Renewable and Sustainable Energy Reviews 48 (2015) 240–255

Contents lists available at ScienceDirect

Renewable and Sustainable Energy Reviews

journal homepage: www.elsevier.com/locate/rser

Green diesel synthesis by hydrodeoxygenation of bio-based

feedstocks: Strategies for catalyst design and development
Naveenji Arun, Rajesh V. Sharma, Ajay K. Dalai n
Catalysis and Chemical Reaction Engineering Laboratories, Department of Chemical and Biological Engineering, University of Saskatchewan, Saskatoon, SK,
Canada S7N 5A9

art ic l e i nf o a b s t r a c t

Article history: Green diesel or straight chain hydrocarbons are produced by the hydrodeoxygenation of bio-based
Received 12 July 2013 feedstocks such as vegetable oils and bio-oils. Usage of green diesel for commercial purpose can be
Received in revised form economical to cater the energy demands of the present generation. Traditionally, transition metal
25 August 2014
sulfides have been used for hydrotreating reactions and more recently usage of transition metallic
Accepted 7 March 2015
carbide, nitride, boride and phosphide catalysts are gaining importance. In this comprehensive review, it
Available online 13 April 2015
is intended to focus on some major developments in catalyst design methodology pertaining to
Keywords: hydrodeoxygenation (HDO) reaction systems. The details related to the different catalysts used for the
Green diesel hydrodeoxygenation reaction are also discussed. Most of the research works have been carried out using
Hydrodeoxygenation
model compounds such as guaiacol and phenol using moderately-highly acidic supports such as Al2O3,
Bio-based feedstocks
ZrO2 and TiO2 and active metals such as palladium, platinum, nickel and molybdenum. Factors to be
Catalyst design
considered for the choice of feed stock, selection of support material, active metals, promoters and their
influence on the hydrodeoxygenation reaction, catalyst design procedure, catalyst preparation method
and the influence of the catalyst treatment methods on the morphology of catalyst and their
performance are discussed in this review paper. Most research works have focused on usage of acidic
support materials and focus on basic support materials is scarce. Choice is promoters is highly
dependent on the promoter–support interaction and it is essential to ensure minimum promoter–
support interaction for the efficient performance of the catalyst.
& 2015 Published by Elsevier Ltd.

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ......................................... 241

2. Hydrodeoxygenation (HDO) process. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ......................................... 241
2.1. Reaction scheme for direct hydrodeoxygenation (Eq. (1)) and decarboxylation (Eq. (2)) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243
3. Feed source for hydrodeoxygenation reaction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ......................................... 243
4. Catalyst design strategy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ......................................... 246
5. Different catalyst used for hydrodeoxygenation process . . . . . . . . . . . . . . . . . . . . . . . . . . . . ......................................... 246
6. Choice of support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ......................................... 248
7. Promoters and their role . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ......................................... 249
8. Metallic carbide and nitride catalysts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ......................................... 250
9. HDO mechanism over different catalyst materials. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ......................................... 251
10. Catalyst activation and deactivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ......................................... 252
11. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ......................................... 253
Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ......................................... 253
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ......................................... 253

n
Corresponding author. Tel.: þ 1 306 966 4771; fax: þ1 306 966 4777.
E-mail address: ajay.dalai@usask.ca (A.K. Dalai).

http://dx.doi.org/10.1016/j.rser.2015.03.074
1364-0321/& 2015 Published by Elsevier Ltd.
 N. Arun et al. / Renewable and Sustainable Energy Reviews 48 (2015) 240–255
1. Introduction
Globally, the outlook and motivation towards the development
and usage of cleaner and renewable source of energy is growing due
to the soaring crude oil prices, diminishing petroleum resources and
environmental health hazards associated with the emissions from
fossil fuels. Increase in production and consumption of energy is
directly related to the growth in population and economic status of
the people. Owing to factors such as capital interests, fluctuations in
oil prices and other geopolitical issues, there is a tremendous growth
in biofuel industry globally [1,2]. Considerable work has been done
focusing on bio-based fuels production from rapeseed oil, soybean
oil, palm oil, karanja oil, waste cooking oil [3,4]. As predicted by
International Energy Agency (IEA), liquid biofuels will dominate the
energy sector by 2030.
Table 1 indicates the rise in world energy consumption from 2008
till date and its future trend [5]. It is seen that there will be a constant
rise in the energy demands in future. Availability of constant energy
source at affordable price is a key factor for a nation's gross domestic
product (GDP) [6,7]. Hence it is essential to find an alternative for the
fossil fuel resources to meet the future energy demands. By 2030,
biofuels production is forecasted to reach 6.5 Mb/d in comparison to
its production in 2010 (1.8 Mb/d) indicating that the need and supply
are forecasted to increase tremendously in future [8].
Fig. 1a and b indicates the energy consumption in Organization
for Economic Co-operation and Development (OECD) region in 2008
and its projection for 2035 [5]. Fig. 1c and d indicates the energy
consumption in the Non-OECD region in 2008 and its projection for
2035 respectively. It could be seen that the energy demands for the
Asian countries will rise tremendously in comparison to Americas
and the European Union. From Table 1 and Fig. 1, it is evident that in
future, an overwhelming demand for alternate fuels will arise and a
replacement to fossil fuels is highly essential. Innovations in technol-
ogy, hike in oil prices, employment effects, reduction in emission of
greenhouse gases and continued support from government through
various policies contribute to the rapid growth of biofuels industry.
Fig. 2 a and b represents the growth in biofuel sector from 2009 to
2013 and it confirms the immediate need for alternate fuels across
the globe. First generation fuels primarily involves fossil fuel
resources such as coal, petroleum and natural gases. Depletion in
fossil fuel resources and the environmental issues associated with
their usage resulted in the search for second generation fuels.
Production of second generation fuels involves feed stocks such as
corn to obtain bio-ethanol and bio-based fuels. However, most of the
feed sources are edible and they lead to food vs. fuel problem. Third
generation fuel involves usage of feed sources such as lignocellulosic
biomass, waste cooking oil and non-edible vegetable oils.
Recently, technologies such as biomass to liquid fuels (BTL) and
hydrotreatment of vegetable oils to produce green diesel are being
focused. Fig. 3 represents the trend in the growth of biofuel sector.
Presently, fossil fuels and second generation fuels such as biodiesel
and bioethanol are being commercially used and it is very evident
that technologies such as BTL and HVO (hydrotreated vegetable
oil) are promising routes to meet the future energy demands.
Table 1
Past and projected world energy consumption [5].
Year OECDn Non-OECDn World totaln
2008 250 260 510
2015 250 300 550
2020 265 340 605
2025 270 400 670
2030 275 450 725
2035 280 500 780
n
quadrillion Btu
241
Methods for upgradation of fuels can be broadly classified into
chemical and physical methods. Chemical upgrading methods
includes catalytic esterification [15] hydrothermal liquefaction,
hydrodeoxygenation and catalytic hydroprocessing [16,17]. Physical
upgrading method includes char removal, hot vapor filtration,
solvent addition and extraction of organic acids [18,19]. In case of
hydrothermal liquefaction, the biomass is converted to fuels and
chemicals under supercritical condition, in the presence of water at
a temperature of 350–374 1C and pressure of 20–25 MPa [20].
Inclusive of hydrodeoxygenation process, other methods such as
catalytic cracking [21,22], emulsification [23,24], steam reforming
and esterification [25] have been investigated for upgradation of
fuels. Li et al. [26] reported on upgradation of bio-oil especially the
low-boiling fraction using supercritical methanol and it was proven
to be an efficient method for bio-oil upgradation. Many other
methods such as super-critical method [27,28], pyrolysis [29],
gasification, fisher-tropsch synthesis [30,31], liquefaction and
hydrotreating are also being used to produce liquid fuels [32].
Presently, hydrotreating process for converting vegetable oils and
animal fats into renewable diesel fuel substitutes is gaining con-
siderable importance.
2. Hydrodeoxygenation (HDO) process
During a catalytic hydrotreating reaction, simultaneous removal of
sulfur, nitrogen, oxygen and metals is possible. Factors such as nature
of feedstock, catalyst and process parameters influence the extent of
particular reaction. Hydrodeoxygenation is a process of removing
oxygenated compounds from a molecule usually in the form of water
using catalyst such as nickel-molybdenum supported on alumina and
zeolite materials where hydrogen is used to cleave carbon–carbon or
carbon–heteroatom bonds in a molecule. Oxygen content of the bio-
based fuel plays a major role in assessing the fuel properties. The high
oxygen content of vegetable oils (upto 50 wt%) has adverse effects
such as low heating value, thermal and chemical instabilities, corro-
sivity, immiscibility with fossil fuels and increase in tendency towards
polymerization [33].
Hydrodeoxygenation can be a promising process to remove
oxygen content of the fuel thus aiding in the production of bio-
based transportation fuels [34]. Bio-oils obtained from wood and
wood residues can be potential feedstocks for HDO reactions due to
their high oxygen content (435 wt%) and they can be upgraded by
physical or chemical methods [35]. Research efforts to study the
chemistry of hydrodeoxygenation of biomass-derived oil are gaining
considerable importance over the last 20 years. Hydrodeoxygenation
process has been reviewed by many groups in the past (see Table 2).
Recently, Mohammad et al. [14] published a review on the overview
of catalytic hydrodeoxygenation to produce paraffin based-biofuels.
They have illustrated the importance of different feed stocks and
their advantages and disadvantages. In the present review, insight
into the chemistry behind the choice of feedstocks, catalyst devel-
opment and design strategies for the better understanding of HDO
process are provided. Hydrodeoxygenation of bio-oil is reported to be
the most suitable method to upgrade bio-oil [36,37]. Industrialization
of HDO process was first done by Neste oil in Porvoo, Finland using
vegetable oils as primary feed stocks. Although HDO is an efficient
method to obtain fuel from bio-based feed stocks, it has not received
considerable importance as the petroleum feeds generally contains
98 wt% carbon and hydrogen, 1.8 wt% sulfur and only 0.1 wt% oxygen
[38]. Hydrodeoxygenation of feedstock's that contains triglyceride
produces diesel-like deoxygenated hydrocarbons, with high cetane
number that is fully compatible with petro-diesel [39]. Fig. 4
indicates the factors affecting the choice of feed source, reaction
conditions, product formation and commercialization of hydrodeox-
ygenation process. Out of these, optimization of catalyst can be the
242 N. Arun et al. / Renewable and Sustainable Energy Reviews 48 (2015) 240–255

Fig. 1. World energy consumption by country grouping [5].

Fig. 2. World biodiesel production [8].

Table 2
Important reviews on hydrodeoxygenation systems.

S.No Significance Reference

1 Extensive information on the mechanism and kinetics of HDO reactions using different feed stocks [10]
2 Information on bio-oils hydrodeoxygenation and life cycle assessment [11]
3 Information on hydrodeoxygenation of biomass pyrolysis oil [12]
4 Information on hydrodeoxygenation of pyrolysis bio-oil on different catalysts [13]
5 Information on choice of feed stocks and future outlook [14]
6 Information on catalyst design strategy for hydrodeoxygenation of bio-based feedstocks and outlook on bioenergy sector In this study.

most challenging task. Choice of feed stock should be based on the
environmental conditions of the locality. In a province like Saskatch-
ewan, canola oil is available abundantly and it can be a promising
feed source. For efficient design of the hydrodeoxygenation process,
choice of catalyst, optimization of reaction temperature, hydrogen
pressure and catalyst loading is highly essential. Catalyst should be
designed to hinder side reactions and have higher selectivity towards
desired products. Finally, information on supply-demand structure
for the diesel substitute products in a local community and process
economics are mandatory for the commercialization of the hydro-
deoxygenation process.
The possible routes for hydrodeoxygenation include direct hydro-
deoxygenation, decarbonylation and/or decarboxylation [40]. In direct
hydrodeoxygenation, organo–oxygen molecules present in feed stocks
 N. Arun et al. / Renewable and Sustainable Energy Reviews 48 (2015) 240–255 243

Fig. 3. Trend in the growth of biofuel sector [9].

