1
FINAL TECHNICAL REPORT
January 1, 2013, through September 30, 2013
Project Title: ADVANCED OXY-COMBUSTION TECHNOLOGY
DEVELOPMENT AND SCALE-UP FOR NEW AND EXISTING
COAL-FIRED POWER PLANTS
ICCI Project Number: DEV12-2
Principal Investigator: David Rue, Gas Technology Institute (GTI)
Project Manager: Debalina Dasgupta, ICCI
ABSTRACT
The Gas Technology Instittue (GTI) has developed a pressurized oxy-coal fired molten
bed boiler (MBB) concept, in which coal and oxygen are fired directly into a bed of
molten coal slag through burners located on the bottom of the boiler and fired upward.
Circulation of heat by the molten slag eliminates the need for a flue gas recirculation loop
and provides excellent heat transfer to steam tubes in the boiler walls. Advantages of the
MBB technology over other boilers include higher efficiency (from eliminating flue gas
recirculation), a smaller and less expensive boiler, modular design leading to direct
scalability, decreased fines carryover and handling costs, smaller exhaust duct size, and
smaller emissions control equipment sizes. The objective of this project was to conduct
techno-economic analyses and an engineering design of the MBB projeces and to support
this work with thermodynamic analyses and oxy-coal burner testing.
Techno-economic analyses of GTI’s pressurized oxy-coal fired MBB technology have
found that the overall plant with compressed CO2 has an efficiency of 31.6%. This is a
significant increase over calculated 29.2% efficiency of first generation oxy-coal plants.
Cost of electricity (COE) for the pressurized MBB supercritical steam power plant with
CO2 capture and compression is calculated to be 134% of the COE for an air-coal
supercritical steam power plant with no CO2 capture. This compares positively with a
calculated COE for first generation oxy-coal supercritical steam power plants with CO2
capture and compression of 164%. The COE for the MBB power plant is found to meet
the U.S. Department of Energy (DOE) target of 135%, before any plant optimization. The
MBB power plant is also determined to be simpler than other oxy-coal power plants with
a 17% lower capital cost. No other known combustion technology can produce higher
efficiencies or lower COE when CO2 capture and compression are included.
A thermodynamic enthalpy and exergy analysis found a number of modifications and
adjustments that could provide higher efficiency and better use of available work.
Conclusions from this analysis will help guide the analyses and CFD modeling in future
process development. The MBB technology has the potential to be a disruptive
technology that will enable coal combustion power plants to be built and operated in a
cost effective way, cleanly with no carbon dioxide emissions. A large amount of work is
needed to quantify and confirm the great promise of the MBB technology. A Phase 2
2
proposal was submitted to DOE and other sponsors to address the most critical MBB
process technical gaps. The Phase 2 proposal was not accepted for current DOE support.
3
EXECUTIVE SUMMARY
The techno-economic Study Design Basis & Methodology report was issued to NETL by
GTI and the project partner Nexant. This report included power plant design criteria, cost
estimation methodology, and financial bases that were used in the engineering design and
economic analysis of a power plant based on the molten-bed oxy-coal boiler. NETL
reviewed and commented on the report. The report was revised accordingly by GTI and
Nexant.
The designed PC power plant is a pressurized oxy combustion supercritical steam-electric
generating plant with 90% carbon capture and generating a nominal 550 MWe. The
combustion block contains the following major systems that are directly associated with
pressurized oxy-combustion of coal:
Pulverized Coal Pressurized Storage and Transport/Feed System
Pressurized Submerged Combustion Molten Bed Furnace/Boiler
Pressurized Convective Superheater, Reheater, Economizer, and Condenser
CO2 -rich Flue Gas Treating and Conditioning Facilities
Product CO2 Recovery, Purification and Compression Facilities
O2 Booster Compressor
The MBB plant will be optimized to have higher efficiency and lower capital cost. Since
engineering design of the plant is based on the MBB oxy-coal technology, the
MBB/steam generator is also a subject of engineering design and optimization. GTI is
currently approaching boiler makers about getting their advice on the boiler design.
The ASPEN Plus model of the MBB process has been completed. Plant efficiency is
found to increase from 29.2% for a first generation oxy-coal power plant to 31.6% for the
MBB case. Further optimization is expected to lead to some further increase in plant
efficiency. The cases assumed firing the same quantity of Illinois #6 coal. This allows
direct comparisons of cases. Higher efficiency for the MBB case leads to higher gross
and net power production. Overall auxiliary load is lower for the MBB. Operation at 150
psig reduces the compression demand for CO2 by 38.5 MWe which is somewhat more
than the added power demand of 35.4 MWe for oxygen compression. Still, the CO2
compression saving is greater than the added energy to compress oxygen. The single
highest MBB power plant demand is for makeup water. This 4% increase results from the
current decision to not clean and recycle the more saline reject water because of high
cost.
The overall power plant cost summary leading to determination of the COE is shown in
Table 1. The MBB case in the current work is found to have a lower total plant cost and
lower total overnight cost. COE in $/MWh is 17% lower for the MBB case with
individual pressure shells than for the DOE 5C atmospheric pressure oxy-coal
supercritical steam power plant with CO2 capture and compression. The COE for the
MBB case with CO2 capture and compression is found to be 134% of the COE for the
baseline air-coal supercritical steam power plant with no CO2 capture. The Table 1 COE
4
increase for MBB with separate pressure shells is shown as 131%. But this reflects a
larger plant case with constant fuel input. Scaling to 550 MWe increases the COE to
108.2 mils/kWh, or 134% of the baseline air-coal case. This is a significant improvement
over the approximately 150% COE for the DOE oxy-coal 5C case compared with the air-
coal baseline case. The 34% increase in COE for the MBB-based power plant, even with
no optimization, is very close to the DOE target of a COE increase of only 35%.
Optimization analyses in Phase 2 will improve efficiency and have an excellent chance of
meeting the challenging target of a COE increase of only 35%.
Table 1. Cost Summary for Oxy-Coal Supercritical Steam Power Plants with Carbon
Capture and Compression: DOE 5C Case, Nexant Calculation of DOE 5C Case,
Proposed Molten Bed Boiler Case
Prof. Dale Tree of BYU conducted an initial thermodynamic analysis of the MBB along
with the radiant and convective sections above the boiler. The work included both first
and second law energy and exergy analyses. The results indicate locations in the process
where available heat is present and where there is the potential to produce work. Energy
and exergy analyses are useful in guiding engineers seeking to optimize processes and
identify ways in which to improve the overall efficiency of processes.
Using the mass flow rates and gas species concentrations produced by NEXANT for
NETL Case 5C, the exergy of each flow for the fuel/air and steam into and out of the
boiler was calculated. The results show a decrease in exergy and enthalpy for the gas
phase of 6,120.4 and 5,988.6 MMBtu/hr respectively. The increase in steam exergy and
enthalpy were found to be 3,290.6 and 5,931.0 MMBtu/hr. The net change in exergy and
enthalpy for the combined gas and steam flows into and out of the boiler showed a net
loss of 57.6 MMBtu in enthalpy which is a small fraction (1%) of the total energy
exchanged between the combustion gas and the steam. This loss can be attributed to heat
transfer to the surrounding that does not enter the steam.
This is not the case for exergy. Steam exergy increase was found to be lower than the loss
in gas phase exergy. Exergy destruction occurs primarily from two physical processes in
the boiler 1) Chemical reactions which convert chemical energy to heat and 2) Heat
transfer from a high temperature gas to the molten bed and then to the steam. Exergy
Case 11 Case DOE 5C Case 5C-N Case GTI Case Ind-GTI
Supercritical PC
w/o CO2 Capture
(3,500 psig, 1,100 ◦F, 1,100 ◦F)
550.0 548.7 559.2 592.9 592.9
409,528 549,471 549,471 549,471 549,471
Net Efficiency 39.3% 29.2% 29.8% 31.6% 31.6%
TOC $MM 1348 2161 2229 2065 1916
OCfix $MM/yr 38.8 57.8 59.3 55.5 52.0
OCv ar w/o fuel $MM/yr 37.3 46.3 48.7 47.1 45.6
Fuel $MM/yr 123.0 165.0 165.0 165.0 165.0
COE mills/kWh 80.95 123.7 124.2 111.4 106.1
100% 153% 153% 138% 131%
Net Power MWe
Coal Feed Rate, LB/hr
% Case 11 COE
GTI MBB Cost of Electricity Summary, June 2011 Cost Base
Case
Description Supercritical Oxycombustion Based
with CO2 Capture and Purification
(3,500 psig, 1,110 ◦F, 1150 ◦F)
5
destruction can be reduced by 1) increasing the temperature of combustion products and
2) decreasing the temperature difference between the combustion products and steam.
