G Model

CATTOD-9959; No. of Pages 8 ARTICLE IN PRESS

Catalysis Today xxx (2016) xxx–xxx

Contents lists available at ScienceDirect

Catalysis Today
journal homepage: www.elsevier.com/locate/cattod

Recent advances on fast hydropyrolysis of biomass

Fernando L.P. Resende
School of Environmental and Forest Sciences, University of Washington, Seattle, WA 98195, United States

a r t i c l e i n f o a b s t r a c t

Article history: Even though hydropyrolysis of biomass has been studied for many years, it was characterized by long
Received 7 August 2015 residence times and low heating rates. On the other hand, fast hydropyrolysis, the rapid decomposition
Received in revised form 8 January 2016 of an organic material under a hydrogen atmosphere, has been primarily reported only over the last
Accepted 8 January 2016
five years. There is growing interest in the topic, and this brief article reviews fast hydropyrolysis of
Available online xxx
biomass, describing previous findings, current challenges, and research opportunities for the future. The
current literature shows that catalytic fast hydropyrolysis produces primarily aromatic hydrocarbons,
Keywords:
but alkanes and naphthenes can also be produced under appropriate conditions if a secondary unit is
Hydropyrolysis
Fast hydropyrolysis
added for hydrotreating (ex-situ upgrading). Compared to catalytic fast pyrolysis, the higher yields of
Fast pyrolysis hydrocarbons and much slower catalyst deactivation due to coking promising. Yields in the range of
Hydrotreating 80–95 gal/t can be obtained, and the process economics is equivalent to those of other biofuel processes,
Biomass such as fast pyrolysis followed by hydrotreating/hydrocracking.
HDO © 2016 Elsevier B.V. All rights reserved.

1. Introduction of the original biomass can be collected as bio-oil. About 12 wt% of

char, and 13 wt% of permanent gases (CO, CO2 , CH4 ) are also pro-
Liquid fuels for transportation constitute a large fraction of the duced [9,10]. The key to maximize liquid product yield is a short
fuels used worldwide. The world consumption of liquid fuels in residence time for the volatiles (usually a few seconds), in order to
2010 was roughly 87 million barrels/day [1]. With a few exceptions, prevent secondary cracking reactions [9].
these liquid fuels are mixtures of hydrocarbons, which are conve- Bio-oil can be used for applications such as heating, electricity
nient for use and have a high hydrogen content and volumetric generation, and chemicals production [10–12]. Its use as a trans-
energy [2,3]. Transportation fuels are currently obtained from fossil portation fuel, however, remains problematic because of properties
resources, but there is great interest in producing a portion of those associated to the high level of oxygen in the fuel. The oxygen con-
fuels from renewable resources. Biomass is one of the few possible tent in the bio-oil is roughly 35–40 wt%, which leads to a higher
sources of renewable carbon, and therefore the most likely source heating value (HHV) of about 17 MJ/kg, compared to 42–44 MJ/kg
of renewable liquid fuels [4] (CO2 is another potential source). The of conventional gasoline and diesel. In addition, bio-oil is not mis-
Energy Independence and Security Act (EISA) of 2007 requires the cible with conventional hydrocarbon fuels, has high acidity (pH
US to produce 36 billion gallons of renewable fuels/year by the year 2–3, total acid number TAN 100–200), which makes it corrosive
2022 [5,6]. This number corresponds to about 25% of the annual to materials of construction—usually TAN lower than 2 is required
consumption of gasoline in the US [7]. for transport vessels and refinery conversion equipment [13]. Bio-
Several processes have been proposed to produce liquid fuels oil is also unstable because of reactive components such as acids,
from lignocellulosic biomass, and fast pyrolysis is one of the most aldehydes and ketones, which cause polymerization over time and
promising ones. Fast pyrolysis is the rapid thermal degradation of increase the viscosity (this process is called “aging”) [9,14]. It is
organic materials in the absence of oxygen, at rates in the range of believed that temperatures as low as −20 ◦ C are required to stop the
500 ◦ C/s, to a final temperature of 500–600 ◦ C [8]. The rapid heat- chemical mechanisms involved in the aging of the oil [15]. Another
ing decomposes the large molecules of biomass into smaller ones, problem is the high water content mixed with the oil, usually about
which are released in the form of volatile compounds. The volatiles 25 wt%, which decreases the heating value and causes ignition delay
are then quickly cooled back to room temperature, generating a [14,16]. Due to these characteristics, the demand for fast pyrolysis
brownish liquid fuel, usually referred to as “bio-oil”. Up to 75 wt% oil is low [17].
Despite the unfavorable properties, bio-oil can be upgraded to
liquid hydrocarbons that are compatible with the infrastructure
used for transportation fuels, and this is performed by removal of
E-mail address: [email protected]

http://dx.doi.org/10.1016/j.cattod.2016.01.004
0920-5861/© 2016 Elsevier B.V. All rights reserved.

Please cite this article in press as: F.L.P. Resende, Recent advances on fast hydropyrolysis of biomass, Catal. Today (2016),
http://dx.doi.org/10.1016/j.cattod.2016.01.004
G Model
CATTOD-9959; No. of Pages 8 ARTICLE IN PRESS
2 F.L.P. Resende / Catalysis Today xxx (2016) xxx–xxx

Table 1
Summary of previous works on fast hydropyrolysis of biomass.

References System Feedstock Reactor temp. (◦ C) Hydrotreating temp. (◦ C) Pressure (bar) Catalyst

[44,51] Py–GC/MS with Miscanthus × gigantheus 600 300–450 Atmosph.—30 bar ZSM-5, Ni/ZSM-5, MCM-41,
hydrotreating Ni/MCM-41 (hydrot.)
[20,26] Py–GC/MS Pine wood 650 – 5.5–28.0 HZSM-5, Co/ZSM-5, Ni-ZSM-5,
Pt-ZSM-5 (hydrop.)
[50] Py–GC/MS Lignin 650 – Atmosp.—25.8 Pd/HZSM-5 (hydrop.)
[49] Fluidized Bed Woody residue 375–450 – 3.4–20.0 Proprietary (hydrop.)
[52,53,61] Fluidized Bed with Rice husk 450–650 – Atmosp. Ni/LY, Ni/Al2 O3 , dolomite,
single shot for feed CoMo/Al2 O3 (hydrop.)
[17,45] Fluidized Bed with Several 343–469 343–399 19.5–22.4 Proprietary (hydrop.)
hydrotreating
[1,46] Cyclone with Cellulose, poplar 480–580 300–375 27–54 Al2O3, 2% Ru/Al2O3, 2%
hydrotreating Pt/Al2O3, 5 wt% Pt—2.5 wt%
Mo/MWCNT (hydrot.)

