorption at 150 °C (max. at 2696 cm ,
accordance with the interpretation of surface reduction versus spectrum (c)). After desorption at 350 and 450 °C, in spectra (d)
bulk reduction of the respective temperature regions.
and (e) the shape of the bands changed and two maxima were
−
1
In order to gain insight into the hydroxylation (hydration) observed at 2710 and 2680 cm . The band at 2696 cm–1 has
process of the CeO2–surface an in-situ FTIR-study was carried been attributed to bridging OD groups coordinated to two CeIV
Figure 2. Profiles of the temperature-programmed reduction (TPR) of different
CeO2 catalysts.
Figure 1. X-ray diffraction patterns of different CeO2 materials.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 5
Please do not adjust margins

Please do not adjust margins
Green Chemistry
Page 6 of 16
ARTICLE
Green Chemistry
presence of hydroxy groups in vicinity of such vacancies (band
View Article Online
DOI: 10.1039/C6GC03511F
–1
at 2680 cm ). In addition, the band for the hydroxy group
without adjacent vacancies is shifted to higher wave numbers.
In conclusion, the in-situ FTIR study demonstrated that hydroxy
groups are present on the CeO2-surface, even at 450 °C, and
that these groups are available for coordination of other
molecules, for instance water or aldehyde molecules. Part of
the hydroxy groups can be removed by dehydration and oxygen
vacancies can be created. Water desorption from the surface is
further supported by thermogravimetry (TG) at atmospheric
pressure and under air flow. In the region from 400 to 500 °C a
weight loss was observed (Figure S2) which was proportional to
the BET surface area.
Catalytic results
A screening of the catalysts was performed in a fixed-bed
continuous flow reactor aiming to high conversion and high
ketone yield. Hence, heptanal was fed in presence (1 : 1 molar
ratio) and absence of water at 450 °C (Table 2, entries 4 and 5,
respectively). A cerium oxide material with an average crystal
size of approximately 11 nm (CeO2-11nm) was used as catalyst.
In both cases conversion was almost complete but in presence
of water the selectivity towards the desired ketone, i.e. 7-
tridecanone, was higher with 66% versus 56%. The selectivity
was improved still further by decreasing the average crystal size
of the CeO2 to approximately 6 nm (CeO2-6nm, Table 2, entry
2). The selectivity towards the desired ketone product was
raised to 79% in the presence of water. Apart from the C13-
ketone, a series of linear ketones with different carbon atom
number was observed (molecular weight distribution and
formation will be discussed below) and the selectivity towards
all ketones reached almost 90%. The aldol condensation
product and products derived from it were observed in total in
approx. 5% yield. Whereas the selectivity seems to be related to
the crystal size, the catalytic activity indicated by the conversion
Figure 3. FT-IR spectra for the D2O adsorption and desorption on CeO2-11nm. (a)
Spectrum after treatment of the sample at 450 °C applying high vacuum (8.6·10−3
Pa) before D2O adsorption (non-deuterated OH bands are visible in the 3600 to
3750 cm−1 region). (b) Spectrum after saturation of the sample with D2O and
evacuation under high vacuum (4.6·10−3 Pa) at room temperature, (c) to (e) Spectra
after desorption under high vacuum at 150 °C (c), 350 °C (d) and 450 °C (e),
respectively.
atoms without any oxygen vacancy in their coordination
sphere22, whereas the band at 2680 cm–1 has been assigned to
the same doubly coordinated species but with one of the cerium
atoms with an oxygen vacancy. Mono-coordinated hydroxy
groups (approx. 2740 cm–1) and hydroxy groups shared by three
cerium atoms (approx. 2655 cm–1) were not detected in
significant quantity.
From the above spectroscopic results it can be concluded that
at room temperature the surface is covered by hydroxy groups
bridging two cerium atoms. Then when heating up the sample
to 350 °C or 450 °C (spectra d and e, respectively), the surface is
dehydrated which creates oxygen vacancy evidenced by the
Table 2. Heptanal conversion and selectivities over different CeO2 catalysts for the ketonization of heptanal in the presence of water. Reaction conditions: 1.0 g of catalyst was
placed in a fixed bed and heptanal (5 mL, 0.2 mL·min─1) and water (0.026 mL·min─1), molar ratio 1 to 1, were reacted in the gas phase in a continuous-flow reactor at 450 °C
under nitrogen flow (163 mL·min─1), heptanal partial pressure 1.5·104 Pa, W/F 706 g·min·mol−
.
1
Selectivitya
Aldol Cond.c
Conv.
[%]
7-C13H26O
[%]
Total ketonesb
[%]
Hydrogen. aldol cond.d
Otherse
[%]
7.0
9.9
9.5
12.8
13.7h
10.8
16.8i
15
Entry
Catalyst
CeO2-5nmf
CeO2-6nm
CeO2-7nmf
CeO2-11nm
CeO2-11nmg
CeO2-12nmf
CeO2-43nm
CeO2-6nmj
[%]
4.4
2.3
5.0
4.3
6.5
6.0
6.7
31.6k
[%]
0.9
0.0
1.4
1.3
4.3
1.2
0.0
31.1
1
2
3
4
5
6
7
8
88.5
99.6
76.6
97.8
94.4
76.2
29.8
40.0
77.3
78.9
74.0
66.4
56.0
69.4
57.7
12.6
87.7
87.8
84.1
81.6
75.5
82.0
76.5
21.8
a Selectivity calculated with respect to the aldehyde substrate taking into account the reaction stoichiometry: two aldehyde molecules form one ketone molecule or
one molecule of the other products. b Total selectivity of ketones including 7-tridecanone; more detailed information on the distribution of the minor ketones can
be obtained from Figure 10. c Selectivity of the aldol condensation product, partly isomerized. d Saturated aldehyde with molecular formula C14H28O, produced by
carbon-carbon bond hydrogenation of the aldol condensation product. e Other products observed in the liquid phase and quantified by GC. f Reaction was carried
out with 0.50 g of catalyst, scaling down all reactant and gas flows to W/F stablished. g Reaction carried out without co-feeding water. Reaction conditions: heptanal
h
i
j
(5 mL, 0.2 mL·min─1) and N2 (198 mL·min─1): Olefins are observed in 5% selectivity. 1-Heptanol was observed with 12% selectivity. 2-pentyl-2-nonenal (aldol
condensation product) employed as feed together with water. Reaction conditions: 1.0 g of CeO2-6nm was placed in a fixed bed and 2-pentyl-2-nonenal (5 mL, 0.147
mL·min─1) was reacted under nitrogen flow (29 mL·min─1), feeding 2-pentyl-2-nonenal and water with a molar ratio of 1 to 8 (0.085 mL·min─1 water); 2-pentyl-2-
nonenal partial pressure 1.5·104 Pa. k Only isomerized aldol-condensation product.