Optimization of the MBB to reduce exergy destruction will require a more detailed boiler
model. The current model assumes complete coal combustion including solid carbon and
volatiles within the melt. After releasing all of the energy from combustion into the melt,
heat is transferred from the melt to the high pressure steam. The products of combustion
(primarily CO2 and H2O) leaving the melt are assumed to be at the molten bed
temperature of 1632°C (2970°F) and heat is transferred from these combustion products
to both the high and intermediate pressure steam. The simplified process is not likely to
be the most efficient or the process that generates the least exergy destruction. Future
modeling of the MBB will consider the combustion process distributed throughout the
molten bed and will allow staged combustion with an overfire region.
GTI previously developed, designed, fabricated, tested, and successfully commercialized
patented oxygen-natural gas burners for the submerged combustion melting (SCM)
furnace. The SCM technology has been used to melt a wide range of mineral materials
including mineral wool, cement kiln dust, electric arc furnace dust, simulated high level
radioactive waste, and many industrial glasses (fiberglass, container glass, etc.). The
prototype burners to operate on coal and oxygen will be derived by modification these
proven burners. The design size for testing the oxy-coal burners is a firing rate of 0.5-1
MMBtu/hr. This is considered a reasonable firing rate for the laboratory demonstration of
oxy-coal combustion in a molten bed.
The initial testing objective was to feed coal at a controlled rate of 20 to 40 lb/hr as
needed for testing. Carrier gas is only 5-10 pounds per hour. The feeder chamber was
modified to enable control of pressure in the tank, to feed coal through a valve at the
bottom of the tank at a uniform rate, and to control carrier gas rate to carry coal through a
transport line into the burner. The feed system was designed to deliver nitrogen carrier
gas at up to 100 psig. Shakedown tests were carried out using sand as fine as 60-80 mesh.
The feeder was placed on a scale. Pressurization of the feed tank was used to push fine
particles through a valve and then through a transfer line. The transfer line traveled up the
center of the burner to the burner tip. Cold flow tests were conducted with fine sand
particles and pulverized coal to determine initial sizing of burner transport line and to
select proper pressures to push fine particles.
Cold flow testing was conducted with several different particles, particle sizes, and
nozzle designs. The goal was to find a nozzle size and shape to deliver the desired rate of
pulverized coal without flow interruption. Initial work was carried out with two different
particle sizes of silica sand. After selection of a workable nozzle design, testing was
conducted with a Powder River Basin (PRB) coal. Engineers set up the simulated coal
feeder and simulated burner with oxygen and natural gas on a test stand. The simulated
burner is water cooled and has inlet natural gas and oxygen lines as well as the coal feed
line. The coal feeder was again placed on a scale to allow determination of coal feed
rates. The natural gas and oxygen flows were controlled by mass flow controllers.
6
Burner tests were conducted by igniting the simulated burner with natural gas and oxygen
and establishing a stable flame. Coal was then introduced to the flame with natural gas
still flowing. The dual fuel flame was operated for some time, and then the natural gas
was shut off to create the oxy-coal flame. Coal feeding with pulverized coal was difficult,
but methods were developed to keep coal flowing into the transport line.
GTI engineers and technicians have successfully demonstrated that the simulated version
of the oxy-coal burner can be operated under controlled conditions. Later tests have been
conducted to vary the oxygen to coal ratio and the coal feed rate. Later project work will
focus on installing the simulated oxy-coal burner on a molten bed furnace and firing the
burner up into the melt. This work will confirm the ability to fire oxygen and coal
directly into a bed of molten slag from below. During that test, engineers from REI will
collect gas samples, analyze those samples, and then conduct corrosion analyses. All
information learned from the molten bed furnace trials will be available to the project
team in planning for development testing in Phase 2.
Corrosion analyses are planned to be carried out with samples collected during trials of
oxy-coal combustion in a molten bed. The molten bed oxy-coal firing trials have not yet
been conducted. Therefore, the corrosion gas analyses and corrosion analyses could not
yet be performed. This work was scheduled for the last quarter of the project. Sampling
was to be carried out by REI engineers during trials at GTI. REI engineers were to
analyze the collected gas samples and conduct the corrosion analyses using models
available from other coal boiler testing and analysis.
7
OBJECTIVES
The overall objective of this project was to demonstrate that the pressurized molten bed
oxy-coal combustion process with >90% carbon capture can operate with a COE increase
under 35% compared with a baseline air-coal power plant with no carbon capture.
Engineering and economic analyses for the comparative air-coal power plant with no
carbon capture and the atmospheric oxy-coal power plant with carbon capture and
compression were reported in recent DOE published reports and were referenced in this
work. The objective was to be met through engineering design and economic analysis of
the proposed pressurized oxy-coal MBB and a 550 MWe power plant based on the MBB
technology. Work also included thermodynamic energy and exergy analyses, oxy-coal
burner testing, and corrosion assessment.
INTRODUCTION AND BACKGROUND
The heart of the GTI pressurized, oxy-coal wet bottom steam generator concept is a
submerged combustion molten bed (SCM) furnace (Figure 1) that offers higher efficiency
than other elevated pressure oxy-coal boilers by greatly reducing the flue gas
recirculation (FGR). The unique combustion and heat transfer design employs a smaller
and less expensive boiler with reduced heat exchanger surface area.
Molten
Bed
Boiler
Air
Steam to Power
Production
CO2
Co
al
Cra
sh
ing
Secondary
Oxygen
Flue Gas Recirculation
Slag Discharge
Coal
Oxygen
Coal
Carbon
Capture
Air
Separation
Unit
Flue Gas
Feedwater
Figure 1. Pressurized Molten Bed Oxy-Coal Boiler
GTI has specified initial parameters and boundary conditions to be provided for the
engineering design mass and energy balance models which have been developed by
Nexant. These include preliminary estimated geometry (dimensions) of the pressurized
oxy-coal boiler, heat transfer surface areas, heat fluxes, coal type and composition, flue
gas composition, pollutant gases composition, slag physical properties, etc. Based on
these parameters, the performance and cost of an integrated pressurized oxy combustion
supercritical pulverized coal (SCPC) power plant with CO2 recovery has been estimated.
The designed PC power plant in this study is a pressurized oxy combustion supercritical
steam-electric generating power plant with >90% carbon capture and generating a net 550
8
MWe. The combustion block contains the following major systems that are directly
associated with pressurized oxy combustion of coal (Figure ):
Pulverized Coal Pressurized Storage and Transport/Feed System
Pressurized Submerged Combustion Molten Bed Furnace/Boiler
Pressurized Convective Superheater, Reheater, Economizer, and Condenser
CO2 -rich Flue Gas Treating and Conditioning Facilities
Product CO2 Recovery, Purification and Compression Facilities
O2 Booster Compressor
Figure 2 shows a preliminary power plant design. Since the engineering design of the
power plant is based on the molten bed oxy-coal technology, the MBB/steam generator is
also a subject of engineering design and optimization.
Figure 2. Pressurized Molten Bed Oxy-Coal Steam Generator Power Plant Flow Diagram
EXPERIMENTAL PROCEDURES
The majority of the project effort involved techno-economic analyses, engineering
design, and thermodynamic analyses of the pressurized, oxy-coal molten bed boiler
technology concept. There was no experimental setup required for this work. Work was
conducted to study the firing of coal with the modified oxy-caol burner. The setup of that
equipment was part of this project. No materials or equipment were purchased in this
project. All materials were purchased under the DOE matching funds to this project.
