A*17 3 OPTICAL CHARCTERIZATION OF TROPOSPHERIC MEROSOLSCU) 1*1CHEMICAL RESEARCH DEVELOPMENT RHO ENGINEERING CENTERABERDEEN PROVING GROUND "D H R CARLON SEP 9?

UCASIFIED CRDEC-TR-97070 F/O ?n/3

mmEmmmmmmEmmmE

1.0f 4. 1 c a

UU ~ 1.32

J. .1

1111% %I~i~Ir

00 -DEVELOPMENT riENGINEERING

0 CENTER* CRDEC-TR-87070

OPTICAL CHARACTERIZATION OFTROPOSPHERIC AEROSOLS

by Hugh R. CarlonRESEARCH DIRECTORATE

September 1987

ELECTEUDEC 03 1%7U.S. ARMY

DEC0 3 187ARMAM=S Q DCHEMICAL COMMAND

Aberdeen Proving Ground. Maryland 21010-5423

Approvd fOPUbUc 101987 24 12

I

Disclaimer

The findings in this report are not to be construed as an officialDepartment of the Army position unless so designated by otherauthorizing documents.

Distribution Statement

Approved for public release; distribution is unlimited.

-44

.4

-4

UNCLASSIFIED, I,/' '

SECURITY CLASSIFICATION OF THIS PAGE

REPORT DOCUMENTATION PAGEi. REPORT SECURITY CLASSIFICATION lb RESTRICTIVE MARKINGS

UNCIASSIFI]D

2a SECURITY CLASSIFICATION AUTHORITY 3 DISTRIBUTION/AVAILABILITY OF REPORT

2b DECLASSIFICATION DOWNGRADING SCHEDULE Approved for public release; distributionis unlimited.

4 PERFORMING ORGANIZATION REPORT NUMBER(S) S MONITORING ORGANIZATION REPORT NUMBER(S)

CMEC-TR-87070

6a NAME OF PERFORMING ORGANIZATION 6b OFFICE SYMBOL 7a. NAME OF MONITORING ORGANIZATION(if applicable)

CMDEC SMCCR-RPT-P

6C. ADDRESS (City, State, and ZIPCode) 7b ADDRESS (City, State, and ZIP Code)

Aberdeen Proving Ground, MD 21010-5423

8a. NAME OF FUNDING/SPONSORING Bb. OFFICE SYMBOL 9. PROCUREMENT INSTRUMENT IDENTIFICATION NUMBERORGANIZATION (if applicable)

CMDEC SMCRM-RSP-P

8c. ADDRESS (City, State, and ZIP Code) 10 SOURCE OF FUNDING NUMBERSPROGRAM PROJECT TASK WORK UNIT

Aberdeen Proving Ground, MD 21010-5423 ELEMENT NO NO NO ACCESSION NO

1L161101 A91A

11 TITLE (Include Security Classification)

Optical Characterization of Tropospheric Aerosols

12 PERSONAL AUTHOR(S)Carlon, Hugh R.

13a. TYPE OF REPORT 13b. TIME COVERED / 14 DATE OF REPORT (Year, Month, Day) 15 PAGE COUNTTechnical FROM81 Jan TOJ2 Dec 1987 September 28

16. SUPPLEMENTARY NOTATION

171 COSATI CODES 18 \SJECT TERMS (Continue on reverse if necessary and identify by block number)FIEL GROUP SUBGROUP Aerosols ? Tropospherep Extinction/0 04 Optical Scattering Isosbestics Mie Theory e'...

0419. STRACT (Continue on reverse if necessary and identify by block number)

Canplex calculations of optical scattering using high-speed electronic computers havebecme so ccmnonplace that the technology has evolved into several specialty fields,e.g., extinction spectroscopy, angular scattering measurements, and nephelcmetry. Asa consequence, the overall technology and our ability to remotely characterize orquantitatively analyze atnospheric aerosols might be further advanced than realized byworkers in the individual disciplines who do not have the time to keep up with advancesin other disciplines. This report examines techniques of transnission (extinction)and turbidity spectroscopy, polarization and angular scattering measurements andrelated developments, and concludes that camputer programs can be written to optimizedesired parameters in interdisciplinary investigations.

20 DISTRIBUTION /AVAILABILITY OF ABSTRACT 21 ABSTRACT SECURITY CLASSIFICATIONaUNCLASSIFIEDUNLIMITED 0 SAME AS RPT OTIC USERS UNCLASSIFIED

22a NAME OF RESPONSIBLE INDIVIDUAL 22b TELEPHONE (Include Area Code) 22c OFFICE SYMBOLTIMOTHY E. HAMPTON (301) 67P-2914 SiCCR-SPS-T

DO FORM 1473.84 MAR 83 APR edition may be used until exhausted SECURITY CLASSIFICATION OF THIS PAGEAll other editions are obsolete U1CLASSIFIMD

PREFACE

The work described in this report was authorizedunder Project No. 1L1611O1A91A, Independent Laboratory In-HouseResearch (ILIR). This work was started in January 1981 andcompleted in December 1982.

