entrained flow gasification of wood pyrolysis oil
TRANSCRIPT
Muhamad Fazly Abdul Patah ([email protected]) & Professor Shusheng Pang Department of Chemical and Process Engineering, University of Canterbury, Christchurch, New Zealand
ENTRAINED FLOW GASIFICATION OF WOOD PYROLYSIS OIL
BACKGROUND: Densification of biomass into pyrolysis oil for gasification is gaining increasing interest as a possible alternative to solid biomass gasification. The major advantages of this approach include costs reduction for transportation and storage, easier feedstock handling as well as easier feeding of the bio-oil for pressurized gasification system. In this project, woody biomass from local radiata pine trees is converted into pyrolysis oil where the oil is then sprayed into a high temperature entrained flow gasifier which was recently developed and commissioned at this university. The oil is atomized into fine droplets as it enters the gasifier; before going through series of conversion reactions to yield hydrogen and carbon monoxide rich gas product.
REACTOR DESIGN: 1. The construction material of the reactor is mainly 253MA stainless
steel alloy which is highly resistant to high temperature corrosion attack; including within oxygen-deficient environment.
2. The intended operation conditions of the reactor is at atmospheric pressure and maximum temperature of 1100oC.
3. The system is pre-heated to set operation temperature using LPG. 4. During gasification, gasifier temperature is solely controlled by the
extent of pyrolysis oil combustion. 5. Fine pyrolysis oil droplets are generated by an external mix gas-
assisted atomizer. This type of atomizer enables safe and independent control of pyrolysis oil and oxygen gas flow rates during gasification.
DISCUSSIONS: 1. Gasification temperature is strongly influenced by flow rates of
pyrolysis oil and oxygen gas. Nevertheless the effect of oil flow rate is found to be more dominant.
2. H2 and CO contents in the producer gas are the highest when the equivalent ratio is set to 0.3. At this condition CO2 yield is the lowest, suggesting an optimum condition for gasification.
3. H2 and CO yields decrease when equivalent ratio is reduced below 0.3, potentially due to poor oil atomization at low oxygen flow rate. This suggests significant influence of spray characteristics on oil-to-gas conversion and the subsequent reactions pathways.
4. As the equivalent ratio increases to values greater than 0.3, oxidation reactions of gasification products become more significant thus causing progressive increase in CO2 on the expense of useful gaseous products such as H2, CO and CH4 gases.
GASIFICATION SYSTEM:
CHALLENGES: The main challenges in the operation of entrained flow gasification of pyrolysis oil are related to the large variations in pyrolysis oil properties; especially its viscosity and moisture content. Significant change in oil properties could deteriorate atomization performance, interrupt flow stability and in many occasions completely block the oil feeding system.
CONCLUSIONS: There are strong influences of pyrolysis oil and oxygen gas flow rates on the equivalent ratio, gasification temperature and oil atomization characteristics. These factors have significantly affect the quality of syngas generated during gasification operation.
Research on this project is on-going and more gasification runs are planned in the future to further investigate influence of different parameters on gasification products and performance.
RESULTS:
Figure 1: Influence of equivalent ratio (ER) on gasifier average temperature at constant oil flow rate
Figure 2: Influence of oil flow rate and ER on gasifier average temperature
ACKNOWLEDGEMENTS: 1. Ministry of Business, Innovation and Employment New Zealand 2. Ministry of Higher Education Malaysia
1. Entrained flow reactor 4. Syngas after-burner 7. Atomizing gas (oxygen) 2. Atomizer and cooling jacket 5. Oil filter and flow meter 8. PLC and control panel 3. Gas burner 6. PT and display 9. Peristaltic pump
ER: 0.21
ER: 0.31
ER: 0.38
ER: 0.45 ER: 0.54 ER: 0.36
500
600
700
800
900
1000
700 800 900 1000 1100 1200 1300 1400 1500
Tem
per
atu
re (
C)
Oxygen Gas Flow Rate (L/h)
High Oil Flow (60 ml/min)Low Oil Flow (35 ml/min)
Figure 3: Syngas compositions at various ER (N2, He and H2O free)
0
10
20
30
40
0.00 0.20 0.40 0.60
Mo
l % H
2
Equivalent Ratio (ER)
Run 1Run 2Run 3
0
10
20
30
40
0.00 0.20 0.40 0.60
Mo
l % C
O
Equivalent Ratio (ER)
Run 1
Run 2
Run 3
0
20
40
60
80
0.00 0.20 0.40 0.60
Mo
l %C
O2
Equivalent Ratio (ER)
Run 1
Run 2
Run 30
2
4
6
0.00 0.20 0.40 0.60
Mo
l % C
H4
Equivalent Ratio (ER)
Run 1
Run 2
Run 3
PYROLYSIS OIL PROPERTIES: Physical Properties: Elemental compositions: Water Content: 15 – 40 wt% C = 27 – 42 wt% Viscosity: 9 – 500 cSt @ 20oC H = ~ 8 wt% pH: ~ 3 O = 50 – 65 wt%
1
2
3
4
5 6
7
8
9
300
400
500
600
700
800
900
1000
0.20 0.25 0.30 0.35 0.40
Tem
per
atu
re (
C)
Equivalent Ratio (ER)
Run 1
Run 2
Pyrolysis oil
tank
LPG tankOxygen
tank
Helium
tank
LPG
burner air
RR
Peristaltic
pump
Raw syngas
Pressure
regulator
Rotameter
Flame
Arrester
Safety valve
Stirrer motor
T
T
T
P
T
Purge
SPE columns
Syringe
R
Pressure
regulator
Solenoid
valve
O2
Oxygen
sensor
R
Gas sampling
port
To after-burner
P
Flow
meter
P
P
T
P
T
Solenoid
valve
T
Ethanol
tank
Pressure
regulator
Rotameter
Cooling
water
inlet
Cooling
water
outlet
Pressure
regulator
PG
PG
PG
Purge Air
Check valve
Peristaltic
pump
Ash filter
Entrained
Flow Gasifier
T
Nitrogen
tank
R
Drain
Tank
filter
Rotameter
Check valve
Characterization of tar and soot formation for an improved co-gasification of black liquor and pyrolysis oil Albert Bach Oller, Kawnish Kirtania, Erik Furusjö, Kentaro Umeki
Energy Engineering Subject, Division of Energy Science, Luleå University of Technology, Sweden. E-mail: [email protected]
Summary Addition of PO in BL is correlated with much lower yields of tar and negligible formation of soot. Addition of PO significantly increases the syngas yield. Addition of PO does not worsen carbon conversion, increase in particle size does.
Major questions Does addition of PO into BL affect … ? Formation of tar and soot. (Undesired for syngas upgrading) Syngas yield. (Essential to evaluate thermal efficiency) Carbon conversion (Crucial to recover Na&S into pulping processes
Background Black liquor gasification (BLG) is a proven technology for syngas production due to
Na catalyst. Alternative to recovery boilers in pulping industry. Economic performance of BLG is expected to improve by mixing pyrolysis oil (PO)
into BL[1] Char reactivity of BL/PO blends was high [2] Pilot-scale co-gasification of BL/PO is on-going (Jafri et at. , separate poster)
1. LTU biosyngas program (Swedish Energy Agency and industrial partners)
2. Swedish Centre for Biomass Gasification (SFC)
Acknowledgments
2.25 mm
1.Blend fuels (weight basis) 2.Spread over a thin surface
(<1mm) 3.Dry at 105 °C for 12 h 4.Grind and sieve
Only alkali salts are found in the filter. Negligible soot formation
9.5MJ/kg Na: 20.6.% 17.1MJ/kg
Tar formation is reduced with the: Addition of PO Temperature increase Decrease in particle size
Light aromatics are the main compounds in the tar
95% confidence level
Temperature significant?
Sample significant?
Mixed significant?
Solid yield yes yes no CO+H2 yield yes yes no Mass balance no no no
Temperature and sample effects are independent and significant
Syngas (CO+H2) does not come at thermodynamic equilibrium at 800 °C
Sample preparation
90-200µm 500-630µm
Drop tube furnace(DTF)
For small particles.
Conditions Reactor: I.D. 54 mm, length 2.3 m Gas flow: 5 L min-1 (5vol.%CO2+N2) Feeding rate: 6.8 g/h Parameters (Factorial design) Temperature: 800-1400 °C Particle size: 90-200, 500-630 µm Sample: BL, BP20 (20wt.% PO+80wt.% BL)
and BP40 (40wt.% PO+60wt.% BL) Measurement Solid residue in char bin & cyclone
(C, H, N, S, Na, K, O) Soot + condensed ash vapors, glass filter Tar collection in iso-propanol at -50 °C,
quantified by GC-MS-FID Gas composition in µGC H2O from WGR equilibrium Mass and carbon balances Mass balance: Average 101.6 St.dev.:16.1 Carbon balance: Average 98.4 St.dev.:13.5
Syngas
Tar
Materials found on Filter
Char Mass map
Analysis of variance
Syngas yield (CO+H2) is increased with; Addition of PO Temperature
H2:CO ratio is decreased with; Addition of PO Temperature
n ≥3
[1] Andersson J, Lundgren J, Furusjö E, Landälv I. Co-gasification of pyrolysis oil and black liquor for methanol production. Fuel 2015;158:451-459 [2] Bach-Oller A, Furusjö E, Umeki K. Fuel conversion characteristics of black liquor and pyrolysis oil mixtures: Efficient gasification with inherent catalyst. Biomass and Bioenergy 2015;79:155–165.
References
Particle size is crucial for carbon conversion 500-630µm particles show unconverted carbon even at 1400 °C Ash evaporates between 1000-1400 °C
0.00
0.02
0.03
0.05
0.07
0.08
0.10
0.12
0.13
500150025003500
Abso
rbance
Wavelength (cm-1)
B
D
A
C
A B C D
C (%wtdry) 68.63 83.63 48.03 87.68
H (%wtdry) 0.33 0.98 0.89 0.98
N (%wtdry) 0.83 0.23 0.25 1.98
S (%wtdry) 2.37 1.86 1.31 0.78
LHV (MJ/kgdry) 23.04 31.26 14.33 25.95
Ash (%wtdry) 27.84 13.54 49.51 8.68
Proximate Analysis
Small scale gasification plants authorized in South Tyrol in the last 4 years
2012
2013
2014
2015
CHARACTERIZATION OF BIOMASS CHAR FROM REAL SCALE GASIFIERSAND ASSESSMENT OF POSSIBLE UTILIZATION PATHWAYSJ. Ahmad, F. Patuzzi, D. Prando, S. Vakalis, L. Fiori, D. Chiaramonti, A. M. Rizzo, M. Baratieri*
Bioenergy & Biofuels LAB Free University of Bozen-Bolzano, Faculty of Science and Technology
Char in South Tyrol*corresponding author: *[email protected]
• 36 small scale biomass gasification plants in South Tyrol
• About 2600 tons/year of char disposed of as a waste
• High cost for disposal (total of approximately 373 k€ per year)
BET Surface Area
which possible applications?
A
B
C
D
• as catalyst• as land fill • as adsorbent • as fertilizer• co-firing in power plants
L1 and L2 are produced in lab at 600 °C and 900 °C. A,B,C and D designate the four different technologies (real scale plants).The other points refer to literature data.
Conclusions
Aromatic (C-H)900-675
• Char is mainly composed of carbon and ash.• LHV results show a high energy content in char.
However, the high ash content could be an issue during char combustion.
• Char surface area is higher than some synthetic catalysts such as impregnated NiO with CeO2 (i.e. 2.16 m2/g).
• Char can be considered as a catalyst or catalyst support, further analyses are required.
• ATR analyses show similar spectra for char C and D, which are produced with similar biomasses and similar technologies. Some analogies can be found in the ATR spectrum of char A, produced with a similar biomass but a different technology. The ATR spectrum of char B, produced with a different biomass typology and a different technology, shows alkyne peak only but not aromatic peaks.
