United States
                   Environmental Protection
                   Agency
Air and Energy Engineering
Research Laboratory
Research Triangle Park NC 27711
                   Research and Development
 EPA/600/S7-88/018  Dec. 1988
&EPA         Project  Summary
                   Mathematical  Modeling  of
                   Single  Droplet Trajectories  in
                   Combustor  Flow  Fields
                   W. S. Lanier and S. B. Robinson
                    In liquid hazardous waste incin-
                   eration, the ballistics of some large
                   single  droplets can limit destruction
                   efficiency. Mathematical modeling
                   and experimental work were  per-
                   formed in this study to examine the
                   behavior of individual fuel droplets
                   sprayed into  a combustor, to
                   determine which parameters could
                   influence incinerator effectiveness. A
                   computer model has been developed
                   to predict the  motion,  heating, and
                   evaporation of such a  droplet  in a
                   heated environment. The model  is
                   based  on droplet behavior governed
                   by  fluid  mechanics and contains
                   formulations to predict  the effect of
                   droplet  spacing  on  drag  and
                   evaporation as well as  the effect of
                   evaporation on drag. The gas  flow
                   field can be  specified with mean
                   velocities or a randomly fluctuating
                   turbulent field, based  on  experi-
                   mental values for the standard
                   deviation of the velocity. Numerical
                   predictions of the Initial heating and
                   evaporation of an isolated burning
                   droplet are  compared to  experi-
                   mental results.  The predictions,
                   which utilized measured temperature
                   and velocity gas fields,  compared
                   well to experimental observations.
                   The correlation of droplet penetration
                   with droplet incineration suggests
                   that incinerator failure modes may be
                   predicted on  the basis of  droplet
                   atomization parameters and gas field
                   conditions.
  This Project Summary was devel-
oped by  EPA's Air and  Energy
Engineering Research  Laboratory,
Research Triangle Park,  NC,  to
announce key findings of the research
project that is fully documented in a
separate report of the same title (see
Project Report ordering information at
back).

Introduction
  The first of several tasks in the EPA's
Fundamental Combustion Research
program (FCR III)  was to examine  more
closely the characteristics of  a single
droplet in a combustion environment.
This  report summarized the modeling
work  performed to predict the trajectory
and the evaporation of a single droplet. It
included  a description of the theory,
development, and  operation  of  the
computer code and is accompanied by a
user's manual. This computer code is to
be used in conjunction with experiments
which are  performed by  EPA  in  an
inhouse research program. This model is
applicable to liquid injection hazardous
waste incineration.

Computer Model
  The model developed for this program
is  expected to  increase the under-
standing of droplet combustion in liquid
injection incineration. It is quite important
to  know the behavior of individual
droplets in a hazardous waste incinerator.
In  these incinerators,  waste must  be

-------
destroyed  to  a degree greater than
99.99%. The incinerator is said to fail  if
this destruction  efficiency  is not
achieved. If there are droplets which are
larger than the mean droplet size, they
tend to concentrate on the outer edge of
the  spray core,  with sometimes
unpredictable behavior. This model can
be used to help predict the  behavior of
these rogue droplets, and to determine
additional  information  that  might be
required  to  fully characterize the
incinerator.
  The computer model for  predicting  a
droplet's  trajectory was adapted from  a
model developed at  University  of
Sheffield  for the U.S. Air Force Office of
Scientific Research. The work presented
by the University of Sheffield  included an
algorithm for  predicting droplet tra-
jectories  in three dimensions, including
droplet evaporation and initial heat up
effects. The model was used to predict
the environment inside a gas turbine
combustor, primarily as  a tool to assist in
combustor design. Starting  with the
University  of  Sheffield equations, the
computer  model  was written  in
FORTRAN  IV. Two   models  were
developed one to characterize a droplet's
path in two-dimensional space  and  the
other  in  three-dimensional  space.
Initially the two-dimensional version was
written for  an  IBM  PC, but the three-
dimensional version,  requiring  much
more storage space, was maintained and
executed on  an IBM  3070  mainframe
computer.  Subsequently,  the three-
dimensional version has  been modified
to be manageable by an IBM  PC/AT.

