G Model

JAAP-3928; No. of Pages 10 ARTICLE IN PRESS

Journal of Analytical and Applied Pyrolysis xxx (2016) xxx–xxx

Contents lists available at ScienceDirect

Journal of Analytical and Applied Pyrolysis

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

Separation of chemical groups from bio-oil water-extract via

sequential organic solvent extraction
Shoujie Ren a , X. Philip Ye a,∗ , Abhijeet P. Borole b
a
Biosystems Engineering & Soil Science, University of Tennessee, Knoxville, TN, 37996, USA
b
Bioscience Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831, USA

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

Article history: The chemical complexity of bio-oil aqueous phase limits its efficient utilization. To improve the effi-
Received 4 November 2016 ciency of the bio-oil biorefinery, this study focused on the separation of chemical groups from the bio-oil
Received in revised form 1 January 2017 water-extract via sequential organic solvent extractions. Due to their high recoverability and low sol-
Accepted 3 January 2017
ubility in water, four solvents (hexane, petroleum ether, chloroform, and ethyl acetate) with different
Available online xxx
polarities were evaluated, and the optimum process conditions for chemical extraction were determined.
Chloroform had high extraction efficiency for furans, phenolics, and ketones. In addition to these classes
Keywords:
of chemical, ethyl acetate had a high extraction efficiency for organic acids. The sequential extraction
Bio-oil aqueous phase
Organic solvent
using chloroform followed by ethyl acetate resulted in 62.2 wt.% of original furans, ketones, alcohols, and
Solvent extraction phenolics being extracted into chloroform, while 62 wt.% acetic acid was extracted into ethyl acetate,
Chemical groups leaving behind a high concentration of levoglucosan (∼53.0 wt.%) in the final aqueous phase. Chemicals
separated via the sequential extraction could be used as feedstocks in a biorefinery using processes such
as catalytic upgrading of furans and phenolics to hydrocarbons, fermentation of levoglucosan to produce
alcohols and diols, and hydrogen production from organic acids via microbial electrolysis.
© 2017 Elsevier B.V. All rights reserved.

1. Introduction polarity and solubility of the chemicals. Because of their high polar-
ity and solubility in water, levoglucosan and organic acids, such as
Bio-oil from biomass pyrolysis is a promising feedstock for the acetic acid, have high distribution coefficients in water, resulting
production of transportation fuels and value-added chemicals [1,2]. in a high concentration in the aqueous phase. Furans, such as fur-
However, some disadvantages make it difficult to directly use bio- fural and furanone, have a distribution coefficient similar to acetic
oil as a transportation fuel. Bio-oil contains hundreds of chemicals acid [6]. Some phenolic compounds, such as syringol and guaiacol,
that are classified according to their functional groups, including have low polarity, but due to their low initial concentration in crude
phenolics, furans, organic acids, ketones, aldehydes, esters, and bio-oil, these compounds also have a high distribution coefficient
anhydrosugars [3,4]. Unlike petroleum oil, most chemicals in bio- to water and a considerable fraction can also be extracted by water
oil are unstable oxygenated compounds, amounting to as high [6]. Thus, the bio-oil aqueous phase is a complex mixture requiring
as 40 wt.% of oxygen [5]. Bio-oil also has a significant content of further separation to be efficiently utilized.
water generally ranging from 15–30 wt.%, depending on the dif- Due to the multitude of chemicals present in the aqueous phase,
ferent types of biomass and pyrolysis processes [5]. Furthermore, a variety of applications of bio-oil aqueous phase have been investi-
organic acids are the major contributors to the acidity of bio-oil, gated, including aqueous phase reforming [10–12], extraction and
causing corrosion that requires special containers for the storage recovery of value-added chemicals, such as acetic acid [13,14], and
and transportation. ethanol production via hydrolysis and fermentation of levoglu-
To recover and utilize the water-soluble chemicals, phase sepa- cosan [12,15]. Recently, a novel application of the aqueous phase for
ration of bio-oil into an organic phase and an aqueous phase by hydrogen production has been reported via a microbial electrolysis
adding water has been investigated [2,6–9]. The extraction effi- (MEC) process [16]. Certain biological processes, such as micro-
ciency of chemicals in bio-oil by water is highly dependent on the bial fermentation, are negatively affected by furans and phenolics
[12,16–18]. Therefore, the removal of non-desirable compounds
from the bio-oil aqueous phase is required.
∗ Corresponding author. Several methods including distillation, solvent extraction, and
E-mail address: xye2@utk.edu (X.P. Ye). column chromatography have been developed for separating and

http://dx.doi.org/10.1016/j.jaap.2017.01.004
0165-2370/© 2017 Elsevier B.V. All rights reserved.

Please cite this article in press as: S. Ren, et al., Separation of chemical groups from bio-oil water-extract via sequential organic solvent
extraction, J. Anal. Appl. Pyrol. (2016), http://dx.doi.org/10.1016/j.jaap.2017.01.004
G Model
JAAP-3928; No. of Pages 10 ARTICLE IN PRESS
2 S. Ren et al. / Journal of Analytical and Applied Pyrolysis xxx (2016) xxx–xxx

recovering the various components and to characterize bio-oil and integrated into a biorefinery for the production of fuels and chem-
the associated aqueous phase [7,13,19–22]. Among these methods, icals (Fig. 1B).
solvent extraction is considered cost-effective, as it can be operated A semi-pilot scale auger pyrolysis system (Proton Power, Inc.,
at room temperature and atmospheric pressure [23]. Garcia-Perez Lenoir City, TN) was used to pyrolyze switchgrass for bio-oil
et al. investigated different organic solvents to fractionate the production. A detailed description of the pyrolysis system was pro-
chemicals for characterization [24]. Wei et al. found that chloro- vided elsewhere [25]. The pyrolysis was conducted at 500 ◦ C with
form had good performance in extracting phenolic compounds [7]. a residence time of 72 s.
However, a considerable amount of furans, ketones, and alcohols The separation of the bio-oil aqueous phase was achieved by
were still detected by gas chromatography and mass spectrometry simply adding water to crude bio-oil [26]. First, the crude bio-oil
(GC/MS) in bio-oil aqueous phase after organic solvent extraction was mixed with four times by weight of distilled water. Then, the
[7]. Moreover, organic acids and levoglucosan were not well sep- mixture was shaken vigorously using a mini vortexer (Model MS1
arated by these organic solvents. Therefore, only using a single S7, Fisher Scientific) until a homogeneous mixture was formed. The
solvent might not sufficiently extract and separate these chemicals mixture was stored at 4 ◦ C overnight, followed by centrifugation
from the bio-oil aqueous phase. using an IEC Model 120 clinical centrifuge (International Equipment
Separating compounds in the bio-oil aqueous phase into differ- Company) at a relative centrifugal force (RCF) of 2400 g for 30 min to
ent chemical groups is practically useful. It minimizes the effects accelerate phase separation. After centrifugation, bio-oil aqueous
of non-desirable chemicals on the application of specific chemi- phase (BOAP I) on top was collected and analyzed, as described
cal groups, develops strategies for the application of the bio-oil below.
aqueous phase, and improves its application efficiency. Due to the Physical properties of crude bio-oil and BOAP I, including den-
very low concentration of individual compounds of furans, alco- sity, pH, viscosity, water content, solid content, ash content, and
hols, ketones, and phenolics in the bio-oil aqueous phase, isolating total acid number, were measured in triplicate. Density was mea-
these chemicals individually would be difficult and costly. There- sured according to the ASTM D1217 (2012) standard [27], and pH
fore, separating these chemicals as a group (Group 1) may be more was measured with an Extech pH meter. A Schott TitroLine Karl
practical. Due to the significant amount of organic acids (Group 2) Fischer volumetric titrator was used to measure water content
and anhydrosugars (Group 3) in the bio-oil aqueous phase, separat- according to ASTM D4377 (2011) [28]. Viscosity was measured at 40
ing these two chemical groups is feasible. Therefore, the purpose ◦ C with serialized Schott Ubbelohde capillary viscometers accord-

