Low temperature aqueous phase hydrogenation of the light oxygenate fraction of bio-oil over supported ruthenium catalysts

Article abstract of DOI:10.1039/c7gc00367f

The low temperature hydrogenation of light oxygenates in the aqueous fraction of bio-oil (AFBO) was studied using both model compounds and a real bio-oil fraction with Ru/TiO₂ and Ru/C catalysts in a continuous flow reactor. The hydrogenation of hydroxyacetone, acetic acid, and formic acid, in single and multi-compound aqueous mixtures, produced 1,2-propanediol, ethanol, and CO₂ respectively, as the main products. The addition of acetic acid increased the rate of hydrogenation of hydroxyacetone by catalyzing the enolization of hydroxyacetone. The main reactions occurring during hydrogenation of the light oxygenates in the aqueous fraction of bio-oil included conversion of hydroxy-ketones into diols and hydrogenation of ketones and aldehydes into mono-alcohols. Differences in the product selectivity between the model carboxylic acids and AFBO were observed. High selectivity to ethanol was found when model-feed mixtures containing acetic acid were hydrogenated. In contrast, the ethanol selectivity for hydrotreating of the AFBO was 13 times lower than the ethanol selectivity for hydrotreating of acetic acid in water. A declining catalyst activity was observed when processing the AFBO.

Full text of DOI:10.1039/c7gc00367f

View Article Online
View Journal
Green
Chemistry
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: H. Bergem, R. Xu,
R. C. Brown and G. Huber, Green Chem., 2017, DOI: 10.1039/C7GC00367F.
This is an Accepted Manuscript, which has been through the
Volume 18 Number
7 7 April 2016 Pages 1821–2242
Royal Society of Chemistry peer review process and has been
accepted for publication.
Green
Chemistry
Accepted Manuscripts are published online shortly after
Cutting-edge research for a greener sustainable future
www.rsc.org/greenchem
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
ISSN 1463-9262
CRITICAL REVIEW
G. Chatel et al.
Heterogeneous catalytic oxidation for lignin valorization into valuable
chemicals: what results? What limitations? What trends?
rsc.li/green-chem

Page 1 of 34
Green Chemistry
View Article Online
DOI: 10.1039/C7GC00367F
Low Temperature Aqueous Phase Hydrogenation of the Light
Oxygenate Fraction of Bioꢀoil over Supported Ruthenium
Catalysts
Håkon Bergem1 , Run Xu2 , Robert C. Brown3 and George W. Huber4
1 Department of Oil and Gas Process Technology, SINTEF Materials & Chemistry, Nꢀ7465,
Trondheim, Norway
2
Hydroprocessing Department, Research Institute of Petroleum Processing, Sinopec, 18
Xueyuan Road, P.O. Box 914ꢀ17, Beijing 100083, P.R. China
3 Department of Mechanical Engineering, Iowa State University, Ames, IA 50011, USA
4
Department of Chemical and Biological Engineering, University of WisconsinꢀMadison,
1415 Engineering Drive, Madison, WI 53706, USA
Abstract
The low temperature hydrogenation of light oxygenates in the aqueous fraction of bioꢀoil
(AFBO) was studied using both model compounds and a real bioꢀoil fraction with Ru/TiO2
and Ru/C catalysts in a continuous flow reactor. Hydrogenation of hydroxyacetone, acetic
acid, and formic acid, in single and multiꢀcompound aqueous mixtures produced 1,2ꢀ
propanediol, ethanol, and CO2 respectively, as the main products. The addition of acetic acid
increased the rate of hydrogenation of hydroxyacetone by catalyzing the enolization of
hydroxyacetone. The main reactions occurring during hydrogenation of the light oxygenates
in the aqueous fraction bioꢀoil included conversion of hydroxyꢀketones into diols and
hydrogenation of ketones and aldehydes into monoꢀalcohols. Differences in the product
selectivity between the model carboxylic acids and APBO were observed. High selectivity to
ethanol was found when modelꢀfeed mixtures containing acetic acid were hydrogenated. In
contrast the ethanol selectivity for hydrotreating of the APBO was 13 times lower than the
ethanol selectivity for hydrotreating of acetic acid in water. A declining catalyst activity was
observed when processing the APBO.
Keywords: bioꢀoil upgrading, light oxygenates, aqueous phase hydrogenation, biochemicals

Green Chemistry
Page 2 of 34
View Article Online
DOI: 10.1039/C7GC00367F
1.0 Introduction
Fast pyrolysis is becoming one of the most promising technologies for harvesting solar energy
stored in biomass by thermochemically transforming it into a high energy density liquid
product [1ꢀ4]. There are numerous advantages of handling liquid fuels in terms of
transportation, storage, transfer, and further chemical processing. The existing infrastructure
for fossil derivedꢀliquid fuels makes the utilization of liquid fuels derived from renewable
resources an obvious choice.
Bioꢀoil from fast pyrolysis of biomass consists of a large number of chemical compounds in
the form of oxygenated organic molecules and water. The oxygenated compounds that are
found in bioꢀoil include pyrolytic lignin (phenolic oligomers), acetic acid, hydroxyacetone,
hydroxyacetaldehyde, anhydrosugars, furans, and several other oxygenated compounds. Bioꢀ
oil polymerizes upon heating and therefore traditional methods like distillation cannot be used
for separation [1]. Bioꢀoil has high oxygen content (up to 50 wt%), high acidity, and phase
separates in storage. The vast number of chemical components in bioꢀoil makes targeted
molecular alterations of the whole bioꢀoil by catalytic hydroprocessing nonꢀselective. The
products include large amounts of coke, tar, and light gases. Catalysts deactivate in the
hydroprocessing of bioꢀoils as reported due to the high reactivity of aldehydes and ketones
[5, 6]. Elliot et al. have reported on the beneficial effect on bioꢀoil stability by initially
hydrogenating these compounds at low temperatures [7]. Stabilized bioꢀoil can subsequently
be processed at higher temperatures and hydrogen pressure [7]. Furthermore, light oxygenates
in the bioꢀoil will produce undesired methane and ethane when hydroprocessing bioꢀoil with
conventional catalysts. Bioꢀoil contains around 25 wt% (dry basis) of light oxygenates
(aldehydes, ketones, alcohols, and carboxylic acids) [8]. It would be highly desirable to
convert all of these into alcohols or polyols.

Page 3 of 34
Green Chemistry
View Article Online
DOI: 10.1039/C7GC00367F
A different approach is to recover the bioꢀoil from fast pyrolysis by selective condensation of
different fractions in distinct stage fractions [9, 10]. These different fractions have distinctive
physical and chemical characteristics. Tailored catalytic processes can then be designed for
conversion of these unique fractions. The need for a combination of separation and
conversion steps as a strategy in bioꢀoil upgrading is highlighted in a recent review by
Resasco and Crossley [11]. The AFBO contains light oxygenates (C1ꢀC6) with functionalities
that can be tuned by hydrogenation into chemicals with desirable properties. Resasco et. al
have also discussed the strategy of eliminating the small acids from bioꢀoil by aqueous phase
ketonization producing larger ketones which subsequently can be hydrogenated into alcohols
[12]. These alcohols can then be used to alkylate with phenolic compounds. The same group
showed that the hydrogenation and alkylation step can be performed in a single reactor with
two sequential catalyst beds [13]. Several investigations have shown that an array of
chemicals including H2, alkanes, and alcohols can be produced by aqueous phase processing
of bioꢀoil [14ꢀ16].
Because of the resemblance between hydroprocessing of petroleum and bioꢀoils, several
studies involving conventional CoMo and NiMoꢀcatalysts have been conducted. Early work
has been comprehensively reviewed [2, 17ꢀ21]. Compared to hydroprocessing of petroleum,
upgrading of bioꢀoils requires lower Liquid Hourly Space Velocity (LHSV), higher pressure,
higher H2ꢀconsumption, and is substantially more exothermic. In addition catalyst stability
for hydrodeoxygenation (HDO) of bioꢀoil is poor compared to hydrotreating of petroleum.
Hydrotreating catalysts typically possess their highest activity for HDO in their sulfide form,
but bioꢀoils do not contain sufficient levels of sulfur to maintain the catalyst structure in the
desired sulfided state during operation. The catalyst support constitutes high surface area
Al2O3 and water combined with high temperature is known to induce sintering and structural
changes forming the hydroxide form (Boehmite) [18]. The presence of acidic sites on the

