A kinetic model for production of glucose by hydrolysis of levoglucosan and cellobiosan from pyrolysis oil

Article abstract of DOI:10.1016/j.carres.2007.07.016

Anhydro sugars, produced during wood pyrolysis, can by hydrolyzed to sugars under acidic conditions. The acid hydrolysis of two common anhydro sugars in wood pyrolysis oils, levoglucosan (1,6-anhydro-β-d-glucopyranose) and cellobiosan (β-d-glucopyranosyl-(1→4)-1,6-anhydro-d-glucopyranose), was investigated. Levoglucosan hydrolysis to glucose follows a first-order reaction, with an activation energy of 114 kJ mol^-1. For cellobiosan hydrolysis, 44% of the cellobiosan is hydrolyzed initially via the β-(1→4) glycosidic bond to form levoglucosan and glucose. The remaining cellobiosan is hydrolyzed initially at the 1,6 anhydro bond to form cellobiose. Both reactions are first order with respect to cellobiosan, with an activation energy of 99 kJ mol^-1. The intermediate levoglucosan and cellobiose are hydrolyzed to glucose.

Full text of DOI:10.1016/j.carres.2007.07.016

Carbohydrate Research 342 (2007) 2365–2370
A kinetic model for production of glucose by hydrolysis
of levoglucosan and cellobiosan from pyrolysis oil
Steve Helle,a,* Nicole M. Bennett,b Karen Lau,b Justin H. Matsuib and Sheldon J. B. Duffb
aEnvironmental Science and Environmental Engineering, University of Northern British Columbia,
3333 University Way, Prince George, BC, Canada V2N 4Z9
bDepartment of Chemical and Biological Engineering, The University of British Columbia, 2360 East Mall,
Vancouver, BC, Canada V6T 1Z3
Received 16 April 2007; received in revised form 18 July 2007; accepted 24 July 2007
Available online 2 August 2007
Abstract—Anhydro sugars, produced during wood pyrolysis, can by hydrolyzed to sugars under acidic conditions. The acid hydro-
lysis of two common anhydro sugars in wood pyrolysis oils, levoglucosan (1,6-anhydro-b-D-glucopyranose) and cellobiosan (b-D-
glucopyranosyl-(1!4)-1,6-anhydro-D-glucopyranose), was investigated. Levoglucosan hydrolysis to glucose follows a first-order
reaction, with an activation energy of 114 kJ molꢀ1. For cellobiosan hydrolysis, 44% of the cellobiosan is hydrolyzed initially via
the bꢀ(1!4) glycosidic bond to form levoglucosan and glucose. The remaining cellobiosan is hydrolyzed initially at the 1,6 anhydro
bond to form cellobiose. Both reactions are first order with respect to cellobiosan, with an activation energy of 99 kJ molꢀ1. The
intermediate levoglucosan and cellobiose are hydrolyzed to glucose.
ꢀ 2007 Elsevier Ltd. All rights reserved.
Keywords: Levoglucosan; Cellobiosan; Glucose; Hydrolysis; Pyrolysis oil
1. Introduction
bution of pyrolysis products depends on the reaction
conditions, with maximum yields of pyrolysis oil
obtained around 500 ꢁC.5 A major product in the pyro-
lysis oil is levoglucosan, formed from cellulose.6,7 Under
many pyrolysis conditions, a significant amount of cello-
biosan is also produced, with yields as high as 30% of
the levoglucosan yield.8
The pyrolysis of biomass provides an alternate method
to acid or enzymatic hydrolysis to overcome one of the
major challenges of producing renewable chemicals from
biomass, namely the extraction and conversion of cellu-
lose to glucose. The anhydro sugars formed during pyro-
lysis, for example, levoglucosan and cellobiosan, may be
converted to sugar by dilute acid hydrolysis.9,10 Previous
work has shown that the addition of sulfuric acid to pyro-
lysis oil generated more glucose than could be accounted
for by the amount of levoglucosan present.9 While this
extra glucose was attributed to unknown carbohydrate
oilgomers such as cellobiosan and other sugars, little is
known about the identity, relative proportions, and
hydrolysis kinetics of these compounds.
