DEPOSITION OF AERIAL SUSPENSIONS OF PESTICIDES By William Eldon Splinter A.THESIS Submitted to the School of Graduate Studies of Michigan State College of Agriculture and Applied Science in partial gulfillment or the requirements for the degree of DOCTOR OF PHILOSOPHY Department of.Agricultural Engineering Year 1955 THESIS WILLIAM ELDON SPLINTER AN §BSTRACT Present-day methods of application of pesticides as dusts, sprays, or fogs, are very inefficient. The basic problem to which this work was devoted was the increase in efficiency of deposition and uniformity of coverage of plant surfaces by insecticides and fungicides for plant protection. The specific phases investigated were; the influence of particle sizes on the effectiveness of an insecticide or a fungicide, and the various forces af- fecting deposition and their relative importance in depo- sition. .A review of experimental results obtained by ento- mologists and plant physiologists was carried out and an analysis of the effect of particle size and application rate on effectiveness of control of insecticides and fungicides was correlated with these experimental findings. ‘A series of experiments were run under controlled laboratory conditions to quantify the effects of gravi- tational, inertial, and electrical forces on dusts of different density and particle size. The effects of these forces on dust deposition were measured at three relative humidities. The effects of inertial forces were investi- gated at four air velocities. An additional experiment iii WILLIAM ELDON SPLINTER AN.ABSTRACT was conducted to determine the magnitude of friction charging of dust. Field tests on plots of onions and celery were conducted to compare the effectiveness of plant disease control of commercial and micronized dusts at various rates per acre, both charged and uncharged. From the review of literature it was found that, in many cases, reduction in particle size of pesticidal dusts increases their effectiveness against insects and fungi. Equations were derived with which it is believed the effect of changes in application rate and particle size on control of plant diseases and insects can be calculated. The effects of the various forces on deposition of fine particles were compared on the basis that particle terminal velocity in any given force field could be used as a criterion of deposition rate. Analysis indicated that the rate of fall of particles below five microns in a gravitational force field would be so small that drifting in the air currents would be excessive. The rate of deposit due to inertial forces was found theoretically and experimentally to increase linearly with air velocity for laminar conditions of flow. For turbulent flow conditions, analysis indicated that increase in air iv WILLIAM ELDON SPLINTER AN ABSTRACT velocity would not give further increase in deposit at higher air velocities. Analysis indicated that inertial forces would be of little value for very small particles. Plant surfaces may be as high as 20° C. above ambient air temperature in direct sunlight. Thermal forces, though of small magnitude, may decrease deposit of very small particles through thermal repulsion. Frictional charging was found to increase with particle weight. Analysis indicated that deposition rate due to friction charging would increase with particle size. Charging of particles by an ionized field was found to increase with particle surface area. Deposition rate due to charging in an ionized field was found analytically and experimentally to be independent of particle size. Of the forces considered, only those due to electrical field forces caused by particles charged in an ionized field were found to be of potential importance toward effective deposition of fine particles. William.Eldon Splinter candidate for the degree of Doctor of Philosophy Final examination, May 9, 1955, 3:00 P;M., Room 218, [Agricultural Engineering Building Dissertation: Deposition of Aerial Suspensions of Pesticides Outline of Studies Major Subject: Agricultural Engineering Minor Subjects: Physics, Mathematics Biographical Items Born, November 24, 1925, North Platte, Nebraska Undergraduate Studies, University of Nebraska, 1945-50 Graduate Studies, Michigan State College, 1950-54 Experience: Graduate Fellow, Michigan State College, 1950-51; Instructor, research,‘Michigan State College, 1952; Graduate Research Assistant, Michigan State College, 1953; Instructor, Michigan State College, 1953-54, Research Associate Professor, North Carolina State College, 1954. Member of Sigma Tau, Pi Mu Epsilon, Sigma Pi Sigma, Society of the Sigma Xi, American Society of Agri- cultural Engineers, and Society of Automotive Engineers. vi TABLE OF CONTENTS INTRODUCTION 0 O O O O O O I O O O O 0 O 0 History of the Project. . . . . . . . The General Problem . . . . . . . . . The Specific Problem. . . . . . . . . INFLUENCE OF PARTICLE SIZE ON EFFECTIVENESS OF PEST IC IDES O O o O Q 0 0 O O I O 0 O O O O Insecticides. . . . . . . . . . . . . Fungicides. . . . . . . . . . . . . . The Effect of Particle Size on Required Dosage per Acre. . . . . . . . . . . Where Control is Proportional to Surface Area of Pesticide . . . Where Control is Proportional to of Particles per‘Unit Leaf Area Where Control is Proportional to of Pesticide per Unit Leaf Area Generalized Theory for Fungicides . . Summary . . . . . . . . . . . . . . . PHYSICAL FACTORS INFLUENCING DEPOSITION OF SUSPENSIONS . . . . . . . . . . . . . . . Viscosity . . . . . . . . . . . . . . Brownian Motion . . . . . . . . . . . l3 l5 l7 18 21 22 26 29 29 31 vii Page Coagulation . . . . . . . . . . . . . . . . . 