Green Chemistry
Page 4 of 34
View Article Online
DOI: 10.1039/C7GC00367F
Al2O3ꢀsupport in combination with potential coke precursors in the bioꢀoil also initiates
polymerization/condensation reactions resulting in blocking of active sites by carbonaceous
materials. Development of new catalyst formulations are needed to enable costꢀefficient
upgrading of bioꢀoil feedstocks.
Supported ruthenium catalysts have received increased attention as active and selective
catalysts for the hydrogenation and HDO of oxygenated compounds [22ꢀ26] and bioꢀoil [27,
28]. Ruꢀmetal on carbon or TiO2 gives active catalysts that are hydrothermally stable under
the reaction conditions. An additional advantage of utilizing TiO2 in a commercial process is
that oxidative regeneration of a deactivated catalyst would be possible while this is not
possible in the case of carbon.
The objective of this paper is to study the low temperature aqueous phase hydrogenation
(APH) of a staged bioꢀoil fraction containing light oxygenates over supported ruthenium
catalysts to gain insight into the molecular level chemical reactions taking place. In addition,
we will identify the effects of reaction conditions (temperature and contact time) on the
product distribution. Experiments with singleꢀ and multiꢀcomponent modelꢀfeeds in a
continuous flow reactor have been performed to identify the reaction pathways for some of
the main compounds in the real bioꢀoil and the mutual effect they impose on each other`s
reactivity. Acetic acid (AA), hydroxyacetone (HA), and formic acid (FA) were chosen as
representative of the light oxygenates found in the aqueous fraction of bioꢀoil collected
through staged recovery of pyrolysis vapors as described by Pollard et al. [9] and Rover et al.
[10]. Longꢀterm catalyst stability when upgrading bioꢀoil is of particular importance because
of carbon formation, poisoning from feed impurities, and leaching of active metal from the
catalyst. This paper also provides increased knowledge on the production of value added
chemicals from bioꢀoil feedstocks.

Page 5 of 34
Green Chemistry
View Article Online
DOI: 10.1039/C7GC00367F
2. Experimental
2.1 Catalyst preparation and characterization
The 3 wt% Ru/TiO2 catalyst was prepared by the incipient wetness method with ruthenium
(III) nitrosyl nitrate (SigmaꢀAldrich catalog number 373567, 1.5 wt% Ru in dilute nitric acid
solution) as the Ru precursor. The titania was supplied by SigmaꢀAldrich (catalog number
718467). The support was calcined at 400 °C for 4 h in a muffle furnace (heating rate 4 °C/h)
prior to use. In the catalyst synthesis, a volume of the precursor containing Ru equivalent to 3
wt% of the weighted titania was added dropꢀwise to the support previously dried at 110 °C for
12 h until incipient wetness was observed. Multiple impregnation steps were needed with
drying of the sample at 110 °C for 12 h after each step. The 5 wt% Ru/C was a commercial
available catalyst (Strem Chemicals) and was used as received.
The BrunnerꢀEmmettꢀTeller (BET) surface area of the catalysts was determined from N2
physisorption at ꢀ196 °C using a Micromeritics ASAP 2020 system. Before the measurements,
the sample was degassed under vacuum at 523 K for 12 h. The pore volume and average pore
diameter for each catalyst were calculated by the Barrett–Joyner–Halenda (BJH) desorption
method.
H2 uptake of the catalysts were carried out at 50°C using a Micromeritics® AutoChem II 2920
system; each sample (200 mg) was reduced in situ at 450°C for 2 h at a heating rating of 1 °C
minꢀ1 in flowing hydrogen (50 mL minꢀ1) prior to chemisorption measurements. After the
reduction was complete, the sample was degassed at 450°C for 2 h in He. The sample cell was
cooled to 50°C. The peak area can be correlated with the amount of adsorbed H2 on the basis
of pulsed H2 experiments. Surface average metal particle sizes were calculated using the
equation
D = 6/Sρ,

Green Chemistry
Page 6 of 34
View Article Online
DOI: 10.1039/C7GC00367F
where S is the metal surface area per gram of metal, and ρ the density of the metal. A crossꢀ
sectional area of 9.03 Å2 per Ru surface atom was assumed for calculating S [29].
Temperatureꢀprogrammedꢀreduction (TPR) of the Ru/TiO2 catalyst was performed in a
Micromeritics Autochem II 2920 instrument.
2.2 Preparation and characterization of model compounds and aqueous fraction of bio-
oil
Model compounds used in this study included acetic acid (AA; purity > 99.7%), formic acid
(FA; purity > 98%), and hydroxyacetone (HA; purity > 90%) supplied by SigmaꢀAldrich and
used without further purification.
Bioꢀoil was produced by pyrolyzing red oak at 500 °C in an 8 kg/h pyrolysis development
unit (PDU) at Iowa State University's BioCentury Research Farm. This system has a bioꢀoil
recovery system that separates bioꢀoil into distinctive stage fractions, as previously described
by Rover et al. [9]. Heavy ends consisting of ligninꢀderived phenolic oligomers and
carbohydrateꢀderived anhydrosugars were removed from the pyrolysis vapor stream by a
condenser and electrostatic precipitator with coolant set to 85 °C; a middle fraction consisting
of mostly furanics and phenolic monomers was removed by a second condenser and
electrostatic precipitator with coolant set at 65 °C; the AFBO employed in the present study,
consisting of waterꢀsoluble light oxygenates, was removed by a condenser operated at 18 °C.
The aqueous fraction was stored in a refrigerator and filtered before use.
Inductively coupled plasma (ICP) analysis of the liquid products was performed on a Plasma
400 ICP (Perkin Elmer).
2.3 Aqueous phase hydrogenation

Page 7 of 34
Green Chemistry
View Article Online
DOI: 10.1039/C7GC00367F
Aqueousꢀphase hydrogenation was performed in a continuous flow reactor operated in upꢀ
flow mode. The experimental setup was placed in a ventilated hood with a safety screen in
front. All pressurized lines were equipped with safety relief valves and hydrogen was feed
from a gas cylinder completed with an automatic shutꢀoff valve (Hoke). The reactor consisted
of a ¼ inch stainless steel tube 580 mm long with 3 mm inner diameter placed inside a
furnace with an aluminum block/filler inserted in the void between the furnace and the tubular
reactor. The aluminum filler was used to obtain a uniform temperature profile along the
catalyst bed by improved heat transfer. In the experiments 2.0 g of catalyst was usually loaded
into the reactor and positioned with glass wool above and below the catalyst bed. In the case
of the extraꢀlow temperature experiments with HA, 0.5 g of catalyst was used. Reduction of
the catalyst was performed by feeding hydrogen at a flow rate of 100 mL minꢀ1 while heating
the reactor to 300 °C at a rate of 30 °C/h. The catalyst was soaked at 300 °C for 2 h before the
temperature was decreased to room temperature before the pressure was increased to the
operating pressure. The liquid was fed to the reactor by a Varian ProStar 210 HPLC pump and
mixed with hydrogen before the reactor inlet. Hydrogen flow rate was 40 mL minꢀ1 metered
by a Brooks mass flow controller. The product mixture passed through a gasꢀliquid separator
before the gas flowed through a backꢀpressure valve (Swagelok) maintaining the system
pressure. Gaseous products were vented or periodically directed to an online gas
chromatograph (Shimadzu GCꢀ2014) by a threeꢀway valve for composition analysis.
Compound separation was obtained with two columns using helium as carrier gas.
Hydrocarbons were separated on a RTꢀQꢀBond capillary column and analyzed with a flame
ionization detector (FID) while permanent gases were separated on a ShinCarbon ST packed
column and analyzed with a thermal conductivity detector (TCD).
Liquid products were sampled from the gasꢀliquid separator once every hour and analyzed
offline by GCꢀFID, HPLC, GCꢀMS, and a Total Organic Carbon (TOC) analyzer. For the

