lign=right>

ELECTRICALLY CHARGED WATER MISTS FOR EXTINGUISHING FIRES Charles H. Berman and Otto P. Andersen, Jr. ELECTRICALLY CHARGED WATER MISTS FOR EXTINGUISHING FIRES
Charles H. Berman and Otto P. Andersen, Jr.
AeroChem Research Laboratory
The Titan Corporation
PO Box 2229
Princeton, NJ 08543-2229
and
Stuart A. Hoenig
Associates in Applied Research, Inc.
80 W. Yvon Dr.
Tucson, AZ 85704-5234
March 1998
Purchase Order Number MDA972-97-M-0010
AeroChem TP-568
Final Report
prepared for:
The Department of Defense
Strategic Environmental Research and Development Program
The views and conclusions contained in this document are those of the authors and should not be
interpreted as representing the official policies, either expressed or implied, of the Defense Advanced
Research Projects Agency or any other part of the U.S. Government. TP-568
1
SUMMARY
A brief experimental fire suppression study found that electrical charging of water mist can
substantially reduce both the time and the amount of water required to extinguish a pool fire.
Another benefit of charging was an increased spraying angle, which occurs even for low pressure
sprays. It was also found that much lower voltages than reported by other workers can be very
effective in charging the mist and extinguishing fires. The specific objectives of the study were to:
(1) compare the motion of charged and uncharged water mist droplets near a flame and (2) determine
the reduced time for fire extinguishment due to electric charging of the water mist. The current
program addressed ceiling sprinkler extinguishment of compartment fires, and the same principles
should hold for aircraft engine nacelle fires. The main technical problem encountered was
achieving a uniform distribution of water mist over a significant surface area both with and without
charging. Promising results were obtained despite this difficulty which can be corrected with
additional work. This study has achieved the objectives of element 4d of the Next Generation Fire
Suppression Technology Program and additional funding should be provided to optimize the
technology and demonstrate its effectiveness on larger scale fires. TP-568
2
I. INTRODUCTION
A.
BACKGROUND
The technical basis for this program originated with Dr. Stuart Hoenig at Associates in Applied
Research, Inc., AARI. He suggested that his experience with water droplet charging and
AeroChems experience with electric field effects on flames could be utilized to demonstrate that
an electrically charged water mist would extinguish a fire more efficiently than an uncharged mist.
Basically, less water would be needed if the charged droplets were attracted to the fire. The original
motivation for the concept was the extinguishment of conventional civilian sector fires by generally
horizontal water streams in regions such as the Southwest and rural areas having limited water
resources. Under NGP sponsorship, the program was directed towards the application of water mist
charging to Navy compartment fires using ceiling sprinkler systems.
Because of the limited amount of time and funding available, the focus of the program was on
demonstrating that electric charging of the water mist produces a beneficial effect on small scale
fires. The results obtained indicate that there is indeed an improved performance. Additional time
and funding are needed to validate and extend these results, and more importantly, to better
understand the basic mechanisms, optimize the charging technique, and test the approach on larger
scale fires.
B.
EFFECTIVENESS OF WATER MIST
Water mist is highly effective in suppressing fires. For example, it has been shown that considerably
less mist (by mass) is needed to extinguish a pool fire [Ndubizsu et al, 1997] compared to nitrogen.
The prime reasons for this effectiveness are the rapid evaporation of the small water droplets which TP-568
3
enhances cooling of the fire, oxygen displacement by the water vapor, and absorption of radiant
energy [Mawhinney et al, 1994]. However, the small size of mist droplets can also be detrimental
if the droplets do not have sufficient momentum to reach the fire due to air currents induced by the
fire plume. For this reason, the idea of using a charged mist to take advantage of the electrical
properties of flames to increase the number of droplets near a fire was proposed, and this was the
major thrust of the current program.
C.
FLAME IONIZATION
It is well known that hydrocarbon flames contain electrical charges because of the process of
chemiionization [Calcote, 1962]. In chemiionization, flame radicals react to form electrons and
positive ions at temperatures much lower than those required for thermal ionization. This is the
principle behind flame rods which detect the presence of a flame in industrial burners through a
measured electrical current. In a normal flame both positive and negative ions (primarily formed
by electron attachment) are formed in addition to the electrons. The high mobility of the electrons
results in the outer region of the flame being negatively charged by the electrons and the bulk of the
flame region being positively charged by the preponderance of positive ions.
Much of what is known about flame ionization has been learned from studies of premixed flames.
Calcote [1962] found that the maximum ion concentration at low pressures varied little for premixed
flames composed of various fuels such as methane, propane and ethylene while significantly higher
levels were attained with acetylene. The ion concentrations were higher at atmospheric pressure, on
the order of 10
11
ions/cc, with the differences between methane [Wortberg, 1965 and Peters et al,
1969] and propane [Calcote, 1963] being relatively small.
The peak ion concentration for an atmospheric pressure, methane-air diffusion flame was found to
be in the same range as that for the atmospheric pressure premixed flames discussed above [Calcote
et al, 1988]. We are not aware of ion concentration measurements for liquid fuels. TP-568
4
For real fuels and real fires (almost always diffusion flames) there are a number of considerations
which could lead to either increases or decreases in the flame ionization, but these cannot be
quantified at this time. For example, real fuels contain impurities (e.g., metals) which can greatly
increase the ion concentration. On the other hand the formation of soot particles in cooler regions
of a fire act as sites to which electrons will attach. This keeps the negative charge closer to these
regions of the fire than would occur if the electrons were free to diffuse away on their own. Thus,
electron attachment may reduce the charge separation (at least in the fire plume) that normally
makes flames appear to be positive.
While it cannot be certain which effects will dominate in real fires, there is no doubt that diffusion
flames in air behave as though they are positively charged since they are strongly pulled by
negatively charged objects. This has been demonstrated when the fuels are ethylene [Payne and
Weinberg, 1959], methane [Berman et al, 1987], propane, oil, wood and paper [Hoenig, 1995].
D.
ELECTROSTATIC INTERACTIONS
The application of an external electric field in the vicinity of the flame results in the motion of
electrons and negative ions towards the positive electrode and the motion of positive ions in the
opposite direction towards the negative electrode. More momentum is generally exchanged by
collisions between the positive ions and the neutral gas molecules as compared to the momentum
exchanged by collisions with the negative species dominated by electrons. This momentum
exchange results in the production of an ionic wind [Chattock, 1899] which is a flow towards the
negative electrode. This ionic wind is responsible for the large flame displacements reported above,
and the effect has been used to control blowoff stability [Berman et al, 1987], which can be
increased or decreased depending on the orientation of the electrodes that produce the electric field.
These effects can be produced with extremely low electrical power levels, on the order of 10
-2
% of
the power released during combustion [Calcote and Berman, 1991]. TP-568
5
1.
Relative Motion of Charged Mist and Fire
There are several different scenarios in which electric charging of the water mist droplets can
interact with real fires. A negatively charged droplet approaching the vicinity of the fire will repel
electrons and negative ions, and there will be an attraction between the n