oxygen atoms. The most commonly studied option in the past was [41]. Fast hydropyrolysis, on the other hand, is analogous to fast
hydrotreating, which consists of re-evaporating the oil and reacting pyrolysis, with the main difference being the presence of a H2
it with H2 at very high pressures (100–170 bar), 310–375 ◦ C [18], atmosphere. Two recent reviews on hydropyrolysis have been pub-
and low space velocities (0.1–0.2 LHSV), and it typically requires lished [4,47] and describe extensively the previous work with slow
a sulfide CoMo or NiMo catalyst supported on alumina [19,20]. hydropyrolysis. Another review from GTI researchers [48] focuses
Hydrotreating usually requires adding about 3–5 wt% of H2 relative on the integrated hydropyrolysis plus hydroconversion (IH2 ) pro-
to the mass of bio-oil, obtaining a hydrogenated liquid correspond- cess they developed. Recent research efforts have been focused
ing to 26–30 wt% of the original biomass [13], or 58 wt% of the on hydropyrolysis systems that resemble fast pyrolysis systems in
liquid bio-oil [21]. However, this process would be expensive and terms of heat transfer rates and vapors residence time, referred as
slow. Single stage hydrotreatment is considered an inappropriate fast hydropyrolysis. The present review discusses the recent find-
method for upgrade, because of the formation of large quantities ings and developments on this process.
of coke, leading to rapid catalyst deactivation and reactor clogging
[18].
In recent years, significant attention has been devoted to cat- 2. Process description
alytic fast pyrolysis as a method for producing hydrocarbons from
biomass. In this case, fast pyrolysis is performed in the presence Fast hydropyrolysis is the decomposition of an organic mate-
of a catalyst, usually a zeolite [22–25]. The pyrolysis volatiles are rial under high heating rates (about 500 ◦ C/s) in a H2 environment
upgrade in-situ via dehydration, decarboxylation, and decarbony- [1]. It can only be performed in systems that allow short vapor
lation reactions. Since a significant amount of the oxygen is released residence times (a few seconds), such as fluidized bed reactors
in the form of water, a large amount of hydrogen atoms are removed [17,45,49], cyclone reactors [1,46], and micropyrolysis systems
from the liquid product, leading to aromatic hydrocarbons and such as Py–GC/MS (pyroprobe) [20,50]. The H2 may be the only gas,
some light olefins. The great advantage of catalytic fast pyrolysis or may be diluted in an inert gas such as N2 . Fast hydropyrolysis was
is that H2 is not required, and a stable liquid product is formed performed for coal in the past [28–30], and to the best of our knowl-
[26]. However, at best only 20–30% of the carbon in the biomass edge, it was reported first for biomass by Steinberg and Division
ends up in the liquid product. Roughly 30% of the carbon goes to [34] in 1986. However, the temperatures employed in that work
coke, causing catalyst deactivation [27]. were very high (600–1000 ◦ C), so that CH4 and CO were the major
In this review, we discuss hydropyrolysis of biomass, which is products. In this review, we are concerned with fast hydropyrol-
the decomposition in the presence of H2 [3]. Hydropyrolysis has ysis in the range 350–650 ◦ C, which focuses on maximizing yields
been studied initially for coal [28–33], and later to generate hydro- of liquid hydrocarbons. The H2 partial pressure is variable, and it
carbons from biomass [34–43]. In hydropyrolysis, the reducing H2 usually ranges from atmospheric to about 30 bar, though studies
gas generates hydrogen radicals which react with volatiles released with 54 bar have been reported [46]. The process can be catalytic or
by the biomass, usually in the presence of a catalyst, removing oxy- non-catalytic—however, in order to ensure proper level of deoxy-
gen which can be released in the form of water, CO, and CO2 , and genation, some groups have employed a hydrotreating unit right
producing hydrocarbons. Additionally, many of the reactive volatile after the fast hydropyrolysis reactor, so that the volatiles coming
intermediates are capped by the hydrogen radicals. These interme- from the first reactor are immediately upgraded in the second unit
diates would otherwise undergo polymerization, so the addition [1,17,45,46]. This is referred to as ex-situ hydrotreating. Table 1
of H2 lowers the possibility of coking on the catalyst [20,44]. One shows a summary of the works in the literature on fast hydropy-
of the advantages of hydropyrolysis is that the pressures required rolysis of biomass.
are typically much lower than what is needed to hydrotreat bio-oil Fast hydropyrolysis generates two liquid phases: an organic
(about 30 bar compared to 100–170 bar for bio-oil hydrotreating) phase containing a mixture of hydrocarbons which is the target
[17,45]. Another interesting characteristic of hydropyrolysis is that product, and an aqueous phase, along with char and permanent
the process is exothermic, generating heat which helps sustain gases. If a catalyst is used, then coke on the catalyst surface may
endothermal pyrolysis reactions that take place in the reactor [17]. also be produced, though it has been reported that coke forma-
Prior to 2010, most of the work reported for hydropyrolysis of tion is small [17]. Fig. 1 shows the possible configurations for
biomass focused on batch and fixed bed reactors operating at low the catalyst, the fast hydropyrolysis reactor, and the hydrotreat-
heating rates and long residence times [34–42] (here, we refer to ing reactor. A fluidized bed reactor is chosen for the purpose of this
this as slow hydropyrolysis). These conditions, however, are not illustration. All the configurations are discussed next: non-catalytic
consistent with the need for low residence times with rapid heat- fast hydropyrolysis, catalytic fast hydropyrolysis, non-catalytic
ing and cooling which are typical of fast pyrolysis systems [46], fast hydropyrolysis with ex-situ hydrotreating, and catalytic fast
and the oxygen content of the oil obtained is still about 20 wt% hydropyrolysis with ex-situ hydrotreating.

Please cite this article in press as: F.L.P. Resende, Recent advances on fast hydropyrolysis of biomass, Catal. Today (2016),
http://dx.doi.org/10.1016/j.cattod.2016.01.004
G Model
CATTOD-9959; No. of Pages 8 ARTICLE IN PRESS
F.L.P. Resende / Catalysis Today xxx (2016) xxx–xxx 3

Fig. 1. Possible configurations for continuous fast hydropyrolysis systems: (a) non-catalytic fast hydropyrolysis; (b) catalytic fast hydropyrolysis; (c) non-catalytic fast
hydropyrolysis with ex-situ hydrotreating; (d) catalytic fast hydropyrolysis with ex-situ hydrotreating.

2.1. Non-catalytic fast hydropyrolysis H2 causes a small drop in the yield of detectable products. This
decrease in yield is more evident as the pressure increases.
This section discusses the work in which fast hydropyrolysis of Some research groups have reported non-catalytic fast hydropy-
biomass was performed without any added catalysts to the reac- rolysis in fludized beds. RTI International [49] studied Loblolly Pine
tor. Some groups have carried out non-catalytic fast hydropyrolysis in the range 375–450 ◦ C and from 3.4 to 20 bar. N2 was the fluidizing
in a Py–GC/MS. Thangalazhy-Gopakumar et al. [26] reported work agent and H2 was added as reactant gas. In their setup, the N2 -only
with pine wood chips at 550 ◦ C and 5.5 bar performed with H2 in experiments led to a large yield of char (29–40 wt%), which was
Py–GC/MS, and observed that the products are the same as those reduced to about half with H2 at any partial pressure between 0.5
obtained in a He atmosphere, but note a reduction in yield of all and 3.0 bar. In addition, they reported that increasing the H2 partial
oxygenated compounds when the pyrolysis gas was changed from pressure in this range increases the liquid and decreases gas yields.
He to H2 . Jan et al. [50] reported that the products of non-catalytic Higher temperature led to lower liquid yields, and no change in oil
fast hydropyrolysis of lignin at 650 ◦ C are essentially the same as composition was observed. Meesuk et al. also worked with a flu-
the fast pyrolysis products, with phenolic compounds being the pri- idized bed [52] and compared fast hydropyrolysis at atmospheric
mary products, and only trace amounts of aromatic hydrocarbons. pressure with fast pyrolysis at 650 ◦ C, and concluded that the addi-
The effect of H2 pressure was small up to 17 bar, though a slightly tion of H2 produces additional gases like CH4 and CO2 , with lower
increase in phenolic yields is observed, suggesting a higher rate of yield of liquid product. They hypothesize that in the presence of H2 ,
decomposition for larger phenolic molecules. Once the H2 pressure the char is gasified via hydrogasification, and the heavy oil fraction
reached 26 bar, all the yields decreased, suggesting that there may is decomposed. The bio-oil yield decreases from 45 wt% under N2
be secondary cracking of volatiles producing additional perma- to 40% under H2 .
nent gases. Melligan et al. [51] worked with Miscanthus × giganteus Venkatakrishnan et al. [46] reported non-catalytic fast hydropy-
at 600 ◦ C from atmospheric pressure to 30 bar, and reported that rolysis of cellulose in a high-pressure cyclone reactor. They found

