mcilvaine “hot topic hourhot topic hour” june 30, 2011 · mcilvaine “hot topic hourhot topic...
TRANSCRIPT
McIlvaine “Hot Topic Hour”McIlvaine Hot Topic Hour June 30, 2011
Fuel Impacts on Design andPerformance of SCR Catalysts
Presenter:John Cochran
Performance of SCR Catalysts
CERAM Environmental, Inc.+1 913 239 9896
Catalyst Deactivates With Time of Exposure to Flue Gasp
Ko (Original Catalyst Activity) ct
ivity
)
Coal/Heavy Pet Coke Blend
Coal Low Dust
Coal High Dust
(Rel
ativ
e A
c Coal High Dust
Biomass
Waste High Dust
K/K
o
K/Ko @ 16 khr
0 2000 4000 6000 8000 10000 12000 14000 16000
Operation [hrs]
Catalyst Design (at hour=0) Must Anticipate Deactivation to End of Guarantee Period (e.g., 16,000 hours) to Size Catalyst Properly
Catalyst Deactivation is Very Site SpecificVery Site Specific
K/Ko vs Time
0 9
1.0
0.7
0.8
0.9
vity
0.4
0.5
0.6
Rel
ativ
e A
ctiv
0.1
0.2
0.3R
0.00 2000 4000 6000 8000 10000 12000 14000 16000 18000 20000
Hours of SCR Operation
Varying Based on Fuel Quality and OperationsVarying Based on Fuel Quality and Operations
Catalyst Deactivation Mechanisms Gaseous Poisons: Gaseous Poisons:
Arsenic Phosphorus
P t i Act
ivity
)
Coal/Heavy Pet Coke Blend
Coal Low Dust
Coal High Dust Potassium Sodium Cadmium
L d
K/K
o (R
elat
ive
A
Biomass
Waste High Dust
Lead Copper Other Elements
0 2000 4000 6000 8000 10000 12000 14000 16000
Operation [hrs]
Fouling by Solid Compounds Gypsum (Calcium Sulfate) and Other Solid Compound Deposition Ammonium Bisulfate (Avoided by Keeping Above Permissive
T t )Temperatures) Occurs During...
Normal SCR OperationS d Sh d i G h h A id i Startups and Shutdowns as Unit Goes Through Acid Dew Point
Catalyst Deactivation Vanadium Titania Based Catalyst Deactivates Based on Site Specifics Vanadium-Titania Based Catalyst Deactivates Based on Site Specifics
Fuel Quality – Arsenic, Phosphorus, Potassium, Sodium, Sulfur, Calcium, etc.
Comb stion Q alit Increased S bstoichiometric Staging Increases Combustion Quality – Increased Substoichiometric Staging Increases Quantity of Gaseous Poisons
# of Startups and ShutdownsV di Tit i B d C t l t D ti t I d d t f Vanadium-Titania Based Catalyst Deactivates Independent of... Catalyst Type – Plate, Honeycomb, Corrugated Fiber Formulation – Different Activities and SO2:3 Conversion Rates Reference Also “Comparison of Deactivation Rates of Different
Catalyst Types” by Ed Healy, Southern Company and Hans Hartenstein, Evonik (now Steag) Presented February 9, 2009
h // i h ld i l / bli /47b 6d6 7 8f479388 20d66579738f8/H %20H i %20 ihttp://www.reinholdenvironmental.com/public/47bc6d6a7e8f479388a20d66579738f8/Hans%20Hartenstein%20presentation%20Deactivation%202009.pdf
Deactivation Resistance Comes From Providing Adequate Reactor Potential (RP=K/Av) – There Are No Magic Potions( ) g
Case Example 1: Two x 500 MW Unit PRB Deactivation Rate History
Being Considered
0.9
1.02 x 500 MW Units - Plate and Honeycomb Catalyst Deactivation Trends
0.7
0.8
ctiv
ity K
/Ko
0 4
0.5
0.6
Rel
ativ
e A
c
0.3
0.4
0 5,000 10,000 15,000 20,000 25,000 30,000 35,000 40,000 45,000 50,000Time of Exposure to Flue Gas, hours
Original Supplier’s (Plate by Others) Estimate of Deactivation RateU d B i f C t l t D i
K/Ko Design Basis of Original Plate Supplier Expected Deactivation Trend