6 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 7 of 16
Green Chemistry
Please do not adjust margins
Green Chemistry
ARTICLE
applied to the catalyst or that the coke elimination procedure
View Article Online
has to be further optimized. A small decliDnOeI:in10t.h10e39s/uCr6faGcCe03a5r1e1aF
might contribute to the activity decrease after reactivation by
calcination (cf. Table 1, entries 9 and 10) whereas a change in
the crystalline phase was not detected (Figure S3).
100
80
60
40
20
0
In conclusion the preliminary catalytic experiments indicated
that nanocrystalline cerium oxide is an excellent catalyst for the
ketonization of aldehydes providing selectivity to ketones up to
90%. This reactivity could be assumed since literature studies
conversion
7-tridecanone
23,24
on CeO2-ZrO2 mixed oxides18 or on bigger crystals of CeO2
indicated a basic catalytic activity for aldehyde ketonization. In
analogy to the reaction catalysed by zirconium oxide, water
seems to play a crucial role in this reaction and, therefore, the
mechanism and the role of water should be compared for both
materials.
25
125
225
325
425
525
625
725
Time on stream / min
Figure 4. Catalytic performance of CeO2-11nm with time on stream in the
ketonization of heptanal in the presence of water. Reaction conditions: heptanal -
water molar ratio 1 : 1, 0.2 mL·min─1 heptanal, 0.026 mL·min─1 water, 163 mL·min─1
N2 and 450 °C. During the reaction 109 g of heptanal were passed per g of catalyst.
Confirmation of the carboxylic acid as reaction intermediate
For zirconium oxide it has been established that the
ketonization of aldehydes proceeds by aldehyde oxidation and
subsequent classical ketonic decarboxylation of the carboxylic
acid as explained before.16 For the cerium oxide case the same
mechanism was assumed as working hypothesis and, therefore,
the key-control experiments and analysis were carried out with
this material. Gaseous effluents were collected at the exit of the
reactor and analyzed off-line (see experimental section for
more details). Exactly two equivalents of hydrogen were
obtained per molecule of ketone (Eq. 5), which supports the
stoichiometric dehydrogenation of the aldehyde. It can then be
concluded when the stoichiometric equation is balanced that
exactly one water molecule is consumed per ketone produced.
Thereby, it has to be kept in mind that, actually two molecules
of water are consumed per two equivalents of aldehyde, but
one of these is produced during the reaction at a later stage.
Almost one equivalent of carbon dioxide was observed per
ketone produced which is in accordance to the ketonic
decarboxylation mechanism of the intermediary carboxylic
acids. A possible reason for the slightly lower value with respect
to the theoretical one might be that a small part of carbon
degree is not clearly related to the textural properties such as
BET surface area or pore volume for a crystal size < 15 nm (cf.
Table 1 and Table 2). The best correlation can be found between
average channel size and conversion degree. In any case, raising
the crystal size significantly to 43 nm (which implicates a low
surface area of the material of 47 m2 g–1) decreases conversion
degree drastically to only 30% under otherwise identical
conditions (Table 2, entry 7).
The water had an effect on product selectivity on short term
and it had an influence on the catalytic performance at longer
term. Whereas the catalyst is fairly stable when co-feeding
water (Figure 4), in absence of water a rapid decline of activity
was observed (Figure 5). When the latter catalyst was recovered
from the reactor, its black colour indicated coke deposition
which could be quantified by TG in presence of air to be a 15%
of catalyst weight. However, calcination in air could only restore
the catalytic activity partly. A conversion of 45% and a total
ketone selectivity of 89% was achieved with the recycled
material. This indicated that additional deactivation pathways
100
conversion
7-tridecanone
80
60
40
20
0
50
100
150
200
250
300
350
400
Time on stream / min
Figure 5. Catalytic performance of CeO2-11nm with time on stream in the
ketonization of heptanal without co-feeding water. Reaction conditions: 0.2
mL·min─1 heptanal, 198 mL·min─1 N2 and 450 °C. During the reaction 65 g of
heptanal were passed per g of catalyst.
Scheme 2. Schematic description of the cross-coupling experiment between
heptanal and hexanoic acid. Cross-coupling product 6-dodecanone can only be
obtained if heptanal is oxidized to heptanoic acid during the reaction.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 7
Please do not adjust margins

Please do not adjust margins
Green Chemistry
Page 8 of 16
ARTICLE
Green Chemistry
condensation product or adduct as an intermediaVtieew.25A–rt3ic0leIfOtnhlinies
was the case, the final ketone has to be fDoOrmI: 1e0d.10a3t9l/eCa6sGtCa0s3f5a11sFt
from the aldol condensation product as from the aldehyde, and
with the same or in better yield. However, when the aldol
condensation product is passed over CeO2 under similar
reaction conditions, in presence or absence of water, the
ketone is obtained only in 5% yield versus 57–79% yield when
starting directly with the aldehyde (Table 2, entry 8). This result
clearly indicates that the ketone product is not produced
directly from the aldol condensation product. Ketones might
however be produced by a back shift of the aldol equilibrium
from the aldol product to the aldehyde substrates, and then via
the corresponding carboxylic acids.
Table 3. Product yields for the cross-coupling product 6-dodecanone and the homo-
coupling products 6-undecanone and 7-tridecanone in the reaction of hexanoic acid
with either heptanal or heptanoic acid. Reaction conditions: 1.0 g of catalyst was
placed in a fixed bed and both reactions were carried out at 450 °C in presence of
water with molar ratios of heptanal : hexanoic acid : water = 1 : 1 : 1 and of
heptanoic acid : hexanoic acid : water = 1 : 1 : 1, reaction mixture (5 mL, 0.147
mL·min–1), water (0.675 mL, 0.02 mL·min–1), N2 (122 mL·min–1). Conversion of
aldehyde and carboxylic acid was complete in all cases except for entry 2 (94%).