Green Chemistry
Page 8 of 34
View Article Online
DOI: 10.1039/C7GC00367F
experiments with model feeds at least 2 samples were collected at each set of operating
conditions to ensure the steady state. The GCꢀFID was a Shimadzu GCꢀ2010 with an Rtxꢀ
VMS capillary column utilizing helium as carrier gas. HPLC analysis was performed with a
Shimadzu LCꢀ20AT equipped with a Biorad Aminex HPXꢀ87H column maintained at 30 °C
and applying 0.005 M H2SO4 as the mobile phase pumped at 0.6 mL minꢀ1. Component
detection was obtained with a UVꢀVIS detector (SPDꢀM20A) and a RI detector (RIDꢀ10A).
The GCꢀMSꢀQP2010S (Shimadzu) was equipped with an RtxꢀVMS capillary column using
helium as carrier gas and a QPꢀ2010. TOC was quantified using a Shimadzu TOCꢀVCPH
analyzer.
3. Results and discussion
3.1 Characterization of catalysts
Physical properties of the prepared Ru catalysts are summarized in Table 1. The two support
materials have distinctly different properties with carbon being a high surface area support
(834 m2/g) with a smaller average pore size (38.4 Å) compared to TiO2 (57 m2/g and 214.7 Å
respectively). Hydrogen uptake on fresh Ru/C catalyst corresponds to a Ru dispersion of
7.8% or an average Ru particle size of 11.7 nm. Metal dispersion and average particle size on
fresh Ru/TiO2 catalyst were calculated to be 65.1% and 1.4 nm respectively. TPR of fresh
Ru/TiO2 catalyst showed one large peak with a maximum at 120 °C and a smaller shoulder
peak with a maximum at 150 °C. No additional H2 uptake was measured above 200 °C. From
the TPR measurements it was concluded that reduction of the catalyst with H2 at 300 °C
before reaction was sufficient to obtain reduced Ru metal.
3.2 Hydrogenation of single component feed
o
Hydrogenation of AA in water was done at 120 C over Ru/TiO2 as shown in Table 2. At
these conditions, 37.5 % of the AA was converted and two liquid products were observed:
ethanol (78.3% selectivity) and ethyl acetate (1.4 % selectivity). Other products were

Page 9 of 34
Green Chemistry
View Article Online
DOI: 10.1039/C7GC00367F
methane, ethane, and CO2 with traces of C3ꢀC5 hydrocarbons. Several investigations have
found Ru to be the most active and selective metal for converting acetic acid into ethanol [24,
30, 31].
Hydrogenation of HA was faster than hydrogenation of AA. Hydrogenation of HA occurred
at temperatures as low as 20oC, as shown in Figure 1. HA conversion increased from 11.7 %
o
to 93.6 % as temperature increased from 20 to 70 C. Under these conditions HA was
selectively hydrogenated into 1,2ꢀpropanediol (PGꢀpropylene glycol) with trace amounts of
methane observed in the gasꢀphase. The PG selectivity was always above 98% under all
conditions investigated.
3.3 Hydrogenation of two-component feed
Hydrogenation of twoꢀcomponent feeds (AA+HA, AA+FA) in water with Ru/TiO2 catalyst
were performed with product distributions as shown in Table 2 and Figures 1ꢀ2. Coꢀfeeding
AA+HA at 120 °C resulted in an acetic acid conversion of 45.5 % while HA was completely
converted at these conditions. A small amount of propanol was detected among the products
(below 1%), which was not observed when the compounds were feed separately. Propanol is
probably formed as a secondary product by hydrodeoxygenation of PG. Complete conversion
of the FA occurred when aqueous solutions of FA + AA were coꢀfed. CO2 was the major
product observed from FA with only small amounts of methanol formed. This suggests that
the FA decomposes to CO2 and H2 rather than hydrogenates as the main reaction pathway.
Hydrogenation of HA with and without AA present in the feed is shown in Figure 1. Coꢀ
feeding 10 wt% AA increased the reactivity of HA. The effect of feeding AA or FA on the
HA conversion is detailed more in Figure 2 which plots the HA conversion as a function of
time on stream. PG was the only observed product for these expeirments. Neither of the two
acids showed any reactivity at these reaction conditions. Increasing the AA concentration

Green Chemistry
Page 10 of 34
View Article Online
DOI: 10.1039/C7GC00367F
increased the rate of hydrogenation of HA. Coꢀfeeding 2 wt% FA together with HA
completely shut off the HA conversion. Subsequently coꢀfeeding 20 wt% AA only partly
restored the HA conversion activity of the Ruꢀcatalyst before a decreasing activity with time
on stream (TOS) was observed.
In elucidating the enhancing effect of AA on the HA reactivity a reaction mechanism
involving acid catalyzed enolization of HA into its enolꢀ and enediolꢀform is proposed as
shown in Scheme 1. Enolization is a proton transfer reaction catalyzed by acids where the rate
is dependent on the acid strength (pK) and concentration [32]. Yaylayan et al. have studied
the enolization reactions of HA in different solvents with FTIR spectroscopy [33]. They found
no evidence of enolization of HA in D2O, while under acidic conditions the enol and
predominantly enediol tautomers were observed. The concentration of enolized species was
found to increase with temperature. In addition to the route involving direction hydrogenation
of the C=O bond in HA (Path 1) another parallel route opens when acid species are introduced
in the feed (Path 2). In Path 2 the enediol is formed as an intermediate before rapid
hydrogenation of the C=C bond takes place forming PG. The observation that the total
conversion of HA increases with AA concentration implies that formation of enediol is the
rateꢀdetermining step along path 2. FA should also catalyze the enolization reaction [32, 33].
The reason for the observed negative effect is believed to be a strong deactivation of the
active Ru sites caused by adsorbates originating from FA. CO formation from FA
decomposition has been shown experimentally to take place on Ru [34]. If CO is formed then
Ru particle size becomes an important factor. FischerꢀTropsch Synthesis was found to be a
highly structureꢀsensitive reaction on Ru clusters smaller than 10 nm [35]. Lower intrinsic
activity, measured as a decreased turnover frequency of CO consumption, was found on Ru
clusters less than 10 nm. This observation was related to stronger CO adsorption and
concomitant blocking of active sites. The small metal clusters on the Ru/TiO2 (Table 1) are