Use of Program
  The  initial  effort for the  FCR  III
program  was to examine the effects of
droplet  spacing  on  droplet  drag.
Experiments were performed at the EPA
to determine this effect, and  empirical
relations  were developed  from the
experimental  results.  The  next  effort
examined the  effect of evaporation on
drag. It was postulated  that as a droplet
evaporates, the environment immediately
surrounding the  droplet will change,
changing the drag force on the droplet.  A
theory was incorporated into the model
to account for the effect of  evaporation
on drag.  The effect of  burning has also
been included into  the evaporation
equation, based on the theory that the
burning rate  of many  droplets can be
expressed  as  a function  of  the burning
rate of an  isolated droplet.  The original
model  accounted only  for  evaporation,
not  burning.   Finally,  the  effect  of
turbulence  in  the background gas was
included in  order to provide a  more
realistic model of actual conditions in an
incinerator.
  Experiments were  performed over  a
wide range  of droplet  and flow  field
conditions. Corresponding  numerical
predictions are  compared to  experi-
mental results in Table 1 and Figure 1.
Experimental  results  were recorded for
location  of droplet  ignition and  final
burnout, whereas the model predicts only
burnout location. Model predictions show
penetration   into  the combustor  to
approximately  the same location  as the
experimental ignition  point for most  test
cases.  However,  the  time predicted for
droplet  heating  to  the  ignition
temperature  is about 30-50% of  the
time predicted for droplet burning, which
is approximately the order of magnitude
seen in experimental testing. It  is  not
currently  understood why  the  model
predicts burnout too  early.  This could
indicate oversimplifications in the model
or perhaps some physical  phenomena
which  are unaccounted for.  Another
possible reason for the inconsistency is
the  fact that a multi-component fuel was
used  in the  experiments  which  was
difficult  to  characterize and  model
correctly.  The model  assumes that the
fuel has a single  boiling  point, with no
evaporation until it heats up to the boiling
temperature.   In addition,  it  assumes  a
homogenous  droplet with  constant
droplet  properties. These assumptions
are  not  totally  valid  for  a  multi-
component fuel with  a variable boiling
point (depending  on  the  percentage of
various constituents) and  hence  the
predicted  droplet penetration  may be
skewed.

Code Limitations
  The  model  allows the user to predict
droplet  trajectories within an incinerator
at  a  variety  of  initial  droplet  and
background  gas conditions.  Never-
theless, there are some limitations to the
code. The geometry of the incinerator is
limited,  in the 3-D case, to a  circular
cylinder. Dimensions  are specified as
program inputs   Eventually this  code
may be expanded to handle  droplet
sprays  and  group  combustion,   but
presently  it is assumed  that  only  one
droplet  is injected at a  time.  Another
limitation is  the characterization  of  the
fuel. It is assumed   that the droplet
properties are constant with temperature,
and that there is a single boiling point for
the  fuel. This is not realistic for all fuels,
but is  an assumption  necessary  to
simplify calculations.

-------
 CO
 Q.


 I
O
s
 ia


|.s
$8.
« a
11



42 c
"So
wo.
CD ._
CC 3
— 0
f|
||
ujS>




w
S
CD
1
cf
1

o
p
c
^^
!
Q













*M
»
&

ill
^s

$•=-
CO O
a: ^
76 'c'
|t
^ s^
q) — v




5"£-
"5 E
 CM co co
o> o o v
CM CO CO CO

GO O) O> O
CM CM CM O O IO
10 5 5 CM
0 O 10 10
t. i. CM »

CO


-
^
270/270 1




CM
00

g
io
CM
IO
*


^
(O
><
o





















s





















-------
    20
  I  0
    20
   20
 I  0
   20
           20    40   60    80
                     (cm)
           20    40   60    80
                     (cm)
    20
  I  0
    20
   20
   20
   20
 I  0
   20
           20   40   60    80
                     (cm)
           20   40    60    80
                     (cm)
Test Point 1
Test Point 2
                                                                    Test Point 3
                                                                    Test Point 4
                                                                    Test Point 5
          20   40    60    80
                     (cm)
A = Experimental Ignition Point   D = Experimental  Burnout Point   X = Model Predict/on
Figure 1.    Droplet trajectories—experimental results and model predictions. (Sheet 1 of 2).

-------
   20
  § O
    20
   20
 I  0
   20
           20    40   60    80
                  (cm)
   20
   20
 6  0
 u
   20
   20
   20
           20    40   60    80
                   fern)
          20   40   60    80
                  (cm)
           20   40    60   80
                   (cm)
7"esf Point 6
                                                                   Test Point 7, 8
                                                                   Test Point 9, 10
                                                                  Test Point 11,12
                                                                  Test Point 13
          20   40    60   SO
                  (cm)

A = Experimental1 Ignition Point   D = Experimental Burnout Point   X = Model  Prediction
Figure  1.    Sheet 2 of 2.

-------
 W.  S. Lanier and S. B.  Robinson are with Energy and Environmental Research
  Corporation, Durham, NC 27707.
 James A. Mulholland is the EPA Project Officer (see below).
 The complete report, entitled "Mathematical  Modeling of Single Droplet Trajec-
  tories in Combustor Flow Fields," (Order No.  PB 88-252 010/AS; Cost: $21.95,
  subject to change) will be available only from:
        National Technical Information Service
        5285 Port Royal Road
        Springfield, VA 22161
        Telephone:  703-487-4650
 The EPA Project Officer can be contacted at:
        Air and Energy Engineering Research Laboratory
        U.S. Environmental Protection Agency
        Research Triangle Park, NC 27711
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use $300

EPA/600/S7-88/018
                    lISlr!RS«««
                                               60604

-------