EDGEWOOD RESEARCH, DEVELOPMENT & ENGINEERING CENTER

AD-A27 2 461
lI Illltl/]lt/ltltti

ERDEC-TR-093•

SCREENING SMOKE PERFORMANCE OF COMMERCIALLY AVAILABLE POWDERS I. INFRARED SCREENING BY GRAPHITE FLAKE

Janon F. Embury

TXC
3 •B

Walker Donald L.
RESEARCH AND TECHNOLOGY DIRECTORATE Curtis J. Zimmermarn

GEO-CENTERS, INC. Fort Washington, MD 20744

93-27681
Approved for public release; distributon Is unlanited.

July 1993

t

U.S. ARMY

CHEMICAL

AND BIOLOGICAL

B AD

DEFENSE AGENCY

Aberdeen Proving Ground, Mmyland 21010.5423

Disclaimer The findings in this report are not to be construed as an official Department of the Army position unless so designated by other authorizing documents.

REPORT DOCUMENTATION PAGE
mstliad0,., e0 Suite. 1104. AtlntoV 1)0 td veift .4 1nd lath 011fir o

JFort"
Final, 91 Aug - 93 Feb

Approved
1foso

Washingtoncomp1t04q Seie" vfd 0ito ateeg~f lo intlln IpucII orii at n waund newts~ dal* fI~c Matey,it#..etl flud~pt. Pmpeteworl snl Aputej" rthnroI.~t (DF04 a0Jell. Wonhingtoi. oc 201161.

1. AGENCY USE ONLY (Leave b aMn 4. TITLE AND SUBTITLE

.REOTAE3.EPTTYENDATES

COVERED S. FUNDING NUMBERS

1993 July Screening Smoke Performance of Commercially Available Powders 1. Inrrared Screening by Graphite Flake
*6. AUTIIOR(S)

PR-I10464609D200

7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES)

F.; Walker, Donald L. (ERDEC); and Zimmermann, Curits J. (01EO-CENTERS, llnc.)
Embury, Janoit

s. tORr~tiNG IORXNIZAI5FREPORT NUMBER

DIR, ERDEC,* ATrN: SCBRD-RTB, APO, MD 21010-5423 CEO-CENTERS, Inc., Fort Washington, MD 20744
9. SPONSORING i ONITORING AGENCY NAME S AND ADDRESM E

ERDEC-TR-093
0. SPONSORING iiMONITOnING AGENCY REPORT NUMBER

11. SUPPLEMENTARY NOTES

*When this work was performed, ERDEC warn known as the U.S. Army Chemimcal Research, Development and Engifletring Center, and the ERDEC authors were assigned to the Res earch Directorate.

IlaDISTIBU1IONAVAMILO SiTIMET
Approved for public release; distribution Is unlimited.
11ý ABSTRACT (Maxirmum 200wordu)

1b. DIMINIUTION CODE

Commercially available graphite powders are evaluated in the U.S. Army Edgewood Research, Development and Engineering Center Smoke Chamber. Figures of merit for weight, volume, and cost constrained applications are described to rate the screening ability of these materials in the visible and in~frared. Fundamental properties called "performance parameters" are Identified (extinction coefficient, deposition velocity, packing density, and dissemination yield) for measurements so that chamber characterization can be used to predict smoke performance in the battlefield,

14. SUBJECT TERMS

IS. UMBER OFPAGES"

Smoke materials IR extinction
I

Infrared screening JR absorption
OF T111S PAGE

Obscurants Powder technology
19. SECU"ITY (LAssirICAIION OF ABSTRACT

16. PRICE coDE
20

31

11. StCURItV CLASSIFICATION or REPORT

III. StCUR11Y CLASsirICATION

IITATION OF AilltNACT

_-NQLAS1IE

UNCLASSIFIED

NSN 1540 01 410W55OU

UNCLASSIFIED

IUL
sinfitJTud
f'or%-tb..,

by ANSI ltd IIIIllS

rom,, 5~a

(fl~v 2 89)

Blank 2 .