of this study was to develop a method to separate these three ing to ASTM D445 (2012) [29]. Ash content was measured according
chemical groups from bio-oil aqueous phase using organic solvents. to ASTM D482 (2013) at 575 ◦ C [30]. The solid content was deter-
Four solvents including hexane, petroleum ether, chloroform, and mined according to Boucher et al. [31]. Total acid number (TAN)
ethyl acetate with different polarities were first evaluated individ- was measured by titrating bio-oil (0.1 g) or BOAP I (0.2 g) in a sol-
ually for chemical extraction. Optimum conditions and extraction vent of water, isopropyl alcohol and toluene (volume ratio of water:
efficiency related to the different chemical groups were deter- isopropyl alcohol: toluene = 1: 99:100) with 0.1 M KOH isopropyl
mined. According to these analyses, a sequential extraction using alcohol solution to an end point of pH 11 according to ASTM D664
chloroform followed by ethyl acetate were further investigated (2011) [32].
for separating Group 1 (furans, alcohols, ketones, and phenolics),
Group 2 (organic acids) and Group 3 (anhydrosugars).
2.3. Chemical extraction from bio-oil aqueous phase

2. Materials and methods The separated BOAP I was used for extraction experiments. Four
organic solvents, hexane, petroleum ether, chloroform, and ethyl
2.1. Materials acetate, were first individually investigated to separate chemicals
from BOAP I. Four different volumetric ratios of solvent to BOAP
Air-dried switchgrass (Panicum virgatum L.) obtained from a I (0.5:1, 1:1, 2:1, and 3:1) were employed to determine the effect
local producer in East Tennessee was used for the bio-oil pro- of solvent-to-feed ratio (S/F ratio) on the chemical extraction. A
duction. The water content of the biomass was 7–8 wt.%. Before fixed volume of BOAP I at 20 mL was weighed and measured into
pyrolysis, the material was ground to less than a 2 mm particle a 100 mL beaker. Then a predesignated volume of organic solvent
size. The switchgrass is composed of 34.1 wt.% cellulose, 25.7 wt.% was added into the beaker. The mixture of BOAP I and the organic
hemicellulose, 18.8 wt.% lignin, 14.2 wt.% extractives, and 2.7 wt.% solvent was magnetically stirred for 30 min. After stirring, the mix-
ash [25]. ture was transferred to a separatory funnel and left undisturbed for
Hexane (a mixture of isomers, purity > 98.5%), petroleum 24 h to allow phase separation. During the stirring and separation,
ether (ACS certified grade), chloroform (purity > 99.0%), and ethyl the beaker and funnel were sealed to minimize solvent evaporation.
acetate (purity > 99.5%) purchased from Thermo Fisher Scientific After separation, the solvent phase and BOAP II were collected and
(Waltham, MA) were used as organic solvents for chemical extrac- weighed. Chemicals extracted to the organic solvent were recov-
tion from bio-oil aqueous phase (BOAP). All these chemicals were ered by evaporating the organic solvent via a rotary evaporator at
used as received. Fifteen external standards, all purchased from 40 ◦ C.
Sigma-Aldrich (St. Louis, MO), were used for quantifying com- First, distribution coefficient and extraction efficiency [33] were
pounds in the BOAP. used to evaluate the four solvents in extracting BOAP I. The distri-
bution coefficient (Di ) is defined as the ratio of equilibrium mass
concentration (g/L) of compounds in the solvent (i indicates six
2.2. Crude bio-oil production and bio-oil aqueous phase groups based on their functional groups: acids, furans, alcohols,
separation ketones, phenolics, and levoglucosan, that is the only anhydro-
sugar detected) to their equilibrium mass concentration in BOAP
A schematic diagram of the experiment for the separation of I, according to Eq. (1).
chemical groups by a sequential extraction is shown in Fig. 1A. The
process includes bio-oil production, aqueous phase separation, and mass concentration of i in solvent
Di = (1)
organic solvent extractions. The separated chemical groups can be mass concentration of i in BOAP I

Please cite this article in press as: S. Ren, et al., Separation of chemical groups from bio-oil water-extract via sequential organic solvent
extraction, J. Anal. Appl. Pyrol. (2016), http://dx.doi.org/10.1016/j.jaap.2017.01.004
G Model
JAAP-3928; No. of Pages 10 ARTICLE IN PRESS
S. Ren et al. / Journal of Analytical and Applied Pyrolysis xxx (2016) xxx–xxx 3

A B
Switchgrass

Auger pyrolysis Resins

at 500 ºC
Phase Organic phase
Distilled water: separation Catalytic upgrading
bio-oil by weight Bio-oil rich in pyrolytic Aromatics
at 4:1 lignin

Group 1 rich in Transportation

Extraction Catalytic hydrotreating fuels
Chloroform BOAP I phenolics,
alcohols,
furans, ketones
Hydrogen
Extraction Group 2 rich in
Ethyl acetate BOAP II
acids
Acetic acid

BOAP III
Group 3 rich in Fermentation
levoglucosan Alcoholic fuels

Fig. 1. (A) Experimental procedures for the production of bio-oil and sequential extraction of chemical groups from the aqueous phase (BOAP I–III), and (B) their envisaged
applications for the production of fuels and chemicals.

