Pyrolysis II Proceedings
Session Chair: Akwasi Boatang
U.S. Department of Agriculture, Agriculture Research Service
High Yield Process for Hydrocarbon
Fuels from Cellulosic Biomass by
Pyrolysis of Organic Acid Salts
M. Clayton Wheeler
University of Maine
Clayton Wheeler received a Ph.D. in Chemical Engineering from the
University of Texas at Austin in 1997.
He then worked as a research engineer for Texaco in the areas of enhanced
oil recovery and Fisher-Tropsch Gas-to-Liquids technologies. After leaving
Texaco, he received a National Research Council Postdoctoral Fellowship to
conduct catalytic gas sensor research at the National Institute of Standards
and Technology.
He is currently an Associate Professor of Chemical and Biological
Engineering and the Thermal Conversion Group Leader of the University of
Maine’s Forest Bioproducts Research Institute.
The group’s areas of expertise include pyrolysis, catalyst development for
pyrolysis and pyrolysis oil hydrodeoxygenation, ketonization of mixed organic
acids, thermal deoxygenation and hydrogenation.
Atmospheric Non catalytic Method for Atmospheric, Non-catalytic Method for Hydrocarbon Fuels from Cellulosic Biomass b P rol sis of Organic Acid SaltsBiomass by Pyrolysis of Organic Acid Salts
M. Clayton Wheeler, University of MainePaige A. Case, Scott J. Eaton, Adriaan R. P. van Heiningen, and William J. DeSisto
Forest Bioproducts Research InstitutepFBRI Technology Research Center
FBRIChemistry Chem. Bio. Eng.Chemistry
Advanced Composites
Forestry
Biology
Margaret Chase Smith Policy Center
Footer goes here2tcbiomass2011
Overview of Method for Biomass Conversion to Drop-in Hydrocarbon FuelsConversion to Drop-in Hydrocarbon Fuels
High yields of carbohydrate conversion to oilHHV=41MJ/kg<1% Oxygen<1% OxygenTAN = 1 mgKOH/g
Hydrolysis and
Dehydration
Neutralize with Ca(OH)
Pyrolysis
with Ca(OH)2
Footer goes here3tcbiomass2011
Removal of Oxygen from Biomass(E l C ll l )(Example: Cellulose)> Hydrodeoxygenation
─ ~81% theoretical energy yield
100% carbon efficiency
OHCHOHCH 2251062 566 +→+
─ 100% carbon efficiency
─ 48% mass efficiency
> Theoretical Decomposition> Theoretical Decomposition
─ ~87% theoretical energy yield2225106 24 COOHCHOHC ++→
─ 67% carbon efficiency
─ 35% mass efficiency
Footer goes here4tcbiomass2011
Ketonization of Organic AcidsKetonization of Organic Acids
O
O
O
ΔCa
O
CaCO3+
> Net removal of 1 CO2 and 1 H2O or 1.5 O atoms/acid:─ Slaking
O
( )22 OHCaOHCaO →+g─ Ketonization─ Calcination
> I b h i l th t R1+R2 +1
( ) 3221221 2 CaCOOHCORROHCaCOOHRCOOHR ++→++
23 COCaOCaCO +→
( )22
> Increases carbon chain length to R1+R2 +1> Ketones can be hydrogenated and dehydrated to olefins which
can then be oligomerized to hydrocarbon mixtures
Footer goes here5tcbiomass2011
Organic Acids from BiomassOrganic Acids from Biomass
> Acidogenic mixed-culture fermentation (Terrabon)─ Carbohydrates are converted to mixtures of C2 to C7 range
predominately aliphatic acids
> Acid hydrolysis and dehydration (Biofine)> Acid hydrolysis and dehydration (Biofine)─ Cellulose converted to levulinic and formic acids (1:1 ratio)
223855106 OHCOHCOHC +→
─ C5 Hemicellulose (xylan) converted to furfural
223855106
Footer goes here6tcbiomass2011
Ketonization of Calcium LevulinateKetonization of Calcium Levulinate450 CO
OH
10
20
Inte
nsity
OO
OH
OH
20 40 600
Retention Time (min)
> Ramp to 450°C with N2 purgeW t l bl d t
T. Schwartz, et al., Green Chem., 2010, 12, 1353–1356.
> Water soluble products> No predicted ketone observed> 35 MJ/kg (calorimetry of extracts)
Footer goes here7tcbiomass2011
> 35 MJ/kg (calorimetry of extracts)
Thermal Deoxygenation (TDO)O O
O
Base BaseCa
O
O
OO CaCO3 +
OO
OO
OOH
O OBase Bas
H+ H+
BaseCa
O
O
O CaCO3
O
O
+H2O
O
OH
O
+
H2O
se
Products can be explained by ─ initial Ketonization
O
─ followed by Pyrolysis• Aldol condensation + dehydration
Footer goes here8tcbiomass2011
• High-temperature free radical reactions
Thermogravimetric Analysis of Reagent Formate and Levulinate SaltsFormate and Levulinate Salts> Cellulose hydrolyzates contain equimolar levulinate and formate
> Ca(HCOO)2 → CO + H2 + CaCO3( )2 2 3
Virtually no carbon in residue of levulinate/formate mixture!
Footer goes here9tcbiomass2011
Formate-Enhanced TDO Produces a H d b OilHydrocarbon Oil
Aqueous +nt
ribut
ion
Aqueous + Gas
arbo
n C
on Oil
Char
% C
a
F i /L li i A id R ti
Char
Carbonate
> Oil contains 82% of theoretical carbon (to 44MJ/kg) and 78% of theoretical mass at Formic/Levulinic = 1/1
Formic/Levulinic Acid Ratio
Footer goes here10tcbiomass2011
78% of theoretical mass at Formic/Levulinic 1/1
13C NMR of Oils Indicate Very Low O C t t Oxygen Content
Footer goes here11tcbiomass2011
Formate Increases Relative Fraction of U t t d C b i OilUnsaturated Carbons in Oil
70Aromatic/Alkene
50
60
13C
NM
R
Alk l
Aromatic/Alkene
30
40
rbon
s fro
m Alkyl
10
20
% o
f Car
Carbonyl00.0 0.5 1.0 1.5
Formate/Levulinate Mole Ratio
Footer goes here12tcbiomass2011
Insignificant Increase of HHV or H/C R ti Ab 5% F tH/C Ratio Above 5% Formate
1.9
44
1.5
1.7
40
42
Rat
io
kg)
H/C
HHV
1.1
1.3
36
38
H/C
Mol
e R
HH
V (M
J/k H/C
0.7
0.9
30
32
34 H
0.5300.0 0.5 1.0 1.5
Formate/Levulinate Mole Ratio
Footer goes here13tcbiomass2011
Efficiencies of Carbohydrate Conversiony2225106 24 COOHCHOHC ++→
222485 67.167.0333. COOHCHOHC ++→
Cellulose:Xylan:
Mass Efficiency (%) Energy Efficiency (%) §
Basis Theoretical Actual/Calculated Theoretical Actual/Calculated
Levulinate/Formate1/1 35 27† 87 63
Cellulose 35 19† 87 44Xylan 35 13‡ 89 32Wood* 26 13 59 27