: A v.*% . This report has been approved for release to the public. Registered users should request additional copies from the Defense Technical Information Center. . . [' NTIS tl. This work was started in August 1991 and completed in February 1993. This report may not be cited for purposes of advertisement. unregistered users should direct such requests to the National Technical Information Service. DTIC QUAL'Y USPB 4 Aiaeu~sion DT I"' TABf NOV . Screening Smoke Material Engineering. 10464609D200. .• . The use of trade names or manufacturers' names in this report does not constitute an official endorsement of any commercial products.. 3I J .PREFACE The work described in this report was authorized under Project No.

Blank .

................................................................................ CONCLUSION ...... AND BUBBLES ....................CONTENTS Page INTRODUCTION ....J......... 20 EXTINCTION COEFFICIENTS OF CONDUCTIVE FLAKES................ GRAPHITE FLAKE MANUFACTURING ........................... FOAM I................................................ FIGURES OF MERIT ........................ ........... LTIERATURE CITED ............. BET SURFACE AREA .........................CLES .......... ERDEC SMOKE CHAMBER AND DATA COMPUTATIONAL ALGORITHMS HOMOGENEOUS ELLIPTICAL HALF CONE SMOKE PLUME WITH AEROSOL DEPOSITION ................................ 7 15 18 ............................ 23 26 29 30 31 5 ..........................

..........LIST OF FIGURES AND TABLES Figure I 2 3 Infrared Extinction Spectra of Asbury 260 and Lonza EKS4 ......................................... 13........6 M' Aerosol Characterization Facility .................. 8 o . 16 21 29 Table 1 Performance Parameters and Figures of Merit for Graphite Screening Materials Chamber Tests Utilizing SRI Nozzle at 60 PSI .. BET Surface Area and Extinction ...

sedimentation. or simply "yield" (Y) and is the first performance parameter that is used to judge smoke screening performance of a material. Impaction. deposition velocity (Q).SCREENING SMOKE PERFORMANCE OF COMMERCIALLY AVAILABLE POWDERS I. INFRARED SCREENING BY GRAPHITE FLAKE INTRODUCTION To measure a material's ability to screen in selected regions of the electromagnetic spectrum it is necessary to characterize the smoke material in terms of its performance parameters and figures of merit. c. It is very difficult to generate such particles with an in situ chemical reaction pulling mass from the air to contribute to aerosol mass. The yield (Y) values correspond to dissemination efficiency of dry powders pneumatically disseminated by the Stanford Research Institute (SRI) sonic velocity nozzle operated at 60 psi. Phosphorous has been a populRr snmoke screening material for a long time because it reacts with water vapor and oxygen in the air to produce an optically dense aerosol mass more than three times greater than the starting material mass (yields greater than three). It can be related to downwind smoke screening performance and takes into account reaerosolization. Phosphorus has an acceptably large alpha in the visible spectral region but not at the longer infrared wavelengths. and Brownian motion deposition of the smoke material. It quantifies the size of the "shadow" per mass of material when deployed as a tenuous cloud where no particle is in the shadow of another particle. Acceptable values for alpha (ax) in the infrared are achievable not by using micron size dielectric spherical particles such as those present in phosphorus smoke. so the maximum possible dissemination yield is unity. indicates the rate of aerosol fallout onto the gpound as a smoke plume is transported downwind. 7 Text continues on page 15. Infrared screening is necessary in these regions in order to defeat detectors such as indium-antimonide and mercury-cadmium-telluride that are pat of thermal imaging systems working in these respective wavelength bands'. The reaction with air allows the formation of aerosol droplets with a mass median diameter near the optimal value of 3/4 micron (p) for maximum alpha for visible screening at refractive indices in the range of 1. where smoke screening is required in the atmospheric infrared windows of 3-5 and 8-14 microns (p) wavelength. . In other words it is the size of the shadow cast by a single particle divided by the particle mass.5. These three performance parameters have been measured in the SERDEC smoke chamber (Figure 2) and are presented in Table 1. foam or bubbles where maximum dimensions are several microns and minimum dimensions (wall thickness in the case of foams and bubbles) become less than about a tenth of a micron. The second performance parameter that measures smoke screening performance is the electromagnetic extinction cross section per mass of aerosol and is generally represented by and referred to as the greek letter alpha. but instead by using conductive particles that are either flakes.33-1. The third performance parameter. The ratio of aerosol mass to transported precursor material mass is called "yield factor". fibers.