Aromatic (C-C)1500-1400
Alkyne (C≡C)2400-2200
ATR analysis
300 400 500 600 700 800 900 10000
100
200
300
400
500
600
700 N2
Air
Steam
Real scale
Lab
Surf
ace
area
Temperature
L1 L2
AD
BC
BET comparison with hard wood
300 400 500 600 700 800 900 10000
100
200
300
400
500
600
700 N2
Air
CO2
Real Scale
Lab
Surf
ace
area
Temperature
L1 L2
A
B
BET comparison with soft wood
C
D
Surf
ace
are
a (
m²/g
)
Surf
ace
are
a (
m²/g
)
Temperature (°C)Temperature (°C)
Fuel wood chips pellet wood chips wood chips
Char extraction after HE* after HE* before HE* before HE*
Reactor downdraft rising co-current
downdraft downdraft
* HE = heat exchanger
0.2
0.4
0.6
0.8
1.0
0 5 10
Hyd
roch
ar
yie
ld (
-)
Time (h)
Experimental values
Model prediction
180 °C
220 °C
250 °C
EXPERIENCES ONHYDROTERMAL CARBONISATIONKINETIC MODELLING BASED ON EXPERIMENTAL RESULTSD. Basso, F. Patuzzi, E. Weiss-Hortala, M. Baratieri*, L. Fiori
Bioenergy & Biofuels LAB Free University of Bozen-Bolzano, Faculty of Science and Technology
The HTC process
*corresponding author: *[email protected]
• Thermochemical conversion process
• Wet biomass in inlet (H2O > 60%)
• Temperature: 180 - 250°C
• Pressure: 10 - 40 bar
H2O liquid (hot pressurized water)
• Residence time: 1 - 8 h
• Hydrochar yields ≈ 50 – 80 %56% (250°C, 8 h) 77% (180°C, 1 h)
• HHV increase ≈ 20%23.6 28.0 MJ/kg
• Carbon enrichment ≈ 20%54.4% 65%
• H content unchanged (6.6%)
• O content decrease ≈ 35%34.2% 22.2%
hydrochar
OFMSW Agricultural/ Forest wastes
Sewagesludge
SlurriesManure
Poultrymanure
Design parameters:T = 300 °C P = 140 barV = 50 ml Material: AISI 316
The HTC experimental apparatus
volatiles (gas + liquid)
solid
products
A B
V2V1
k1
kV1 kV2
Ck2
,
0, 1 2exp 1,2, ,a i
i i
Ek k i V V
RT
Kinetic scheme
- Di Blasi and Lanzetta, 1997- Prins, M.J. et al., 2006
1 2 V1 V2
k0 [s-1] 3.34·107 1.10·1010 9.15·106 1.55·1010
Ea [kJ/mol] 94.5 139.7 93.7 146.2
1 2 V1 V2
1.30·107 4.61·1010 7.22·106 3.86·109
90.9 143.4 92.8 138.0
1 2 V1 V2
3.05·107 1.33·1010 7.83·106 2.45·1010
98.6 146.6 96.0 142.1
Experimental results & Modelling
Grape marcGrape seeds Off-specification compost
0.5
0.6
0.7
0.8
0.9
1.0
0 5 10
Hyd
roch
ar
yie
ld (
-)
Time (h)
Experimental values
Model prediction
180 °C
220 °C
250 °C
0.5
0.6
0.7
0.8
0.9
1.0
0 5 10
Hyd
roch
ar
yie
ld (
-)
Time (h)
Experimental values
Model prediction
180 °C
220 °C
250 °C
Improved Tar Reforming in Producer Gas on Fe and CaO promoted Ni catalyst Tinku Baidya and Robert Cattolica
Department of Mechanical and Aerospace Engineering, University of California San Diego, La Jolla, CA 92093
A new catalyst with Ni-Fe-CaO components was impregnated on Carbo HSP as gasifier bed material
and optimized the composition
Roles of Fe and CaO were investigated resulting in high tar conversion and better coke resistance of
Ni catalyst
Stability of the optimized catalyst was investigated under time on stream for 96 hrs
Gas mixing chamber
Tar tank(Toluene)
Water tank
High pressure water pump Electric heater
MicroGC
Condensers
Furnace Catalyst bed
Flow controllers
Reactor bypass
Pressure transducerThermocoupleCheck valveOn/off valves3-way valveHeat trace
H2
N2
CO
CO2
CH4
O2
C2H4
Inert (N2)
Vent
Vent
Figure 1: Schematic representation of the steam reforming unit with a fixed-bed reactor
Abstract
Materials and Methods
Catalyst Characterization
Figure 2. XRD patterns of freshly reduced (a) 1%Ca, (b) 1.6%(Ni60Ca40), (c) 1.6%(Ni45Fe15Ca40), (d) 1.6%(Ni35Fe25Ca40), (e) 1.6%(Ni20Fe40Ca40), (f) 1%(Fe60Ca40) and (g) spent catalyst of 1.6%(Ni40Fe20Ca40) ; * indicates Ni-Fe alloy peak
Pure Ni and Fe peaks vanished as Ni/Fe ration decreased, with gradually increasing intensity of Ni-Fe alloy at ∼ 40o
The alloy phase was detected to be FeNi3 (indicated by *)
No Ca oxide related peak was detected indicating high dispersion on surface
Catalytic Results
The results confirm that both Fe and CaO promoted Ni catalyst resulting in higher tar conversion activity
Coke resistance power of Ni-Fe-CaO catalyst is much better than Ni or Ni-Fe alone catalysts
Conclusions
Fe promotes activity of Ni only at lower Fe/Ni ratio (Fig 5a)
CaO promotes Ni-Fe for all ratios (Fig 5b)
The morphology of fresh sample indicated fine dispersion of Ni-Fe-CaO catalyst components on the support
Reduced metals are not visible
The visible compounds seems to be of non-crystalline Ca related oxides
Deactivation studies: Time on Steam
Reaction Procedure & Conditions
Feed: H2, N2, CO, CO2, CH4, C2H4, and toluene (model tar).
Reaction temperatures: 700, 750, and 800°C
Flow rate of producer gas: 433 sccm; catalyst loading: 2 g (+20 g Quartz bead).
SEM
Figure 4. SEM images of freshly reduced and spent catalyst of 1.6%(Ni40Fe20Ca40)/Carbo HSP
It was difficult to distinguish metal particles in the catalysts
The particle size and dispersion in the fresh catalyst and spent catalysts indicated minor change comparatively
The approximate size of the metal particles were ∼ 100 nm
TEM
Figure 5. TEM images of freshly reduced and spent catalyst of 1.6%(Ni40Fe20Ca40)/Carbo HSP
Figure 6. Comparison of Tar removal efficiency in (a) Ni vs Ni-Fe; (b) (Ni-Fe vs Ni-Fe-CaO
Role Fe and CaO
Toluene conversion was reduced by from ∼100% to 80% in 48 hrs and after regeneration of catalyst, it goes back to stable 80% in 24 hrs
The Ni-Fe catalyst without CaO was deactivated fast with few hrs of operation
Figure 7. Time-on-Stream & Regeneration study over 1%(Ni65Fe35Ca0), 1.6% (Ni45Fe15Ca40) and 1.6% (Ni40Fe20Ca40) catalysts
Catalyst on CARBO C7H8 (%) C2H4 (%) CH4 (%) COx Sel (%) H2/CO TOF X 103(s-1) on Ni
1.5 %Ni 44 40 3* 77 0.88 1.8
1.6%(Ni60Ca40) 92 88 7 83 1.28 3.9
1.6%(Ni45Fe15Ca40) 80 71 2* 96 1.32 5.6
1.6%(Ni40Fe20Ca40) 68 64 2* 98 1.19 5.5
1.6%(Ni35Fe25Ca40) 54 45 2* 89 1.20 5.6
1.6%(Ni20Fe40Ca40) 31 28 3* 73 1.07 6.2
1.6%(Fe60Ca40) 20 19 4* 60 0.82 -
1%(Ni65Fe35) 40 37 2 86 0.92 3.4
1%Fe 18 25 4* 52 0.72 - Table: C7H8, C2H4 and CH4 conversion efficiency, COx selectivity and H2/CO ratio over Ni-Fe-CaO/Carbo catalysts (750 0C); * indicate increase in CH4 conc.
Optimization of catalyst composition With constant loading of CaO,
Ni/Fe ratios were changed to reach the best composition
C7H8 conversion decreases with decreasing Ni/Fe ratios, but TOF remains almost constant on Ni
COx selectivity is optimum when Ni/Fe ratio remains between 2 and 3
XRD TPO
Figure 3. TPO of spent 1.6%(Ni40Fe20Ca40) catalyst
Low intensity peaks for coke oxidation as indicated by noise level signal shows catalyst is more resistant to deactivation by coke formation in the catalyst bed
Catalyst Preparation: Wet impregnation of Ni(NO3)2.6H2O, Fe(NO3)3.9H2O and Ca(NO3)2.6H2O salts on support. Catalyst support: Carbo HSP (containing mixture of Al2O3, SiO2, TiO2 and Fe2O3) Catalyst nomenclature: For example, 1.6%(Ni40Fe20Ca40)/Carbo HSP catalyst had Ni, Fe and CaO in weight proportions of 40%, 20% and 40% respectively.
CFD Modeling and Analysis of a Dual Fluidized-Bed Biomass Gasifier Hui Liu, Robert Cattolica, and Reinhard Seiser
Department of Mechanical and Aerospace Engineering, University of California San Diego, La Jolla, CA 92093
Abstract A three-dimensional CFD model is developed to simulate the complete-loop of a pilot-scale (6 tons/day, 1 MWth) power plant based on a dual fluidized-bed system which consists of a gasifier, a combustor, a cyclone separator, and a loop-seal. In this model, the gas phase is described by the Large Eddy Simulation (LES), while the particle phase is described by the Multiphase Particle-In-Cell (MP-PIC) method. The momentum, mass, and energy transport equations are integrated with the reaction kinetics to simulate biomass pyrolysis, heterogeneous char reactions, and homogenous gas-phase reactions in the dual fluidized-bed system. The reliability and accuracy of the CFD model are tested at different operating conditions. The producer gas composition and reactor temperature are predicted by this model and are validated with experimental data from the pilot plant operations.
Case Case 1 Case 2 Case 3 The biomass feed rate (kg/h) 228.04 219.55 243.04 The steam supply to the gasifier (kg/h) 85.32 78.21 64.71 The 1st air supply to the combustor (kg/h) 28.89 40.85 26.52 The 2nd air supply to the combustor (kg/h) 240.68 283.37 238.01 The 3rd air supply to the combustor (kg/h) 305.63 312.62 322.76 The steam supply to the cyclone separator (kg/h) 78.21 85.32 64.71
Proximate analysis (wt. %) Moisture Fixed carbon Volatile matter Ash 5.19 20.19 72.53 2.09 Ultimate analysis (wt. %) dry –ash-free C H O Residual 50.96 5.94 42.59 0.51
00.050.1
0.150.2
0.250.3
0.350.4
H2 CO CO2 CH4 C2H4 C2H6
Mol
e Fr
actio
n
Simulation Results_Case 1Experimental Data_Case 1
500550600650700750800850900
Gasifier_bottom Gasifier_bed Gasifier_upper
Tem
pera
ture
(◦C
) T prediction_Case 1T data_Case 1
00.050.1
0.150.2
0.250.3
0.350.4
0.45
H2 CO CO2 CH4 C2H4 C2H6
Mol
e Fr
actio
n
Simulation Results_Case 2Experimental Data_Case 2
500550600650700750800850900950
Gasifier_bottom Gasifier_bed Gasifier_upper
Tem
pera
ture
(◦C)
T prediction_Case 2T data_Case 2
00.050.1
0.150.2
0.250.3
0.350.4
H2 CO CO2 CH4 C2H4 C2H6
Mol
e Fr
actio
n
Simulation Results_Case 3Experimental Data_Case 3
500550600650700750800850
Gasifier_bottom Gasifier_bed Gasifier_upper
Tem
pera
ture
(◦C
)
T prediction_Case 3T data_Case 3
600650700750800850900950
1000
Combustor_bottom Combustor_lower Combustor_upper
Tem
pera
ture
(◦C
)
T prediction_Case1T data_Case 1
600650700750800850900950
1000
Combustor_bottom Combustor_lower Combustor_upperTe
mpe
ratu
re (◦
C)
T prediction_Case 2T data_Case 2
600650700750800850900950
Combustor_bottom Combustor_lower Combustor_upper
Tem
pera
ture
(◦C
)
T prediction_Case 3T data_Case 3
Table 1: Reaction Kinetics
Figure 1: Pilot-Scale Dual Fluidized-Bed Gasification Power Plant
Figure 2: Full Scale CFD Model
Figure 3: Particle Circulation in Dual Fluidized-Bed System Figure 4: Transient Gas Distribution
Figure 5: Gas Distribution in the Steady-State
Conclusions A three dimensional CFD model was established to simulate biomass gasification in a dual fluidized-bed system. The model successfully predicted the bubbling fluidized-bed in the gasifier and the fast fluidized-bed in the combustor. The CFD model was validated at different operating conditions. The predicted gas composition and reactor temperatures showed good agreement with experimental data.
Table 3: Case Settings
Table 2: Biomass Properties
Steam Supply
1st Air Supply
2nd Air Supply
3rd Air Supply
Propane and
Additional Air Supply
Biomass
Producer Gas
Flue Gas
Steam Supply
ANALYSIS OF LIQUID-PHASE INTERMEDIATES GENERATED DURING
AQUEOUS-PHASE REFORMING OF SORBITOL ON PT/AL2O3 Lidia I. Godina, Alexey V. Kirilin
†, Anton V. Tokarev, Dmitry Yu. Murzin*
Laboratory of Industrial Chemistry and Reaction Engineering, Process Chemistry Centre,
Åbo Akademi University, FI-20500 Turku, Finland
Objectives
Reaction pathways
Analysis of liquid phase
Acknowledgements
The SusFuelCat project has received funding from the European Union's Seventh Framework Programme for research, technological development and
demonstration under grant agreement No 310490 (www.susfuelcat.eu).
10 15 20 25 300
2
4
6
8
10
12
14
10 15 20 25 30-14
-12
-10
-8
-6
-4
-2
030 35 40 45 50 55 60 65 70
0,0
0,2
0,4
0,6
0,8
1,0
30 40 50 60 70-1,0
-0,8
-0,6
-0,4
-0,2
0,0
Sorbitol, conversion 62%
mV
Galactitol, conversion 65%
mV
time [min]
Original data
Cumulative fit
Fitted peaks
Determined peaks
Galactitol
Original data
Cumulative Fit Peak
Sorbitol
Determined peaks
mV
Fitted peaks
mV
time [min]
Selectivity to intermediates [mol C %] for the experiments at conversion level 62% (sorbitol) and 65%
(galactitol)
Acids Gal Sorb Sugars / Sugar alcohols Gal Sorb Alcohols Gal Sorb
acetic acid 0.54* 0.82 D-(+)-arabitol 0.56 0.15 butane-1,2-diol 0.47 0.64
fumaric acid 0.00 0.69 D-(+)-glucose 0.003 0.11 ethanol 2.02 3.02
hexanoic acid 0.07 0.04 D-galactose 0.1 0.00 glycerol 1.36 1.75
lactic acid 0.15 0.1 meso-erythritol 1.12 0.95 hexane-1,2-diol 0.10 0.09
pyruvic acid 1.71 0.08 xylitol 0.99 1.05 hexane-2,5-diol 0.18 0.19
tartaric acid 0.00 0.03 Furans / Pyranes Gal Sorb methanol 6.96 5.20
2-acetylfuran 0.02 0.02 pentanol-1 0.06 0.06
Ketones Gal Sorb 2,5-dimethyltetrahydrofuran 0.06 0.04 pentanol-2 0.04 0.02
3-hydroxy-2-butanone 1.49 1.59 5-hydroxymethyl-2-furaldehyde 0.06 0.07 propane-1,2-diol 1.39 1.35
acetone 0.53 0.25 tetrahydrofurfuryl alcohol 0.45 0.17 propanol-1 1.45 1.74
1 – dehydrogenation of C-O bond
2 – hydrogenolysis of C-O bond
3 – hydrogenation of C=O bond
4 – hydrogenation of C=C bond
5 – decarbonylation
6 – cyclization
7 – pinacol-pinacolone rearrangement with
subsequent hydrogenation
8 – aldol/croton condensation or retro-aldol
reaction
9 – diastereomerization on Pt surface
10 – formation of gemi-diol
11 – water-gas shift reaction (WGS)
Aqueous-phase reforming (APR) is a promising technology of hydrogen or
alkanes production from renewable sources. Different substrates, such as alcohols,
sugars, polyols, or even aqueous fraction of pyrolytic bio-oil, could be converted at
moderate temperatures (190-250°C) and elevated pressures to maintain the liquid
phase. This catalytic process appeared to be more energy-efficient compared to
natural gas steam reforming [1].