The use of trade names or manufacturers' names inthis report does not constitute an official endorsement of anycommercial products. This report may not be cited for purposesof advertisement.

Reproduction of this document in whole or in part isprohibited except with permission of the Commander, U.S. ArmyChemical Research, Development and Engineering Center, ATTN:SMCCR-SPS-T, Aberdeen Proving Ground, Maryland 21010-5423.However, the Defense Technical Information Center and theNational Technical Information Service are authorized to reproducethe document for U.S. Government purposes.

This report has been approved for release to thepublic.

Acknowledgment

Valuable contributions to this report were made byB. V. Gerber. He suggested studies on which parts of this reportare based and is underwriting the expansion of our experimentaland computational capabilities.

Aooession For

WTIS GRA&IDTIC TAB 0Unannounoed 0Justifioatlon _

Distributionz/

K vatilability Code

Avail pa/or

V1 , 1 p-vw m -M -M M - .

CONTENTS

Page

1. INTRODUCTION AND BACKGROUND........................... 9

2. AEROSOL EXTINCTION MEASUREMENTS....................... 10

3. POLARIZATION AND ANGULAR SCATTERINGMEASUJREMENTS......................................... 18

4. SUMMARY AND CONCLUSIONS............................... 23

LITERATURE CITED............................... ....... 25

5

LIST OF FIGURES

Figure Page

1. Values ofaXD/L vs. Di Calculated from theMie Program for Water Using the Data of Haleand Querry ................... .................... 12

2. Comparison of Computed Curves Relating Visibleto Infrared Optical Transmittance of WaterFog to Experimental Data Points for Cooling,Steam-Generated Water Fog Clouds ............. 13

3. Curves for Three Infrared WavelengthsComprising the Broadband Composite ofWavelengths Shown as the Abscissa ofFigure 2 .......................................... 14

4. Calculated Curves from the Data of Hale andQuerry for Water Droplets Showinga) vs. D"..................................... 15

5. Computed Values of ax for Liquid DropletAerosols Comprising Several Concentrationsof Orthophosphoric Acid (H3PO4 ) in Water ..... 17

6. Plot of Equation (3) forvX Values of 0.1-4.0Shown on the Curves and Kx Values Shownon the Abscissa ........ ......... ................... 19

7. Fractional Transmittance for a C x L Product= 5.0 of Water Fog vs. Droplet Diameterfor Wavelengths .............................. 20

8. aX vs. DM Calculated from Mie Program UsingData of Hale and Querry at X = 10 km forWater Droplets ............................... 20

9. Polarization RatiopO)vs. Scattering Angle efor DOP ...................................... 22

I7

~ ~ ~ N~ :-. ~%'V: -

OPTICAL CHARACTERIZATION OF TROPOSPHERIC AEROSOLS

1. INTRODUCTION AND BACKGROUND

With the increasing availability over the past twodecades of high-speed electronic computers, complex calculationsof optical extinction and scattering of electromagnetic radiationby atmospheric aerosols using the Mie theoryl have becomecommonplace. The technology has advanced so swiftly thatspecialists measuring cloud extinction of aerosols often areunaware of advances being made simultaneously by other specialistsmeasuring angular scattering patterns of aerosols with nephelom-eters. Such measurements and calculations can be done in numerouswavelength regions, but the technology which concentrated firstupon the visible wavelengths now is expanding rapidly, especiallyinto the regions of relatively high atmospheric transparencycalled 'windows' in the infrared. This technological expansioncan be measured by a sampling of papers appearing in journalsduring the past year, discussin aerosol measurements orcalculations of extinction,2,3jM5,6,7,8,9 angular lightscattering or nephelometry,5,10,11,12,13,14, I lidar 16,17,18,19and related aerosol technology including thermal emissivity 20 andcalibration of optical particle counters. 2 1

Several trends have been noted in all of this work.First, the utility and information content of simple extinctionmeasurements has tended to be overlooked in the rush to exploitscattering techniques. Second, the polarization techniques do notseem to include applicable techniques, some new, that can beapplied both to extinction and to polarization measurements.Third, if taken in combination at this point in time, the overalltechnology might be further advanced than is realized by workersin the individual disciplines. For example, remote characteri-zation or quantitative analysis of tropospheric aerosols seems adistinct possibility, both by conventional transmission(extinction) spectroscopy and by angular scattering or back-scattering techniques. This report examines these possibilitiesby first reviewing the technology of aerosol extinctionspectroscopy.