*0L 2~ ui S2 ? C C C.) WV)C ) 9 C.) .

Vý0 wC4 C4J~b w~C V.W)t- mmwui CE~ ~~ ~ ) I 0o§S4 __q___V-_ ____ ____ 000 q00 C q* Y0 9Y- 9.N~4 W(DU Cf 0 n0(1-r. Co 000 o ~I C44 - ci. 0 - r0 C4 0l (0 IDj 00 04 CY 0Y 0 c NY 09 0) 0 0 NJ NN §) S cc NN N9 ..

CD _ ____ _ _ _ _ _ _ _ __ _ _ ____ _ _SI'rSIC __ ____ ___ V_ _____PL C' ~ _ _ _ w 10 . N. *i 4y 0 CO . . O ~~w Cl * 0 wt~Mqr-0% V 0c.5- '- 0000 'al 00 0000 CRaq ~ V-V-W. 10" O cq 0 CR a~ C') 0 CD INC) C Y 04 W.

X I w Z MIZN wN Zr~ m0 a 5 0' 6~Z .9 q GQ 4V0t Iwo MY CO q "q* CD) CO~ W0C% M~' N co co 0 0t 0 o m V i )ac NY cm Y C'J C') CM 4 N4 N N4C N N Z. C) C) C.)3.V- V)9)- C ) VO Cj0 NO@ 0 ' 04 000)' .

C a c . co0 0 _oW)C C3 c. SO CY CD O N Cý z C. Iwo "* v o ocy DOO' N V (DNO qtI% 0ND~ V) C# 4 U)W)W (OSO Ih~f q -: C O U In W) N4 (04 W) V) -f -f . V.-)vr W)W) V.4 6. w C 12 .q R p-Y*0a - t ý clC C4 OR~ ci 4pNN @0!@ _ *0C t C ____ o0 ____ ~ m ~ D i-i I oot 'Co. w.

w- V- C~ or~ C) inIw'mo CC4%4 If I ) QC ___IV)_oV__) ED6 R CO~ C_ _ _ ccla 4 ciCi co ciV0 NY CY4 Nr I U~~~ 0 ) w 'L CL U* Iw _____ -R4* C ~~1- .

mqfq o I WE co _ _ _ __ _ _ _ ' I14 .WS _ _ _ _ Man _ _ LO q* (D LL8 _ _ _ _II w 4ow q.

fiber. One can see an example of this plateau/particle size effect in Figure I and Table I when comparing the a of Asbury 260 and Lonza E-KS-4 in the near. At longer wavelengths beyond the plateau ca drops quickly with increasing wavelength and the geometric optics treatment Is no longer applicable. size distribution and absorption due to conductivity all act to remove resonance structure that is so apparent in the cross section spectra of monodisperse nonabsorbing dielectric spheres. of conductive flake.EXTINCTION COEFFICIENTS OF CONDUCTIVE FLAKES. foam and hollow spheres at wavelengths less than about three times the major particle dimension and for minimum dimensions greatee than the skin depth. foam or hollow spheres are possible because of the absence of resonance structure. Thus the theory of geometric optics serves as a reasonable 2pproximation. Orientation distribution. Here the efficiency factor (Q) for extinction. FOAM PARTICLES AND BUBBLES Simple theoretical treatments of the electromagnetic extinction cross section per mass. an approximation that becomes more accurate at wavelengths much smaller than particle dimensions.is approximately equal to two (Q-2). mid and far infrared spectrum we want particles that lie on this high plateau with major dimensions greater or equal to about four microns that extend the plateau across the atmospheric infrared windows out approaching 14 microns wavelength. a lies on a high plateau independent of conductivity and complex refractive index. mid and far infrared regions. However. foams and bubbles throughout the plateau. on through the resonance region where resonance structure is washed out by size and orientation distribution yielding Q and a values only slightly greater than at the shorter wavelengths where the geometric optics theory is more strictly valid. The higher a curve is due to a smaller flake major dimension particle size (mass median diameter) and a thinner flake thickness (minor dimension). At wavelengths on the order of particle dimensions the efficiency factor represents an average over the largest resonances that appear in single particle cross section spectra so Q is somewhat higher especially when averaged over the lowest order resonant modes closest to the plateau edge. 15 . At longer wavelengths greater than about three times the particles major dimension a decreases below the plateau level eventually with a reciprocal wavelength squared dependence when the absorption cross section dominates and conductivity is represented by the dispersion relation of a free electron metal. the ratio of electromagnetic extinction cross section divided by geometric cross section. as apparent from continuous cross section spectral scans such as Figure I the peak value for a is broad and no more than about 20% greater than the value for az at shorter wavelengths. beyond the typical region where all particle dimensions are much greater than the radiation wavelength. a. For conductive flake. The level of the plateau will now be shown to be inversely proportional to flake thickness and foam or bubble wall thickness using the geometric optics theory to explain extinction of radiation by polydispersions of randomly orientated absorbing flakes. For best screening performance in the near.