Product yield
C7
6-
6-
7-
undecanone dodecanone tridecanone
Entry substr. Catalyst
[%]
40
40
25
22
[%]
16
22
48
47
[%]
22
26
22
24
1
2
3
4
ald.
ald.
acid
acid
m-ZrO2
CeO2
m-ZrO2
CeO2
In conclusion, we present two convincing experimental
observations that the ketonization proceeds via aldehyde
oxidation into the carboxylic acid and not via the aldol
condensation product as intermediate.
Kinetics of the reaction and effect of water
dioxide is dissolved in the water of the gas burette with which
the gas was collected.
In the case of zirconium oxide water has a beneficial effect on
the selectivity towards the desired ketone, i.e. 7-tridecanone,
but it decreases the reaction rate.16 Now, the effect of water on
the reaction kinetics in the case of cerium oxide should be
evaluated. With zirconium oxide and in absence of water the
aldol condensation is assumed to provide the water required for
the ketonization. For cerium oxide, however, it has been
observed before that the catalytic activity for the aldol
condensation is much less pronounced and a deficit in water
concentration, with respect to the stoichiometric amount
necessary for the oxidation reaction, can be expected. Even
more, in absence of water a rapid cease of the catalytic activity
The carboxylic acid intermediate should be further confirmed
by cross-coupling experiments. If we assume that the carboxylic
acid is formed from the aldehyde and it is an intermediate for
the global reaction, and then carry out the reaction of heptanal
in the presence of other carboxylic acids with different chain
length, cross-coupling products have to be produced from the
in-situ synthesized carboxylic acid and the ones added initially
(Scheme 2). Hence, heptanal was reacted with hexanoic acid
and, indeed, the corresponding cross-coupling product, i.e. 6-
dodecanone, was observed in 22% yield (Table 3). Product
distribution between homo-coupling and cross-coupling
products was similar for reactions catalysed by CeO2 and m-
ZrO2, but in both cases different from the one observed when
reacting both molecules as carboxylic acid. When doing so, the
main product is the cross coupling product and the distribution
is almost the statistical one of 1 : 2 : 1. The deviation observed
when reacting the aldehyde has been explained on the bases of
different reaction rates.16 The second step of the aldehyde
ketonization is faster than the first one. As a consequence,
carboxylic acid present at an early reaction stage reacts fast to
the homo-coupling product and its concentration is decreasing
rapidly. In contrast, the aldehyde is converted slowly into the
corresponding carboxylic acid which then reacts either with the
lower concentrated carboxylic acid or in homo-coupling. This
explanation is well in line with the results observed (cf. Table 3).
As a consequence, both observations, i.e. the detection of the
cross-coupling product and the product distribution, are strong
evidences for the carboxylic acid as reaction intermediate from
the aldehyde to the ketones.
0.007
2-5 nm
CeO2-6nm
2-10 nm
CeO2-11nm
ZrO2-11 nm
ZrO2-11nm
0.006
0.005
0.004
0.003
0.002
0.001
0.000
0
1
2
3
4
5
6
7
heptanal partial pressure / 104 Pa
Figure 6. Comparison of experimental initial reaction rates in heptanal ketonization
catalysed with CeO2-6nm, CeO2-11nm and ZrO2-11nm in presence of water (1.5·104
Pa) as a function of heptanal partial pressure. Reaction conditions for CeO2-6nm
and CeO2-11nm: heptanal (5 mL), water (0.019 mL·min–1), catalyst (10, 25, 40 mg),
heptanal partial pressure 1.5·104 Pa (0.638 mL water, 0.147 mL·min−1 heptanal, 118
mL·min−1 N2), 3.0·104 Pa (0.319 mL water , 0.293 mL·min−1 heptanal, 93 mL·min−1
N2), 4.5·104 Pa (0.213 mL water , 0.440 mL·min−1 heptanal, 68 mL·min−1 N2), .6.0·104
Pa (0.159 mL water , 0.587 mL·min−1 heptanal, 42 mL·min−1 N2). Reaction conditions
for ZrO2-11nm: heptanal (5 mL), water (0.022 mL·min–1), heptanal partial pressure
1.5·104 Pa (0.638 mL water, 0.169 mL·min–1 heptanal, 6, 15, 23 mg of catalyst, 136
mL·min–1 N2), 3.0·104 Pa (0.319 mL water, 0.350 mL·min–1 heptanal, 12, 25, 40 mg
of catalyst, 110 mL·min–1 N2), 4.5·104 Pa (0.213 mL water, 0.518 mL·min–1 heptanal,
15, 25, 40 mg of catalyst, 80 mL·min–1 N2) and 6.0·104 Pa (0.159 mL water, 0.69
mL·min–1 heptanal, 20, 30, 45 mg of catalyst, 50 mL·min–1 N2).
Alternative mechanisms proposed in the literature for the
reaction of aldehydes to the ketone involve the aldol
8 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 9 of 16
Green Chemistry
Please do not adjust margins
Green Chemistry
ARTICLE
was observed (cf. Figure 5). Therefore, it can be concluded that This effect was more pronounced for the CeO2-6nm material
View Article Online
DOI: 10.1039/C6GC03511F
the kinetics of the reaction in presence of both metal oxides, i.e. than for the CeO2-11nm one.
m-ZrO2 and CeO2, have to be different.