Extraction efficiency (Xj ) is defined as the mass percentage Table 1

Properties of crude bio-oil and BOAP I (Deionized water added at 4:1 by weight for
(wt.%) of 15 individual chemical compounds (j) transferred from
BOAP I separation).
BOAP I to the organic solvent after the extraction process according
to Eq. (2). Properties Crude bio-oil BOAP I

mass of j in BOAP I − mass of j in BOAP II Water content (wt%) 41.1 ± 0.5 90.8 ± 1.4
Xj = × 100% (2) Total solid (wt%) 0.77 ± 0.05 Not detected
mass of j in BOAP I pH value 2.4 ± 0.1 2.6 ± 0.1
Density (g/mL) 1.14 ± 0.01 1.02 ± 0.00
According to the results evaluating the four solvents, chloroform
Ash (wt%) 0.09 ± 0.01 0.02 ± 0.01
and ethyl acetate were chosen for sequential extraction to sepa- Viscosity at 40 ◦ C centistokes (cSt) 3.23 ± 0.03 0.71 ± 0.01
rate the chemical groups from BOAP I, which was first extracted by TAN, mg KOH/g 130.3 ± 1.9 25.3 ± 0.1
chloroform to recover furans, ketones, and phenolics. After chloro-
form extraction, BOAP II and the solvent phase were collected and
weighed. BOAP II was further extracted using ethyl acetate. The tified using a gas chromatography-flame ionization detector
organics extracted to chloroform and ethyl acetate were recovered (GC-FID) with a HP-5 column (30 m × 0.32 mm × 0.25 ␮m): 2(5H)-
by evaporating the organic solvents via a rotary evaporator at 40 ◦ C. furanone, 1-hydroxy-2-butanone, 1,3-propanediol, 3-methyl-1,2-
BOAP II and III were collected and analyzed. The experiment was cyclopentanedione, guaiacol, creosol, 2,6-dimethoxyphenol, 3-
performed in triplicate. ethylphenol. The same temperature program as that with GC/MS
was used in the GC-FID. The compounds extracted to organic sol-
2.4. Chemical identification and quantification by GC/MS, GC-FID vent phase were also quantified by the GC-FID. Compounds were
and HPLC quantified using external standards in both the HPLC and GC-FID
analysis.
Chemical compounds in BOAP I were identified using
gas chromatography/mass spectrometry (GC/MS). A Shimadzu 3. Results and discussion
GC/MS (QP2010S) with a Restek Rtx-5MS capillary column
(30 m × 0.25 mm × 0.25 ␮m) was used. The column temperature 3.1. Characterization of bio-oil and BOAP I
was programmed at 45 ◦ C for 3 min and increased to 150 ◦ C at
5 ◦ C/min; then, it was further increased to 260 ◦ C at 10 ◦ C/min and The physical properties of crude bio-oil and BOAP I are shown in
held for 7 min at the final temperature. The inlet was set at 240 ◦ C, Table 1. The crude bio-oil obtained from the pyrolysis was an even
and sample injection was made in a split mode (1:20). The com- mixture containing about 41.1 wt.% water. After aqueous phase
pounds were identified by comparing their mass spectra with those separation, BOAP I showed a light yellow color containing about
from the National Institute of Standards and Technology (NIST) 90.8 wt.% water. After the separation, most polar compounds were
mass spectral data library. extracted by water to BOAP I. The oligomers derived from lignin
The acids, levoglucosan, hydroxymethylfurfural, furfural, phe- remained in the black and viscous organic phase. Due to the dilu-
nol, and 1,2-benzenediol in BOAP I–III were quantified using a high tion by water, the pH in BOAP I slightly increased compared with
pressure liquid chromatography system (HPLC, Jasco 2000Plus, the crude bio-oil, while the density, ash content, viscosity, and TAN
Jasco Analytical Instruments, Easton, MD) equipped with a MD- decreased.
2018 plus photodiode array detector (PAD), a RI-2031 Plus GC/MS analysis showed that more than 50 compounds were
intelligent RI detector, and an AS-2055 plus auto sampler [26]. detected in BOAP I. The relative peak areas in GC/MS chromatogram
The liquid chromatography was conducted at 50 ◦ C using a Bio-Rad were considered a useful indication of the relative abundance of
column HPX-87H (300 × 8 mm). The injected sample volume was chemicals, which has been used in bio-oil analysis [38–41]. Among
20 ␮L. The mobile phase was 5 mM H2 SO4 in deionized water with these compounds, levoglucosan, acetic acid, furfural, and phe-
a flow rate of 0.6 mL/min. nol were the most abundant as detected by GC/MS spectrometry.
The following compounds in BOAP I–III, which have been According to their functional groups, the detected compounds were
reported in switchgrass bio-oil analysis [34–37], were quan- classified into anhydrosugars, acids, furans, alcohols, phenolics,

Please cite this article in press as: S. Ren, et al., Separation of chemical groups from bio-oil water-extract via sequential organic solvent
extraction, J. Anal. Appl. Pyrol. (2016), http://dx.doi.org/10.1016/j.jaap.2017.01.004
G Model
JAAP-3928; No. of Pages 10 ARTICLE IN PRESS
4 S. Ren et al. / Journal of Analytical and Applied Pyrolysis xxx (2016) xxx–xxx

Fig. 2. Chemical distribution (relative peak area percentage analyzed by GC/MS) in BOAP I.