In Q NJ Tin u NZ T- 46 > InI Ii a I to kn 0 ID in 6 o t 40a 16 .

is the flake thickness. a well known result puts the geometric cross section <G>=-S/4 equal to one fourth the surface area. with the definition of a.. Because adjacent cells share walls. then the geometric optics solution puts Q=2 and if we assume particles are randomly orientated and convex. the electromagnetic extinction cross section per mass pV of material where p is the particle density and V is the particle volume. Thus we find that a averaged over random orientation becomes <(>ae 2S/4 = S pV 2pV For our flake particles S/V-. For bubbles and spherical foam particles of radius R we have S/V=3/R at a reduced density p' which becomes in the case of a bubble with wall thickness t4 41cR 3 3 A foam particle of radius R may be considered to consist of a large number N of roughly spherical cells having radius r and cell wall thickness t.2. the ratio of the electromagnetic extinction cross section divided by the geometric cross section G. Noting that N=R 3/r' we find a reduced density p' for the foam particle 17 . averaged over random orientation angles'Q (a= acx 1 pv (QG)O QG)1"W If we assume major particle dimensions are greater than about one third of the wavelength. where t.Combine the earlier definition of Q. total foam particle mass may be expressed as the sum of the masses of N bubbles having radius r and wall thickness tW2 plus a single envelope bubble of radius R a•jd wall thickness t...2/t.

Graphite flakes are not toxic to aquatic life in the environment. density p in g/cmO and a in m2/g. they have to be transported as foam and hollow sphere. Natural graphite is known to occur in three different forms. foam particles or bubbles with major dimension greater than one third the maximum wavelength to be screened with the smallest possible minimum dimension in order to maximize a throughout the spectral region. Thus among the various types of conductive particles that screen effectively in the infrared. It has been shown that explosive dissemination of carbon fibers results in the presence of fiber fragments and combustion products in the respirable range3 . Conductive fibers longer than several microns with diameters on the order of and less than a tenth of a micron are available commercially as graphite from Hyperion and experimental quantities of iron whiskers with this morphology have been made by 3M. Furthermore. Carbon having the true graphitic structure is available commercially from naturally occurring feedstock (embedded in minerals) and synthetic fabrication. from a liquid for example. in contrast to brass flakes. brass and aluminum flakes . Conductive flakes are available commercially and graphite flakes have 4 distinct advantages over the earlier infrared screeners. as a refractory material graphite will not undergo cold welding during explosive dissemination.s with solid walls at prohibitively low packing densities. only graphite flake is acceptable on all counts.Thus the extinction coefficients comparing these three types of particles averaged over random orientations become for flakes for bubbles 1 for foams A convenient and self consistent set of units puts particle dimensions in microns. each produced from different geological conditions. and graphite flake aerosol does not suffer a flashing problem near a dissemination source as aluminum flake does because of its reactivity. Graphite flake falls into two categories. We need conductive (even very strongly absorbing dielectrics are not sufficiently broadband) flakes. natural (mined) and synthetic. 18 . Graphite fiber is not desirable from a toxicological point of view because it does not break up in the lungs and the morphology is similar to that of asbestos. GRAPHITE FLAKE MANUFACTURING Unless the conductive foam and bubble aerosol particles are produced in the field.