Fig. 4. Overall hydrodeoxygenation reaction system.

react with hydrogen at high temperature (250–400 1C) and pressure
(3–10 MPa) and hydrocarbons are obtained as main products.
During direct HDO, the CQO double bond is cleaved initially,
resulting in alcohol formation. Furthermore, hydrogenation of C–O
bond occurs forming alkane. During decarboxylation reaction, oxy-
gen is removed as CO2 by the direct attack on C–C single bond [40].
Reactions involved during hydrodeoxygenation are represented by
Eqs. (1) and (2). Eq. (1) is the most desired route (direct HDO) as the
number of carbons is preserved and the by-product formed is water.
2.1. Reaction scheme for direct hydrodeoxygenation (Eq. (1)) and
decarboxylation (Eq. (2))
  H2 
CH3  ðCH2 Þm  COOH ⟹ CH3  ðCH2 Þm
H2   fuels, and (iii) profitable agricultural economy.
 CH2 OH⟹ CH3  ðCH2 Þm  CH3 þ H2 O ð1Þ
  H2   preferred model compounds as it is an oxygenated compound with
CH3  ðCH2 Þm  COOH ⟹ CH3  ðCH2 Þm  1 CH3 þ CO2 ð2Þ
Generally, HDO reactions produce H2O (Direct HDO), CO (decar-
bonylation) and CO2 (decarboxylation) and desired hydrocarbon
compounds as the products and the product distribution depends
on the reaction route and catalyst type.
3. Feed source for hydrodeoxygenation reaction
Vegetable oils are ideal sources as they have a chemical structure
having long chain fatty acid with 16-24 atoms of carbon. Choice of feed
stocks depends on crop growing pattern of local regions and proper
choice of feedstock can aid in reduction of tariffs and transportation
costs. Soybean oil is a potential source in United States due to its higher
production rate [41]. Production of rapeseed oil, palm oil and sunflower
oil dominates in Europe and hence, they can be the potential feed
stocks for biofuel production in European nations (Fig. 5). Based on the
locality, similar considerations should be made while choosing a
potential feed stock for HDO reactions. Usage of renewable bio-based
feedstocks as fuel resource have many advantages pertaining to
economic, social and environmental aspects: (i) Stable supply of raw
materials. (ii) Reduction in carbon footprint from the usage of liquid
Most of the researchers have used model compounds to investi-
gate hydrodeoxygenation reactions. Phenol is one of the highly
simple structure [42]. Feeds such as guaiacol [43–48], methyl
heptonate [49], sorbitol [50], vanillin [51] and acetic acid [52] have
been tried and few papers have reported on HDO of real feedstocks
such as bio-crude [53], vegetable oils [54,55], waste vegetable oil
[56], canola oil [57] and bio-oils [58–62]. Specifically, many works
244 N. Arun et al. / Renewable and Sustainable Energy Reviews 48 (2015) 240–255

Fig. 5. Feedstocks for hydrodeoxygenation reaction in different regions of the world.

Fig. 6. Biomass to renewable hydrocarbon fuels [71].

have reported the usage of canola [63], palm [64], rapeseed [65],
sunflower [66] and karanja [67] oils as potential feedstocks. Since
most of the vegetable oils are edible, it leads to food vs. fuel problem.
In this regard, waste cooking oil has gained great importance as a
feed oil to produce fuel [68,69]. To improve the social and economic
feasibility of bio-based fuels, usage of affordable and low quality
feedstocks are recommended [70].
Production of third generation fuels are mostly based on bio-oils
and vegetable oil. Bio-oils have gained considerable importance as a
potential feed stock for the production of hydrocarbons by catalytic
upgradation. Lignocellulosic biomass is a promising renewable
energy source and can be used to produce bio-based fuel. From
the results on characterization of different biomass, it was con-
cluded that pinewood is the most reliable source owing to its large
hemicellulose and cellulose content and high calorific value [32].
Fig. 6 illustrates the conversion of different biomass to diesel fuel
products [11]. Biomass can be classified as carbohydrates, lignin and
fats/oils. Carbohydrates consist of cellulosic and hemicellulosic
materials. Fats comprises of triglycerides and fatty acids. As name
implies, lignin are basically derived from wood and lignin. Lignin
can be hydroprocessed to form phenols and tar which can be
further hydroprocessed to produce aromatics and naphthalenes
[71]. Similarly, sugars from carbohydrates of a biomass can be
converted to sugar alcohols by hydrogenation and further hydro-
processed to yield gasoline products.
Fig. 7 indicates the oil content in seeds and oleic acid content of
different feed oils that can be potentially used as feedstock for
hydrodeoxygenation reaction. As stated earlier, the choice of feed
stock is a factor of regional preference. Due to abundant availability
of canola oil (more than 3.3 million metric tons) and green seed oil
 N. Arun et al. / Renewable and Sustainable Energy Reviews 48 (2015) 240–255 245

100

Oil content in seed/kernel (%)

80

60
Oil content in
seed/kernel of
different 40
feedstocks
20

100

80
Oleic acid content (%)

60
Oleic acid
composition
of different 40
feedstocks

20

Fig. 7. Properties of different feed oils [72].

Source: http://www.chempro.in/fattyacid.htm

in Saskatchewan, canola and green seed (non-edible) canola oil can Table 3
be the potential feed stocks for HDO process in Canadian biofuels Fatty acid composition of canola and greenseed oil (Reproduced with permission
industry. from publisher) [73].
Understanding the reaction mechanisms using real feed stocks
Major Fatty acids Methyl ester of Canola oil Methyl ester of Green Seed oil
can be challenging and HDO studies using model compounds is
preferred initially to elucidate the reaction pathway before eval- Palmitic 4.21 4.38
uating real feed stocks. Studies on HDO of oleic acid can provide Stearic 2.03 1.98
useful information on the reaction mechanism and can help Oleic 62.33 62.74
Linoleic 19.13 19.99
researchers to design a catalyst that will work efficiently for real
Linolenic 9.18 9.60
feed stocks such as canola oil and waste vegetable oils. Eicosenoic 1.26 0.00
Table 3 (reproduced with permission from Elsevier) indicates the Erucic 1.87 1.31
fatty acid composition of canola and green seed oil. Primary fatty acid
constituent is oleic acid. Linoleic and linolenic acids are the next
major constituents of canola and greenseed oil [73].
Greenseed (non-edible) canola oil has high content of chloro- canola oil are chlorophylls and pheophytins. Due to the high
phyll and due to its non-edible nature, it is considered as a “waste content of chlorophyll in greenseed oil, it appears green and it is
product” [74,75]. The major pigments in canola and greenseed considered non-edible making it a potential source for HDO
246 N. Arun et al. / Renewable and Sustainable Energy Reviews 48 (2015) 240–255

reaction to obtain bio-based diesel products. The high chlorophyll 5. Different catalyst used for hydrodeoxygenation process
content of green seed canola oil is reported to inhibit the activity of
hydrogenation catalyst. Moreover, the presence of pigments has Many literature have reported on hydrotreating reactions using
been reported to have detrimental effects on the stability of oil and conventional metal sulfide catalysts [85] and noble metal catalysts
the biodiesel and bleaching of the oil is proven to remove the [86]. However, noble metal catalysts are costlier than transitional
pigments [75]. Cerny et al. [76] carried out hydrodeoxygenation metal sulfide catalysts and the cost factor limits the applicability of
reaction using zeolite USY catalyst and different feed stocks such as noble metals as catalysts for HDO reactions [87]. Table 4 sum-
canola oil, palm oil, oleic acid, stearic acid, cis-9-octaden-1-ol and marizes different hydrodeoxygenation reaction systems that have
octadecan-1-ol to study the influence of the nature of carbon- been investigated over the last two years.
oxygen bond and the degree of unsaturation present in the reactant Transition metallic sulfides of nickel, cobalt and molybdenum are
[76]. It was concluded that the usage of saturated alcohols resulted proven to be efficient catalysts for hydrotreating reactions such as
in the formation of the most desired products (iso-alkane). Using hydrodesulfurization (C–S) and hydrodenitrogenation (C–N). Hence, it
unsaturated feedstocks instead of the saturated ones resulted in the can be assumed that the same active metals could work for hydro-
increased formation of aromatics (410%). deoxygenation (C–O) reactions. Sulfidation process is carried out at
high temperature (280–350 1C) using hydrogen sulfide (H2S) or carbon
disulfide (CS2). Senol et al. [88] reported that the usage of H2S as
sulphiding agent is more beneficial than CS2 due to decrease in
4. Catalyst design strategy
hydrogen consumption and coke formation. High oxygen (430 wt%)
and low sulfur content (o3 wt%) of the feed (vegetable oil) can cause
Design of a catalyst can be stated as the systematic application
oxidation of the active catalyst phase and thus deactivating the sulfide
of available information for the selection of a catalyst in a
catalyst. In Canada, diesel fuel should have less than 15 ppm of sulfur
particular reaction. It is a challenging task because, accurate and
content for transportation usage (Canadian Environmental Protection
specific methods for catalyst development and design for specific
Act, 1999). Hydrotreatment of fossil fuels and model compounds have
reaction system are generally scarce and in most cases, it works on
been studied and it was shown that non-sulphided catalysts can be
trial and error method. Few authors have employed computational
successful in removing many impurities such as oxygen, sulfur and
techniques for accelerated catalyst design [77–83]. Generally, a
nitrogen. Furimsky [89] emphasized that reduced catalysts have strong
variety of catalysts can be predicted to perform well for particular
tendency to remove impurities (nitrogen, sulfur and oxygen) by
reaction and can be checked only by experiments. The number of
hydrotreating process.
catalyst to be tested can be reduced by proper application of
Li and Huber [90] studied on aqueous-phase hydrodeoxygena-
catalyst design procedure. In catalyst design, it is often recom-
tion of sorbitol using Pt/SiO2–Al2O3 catalyst and have concluded
mended to alter the already existing catalyst for a particular
that dehydration reaction occurs on the bronsted acid sites of the
process depending on the needs of the user than to develop an
catalyst. Hydrogenation reaction is exothermic and hence favorable
entirely new catalyst material [84].
at low temperature and in conventional processes, temperature
Fig. 8 indicates the steps involved in primary design of catalyst.
range of 573–618 K and hydrogen partial pressure in the range of
Methods for designing a totally novel catalyst and altering an
3.5–6.8 MPa are preferred. Severity of operating conditions depends
available catalyst for specific purpose differ slightly. Before design-
on the nature of the feed. Especially, heavier feed requires high
ing a completely novel catalyst, technical, economic and environ-
temperature and pressure for proper hydrogenation. Zhao et al. [91]
mental evaluations are mandatory. Catalysts developed should be
have performed hydrodeoxygenation of phenols to cycloalkanes
technically feasible with no harm to environment and should be
using dual-functional catalyst system (Pd/C and H3PO4). They have
cost-effective. However, if a catalyst needs to be designed for
studied the influence of catalyst support, pH value of aqueous
already existing mechanism, the design could be carried out based
solution, metal sites, and temperature on the HDO activity. In HDO
on some chemical grounds. Fig. 9 illustrates the different support
reactions of vegetable oils or bio-oils, preference towards hydro-
materials and active metals that are being employed for hydro-
genolysis of carboxylic groups or carbonyl groups is high in
deoxygenation reaction. Zeolite materials are used quite extensively
comparison to saturation of C–C bonds as the latter can reduce
as they offer high acidity that enables cracking of C¼O bond in feed
the products octane number [92]. Bykova et al. [45] have used
oils and aids hydrogenation. Recently, mesoporous materials are
nickel based sol–gel catalyst for the hydrodeoxygenation of guaia-
gaining importance as they offer moderate acidity and also the
col. Different loadings of nickel and copper on stabilizing compo-
required surface area and pore diameter for the reaction of real feed
nents such as γ-Al2O3, CeO2, ZrO2, SiO2 and La2O3 were employed to
stocks that are relatively heavier molecules.