When co-feeding higher amounts of water, i.e. >0.5
The kinetics of the reaction catalysed by cerium oxide was equivalents, a decrease was observed for the ketone formation
evaluated following the initial-rates methodology. In rate (Figure 7), in a similar manner as for zirconium oxide as
preliminary experiments it was observed that working at higher catalyst.16 It has been proposed that competitive adsorption of
heptanal flow than 0.147 mL·min–1 and lower catalyst particle water to active sites caused this deactivation, impeding an
size than 0.4 mm the process was not controlled by either efficient approach of the aldehyde substrate. However, as a
external or internal diffusion problems of reactants. At low clear difference, when the reaction was carried out over
conversion (<12%) the rate of ketone formation from heptanal zirconium oxide in absence of water (i.e. water partial pressure
was measured, always co-feeding water at a constant partial in the feed = 0) ketone formation was highest. This was related
pressure. When the initial rates of the ketonization with to an efficient water supply by the aldol condensation side
increasing substrate partial pressure are compared for three reaction which does not occur with cerium oxide, which is much
materials, namely cerium oxide with a crystal size of 6 and 11 less active than m-ZrO2 for catalysing the aldol condensation. As
nm and a zirconium oxide with a crystal size of 11 nm, it can be a consequence, with cerium oxide a certain amount of water
seen that the reaction is significantly faster with cerium oxide has to be added to the feed to achieve the maximum ketone
than with zirconium oxide (Figure 6). In addition, for cerium formation rate.
oxide, the reaction rate is accelerated even more with the The dependence of the ketone formation on the presence of
smaller crystal-size material. This has to be expected since the water in the ketonization of aldehydes over cerium oxide is a
material with the smaller crystal size also possesses an clear difference with respect to the zirconium case and this has
increased BET surface area with 234 versus 114 m2 g–1 for CeO2- also to be reflected in the rate equation, although the reaction
6nm and CeO2-11nm, respectively (Table 1, entries 2 and 4), and scheme is the same for both catalysts. In fact, it has been
a higher surface area is generally related to higher catalytic confirmed that the same overall mechanism applies in both
activity. The shape of the kinetic curve, i.e. the dependence on cases which is displayed in Scheme 1. Hence, the aldehyde is
the partial pressure of the aldehyde, is similar in all three cases oxidized in a first step and the corresponding acid converted
(Figure 6).
into the ketone in a second step by a classical ketonic
The initial rate was also studied in function of the partial decarboxylation. The carboxylic acid intermediate has been
pressure of water in the feed (Figure 7). Interestingly, the figure confirmed as such by cross-coupling experiments (Scheme 2)
demonstrates clearly that water is required for the reaction. and the hydrogen which has to be produced for the carboxylic
When the catalyst was freshly calcined to reduce surface acid formation was quantified in the right stoichiometry (Eq. 5).
hydroxy groups to a minimum (and therewith potential water In addition, the aldol condensation product was excluded as
formation/availability by surface dehydration) and water was intermediate (Table 2, entry 8).
not co-fed, the ketone formation rate approached zero. With For the zirconium oxide-catalysed reaction the first step, i.e. the
0.5 equivalent of water the highest initial rate was achieved and aldehyde oxidation, has been identified as rate-determining
increasing the water amount further the rate decreased again. step (Figure S4) with an activation energy of 198 kJ·mol–1, versus
110 kJ·mol–1 for the ketonic decarboxylation.16 When the
apparent activation energies were determined over cerium
oxide a value of 110 kJ·mol―1 (Figure S5) was observed for the
0.004
0.003
0.002
110 kJ/mol
144 kJ/mol
0.001
Heptanal
0.000
0
1
2
3
4
5
7-tridecanone
Water partial pressure / 104 Pa
Heptanoicacid
Figure 7. Initial reaction rates in heptanal ketonization catalysed with CeO2-6nm at
heptanal partial pressure of 1.5·104 Pa, in the presence of different amounts of
water (0.8, 1.5, 3.0 and 4.5·104 Pa). Reaction conditions: heptanal (5 mL, 0.147 mL
min–1), catalyst (10, 25, 40 mg), heptanal : water molar ratio 1 : 0.5 (0.009 mL min−1
water, 130 mL min−1 N2), 1 : 1 (0.019 mL min−1 water, 118 mL min−1 N2), 1 : 2 (0.038
mL min−1 water, 92 mL min−1 N2) and 1 : 3 (0.056 mL min−1 water, 67 mL min−1 N2),
450 °C; without water the reaction was carried out with 144 mL min−1 N2.
Figure 8. Apparent activation energy for the ketonization of heptanal and of the
ketonic decarboxylation of heptanoic acid taking into account the different heat of
formation of the organic molecules. The latter values were taken from literature
(heptanal45 and heptanoic acid46)and the heat of formation of 7-tridecanone from
the Cheméo database (https://www.chemeo.com/cid/56-631-9/7-Tridecanone;
obtained by the Joback method).47
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 9
Please do not adjust margins

Please do not adjust margins
Green Chemistry
Page 10 of 16
ARTICLE
Green Chemistry
ketonization of heptanal and 144 kJ·mol―1 for the ketonic the reaction site is ready again for aldehyde adsorption. In
View Article Online
DOI: 10.1039/C6GC03511F
decarboxylation of heptanoic acid in absence of water (Figure contrast, when working with cerium oxide, water is scarcer so
S6) and 143 kJ·mol―1 in presence of water (1 : 1 molar ratio, that a defect site might be created for a short term which can
Figure S7). Hence, on a first view that activation energy of the then be occupied first by an aldehyde molecule or a water
second step seemed to be higher than the one of the initial molecule. Once both substrates are coordinated, the formation
reaction step. However, when the different values for the heat of the surface-stabilized aldehyde hydrate by nucleophilic
of formation of the molecules were taken into account, the attack of the hydroxy group to the carbonyl group should be fast
activation step of the first step is still slightly higher than the again. However, the fact that both molecules have to be
second one and, therewith, the rate determining step (Figure incorporated from the gas phase causes a dependence for the
8). Also the product distribution of the cross coupling concentration of the surface-stabilized aldehyde hydrate on the
experiment of heptanal and hexanoic acid has been interpreted gas phase concentration for both reagents.
by means of a faster reaction of the intermediate acid than of All adsorption and desorption steps of the catalytic cycle on
the aldehyde. As additional evidence it can be mentioned that cerium oxide, including the nucleophilic attack of the surface
heptanoic acid, although being a reaction intermediate, is not hydroxy group onto the carbonyl group, are fast processes so
accumulated during the reaction but consumed immediately in that the rate-determining step has to be the hydride elimination
the second reaction step.
from the adsorbed aldehyde. The concentration of this species
Having identified the rate determining reaction step, all single is proportional to the occupancy of the two reaction sites and,
steps of the reaction were evaluated upon their influence on therewith, to the gas phase concentrations of both substrates.
the reaction rate. As the products formed in the first reaction Consequently, also the overall rate of the reaction depends on
step, namely carboxylic acid and molecular hydrogen, are the their gas phase concentration, and the rate equation should
same for both metal oxides, i.e. ZrO2 and CeO2, it is assumed display the water and the aldehyde concentration in the
that the same intermediates and transition states occur, i.e. the counter:
hydrogen is produced from a surface-stabilized aldehyde
(Eq. 6)
hydrate (Figure 9, upper right corner). After hydrogen
formation the carboxylic acid product has to be desorbed and
this step is supposed to be different for the two metal oxides.
