fine pm emission factors for gas turbine engines 2014/pmemissions... · fine pm emission factors...
TRANSCRIPT
ICEPAG 2014
April 1-3
Newport Beach, California
Fine PM Emission Factors for
Gas Turbine Engines
Glenn C. England
Principal Consultant
ENVIRON International Corporation
Irvine, California 92629
Overview
• Particulate matter (PM) & regulations
• PM Emissions from gas-fired engines
• Emission factors
• Measurement influences & impacts
April 1-3, 2014 ICEPAG, Newport Beach, CA 2
Why Are Fine PM Emissions from Gas
Turbines Relevant?
• Increasing natural gas use for power generation
• Typical 500 MW NGCC plant “emits” ~30-50 tpy
All submicron aerosols
• New Source Review
10 tons/year PM2.5 Prevention of Significant Deterioration
threshold
• triggers modeling, technology, offset requirements = time, $
Includes filterable and condensable PM
• Impediment to licensing new power plants in
California due to lack of PM10/2.5 offsets
• Evolving interest in human health effects of
“ultrafine” PM and nanoparticle emissions
April 1-3, 2014 ICEPAG, Newport Beach, CA 3
Hot Filter
Iced Impingers
Cyclones
Impactors Particle counting & size
classification (e.g., SMPS)
“Filterable PM” “Condensable PM”
Traditional stationary source methods
Dilution Methods Traditional mobile source measurement methods & research methods for stationary sources
Particulate Matter
>10 µm 10-2.5 µm
“Coarse PM”
2.5 µm 0.1 µm
(100 nm) 50 nm
PM10 10 µm PM2.5
“Fine PM” “Ultrafine PM” “Nanoparticles”
TSP
mass
number 10,000 nm 2,500 nm 100 nm 50 nm
1960’s 1980’s 1990’s 2000’s 2010’s
April 1-3, 2014 ICEPAG, Newport Beach, CA 4
PM10 and PM2.5
• Definitions - 40 CFR 51, §51.50:
Filterable PM2.5/PM10 : Particles that are directly emitted
by a source as a solid or liquid at stack or release
conditions and captured on the filter of a stack test train.
Condensable PM: Material that is vapor phase at stack
conditions, but which condenses and/or reacts upon
cooling and dilution in the ambient air to form solid or
liquid PM immediately after discharge from the stack.
• Particulate matter is defined by the test methods
used to measure it!
April 1-3, 2014 ICEPAG, Newport Beach, CA 5
Natural Gas-Fired Combined Cycle
• Gas turbines & steam
turbines
Brayton+Rankine
“2 on 1” or “3 on 1” blocks
~500-800 MW/block
Duct burners
Thermal efficiencies ~55->60%
April 1-3, 2014 ICEPAG, Newport Beach, CA
• Emissions controls
Lean pre-mix combustors
SCR
• 2-5 ppm NOX limits
CO oxidation catalysts
G
Combustion
Turbine-
Generator
Heat Recovery
Steam Generator
Natural gas
TG
Steam Turbine-
Generator
C T
O2, CO
2
H2O
N2
Duct Burners
CO Catalyst
SCR
6
Air
IGCC with CO2 Separation
Scrubber Sour
Shift
Gasification Block
Gasifier
G
Power Block
O2 Plant
Combustion
Turbine-
Generator
Heat Recovery
Steam
Generator
Coal, Petcoke,
etc.
Air
N2
O2
Low-carbon fuel gas
CO2
Sequestration
Sulfur
Recovery
Slag
T G
Steam Turbine-
Generator
C T
Duct Burners
CO Catalyst
SCR
O2,
H2O
N2
Mercury
Adsorption
Mercury
Recovery
April 1-3, 2014 ICEPAG, Newport Beach, CA 7
Sulfur
Absorption
CO2
Absorption
PM2.5 Sources
• Incomplete fuel combustion
(carbon conversion to soot,
organics)
• Combustion and post-
combustion conversion of
sulfur (& chlorine) to acid
aerosols & salts
• Vaporization and
condensation of volatile
elements
• Transmission of non-volatile
elements
• Corrosion/erosion of system
components
• Aerosol nucleation,
condensation, accumulation
and breakup
Gas Turbine
(compressor,
combustor, turbine)
Air solids
Air sulfur
Fuel carbon
Fuel sulfur
Fuel solids
SCR reagent
solids
Nitrogen solids
Steam solids
Evaporative
cooler solids
HRSG
(duct burners,
catalysts, boiler)
PM2.5
April 1-3, 2014 ICEPAG, Newport Beach, CA 8
PM2.5, lb/hr
Rela
tive fre
quency
Effect of CO2 (and S) Separation
PM2.5, lb/hr
Rela
tive fre
quency
Total PM2.5
Solids (including chloride)
Sulfate
OC
EC
Ammonium
Nitrate
NGCC - Natural Gas IGCC - Low Carbon Fuel
April 1-3, 2014 ICEPAG, Newport Beach, CA 9
PM Measurements - traditional
• Hot filter/impingers
Collect filterable & condensable separately (?)