Fig. 8. Algorithm for catalyst design [84].

 N. Arun et al. / Renewable and Sustainable Energy Reviews 48 (2015) 240–255 247

Fig. 9. Catalyst development for hydrodeoxygenation reaction.

Table 4
Summary on the recent studies of different hydrodeoxygenation reaction systems.

S. Model compound/feed Catalyst system Max. conversion/products References

No
Active Support material
metal

1 Dibenzofuran NiP – 99% (T ¼ 300 1C, PH2 ¼ 3 MPa) [122]

2 Phenol, cresols, guaiacol and Platinum Zeolite-beta 99% (T ¼ 180 1C, PH2 ¼5 MPa) [123]
eugenol (Pt)
3 Soybean oil Pt, Ni, Co– Al2O3, SiO2–Al2O3 95% (T ¼ 400 1C, PH2 ¼ 9.2 MPa) [124]
Mo
1
4 Jatropha oil Ni-HPW Al2O3 99.85% (T ¼ 360 1C, P¼ 3.3 MPa, LHSV¼ 1.0 h ) [125]
5 Methyl ester Ni, Co SBA-15, Al–SBA-15 98% (C17 paraffin's, Cat-Ni/Al–SBA-15; T ¼ 340 1C, PH2 ¼ 3 MPa) [126]
6 Guaiacol Fe SiO2, Activated Carbon 100% (10% Fe/AC, T ¼673 K, P¼ 1 atm) [99]
7 Bio-oil Ni ZSM-5, Al2O3 91.8%(T ¼ 240 1C, PH2 ¼ 4.0 MPa, 10 wt% Ni/ZSM-5) [103]
8 Guaiacol Ni, Ni–Cu ZrO2–SiO2 100% (T ¼300 1C, PH2 ¼ 5 MPa, t¼ 8 h, NiCu (5 wt%)/ZrO2–SiO2) [127]
9 Methyl isobutyl ketone Pt, Pd, Cu SiO2, C, HZSM-5 100% (T ¼100 1C, t ¼ 1 h, 0.5% Pt/HZSM-5) [128]
10 Jatropha oil Co–Mo Al2O3 99.99% (T ¼ 360 1C, PH2 ¼ 80 bar) [129]
11 m-Cresol Ga H-Beta, ZSM-5, HZSM-5 and 84.52% (T ¼ 450 1C, PH2 ¼1 atm, Ga (3 wt%)/H-Beta). Conversion reached [130]
Silica 95% when T ¼ 550 1C
13 Guaiacol Ni–Cu Al2O3, SiO2, CeO2–ZrO2, ZrO2– 97.5% (T ¼320, P¼ 17 MPa, t¼ 1 h, Ni(55 wt% )SiO2 [45]
SiO2–Al2O3
14 Propanoic acid Ru, Ru– ZrO2 95% (T ¼ 210 1C, P¼ 6.4 MPa, Mo/Ru¼ 0.2, Ru–Mo/ZrO2) [131]
Mo
1
15 Oxygenates in aq. Fischer Ru ZrO2, TiO2, SiO2, Al2O3 94.4% (T ¼ 200 1C, PH2 ¼9.8 MPa, LHSV¼3.0 h , Ru/TiO2) [132]
Tropsch process
16 Glycerol Pt γ-Al2O3, mixed silica alumina, 11.9% (T ¼ 225 1C, PH2 ¼ 29 bar, Pt (0.8 wt%)/Al2O3 [133]
SiO2