The Beer-Lambert equation for atmospheric aerosols canbe written:

-2n (Tx) aX CL (1)

where for a given wavelength X(pm), TX is the fractional trans-mittance, a) is the optical mass extinction coefficient of theaerosol (m2g-l), C is the aerosol mass concentration (gm-3 ) and

9

L is the optical path length (i). Some form of Equation 1 hasbeen used by spectroscopists virtually since the birth of spec-troscopy and the equation has become so well known that the powerof simple extinction measurements of aerosols often is not fullyappreciated. For example, using suitable multi-wavelengthtechniques, extinction measurements can contain sufficientinformation to allow direct determination of the mean size of aspherical droplet aerosol distribution or even remotely tocharacterize chemical reactions in cloud droplets at isosbesticwavelengths, without the use of Dolarized radiation or other moresophisticated techniques. 2 2,23,24,25

It is not surprising, however, that polarizationtechniques have received heavy emphasis since their exploitationin the early 1940's to characterize military fog-oil smoke meandroplet diameter based, in part, in earlier work at the U.S.Bureau of Standards and by Stratton and Houghton using the Mietheory.26,1 The polarization ration then was measured only fortransparent spheres for which, if the complex index of refractionmw is given by (n - ik)) where 7 is the real index andk) is the imaginary index (closely related to the absorptioncoefficient), then kX = 0. Polarization measurements can be usedto determine particle size distribution in addition to mean sizeof aerosol droplet distributions. This was recognized and thetechnique was developed in the 1960's. 2 7 ,2 8 ,2 9 By 1969, thetechniques of particle size distribution measurements had beenextended, as discussed in Kerker's classical book on lightscattering. This book gives an excellent review of the technologyas it existed then.

0

In recent years, there is an increased awareness thataerosol particle absorption, e.g., in the infrared where k 0 o,can be interpreted in new ways utilizing not only extinctionmeasurements 3r polarization measurements, but the twotogether.22,2i,2 ,30 The author concurs that this is a promisingfield for new research and this Laboratory is developing anexpanded capability for scattering and polarization measurementsto complement ongoing research in extinction spectroscopy ofaerosols.

2. AEROSOL EXTINCTION MEASUREMENTS

It is known that the variation of kX from values nearzero in the visible wavelengths to significant ones in theinfrared wavelengths provides a method for the determination ofaerosol mean droplet size and mass concentration from simpleextinction measurements if two or more wavelengths are selectedproperly for observation. For example, at the He:Ne wavelengthX = 0.63 Am,k;-O and typical water fog droplets have diameters DP(in um) such that Dp>>X or, to use Kerker's criterion, the sizeparameter rDp/X 2.0 so that the conditions for geometricscattering exist. This leads to a constancy of the product a? . DL3.0-3.2, where a, is defined in Equation 1. At the same time,

10

the product aX• DTp decidedly is not constant with Dg at wate-lengths whk.re k#o in the infrared. This is illustrated in Figure1, calculated from the Mie theory. In Figure 1, curves are shownfor the infrared wavelengths of 8.5, 10.5 and 12.57 (or 4r) Am, aswell as for a composite of wavelengths in the 8.5 - 12.57 Amregion where extinction can be measured by a simple broadbandtra'Ismissometer operating at 8-13 Am in the atmospheric windowregion. While operation at a specific wavelength such as the 10.6Am CO2 laser line for comparison to a visible wavelength such asthe He:Ne line at X= 0.63 Am has advantages, a broadband infraredreference band also can be used. In fact, related findings, usinga filtered light source in the visible at X= 0.515 gm and a 9-12Am broadband transmissometer to study the formation anddissipation of water fogs, were reported at least as early as1970. 3 1

When the techniques of Carlon et al.,22,24 are extendedto use the Mie theory to calculate functions of the transmittancesTX at = 0.63 gm and the broadband or composite wavelength band8.5-12.57 gm versus water droplet diameter DU, curves like thoseshown in Figure 2 are obtained. It is seen that the calculatedcurves are widely separated. Thus, a sensitive method is providedfor the determination of mean droplet diameter. Experimentalpoints for a steam-generated water fog are shown in Figure 2.These give excellent agreement with calculated curves, for exampleDA = 8 gm. This result recently was cited by Weinman et al., 3 2 asapproximating Deirmendjian's3 3 md'del C-i (or vice versa), where DAis taken as the mass median diameter. These results seemcredible. What is uncertain is the kind of distribution to whichthis diameter is most applicable. Kerker 3O has given aninformative discussion of particle size distributions, and oftransmission (extinction) and turbidity measurements. In thisreport, turbidity can be taken as the product aX.C in Equation 1,having the units m-1 .