GTI previously developed, designed, fabricated, tested, and successfully commercialized
patented oxygen-natural gas burners (Figure 3) for the SCM. The SCM technology has
been used to melt a wide range of mineral materials including mineral wool, cement kiln
Acid Condenser
Carbon
Sequestration
Unit
Flue
GasCoal
Crashing
/Sizing
Coal
Secondary
Oxygen
HP
Turbine
Slag Discharge
IP
Turbine
LP
TurbineGenerator
Sequestrated
CO2
Vented Gas
Feedwater
Reheat
Bleed1
Heater1 Heater2
Bleed2
Deaerator
Bleed3
HP Pump
Cooling
Water
Condenser
LP Pump
Coal
Steam
Storage/
Transport
Coal Preparation
Air
Compressor
Oxygen
Oxygen
Nitrogen
Oxygen
Separation
Flue Gas Recirculation
Molten Bed
Boiler
9
dust, electric arc furnace dust, simulated high level radioactive waste, and a wide range of
industrial glasses (fiberglass, container glass, etc.). The prototype burners to operate on
coal and oxygen will be derived from modification of these already proven burners. The
design size for testing the oxy-coal burners is a firing rate of 0.5-1 MMBtu/hr. This is
considered to be a reasonable firing rate for the laboratory demonstration of oxy-coal
combustion in a molten bed.
The project team decided to conduct the oxy-coal burner testing at GTI instead of BYU
as planned. The BYU facilities do not have a pneumatic coal feed system able to supply
pulverized coal to the oxy-coal MBB. GTI engineers have adapted a commercial sand
blasting unit for coal delivery. A photograph of one early configuration of this unit is
shown in Figure 4. The initial testing objective was to feed coal at a controlled rate of 20
to 40 lb/hr as needed for testing. Carrier gas is only 5-10 pounds per hour. The feeder
chamber was modified to enable control of pressure in the tank, to feed coal through a
valve at the bottom of the tank at a uniform rate, and to control carrier gas rate to carry
coal through a transport line into the burner. The feed system was designed to deliver
nitrogen carrier gas at up to 100 psig. Shakedown tests were carried out using sand as
fine as 60-80 mesh.
Oxygen
Natural
Gas
Cooling
Water out
Cooling
Water in
Cooling
Water out
Cooling
Water in
(a) (b)
Figure 3. GTI’s Water-Cooled Burners for Oxy-Fired Submerged Combustion Melter:
Burner With Center Nozzle for Natural Gas; (b) Burner With Peripheral Nozzles for
Natural Gas
Cooling
Water in
Cooling
Water out
10
Figure 4. Coal Feeder for Pneumatic Transport of Coal into the Oxy-Coal Burner
Figure 5a shows the setup for initial oxy-coal burner nozzle testing. The feeder was
placed on a scale. Pressurization of the feed tank was used to push fine particles through
a valve and then through a transfer line. The transfer line traveled up the center of the
burner to the burner tip. The photograph in Figure 5b shows the burner tip inside a
protective Plexiglas housing. Cold flow tests were conducted with fine sand particles and
pulverized coal to determine initial sizing of burner transport line and to select proper
pressures to push fine particles.
The sketch in Figure 5c shows a simplified cutaway of the burner concept. In this initial
simulated burner, a jet of pulverized coal is introduced pneumatically through the center
of the burner. An outer annulus provides oxygen for combustion. An inner annulus
between the coal and oxygen provides natural gas to the burner. The natural gas is
included for several important reasons. The burner can be started with gas and oxygen,
and then switched to coal. Second, the natural gas can serve as backup fuel in the event
that coal feed is interrupted for any reason. Finally, the burner can, when desired, be
easily operated as a dual coal-gas burner with any ratio of coal to natural gas. This dual
fuel capability adds flexibility to the MBB technology enabling operators to optimize for
emissions production and fuel cost.
11
(a) (b) (c)
Figure 5. Initial Setup to Test Oxy-Coal Nozzle, (a) Feeder with Nozzle on Test Stand,
(b) Nozzle Tip Seen Through Plexiglas Tube, (c) Burner Concept
RESULTS AND DISCUSSION
The Pressurized Coal Transport/Feed System should be developed for the power plant.
Several approaches for the Transport/Feed System have been analyzed; some of them (1)
use coal-water mixture/slurry to feed the mixture through the oxy burner, (2) feed
pressurized coal through the burner by injecting the coal together with oxygen or
steam/oxygen mixture, and (3) employ specially designed coal feeders. GTI is analyzing
the coal transport/feed systems and will discuss them with the project partners in order to
select the preferred system for engineering design and economic analysis.
The ASPEN Plus model of the MBB process has been completed. A full discussion of the
plant layout, material and energy balances, utilities, and the supercritical steam cycle is
presented in the Design Report sent to DOE. The block flow diagram shown in Figure 6
illustrates all the major streams, stream sizes, and stream compositions.
Coal
Oxygen
Water
Nitrogen/NG
12
Figure 6. Overall Block Diagram with Stream Table
Vapor, Mole%: CO2 -
92.22
-
61.17
61.16
61.18
61.18
61.18
90.91
-
59.71
92.22
-
-
-
-
-
-
H2O Vapor -
0.17
-
32.27
32.28
32.28
32.28
32.28
1.59
-
1.17
0.17
-
-
-
-
-
- N2 -
2.62
1.62
1.77
1.77
1.77
1.77
1.77
2.58
-
14.78
2.62
-
1.62
1.62
-
-
-
Ar -
3.91
3.40
2.63
2.63
2.63
2.63
2.63
3.85
-
18.68
3.91
-
3.40
3.40
-
-
- O2 -
1.07
94.98
1.17
1.16
1.14
1.14
1.14
1.05
-
5.60
1.07
-
94.98
94.98
-
-
-
NOx (as ppmV) -
131.04
-
2
88
88
88
88
129
-
625
131
-
-
-
-
-
- SOx (as ppmV) -
77.17
-
8,872
8,872
8,873
8,873
8,873
76
-
1
77
-
-
-
-
-
-
H2 (as ppmV) -
-
-
-
10.33
0.08
0.08
0.08
0.12
-
-
-
-
-
-
-
-
- CL2 (as ppmV) -
-
-
-
0.02
0.06
0.06
0.06
0.06
-
-
-
-
-
-
-
-
-
HCL (as ppmV) -
-
-
939
939
939
939
939
-
-
-
-
-
-
-
-
-
- CO (as ppmV) -
-
-
-
62.17
0.27
0.27
0.27
0.40
-
-
-
-
-
-
-
-
-
NH3 (as ppmV) -
-
-
0.00
-
-
-
-
-
-
-
-
-
-
-
-
-
- Hg (as ppmV) -
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Total Mole% -
100.00
100.00
100.00
100.00
100.00
100.00
100.00
100.00
-
100.00
100.00
-
100.00
100.00
-
-
- Vapor Total MPH -
488
36,945
48,415
48,416
48,407
48,407
48,407
32,904
-
4,595
488
-
518
37,463
-
-
-
Vapor Total LB/Hr -
21,123
1,189,810
1,707,130
1,707,130
1,707,130
1,707,130
1,707,130
1,411,596
-
183,369
21,123
-
16,668
1,206,480
-
-
- Liquid, Mole%:
CO2 97.73
-
0.60 H2O -
100.00
99.36
Misc Gases 2.27
-
0.04 Total Mole% 100.00
100.00
100.00
Liquid Total MPH 27,131
528
21,597 Liquid Total LB/Hr 1,189,127
9,506
392,579
Solids, Wt%: MAF Coal 79.18
79.18
Ash 9.70
9.70
100.00
100.00
100.00
100.00
100.00
100.00 Carbon -
-
A 4/23/2013 AKL
CaCO3 -
-
-
-
-
-
11.82
-
-
Rev. Date BY CaSO4.2H2O -
-
-
-
-
-
88.18
-
-
H2O (Solid-Bound) 11.12
11.12 Total Wt% 100.00
100.00
100.00
100.00
100.00
100.00
-
-
-
-
-
100.00
-
-
100.00
100.00
-
Solid Total LB/Hr 549,471
549,471
2,664
2,664
2,664
2,664
-
-
-
-
-
83,672
-
-
50,617
2,664 Stream Mass Flow, Lb/Hr 549,471
570,594
1,189,810
1,709,794
1,709,794
1,709,794
1,709,794
1,707,130
1,411,596
1,189,127
183,369
21,123
93,178
16,668
1,206,480
50,617
2,664
392,579
Stream Heats of Formation, MMBtu/Hr (496.23)
(572.77)
56.60
(5,072.15)
(5,750.40)
(6,431.14)
(6,523.15)
(6,522.43)
(5,106.21)
(4,606.27)
(467.70)
(76.54)
(487.36)
0.74
(5.71)
(14.74)
(0.72)
(2,640.62) Stream Temperature, oF 59
59
295
2,970
1,880
650
460
460
134.9
93
142
70
123
281
55
356
460
124
Job Rev.
