


Our sponsors: 
FAA,
NASA, DoE, AFOSR, Industries 
Projects: 



Ozone Assisted Combustion
If ozone is doped into
air and ethylene is used as fuel in a diffusion jet flame,
ozonolysis reaction starts in the mixing layer. The heat
lease and radical production from ozonolysis reaction will
cause autoignition of the diffusion jet flame. For lifted
jet flame, if ozone is added, autoignition kernels will form
in flame front and cause flame attachment. From the below
images from high speed camera, it can be clearly shown the
flame propagation is controlled by autoignition kernel which
is separated from flame front.
The time differences between frame (a) and the rest frames
are (b) 0.2 ms, (c) 0.4 ms, (d) 0.6 ms, (e) 0.8 ms, (f) 1.2
ms, (g) 2.0 ms, (h) 4.0 ms, (i) 6.0 ms, and (j) 10.0 ms.

Autoignition
and Combustion Stability of High Pressure
Supercritical
CO_{2}
OxyCombustion
The successful development of directly heated supercritical
carbon dioxide gas turbine power systems utilizing
pressurized oxycombustion, that have high efficiency while
capturing nearly 100% of the CO_{2} produced, is
critically dependent on the design of combustors. Two key
issues that need to be considered in the design of
combustors are basic burning properties of the mixture
(e.g., autoignition times), as well as the high pressure
system dynamics (e.g., combustion stability). The objective
of this project is to experimentally investigate (for the
first time) the autoignition delays at practical conditions
for SCO_{2} oxycombustion (150 to 330 atm). An
optimized chemical mechanism will then be developed for the
design of combustors for SCO_{2} oxycombustion.
Large Eddy Simulation (LES) will utilize the mechanism
developed in this proposed program to simulate autoignition
and flame stability in a dump combustor under these
operating conditions.
Project presentation:
pdf

Combustion Instability Control Using
Plasma 
Laser Absorption Spectroscopy for
Combustion Chemistry 
Dynamic Adaptive Kinetics for Turbulent
Combustion
In this project, a new regimeindependent framework for 3D
DNS of turbulent combustion with detailed kinetics is
developed by incorporating onthefly adaptive kinetics
(OAK), correlated transport (CoTran) techniques, and an
efficient pointimplicit ODE solver (ODEPIM) into a
conventional DNS platform. All three methods are modified
and optimized to adapt to 3D turbulent combustion and
parallel high performance computing (HPC). A canonical
turbulent premixed flame configuration corresponding to the
thin reactionzone regime is considered, where an initially
planar premixed flame front interacts with a decaying
isotropic turbulence. The computational domain consists of a
cube with length 0.015 m. With the new numerical frame work,
calculation of chemistry can be accelerated 46 times,
calculation of transport can be accelerated 72 times, and
overall acceleration of calculation is 20 times.
Temperature (upper) and vorticity (lower) at the center
plane (z = 0.75 cm) using conventional DNS (left) and
proposed framework (right)
Mass fraction (upper) and reaction rate (lower) of fuel at
the center plane (z = 0.75 cm) from conventional DNS (left)
and proposed framework (right)
spatial distribution of the number of active species at the
center plane (z = 0.75 cm)
Average CPU time distribution per time step (sec) for four
methods (from left to right): DVODE, ODEPIM, ODEPIM+OAK, and
ODEPIM+OAK+CoTran

Numerical Modeling of Plasma Assisted
Combustion System 
The 1D numerical model contains the Poisson
equation for electric potential, electron energy equation, and
species continuity equations for all charged and neutral species
given by Eqs. (1)(3) respectively,
where electron energy density
is
given by the product of electron density and
electron energy.
The transport of energy and species is calculated by the drift
(mobility)diffusion model. Reduced electric field (E/N) and energy
input can then be calculated rather than prespecified as in 0D
model.
The gas flow is modeled by solving
mass, momentum and total energy conservation equations
simultaneously, as given by Eqs. (4)(6) respectively. Unlike
equilibrium plasma, which transfers electrical energy only into
sensible enthalpy, Joule heating in nonequilibrium plasma transfers
electrical energy into the total energy of the gas mixture. This
means that not all energy from Joule heating contributes to gas
temperature rise. Heat release from chemical reactions is implicitly
included in the unsteady term of total energy with the form of
chemical energy converting to sensible enthalpy.
The simulations were conducted at
pressure p=60 Torr, and initial temperature T=300 K, to match the
experiments. The initial mixture composition was C_{2}H_{4}:Ar:O_{2}
= 0.062:0.75:0.18 in molefractions.
Zero potential is set at the left
boundary, and gap voltage at
the right boundary. is
obtained from the applied voltage by
the equation (7), where dielectric constant is
4.8 for quartz, and 3.2 for silicone rubber.
Zero flux wall boundary condition is
used for neutral species. The wall boundary fluxes for electrons,
positive ions, negative ions and electron energy are given by Eq.
(8)(11) respectively.
where secondary electron emission coefficient
is
taken to be 0.05. Temperature of secondary electrons ejected from
the electrode surface is
assumed to be 1 eV. =1
if ,
and =0
otherwise.
A zero flux boundary condition is
also imposed for the mass and momentum conservation equations.
Analytic selfsimilar solutions of transient temperature
distribution in a semiinfinite solid with constant heat flux are
used as the boundary condition for gas temperature:
where thermal conductivity of quartz
is
1.4 Wm^{1}K^{1}. In practice, this boundary is
closer to isothermal than adiabatic conditions.
See more details and
comparison with experimental data
here.