Transformation of feed in APR proceeds through formation of many liquid-phase
intermediates and leads to formation the main gas products: alkanes, H2, and CO2.
Reaction pathways investigation is crucial for kinetic modeling and, particularly, for
catalyst development.
• HPLC (Agilent 1100) system
• Aminex HPX-87H column
• Refractive index detector
• Isocratic conditions:
• 45°C,
• flow rate 0.6 mL/min
• aqueous 5 mM H2SO4
• Spiking with 89 substances
• Peak fitting application
[1] Davda, R. R., Shabaker, J. W., Huber, G. W., Cortright, R. D. & Dumesic, J. A.. Appl.
Catal. B Environ. 56, 171–186 (2005)
[2] Godina, L. I., Kirilin, A. V., Tokarev, A. V. & Murzin, D. Y.. ACS Catal. 2989 (2015)
[3] Deutsch, K. L.; Lahr, D. G.; Shanks, B. H. Green Chem. 2012, 14, 1635.
[4] Liu, B. Catalytic Generation of Hydrogen and Chemicals from Biomass Derived Polyols;
ProQuest, (2008); p. 159.
†Dow Benelux B.V., P.O. Box 48, building 443, 4530 AA, Terneuzen, Netherlands
Experimental
20 40 60 80 1000
20
40
60
80
100
Car
bon
cont
ent (
%)
Conversion (%)
feedstock
Carbon distribution
versus conversion
0 20 40 60 80 100
60
80
100
120
140
Sorbitol
Galactitol
Sequentional experiment
Selectivity to H2
Sel
ectiv
ity [%
]
Conversion [%]
Selectivity
to CO2
0 20 40 60 80 100-5
0
5
10
15
20
25
30
Sorbitol
Galactitol
Sequentional experiment
Total selectivity to alkanes
Selectivity to CO
Sel
ectiv
ity [%
]
Conversion [%]
0 50 100 1500
20
40
60
80
100
conve
rsio
n [%
]
TOS [h]
Sorbitol (sorb)
Galactitol (gal)
Sorbitol (s+g)
Galactitol (s+g)
Conversion versus
time-on-stream
Selectivity to H2 and CO2 versus time-
on-stream
Selectivity to alkanes and CO versus
time-on-stream
Selectivity to
product X (%) =
n(CX )
n(Cgas )×100
Selectivity to H2 (%) =n(H2 )
n(Cgas )×
1
RR×100
C6H14O6 6CO + 7H2
+ 6H2O 6CO2 + 13H2; RR = H2/CO2 = 13/6
ν(H2) – moles of H2 formed
ν(Cgas) – moles of carbon in gas-phase products
ν(Cx) – moles of carbon in CO2 or alkane formed
Theoretical stoichiometry of H2, CO2, and alkanes formed:
0
0,02
0,04
0,06
0,08
0,1
0,12
0 50 100
alcohols
acids
carbonyles
polyols
sugars
heterocycles
Wood
Lignocellulosic
biomass
Paper
APR fermentation ethanol
pyrolysis
Bio-oil
aqueous
fraction
Animal fat Fuel
hydrolysis sugars polyols APR
APR
APR
hemicellulose sugars polyols APR
glycerol
Hydrogen
and alkanes
H2
N2 (1% He)
mass flow
controller furnace
sampling
GC analysis
P effluent
pressure
controller
safety valve
MFC
catalyst
MFC
MFC
collector
HPLC pump
feed solution
HPLC analysis
Total Organic Carbon
(TOC) analysis
• Commerciall 1% Pt/Al2O3 (125-250 μm)
• Reduction in situ (H2, 250°C, 2 hours)
• Stainless-steel continuous reactor
• Catalyst loading: 1 g
• 225°C, 29.7 bar
• Carrier-gas: N2 (1% He), 25 ml/min
*Detection limit strongly depends on the type of substance and can be in the range
8−800 mmol/L, assuming a signal-to-noise ratio (S/N) of 3
galactitol or sorbitol Pt/Al2O3
225°C, 29.7 bar N2
H2
CO2
alkanes
• Aqueous solutions of polyols: sorbitol (3.6 wt %), galactitol
(3.6 wt %), and a mixed experiment with sequential supply
of sorbitol, galactitol and their 1:1 mixture
• GC-analysis: online, micro-GC (Agilent Micro-GC 3000A)
• HPLC-analysis: offline
• TOC analysis: carbon balance ≈ 90-95%.
Results
Conclusions • Aqueous-phase reforming of two epimeric polyols was thoroughly
studied in order to introduce reaction pathways and reveal
influence of chirality [2]
• Spiking technique and peak fitting were implemented for the
samples obtained at different conversion levels, significantly
improving identification reliability compared to previous studies
[3,4]
• 27 peaks were identified, corresponding to up to 85 mol % of carbon
in the liquid-phase intermediates
• The results provide sufficient basis for a reliable reaction network
and subsequent kinetic modeling
• Diastereomerization of carbohydrates was proposed to explain the
formation of both glucose and galactose from sorbitol, and the
differences in selectivities to some primary intermediates (xylitol,
arabitol, glucose and galactose).
• Heptane formation was explained via aldol and croton condensation
• Substrate chirality was shown to have no influence on gas-phase
products distribution, and slightly changed concentration of some
intermediates
• Metal screening of Sibunit-supported mono- and bimetallic catalysts was performed in APR of xylitol for
the first time
• Catalytic tests revealed that Pt is an optimum active and selective hydrogen-generating metal among
carbon-supported bimetallic and monometallic catalysts: Pt, Ni, Re, Ru, Pt-Ni, Pt-Co, Pt-Re, Pt-Ru
• Selectivity to hydrogen decreased along with increasing xylitol conversion for all catalysts. Conversion
level has approximately no effect on total selectivity to alkanes for all catalysts
• Metal choice determines the alkane distribution.
• Ru, Ni and Re based catalyst exhibited the highest overall selectivity to alkanes among all catalysts,
however, they were not active or deactivated very fast (after 20 hours time-on-stream).
• Bimetallic Pt-Re catalysts showed an outstandingly high activity, however, it shifted reaction pathways to
alkanes production. Addition of Ru or Co worked in the same way, keeping the same activity level.
Performance of Pt-Ni catalyst is very close to a monometallic Pt.
AQUEOUS-PHASE REFORMING OF XYLITOL ON CARBON-SUPPORTED CATALYSTS Lidia I. Godina1, Alexey V. Kirilin
†, Anton V. Tokarev1, Yulia S. Demidova2,3, Jesus Lemus4, Luisa Calvo4, Tim Schubert5, Miguel A.
Gilarranz4, Irina L. Simakova2,3, Dmitry Yu. Murzin1* 1 Abo Akademi University, Laboratory of Industrial Chemistry and Reaction Engineering, Biskopsgatan 8, Abo, 20500, Finland
2 Boreskov Institute of Catalysis, Novosibirsk, 630090 (Russia) 3 Novosibirsk State University, Novosibirsk, 630090 (Russia) 4 Universidad Autonoma de Madrid, Madrid, 28049 (Spain)
5 FutureCarbon GmbH, Gottlieb-Keim-Strasse 60, 95448 Bayreuth, Germany
Objectives
Conclusions
Acknowledgements
The SusFuelCat project has received funding from the European Union's Seventh Framework Programme for research, technological
development and demonstration under grant agreement No 310490 (www.susfuelcat.eu).
Literature
(1) Davda, R. R.; Shabaker, J. W.; Huber, G. W.; Cortright, R. D.; Dumesic, J. A. Appl. Catal. B Environ. 2005, 56, 171–186.
(2) Huber, G. W.; Dumesic, J. A. Catal. Today 2006, 111, 119–132. †Dow Benelux B.V., P.O. Box 48, building 443, 4530 AA, Terneuzen, Netherlands
Selectivity to
product X (%) =
n(CX )
n(Cgas )×100
Selectivity to H2 (%) =n(H2 )
n(Cgas )×
1
RR×100
C5H12O5 5CO + 6H2
+ 5H2O 5CO2 + 11H2
ν(H2) – moles of H2 formed
ν(Cgas) – moles of carbon in gas-phase products
ν(Cx) – moles of carbon in CO2 or alkane formed
Theoretical stoichiometry of H2, CO2, and alkanes formed:
RR = H2/CO2 = 11/5
Catalyst Precursor Metal particle dm (TEM)
2.5 wt % Pt/AC* Pt(NO3)4
3 wt % Pt/C** H2PtCl6 1.0 nm
3 wt % Ni/C** Ni(NO3)2 9.2 nm
3 wt % Re/C** HReO4
3 wt % Ru/C** RuCl3 1.2 nm
3+3 wt % Pt-Ni/C** H2PtCl6 + Ni(NO3)2*** Bimodal distribution: Pt ≈ 20 nm, Ni ≈ 2 nm
3+3 wt % Pt-Re/C** H2PtCl6 + HReO4*** 1.3 nm
3+3 wt % Pt-Ru/C** H2PtCl6 + RuCl3*** 2.1 nm
3+3 wt % Pt-Co/C** H2PtCl6 + CoCl2*** 1.6 nm
• Stainless-steel continuous reactor
• Catalyst loading: 0.5 g, diluted 1:1 with sand
• Feed: 10 wt % aqueous solution of xylitol, 0.1-1.0 ml/min
• 225°C, 29.7 bar
• Carrier-gas: N2, 25 ml/min
• GC-analysis: online, micro-GC (Agilent Micro-GC 3000A)
• HPLC-analysis: offline, HPLC (Agilent 1100), Aminex HPX-87H
column (0.005M H2SO4, 45°C, 0.6 ml/min)
• TOC analysis: carbon balance ≈ 90-95%.
* support is powdered activated carbon
** support is Sibunit, 100-200 μm
*** co-impregnation
Results
0
20
40
60
80
100
0 5 10
Co
nve
rsio
n [
%]
WSHV [1/h]
Conversion vs WSHV
Pt/C (3%)
Ni/C (3%)
Pt-Ni/C (3+3%)
Re/C (3%)
Pt-Re/C (3+3%)
Pt-Ru/C (3+3%)
Ru/C (3%)
Pt/AC (2.5%)
Pt-Co/C (3+3%)0
20
40
60
80
100
0 20 40 60 80 100
Sele
cti
vit
y [
%]
conversion [%]
Selectivity to H2 vs conversion
Pt/C (3%)
Ni/C (3%)
Pt-Ni/C (3+3%)
Re/C (3%)
Pt-Re/C (3+3%)
Pt-Ru/C (3+3%)
Ru/C (3%)
Pt/AC (2.5%)
Pt-Co/C (3+3%)0
20
40
60
80
100
0 20 40 60 80 100
Sele
cti
vit
y [
%]
conversion [%]
Total selectivity to alkanes vs conversion
Pt/C (3%)
Ni/C (3%)
Pt-Ni/C (3+3%)
Re/C (3%)
Pt-Re/C (3+3%)
Pt-Ru/C (3+3%)
Ru/C (3%)
Pt/AC (2.5%)
Pt-Co/C (3+3%)
0
5
10
15
20
25
1 2 3 4 5 6 7
Se
lec
tivit
y [
%]
Number of carbon atoms in alkane
1,0 %
2,3 %
6,6 %
9,0 %
18,3 %
conversion
0
20
40
60
80
100
1 3 5 7
Se
lec
tivit
y [
%]
Number of carbon atoms in alkane
0
5
10
15
20
25
1 3 5 7
Se
lec
tivit
y [
%]
Number of carbon atoms in alkane
0
5
10
15
20
25
1 2 3 4 5 6 7
Se
lec
tivit
y [
%]
Number of carbon atoms in alkane
65,2 %
47,9 %
16,2 %
9,1 %
conversion
0
5
10
15
20
25
1 2 3 4 5 6 7
Se
lec
tivit
y [
%]
Number of carbon atoms in alkane
0
20
40
60
80
100
1 2 3 4 5 6 7
Se
lec
tivit
y [
%]
Number of carbon atoms in alkane
0
5
10
15
20
25
1 2 3 4 5 6 7
Se
lec
tivit
y [
%]
Number of carbon atoms in alkane
C3H8
C2H6
C3H8
C3H8
CH4
C2H6 C6H14 C4H10
Pt Ni Pt + Ni
Ru Pt + Ru
Re Pt + Re
Pt-Re
Pt and Pt-M Ni, Re, Ru
Pt/AC
Pt-Re
Ni, Re
Ru
Pt-Re
Ni, Re, Ru
Pt
and Pt-M
Pt/AC
Pt/AC
Pt and Pt-M
Hydrogen can be effectively produced from renewable sources, such as
biomass, via a promising Aqueous-Phase Reforming (APR) technology1,2.
This process occurs in liquid phase at comparably low-temperatures (190-
250°C). Various feed can be converted to hydrogen and hydrocarbons. One of
the advantages of APR is process flexibility, allowing determination of product
(hydrogen or alkanes) by a proper catalyst choice.
• Metal screening should be performed for the most hydrothermally stable
carbon-supported catalysts, such as mesoporous carbon Sibunit with high
mechanical strength
• Additions of such metals as Ni, Co, Re and Ru could enhance Pt-based
catalysts performance
xylitol
alkanes
Pt-Re/C
Pt/C
Ni/C Pt-Co/C
Ru/C
Pt-Ni/C
190-250°C
Elevated pressure
Liquid phase
Support preparation
• Boiled in water to remove the dust
• Boiled in 5 wt % HCl to remove metal impurities
• Sieved to proper fraction (100-200 μm)
Catalyst preparation
• Incipient wetness impregnation with aqueous solutions of
corresponding metal precursors
• Drying at 110°C overnight (270°C for Ni/C and Pt-Ni/C)
• Reduction and chlorine removal in H2 flow with heating rate 2°C
/min, 2-7 h at 270-450°C
H2
N2 (1% He)
mass flow controller furnace
sampling
GC analysis
P effluent
pressure controller
safety valve
MFC
catalyst
MFC
MFC
collector
HPLC pump
feed solution
HPLC analysis
Total Organic Carbon (TOC) analysis
Experimental
Catalyst characterisation
• TEM
• TPR-DTA-MS
• XPS
• EDX
• XRD
Syngas Derived Oxygenate Intermediate
to Hydrocarbon Fuel
Michel Gray, Heather Job, Karthikeyan K. Ramasamy
File
Na
me
//F
ile D
ate
//
P
NN
L-S
A-#
####
Acknowledgement: The authors would like to thank the Department of
Energy’s Bioenergy Technologies Office for supporting this work. Pacific
Northwest National Laboratory is operated by Battelle Memorial Institute for
the U. S. Department of Energy under Contract No. DE-AC05-76RL01830.