Stream Pressure, psia 157
157.0
157
147
147
147
142
142
141.3
2,215
20
299
15
70
23
145
142
35
No. No. Stream Vapor Vol Flow, MMSCFD -
4.45
336.50
440.97
440.97
440.89
440.89
440.89
299.69
-
41.86
4.45
-
4.71
341.21
-
-
-
Stream Liquid Vol Flow, GPM -
-
-
-
-
-
-
-
-
3,049
-
-
20
-
-
-
-
807
-
A02067 A
GTI High Pressure Molten Slag Boiler Oxy-Combustion Technology
Issued for Review Revision
NETL Carbon Capture Study With Oxy-Combustion Power Plant
BFD-001
Overall Block Flow Diagram Initial Design with High Pressure WFGD
DRAWING No.
Feed Pressurization & Conveying Coal Receiving &
Sizing Molten Bed Boiler Radiant Boiler Convective Boiler Low Level Heat Recovery
Air Separation Unit
CO2 Purification & Compression
Power Generation
Fines & Particulate Separation WFGD
O 2 Compression Water Treatment Waste Water Treatment Coolong Water
System Effluent Misc Liq Wastes
Supercritical CO2 to
Sequestration
Inert Purge
Gypsum Waste Ash, Carbon & Hg
Activated Carbon Injection
Blowdown Evaporation
Purge BFW Makeup
CW Makeup Well Water Muni Water
Process Makeup
SHSC Steam
PreHt BFW HP BFW
ReHt IP IP Stm
Generator Output 819 MWe
Coal Feed
Ambient Air
Slag
1
3
2 4
CO2 to Feed Conveying
O2 to WFGD Forced Oxidation
5 8 7 6 9 10
11
12
15
14 13 16 17
4 1 3 2 11 10 9 8 7 6 5 15 14 13 12 17 16
N 2 Vent
1
18 Excess Water
18
13
Table 2 summarizes the overall performance of the GTI oxy-combustion SCPC MBB and
compares it with the NETL 2007/1291 5C oxy-combustion case. Plant efficiency is found
to increase from 29.2% to 31.6% for the MBB case. Further optimization is expected to
lead to some further increase in the plant efficiency. The first three cases are assumed to
fire the same quantity of Illinois #6 coal. This allows straight forward comparisons
between the cases. The fourth case is the MBB scaled to same net power of ~550 MWe.
This case allows comparisons to be made on a net power basis. Higher efficiency for the
MBB case leads to higher gross and net power production. Overall auxiliary load is lower
for the MBB case. Operation at 150 psig reduces the compression demand for CO2 by
38.5 MWe which is somewhat more than the added power demand of 35.4 MWe for
oxygen compression. Still, the CO2compression saving is greater than the added energy
cost to compress oxygen. The single higher MBB power plant demand is for makeup
water. This 4% increase results from the current decision to not clean and recycle the
more saline reject water because of high cost. This decision will be revisited in Phase 2 to
find ways to reclaim a portion of this reject water and lower overall make up water
demand.
Plant capital costs for the DOE oxy-coal 5C case, Nexant’s re-calculation of the DOE 5C
case, and the current MBB case are shown in Table 2. Capital costs are determined for all
plant systems. Where DOE base case capital cost data was available and systems were
similar, Nexant used the DOE case capital cost data to keep comparisons on the same
bases. Other capital costs were obtained in consultation with vendors. Burner system
costs are based on GTI experience. The MBB and radiant section along with the pressure
shell were priced using algorithms provided under a non-disclosure agreement by
engineers at Alstom Power.
14
Table 2. Overall Performance Table and Comparison with Case 5C
Notes
MBB cases with a single pressure shell and with individual pressure shells were
considered. The MBB case with individual pressure shells is found to have a capital cost
savings of approximately 13% compared with the baseline oxy-coal case 5C. When
Supercritical Cycle Cases -- 3500 psig/1110 F/1150 F 2007/1291 Nexant 5C Rerun GTI MBB 550 MWe GTI MBB
Normalized to DOE Case 5C Coal Rate SCPC with CO2 SCPC with CO2 SCPC with CO2 SCPC with CO2
Capture/Purification Cap/Pur for Capture/Purification Capture/Purification
for Atmospheric for Atmospheric for Pressurized for Pressurized
Oxycombustion Oxycombustion5
Oxycombustion5
Oxycombustion5
POWER SUMMARY (Gross Power at Generator Terminals, kWe) Case 5C Case 5C-N Case GTI-MBB
550 MWe Net
GTI-MBB
COAL FEED SUMMARY
As-Received Coal Feed, lb/h 549,471 549,471 549,471 512,000
Thermal Input (HHV), MMBtu/hr 6,410 6,410 6,410 5,973
Nominal Combustor Pressure, psig Atmospheric Atmospheric 150 150
TOTAL POWER, kWe2,3
: 785,900 797,998 817,820 761,965
BOILER STEAM GENERATION DUTY (> 650 F), MMBtu/hr
HP Steam Generation 4,616 4,676 4,736 4,413
IP Reheat Duty 1,063 1,082 1,090 1,015
Total Boiler Steam Generation Duty 5,679 5,758 5,826 5,428
HP Steam Flow Rate, lb/hr 4,863,468 4,925,901 4,989,320 4,648,888
BFW PREHEAT VIA WASTE HEAT RECOVERY, MMBtu/hr
Heat Recovery via Flue Gas Cooling (650 F → WFGD Inlet Temp1) Air Preheat Air Preheat 92 86
Heat Recovery via Slag Cooling 0 0 27 25
Total BFW Preheat Duty 0 0 119 111
AUXILIARY LOAD SUMMARY6, kWe:
Coal Handling and Conveying 500 500 500 470
Limestone Handling & Reagent Preparation 1,210 1,210 1,210 1,127
Pulverizers 3,740 3,740 4,408 4,108
Ash Handling 720 720 720 671
Slag Cooling 0 500 466
Primary Air Fans 1,170 1,237 0 0
Forced Draft Fans 1,500 1,644 0 0
Induced Draft Fans 7,850 8,594 0 0
Air Separation Unit Main Air Compressor 125,680 125,680 125,680 117,109
Air Separation Unit Auxiliaries 1,000 1,000 1,000 932
Oxygen Compressor -- 177 35,310 32,073
SCR -- -- -- --
Baghouse 90 90 90 90
FGD Pumps and Agitators 4,050 4,178 4,161 3,877
Econamine FG Plus Auxiliaries -- -- -- --
Econamine Condensate Pump -- -- -- --
CO2 Compression 73,390 72,313 34,009 31,690
Condensate Pumps 1,050 965 999 931
Condensate Booster Pump -- -- -- --
Boiler Feedwater Booster Pumps -- -- -- --
Miscellaneous Balance of Plant 2,000 2,000 2,000 2,000
Steam Turbine Auxiliaries 400 400 400 400
Circulating Water Pumps 6,200 7,116 6,822 6,362
Cooling Tower Fans 3,620 4,155 3,983 3,236
Transformer Losses 3,000 3,046 3,122 2,909
TOTAL AUXILIARIES, kWe 237,170 238,764 224,915 208,450
NET POWER, kWe 548,730 559,234 592,905 553,515
Net Plant Efficiency (HHV) 29.2% 29.8% 31.6% 31.6%
Net Plant Heat Rate (Btu/kWh) 11,682 11,462 10,811 10,791
CONDENSER COOLING DUTY (MMBtu/hr) 2,890 2,939 3,078 2,868
ESTIMATED TOTAL COOLING DUTY (MMBtu/hr)4
3,282 3,767 3,611 3,368
CONSUMABLES
Makeup Water, gpm 6,096 8,000 8,001 7,461
1 GTI Pressurized MBB Flue Gas is cooled to 460 F, above sulfuric acid dew point
2 DOE Oxycombustion Case 5C seems to be using ~2% coal HHV heat loss in boiler, compared with 1% as prescribed in QGESS Process Modeling Parameters report
3 GTI Pressurized MBB Case assumes 1% coal HHV heat loss in boiler, as prescribed in QGESS Process Modeling Parameters report
4 Total Cooling Duty for Case 5C back-calculated based on parameters established in QGESS Process Modeling Parameters report
5 Case 5C-N and GTI-MBB performance run by Nexant models
6 Estimated performance based on Nexant design of 5C-N case and GTI-MBB Case
15
corrected for the higher electricity production, the MBB case is found to have an overall
power plant capital cost savings of 17%.