aldehydes, ketones, esters, and others (nitrogen-containing com-

pounds). The distribution of these chemicals in BOAP I according to
their relative GC/MS peak area is shown in Fig. 2.
Previous study has pointed out that levoglucosan, organic acids,
alcohols, furans such as furfural, and acetol have high distribution
coefficient in water [6]. After the bio-oil aqueous phase separation,
the majority of these compounds can be extracted to the aque-
ous phase. Although phenolic compounds have a lower polarity
than acids and furans, their low initial concentration in the crude
bio-oil affords them a high distribution coefficient in water [6].
GC/MS analysis detected a number of phenolic compounds with
small peaks in BOAP I, but these small peaks added up to a total
relative area of about 28 area%. A number of aldehydes, ketones,
and esters with relatively small area percentage were also found in
BOAP I.
Taking their high peak area percentages into account, 15 com-
pounds were further quantified using GC-FID and HPLC, including
two acids, three furans, two alcohols, six phenolics, one ketone,
and one anhydrosugar (levoglucosan). These quantified chemicals Fig. 3. The mass percentage of total chemicals extracted by different organic sol-
accounted for about 53.9 wt.% of total chemicals in BOAP I. The vents and S/F ratios.
concentrations of these 15 compounds in BOAP I are presented
in Table 2. Acids and anhydrosugars are the two major chemi-
cal groups extracted to BOAP I by water. Among the acids, acetic ity for chemicals than hexane and petroleum ether, in agreement
acid was detected at the highest concentration, which was about with a previous report [7]. Ethyl acetate has the highest affinity for
16.54 g/L. The concentration of levoglucosan was about 18.99 g/L chemicals in BOAP I, followed by chloroform. These two can extract
in BOAP I. These two compounds were the most abundant chem- over 50 wt.% of total chemicals in BOAP I, due to their high polarity
icals observed in BOAP I. The quantification of other chemicals in (polarity of chloroform at 4.1 and polarity of ethyl acetate at 4.4). In
BOAP I showed that 1-hydroxy-2-butanone and hydroxymethyl- BOAP I, most chemicals present are oxygenated compounds, hav-
furfural with concentration of 2.25 g/L and 1.37 g/L, respectively, ing high polarity and solubility in water [6]. Therefore, hexane and
are the major chemicals following levoglucosan, acetic acid, and petroleum ether with very low polarity and solubility in water limit
propionic acid. The total quantified concentration of furans and the dispersion of the oxygenated compounds. These results agree
alcohols were about 3.17 g/L and 2.6 g/L, respectively. While the with previous studies suggesting that to recover more chemicals
individual phenolic compound had a low concentration in BOAP I, from BOAP I, solvents with high polarity, such as ethyl acetate and
the total concentration of quantified phenolic compounds added chloroform would be more preferable than hexane and petroleum
up to 1.5 g/L. These 15 quantified compounds (6 groups) were used ether with low polarity.
in the next step to evaluate the extraction efficiency of organic The effect of different volumetric ratios of organic solvent to
solvents from BOAP I. BOAP I (S/F ratio) was also investigated (Fig. 3). With a S/F ratio
of 0.5, all the four solvents showed poor extraction performance
3.2. Effects of organic solvents on the extraction of total chemicals for chemicals, with the maximum extracted amount being less
than 22 wt.%. Increasing the S/F ratio to 1, the amount of chemi-
Fig. 3 shows the total mass percentages of the 15 chemicals cals extracted using hexane and petroleum ether did not change
extracted by the four different organic solvents. Among the four, significantly. However, the amount of extracted chemicals was
hexane has the lowest affinity for chemicals in BOAP I, extract- significantly increased using chloroform and ethyl acetate from
ing less than 30 wt.%. Petroleum ether has slightly higher affinity 14.7 wt.% to 47.8 wt.% and 21.1 wt.% to 54.1 wt.%, respectively.
for chemicals than hexane, with the highest amount of chemicals When the S/F ratio further increased to 2, the amount extracted by
extracted at 33.2 wt.% from BOAP I. Chloroform has a higher affin- hexane, petroleum ether and ethyl acetate, except for chloroform

Please cite this article in press as: S. Ren, et al., Separation of chemical groups from bio-oil water-extract via sequential organic solvent
extraction, J. Anal. Appl. Pyrol. (2016), http://dx.doi.org/10.1016/j.jaap.2017.01.004
G Model
JAAP-3928; No. of Pages 10 ARTICLE IN PRESS
S. Ren et al. / Journal of Analytical and Applied Pyrolysis xxx (2016) xxx–xxx 5

Table 2
Concentrations of 15 compounds quantified in BOAP I, BOAP II (aqueous fraction produced post chloroform extraction), and BOAP III (aqueous fraction remaining after ethyl
acetate extraction). Data reported are average numbers based on three replicates.

Classifications Major compounds Concentration in BOAP I (g/L) Concentration in BOAP II (g/L) Concentration in BOAP III (g/L)

Acids Acetic acid 16.54 14.64 6.26

Propionic acid 6.37 6.35 3.93
Anhydrosugars Levoglucosan 18.99 17.46 16.62
Furans Furfural 0.98 0.05 BDL
2(5H)-Furanone 0.82 0.31 0.04
Hydroxymethylfurfural 1.37 0.62 0.17
Alcohols 1-Hydroxy-2-butanone 2.25 1.17 0.51
1,3-Propanediol 0.35 0.29 0.18
Ketones 3-Methyl-1,2-cyclopentanedione 0.40 0.27 BDL
Phenolics 1,2-Benzenediol 0.63 0.54 0.01
Phenol 0.15 BDL BDL
Guaiacol 0.20 BDL BDL
Creosol 0.15 BDL BDL
2,6-Dimethoxyphenol 0.17 BDL BDL
3-Ethylphenol 0.20 BDL BDL

BDL: The concentrations are below the detection limit.