investigate the HDO reaction. Nickel based sol–gel catalysts have
been reported to outperform the conventional sulfide Ni(Mo), Co
(Mo) and noble metal catalysts owing to factors such as higher
stability towards coke formation, low cost, stability towards leach-
ing of active components. Catalyst with 55.4 wt% of nickel with no
copper loading stabilized using SiO2 is reported to have higher
surface area (ABET ¼216 m2/g; d¼20 Å) than the other synthesized
catalysts (66–142 m2/g). This could be attributed to the blocking of
some of the pores during the loading of copper. Presence of copper
aids in the reduction of nickel oxide at lower temperature and
hinders coke formation. Nickel supported on SiO2 (without copper)
gave better conversion of guaiacol (97.5%) and HDO (97.3%). How-
ever, degree of coking was high in comparison to other catalysts
with copper loading. One specialty about this study is the use of
pure guaiacol without any solvent.
Ardiyanti et al. [93] used bimetallic Ni–Cu supported on δ-Al2O3
catalysts for the hydrodeoxygenation of fast-pyrolysis oil. δ-Al2O3 is
thermally more stable and less acidic than γ-Al2O3 and this helps in
reduced coke formation. The loadings of nickel and copper were
changed and their effects on hydrodeoxygenation rate were mon-
itored. Leaching levels and carbon deposition were lowest when the
nickel loading was maximum (16 wt%) indicating that usage of
promoter's aid in increasing the resistance towards coking. Lee
et al. [94] studied the hydrogenation of model (oxygenated) com-
pounds such as acetaldehyde, xylose and furfural using Palladium
(Pd), Platinum (Pt), Ruthenium (Ru), Rhodium (Rh), Nickel (Ni) and
Cobalt (Co) supported on Alumina. Moreover, they used the same
catalysts to investigate tetrahydrofurfuryl alcohol hydrogenolysis.
From their results, it was concluded that Rh seems to outperform
all the other mono-metallic systems for hydrogenation reaction.
Additionally, from their studies on hydrogenolysis of THFA and
Xylitol, it was confirmed that Rh is a promising active metal for both
hydrogenation of CQO and CQC bond and hydrogenolysis of C–C
bond. However, Pt, Ni and Co are reported to be more suitable for
248 N. Arun et al. / Renewable and Sustainable Energy Reviews 48 (2015) 240–255
hydrogenolysis of C–O–C bond. During the hydrogenolysis of THFA,
20 wt% of Ni/Al2O3 is reported to give a conversion of 17.3% with 0%
selectivity towards 1,2-pentanediol and 59.2% selectivity to 1, 5 pen-
tanediol. However, 3 wt% Pt/Al2O3 gave 9.3% conversion and 38.9%
selectivity towards 1,2-pentanediol and 0% selectivity to 1, 5 penta-
nediol. Echeandia et al. [95] studied on the synergetic effects of
tungsten precursors in the hydrodeoxygenation of phenol over Ni–W
supported on active carbon. It was stated that tungsten from
heteropolyacids (HPA) performed better and activated carbon gave
better conversion than the conventional alumina support. Chances of
methanation are high during HDO reactions. CO and CO2 produced
by HDO reactions through decarbonylation and decarboxylation can
react with H2 (water–gas shift and methanation reactions) to form
methane. Methane is a worse green house gas than CO2 and hence
this reaction is highly undesirable. Chen et al. [134] studied on the
effect of Ni/Mo ratio on deoxygenation of model compound (methyl
laurate) using NiMo phosphide catalysts. Electron transfer between
the promote and the active metal plays a vital role in the hydro-
deoxygenation activity and the transfer is governed by the Ni/Mo
ratio in active phase. During phosphidation of catalyst presence of
MoNiP2 is confirmed by XPS analysis and it is due to the incorpora-
tion of Mo atoms into Ni2P increasing the chemical binding of Ni-Mo
species. It is also reported in the same paper that occurrence of
MoNiP2 phase is beneficial as it aids the catalytic activity of the NiMo
phosphide catalyst. Catalyst for HDO reaction should be particularly
designed so that these undesired side reactions are inhibited.
6. Choice of support
Choice of support is also a key factor for determining the
hydrodeoxygenation activity of different catalysts. The most com-
mon and conventional support is γ-Al2O3 [96]. γ-Al2O3 has many
beneficial properties such as high stability and moderate/slight
acidity. However, catalyst deactivation due to coke formation is a
major problem especially with the use of alumina support. Alter-
native materials such as SiO2, active carbon, TiO2, ZrO2, Nb2O5,
zeolites and various metal oxides were used in HDO reactions [97].
Hydrodeoxygenation reactions have been traditionally carried out
using moderate-highly acidic support materials such as oxides of
alumina, silica, zirconia and all these materials have high ionic
potential. Support materials such as calcia and magnesia are basic
materials with low ionic potential and usage of basic supports and
thoria (acidic support) are scarce with respect to HDO reactions.
One interesting fact about using activated carbon as support is
that, their activity depends on the preparation method and hence,
tailoring of their activity is a possibility. This unique property can
make activated carbon a suitable choice as a support material for
HDO reactions. Low-temperature carbons have acidic hydrophilic
surface while high-temperature carbons tend to have basic and
hydrophobic surfaces.
Deoxygenation of methyl ester to olefins is shown in Fig. 10.
Reaction pathway (a) is desired which involves conversion of methyl
ester to α-olefin [98]. Chiappero et al. [98] have reported the
occurrence of both the desired and undesired pathways using PtSnK
supported on silica. However, selectivity towards formation of α-
olefin was reported to be higher (450%) compared to olefin isomers.
Bui et al. [43] compared the support effects (zirconia, titania and
Al2O3) for CoMoS catalyst on the hydrodeoxygenation of guaiacol.
Guaiacol is a model compound used to represent lignin. It was
concluded that the performance of zirconia support was better than
titania and γ-Al2O3 in terms of HDO conversion and product selectiv-
ity. NiMo supported on ZrO2, TiO2 and γ-Al2O3 were characterized
using Raman spectroscopy to confirm the occurrence of desired metal-
oxide species. Zirconia support offers both basic and acidic nature;
basic nature of the support material helps to reduce coke formation
Removal of CO, CO2 ,CH 4
O
a
O
b Olefin
Methyl Ester
Olefin isomer
Fig. 10. HDO Pathway for methyl ester (a) desirable and (b) undesirable pathways [98]
(Reproduced with permission from publisher).
and acidic nature helps to aid hydrogenation and subsequent C(sp2)–O
bond cleavage and deoxygenation reactions. Olcese et al. [99] have
compared the performance of Fe/SiO and Fe/Activated carbon for
hydrodeoxygenation of guaiacol. For the same iron loading, the
guaiacol conversion was higher with the usage of silica support than
activated carbon. However, the selectivity towards desired product
was higher with the usage of activated carbon support material.
Catalyst design for HDO reactions can be done by altering the
host lattice (support) and seeing the influence on the guest metal
ion on catalytic activity. When the host lattice changes, the inter-
action and the co-ordination environment of the guest ion changes.
Alternatively, studies can be done by keeping the host lattice same
and varying the concentration and/or nature of the guest ion
(change in the active metals). Primary advantage of using porous
support is that the extent of availability of active metal for catalytic
reactions can be controlled. Preparation methods can govern the
diffusion limitations to obtain well-distributed metal catalysts.
Molybdenum forms polymolybdate species at any loading on
silica supports due to the similarity in isoelectric points of silica
and MoO3. Generally active metals such as Ni or Co are used and Ni
has better activity towards hydrogenation than Co. A lower activity
towards HDO can be directly linked to lower acidity of the catalyst
support materials. Nava et al. [100] also stated that catalyst acidity
and extent of coke formation have no correlation which seems
contradicting to their statement where lower acidic strength of
MCM-41 was related to the lower coke formation in comparison to
zeolite which has considerable higher acidity.
Influence of sulfidation rate has lesser influence than the acidity
of support. Hence, degree of carbidation or nitridation of exposed
active metal species may not play a vital role in HDO activity in
comparison to the acidity of the support. In comparing the supports
SBA-15, SBA-16, MCM-41 and DMS-1, Nava et al. [100] have stated
that HDO selectivity towards desired products (paraffin's and olefins)
was better in DMS-1 support material. Formation of Bronsted acid
sites is important for hydrogenolysis during HDO reactions and the
formation depends on silanol groups structure when silica–alumina
is used as the support material. Incorporation of aluminum increases
the catalyst acid strength. As incorporation of aluminum occurs near
silanol groups, the OH groups on surfaces acts as bronsted acid sites.
It is widely accepted that activity of catalyst for hydrogenation can be
increased by promoting the acidity of catalyst by introduction of
acidic solvents and/or supports.
Kubicka et al. [101] studied on the deoxygenation of rapeseed oil
using support material such as SiO2, TiO2 and Al2O3 using nickel
(3.3 wt%) and molybdenum (15 wt%) as catalyst. HDO reactions were
performed at a temperature range of 260–300 1C and hydrogen
pressure maintained at 3.5 MPa. It can be observed from their work
that product selectivity depends on the nature of support material to
a considerable extent. TiO2 supported NiMo catalyst exhibited high
selectivity towards hydrodeoxygenatlon product in comparison to
other support materials. In the case of TiO2 support, the active phase
was not dispersed well and the active phase cluster size was high.
 N. Arun et al. / Renewable and Sustainable Energy Reviews 48 (2015) 240–255
However, SiO2 exhibited better dispersion of active phase in compar-
ison to Al2O3 and TiO2 and this well dispersion is reported to be the
main reason for the preferential decarboxylation over SiO2. From this
work, it is clearly observed that metal-support interaction and the
support properties plays a crucial role in the choice of reaction
pathway and also the formation of desired products.
Yang et al. [102] studied on the hydrodeoxygenation of anisole over
nickel supported on materials such as carbon, γ-Al2O3, SBA-14, Al–
SBA-15, CeO2, TiO2 in a fixed bed reactor system. Major products
formed by the hydrodeoxygenation of anisole are hexane, cyclohex-
ane, benzene and toluene. Conversion of anisole and selectivity
towards products differed significantly depending on the different
supports. In the case of Al–SBA-15 and SBA-15, the conversion of
anisole showed less difference (o3%). However, the selectivity
towards the products differed significantly (45%). Conversion and
selectivity were reported at two reaction temperatures (290 1C and
310 1C) and LHSV (20.4 h  1 and 81.6 h  1). It was reported that, at
higher temperature (310 1C), the selectivity towards hexane was high
(LHSV-20.4 h  1; Cat: Ni/SBA-15; Shexane ¼40%). At a temperature of
290 1C, the selectivity was 22% at the same process conditions. Zhang
et al. [103] synthesized nickel (different loadings) supported on ZSM-5
and Al2O3. ZSM-5 represents a highly acidic support material, while
Al2O3 represents moderately acidic support material. Conversion of
bio-oil using nickel supported on ZSM-5 and Al2O3 (10 wt%) were
91.8% and 20.9% respectively. However, characterization of the catalyst
materials to analyze the metal-support interaction is essential to
support the conclusions substantially. Information on active metallic
surface area and metal dispersion is recommended to help find the
justification for the significant difference in the conversion of bio-oil.
Zhao et al. [104] synthesized nickel supported on HZSM-5 and Al2O3–
HZSM-5 and evaluated them for the hydrodeoxygenation of phenol.
The purpose of this study was to understand the influence of Al2O3
binder on the Ni/HZSM-5 catalyzed hydrodeoxygenation of phenol.
Catalyst samples were characterized using XRD, TEM, lR, TPR, EXAFS
and XANES and it was concluded that, Ni supported on Al2O3–HZSM-5
had higher concentration of nickel atoms in comparison to Ni/HZSM-5.
Moreover, usage of binder like Al2O3, resulted in strong and higher
adsorption capacity of reactant molecules on Al2O3 leading to better
product conversion. Catalyst deactivation due to sintering is also
reduced by the usage of binders. Nickel oxide forms a strong chemical
bond with the binder (NiO–Al2O3) and forms ‘pseudo spinels'. This
strong chemical interaction makes the reduction of NiO/Al2O3–HZSM-
5 tougher than NiO/HZSM-5 as confirmed from their TPR results.
Over the years, usages of basic support materials have not
gained considerable attention. Acid–base interaction between the
basic support and the metal solution can promote dispersion of
active metal species on catalyst support. Secondly, coke formation
can be inhibited by the basic nature of the support. Literature on
hydrodeoxygenation reaction using MgO-supported catalysts are
generally scarce. Yang et al. [53] studied on hydrodeoxygenation of
bio-crude using phenol as model compound in super-critical
hexane at 300–450 1C and hydrogen pressure of 5.0 MPa using
sulfided CoMo/MgO and CoMoP/MgO catalyst. It was concluded
that the basic support is active for hydrodeoxygenation reaction
only at reaction temperature greater than 350 1C. XPS analysis was
performed on fresh and spent CoMo and CoMoP catalysts. It was
found that coke formation on the MgO supported catalyst was
considerably less (o20%). Since, the bond energy of CQO is high in
fatty acids, support materials with moderate to high acidity are
generally preferred. However, cracking reactions renders products
with less carbon number. One of the major roles of support is
adsorption of reactant molecules in close proximity to active metal
particles. During HDO, H2 molecules adsorb on active metal and are
supplied to the reacting molecule by spill-over mechanism. Adsorp-
tion of the reacting molecule can vary depending on the acid–base
nature of the support material and this affects selectivity towards
249
certain products. Velu et al. [105] have stated that phenol (popular
model compound for HDO reaction) adsorbs in non-planar fashion
on basic supports and it adsorbs in co-planar fashion with respect to
the surface on acidic supports. Support materials such as CeO2 and
ZrO2 are gaining importance owing to their main advantages like
higher thermal stability, better dispersion of metals, ability to store
oxygen and redox properties.
Hydrodeoxygenation of bio-based feedstocks deals with bulkier
molecules and hence support materials such as MCM-41 which has
high surface area in the range of 600–1000 m2/g and homogenous
pore diameters from 1.5 to 10 nm are highly recommended. Acidity
of support material plays a crucial role in degree of coke formation.
Support materials such as zeolites are highly acidic and can lead to
higher degree of coke formation. However, moderate acidity is
essential for the breaking of carboxylic acid group when vegetable
oils are used as feed stock during hydrodeoxygenation. Using
support material such as MCM-41 which has lower strength of
acidic OH groups produce minimum amount of coke in comparison
to strongly acidic support materials. SBA-15 can be an effective
support for hydrotreating reactions. SBA-15 with hexagonal pores in
a 2D array has thicker walls (3–9 nm) than MCM-41 and this makes
them thermally stable than MCM-41 [106]. Belonging to the same
series, SBA-16 was synthesized and scarcely used in hydrotreating
reactions. SBA-16 is cubic cage-structured mesoporous silica mate-
rial and it can be a promising support for hydrodeoxygenation of
bio-based oils because, it has body-centered structure in which each
mesopore is connected to eight neighboring mesopores and this
allows easier diffusion of bulkier molecules into pores [100]. Nava
et al. [100] published a significant article to report on usage of
disordered mesoporous silica (DMS-1) having surface area in range
of 600–1000 m2/g.
Wide range of acidic, neutral and basic supports have been used
for hydrodeoxygenation reactions. Most of the studies have focused
on using acidic (ZrO2) or moderately acidic supports such as Al2O3
and SiO2. Usage of neutral supports such as SBA-15, MCM-41 and
KIT-6 are gaining importance. Overall, studies on basic support are
scarce and this is due to their lower activity at temperatures less
than 350 1C. On a commercial scale, carrying out reactions at
reaction temperatures above 350 1C may not be viable owing to
the huge energy supply. Acidic supports have the potential to break
the CQO bond in fatty acids and can perform hydrodeoxygenation
at temperatures less than 350 1C. However, cracking and coke
formation are crucial parameters that need to be addressed. Other
parameter that should be considered during the choice of support is
its porosity as the real feed stocks such as vegetable oils are bulky
molecules, and porosity of support material plays a crucial role for
the efficient HDO of real feed stocks.
Support material which offers intermediate metal–oxygen bond
strength are considered to be the best support materials. Strong
metal–oxygen bonding causes difficulty in the creation of oxygen
vacant sites on the catalyst material to adsorb oxygen containing
compounds. Weak metal–oxygen bonding makes the abstraction of
oxygen from the feed challenging [13]. Basic support offers higher
resistance towards sintering and deactivation, but the conversion
and hydrodeoxygenation rate are reported to be less in comparison
to acidic support materials. Though acidic support materials offer
higher conversions, they suffer from deactivation significantly.
Conclusively, moderate to slightly acidic supports such as TiO2,
ZrO2 and Al2O3 have gained great importance and their usage is
highly recommended for hydrodeoxygenation systems.
7. Promoters and their role
Over the years, cobalt, tungsten and nickel have been used as
promoters for hydrotreating reactions [107,108]. Usage of bimetallic
250 N. Arun et al. / Renewable and Sustainable Energy Reviews 48 (2015) 240–255
or multi-metallic catalysts is gaining considerable importance as
they show two distinct effects: i) ‘Ensemble effect’ caused by the
influence of geometry and ii.) ‘Ligand effect’ due to the influence of
electronic interaction [84]. Usages of bimetallic catalysts are
reported to show better catalytic activity than single metal catalysts.
Amount of promoter used is a crucial parameter and as reported by
[109], excess usage of Ni (promoter) can possible result in formation
of Ni–γ-Al2O3 (Promoter–support interaction) and thus lessening
the availability of active sites (Ni–Mo–S) for catalytic reactions.
Excess promoters can accumulate on the desired MoS2 phase and
make the catalyst inactive. In hydrotreating reactions using molyb-
denum sulfide catalysts, promoters increases the catalytic activity of
the active metal used by donating electrons to the metal, thus
weakening the metal-S bond. Similar trend can be assumed for the
metallic carbide and nitride catalysts.
At lower temperature (80–100 K), the promoter Ni atoms remain
in metallic state at the top of Mo substrate and is not able to remove
S from Mo to form NiSy state. At higher temperatures of 600 K,
metal and substrate intermix significantly and it is essential for the
formation of NiMoSx complex in the case of metallic sulfide catalyst.
Similar effects can be proposed for the carbide and nitride catalysts.
From the same paper, it can be observed that, adding promoters
such as nickel reduces the binding energy of Mo 3d levels for MoSx
catalyst. In Fermi levels located in valence regions, availability of
density of states (DOS) rises and this change is one of the major
reason for the increase in catalytic activity when doping with a
promoter in comparison to catalytic activity without a promoter.
Use of promoters such as boron, creates new Lewis and
Bronsted acid sites, increasing the overall acidity of the catalyst
material. Addition of phosphorous also has similar effect as boron
[110]. Presence of boron alters the oxide structure of molybdenum.
It helps in increasing the octahedral molybdates density in
comparison to tetrahedral MoO24  species and thus promoting
the activity of the catalyst. Moreover, promoters alter the catalyst
geometry by increasing the number of stacks. Usage of promoters
such as cobalt (Co) and lanthanum (La) were reported to increase
selectivity of catalyst (490%) towards oxygen free products [111].
Romero et al. [112] studied on the hydrodeoxygenation of 2-
ethylphenol using unpromoted Mo/Al2O3, CoMo/Al2O3 and NiMo/
Al2O3 catalysts. It was concluded that using promoters increased the
deoxygenation rate. Moreover, nickel was reported to promote only
hydrogenation, but cobalt promoted both hydrogenation and direct
deoxygenation reaction. From their results, it could be seen than the
incorportation of active metal and the promoter seems to increase the
BET surface area of the support material. Generally, clogging of pores
occurs while loading and the increase in the BET area seems to be
quite contradicting. Yang et al. [53] have stated that activity of MoS2
based catalysts for hydroprocesses can be improved especially on HDN
by doping of phosphorous that act as promoter. Villarroel et al. [113]
studied on the promotional effects of promoters like Mn, Fe, Co, Ni, Cu
and Zn through spillover hydrogen for hydrodesulphurization reaction.
The promoters are reported to favor the migration of hydrogen
spillover. One of the major factors that need to be considered during
the choice of promoter is promoter–support interaction. Based on the
catalyst activation conditions and the oxidation state of the metal,
some promoters (ex. Nickel) form strong chemical bond with support
and this reduces the activity of the catalyst.
8. Metallic carbide and nitride catalysts
Review by Furimsky [89] illustrates many important properties
of metallic carbide and nitride catalysts. Metallic carbides and
nitrides of transition metals such as molybdenum and nickel have
been proven to be beneficial for hydrogenation and deoxygenation
reactions. Hydrodeoxygenation activity of metallic carbide or
nitride catalyst will depend on factors such as the anionic vacancies
that are generated by catalyst reduction and the Bronsted acidity of
the support material. Noble catalyst are generally expensive and
this causes limitations in their usage for hydrodeoxygentaion
reactions [87]. Preparation method for catalysts plays a crucial role
as structure properties and chemical composition of catalysts
determines the performance of catalysts.
Temperature programmed reduction (TPR) method developed by
Lee et al. [114] is the most prominent and used method for synthesiz-
ing metallic carbide catalyst and carburization temperature is generally
within 973 K and 1000 K. The oxide precursors are converted to the
active form (sulfide, carbide, phosphide or nitride) before being used in
hydrodeoxygenation of biobased liquid feedstocks. Reports on usage of
phosphide catalyst for the hydrodeoxygenation reaction is very scarce
and the reason is still unknown [115,116]. Catalyst obtained from in-
situ decomposition showed better hydrogenation activity than catalysts
obtained by ex-situ decomposition for the hydrodesulfurization of
dibenzothiphene. In hydrotreating reactions, well dispersed catalysts
can perform better than crystalline catalysts. Degree of stacking has a
crucial effect on the catalytic performance and Yang et al. [117] have
reported that selectivity between hydrogenolysis and hydrogenation
can be affected by the extent of stacking.
Factors such as rate of heating, nitridation temperature and its
duration, hydrocarbon/H2 ratio, carbon and metals sources should
be addressed during development of novel catalysts. Surface carbon
deposition is a major problem during the preparation of carbide
catalysts, if temperature and carburization conditions are not
chosen. Passivation of metallic carbide and nitride catalysts is
important as they are pyrophoric. Usually, 1 % O2 is used for the
passivation process [89]. In the past, metal nitride catalysts have
attracted lot of attention. During the catalyst activation, pure NH3 or
ammonia and helium mixture are used. Mixture of methane and
hydrogen or ethane and hydrogen are used for catalyst reduction
and carburization. Extent of carburization affected the active site
density on Mo2C and hydrocarbon content in H2 mixture influences
carbide structure [89,114,118].
Carbidation or nitridation of metal–oxygen species will produce
water similar to the production of ambient water during sulfidation of
metal–oxygen species. Presence of water weakens the metal–oxygen
bond and decreases the activation energy for carbidation or nitrida-
tion and it also provides more surface protons resulting in lower
activation energy for Bronsted site assisted sulfidation. Transition
metals can be made to resemble noble metals in terms of physico-
chemical properties by means of reduction (carburization/nitridation).
Since group VI metal carbides and nitrides resemble noble metals,
they have fewer tendencies towards recarburization in comparison to
carbides and nitrides of Group V metals. Molybdenum carbides have
hexagonal close-packed structure whereas Molybdenum nitrides
have body-centered cubic crystal structure.
Interaction between Ni–S species has adverse effect on reactiv-
ity of active metal (Nickel). Similarly, in case of hydrodeoxygena-
tion reaction systems, it can be assumed that interaction between
Ni–O species should be lessened. On sites with active metals,
cleavage of C–C bond occurs while acid sites favours cleavage of C–
O bonds occurs by dehydration reactions [119]. Frauwallner et al.
[120] synthesized molybdenum carbide catalyst using toluene as
the carbon source. Raman spectroscopic analysis of Molybdenum
oxides (MoO3 and MoO2) shows sharp nature of the peak indicat-
ing that the catalyst had a highly ordered structure. During the
reduction of precursor material (MoO3), the removal of oxygen
results in generation of oxygen vacancies (anionic vacancies).
Mo2C and MoO3 were compared based on Raman shift in the
600–1100 cm  1 region of Raman spectroscopy. Broadening of
peaks was observed in the case of metallic carbide catalysts
indicating that there is decrease in crystallinity as the concentra-
tion of oxygen vacancies is increased.
 N. Arun et al. / Renewable and Sustainable Energy Reviews 48 (2015) 240–255 251