Please cite this article in press as: F.L.P. Resende, Recent advances on fast hydropyrolysis of biomass, Catal. Today (2016),
http://dx.doi.org/10.1016/j.cattod.2016.01.004
G Model
CATTOD-9959; No. of Pages 8 ARTICLE IN PRESS
4 F.L.P. Resende / Catalysis Today xxx (2016) xxx–xxx

that the addition of H2 does not alter the composition of the hydrogenation and deoxygenation reactions that convert pheno-
liquid products at 480 ◦ C. When the temperatures reach 580 ◦ C, lic compounds into aromatic hydrocarbons. The presence of Pd did
the molecules appear to have enough energy to react, so that H2 not affect the yields below 17.2 bar H2 partial pressure. The most
consumes light oxygenates like formic acid and glycolaldehyde, abundant aromatic compounds are toluene, xylene, and polyaro-
producing water [46]. In addition, they report that increase in the matic hydrocarbons such as naphthalene and methylnaphthalene.
H2 pressure from 27 to 54 bar decreases liquid yields and increases At these conditions, no hydrogenation of aromatics to cycloalka-
char formation. In this case, the decrease was attributed to slower nes was observed. Hydrogenation is an exothermic reaction, and
gas velocities inside the cyclone reactor at high pressure, which it is thermodynamically unfavorable at high temperatures. The
could lead to less effective ablation of the cellulose particles against equilibrium constant for benzene hydrogenation to cyclohexane
the reactor wall. decreases more than 21,000 times from 300 ◦ C to 650 ◦ C, so negli-
Overall, some observations can be made about the works for gible hydrogenation to cycloalkanes takes place at 650 ◦ C (the fast
non-catalytic fast hydropyrolysis: (1) the addition of H2 gas to a hydropyrolysis temperature investigated). The equilibrium con-
fast pyrolysis system, without added catalysts, is not sufficient for stant for the hydrogenation of various aromatic hydrocarbons is
an extensive hydrogenation of the oxygenated volatiles, since the shown in Fig. 2. When the temperature was dropped to 400 ◦ C,
same products are observed, though some level of deoxygenation is the authors observed a 3 wt% yield of cycloalkanes (25% selectiv-
observed depending on the conditions. This highlights the impor- ity). Ideally, the hydrogenation should be performed at even lower
tance of the catalyst for the effective conversion of biomass into temperatures (300 ◦ C). However, this would decrease the rates of
hydrocarbons; (2) severe conditions (above 500 ◦ C and especially decomposition for lignin.
above 30 bar) appear to cause H2 molecules to react with pyrolysis GTI reports that a single fluidized bed for fast hydropyrolysis
volatiles, generating lower molecular species. As the H2 pressure with a proprietary catalyst achieves significant rates of deoxygena-
increases, additional cracking reactions are promoted. This trend is tion. They have added, though, a hydrotreating unit which appears
especially evident in the case of lignin, with a maximum in the yield to polish the composition of the final product [17]. Their product
of phenol which indicate consumption of high molecular weight analysis focused on fuel properties, though, so the actual hydrocar-
phenolics to generate phenol, and subsequent cracking of phe- bons produced are not reported.
nol to produce permanent gases. Most works indicate decrease in RTI International [49] has performed catalytic fast hydropyroly-
condensable products as the pressure increases, with the yield of sis of sawdust in a fluidized bed reactor. They tested two proprietary
permanent gases increasing. (3) The specifics of the reactor design catalysts, which were mixed with SiC as inert bed material. N2 was
is also a factor in determining the effect of H2 pressure. For reac- used as fluidizing agent, and H2 was added as the reactive gas.
tors that depend on high gas velocity like a cyclone reactor, a large They reported that both catalysts increase gas yield and decrease
increase in pressure with compression of the gas is likely to cause solids yield. The char yield also decreases with temperature. The
negative effects on the yield of liquid products. appearance of the liquid is a clear yellow, in contrast with the
dark brown aspect of fast pyrolysis oil, and the ratio CO/CO2 is
2.2. Catalytic fast hydropyrolysis a strong function of the catalyst used. Their work also included
longer runs (21 h, interrupted, for a total of 10 days), at 375 ◦ C and
This section discusses the work in which fast hydropyrolysis 20.6 bar. The organic liquid was produced at 8.7 wt% yield, with
was performed with a catalyst added to the reactor (with no ex- 4.2 wt% oxygen content. In addition, 25.3 wt% water, 22 wt% solids,
situ upgrade). Thangalazhy-Gopakumar et al. reported pyroprobe and 26.8 wt% gas were produced. The organic liquid is primarily
experiments with pine wood which showed that the addition of a mixture of aromatic hydrocarbons (94 wt%), including benzene
HZSM-5 (catalyst:biomass ratio 5:1) at 5.5 bar produces aromatic derivatives, substituted naphtalenes, and PAHs [49].
hydrocarbons [26]. However, the yields are about the same that are Meesuk et al. [52] reported catalytic fast hydropyrolysis of rice
obtained with HZSM-5 in a helium atmosphere (45 wt%), indicating husk in a fluidized bed at atmospheric pressure. The catalyst was
that the role of H2 was small. mixed with the bed material, and the biomass was fed into the
They also reported that increase in pressure from 5.5 to reactor in a single shot. They compared catalytic fast hydropyroly-
27.5 bar did not increase the yields of aromatic hydrocarbons [20]. sis with catalytic fast pyrolysis, and concluded that the gas yield is
This result contradicted expectations, because higher H2 pressure higher for hydropyrolysis. The oxygen content of the oil was 31 wt%
should increase the quantity of hydrogen free radicals, leading to without a catalyst, and decreased to 26 wt% with Ni/Al2 O3 , 20% with
increased rates of hydrogenation reactions. One possible expla- Ni/LY, 26% with dolomite, and 10% with CoMo/Al2 O3 . The prod-
nation for this unexpected trend is the lack of sufficient contact ucts are aromatic hydrocarbons. The order of catalyst performance
between the volatiles released, the catalyst, and the H2 gas in under the same conditions is CoMo/Al2 O3 > Ni/LY (nickel loaded
the pyroprobe configuration. The fixed bed-like configuration of Loy Yang brown coal) > Ni/Al2 O3 > dolomite > sand. However, Ni/LY
the pyroprobe quartz tube does not promote adequate mixing is more favorable than CoMo/Al2 O3 in terms of cost. With Ni/LY,
of reactants, as would be observed in other systems, like a flu- the maximum yield of liquid product is 47 wt% at 500 ◦ C, with 75%
idized bed. The only effect of increasing pressure appears to be of the bed volume of catalyst relative to the sand. They suggest
on the selectivity: toluene increased and m/p-xylenes decreased that the increase in temperature increases the rates of secondary
[20] which suggests that H2 may be reacting with one of the reactions such as cracking, hydrocracking, and deoxygenation, thus
methyl groups in m/p xylene to form CH4 . It was speculated that decreasing the liquid yield and increasing gas and water yields [53].
H2 would be more active if a hydrogenation catalyst (a metal) had Several observations can be made from the catalytic fast
been added to the HZSM-5 support [26], so in subsequent work hydropyrolysis works: (1) the addition of a catalyst significantly
they doped the HZSM-5 support with Ni, Co, Mo (5 wt%) and Pt improves deoxygenation of the liquid product relative to the non-
(0.5 wt%). The catalysts were equally effective in generating aro- catalytic case, with oxygen levels as low as 4 wt% reported [49];
matic hydrocarbons [20], especially above 20 bar. Hydrogenation (2) the major products of catalytic fast hydropyrolysis are aromatic
of aromatic hydrocarbons was not observed. Similar results were hydrocarbons in the organic liquid phase; (3) an aqueous phase is
reported by Jan et al. [50] for the hydropyrolysis of lignin using formed due to the dehydration reactions; (4) the addition of a sup-
HZSM-5. Pd supported on HZSM-5 (1 wt%) produced 44% more ported metal increases the rate of aromatics production, as long
aromatic hydrocarbons than HZSM-5 at a catalyst-to-lignin ratio as the H2 pressure is more than 15 bar; (5) it does not appear to
of 20:1, 650 ◦ C, and 17.2 bar. So, the metal appears to perform be thermodynamically possible to achieve high rates of thermal

