Transcript Document

Agenda: more bad particles
• Secondary aerosols
– Biogenic example – alpha pinene
– Anthropogenic example – SOx and NOx
• Particle control technologies
– Brief overview of the rules
– Overview of the technologies
– What to use for what type of particles
• Basic design strategies
• Cyclones
ATTACK OF THE KILLER
TREES
a-pinene and contributions to
ambient secondary organic aerosol
in Maryland
Sheryl Ehrman, Rich Calabrese, Yu Jin Choi, Anu
Nadarajan (M.S., January 2002)
ATTACK OF THE KILLER
TREES
a-pinene and contributions to
ambient secondary organic aerosol
in Maryland
Motivation
• Presence of high concentrations of ambient
aerosols has many adverse effects:
– Haze, building and plant damage, illness, death
• Implementation of effective controls requires
knowledge of what to control
• Biogenic emissions (that we can’t control – unless
we chop down all plants) have long been known to
be precursors to organic aerosol
• Because aerosols are not directly emitted, new
acronym - secondary organic aerosol = SOA
Objective
• Preliminary study of formation of SOA from
biogenic emissions
• Estimate contribution of an important compound,
a-pinene, to SOA formation
• Explore relationships between variables such as
emission rate, temperature, time, on SOA
formation from a-pinene
• Explore sensitivity of SOA formation from apinene to changes in model parameters.
The Suspect Compound
a-pinene, CAS 7785-26-4
Background image with permission
Plan of Attack – long term
• Develop comprehensive model of SOA
formation from biogenic and anthropogenic
compounds
• Evaluate/improve model via comparisons
with experiments in smog chambers
• Evaluate model/improve model with field
data
• Use results to develop effective controls to
reduce SOA concentrations
What we need
• Emissions information for all compounds of
interest
• Detailed chemical mechanism for SOA
formation, including gas-to-particle
conversion and partitioning between
gas/particle phases for semivolatiles
• Information about loss rates
• Meteorology/transport
What we have to work with
• Emissions information for some compounds of interest
• Not so detailed chemical mechanism for SOA formation
from very few compounds, including gas-to-particle
conversion and partitioning between gas/particle phases
(simple yield measurements from smog chamber studies)
for semivolatiles
• Information about loss rates as a function of particle size,
but not composition because we are not sure about
composition. We don’t treat wet deposition.
• Meteorology/transport information
What we did with it
• “Knowledge adds, wisdom lets slide”- Paul
Westerberg
• We kept it simple and started with a box model
– Simple chemistry, simple treatment of gas-to-particle
conversion, simple loss
• Previous studies – steady state, yet haze episodes
fortunately do not reach steady state!
• We learned some interesting things from our
dynamic study
The dynamic box model
OH(t), T(t)
k(t,T) – reaction rate
with a-pinene
K(t,T) – equilibrium
constant for partitioning
Emissions (t,T) (mg/hr)
Simplified chemistry:
MP (t) (mg/m3)
Depositional Loss
a-pinene  stuff that mostly condenses (with partitioning coefficient K1)
+ semivolatile stuff (with partitioning coefficient K2)
More details about the box
Input Parameters For Baseline Run
QUANTITY
VALUE
REFERENCE
EM (t),
(mg/m3)
Time-dependent
CMAQ Estimates
Thanks Jeff!
OH(t), (ppm)
Time-dependent
CMAQ Estimates
Thanks Jeff!
T, (K)
Time-dependent
CMAQ Estimates
Thanks Jeff!
k, (hr-1)
17873.OH(t).exp
(444/T)
Kamens and
Jaoui, 2001
k’ (hr-1)
10-6
Wesely and Hicks,
1985
K1, K2
(m3/mg)
0.171, 0.004
Odum et al., 1996
a1, a2
0.038, 0.0326
Odum et al., 1996
.
Model inputs: OH and EM
Money shot I
EST
7PM
6AM
N
50
7PM
D
6AM
N
7PM
D
6AM
N
7PM
D
6AM
N
7PM
D
SOA concentration (m g/m3)
45
40
35
30
25
20
15
10
5
0
0
12
24
36
48
t (hr)
60
72
84
96
Money shot II
25
3
Concentration (m g/m )
20
15
10
5
0
20%
40%
60%
% Baseline Emissions
mg/hr
80%
100%
Alpha-Pinene Conc.