Calcination at high temperature can cause collapse of ordered term high-low surface area seems to be relative and clear and specific
mesoporous structure. It should be noted that catalytic performance of range of surface area for enhanced HDO activity is yet to be documented.
transition or noble metals are better with mesoporous oxide support Depending on the catalyst loading and calcination temperature, the
with high surface area in comparison to commercial supports with low catalyst precursor exists in different forms: (i). tetrahedral species in
surface area and the former can offer higher degree of dispersion with isolated state, (ii). polymolybdates or (iii). Oxides of molybdenum when
increased stability of metal particles upon calcination and reduction. The molybdenum is used as active metal. HDS activity of a sulphided catalyst
is highly dependent on the surface concentration of S2 . Hence, similar
correlation can be observed in case of metallic nitrides (N3 ) and carbide
(1) +H 2
RCOR+CO 2 RCHOHR catalysts.
+RCOOH
RCOOCH2R

(2) +H 2
RCOOH RCHO RCH 2OH
9. HDO mechanism over different catalyst materials
+H 2
RCH 3
-H 2O
(3) +H 2 Understanding the reaction mechanism using model compounds
R+CO 2 /CO CH 4
such as oleic acid and stearic acid will help us to develop efficient
Fig. 11. Main reaction pathways for HDO of carboxylic acids, R represents alkyl catalysts for real feed stocks. HDO of carboxylic acid proceeds by three
groups [13] (Reproduced with permission from publisher). routes: Route 1 involves ketonization by the cleavage of C–O bond

Catalyst HDO reaction mechanism

Noble metal catalysts
(Mechanism a)

Noble metal catalysts

(Mechanism b)

Transition metallic
phosphide catalysts

Transition metallic
carbide and nitride
catalysts
(M1: unsaturated
metals, M2: C or N)

Fig. 12. HDO mechanism over different catalyst materials [13].