Please cite this article in press as: F.L.P. Resende, Recent advances on fast hydropyrolysis of biomass, Catal. Today (2016),
http://dx.doi.org/10.1016/j.cattod.2016.01.004
G Model
CATTOD-9959; No. of Pages 8 ARTICLE IN PRESS
F.L.P. Resende / Catalysis Today xxx (2016) xxx–xxx 5

Fig. 2. Equilibrium constant for hydrogenation of various aromatic hydrocarbons (adapted from Ref. [50]).

decomposition for all components in lignocellulosic biomass and They hypothesize that the predominant reactions are dehydration
complete HDO reactions of the volatiles in a single unit, because on Al2 O3 , methanation on Ru, and hydrogenation on Pt. In 5 wt%
the temperatures required for each process are different. Pt—2.5 wt% Mo/MWCNT, the role of Pt is hydrogenation, the role of
Mo is to promote C O scission reactions, and MWCNT is an inert
support which is stable at high temperatures. The deoxygenation
2.3. Non-catalytic fast hydropyrolysis with ex-situ hydrotreating ability increased in the order Al2 O3 < Ru/Al2 O3 < Pt/Al2 O3 < 5 wt%
Pt—2.5 wt% Mo/MWCNT. They observed that Ru and Pt reduced
In this section, we discuss the literature work in which the catalyst coking because hydrogenation takes place on the metal
hydropyrolysis reactors do not contain a catalyst, but the product site. Ru and Pt promoted CC cleavage, which contributed to the
vapors immediately flow through a hydrotreating reactor contain- formation of CO in the case of Pt, and CH4 in the case of Ru, decreas-
ing a catalytic bed for upgrade of the products. This is the preferred ing the yields of liquid product, which was composed primarily of
setup adopted by Agrawal’s group at Purdue University [1,46]. C4–C9 hydrocarbons in the form of straight and branched chains, or
This novel process is referred to as H2Bioil, and it is carried out cycloalkanes. An important note is the low yield of char, only 3 wt%,
at 27–54 bar of H2 pressure. In this system, the fast hydropyrol- possibly a combination of ablative heat transfer in the cyclone reac-
ysis reactor is a cyclone at 480–580 ◦ C, and the product vapors tor and the presence of H2 . The yields were largely different for
undergo hydrodeoxygenation (HDO) reactions in the catalytic bed poplar and cellulose. In both cases, there is no catalyst coking.
at 300–375 ◦ C [46]. They note that a one-step process would be ideal Melligan et al. reported fast hydropyrolysis of Miscant-
to produce drop-in liquid fuels from biomass, however, the com- hus × gigantheus in a Py–GC/MS [44] in pressures up to 30 bar,
bination of high temperature (400 ◦ C) and short residence times with the vapors hydrotreated on a secondary unit containing a 10%
(about 1 s) would make it unlikely that a satisfactory level of deoxy- Ni/ZSM-5 catalyst. Ni/HZSM-5 has been shown in the past to be the
genation could be achieved in a single step [46]. The key is to most active among 40 catalysts for catalytic fast pyrolysis. Nickel
upgrade the reactive oxygenated molecules before the occurrence is very active for the hydrogenation of aromatics into cycloalkanes,
of secondary reactions that take place during the condensation of which are high energy density compounds and are important com-
bio-oil. They also point out that separating the fast hydropyroly- ponents of jet fuels. In addition, they also tested ZSM-5, MCM-41,
sis and the hydrotreating catalyst allows for better process control, and Ni-MCM-41 (2.5% and 10%). The main products were aro-
since each process is optimized at a different temperature [1]. An matic hydrocarbons (benzene and toluene). They report that both
HDO temperature higher than 300 ◦ C leads to higher rates of CC nickel-based catalysts consume high molecular weight phenolics,
bond scission, causing loss of carbon in the form of CO. The fast producing light phenols, and this is more evident at the 10% loading.
hydropyrolysis reactor was operated at 480 ◦ C to minimize losses in They observed that increasing the H2 pressure to 30 bar
the form of permanent gases like CO, CO2 , CH4 and light oxygenates decreases the amount of products detected, in particular acetic acid.
[1]. As discussed in the previous section, this issue was also reported They attribute this to the increase in the rates of dehydration and
by Jan et al. [50] for the fast hydropyrolysis of lignin, when it was decarboxylation reactions due to the presence of H2 . They report
observed that the high temperature required for decomposition the production of long chain alkanes in the range C13–C17. The
of the lignin structure (650 ◦ C) is much higher than the tempera- increase in H2 pressure increases the yields of phenolic compounds,
ture needed for the hydrogenation of aromatics to cycloalkanes to confirming the trend discussed for non-catalytic fast hydropyrol-
occur. The need of a secondary hydrotreating unit to complete HDO ysis. They suggest that once the H2 pressure reaches 10 bar, the
reactions is evident. hydroxyl group in phenol is removed to generate benzene, with
The experiments with cellulose and poplar in the cyclone reac- aliphatic hydrocarbons also being produced. At 20 bar, benzene is
tor showed that 55–70 wt% of the carbon in the biomass ends up hydrogenated to produce cyclohexane [44]. ZSM-5 is more efficient
in the liquid hydrocarbon product [1,46]. The process is estimated then MCM-41 in increasing the aromatics, but cycloalkanes were
to require 1240 L H2 at STP/L product (or 0.11 kg H2 /L product). identified in very small quantities [51].
This value is slightly higher than the experimental value of H2 Some important observations can be drawn from the studies
required to upgrade fast pyrolysis oil via hydrotreating (1144 L H2 at with non-catalytic fast hydropyrolysis with ex-situ upgrading: (1)
STP/L product). They estimate the energy efficiency to be 70%. The a secondary hydrotreating unit is essential for the complete HDO
HDO catalysts tested were Al2 O3 , 2% Ru/Al2 O3 , and 2% Pt/Al2 O3 , reactions for the production of hydrocarbon fuels. This is true
5 wt% Pt—2.5 wt% Mo/MWCNT (multiwalled carbon nanotubes).