Levoglucosan (1,6-anhydro-b-D-glucopyranose) and
cellobiosan (b-D-glucopyranosyl-(1!4)-1,6-anhydro-D-
glucopyranose or 1,6-anhydro-b-cellobiose) are among
the organic compounds formed from the burning of
biomass, along with the other derivatives from the
breakdown of cellulose, aliphatic and oxygenated com-
pounds, and terpenoids from vegetation biopolymers.1
Levoglucosan and cellobiosan can be used as tracer
compounds to determine smoke distribution.2,3 Levo-
glucosan does not undergo acid-catalyzed hydrolysis un-
der typical atmospheric conditions.4
Levoglucosan and cellobiosan are also present in bio-
mass pyrolysis oils, which are produced from the degra-
dation of biomass materials at high temperatures under
nonoxidative conditions. Pyrolysis forms many prod-
ucts, broadly classified as char, oil, and gas.5 The distri-
*
Corresponding author. E-mail: helle@unbc.ca
0008-6215/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carres.2007.07.016

2366
S. Helle et al. / Carbohydrate Research 342 (2007) 2365–2370
dA
dt
dD
dt
In this study, we have investigated the rates of levo-
¼ ꢀk1A
¼ k1A
ð2Þ
ð3Þ
glucosan and cellobiosan hydrolysis in water, sulfite pulp
mill pulping liquor, and in pyrolysis oil extract. The
effects of acid concentration and temperature on hydro-
lysis rates were measured. We present a kinetic model
for the production of fermentable sugars from the acid
hydrolysis of pyrolysis extracts.
where A and D are the molar concentrations of levo-
glucosan and glucose, respectively, and k1 is the first-
order rate constant for levoglucosan hydrolysis. If the
initial glucose and levoglucosan concentrations are zero
and A0, respectively, then Eqs. 1 and 2 integrate to the
following:
2. Results and discussion
2.1. Levoglucosan hydrolysis kinetics
A
A0
D
¼ eꢀk t
ð4Þ
ð5Þ
1
Levoglucosan hydrolysis is known to follow first-order
kinetics.10
¼ ð1 ꢀ eꢀk t
Þ
1
A0
Levoglucosan ðAÞ ! Glucose ðDÞ
ð1Þ
k1
Sample data for levoglucosan hydrolysis in water are
presented in Figures 1 and 2. Figure 1 demonstrates
the effect of temperature on the hydrolysis rate of levo-
glucosan in dilute sulfuric acid, with the acid concentra-
Assuming first-order kinetics, the rate of disappearance
of levoglucosan and the rate of production of glucose
are given in Eqs. 2 and 3,
10
8
10
8
6
6
4
4
2
2
0
0
100
0
20
40
60
80
0
20
40
60
80
Time (minutes)
Time (minutes)
Figure 1. Effect of temperature on levoglucosan kinetics. Hydrolysis at 90 ꢁC (j), 100 ꢁC (m), and 110 ꢁC (d). All hydrolyses at 500 mM sulfuric
acid.
10
8
10
8
6
6
4
4
2
2
0
0
100
0
20
40
60
80
0
20
40
60
80
Time (minutes)
Time (minutes)
Figure 2. Effect of sulfuric acid concentration on levoglucosan kinetics. Hydrolysis at 120 mM H2SO4 (j), 250 mM H2SO4 (m), and 500 mM H2SO4
(d). All hydrolyses at 110 ꢁC.