Table 3. Capital Cost Comparison for Oxy-Coal Supercritical Steam Power Plants with
Carbon Capture and Compression: DOE 5C, Nexant Calculation of DOE 5C, MBB Case
Acct DOE Nexant Nexant Nexant
No. Item/Description Case 5C Case 5C-N GTI-MBB Ind GTI-MBB
1 COAL & SORBENT HANDLING
SUBTOTAL 1. $54,243 $54,243 $54,243 $54,243
2 COAL & SORBENT PREP & FEED
SUBTOTAL 2. $26,097 $26,097 $41,268 $41,268
3 FEEDWATER & MISC BOP SYSTEMS
SUBTOTAL 3. $107,585 $110,692 $111,238 $112,210
4 PC BOILER & ACCESSORIES
SUBTOTAL 4. $723,675 $792,922 $779,715 $656,261
5 FLUE GAS CLEANUP
SUBTOTAL 5. $154,325 $124,791 $88,194 $88,194
5B CO2 REMOVAL & COMPRESSION
SUBTOTAL 5B. $167,959 $174,103 $79,762 $79,762
6 COMBUSTION TURBINE/ACCESSORIES
SUBTOTAL 6. $0 $0 $0 $0
7 HRSG, DUCTING & STACK
SUBTOTAL 7. $39,153 $39,286 $35,704 $35,704
8 STEAM TURBINE GENERATOR
SUBTOTAL 8. $178,750 $180,583 $183,506 $183,506
9 COOLING WATER SYSTEM
SUBTOTAL 9. $45,467 $49,986 $48,625 $48,625
10 ASH/SPENT SORBENT HANDLING SYS
SUBTOTAL 10. $17,633 $17,137 $15,441 $15,441
11 ACCESSORY ELECTRIC PLANT
SUBTOTAL 11. $120,124 $120,537 $118,254 $118,249
12 INSTRUMENTATION & CONTROL
SUBTOTAL 12. $32,579 $32,607 $32,355 $32,355
13 IMPROVEMENTS TO SITE
SUBTOTAL 13. $18,164 $18,139 $17,858 $17,569
14 BUILDINGS & STRUCTURES
SUBTOTAL 14. $71,108 $71,448 $70,886 $70,311
TOTAL PLANT COST $1,756,862 $1,812,573 $1,677,049 $1,553,698
Owner's Costs
Preproduction Costs
6 months All Labor $11,313 $11,546 $10,977 $10,462
1 Month Maintenance Materials $1,734 $1,789 $1,655 $1,534
1 1 Month Non-Fuel Consumables $1,638 $1,778 $1,748 $1,755
1 Month Waste Disposal $490 $490 $513 $513
25% of 1 Months Fuel Cost at 100% CF $3,437 $3,437 $3,437 $3,437
2% of TPC $35,137 $36,251 $33,534 $31,074
Total $53,749 $55,291 $51,864 $48,774
Preproduction Costs
60 day supply of fuel and consumables at 100% CF $30,014 $30,289 $30,276 $30,290
0.5% of TPC (spare parts) $8,784 $9,063 $8,383 $7,768
Total $38,798 $39,352 $38,659 $38,058
Initial Cost for Catalyst and Chemicals $0 $0 $0 $0
Land $900 $900 $900 $900
Other Owner's Cost $263,529 $271,886 $251,502 $233,055
Financing Costs $47,435 $48,939 $45,270 $41,950
Total Overnight Costs (TOC) $2,161,273 $2,228,941 $2,064,878 $1,916,435
TOTAL PLANT COST, 2011 $1000
16
Construction of a pressurized MBB is a large capital expense. Figure 7 shows the effect
of heat flux rate on the capital cost of the boiler plus radiant section in $/kW. Assuming a
heat flux of only 25-50% of the value estimated by GTI for the MBB leads to a capital
cost in the same range as the case 5C which operates at atmospheric pressure.
The boiler pressure shell is a large expense. Over most of the projected possible heat flux
range, the pressure shell accounts for more than half of the overall boiler and radiant
section cost. Even with this high added cost, the overall capital cost in $/kW for the MBB
with pressure shell is similar to the atmospheric pressure case 5C when heat flux is at
least 50% of the heat flux estimated by GTI. The project team is considering options for
lowering the cost of the boiler pressure housing. Figure 8 shows the rough dimensions of
a single housing enclosing MBB modules producing 550 MWe net electricity. This large
shell must be thick and is subsequently expensive. Consideration is being given to
building smaller pressure housings around modular MBBs adding up to 550 MWe of net
power production. This could save significant capital cost and provide means to service
boilers as need arises since boilers could be shut down on an individual basis.
Figure 7. Capital cost in $/kW as Effected by Heat Flux and added Cost for the Boiler
Pressure Shell. Data is also included for the DOE Oxy-Coal Case 5C for Comparison
MBB Cost (2011 $) vs Average Heat Flux (Btu/hr-SF)Average Heat Flux = Total MBB Absorbed Duty/Total MBB Tube Surface Area
0
200
400
600
800
1000
1200
0 10,000 20,000 30,000 40,000 50,000 60,000 70,000 80,000
Average MBB Heat Flux (Btu/hr-SF)
Bo
iler
Co
st, $
/kW
Nexant In-House USC PC Boiler(2009,815MW Gross)
DOE Case 12 SC PC Boiler (831 MW Gross)
DOE 5C Oxy Boiler
GTI MBB (819MW Gross, 30000 Btu/hr-ft2Radiant Heat Flux, excl Ext Shell)
GTI MBB (819 MW Gross, 30000 Btu/hr-ft2Radiant Heat Flux, incl Ext Shell)
@ 50% GTI MBB
Heat Flux
@ 25% GTI MBB
Heat Flux
@ 12.5% GTI
MBB Heat Flux
@ 100% GTI MBB
Heat Flux
97' D x 280' T/T Shell
97' D x 200' T/T Shell
97' D x 160' T/T Shell
97' D x 140' T/T Shell
17
Figure 8. Calculated Dimensions of a Single Pressure Shell Enclosing Modular Molten
Bed Oxy-Coal Boilers Producing a Total Net Power of 550 MWe
The overall power plant cost summary leading to determination of the COE is shown in
Table 4. The MBB case in the current work is found to have a lower total plant cost and
lower total overnight cost. COE in $/MWh is 17% lower for the MBB case than for the
DOE 5C atmospheric pressure oxy-coal supercritical steam power plant with CO2 capture
and compression. The COE for the MBB case with CO2 capture and compression is
found to be 134% of the COE for the baseline air-coal supercritical steam power plant
with no CO2 capture. Table 3 oxy-coal cases are calculated with the same coal feed rate
which leads to a higher net power production from the more efficient MBB cases (single
and individual pressure shells). When the MBB cases are scaled to the 550 MWe plant
size, the COE increases from 106.2 to 108.2 mils/kWh, or from 131% to 134% of the
base case COE. This is a significant improvement over the approximately 150% COE for
the DOE oxy-coal 5C case compared with the air-coal baseline case. The 34% increase in
COE for the MBB-based power plant, even with no optimization, is very close to the
DOE target of a COE increase of only 35%. Optimization analyses in Phase 2 will
improve efficiency and have an excellent chance of meeting the challenging target of a
COE increase of only 35%.
97.0'
21
.7'
MBB Section
24
.0'
Radiant Section
54
'
12
.1'
12
.1'
Convection/Economizer Section
18
.0'
12
.5'
14
0'
18
Table 4. Cost Summary for Oxy-Coal Supercritical Steam Power Plants with Carbon
Capture and Compression: DOE 5C Case, NETL Calculation of DOE 5C Case, Proposed
Molten Bed Boiler Case
Prof. Dale Tree of BYU conducted an initial thermodynamic analysis of the MBB along
with the radiant and convective sections above the boiler. The work included both first
and second law energy and exergy analyses. The results indicate locations in the process
where available heat is present and where there is the potential to produce work. Energy
and exergy analyses are useful in guiding engineers seeking to optimize processes and
identify ways in which to improve the overall efficiency of processes.