Dry & weigh filters, evaporate & dry probe/cyclone rinses
Organic extraction, evaporate & weigh impinger residues
April 1-3, 2014 ICEPAG, Newport Beach, CA 10
EPA Methods
201A/202 (2010)
EPA Method 5
State & local
variations
High bias due to measurement artifacts can be significant
(dissolved gases e.g. O2, SO
2, NH
3, VOC, HCl → solid residues)
Water Bath
Empty “dry”
impingers
Silica Gel
Condenser
(≤30 ºC)
Heated or
in-stack
filter CPM Filter
(20-30 °C)
≤20°C Outlet
Ice
Bath
In-Stack Cyclones
10 &/or 2.5 µm
Gravimetric analysis
tolerance 1 mg each
fraction (5 mg total)
AP-42 PM Emission Factors - GTs
• Ch. 3.1, gas turbines, natural gas (April 2000)
EPA Methods 5 and WDNR 5
5 tests, 4 units 1994-1996
• Total PM 0.0065 lb/MMBtu
cPM is 71% of total
Uncertainty >100%
April 1-3, 2014 ICEPAG, Newport Beach, CA 11
0
2
4
6
8
10
1 2 3 4 5
mg
/d
scm
cPM
fPM
Measured Gas Turbine PM Emissions
April 1-3, 2014 ICEPAG, Newport Beach, CA 12
0.00
0.01
0.02
0.03
0.04
0.05SC
SC
SC
SC
SC
SC
SC
SC
SC
CC
, no D
B
CC
, no D
B
CC
, no D
B, N
G/RG
CC
, no D
B, N
G/RG
CC
, D
B off, 53%
CC
, D
B off, 75%
CC
, D
B off, 100%
SC
, LG
66 MW 64 MW 45 MW 24 MW 13.5 MW 13.3 MW 5.7
MW
lb
/M
MBtu
Impinger-total
Impinger-organic
Impinger-inorganic
Probe rinse
Filter
CC=combined cycle
SC=simple cycle
DB=duct burners
NG=natural gas
LG=landfill gas
RG=refinery fuel gas
Impinger catch dominates measured emissions
Measured Gas Turbine PM Emissions
April 1-3, 2014 ICEPAG, Newport Beach, CA 13
0
0.0002
0.0004
0.0006
0.0008
0.001
0.0012
0.0014
0.0016
0.0018
CC,
no
DB
CC,
DB
on
CC,
DB
off
CC,
DB
off
CC,
DB
on
CC,
DB
off
CC,
DB
on
CC,
DB
off
CC,
DB
on
CC,
DB
off
CC,
DB
on
405
MW
181 MW 176 MW
lb
/M
MBtu
Impinger-total
Impinger-organic
Impinger-inorganic
Probe rinse
Filter
CC=combined cycle
DB=duct burners
Lower emission rate for larger units (?)
Natural gas
Condensable PM Speciation
Method 202 (1991)
April 1-3, 2014 ICEPAG, Newport Beach, CA 14
0.000
0.002
0.004
0.006
0.008
0.010
0.012
4S
RB
+N
SC
R
RIC
E
4S
LB
RIC
E
Alp
ha
(he
ate
r)
Bra
vo
(p
ow
er
pla
nt)
Ch
arlie
(he
ate
r)
A (
boile
r*)
B (
hea
ter*
)
C (
ste
am
ge
nera
tor)
lb/M
MB
tu
O-CPM I-CPM (SO4=)
I-CPM (NH4+) I-CPM (Cl-)
I-CPM (Na) I-CPM (Ca)
I-CPM (K) I-CPM (other species)
I-CPM (other unspeciated)
157%
98%
130%
85%
133%
77%
128%
141%
Gas fuels
HeaterNGCC Boiler HeaterHeater Boiler4SRB
RICE
4SLB
RICE
Dilution Methods
April 1-3, 2014 ICEPAG, Newport Beach, CA 15
EPA CTM 039 (2004)
• Stirred jet mixing cone
• Dilution ratio 10-40
• Dilution air <50% RH, 80°F
• Mixing residence time 0.2
sec
• Diluted sample <85°F
• Water & acetone recovery
rinses required
• Method 5 gravimetric
analysis
• 0.1 mg balance
• Evaporate & dry rinses
• Dry & weigh filter
In-Stack Cyclones
10 &/or 2.5 mm
Dilution Methods
April 1-3, 2014 ICEPAG, Newport Beach, CA 16
Dilution Air
In-Stack Cyclones
10 &/or 2.5 mm
Heated
Probe
Flow
Meters
Mixing
Section
Aging
Section
Bypass
ULPA Filter
Charcoal Filter
Stack
Gas
Fan
PM2.5 Cyclone
(optional)
Flow
Valve
• (Parallel jet mixing)
• Dilution ratio ≥20
• Aging residence time 10 sec
• Diluted sample RH <70%
• PM2.5 cyclone after aging
• Ambient air gravimetric analysis
• 0.001 mg balance, equilibrate @ 20-25°C, 45-55% RH
• Recovery rinses (test or run)