(increased only 2 wt.%), significantly increased as compared to that varied greatly for the different compounds of furans. Extrac-
at S/F ratio of 1. This result indicates that the S/F ratio of chloroform tion efficiencies for hydroxymethylfurfural were observed at less
to BOAP I at 1 is desirable for chemical extraction from an eco- than 8 wt.%. Hexane and petroleum ether had good extraction
nomic viewpoint, consistent with a previous study [7]. When the performance for 2(5H)-furanone with extraction efficiencies at
S/F ratio further increased to 3, there were no significant changes in about 58 wt.% and 59 wt.%, respectively. Extraction efficiencies
extracted chemicals for all the tested solvents. Therefore, the opti- for furfural using hexane and petroleum ether were 13.3 wt.%
mum S/F ratio were 2 for hexane, petroleum ether, or ethyl acetate and 19.9 wt.% at the S/F ratio of 0.5, respectively; when the S/F
and 1 for chloroform. ratio was increased to 2, a significant increase in extraction effi-
ciency for furfural was observed using both solvents. Further
3.3. Effects of solvents on the separation of 15 individual increasing the S/F ratio to 3 using hexane and petroleum ether
chemicals increased the extraction efficiencies for furfural by about 7 and
10 wt.%, respectively. However, the highest extraction efficien-
After organic solvent extraction of BOAP I, 15 chemicals in BOAP cies using hexane and petroleum ether for furfural were achieved
II were quantified using GC-FID and HPLC to determine the effects at 54.5 and 53.6 wt.%, respectively. Compared to hexane and
of different organic solvents and S/F ratios on the extraction of these petroleum ether, chloroform and ethyl acetate had a better extrac-
chemicals. The extraction efficiency of individual chemical was cal- tion performance for these chemical groups (two alcohols, one
culated according to Eq. (2). The results are presented in Fig. 4 for ketone, and three furans). The extraction efficiencies using chloro-
Group 2 (two organic acids) and Group 3 (one anhydrosugar) chem- form reached to about 29 wt.% for 1-hydroxy-2-butanone, 10 wt.%
icals, and in Figs. 5 and 6 for Group 1 (twelve chemicals of furans, for 1,3-Propanediol, 68 wt.% for 3-methyl-1,2-cyclopentanedione,
alcohols, ketones, and phenolics). 54 wt.% for hydroxymethylfurfural, 63 wt.% for 2(5H)-furanone, and
As Fig. 4 shows, the extraction efficiencies for anhydrosugar 95 wt.% for furfural at the low S/F ratio of 0.5. When the S/F
(levoglucosan), and organic acids (acetic acid and propionic acid) ratio increased to 1, the extraction efficiencies for 1-hydroxy-2-
in BOAP I by hexane and petroleum ether were less than 10 wt.%. butanone, 1,3-Propanediol, 3-methyl-1,2-cyclopentanedione, and
Increasing loading of the two solvents did not significantly increase hydroxymethylfurfural greatly increased to 48.1 wt.%, 19 wt.%,
the extraction efficiency. Chloroform had low extraction efficien- 81 wt.%, and 69.6 wt.%, respectively. When the S/F ratio increased
cies for anhydrosugar (less than 8.8 wt.%) and organic acids (less to 2, the extraction efficiencies for these four compounds further
than 14 wt.% for acetic acid and less than 3.9 wt.% for propionic acid) increased to 57 wt.%, 36 wt.%, 89 wt.% and 82 wt.%, respectively.
in BOAP I at low S/F ratios (volume ratio at 0.5 and 1); when the Further increasing the S/F ratio to 3, the extraction efficiency for
S/F ratio increased to 3, the extraction efficiencies for levoglucosan, 1,3-Propanediol increased to 56 wt.% while only a slight increase
acetic acid, and propionic acid increased to 11.1 wt.%, 23.8 wt.%, and was observed for the other three compounds. No significant change
34 wt.%, respectively. Ethyl acetate showed a performance similar in the extraction efficiency was observed for 2(5H)-furanone and
to chloroform for the extraction of levoglucosan. The extraction furfural when the S/F ratio was increased from 0.5 to 3. The extrac-
efficiency for acetic acid was low using ethyl acetate when the S/F tion efficiency for 2(5H)-furanone by ethyl acetate was higher
was at 0.5. However, when the S/F ratios were increased to 2 and 3, than that by chloroform. However, the extraction efficiency for
the extraction efficiency of acetic acid was significantly increased to 1-hydroxy-2-butanone, 3-methyl-1,2-cyclopentanedione, hydrox-
55 wt.% and 61.2 wt.%, respectively. This result suggests that the S/F ymethylfurfural, and furfural by ethyl acetate was lower than that
ratio has great effect on acetic acid extraction. Because ethyl acetate by chloroform, especially at low solvent loading (S/F ratio at 0.5 and
can be used as a hydrogen bond acceptor in extraction [42], the 1).
increase in its loading provides more receptors for hydrogen bond, Hexane and petroleum ether had good performance for
thus breaks the hydrogen bond between acetic acid and water. the extraction of phenolic compounds (3-ethylphenol, guaiacol,
Fig. 5 illustrates the extraction efficiencies for two alcohols 2,6-dimethoxyphenol, and creosol) as shown in Fig. 6. When
(1-hydroxy-2-butanone, 1,3-Propanediol), one ketone (3-methyl- the S/F ratio was increased to 2, all creosol in BOAP I was
1,2-cyclopentanedione), and three furans (hydroxymethylfurfural, extracted to solvents. At this S/F ratio, extraction efficiencies for
2(5H)-furanone, furfural) in BOAP I using the four organic sol- 3-ethylphenol, guaiacol, and 2,6-dimethoxyphenol were about
vents. Both hexane and petroleum ether had poor performance, 73.5 wt.%, 80.1 wt.%, and 36.8 wt.%, respectively, by hexane, and
even with increased loading, for the extraction of alcohols and 68.5 wt.%, 74.4 wt.% and 51.2 wt.%, respectively, by petroleum ether.
the ketone. The extraction efficiencies using these two solvents

Please cite this article in press as: S. Ren, et al., Separation of chemical groups from bio-oil water-extract via sequential organic solvent
extraction, J. Anal. Appl. Pyrol. (2016), http://dx.doi.org/10.1016/j.jaap.2017.01.004
G Model
JAAP-3928; No. of Pages 10 ARTICLE IN PRESS
6 S. Ren et al. / Journal of Analytical and Applied Pyrolysis xxx (2016) xxx–xxx

Fig. 4. Extraction efficiencies for levoglucosan, acetic acid, and propionic acid in BOAP I by different solvents and S/F ratios.

Fig. 5. Extraction efficiencies for alcohols, ketones, and furans in BOAP I by different solvents and S/F ratios.

Please cite this article in press as: S. Ren, et al., Separation of chemical groups from bio-oil water-extract via sequential organic solvent
extraction, J. Anal. Appl. Pyrol. (2016), http://dx.doi.org/10.1016/j.jaap.2017.01.004
G Model
JAAP-3928; No. of Pages 10 ARTICLE IN PRESS
S. Ren et al. / Journal of Analytical and Applied Pyrolysis xxx (2016) xxx–xxx 7

Fig. 6. Extraction efficiencies for phenolics in BOAP I by different solvents and S/F ratios.