Exergy is the potential of a given system (mass) to produce work. Discussions of exergy
can be found in most textbooks of Thermodynamics such Moran et al. (2011)1 or Cengel
and Boles (2008).2 The exergy of a system is found by determining the work that can be
done by a system as it transitions from its initial state to a state where it can no longer
produce work. For example, a 1 kg mass of copper moving at 10 m/s has kinetic energy
which can be converted to work. In the case of kinetic energy (and all forms of
mechanical energy), all of the energy in the system can potentially be converted to useful
work. Therefore, the maximum amount of work possible is found by comparing the
kinetic energy of the copper at 10 m/s with the kinetic energy of the copper at rest or at
zero velocity. Determining the exergy of the thermal or chemical energy of system is
more complex.
The exergy for a steady-flow system with negligible kinetic and potential energy is given
by Equation 1, where h is specific enthalpy, T is temperature, and s is the specific entropy
(Moran et al., 2011). 1 The subscript, 0, represents the dead state which for thermal
systems is defined by the lowest temperature and pressure reservoir available for the
system. A reservoir is a mass that will not change temperature or pressure when energy is
added. In this work the dead state will be assumed to be at a temperature of 25°C
(77.5°F) and pressure of 101.325 kPa (14.7 psia) represented ambient conditions.
∑ ( )
∑ ( )
(1)
Case 11 Case DOE 5C Case 5C-N Case GTI Case Ind-GTI
Supercritical PC
w/o CO2 Capture
(3,500 psig, 1,100 ◦F, 1,100 ◦F)
550.0 548.7 559.2 592.9 592.9
409,528 549,471 549,471 549,471 549,471
Net Efficiency 39.3% 29.2% 29.8% 31.6% 31.6%
TOC $MM 1348 2161 2229 2065 1916
OCfix $MM/yr 38.8 57.8 59.3 55.5 52.0
OCv ar w/o fuel $MM/yr 37.3 46.3 48.7 47.1 45.6
Fuel $MM/yr 123.0 165.0 165.0 165.0 165.0
COE mills/kWh 80.95 123.7 124.2 111.4 106.1
100% 153% 153% 138% 131%
Net Power MWe
Coal Feed Rate, LB/hr
% Case 11 COE
GTI MBB Cost of Electricity Summary, June 2011 Cost Base
Case
Description Supercritical Oxycombustion Based
with CO2 Capture and Purification
(3,500 psig, 1,110 ◦F, 1150 ◦F)
19
The exergy of a fuel initially at the dead state is zero unless the fuel is allowed to react
and produce products. The exergy of a fuel can be determined by finding the difference
between the exergy of fuel plus oxidizer and the exergy of complete products of
combustion at the dead state as shown in Equation 2. For example, the exergy of methane
without an oxidizer at T = 25°C, P=101 kPa, is zero because the methane can produce no
work unless the chemical energy is released. The chemical energy can be released by
reacting methane with oxygen. The exergy of methane and a stoichiometric mixture of air
at the dead state can be found by determining the exergy of methane plus air and
subtracting the exergy of complete products of combustion (CO2, H2O and N2). In all
cases reported, the exergy of a stream containing fuel has been calculated by determining
the exergy of that stream when oxidized by a stoichiometric mixture of air. Typically, the
exergy of fuels are similar in magnitude to their heating values. After combustion
however, the exergy is significantly decreased.
∑ ( )
∑ ( )
∑ ( )
(2)
As can be seen in Equations 1 and 2, in order to calculate the exergy, the specific entropy
of each component of fuel and products must be determined. All of the reactants and
product for coal combustion can be considered and ideal gas with the exception of the
coal. The ideal gas properties were calculated using a commercial thermodynamics code,
Engineering Equation Solver (EES).
The entropy of coal was obtained based on a correlation developed by Ikumi et al.
(1982)3 and is summarized below. The coal entropy is the sum of the entropies of the
organic solid, organic sulfur, pyritic sulfur, ash and mixing entropies as shown in
Equation 3. The organic solid entropy is found by the correlation in Equation 4 which
requires the molar ratios of hydrogen (NH), oxygen (NO) and nitrogen (NN) to carbon
(NC). The sulfur, ash, and moisture entropies were calculated as the product of the
number of moles and the specific entropy of each constituent. Values for specific entropy
were obtained from Ikumi et a. (1982)3as shown in Equations 5 and 6. In order to
determine the number of moles of ash, a molar mass of 76 was assumed. Finally the
entropy of mixing is a function of the moles and mole fractions of each component in the
mixture as shown in Equation 7. Using this approach the specific entropy of the Illinois
#6 coal was determined to be 0.3626 Btu/lbm-R.
(3)
(
) ( )
(4)
( ) (5)
20
( ) (6)
∑ ( )
(7)
Using the mass flow rates and gas species concentrations produced by NEXANT for
NETL Case 5c, the exergy of each flow for the fuel/air and steam into and out of the
boiler has been calculated and are reported in Table . Flow streams for fuel and air are
identified by number which correlates to the block flow diagram of Figure . The
enthalpies shown are total enthalpies calculated by the EES program for the mixtures at
the temperature and pressure shown in Table 2. Overall Performance Table and
Comparison with Case 5C and are slightly different but very close to the values
calculated using ASPEN.
The decrease in exergy and enthalpy for the gas phase are shown in Table 4 to be 6,120.4
and 5,988.6 MMBtu/hr respectively. The increase in steam exergy and enthalpy were
found to be 3,290.6 and 5,931.0 MMBtu/hr. The final row of Table reports the net
change in exergy and enthalpy for the combined gas and steam flows into and out of the
boiler. For enthalpy, there is a net loss of 57.6 MMBtu which is a small fraction (1%) of
the total energy exchanged between the combustion gas and the steam. This loss can be
attributed to heat transfer to the surrounding that does not enter the steam.
Table 5. Energy and Exergy Flows for Air and Steam Into and Out of the Boiler
Number Description Flow Exergy
(MMBtu/hr)
Enthalpy
(MMBtu/hr)
1 Coal feed 6,343.3 -501
2 Coal feed + carrier gas 6,559.7 -553
3 Total oxidizer flow 0.158 56.6
4 Molten bed products 1475.5 -5,043
5 Radiant boiler exit 935.9 -5,717
6 Convection boiler exit 483.6 -6,393
7 Low heat recovery exit 439.4 -6,485
7-(1+2) Decrease in gas phase 6,120.4 5,988.6
100 High pressure feed water entering
boiler
691.6 2,661
101 High pressure steam exiting the
boiler
3,337.7 7,492
102 Low pressure reheat steam
entering boiler
1,958.9 5,412
103 Low pressure steam exiting boiler 2,603.4 6,509
(100 + 102) –
(101 + 103)
Increase in steam 3290.6 5931.0
Balance Heat lost or Exergy Destroyed 2829.85
(30.7%)
57.6 (1%)
21
A large decrease in exergy of the gas phase is seen to occur between streams 2 and 4
which includes both the combustion process and heat transfer to the high pressure steam
tubes. Exergy destruction occurs primarily from two physical processes in the boiler 1)
Chemical reactions which convert chemical energy to heat and 2) Heat transfer from a
high temperature gas to the molten bed and then to the steam. Exergy destruction can be
reduced by 1) Increasing the temperature of combustion products and 2) Decreasing the
temperature difference between the combustion products and the steam.
Optimization of the MBB to reduce exergy destruction will require a more detailed model
of the boiler itself. The current model of the MBB is shown in Figure 9a. The current
model assumes complete combustion of the coal including solid carbon and volatiles
within the melt. After releasing all of the energy from the combustion process into the
melt, heat is transferred from the melt to the high pressure steam. The products of
combustion (primarily CO2 and H2O) leaving the melt are assumed to be at the molten
bed temperature of 1632°C (2970°F) and heat is transferred from these combustion
products to both the high and intermediate pressure steam. The simplified process is not
likely to be the most efficient or the process that generates the least exergy destruction.