Justification for the use of composite of wavelengths(Figure 1 and abscissa, Figure 2) can be found by examiningcalculated curves and steam generated fog data points for theindividual wavelengths X= 8.5, 10.5 and 12.57 (or 4 r) gm andplotting curves like Figure 2 as is done in Figure 3. Viewing theplots from top to bottom in Figure 3, it can be seen that the fogdata gave best agreement with calculation for a mass mediandiameter (MMD) of 6 jim at X= 8.5 gm, MMD = 10 gm at X= 10.5 gmand MMD = 6 Am at X = 4 Am. In his discussions of particle sizedistribution functions and transmission measurements, Kerker 30

cautions that mean particle diameters obtained using methods likethose discussed in the present paper may be adequate for mostpurposes if the distribution is sufficiently narrow, but strictlyare valid only for monodisperse systems. By using a composite ofwavelengths to obtain a mean diameter, one is averaging both overwavelength intervals and over particle size distributions. Theresults indicate that the technique works very well, especiallyunder dynamic droplet growth or evaporation conditions where what

1 11 o .,, . ' ..,..." " .% - ." '. ' .:. "-".-.

5

X 8.5 jum4 -/ COMPOSITE

X = 0.63 pm /

3 ("CONSTANT")

t5 X = 12.57 um

2 -- X = 10.5 jum

1

0 _0 4 8 12 16 20

DO

Figure 1. Values of aX DA vs. DAL Calculated from theMie Program for Water Using the Data of Haleand Querry.38 The Average Value of the'Constant' Curve Is Shown; Actually theFunction Is a Damped Oscillation.

12

100- 0.01

CL 580-

20 - 30

E 60- 0.1 ."

W 15

40 10 Dj 0.5- 1.0

0 EXPERIMENTAL208 DATA POINTS,

5 STEAM 30 - 350 C

00 20 40 60 80 100

% T 8.5 - 4 pm

Figure 2. Comparison of Computed Curves RelatingVisible to Infrared Optical Transmittanceof Water Fog to Experimental Data Points forCooling, Steam-Generated Water Fog Clouds;Note that Data Points Cluster Along the CurveDp = 8 Am in Good Agreement with GravimetricData from the Same Experiment.

13

V V %

100

-00

E '

o ~~./ - O o

0-

0 %T 05 m 100

O10 b C.

100

0-0 %T 10.Swmm 100

'.

X 4uj CL b

100- 0fl,

Figure 3. Curves for Three Infrared WavelengthsComprising the Broadband Composite ofWavelengths Shown as the Abscissa ofFigure 2.

14

is wanted is a real-time indication of approximate droplet sie,e.g., in studies of developing or dissipating water fogs. 3 1 ,

The highly variable dependence of aX upon X is due tovariations in (n - ik)X in addition to which aX varies with DA aswas discussed previously. This is illustrated in Mie calculationsfor the extinction coefficient of water droplets (aX , ordinate)in Figure 4 vs. Dg for several wavelengths. At certainwavelengths, e.g., X= 12.5 gm, the combined contributions ofrefractive (.77X) and absorptive (kX) components of water dropletextinction lead to values of x that virtually are independent ofdroplet size. Experiments seem to confirm that a 12.5 is nearlyconstant with Dg, suggesting that the liquid water content C inEquation 1 of a tropospheric optical path could be monitored by asimple 12.5 gm transmissometer with good precision for dropletsizes of up to about 15 Mm.24 ,3 5 In this application, however,care must be taken to account for absorption due to hydrogenbonding in molecular clusters (water clusters) in the vaporleading to absorption easily confused with droplet absorption. 2

10.0--7.0

E 3.0

1.5 - X =0.63 /z

O 0 - 1. 5

U-

012

Z 0 .1 6 .

2.10 .07 -10.0

.0 8.6X

II0 a, 0*

0.1 1.0 10.0 100.0

DIAMETER (pm)

Figure 4. Calcul ted Curves from the Data of Hale andQuerry 38 for Water Droplets Showing axvs. Dg; Note Constancy of a 12.5 at AllDiameters Up to Dg - 15 ,m.