When the S/F ratio increased to 3, the changes of 3-ethylphenol, To further understand the effects of solvent on the chemical
guaiacol, and 2,6-dimethoxyphenol were insignificant. Hexane extraction, the distribution coefficients of six chemical groups were
showed a poor extraction performance for 1,2-benzenediol and summarized in Fig. 7. The distribution coefficient of furans, alco-
phenol; the highest extraction efficiencies for these two chem- hols, or ketones in hexane and petroleum ether was less than 0.6,
icals were 7 wt.% and 20 wt.%, respectively, when the S/F ratio much lower than that in chloroform or ethyl acetate. Due to the
was increased to 3. Petroleum ether had a slightly better per- large difference in polarity between the solvent and the chemicals
formance for 1,2-benzenediol and phenol than hexane. But the to be extracted, the distribution coefficients of levoglucosan and
highest extraction efficiency was observed at 12.6 wt.% for 1,2- acids in hexane and petroleum ether were very low, generally less
benzenediol and 35.5 wt.% for phenol. Compared to hexane and than 0.1 (Fig. 7). These results indicate that hexane and petroleum
petroleum ether, chloroform and ethyl acetate showed superior ether cannot be efficiently used to extract furans, alcohols, ketones,
extraction efficiency for phenolic compounds (Fig. 6). All guaiacol, anhydrosugars, and organic acids. The highest distribution coeffi-
2,6-dimethoxyphenol, and creosol were extracted to solvents even cient of alcohols and furans in chloroform was observed at 0.79
at the low S/F ratio of 0.5, suggesting that the two had very high and 4.49, respectively. Compared to those in chloroform, the dis-
affinity for the three compounds. Extraction efficiency for phenol tribution coefficients of alcohols and furans in ethyl acetate were
and 3-ethylphenol using chloroform was over 94 wt.% and 87 wt.% lower. Therefore, chloroform would be the preferred choice for
respectively, while the extraction efficiency for these two chemi- the extraction of alcohols and furans from the bio-oil aqueous
cals using ethyl acetate was at 100 wt.% and 90 wt.%, respectively. phase. Levoglucosan and organic acids such as acetic acid can form
Chloroform had low affinity for 1,2-benzenediol as indicated by the hydrogen-bonding with water in the bio-oil aqueous phase [6,43].
extraction efficiency of only about 36 wt.%. However, ethyl acetate Although chloroform has a high polarity, it barely breaks the hydro-
showed a high affinity for 1,2-benzenediol resulting in an extrac- gen bond in extraction. However, in the case of ethyl acetate with
tion efficiency of over 96 wt.% at a S/F ratio of 2. a slightly higher polarity, when the S/F ratio was increased from
According to the above analyses, the extraction efficiency by the 0.5 to 2, the hydrogen bond between acids and water could be
organic solvents for chemicals in BOAP I was not only affected by the broken, thereby increased the distribution coefficient of acids in
different solvents and S/F ratios, but also subjective to the different ethyl acetate from 0.01 to 0.67 (Fig. 7). Therefore, ethyl acetate at
chemical groups. Generally, chloroform had high extraction effi- an S/F ratio of 2 would be suitable for extracting acetic acid from
ciency for furans, alcohols, and phenolics, while ethyl acetate had BOAP I. The distribution coefficients of phenolics in chloroform
high extraction efficiency for organic acids in addition to furans and were greater than 1.5 at the S/F ratios of 0.5 and 1, and they were
phenolics, indicating non-selective extraction by ethyl acetate. The greater than 10 at all the investigated S/F ratios in ethyl acetate
optimum S/F ratio and extraction efficiency by the four solvents for (Fig. 7), much higher than those in hexane and petroleum ether,
the 15 quantified chemicals are summarized in Table 3. indicating that both chloroform and ethyl acetate are suitable for
extracting phenolics from the bio-oil aqueous phase.

Please cite this article in press as: S. Ren, et al., Separation of chemical groups from bio-oil water-extract via sequential organic solvent
extraction, J. Anal. Appl. Pyrol. (2016), http://dx.doi.org/10.1016/j.jaap.2017.01.004
G Model
JAAP-3928; No. of Pages 10 ARTICLE IN PRESS
8 S. Ren et al. / Journal of Analytical and Applied Pyrolysis xxx (2016) xxx–xxx

Table 3
Extraction efficiency of different solvents for 15 quantified chemicals at the optimum conditions. Data reported are average numbers based on three replicates.

Major compounds Extraction efficiency (wt.%)

Hexane (2:1)a Petroleum ether (2:1) Chloroform (1:1) Ethyl acetate (2:1)

Group 1 Furans Furfural 56.8 43.9 98.1 97.3

2(5H)-Furanone 47.3 59.4 65.8 93.1
Hydroxymethylfurfural 1.8 5.9 69.6 60.8
Alcohols 1-Hydroxy-2-butanone 23.3 16.5 48.1 50.4
1,3-Propanediol 7.7 17.9 18.6 35.0
Ketones 3-Methyl-1,2-cyclopentanedione 31.7 13.5 81.0 90.1
Phenolics 1,2-Benzenediol 5.1 12.6 21.1 96.7
Phenol 9.1 32.6 79.8 100.0
3-Ethylphenol 73.5 68.5 71.8 90.3
Guaiacol 80.7 74.4 100.0 100.0
2,6-Dimethoxyphenol 36.1 51.2 100.0 100.0
Creosol 100.0 100.0 100.0 100.0
Group 2 Acids Acetic acid 1.3 9.0 14.0 55.0
Propionic acid 0.4 1.2 3.9 23.3
Group 3 Anhydro- Levoglucosan 7.1 5.9 8.8 7.6
sugars
a
The optimum volume ratio of solvent-to-feed.

Fig. 7. Distribution coefficients for chemical groups from BOAP I in different solvents. Distribution coefficients reported are average numbers based on three replicates.

Based on the above analysis, we further developed a sequen- before and after extraction, and the extracted organics are pre-
tial extraction method to separate chemical groups (Fig. 1). First by sented in Table 2 and Fig. 8. In this study, about 53.9 wt.% of
using chloroform, we can separate furans, alcohols, ketones, and total chemicals in BOAP I was quantified. A number of chemi-
phenolics together (Group 1). Then by using ethyl acetate, we can cals (46.1 wt.% of total chemicals) with very low concentration
separate organic acids (Groups 2). Finally, we can recover anhy- listed in alcohol, ketones, aldehydes, etc., in BOAP I were not
drosugars (Group 3) from the final aqueous phase of BOAP III. This quantified. About 62.2 wt.% of the total of furans, ketones, alco-
sequential extraction was investigated and the results are reported hols, and phenolics and 85 wt.% unquantified chemicals in BOAP
in the following section. I were extracted to chloroform. After chloroform extraction, BOAP
II was mainly composed of organic acids and levoglucosan, which
3.4. Sequential extraction by chloroform and ethyl acetate accounted for about 43.4 wt.% and 36.1 wt.%, respectively (Fig. 8).
Other components were about 2.0 wt.% furans, 3.0 wt.% alcohols,
After chloroform extraction, the number of chromatography 1.1 wt.% phenolics, 0.56 wt.% ketones, and 13.8 wt.% unquantified
peaks of BOAP II was reduced and major peaks observed were lev- chemicals. The sequential extraction by chloroform followed by
oglucosan, acetic acid, propionic acid, and 1,2-benzenediol. After ethyl acetate well recovered furans, ketones, and phenolics from
ethyl acetate extraction of BOAP II, the major peak in BOAP III was aqueous phase, leaving behind no ketones and only about 0.7 wt.%
only levoglucosan. The chemical compositions of aqueous phase furans, and 0.04 wt.% phenolics in BOAP III after extraction (Fig. 8).

Please cite this article in press as: S. Ren, et al., Separation of chemical groups from bio-oil water-extract via sequential organic solvent
extraction, J. Anal. Appl. Pyrol. (2016), http://dx.doi.org/10.1016/j.jaap.2017.01.004
G Model
JAAP-3928; No. of Pages 10 ARTICLE IN PRESS
S. Ren et al. / Journal of Analytical and Applied Pyrolysis xxx (2016) xxx–xxx 9

Pyoungchung Kim and Nicole Labbe at the University of Tennessee

Center of Renewable Carbon for help in the production of bio-oil.