Note: Blue-Metal 1; Brown-Metal 2.
252 N. Arun et al. / Renewable and Sustainable Energy Reviews 48 (2015) 240–255
forming ketones which undergoes further hydrogenation to produce
alcohols. Alcohols undergo further deoxygenation (dehydration and
hydrogenation) to produce paraffins. Route 2 involves the formation of
aldehyde through hydrogenolysis by the C–O bond cleavage forming
alcohols. Alcohols can undergo dehydration and hydrogenation to
yield alkanes or they can react with the feed (carboxylic acids) forming
esters. Route 3 involves the breaking of C–C bond forming alkanes. CO2
and CO are formed as byproducts through decarboxylation and
decarbonylation reactions respectively (Fig. 11).
Based on some earlier works, He and Wang [13] proposed HDO
mechanism over noble metal supported catalysts, transition metal-
lic phosphide, carbide and nitride catalysts (Fig. 12). With publish-
er’s permission, Fig. 11 was reproduced from He and Wang [13] and
with publisher’s permission, Fig. 12 was redrawn based on the
illustrations presented in the same paper. In their review work, they
have addressed on the influence of catalyst properties on the
hydrodeoxygenation of pyrolysis bio-oils. Over the noble metal
catalysts, the reaction mechanism can proceed in two ways
(mechanism a and b). In mechanism a, hydrogen is adsorbed on
the noble metal sites and the feed compound gets adsorbed and
activated on either noble metal sites or the metal-support interface.
Adsorbed hydrogen reacts with the adsorbed feed compound by the
spill-over mechanism and forms deoxygenated products and H2O
by the cleavage of C–O bond. Mechanism (b) occurs in the case of
reducible metal oxide species. When the metal oxides are reduced
in the presence of H2, oxygen vacancies are created and the
vacancies are filled by the oxygen from the reacting molecules.
δ
M þ and Bronsted sites acts as the main active phases for
hydrodeoxygenation reaction using transition metallic phosphide
catalysts. Activation of oxygen containing compounds and the
δ
adsorbed hydrogen occurs on the M þ sites and the adsorbed feed
δ
compound reacts with H atoms from M þ sites and Bronsted PO–
H sites to form deoxygenated products.
In the case of metallic nitride catalysts, activity depends on
metal oxidations state and the N deficiency. Transition metallic
carbide catalysts offers the permeation of carbon atoms and it
increases the d-band electron density by lengthening the distance
of metal–metal species of transition metals.
Kim et al. [50] studied on the hydrodeoxygenation of sorbitol
using Pt/Zr phosphate and Pt–ReOx/C catalyst and analyzed the
influence of the support material on extent of hydrodeoxygenation.
It is reported that Pt–ReOx/C catalyst shower 34 times higher
reaction rate than Pt/Zr-P. Over the carbon surface, the Pt particles
had a less diameter (1.2 nm) in comparison to zirconium phosphate
Fig. 13. Probable structure of Pt/Zr-P and Pt–ReOx/C under reaction conditions [50] (Reproduced with permission from publisher).
surface (26 nm). Particle size of the active metal plays a crucial role
in providing active sites for the adsorption of the reactant mole-
cules. Activity of Pt/Zr-P for C–O vs. C–C bond cleavage is reported to
be higher than Pt–ReOx/C catalyst. From the diagram (Fig. 13), it can
be seen that the acid sites at the location of Pt/Zr-P species are
separated from the sites where metal Pt sites are present. However,
acid sites on Pt–ReOx/C are in close proximity to metal Pt sites. Acid
sites’ location in relation to metal site is important to determine
catalyst performance [50]. As a conclusion, it could be stated that,
for better C–O bond cleavage, the catalyst should be designed in
such a way that the metal site and acid site are atomically separated
(Fig. 13).
Increase in acidity aids hydrogenolysis reaction which is very
crucial for breaking the C–O bond in bio-based oils/feedstocks during
hydrodeoxygenation. As stated by Ferdous et al. [109], HDS, HDN or
HDO activity depend on the extent of dispersion of active metal on
the support material and also on the state of the active metal phase.
Energies involved in the bonding of C–O, C–N and C–S is essential to
design the catalyst. The CQO bond is stronger (745 kJ/mole) than
CQN (615 kJ/mole) and C–S bonds (477 kJ/mol). In case of single
bonds, the energy of the bonds are as follows
C–O: 351 kJ/mole; C–N: 276 kJ/mole; C–S: 255 kJ/mole
All these values indicate that the order of difficulty in hydro-
treating can be arranged as follows: HDO4HDN 4HDS. Little
information is available on the difficulty level of hydrodemetalli-
zation reactions.
10. Catalyst activation and deactivation
Rate of hydroprocessing reaction depends on active hydrogen
available on catalyst surface [89]. Hydrogen activation happens at the
metallic sites and activation process can be hindered if the adsorp-
tion and hydrogen activation is slower than the adsorption rate of
reactant molecules and product must desorb rapidly so that another
cycle of adsorption of reactant–activation of molecules–transfer to
active site–desorption of products occurs. Information available on
metal carbides is less than metal nitrides and this may be due to the
complex procedure involved with the preparation of metal carbides.
According to Furimsky [89], “Molybdenum carbides and nitrides
are capable of adsorbing, activating and transferring active surface
hydrogen to reactant molecules. Presence of both metallic molybde-
num and oxygen in the catalyst surface will modify hydrogen
adsorption/activation compared with the pure carbides and nitrides.
 N. Arun et al. / Renewable and Sustainable Energy Reviews 48 (2015) 240–255
The crystal structure of metal carbides and nitrides is favorable for
the migration of active hydrogen from the surface to sub-surface.
However, the complete saturation of the sub-surface with hydrogen
may prevent loss of the surface hydrogen due to its migration to the
sub-surface. Generally, promoted (bi-metal) carbides and nitrides are
reported to exhibit higher catalytic activity than single metal Mo
carbides and nitrides. The role of support on the activity of Mo
carbides is yet to be investigated". Unsupported and supported Mo2C
and Mo2N retained their crystallographic structure for long time
( 10 days) during hydroprocessing reactions and hydrogen is
contained in crystallites surface and sub-surface region. Decrease in
surface area and increase in particle size cause considerable incre-
ment in the catalyst activity per unit surface. Information on carbides
and nitrides of metals except molybdenum is limited.
Catalyst stability is an important parameter governing the
design of catalyst. Mostly, catalyst activity decreases with time
and it is due to the localized blocking of active edges of catalysts.
Additionally, coke formation on catalyst surface leads to fouling. In
general, catalyst acidity and its deactivation rate are directly related.
Catalyst deactivation during hydrodeoxygenation can pose serious
problem on the reusability of the catalyst. Deactivation could be
temporary or permanent and permanent deactivation requires the
replacement of catalyst. It could be caused by catalytic poisoning or by
loss of surface area (sintering). Temporary deactivation is generally
caused by deposition of carbon or sulfur and the catalyst can be
restored by gasification using air/stream mixtures. Properly designed
catalyst will represent minimum deactivation. All catalyst undergoes
deactivation (reversible/irreversible) at the hydrodeoxygenation reac-
tion conditions. However, the deactivation rate changes with catalyst.
In comparison to carbide and nitride catalysts, rate of catalyst deactiva-
tion due to coke formation is higher in the case of sulphided catalysts
[121]. Using H2S instead of CS2 as sulfur source in preparing sulphided
catalysts shifted the selectivity from C7 to C6 hydrocarbons. Leiva et al.
[135] investigated on the influence of water on the hydrodeoxygena-
tion of model compounds (2-methoxyphenol and phenol) using ReS2/
SiO2 catalyst. Influence of water on the hydrodeoxygenation activity of
transition metallic catalysts is still under debate. Some reports indicate
the benign effects of water while few literatures reports on the
decrease in catalytic activity. Water content is reported to reduce the
dispersion of active metal and also inhibition of intermediate reaction
pathways. Additionally, degree of coke formation on catalyst depends
on temperature and pressure of reaction. Higher reaction pressure can
help in reduced coke formation. At any reaction temperature between
350 1C and 375 1C, the degree of coke formation is minimum. At lower
temperatures, preference for hydrodeoxygenation reaction can be
subdued in comparison to other polymerization reactions if aromatic
compounds are used as feed stocks. Generally at higher temperatures,
thermal reactions which produce coke seems to occur and thus coke
yield can increase.
11. Conclusions
Overall, studies on hydrodeoxygenation of bio-based feedstocks to
produce green diesel (paraffins and olefins) is gaining attention in this
decade. Choice of feed stocks for the study should be based on the
local availability of the feedstock and its chemical properties such as
degree of unsaturation and the fatty acid content. Degree of unsatura-
tion in a feed stock plays a vital role in altering the reaction pathway
and hence the product selectivity. Detailed study on reaction pathway
using model compounds such as fatty acids (oleic acid, palmitic acid)
or triolein is essential before using real feed stocks. From this review, it
is very evident that acidic support materials such as ZrO2, TiO2, zeolites
are preferred over basic and neutral support materials. Tailoring the
acidity of mesoporous support materials such as SBA-15, HMS, KlT-6
and MCM-41 using aluminum, titanium or zirconium is essential as
253
these materials possess the desired physio-chemical properties but
have low acidity. Noble metals such as palladium, platinum and
transition metals such as nickel, molybdenum and cobalt are the most
preferred active metals and bimetallic systems generally show higher
activity for hydrodeoxygenation.
Metals such as tungsten (W), vanadium (V) and iron (Fe) can
work for hydrodeoxygenation reaction as they have high affinity
towards oxygen and offer different oxidation states. Reports on the
application of W, V and Fe for HDO reactions are scarce. Active sites
of a catalyst are related to morphology and structure of catalyst and it
is possible to establish a quantitative relation between hydrotreating
activity and number/type of active sites. Hence, designing and
controlling the exposed active sites can aid in altering the catalyst
reactivity. Reduction temperature of a catalyst can play a vital role in
the catalytic activity. Introduction of rare earth elements during
catalyst preparation can improve the thermal stability of the catalyst.
In hydrodeoxygenation reactions, promoters play a major role in
increasing the conversion rate and product selectivity. The most
important factor to be considered for the choice of promoter is
promoter–support interaction. Synthesized catalyst should have
minimum promoter–support interaction to ensure that the forma-
tion of active sites are not hindered. For hydrodeoxygenation reac-
tions, it is recommended that metallic carbides and nitride catalyst
should be preferred over sulfide catalysts as the sulfur content in the
obtained products is of great environmental concern. Hydrodeox-
ygenation reaction is a process of great commercial significance and
research work should continue until the process is commercially
viable worldwide. Studies on engine tests using green diesel synthe-
sized through HDO route, emission studies and process economics
can complement the research and development activities related to
the upgradation of bio-oils and bio-based feedstocks.
Acknowledgments
Financial support from Natural Sciences and Engineering