Used as Basis for Catalyst Design
Case Example 1: Two x 500 MW Unit PRB Deactivation Rate History
Being Considered
0.9
1.02 x 500 MW Units - Plate and Honeycomb Catalyst Deactivation Trends
0.7
0.8
ctiv
ity K
/Ko
0 4
0.5
0.6
Rela
tive
Ac
0.3
0.4
0 5,000 10,000 15,000 20,000 25,000 30,000 35,000 40,000 45,000 50,000Time of Exposure to Flue Gas, hours
Deactivation Rate Assumed by Original Plate Catalyst Supplier Proven to be Overly Optimistic
K/Ko Design Basis of Original Plate Supplier Expected Deactivation Trend
Plate Catalyst Test Results (Original Supplier)
Proven to be Overly Optimistic
Case Example 1: Two x 500 MW Unit PRB Deactivation Rate History
Being Considered
0.9
1.02 x 500 MW Units - Plate and Honeycomb Catalyst Deactivation Trends
0.7
0.8
ctiv
ity K
/Ko
0 4
0.5
0.6
Rel
ativ
e A
c
0.3
0.4
0 5,000 10,000 15,000 20,000 25,000 30,000 35,000 40,000 45,000 50,000Time of Exposure to Flue Gas, hours
Deactivation Rate Assumed by Original Plate Catalyst Supplier Proven to be Overly Optimistic
K/Ko Design Basis of Original Plate Supplier Expected Deactivation Trend
Plate Catalyst Test Results (Original Supplier) Honeycomb Catalyst Test Results (2 Suppliers)
Proven to be Overly Optimistic Subsequent Replacements With Honeycomb Catalyst From 2
Different Suppliers Confirms Both Types Deactivate Based on The Same Trend
Case Example 2: 600 MW Unit Burning Eastern Bituminous High Arsenic/Low Calcium Coal
0.9
1
Relative Activity Vs Time600 MW Unit With High Arsenic and Low CaO
Fuels Burned Different Than
0 6
0.7
0.8
0.9
K/K
o)
Originally Specified Resulting in Extreme Deactivation
0 3
0.4
0.5
0.6
elat
ive
Act
ivity
(K
0
0.1
0.2
0.3
Re
Honeycomb Catalyst Test ResultsCorrugated Fiber Catalyst Test ResultsProjected DeactivationExpected Activity for Specified Fuel (16khr)Current Deactivation Trend (16khr)
00 5,000 10,000 15,000 20,000
Exposure to Flue Gas (hrs)
Original Deactivation Rate Underestimated Based on Change in Fuel S ifi iSpecification
Catalyst Test Results For Honeycomb and Corrugated Fiber Catalyst Confirms Both Types Deactivate Based on The Same Trend
Why is Estimating CatalystDeactivation So Important
If Deactivation is Underestimated Catalyst is Undersized I bl f M ti NO R l d A i Sli P f Incapable of Meeting NOx Removal and Ammonia Slip Performance
at Some Point During the Guarantee Period Deficient Performance is Either Tolerated or an Early Outage
(Unscheduled) is Required for Catalyst Addition(Unscheduled) is Required for Catalyst Addition Catalyst Management Costs are Underestimated
Understanding and Managing Reactor Potential Critical to Minimize RiskPotential Critical to Minimize Risk
Examples Help to Illustrate Risk
Reactor Potential
Can be represented by
P = K/AV
K = catalyst activity, Nm3/m2h or Nm/hAV = catalyst area velocity, Nm/h (normalized operating gas flow, Nm3/h divided by total installed catalyst surface area, m2)
DeNOx Demand Reactor Potential DeNOx Demand = The reactor potential required
to meet NOx removal and ammonia slip requirements at the specified operating conditionsrequirements at the specified operating conditions
Calculated based on NOX removal requirements, NH3 slip, SCR distributions, and boiler operating conditions (flow temperature pressure etc )conditions (flow, temperature, pressure, etc.)