Filter
(and other
media)
ISO 25597 (2013)
Flow
Meters
Dilution Methods vs. 201A/202-1990
• Dilution filter results consistently much lower than
M201A/202-1990 results in side-by-side tests at 7
different gas-fired sources
April 1-3, 2014 ICEPAG, Newport Beach, CA
0.0
0.5
1.0
1.5
2.0
2.5
3.0
201A/202 Dilution (filter) Dilution (filter +rinses)
mg
/ds
cm
3 NGCC's (2001-3)
1 NGCC (2008)
Two different dilution
methods
Modified EPA CTM 39 (2008)
“ISO 25597” (2001-2003)
M202 (1991) iced
impingers
sulfate artifact adds significant
positive bias
Blanks and samples
typically indistinguishable
Concentrations at
measurement “noise” level
17
PM2.5 Emission Factors – Dilution Method
OC accounts for >90% of PM2.5 mass
(but measurement background indicates artifact)
April 1-3, 2014 ICEPAG, Newport Beach, CA 18
0
20
40
60
80
100
120
Me
as
ure
men
t U
nc
ert
ain
ty,
%
Concentration
Result is quantifiable
Re
sult
is g
reat
er
than
ze
ro,
bu
t n
ot
qu
anti
fiab
le
Re
sult
can
no
t b
e d
isti
ngu
ish
ed
fro
m z
ero
LOD LOQ
Detection & Quantification
April 1-3, 2014 ICEPAG, Newport Beach, CA 19
Detection Limit
Quantitation Limit
Detection Limits
• 500 MW NGCC Block
• Fuel sulfur < 1ppm
low driving force for SO2
artifact
• EPA Methods 201A/202
• Paired trains, 4 runs
8 samples, 2 train blanks
Blanks & samples similar
• Measured emissions are
below detection &
quantitation limits
Facility permit limit between
LOD and LOQ
April 1-3, 2014 ICEPAG, Newport Beach, CA 20
NGCC Stack PM2.5: Dilution Results
-0.08
-0.06
-0.04
-0.02
0
0.02
0.04
0.06E
ch
o A
-HI-
R1
Ec
ho
A-H
I-R
2
Ec
ho
A-H
I-R
3
Ec
ho
A-H
I-R
4
Ec
ho
A-L
O-R
1
Ec
ho
A-L
O-R
2
Ec
ho
A-L
O-R
3
Ec
ho
B-H
I-R
1
Ec
ho
B-H
I-R
2
Ec
ho
B-H
I-R
3
Ec
ho
B-H
I-R
4
Ec
ho
B-L
O-R
1
Ec
ho
B-L
O-R
2
Ec
ho
B-L
O-R
3
mg
/ds
cm
Stack-background
AMS
A= DRI sampler
B = Compact sampler
Background PM2.5 (estimated from ambient monitoring
station PM2.5 (AMS), dilution ratio(DR), and dilution air
purifier PM2.5 penetration (PF)) is subtracted from stack
PM2.5
Stack and ambient air PM2.5 concentrations are similar
April 1-3, 2014 ICEPAG, Newport Beach, CA 21
Summary
• Current AP-42 factors for gas-fired sources
Few sources
High uncertainty
Old, problematic methods
• EPA Methods 201A & 202 (& similar methods) remain
problematic for gas-fired sources
Lack sufficient sensitivity
Blank levels are significant – not due to reagent contamination!
New Method 202 reduces but doesn’t eliminate artifacts
• Dilution methods offer advantages for gas-fired sources
Greater accuracy due to absence of SO2 and VOC artifacts &
greater analytical sensitivity
Resulting PM2.5 emission factors are ~1/10 or lower of those
based on hot filter/cooled impinger methods e.g. AP-42
April 1-3, 2014 ICEPAG, Newport Beach, CA 22
Summary (cont’d)
• Dilution method evaluation & refinement
Refine & validate test conditions & procedures
Reduce background levels in dilution air & probe recovery
solvents
Reduce organic carbon artifacts
Reduce relative variability of results
• More tests
Document method performance in field
Develop robust emission factors
April 1-3, 2014 ICEPAG, Newport Beach, CA 23
Questions?
ICEPAG, Newport
Beach, CA April 1-3, 2014
Glenn England
949-798-3643
24