Page 11 of 34
Green Chemistry
View Article Online
DOI: 10.1039/C7GC00367F
thus susceptible to CO poisoning at these lowꢀtemperature conditions. After removing FA
from the feed a full recovery of the catalytic activity is not observed suggesting a continuing
partial CO poisoning of the active sites.
3.4 Hydrogenation of three-component feed
The effect of temperature on the hydrogenation of a threeꢀcomponent feed, consisting of AA,
HA and FA, is shown in Table 2 and Figure 3. The conversion of AA increased from 3 % to
60 % when the temperature was increased from 100 to 150 °C. The apparent activation energy
(Eapp) for AA conversion was 46.2 kJ/mol. This is comparable to the activation energy of 40.0
kJ/mol reported by Olcay et al. for the hydrogenation of AA on Ru/C in a continuous packedꢀ
bed reactor [31]. HA and FA showed similar conversion at 100 °C (64 % and 66 %
respectively), while both were completely converted at 120 °C. The selectivity into liquid
products was 96 % at 100 °C, decreasing to 67 % at 150 °C with a corresponding increase in
the gas products. The dominant liquid product at low temperature (100 ° C) was PG (95 %).
Minor liquid products (< 1 %) at 100 °C were ethanol, methanol, propanol, and ethyl acetate.
When the reactor temperature increased to 150 °C PG liquid selectivity decreased to 32 %.
Ethanol selectivity initially increased to 20 % at 120 °C before decreasing to 18 % at 150 °C.
In the same temperature interval propanol selectivity increased from 0.1 % to 1.8 %, while
methanol selectivity decreased from 0.5 % to below 0.1 %. Ethyl acetate selectivity increased
to 0.4 % at 120 °C but did not increase further as temperature increased to 150 °C. Methane,
CO2, ethane, and propane were the main gasꢀproducts formed with small amounts of C4ꢀC6
also observed. The selectivity to gasꢀphase products increased with temperature from 0.3 % at
100 °C to 38.2% at 150 °C. A reactor temperature of 150 °C resulted in pressure buildup and
visual evidence of carbonaceous material on the catalyst and reactor internals. At

Green Chemistry
Page 12 of 34
View Article Online
DOI: 10.1039/C7GC00367F
temperatures below 120 °C catalyst activity was stable for more than 50 hours TOS with no
signs of coke formation when pumping the multiꢀcomponent feed.
3.5 Hydrogenation of AFBO
The APBO was hydrogenated with Ru/C at 120 °C and 5.18 MPa pressure as shown in Table
S1. Hydroxymethylfurfural (HMF), furfural, acetaldehyde, and phenol were completely
converted at the shortest TOS (6 hours). Products from these reactants were HMF alcohol,
furfuryl alcohol, ethanol, and cyclohexanol, respectively.
High conversions of
hydroxyacetaldehyde, FA, propionic acid, butyric acid, and HA were also obtained after 6
hours TOS while only partial conversion of AA took place. Products after hydrogenation were
monoꢀalcohols: methanol > ethanol > hexanol > propanol > butanol, and diols: ethylene
glycol > PG > 1,4ꢀbutanediol > 1,4ꢀpentanediol > 2,3ꢀbutanediol > 1,5ꢀpentanediol > 1,3ꢀ
butanediol. Main gas phase products were methane and CO2 with small amounts of higher
hydrocarbons (C2ꢀC6). The composition of the products decreased and the reactants increased
after 25 hrs of TOS compared to the 6 h TOS point. The most significant changed was
observed for AA, propionic acid, butyric acid, and HA. Hydrogenation of FA,
hydroxyaldehyde, HMF, furfural, and phenol were affected the least.
Hydrogenation of AFBO was studied more in depth using Ru/TiO2 catalyst in the temperature
range of 100 °C to 140 °C at 6.21 MPa pressure, as shown in Table 3. Experiments were run
for at least 8 hours at each set of operating conditions before liquid products were collected
for analysis. GCꢀMS analysis of the liquid samples made quantification of a wider range of
the compounds possible compared to the experiments with Ru/C. Feed compounds detected
by GCꢀMS but not quantified included phenol, 2ꢀmethoxyphenol, 2,3ꢀbutanedione, 3ꢀmethylꢀ
1,2ꢀcyclopentadione, and 3ꢀmethylꢀ2ꢀfuranone. Several compounds found in the feed were
completely converted even at 100 °C. These included acetone, acetaldehyde,

Page 13 of 34
Green Chemistry
View Article Online
DOI: 10.1039/C7GC00367F
propionaldehyde, 2ꢀpropenꢀ1ꢀol, 1ꢀhydroxyꢀ2ꢀbutanone, 3ꢀhydroxyꢀ2ꢀbutanone, 2ꢀhexanone,
and 2ꢀfuranone. Feed compounds including HMF, furfural, FA, hydroxyacetaldehyde, and
HA were less reactive, requiring a reaction temperature of 140 °C to reach near complete
conversion. The least reactive compounds in the bioꢀoil were organic acids including AA and
propionic acid. The amount of butyric acid in the product decreased by 65 % relative to the
feed at 140 °C, while the amount of AA decreased by 30 % with increasing temperature up to
120 °C. The AA amount in the product then increased again when the temperature was set to
140 °C. The refractory behavior of AA under LTH of bio oil with Ru/C has been documented
by Elliot and Oasmaa [36, 37]. They reported only limited conversion at 150 °C after 4 h in a
batch reactor with ethanol as the main product. In their work higher temperatures resulted in
decreasing ethanol yield and an increase in methane production. Propionic acid content
decreased initially at a reactor temperature of 100 °C, but increased at higher temperatures. 2ꢀ
butanone found in the feed was stable up to 120 °C before being partially converted when the
reactor temperature increased to 140 °C. Methanol and methyl acetate were present in the bioꢀ
oil feed but, unlike the other feed compounds, their concentration increased in the
hydrogenated product.
Table 3 also lists the product distribution. The most abundant liquid products were methanol
> PG >ethylene glycol > 1,4ꢀbutanediol > 1,4ꢀpentanediol > 1,2ꢀbutanediol and 2,3ꢀ
butanediol. Other products formed in minor amounts were ethanol, nꢀpropanol, nꢀbutanols,
cyclopentanol, 1,2ꢀbutanol, butyrolactone, and furfuryl alcohol, and 1ꢀhexanol. A small
amount of ethyl acetate was also observed. A pronounced effect of reaction temperature on
the product distribution was observed. Methanol concentration was highest at 100 °C before
decreasing at higher temperatures. The same behavior was observed with 1,2ꢀethanediol and
1,4ꢀbutanediol (above 120 °C). Sanna et al. reported similar shifts in product selectivity for

Green Chemistry
Page 14 of 34
View Article Online
DOI: 10.1039/C7GC00367F
HDO of the water soluble fraction of bioꢀoil over Ru/C catalyst [38]. Sanna et al. observed
that the HA concentration decreases and the AA, furufral and phenolic concentrations did not
change when no catalyst was present in the reactor. Selectivity to gasꢀphase products was
below 10% in the investigated temperature range with the most abundant products being
methane and CO2.
An interesting question is how close the product yields of these monoꢀalcohols and diols are
to the potential product yields that can be obtained for a given feed composition. Table 4
summarize the theoretical potential yields of different alcohols assuming complete and
selective conversion of the listed feed molecules into conceivable products. A product yield
exceeding the theoretical yield indicates that the product molecule is formed in parallel
reactions originating from additional feed compounds or intermediate products. Ethanol was
only formed in yields lower than 3% of the theoretical yield throughout the entire investigated
reaction temperature window. The listed butanediols and 1,4ꢀpentanediol were produced in
high yields close to and above the theoretical values at the operating conditions. The yield of
PG increased from 6% of the theoretical yield at 100 °C to 90% at 140 °C. In contrast to other
diols, ethylene glycol yield decreased with temperature, going from 36% at 100 °C to 5% at
140 °C.
In summary, the main reactions occurring during hydrogenation of light oxygenates in AFBO
included conversion of hydroxyꢀketones into diols and hydrogenation of ketones and
aldehydes into monoꢀalcohols. HDO of carboxylic acids to produce monoꢀalcohols proceeded
with low selectivity. The resulting decrease in carbonyl content after hydrogenation improved
stability of the bioꢀoil, which might be a useful pretreatment ahead of further upgrading.
3.6 Comparison of model-feed to aqueous fraction bio-oil processing