Independent of catalyst design life (i.e. same value for 16,000 or 24,000 hour catalyst life)I d d t f C t l t T F l ti Independent of Catalyst Type, Formulation, or Manufacturer
DeNOx Demand, Reactor Potential, & Catalyst Design
1. Determine DeNOx Demand (Preq)
1. DeNOx demand (Preq) is the amount of reactor potential necessary to achieve NOx removal and ammonia slip performance based on fixed operating conditions (flows temp etc )and ammonia slip performance based on fixed operating conditions (flows, temp, etc.)
DeNOx Demand, Reactor Potential, & Catalyst Design
2 E ti t C t l t D ti ti (K /K )
1. Determine DeNOx Demand (Preq)
2. Estimate Catalyst Deactivation (Ke/Ko) Varies as Function of Design Life
1. DeNOx demand (Preq) is the amount of reactor potential necessary to achieve NOx removal and ammonia slip performance based on fixed operating conditions (flows temp etc )and ammonia slip performance based on fixed operating conditions (flows, temp, etc.)
2. Catalyst deactivation is estimated based on fuel quality, combustion parameters, and design life
DeNOx Demand, Reactor Potential, & Catalyst Design
3. Determine Initial Reactor Potential
2 E ti t C t l t D ti ti (K /K )
3. Determine Initial Reactor Potential Required (Po)
1. Determine DeNOx Demand (Preq)
2. Estimate Catalyst Deactivation (Ke/Ko) Varies as Function of Design Life
1. DeNOx demand (Preq) is the amount of reactor potential necessary to achieve NOx removal and ammonia slip performance based on fixed operating conditions (flows temp etc )and ammonia slip performance based on fixed operating conditions (flows, temp, etc.)
2. Catalyst deactivation is estimated based on fuel quality, combustion parameters, and design life3. Based on DeNOx demand and deactivation the initial reactor potential (Po) is determined
DeNOx Demand, Reactor Potential, & Catalyst Design
3. Determine Initial Reactor Potential
2 E ti t C t l t D ti ti (K /K )
3. Determine Initial Reactor Potential Required (Po) 4. Determine Catalyst Volume
fn(Ko, geometry, SO2:3, gas, etc)
1. Determine DeNOx Demand (Preq)
2. Estimate Catalyst Deactivation (Ke/Ko) Varies as Function of Design Life
1. DeNOx demand (Preq) is the amount of reactor potential necessary to achieve NOx removal and ammonia slip performance based on fixed operating conditions (flows temp etc )and ammonia slip performance based on fixed operating conditions (flows, temp, etc.)
2. Catalyst deactivation is estimated based on fuel quality, combustion parameters, and design life3. Based on DeNOx demand and deactivation the initial reactor potential (Po) is determined4. Catalyst volume is determined based on Po, catalyst activity, geometry, SO2 to SO3 conversion y , y y, g y, 2 3
rate, and various gas conditions and constituents
Comparison of Catalyst Design Cases
1.6
)Proposal 1: 'Conservative' Design Case (0.65 K/Ko @16 khr)
1.4
or Poten
tial (K/AV Required Reactor Potential
Guaranteed Life
1.2
Relativ
e Re
acto
1.0
Proposal 1 Has the Most Conservative K/Ko Basis (0.65 @ 16,000 hr)
0.8
0 4000 8000 12000 16000 20000
Hours of Operation
p ( @ )
Comparison of Catalyst Design Cases
1.6
)Proposal 1: 'Conservative' Design Case (0.65 K/Ko @16 khr)
Proposal 2: 'Aggressive' Design Case (0.72 K/Ko @16 khr)Variance in Reactor Potential Consists of Volume, Surface Area and/or K Differences
1.4
or Poten
tial (K/AV
Required Reactor Potential
Guaranteed Life
Area and/or K Differences
1.2
Relativ
e Re
acto
1.0
Proposal 2 is Based on a More Aggressive Deactivation Rate (0.72 K/Ko)
0.8
0 4000 8000 12000 16000 20000
Hours of Operation
p gg ( ) Approximately 10% Difference in Catalyst Volume Who is Right?
