Heat and power applications of advanced biomass gasifiers
Rutherford
Heat and Power Applications of Advanced Biomass
Gasifiers in the New Zealand Wood Industry
J.P Rutherford and C.J Williamson
Department of Chemical and Process Engineering
University of Canterbury
Private Bag 4800 Christchurch 8020
NEW ZEALAND
E-mail: jpr62@student.canterbury.ac.nz
Abstract
Biomass gasification offers an appealing
cogeneration option for the energy intensive wood
industry. The appeal of biomass gasification stems
from the fact that gasification transforms a solid,
often waste, fuel into a gaseous fuel which retains
75-88% of the heating value of the original
(Higman and Burgt, 2003). A gaseous fuel offers
easier handling and the ability to be utilized in
either a gas engine or a gas turbine. Conventional
biomass cogeneration plants utilize steam turbines
and manage an electrical efficiency of 15-28%,
while integration of a gasifier with a gas turbine or
engine allow efficiencies of 25-40% (Franco and
Giannini, 2005).
2. FICFB GASIFICATION
The FICFB gasifier produces a high hydrogen gas
yield due to the use of steam as the gasifying
agent. The endothermic nature of the gasification
reactions combined with the use of steam as a
gasifying agent requires that there is heat transfer
to the gasification reactor in order for the
gasification to take place. This is achieved through
a twin bed system. The bubbling fluid bed (BFB)
gasification reactor is combined with a circulating
fluid bed (CFB) combustor. The CFB heats an
inert heat carrying medium (sand) which flows
from the CFB to the BFB providing the heat of
reaction. A diagram of the system is shown below.
This paper presents a chemical equilibrium model
for a fast internally circulating fluid bed biomass
gasifier and explores the feasibility for integrating
the gasifier into an energy plant for a medium
density fiberboard (MDF) plant. Two heat and
power applications are considered, integration
with gas engine and integration with a gas turbine
combined cycle.
Flue Gas
Product Gas
CFB
Combustion
Hot Sand
BFB
Gasification
N Steam H 2 O
Biomass &
Nitrogen Gas
LPG
1. INTRODUCTION
In 2004 the BIGAS Consortium research group
was established to develop biomass gasification
technology for New Zealand. This work is
undertaken as part of objective 4 of this research
group. The aim of objective 4 is to provide a
modeling tool to evaluate the economic feasibility
of FICFB gasification for woody biomass energy
plant. Progress to date is reported in this paper.
The approach taken has been to model the gasifier
using chemical equilibrium so that a product gas
composition and heating value can be estimated. In
conjunction process flow-sheets and costing
models have been created for two possible energy
plant concepts.
Air
Cool Sand
& Char
Steam
Figure 1: Diagram of FICFB gasifier
The BFB reactor is screw-fed woody biomass
accompanied by a nitrogen purge gas. The nitrogen
purge gas is used to ensure positive gas flow into
the gasifier hence reducing the risk of fire in the
feed hopper or release of product gas through the
feed system. The biomass is fed in above the fluid
bed. Drying and devolatilization of the biomass
occur immediately upon the biomass entering the
reactor. The heterogeneous char-gasification
reactions have longer reaction rates (Kinoshita and
1
Heat and power applications of advanced biomass gasifiers
Wang, 1993, Fiaschi and Michelini, 2001) and will
occur throughout the BFB. The BFB has a sand
bed fluidized with steam. During gasifying the bed
will also contain significant amounts of char. The
sand and char bed material flow from the BFB