For More Information, Contact:
Michel Gray
Pacific Northwest National Laboratory
(509) 375-4549
Introduction Ethanol can be produced from renewable resources (e.g. biomass or
municipal solid waste) via biochemical and thermochemical routes, in a
very efficient manner.
Ethanol production is expected to be greater than 30 billion gallons
beyond 2017. Due to blending limitation and technology advancements
ethanol is expected to be in surplus in future years.
Bi-functional mixed oxides have shown great potential for the catalytic
conversion of ethanol to 1-butanol.
Butanol has the potential to serve as a building block to produce
infrastructure compatible transportation fuels and commodity chemicals.
Results and Discussion
Conclusion and Future Work
Reaction mechanism for ethanol conversion to 1-butanol over bi-
functional mixed oxides derived from hydrotalcite.
Experiments on ethanol coupling reactions were conducted using a
down-flow fixed bed reactor. Experiments were conducted at
atmospheric pressure and 300 - 350 °C.
Ethanol to 1-butanol occurs in a complex sequence of steps involving
dehydrogenation, dehydration, aldolation, and hydrogenation reactions
requiring acid-base bi-functional catalytic material.
0
20
40
60
80
100
0 250 500 750 1000 1250
Pe
rce
nta
ge
Time, hrs
Selectivity to C4+ compounds
Ethanol Conversion
0
5
10
15
20
25
30
35
40
45
50
Baseline Catalyst Modified Catalyst
Eth
an
ol
Co
nve
rsio
n a
nd
P
rod
uct
Sele
cti
vit
y (
%)
Ethylene
DEE
Acetaldehyde
Butanol
Hexanol
C4+ Others
Conversion
Product distribution from the ethanol condensation
experiments on baseline and modified catalyst.
Catalyst life time performance Ethanol condensation experiment
over bifunctional acid-base catalyst derived from Hydrotalcite.
For >1000 hours ~40 percent ethanol conversion was achieved
with ~65 percent selectivity to C4+ compounds.
Small levels of promoter added to modify the sites responsible
for the dehydration pathway and promote the dehydrogenation.
Presence of promoter reduced diethyl ether (DEE) and ethylene
generation and increased ethanol conversion.
Ethanol is shown to be converted to butanol and higher alcohols
with minimal unwanted by products.
The hydrotalcite derived mixed oxide catalyst is shown to be
very stable for the ethanol condensation chemistry
C4+ alcohols can be converted to jet fuel range olefins in a
single step dehydration -oligomerization process over solid acid
catalysts.
AbstractA large share of the agricultural biomass resource remains underutilized while local and national policies continueto mandate increasing levels of renewable energy. Only a small fraction of the available agricultural residues arecurrently utilized for energy, even where sustainability would support it. This virtually untapped resource for energyhas struggled because it is low in energy density, diffusely located, and costly to transport to large, centralizedenergy facilities. The use of smaller on-site gasification systems to produce bio-syngas from agricultural biomassresidues shows promise for increasing resource utilization for displacement of fossil fuels. Bio-syngas generated inon-site systems can provide a direct energy substitution for natural gas in gas devices like boilers, dryers, heaters,engines, etc. Proliferation of these systems would help industrial and agricultural processing facilities trying todecrease their carbon emissions footprint, a large part of which comes from on-site natural gas usage. This projectshows the technical, economic, and environmental feasibilities associated with on-site conversion of residuebiomass to bio-syngas for fuel in industrial and agricultural applications. The project samples and characterizes thephysical, chemical, and thermal properties of the biomass residues and resulting bio-syngas from the CircleDraft®gasifier system being deployed for commercial operation by West Biofuels, LLC, and its development partner, INSEREnergia, S.P.A. The project develops the optimal feedstock and operational conditions required to deliver bio-syngasthat is compatible with common natural-gas devices used in industry. In addition, the market potential of the bio-char produced as a byproduct of the gasification process is reviewed. The optimized results are used to evaluate theoverall costs of using bio-syngas in comparison with current and forecast industrial natural-gas prices, to determinethe economic feasibility for commercial projects. The net environmental benefits including potential greenhouse gasemission reductions and carbon sequestration will be reviewed.
Project Goal: The goal of this project is to determine the feasibility of using currently underutilized agriculturalbiomass residues for conversion to synthetic gas to replace or blend with natural gas for direct, on-site industrialand agricultural use
References• Faussone, G.C. 2012. Biomass Gasification With CircleDraft Process. Published online in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/ep.11608• Faussone, G.C. 2014. Personal Communication.• Tittmann, P., N. Parker, Q. Hart and B. Jenkins. 2008. Economic potential of California biomass resources for energy and biofuel. PIER Collaborative Report, California Energy Commission, Sacramento, CA.• U.S. Energy Information Administration (EIA). 2014. California Natural Gas Statistics. http://www.eia.gov/dnav/ng/hist/n3035ca2A.htm• USDA NASS. 2013. California Field Crop Review 34(2), 21 February, www.nass.usda.gov/ca.• Williams, R.B., et al. 2008. An Assessment of Biomass Resources in California, 2007. PIER Collaborative Report, California Energy Commission, Sacramento, CA.
ConclusionAn 2-MW thermal biomass gas production facility has been constructed and is being operated in California fordemonstration on both forest and agricultural biomass feedstock. The produced gas is intended to be used as areplacement for natural gas in typical gas appliances used by the processing industries. While the project is ongoing,early results indicate that the performance targets listed above are within the capabilities of the technology.
Acknowledgement: Funds for this project were provided by West Biofuels and the California Energy Commission under Grant Agreement Number: #14-01-11G
ON-SITE BIOMASS GASIFICATION AS A NATURAL GAS SUBSTITUTE
Dr. Matthew Summers, Dr. Chang-hsien Liao, Mr. Matthew Hart, West Biofuels LLC, Woodland, CA 95776 USADr. Gian Claudio Faussone, INSER Energia S.p.A., C.so Appio Claudio, 229/5, Torino 10146 ITALY
Dr. Reinhard Seiser, Dr. Robert Cattolica, University of California at San Diego, La Jolla, CA 92093 USA
Project Tasks Performance / Cost Objectives1) Experimental design and selection and characterization of available biomass residues from agricultural sector
At least 3 specific biomass residues will be selected for potential suitability for gasification via the CircleDraft® gasifier system that meet criteria for high-quality gasification. The experimental design will be developed to determine the performance and the gas quality from each feedstock at various operating conditions.
Performance targets: Statewide Availability: 1 million tons/year aggregate; Cost: < $30 per ton; Moisture Content: < 25%; Ash Content: < 5%; Higher Heating Value: > 7,000 BTU/lb; Ash Melting Temperature: >1100°C; Particle Size: < 3 in.
2) Setup measurement systems to characterize performance and gas quality on a commercial CircleDraft® gasifier system
A CircleDraft® gasifier system will be provided by West Biofuels that nominally produces 6.75 MMBTU/hr of synthetic gas from 1000 pounds per hour of dry wood chips and will be outfitted with sensors and sampling ports so that the performance and gas quality can be analyzed for the selected residue feedstock.
Performance targets: Feedstock: mass flow measurement to ± 5% and composition to ± 3%; Gas: flow measurement to ± 10% and characterize 99% of major and trace gases; Energy: heating value to ±3% for feedstock and synthetic gas.
3) Conduct testing of biomass feedstock for performance and gas quality with CircleDraft® gasifier system
Tests with each feedstock will be conducted with the CircleDraft®
gasifier system over a range of selected operating conditions from the experimental design and the performance of the gasifier will be measured and the gas quality characterized.
Performance targets: Gasifier efficiency: >70%; Gas heating value: 150-300 BTU/SCF; Gas inert content: <60%; Gas impurities: Less than air quality and manufacturer requirements;
4) Analysis of technical and economic feasibility of natural gas replacement
Based on the optimal results from Task 3, the technical and economic feasibility of industrial or agricultural on-site gasifier projects for the replacement of natural gas will be developed.
Performance targets: Compatibility with natural gas devices (boilers/engines); Gasifier system capital cost: <$100,000 per MMBTU/hr capacity; Operating hours: 7000/year; Synthetic gas cost: $6-$12/MMBTU; Project simple payback: 3-6 years;
Agricultural Biomass Potential
TypeCurrent Industry
Energy Usage
BCHP Energy Production Potential*
Net Energy Potential
Power (GWh)
744 1366 622
Power Value($)
-$120 M @ $0.16/kWh
$188 M$68M @
$0.11/kWhNatural Gas
(BCF)14.9 49 34.1
Nat Gas Value ($)
-$119 M @ $8/1000 ft3 $255 M
$136 M @ $4/1000 ft3
* From Estimated 825,000 BDT of almond shell and 750,000 BDT of prunings.
California Biomass Feedstock Potential
California Almond Industry Energy Potential vs. Current Usage
Almond Shell Almond Pruning
The CircleDraft® Gasification Process
CircleDraft® Gasifier
Process Unit Specifications
Description Value
Fuel Type Chips, Shells, Residue Pellets & Cubes
Oxidant Humidified Air
Fuel Heating Value 8,500 BTU/dry lb
Moisture 25%
Specific Consumption 150 lbs/MMBTU
Gas Production Rate 45 SCF/dry lb
Gas Heating Value 150 BTU/SCF
Fuel Consumption ~ 1000 lbs/hr
Total Gas Output 6.75 MMBTU/hr
Electrical Consumption 90 kW
Working Temperatures 350 °C -1050 °C
Pressure Max: 20 in H2O
Thermal Efficiency 75%
Erection of Gasifier Structure Fabrication of Lower Grate Housing
Upper Reactor with Feeder Lower Cone Hot Face Refractory
Lower Cone Prior to Grate Installation Modular Control Room Fabrication
Plant Construction
Synthetic Gas Characteristic
CircleDraft® Range
H2 10 - 35%
CO 15 - 48%
CH4 1 - 4%
C2H4 0.1 - 0.5%
C3+ 0.1 - 0.3%
CO2 0.1 – 15%
N2 15 - 49%
H2/CO Ratio 0.5 – 1.5
HHV (MJ/Nm3) [BTU/scf] 5.6 – 10.3 [150 - 300]
Wobbe Index (MJ/Nm3) 22 – 38
Tars (mg/Nm3) 5 - 25
Particulate (mg/Nm3) 1 – 20
H2S (ppm) 5 – 50
Product Syngas Composition
3D CAD and Actual Photo of Gas Production PlantControl Panel for Gas Production Plant
Experimental Design
BL only 10% PO “Standard” 15% PO 20% PO Cold Hot
Pyrolysis Oil (kg/hr) N/A 100.0 145.0 185.0 145.0 145.0
Black liquor (kg/hr) 1102.7 887.5 812.5 732.8 805.1 805.7
Oxygen to Fuel ratio (kg/kg)
0.291 0.306 0.309 0.320 0.301 0.325
Gasifier Pressure
(bar) 26.7 26.7 26.7 26.7 26.7 26.7
λ 0.373 0.383 0.379 0.388 0.373 0.402
• Co-gasification of three black liquor and pyrolysis oil mixes with up to 20% wt PO at different oxygen-to-fuel ratios was demonstrated in pilot-scale.
• Syngas from the co-gasification experiments was found to contain low levels of impurities and was of acceptable quality for use in the downstream methanol synthesis plant.
• The introduction of the more energy rich pyrolysis oil
increased the process cold gas efficiency and improved the process carbon distribution without adversely affecting carbon conversion.
• Carbon conversion efficiency was greater than 98.8% at all
operating points; the overall mass and carbon balance closures did not deviate by more than 2% and 5% respectively.
A schematic representation of the gasifier and the gas cooler system; solid lines represent both mass and heat flows while dashed lines represent heat flows only. Note that the black liquor used in the process is taken from the nearby pulp mill. A dedicated pyrolysis feeding system has been installed together with a mixer to ensure fuel mix homogeneity.
Entrained-flow Co-gasification of Black Liquor and Pyrolysis Oil Concept Verification and Assessment of Gasifier Performance
Yawer Jafri, Erik Furusjö, Kawnish Kirtania, Rikard Gebart; [email protected] Energy Science, Luleå University of Technology, SE 971 87 Luleå, Sweden
The experimental campaign presented in this poster was designed to measure product composition, perform material balances and quantify process performance on a pyrolysis oil/black liquor mix at a) different mixing ratios, b) a range of temperature within the pilot plant’s operating envelope. For benchmarking purposes, the co-gasification runs were alternated with black liquor only experiments at comparable process settings and allowed to reach steady-state. In the table below “Cold” and “Hot” denote co-gasification runs at low and high oxygen to fuel ratios. A mixing ratio of 15% was picked as “Standard” for co-gasification runs. Plant thermal load was held constant at around 2.8 MW (on HHV basis).
Cold Gas Efficiencies
𝜂𝜂𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶 = 1 − 𝑚𝑚𝐶𝐶,𝑇𝑇𝑇𝑇𝐶𝐶
𝑚𝑚𝐶𝐶,𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹∗ 100
Carbon containing syngas species and inorganic carbon in green liquor are the two main products from the process. In the above definition, 𝑚𝑚𝐶𝐶,𝑇𝑇𝑇𝑇𝐶𝐶 is the mass flow rate of green liquor’s organic fraction (including any unconverted solid char) and 𝑚𝑚𝐶𝐶,𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹 is the mass flow of Carbon in the fuel.