§ HHV(MJ/k ) b(17 5) li i (25) f t il(44) d t l il(41)§ HHV(MJ/kg) = carb(17.5), lignin(25), perfect oil(44), and actual oil(41)† 70% levulinic acid yield, 78% oil yield‡70% furfural yield, 70% conversion to levulinic acid, 78% oil yield
Footer goes here14tcbiomass2011
* Ash-free, 50% C6, 25% C5, 25% Lignin
Independent Testing of Washed Crude TDO Oil Produced from Biomass Hydrolyzate
Boiling Point Distribution ( S 69)
Total Acid Number 1.02 mg KOH/gRamsbottom Carbon Residue 0 48 wt%
80
90
100
(ASTM D7169)
OXYGENATES (0.86%) GC Low Ox method (ppm)
Ramsbottom Carbon Residue 0.48 wt%
TRACE METALS
ICP (ppm)
Aluminum <0.1 A i <0 1
50
60
70
80
ecov
ered
Acetaldehyde 43 Methanol 18
Ethanol 53 Propanols 44 n-Butanol 14
Butyraldehyde 16
Arsenic <0.1Barium <0.1
Beryllium <0.1 Bismuth <0.1
Boron <0.1 Cadmium <0 1
10
20
30
40
%R
eButyraldehyde 16 Methyl T-Butyl Ether 24
Ally Alcohol 4 tert-Amyl Alcohol 15
Vinyl Acetate 8338
Cadmium <0.1Calcium 4.5
Chromium <0.1 Cobalt <0.1
Copper 0.4 Iron 8 8
0
10
100 200 300 400 500Boiling Point, °C
Iron 8.8Lead 0.3
Magnesium <0.1 Potassium 0.2
Sodium 0.2
Footer goes here15tcbiomass2011
Comparison of Boiling Point DistributionsCrude TDO OilRetorted Oil ShaleOil Sands HGOLight Arabian Crude#2 Diesel
ure
(°F)
Tem
pera
tuT
> ~80% Overlap with #2 dieselRecovery (mass%)
Footer goes here16tcbiomass2011
80% Overlap with #2 diesel
PONA Analysis (ASTM D6730)35
30
35
P ffi (2 7%)Calculated PropertiesO t (R+M)/2 81 9
Sub-400°F "Gasoline" Cut (28% of sample)
20
25
cove
red Paraffins (2.7%)
I-Paraffins (30.0%)Olefins (16.8%)Cyclo-Paraffins (19 0%)
Octane, (R+M)/2 = 81.9Specific Gravity = 0.793H/C ratio = 1.75Benzene = 0 8 vol%
15
e %
Rec Cyclo Paraffins (19.0%)
Aromatics (31.5%)Benzene = 0.8 vol%
5
10
Volu
me
03 4 5 6 7 8 9 10 11 12 13 14 15
Carbon Number
Footer goes here17tcbiomass2011
Conclusions/ImplicationsConclusions/Implications
> Two-step pathway to fungible hydrocarbon oil (41 MJ/kg)!
> 0.9 BOE / ODMT Wood demonstrated (1.9 BOE potential)
> Atmospheric pressure TDO
> No hydrogen
> No precious metal or other solid catalysts
> Cation may be regenerated in pulp mill based biorefinery
> Pyrolytic “deoxyhydrogenation” tolerant to impuritiesy y y y g p
> Future improvements likely from lignin (1 BOE potential) and improved carbohydrate yields
Footer goes here18tcbiomass2011
AcknowlegementsAcknowlegements> Co-Authors:
─ Paige A. Case
─ Scott J. Eaton
─ Adriaan R. P. van Heiningen
─ William J. DeSisto
> Significant Contributions─ Thomas J. Schwartz
─ Sedat H. Beis Mild hydro-t t t─ G. Peter van Walsum
─ Hemant P. Pendse
> Funding
treatment
─ UMaine Forest Bioproducts Research Institute
─ U.S. Dept. of Energy DE-FG02-07ER46373 and DE-FG36-08GO18165
Footer goes here19tcbiomass2011
Questions?Questions?
50°C 75°C 125°C 175°C150°C100°C85°C50°C122°F
75°C167°F
125°C257°F
175°C347°F
150°C302°F
100°C212°F
85°C185°F
Footer goes here20tcbiomass2011
Co-Processing of Standard Gas Oil
and Stable Pyrolysis Oil to
Hydrocarbon Fuels
Foster A. Agblevor
Utah State University
Foster A. Agblevor
- Biological Engineering, Utah State University, Logan UT.
- 1996-2010: Professor, Biological Systems Engineering, Virginia Polytechnic
Institute and State University, Blacksburg, VA.
- Research Interest: thermochemical and biochemical conversion of biomass
to fuels and chemicals.
Co-Processing of standard gas oil and
stable pyrolysis oils to hydrocarbon fuels
Foster A Agblevor, O. Mante*, R.
McClung**, F. Battaglia*, Ted Oyama*,
Biological Engineering, Utah State
University, Logan, UT
*Virginia Tech, Blacksburg, VA
**BASF Inc
tcbiomass11.presentation.Chicago.2011
Objectives
• Develop low temperature catalytic process
to produce stable pyrolysis oils
– Develop suitable catalysts for the process
– Develop integrated processing of biocrude
and petroleum crude oils to produce ―drop in
fuels‖
– Develop fundamentals of fractional catalytic
pyrolysis
O
O
O
OH
OH
OH
HO
HO
OHO
O
O
O
OH
OH
OH
HO
HO
OHO
O
O
O
OH
OH
OH
HO
HO
OHO
O
O
O
OH
OH
OH
HO
HO
OHO
O
O
O
OH
OH
OH
HO
HO
OHO
O
O
O
OH
OH
OH
HO
HO
OHO
O
O
O
OH
OH
OH
HO
HO
OHO
O
O
O
OH
OH
OH
HO
HO
OHO
OH
O
HO
H3CO
OH
OCH3
OCH3
O
O
O
OH
OCH3
OCH3
H3CO
OO
HO
H3CO
HO
OCH3
OCH3
OH
O
HO
H3CO
OH
OCH3
OCH3
O
O
OH
OCH3
OCH3
OCH3
O
O
O
OH
HO
O
O
O
O
OH
HO
OH
OH
O
O
O
OH
HO
OH
OH
O
O
O
OH
HO
OH
OH
O
O
O
OH
OH
OH
HO
HO
OHO
O
O
O
OH
OH
OH
HO
HO
OHO
O
O
O
OH
OH
OH
HO
HO
OHO
O
O
O
OH
OH
OH
HO
HO
OHO
O
O
O
OH
OH
OH
HO
HO
OHO
O
O
O
OH
OH
OH
HO
HO
OHO
O
O
O
OH
OH
OH
HO
HO
OHO
O
O
O
OH
OH
OH
HO
HO
OHO
Lignin: 15-25%
Complex aromatic structure
Very high energy content
Cellulose: 38-50%
Polymer of glucose, very good biochemical feedstock
Biomass
Constituents
Hemicellulose: 23-32%
Polymer of 5 & 6 carbon sugar
Solid
Gas or Liquid
Pyrolysis molecular beam mass spectra of biomass
MCC
Poplar wood
13C-NMR spectra of FCP heavy oils
Materials and Methods
• Materials
– Hybrid poplar wood ground to pass 1-mm
screen
– Proprietary catalyst/silica sand
– Fluidizing gas –Nitrogen, pyrolysis gas
• 2-inch Bubbling fluidized bed reactor
• Pilot scale pyrolysis unit
Biomass Catalytic Pyrolysis Unit
1- Fluidized bed reactor,
3- Thermocouple, 4- Mass flow controller, 5- jacketed air-cooled
feeder tube, 6- Hopper, 7- Screw feeder, 8- Computer, 9- Heating tape,10-Hot gas filter, 11-Reservoir, 12-Condenser, 13-ESP,14-AC power supply, 15-Filter, 16-Wet gas meter, 17-Gas
chromatograph)
Materials and Method
• Method
• Pyrolysis temp = 450-600 C
• Vapor residence time = 1 s
• Electrostatic precipitator at 18-30 kV
• Run time – 2-3 hours
• Biomass feed rate 100 g/h.