through a chute fluidized with either air or steam
into the CFB. Inside the CFB, the char and any
additional fuel in the form of LPG is combusted.
The CFB is a sand bed fluidized with air. Air rates
are maintained to provide excess air conditions.
The CFB air velocity is significantly greater than
the steam velocity in the BFB (7ms-1 compared to
1.5ms-1) and hence the sand is entrained up and out
of the CFB. The sand entrained out of the CFB is
separated from the flue gases by a cyclone and fed
back through a siphon into the BFB. The hot sand
settles at the bottom of the siphon preventing flow
of the BFB product gas out through the siphon.
The sand is then fluidized with either air or steam
up and over into the BFB. The sand, having passed
through the combustion reactor, is hotter than the
BFB bed and cools providing the heat for the
gasification reactions. The product gas from the
BFB flows out of the top of the BFB and through a
cyclone, to separate particulates, before being
burnt in an afterburner. When the FICFB is
integrated into a process the afterburner would be
replaced with either a boiler system, chemical
reactor, gas engine or a gas turbine.
3.2.
Rutherford
Parameters
N char is the mols of carbon in the wood which do
not take part in the BFB reactions. It is assumed
that this carbon is completely combusted in the
CFB
N Steam is the mols of steam entering the BFB
N moisture is the mols of moisture entering the BFB
with the wood
NWood is the mols of carbon in the wood entering
the BFB. This is equivalent to the number of mols
of wood given wood is in the form CHH/COO/C
N Purge is the mols of nitrogen purge gas entering
the BFB
H/C is the hydrogen to carbon ratio of the wood
O/C is the oxygen to carbon ratio of the wood
3. CHEMICAL EQUILIBRIUM MODEL
The various flows around the gasifier are
simplified and modeled as shown in the Figure 2.
A steady-state equilibrium model has been
developed to predict the composition of the
product gas from a FICFB gasifier so that
preliminary feasibility studies can be undertaken
for the integration of a FICFB gasifier into
different heat and power applications.
3.1.
Variables
yCH 4 is the mol fraction of methane in product gas
yCO2 is the mol fraction of carbon dioxide in the
product gas
yCO is the mol fraction of carbon monoxide in the
product gas
y H 2 is the mol fraction of hydrogen in the product
gas
y H 2O is the mol fraction of steam in the product gas
N gas is the mols of product gas
Figure 2: FICFB Model Diagram
3.3.
Modeling Approach
The model of the FICFB gasifier is based around
equation 3, which represents the governing
reaction for the BFB gasifier. Equations for three
of the unknown variables are found using
elemental balances for C, H and O shown
(equations 4-6), which can be rearranged to show
the dependence of the molar flow of CO, CO2 and
CH4 on the molar flow of H2 and H2O (equations
7-9). The mol fractions of H2 and H2O are found
through chemical equilibrium (equations 10-11)
using the reactions below:
CO2  H 2  CO  H 2 O
k1 = equilibrium constant
(1)
CH 4  H 2 O  CO  3H 2
k2 = equilibrium constant
(2)
The molar flow of product gas can be found by
equation 12.
2
Heat and power applications of advanced biomass gasifiers
Rutherford
Governing Reaction
NWood CH ( H / C ) O(O / C )  ( N Steam  N moisture ) H 2 O  N purge N 2  N gas  yCH 4  yCO2  yCO  yH 2  yH 2O  y N2   N char C
(3)
Carbon Balance
N wood = N CH4 +N CO2  NCO  N char
(4)
Hydrogen balance
H
N wood +2N steam +2N moisture = 4N CH 4  2 N H 2O  2 N H 2
C
(5)
Oxygen Balance
O
N wood +Nsteam +N moisture = 2N CO2  NCO  N H 2O
C
(6)
Methane
 H/C.N wood Nsteam  N moisture N H2O  N H2 
NCH 4  
+