Page 15 of 34
Green Chemistry
View Article Online
DOI: 10.1039/C7GC00367F
The reactivity of model compounds is compared to that of AFBO in Table 5. AA was the least
reactive compound with FA being the most reactive one. Complete conversions of FA and
HA were obtained at most of the reaction conditions when studying twoꢀ and threeꢀ
component model feeds. The complex composition of the AFBO also induces differences in
the observed product selectivity. There was a striking difference in the reactions for AA
between the model and real feeds. Ethanol was the dominant product from AA with
selectivity above 60% in the model feeds. In the AFBO the ethanol selectivity from AA
hydrogenation was 6% and 26% on Ru/TiO2 and Ru/C respectively. This is based on the
assumption that the produced ethanol originates from AA hydrogenation. The reaction
network for aqueous phase hydrogenation of AA on Ru proposed by Wan [24] and Olcay [30]
involves formation of acetyl species followed by ethanol formation via acetaldehyde when
hydrogen is present. Our experimental data indicates that this reaction path is suppressed in
AFBO and that an equally important source of ethanol is the direct hydrogenation of
acetaldehyde originally present in the feed. The question then is what other chemical reactions
are contributing in the transformation of AA. Esterification of AA with methanol forming
methyl acetate seems to be responsible for the increased amount of this compound in the
product. Formation of methyl acetate during bioꢀoil upgrading under hydrogen pressure and
with a Ru/Al2O3 catalyst was also reported by Ying et al. [39]. As stated earlier esterification
of AA with ethanol producing ethyl acetate only take place to a limited degree.
Methanol formation was catalyzed by both Ru/TiO2 and Ru/C during the hydrogenation of
APBO. Scheme 2 shows reaction pathways suggesting how methanol can be formed from
hydroxyacetaldehyde over the Ru catalysts. Hydroxyacetaldehyde is first hydrogenated into
ethylene glycol (reaction pathway I) followed by cleavage of the CꢀC bond and formation of
methanol as a secondary product. Direct cleavage of the CꢀC bond in hydroxyacetaldedyde
with formation of methanol and formaldehyde products (reaction pathway II) is also

Green Chemistry
Page 16 of 34
View Article Online
DOI: 10.1039/C7GC00367F
suggested but no formaldehyde was detected among the products. Several investigations of
aqueous phase hydrogenation of hydroxyacetaldehyde have found ethylene glycol as the sole
product [14, 40, 41]. The less than stoichiometric amount of ethylene glycol produced
compared to the amount of converted hydroxyacetaldehyde (Table S1 and 3) found in the
present study shows that ethylene glycol reacts further at the employed reaction conditions.
These earlier studies were conducted at lower temperatures (75ꢀ90°C). High CꢀC bond
cleavage activity of Ru has been reported for ethylene glycol aqueous phase reforming
conducted at higher temperatures [15]. Chemical composition analysis of the studied AFBO
showed the presence of levoglucosan as a feed molecule [9]. Levoglucosan can be converted
into sorbitol which could then undergo further hydrogenolysis into C2ꢀC4 diols and monoꢀ
alcohols [38].
3.7 Catalyst Stability
The AA, FA and HA conversion decreased with time when processing AFBO, as shown in
Figure 4. From 21 to 106 hours TOS the conversion of AA decreased from 30% to 1% (97%
reduction). FA conversion decreased from 80% to 27% (66% reduction), and HA conversion
from 40% to 6% (85% reduction). Characterization of the catalysts after processing APBO
were performed (Table 1). Chemisorption measurements on the catalysts after reaction
resulted in decreased H2 uptake for both catalysts. The H2 uptake on the used Ru/C catalyst
was reduced by 89% while the corresponding reduction on the Ru/TiO2 catalyst was 63%.
The BET surface area was reduced by 81% and 5% on the used Ru/C and Ru/TiO2,
respectively. The effect of reaction conditions on catalyst pore volume was much more
pronounced for Ru/C compared to Ru/TiO2 with a reduction of 59% compared to 5%. The
high surface area carbon support contains pores of considerably smaller size compared to the
TiO2 support. The observed increase in average pore size after reaction is a consequence of
catalyst fouling due to carbonaceous deposits.

Page 17 of 34
Green Chemistry
View Article Online
DOI: 10.1039/C7GC00367F
There was a notable effect of temperature on Ru/TiO2 catalyst stability. When processing
modelꢀfeeds, gasꢀmake increased sharply with temperature. Pressure drop increased across the
reactor when running the reactions at 150 °C, Within a few hours of operation the reactor
plugged due to coke formation. The observed decline in catalyst activity at lower
temperatures is probably not attributed to coke buildup on the catalyst surface alone. A
contributing factor seems to be acid leaching of ruthenium metal from the catalyst caused by
the low pH of the feeds. ICP measurements of the liquid products from the multiꢀcomponent
feed (AA+HA+FA) for the Ru/TiO2 catalyst detected Ru with highest concentration (110
ppm) at short TOS (24 h) decreasing to 30 ppm after 68 h TOS. Based on an average leaching
rate, this amounts to roughly 25 % of the Ru being leached out of solution during the
experiment. This substantial loss of Ru is expected to influence catalyst performance during
90 hours TOS.
Biomass contains trace amounts of elements like potassium, phosphorus, chlorine, and sulfur,
which will appear in the products of fast pyrolysis [1]. Analysis of woodꢀderived char and
bioꢀoil shows that most of the ashꢀforming elements are retained in the char (above 90%)
while much of the sulfur and chlorine can be found in the bioꢀoil [42, 43, 44]. Based on
elemental analysis of the AFBO used in our study [9] a concentration of approximately 68
ppm in the feed was previously measured. Sulfur species adsorb strongly on most metals
creating stable surface sulfides [45]. This modifies the adsorption characteristics of other
molecules by blocking chemisorption sites. Lowering of the H2ꢀadsorption capacity and
decreased hydrogen binding energy take place on presulfided metal surfaces. Catalyst
deactivation by sulfur poisoning was found to be significant during low temperature
hydrogenation of bioꢀoil over a Ru/TiO2 catalyst [46]. The bioꢀoil contained 68 ppm sulfur as
reported elsewhere [9]. Lowered hydrogenation activity resulted in increased catalyst fouling
by carbonaceous species. Reducing the sulfur content in the bioꢀoil feed to 39 ppm resulted in

Green Chemistry
Page 18 of 34
View Article Online
DOI: 10.1039/C7GC00367F
higher hydrogenꢀtoꢀcarbon ratio and lower carbonyl concentration in the hydrogenated
product. Our results supports the role of sulfur acting as a poison affecting the hydrogenation
function of the catalysts. Decreasing conversion with TOS and significant reduction in H2
uptake on the used catalysts from blocking of active Ru sites by sulfur and carbonaceous
species are observed. The latter being most detrimental on the carbonꢀsupport due to its
smaller pores. Declining catalyst activity when processing the aqueous phase bioꢀoil could
also be due in part to CO poisoning of the catalyst surface from formic acid decomposition.
The observed shift in Table 3 from CH4 and CO2 being dominant to CO indicates declines in
waterꢀgasꢀshift (WGS) and methanization reactions with time.
4. Conclusions
We have performed experiments on hydrogenation over Ru catalysts of singleꢀ and multiꢀ
component model compounds and the aqueous fraction of bioꢀoil collected through staged
recovery of pyrolysis vapors. The order of increasing rate of hydrogenation is the same
whether processing singleꢀ or multiꢀcomponent model compounds or APBO with the rate
decreasing as: formic acid> hydroxyacetone > acetic acid. When present in the model feeds
formic acid undergoes dehydrogenation as the main reaction route forming CO2 and H2. The
product selectivity to methanol is below 1 %. Hydrogenation of hydroxyacetone gives
propylene glycol as the dominant product. Hydrogenation of acetic acid produces ethanol at
high selectivity with ethyl acetate as a minor product.
Coꢀfeeding acetic acid and hydroxyacetone increase the hydrogenation rate of
hydroxyacetone. A mechanism involving the acidꢀcatalyzed enolization of hydroxyacetone
into its enolꢀform (propꢀ2ꢀeneꢀ1,2ꢀdiol) is proposed. The observation that the total conversion
of hydroxyacetone increases with acetic acid concentration implies that formation of the