• Catalyst = 100-150 g
Materials and Method
• Analysis
– FTIR and 13C NMR analysis of oils
– GC/MS analysis of liquid products
– High temperature simulated distillation
– GC analysis of gases
– FCC co-cracking
– 14C analysis of cracked products
Results and Discussion
Product yields for stable oils
Biomass FCP oil (wt%) Char (wt%) Gas (wt%) Oil pH
Hybrid poplar 33.3 12.2 55.0 3.4
Pinewood 43.3 35.1 21.5 3.3
Oakwood 41.8 33.8 24.6 4.4
Corn stover 40.1 24.8 35.5 4.2
Switchgrass 35.5 27.6 36.7 4.2
Hybrid poplar FCP oil
Molecular weight distribution of poplar FCP oil
Mn = 120 Mw = 257Mz = 555Mp = 134Mn/Mw = 2.136
Properties SPO
C (%) 71.16±1.81
H (%) 6.82±0.14
O (%) 21.85±1.95
N (%) <0.5
S (%) <0.05
Ash (%) <0.08
pH 3.73±0.03
Viscosity of fresh oil (cP) 11.24±0.20
Moisture content of fresh oil (%) 8.59±0.40
Viscosity of stored oil (cP) 12.70±0.20
Moisture content of stored oil (%) 8.66±0.20
Density fresh oil (g/ml) 1.116±0.001
Density stored oil (g/ml) 1.117±0.001
TAN (mgKOH/g oil) 41.01±0.82
HHV (MJ/kg) 30.5
Properties of stable pyrolysis oils (SPO).
13C-NMR SPECTRA OF CONVENTIONAL AND STABLE WOOD BIOOILS
0
200
400
600
800
1000
1200
1400
0 20 40 60 80 100
Tem
pera
ture
( d
eg
.F)
Wt.% distilled
Standard gasoil and biooils distillation curves—(BASF)
VPI-1
VPI-3
VPI-6
VPISU001
Pilot Scale
Std. Gas OIl
Catalyst ID VPI-1 VPI-3 VPI-6 HZSM-5 Pilot scale Disposition
<430oF (heavy naphtha(wt%))
17.5 21.5 17.5 21.5 21.5 Naphtha hydrotreater
430-650 oF[middle distillate --diesel and heating oils] (wt%)
42.5 57.5 45.5 46.5 45.0 Diesel or heating oil hydrotreater
>650oF [FCC unit feed] (wt%)
40 22.5 37 32 33.5 FCC feed hydrotreater
Distillate fractions of stable biomass pyrolysis oils (biosyncrude)
FCC Co-Cracking Data—(BASF)
Standard
4350VPI-4 VPI-4ST VPISU001
H2 (%) 0.61 0.53 0.44 0.56
Total C2- (%) 2.98 2.99 2.92 2.94
LPG (%) 16.00 16.19 16.00 15.95
Gasoline (%) 43.97 44.01 44.44 44.35
LCO (%) 17.06 16.93 17.23 17.23
HCO (%) 12.94 13.07 12.77 12.77
Coke (%) 7.06 6.81 6.64 6.76
Conversion (%) 70.00 70.00 70.00 70.00
Cat/Oil 6.00 6.08 5.96 5.81
200 150 100 50 0 PPM
13C-NMR SPECTRUM OF FCC CRACKED BIOOIL/GASOIL BLEND
Aromatic SPO
Co-cracked SPO
Aliphatic SPO
14C analysis of SPO/SIGO Blend cracked products
• Beta Analytical Inc, Miami, FL
• ASTM D6866 method used for the analysis
• The cracked SPO/SIGO product contained 3% biocarbon.
Parameter As received
basis
Calorific value 24.94 MJ/kg
Karl Fischer water content (%) 2.89
Ash (%) <0.09
Carbon (%) 65.96
Hydrogen (%) 7.11
Oxygen (%) 26.36
Nitrogen (%) 0.54
Sulfur (%) <0.5
Chlorine (ppm) 54
Viscosity (cP) 1024
pH 2.8
Characteristics of FCP Heavy oil.
Summary
• We have developed a process to produce stable FCC-crackable pyrolysis oils.
• The viscosity of the stable pyrolysis oil increased from 10.1 cSt to 11.4 cSt after ten months of storage at ambient laboratory conditions
• The stable pyrolysis oils were completely distillable without char formation (no residuum)
• 85/15 blend of Standard Gulf coast FCC feed and stable pyrolysis oils were crackable without any problem. The blend produced slightly less coke than the Standard Gulf coast FCC feed cracked under similar conditions
Summary
• How transformational is the technology?
• Pyrolysis oils could be co-processed in
existing petroleum refineries
• Existing petroleum refineries could
claim ―green credits‖ by processing the
biosyncrude.
• We have also developed technology to
produce heavy oils to complement
stable oils
Grassoline for your car?
Acknowledgement
• DOE for funding the pyrolysis oil stability
studies.
• Contract# DE-FG36-08GO18214-1
Thank you!!
• Questions?
The Role of the Lignocellulosic
Composition on the Pyrolysis of
Waste Crops: Fractionation and
Kinetic Study
Marion Carrier
Stellenbosch University
Marion Carrier obtained a degree in chemical engineering from the CPE
Lyon Engineering School as well as a Masters degree in Analytical
Chemistry from the University of Lyon.
In 2007, she completed her PhD in Chemistry at the University of Lyon in
collaboration with IRCELYON-CNRS (Lyon), INRA (Thonon-les-bains) and
Cemagref (Lyon) laboratories. The focus of her PhD was the treatment of
wastewaters from wine fermentation by advanced oxidation processes:
photocatalysis, sonolysis and catalyzed wet air oxidation.
Marion joined the Department of Process Engineering in November 2008 as a
postdoctoral research fellow and she has been granted by the NRF
Innovation fellowship in 2010 to pursue her research on the conversion of
lignocellulosic biomass to biofuels and chemicals with pyrolysis and
gasification.
The role of the
lignocellulosic composition
on the pyrolysis of waste
crops: Fractionation and
kinetic study
Dr Marion Carrier, Stellenbosch University, South-Africa
Footer goes here2
tcbiomass2011
Lignocellulosic composition
Lignin
Really complex
3 main
compounds
Hemicelluloses 20-40 %
Polymerisation degree 50-300
Hexosanes/Pentosanes blocks
Cellulose 43-60 %
Polymerisation degree 10000
Glucose blocks linked by ether bonds Mineral content
Si, K, Mg, Ca, Na, P, Fe, Al, Zn
Linked with cellulose
Footer goes here3
tcbiomass2011
How to predict the influence of biomass
nature on the pyrolysis process?
> Natural or synthetic compounds?
> Which wet chemical extraction methods?
> Which type of kinetic study?
Footer goes here4
tcbiomass2011
OBJECTIVE
Understand the thermal behaviour of the lignocellulosic
composition to reach efficient pyrolysis conversion
> by understanding the role of interactions between
macrocomponents,
> by establishing a mathematical correlation between the
lignocellulosic composition and thermal properties.
Footer goes here5
tcbiomass2011
> Using synthetic biomass studies, no detectable
interaction among the commercial components during
pyrolysis.
> A possible synergistic effect arising from the coincidence
of the ingredients in the parent sample in the favor of
lowering the activation energy.
Natural or synthetic compounds?
Raveendran et al., Fuel (1996) 75 (8) 987-998
Haykiri-Acma et al., Fuel Processing technology (2010) 91 759-764.
Footer goes here6
tcbiomass2011
Which wet chemical analysis to
fractionate the whole structure?
NDF: Neutral Detergent Fibre
ADF: Acid Detergent Fibre
ADL: Acid Detergent Lignin
Weende
HOLO: Holocellulose
α: α-cellulose
LIG: Klason lignin
Sluiter et al., J.Ag. Food Chem. (2010) 58 (16) 9043-9053.