S. Helle et al. / Carbohydrate Research 342 (2007) 2365–2370
2367
tion held constant. For each hydrolysis condition, the
first-order rate constant was determined by simulta-
neous nonlinear regression of the levoglucosan and
glucose concentration data using Eqs. 4 and 5. The
resulting first-order model curve fits are shown in Fig-
ures 1 and 2.
and in SSL/extract. The rate constants for hydrolysis at
500 and 250 mM H2SO4 are presented in Figure 3, along
with the Arrhenius model (Eq. 6). The activation energy
and the exponent n obtained in this study were similar to
the values of 97 kJ molꢀ1 and 1 found using hydrochloric
acid and lower temperatures (25–50 ꢁC).10
Increasing the temperature by 10 ꢁC increments re-
sulted in close to twofold increases in the hydrolysis rate
(Fig. 1). Over the temperature range studied, the reac-
tion rate increased exponentially with increasing temper-
ature, following an Arrhenius relationship. Increasing
the sulfuric acid concentration also increased the levo-
glucosan hydrolysis rate. For example, with the temper-
ature held constant at 110 ꢁC, increasing the acid
concentration from 250 mM to 500 mM had approxi-
mately the same effect as increasing the temperature
from 100 ꢁC to 110 ꢁC with sulfuric acid concentration
held constant at 500 mM. Increasing the acid concentra-
tion from 110 mM to 500 mM had approximately the
same effect as increasing the temperature from 90 ꢁC
to 110 ꢁC. Over the ranges tested, a 10 ꢁC change in
temperature had approximately the same effect as a
doubling (or halving) of the acid concentration. The
following relationship was used to relate the rate con-
stant, k1 (sꢀ1), to temperature and acid concentration:
First-order rate constants were calculated from a
number of levoglucosan hydrolysis experiments per-
formed in spent sulfite pulping liquor, water extract of
pyrolysis oil, and spent sulfite pulping liquor extract of
pyrolysis oil, with and without acid addition. Spent
sulfite liquor contains the hemicellulose sugars and the
lignin that are removed from the wood during the sulfite
pulping process. Several sulfite pulp mills ferment the
spent sulfite liquor to produce ethanol, although the
process is limited by the low concentration of sugars.
Spent sulfite liquor was used to extract levoglucosan
from pyrolysis oil with the goal of increasing the sugar
concentration in the pulping liquor prior to fermenta-
tion. When spent sulfite liquor (or water) is mixed with
pyrolysis oil, the levoglucosan and other water-soluble
compounds in the pyrolysis oil transfer to the spent
sulfite liquor (or water), leaving a residue of the hydro-
phobic lignin compounds. For a given sulfuric acid
concentration, the levoglucosan hydrolysis rate was
greater in spent sulfite liquor (pure levoglucosan added
to spent sulfite liquor), spent sulfite liquor pyrolysis oil
extract, or water pyrolysis oil extract than the levogluco-
san hydrolysis rate in dilute sulfuric acid. For hydrolysis
in spent sulfite liquor or pyrolysis oil extract, the activa-
tion energy was found to be 87 kJ molꢀ1, which is signi-
ficantly lower than the activation energy for hydrolysis in
dilute sulfuric acid. In addition, acid concentration had
less of an effect on the hydrolysis rate in pyrolysis oil
extract compared to hydrolysis in water (n was 0.67 for
hydrolysis in pyrolysis oil extract compared to 1 for
hydrolysis in water). The parameter A was 3.3 · 106.
The difference between hydrolysis in sulfuric acid/dis-
tilled water and sulfuric acid/(pyrolysis oil extract or
spent sulfite liquor) was greatest at low temperatures
and high acid concentrations.
n
k1 ¼ ½H2SO4ꢁ AeꢀE =RT
ð6Þ
1
where [H2SO4] is the acid concentration in mmol Lꢀ1
,
n and A are constants, R is the universal gas constant
(8.314 J molꢀ1 Kꢀ1), E1 is the activation energy (in
J molꢀ1), and T is the temperature (in Kelvin). For levo-
glucosan hydrolysis in dilute sulfuric acid, the activa-
tion energy was 114 kJ molꢀ1 and A was 1.0 · 1010.