𝐶𝐶𝐶𝐶𝐶𝐶𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹 =𝐿𝐿𝐿𝐿𝐿𝐿𝐶𝐶𝑇𝑇 + 𝐿𝐿𝐿𝐿𝐿𝐿𝐻𝐻2
𝐿𝐿𝐿𝐿𝐿𝐿𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹 𝐶𝐶𝐶𝐶𝐶𝐶𝑃𝑃𝐶𝐶𝑃𝑃𝐹𝐹𝐶𝐶 =
𝐿𝐿𝐿𝐿𝐿𝐿𝑆𝑆𝑆𝑆𝐶𝐶𝑆𝑆𝐶𝐶𝑆𝑆𝐿𝐿𝐿𝐿𝐿𝐿𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹
𝐶𝐶𝐶𝐶𝐶𝐶𝑆𝑆𝐹𝐹𝐹𝐹𝑆𝑆𝐹𝐹𝐶𝐶𝐹𝐹𝐶𝐶𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹 =𝐿𝐿𝐿𝐿𝐿𝐿𝐶𝐶𝑇𝑇 + 𝐿𝐿𝐿𝐿𝐿𝐿𝐻𝐻2𝐿𝐿𝐿𝐿𝐿𝐿𝑆𝑆𝐹𝐹𝐹𝐹𝑆𝑆𝐹𝐹𝐶𝐶𝐹𝐹𝐶𝐶𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹
𝐶𝐶𝐶𝐶𝐶𝐶𝑆𝑆𝐹𝐹𝐹𝐹𝑆𝑆𝐹𝐹𝐶𝐶𝐹𝐹𝐶𝐶𝐹𝐹𝐹𝐹 𝑃𝑃𝐶𝐶𝑃𝑃𝐹𝐹𝐶𝐶 =𝐿𝐿𝐿𝐿𝐿𝐿𝑆𝑆𝐹𝐹𝐹𝐹𝑆𝑆𝐹𝐹𝐶𝐶𝐹𝐹𝐶𝐶𝐹𝐹𝐹𝐹𝑆𝑆𝑆𝑆𝐶𝐶𝑆𝑆𝐶𝐶𝑆𝑆𝐿𝐿𝐿𝐿𝐿𝐿𝑆𝑆𝐹𝐹𝐹𝐹𝑆𝑆𝐹𝐹𝐶𝐶𝐹𝐹𝐶𝐶𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹
Syngas from the co-gasification of black liquor and pyrolysis oil is primarily intended for catalytic upgrading to bio-chemicals or biofuels. Hence, the useful chemical energy in the syngas for this purpose is quantified as CGEFuel which only takes into account the heating values of syngas CO and hydrogen components. In a pulp biorefinery implementation of the concept, sulfur in the fuel needs to be recovered and returned to the pulp mill in a reduced state (sulfide). Hence, in recognition of this, the CGEs are also determined on sulfur-free basis to provide a more representative basis of comparison. All LHVs are computed on wet-basis.
Entrained-flow gasification of black liquor (BL) can facilitate the conversion of today’s pulp mills into tomorrow’s multi-product biorefineries. The technology has been demonstrated at a 3 MWth pilot plant in Piteå (pictured above) which has been operated for over twenty-five thousand hours. BL is a fuel with a high sodium content and a relatively low heating value. However, its availability is tied to pulp production which represents a potential bottleneck. This may be sidestepped by harnessing the catalytic effect of its alkaline component to mix in a more easily transportable fuel such as pyrolysis oil (PO). Refer to the poster by Bach et al. for a labscale study on co-gasification of PO and BL.
LTU Green Fuels Pilot facility in Piteå, Sweden
Water streams into the gasifier provide a means for regulating the level of liquid in the dissolver pool at the bottom of the gasifier vessel. This liquid is referred to as “green liquor” and it is made up of gasifier smelt (containing black liquor’s ash component) and water.
Carbon enters the gasifier in the fuel and leaves in the syngas and the green liquor. CO, CO2 and CH4 are the main carbon containing components in the former. C6H6 constitutes the bulk of the minor gases (higher hydrocarbons). It can be seen (below) that the addition of pyrolysis oil to black liquor leads to a shift in the carbon distribution. At higher mixing ratios more of the additional carbon ends up as CO in the syngas at the expense of CO2. Also notable is the strongly linear correlation between temperature (controlled by λ) and CH4.
The wt% of alkali metals such as Na and K can lay around 16-17% in industrial-grade black liquor from Kraft pulping. Together with sulfur, the recovery of these elements for re-use in the pulping process would be an integral function of a gasifier in a pulping biorefinery. All the alkali metals entering the system end up in the green liquor and are returned to the pulp mill. In contrast to a conventional recovery boiler, the out-going sulfur is split into two: part of it leaves as hydrogen sulfide in the syngas while the bulk ends up as alkali metal sulfides in green liquor (see above).
𝜂𝜂_𝐶𝐶𝑎𝑎𝑟𝑟𝑏𝑏𝑜𝑜𝑛𝑛 at all settings was greater than 98.8%. The green liquor organic C concentration in the “Cold” run sample was uncharacteristically low, resulting in a higher than expected 𝜂𝜂_𝐶𝐶𝑎𝑎𝑟𝑟𝑏𝑏𝑜𝑜𝑛𝑛 .This may be due to an erroneous measurement or a failure of the measurement method to account for the porous unconverted residue observed in a sample after the experiment. Characterisation of individual components in green liquor’s organic content requires the development of a bespoke methodology. Neither feedstock composition nor temperature appeared to have a significant influence on carbon conversion within the studied operational envelope.
It is evident from the graph above that co-gasification runs registered significantly higher CGEs. The two “Power” CGE values were lower for the “Hot” run owing largely to the consumption of CH4. On the other hand, “Fuel” efficiencies remained largely unchanged suggesting that a decrease in CH4 concentration below a certain point does not necessarily lead to more CO in the syngas. Given a fixed PO fraction, “Standard” appears to represent a close to optimal point within the operating envelope for biofuel production.
Performance Indicators Process Overview Background
Process Element Distribution Characteristic Mass Flow Distribution
Carbon Conversion and Cold Gas Efficiencies Syngas Composition Conclusions and Future Work
It can be seen above that the introduction of pyrolysis oil leads to an increase in CO and H2 fractions at the expense of CO2. Since pyrolysis oil is virtually free of sulfur and metals, the sulfur concentration in the syngas also goes down. CH4 exhibits a very clear temperature dependence. It is worth noting that the syngas produced in the process is virtually tar-free. The concentration of benzene, the most common hydrocarbon observed, hovered around 100 ppm.
Carbon Conversion Efficiency
The commissioning of an onsite pyrolysis oil storage facility (right) will allow the performance of multi-week co-gasification campaigns with subsequent methanol and DME production at the LTU Green Fuels Gasification Plant. These will thus enable the study and quantification of long-term effects at a range of process settings.
Methanol and dimethyl ether synthesis plant downstream of the gasifier.
Typical mass (above) flow distribution [“standard” operating point].
The inorganic and organic carbon fractions in the green liquor sludge that is returned to the pulp mill constitute less than 0.02% of the total
output Carbon.
Acknowledgement: This work was financed by the Swedish Energy Agency and a consortium of partners, namely, Haldor Topsoe, Volvo, Bio Green, Innventia, Preem, Länsstyrelsen Norrbotten, Smurfit Kappa, Aga Gas, Flogas Sverige, Chemrec, Sveaskog, Södra and Adita Birla Domsjö.
Pyrolysis oil used in the campaign was supplied by Fortum Oyj. Fuel compositions (left) are on wet basis.
Pyrolysis Oil Black Liquor Carbon (wt%) 37.6 21.9
Hydrogen (wt%) 7.8 6.6 Oxygen (wt%) 54.5 46.7
Sodium (wt%) N/A 17.6 Sulfur (wt%) N/A 4.2
LHV_wet (MJ/kg) 15.1 7.7
35
37
39
41
43
45
BL only 10% PO 15% PO 20% PO Cold Standard Hot
Car
bon
frac
tion
-%
Operating Point
Carbon Fraction in CO and CO2 C in CO/Fuel Carbon (%) C in CO2/Fuel Carbon (%)
at "Standard" Temperature at 15% PO "Standard"
0
0,5
1
1,5
2
2,5
3
3,5
BL only 10% PO 15% PO 20% PO Cold Standard Hot
Car
bon
frac
tion
-%
Operating Point
Carbon Fraction in CH4 and Minor Species C in CH4/Fuel Carbon (%) C in minor species/Fuel Carbon (%)
at "Standard" Temperature at 15% PO "Standard"
0,0
0,5
1,0
1,5
2,0
2,5
3,0
3,5
20222426283032343638
BL only 10% PO 15% PO 20% PO Cold Standard Hot
% S
ynga
s (H
2S, C
H4)
Syng
as %
(CO
2, C
O, H
2)
Operating Point
Syngas Composition CO2 CO H2 CH4 H2S
At “Standard” Temperature At 15% PO “Standard”
0,40
0,45
0,50
0,55
0,60
0,65
0,70
0,75
BL only 10% PO 15% PO 20% PO Cold Standard Hot
%
Operating Point
Cold Gas Efficiencies CGE_Power CGE_Fuel CGE_SulfurFreePower CGE_SulfurFreeFuel
at “Standard” Temperature at 15%PO “Standard”
Syngas to Olefinic Compounds over
Co-Ni Bimetallic Catalyst
Heather Job, Michel Gray, Karthikeyan K. Ramasamy
File
Na
me
//F
ile D
ate
//
P
NN
L-S
A-#
####
Acknowledgement: The authors would like to thank the Department
of Energy’s Bioenergy Technologies Office for supporting this work. Pacific
Northwest National Laboratory is operated by Battelle Memorial Institute for
the U. S. Department of Energy under Contract No. DE-AC05-76RL01830.
For More Information, Contact:
Heather Job
Pacific Northwest National Laboratory
(509) 375-4529
Introduction
Results and Discussion
0
50
100
150
200
250
300
350
400
450
5.8%Co 2.9%Co-2.9%Ni 5.8%Ni
Sp
ace T
ime Y
ield
, g
/kg
cat/
hr
Methane
Ethane
Propane
C4 Alkanes
C5 Alkanes
C6 Alkanes
Hydrocarbon Liquids
Total Oxygenates
C6 Alkenes
C5 Alkenes
Butenes
Propenes
Ethene
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
280⁰C 285⁰C 290⁰C 295⁰C 300⁰C
CO
Co
nv
ers
ion
an
d P
rod
uc
t S
ele
cti
vit
y
CO2
CH4
Ethane
Propane
C4 Alkanes
C5 Alkanes
C6 Alkanes
Hydrocarbon Liquids
Total Oxygenates
C6 Alkenes
C5 Alkenes
Butenes
Propenes
Ethene
CO Conversion
0
1
2
3
4
5
6
7
8
9
10
0
0.05
0.1
0.15
0.2
0.25
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Ole
fin
/Para
ffin
Se
lec
tiv
ity
Carbon Number
Olefin Paraffin Olefin/Paraffin
Conclusion and Future Work
The projected depletion of crude oil resources and the potential
contribution of fossil energy resources to greenhouse gas
emissions contributes to the need to develop renewable energy
technologies.
Syngas hydrogenation is one of the most technologically feasible
processes for producing transportation fuels from a range of feed
stocks such as natural gas, coal, municipal solid waste, and biomass.
STY and the product distribution of the syngas hydrogenation over monometallic Co, Ni and
bimetallic Co-Ni catalysts, H2 to CO ratio 2:1, 285°C, GHSV 12000 L/Kg/hr, and 1200 psi.
Selectivity and product distribution at different reaction temperatures over the 2.9%Co-
2.9%Ni bimetallic catalyst, H2 to CO ratio 2:1, GHSV 12000 L/Kg/hr, and 1200 psi.
Paraffin and olefin composition generated from CO hydrogenation over 2.9%Co-
2.9%Ni catalyst at 300 ºC, H2 to CO ratio 2:1, GHSV 12000 L/Kg/hr, and 1200 psi .
Bench scale packed bed flow through reactor system used to test the
syngas conversion over Co-Ni based catalyst prepared via
Impregnation.
Monometallic Ni catalyst is considered to be very active for the
methanation reaction, but it shows almost no activity in this study
because the operating conditions used were not optimum.
Monometallic Co catalyst generated primarily paraffinic compounds
(in C5+ range) along with high methane selectivity.
The bimetallic Co-Ni catalyst generated less methane (approximately
15% carbon selectivity) as well as a higher olefinic to paraffinic ratio.
Increasing temperature between 280 to 300°C linearly increases
the STY from 263 to 600 g/Kgcat/hr and the CO conversion from
11 to 26 % with out much change in the product composition.
The calculated (ASF) α value for the C5+ product composition is
around 0.37 compared to the 0.5 and 0.7 for the Fe Catalyst
(lower α value predicts increased small olefin selectivity).
The olefin to paraffin ratio was close to 8 for the C3 hydrocarbon,
and was around 1.6 for C4 through C11. For C11+, the olefin to
paraffin ratio starts to drop slowly.
Co-Ni bimetallic catalysts show promise for converting low value
feedstocks to valuable olefinic compounds via syngas route.
Combining syngas hydrogenation with acidic catalysts in a multi-
step conversion process reduces the oxygenate composition torespective olefinic compounds.
OPERATION A 1-MWth FICFB BIOMASS GASIFIER IN CALIFORNIA
AbstractThere is a large potential to deploy Biomass Combined Heat and Power (BCHP) facilities inCalifornia, where large agricultural processing is taking place and agricultural waste materials areavailable. A 1-MW (thermal input) Fast Internally Circulating Fluidized Bed (FICFB) gasificationsystem has been constructed and operated at the Woodland Biomass Research Center (WBRC), inCalifornia for demonstration purposes on agricultural feedstock and other waste biomass streamsthat are available regionally. It is based on the same design that has been operating in Güssing,Austria on forest wood chips. In this project, West Biofuels collaborated with University ofCalifornia (San Diego, Davis) and Vienna Technical University to design and operate this FICFBsystem. As part of the BCHP project, a 250-kW power generator (natural gas IC engine) has beenoperated on producer gas to perform studies of engine emission control. An activated-carbonfilter system was integrated on the combustor side of the FICFB plant to study NOx emissioncontrol of the combustor emissions. A new synthetic bed material has been demonstrated as afluidization medium. The integrated heat recovery system harvests the waste heat for the steamproduction, steam super-heating, and air preheating for an efficient gasification process. Theoperating parameters, recovered heat rates, gas production and composition, and exhaustemissions were recorded during multiple test runs. The ranges of the main components in thesyngas, before gas cleaning and on a dry-basis, were: hydrogen: 22-40 vol%; carbon monoxide:21-40 vol%; methane: 6-14 vol%; nitrogen: 0.8-7 vol%; carbon dioxide: 12-22 vol%; benzene:600-6000 ppmv; Toluene: 10-100 ppmv. Tars (>toluene) were in the range of 10-200 mg/Nm3.