Footer goes here7
tcbiomass2011
Results
Method Lignin(wt% dry)
Holocellulose(wt% dry)
Calculated
hemicelluloses(wt% dry)
-cellulose(wt% dry)
Extractibles(wt% dry)
“wood-
industry”35 ± 2 (a) 68 ± 5 (d) 34 ± 9 (e) 34 ± 4 (f) 3 ± 2
“food-
industry” 22 ± 5 (c) 4 ± 2 (g) 26 ± 10 (h)
“food-
industry” 18 ± 3 (b) 13 ± 2 (k)
37 ± 6 (j)
35 ± 9 (l)
“wood-
industry”25 ± 2 (a) 71 ± 5 (d) 28 ± 10 (e) 43 ± 5 (f) 5 ± 3
“food-
industry” 16 ± 4 (c) 9 ± 4 (g) 41 ± 16 (h)
Extraction methods used: (a) LIG, (b) ADL, (c) ADLB, (d) (HOLO), (e) (HOLO-α), (f)
(α), (g) (NDFB-ADFB), (h) (ADFB-ADLB), (i) unknown, (j) Weende, (k) NDF-ADF
and (l) NDF-ADL.
Yields of lignin, holocellulose, hemicelluloses and cellulose from fern (dry matter, 91 %) and
wood (dry matter, 97 %).
Footer goes here8
tcbiomass2011
Results
-1
0
1
2
3
4
5
6
0 200 400 600 800
DT
G d
m/d
m0
(%)
Temperature (°C)
α-Cellulose
Lignin
Holocellulose
Crude poplar
Washed
DTG curves of the crude and washed biomasses, and macromolecules (5 ºC min-1): holocellulose, α-cellulose, and lignin extracted from the washed Populus nigra L.
Footer goes here9
tcbiomass2011
Results
Hemicellulosesy = 49x
R2 = 0.99
y = 32xR2 = 0.90
0
100
200
300
400
0 5 10
Aera
(au)
Mass (mg)
α-cellulose
Cellulose
pulps-Pine
α-cellulose
(wt.% dry) (1)
wood
(wt.% dry) (2)
fern
(wt.% dry) (2)
110106 87.6 79 ± 7 74 ± 6
5092 90.3 97 ± 7 91 ± 6
5094 92.8 96 ± 8 91 ± 7
5096 93.6 99 ± 8 93 ± 7
Cellulose
pulps-Pine
Hemicelluloses
(wt.% dry) (3)
wood
(wt.% dry) (2)
fern
(wt.% dry) (2)
110106 11.7 8 ± 4 17 ± 10
5092 7 6 ± 3 13 ± 7
5094 4.7 6 ± 4 13 ± 9
5096 3.6 5 ± 4 10 ± 9
Carrier et al., Biomass and Bioenergy, 35 (2011) 298-307.
(1) R10 method
(2) Calibration curves
(3) 100-R18 method
Calibration curves for the wood.
Footer goes here10
tcbiomass2011
Which kinetic method to apply for
solid state reactions?
„Model-free‟ or isoconversional methods
Model-fitting
Differential method or Integral method
Arrhenius rate expressionk(T): reaction rate constant
T: Temperature in K
A: Frequency factor or pre-
exponential
Ea: Activation energy
R: Universal gas constant
Kinetic expressionα: degree of conversiont: timek(T): reaction rate constant
f(α): conversion function
White et al., Journal of Analytical and Applied Pyrolysis, 91 (2011) 1-33
Vyazovkin et al., Thermochimica Acta, 520 (2011) 1-19.
Kinetic triplet: A, Ea and f(α)
Friedman‟s method
Footer goes here11
tcbiomass2011
Results
Experimental results
Sugar cane bagasse
0
10
20
30
40
50
0
50
100
150
200
250
0 0.5 1
Conversion α
ln(A
*f(α
)) (
1/s
)
E (
kJ
/mo
l)
010203040506070
0
100
200
300
0 0.5 1
Conversion α
ln(A
*f(α
)) (
1/s
)
E (
kJ
/mo
l)
Hemicelluloses
0
10
20
30
40
50
60
0
100
200
300
0 0.5 1
Conversion α
ln(A
*f(α
)) (
1/s
)
E (
kJ
/mo
l)
Cellulose
0
10
20
30
40
50
60
0
100
200
0 0.5 1
Conversion α
ln(A
*f(α
)) (
1/s
)
E (
kJ
/mo
l) Lignin
Footer goes here12
tcbiomass2011
Results
Comparison of predicted and experimental curves of bagasse pyrolysis at 50°C min-1
Aboyade et al., Thermochimica Acta 517 (2011) 81-89.
0
0.005
0.01
0.015
100 200 300 400 500 600
DT
G (
s-1
)
Temperature (oC)
Cellulose
predictedHemicelluloses
predicted
Lignin predicted
Total predicted
Experimental
Footer goes here13
tcbiomass2011
Conclusions
> Variation inherent to the selection of the extraction method (∆% = 20 %).
> Alkaline extraction recommended for the hemicellulose
content determination.
> α-cellulose and hemicellulose contents determination
successful.
> Kinetic parameters determined by Friedman‟s method
applicable to the prediction of experimental data.
> TGA technique is suitable to point out and quantify
synergistic effects.
Footer goes here14
tcbiomass2011
Future works
> Improve the lignin extraction technique.
> Determinate the calibration curve for the lignin.
> Compare the kinetic parameters of extracted products
with the ones obtained from predicted products.
> Evaluate the accuracy and precision of predictions.
Footer goes here15
tcbiomass2011
Acknowledgements
> Supercritical team from CNRS-ICMCB, Bordeaux, France
(Cyril Aymonier, Anne Serani-Loppinet, Michel Mench).
> Process Engineering department from Stellenbosch
University, South Africa (Hansie Knoetze, Johann
Görgens, Wale Aboyade).
> The National Research Foundation for sponsoring the
study.
THANK YOU
Lignin Pyrolysis in a Fluidized Bed
Reactor with Fractional
Condensation
Cedric Briens
ICFAR at the University of Western Ontario
Lignin Pyrolysis in a Fluidized Bed Reactor
Pietro Palmisano, Federico Berruti,
Valentina Lago, Franco Berruti,
Cedric Briens
Why pyrolyze lignin?
• Organosolv lignin:– Newer process
• Kraft lignin:– Kraft process used for most of worldwide paper
production– Black Liquor side product is currently burned– Lignin can be extracted from Black Liquor
• 50 million tonnes of lignin available worldwide• Most of it is burned
Why pyrolyze lignin?
Cracking of lignin → interesting aromatic products?
Objective
• Pyrolyze lignin (> 200 g / run)
• Lignin cannot be pyrolyzed in standard equipment
• Develop equipment for reliable pyrolysis of lignin
Outline
• Equipment
• Why is lignin difficult to pyrolyze?
• Solutions
• Results
• Conclusions
• What is next
Equipment
Solid feeder
FluidizedBedReactor
Condensing zone
Gas sampling
N2 for fluidization
Reactor diameter = 7.5 cm (3 in)
1.5 kg of sand with dpsm = 180 μm
Lignin
• Very fine particles ( < 30 μm)
• Particle density = 575 kg/m3
Why is lignin difficult to pyrolyze?
• Lignin starts melting at low temperature: 150 – 200 ºC
• Lignin needs a high reactor temperature to crack fully
1) Solids feeder plugs
Lignin
Fluidized sand
Why is lignin difficult to pyrolyze?
• Lignin starts melting at low temperature: 150 – 200 ºC
• Lignin needs a high reactor temperature to crack fully
1) Solids feeder plugs
2) Low density foam forms
at the bed surfaceLignin foam
Lignin
Fluidized sand
3 in
Solution 1: Pulsed feeder
N2 Pulse
N2 continuous
Solution 1: Pulsed feeder
Foam forms at bed surface (only with Kraft Lignin)
Bio-oil vapours and gases
What causes foam at the top?