Page 19 of 34
Green Chemistry
View Article Online
DOI: 10.1039/C7GC00367F
intermediate enol is the rateꢀdetermining step. Coꢀfeeding 2 wt% formic acid together with
hydroxyacetone completely shuts off hydroxyacetone conversion.
The main reactions occurring during hydrogenation of the light oxygenates in AFBO include
conversion of hydroxyꢀketones into diols and hydrogenation of ketones and aldehydes into
monoꢀalcohols. HDO of carboxylic acids into corresponding monoꢀalcohols proceeds with
low selectivity. The resulting decrease in carbonyl content after hydrogenation contribute
towards stabilizing the bioꢀoil if performed as an intermediate step before further processing
either alone or in combined streams in an integrated chemical facility.
This study shows differences in product selectivity for AA when present as singleꢀ and multiꢀ
compound in water or in AFBO. High selectivity to ethanol is found when modelꢀfeed
mixtures containing AA are hydrogenated. This is in contrast to APBO hydroprocessing
where the ethanol selectivity (at comparable AA conversion) is low.
Catalyst activity and stability were influenced by several factors. Higher temperature (~150
°C) resulted in catalyst fouling from carbon deposition. Reduction in BET surface area and
pore volume with reaction time was found to be more pronounced for Ru/C catalyst compared
to Ru/TiO2. Loss of Ru from the catalyst as a result of acid leaching was observed. Ru metal
content on the catalyst was observed to decrease by about 25 % after 90 hours on stream. An
increased rate of catalyst deactivation was observed during hydrogenation of AFBO compared
to modelꢀfeed mixtures. Catalyst poisoning by sulfur present in the AFBO is suggested to be
an important factor.

Green Chemistry
Page 20 of 34
View Article Online
DOI: 10.1039/C7GC00367F
Acknowledgment
We appreciate the support of Lysle Whitmer, Marge Rover, and Ryan Smith at Iowa State
University who prepared and analyzed the bioꢀoil samples used in these experiments.
1.
2.
3.
4.
5.
6.
Huber, G.W., S. Iborra, and A. Corma, Synthesis of transportation fuels from biomass:
Chemistry, catalysts, and engineering. Chemical Reviews, 2006. 106(9): p. 4044-4098.
Bridgwater, A.V., Review of fast pyrolysis of biomass and product upgrading. Biomass and
Bioenergy, 2012. 38(0): p. 68-94.
Mohan, D., C.U. Pittman Jr, and P.H. Steele, Pyrolysis of wood/biomass for bio-oil: A critical
review. Energy and Fuels, 2006. 20(3): p. 848-889.
Czernik, S. and A.V. Bridgwater, Overview of applications of biomass fast pyrolysis oil. Energy
and Fuels, 2004. 18(2): p. 590-598.
Oasmaa, A. and E. Kuoppala, Fast pyrolysis of forestry residue. 3. Storage stability of liquid
fuel. Energy and Fuels, 2003. 17(4): p. 1075-1084.
Chen, W., et al., Low temperature hydrogenation of pyrolytic lignin over Ru/TiO2: 2D HSQC
and 13C NMR study of reactants and products. Green Chemistry, 2016. 18(1): p. 271-281.
Elliott, D.C. and E.G. Baker, Process for upgrading biomass pyrolyzates. 1989: USA.
Brown, R.C. and T.R. Brown, Thermochemical Processing of Lignocellulosic Biomass, in
Biorenewable Resources. 2014, John Wiley & Sons, Inc. p. 195-236.
7.
8.
9.
Pollard, A.S., M.R. Rover, and R.C. Brown, Characterization of bio-oil recovered as stage
fractions with unique chemical and physical properties. Journal of Analytical and Applied
Pyrolysis, 2012. 93(0): p. 129-138.
10.
11.
Rover, M.R., et al., The effect of pyrolysis temperature on recovery of bio-oil as distinctive
stage fractions. Journal of Analytical and Applied Pyrolysis, 2014. 105(0): p. 262-268.
Resasco, D.E. and S.P. Crossley, Implementation of concepts derived from model compound
studies in the separation and conversion of bio-oil to fuel. Catalysis Today, 2015. 257(P2): p.
185-199.
12.
13.
14.
15.
Pham, T.N., D. Shi, and D.E. Resasco, Evaluating strategies for catalytic upgrading of pyrolysis
oil in liquid phase. Applied Catalysis B: Environmental, 2014. 145: p. 10-23.
Nie, L. and D.E. Resasco, Improving carbon retention in biomass conversion by alkylation of
phenolics with small oxygenates. Applied Catalysis A: General, 2012. 447-448: p. 14-21.
Vispute, T.P. and G.W. Huber, Production of hydrogen, alkanes and polyols by aqueous phase
processing of wood-derived pyrolysis oils. Green Chemistry, 2009. 11(9): p. 1433-1445.
Davda, R.R., et al., A review of catalytic issues and process conditions for renewable hydrogen
and alkanes by aqueous-phase reforming of oxygenated hydrocarbons over supported metal
catalysts. Applied Catalysis B: Environmental, 2005. 56(1-2 SPEC. ISS.): p. 171-186.
Bond, J.Q., et al., Production of renewable jet fuel range alkanes and commodity chemicals
from integrated catalytic processing of biomass. Energy & Environmental Science, 2014. 7(4):
p. 1500-1523.
16.
17.
18.
19.
Furimsky, E., Catalytic hydrodeoxygenation. Applied Catalysis A: General, 2000. 199(2): p.
147-190.
Elliott, D.C., Historical Developments in Hydroprocessing Bio-oils. Energy & Fuels, 2007. 21(3):
p. 1792-1815.
Mortensen, P.M., et al., A review of catalytic upgrading of bio-oil to engine fuels. Applied
Catalysis A: General, 2011. 407(1–2): p. 1-19.