References• West Biofuels, LLC. 2015. “Demonstration of Advanced Biomass Combined Heat and Power Systems in the Agricultural Processing Sector”. Final Report
for Contract Number PIR-11-008. California Energy Commission. Publication #CEC-500-2015-XXX. • H.Hofbauer, R. Rauch, R.Koch, “Biomass CHP Plant Gussing-A Success Story”• R.Rauch, H.Hofbauer, “Steam Gasification of Biomass at CHP Plant Gussing-Status of the Demonstration Plant”
Bed material name CARBO® HSP (30/60)
Mean diameter,(µm)
488
Chemical composition
Al2O3 (wt%) 82%
SiO2 (wt%) 7%
Fe2O3 (wt%) 7%
TiO2 (wt%) 3%
Others (wt%) 1%
Bulk density(kg/m3)
2,010
Heat Capacity Cp, (J/kg.K)
761.6
Table 1: Bed material properties
Almond Prunings
HHV, as received (dry), (MJ/kg) 16.75 (18.64)
Moisture (wt%) 5.19
Proximate Analysis (dry basis)
Volatile matter (wt%) 76.50
Fixed carbon (wt%) 21.30
Ash (wt%) 2.20
Ultimate analysis (wet basis)
C (wt%) 46.92
H (wt%) 5.47
O (wt%) 39.22
N (wt%) 0.44
S (wt%) 0.03
Table 2: Proximate and Ultimate analysis of example feedstock
Operational parameters
S/B ratio (wet biomass) 0.22 - 0.54
Biomass feed rate 68.1 - 182.5 kg/h
Thermal input (Biomass LHV + Auxiliary)
710 - 1170 kW
Total bed material mass 1800 kg
Gasifier temperature 810 - 860 °C
Gasifier exit pressure 0 - (-15) mBar
∆T, (Tcombustor –Tgasifier ) 60 - 100 °C
Table 3: Operational parameters and overall dimensions of the 1MW FICFB system in Woodland, CA
Gas composition (vol %)
Syngas Output 118.63 kg/hr
H2O(g) 7%
H2 29 40%
CO 12 29%
CO2 14 20%
CH4 1 11%
C2H4 0.1 2.7%
C2H6 0 0.5%
C3H8 0 0.04%
N2 2 8%
O2 0.1 1.0%
C6 + C7 others 0 0.5%
ConclusionAn 1-MW thermal FICFB biomass gasification was built and has been operated in California fordemonstration on non-forest feedstock and for process improvement including materials,consumables, and emissions controls. The system tests demonstrated successful syngas productionfor use in an IC engine from non-forest biomass and other important improvements to the processto make it viable in North America. Other kinds of feedstock and bed materials will be tested forthis pilot-plant in the future and it will be used for studies on syngas upgrading to mixed alcohols,synthetic natural gas, and other higher value energy products.
Dr. Chang-hsien Liao; Dr. Matthew Summers West Biofuels LLCWoodland, CA 95776
Dr. Reinhard Seiser, Dr. Robert CattolicaDepartment of Mechanical and Aerospace EngineeringUniversity of California, San Diego, La Jolla, CA 92093
Dr. Reinhard Rauch; Vienna Technical University, Austria
Fig.1: 1MW FICFB system in Woodland, CA, USA
Based on LHV (NCV) Average
Gasifier Cold Gas Efficiency 68%
Total Input Power 975 kW
Total Output Power 745 kW
Unrecovered Heat Loss 229 kW
Unrecovered Heat Loss 23%
Gross Overall Efficiency 77%
*: Barracuda CFD result: 3.7kg/s, as grid number is smaller, it may be slightly over estimated by 20%
Gasifier Dimensions
Gasifier height 9 m
Combustor ID 24 cm
Gasifier ID (up/low) 107/36 cm
Circulation rate (estimated from CFD analysis)
~ 3.0 kg/s *
Location Gravimetric analysis (g/Nm3)
SPA method(mg/Nm3)
DCM impinger(mg/Nm3)
Average(g/Nm3)
Pre-filter 16.19 NA NA 16.19
Post-scrubber NA 7961 5977 6.97
Recoverable heat rate
Average (kW)
Steam generator 107 kW
Steam super heater 23 kW
Preheat air 68 kW
Hot water 156 kW
Char dust NA*
Biodiesel Emulsion NA *** The discharged char dust from the hot filter has notintegrated into the system yet.** The discharged biodiesel emulsion from the scrubber tankhas not integrated into the combustor yet.
Table 6: Tar measurement before and after gas clean-up system
Fig.2: Process flow diagram of BCHP system
Fig.5: Heat recovery system for the FICFB
Fig. 6: Gas cleanup system
Table 7: Heat recovery results
Table 4: Gas composition after gas clean-up
Fig.3: Operating temperatures and pressure drops for FICFB
Fig.4: Example of tar measurement using SPE and DCM impinger
Acknowledgement: Funds for this project were provided by West Biofuels and by the California Energy Commission under Agreement Number: PIR-11-008
Table 5: Measured energy parameters for FICFB
Chemical composition of crop straw ashesThe chemical composition of three kinds of ash at the ashing temperature of 525°C and 815°C were presented in Table 3,
including the composition of Baodian coal ash of 815°C. The result shows that the crop straw ashes contain the following main elements:
Si, Ca, Mg, K, Na, S, Cl, P and only a little Al, Fe, Ti. There are significant differences in two main elements (Si and K) among these
straw ashes. For the ash of 525°C, the Si content in rice straw ash is the largest, followed by corn straw ash and cotton straw ash. The
content of K (in the form of K2O) in corn straw ash and cotton straw ash is up to 37.46% and 36.69% and occupies a larger proportion
than that in rice straw ash. Obviously, all above differences in elemental contents are ascribed to their different growing conditions and
the difference of species. The above results suggest that rice straw ash with high silica content is suitable for producing ceramic products
and the high content of potassium in corn straw ash and cotton straw ash is suitable for soil amendment. Interestingly, we should notice
that the elements of potassium and chlorine decreased rapidly with the increase of ashing temperature, which can be confirmed by
subsequent XRD analysis.
In this experimental study, the ash fusion characteristics of three kinds of crop straw (rice straw, corn straw, cotton straw) at different
ashing temperatures were investigated and the main findings derived from the present investigation can be summarized as follows:
(1) By comparing the ashing temperature of GB and ASTM norms, the low temperature is more accurate for preparing crop straw biomass
ash due to the release of alkali metals, especially potassium between 700 ℃ and 800 ℃ in the form of KCl and (KCl)2.
(2) The fusion temperatures of three kinds of straw ash are low (FT order: cotton straw ash > rice straw ash > corn straw ash), and not
affected significantly by the different ashing temperatures. The rich contents of many alkali metal and alkali earth metal elements are the
main cause of the low fusion temperatures. These three kinds of straw ashes, especially corn straw ash are very easily to melt, slag, and
foul in the thermochemical conversion process.
(3) According to the results of thermal analysis, weight losses before 1200 ℃ from the ash samples may be caused by the release of alkali
metals, and the decomposition of carbonate and sulfate.
Ashing and Ash Fusion Characteristics of Crop Straw at Different Ashing TemperaturesXia Liu, Xue-li Chen, Xiao-jun Tian, Han-ding Chen
Key Laboratory of Coal Gasification of Ministry of Education, Shanghai Engineering Research Center of Coal Gasification,
East China University of Science and Technology, P. R. China
On account of environmental requirement, the interest in using agricultural wastes, trash and other biomass fuels for power generation
and chemical production is continuously growing throughout the world. Biomass, after coal, oil and nature gas has been becoming one
of the largest primary energy sources in the world. Thermochemical conversion processes such as gasification, liquification, pyrolysis
and combustion are most commonly used for converting biomass into higher heating value fuels [1,2]. In China, where the energy
demand nowadays is mainly covered by conventional energy sources, a large amount of biomass in the form of agricultural residues
(150 million tons of standard coal, annual production) such as rice straw, wheat straw, corn straw and cotton straw is used ineffectively
[3,4].
Different biomass shows differences in the concentrations of the ash forming elements (e.g. Si, Al, Fe, Ca, K, Na, S, Cl). During
thermochemical conversion processes, these elements can cause ash related problems such as slagging, fouling, and corrosion in the
boiler and convection area, not only reducing the utilization efficiency of equipment but also shortening their life [5-9]. To solve these
problems, several authors proposed that various mineral additives such as limestone, kaolin, calcined dolomite, and ophite should be
used [10,11]. In addition, there are some other methods that can enhance the operation of biomass-fired boiler such as co-combustion
[12,13] and leaching [14]. Presently, there have been some correlative reports on fusion characteristics of biomass ash [15-18], but the
investigations appear quite dispersive. Wang et al [15] reported that the melting point of seaweed ash was low, and neither Chinese
standard nor the US ASTM standard was suitable for seaweed biomass. Xiao et al [16] suggested that the optimal temperature for an
exact determination of biomass ash properties was 600 ℃. Although some studies have been carried out, the influence of ashing
temperature on ash fusion temperatures is still uncertain or incompletely understood.
There are no specific standards for biomass ash analysis in China, so the standards for coal ash analysis are usually used to determine
the property of biomass. In fact, there are considerable differences in ash characteristics between coal and biomass. Considering the
limit of research on crop straw ashes, rice straw, corn straw and cotton straw as the representative samples were investigated. In this
paper, their chemical compositions, mineral compositions, slagging and fouling characteristics at different ashing temperature, and
thermochemical characteristics were discussed in order to provide some useful references for the exploration and utilization of crop
straw biomass.
Background
Results and discussion
Experiment
Conclusions
Methods
Chemical analysis of major elements in ash was conducted by X-ray fluorescence spectrometer (XRF, Shimadzu Corporation).
Mineralogical analysis of ash was performed using X-ray diffraction analyzer (XRD, D/MAX 2550 VB/PC, RIGAKU, Japan) with
application of Cu kα radiation and graphite monochromator (U=40kv, I=100mA). The XRD scans were conducted between 10 and 80
(2θ°), with a step size of 0.02°/s. Major crystal phases were identified using the standard cards of JCPDSICDD (PDF-2 database
Sets).
Thermal analysis of crop straw ash was conducted in a WRT-3P Thermal Analyzer. 0.5mg each of three straw ash samples was heated
at a rate of 10 ℃ /min from room temperature to 1300 ℃. The pure nitrogen (100mL/min) maintains inert condition in the oven and
removes any gaseous products that may be released.
Ash content of crop straw
MaterialsThree kinds of crop straw (rice straw, corn straw, cotton straw) procured from the suburb of Shanghai in China were used as raw
materials. The air-dried biomass samples were ground and sieved. The obtained particles between 106µm and 425µm were chosen as
the experimental material. The results of the proximate analysis, ultimate analysis and calorific values of biomass samples have been
summarized in Table 1. The analysis data of Baodian coal which is widely used in China for combustion and gasification were also
included for comparison. From the table, it can be found that compared with coal, the straw biomass has much higher volatile and much
lower fixed carbon. The Baodian coal has higher calorific value than straw biomass. The results also indicate that the rice straw has
higher ash contents than the other two straw.
There are no specific standards for preparation of biomass ash in China. The ashes of three kinds of crop straw were prepared in a
muffle furnace using open ceramic container according to the procedures proposed by GB/T212-2007 (for coal in China) and ASTM
standard respectively. For GB/T212-2007, temperature increases at 20 ℃ /min to 500 ℃ and then at 10 ℃/min to a maximum of 815 ℃.
Temperature was hold at 500C for 30min and again at 815 ℃ for 1h. The furnace temperature was then dropped from the maximum 815
℃ to room temperature. For ASTM, temperature increases at 20 ℃ /min to 100 ℃ and then at 2 ℃ /min to a maximum of 525 ℃.
Temperature was hold at 400 ℃ for 3h and again at 525 ℃ for 4h. The furnace temperature was then dropped at 10 ℃ /min from the
maximum 525 ℃ to room temperature. The ash samples were stored in airtight containers during the whole study.
The ash content of biomass at different ashing temperatures was listed in Table 2. The increasing ashing temperature will result in
reduction of ash content, which might be caused by the insufficient combustion of crop straw at the lower temperature and the release
of alkali metals at the higher temperature [16]. Residual carbon in these crop straw ash samples of 525 ℃ was measured by element
analyzer. The results showed that the content of carbon in ashes was less than 0.5wt%, which means the influence of residual carbon on
the mass loss can be ignored. Observed with the naked eye, three kinds of straw ash of 525 ℃were soft and gray, in contrast to the
Ash preparation
Ash fusibility was analyzed by 5E-AFⅢ-auto analyzer (Changsha KaiYuan
Instruments Co.Ltd) under reducing atmosphere according to GB219-2007. The
reducing atmosphere was gained by the incomplete combustion of black lead and
active carbon in corundum tube inside the 5E-AFⅢ-auto analyzer. The test
involved heating a sample cone of specified geometry at a rate of 15 ℃ /min
before 900 ℃, then 5 ℃ /min up to 1600 ℃ and recording the temperatures that
characterize the ash melting: deformation temperature (DT), softening
temperature (ST), hemispherical temperature (HT) and flow temperature (FT) as
depicted in Figure 1.
larger proportion than that in rice straw ash. Obviously, all above differences in elemental
contents are ascribed to their different growing conditions and the difference of species.
The above results suggest that rice straw ash with high silica content is suitable for
producing ceramic products and the high content of potassium in corn straw ash and cotton
straw ash is suitable for soil amendment. Interestingly, we should notice that the elements
of potassium and chlorine decreased rapidly with the increase of ashing temperature,
which can be confirmed by subsequent XRD analysis.
Mineral composition of crop straw ashes
The XRD analysis results indicating the qualitative presence of crystalline minerals in each ash sample were illustrated in Figure 3.
As seen in Figure 3(a), X-ray diffraction pattern reveals that in addition to KCl, the main part of the ash at 525°C was composed of non-
crystalline, amorphous compounds (2θ =20~30°). When the ashing temperature increased up to 815°C, the characteristic peak of KCl
disappeared, and the predominant mineral was quartz (2θ = 21.984°, 31.461°) corresponding to the high SiO2 content (75.72%, Table
3).
It was found in Figure 3(b) that the main crystal phase detected in corn straw ash of 525°C and 815°C was both KCl. But the
diffraction peak intensities of KCl decreased with the increase of ashing temperature, because it was released partly into gas phase
between 525°C and 815°C.