Research Council of Canada (NSERC) is greatly acknowledged.
References
[1] Romero Y, Richard F, Reneme Y, Brunet S. Hydrodeoxygenation of benzo-
furan and its oxygenated derivatives (2,3-dihydrobenzofuran and 2-ethyl-
phenol) over NiMoP/Al2O3. Appl Catal A: Gen 2009;353:46–53.
[2] Azadi P, Inderwildi OR, Farnood R, King DA. Liquid fuels, hydrogen and chemicals
from lignin: a critical review. Renew Sustain Energy Rev 2013;21:506–23.
[3] Herskowitz M, Landau MV, Reizner Y, Berger D. A commercially-viable, one-
step process for production of green diesel from soybean oil on Pt/SAPO-11.
Fuel 2013;111:157–64.
[4] Kim SK, Brand S, Lee H, Kim Y, Kim J. Production of renewable diesel by
hydrotreatment of soybean oil: effect of reaction parameters. Chem Eng J
2013;228:114–23.
[5] International Energy Outlook; 2011. DOE/EIA-0484.
[6] Sheldon RA. Utilisation of biomass for sustainable fuels and chemicals:
molecules, methods and metrics. Catal Today 2011;167:3–13.
[7] Miskolczi N, Buyong F, Angyal A, Williams PT, Bartha L. Two stages catalytic
pyrolysis of refuse derived fuel: production of biofuel via syncrude. Bioresour
Technol 2010;101:8881–90.
[8] World Energy Outlook; 2009. International Energy Agency (IEA).
[9] DaimelerChrysler, biofuels conference. Berlin, 16–17 May; 2006.
[10] Furimsky E. Catalytic hydrodeoxygenation. Appl Catal A: Gen 2000;199:147–90.
[11] Choudhary TV, Phillips CB. Renewable fuels via catalytic hydrodeoxygena-
tion. Appl Catal A: Gen 2011;397:1–12.
[12] Bu Q, Lei H, Zacher AH, Wang L, Ren S, Liang J, et al. A review of catalytic
hydrodeoxygenation of lignin-derived phenols from biomass pyrolysis.
Bioresour Technol 2012;124:470–7.
[13] He Z, Wang X. Hydrodeoxygenation of model compounds and catalytic
systems for pyrolysis bio-oils upgrading. Catal Sustain Energy 2012;1:28–52.
[14] Mohammad M, Kandaramath T, Yaakob Z, Chandra Y. Overview on the
production of paraffin based-biofuels via catalytic hydrodeoxygenation.
Renew Sustain Energy Rev 2013;22:121–32.
[15] Xu JM, Jiang JC, Sun YJ, Lu YJ. Bio-oil upgrading by means of ethyl ester
production in reactive distillation to remove water and to improve storage
and fuel characteristics. Biomass Bioenergy 2008;32:1056–61.
254 N. Arun et al. / Renewable and Sustainable Energy Reviews 48 (2015) 240–255
[16] Elliott DC, Hart TR. Catalytic hydroprocessing of chemical models for bio-oil.
Energy Fuels 2009;23:631–7.
[17] Donnis B, Egeberg RG, Blom P, Knudsen KG. Hydroprocessing of bio-oils and
oxygenates to hydrocarbons. Understanding the reaction routes. Top Catal
2009;52:229–40.
[18] Mahfud FH, Geel FPV, Venderbosch RH, Heeres HJ. Acetic acid recovery from
fast pyrolysis oil. An exploratory study on liquid–liquid reactive extraction
using aliphatic tertiary amines. Sep Sci Technol 2008;43:3056–74.
[19] Luo J, Xu Y, Zhao L, Dong L, Tong D, Zhu L, Hu C. Two-step hydrothermal
conversion of Pubescens to obtain furans and phenol compounds separately.
Bioresour Technol 2010;101:8873–80.
[20] Toor SS, Rosendahl L, Rudolf A. Hydrothermal liquefaction of biomass: a
review of subcritical water technologies. Energy 2011;36:2328–42.
[21] Nokkosmaki MI, Kuoppala ET, Leppamaki EA. Krause AOI. Catalytic conver-
sion of biomass pyrolysis vapours with zinc oxide. J Anal Appl Pyrolysis
2000;55:119–31.
[22] Graca I, Lopes JM, Ribeiro MF, Ribeiro FR, Cerqueira HS. Almeida MBB.
Catalytic cracking in the presence of guaiacol. Appl Catal B: Environ
2011;101:613–21.
[23] Chiaramonti D, Bonini M, Fratini E, Tondi G, Gartner K, Bridgewater AV,
Grimm HP, Soldaini I, Webster A, Baglioni P. Development of emulsions from
biomass pyrolysis liquid and diesel and their use in engines – Part 1:
emulsion production. Biomass Bioenergy 2003;25:85–99.
[24] Chiaramonti D, Bonini M, Fratini E, Tondi G, Gartner K, Bridgewater AV,
Grimm HP, Soldaini I, Webster A, Baglioni P. Development of emulsions from
biomass pyrolysis liquid and diesel and their use in engines – Part 2: tests in
diesel engines. Biomass Bioenergy 2003;25:101–11.
[25] Zhang Q, Chang J, Tiejun Xu Y. Upgrading bio-oil over different solid
catalysts. Energy Fuel 2006;20:2717–20.
[26] Li W, Pan C, Zhang Q, Liu Z, Peng J, Chen P, Lou H, Zheng X. Upgrading of low-
boiling fraction of bio-oil in supercritical methanol and reaction network.
Bioresour Technol 2011;102:4884–9.
[27] Dickinson JG, Poberezny JT, Savage PE. Deoxygenation of benzofuran in
supercritical water over a platinum catalyst. Appl Catal B: Environ 2012;123-
124:357–66.
[28] Duan P, Savage RE. Catalytic hydrotreatment of crude algal bio-oil in
supercritical water. Appl Catal B: Environ 2011;104:136–43.
[29] Mercader FM, Groeneveld MJ, Kersten SRA, Way NWJ, Schaverien CJ,
Hogendoorn JA. Production of advanced biofuels: co-processing of upgraded
pyrolysis oil in standard refinery units. Appl Catal B: Environ 2010;96:57–66.
[30] Lappas AA, Bezergianni S, Vasalos IA. Production of biofuels via co-
processing in conventional refining processes. Catal Today 2009;145:55–62.
[31] Damartzis T, Zabaniotou A. Thermochemical conversion of biomass to
second generation biofuels through integrated process design – a review.
Renew Sustain Energy Rev 2011;15:366–78.
[32] Naik S, Goud VV, Rout PK, Jacobson K, Dalai AK. Characterization of Canadian
biomass for alternative renewable biofuel. Renew Energy 2010;35:1624–31.
[33] Wang W, Yang Y, Luo H, Liu W. Effect of additive (Co, La) for Ni–Mo–B
amorphous catalyst and its hydrodeoxygenation properties. Catal Commun
2010;11:803–7.
[34] Wang Y, He T, Liu K, Wu J, Fang Y. From biomass to advanced bio-fuel by
catalytic pyrolysis/hydro-processing: hydrodeoxygenation of bio-oil derived
from biomass catalytic pyrolysis. Bioresour Technol 2012;108:280–4.
[35] Zhang L, Liu R, Yin R, Mei Y. Upgrading of bio-oil from biomass fast pyrolysis
in China: a review. Renew Sustain Energy Rev 2013;24:66–72.
[36] Huber GW, Iborra S, Corma A. Synthesis of transportation fuels from biomass:
chemistry, catalysts, and engineering. Chem Rev 2006;106:4044–98.
[37] Majhi A, Sharma YK, Bal R, Behera B, Kumar J. Upgrading of bio-oils over
PdO/Al2O3 catalyst and fractionation. Fuel 2013;107:131–7.
[38] Yunquan Y, Hean L, Gangsheng T, Smith KJ, Thian TC. Hydrodeoxygenation of
phenolic model compounds over MoS2 catalysts with different structures.
Chin J Chem Eng 2008;16:733–9.
[39] Sebos I, Matsoukas A, Apostolopoulos V, Papayannakos N. Catalytic hydro-
processing of cottonseed oil in petroleum diesel mixtures for production of
renewable diesel. Fuel 2009;88:145–9.
[40] Dupont C, Lemeur R, Daudin A, Raybaud P. Hydrodeoxygenation pathways
catalyzed by MoS2 and NiMoS active phases: a DFT study. J Catal 2011;279:276–86.
[41] Sirvastava A, Prasad R. Triglycerides-based diesel fuels. Renew Sustain
Energy Rev 2000;4:111–33.
[42] Mori A, Mizusaki T, Ikawa T, Maegawa T, Monguchi Y, Sajiki H. Palladium on
carbon-diethylamine-mediated hydrodeoxygenation of phenol derivatives
under mild conditions. Tetrahedron 2007;63:1270–80.
[43] Bui VN, Laurenti D, Delichere P, Geantet C. Hydrodeoxygenation of guaiacol:
Part II: support effect for CoMoS catalysts on HDO activity and selectivity.
Appl Catal B: Environ 2011;101:246–55.
[44] Filley J, Roth C. Vanadium catalyzed guaiacol deoxygenation. J Mol Catal A:
Gen 1999;139:245–52.
[45] Bykova MV, Ermakov DY, Kaichev VV, Bulavchenko OA, Saraev AA, Lebedev
MY, Yakovlev VA. Ni-based sol–gel catalysts as promising systems for crude
bio-oil upgrading: guaiacol hydrodeoxygenation study. Appl Catal B: Environ
2012;113–114:296–307.
[46] Ghampson IT, Sepulveda C, Garcia R, Radovic LR, Fierro JLG, DeSisto WJ,
Escalona N. Hydrodeoxygenation of guaiacol over carbon-supported molyb-
denum nitride catalysts: effects of nitriding methods and support properties.
Appl Catal A: Gen 2012;439–440:111–24.
[47] Bui VN, Laurenti D, Afanasiev P, Geantet C. Hydrodeoxygenation of guaiacol
with CoMo catalysts. Part I: promoting effect of cobalt on HDO selectivity and
activity. Appl Catal B: Environ 2011;101:239–45.
[48] Olcese RN, Bettahar M, Petitjean D, Malaman B, Giovanella F, Dufour A. Gas-
phase hydrodeoxygenation of guaiacol over Fe/SiO2 catalyst. Appl Catal B:
Environ 2012;115– 116:63–73.
[49] Senol OI, Ryymin EM, Viljava TR. Krause AOI. Reactions of methyl heptanoate
hydrodeoxygenation on sulphided catalysts. J Mol Catal A: Chem 2007;268:1–8.
[50] Kim YT, Dumesic JA, Huber GW. Aqueous-phase hydrodeoxygenation of
sorbitol: a comparative study of Pt/Zr phosphate and Pt–ReOx/C. J Catal
2013;304:72–85.
[51] Ramirez A, Royo C, Romeo E, Faria J, Resasco DE. Modelling of experimental
vanillin hydrodeoxygenation reactions in water/oil emulsions. Effects of
mass transport. Catal Today 2013;210:89–97.
[52] Joshi N, Lawal A. Hydrodeoxygenation of acetic acid in a microreactor. Chem
Eng Sci 2012;84:761–71.
[53] Yang Y, Gilbert A, Xu C. Hydrodeoxygenation of bio-crude in supercritical
hexane with sulfide CoMo and CoMoP catalysts supported on MgO: a model
compound study using phenol. Appl Catal A: Gen 2009;360:242–9.
[54] Kubicka D, Kaluza L. Deoxygenation of vegetable oils over sulphided Ni, Mo
and NiMo catalysts. Appl Catal A: Gen 2010;372:199–208.
[55] Templis C, Vonortas A, Sebos I, Papayannakos N. Vegetable oil effect on gasoil HDS
in their catalytic co-hydroprocessing. Appl Catal B: Environ 2011;104:324–9.
[56] Toba M, Abe Y, Kuramochi H, Osaka M, Mochizuki T, Yoshimura Y. Hydro-
deoxygenation of waste vegetable oil over sulfide catalysts. Catal Today
2011;164:533–7.
[57] Monnier J, Sulimma H, Dalai AK, Caravaggio G. Hydrodeoxygenation of oleic
acid and canola oil over alumina-supported metal nitrides. Appl Catal A: Gen
2010;382:176–80.
[58] Tang Y, Yu WJ, Mo LY, Lou H, Zheng XM. One-step hydrogenation esterifica-
tion of aldehyde and acid to ester over bifunctional Pt catalysts: a model
reaction as novel route for catalytic upgrading of fast pyrolysis bio-oil.
Energy Fuels 2008;22:3484–8.
[59] Xu Y, Wang T, Ma L, Zhang Q, Liang W. Upgrading of the liquid fuel from fast
pyrolysis of biomass over MoNi/γ-Al2O3 catalysts. Appl Energy
2010;87:2886–91.
[60] Gunawan R, Li X, Lievens C, Gholizadeh M, Chaiwat W, Hu X, et al.
Upgrading of bio-oil into advanced biofuels and chemicals. Part I. Transfor-
mation of GC-detectable light species during the hydrotreatment of bio-oil
using Pd/C catalyst. Fuel 2013;111:709–17.
[61] Pham TN, Shi D, Resasco DE. Evaluating strategies for catalytic upgrading of
pyrolysis oil in liquid phase. Appl Catal B: Environ 2014;145:10–23.
[62] Popov A, Kondratieva E, Mariey L, Goupil JM, El Fallah J, Gilson J-P, et al. Bio-
oil hydrodeoxygenation: adsorption of phenolic compounds on sulfided (Co)
Mo catalysts. J Catal 2013;297:176–86.
[63] Kulkarni MG, Dalai AK, Bakhshi NN. Transesterification of canola oil in mixed
methanol/ethanol system and use of esters as lubricity additive. Bioresour
Technol 2007;98:2027–33.
[64] Ong HC, Mahlia TMI, Masjuki HH, Norhasyima RS. Comparison of palm oil,
Jatropha curcas and Calophyllum inophyllum for biodiesel: a review. Renew
Sustain Energy Rev 2011;15:3501–15.
[65] Biktashev SA, Usmanov RA, Gabitov RR, Gazizov RA, Gumerov FM, Gabitov FR,
Abdulagatov IM, Yarullin RS, Yakushev I. Transesterification of rapeseed and palm
oils in supercritical methanol and ethanol. Biomass Bioenergy 2011;35:2999–3011.
[66] Sankaranarayanan TM, Pandurangan A, Banu M, Sivasanker S. Transester-
ification of sunflower oil over MoO3 supported on alumina. Appl Catal A: Gen
2011;409:239–47.
[67] Naik M, Meher LC, Naik SN, Das LM. Production of biodiesel from high free
fatty acid Karanja (Pongamia pinnata) oil. Biomass Bioenergy
2008;32:354–7.
[68] Patil P, Deng S, Rhodes JI, Lammers PJ. Conversion of waste cooking oil to
biodiesel using ferric sulfate and supercritical methanol processes. Fuel
2010;89:360–4.
[69] Bautista LF, Vicente G, Rodriguez R, Pacheco M. Optimisation of FAME
production from waste cooking oil biodiesel use. Biomass Bioenergy
2009;33:862–72.
