Research on CO2 Capture, Hydrogen Generation and Biomass Feeding

1
Research on CO2 Capture, Hydrogen Generation and
Biomass Feeding

• John Grace
• Chemical and Biological Engineering

UBC CERC, Feb. 5, 2008
2
The Context

• Energy and Environment are likely to be the
  Defining Issues of this Century. They are closely
  interlinked.
• Canada has so far not engaged in a meaningful way
  with Climate Change. We have totally failed to
  make either the policy changes needed, to make
  the public understand the nature of the
  sacrifices needed, or to truly understand the
  implications for technology development.
• Leadership is needed from many. CERC could play
  a major role on the technology side.

3
This Talk

• Unifying theme Climate Change
• First Part Looping Cycles for CO2 Capture.
• Second Part Hydrogen production aided by in situ
  CO2 capture and/or hydrogen removal.
• Third Part Biomass feeding.

4
Capture vs Sequestration

• This talk will focus on Capture of CO2 i.e. how
  do we separate ? 80 of the CO2 into a stream
  which has a CO2 purity of ? 95.
• Ultimate Sequestration step, is clearly also
  important (whether by Enhanced Oil Recovery or
  Geological storage, etc.), but the Capture step
  typically requires 70 of the total cost,
  depending on source of CO2, distance between
  source and injection site, nature of reservoir,
  etc.

5
Options for CO2 Capture

• Pre-Combustion e.g. during Gasification or Steam
  reforming
• Coupled with Combustion e.g. Carbonation during
  Combustion, Oxy-fuel combustion
• Post-Combustion e.g. Absorption via MEA or other
  liquid sorbents, Adsorption on solids, Membranes

6
ZECA Cycle for CO2 Separation with Combustion
CO2, (H2O)
N2, H2O, (O2)
Combustor with CO2 Capture
Calciner/ Regenerator
CaCO3
CaO CO2 ? CaCO3
CaCO3 ? CaO CO2
CaO
(Endothermic)
(Exothermic)
(Benefits by high P)
(Benefits by low P)
H2O
Fuel Air
7
Solid Sorbents

• This is an area of considerable research
  activity there are a number of proprietary
  sorbents under development and testing.
• All high-temperature sorbents must be compared
  with limestone (CaCO3) given its low cost and
  wide availability.
• Lithium-based and zirconium-based sorbents are
  receiving particular attention due to lower
  temperatures for calcination and improved
  reversibility on repeated cycling, but their cost
  is much higher than for limestone.

8
Mass flow controller
Schematic of custom-built atmospheric pressure
thermogravimetric reactor (ATGR) system at UBC
9
Product layer diffusion control
(0-order)
Surface reaction control (1st order)
Kinetic study CaOCO2 ? CaCO3 38-45 ?m
Strassburg limestone
10
Reversibility for Strassburg limestone
Calcination in 100 N2 Carbonation at 850?C in
100 CO2. Fast stage of carbonation finished for
each cycle.
11
(a) 5 ?m
(b) 5 ?m
SEM photos showing surface texture for samples
derived from 212-250 µm Strassburg limestone
(a) Carbonate after 1020 cycles with
carbonation time of 3.5 min and calcination time
of 4 min followed by 24 h of carbonation.
(b) Same sample as in (a) after calcination.
12
Evolution of Pore size distribution over various
numbers of calcination/carbonation cycles.
13
Simultaneous Carbonation and Sulphation
Experimental TGR results for 212-250 ?m
limestone, calcination at 850?C sorption at
850?C with 2900 ppm SO2 and 80 CO2.
14
Total calcium utilization to CaCO3 and CaSO4
Calcium utilization to CaCO3
Calcium utilization to CaSO4
Calcium utilization during co-capture of CO2 and
SO2 (212-250 ?m Strassburg limestone)
15
CFBC
Favoured process for CO2 and SO2 removal with
calcium-based sorbents Sorbent calcination and
carbonation cycling completed prior to sulphation
in the CFBC reactor.
16
Cyclic CaO Utilization Efficiency at different
temperatures
CaO Utilization
750?C
800?C
825?C
850?C
Number of Cycles

• CaCO3 from previous carbonation stage divided by
  total calcium present, as a
• function of number of calcination/carbonation
  cycles
• Thames limestone Carbonation 9 min in pure CO2
  Calcination 8 min in pure N2.