As shown in Figure 3(c), like rice straw ash, when the ashing temperature increased to 815°C, KCl can’t be detected. Ca2SiO4
(2θ=32.053°, 32.136°), MgSiO3 (2θ=31.093°) were both detected in ash of 525°C and 815°C. Because of the high content of K2O
(22.80%) and SO3 (8.90%) ash of 815°C, strong characteristic peaks of K2SO4 were detected at 21.259°, 21.341°, 29.746°,
30.774°, 43.275°, 43.428°.
ashes of 815 ℃, which were hard and of which partial ash particles adhered to the surface
of the ceramic container. Figure 2 shows the surface morphology characteristics of the rice
straw particles. A lot of soft things (not confirmed) disappeared and the whole particle. A
lot of soft things (not confirmed) disappeared and the whole particle surface became
smooth with the increase of ashing temperature. These phenomena showed that the
ashingtemperature of 815°C was too high for crop straw biomass.
Ash fusion temperaturesReal-time graphical representations during the ash fusion tests were represented in Figure 6. During fusion process, the changes of ash
cone height, expressed as the ratio of ash cone height to it initial height at 900°C (ŋ, %), were obtained and were plotted against the
corresponding temperatures. Cone of cotton straw ash and corn straw ash showed a similar fusibility trend, while different from that of
rice straw ash. As is evident from the Table 6, rice straw ash exhibited higher DT-FT intervals than corn straw ash and cotton straw ash,
which was most probably because of higher content of quartz in rice straw ash. From the ash fusion data of crop straw species, it can be
seen that the ash fusion temperatures were not affected obviously by different ashing temperature although partial alkali metals were
released into the gas phase in the temperature range of 700-800°C. The order of FT of three kinds of straw ash was: cotton straw ash>
rice straw ash> corn straw ash, which was different from the order of RB/A (Table 5). From Table 5, the fusion temperature of coal ash
may be influenced significantly by the addition of straw biomass because of the high contents of alkali metals and alkali earth metals in
straw ashes.
Thermal analysis of crop straw ashThe results of thermal analysis with different crop straw ash samples were shown in Figure 5. In the temperature range of 525-815 ℃, a
small mass loss of rice straw ash that accounts for 2.3% of the initial mass can be found, which is attributed to the release of alkali metals
especially K in the form of KCl and (KCl)2 as shown in Figure 4. Due to the higher amounts of potassium and chlorine in corn straw ash,
a large mass loss (7%) of corn straw ash was observed between 525 ℃ and 815 ℃. At the temperature segment from 525 ℃ to 815 ℃, a
significant mass loss (16.5%) can be seen from TG curve of cotton straw ash. Interestingly, although the cotton straw ash has a very high
content of potassium similar with corn straw ash, the low content of chlorine in cotton straw ash may hinder the release of KCl or (KCl)2.
The observed significant mass loss of cotton straw ash may be caused mainly by the decomposition of calcium carbonate, according to
Table 3 (CaO, 18.43wt%) [22].
Removal of tars is critical to the design and operation of biomass gasification
systems as internal combustion engines, gas turbines, fuel cells, and fuel
synthesis reactors all have a low tolerance for tar. Water scrubbing is an
existing technique commonly used in gasification plants to remove
contaminants and tar, however using water as the absorbent is non-ideal as tar
compounds have low or no water solubility. Hydrophobic solvents can improve
scrubber performance and this study evaluates tar solubility in selected solvents
using slip-streams of untreated syngas from a laboratory fluidized bed gasifier
operated on almond shell feedstock (Figure 1). A group of solvents were
selected that encompassed varying degrees of similarity to verify if Hansen’s
Solubility Parameters (HSP) predicts good solvents as well as poor solvents. Tar
solubility is compared with Hansen's solubility theory to examine the extent to
which the tar removal is predicted. The experimental results were compared
with the theoretical HSP predictions. This work aims to provide a better
understanding of tar collection and solvent selection for wet scrubbers, and
provide information for designing improved tar management systems for
biomass gasification.
Introduction
The main objective of this study was to evaluate the impact that scrubber
solvent has on tar removal efficiency by accomplishing the following:
1. Select scrubber solvents using HSP theory to minimize both tar dew-
point and total concentration in the cleaned syngas.
2. Experimentally evaluate solubility in selected solvents using slip-
streams of raw syngas from a laboratory reactor.
3. Compare the experimental results with the theoretical predictions.
Objectives
Hansen solubility parameters have been used to predict solubility especially in
the area of polymers (Hansen, 2007; Levin and Redelius, 2008; Lindvig et al,
2002; Segarceanu and Leca, 1997). If solubility parameters can be used to
predict solubility for tar compounds, this could facilitate selection of an
appropriate solvent to be used within the packed bed wet scrubber for tar
removal.
The basic premise of HSP is that chemicals will dissolve in solvents whose
solubility parameters are similar to their own. A more rigorous thermodynamic
justification is presented in Hansen (2007). Using Hansen’s database of
dispersive, polar, and hydrogen bonding solubility parameters for solvent and
tar compounds, the selected solvents were ranked by relative distance, Ra:
where ζD is the dispersive parameter,
ζP is the polar parameter,
ζH is the hydrogen bonding parameter,
subscript 1 indicates the solvent, and 2 indicates the tar mixture
Methods Results
Over the set of all trials, the gasifier produced 11.8 ± 4.3 g m-3 tar before all
impingers, reported as mean value ± standard deviation. The tar was composed
of 3.6 ± 1.5 g m-3 benzene, 2.3 ± 0.5 g m-3 toluene, 0.2 ± 0.1 g m-3 ethylbenzene,
0.4 ± 0.1 g m-3 xylene, 0.6 ± 0.1 g m-3 styrene, 0.4 ± 0.1 g m-3 naphthalene, and
4.6 ± 2.3 g m-3 unidentified tar. Because of the high variation in tar yield from
trial to trial, tar concentration was compared by solvent on each trial basis and
ranked by the mean of total tar concentration. Solvents were ranked by total tar
concentration for each gasifier run using ANOVA and Tukey test (α=0.05).
Solvents were additionally ranked across all experiments by comparing ratio of
tars measured after the impinger to tars measured before the impinger. Table 2
shows a comparison of the HSP ranking with the experimental results..
Conclusions• HSP did an excellent job of predicting the performance ranking of the solvents tested.
• Benzyl benzoate, biodiesel, and soy oil were predicted and found experimentally to be the
best solvents.
• Diesel and water were found to be poor solvents for tar removal.
• The most noticeable differences between experiments and predictions were for the
solvents hexane and acetone. The higher vapor pressure of these two solvents led to
greater evaporation and higher variation of solvent volume in the impinger during the
post-solvent tar measurement.
1UC Davis Energy Institute, 2Department of Mechanical and Aerospace Engineering, 3Department of Biological and Agricultural EngineeringUniversity of California, Davis
4Universidade Estadual Paulista-Unesp Jaboticabal, Brazil
Z. McCaffrey1,2, P. Iamaguti4, M. Long1,3, L. Wang1,3, B. Jenkins1,3
Solvent selection for tar removal in almond shell gasification
Experimental Setup: A 150 ºC syngas slip-stream entered the multisolvent
manifold and was separated into 4 parallel streams of equal flow (Figure 2).
Each of the parallel streams passed through an impinger of solvent and then
through a backup impinger. Tar was sampled before and after the solvent
impinger to measure tar removal. Flow rate for each of the four streams was
maintained using a rotameter and orifice plate. Flow of the entire slipstream was
maintained and monitored using a diaphragm pump and dry gas meter.
Figure 2: Schematic of multisolvent scrubber
Figure 1: Laboratory gasifier and purification system
Rank Solvent Ra Solvent Exp. Rank
1 Benzyl benzoate 5.2 Benzyl benzoate 1
2 Biodiesel 6.0 Soy oil 1
3 Linolenic acid (Soy oil) 7.7 Biodiesel 2
4 Hexane 8.9 IPA 2
5 Acetone 12.7 Acetone 3
6 IPA 16.1 Ethanol 4
7 Ethanol 19.7 Hexane 4
8 Ethylene glycol 25.8 Ethylene glycol 5
9 Glycerol 26.9 Glycerol 5
10 Water 43.0 Water 6
*Diesel 6
Table 2: (Left 3 colomns) HSP ranking of solvents as relative distance (Ra) to tar
mixture, and (right 2 columns) ranking of the experimental results
ReferencesHansen, C. M., 2007, Hansen solubility parameters: a user’s handbook, 2nd Edition, CRC Press: Boca Raton.
Levin, M. and Redelius; 2012, Determining the Hansen Solubility Parameter of Three Corrosion Inhibitors
and the Correlation with Mineral Oil, Energy & Fuels, vol 26, pp 7243-7250.
Lindvig, T., Michelsen, M. L., Kontogeorgis, G. M. J., 2002, Phase equilibria for complex polymer solutions,
Fluid Phase Equilibria, vol 194-197, pp.663-673.
Segarceanu and Leca, 1997, Improved method to calculate Hansen solubility parameters of a polymer,
Progress in Organic Coatings, Volume 31, Number 4, August, pp. 307-310.
) ) -(+)-(+)-( (4=Ra 2
H2H1
2
2PP1
2
D2D1
* HSP parameters were not available for the complex mixture of petroleum diesel
Solvent ζD ζP ζH
1 Biodiesel (FAME) 16.4 2.6 4.5
2 Benzyl Benzoate 20.0 5.1 5.2
3 Linolenic acid (Soy oil) 15.4 1.5 4.6
4 Hexane 14.9 0.0 0.0
5 Acetone 15.5 10.4 7.0
6 Isopropanol (IPA) 15.8 6.1 16.4
7 Ethanol 15.8 8.8 19.4
8 Ethylene Glycol 17.0 11.0 26.0
9 Glycerol 17.4 11.3 27.2
10 Water 15.5 16.0 42.3
Table 1: Solubility parameters for selected solvents
Table 1 shows the solubility parameters for the selected solvents.
AcknowledgementsWe would like to acknowledge J. Gardner, A. Liu, D. Myers, K. Tolentino, and J. Zhao at UC
Davis for their assistance in conducting the laboratory experiments and R. Cattolica and R.
Seiser from UC San Diego for their contributions and support. The support of the California
Energy Commission is also gratefully acknowledged.
TU Berlin
Institute for Energy Technology
Energy Process Engineering and Renewable Energy
Conversion Technologies (EVUR)
Julian Borgmeyer, M.Sc.Fasanenstr. 89, 10623 Berlin, Germany
Phone.: +49(0)30 314-25972 | Email: [email protected]
Tar and tar related problems are still remaining the foremost obstacles in
development and especially in the implementation of gasification technologies into
today‘s energy supply systems.
Aromatic and polycyclic aromatic hydrocarbons (PAH) are by-products in most high
temperature thermochemical conversion processes. They lower the efficiency of
these processes and form tarry deposits when the gases are cooled and vapors
begin to condense.
It would be desirable to have a tool that is not only capable of monitoring the PAH
composition and load of the product gas of a gasifier but is also sufficiently fast to be
used for process control to minimize the formation of PAHs in the reactor.
In the past, several analytical methods for the characterization of ‘tar’ were introduced
and applied. But up to now, no easy-to-use and reliable tool is available for
monitoring tar with a sufficiently low dead time [1].
Presently, at the TU Berlin further basic research on ‘tar’-fluorescence as well as
development of a compact, robust on-line tool for tar measurement, monitoring and
process control requiring little operating effort are being conducted within two
projects.
Continuous On-Line Tar Monitoring for Process
Control by Application of Optical Emission
Spectroscopy
Julian Borgmeyer*, Eva Brüning, Shaimaa Mahdi, York Neubauer
Fig 1: Scheme of the Jablonski-diagram
showing the principle of absorption and
emission of photons leading to
fluorescence [2].
Fig 2: Emission spectrum of fluorene excited with
240 nm laser light.
[1] Sun, R., Zobel, N., Neubauer, Y., Cardenaz-Chavez, C., Behrendt, F., Analysis of gas-phase
polycyclic aromatic hydrocarbon mixtures by laser-induced fluorescence. Optics and Lasers in
Engineering , 2010, 48, 1231 - 1237
[2] Cullum, B. M.; Chi, Z., Vo-Dinh, T., High-Temperature Fluorescence Measurements and
Instrumentation for Polyaromatic Hydrocarbons (PAH): A Review. Polycyclic Aromatic
Compounds, 2000, 18, 25-47
We would like to thank Ms. Sandra Walther, Mr. Uwe Röhr and Mr. Colin Muxlhanga
for their input and persistent support during setup and operation of the
aforementioned equipment.
Fig. 3: Left: Scheme of the lab setup for the fundamental research on PAH compounds.
Right: Measurement Cell. View from the laser, gas flowing from left to right, spectrometer out on
top.
Figure 3 shows the experimental setup for the analyzation of test gas. The test gas is
being provided by a test gas setup where a syringe pump supplies PAHs dissolved in
toluene into a hot N2 gas flow controlled by a mass flow controller. This setup
provides a very stable PAH load of the carrier gas.
Objective Laser-Induced Fluorescence Spectroscopy (LIF)
Fundamental Analysis of PAH Fluorescence in Hot Gases
Application of the on-line tar monitoring tool for
process control is funded by the German Federal
Ministry of Food and Agriculture (FKZ 22401814)
Acknowledgement
Excitation Emission Matrices (EEM) for Tar Characterization
By varying the excitation wavelength, it is possible to find appropriate excitation
wavelengths to measure different species in the product gas. Figure 4 shows EEMs
of different PAHs and a spectrum of a mix of all these PAHs at the same
concentrations.
Using different wavelengths provides much more information about the components
of a PAH-loaded gas than using just a single-wavelength light source. It is possible to
discriminate components in the gas mixture, e.g. excitation at nearly 300 nm affects
only fluorene, while only phenanthrene is active at 240 nm. The lower signals in the
sum spectrum are caused by absorption.
Light Emitting Diodes for Fluorescence Spectroscopy
Light emitting diodes have several advantages over lasers when it comes to industry
applications. Besides being more stable and reliable especially in a rough
environment they are also much cheaper.
It is possible to combine multiple diodes with different wavelengths to an array to take
advantage of the diverse absorption-emission behavior of different PAHs.
Figure 6 shows the emission spectrum of a 275 nm LED and the corresponding
fluorescence spectrum of naphthalene as a proof of concept for this technology.
Fig. 6: Left: emission spectra of a 275 nm 1 mW UV-LED. Middle: focused UV-LED (dot diameter
approx. 10 mm). Right: Corresponding emission spectrum of naphthalene in N2 carrier gas at 300°C.
Development of an On-Line Analysis Tool
The fundamental research outlined above will be utilized in an on-line analysis tool
suitable for industrial applications in biomass gasification.