[70] Jacobson K, Gopinath R, Meher LC, Dalai AK. Solid acid catalyzed biodiesel
production from waste cooking oil. Appl Catal B: Environ 2008;85:86–91.
[71] Ben H, Mu W, Deng Y, Ragauskas AJ. Production of renewable gasoline from
aqueous phase hydrogenation of lignin pyrolysis oil. Fuel 2013;103:1148–53.
[72] http://www.chempro.in/fattyacid. htm [Date accessed 12.05.13].
[73] Dmytryshyn SL, Dalai AK, Chaudhari ST, Mishra HK, Reaney MJ. Synthesis
and characterization of vegetable oil derived esters: evaluation for their
diesel additive properties. Bioresour Technol 2004;92:55–64.
[74] Issariaykul T, Dalai AK. Biodiesel production from greenseed canola oil.
Energy Fuels 2010;24:4652–8.
[75] Mungroo R, Goud VV, Naik SN, Dalai AK. Utilization of green seed canola oil
for in situ epoxidation. Eur J Lipid Sci Technol 2011;113:768–74.
[76] Cerny R, Kubu M, Kubicka D. The effect of oxygenates structure on their
deoxygenation over USY zeolite. Catal Today 2013;204:46–53.
[77] Liu Y, Liu Y, Liu D, Cao T, Han S, Xu G. Design of CO2 hydrogenation catalyst
by an artificial neural network. Comput Chem Eng 2001;25:1711–4.
[78] Huang K, Zhan XL, Chen FQ, Lu DW. Catalyst design for methane oxidative
coupling by using artificial neural network and hybrid genetic algorithm.
Chem Eng Sci 2003;58:81–7.
 N. Arun et al. / Renewable and Sustainable Energy Reviews 48 (2015) 240–255
[79] Huang K, Chen FQ, Lu DW. Artificial neural network-aided design of a multi-
component catalyst for methane oxidative coupling. Appl Catal A: Gen
2001;219:61–8.
[80] Gunay ME, Yildirim R. Analysis of selective CO oxidation over promoted Pt/
Al2O3 catalysts using modular neural networks: combining preparation and
operational variables. Appl Catal A: Gen 2010;377:174–80.
[81] Omata K, Kobayashi Y, Yamada M. Artificial neural network-aided design of
Co/SrCO3 catalyst for preferential oxidation of CO in excess hydrogen. Catal
Today 2006;117:311–5.
[82] Serra JM, Corma A, Argente E, Valero S, Botti V. Neural networks for
modeling of kinetic reaction data applicable to catalyst scale up and process
control and optimization in the frame of combinatorial catalysis. Appl Catal
A: Gen 2003;254:133–45.
[83] Kito S, Satsuma A, Ishikura T, Niwa M, Murakami Y, Hattori T. Application of
neural network to estimation of catalyst deactivation in methanol conver-
sion. Catal Today 2004;97:41–7.
[84] Trimm DL. Design of industrial catalysts. Elsevier Scientific Publishing
Company, The Netherlands; 1980.
[85] Senol OI, Viljava TR. Krause AOI. Hydrodeoxygenation of aliphatic esters on
sulphided NiMo/γ-Al2O3 and CoMo/γ-Al2O3 catalyst: the effect of water.
Catal Today 2005;106:186–9.
[86] Wawrzetz A, Peng B, Hrabar A, Jentys A, Lemonidou AA, Lercher JA. Towards
understanding the bifunctional hydrodeoxygenation and aqueous phase
reforming of glycerol. J Catal 2010;269:411–20.
[87] Wildschut J, Melián-Cabrera I, Heeres HJ. Catalyst studies on the hydrotreat-
ment of fast pyrolysis oil. Appl Catal B: Environ 2010;99:298–306.
[88] Senol OI, Viljava TR. Krause AOI. Effect of sulphiding agents on the Hydro-
deoxygenation of methyl esters on sulphided catalysts. Appl Catal A: Gen
2007;326:236–44.
[89] Furimsky E. Metal carbides and nitrides as potential catalysts for hydro-
processing. Appl Catal A: Gen 2003;240:1–28.
[90] Li N, Huber GW. Aqueous-phase hydrodeoxygenation of sorbitol with Pt/
SiO2–Al2O3: Identification of reaction intermediates. J Catal 2010;270:48–59.
[91] Zhao C, He J, Lemonidou AA, Li X, Lercher JA. Aqueous-phase hydrodeox-
ygenation of bio-derived phenols to cycloalkanes. J Catal 2011;280:8–16.
[92] Huang J, Jiang Y, Vegten N, Hunger M, Baiker A. Tuning the support acidity of
flame-made Pd/SiO2–Al2O3 catalysts for chemoselective hydrogenation. J
Catal 2011;281:352–60.
[93] Ardiyanti AR, Khromova SA, Venderbosch RH, Yakovlev VA, Heeres HJ.
Catalytic hydrotreatment of fast-pyrolysis oil using non-sulfided bimetallic
Ni–Cu catalysts on a δ-Al2O3 support. Appl Catal B: Environ 2012;117–
118:105–17.
[94] Lee J, Xu Y, Huber GW. High-throughput screening of monometallic catalysts
for aqueous-phase hydrogenation of biomass-derived oxygenates. Appl Catal
B: Environ 2013;141:98–107.
[95] Echeandia S, Arias PL, Barrio VL, Pawelec B, JLG Fierro. Synergy effect in the
HDO of phenol over Ni–W catalysts supported on active carbon: effect of
tungsten precursors. Appl Catal B: Environ 2010;101:1–12.
[96] Elliott DC. Historical developments in hydroprocessing bio-oils. Energy Fuels
2007;21:1792–815.
[97] Serrano-Ruiz JC, Braden DJ, West RM, Dumesic JA. Conversion of cellulose to
hydrocarbon fuels by progressive removal of oxygen. Appl Catal B: Environ
2010;100:184–9.
[98] Chiappero M, Do PTM, Crossley S, Lobban LL, Resasco DE. Direct conversion
of triglycerides to olefins and paraffins over noble metal supported catalysts.
Fuel 2011;90:1155–65.
[99] Olcese R, Bettahar MM, Malaman B, Ghanbaja J, Tibavizco L, Petitjean D,
Dufour A. Gas-phase hydrodeoxygenation of guaiacol over iron-based
catalysts. Effect of gases composition, iron load and supports (silica and
activated carbon). Appl Catal B: Environ 2013;129:528–38.
[100] Nava R, Pawelec B, Castano P, Alvarez-Galvan MC, Loricera CV. Fierro JLG.
Upgrading of bio-liquids on different mesoporous silica-supported CoMo
catalysts. Appl Catal B: Environ 2009;92:154–67.
[101] Kubička D, Horáček J, Setnička M, Bulánek R, Zukal A, Kubičková I. Effect of
support-active phase interactions on the catalyst activity and selectivity in
deoxygenation of triglycerides. Appl Catal B: Environ 2014;145:101–7.
[102] Yang Y, Ochoa-Hernández C, O’Shea VA, Pizarro P, Coronado JM, Serrano DP.
Effect of metal-support interaction on the selective hydrodeoxygenation of
anisole to aromatics over Ni-based catalysts. Appl Catal B: Environ
2014;145:91–100.
[103] Zhang X, Wang T, Ma L, Zhang Q, Jiang T. Hydrotreatment of bio-oil over Ni-
based catalyst. Bioresour Technol 2013;127:306–11.
[104] Zhao C, Yu Y, Jentys A, Lercher J. Understanding the impact of aluminum
oxide binder on Ni/HZSM-5 for phenol hydrodeoxygenation. Appl Catal B:
Environ 2013;132-133:282–92.
[105] Velu S, Kapoor MP, Inagaki S, Suzuki K. Vapor phase hydrogenation of phenol
over palladium supported on mesoporous CeO2 and ZrO2. Appl Catal A: Gen
2003;245:317–31.
[106] Nava R, Ortega RA, Alonso G, Ornelas C, Pawelec B. Fierro JLG. CoMo/Ti-SBA-15
catalysts for dibenzothiophene desulfurization. Catal Today 2007;127:70–84.
255
[107] Ferrari M, Delmon B, Grange P. Influence of the impregnation order of
molybdenum and cobalt in carbon-supported catalysts for hydrodeoxygena-
tion reactions. Carbon 2002;40:497–511.
[108] Senol OI, Viljava TR. Krause AOI. Hydrodeoxygenation of methyl esters on
sulphided NiMo/γ-Al2O3 and CoMo/γ-Al2O3 catalysts. Catal Today
2005;100:331–5.
[109] Ferdous D, Dalai AK, Adjaye J. A series of NiMo/Al2O3 catalysts containing
boron and phosphorus Part II. Hydrodenitrogenation and hydrodesulfuriza-
tion using heavy gas oil derived from Athabasca bitumen. Appl Catal A: Gen
2004;260:153–62.
[110] Lewandowski M, Sarbak Z. Effect of boron addition on hydrodesulfurization and
hydrodenitrogenation activity of NiMo/Al2O3 catalysts. Fuel 2000;79:487–95.
[111] Wang W, Yang Y, Luo H, Hu T, Liu W. Amorphous Co–Mo–B catalyst with high
activity for the hydrodeoxygenation of bio-oil. Catal Commun 2011;12:436–40.
[112] Romero Y, Richard F, Brunet S. Hydrodeoxygenation of 2-ethylphenol as a
model compound of bio-crude over sulfided Mo-based catalysts: promoting
effect and reaction mechanism. Appl Catal B: Environ 2010;98:213–23.
[113] Villarroel M, Baeza P, Escalona N, Ojeda J, Delmon B, Gil-Llambias FJ. MD//Mo
and MD//W [MD ¼ Mn, Fe, Co, Ni, Cu and Zn] promotion via spillover hydrogen
in hydrodesulfurization. Appl Catal A: Gen 2008;345:152–7.
[114] Lee JS, Oyama ST, Boudart M. Molybdenum carbide catalysts: I. Synthesis of
unsupported powders. J Catal 1987;106:125–33.
[115] Peterson AA, Vogel F, Lachance RP, Fröling M, Antal M, Tester JW. Thermo-
chemical biofuel production in hydrothermal media: a review of sub- and
supercritical water technologies. Energy Environ Sci 2008;1:32–65.
[116] Yang Y, Ochan-Hernandez C, O’Shea VA, Coronado JM, Serrano DP. Ni2P/SBA-
15 as a hydrodeoxygenation catalyst with enhanced selectivity for the
conversion of methyl oleate into n-octadecane. ACS Catal 2012;2:592–8.
[117] Yang YQ, Tye CT, Smith KJ. Influence of MoS2 catalyst morphology on the
hydrodeoxygenation of phenols. Catal Commun 2008;9:1364–8.
[118] Choi JS, Bugli G, Mariadassou GD. Influence of the degree of carburization on
the density of sites and hydrogenating activity of molybdenum carbides.
J Catal 2000;193:238–47.
[119] Ryymin EM, Honkela ML, Viljava TR. Krause AOI. Competitive reactions and
mechanisms in the simultaneous HDO of phenol and methyl heptonate over
sulphided NiMo/γ-Al2O3. Appl Catal A: Gen 2010;389:114–21.
[120] Frauwallner ML, Lopez-Linares F, Lara-Romero J, Scott CE, Ali V, Hernandez E,
Pereira-Almao P. Toluene hydrogenation at low temperature using a molyb-
denum carbide catalyst. Appl Catal A: Gen 2011;394:62–70.
[121] Xiong WM, Fu Y, Zeng FX, Guo QX. An in situ reduction approach for bio-oil
hydroprocessing. Fuel Process Technol 2011;92:1599–605.
[122] Cecilia J, Infantes-Molina A, Rodríguez-Castellón E, Jiménez-López Oyama ST.
Oxygen-removal of dibenzofuran as a model compound in biomass derived
bio-oil on nickel phosphide catalysts: role of phosphorus. Appl Catal B:
Environ 2013;136-137:140–9.
[123] Horáček J, Šťávová G, Kelbichová V, Kubička D. Zeolite-Beta-supported
platinum catalysts for hydrogenation/hydrodeoxygenation of pyrolysis oil
model compounds. Catal Today 2013;204:38–45.
[124] Ki S, Young J, Hong S, Lee Y, Kim J. Supercritical CO2 purification of waste
cooking oil for high-yield diesel-like hydrocarbons via catalytic hydrodeox-
ygenation. Fuel 2013;111:510–8.
[125] Liu C, Liu J, Zhou G, Tian W, Rong L. A cleaner process for hydrocracking of
jatropha oil into green diesel. J Taiwan Inst Chem Eng 2013;44:221–7.
[126] Ochoa-Hernández C, Yang Y, Pizarro P, O’Shea VA, Coronado JM, Serrano DP.
Hydrocarbons production through hydrotreating of methyl esters over Ni
and Co supported on SBA-15 and Al-SBA-15. Catal Today 2013;210:81–8.
[127] Zhang X, Wang T, Ma L, Zhang Q, Yu Y, Liu Q. Characterization and catalytic
properties of Ni and NiCu catalysts supported on ZrO2–SiO2 for guaiacol
hydrodeoxygenation. Catal Commun 2013;33:15–9.
[128] Alotaibi M, Kozhevnikova EF, Kozhevnikov IV. Efficient hydrodeoxygenation
of biomass-derived ketones over bifunctional Pt-polyoxometalate catalyst.
Chem Commun 2012;48:7194–6.
[129] Anand M, Sinha AK. Temperature-dependent reaction pathways for the
anomalous hydrocracking of triglycerides in the presence of sulfided Co–
Mo-catalyst. Bioresour Technol 2012;126:148–55.
[130] Ausavasukhi A, Huang Y, To AT, Sooknoi T, Resasco DE. Hydrodeoxygenation of m-
cresol over gallium-modified beta zeolite catalysts. J Catal 2012;290:90–100.
[131] Chen L, Zhu Y, Zheng H, Zhang C, Li Y. Aqueous-phase hydrodeoxygenation of
propanoic acid over the Ru/ZrO2 and Ru–Mo/ZrO2 catalysts. Appl Catal A: Gen
2012;412:95–104.
[132] Chen L, Zhu Y, Zheng H, Zhang C, Zhang B, Li Y. Catalytic degradation of oxygenates
in Fischer-Tropsch aqueous phase effluents to fuel gas via hydrodeoxygenation
over Ru/AC catalyst. J Chem Technol Biotechnol 2012;87:112–22.
[133] Ciftci A, Peng B, Jentys A, Lercher J, Hensen EJM. Support effects in the
aqueous phase reforming of glycerol over supported platinum catalysts. Appl
Catal A: Gen 2012;431–432:113–9.
[134] Chen J, Yang Y, Shi H, Li M, Chu Y, Pan Z, Yu X. Regulating product
distribution in deoxygenation of methyl laurate on silica-supported Ni–Mo
phosphides: effect of Ni/Mo ratio. Fuel 2014;129:1–10.
[135] Leiva K, Sepúlveda C, García R, Fierro JL G, Escalona N. Effect of water on the
conversions of 2-methoxyphenol and phenol as bio-oil model compounds
over ReS2/SiO2 catalyst. Catal Commun 2014;53:33–7.