2 possibilities:
1) Injected particles do not mix well with sand particles
2) Reacting lignin particles rise to the bed surface
Solution 2: Premix lignin with sand
Sand premixed with lignin before injection into the hot bed (sand/lignin = 5/1 wt/wt).
3 in
Solution 3: Use a mechanical stirrer
Mechanical stirrer in the fluidized bed to bringsolids from the bed surface down into the bed
Solution 3: Use a mechanical stirrer
No mixer. 200g of lignin fed.
Mixer. 80 rpm.200g of lignin fed.
Lignin particles rose to the surface and agglomerated
Strong top to bottom mixing: no foam
Reactor temperature, oC
440 460 480 500 520 540 560
yie
ld,
wt%
0
10
20
30
40
50
60
70
80
char
bio-oil
gas
Product yields: Kraft lignin1.8 s vapor residence time
Reactor temperature, oC
440 460 480 500 520 540 560 580 600 620
yie
ld,
wt%
0
10
20
30
40
50
60
char
bio-oil
gas
Product yields: Kraft lignin0.4 s vapor residence time
Reactor temperature, oC
440 460 480 500 520 540 560
yie
ld,
wt%
20
25
30
35
40
45
50
0.4 s
1.8 s
Effect of vapor residence time on liquid yield
Reactor temperature, oC
400 450 500 550 600 650
HH
V, kJ/g
8
10
12
14
16
18
20
22
0.4 s
1.8 s
Heating value of bio-oil
Reactor temperature, oC
400 450 500 550 600 650
(liq
uid
HH
V)
/ (lig
nin
HH
V),
%
5
10
15
20
25
30
35
40
0.4 s
1.8 s
% energy recovered in bio-oil
Bio-oils from Organosolv and Kraft lignins
Bio-oil from Kraft lignin
Bio-oil from Organosolv lignin
Yield at 550 ºC, 1.8 s 31% 35%
Pesticide activity X 3
Conclusions
• First practical technology for fast Kraft lignin pyrolysis
• Low but reasonable liquid yields
• Energy from lignin recovered in liquid < 40%
• Strong effect of vapor residence time
What is next
• Detailed chemical analysis of bio-oil
• Separation methods to extract valuable compounds
• Study of effect of vapor residence time
• Replace mechanical stirrer with draft/lift tube
Acknowledgements
• Technical support:– Clayton Stanlick and Rob Taylor (mechanical stirrer)
– Mohammad Hossain and Dr. Ian Scott, Agriculture and AgriFood Canada (Pesticide testing)
• Lignoworks network for financial support:– NSERC
– FPInnovations
– Lignol
– Weyerhaueser
Prospects for a Thermolytic
Sugar Platform
Robert Brown
Bioeconomy Institute at Iowa State University
Dr. Robert Brown is Anson Marston Distinguished Professor of
Engineering and Gary and Donna Hoover Chair in Mechanical Engineering at
Iowa State University (ISU).
He also holds courtesy academic appointments in the Departments of
Chemical and Biological Engineering and Agriculture and Biosystems
Engineering. He is the director of ISU’s Bioeconomy Institute.
His research focuses on the thermochemical processing of biomass and
fossil fuels into energy, fuels, and chemicals.
Prospects for a ThermolyticProspects for a Thermolytic Sugar Platform
Robert C. Brown
g
Center for Sustainable Environmental TechnologiesIowa State University
Ames IAAmes, IA
TC Biomass ConferenceChicago, IL
September 27‐30, 2011
AcknowledgmentsIowa State University’s thermochemical research program is conducted atIowa State University s thermochemical research program is conducted at the Center for Sustainable Environmental Technologies (CSET). Current staff include Ryan Smith, Marge Rover, Pat Johnston, Lysle Whitmer, and Jordan Funkhouser and postdoctoral research associates Sunitha Sadula and Tristan Brown. Faculty collaborators include Profs. Brent Shanks, Guiping Hu, and Xianglan Bai at ISU and Profs. Wolter Prins and Frederik Ronsse at Ghent University, Belgium. I am indebted to several current and former graduate students who have contributed to this presentation:former graduate students who have contributed to this presentation: Pushkaraj Patwardhan, Mark Wright, Yanan Zhang, Najeeb Kuzhiyil, AJ Pollard, Dustin Dalluge, Andrew Olthoff, and Andrew Friend.
CSET’ h h i l j d b h C PhilliCSET’s thermochemical projects are supported by the ConocoPhillips Company, ADM, the U.S. Department of Energy, the National Advanced Biofuels Consortium, the U.S. Department of Agriculture, the state of Iowa, and gifts from Gary and Donna Hoover and John Pappajohn.Iowa, and gifts from Gary and Donna Hoover and John Pappajohn.
The goal of this work is to develop a thermal process for the production of “cheap sugars” from cellulosic biomasscheap sugars from cellulosic biomass
Fast PyrolysisRapid thermal decomposition of organic compounds in the absence of oxygen to produce
Conventional Products from Fast Pyrolysisyg p
predominately liquid product
Composition of Conventional Bio‐Oil from Oak Wood (wt%)
y y
Aqueous phase (carbohydrate‐derived)
Oil from Oak Wood (wt%)
Saccharides 14.1
Aldehydes 16.3
Pyrolytic lignin (water insolubles)
Char*
Syngas
Biochar
Furans 1.4
Ketones 3.2
Carboxylic acids 4.7
Syngas
*Also called biochar
Phenolic Monomers 7.3
Other GC/MS Detected 2.2
Phenolic oligomers 22.4
Water 28.4
Source: ISU (2010)
Cellulose Decomposition: Conventional WisdomConventional Wisdom
Fast pyrolysis yields predominately anhydrosugar, which decomposes into low molecular weight products or in
Li ht t
decomposes into low molecular weight products or, in the presence of char, dehydrates to char and water.
Fast
Decomposition
Levoglucosan
Light oxygenates
Tarry vaporsVaporization
Cellulose
SlowChar + water
Char + waterChar-catalyzed
dehydration
Vaporization
Ch ll i C i l Wi d bChallenging Conventional Wisdom about Carbohydrate Pyrolysis (Working Hypotheses)
• Pure holocellulose depolymerizes to monomeric anhydrosugars;
• Alkali and alkali earth metals (AAEM) catalyze decomposition of holocellulose to undesirable “light oxygenates” inundesirable light oxygenates in competition with depolymerization;
• AAEM can be “passivated” through biomass Pyrolytic Molasses
pretreatments;
• Sugar yields are strongly influenced by secondary processes of evaporation andsecondary processes of evaporation and polymerization of anhydrosugars
/Fundamental Studies with Py‐GC/MS or FIDFrontier ~ 500 μg
FeedstockMicro-pyrolyzer
500 μg
He
Capillary Separation
Mass Spectrometer
( )Capillary Separation Column (MS)
Gas Chromatograph (GC)
Effects of Cations on Deconstruction of Cellulose
20Formic acid
60 LevoglucosanNaCl
NaClKCl
10
20
Wt %
20
40
wt %
NaClKClMgCl2CaCl2
MgCl2CaCl2
00 0.1 0.2 0.3 0.4
moles of salt/g of cellulose
00 0.1 0.2 0.3 0.4
moles of salt/g of cellulose
3
4
%
Acetolmoles of salt/g of cellulose
3
4
5
%
Furfural
NaClNaClKCl
0
1
2
Wt %
0
1
2Wt
KClMgCl2CaCl2
MgCl2CaCl2
Patwardhan et al., Bioresources Technology (2010) 4646-4655.