Page 21 of 34
Green Chemistry
View Article Online
DOI: 10.1039/C7GC00367F
20.
21.
22.
Lødeng, R., et al., Chapter 11 - Catalytic Hydrotreatment of Bio-Oils for High-Quality Fuel
Production, in The Role of Catalysis for the Sustainable Production of Bio-fuels and Bio-
chemicals, K.S.T.A.L. Stöcker, Editor. 2013, Elsevier: Amsterdam. p. 351-396.
Boullosa-Eiras, S., et al., Chapter 2 Potential for metal-carbide, -nitride, and -phosphide as
future hydrotreating (HT) catalysts for processing of bio-oils, in Catalysis: Volume 26. 2014,
The Royal Society of Chemistry. p. 29-71.
Chen, L., et al., Aqueous-phase hydrodeoxygenation of carboxylic acids to alcohols or alkanes
over supported Ru catalysts. Journal of Molecular Catalysis A: Chemical, 2011. 351(0): p. 217-
227.
23.
24.
Luo, W., et al., Ruthenium-catalyzed hydrogenation of levulinic acid: Influence of the support
and solvent on catalyst selectivity and stability. Journal of Catalysis, 2013. 301(0): p. 175-186.
Wan, H., R.V. Chaudhari, and B. Subramaniam, Aqueous phase hydrogenation of acetic acid
and its promotional effect on p-cresol hydrodeoxygenation. Energy and Fuels, 2013. 27(1): p.
487-493.
25.
26.
27.
28.
29.
30.
31.
32.
33.
Wan, H., et al., Kinetic investigations of unusual solvent effects during Ru/C catalyzed
hydrogenation of model oxygenates. Journal of Catalysis, 2014. 309(0): p. 174-184.
Newman, C., et al., Effects of support identity and metal dispersion in supported ruthenium
hydrodeoxygenation catalysts. Applied Catalysis A: General, 2014. 477(0): p. 64-74.
Wildschut, J., et al., Hydrotreatment of fast pyrolysis oil using heterogeneous noble-metal
catalysts. Industrial and Engineering Chemistry Research, 2009. 48(23): p. 10324-10334.
Wildschut, J., et al., Insights in the hydrotreatment of fast pyrolysis oil using a ruthenium on
carbon catalyst. Energy and Environmental Science, 2010. 3(7): p. 962-970.
Taylor, K.C., Determination of ruthenium surface areas by hydrogen and oxygen
chemisorption. Journal of Catalysis, 1975. 38(1–3): p. 299-306.
Olcay, H., et al., Aqueous-phase hydrogenation of acetic acid over transition metal catalysts.
ChemCatChem, 2010. 2(11): p. 1420-1424.
Olcay, H., Y. Xu, and G.W. Huber, Effects of hydrogen and water on the activity and selectivity
of acetic acid hydrogenation on ruthenium. Green Chemistry, 2014. 16(2): p. 911-924.
Lienhard, G.E. and T.C. Wang, On the mechanism of acid-catalyzed enolization of ketones.
Journal of the American Chemical Society, 1969. 91(5): p. 1146-1153.
Yaylayan, V.A., S. Harty-Majors, and A.A. Ismail, Monitoring carbonyl-amine reactions and
enolization of 1-hydroxy-2-propanone (acetol) by FTIR spectroscopy. Journal of Agricultural
and Food Chemistry, 1999. 47(6): p. 2335-2340.
34.
35.
Sun, Y.K., and W.H. Weinberg, Catalytic decomposition of formic acid on Ru(001): Transient
measurements. Journal of Chemical Physics, 1991. 94: p. 4587-4599.
Gonzalez Carballo, J.M., Y. Jang, A. Holmen, S. Garcia-Rodriguez, S. Rojas, M. Ojeda, and J.L.G.
Fierro, Catalytic effects of ruthenium particle size on the Fischer-Tropsch Synthesis. Journal of
Catalysis, 2011. 284: p. 102-108.
36.
37.
38.
39.
40.
Elliott, D.C. and T.R. Hart, Catalytic hydroprocessing of chemical models for bio-oil. Energy
and Fuels, 2009. 23(2): p. 631-637.
Oasmaa, A., et al., Characterization of Hydrotreated Fast Pyrolysis Liquids. Energy & Fuels,
2010. 24(9): p. 5264-5272.
Sanna, A., T.P. Vispute, and G.W. Huber, Hydrodeoxygenation of the aqueous fraction of bio-
oil with Ru/C and Pt/C catalysts. Applied Catalysis B: Environmental, 2015. 165: p. 446-456.
Ying, X., et al., Upgrading of fast pyrolysis liquid fuel from biomass over Ru/γ-Al2O3 catalyst.
Energy Conversion and Management, 2012. 55: p. 172-177.
Mahfud, F.H., F. Ghijsen, and H.J. Heeres, Hydrogenation of fast pyrolyis oil and model
compounds in a two-phase aqueous organic system using homogeneous ruthenium catalysts.
Journal of Molecular Catalysis A: Chemical, 2007. 264(1–2): p. 227-236.
Bindwal, A.B., A.H. Bari, and P.D. Vaidya, Kinetics of low temperature aqueous-phase
hydrogenation of model bio-oil compounds. Chemical Engineering Journal, 2012. 207-208: p.
725-733.
41.

Green Chemistry
Page 22 of 34
View Article Online
DOI: 10.1039/C7GC00367F
42.
43.
Trinh, T.N., et al., Comparison of lignin, macroalgae, wood, and straw fast pyrolysis. Energy
and Fuels, 2013. 27(3): p. 1399-1409.
Jung, S.-H., S.-J. Kim, and J.-S. Kim, Characteristics of products from fast pyrolysis of fractions
of waste square timber and ordinary plywood using a fluidized bed reactor. Bioresource
Technology, 2012. 114(0): p. 670-676.
44.
45.
46.
Leijenhorst, E.J., W. Wolters, L. van de Beld, and W. Prins, Inorganic element transfer from
biomass to fast pyrolysis oil: Review and experimentsr. Fuel Processing Technology, 2016.
149: p. 96-111.
Bartholomew, C.H., P.K. Agrawal, and J.R. Katzer, Sulfur poisoning of metals. Adv. Catal.,
1982. 31: p. 135-242.
Wang, H., S.-J. Lee, M.V. Olarte, and A.H. Zacher, Bio-oil stabilization by hydrogenation over
reduced metal catalysts at low temperature. ACS Sustainable Che.m Eng., 2016. 4: p. 5533-
5545.

Page 23 of 34
Green Chemistry
View Article Online
DOI: 10.1039/C7GC00367F
Table 1 Metal loading, H2 uptake, metal dispersion analysis, surface average metal particle
size, BET surface area, pore volume and average pore size of Ruthenium catalysts.
Property
Ru/C fresh
5
Ru/C useda
Ru/TiO2 fresh
Ru/TiO2 useda
metal loading (wt%)
H2 uptake (µmol/g)
metal dispersion (%)
metal particle size (nm)
BET surface area (m2/g)
pore volume (cm3/g)
average pore size (Å)
-
3
-
70.7
-
38.5
7.8
4.2
-
193.2
65.1
1.4
11.7
834
-
-
162
0.33
81.4
57
54
0.80
38.4
0.31
214.7
0.27
196.5
a Characterization of catalyst after reaction with AFBO.

Green Chemistry
Page 24 of 34
Table 2 Hydrogenation of model compounds in aqueous phase over 3 wt% Ru/TiO2 catalyst. (AA) acetic acid, (HA) hydroxyacetone, (FA)
formic acid. Reaction conditions: 6.21 MPa, H2 flow rate 40 mL minꢀ1.
Feedstock
AA
33
120
1.5
37.5
-
HA
14
70
AA+HA AA+FA AA+HA+FA AA+HA+FA AA+HA+FA AA+HA+FA
Time on stream (h)
Reaction temperature, °C
WHSV, h-1
11
22
11
79
11
45
120
120
120
100
120
150
20
1.5
1.5
1.5
0.75
0.75
0.75
Acetic acid conversion, %
Hydroxyacetone conversion, %
Formic acid conversion, %
Carbon in gas effluent, %
Carbon in liquid effluent, %
Carbon balance, %
Carbon selectivity, %
CO2
-
93.6
-
45.5
100.0
-
33.4
-
19.2
100.0
100.0
8.9
3.0
63.5
65.7
0.3
40.8
100.0
100.0
13.9
60.4
99.8
100
-
100.0
10.2
79.1
89.3
7.1
0.1
5.9
38.2
67.2
105.4
91.1 97.1
98.2 97.2
96.2
102.1
96.6
95.6
95.9
84.9
105.5
98.8
5.0
8.3
6.5
0.1
0.1
0.1
0
0
0.1
0
2.3
3.7
2.6
0.3
0
9.6
16.0
6.9
1.1
0.4
0.2
0.1
0.3
64.0
0.1
0
5.0
7.8
0.0
1.0
0.3
0.1
0.0
2.3
0.0
0.5
0.6
0.1
94.7
0.1
6.6
10.4
4.0
14.2
22.2
8.2
Methane
Ethane
2.0
Propane
0
0.7
0.9
2.4
Butane
0
0.5
0.4
0.4
Pentane
0
0.1
0
0.3
0.3
0.3
Hexane
0
0.2
0.2
0.2
Methanol
0
0
0
0.2
0.1
0.0
Ethanol
78.3
0
0
30.4
0.9
58.9
0.5
13.1
0.4
20.1
0.7
17.9
1.8
Propanol
0
1,2-propanediol
Ethyl acetate
0
99.9
0
69.4
0.4
56.1
0.4
31.9
0.4
1.4
1.0