15

V-%WW.R. 2, -% WI -T

At wavelengths called 'isosbestic points', 2 5 the complexindices (n - ik)X of some liquids are such that aX remains con-stant even though the liquid solutions comprising the dropletaerosol vary widely in their chemical composition, e.g., fordroplets of orthophosphoric and related acids that might be formedby burning phosphorus in air of varying relative humidity suchthat the solute (acid) concentrations vary from near zero to near85 by weight. In Figure 5, X = 11.4 Mm is an isosbestic pointcalculated from refractive index measurements of pure ortho-phosphoric acid (H3PO4 ) in water.36,37 Real atmospheric smokes,obtained by burning phosphorus, may contain mixtures of acidsdepending on the rates of combustion in the presence of varyinghumidities, and so may not give experimental spectra like thosefor pure H3 PO4 in Figurq 5 wherea 1 1 .4 remains essentiallyconstant at 0.13-0.15 m g- 1, while acid concentration varies fromnear zero to 85. by weight. This will be investigated. Becauseof the constancy ofa 11.4 compared to the wide range of values of,say a 9.7 (see Figure 5) with solute concentration, remotecharacterization of the aerosol becomes a possibility if somethingis known about the droplet constituents. It can be shown that:

A m 4nkkf(mx)/Xp (2)

where p(g cm- 3 ) is the droplet solution (acid) density i.e., massdensity, and:

9 n (3)f (mx) [ n + k 2 ) 2 + 4 (n 2 - k2) + 4

16

0.3

5 0.6-

9

'S w 0.40

U

x

10

0.0 -______

7 a 9 10 11 12X, WAVELENGTH Um)

Figure 5. Computed Values of aX for Liquid DropletAerosols Comprising Several Concentrationsof Orthophosphoric Acid (H3 PO 4 ) in Water,from Optical Constants of Querry and Tyler.

3 7

17

where m\ = (n - ik)x and the function of Equation 3 is plotted inFigure 6 for values of 9,\ marked on the curves and ranging from0.1 to 4.0, and for values for kX shown on the abscissa. It isemphasized that the above discussion applies to Rayleigh-scattering smokes where D, <<) as clearly is the case fororthophosphoric or other acid aerosols (smokes) produced byburning phosphorus in moist air.

Often it is useful to plot extinction coefficients orfractional transmittances versus Dp rather than versus the sizeparameter r DU/X, for specific wavelengths. Two such calculatedplots are shown for water droplets in Figure 7 and 8.38 In Figure7, the curves are calculated for a product C x L = 5.0 (Equation1) and for wavelengths, reading from top to bottom on the left-hand ends of the curves, of X = 0.63, 8.5, 10.5 and 4 ram (thesame wavelengths considered in Figures 1 - 3). In Figure 8, thecurves are calculated for X = 10 lm to show the functions vs. DLof total extinction, and of the scattered and absorbed componentsthat are summed to give total extinction, i.e.,

aT = a S = + aA (4)

where the subscripts indicate total (T), scattering (S) andabsorptive (A) extinction.

Sassen 23 recognizes that aerosol cloud extinctionmeasured in conjunction with angular scattering measurements couldlead to remote sensing of cloud composition by using two wave-lengths like X = 0.63 and 10.6 11m.73 But the present discussionindicates that such remote characterization could solely resultfrom cloud extinction measurements at two closely-spaced infraredwavelengths, one of them an isosbestic point. Additional datamight be obtained by combining extinction and angular scatteringmeasurements at an isosbestic wavelength. Optical parameters cancarry only so much information, but the idea is provocative enoughto warrant an investigation.

Other investigations for remote characterization oftropospheric aerosol clouds are possible, such as those based oneffects peculiar to Christiansen wavelengths. 3 9 A completediscussion is given by Carlon where the latter reference discussesaerosols ranging in size from molecular clusters to water clouddroplets. 2,39

3. POLARIZATION AND ANGULAR SCATTERING MEASUREMENTS

Kerker 30 has given an excellent review of techniques todetermine particle size distributions using these measurements andI will not elaborate except to raise a few pertinent points.Sassen23 has verified that angular scattering patterns calculatedfrom the Mie theory can be verified rather closely by experimentnot only for spherical particles but for irregularly-shaped onesas well. Thus polarization or angular scattering techniques