As on the zirconium oxide surface water is abundant, the
carboxylic acid can be replaced directly by a water molecule and
In the denominator two different centres are reflected, one
adsorbing water and the second one the aldehyde. In addition,
both centres might be blocked by water. For a detailed
derivation of the formula see the ESI.
The rate equation was evaluated employing the data obtained
when carrying out the reaction in different conditions at low
conversion. Origin software was used to fit equation 6 with the
experimental data (Non-linear fit, Levenberg-Marquardt
iteration algorithm, coefficient of determination R2 = 0.998; cf.
Figure S8 and Figure S9). These results support the proposed
kinetic equation and, as a consequence, they confirm the
hypothesis that the dehydrogenation of the surface-stabilized
aldehyde hydrate is the rate-determining step for the
ketonization of aldehydes over cerium oxide.
The reaction network
After having established heptanoic acid as an intermediate for
the ketonization of heptanal and the ketonic decarboxylation as
second reaction step, the complete reaction network has been
worked out to elucidate the formation of the side products. As
side products the following compounds were observed: ketones
with an alkyl chain length different from the main product,
isomerized aldol condensation products and products with a
molecular weight of 212 g/mol, which might be derived from
hydrogenation of the aldol condensation product.
The ketone side product observed in highest concentration was
2-octanone (Figure 10). From the product distribution it can be
assumed that its production should be different from the other
ketones as it is the only one with an even number of carbon
atoms and its formation can be explained easily. It might be
produced by the McLafferty rearrangement from the main
product, i.e. 7-tridecanone, as described in Scheme S1. This
Figure 9. Proposed reaction mechanism for the oxidation of heptanal to heptanoic
acid in the presence of CeO2. At an oxygen vacancy (lower left corner) water and
aldehyde are adsorbed in close vicinity to two different sites. One proton from the
water is transferred to the oxide surface. Once both substrates adsorbed to the
surface, a surface-stabilized analogue of the aldehyde hydrate is formed which then
is able to transfer a hydride species to the surface. The latter can combine with a
proton and evolve as molecular hydrogen. With a further surface proton (from the
initial water adsorption) the carboxylic acid product can be desorbed and the initial
free dual site is recovered.
10 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 11 of 16
Green Chemistry
Please do not adjust margins
Green Chemistry
ARTICLE
in Scheme 3. Acidic and/or basic sites might be responsible for
View Article Online
CeO2 Heptanal CeO2Heptanal-agua ZrO2 Heptanal ZrO2 Heptanalagua1:1
CeO2-(Hept) CeO2-(Hept-water) ZrO2-(Hept) ZrO2-(Hept-water)
this isomerization. Please notice, the dDoOubI:l1e0.1b0o3n9/dC6isGCs0t3il5l11inF
conjugation with the carbonyl group and the formation
enthalpy should be similar for both compounds so that
thermodynamics favour a statistical distribution for the double
bond positions. After shifting back the aldol equilibrium from
the isomerized aldol condensation product towards the
aldehydes, two molecules of heptanal were converted into one
molecule of pentanal and one of nonanal, (Scheme 3) and an
alkyl chain length disproportionation had occurred. The
corresponding oxidation of the isomerized aldehydes into the
carboxylic acids should occur in analogy to the heptanal
oxidation. With this reaction sequence of aldol condensation,
double bond isomerization and retro-aldol reaction a plausible
way to chain-length isomerized aldehydes or carboxylic acids is
illustrated.
6
5
4
3
2
1
0
C8H16
O
C9H18
O
C11H22
O
C15H30
O
C17H34O
Distribution of ketones
Figure 10. Yields of the different ketone side products in different reactions. C8H16O:
2-octanone; C9H18O: 3-nonanone; C11H22O: 5-undecanone; C15H30O: 7-
pentadecanone; C17H34O: 9-heptadecanone.
fragmentation reaction to produce 2-alkanones is typically
observed when large chain ketones are produced under severe
conditions as for instance by pyrolysis of calcium decanoate.31
Other ketones were detected and identified by GC-MS, such as
e.g. 7-pentadecanone, for which fragmentation was not a
possible formation pathway since the number of carbon atoms
in this compound was increased with respect to the main
product. First of all, the product identification was confirmed
unambiguously by comparison with a synthesized authentic
sample. Hence, 7-pentadecanone was synthesized by the
ketonic decarboxylation of heptanoic and nonanoic acid over
zirconium oxide (see ESI). And indeed, mass spectrum and
retention time in gas chromatography of the compound
detected in the reaction mixture and of the authentic sample
were identical. Then, as a working hypothesis it was proposed
that the observed product was produced in the same way as the
authentic sample was synthesized, i.e. by cross ketonic
decarboxylation of nonanoic acid and heptanoic acid.
Therefore, nonanoic acid has to be formed under the present
reaction conditions.
The chain length isomerization reaction might be understood in
analogy to the isomerizing olefin metathesis which leads to
regular distributions with a mean chain length.32 However, to
the best of our knowledge this concept has not been reported
before for aldehydes. Although in the present case the yield in
these chain length isomerization reaction is still low, it is an
interesting concept which should be further developed in the
future.
If the isomerization occurs as described in Scheme 3, products
derived from pentanal should also be detected in the product
mixture. Even more, second generation aldehydes such as
propanal or products derived from them, might also be
observed (in low concentration). Indeed this was the case as the
following ketones were detected in the product mixture by GC-
MS: 3-nonanone, 5-undecanone, 7-pentadecanone and 9-
heptadecanone (Figure 10). These ketones have to be formed
from propanal, pentanal, heptanal (the actual starting
aldehyde) and nonanal, after having been oxidized to their
corresponding acids. This product pattern clearly supports the
proposed formation pathway of the ketone side products via
isomerization of the aldol condensation product and retro-aldol
reaction. A further experimental observation of the present
work which also supports this proposed pathway is the
increased selectivity towards other ketones (9.2%) with respect
to the target compound 7-tridecanone (12.6%), when the aldol
condensation product was reacted (Table 2, entry 8) and the
As a potential source for nonanoic acid the aldol condensation
product isomers were identified, which have been detected in
the product mixture. It was assumed that the double bond in
this molecule can be isomerized on the cerium oxide as depicted
Scheme 3. Proposed mechanism for the chain length isomerization of the aldehyde
substrates which is then responsible for the change in the ketone skeleton. Two
aldehyde molecules with n carbon atoms form the aldol condensation product. In
the latter, the double bond is isomerized into the alternative position, still in
conjugation with the carbonyl group. With the retro aldol reaction two different
aldehyde molecules are obtained with n+2 and n–2 carbon atoms, respectively.