Please cite this article in press as: F.L.P. Resende, Recent advances on fast hydropyrolysis of biomass, Catal. Today (2016),
http://dx.doi.org/10.1016/j.cattod.2016.01.004
G Model
CATTOD-9959; No. of Pages 8 ARTICLE IN PRESS
6 F.L.P. Resende / Catalysis Today xxx (2016) xxx–xxx

regardless of the presence of a catalyst in the fast hydropyroly- tors, as opposed to only in the hydrotreating unit, remains unclear.
sis reactor; (2) the addition of a hydrotreating unit increases the From a process perspective, though, it is advantageous to have a
range of possible products from fast hydropyrolysis. Rather than catalyst in the first unit, because the hydrogenation reactions are
simply making aromatics in the liquid, the HDO reactions also lead largely exothermic. The heat of reaction released in the first unit
to alkanes and cycloalkanes, though the yield of cycloalkanes in can supply its own energy needs for volatilization of the biomass
most works is still low; (3) catalyst selection is essential to control and the heating needs of the second unit, decreasing overall process
product distribution, and multi-functional catalysts are required heating requirements. In addition, this would decrease the extent
to perform the multitude of reactions involved. Clearly, an effec- of hydrogenation in the second unit, avoiding the need for mul-
tive hydrogenation catalyst is needed, with Pt, Pd, and Ni providing tiple quenches as well. These considerations are important for an
good results in past work. A support that favors dehydration reac- industrial scale unit.
tions will remove oxygen atoms in the form of water as opposed
to CO and CO2 , avoiding loss of carbon to the gas phase. However,
in this case the loss of hydrogen atoms is substantial, and the H2
feed rate needs to increase. Otherwise, the liquid product will be 3. Economic and environmental considerations
hydrogen-deficient and only aromatics will be produced, possibly
along with light olefins, similar to what takes place in catalytic fast Just as with any other process, the question on whether fast
pyrolysis. On the other hand, a catalyst that promotes decarbonyla- hydropyrolysis is an economic viable process is an important one,
tion and decarboxylation will decrease H2 requirements by keeping especially in light of the requirement for H2 gas and its operation
the hydrogen in liquid product, but will also decrease the carbon at relatively high pressures. The groups at Purdue [54] and GTI
recovery in the liquid phase [20]. [13] have worked to address this important question. As reference
for comparison, both studies used data from the Pacific Northwest
2.4. Catalytic fast hydropyrolysis with ex-situ hydrotreating National Lab [55] which estimates the price of gasoline and diesel
from fast pyrolysis followed by hydrotreating and hydrocracking,
In this section, we discuss the work in which the fast hydropy- which is $2.04/gal in a 2007 basis. In this review, we are assum-
rolysis reactor includes a catalyst, and the produced vapors undergo ing a 15.1% cumulative rate of inflation from 2007 to 2015, so all
upgrading in a secondary packed bed unit. Very little work has results are converted to 2015 dollars. This way, the cost of the PNNL
been published for this case, and GTI is the only one that reported process is $2.35/gal, for a biomass cost of $ 58.35/metric tpn.
results for catalytic fast hydropyrolysis with ex-situ upgrading. The The IH2 process proposed by GTI assumes that all the H2 required
process is referred to as IH2 [17,45]. In this process, fast hydropy- can be generated internally, from reforming of C1–C3 hydrocarbons
rolysis is performed in a fluidized bed reactor (H2 is the fluidizing and CO that are released in the product gas. They performed an
agent) containing a proprietary catalyst provided by CRI Catalyst ASPEN simulation for 2000 dry metric tons of woody biomass/day,
Company (Houston, TX). Next, the product vapors flow through a and the minimum fuel selling price is $1.93. Most of the cost is asso-
hydrotreating stage containing another proprietary catalyst pro- ciated to the cost of feedstock and handling ($1.07). It was assumed
vided by CRI. GTI anticipates that, at commercial scale, the C1–C3 that 8% of the catalyst in the fast hydropyrolysis reactor is replaced
hydrocarbons and CO produced can be reformed to generate all each week. This number is not accounting for retail taxes, tax cred-
the H2 required for the process [17]. The process is performed its, costs for distribution, blending and marketing. This cost is about
at 20–35 bar H2 pressure, with the fast hydropyrolysis reactor at $0.58 lower than the cost of producing fast pyrolysis bio-oil [13,56].
343–469 ◦ C, and the hydrotreating temperature ranging from 343 The yield of liquid fuels is 79 gal/dry ton of woody biomass, which
to 399 ◦ C. Proof-of-concept tests were carried out in a 2 kg/h plant, would result in an annual production rate of 61 million gallons. The
and the total yield of organic liquid product (a mixture of hydro- total capital investment is estimated to be roughly $304 millions.
carbons in the gasoline–diesel range) is about 28 wt%. 35–40 wt% The capital costs of a similar plant producing gasoline and diesel
water is obtained in a separate phase, and about 10 wt% char is pro- from hybrid poplar wood chips via fast pyrolysis, hydrotreating,
duced as well. Several fuel properties are reported: 0.3 total acid and hydrocracking is estimated to be $349 millions. The process
number (TAN), 84 research octane number (RON), boiling range in costs are, therefore, low enough for IH2 to be competitive with
the same range as diesel and gasoline, and less than 0.4 wt% oxygen. petroleum-derived liquid hydrocarbon fuels [13].
About 55% of the energy content of the feedstock is recovered in the The economics of fast hydropyrolysis was also evaluated by
gasoline and diesel fuel products. They also note that the hydrocar- Singh and Mallapragada at Purdue [54] for the H2Bioil process.
bons produced are not soluble in water so the water produced is In their analysis, they assumed that the H2 can originate from
essentially carbon-free. coal, natural gas, or nuclear energy. The liquid yield in the H2Bioil
After the proof-of-concept stage, GTI also reported the opera- process is estimated to range from 83 to 97 gge/t (gasoline gal-
tion of a pilot-scale unit running at 50 kg/day at 24 bar for more lon equivalent). A conventional fast pyrolysis system produces
than 750 h continuously [45]. with very stable operation. The coke 92–110 gge/t, and fermentation produces 50–70 gge/t [3].
deposits on the catalyst very slowly (takes place over months). The The minimum selling price for the liquid product from H2Bioil
fuel properties were similar to the proof-of-concept stage. This test would be $2.83/gal gasoline/diesel blendstock. The factors that are
still did not include the reformer for internal H2 generation, but most important to the cost are the crude oil price, H2 cost, feedstock
the composition of the vapors is consistent with what is considered cost, and capital cost [54]. It needs to be kept in mind that Purdue’s
compatible with commercially available reformers. analysis did not explore economics of scale, which the other stud-
Some observations can be made about this work: (1) the major ies did. The product from the H2Bioil process is competitive with
products of catalytic fast hydropyrolysis with ex-situ hydrotreating cellulosic ethanol, fast pyrolysis liquid fuel, and gasification based
are hydrocarbons in the gasoline and diesel range, which is also con- FT diesel. Table 2 summarizes the works for economic analysis.
firmed by the fuel properties reported. Based on the results, it can The greenhouse gas emissions (GHG) were also evaluated for
be speculated that the composition of the liquid product includes the IH2 process. They conducted a life cycle assessment (LCA) and
hydrocarbons other than aromatics, however, the detailed compo- showed that the process leads to large GHG savings for most feed-
sition of the organic liquid phase was not reported; (2) deactivation stocks (corn stover, sugarcane bagasse, forest residues) compared
is not an an issue for either of the catalysts over 750 h; (3) from a to fossil fuels (91–96%). The only exception is microalgae, mostly
catalysis perspective, the effect of having catalysts in both reac- because of the high amount of energy required for drying. Interest-