17
Long-term Cyclic CaO Utilization Efficiency
CaO Utilization
Dolomite
Limestone
Number of Cycles

• Calcium present from previous carbonation stage
  divided by total calcium present as a fraction of
  number of cycles
• Thames limestone / Arctic dolomite / 850 ?
• Carbonation 9 min in pure CO2 / Calcination 8
  min in pure N2

18
Some Conclusions from this part of the work

• SO2 impedes CO2 capture, even when SO2
  concentration ltlt CO2 concentration.
• Sequential capture of SO2 and CO2 is possible. It
  is best to capture CO2 before SO2.
• Co-capture of H2S and CO2 is feasible.
• 1000-cycle tests indicate that capture efficiency
  (with no SO2) levels off at 4 to 14.
• Dolomite does better than limestone initially,
  but not after many cycles.

19
Hydrogen

• Major industrial commodity used in making
  ammonia, upgrading of hydrocarbons, food products
  and pharmaceuticals.
• Fuel cells based on hydrogen promise to be more
  efficient and less polluting.
• Most H2 is now produced by steam reforming of
  methane in fixed beds in huge furnaces.
• Electrolytic production is only viable for very
  small scale production.
• Better methods of making hydrogen are needed.

20
Relevant Chemical Reactions for H2

• Reforming and Water-Gas Shift Reactions
• CH4 H2O ? CO 3H2 ?H298 206 kJ/mol
• CO H2O ? CO2 H2 ?H298 - 41
  kJ/mol
• CH4 2H2O ? CO2 4H2 ?H298 165 kJ/mol
• Oxidation Reactions
• CH4 0.5O2? CO 2H2 ?H298 - 36 kJ/mol
• CH4 2O2 ? CO2 2H2O ?H298 -802 kJ/mol
• Carbonation/Calcination Reactions
• CaO CO2 ? CaCO3 ?H298 -168 kJ/mol

21
Shifting the SMR Equilibrium

• If either of the product gases, Hydrogen or CO2
  can be removed, the equilibrium shifts forward to
  produce more H2, more CO2 and less CO.
• Hydrogen is the smallest molecule and can be
  removed through membrane filter, e.g. Pd.
• CO2 can be removed by a suitable side reaction,
  such as carbonation of CaO, the reverse of
  calcination CaO CO2 ? CaCO3.

22
Effect of in situ Hydrogen Removal
No H2 Removal
Equilibrium methane conversion as a function of
temperature at different in-situ hydrogen removal
rates. Steam-to-methane molar ratio 3.
Pressure 1000 kPa. No oxygen present.
23
Pure hydrogen production rates for different
membrane thicknesses, different permeation areas
per unit volume of reactor, and different flow
regimes. (All results are for the fast
fluidization flow regime, except where
indicated.)
Membrane thickness and total surface area are
very important in determining the H2 permeation
rate
1 µm
5 µm
15 µm
25 µm
50 µm
24
Fluidized Bed Membrane Reactor
Operating conditions Pressure 10-30
bars Temperature 520-600?C Molar H2O/CH4 feed
ppp ratio 2.5-3.5
25
Advantages of FBMR Reactor

• Equilibrium shift increases conversion.
• Adverse effect of pressure nearly neutralized.
• Hydrogen of high purity (e.g. 99.99).
• Process intensification 3 vessels in 1.
• Lower temperatures of operation.
• Small catalyst particles ? high effectiveness
  factors.
• Improved rates of heat and mass transfer.
• Reduced pressure drops.
• Reduced coking of catalyst.
• On-line replacement of catalyst.
• Eliminates NOx emissions (from furnace).

26
Disadvantages of FBMR Reactor

• Product hydrogen is at a low pressure it must be
  compressed for most applications.
• Catalyst undergoes attrition and entrainment.
• Membranes are expensive and must withstand the
  physical and chemical environment.
• Large internal (membrane) surface area required
  in limited reactor volume.

27
The Opportunity Fuel Cell Infrastructure

• Very small scale Electrolytic H2 generation
• Very large scale Conventional SMR
• Intermediate scale Opportunity for FBMR,
  especially in urban areas where there is already
  an infrastructure for natural gas and water, so
  that H2 can be made on site at fuelling stations.
• Future opportunity Alternative feedstocks such
  as propane, or even biomass (after gasification).

28
Initial Challenges for FBMR Process
Commercialization

• Heat Input Issues
• Catalyst Issues
• Membrane Issues
• Configuration and Mechanical Design
• (These all interact with each other.)

