Modeling Toxic Industrial Chemicals (TICs) and CWAs

1
Modeling Toxic Industrial Chemicals (TICs) and
CWAs Using an Atmospheric Chemistry Module in
SCIPUFF
Douglas S. Burns, Veeradej Chynwat, Jeffrey J.
Piotrowski, Kia Tavares, and Floyd L. Wiseman
ENSCO, Inc., Melbourne, FL
Run Photochemical Box Model (PBM)
Background The goal of this DTRA sponsored
project was to generate a set of polynomial
functions that describe the atmospheric
degradation rates for typical Toxic Industrial
Compounds (TICs). Atmospheric degradation rates
generally depend upon a variety of meteorological
parameters such as the time of day, zenith angle,
relative humidity, temperature, and ambient
conditions. The largest rate processes for most
organic species include the reactions with
hydroxyl radical (OH), ozone (O3), and nitrate
radical (NO3). Current atmospheric models, such
as the Carbon Bond Mechanism (CBM), account for
the reactions with these indigenous species, as
well as reactions with other species, and often
these models contain over a hundred chemical
reactions and close to a hundred different
chemical species. The computer code for
generating the concentration profiles for the
TICs/CWAs requires solving a large system of
ordinary differential equations (ODEs), that is
computationally intensive. The benefit of
replacing this large set of ODEs with a set of
polynomial functions is in saving both cpu and
wall clock run time. The drawback is some loss
of precision by using the simpler
polynomials. ENSCO, Inc. has developed polynomial
functions for describing the atmospheric
degradation rate constants for a series of
Alkenes (e.g., 1-butene and 2-methylpropene),
H2S, and Sarin. The parameters in the
polynomials are obtained by fitting the
polynomials against model-generated data. The
polynomials are then incorporated into the
SCIPUFF TD Model as part of an atmospheric
chemistry module in HPAC. Tests were conducted
to ensure that model run time is not affected by
incorporating the chemistry module. 1-butene is
fairly reactive, as are many other TICS, and so
the effect of including a chemistry module will
be to reduce the toxic footprint for a release of
1-butene. The methodology described here has
been developed and can be readily applied to any
atmospheric pollutant.
Develop Chemical Data for TICs / CWAs

TIC Atmospheric Reactions of 1-butene
or
H2O


NOX
kOH kNO3 kO3

CH2CHCH2CH3 OH ??? 0.94 C2H5CHO (1-butene)
CH2CHCH2CH3 NO3 ??? 0.6 CH3CH2CH(OH)CH2ONO
2 0.12 C2H5CHO 0.11 H2CO
CH2CHCH2CH3 O3 ??? 0.35 C2H5CHO 0.63
H2CO 0.41OH Rate -(kOHOH kNO3NO3
kO3O3) 1-butene Rate -keff 1-butene
VOCs

ZA f(Lat, Lon, DOY, Time of day)
CO

O3
Location (lat, lon)

Example Results
CWA Atmospheric Reactions of Sarin (GB)
kOH kNO3
Concept
TD Only vs TD Chemistry
CH3-P(O)(-F)(-CH(CH3)2) OH ??? Degradation
Product (Sarin) CH3-P(O)(-F)(-CH(CH3)2)
NO3 ??? Degradation Product Rate
-(kOHOH kNO3NO3) Sarin Rate -keff
Sarin
Pollutant OH, NO3, O3, etc.
CWAs
kOH, kNO3, kO3, etc.
Degradation products
Transport, diffusion, and chemical reactivity

• Strategy
• Implement Algorithm into SCIPUFF / HPAC
• Develop a chemistry module (currently called
  degrade.dll) that interfaces with the dispersion
  algorithms in SCIPUFF.
• Develop data for algorithm for the atmospheric
  degradation of TICs / CWAs
• Requirements specified for development of the
  chemistry algorithm
• The algorithm must run rapidly and not impact
  model wall clock run-time
• The algorithm must account for all modeling
  scenarios encompassing a wide range of
  meteorological conditions (i.e., changes in cloud
  cover, temperature, ambient conditions, etc.).
• The algorithm must be robust enough to account
  for diurnal changes to the degradation rates of
  TICs and CWAs.
• The algorithm should account for the potential
  generation of intermediate toxic compounds. In
  some cases it may be possible for the degradation
  products to be more toxic than the pollutant that
  was released.
• Data Development
• Translate C(t) data to keff using the following
• Fit keff data to a series of polynomials and
  determine the best polynomial function.
• The polynomial algorithms are a function of
  meteorological parameters (i.e., T, SE, CC, time
  of day, H2O, etc., and ambient concentrations
  of indigenous atmospheric species (OH, O3, VOCs,
  NO3). Ambient concentrations are associated with
  five super classes of Land Use (Urban, Grassland,
  Forest, Desert, and Water)

Generate c(t) data

Calculated Plume is TIC Dependent
Mechanism ID Rxns, EA, k(T) (w/ OH, NO3, H2O,
O3, etc.)
Implement TIC / CWA Data in Detailed Model such
as the CBM



Decomposition of Butene as f(Latitude)
Run Photochemical Box Model (PBM) as a f(met
parms)
Butene as f(time of day) T 290 K Land use
Urban Latitude 25 50 N


Butene ppm
50
Obtain cTIC(t) as f(met parms)
Populate SCIPUFF w/ keff data
Derive Empirical keff (met parms)
25
Time of day hrs from midnight
Implementation of Algorithm in SCIPUFF
Tracking of Reactant and Product(s)
Plume contours of 1-butene and propanal (product)
at 4 and 8 hrs after release (Urban scenario for
2 hour continuous release of 1-butene)
Translate to keff
1-butene
propanal

• Summary
• An atmospheric chemistry capability that does not
  impact model run time was incorporated into
  SCIPUFF.
• Algorithms were developed for the atmospheric
  degradation of nine alkenes, H2S, and Sarin.
• The effective degradation rates (keff) for TICs
  and CWAs are describe by a series of developed
  polynomial functions that are a function of
  important meteorological parameters
• Temperature, solar elevation, cloud cover, time
  of day, moisture level, latitude, ambient
  conditions (Land use is used as a surrogate for
  air quality)
• The algorithm can account for the formation of
  degradation products (e.g., propanal from
  1-butene)
• Future plans are to increase the data base to
  include other TICs and CWAs.