References

[1] E.d. Jong, A. Higson, P. Walsh, M. Wellisch, Bio-based chemicals: value added
products from biorefineries, in: IEA Bioenergy/Task 42 Biorefinery, 2012,
http://www.ieabioenergy.com/publications/bio-based-chemicals-value-
added-products-from-biorefineries/.
[2] C.B. Rasrendra, B. Girisuta, H.H. van de Bovenkamp, J.G.M. Winkelman, E.J.
Leijenhorst, R.H. Venderbosch, M. Windt, D. Meier, H.J. Heeres, Recovery of
acetic acid from an aqueous pyrolysis oil phase by reactive extraction using
tri-n-octylamine, Chem. Eng. J. 176–177 (2011) 244–252.
[3] Q. Zhang, J. Chang, T.J. Wang, Y. Xu, Review of biomass pyrolysis oil properties
and upgrading research, Energy Convers. Manage. 48 (2007) 87–92.
Fig. 8. Chemical compositions of aqueous phase before and after sequential extrac- [4] C.A. Mullen, G.D. Strahan, A.A. Boateng, Characterization of various
tion and the extracted organics in solvents. fast-pyrolysis bio-oils by NMR spectroscopy, Energy Fuel 23 (2009)
2707–2718.
[5] D. Mohan, C.U. Pittman, P.H. Steele, Pyrolysis of wood/biomass for bio-oil: a
critical review, Energy Fuel 20 (2006) 848–889.
The sequential extraction also concentrated levoglucosan result- [6] C.R. Vitasari, G.W. Meindersma, A.B. de Haan, Water extraction of pyrolysis
ing in about 53 wt.% levoglucosan in BOAP III. However, we also oil: the first step for the recovery of renewable chemicals, Bioresour. Technol.
102 (2011) 7204–7210.
observed that BOAP III contained about 32.5 wt.% organic acids. In [7] Y. Wei, H.W. Lei, L. Wang, L. Zhu, X.S. Zhang, Y.P. Liu, S.L. Chen, B. Ahring,
this study, BOAP II was extracted only once by ethyl acetate. Multi- Liquid-liquid extraction of biomass pyrolysis bio-oil, Energy Fuel 28 (2014)
ple extractions using ethyl acetate may be implemented to extract 1207–1212.
[8] R. Xu, L. Ferrante, C. Briens, F. Berruti, Bio-oil production by flash pyrolysis of
larger fraction of acids from BOAP II. The organics extracted by chlo- sugarcane residues and post treatments of the aqueous phase, J. Anal. Appl.
roform were mainly furans, ketones, alcohols, phenolics. Due to Pyrol. 91 (2011) 263–272.
the very low concentration of these compounds, it will be difficult [9] L.K.E. Park, S.J. Ren, S. Yiacoumi, X.P. Ye, A.P. Borole, C. Tsouris, Separation of
switchgrass bio-oil by water/organic solvent addition and pH adjustment,
to further purify them individually. The preferred application of Energy Fuel 30 (2016) 2164–2173.
these compounds could be as a feedstock for the biorefinery pro- [10] H.Y. Li, Q.L. Xu, H.S. Xue, Y.J. Yan, Catalytic reforming of the aqueous phase
cess to generate hydrocarbons fuels. The organics extracted by ethyl derived from fast-pyrolysis of biomass, Renew. Energy 34 (2009) 2872–2877.
[11] P.N. Kechagiopoulos, S.S. Voutetakis, A.A. Lemonidou, I.A. Vasalos, Hydrogen
acetate contains over 57.6 wt.% organic acids, in which acetic acid production via reforming of the aqueous phase of bio-oil over Ni/olivine
accounts for about 75 wt.%. It will be costly to produce glacial acetic catalysts in a spouted bed reactor, Ind. Eng. Chem. Res. 48 (2009) 1400–1408.
acid from this stream. Therefore, these organics could serve as a [12] P.N. Kechagiopoulos, S.S. Voutetakis, A.A. Lemonidou, I.A. Vasalos, Hydrogen
production via steam reforming of the aqueous phase of bio-oil in a fixed bed
source for hydrogen production using microbial electrolysis [16].
reactor, Energy Fuel 20 (2006) 2155–2163.
[13] F.H. Mahfud, F.P. van Geel, R.H. Venderbosch, H.J. Heeres, Acetic acid recovery
from fast pyrolysis oil. An exploratory study on liquid–liquid reactive
4. Conclusions extraction using aliphatic tertiary amines, Sep. Sci. Technol. 43 (2008)
3056–3074.
[14] Y. Feng, D. Meier, Extraction of value-added chemicals from pyrolysis liquids
Four organic solvents were evaluated for chemical extraction with supercritical carbon dioxide, J. Anal. Appl. Pyrol. 113 (2015) 174–185.
from bio-oil aqueous phase. The amount of chemicals extracted by [15] J.N. Lian, M. Garcia-Perez, S.L. Chen, Fermentation of levoglucosan with
these four solvents from bio-oil aqueous phase was in the order oleaginous yeasts for lipid production, Bioresour. Technol. 133 (2013)
183–189.
of hexane < petroleum < chloroform < ether ethyl acetate at the [16] A.J. Lewis, S. Ren, X. Ye, P. Kim, N. Labbe, A.P. Borole, Hydrogen production
same S/F ratio. Further, chloroform had similar extraction efficiency from switchgrass via a hybrid pyrolysis-microbial electrolysis process,
with ethyl acetate for furans, alcohols, ketones and phenolics. Ethyl Bioresour. Technol. 195 (2015) 231–241.
[17] A. Thygesen, F.W. Poulsen, I. Angelidaki, B. Min, A.B. Bjerre, Electricity
acetate also had high extraction efficiency for acetic acid when the
generation by microbial fuel cells fuelled with wheat straw hydrolysate,
S/F ratio was at 2. According to the distribution coefficients and Biomass Bioenergy 35 (2011) 4732–4739.
extraction efficiencies obtained for the different chemical groups, [18] Z.Y. Chi, M. Rover, E. Jun, M. Deaton, P. Johnston, R.C. Brown, Z.Y. Wen, L.R.
a sequential extraction using chloroform followed by ethyl acetate Jarboe, Overliming detoxification of pyrolytic sugar syrup for direct
fermentation of levoglucosan to ethanol, Bioresour. Technol. 150 (2013)
is recommended. The first step of extraction using chloroform sep- 220–227.
arated most of the furans, ketones, alcohols, and phenolics from [19] J.A. Capunitan, S.C. Capareda, Characterization and separation of corn stover
bio-oil aqueous phase. The second step of extraction employing bio-oil by fractional distillation, Fuel 112 (2013) 60–73.
[20] Y. Elkasabi, C.A. Mullen, A.A. Boateng, Distillation and isolation of commodity
ethyl acetate concentrated organic acids (over 62 wt.% acetic acid in chemicals from bio-oil made by tail-gas reactive pyrolysis, Acs. Sustain Chem.
BOAP I). In the final aqueous phase, levoglucosan was concentrated Eng. 2 (2014) 2042–2052.
at 53 wt.%. These results suggest that sequential extraction with [21] F.X. Zeng, W.J. Liu, H. Jiang, H.Q. Yu, R.J. Zeng, Q.X. Guo, Separation of
phthalate esters from bio-oil derived from rice husk by a
different solvents can be used to separate chemical groups in bio- basification-acidification process and column chromatography, Bioresour.
oil aqueous phase and these extracted chemical groups could serve Technol. 102 (2011) 1982–1987.
as feedstocks for the specific production of chemicals, hydrogen, or [22] J.S. Kim, Production, separation and applications of phenolic-rich bio-oil – a
review, Bioresour. Technol. 178 (2015) 90–98.
hydrocarbons. [23] L.F. Zilnik, A. Jazbinsek, Recovery of renewable phenolic fraction from
pyrolysis oil, Sep. Purif. Technol. 86 (2012) 157–170.
[24] M. Garcia-Perez, A. Chaala, H. Pakdel, D. Kretschmer, C. Roy, Characterization
Acknowledgements of bio-oils in chemical families, Biomass Bioenergy 31 (2007) 222–242.
[25] P. Kim, A. Johnson, C.W. Edmunds, M. Radosevich, F. Vogt, T.G. Rials, N. Labbe,
Surface functionality and carbon structures in lignocellulosic-derived
We acknowledge funding for this work from the U.S. Depart- biochars produced by fast pyrolysis, Energ. Fuel 25 (2011) 4693–4703.
ment of Energy, BioEnergy Technologies Office under the Carbon, [26] S.J. Ren, X.P. Ye, A.P. Borole, P. Kim, N. Labbe, Analysis of switchgrass-derived
Hydrogen and Separations Efficiency (CHASE) in Bio-Oil Conversion bio-oil and associated aqueous phase generated in a semi-pilot scale auger
pyrolyzer, J. Anal. Appl. Pyrol. 119 (2016) 97–103.
Pathways program, DE-FOA-0000812. The manuscript is coau- [27] ASTM D1217-12, Standard test method for density and relative density
thored by UT-Battelle, LLC, under Contract DEAC05-00OR22725 (specific gravity) of liquids by bingham pycnometer, ASTM, International,
with the U.S. Department of Energy. The authors also thank Drs. West Conshohocken, PA, 2012, www.astm.org.