00 0.1 0.2 0.3 0.4
moles of salt/g of cellulose
00 0.1 0.2 0.3 0.4
moles of salt/g of cellulose
M h i f C ll l D iti Vi C tiMechanism of Cellulose Decomposition Via CationsCoordination bonding of cations on cellulose
induces homolytic fission of glucose rings
Cellulose
Ca2+
y g gK+
Depolymerized fragment
Intermediate
LevoglucosanPonder et. al., J Anal. App.Pyrolysis, 1991, Yang et al. Chem. Res. Chinese U. 2006
Passivating Alkali in BiomassPassivating Alkali in Biomass
Pretreating switchgrass with most mineral acids (with the exception of nitric acid)
TC Biomass POSTER: Kuzhiyil et al. “Passivating Alkali Metal during Pyrolysis of Biomass”
Pretreating switchgrass with most mineral acids (with the exception of nitric acid) significantly increases yields of anhydrosugars and decreases yields of light oxygenates. Organic acids and nitric acid had little affect on pyrolysis.
Comparison of Different Pretreatments
16.0018.0020.00
basi
s)
pSwitchgrass ControlAcetic AcidFormic AcidNitric Acid
10.0012.0014.00
mas
s (w
et Hydrochloric Acid
Phosphoric AcidSulfuric Acid
4.006.008.00
t% o
f bio
m
0.002.00
Light Oxygenates Anhydrosugars Furans Phenols
wt
Th i i l l l f id i f iThere is an optimal level of acid infusionTC Biomass POSTER: Kuzhiyil et al. “Passivating Alkali Metal
during Pyrolysis of Biomass”during Pyrolysis of Biomass
The AAEM passivation hypothesis predictsThe AAEM passivation hypothesis predicts optimal acid infusions
TC Bi POSTER K hi il t l “P i ti Alk li M t lTC Biomass POSTER: Kuzhiyil et al. “Passivating Alkali Metal during Pyrolysis of Biomass”
Correlation between amount of minerals in biomass and amount of acid required to achieve optimal levoglucosan yieldof acid required to achieve optimal levoglucosan yield
Role of Secondary Reactions in Determining Yield of Levoglucosan
• Rapid quenching of primary products thought essential to prevent decomposition ofessential to prevent decomposition of levoglucosan
Ch th ht t t l d h d ti f• Char thought to catalyze dehydration of levoglucosan
Decomposition of levoglucosan suppressed by acid washing of char, which suggests that alkali is the catalytic agent
TC Biomass POSTER: Ronsse et al. “Secondary reactions of levoglucosan and char in the fast pyrolysis of cellulose”
unwashed
acid washed
M T f M Li it S Yi ld f AAEMMass Transfer May Limit Sugar Yields from AAEM Passivated Biomass
TC Biomass POSTER: Bai et al. “The Role of Levoglucosan
0.010
0.012
5 Deg. C/minLG evaporation peak
/s)
Dehydration peak
gPhysiochemistry on Cellulose Pyrolysis”
Cellulose
0.004
0.006
0.008
2 mg open cup
10ss Loss Ra
te (m
g/
y p
Differential Thermogravimetry
L l
Pyrolysis
0.000
0.002
50 100 150 200 250 300 350 400 450 500
10 mg open cup
Mas Levoglucosan
Temperature (⁰C) Vaporization Polymerization
DehydrationResidue of cellulose y
CharTar
pyrolysis in TGA sample cups
ll l d dCellulose Decomposition: Updated
Fast pyrolysis consists of two stages of competitive
• Depolymerization and alkali‐catalyzed decomposition of cellulose
Fast pyrolysis consists of two stages of competitive processes:
• Vaporization and polymerization/dehydration of levoglucosan
Alk li t l d d iti
Fast
Alkali-catalyzed decompositionLight oxygenates
LG polymerization/ dehydration
Depolymerization Furans + char + waterCellulose
SlowChar + water
LG vapors
Depolymerizationto levoglucosan
LG evaporation
Rethinking Strategy for Upgrading Recover Bio Oil asRethinking Strategy for Upgrading: Recover Bio‐Oil as Stage Fractions
TC Biomass POSTER: Pollard et al. “Analysis of Bio-oil Produced in a C o ass OS o a d et a a ys s o o o oduced aFractionating Bio-Oil Recovery System during Pyrolysis of Red Oak,
Switchgrass, and Cornstover”
Biomass feederSF 1
SF 2: ESPs SF 4: ESP
Pyrolyzer
SF 1: Condenser
SF 3: C d
CyclonesCondenser
SF 5:Condenser
R thi ki St t f Bi Oil U diRethinking Strategy for Bio-Oil UpgradingTC Biomass POSTER: Pollard et al. “Analysis of Bio-oil Produced in a
Fractionating Bio-Oil Recovery System during Pyrolysis of Red Oak, Switchgrass, d C t ”
Current pilot plant at ISU’s BioCentury Research Farm produces five distinct t f ti f bi il
and Cornstover”
stage fractions of bio-oil
Bio-Oil from Red Oak Stage Fraction 1
Stage Fraction 2
Stage Fraction 3
Stage Fraction 4
Stage Fraction 5
Yield (wt% of biomass)) 21.0 26.5 5.5 11.1 36.0Moisture (wt% bio-oil) 7 8 9 15 63Water Insoluble (wt% bio-oil) 44 45 8 13 <1Levoglucosan (wt% bio-oil) 4.8 3.2 1.0 1.5 0.2Levoglucosan (wt% bio oil) 4.8 3.2 1.0 1.5 0.2Total Acid Number 35 32 79 117 117Oxygen content (wt% m.f. basis) 28.6 29.1 41.1 39.6 53.3Higher Heating Value (MJ/kg) 24.2 24.4 20.2 18.7 7
Heavy ends (sugar and phenol oligomer-rich fractions of bio-oil)
Viscosity (cSt @ 40°C)* 4400 3400 36 50 1.3
Separating Heavy Ends into SugarSeparating Heavy Ends into Sugar and Phenolic Oligomers
TC Biomass POSTER: Rover et al “Sugar Recovery from the
Sugar solution (20-40 wt%)
TC Biomass POSTER: Rover et al. Sugar Recovery from the Heavy Ends of Fractionated Bio-Oil”
Raffinate (mostly phenolic oligomers)
Estimating the Cost of Pyrolytic SugarsEstimating the Cost of Pyrolytic SugarsTC Biomass POSTER: Zhang et al. “Biomass Fast
Pyrolysis and Upgrading for Production of Cheap Sugars
Product Yield Market ValueHydrogen 8.23 million kg/yr $2/kgGasoline 8.60 million gal/yr $3.26/galDiesel 7.25 million gal/yr $3.26/galProduct Yield Production CostSugar 128 millionSugar
(glucose syrup) 128 million
kg/yr$0.54/kg
Conclusions• AAEM passivation increases levoglucosan fromAAEM passivation increases levoglucosan from biomass pyrolysis to levels comparable to pyrolysis of pure carbohydrate;
• Levoglucosan undergoes competing processes of evaporation (recovered as “tar”) and polymerization (dehydrates to char and furans);(dehydrates to char and furans);
• Strategy of recovering bio‐oil as stage fractions allows separation of sugars;p g
• Cost analysis of products from non‐optimized process are promising.
Other presentations from CSET students and staff
• PLENARY/– Dalluge et al., “Pyrolytic Pathways to Increasing the Lignin‐Derived Monomer/Oligomer Ratio in
Bio‐oil.”