When the hydrolyses were carried out in aqueous matri-
ces, the hydrolysis rate increased approximately linearly
with increasing acid concentration (n = 1). Slight devia-
tions from linearity were observed for hydrolysis in SSL
1
0.1
2.2. Cellobiose hydrolysis kinetics
Cellobiose is not normally listed as a component of
pyrolysis oils, but it is a major intermediate for cello-
biosan hydrolysis (see below). Cellobiose hydrolysis
follows a first-order reaction similar to levoglucosan
hydrolysis:11
0.01
0.001
Cellobiose ðCÞ ! Glucose ðDÞ
ð7Þ
ð8Þ
ð9Þ
0.0025
0.0027
0.0029
0.0031
k2
1/T (1/K)
dC
dt
¼ ꢀk2C
Figure 3. First-order levoglucosan hydrolysis rate constant at different
sulfuric acid concentrations, and in different media. (d) 250 mM
H2SO4 in water; (j) 500 mM H2SO4 in water; (s) 500 mM H2SO4 in
SSL and/or pyrolysis oil extract.
dD
¼ 2k2C
dt

2368
S. Helle et al. / Carbohydrate Research 342 (2007) 2365–2370
where C is the cellobiose concentration (mol Lꢀ1) and k2
is the first-order cellobiose hydrolysis rate constant.
Integrating from an initial cellobiose concentration of
C0 and an initial glucose concentration of zero yields
the following:
where B is the cellobiosan concentration (mol Lꢀ1), k3 is
the rate coefficient for cellobiosan hydrolysis to cellobi-
ose and k4 is the rate coefficient for cellobiosan hydroly-
sis to glucose and levoglucosan. The first-order rate
coefficient for the removal of cellobiosan (k5) is equal
to k3 + k4. For the case with initial cellobiosan equal
to B0, and zero initial levoglucosan, cellobiose, and glu-
cose, Eqs. 12–15 may be integrated to yield the
following:
C
C0
D
¼ eꢀk t
ð10Þ
2
¼ 2ð1 ꢀ eꢀk t
Þ
ð11Þ
2
C0
These equations were found to describe cellobiose
hydrolysis well. At a given acid concentration and tem-
perature, cellobiose hydrolysis was slower than levo-
glucosan hydrolysis. For example, at 110 ꢁC and
500 mM sulfuric acid, k1 was 0.00135 sꢀ1, while k2 was
0.00062 sꢀ1. Sample data are presented in Figure 5b.
The activation energy for cellobiose hydrolysis was
found to be 103 kJ molꢀ1, similar to the previously
A
k4
1
3
4
¼
ðeꢀk t ꢀ eðꢀk ꢀk Þt
Þ
ð16Þ
ð17Þ
ð18Þ
B0 k3 þ k4 ꢀ k1
B
B0
C
¼ eðꢀk ꢀk Þt
3
4
k3
2
3
4
¼
ðeꢀk t ꢀ eðꢀk ꢀk Þt
Þ
B0 k3 þ k4 ꢀ k2
D
B0
ðk4 ꢀ k2Þðk4 ꢀ 2k1Þ þ k3ðk4 ꢀ 2k2Þ
eðꢀk ꢀk Þt
3
4
reported value of 110 kJ molꢀ1 11
.
¼ ꢀ
ðk3 þ k4 ꢀ k1Þðk3 þ k4 ꢀ k2Þ
k4
2k3
1
2
ꢀ
eꢀk t
ꢀ
eꢀk t þ 2 ð19Þ
2.3. Cellobiosan hydrolysis kinetics
k3 þ k4 ꢀ k1
k3 þ k4 ꢀ k2
As measured by ion chromatography using a pulsed
ampirometric detector, cellobiosan hydrolysis products
included cellobiose, levoglucosan, and glucose. The
mass balance closes within the error of analysis, indicat-
ing that there are no other major products formed. The
proposed reaction scheme is presented in Figure 4. The
initial hydrolysis reaction may be either the hydrolysis of
the b-(1!4) glycosidic bond, producing levoglucosan
and glucose, or the hydrolysis of the 1,6-anhydro bond
to produce cellobiose. Cellobiose and levoglucosan are
subsequently hydrolyzed to glucose (Eqs. 1 and 7).