Please cite this article in press as: S. Ren, et al., Separation of chemical groups from bio-oil water-extract via sequential organic solvent
extraction, J. Anal. Appl. Pyrol. (2016), http://dx.doi.org/10.1016/j.jaap.2017.01.004
G Model
JAAP-3928; No. of Pages 10 ARTICLE IN PRESS
10 S. Ren et al. / Journal of Analytical and Applied Pyrolysis xxx (2016) xxx–xxx

[28] ASTM D4377-00(2011), Standard test method for water in crude oils by [36] C.A. Mullen, A.A. Boateng, Chemical composition of bio-oils produced by fast
potentiometric Karl Fischer titration, ASTM International, West pyrolysis of two energy crops, Energy Fuel 22 (2008) 2104–2109.
Conshohocken, PA, 2011, www.astm.org. [37] S. Thangalazhy-Gopakumar, S. Adhikari, R.B. Gupta, S.D. Fernando, Influence
[29] ASTM D445-12, Standard test method for kinematic viscosity of transparent of pyrolysis operating conditions on bio-oil components: a microscale study
and opaque liquids (and calculation of dynamic viscosity), ASTM in a pyroprobe, Energy Fuel 25 (2011) 1191–1199.
International, West Conshohocken, PA, 2012, www.astm.org. [38] V.L. Budarin, J.H. Clark, B.A. Lanigan, P. Shuttleworth, S.W. Breeden, A.J.
[30] ASTM D482-13, Standard test method for ash from petroleum products, ASTM Wilson, D.J. Macquarrie, K. Milkowski, J. Jones, T. Bridgeman, A. Ross, The
International, West Conshohocken, PA, 2013, www.astm.org. preparation of high-grade bio-oils through the controlled, low temperature
[31] M.E. Boucher, A. Chaala, C. Roy, Bio-oils obtained by vacuum pyrolysis of microwave activation of wheat straw, Bioresour. Technol. 100 (2009)
softwood bark as a liquid fuel for gas turbines. Part I: properties of bio-oil and 6064–6068.
its blends with methanol and a pyrolytic aqueous phase, Biomass Bioenergy [39] F. Ates, M.A. Isikdag, Evaluation of the role of the pyrolysis temperature in
19 (2000) 337–350. straw biomass samples and characterization of the oils by GUMS, Energy Fuel
[32] ASTM D664-11a, Standard test method for acid number of petroleum 22 (2008) 1936–1943.
products by potentiometric titration, ASTM, International, West [40] P.G. Duan, P.E. Savage, Catalytic treatment of crude algal bio-oil in
Conshohocken, PA, 2011, www.astm.org. supercritical water: optimization studies, Energy Environ. Sci. 4 (2011)
[33] S.H. Ha, N.L. Mai, Y.M. Koo, Butanol recovery from aqueous solution into ionic 1447–1456.
liquids by liquid–liquid extraction, Process Biochem. 45 (2010) 1899–1903. [41] M.C. Samolada, A. Papafotica, I.A. Vasalos, Catalyst evaluation for catalytic
[34] R.H. He, X.P. Ye, B.C. English, J.A. Satrio, Influence of pyrolysis condition on biomass pyrolysis, Energy Fuel 14 (2000) 1161–1167.
switchgrass bio-oil yield and physicochemical properties, Bioresour. Technol. [42] J. Mortimer, R.W. Taft Kamlet, The solvatochromic comparison method. I. The
100 (2009) 5305–5311. ␤-scale of solvent hydrogen-bond acceptor (HBA) basicities, J. Am. Chem. Soc.
[35] T. Imam, S. Capareda, Characterization of bio-oil, syn-gas and bio-char from 98 (1976) 377–383.
switchgrass pyrolysis at various temperatures, J. Anal. Appl. Pyrol. 93 (2012) [43] L. Pu, Y.M. Sun, Z.B. Zhang, Hydrogen bonding of single acetic acid with water
170–177. molecules in dilute aqueous solutions, Sci. China Ser. B 52 (2009) 2219–2225.

Please cite this article in press as: S. Ren, et al., Separation of chemical groups from bio-oil water-extract via sequential organic solvent
extraction, J. Anal. Appl. Pyrol. (2016), http://dx.doi.org/10.1016/j.jaap.2017.01.004