• POSTERS– Broer et al “Biomass Gasification and Syngas Clean‐up;”Broer et al., Biomass Gasification and Syngas Clean up;
– Brown et al., “Technoeconomic Analysis of Biobased Chemicals Production via Integrated Catalytic Processing;”
– Creager et al., “High Pressure, Oxygen Blown Entrained‐Flow Gasification of Bio‐oil;”
– Del Campo et al., “Characterization of Fast Pyrolysis Biochars for Safe Application;”
– Friend et al., “Co‐firing Pellets of Coal and Pyrolysis‐derived Binder;”
– Rover et al., “Analysis of Sugars and Phenolic Compounds in Bio‐oil;”
Sadula et al “Stability of Bio Oil as Measured by Gel Permeation Chromatography;”– Sadula et al., Stability of Bio‐Oil as Measured by Gel Permeation Chromatography;
– Wang et al., “Pyrolysis of Lipid‐rich Biomass for Fuel and Chemicals Production;”
– Whitmer et al., “Gas Cleaning Systems for Syngas and Bio‐oil Production;”
– Woolcock et al., “Analysis of Tar Compounds at Trace Levels in Cleaned Syngas;”, y p y g ;
– Zhang et al., “Techno‐Economic Analysis of Co‐Production of Hydrogen and Transportation Fuels from Corn Stover.”
Pyrolytic Pathways to Increasing the
Lignin-Derived Monomer / Oligomer
Ratio in Bio-oil
Dustin Dalluge
Iowa State University
Dustin Dalluge is a Graduate Research Assistant working toward his
Ph.D. in Mechanical Engineering and Biorenewable Resources & Technology
at Iowa State University under the supervision of Dr. Robert Brown.
Dustin received his A.S. in Engineering at North Iowa Area Community
College and B.S. in Mechanical Engineering from Iowa State University.
His research areas are in fast pyrolysis, feedstock pretreatments, bio-oil
fractionation, and pyrolysis fundamentals.
Pyrolytic Pathways to
Increasing the Lignin-
Derived Monomer/Oligomer
Ratio in Bio-oil
Dustin Dalluge, Department of Mechanical Engineering, Iowa State University
Dr. Robert C. Brown, Center for Sustainable Environmental Technologies, Iowa State University
Footer goes here2
tcbiomass2011
Project Goals
• Understand the formation of the lignin-derived, water
insoluble compounds in bio-oil
• Develop methods to produce bio-oil with a higher ratio of
monomers to oligomers
> Leads to bio-oil with improved properties for
upgrading to transportation fuels and chemicals
Footer goes here3
tcbiomass2011
Advantages of Phenolic Monomers
• Higher C/O ratio = higher energy density than carbohydrate monomers
• Many phenolic monomers are liquids and soluble in solvents
> Easier to upgrade than oligomers
> Less sticky and complex than oligomers
• Phenolic monomers could be upgraded separately from sugars
Footer goes here4
tcbiomass2011
Defining The Resource Base
23-53% Cellulose12-32% Lignin
12-25% Hemicellulose
Biomass
0.5-16% Ash
(Wyman, 1996.)
Footer goes here5
tcbiomass2011
Lignin Facts
• Enzyme-mediated polymerization of three precursors:
> p-Coumaryl alcohol (H-lignin)
> Coniferyl alcohol (G-lignin)
> Sinapyl alcohol (S-lignin)
Most common linkages:
β-O-4 (60%+) > α-O-4 > β-5 > β-1 > others
H O
O
OH
OH
O
O
C OH
OCH3
O
HO
O
O
HO
HO
OH
O O
OH
OH
OHHO
HOO
OCH3
HO
O OH
OH
OOHC
OCH3
HO
OH
OC
OH
O
OCH
HO
HO
O
OCH3HO
HO
OH
O
OCH3
O
OO
H
HOO
OCH3
3
3CH O
3CH O
3CH O
3CH O3CH O
3CH O
3CH O
3CH O
3CH O
3CH O
(Amen-Chen, 2001)
Footer goes here6
tcbiomass2011
High
Molecular
Lignin, 25%
Reaction
Water, 10%
Moisture,
10%GC-
detectable,
10%
Quantified via
GC, 30%
HPLC
detectable,
15%
Typical Bio-oil Composition
(Mohan, 2006)
Contains around 25% high molecular weight lignin-derived compounds
Also called “pyrolytic lignin” or “water insolubles”
Footer goes here7
tcbiomass2011
Formation of Lignin-Derived Oligomers
H O
O
OH
OH
O
O
C OH
OCH3
O
HO
O
O
HO
HO
OH
O O
OH
OH
OHHO
HOO
OCH3
HO
O OH
OH
OOHC
OCH3
HO
OH
OC
OH
O
OCH
HO
HO
O
OCH3HO
HO
OH
O
OCH3
O
OO
H
HOO
OCH3
3
3CH O
3CH O
3CH O
3CH O3CH O
3CH O
3CH O
3CH O
3CH O
3CH O
Approximate size of
lignin-derived oligomer
in bio-oil
Hypothesis 1:
Pyrolysis does not sufficiently
depolymerize lignin
Hypothesis 2:
Monomers formed during
pyrolysis repolymerize in
secondary reactions or after
condensation of vapors
(Piskorz, 1999)
Footer goes here8
tcbiomass2011
Vinyl Phenols in Bio-oil
• Vinyl phenols appear as the most dominant products in
primary lignin pyrolysis (GC-MS analysis of vapors)
• Very few vinyl phenols are found in the analysis of bio-oil
(condensed vapors and aerosols)
4-vinylphenol 2-methoxy-4-
vinylphenol
2,6-dimethoxy-4-
vinylphenol
(Patwardhan, 2011)
Footer goes here9
tcbiomass2011
Polymerization of Vinyl Phenols
• Vinyl phenols formed during lignin pyrolysis will be very
reactive
• Gas-phase polymerization of vinyl phenols with other
phenol monomers will produce phenolic oligomers
• Phenolic oligomers have little vapor pressure and will
condense to aerosols in the gas flow
Footer goes here10
tcbiomass2011
Disappearance of Vinyl Phenols
Theoretical Yield1
(wt% of lignin)
Actual Recovery inCorn Stover Bio-oil2
(wt% of lignin)
Total Monomeric
Phenols20.0% 10.2%
Vinyl Phenols 7.3% 0.0%
1 – Primary products of corn stover lignin micropyrolysis (Patwardhan, 2011)
2 – Recovery of whole bio-oil from a fluidized bed reactor (Pollard, 2009)
– Assuming lignin content of 18.7% in corn stover (U.S. DOE)
Note: Corn stover bio-oil also contained 12.5% water insolubles based on biomass
weight (24.8 wt.% of bio-oil) – consisting largely of phenolic oligomers
Footer goes here11
tcbiomass2011
Phenolic Monomer Recombination
0
5
10
15
20
25
30
35
0 1000 2000
Are
a %
Mol. Wt. (Da)(Relative to Polystyrene standards)
CCS Mix
CCS Mix Py
CCSA Mix Py
• Gel Permeation Chromatography (GPC) analysis on mixtures of lignin precursors
─ Coniferyl alcohol, coumaryl alcohol and sinapyl alcohol (CCS Mix)
─ Condensed products from pyrolysis of at 500oC of CCS Mix (CCS Mix Py)
─ Condensed products from pyrolysis at 500oC of CCS Mix and acetic acid. (CCSA Mix Py)
• Pyrolysis of precursors promotes polymerization
• Polymerization greatly enhanced by acetic acid
(Patwardhan, 2011)
Footer goes here12
tcbiomass2011
Attempts to Neutralize Acid
Catalysts
• Hypothesis: Rapid conversion of acetic acid to acetate will
prevent it from reacting with phenols released during
pyrolysis
• Test: Pyrolyze lignin in an arrangement that forces
pyrolysis vapors through a shallow bed of basic salt
(NaHCO3 or KHCO3); Analyze vapors via GC-MS
Footer goes here13
tcbiomass2011
Experimental Setup
Lignin
Quartz Wool
Quartz Wool
Suspended
Salt Bed
Salt Bed
Lignin
Quartz Wool
Control
Vapors
Footer goes here14
tcbiomass2011
0
2000
4000
6000
8000
10000
12000
14000
16000
18000
20000A
rea
(uV
*m
in)
Control
NaHCO3 Suspended Bed
KHCO3 Suspended Bed
Changes in Phenol Composition by Passing Lignin
Pyrolysis Vapors Through Bed of Basic Salt
• Complete neutralization of
acetic acid
• Increased phenolic monomers
• Vinyl phenols reduced
Footer goes here15
tcbiomass2011
What is the Effect of Directly Mixing
Alkali and Alkaline Earth (AAEM) Salts
with Lignin Prior to Pyrolysis?