29
H2 flux at 560?C through Pd-Ag Membrane
Double-Sided Pd-Ag Membrane
30
Communicating chamber geometry in current
Fluidized Bed Membrane Reactor developed by UBC
and MRT
31
Relevant Reactions Sorbent Enhanced Reforming

• Reforming and Water-Gas Shift Reactions
• CH4 H2O ? CO 3H2 ?H298 206 kJ/mol
• CO H2O ? CO2 H2 ?H298 - 41
  kJ/mol
• CH4 2H2O ? CO2 4H2 ?H298 165 kJ/mol
• Oxidation Reactions
• CH4 0.5O2? CO 2H2 ?H298 - 36 kJ/mol
• CH4 2O2 ? CO2 2H2O ?H298 -802 kJ/mol
• Carbonation/Calcination Reactions
• CaO CO2 ? CaCO3 ?H298 -168 kJ/mol

32
Schematic of Sorbent-Enhanced Membrane-Assisted
Steam Methane Reforming
33
Advantages of Sorbent-Enhanced Reforming

• Further positive shift in the thermodynamic
  equilibrium by removing a product (here CO2).
• Heat input Carbonation is exothermic, with
  (coincidentally) magnitude of heat of reaction
  very similar to that of the endothermic reforming
• Concentration of CO2, a key step in greenhouse
  gas sequestration, accounting typically for 70
  of the cost of sequestering CO2.

34
Additional Benefit of Removing CO2 beyond
Hydrogen Removal
90 CO2 Removed
No CO2 Removal
35
Most Important Gasification Chemical Reactions

• Gasification and Water-Gas Shift Reactions
• C H2O ? CO H2 ?H298 173
  kJ/mol (steam-carbon)
• CO H2O ? CO2 H2 ?H298 - 41
  kJ/mol (water-gas shift)
• CO2 C ? 2CO ?H298 214 kJ/mol
• (Boudouard)
• Carbonation/Calcination Reactions
• CaO CO2 ? CaCO3 ?H298 -168
  kJ/mol

36
Relevant SMR Chemical Reactions

• Reforming and Water-Gas Shift Reactions
• CH4 H2O ? CO 3H2 ?H298 206 kJ/mol
• CO H2O ? CO2 H2 ?H298 - 41
  kJ/mol
• CH4 2H2O ? CO2 4H2 ?H298 165 kJ/mol
• Carbonation/Calcination Reactions
• CaO CO2 ? CaCO3 ?H298 -168 kJ/mol

37
Advantages of Capturing CO2 in Reforming and
Gasification rather than with Combustion

• Equilibrium shift removing product CO2 shifts
  forward the reforming gasification reactions,
  improving the yield of the desired products and
  decreasing the yield of CO.
• Carbonation is exothermic, with a heat of
  reaction of almost equal magnitude to the
  endothermic heat of reaction needed for the
  reforming and gasification reactions.
• Cyclic performance of the sorbents is better for
  reducing than for oxidizing conditions.
• Calcium is known to reduce tar formation.

38
Batchwise Tests in a fluidized bed 1st Cycle,
limestone and reforming catalyst H2O/CH4 3.
39
Cycle 2
Cycle 3
Cycle 4
Batchwise Tests in a fluidized bed Cycles 2-4,
limestone and reforming catalyst H2O/CH4 3.
40
Long-term cyclic calcination/carbonation test
results
212-250 ?m Strassburg limestone
Carbonation 850oC, 100 CO2 time 9
min Calcination 850oC, 100 N2 time 8 min
41
Calcination at 850?C in 100 N2 Carbonation at
850?C in 100 CO2
9 min carbonation time
4.5 min carbonation time
42
Cyclic CaO Utilization Efficiency at different
temperatures
CaO Utilization
750?C
800?C
825?C
850?C
Number of Cycles

• CaCO3 from previous carbonation stage divided by
  total calcium present, as a
• function of number of calcination/carbonation
  cycles
• Thames limestone Carbonation 9 min in pure CO2
  Calcination 8 min in pure N2.

43
Long-term Cyclic CaO Utilization Efficiency
Thames limestone Arctic dolomite Particle
size 180µm 250µm Carbonation 9 min in 100
CO2 Calcination 8 min in 100 N2
CaO Utilization
Dolomite
Limestone
Number of Cycles

• Calcium present from previous carbonation stage
  divided by total calcium present as a fraction of
  number of cycles at 850?

44
Major Challenges in FBMR and SE-SMR

• Thinner, pinhole-free, robust membrane surfaces,
  e.g. 3-5 ?m compared with current 15-25 ?m
  foil, would immediately lead to
• Less membrane surface area needed
• Less reactor congestion and more compact
  reactor
• Simpler design and assembly
• Lower capital and operating (replacement)
  costs.
• Compatibility between the Ca-sorbent and both the
  catalyst and the membranes.
• Demonstration of continuous sorbent-enhanced
  reforming over many calcination/carbonation
  cycles.
• Pressure swing vs temperature swing operation.

45
Biomass as a Fuel for Combustion

• Traditional fuel, providing 5 of world total
  energy.
• Excluded from Greenhouse Gas accounting.
• Wood wastes (e.g. hogfuel, bark, sawdust, fibres,
  black liquor) Agricultural crops or wastes
  (straw, rice husks, corn stalks, bagasse),
  Refuse-derived fuels, Other.
• Commercial BFB and CFB boilers for such fuels are
  now quite common, separately fed or co-fed with
  fossil fuels.