Scheme 4. Proposed mechanisms for the formation of products with a molecular
weight of 212 g/mol. The unsaturated alcohol may be produced, together with the
unsaturated carboxylic acid, by a Cannizzaro disproportionation. The saturated
aldehyde with the same molecular weight may be formed by the C–C double bond
hydrogenation.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 11
Please do not adjust margins

Please do not adjust margins
Green Chemistry
Page 12 of 16
ARTICLE
Green Chemistry
isomers of the aldol condensation product accounted for 31.6% Alternatively, a reduced product may be obtained by Cannizzaro
View Article Online
DOI: 10.1039/C6GC03511F
disproportionation of the aldol condensation product, an allylic
selectivity.
After having explained the formation of ketone products with alcohol together with the corresponding acid (Scheme 4, upper
different carbon skeleton, the origin of the product with a pathway). However, when the allyl alcohol was produced by
molecular weight of 212 g·mol–1 was examined. This mass NaBH4 reduction or the unsaturated acid by autoxidation the
corresponds to a hydrogenated aldol condensation product. corresponding compounds were not detected in the product
However, two isomers are possible: the allyl alcohol produced mixture. This excludes reduction by occurrence of the
by aldehyde reduction or the saturated aldehyde produced by Cannizzaro disproportionation and confirms unequivocally the
C–C double bond hydrogenation. As a first hypothesis C–C bond double bond reduction of the aldol condensation product under
hydrogenation was assumed (Scheme 4, lower pathway). the present reaction conditions.
Recently, it has been reported in literature that bare cerium Aldehydes might also be decarbonylated, forming an alkane (or
oxide catalyses the hydrogenation of triple bonds.33,34 In arene) and carbon monoxide. Although in general metal
addition, on bare zirconium oxide hydride species have been catalysis is required for this reaction,35–38 the high temperature
evidenced during the ketonization of aldehydes16 and due to employed in the present case may also facilitate it. Therefore,
the analogy of the mechanism for this reaction on cerium oxide the corresponding product hexane was monitored. From Figure
it can be assumed that hydride species are also present on the S10 it can be seen that the yield of this product was always
cerium oxide surface. With the aim to proof this hypothesis the below 2% yield. Therefore, it can be concluded that
double bond in the aldol condensation product was decarbonylation is not an important reaction pathway for the
hydrogenated by molecular hydrogen in the presence of a Pd/C reaction of heptanal over CeO2.
catalyst (Scheme 4, lower pathway). And indeed, GC retention The complete reaction network can be established as depicted
time and the GC-MS mass spectrum confirmed that the product in Scheme 5. In general, the selectivity towards aldol
observed in the mixture with a mass of 212 g·mol–1 was the condensation products in the presence of water is low (<5%, see
double-bond hydrogenated aldol condensation product.
e.g. Table 2, entries 1 to 4), but isomerized aldol condensation
products are observed in higher selectivity when the aldol
condensation product is fed directly. Ketones derived from
fragmentation and chain length isomerization are avoided
especially with small crystal size material (<10 nm). Here the
selectivity towards these products is limited to selectivities of
10% (Table 2, entries 1 to 3).
The ketonization and hydrodeoxygenation cascade
One particular application of the ketonization of aldehydes is
the up-grading of biomass-derived mixtures to bio-fuels.
Thereby, complete hydrodeoxygenation is desired. To achieve
this an additional reaction step has to be carried out
subsequently to convert the ketone into an alcohol, then into
an olefin and, last, into the alkane. This cascade reaction has
been established starting from carboxylic acids.39–41 However,
when employing aldehydes the reaction possesses a special
attraction since during the reaction of the aldehyde two
equivalents of hydrogen are produced per molecule of ketone
and this is exactly the amount required to convert the ketone
into the alkane. Even more, in the hydrodeoxygenation one
molecule of water is produced which replaces the one
consumed for the oxidation of the aldehydes. In summary,
balancing the overall equation of all five reaction steps two
aldehyde molecules are converted into a linear alkane with 2n–
1 carbon atoms and one molecule of carbon dioxide (Scheme
6).
For the hydrodeoxygenation, literature conditions were
selected i.e. alumina supported platinum as catalyst and a
reaction temperature of 300 °C. In preliminary tests it was
found that the reaction at atmospheric pressure, applying
either a nitrogen or a hydrogen flow, the conversion of the
ketone was not complete. Therefore, the reaction was carried
out applying a 2·106 Pa hydrogen pressure. Under these
conditions it can be seen that the conversion of the ketone was
Scheme 5. Complete reaction network for the transformation of heptanal over
cerium oxide. During the ketonization reaction heptanal is oxidized into heptanoic
acid which is subsequently transformed into the symmetrical ketone 7-tridecanone.
The latter ketone can undergo McLafferty rearrangement causing a fragmentation
of the ketone and formation of 2-octanone. In parallel to the ketonization pathway
the aldehyde is in equilibrium with its aldol condensation product. The latter can
undergo double bond isomerization and hydrogenation. When the hydrogenation
occurs the saturated aldehyde is produced. Isomerized aldol condensation
products are observed in the product mixture (detected by GC-MS). However, they
can also react back to the aldehyde stage and produce aldehydes with different
chain length. The ketonization of these aldehydes provides ketones with different
chain length or with a different carbonyl group position.