Please cite this article in press as: F.L.P. Resende, Recent advances on fast hydropyrolysis of biomass, Catal. Today (2016),
http://dx.doi.org/10.1016/j.cattod.2016.01.004
G Model
CATTOD-9959; No. of Pages 8 ARTICLE IN PRESS
F.L.P. Resende / Catalysis Today xxx (2016) xxx–xxx 7

Table 2
Summary of economic analysis works related to fast hydropyrolysis.

References Process Cost of Plant size (dry Capital Yield (gge/t) H2 source
fuel/gallon ($) MTs of investment
feedstock) (million $)

[55] Fast pyroly- 2.35 2000 349 92–110 Natural gas

sis + hydrotreating/
hydrocraking of oil
[13] IH2 1.93 2000 304 79 Assumed to be generated internally
[54] H2Bioil 2.83 550 123 83–97 Natural gas, coal, nuclear, wind, and solar photovoltaic

ingly, microalgae was the feedstock that also provided the highest these compounds. In order to produce species other than aromatic
hydrocarbon yield in the IH2 process [56]. hydrocarbons, there is clearly a need for a secondary unit (ex-situ
These studies confirm that fast hydropyrolysis-based processes upgrading), typically at a lower temperature than the second unit.
have similar costs to other biofuel production routes. If no addi- The location of the catalyst in the units is also a topic interesting for
tional H2 is required, as proposed in the IH2 process, the scenario future research. For instance, when ex-situ upgrading is performed,
is even more favorable, but this is yet to be demonstrated experi- it remains unclear what is the role of a catalyst in the first reactor.
mentally. In addition, large benefits in terms of GHG emissions are Despite the fact that most research with fast hydropyrolysis of
anticipated. biomass is still recent, major breakthroughs have been reported,
such as the ideas reported by Agrawal and coworkers [2,3,60] on
4. Discussion the use of H2 as source of hydrogen atoms for the liquid fuel, and
the gasoline and diesel range hydrocarbons produced by the IH2
The previous works have demonstrated that fast hydropyroly- process [17]. There is, however, at this point, a serious lack of fun-
sis produces a mixture of hydrocarbon fuels. The yield, composition damental understanding on the fast hydropyrolysis process, and
and properties of this liquid fuel depend greatly on the catalyst(s) many of the works reported do not even include compositional
selected, the use of an HDO unit to upgrade the vapors downstream, analysis of the organic liquid product. The identification of reac-
and even the type of reactor selected for the fast hydropyroly- tion pathways that lead to desirable products, and the synthesis
sis stage. The use of supported metal catalysts in H2 has led to of multi-functional catalysts that promote those pathways is an
an increase in aromatic yields compared to what is obtained by area where substantial amount of research is required. For instance,
catalytic fast pyrolysis. Another exciting accomplishment was the a catalyst that favors decarboxylation over decarbonylation and
operation of a catalytic fast hydropyrolysis reactor by GTI for over minimizes the CO/CO2 ratio in the gas would be useful. This is
750 h without signs of coking/deactivation. However, this process because two atoms of oxygen would be released for each carbon
uses a proprietary catalyst and it is unknown if other catalysts cur- atom released, as opposed to just one. This way, deoxygenation
rently studied would also be able to sustain their performance over would be accomplished minimizing the carbon loss to the gas, and
long periods of time. Nonetheless, the H2Bioil process developed at therefore maximizing the carbon in the organic liquid.
Purdue seems to provide an excellent option to avoid deactivation,
since the fast hydropyrolysis reactor does not contain a catalyst. 5. Conclusion
This way, the carbonaceous residues from pyrolysis and minerals
from biomass remain in the first reactor and do not reach the HDO There has been significant advances on research with fast
unit. But if deactivation is not a problem, then a single catalytic hydropyrolysis of biomass during the last five years. Fast hydropy-
fast hydropyrolysis unit may be sufficient to generate aromatic rolysis produces higher hydrocarbons yields than catalytic fast
hydrocarbons. pyrolysis and appears to have less problems with coking/catalyst
Aromatic hydrocarbons are useful products because they can deactivation due to the presence of H2 . The primary products
be blended with gasoline to increase octane number. Neverthe- reported are aromatic hydrocarbons, but a range of other products
less, blending aromatics to gasoline is a limited option, because the (alkanes and cycloalkanes) can be produced in the presence of a
combustion of aromatics is generally unclean and leads to emis- secondary upgrading unit (ex-situ upgrading). In some cases, the
sion problems. In the US, gasoline is limited to 20 wt% aromatics. organic liquid is reported to contain less than 0.5 wt% oxygen, and
The problem is aggravated by the presence of benzene in the liquid yields as high as 70 wt% have been reported on a carbon basis. There
mixture. Benzene is a known carcinogen identified by the U.S. is debate about the source of the H2 for the process, but studies
Environmental Protection Agency (EPA) as a toxic air pollutant, have shown that the economics of fast hydropyrolysis is competi-
and its allowed concentration in gasoline is limited to only 0.8 wt% tive with those of other biofuel processes. Future research should
[19]. In addition, the insertion of aromatics into oil refinery pro- focus on answering fundamental questions about the process.
cesses such as catalytic cracking is also challenging because of the
high chemical and thermal stability of the aromatic ring [57,58].
For these reasons, it is interesting to operate fast hydropyrolysis References
systems in conditions that favor other hydrocarbon products, such
as olefins, cycloalkanes, and alkanes. Cycloalkanes, for instance, [1] V.K. Venkatakrishnan, W.N. Delgass, F.H. Ribeiro, R. Agrawal, Oxygen removal
from intact biomass to produce liquid fuel range hydrocarbons via
would be an excellent feedstock for insertion into bio-refineries. fast-hydropyrolysis and vapor-phase catalytic hydrodeoxygenation, Green
They can be fed to Fluid Catalytic Cracking (FCC) reactors, pro- Chem. 17 (2015) 178–183.
ducing a high-octane gasoline. Alternatively, cycloalkanes can be [2] R. Agrawal, N.R. Singh, F.H. Ribeiro, W.N. Delgass, Sustainable fuel for the
transportation sector, Proc. Natl. Acad. Sci. U. S. A. 104 (2007) 4828–4833.
additives for aviation fuels such as Jet A. They usually compose [3] N.R. Singh, W.N. Delgass, F.H. Ribeiro, R. Agrawal, Estimation of liquid fuel
20–40 wt% of jet fuels, are thermally stable and their combustion is yields from biomass, Environ. Sci. Technol. 44 (2010) 5298–5305.
clean. These characteristics allow cycloalkanes to be components [4] B. Balagurumurthy, T.S. Oza, T. Bhaskar, D.K. Adhikari, Renewable
hydrocarbons through biomass hydropyrolysis process: challenges and
of high-grade fuels [59]. Some of the works in fast hydropyrolysis opportunities, J. Mater. Cycles Waste Manag. 15 (2012) 9–15.
have reported production of cycloalkanes, but the yield is still low. [5] Energy Independence and Security Act of 2007, n.d. http://www.govtrack.us/
Future research could focus on how to obtain higher selectivity to congress/bills/110/hr6 (accessed 17.07.15).