46
Barriers to Biomass Utilization

• Availability Biomass is in limited supply.
• Collection Collection costs are high.
• Feeding Particles are large, wet, compressible,
  pliable, heterogeneous difficult to feed.
• Processing Biomass particles are not ideal
  materials for fluidization or even for fixed
  beds.
• Losses Low density causes entrainment, affecting
  reactor design and heat transfer.
• Environmental Issues emissions of polyaromatic
  hydrocarbons, dioxins/furans, K, Na, NOx, etc.,
  mostly due to contaminants.

47
Some Challenges in Biomass Utilization

• Which Biomass? Wood, Agricultural, Wastes
• How to Transport to Central Location?
• Which Pre-Treatment? Dry, Mill, Pelletize
• Feeding Method?
• What Type of Reactor? Fixed/moving, Bubbling
  Fluidized, Circulating Fluidized Bed, Entrained
• Operating Temperature and Pressure?
• Air, Oxygen or Steam Gasification?
• Non-Catalytic or Catalytic?
• Product? Power, Heat, Steam, Syngas, H2, CH3OH

48
Harvest
Clean-up
Application
Transport
Reactor
Feed
Store
Dry
Densify
Screen
Mill
Wood Biomass Utilization Steps
49
Hog fuel
50
UBC Pilot-scale hog fuel combustor
51
Circulating Fluidized Bed Gasification Facility
at UBC Designed to gasify sawdust, hogfuel and
other waste materials. Can also be used for CFB
combustion.
52
Biomass Feeding Research

• Feeding is commonly the most difficult part of
  biomass operations A process is only as good as
  the ability to feed reactants in a sustained,
  uniform and reliable manner.
• Major problem involves bridging and slippage.
• Two-part study
• (a) Influence of moisture content, particle size
  and shape, etc. on an existing screw feeder with
  12 types of biomass particles Studied torque and
  onset of bridging.
• (b) Flow loop to study basic mechanisms of
  bridging and how to prevent it.

53
Biomass Properties and Feeding

• Biomass particles are unusual materials low
  density, large, wide size distribution, irregular
  shapes, compressible, high moisture content.
• Feeder choices are closely related to reactor
  types, pressure and biomass properties.
• Feeder is commonly the most problematic component
  of the entire system.

1
54
Objectives of Feeder Study

• To define what limits screw feeding, in
  particular the mechanisms of blockage and
  slippage.
• To clarify the range of operation for different
  types of biomass and the influence of material
  properties such as particle size, particle shape,
  and moisture content.
• To improve the design and operation of biomass
  feeding systems.

2
55
Biomass Feeding Experimental Setup
12 biomass materials tested up to 60 moisture
and up to 9.9 mm mean diameter.

• Schematic of Biomass Feeding System

3
56
Wood pellets, sawdust and hog fuel compositions
 
57

• Particle Size Effect

Wood Pellets
8.0 11.6 mm
3.4 4.8 mm
2.0 3.4 mm
8
58
Blockage Mechanism

• Blockage may occur due to mechanical or cohesive
  blockage.
• Mechanical blockage is related to particle size,
  size distribution, particle surface friction,
  particle shape, density, strength and
  compressibility.
• Cohesive blockage is related to van der Waals
  forces, moisture content and electrical charges.

13
59
Some Conclusions from Feeding Study

• Fines promote blockage.
• Even a small pressure difference between hopper
  and receiving vessel can significantly reduce
  blockage.
• Sawdusts and hog fuel with higher moisture
  contents (gt30) are more likely to bridge in the
  hopper.
• Hopper level affects blockage.
• Tapered casing improves plug seal but increases
  the tendency to block.
• Larger particles, irregular shapes, larger
  density and rougher surfaces make particles more
  likely to block

16
60
Acknowledgements

• UBC Colleagues J. Lim, P. Watkinson, X. Bi, D.
  Posarac
• Former graduate students, I. Abba, A. Adris, T.
  Boyd, J. Dai, K. Johnsen, X. Li, S. Roy, P. Sun
• Current graduate students A. Mahecha-Botero, M.
  Rakib, A. Vigneault
• Others B. Anthony, Z. Chen, M. Dogan, S.
  Elnashaie, K. Laursen, A. Li, Y. Li, B. Pruden,
  H. Ryu, J. Song
• Companies Membrane Reactor Technologies, Alstom
  Power, Noram Engineering, Tokyo Gas
• Funding agencies, NSERC, NRC, NRCan, CFI, Canada
  Research Chairs, Fuel Cells Canada, BC Science
  Council