5 depending on the different structures '. further processing of the commercial flake including ball and/or jet milling. It w'uld be of great interest to determine the ratio of crystalline to amorphous silica present in a natural graphite material and determine via inhalation studies if there exists an inhalation toxicity associated with natural graphite flake. The natural graphites are known to contain a small amount of silica as a contaminant. It is important to remember that only naturally occurring flake graphite represents a true flake (analogous to mica) and is capable of being intercalated and exfoliated to yield very thin flakes having high aspect ratios7 . the concentration of which depends on the geological origin of the graphite.26g/cm 3 and has 5 a Mohs scale of hardness of -0. The calculated density for graphite is -2. orientation of the crystallites and the nature of the milling operations. Air classification can be utilized to provide the milled powders in a particular particle size and narrow size distribution. air classification. The resulting produced graphitic solid is then coarse ground and finally processed via ball and/or jet milling to yield a fine "flake" like powder. This material being deposited as a result of decomposed carbon-rich materials that later transformed to graphite under pyrogenic conditions. Conversion of the feed stock to produce synthetic graphite may have a yield as high as 85%. A small fraction of this silica may be in the crystalline form. Stringent control over particle size reduction leads to uniformly sized powders.5-1. Vein graphite. powder densification and grinding for deagglomeration could prove more advantageous to one graphite form than the other'. Because of the morphology differences between natural and synthetic graphite. The third distinct form is amorphous graphite which results from graphitization of coal deposits due to the presence of molten magma. These different forms of graphite are rarely found together due to their different geological origins. which must be restricted to very low levels under 1/2% because of inhalation toxicity. also known as lump graphite. Synthetic graphite's "flake" like morphology is a result of many factors Including the petroleum fraction calcined.. Synthetic graphite is produced typically from petroleum coke (a byproduct of the petroleum cracking process) or anthracite coal calcined to high temperatures (electrothermal graphitization). with the flake more true in shape being analogous to mica. Natural graphite can be mined as both crystalline flake and vein graphite. 19 . is found as veins throughout rock structures.Natural flake graphite is found in metamorphose rocks that have undergone plastic flow. Identification of synthetic graphite is rather simple since it will have a carbon content of at least 99% and the residual ash after ignition will contain silicon carbide5 .

) and Cv(t 2 taken over time intervals tj to t.+v and t2 to t+r respectively allows computation of C. and uD as follows. Dissemination yield and deposition velocity are computed based on a series of two 0 filter concentration measurements.. we find the instantaneous concentration C as a function of time t. 0 p. Two FTIR spectrometers measure aerosol transmittance over the spectral regions 0.Ce -%)tDO This expression in combination with two filter concentration measurements Cp(t.4 p-l.. deposition velocity % and chamber height H C . A mixing fan is operated continuously in the chamber at a low speed to maintain uniform concentration and provide a level of turbulence driving reaerosolization and impaction approximating those components of aerosol deposition in the battlefield. yield (Y) and deposition velocity (t. A photodiode array spectrometer measures aerosol transmittance over the wavelength range of 0 .2 2 p.9p-3p and 2 . Because the air pump flow rate through the filter is monitored with a rotometer and held constant at dV/dt we have C. The aerosol sedimentation component of deposition will of course be independent of whether the aerosol Is in a chamber or on the battlefield.ERDEC SMOKE CHAMBER AND DATA COMPUTATIONAL ALGORJTHMS The 14 cubic meter smoke chamber used to measure the performance parameters such as the electromagnetic extinction cross section per mass of aerosol (a). a rotometer and vacuum pump are used to make aerosol concentration measurements at a flow rate of 20 liters per minute..*~~ i. The filter concentration measurement accumulates incremental mass dm contained in incremental volume dV over time t through t+c. The Stanford Research Institute sonic pneumatic nozzle is operated at 60 psi to disperse and deaggregate powders to produce an aerosol of primary particles'. At reduced concentrations a quartz crystal microbalance (QCM) and an aerodynamic particle sizer (APS) measure aerodynamic particle size distribution. Glass fiber filters.) is shown in Figure 2 with the full smoke characterization instrument configuration.* d f-d. initial coricentration C. Using the stirred settling model" to describe the situation where concentration is maintained uniform throughout the chamber by a mixing fan. dV&n d -r 20 .5 p.