18

~~~ , ,, A AL

.3

5 .. 1

.01

0 1 -P2 2 3 4 5 6

k

Figure 6. Plot of Equation (3) for %Values of 0.1-4.0Shown on the Curves and kX Values Shownon the Abscissa; for Rayleigh-ScatteringAerosols 'Only' (DpL Ix).-

19

Vo -omzo \qp wK1L II-WX 1VKMWWW[1

0.90 . -\

0.7 .°........... ....... I . . . ,04 .. ......

0.6

20.3

I*I- I I,..

0 .404 % je*0.3 t I

0.0 ".11'9"

10-4

10-3 10-2 10'1 1 10 10

2

DIAMETER (pa)

Figure 7. Fractional Transmittance for a C x L Product= 5.0 (Equation 1) of Water Fog vs. DropletDiameter for Wavelengths, Reading from Topto Bottom of the Left-Hand Ends of the Curves,ofX = 0.63, 8.5, 10.5, and 41 gm, which Arethe Wavelengths also Considered in Figures1 - 3.

WATER, X = 10m

• Z 10-1

I.

LU0 cl 10-2

U 4

x J I I I imil p I I II i iil

1 101 102

DIAMETER (pm)

Figure 8. axvs. T) Calculated from Mie ProgramUsing Data of Hale and Querry 3 8 atX = 10 gm for Water Droplets; theSubscripts T, S, and A Refer to Equation(4) and Definitions.

20

combined with some of the recent advances in extinction measure-ment technology discussed here should provide new combinations forthe remote characterization of tropospheric aerosols.

The polarization method for evaluating the distributionof sphere sizes, utilizes polarization of scattered radiationobtained from a monochromatic light source at various angles ofobservation. 30 The intensity of the component of the scatteredlight, whose electric vector vibrates perpendicular and parallelto the plane of observation, 11(8) and 12(8), is measured at anumber of angles, where8 is the scattering angle measured from thedirection of the incident light beam. A polarization ratio, P(9),is defined such that:

P(9) = 12 (8)/ i i () (5)

For example, P(8) can be calculated as a function offor any assumed particle distribution such as the zeroth orderlogarthmic distribution (ZOLD) and for any standard deviation,00, ranging from 00 = 0 (i.e., a monodisperse aerosol), to valuesof 0.3 or more where the curves flatten and the method approachesits limit of usefulness. Figure 9 shows the family of curvescalculated for the He:Cd wavelength X = 0.4416 m, DM = 0.3 am,%X= 1.484 and k X = 0 so that m> = (1.484 - i.o)x and the densityis p = 0.98 g cm - . The curves are for dioctyl phthalate (DOP), aliquid commonly used for test aerosols.2 Agreement betweencalculation and measurement using the polarization method is goodto excellent, often within a few percent.

The effect of the wavelength of the monochromaticlight upon the function shown hn Figure 9 is of special interest.This has been investigated27,2 for visible and near infraredwavelength, including cases such as vanadium pentoxide (V205 ) forwhich strong absorption occurs at shorter visible wavelengths,i.e.,kX 0. 9 Special techniques using extinction measurements inthe infrared where kX* 0, suggest that combinations of visible andinfrared wavelengths can optimize desired parameters usingpolarization techniques. Since the size parameter r DM /Xdepends on wavelength, a change in the wavelength of the incidentlight will cause a shift in the position of the curve peaks inFigure 9, larger values of X shift the peaks to the right.Multi-wavelength observations have been used to show that particlesize distributions obtained at different wavelengths areconsistent, but when the wavelength spacing is as great as X= 0.63km and 10.5 pm, additional work is needed to confirm this and alsoto determine whether the useful range of 00 can be extended beyond0.3 in measurements using the polarization method.30

The scattering ratio method for the determination ofparticle size distribution is a variation where the angle ofobservation is held constant and the polarization ratio is

21

100

MONODISPERSE AEROSOL-,,.

00= 0.02

00 =0.05

S10:

2- 00 0.10 /

0

N 0030-4g

UN020d

.1~0 0.15 II

0 25 50 75 100 125 150 175

SCATTERING ANGLE (0)I

Figure 9. Polarization Ratio P(e) vs. Scattering Angle efor DOP.

22

measured or calculated over a spectrum or over a number ofdiscrete wavelengths.4 0 ,4 1 ,4 2 The method was perfected by usingmonodisperse, polystyrene, latex spheres and known combinations ofsingle sphere sizes to give mixtures of known size distributionschecked by electron microscopy. The scattering ratio method givesan alternative way of studying the variation of 00 at some D/i withX at some constant e. The same thing can be done by using thepolarization method directly and plotting, e.g., Figure 9 with Xas the abscissa at constant e.