4
2
2


(7)
Carbon Monoxide


 H_C

NCO =  2(N wood -N char )- 
+O_C  *N wood -2(Nsteam +N moisture )+2N H2O +N H2 
 2



(8)
Carbon Dioxide
3(Nsteam  N moisture ) 3N H2O +N H2 

H/C 
N CO2 =  O/C+
N wood -(N wood -Nchar )+


4 
2
2


(9)
Steam
y H2O 
y CO2 y H2
y CO k1
Hydrogen
y H2  3
y CH4 y H2O
y CO k 2
Product Gas Yield
N gas =
(10)
 Preac 


P
 0 
11-3-1
(11)
N CH4 +N CO2  N CO  N purge
1  y
H 2O
 yH 2
(12)

_______________________________________________________________________________________
Equations 7-12 provide a system of linear and nonlinear equations which Microsoft Excel Solver is
used to solve.
The model presented above was then further
adapted to include the possibility of solid carbon as
a product. At low temperatures and low H2O to
biomass ratios it is possible for incomplete carbon
conversion. Hence some carbon will remain in
solid form and not be gasified. To enable the
model to deal with solid carbon the reaction set
was adapted to include 3 reactions (equations 1315) in situations where carbon was present as a
product.
C  H 2 O  CO  H 2
(13)
C  H 2  CH 4
C  CO2  2CO
(14)
3.4.
(15)
Results
Figure 3 and 4 present the model results from
gasifying one kmol of wood, modeled as
CH1.43O0.62, with varying H2O to biomass ratios and
temperatures. Chemical equilibrium is dependant
only on the elemental abundances in the reactor,
hence moisture content of the wood has not been
3
Heat and power applications of advanced biomass gasifiers
specified in reporting these results but is included
in the H2O to biomass ratio shown in the Figures 3
and 4. If char circulation rate was considered it has
the effect of removing carbon from the reactor,
which acts in a similar way to increasing the H2O
to biomass ratio. In order to clearly illustrate the
trends evident from thermodynamic modeling char
circulation and nitrogen flow have been set to zero.
525
475
425
375
325
275
1350
1250
1150
Temp
1050
(K)
950
225
LHV of Product Gas
(MJ/kmol of C in System)
575
0.6
0.4
0.2
0
1
0.8
175
Molar H2O to Biomass Ratio
carbon conversion due to shifting the endothermic
char gasification reactions (13-15) to the right.
However increased temperature decreases lower
heating values at steam to biomass ratios above
complete carbon conversion due to the water-gas
shift (equation 1) promoting the formation of H2 at
the expense of CO.
Chemical equilibrium modeling suggests the
optimal operating point is at high temperature and
at a H2O to biomass ratio that results in complete
carbon conversion but without excessive dilution
of the product gas. At a temperature of 1300K and
a H2O to biomass of 0.51 (complete carbon
conversion) the equilibrium composition of the
product gas is stated below:
H2
CO
CH4
CO2
H2O
50%
36%
4%
8%
2%
(mol/mol)
(mol/mol)
(mol/mol)
(mol/mol)
(mol/mol)
LHV
LHV
255
11
MJ/kmol
MJ/Nm3
It should be noted that this model represents pure
chemical equilibrium and has not been validated
against experimental results. This process will be
undertaken once results become available from the
University of Canterbury FICFB gasifier.
1
0.8
0.6
0.4
0.2
265
255
245
235
225
215
205
195
185
175
0
LHV of Product Gas
(MJ/kmol of Product Gas)
Figure 3: Lower heating value of product gas
(MJ/kmol of C in system)
Rutherford
1350
1250
1150
Temp
1050
(K)
950
Molar H2O to Biomass Ratio
Figure 4: Lower heating value of product gas
(MJ/kmol of gas)
Two major trends are evident in Figures 3 and 4.
At low H2O to biomass ratios not all of the carbon
in the system is being gasified. Increasing the H2O
to biomass ratio results in greater carbon
conversion, hence greater gas yield and increased
lower heating value of the product gas per kmol of
carbon in the system. This is shown in Figure 3.
Once the H2O to biomass ratio exceeds 0.5-0.7
complete carbon conversion is attained and
increasing the H2O to biomass ratio results in
dilution of the product gas, reducing the lower
heating value per kmol of gas. This is shown in
Figure 4.
Increasing temperature increases the lower heating
value at steam to biomass ratios below complete
4. PROCESS INTEGRATION
The FICFB gasification model allows the gas yield
and heating value of the product gas to be