SOA Conc.
Sensitivity Analysis Results
ANALYSIS
a-Pinene Concentration
Increase
Emissions Decrease
5%, 10%, 25%, 50%, 70%
OH Conc. Decrease
5%, 10%, 25%; (Inc.15%)
XX-Significant Change; X-Insignificant Change.
Decrease
None
SOA Concentration
Increase
Decrease
XX
XX
X
X
Temperature Decrease
5oC, 10oC; (Inc. 5oC)
X
XX
Enthalpy Decrease
17.5 kcal/mol, 25 kcal/mol
X
XX
Deposition Velocity Decrease
0.01 cm/s; (Inc. 1 cm/s)
X
XX
None
Conclusions from our study
•
•
•
•
SOA from a-pinene appears to form quickly
Snowball sort of effect (non linear with time)
Also, non linear with emissions!
When temperature decreases, haze from a-pinene
increases
– Bit counterintuitive, but chem rxn rate to form SOA
decreasing doesn’t beat more semivolatiles condensing
• Need for high time resolution measurements of
SOA and of a-pinene is clear (ongoing with new
students and with collaborations elsewhere)
What we can’t conclude from our study
Other SOA precursors?
• Other hydrocarbons…(anthropogenic, biogenic)
• Sulfur Dioxide – simplified (possibly outdated)
mechanisms
Soot or metal
oxide surface
SO2 +O3  SO3 + O2 or SO2 + ½ O2  SO3
heterogeneously
then SO3 + H2O  H2SO4  (H2SO4)n , an aerosol droplet
• Nitrogen Oxides – more complicated relationship, NOx
contributes to ozone formation, and reacts with water to form nitric
acid. Rxns of NOx with VOC’s lead to nitrated hydrocarbons, some of
which can condense to make particles, others are just gas phase but
irritants
• Ref – Spiro and Stigliani, Environmental Science in Perspective,
SUNY Press, Albany 1980
Brief history of air regulation
• Pre 1955 – no regulation (!) just keep neighbors
happy
• 1955 – Air Pollution Control Act
– Various fed agencies were authorized to investigate
cause/effect relationships
– States and local governments are responsible for
maintaining and improving air quality within their own
jurisdiction (this is important, still holds) Transport is a
big issue for states; they sue each other!
• 1963 – Clean Air Act, more research at fed. Level,
more funds to state and local govt’s to improve air
pollution control
More historical stuff
•
•
1963 act amended many times
Major amendments 1967
1.
2.
3.
4.
5.
•
Establishment of air quality control regions (AQCRs)
Development of specific mandated air quality control
criteria and control technologys
Adoption of National Ambient Air Quality Standards
(NAAQS) to protect public health and welfare
Requirements for State Implementation Plans (SIPs)
to ensure air quality meets NAAQS
Further development of crazy system of acronyms
Particles regulated as total suspended particles
(all particles less than 55 microns), vague…
 Poor particles, who will care about them?
The particles in more detail?
• 1987, EPA changed the NAAQS to have separate
standards for TSP and PM10
• PSD (prevention of significant deterioration in
attainment areas) regulation was changed to reflect
ONLY PM10 in 1993 (figuring all the bigger stuff
will settle out, and not make us sick anyways)
• PM2.5, regulations proposed but not yet in place
• National Regulations for PM10:(states can be higher)
– PM 10 Annual average concentration = 50
micrograms/m3
– 24 hour average standard 150 micrograms/m3
How are the regs related to the
controls at the source?
• Permits!
• Local authorities require permit for operation
• In 1990, CAA established national operating
permits in addition to local permits
• Permit sets emission limits
• Some types of permits (sulfur emissions for
example are traded)
• Permits designed to bring area into compliance or
to maintain attainment with standards
• Someone at state/local agency figures this out…
Two types:
• Permit to Construct (PTC)
– Not required at federal level unless it’s a major source
• Permit to Operate (PTO)
– Specifies maximum allowed emissions rates
– Exactly how facility will be operated
– How certification of compliance will be determined
(operators police self with penalties for not doing what
permit says)
• Compliance proved via stack testing, other testing,
calculations (there’s no way it could escape over
the fence!) including mass balances, etc..