Future modeling of the MBB illustrated in Figure 9b will consider the combustion
process distributed throughout the molten bed and will allow staged combustion with an
overfire region (Figure 10). The future model will include information from a three
dimensional finite difference model of the molten bed (FLUENT) and allow for modeling
of heat transfer in the bed to steam tubes. Combustion within the bed will be modeled
including the fraction of volatile release and the fraction of solid and gaseous fuel
oxidized within the bed. The flow rates of primary, secondary and overfire oxidizer will
be determined. This staging of heat release will provide more flexibility for the sizing of
boiler components and may be used to reduce exergy destruction.
22
(a) Current model of molten bed combustor (b) Future model of molten bed combustor
Figure 9. Block Diagram of Molten Bed Combustor Model Used for Exergy Analysis
Radiant Boiler
Coal – RFG, 2
O2, 3
4
6
Low Level Heat Recovery
Complete Products CO2 + H2O, T= 2970 F
Convective Boiler
5
High Pressure Feed Water
100
101
High Pressure Steam To Boiler
Reheat, IP, From Turbine 102
103
Reheat, IP, To Turbine
Radiant Boiler
Coal – RFG, 2
O2, 3
4
6
Low Level Heat Recovery
Fuel Rich Products and Volatiles
Convective Boiler
5
.
... .. ..
.
....
...
..
.
... .. ..
.
....
...
..
.
... .. ..
.
....
...
..
.
... .. ..
.
....
...
...... .. ..
.
....
...
...
... .. ..
.
....
...
...
... .. ..
.
....
...
..
.
... .. ..
.
....
...
..
.
... .. ..
.
....
...
...
... .. ..
.
....
...
..
.
... .. ..
.
....
...
.. .
... .. ..
.
....
...
..
.
... .. ..
.
....
...
..
.
... .. ..
.
....
...
..
.
... .. ..
.
....
...
..
.
... .. ..
.
....
...
..
Overfire Oxidizer
Complete Products of Combustion
Overfire Oxidizer
23
Figure 10. Energy and Exergy Analysis Block Flow Diagram of the Molten Bed Oxy-
Coal Boiler with Radiant and Convective Zones
24
Cold flow testing was conducted with several different particles, particle sizes, and
nozzle designs. The goal was to find a nozzle size and shape that would deliver the
desired rate of pulverized coal without flow interruption. Initial work was carried out
with two different particle sizes of silica sand. After selection of a workable nozzle
design, testing was conducted with a PRB coal.
The first tests were conducted with -40+60 mesh glass beads with the unmodified blaster
operating at different pressures. Figure 11 shows that the particle rate through the nozzle
was well above desired levels and increased with increasing pressure.
Time, minutes
Figure 11. Flow of Material Using Blaster Factory Settings
To reduce the feed rate, several modifications were made to the blaster. The rubber hose
was replaced with a 0.5 in. OD stainless steel tube. The ceramic nozzle was replaced
with a restriction by capping the tube with a 0.039 In. ID stainless steel cap. Testing
results with these changes are shown in Figure 12. Particle flow rate was significantly
reduced as a result of these changes.
0
5
10
15
20
25
30
35
40
45
0 1 2 3 4 5 6 7
25 psig
20 psig
15 psig
50 psig
25
Time, minutes
Figure 12. Data From Blaster Modified With Stainless Steel Tranfer Line and End Cap
Interruptions in flow occurred with the end cap restriction. The end of the nozzle was
replaced with a long tapering nozzle with 0.059 in. ID opening at the tip. The intent was
to eliminate opportunities for solids build-up in the nozzle. Tests with -40+60 mesh glass
at 40 psig found that flow rates were again higher, and much higher than desired. Also,
the orientation of the nozzle had no impact on the flow rate of glass beads.
Time, minutes
Figure 13. Blaster With Stainless Steel Tranfer Line and Long Tapered Nozzle
0
1
2
3
4
5
6
0 1 2 3 4 5 6 7
60 psig
50 psig
40 psig
0
2
4
6
8
10
12
14
16
18
20
22
0 1 2 3 4 5 6 7 8 9 10 11
40 psig down #1
40 psig down #2
40 psig up
26
Figure 14 shows a cross-sectional view of the a short tapered nozzle that replaced the
long tapered nozzle and installed within the transport line of the simulated oxy-coal
burner. The initial design imagined a transport line with a fairly large diameter and a
nozzle that decreases in size at the tip right before coal is transported into the flame.
Figure 15 shows the flow of material (40-60 mesh glass beads) through the first
configuration using a 0.5 in. rubber hose transfer line and 0.136 in. diameter ceramic
nozzle. This was the factory configuration for the blasting unit. Particle flow rates were
higher than desired.
Figure 14. Short Nozzle Installed in Submerged Combustion Burner
Time, minutes
Figure 15. Data From Blaster Modified With Stainless Steel Tranfer Line and Short
Tapered Nozzle
0
4
8
12
16
20
24
28
32
36
40
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
40 psig short nozzle
60 psig short nozzle
27
The nozzle with a decreasing cross section at the tip tended to plug every few minutes.
The plugs were easy to clear, but were both unpredictable and unreliable for steady
operation. Engineers redesigned the transport line as a smaller diameter tube with a
constant diameter all the way from the feeder to the burner tip. Figure 16 shows a cross-
sectional view of the small diameter tube installed within the simulated oxy-coal burner.
Figure 17 shows the flow of the larger material (40-60 mesh glass beads) through the
small diameter tube at three different pressures. The flow rate at the lower pressure was
46 lb/hr. The flow rate at the middle pressure was 60 lb/hr. The flow rate at the highest
pressure was 74 lb/hr. Figure 15 shows the flow of the smaller material (60-120 mesh
glass beads) through the small diameter tube at two different pressures. The flow rate at
the lower pressure was 87 lb/hr. The flow rate at the higher pressure was 111 lb/hr. The
smaller particles had a higher flow with the same back pressure. Pulverized coal at under
200 mesh is expected to have a much higher flow through the same size transport line.
Figure 16. Small Diameter Tube Installed in Submerged Combustion Burner
Time, minutes
Figure 17. Flow of the Larger (40-60 mesh) Material through the Small Diameter Tube
0
2
4
6
8
10
12
14
0 2 4 6 8 10 12 14 16 18 20 22 24
60 psig 1/8" tubing
80 psig 1/8" tubing
100 psig 1/8" tubing
28
A series of tests were conducted next with the 40-60 mesh glass beads, a separate purge
line, no nozzle, and 0.125 in. stainless steel tubing of different thicknesses. Figure 18
shows the that adding a throttling valuve in the transfer line had only a minor effect on
particle flow to the nozzle. Results at differect pressures are shown in Figure 19. Tests
found the flow rates of particles increased with increasing pressure, but plugging
occurred at higher pressures.
Time, minutes
Figure 18. Effect of a Throttling Valve on Particle Flow Rate
Time, minutes
Figure 19. Effect of Thicker-Walled Tubing With Smaller Cross Section on Throttleed
Particle Flow Rate
After completion of the cold flow tests with various nozzle configurations, engineers set
up the simulated coal feeder and simulated burner with oxygen and natural gas on a test
stand. The simulated burner, shown in Figure 20 on the test stand, is water cooled and has
0
1
2
3
4
5
6
7
8
0 1 2 3 4 5 6
40 psig 0.020 wall fully open
40 psig 0.020 wall throttled
40 psig 0.020 wall more throttled
0
1
2
3
4
5
0 1 2 3 4 5 6 7
40 psig 0.028 wall
60 psig 0.028 wall
80 psig 0.028 wall
100 psig 0.028 wall
29
inlet natural gas and oxygen lines as well as the coal feed line. The coal feeder was again
placed on a scale to allow determination of coal feed rates. The natural gas and oxygen
flows were controlled by mass flow controllers.
The first tests were conducted with 40-60 mest glass beads and 60-120 mesh glass beads.
Results are shown in Figures 21 and 22. Figure 21 shows tests with 40-60 mesh glass
beads. A separate purge line was used. The stainless steel particle transfer line was a
0.25 in. thick wall tube. The outside end of the transfer tube was tapered to fit in the
burner sparger. An annular flow was established around the outside of the sparger tube.
Particle feed rates were similar to those achieved with the single nozzle. This confirmed
that the oxy-coal burner design is practical.