Turbidimetric particle size distribution methodsinvariably are similar to those of Wallach et al. 4 0 Very recentlyNelson4 3 proposed a method by which calculated scattering crosssections from the Mie theory are compared to experimentalextinction data using monodispersed, polystyrene, latex spheres indistilled water as a calibration means. A computer program isused to null the error versus wavelength, forcing a unique solutionfor the size distribution which is taken as bimodal to account forcoagulation. The method also is said to yield particle volumeconcentration and the scattering cross section function, (Q)X.Extrapolation of the technique to tropospheric aerosols remains tobe demonstrated.

4. SUMMARY AND CONCLUSIONS

This report has concentrated on advances that have beenmade in the remote characterization of tropospheric aerosols byextinction and turbidimetric methods over the decade that haselapsed since the publication of Kerker's classic text. 3 0 Thetechnology has been updated, including results not previouslypublished, with improved techniques in polarization and angularscattering methods. Perhaps most importantly, the author hastried to support the argument that if all methods presentlyavailable in this technology are taken into account at this time,our ability to remotely characterize or quantitatively analyzeatmospheric aerosols is further advanced than workers in theindividual light measurement disciplines realize.

It is concluded that by combinations of well establishedand recently developed light scattering techniques, investigatorscan undertake promising new areas of research. Some examplesinclude combinations of extinction and polarization measurementswith computer programs to study isosbestic or Christiansenwavelengths, and optimization of desired parameters as functionsof wavelength or particle size distribution.

23

LITERATURE CITED

1. Mie, G., Ann. Phys. (Leipzig) Vol. 25, p 377

(1908).

2. Carlon, H. R. "Contribution of Particle Absorp-tion to Mass Extinction Coefficients (0.55-14,m) of Soil-DerivedAtmospheric Dusts: Erratum," p 1165; "Mass Extinction Coeffi-cients Estimated For Nonabsorbing Spherical Aerosol Particles inthe Geometric Scattering Regime," p 1891; "Christiansen Effect inIR Spectra of Soil-Derived Atmospheric Dusts: Addenda," p 1892;"Aerosol Spectrometry in the Infrared," p 2210, Appl. Opt. Vol.19, (1980).

3. Carlon, H. R., Kimball, D. V., and Wright, R. J.,"Laser Monitoring of Mass Concentrations of Monodisperse TestAerosols," Appl. Opt. Vol. 19, p 2366 (1980).

4. Bruce, C. W., Yee, Y. P., and Jennings, S. G.,"In Situ Measurement of the Ratio of Aerosol Absorption to Extinc-tion Coefficient," Appl. Opt. Vol. 19, p 1893 (1980).

5. Tam, W. G., and Zardecki, A., "Off-Axis Propagationof a Laser Bean in Low Visibility Weather Conditions," Appi. Opt.Vol. 19, p 2822 (1980).

6. Tam, W. G., "Multiple Scattering Corrections forAtmospheric Aerosol Extinction Measurements," Ap21. Opt. Vol. 19,p 2090 (1980).

7. Janzen, J., "Extinction of Light by Highly Non-spherical Strongly Absorbing Colloidal Particles: Spectrophoto-metric Determination of Volume Distributions for Carbon Blacks,"Appl. Opt. Vol. 19, p 2977 (1980).

8. Yue, G. K., and Deepak, A., "Modeling of Growth,Evaporation and Sedimentation Effects on Transmission of VisibleIR Laser Beams in Artificial Fogs," Appl. Opt. Vol. 19, p 3767(1980).

9. Ariessohn, P. C., Self, S. A., and Eustis, R. H.,"Two-wavelength Laser Transmissometer for Measurements of the MeanSize and Concentration of Coal Ash Droplets in Combustion Flows,"Appl. Opt. Vol. 19, p 3775 (1980).

10. Wyatt, P. J., "Some Chemical, Physical, and Optical

Properties of Fly Ash Particles," Appl. Opt. Vol. 19, p 975 (1980).

11. Kapustin, V. N., Rozenberg, G. V., Ahlquist, N. C.,Covert, D. S., Waggoner, A. P., and Charlson, R. J., "Characteriza-tion of Nonspherical Atmospheric Aerosol Particles With Electro-optical Nephelometry," Appl. Opt. Vol. 19, p 1345 (1980).

25

g

12. Weil, H., and Chu, C. M., "Scattering and Absorptionby Thin Flat Aerosols," Appl. Opt. Vol. 19, p 2066 (1980).

13. Horvath, H., "Determination of the Scattering Functionof Atmospheric Aerosols with a Telephotometer," Appl. Opt. Vol.19, p 2651 (1980).

14. Hansen, M. Z., and Evans, W. H., "Polar Nephelometerfor Atmospheric Particulate Studies," Appl. Opt. Vol. 19, p 3389(1980).