Particle control technologies
• No one magic bullet, often combo of separators is
needed
• Brainstorm on what phenomena can be used to
eliminate particles at the source
– Big particles (resuspension dust, greater than 20
microns)
– Smaller particles (1 to 10 microns)
– Nanoparticles
Note: Exam next week will cover all we have discussed
up to this slide, focus on fundamental phenomena and
problem analysis as well as some quantitative problem
solving
Control design approach from
intro to ChE perspective
Basic technologies to remove particles
•
•
•
•
•
Electrostatic precipitators – Zap it!
Filters – Suck it!
Scrubbers – Wash it!
Cyclones – Fling it!
Tradeoffs? Costs, lifetimes, maintenance,
efficiency, appearance (just kidding)
ESPs
• Applies electrical force to get particles out of gas
stream
• ESPs can handle large volumetric flow rates
– Low pressure drops
– High efficiencies
• Downsides?
– Not cheap
– Inflexible to changes in the process (designed for
certain efficiency on certain sized particles)
Filters
• In industry, typically: baghouse filters, looks like
bank of really big socks!
• Air carrying dust particles is forced through a
cloth bag. As air passes, dust accumulates on
cloth, leaving air cleaner. Dust is periodically
removed by shaking or reversing air flow
• Goods – filters have high efficiencies, can handle
lots of different kinds/sizes of dusts
• Bad – costly, and limited to dry, low temperature
conditions
Scrubbers
• Impacts and intercepts dust particles via collisions
with water droplets
• Solid particles can be separated from water
stream, or whole stream can be treated in some
other way before discharge (or if you live where
there are no water regulations, and you are not
nice..)
• Goods – can scavenge some gases too (mass
transport!), high efficiencies
• Bads – costly, and you have a wastewater stream
to deal with
Cyclones
• Simplest is no turning, gravity settling, great for
big chunks
• More sophisticated: cyclones use centrifugal force
to fling large particles out towards wall of tube
• Goods – not as expensive, great for large particles,
easy to maintain
• Bads – efficiency poor for small particles (where
inertia does not dominate particle motion)
• Cyclones often used as pre-cleaners
Important dimensionless numbers
• Resistance coefficient = Euler number
– Relates cyclone pressure drop to some characteristic
velocity:
Eu 
p
(  f v 2 / 2)
• Gives ratio to pressure forces to inertial forces
• Constant for given cyclone geometry, important for scaleup
• What characteristic velocity should we use?
– One based on gas flow rate/x sectional area
4q
v
 D2
Stokes 50 number (Stk50)
• Physical significance: Ratio of centrifugal force
(less buoyancy) to the drag force acting on a
particle of diameter x50
2
x50
 pv
Stk50 
18 m D
• m - gas viscosity, density is particle density, D is
cyclone body diam., char.velocity is same as
previous.
• x50 =
• Cut size… at this diameter, 50 percent of particles
will be collected
Cool rule of thumb:
• For well-designed cyclones, there is a direct
correlation between Eu and Stk50
12
Eu 
Stk50
• This is pretty neat because you could find
x50 by knowing specifics about cyclone
Design stuff
Other factors
• Dust loading – if there is a high concentration of
particles in gas, this leads to higher separation
efficiencies because of particle enlargement via
agglomeration. Humidity increases magnitude of
this effect.
• Abrasion – critical zones for abrasion = just
beyond inlet opening, and in the conical part near
dust discharge
• Blockage – usually caused by overloading of
solids outlet orifice, caused by mechanical defects,
changes in chemical or physical properties of
solids (sudden dose of humidity)
More factors
• Attrition – break up of solids could occur as particles are
smacked against the wall; large particles affected more
than smaller ones. Big concern in fluidized beds where
cyclones are used to sort out the particles that escaped, to
return them to the bed.
(draw this)
• What would happen to size distribution of fluidized bed
particles over long times in this case?