With each hydrolysis step modeled as a first-order
reaction, the rate equations developed from the pathway
shown in Figure 4 are as follows:
Sample hydrolysis data for cellobiosan are presented in
Figure 5c, along with hydrolysis of levoglucosan
(Fig. 5a) and cellobiose (Fig. 5b) under the same condi-
tions. For determining the model parameters, k5 was
calculated from the change in cellobiosan concentration
with time, then k3 and k4 were calculated by simulta-
neous nonlinear regression of Eqs. 16–19 to the cellobio-
san, glucose, levoglucosan, and cellobiose data using
previously determined values of k1 and k2.
For the hydrolysis experiment presented in Figure 5,
k1 = 0.000539 sꢀ1, k2 = 0.000218 sꢀ1, k3 = 0.000183 sꢀ1
,
k4 = 0.000158 sꢀ1, and k5 = 0.000342 sꢀ1. The levogluco-
san hydrolysis rate was two-and-a-half times greater than
the cellobiose hydrolysis rate and one-and-a-half times
greater than the cellobiosan hydrolysis rate. For cellobio-
san, hydrolysis to cellobiose proceeded at a greater rate
than hydrolysis to levoglucosan and glucose. The greater
rate of cellobiose formation, coupled with the slower cel-
lobiose hydrolysis reaction, resulted in a greater accumu-
lation of cellobiose than levoglucosan. The amount of
cellobiosan hydrolyzed through cellobiose can be calcu-
lated from the ratio of k3 to k4, which is equal to 1.2. This
dA
¼ ꢀk1A þ k4B
ð12Þ
ð13Þ
ð14Þ
ð15Þ
dt
dB
dt
dC
dt
dD
dt
¼ ꢀk3B ꢀ k4B ¼ ꢀðk3 þ k4ÞB ¼ ꢀk5B
¼ k3B ꢀ k2C
¼ k1A þ k4B þ 2k2C
OH
O
O
O
HO
HO
HO
HO
k4
k1
OH
OH
OH
OH
OH
2
4
O
OH
O
HO
HO
HO
Degradation
products
HO
O
OH
HO
OH
O
OH
OH
O
1
2
O
k3
k2
OH
O
HO
HO
HO
OH
O
OH
OH
3
Figure 4. Cellobiosan (1) hydrolysis scheme. Cellobiose (2), glucose (3), and levoglucosan (4) are formed by acid hydrolysis.

S. Helle et al. / Carbohydrate Research 342 (2007) 2365–2370
2369
t
10
8
Dimensionless time: s ¼
ð20Þ
a
b
c
1=ðk3 þ k4Þ
A
B0
B
B0
Dimensionless concentrations: a ¼
;
b ¼
;
6
4
C
B0
D
B0
c ¼
;
d ¼
ð21Þ
2
0
10
As demonstrated in Figure 6, the hydrolysis pattern was
similar over a range of hydrolysis conditions. This com-
bined data set can be used to obtain an average activa-
tion energy and exponent for the effect of acid
concentration. Using the combined data set, the activa-
8
6
4
2
tion energy for cellobiosan hydrolysis is 99 kJ molꢀ1
,
which is similar to the activation energy obtained for
k2, k3, and k4. Levoglucosan hydrolysis will not have a
great impact on the analysis due to its low concentra-
tions compared to the other compounds. When using
the entire data set, the effect of acid concentration was
smaller, with n = 1.1, compared to a value of 1.3 for cel-
lobiosan hydrolysis.