12 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 13 of 16
Green Chemistry
Please do not adjust margins
Green Chemistry
ARTICLE
View Article Online
DOI: 10.1039/C6GC03511F
Scheme 6. Cascade reaction from heptanal to tridecane. In the first step the aldehyde hydrate is dehydrogenated over CeO2 at 450 °C, i.e. per two aldehyde molecules two water
molecules are consumed and two hydrogen molecules and two carboxylic acids produced. In the second step two molecules of heptanoic acid are joined into 7-tridecanone, co-
producing one molecule each of water and carbon dioxide. In the third step, changing to Pt/Al2O3 as catalyst and applying a hydrogen pressure of 2·106 Pa and a temperature of
300 °C, the ketone is reduced into the corresponding alcohol. In the fourth step the alcohol is dehydrated under the same reaction conditions. Finally, in the fifth step the olefin
is hydrogenated into linear tridecane.
higher than 95% (Figure 11). Tridecane, the linear alkane treatment of methyl ricinoleate or castor oil.42,43 The analysis
produced from 7-tridecanone, was obtained in 60% yield and an indicated that almost 90% of the crude product mixture of the
additional 30% of other alkanes. The latter were derived from five-step cascade falls within the boiling point range of the
the hydrodeoxygenation of other ketones (e.g. C8H18, C9H20, diesel fraction (Figure 12). In addition, due to the high content
C11H24, C15H32 and C17H36) and also of the aldol condensation- of linear alkanes and mono-methylalkanes the quality as diesel
derived products (C14H30; Scheme 7). In addition, n-heptane was fuel is excellent. Therefore, it can be concluded that combing
observed which was probably derived from non-ketonised the ketonization of aldehydes over cerium oxide with a
heptanal. The exact composition of the alkane fraction is subsequent hydrodeoxygenation step aldehydes can be
depicted in Figure S11.
transformed into a high quality diesel fuel. For this
The whole product mixture was evaluated as bio-fuel by transformation (almost) no external hydrogen is consumed.
simulated distillation. This is particularly interesting since
heptanal is a residual aldehyde produced during the thermal
Conclusions
It has been shown that cerium oxide with a small crystal size
(<15 nm) is an active and selective catalyst for the ketonization
of aldehydes. In contrast to zirconium oxide, the catalytic
activity for the competing aldol condensation is low which on
one hand allows to obtain higher selectivity but which on the
other hand converts water into a scarce compound on the
catalyst surface. Since water is consumed during the reaction, it
has to be co-fed to maintain catalytic activity.
80
60
40
20
0
Tridecane
Alkenes
Other alkanes
Carbonyl compounds
From the distribution pattern of the ketone by-products
observed it can be concluded that cerium oxide catalyses in
parallel an aldehyde chain length isomerization. As
a
50
100
150
200
250
Time on stream / min
Figure 11. Product yields for the cascade reaction of heptanal to tridecane. Other
alkanes are specified in Figure S11. Alkenes cover the whole range from C6 to C20
and carbonyl compounds are mainly 7-tridecanone, 2-octanone and the aldol
condensation product of heptanal. Reaction conditions: heptanal ketonization and
7-tridecanone hydrogenation reactions were carried out in two vertically
connected fixed-bed reactors that were packed separately with 1 g of CeO2-11nm
and 1.5 g of Pt/Al2O3 and heated at 450°C and 300 °C respectively. Heptanal (5 mL,
0.1 mL min–1) and water (0.638 mL, 0.013 mL min−1) were fed separately into the
reactor with heptanal : water molar ratio 1 : 1 under 2·106 Pa hydrogen pressure
and a flow of 220 mL min−1. The mass balance for the organic phase was
approximately 95% for every reaction (calculated with respect to the theoretical
yield, subtracting the weight of one molecule of carbon dioxide).
Scheme 7. Hydrodeoxygenation of the main products: 7-tridecanone is
transformed into tridecane, 2-octanone into octane and 2-pentyl-2-nonenal into 6-
methyltridecane.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 13
Please do not adjust margins

Please do not adjust margins
Green Chemistry
Page 14 of 16
ARTICLE
Green Chemistry
4
D. Gabriëls, W. Y. Hernández, B. F. Sels, P. Van Der Voort
and A. Verberckmoes, Catal. Sci. Technol., 2015, 5, 3876–
View Article Online
DOI: 10.1039/C6GC03511F
3902.
5
G. W. Huber, J. N. Chheda, C. J. Barrett and J. A. Dumesic,
Science, 2005, 308, 1446–1450.
6
S. De, B. Saha and R. Luque, Bioresour. Technol., 2015, 178,
108–118.
7
H. R. Beller, E.-B. Goh and J. D. Keasling, Appl. Environ.
Microbiol., 2010, 76, 1212–1223.
8
G. Kumar, J. L. Johnson and P. A. Frantom, Biochemistry,
2016, 55, 1863–1872.
9
I. L. Simakova and D. Y. Murzin, J. Energy Chem., 2016, 25,
208–224.
10
11
12
13
A. Corma Canos, S. Iborra and A. Velty, Chem. Rev., 2007,
107, 2411–2502.
G. W. Huber and A. Corma, Angew. Chemie - Int. Ed., 2007,
46, 7184–7201.
D. Mohan, C. U. Pittman and P. H. Steele, Energy and Fuels,
2006, 20, 848–889.
Q. Zhang, J. Chang, T. Wang and Y. Xu, Energy Convers.
Manag., 2007, 48, 87–92.
Figure 12. Simulated distillation of the product mixture of the reactions of heptanal
over CeO2-11nm to produce 7-tridecanone as main compound, followed by
hydrogenation over Pt/Al2O3 to produce tridecane and other alkanes. Horizontal
lines indicate temperature limits for the boiling point ranges of diesel fuels (171-
359 °C).
14
15
A. Demirbas, Appl. Energy, 2011, 88, 17–28.
D. K. Shen and S. Gu, Bioresour. Technol., 2009, 100, 6496–
6504.
L. M. Orozco, M. Renz and A. Corma, ChemSusChem, 2016,
9, 2430–2442.
A. Pulido, B. Oliver-Tomas, M. Renz, M. Boronat and A.
Corma, ChemSusChem, 2013, 6, 141–151.
A. Gangadharan, M. Shen, T. Sooknoi, D. E. Resasco and R.
G. Mallinson, Appl. Catal. A Gen., 2010, 385, 80–91.
M. B. Smith and J. March, March’s Advanced Organic
Chemistry: Reactions, Mechanisms, and Structure, John
Wiley and Sons, John Wiley & Sons, Inc., 7th edn., 2013.