Page 25 of 34
Green Chemistry
View Article Online
DOI: 10.1039/C7GC00367F
Table 3 Aqueous phase hydrogenation of the light oxygenate fraction of bio oil over 3 wt%
Ru/TiO2 catalyst. Reaction conditions: Pressure 6.21 MPa and H2 flow rate 40 mL minꢀ1.
(n.d.: not detected).
Temperature (°C)
WHSV (h-1)
100
1.5
9h
1416.3
0.5
120
1.5
21h
1355.0
0.3
n.d.
45.1
n.d.
n.d.
95.5
n.d.
140
1.5
31h
120
0.75
68h
1718.2
0.3
0.6
117.0
n.d.
n.d.
137.8
n.d.
n.d.
6.7
120
0.5
93h
1706.5
0.3
0.6
121.7
n.d.
n.d.
156.8
n.d.
120
1.5
106h
1913.8
0.6
Species (mmol L-1)
Feed
1936.6
0.9
61.6
230.8
12.4
28.7
85.6
4.6
Acetic acid
Hydroxymethylfurfural
Furfural
Formic acid
Acetone
Acetaldehyde
Methyl acetate
Propionaldehyde
2-propen-1-ol
2-butanone
1675.1
n.d.
n.d.
2.0
n.d.
n.d.
107.4
n.d.
n.d.
1.6
3.2
2.2
120.1
n.d.
n.d.
115.7
n.d.
n.d.
3.6
169.1
n.d.
4.6
96.7
1.2
15.1
3.9
n.d.
3.7
n.d.
7.4
n.d.
5.7
1-OH-2-butanone
3-OH-2-butanone
2-hexanone
3-hexanone
2-furanone
Cyclopentanol
Methanol
Ethanol
Hydroxyacetaldehyde
Hydroxyacetone
Propionic acid
Butyric acid
1-propanol
2-propanol
1,2-propanediol
Ethylene glycol
Ethyl acetate
1-butanol
25.8
21.4
3.0
2.9
97.7
0.9
423.8
4.1
610.7
477.5
104.9
211.2
n.d.
n.d.
n.d.
1.5
n.d.
3.5
823.1
42.8
28.8
477.1
78.1
178.8
11.2
2.0
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
21.0
495.8
45.5
n.d.
20.2
133.3
74.4
14.1
10.2
428.7
27.4
n.d.
12.2
8.7
n.d.
n.d.
n.d.
n.d.
n.d.
1.1
522.3
35.0
n.d.
419.1
122.7
203.4
11.5
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
9.6
470.2
35.8
n.d.
363.3
124.8
171.2
10.5
n.d.
n.d.
n.d.
n.d.
n.d.
19.8
1.6
505.0
26.7
31.8
451.4
112.0
189.8
10.1
n.d.
n.d.
61.3
0.9
n.d.
12.3
591.4
34.8
12.9
285.9
113.4
161.7
13.8
3.7
166.6
194.5
n.d.
12.7
2.9
28.8
221.3
2.0
10.0
77.4
1.7
14.2
55.6
1.9
9.8
11.2
10.8
9.0
2-butanol
1,2-butanediol
1,4-butanediol
2,3-butanediol
Butyrolactone
1,4-pentanediol
Furfuryl alcohol
1-hexanol
Carbon identified in liquid
TOC in liquid
Carbon in liquid phase (%)
Carbon in gas phase (%)
C sel. in gas phase (%)
Methane
19.5
79.7
13.3
17.9
66.4
16.4
5.0
8348
12139
84
25.0
152.9
31.7
18.0
60.0
8.2
19.7
81.1
29.5
18.1
11.9
30.7
14.6
15.2
29.9
126.5
108.1
10.4
66.6
31.8
16.2
44.4
74.5
5.8
3.2
8057
12419
86
6.5
7778
12820
89
5.0
8236
13311
92
9617
14453
7999
13438
93
7951
12784
89
10
5
10
10
10
5
47
4
53
4
52
5
38
4
33
3
21
4
Ethane
C3-5 alkanes
CO2
CO
19
30
0
11
32
0
11
32
0
31
17
10
33
18
13
40
11
24

Green Chemistry
Page 26 of 34
Table 4 Potential and actual product yield for hydrogenation of light fraction of bioꢀoil over 3 wt% Ru/TiO2 catalyst. Reaction conditions:
Pressure 5.18 MPa, WHSV 1.5 hꢀ1, and H2 flow rate 40 mL minꢀ1.
Aqueous phase feed
Species
Acetic acid
Potential product yielda
Yield (%)b
120°C
1.6
Wt%
wt%
100°C
2.0
140°C
2.1
11.63 Ethanol
8.92
3.79
3.64
0.88
0.24
0.19
0.64
Hydroxyacetaldehyde
Hydroxyacetone
2-furanone
3.67 Ethylene Glycol
3.54 Propylene Glycol
0.82 1,4-butanediol
0.23 1,2-butanediol
0.19 2,3-butanediol
0.59 1,4-pentanediol
36.1
6.0
31.9
4.5
34.9
89.6
129.5
-
81.8
75.0
63.2
107.8
156.8
95.8
1-OH-2-butanone
3-OH-2-butanone
Furfural
152.6
96.9
510.5
121.9
a.
b.
Potential product yield = mass of feed x MW of product/MW of feed.
Yield = mass of product produced divided by mass of potential product yield.

Page 27 of 34
Green Chemistry
View Article Online
DOI: 10.1039/C7GC00367F
Table 5 Conversion of individual compounds in model feed and the aqueous phase bio oil
over the Ru catalysts. H2 flow rate 40 mL minꢀ1.
Feedstock
AA
Ru/TiO2
33
AA+HA
AA+FA AA+HA+FA Bio oil
Bio oil
Ru/C
6
Catalyst
Ru/TiO2 Ru/TiO2
Ru/TiO2
24
Ru/TiO2
21
Time on stream (h)
11
22
Reaction
120
120
120
120
120
120
temperature, °C
Pressure, MPa
WHSV, h-1
6.21
1.5
6.21
1.5
6.21
1.5
6.21
1.5
6.21
1.5
27
38
79
6
5.18
1.5
AA conversion, %
HA conversion, %
FA conversion, %
Sethanol , %1
38
46
33
36
33
100
66
86
96
68
99
97
26
100
65
78
1
:

Green Chemistry
Page 28 of 34
View Article Online
DOI: 10.1039/C7GC00367F
Figure 1 Hydrogenation of hydroxyacetone in aqueous phase over 3 wt% Ru/TiO2. Reaction
conditions: Pressure 6.21 MPa, WHSV 20 hꢀ1, H2 flow rate 40 mL minꢀ1. (∆) 5 wt%
hydroxyacetone, (▲) 5 wt% hydroxyacetone+10 wt% acetic acid.

Page 29 of 34
Green Chemistry
View Article Online
DOI: 10.1039/C7GC00367F
Figure 2 Effect of acetic acid and formic acid on hydrogenation of hydroxyacetone over 3
wt% Ru/TiO2. Reaction conditions: Pressure 6.21 MPa, temperature 50 °C, WHSV 20 hꢀ1, H2
flow rate 40 mL minꢀ1. Legend: (open triangle) 5 wt% hydroxyacetone, (black triangle) 5 wt%
hydroxyacetone and 10 wt% acetic acid, (horizontally lined triangle) 5 wt% hydroxyacetone
and 20 wt% acetic acid, (diagonally lined triangle) 5 wt% hydroxyacetone and 2 wt% formic
acid.