Case A: Proposal 1 Deactivation Rate Correct
1.6
)Proposal 1: 'Conservative' Design Case (0.65 K/Ko @16 khr)
Proposal 2: 'Aggressive' Design Case (0.72 K/Ko @16 khr)
1.4
or Poten
tial (K/AV 0.72 K/Ko Based Proposal 2 Operating at 0.65 K/Ko @16 khr
Required Reactor Potential
Guaranteed Life
1.2
Relativ
e Re
acto
1.0
Reduced Catalyst Life Since Proposal 2 Was Undersized
Should a K/Ko of 0.65 Actually Occur Proposal 2 Would be Undersized
0.8
0 4000 8000 12000 16000 20000
Hours of Operation
y pand Meet Performance for Less Than 11,000 Hours
Early Outage Required or Reduced Performance Must be Accepted
Case B: Proposal 2 Deactivation Rate Correct
1.6
)Proposal 1: 'Conservative' Design Case (0.65 K/Ko @16 khr)
Proposal 2: 'Aggressive' Design Case (0.72 K/Ko @16 khr)
1.4
or Poten
tial (K/AV 0.65 K/Ko Based Proposal 1 Operating at 0.72 K/Ko @16 khr
Required Reactor Potential
Guaranteed Life
1.2
Relativ
e Re
acto
Increased Catalyst Life
1.0
Should a K/Ko of 0.72 Actually Occur Proposal 1 Would be Oversized
0.8
0 4000 8000 12000 16000 20000 24000 28000
Hours of Operation
y pand Meet Performance for More Than 24,000 Hours (>3 Years)
Catalyst Management Costs Greatly Reduced
Unstaged PRB Fired Units Catalyst Deactivation
Group 1: PRB Unit Catalyst Activity Test Results
1.0
1.2
0.6
0.8
tive
Act
ivity
K/K
o
0.2
0.4Rel
at
0.00 5,000 10,000 15,000 20,000 25,000 30,000 35,000 40,000 45,000 50,000
Exposure to Flue Gas, hours
Plant 7 PC Plant 8 PC
Plant 9 Cyclone Plant 10 PC
“Unstaged” PRB Units Indicate a K/Ko @ 16,000 hours of 0.6 to 0.8
Plant 9 Cyclone Plant 10 PC
Plant 11 PC Plant 12 Cyclone
Plant 13 Cyclone Plant 14 Cyclone
General Deactivation Trend - Unstaged PRB Units
g @ Wide Variation of Results Dependent on Many Operations and Fuel
Variables
Catalyst Deactivation for "Deeply Staged" PRB Units
Group 2: PRB Unit Catalyst Activity Test Results
1.0
1.2
o
Catalyst Deactivation for "Deeply Staged" PRB Units
0.6
0.8
ativ
e A
ctiv
ity K
/K
0.2
0.4Rel
a
Plant 1 PC Plant 2 PCPlant 3 Cyclone Plant 4 CyclonePlant 5 Cyclone Plant 6 PCOther Deeply Staged General Trend Plant 6 PC General Trend
0.00 5,000 10,000 15,000 20,000 25,000
Exposure to Flue Gas, hours
p y g
Combustion Conditions Greatly Affect PRB Application Deactivation Rates Combustion Conditions Greatly Affect PRB Application Deactivation Rates Broad Consensus of Results for “Deeply Staged” Units Confirm Severe Deactivation Highly Risky if Plant 6 Alone Was Selected as a Reference Unit to Support Proposal Sizing A Plant 6 Based Catalyst Design Will Last Less Than One Year on a 24,000 Hour Guarantee With y g ,
Deep Deactivation Seen for Broader Experience
Summary
Deactivation Rates Vary Widely Dependent on Site Specifics Fuel Quality, Combustion Parameters, and Boiler Duty Cycle Greatly
Affect Catalyst Deactivation Ratesy Vanadium-Titania Based Catalysts All Deactivate at the Same Rate Based on
Site Specifics Underestimating or Aggressive Sizing Compromises SCR Performance and Underestimating or Aggressive Sizing Compromises SCR Performance and
Effective Catalyst Management Risk of Early Outage for Catalyst Additions Risk of Deficient Performance
Initial SCR Project Design Should Carefully Consider Reactor Potential to Determine the Risk Profile of Various Proposalsp Aggressive Catalyst Designs Can Result in Operations Difficulties and
Increased Cost