Figure 20. Setup for Hot Testing of the Simulated Oxy-Coal Burner for the Molten Bed
Boiler
30
Time, minutes
Figure 21. 40-60 Mesh Glass Beads Charged Through Oxy-Coal Burner
The same cold flow tests through the burner were conducted with finer glass beads sized
to 60-120 mesh. Results in Figure 22 show similar results to the earlier nozzle tests and
to the burner tests with 40-60 mesh glass beads. The flow rate, as expected, was higher
than the rate with the larger 40-6- mesh beads. Charging interruptions could be
addressed by agitating the material in the feed vessel.
Time, minutes
Figure 22. 60-120 Mesh Glass Beads Charged Through Oxy-Coal Burner
Burner tests were conducted by igniting the simulated burner with natural gas and oxygen
and establishing a stable flame. Coal was then introduced to the flame with natural gas
still flowing. The dual fuel flame was operated for some time, and then the natural gas
0
2
4
6
8
10
12
14
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
40 psig 0.083 wall
60 psig 0.083 wall
80 psig 0.083 wall
0
2
4
6
8
10
0 1 2 3 4 5 6 7
40 psig; with "O2" flow
60 psig; with "O2" flow
31
was shut off to create the oxy-coal flame. Coal feeding with pulverized coal was difficult,
but methods were developed to keep coal flowing into the transport line.
Hot test photographs are shown in Figure 23. The photograph on the left shows an
oxygen-natural gas flame with high excess oxygen and a natural gas rate of 0.1
MMBtu/hr. The photograph on the right shows the flame after introduction of the
pulverized coal. This flame is much larger because the coal rate was well above 300 lb/hr
(3.6 MMBtu/hr) in this test. This coal rate is well above the design rate for the simulated
oxy-coal burner. Further work was focused on reliably feeding coal to the flame at a
much lower rate.
Figure 23. Simulated Oxy-Coal Burner Operating With Oxygen-Natural Gas and
Oxygen-Coal
Combustion tests were conducted with 70% -200 mesh pulverized coal in which the feed
system was modified to include a 0.125 in. jet for coal elutriation, a 0.375 in. OD
stainless steel transfer tube with no nozzle, and a burner sparger installed in the center of
the burner. Natural gas flow was introduced in an annulus around the sparger and left on
at low fire during coal feed to maintain stable combustion. Oxygen flow was introduced
in an annulus around the outside of the natural gas annulus. Combustion was established,
but the coal flow rate was far too high. The fine size of pulverized coal leads to higher
than desired feed rates.
The next efforts focused on using particles of the same size as pulverized coal to improve
the ability to feed the desired rates (50 to 100 lb/h) of pulverized coal. The closest
32
similants were determined to be all purpose flour and talcum powder. Cold flow tests
through the burner were conducted with these materials. They had the same particle size
range as coal, but their behavior in the feeder was very different. The flour was fed for a
period of trime, but the transfer tube became plugged. No amount of feeding could be
sustained because channeling occurred in the solids feed tank. This issue could be
resolved if an agitator was added to the feed tank. The talcum powder tests had the same
result. Channeling occurred in the feed tank, and very low particle feed rates were
obtained. Results of these cold flow tested confired that the feed tank must be agitated so
uniform particle flow rates can be obtained.
GTI engineers and technicians have successfully demonstrated that the simulated version
of the oxy-coal burner can be operated under controlled conditions. Later tests have been
conducted to vary the oxygen to coal ratio and the coal feed rate. Future work beyond this
project will confirm the ability to fire oxygen and coal directly into a bed of molten slag
from below. This will require better control of coal feed rates of 50-100 lb/h. The project
team believes these rates are possible is proper design of an agitated feed tank and
appropriated sized transfer line and diltion gas. The oxy-coal burner itself functioned
properly and does not require any further modification at this time. During future testing
beyond this project, engineers from REI will collect gas samples, analyze those samples,
and then conduct corrosion analyses. All information learned from the molten bed
furnace trials will be available to the project team in planning for integrated boiler
development testing in later stages of molten bed boiler technology development.
CONCLUSIONS AND RECOMMENDATIONS
This project has focused on the initial work to determine the benefits of the pressurized
molten bed oxy-coal combustion technology. The overall objective is to develop a
pressurized oxy-coal supercritical steam power plant with CO2 capture and compression
that has a COE that is no greater than 135% of the COE for an air-coal supercritical
steam power plant with no carbon capture. Results of the Phase 1 project have found that
without optimization the MBB technology can nearly meet this ambitious objective with
a COE of 134% of the COE for the baseline air-coal plant. The work in this project has
been focused in several key areas as the most critical first steps. Work included:
Thermo-chemical design and economic analysis
Thermodynamic analysis
Oxy-coal burner testing
Corrosion analysis
The engineering design of the MBB power plant with individual pressure shells found
that overall capital cost is approximately 17% lower than the capital cost for the DOE
baseline case 5C, an atmospheric oxy-coal plant with CO2 capture and compression.
MBB power plant capital costs, utility costs, fuel cost, auxiliary demands, and other
inputs were used to determine the overall plant COE. The pressurized oxy-coal MBB
power plant with CO2 compression was found to have a COE of 134% of the COE of the
DOE baseline atmospheric air-coal case with no CO2 capture. This is a very promising
33
result because the DOE target is a COE increase of 135%, a value very close to our
estimated COE increase even before optimization. The COE increase of 134% compares
very favorably with an increase of 164% of COE calculated for a first generation oxy-
coal power plant with CO2 capture and compression.
A thermodynamic enthalpy and exergy analysis was conducted around the MBB, radiant
zone, and convective zone. A number of modifications and adjustments were found that
could provide higher efficiency and better use of available work. Conclusions from this
analysis will help guide the analyses and CFD modeling planned for future development.
Oxy-coal combustion testing work began with the design of systems to simulate an oxy-
coal burner. Cold flow and hot tests were conducted. Results demonstrated the oxy-coal
flame can be sustained and controlled. The simulated burner was fired with pulverized
coal and oxygen.
The MBB has the potential to be a disruptive technology that will enable coal combustion
power plants to be operated in a cost effective way, cleanly with no CO2 emissions. Work
is needed to quantify and confirm the great promise of the MBB technology. Results
provide the core information needed to design, build, and test the first integrated, pilot-
scale MBB with supercritical steam production. The next step will involve MBB
technology scale-up, feeder design and testing, operation at pressure, collection of
operating data and exhaust gas samples, and planning for demonstration scale testing.
The project team believes that this multi-step approach is the most efficient approach to
develop the MBB oxy-coal combustion technology and the best way to maximize the
expenditure of increasingly limited research funds.
REFERENCES
1. Moran, M.J., Shapiro, H.N., Boettner, D.D., and Bailey, M.B. (2011) Fundamentals
of Engineering Thermodynamics, Seventh Ed., Wiley, 2011.
2. Cengel, Y. A. and Boles, M.A., Thermodynamics An Engineering Approach, Seventh
Ed., McGraw Hill, 2008.
3. Ikumi, S., Luo, C.D., and Wen, C.Y., “A Method of Estimating the Entropies of Coal
and Coal Liquids” The Canadian Journal of Chemical Engineering, Vol. 60, pp 551-
555, 1982.
34
DISCLAIMER STATEMENT
This report was prepared by David Rue, Gas Technology Institute, with support, in part,
by grants made possible by the Illinois Department of Commerce and Economic
Opportunity through the Office of Coal Development and the Illinois Clean Coal
Institute. Neither David Rue, Gas Technology Institute, nor any of its subcontractors, nor
the Illinois Department of Commerce and Economic Opportunity, Office of Coal
Development, the Illinois Clean Coal Institute, nor any person acting on behalf of either:
(A) Makes any warranty of representation, express or implied, with respect to the
accuracy, completeness, or usefulness of the information contained in this report,
or that the use of any information, apparatus, method, or process disclosed in this
report may not infringe privately-owned rights; or
(B) Assumes any liabilities with respect to the use of, or for damages resulting from
the use of, any information, apparatus, method or process disclosed in this report.
Reference herein to any specific commercial product, process, or service by trade name,
trademark, manufacturer, or otherwise, does not necessarily constitute or imply its
endorsement, recommendation, or favoring; nor do the views and opinions of authors
expressed herein necessarily state or reflect those of the Illinois Department of
Commerce and Economic Opportunity, Office of Coal Development, or the Illinois Clean
Coal Institute.
Notice to Journalists and Publishers: If you borrow information from any part of this
report, you must include a statement about the state of Illinois' support of the project.