estimated from simple gasification parameters.
Once the characteristics of the product gas are
known the effect of integration of the FICFB
gasifier with downstream plant can be assessed. In
this paper integration with either a gas engine or
with a gas turbine combined cycle is considered.
Figure 5-6 show basic diagrams of these processes.
Feed
Handling
FICFB
Gasifier
Gas
Cleaning
Gas
Engine
Figure 5: Gasification-Gas Engine Energy Plant
Feed
Handling
FICFB
Gasifier
Gas
Cleaning
Gas Turbine
Combined
Cycle
Figure 6: Gasification-Combined Cycle Plant
4
Heat and power applications of advanced biomass gasifiers
Undertaking economic feasibility assessments of
these processes draws on work from each of the
objectives of the BIGAS Consortium. Energy
demand modeling of an MDF plant (Li and Pang,
2006) provided estimates of the energy demands
from a MDF plant. Estimates for a MDF plant are
shown below:
Table 1: Energy Demand of a MDF Plant
MDF Panel Output
Electricity required
Thermal Oil required
9 bar saturated steam
4 bar saturated steam
380°C Flue gas required
120,000m3/yr
4.8 MW
2.6 MW
2.5 tonnes/hr
4.6 tonnes/hr
74 tonnes/hr
Objective 3 also provides information on woody
biomass availability and cost for different New
Zealand regions. Objective 2 provides technical
information on the operational limits of the FICFB
gasifier and information for validation of
composition modeling (Brown et al., 2006). Future
work under objective 2 will provide detailed
process designs for the gas cleaning equipment
required to integrate the gasifier with an engine or
turbine.
4.1.
Gas Engine Plant
A typical industrial cogeneration scale gas engine
is a turbocharged, intercooled, spark ignition
engine. They can either be operated at
stoichiometric air for maximum power or at lean
burn conditions which minimize NOx emissions.
For modeling purposes it is assumed that the
engine is operated in lean-burn conditions, with an
air to fuel ratio of 1.6 times the stoichiometric airfuel ratio (Major, 1995).
Gas engines are available in a range of scales.
Jenbacher produce cogeneration engines with
scales ranging from 500 kWel to 1.2 MWel (Herdin,
2006). For larger electrical outputs, it is usual to
use a number of gas engines in parallel. The capital
cost of the 1.2 MWel natural gas engine is quoted at
NZ$1180 /kWel1 and operates with an electrical
efficiency of 46.7% based on LHV (Herdin,
2006). For larger scales it is typical to use a
number of gas engines in parallel. This limits the
economies of scale possible with gas engines.
A typical energy balance for a gas engine is
presented in Table 2. One should note the
1
Quoted as €571 2004/kWel
Rutherford
prevalence of low-grade heat due to the
requirements of cooling the engine. An MDF plant
has limited requirements for low grade heat.
Table 2: Energy balance for a Jenbacher 620
GS Series Engine
Energy Form
Power Output
90°C Hot water from
Engine Cooling
180°C Steam from
engine exhaust gases
Losses
Percentage
41.9%
29.4%
14.1%
13.8%
The efficiencies presented are those for gas
engines operating on natural gas. Reported
electrical efficiencies running on non-natural gas
fuels are closer to 30% (Herdin et al., 2003). The
reduction in efficiency is due to the gasifier where
the product gas retains 75-88% of the heating value
of the feed (Higman and Burgt, 2003) and
requirements to cool the gas to ~40°C so that it can
be injected into the engine (Jenbacher, 2002).
4.2.
Gas Turbine Combined Cycle Plant
A typical gas-fired natural gas turbine combined
cycle (GTCC) unit consists of a single fuel gas
turbine, unfired multi-pressure heat recovery steam
generator with no bypass stack, multi-pressure
condensing steam turbine, electric generators, stepup transformer, water cooled heat rejection. The
smallest of these units quoted in the Gas Turbine
World Handbook (Gas Turbine World, 2005) has
an electrical output of 7.3 MW and operates with
an electrical efficiency of 39.5% based on LHV.
Combined cycles exhibit economies of scale. The
free-onboard (FOB) at the factory NZ$2005 price
of a 7.3 MWel system described above is
$1200/kWel2. A 50 MWel plant will have a capital
cost per kWel 85% of this and a 100 MWel will