Please cite this article in press as: F.L.P. Resende, Recent advances on fast hydropyrolysis of biomass, Catal. Today (2016),
http://dx.doi.org/10.1016/j.cattod.2016.01.004
G Model
CATTOD-9959; No. of Pages 8 ARTICLE IN PRESS
8 F.L.P. Resende / Catalysis Today xxx (2016) xxx–xxx

[6] One Hundred Tenth Congress of the United States of America. https://www. [36] A.E. Pwin, H.F. Gerqel, Oil production from an arid-land plant fixed bed
congress.gov/110/bills/s2840/BILLS-110s2840enr.pdf (accessed 08.03.015) pyrolysis and hydropyrolysis, Fuel 75 (1996) 1307–1312.
(2007). [37] R.V. Pindoria, J.-Y. Lim, J.E. Hawkes, M.-J. Lazaro, A.A. Herod, R. Kandiyoti,
[7] D.R. Brodeur-Campbell, M.J. Jensen, J.R. Eatmon, T.J. Sutherland, J.W. Structural characterization of biomass pyrolysis tars/oils from eucalyptus
Shonnard, Political feasibility, environmental sustainability, and economic wood waste: effect of H2 pressure and sample configuration, Fuel 76 (1997)
efficiency analysis of conventional and advanced biofuels technologies, in: 1013–1023.
U.S. Proc. Glob. Conf. Sustain. Prod. Dev. Life Cycle Eng., Busan, Korea, 2008, [38] R. Pindoria, A. Megaritis, A. Herod, R. Kandiyoti, A two-stage fixed-bed reactor
pp. 46–51. for direct hydrotreatment of volatiles from the hydropyrolysis of biomass:
[8] A. Bridgwater, Principles and practice of biomass fast pyrolysis processes for effect of catalyst temperature, pressure and catalyst ageing time on product
liquids, J. Anal. Appl. Pyrolysis 51 (1999) 3–22. characteristics, Fuel 77 (1998) 1715–1726.
[9] A. Bridgwater, Renewable fuels and chemicals by thermal processing of [39] R.V. Pindoria, A. Megaritis, R. Kandiyoti, R.C. Messenbhk, D.R. Dugwell,
biomass, Chem. Eng. J. 91 (2003) 87–102. Comparison of the pyrolysis and gasification of biomass: effect of reacting gas
[10] A.V. Bridgwater, G.V.C. Peacocke, Fast pyrolysis processes for biomass, Renew. atmosphere and pressure on Eucalyptus wood, Fuel 77 (1998) 1247–1251.
Sustain. Energy Rev. 4 (2000) 1–73. [40] R.V. Pindoria, I.N. Chatzakis, J.-Y. Lim, A.a. Herod, D.R. Dugwell, R. Kandiyoti,
[11] S. Czernik, A.V. Bridgwater, Overview of applications of biomass fast pyrolysis Hydropyrolysis of sugar cane bagasse: effect of sample configuration on
oil, Energy Fuels (2004) 590–598. bio-oil yields and structures from two bench-scale reactors, Fuel 78 (1999)
[12] M. Stöcker, Biofuels and biomass-to-liquid fuels in the biorefinery: catalytic 55–63, http://dx.doi.org/10.1016/S0016-2361(98)00102-1.
conversion of lignocellulosic biomass using porous materials, Angew. Chem. [41] J. Dilcio Rocha, C.a. Luengo, C.E. Snape, The scope for generating bio-oils with
Int. Ed. Engl. 47 (2008) 9200–9211. relatively low oxygen contents via hydropyrolysis, Org. Geochem. 30 (1999)
[13] E.C.D. Tan, T.L. Marker, M.J. Roberts, Direct production of gasoline and diesel 1527–1534.
fuels from biomass via integrated hydropyrolysis and hydroconversion [42] O. Onay, A.F. Gaines, O.M. Kockar, M. Adams, T.R. Tyagi, C.E. Snape,
process-A techno-economic analysis, Environ. Prog. Sustain. Energy 33 (2014) Comparison of the generation of oil by the extraction and the hydropyrolysis
609–617. of biomass, Fuel 85 (2006) 382–392.
[14] A. Oasmaa, S. Czernik, Fuel oil quality of biomass pyrolysis oils state of the art [43] R. Galiasso, Y. González, M. Lucena, New inverted cyclone reactor for flash
for the end users, Energy Fuels (1999) 914–921. hydropyrolysis, Catal. Today 220–222 (2014) 186–197.
[15] J.P. Diebold, A Review of the Chemical and Physical Mechanisms of the [44] F. Melligan, M.H.B. Hayes, W. Kwapinski, J.J. Leahy, A study of hydrogen
Storage Stability of Fast Pyrolysis Bio-oils, Golden, CO (2000). pressure during hydropyrolysis of Miscanthus × giganteus and online catalytic
[16] D.C. Elliott, Water, alkali and char in flash pyrolysis oils, Biomass Bioenergy 7 vapour upgrading with Ni on ZSM-5, J. Anal. Appl. Pyrolysis 103 (2013)
(1994) 179–185. 369–377.
[17] T.L. Marker, L.G. Felix, M.B. Linck, M.J. Roberts, Integrated hydropyrolysis and [45] T.L. Marker, L.G. Felix, M.B. Linck, M.J. Roberts, P. Ortiz-toral, J. Wangerow,
hydroconversion (IH2 ) for the direct production of gasoline and diesel fuels or Integrated hydropyrolysis and hydroconversion (IH2 ) for the direct
blending components from biomass, part 1: proof of principle testing, production of gasoline and diesel fuels or blending components from biomass,
Environ. Prog. Sustain. Energy 31 (2012) 191–199. part 2: continuous testing, Environ. Prog. Sustain. Energy 33 (2014) 762–768.
[18] D.C. Elliott, Historical developments in hydroprocessing bio-oils, Energy Fuels [46] V.K. Venkatakrishnan, J.C. Degenstein, A.D. Smeltz, W.N. Delgass, R. Agrawal,
21 (2007) 1792–1815. F.H. Ribeiro, High-pressure fast-pyrolysis, fast-hydropyrolysis and catalytic
[19] G.W. Huber, S. Iborra, A. Corma, Synthesis of transportation fuels from hydrodeoxygenation of cellulose: production of liquid fuel from biomass,
biomass: chemistry, catalysts, and engineering, Chem. Rev. 106 (2006) Green Chem. 16 (2014) 792.
4044–4098. [47] B. Balagurumurthy, T. Bhaskar, Hydropyrolysis of lignocellulosic biomass:
[20] S. Thangalazhy-Gopakumar, S. Adhikari, R.B. Gupta, Catalytic pyrolysis of state of the art review, Biomass Convers. Biorefin. 4 (2013) 67–75.
biomass over H + ZSM-5 under hydrogen pressure, Energy Fuels 26 (2012) [48] M. Linck, L. Felix, T. Marker, M. Roberts, Integrated biomass hydropyrolysis
5300–5306. and hydrotreating: a brief review, Wiley Interdiscip. Rev. Energy Environ. 3
[21] A.V. Bridgwater, Production of high grade fuels and chemicals from catalytic (2014) 575–581.
pyrolysis of biomass, Catal. Today 5861 (1996). [49] D.C. Dayton, J. Carpenter, J. Farmer, B. Turk, R. Gupta, Biomass hydropyrolysis
[22] A. Aho, N. Kumar, K. Eränen, T. Salmi, M. Hupa, D.Y. Murzin, Catalytic pyrolysis in a pressurized fluidized bed reactor, Energy Fuels 27 (2013) 3778–3785.
of woody biomass in a fluidized bed reactor: influence of the zeolite structure, [50] O. Jan, R. Marchand, L.C.a. Anjos, G.V.S. Seufitelli, E. Nikolla, F.L.P. Resende,
Fuel 87 (2008) 2493–2501. Hydropyrolysis of lignin using Pd/HZSM-5, Energy Fuels 29 (2015)
[23] A.A. Lappas, M.C. Samolada, D.K. Iatridis, S.S. Voutetakis, I.A. Vasalos, Biomass 1793–1800, http://dx.doi.org/10.1021/ef502779s.
pyrolysis in a circulating fluid bed reactor for the production of fuels and [51] F. Melligan, M.H.B. Hayes, W. Kwapinski, J.J. Leahy, Hydro-pyrolysis of
chemicals, Fuel 81 (2002) 2087–2095. biomass and online catalytic vapor upgrading with Ni-ZSM-5 and
[24] R. French, S. Czernik, Catalytic pyrolysis of biomass for biofuels production, Ni-MCM-41, Energy Fuels 26 (2012) 6080–6090.
Fuel Process. Technol. 91 (2010) 25–32. [52] S. Meesuk, J.-P. Cao, K. Sato, Y. Ogawa, T. Takarada, Fast pyrolysis of rice husk
[25] C. Mullen, A. Boateng, Catalytic pyrolysis–GC/MS of lignin from several in a fluidized bed: effects of the gas atmosphere and catalyst on bio-oil with a
sources, Fuel Process. Technol. 91 (2010) 1446–1458. relatively low content of oxygen, Energy Fuels 25 (2011) 4113–4121.
[26] S. Thangalazhy-Gopakumar, S. Adhikari, R.B. Gupta, M. Tu, S. Taylor, [53] S. Meesuk, J.-P. Cao, K. Sato, Y. Ogawa, T. Takarada, The effects of temperature
Production of hydrocarbon fuels from biomass using catalytic pyrolysis under on product yields and composition of bio-oils in hydropyrolysis of rice husk
helium and hydrogen environments, Bioresour. Technol. 102 (2011) using nickel-loaded brown coal char catalyst, J. Anal. Appl. Pyrolysis 94 (2012)
6742–6749. 238–245.
[27] F.A. Agblevor, O. Mante, N. Abdoulmoumine, R. McClung, Production of stable [54] N. Singh, D. Mallapragada, Economic analysis of novel synergistic biofuel
biomass pyrolysis oils using fractional catalytic pyrolysis, Energy Fuels 24 (H2Bioil) processes, Biomass Convers. Biorefin. 2 (2012) 141–148.
(2010) 4087–4089, http://dx.doi.org/10.1021/ef1004144. [55] S. Jones, S.B. Valkenburg, C. Walton, C.W. Elliott, D.C. Holladay, J.E. Stevens,
[28] M. Steinberg, The flash hydropyrolysis and methanolysis of coal with D.J. Kinchin, C. Czernick, Production of Gasoline and Diesel from Biomass via
hydrogen and methane, Int. J. Hydrogen Energy 12 (1987) 251–266. Fast Pyrolysis, Hydrotreating and Hydrocracking: A Design Case (2009).
[29] K. Nikkhah, Hydropyrolysis of Various Materials and Coals with Catalysts, [56] E. Maleche, R. Glaser, T. Marker, D. Shonnard, A Preliminary life cycle
University of Saskatchevan, 1992. assessment of biofuels produced by the IH2 TM process, Environ. Prog.
[30] P. Maa, B. Rouge, R.R. Chianelli, C. Sommerville, Catalytic Hydropyrolysis of Sustain. Energy. 33 (2014).
Carbonaceous Material with Char Recycle, 5096569 (1992). [57] R. Sadeghbeigi, Fluid Catalytic Cracking Handbook: Design Operation and
[31] B. Li, S.C. Mitchell, C.E. Snape, Effect of heating rate on normal and catalytic Troubleshooting of FCC Facilites, Gulf Publishing Company, Houston, TX, 2000.
fixed-bed hydropyrolysis of coals, Fuel 75 (1996) 1393–1396. [58] J.W. Wilson, Fluid Catalytic Cracking: Technology and Operation, Pennwell
[32] M. Chareonpanich, T. Boonfueng, J. Limtrakul, Production of aromatic Publishing Company, Tulsa, OK, 1997.
hydrocarbons from Mae-Moh lignite, Fuel Process. Technol. 79 (2002) [59] J.E. Sinor Consultants, Investigation of Byproduct Application to Jet Fuel: Final
171–179. Report, Niwot, CO (2011).
[33] M. Canel, Z. Mısırlıoǧlu, A. Sınaǧ, Hydropyrolysis of a Turkish lignite [60] R. Agrawal, N.R. Singh, Synergistic routes to liquid fuel for a
(Tunıbilek) and effect of temperature and pressure on product distribution, petroleum-deprived future, AIChE J. 55 (2009) 1898–1905.
Energy Convers. Manag. 46 (2005) 2185–2197. [61] S. Meesuk, J.-P. Cao, K. Sato, Y. Ogawa, T. Takarada, Study of catalytic
[34] M. Steinberg, P.S. Division, Flash pyrolysis of biomass with reactive and hydropyrolysis of rice husk under nickel-loaded brown coal char, Energy
non-reactive gas, Biomass 9 (1986) 293–315. Fuels 25 (2011) 5438–5443.
[35] A.J. Guell, C.Z. Li, A.A. Herod, Effect of H2 pressure on the structures of bio-oils
from the mild hydropyrolysis of biomass, Biomass Bioenergy 5 (1993)
155–171.

Please cite this article in press as: F.L.P. Resende, Recent advances on fast hydropyrolysis of biomass, Catal. Today (2016),
http://dx.doi.org/10.1016/j.cattod.2016.01.004