15. Hansen, M. Z., "Atmospheric Particulate AnalysisUsing Angular Light Scattering," Appl. Opt. Vol. 19, p 3441 (1980).

16. Gardner, C. S., Sechrist, C. F., Jr., andShelton, J. D., "Lidar Observations of the Mount St. Helens DustLayers Over Urbana, Illinois," Appl. Opt. Vol. 19, p A192 (1980).

17. Randhawa, J. S., and Van der Laan, J. E., "LidarObservations During Dusty Infrared Test-l," Appl. Opt. Vol. 19,p 2291 (1980).

18. Schuster, B. G., and Kyle, T. G., "Pollution PlumeTransport and Diffusion Studies Using Fluorescence Lidar," Appl.Opt. Vol. 19, p 2524 (1980).

19. Carswell, A. I., and Pal, S. R., "PolarizationAnisotrophy in Lidar Multiple Scattering From Clouds," Appl. Opt.Vol. 19, p 4123 (1980).

20. Walker, P. L., "Thermal Emissivity of aPolydisperse Aerosol Medium," Appl. Opt. Vol. 19, p 2271 (1980).

21. Jeck, R. K., "Calibration and Testing of OpticalSingle-Particle-Size Spectrometers with Monofilament Fibersas Substitute Particles," Appl. Opt. Vol. 19, p 657 (1980).

22. Carlon, H. R., Milham, M. E., and Frickel, R. H.,"Determination of Aerosol Droplet Size and Concentration FromSimple Transmittance Measurements," Appl. Opt. Vol. 15, p 2454(1976).

23. Sassen, K., "Infrared (10.6-um) Scattering and

Extinction in Laboratory Water and Ice Clouds," Appl. Opt.

Vol. 20, p 185 (1981).

24. Carlon, H. R., Anderson, D. H., Milham, M. E.,Tarnove, T. L., Frickel, R. H., and Sindoni, I., "InfraredExtinction Spectra of Some Common Liquid Aerosols, Appl. Opt.Vol. 16, p 1598 (1977).

25. Carlon, H. R., "Isosbestics in Infrared AerosolSpectra: Proposed Applications for Remote Sensing," InfraredPhys. Vol. 21, p 93 (1981).

26

26. Stratton, J. A., and Houghton, H. G., "ATheoretical Investigation of the Transmission of Light ThroughFog," Phys. Rev. Vol. 38, p 159 (1931).

27. Kerker, M., Matijevic, E., Espenscheid, W. F.,Farone, W. A., and Kitani, S., "Aerosol Studies by LightScattering. I. Particle Size Distribution by Polarization RatioMethod," J. Colloid Sci. Vol. 19, p 213 (1964).

28. Espensheid, W. F., Matijevic, E., and Kerker, M.,"Aerosol Studies by Light Scattering. III. Preparation andParticle Size Analysis of Sodium Chloride Aerosols of NarrowSize Distribution," J. Phys. Chem. Vol. 68, p 2831 (1964).

29. Jacobsen, R., and Kerker, M., "Optical Propertiesof Vanadium Pentoxide," J. Opt. Soc. Vol. 57, p 751 (1967).

30. Kerker, M., The Scattering of Light and OtherElectromagnetic Radiation, Academic Press, New York, 1969.

31. Carlon, H. R., "Infrared Emission by Fine WaterAerosols and Fogs," Appl. Opt. Vol. 9, p 2000 (1970).

32. Weinman, J. A., Harshvardhan, and Olsen, W.S.,"Infrared Radiation Emerging From Smoke Produced By Brush Fires,"Appl. Opt. Vol. 20, p 199 (1981).

33. Deirmendjian, D., Electromgnetic Scattering onSpherical Polydispersions, Elsvier, New York, 1969.

34. Carlon, H. R., and Shaffer, R. E., J. ColloidInterface Sci. Vol. 82, p 203, (1981).

35. Carlon, H. R., U.S. Patent 4, 154, 089, 1979.

36. Querry, M. R., Tyler, I. L., and Holland, W. E.,Bull. Am. Phys. Soc. Vol. 22, p 641 (1977).

37. Querry, M. R., and Tyler, I. L., J. Opt. Soc. Am.Vol. 68, p 1404 (1978).

38. Hale, G. M., and Querry, M. R., "Optical Constantsof Water in the 200-nm to 200-mu Wavelength Region," Appl. Opt.Vol. 12, p 555 (1973).

39. Carlon, H. R., "Christiansen Effect in II Spectraof Soil-Derived Atmospheric Dusts," Appl. Opt. Vol. 18, p 361n(1979).

40. Stevenson, A. F., Heller W., and Wallach, M. I. .,"Theoretical Investigation on the Light Scattering of ColloidalSpheres. XI. Determination of Size Curves From Spectra of thf,Scattering Ratio or From Depolarization Spectra," J. of Chem. Phys.Vol. 34, p 1789, p 1796 (1961).

27

41. Heller, W., and Wallach, M. L., "ExperimentalInvestigations on the Light Scattering of Colloidal Spheres.V. Determination of Size Distribution Curves by Means of Spectraof the Scattering Ratio," J. Phys. Chem. Vol. 67, p 2577 (1963).

42. Wallach, M. L., and Heller W., " ExperimentalInvestigations on the Light Scattering of Colloidal Spheres.VI. Determination of Size Distribution Curves by Means ofTurbidity Spectra," J. Phys. Chem. Vol. 68, p 924 (1964).

43. Nelson, H. F., "Radiative Scattering Cross Sections:Comparison of Experiment and Theory," Appl. Opt. Vol. 20, p 500(1981).

'.'

MblR1m

gI4fll0

& e~i ~v-V-

.w ~ ~ ~ w www www w