have a capital cost 66% of this (Gas Turbine
World, 2005).
Gas turbines suffer from similar efficiency
reductions as gas engines once integrated with a
gasifier. A benefit of gas turbine combined cycles
is when integrated into a process they can provide
high-grade heat at the expense of steam generation.
Typical exhaust temperatures of gas turbines are
500-550°C (Traverso et al., 2004). However it is
generally considered that gas turbines have stricter
gas cleaning requirements than gas engines
(Scharpf and Carrington, 2005).
2
Quoted as $US751 2004/kWel
5
Heat and power applications of advanced biomass gasifiers
Adaptation of gas turbines to product gas may
require modification to the combustion chamber in
order to be suitable for burning lower calorific
value fuel. Standard gas turbines are designed for
natural gas, which has a HHV of around 39
MJ/Nm3 (Baines, 1993) compared to 11 MJ/Nm3
for FICFB producer gas. Rodriques et al (2003)
suggests that this could add between 3 and 20% to
the capital cost of a gas turbine. These
modifications are not novel, GE has developed
combustion chambers specifically for lower
heating value fuels from gasification and has
gained 340,000 hours experience in operating these
turbines (Jones and Shilling, 2003)
5. Conclusions
The discussion presented is a preliminary synopsis
of two of the more appealing uses of a FICFB
gasifier. The prime mover in each case had similar
capital costs. The major differences were that, at
the scales discussed, gas engines offered a higher
electrical efficiency but at the expense of highgrade heat. Selection of mover will likely depend
on matching the plant with process heat demand,
the quality of gas required for the mover with what
is reliably possible from gasification and by
auxiliary costs involved with each mover.
Significant work is still required on characterizing
the product gas, developing gas cleaning
procedures and adaptation of the prime movers to
use with the product gas. Once this work has been
completed more detailed discussions about the
integration of gasification energy plants into wood
processing plants can be made.
Rutherford
Fiaschi, D. and Michelini, M. (2001), A two-phase onedimensional biomass gasification kinetics
model, Biomass and Bioenergy, 21, 121-132.
Franco, A. and Giannini, N. (2005), Perspectives for the
use of biomass as fuel in combined cycle power
plants, International Journal of Thermal
Sciences, 44, 163-177.
Gas Turbine World (2005) Gas Turbine World
Handbook, Pequot Publishing Inc.
Herdin, G. (2006), GE Jenbacher Gas Engine Costs,
(Ed, Rutherford, J.) Achenseetr.
Herdin, G., Robitschko, R., Klausner, J. and Wagner,
M. (2003), GEJ Experience with Wood Gas
Plants, GE Jenbacher, Achenseestr.
Higman, C. and Burgt, M. (2003) Gasification, Gulf
Professional Publishing, Burlington.
Jenbacher, G. (2002), Fuel gas quality. Technical
Instruction No: 1000-0300, (Ed, Dieter, C.),
pp. 1-10.
Jones, R. and Shilling, N. (2003), IGCC Gas Turbines
for Refinery Applications, (Ed, Systems, G. P.)
GE, Schenectady.
Kinoshita, C. and Wang, W. (1993), Kinetic Model of
Biomass Gasification, Solar Energy, 51, 19-25.
Li, J. and Pang, S. (2006), Modelling of energy demand
in a MDF plant, In Chemeca, Vol. ID112
Auckland.
Major, G. (1995) Learning from experiences with smallscale cogeneration, CADDET, Sittard.
Scharpf, E. and Carrington, G. (2005), Wood-derived
producer gas cleanup and tolerances, Delta S
Technologies, University of Otago, Dunedin.
Traverso, A., Cazzola, W. and Lagorio, G. (2004),
Widget-Temp: A novel web-based approach for
thermoeconomic analysis and optimization of
conventional and innovative cycles, In ASMEIGTI Turbo ExpoAMSE, Vienna, pp. 1-9.
6. ACKNOWLEDGEMENTS
The authors acknowledge the support of New
Zealand Foundation for Research Science and
Technology (FRST) and the assistance of Jock
Brown, Rick Dobbs, Jingge Li, Kimberly
Robertson and Shusheng Pang from the BIGAS
Consortium.
7. REFERENCES
Baines, J. (1993) New Zealand Energy Information
Handbook, Taylor Baines and Associates,
Christchurch.
Brown, J., Dobbs, R. and Gilmore, I. (2006), Biomass
Gasification in a Fast Internal Circulated
Fluidised Bed Gasifier, In ChemecaAuckland.
6