0
10
8
6
4
2
0
Acid hydrolysis of pyrolysis oil extracts often results
in a greater than theoretical glucose yield based on ini-
tial levoglucosan.9 The work presented here has several
implications. First, for a given acid concentration,
hydrolysis rates for anhydro sugars are significantly fas-
ter in pyrolysis oil extract and pulp mill wastewater
compared to the rates in dilute acid. Secondly, this work
has shown that other anhydro sugars that are typically
present in pyrolysis oils also hydrolyze to glucose under
typical dilute acid conditions, although at a slower rate
compared to levoglucosan. If acid hydrolysis conditions
are chosen based only on the levoglucosan concentra-
tion in the pyrolysis oil extract, the sugar yield will be
lower than otherwise possible. Thirdly, the effect of
pyrolysis conditions on the generation of cellobiosan
and other oilgomers should be studied if the end goal
is to maximize the production of fermentable glucose
from bio-oil.
0
10
20
30
40
50
60
70
80
90
100
Time (minutes)
Figure 5. (a) Levoglucosan hydrolysis. Lines represent Eqs. 4 and 5.
(b) Cellobiose hydrolysis. Lines represent Eqs. 10 and 11. (c)
Cellobiosan hydrolysis. Lines represent Eqs. 16–19. T = 110 ꢁC,
[H2SO4] = 250 mM.
indicates that 1.2 mol of cellobiose are formed for every
mole of levoglucosan formed, or, 54% of the cellobiosan
is hydrolyzed via cellobiose, and 46% via levoglucosan.
Eq. 6 was used to determine the relationship between
the first-order rate constants and the temperature and
acid concentration. The activation energies and the
exponent n for k1, k2, k3, and k4 are listed in Table 1.
The activation energies for k2, k3, and k4 are similar,
and lower than the activation energy for k1. Acid con-
centration had a greater effect on k2, k3, and k4 com-
pared to k1. On average the levoglucosan hydrolysis
rate was 55% greater than the cellobiosan hydrolysis
rate, and 140% greater than the cellobiose hydroly-
sis rate. The ratio of cellobiosan hydrolyzed via cellobi-
ose remained constant at approximately 55% for all of
the hydrolysis conditions tested.
1
0.8
Cellobiosan hydrolysis data, collected over a range of
acid concentrations (from 110 to 750 mM sulfuric acid)
and temperatures (from 80 to 120 ꢁC), were normalized
using Eqs. 20 and 21.
0.6
0.4
0.2
0
Cellobiosan
Cellobiose
Levoglucosan
Glucose
Table 1. Kinetic coefficients for cellobiosan hydrolysis
E (kJ molꢀ1
)
n
0
1
2
3
4
5
k1
k2
k3
k4
114
103
98
1.0
Dimensionless time (time*(k +k ))
3
4
1.35
1.30
1.26
Figure 6. Normalized cellobiosan hydrolysis data for [H2SO4] = 110–
750 mM and T = 80–120 ꢁC.
99

2370
S. Helle et al. / Carbohydrate Research 342 (2007) 2365–2370
3. Experimental
added postcolumn to maintain a pH of 10.4 in the detec-
tor. The injection volume of the sample was 25 lL and
the column temperature was a constant 35 ꢁC. The levo-
glucosan calibration curve was nonlinear, and a mini-
mum of four standards were run spanning the
concentration range used. The other calibration curves
(glucose, cellobiose, and cellobiosan) were linear. For
samples that contained pyrolysis oil or spent sulfite
liquor, levoglucosan, and glucose were measured as
O-peracetylated aldononitrile derivatives by gas
chromatography.12
Levoglucosan was obtained from Sigma–Aldrich. Cello-
biosan was from Sussex Research Laboratories, Inc.
(Ottawa, Ontario). Pyrolysis oil was from VTT Pro-
cesses, prepared from Scots Pine feedstock in August
2004. The pyrolysis oil was stored at a low pH
(pH ꢂ 2.5, as received) in glass containers at 4 ꢁC with
no light exposure. Softwood spent sulfite pulping liquor
(SSL) was obtained from Tembec Inc., Temiscaming,
Quebec. SSL (240 g/L solids) was shipped from Tembec
Inc., Temiscaming Qc., and stored at a low pH (pH ꢂ 2,
as collected) at 4 ꢁC.