Fig. 5: Scheme of the work-in-progress on-line analysis tool that will be placed into operation in
early 2016.
After entering the measurement system, the process gas is being conditioned to
350°C by a preheater. A control valve regulates the process gas flow through the
system. The windows of the measurement cell are purged with nitrogen to prevent tar
deposits that would absorb light. The process gas is excited using a variable-
wavelength LED Array with one wavelength at a time. The irradiated power that
reaches the cell is constantly monitored by a power meter. A spectrometer evaluates
the fluorescence of the process gas. After the measurement cell the process gas is
fully oxidized with air using a heated oxidization catalyst and an oxygen sensor to
ensure an excess of air. The pressure in the whole system is held slightly below
gasifier pressure using a jet pump with pressurized air. The oxygen sensor after the
jet pump allows the calculation of the process gas flow to control the regulating valve
at the inlet.
The setup will be used in an industry-led project to improve the operation of a
biomass-based transport fuel production plant in Gothenburg, Sweden.
Fig. 4: Excitation Emission Matrices of Naphthalene, Fluorene, Phenanthrene and a 3-component
mixture of these at typical concentrations at a biomass gasifier.
Fundamental research on ‘tar’ fluorescence is
funded by the German Federal Ministry of
Education and Research (FKZ 03SF0442)
Variable λLaser
Measurement Cell
Process GasN2 for window
purging
Beam Dump
Spectrometer
Beam Splitter
Power Meter
Absorption
Vibrational Relaxation
Fluorescence Fluorescence
Absorption
S0
S2
S1
Oxygen Sensor
Measurement Cell
GasifierFlow Control & Preheating
CatalyticOxidization
Jet Pump Vent
Flow ControlN2
Air
Oxygen Sensor
Flow Control & Preheating
Flow Control & Preheating
LED Array SpectrometerPower Meter
1 Technische Universität Berlin
Institute for Energy Engineering, Energy Process Engineering and Renewable Energies, Fasanenstr. 89, 10623 Berlin, Germany, Email: [email protected], [email protected] 2 Universität Rostock (University of Rostock) Institute of Electrical Engineering, Tannenweg 22, 18059 Rostock, Germany, Email: [email protected]
INTRODUCTION Gasification of various solid biogeneous feedstocks into widely useable fuel- and synthesis gases needs well working gas cleaning and upgrading at reasonable capital and operational cost. Due to the limitations in scale of biomass based gasification plants (transport of feedstock, available recipients of excess heat) specific cost for these unit operations tend to be high, often too high for the economics of this technology. In this approach of applying non-thermal “cold” plasma (dielectric barrier discharge - DBD) into the gasification process itself or in its downstream we want to enhance tar species conversion and try to intensify the conversion of steam or CO2 on char surfaces. With this process intensification we aim for an increased gasification efficiency, higher synthesis gas yields and finally at an improved gas quality. SETUP
ELECTRIC DISCHARGE BEHAVIOR • When high-voltage (HV) is applied to the DBD reactor electric discharge events occur • Char particles in a fixed bed show discrete current pulses, mainly in 1st and 3rd quarter
(FIGURE 3) • Char particles under fluidized condition show countless discharge pulses in each quarter
(FIGURE 4) • Fluidized char particles cause compared to fixed bed condition higher values of
discharge current, even at lower applied voltage
PROOF-OF-PRINCIPLE
EXPERIMENTAL CONDITION • The proof-of-principle is performed under ambient condition (ambient temperature and
pressure) • Char particles are investigated in a fixed bed and under fluidized condition (FIGURE 5 and
6) • A mixture of CO2 (1.5 vol.-%) and N2 is fed into the char filled DBD reactor and HV is
applied • The gas phase at the outlet of the DBD reactor is analyzed by an Antaris IGS FT-IR
RESULTS • In both condition, fixed bed and fluidized bed, CO is a gaseous product (FIGURE 8) • Fluidization of char particles results in well over twice as much CO
FIGURE 1: Schematic representation of the setup used for proof-of-principle (DBD reactor)
NON-THERMAL PLASMA application for the enhancement of heterogeneous gasification reactions
Philipp Schröder1, Saravanakumar Arumugam2, York Neubauer1
FIGURE 2: Pictorial representation of the DBD reactor. (1) Control unit & data logging. (2) Signal generator. (3) Oscilloscope. (4) Amplifier. (5) Output transformer. (6) DBD reactor. (7) Emergency switch.
PROJECT AIMS • Proof-of-principle under laboratory condition (FIGURE 1 and 2) • Parametric studies on
• Intensification of gas and solid phase interaction (reaction of char surface and its surrounding gas phase) at various temperatures
• Enhancing reactions with steam for increased conversion in heterogeneous and homogeneous reactions
• Decomposition of condensable hydrocarbon species (minimizing tar forming species)
• Evaluate primary energy input: Non-thermal plasma generated ions and radicals to enhance conversion reactions – The electrons are applied directly into the reacting media – !!! note: No energy demanding plasma-torch is used !!!
SAMPLE MATERIAL • Particles used in the DBD reactor are char particles (FIGURE 7) generated during
the gasification of milled natural pine woodchips in a bubbling fluidized bed • Char particles are rich in carbon and have porous structures • Alkaline compounds in the char particles are expected in supporting
heterogeneous catalytic reactions • Char is a highly conductive material
CONCLUSIONS FROM FIRST RESULTS • Heterogeneous reactions of highly-conductive char with gaseous species can be
initiated at ambient condition by the use of a non-thermal DBD plasma • Proof-of-principle shows the conversion of CO2 into CO • Char particles in a fluidized bed lead to higher concentration of CO in the product
gas
ACKNOWLEDGEMENT We want to acknowledge and would like to express our gratitude to the German Federal Ministry of Education and Research for financial support of our current work in the junior research group “NWG-TCKON” (FKZ: 03SF0442).
FIGURE 7: Pictorial representation of char particles
FIGURE 3: DSO screenshot of DBD reactor with char particles in a fixed bed. Applied voltage: 4 kV. Discharge current: 38 mA.
FIGURE 4: DSO screenshot of DBD reactor with char particles in a fluidized bed. Applied voltage: 3 kV, discharge current: 55 mA.
H2O H2O CO2 CO CO2
FIGURE 8: Transmittance spectrum of gas phase sampled at the gas outlet of the DBD reactor
FIGURE 5: DBD reactor with char particles in a fixed bed
FIGURE 6: DBD reactor with char particles in a fluidized bed
Improving the performance of CFD simulation for entrained flow biomass gasifiers by considering intra-particle heat transfer
Kentaro Umeki*, Mikael Risberg
Division of Energy Science, Luleå University of Technology, Sweden. E-mail: [email protected]
Summary Effect of intra-particle heat transfer is significant for large particles (>ca. 200 µm). Modification of devolatilization kinetics is a simple way to include intra-particle heat transfer effect on devolatilization kinetics. The consideration of intra-particle heat transfer showed significant influence on CFD performance.
Optimizing particle size is an important challenge of entrained flow gasifier of biomass. Overall energy efficiency faces trade-offs between energy penalty in milling process and efficiency increase in gasifier when using small samples.
CFD simulation has become an efficient tool to optimize design and operation of gasifiers and burners.
Pulverized flame near the burners are multi-phase, multi-physics, and multi-scale phenomena, but current CFD models do not consider.
1. Swedish National Strategic Research Program, Bio4Energy 2. Bio4Gasification, Swedish Centre for Biomass Gasification (SFC)
Acknowledgments
Background Aim of the study
Isothermal and non-isothermal approaches
Fig. 1 Multi-scale nature of pulverized burners
Develop computationally efficient sub-models for small scale phenomena. Focus on intra-particle heat transfer during heat-up and devolatilization.
Fig. 2 Physical and chemical aspects of multi-scale phenomena
𝜌𝑝𝑉𝑝𝑐𝑝
𝑑𝑇𝑝
𝑑𝑡= ℎ𝐴𝑝 𝑇𝑔 − 𝑇𝑝 + −
𝑑𝜌𝑝
𝑑𝑡𝑉𝑝𝐻𝑑𝑒𝑣
𝜌𝑝𝑐𝑝𝜕𝑇𝑝
𝜕𝑡=1
𝑟
𝜕
𝜕𝑟𝑟𝜆
𝜕𝑇𝑝
𝜕𝑟+ −
𝑑𝜌𝑝
𝑑𝑡∙ 𝐻𝑑𝑒𝑣
Fig. 4 Effect of internal heat transfer Fig. 5 Application of relaxation factor
Fig. 6 CFD simulation with/without intra-particle heat transfer model
Particle scale mass and heat transfer model was developed and coupled with devolatilization kinetics. Isothermal and non-isothermal approaches were compared at various particle sizes and gas temperature. Non-isothermal approach is similar to the reality, but requires much higher computational load. Isothermal approach with relaxation factors to heat transfer coefficient and kinetic parameter was examined
as alternative approach to non-isothermal model.
Application to CFD simulation
Fig. 3 Temperature inside the particle
Isothermal approach
Non-isothermal approach
Isothermal approach
Non-isothermal approach
Experiments
Equipment Experiments were conducted on a 100kW dual fluidized bed (DFB)
steam gasifier which has two columns,
bubbling fluidized bed (BFB) column
circulating fluidized bed (CFB) column
Biomass gasification process occurs
in the BFB column where the biomass
and steam is fed.
Hot, circulating bed material (sand)
flows into the BFB to provide heat for
the endothermic gasification process
and the producer gas flows out from
the column top.
The solid char generated from the gasification process flows out from
the BFB column bottom, together with the bed material, to the CFB
column where the char is combusted. Thus the bed material is
heated.
Operation conditions
Gas and Tar analysis • Gas product was analyzed using Agilent 3000A Micro-GC.
• Tar were analyzed using a newly developed technique, Varian CP-
3800 GC-FID.
• 32 tar compounds were identified in 4 classes.
Acknowledgments • The financial support from New Zealand Ministry of Business, Innovation &
Employment
• Dr. Woei Saw and Dr. Chris Penniall for assistance in the experiments.
• All technicians from CAPE for the mechanical support
• The anonymous reviewers for their insight comments on the paper.
Results 1. Products from the initial devolatilization
2. Correlations between the products of initial devolatilization and the gas composition and tar content
in the final producer gas
3. Influence of steam input on the products from the devolatilization products
Conclusions Relationships between initial devolatilization and
subsequent gasification have been investigated
for product gas composition and yield. It
indicated that the products form devolatilization
significantly influences the final products via the
subsequent steam char reactions.
Effects of operation conditions were also
examined and it is found that higher
temperature and longer residence time
promoted the gas production and reduced the
tar production and concentration.
Ziyin Zhang and Shusheng Pang
Department of Chemical and Process Engineering (CAPE), University of Canterbury, New Zealand
Future plan • Test different feedstock including agriculture residues and
coal/biomass blends.
• Apply catalytic bed materials with target to reduce tar content
and to increase hydrogen content in the producer gas.
Research Objectives The objective of this research is to fundamentally study the initial
devolatilization in steam gasification in a 100kw dual fluidized bed
(DFB) gasifier.
to investigate the products distribution in initial devolatilization;
to develop correlations between the products from initial
devolatilization and final producer gas in steam gasification
BFB
Gasification
BFB
Gasification
CFB
Combustion
CFB
Combustion
Producer
gas Flue
gas
Sand
(heat)
Char & sand Air
Biomass
Steam
BFB temperature 700-800 °C
CFB temperature 750-850 °C
Feedstock Radiata Pine 30 kg/hr
Bed material Silica sand 20 ~ 30 kg
Class Tar class name
C2 Heterocyclic
C3 Light aromatic (1 ring)
C4 Light PAH compounds (2-3 rings)
C5 Heavy PAH compounds (4-7 rings)
CFB
BFB
Biomass Steam
Fig. 1 - 100kW DFB steam gasifier in CAPE Fig. 1 - 100kW DFB steam gasifier in CAPE
Introduction Devolatilization of biomass is the initial stage of gasification process,
it occurs around at 300-500˚C. It is similar to fast pyrolysis in which
the biomass is decomposed to volatile, char and primary tar.
Products form the initial devolatilization significantly influences the
final products of biomass via the subsequent reactions among steam,
char, gas and tar as follows:
• Homogenous reaction: 1 𝐶𝑂 + 𝐻2𝑂 ↔ C𝑂2 + 𝐻2
• Heterogeneous reaction: 2 𝐶 + 𝐻2𝑂 → C𝑂 + 𝐻2,
3 𝐶 + 2𝐻2 → C𝐻4, (4) 𝐶 + 𝐶𝑂2 → 2CO
• Tar cracking reaction: (5) 𝐶𝑚𝐻𝑛 + 𝐻2𝑂 → 𝐶𝑝𝐻𝑞 + 𝐶𝑂2 + 𝐻2
Volatile Volatile
Char Char
Tar Tar
Steam
Dry Devolatilization Steam char gasification
Producer
Gas
Producer
Gas
Tar Tar
Steam gasification process
Biomass
Fig.2 - Tar samples Fig.2 - Tar samples
Fig. 7 – Tar concentration in product gas
from devolatilization (D) and that from steam
gasification (G) at different temperatures
As reaction temperature increased or mean gas residence time increased,
• More volatile was released from biomass – increasing gas production
• Primary tar components were cracked by heat – reducing tar production
As reaction temperature increased or mean gas residence time increased,
• More volatile was released from biomass – increasing gas production
• Primary tar components were cracked by heat – reducing tar production
Fig. 3 - Effect of temperature on the gas and tar
production from devolatilization
Fig.4 - Effect of residence time on the gas and
tar production from devolatilization
Fig. 8 - Effect of steam to biomass (S/B) ratio on
yields of gas species in final steam gasification Fig. 9 - Effect of S/B ratio on the tar concentration
in steam gasification
The reactions between steam and products (gas, tar and char)
from devolatilization lead the production of gases and the
consumption of tars. The extents of changes in different gas
species from devolatilization to final gasification were different,
as the gasification reactions (1-5) had different kinetics which are
also affected by reaction temperature.
The reactions between steam and products (gas, tar and char)
from devolatilization lead the production of gases and the
consumption of tars. The extents of changes in different gas
species from devolatilization to final gasification were different,
as the gasification reactions (1-5) had different kinetics which are
also affected by reaction temperature.
Fig.5 – Gas yields from initial devolatilization
and from steam gasification at different
temperature
Fig 6 – Comparison of gas product from devolatilization and that from
steam gasification at different temperatures
Devolatilization Steam gasification