• Previous research at ISU has shown that cations strongly
increase ring fragmentation during cellulose pyrolysis
• Experiments were conducted with several AAEM cations:
Potassium, sodium, calcium, and magnesium
• Experiment – To look at effect of only the cations, the
anion was kept constant as acetate, which forms a
thermally unstable salt, susceptible to decomposition
during pyrolysis
Footer goes here16
tcbiomass2011
Experimental Setup
Lignin +
1wt% Salt
Quartz Wool
Lignin/Salt
Mixture
Lignin
Quartz Wool
Control
Vapors
Footer goes here17
tcbiomass2011
0
5000
10000
15000
20000
25000
Are
a (µ
V*
min
)
Lignin
Magnesium Acetate
Calcium Acetate
Potassium Acetate
Sodium Acetate
Effect of Cations on Lignin Pyrolysis
Presence of ~1.0wt% K or Na
almost doubled the yield of
phenol, 2-methoxyphenol, and
2,6-dimethoxyphenol
Footer goes here18
tcbiomass2011
What is the Effect of Thermally
Stable Salts on Lignin Pyrolysis?
• Previous research at ISU has discovered that thermally
stable AAEM salts are less prone to cause ring
fragmentation during cellulose pyrolysis
• Do thermally stable salts also affect lignin
depolymerization?
• Experiment: Infuse 1.0 wt% potassium in the form of
several salts into corn stover lignin – pyrolyze
Footer goes here19
tcbiomass2011
0
5000
10000
15000
20000
25000
Are
a (µ
V*
min
) Control
Sulfate
Chloride
Acetate
Lignin does not as readily depolymerize
in the presence of thermally stable salts
Potassium sulfate has no
significant effect on lignin
pyrolysis
Footer goes here20
tcbiomass2011
Two Approaches
1. Alkali, especially potassium, has shown to effect the
pyrolysis products of lignin when infused into the solid
2. Adding a bed of alkali salt above the lignin is shown to
dramatically shift the products toward recoverable
monomeric phenols
> Experiment: Test the effect of alkali on solid lignin as well as the
pyrolysis vapors – test the effect of the two simultaneously
Footer goes here21
tcbiomass2011
Experimental Setup
Lignin +
1wt% Salt
Quartz Wool
Lignin/1%K
(Kacetate) Mixture
Lignin
Quartz Wool
Quartz Wool
KAcetate
Suspended
Bed
Salt Bed
Lignin
Quartz Wool
Control
Lignin +
1wt% Salt
Lignin/1%K
(Kacetate) +
Kacetate
Suspended Bed
Quartz Wool
Quartz Wool
Salt Bed
Footer goes here22
tcbiomass2011
0
5000
10000
15000
20000
25000A
rea
(uV
*m
in)
Control
Lignin/1%K (Kacetate) Mixture
KAcetate Suspended Bed
Lignin/1%K (Kacetate) + KAcetate Suspended Bed
Cumulative Effects of Alkali
• Decreased yield of reactive vinyl phenols
• Increased yield of recoverable
monomeric phenols
Footer goes here23
tcbiomass2011
Could AAEM Influence Char
Formation During Lignin Pyrolysis?
• Thermally stable salts shown to substantially reduces the catalytic activity of AAEM during pyrolysis.
• Thermally unstable salts increase production of phenolic monomers
• Thermally stable salts have little to no effect on lignin pyrolysis
• We hypothesize that (thermally unstable) AAEM found in biomass contributes to the relative ease of pyrolyzing the lignin contained in the biomass compared to pure lignin pyrolysis
Footer goes here24
tcbiomass2011
Circumstantial Evidence That AAEM Plays
an Important Role in Lignin Pyrolysis
Top: Photos of char from
attempting to pyrolyze lignin in a
fluidized bed reactor (Nowakowski
et al., 2011)
Bottom: Agglomerated bed after
a test with sulfuric acid pretreated
red oak and resulting char
fragments
Footer goes here25
tcbiomass2011
Conclusions
• Vinyl phenols are very reactive and contribute to
formation of lignin oligomers
• Acids tend to catalyze the polymerization of phenolic
monomers in the vapor phase
• Alkali and alkaline earth metals (AAEM) inherent in
biomass appear to have significant effects on lignin
pyrolysis
• Thermally stable AAEM salts significantly reduce release
of phenolic monomers during pyrolysis
Footer goes here26
tcbiomass2011
Acknowledgments
• ConocoPhillips Company’s sponsorship of this study
• CSET staff – Patrick Johnston, Marge Rover, Sunitha
Sadula, Ryan Smith
• Fellow graduate students at CSET
• CSET undergraduate hourly research assistants
Footer goes here27
tcbiomass2011
Footer goes here28
tcbiomass2011
References
Amen-Chen, C., H. Pakdel, et al. (2001). "Production of monomeric phenols by thermochemical
conversion of biomass: a review." Bioresource Technology 79(3): 277-299.
Bayerbach, R. and D. Meier (2009). "Characterization of the water-insoluble fraction from fast
pyrolysis liquids (pyrolytic lignin). Part IV: Structure elucidation of oligomeric molecules." Journal of
Analytical and Applied Pyrolysis 85(1-2): 98-107.
J. Piskorz, P. M., D. Radlein (1999). Pyrolysis of Biomass - Aerosol Generation: Properties,
Applications, and Significance for Process Engineers. Biomass - A Growth Opportunity in Green
Energy and Value-Added Products. R. P. O. a. E. Chornet. Oakland, California, Elsevier Science
Ltd. 2: 1153-1167.
Mohan, D., C. U. Pittman, et al. (2006). "Pyrolysis of Wood/Biomass for Bio-oil: A Critical Review."
Energy & Fuels 20(3): 848-889.
Mullen, C. A. and A. A. Boateng (2011). "Characterization of water insoluble solids isolated from
various biomass fast pyrolysis oils." Journal of Analytical and Applied Pyrolysis 90(2): 197-203.
Patwardhan, P. R. (2010). Understanding the product distribution from biomass fast pyrolysis.
Doctor of Philosophy, Iowa State University.
Pollard, A. J. S. (2009). Comparison of bio-oil produced in a fractionated bio-oil collection system.
Master of Science, Iowa State University.
USDOE (2004). Biomass Feedstock Composition and Property Database. U. S. D. o. Energy. 2011.
Wyman, C. (1996). Handbook on bioethanol: production and utilization, Taylor & Francis.
Footer goes here29
tcbiomass2011
Lignin Pyrolysis
> Organosolv corn stover lignin was obtained from Archer
Daniels Midland company
> Lignin was washed in 0.1N hydrochloric acid to remove
mineral impurities (>0.1% ash remaining) and subsequent
DI water washes
> The dried lignin was then impregnated with 1 wt% cation
in the form of potassium, sodium, magnesium, and
calcium acetate to represent common biomass cations
Footer goes here30
tcbiomass2011
Methodology
> A Frontier Laboratories PY-2020iS furnace type pyrolyzer
fitted with an A3-1020E autosampler was used for
analytical pyrolysis
> A Brucker 430-GC with an FID detector was used to for
quantification of select lignin pyrolysis products
> Select products were identified using an identical
instrument interfaced to an MS and were then checked on
the FID system with chemical standards
> GC column used was a VF-1701ms
> 0.25mm ID, 60 meter length
Footer goes here31
tcbiomass2011
Micropyrolyzer Setup