The hydrolysis experiments were conducted in sealed
glass test tubes placed in heating blocks. For levogluco-
san, cellobiosan, and cellobiose hydrolysis, solutions
were prepared in distilled water or spent sulfite liquor.
Hydrolysis experiments were also performed using pyro-
lysis oil extract, prepared by adding pyrolysis oil to
water, mixing vigorously for 5 min, then centrifuging
to separate the aqueous extract from the residue. The
solutions to be hydrolyzed were preheated prior to the
addition of 72% sulfuric acid. After a set amount of time
(5–240 min), the test tubes were removed from the heat-
ing block and placed in cold water to stop the reaction.
A range of acid concentrations (0–0.5 M H2SO4) and
temperatures (50–130 ꢁC) were used. A minimum of five
test tubes were used for each hydrolysis condition (acid
concentration and temperature). All additions were
measured using an analytical balance.
For experiments in water, levoglucosan, glucose, cello-
biose, and cellobiosan were measured by ion chroma-
tography, with detection by a pulsed amperometric
detector. A Carbopac PA1 column was used, with a
Dionex HPLC system equipped with an AS 50 auto-
Sampler, GP50 gradient pump and ED50A electrochem-
ical detector. The mobile phase was distilled and
deioinized water (Milli-Q Ultra-pure water purification
system, Millipore, MA, USA). Sodium hydroxide was
References
1. Gao, S.; Hegg, D. A.; Hobbs, P. V.; Kirchstetter, T. W.;
Magi, B. I.; Sadilek, M. J. Geophys. Res. 2003, 108, 8491–
8506.
2. Simoneit, B. R. T.; Schauer, J. J.; Nolte, C. G.; Oros, D.
R.; Elias, V. O.; Fraser, M. P.; Rogge, W. F.; Cass, G. R.
Atmos. Environ. 1999, 33, 173–182.
3. Decesari, S., 1; Fuzzi, S.; Facchini, M. C.; Mircea, M.;
Emblico, L.; Cavalli, F.; Maenhaut, W.; Chi, X.; Schkol-
nik, G.; Falkovich, A.; Rudich, Y.; Claeys, M.; Pas-
hynska, V.; Vas, G.; Kourtchev, I.; Vermeylen, R.; Hoffer,
A.; Andreae, M. O.; Tagliavini, E.; Moretti, F.; Artaxo, P.
Atmos. Chem. Phys. 2006, 6, 375–402.
4. Fraser, M. P.; Lakshmanan, K. Environ. Sci. Technol.
2000, 34, 4560–4564.
5. Bridgwater, T. J. Sci. Food Agric. 2006, 86, 1755–1768.
6. Bridgwater, A. V.; Czernik, S.; Piskorz, J. In Progress in
Thermochemical Biomass Conversion; Bridgwater, A. V.,
Ed.; Blackwell Science, Ltd: Oxford, UK, 2001; pp 977–
997.
7. Li, L.; Zhang, H. Energy Sources 2004, 26, 1053–1059.
8. Piskorz, J.; Majerski, P.; Radlein, D.; Vladars-Usas, A.;
Scott, D. S. J. Anal. Appl. Pyrolysis 2000, 56, 145–166.
9. Yu, Z.; Zhang, H. Bioresour. Technol. 2003, 90, 95–100.
10. Vidrio, E. McNair Scholars J 2004, 5, 90–103.
11. Mosier, N. S.; Ladisch, C. M.; Ladisch, M. R. Biotechnol.
Bioeng. 2002, 79, 610–618.
12. Helle, S. S.; Murray, A.; Lam, J.; Cameron, D. R.; Duff, S.
J. B. Bioresour. Technol. 2004, 92, 163–171.