-. 2!J .I .JlIi SIII Iti~ ItI U !I I t--.

We identify dn/dV as the instantaneous concentration C so substituting the stirred settling expression fC~e fco e dt di C" OH"• Taking the ratio of the second filter concentration divided by the first.cv m The extinction coefficient is computed using Beers law along with measurements of concentration C and transmittance T with knowledge of transmissometer path length L through the smoke a CL T 22 .( ti) Solving for the initial concentration H HJIr IeN the dissemination yield Y is found knowing chamber volume V and measuring mass disseminated m . the duration of both measurements being c C.( 0t- C.( ti) so that the third screening smoke performance parameter deposition velocity is In C.

Nfa(O)c(Od fc)C(odl fc(odz fc(t)dl Pft We now develop an expression for smoke areal density CL downwind by using an elliptic half cone smoke plume model in which there is complete mixing. but we should not think of CL as anything more than the areal density of smoke over the line of sight.HOMOGENEOUS ELLIPTICAL HALF CONE SMOKE PLUME WITH AEROSOL DEPOSITION We can judge the performance of a screening smoke material based on the optical depth ECL of a smoke plume produced from that material and the duration c of the plume. E is the average extinction coefficient along the line of sight. This is evident when we define averages . If the cloud boundaries were defined and L could be specified then C could be thought of as concentration C averaged over inhomogeneities along the line of sight. "5CLmfa()C(md The integration over variable 1 is along the line of sight. The cross sectional area of the elliptical half-cone is iHL/2 where H is the plume height and L is the width. Mass conservation requires that the decrease in plume incremental mass m as it progresses downwind in direction x satisfy the expression 23 . but this is true only in a chamber. We approximate the areal density CL of a smoke plume along a horizontal crosswind path and along a vertical crosswind path through the plume centroid path by developing an expression here for a homogeneous plume whose boundaries are defined by an elliptical half cone. This gives nearly the same result as the Gaussian Plume model incorporating deposition and is simpler to describe.

... aH. plume height and width are functions of downwind distance x LDCL .(HL4.CL--.+CL~ab Mx b. 1 +CHaLbLx'I-') vw Rewriting 2 dx dClax= 2 VIVw C X abx•b (bH+b) X 24 L MIl_] .x b where the coefficients aL. and Hwa.dx The deposition velocity vD dx 2 governs the mass loss over the ground surface area Ldx • =-V M. bL and b..DCLdX Making the substitution dt-dx/u• where uw is the wind velocity dx vw Setting the above two expressions equal to one another VW dx 2 AL Because the concentration. .(HL--. .I. depend on Pasquill category and terrain roughness. Thus V. dC -C.CL=..ZdH )C- vw 2 dx ax a Using the relationships that have been developed for elliptical half-cone plumes with a Gaussian concentration profile L-ax.ý-7 .