H.-X. Mai, L.-D. Sun, Y.-W. Zhang, R. Si, W. Feng, H.-P.
Zhang, H.-C. Liu and C.-H. Yan, J. Phys. Chem. B, 2005, 109,
24380–24385.
consequence, ketones with different chain length were
observed, including isomers with more carbon atoms than the
one expected as product with 2n–1 carbon atoms. In addition,
hydride transfer to double bonds activated by carbonyl groups
in conjugation has been detected.
The ketonization of aldehydes over cerium oxide can be
combined with a hydrodeoxygenation. Thereby, a cascade
reaction is designed consisting of five subsequent steps
occurring in two catalytic beds. With this concept, the hydrogen
produced during the aldehyde oxidation can be consumed
during the hydrodeoxygenation. In the balanced overall
reaction equation two aldehyde molecules are transformed into
a linear alkane on the expense of one carbon dioxide molecules
as co-product. When the reaction is carried out with heptanal,
the latter is transformed into a very good quality diesel fraction
with almost 90% selectivity.
16
17
18
19
20
21
22
J. C. Conesa, Surf. Sci., 1995, 339, 337–352.
A. Badri, C. Binet and J. Lavalley, J. Chem. Soc., Faraday
Trans., 1996, 92, 4669–4673.
23
24
25
26
27
28
29
30
31
H. Idriss, C. Diagne, J. P. Hindermann, A. Kiennemann and
M. A. Barteau, J. Catal., 1995, 155, 219–237.
M. Kobune, S. Sato and R. Takahashi, J. Mol. Catal. A
Chem., 2008, 279, 10–19.
R. S. Murthy, P. Patnaik, P. Sidheswaran and M. Jayamani,
J. Catal., 1988, 109, 298–302.
Y. Kamimura, S. Sato, R. Takahashi, T. Sodesawa and T.
Akashi, Appl. Catal. A Gen., 2003, 252, 399–410.
S. Sato, R. Takahashi, T. Sodesawa, K. Matsumoto and Y.
Kamimura, J. Catal., 1999, 184, 180–188.
V. I. Komarewsky and J. R. Coley, Adv. Catal., 1956, 8, 207–
217.
J. B. Claridge, M. L. H. Green, S. C. Tsang and A. P. E. York, J.
Chem. Soc. Faraday Trans., 1993, 89, 1089–1094.
S. Ananthan, N. Venkatasubramanian and C. N. Pillai, J.
Catal., 1984, 89, 489–497.
Acknowledgements
The authors thank MINECO (CTQ2015-67591-P and Severo
Ochoa program, SEV-2012-0267) and Generalitat Valenciana
(PROMETEO II/2013/011 Project) for funding this work. L.M.O.
is grateful to the COLCIENCIAS institute for a Francisco-Jose-de-
Caldas (512/2010) doctoral fellowship. Assistance and advice
received from the Electron Microscopy Service at the
Universitat Politècnica de València is also acknowledged.
Notes and references
1
2
3
J. Q. Bond, D. M. Alonso, D. Wang, R. M. West and J. A.
Dumesic, Science, 2010, 327, 1110–1114.
B. G. Harvey and H. A. Meylemans, Green Chem., 2014, 16,
770–776.
R. L. Wingad, E. J. E. Bergström, M. Everett, K. J. Pellow and
D. F. Wass, Chem. Commun., 2016, 52, 5202–5204.
R. A. Hites and K. Biemann, J. Am. Chem. Soc., 1972, 94,
5772–5777.
14 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 15 of 16
Green Chemistry
Please do not adjust margins
Green Chemistry
ARTICLE
32
D. M. Ohlmann, N. Tschauder, J.-P. Stockis, K. Gooßen, M.
Dierker and L. J. Gooßen, J. Am. Chem. Soc., 2012, 134,
13716–13729.
View Article Online
DOI: 10.1039/C6GC03511F
33
34
G. Vilé, B. Bridier, J. Wichert and J. Pérez-Ramírez, Angew.
Chemie - Int. Ed., 2012, 51, 8620–8623.
J. Carrasco, G. Vilé, D. Fernández-Torre, R. Pérez, J. Pérez-
Ramírez and M. V. Ganduglia-Pirovano, J. Phys. Chem. C,
2014, 118, 5352–5360.
35
36
37
38
39
40
41
42
43
E. I. Gürbüz, D. D. Hibbitts and E. Iglesia, J. Am. Chem. Soc.,
2015, 137, 11984–11995.
P. Malcho, E. J. Garcia-Suarez, U. V. Mentzel, C. Engelbrekt
and A. Riisager, Dalton Trans., 2014, 43, 17230–17235.
A. Modak, A. Deb, T. Patra, S. Rana, S. Maity and D. Maiti,
Chem. Commun., 2012, 48, 4253–4255.
S. R. Yenumala, S. K. Maity and D. Shee, React. Kinet. Mech.
Catal., 2016, 1–20.
J. C. Serrano-Ruiz, D. Wang and J. A. Dumesic, Green
Chem., 2010, 12, 574–577.
A. Corma, M. Renz and C. Schaverien, ChemSusChem, 2008,
1, 739–741.
A. Corma, B. Oliver-Tomas, M. Renz and I. L. Simakova, J.
Mol. Catal. A Chem., 2014, 388–389, 116–122.
T. J. Farmer and M. Mascal, in Introduction to Chemicals
from Biomass, John Wiley & Sons, Ltd, 2015, pp. 89–155.
H.-B. Hu, K.-W. Park, Y.-M. Kim, J.-S. Hong, W.-H. Kim, B.-K.
Hur and J.-W. Yang, J. Ind. Eng. Chem., 2000, 6, 238–241.
A. L. Patterson, Phys. Rev., 1939, 56, 978–982.
G. da Silva and J. W. Bozzelli, J. Phys. Chem. A, 2006, 110,
13058–13067.
44
45
46
47
S. P. Verevkin, J. Chem. Eng. Data, 2000, 45, 953–960.
K. G. Joback and R. C. Reid, Chem. Eng. Commun., 1987, 57,
233–243.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 15
Please do not adjust margins

Green Chemistry
Page 16 of 16
View Article Online
DOI: 10.1039/C6GC03511F
Atom-efficient transformation of two aldehyde molecules into a linear alkane on the expense
of one carbon dioxide molecule.