.. and breadth L. ie.L.& Solving for initial concentration and substituting this into the expression for downwind concentration 25 . so that rhY=vwCo-. The quantity v.. the flowratc out of the generator rh multiplied by dissemination yield Y.H. may be written Lo. represents the wind speed and initial plume height H. downwind 0 close to the generator 2 In =(b +b)ln..(l'-b)vw Mass concentration requires at x._2 X I nC x a. that the mass flowrate into the air.x.Solving in terms of an initial concentration C at an arbitrary initial distance x.a. satisfy tie following equation where the depositiun velocity of aerosolized material has not yet had sufficient time to reduce acrosolized mass because of the proximity to the generator.

while for volume limited applications "r=VprWwhere V is the maximum volume that can be hwansported and p is the density of smoke material contained in volume V. windspeed. We can write the crosswind horizontal optical depth wy . aerosol deposition velocity % onto the terrain and a parameter y that depends on downwind distance. dissemination yield Y..aaLX 2 The crosswind horizontal optical depth of the plume near the ground is czCL Q' u eAP and the vertical optical depth at the plume center is 22 FIGURES OF MERIT We write the duration of the plume =M/fih in terms of the average mass flow rate th and maximum mass M that can be =ansported for weight limited applications. Pasquill category and terrain roughness. 2 x'". plume effective height H.2W- as a function of wind velocity v.. 2a.lv.-I.(l6-b) 6w 26 . aCL..

rockets. V (atype. packed material density p in the storage volume V transported and the cost per pound." I VP . volume V and cost per pound of material $ are found by requiring that the optical depth produce a transmittance level T. and for volume limited applications such as grenades. The area screened per mass M.. and %.l The portions of %. Y and UD that we measure in the chamber. that depend on material properties have been put in parentheses and will be called the weight limited Ow and volume limited Ov figures of merit. we approximate the above two figures of merit as well as a financial limited figure of merit 4' that gives the square meters of screening per dollar of cost. volume or cost of material for weight. it would be proportional to the same figures of merit for a uniform layer of smoke having depth L and area A. Because we can not remove the dependence of y upon variables such as windspeed and downwind distance and because within the first few hundred yards downwind yDO is a reasonable approximation for all materials here. we substitute values for duration in weight limited applications such as large area screening Hv M Hv. the dimensions of czYp are screening area per volume transported and the dimensions of the financial figure of merit aY divided by unit cost is square meters of screening per dollar spent. for effective screening.*454aY/cost per pound [square meters/dollar] These figures of merit depend on the performance parameters cz. mass of aerosol. respectively. Thus the dimensions of ofY are screening area per mass transported. If a goal function were defined as area screened per mass. smoke pots and mortars W. yield dimensions are mass of aerosol per mass transported and packing density dimensions are mass transported per volume transported. An upper limit on p is the density of the particle itself and that Dimensions of extinction coefficient are screening area per is the value tabulated.Defining a goal function i=yaCLc as the product of crosswind optical depth multiplied by duration. 27 . ww-aY (square meters/gram transported] OvwaYp [square meters/cubic centimeter transported] 0•. volume or cost constrained situations.

aY =M Ta =€ Note that for a uniform layer of smoke the area screened per mass. volume or cost for a uniform layer of smoke..*w CCL L I (A T. This expression provides the value for CL which is then substituted into expressions for area screened per mass.1 or equivalently when the required optical depth is unity. Actually the requirement is closer to a transmittance of 5% which corresponds to an optical depth of three in which case the figure of merit should be divided by three to give area screened per mass.aYCL linT. Ow 2. volume and cost to obtain (Al A I Y. volume or cost of material is just the respective figure of merit when the wansmittance required for screening is e.In.. 28 ..

..0 0 11 11T...II i W•"' .II•... I i ll..00 KS-4 E 2. . Using the geometric optics expression developed earlier relating ac to envelope surface area S.00 30..iI .50 0 .60 5. .00 E co-6154 a SOLg N-3203 U" 1..00 10 S-5539 s-go N-71bT 3 2 0 4 N-500 70. "--..(BET.50 o ..00 25. This was explored as a possible alternative to chamber testing with the just mentioned negative result. No doubt this can be attributed to variability in the porosity (concavity and small pits in the surface) that will increase BET surface area but not a.. This is evident in Figure 3. BET Surface Area and Extinction 29 .. It is interesting to note that some of the better graphite screening materials have BET surface areas around 25 m2/g indicating that the porous area dominates over the "envelope" surface area indicated by a.00 BET Surface Area (m 2/g) Figure 3. . we can write the approximate relationship "W.Porous Area)/2 which shows the effect of porous area. v N-75 0 KS-" CO-7525 N-1SO C: S-* 5-450 N-3442 2. I .BET SURFACE AREA The approximate BET surface area contained in the first column of Table I does not correlate veryjweU with the electromagnetic cross section a. I I'l I.00 10. I .00 15.. I I..50 0 . 3.00 20.. 0. II ' .

1.)of 1. dissemination yield. rockets. deposition velocity and powder packing density) has been presented. A wide variety of commercially available graphite powders have been tested in the ERDEC smoke chamber using the SRI sonic pneumatic nozzle at a pressure of 60 psi for dissemination. 30 . The third figure of merit gives the square meters of screening per dollar of smoke material cost and is useful in situations such as training with large area smoke screening. Testing of these materials in the ERDEC smoke chamber to obtain dissemination yield. mortars and smoke pots.CONCLUSION The concept of describing competing smoke materials in terms of four measurable performance parameters (extinction coefficient. Three figures of merit based on these four performance parameters have been introduced. The first figure of merit gives the square meters of smoke screening per mass of smoke material transported and is useful in weight limited applications such as the large area smoke generators. Performance parameters and their product derived figures of merit are tabulated in Table 1 so that materials can be compared over the visible. For example if a minimum weight limited figure of merit (0. Here for example the weight constraint of the large area smoke generator vehiule would have to be met first by specifying a minimum value for the first figure of merit (weight fimited) and then comparing all materials satisfying this constraint based on the third figure of merit (financial limited). 3 -5p and 8-14p spectral regions. Graphite flake manufacturing processes were described. but also quantitatively to predict cloud opacities downwind or screening areas. Toxicity of course might become an issue if the 16. the most cost effective material from Table 1 would be Asbury 999 with a financial figure of merit (0p) of 1362m%/. deposition velocity and spectral extinction coefficients from the visible through 20 microns wavelength was also described. artillery rounds. All three are proportional to smoke plume optical depth downwind and can be used not only to rank performance.5p region were required so that a sufficient quantity could be transported to accomplish a large area screening mission.7% non-carbon component is more than 1/2% crystalline quartz.50m/g in the 3. Chamber data processing algorithms and downwind homogeneous plume transport including deposition were derived.06p. The second figure of merit gives the square meters of screening per volume of smoke material transported and is useful in volume limited applications such as grenades.

. Wiley-Interscience. 305. Embury. Chiyoda. Virginia(1983). unpublished results. US Patent. Macmiliam Company. "Natural Graphite. R. S. Deepak Publishing. Y.195(1984). and Takeuchi. D. 8. 9.. 4. 3. H. CJ. Deepak. 2nd ed..8588(1991). Endoh. and Zimmermann.. Dissemination Techniques for Aerosols. and Zissis.. ed. 71-77(1983) Smoke/Obscurant Symposium XVI.. 66. J.355(1964).. "Extinction by Clouds Consisting of Polydisperse and Randomly Orientated Nonspherical Particles at Arbitrary Wavelengths". April 14-16. Pergarnon Press. 7."Alpine Handbook of Mechanical Processing Technology". Shaffer...A. Fuchs.." in Encyclopedia of Chemical Technology. GJ.L. 6. A. New York. Wolfe. Environmental Research Institute of Michigan (1978). 31 . 2. W. 4..484. 5.780 Third Avenue.B. Kuga. Inc. Seeley. The Infrared Information and Analysis (IRIA) Center. Embury. K. 71(1992).. New York(1964). J. The Mechanics of Aerosols. S. Optical Engineering. 22(1). Walker.LITERATURE CITED 1. 10.. Vol. 4. A. Suite 3201 New York. N. NY 10017. The Infiared Handbook. Powder Technology.. Hosokawa Micron International. "Powder Processing of Synthetic Graphite Flakes".

Master your semester with Scribd & The New York Times

Special offer for students: Only $4.99/month.

Master your semester with Scribd & The New York Times

Cancel anytime.