32 Gravitational Forces. . . . . . . . . . . . . 33 Inertial Forces . . . . . . . . . . . . . . . 35 Electrical Forces . . . . . . . . . . . . . . 42 Terminal Velocity of Charged Particles in Electric Field . . . . . . . . . . . 48 Calculation of Terminal Velocity of Charged Particle Under Electrical Field Forces. . . . . . . . . . . . . . 50 Thermal Forces. . . . . . . . . . . . . . . . 52 Summary . . . . . . . . . . . . . . . . . . . 57 EXPERIMENTAL WORK. . . . . . . . . . . . . . . . . 60 Laboratory Experiments. . . . . . . . . . . . 60 Laboratory Dust Chambers . . . . . . . . 60 Metering of Dust . . . . . . . . . . . . 65 Auxiliary Fan for Inertial Tests . . . . 65 Measurement and Control of Relative Humidity. . . . . . . . . . . . . . . . 66 Measurement of Dust -.Air Density. . . . 68 Measurement of Cloud Potential . . . . . 68 Gravitational Deposit. . . . . . . . . . 72 Inertial Deposit . . . . . . . . . . . . 72 Electrical Deposit . . . . . . . . . . . 74 Dust Particle Sizes. . . . . . . . . . . 76 Friction Charging. . . . . . . . . . . . 78 viii Page Experimental Design. . . . . . . . . . . 78 Field Tests . . . . . . . . . . . . . . . . . 80 Experimental Dust HOpper . . . . . . . . 80 Experimental Design. . . . . . . . . . . 83 EXPERIMENTAL RESULTS . . . . . . . . . . . . . . . 85 Gravitational Deposits. . . . . . . . . . . . 85 Inertial Deposits . . . . . . . . . . . . . . 88 Electrical Deposits . . . . . . . . . . . . . 94 Friction Charging of Dust . . . . . . . . . . 103 Field Experiments . . . . . . . . . . . . . . 107 INTERPRETATION OF RESULTS. . . . . . . . . . . . . 113 Gravitational Deposits. . . . . . . . . . . . 113 Inertial Deposits . . . . . . . . . . . . . . 117 Filter Deposit. . . . . . . . . . . . . . . . 118 Cloud Potentials Due to Charging. . . . . . . 118 Sphere Deposit. . . . . . . . . . . . . . . . 119 Friction Charging of Dusts. . . . . . . . . . 120 Charging of Dust in the Field . . . . . . . . 121 Comparison of Terminal Velocities of the Particle's Field Forces. . . . . . . . . . . 121 CONCLUSIONS. . . . . . . . . . . . . . . . . . . . 125 APPENDIX I - Calculation of . . . . . . . . . . 130 APPENDIX II - Derivation of Potential Distribution in Charged Cloud Between Two Parallel Plates. . . . . . . . . . . 133 ix Page APPENDIX III - Experimental Results. . . . . . . . 137 APPENDIX IV - Calculation of Charge Per Particle . 157 LITERATURE: C ITED O O O O O O O 0 O O O O O O O O O 161 Table II III .4 VII VIII IX XII LIST OF TABLES Particle Size Effect of Paris Green Dust on Mexican Bean Beetles. . . . . . . . . Relationship of Particle Surface Area to Effective Area of Control for Particles of Phygon. . . . . . . . . . . . . . . . Values of Reynolds' Number and Terminal Velocity Equations for Particles . . . . Average Diameters and Densities of Dusts Analysis of Variance of Gravitational Deposit on Horizontal Aluminum.Disks . . Analysis of Variance of Inertial DBPOSitS O O O O O O O O O O O O O O O 0 Average Filter Deposits for Gravitational and Inertial Tests . . . . . . . . . . . Analysis of Variance of Sphere Deposit Due to Electrical Forces . . . . . . . . Analysis of Variance of Cloud Potentials Analysis of Variance of Millipore Filter Deposit . . . . . . . . . . . . . Analysis of Variance of Charge Due to Friction . . . . . . . . . . . . . . . . Analysis of Variance, Charged versus Uncharged Dusts. . . . . . . . . . . . . Page 11 25 36 77 85 9O 94 95 101 103 105 109 Table XIII XVI XVII XVIII XIX XXII XXIII XXVI xi Summary of Results . . . . . . . . . . . Terminal Velocity of Particles in Force Field. . . . . . . . . . . . . . . . . . Dust Particle Size Analysis. . . . . . . Gravitational Deposit on Aluminum Disk . Inertial Deposit on Aluminum Disk. . . . Filter Deposits for Inertial Tests . . . Electrical Deposit Cloud Potential . . . Average Potential One Inch from Grounded Sphere . . . . . . . . . . . . . . . . . Filter Deposit for Charged Tests . . . . Electrical Deposit on Aluminum Spheres . Potential of Dust Cloud Charged by Friction . . . . . . . . . . . . . . . . Average Potential One Inch From Grounded Screen - Friction Charging . . . . . . . Average Yields of Celery . . . . . . . . Average Yields of Onions . . . . . . . . Page 114 123 138 139 141 143 145 149 150 151 152 154 155 156 Figures 1 \OCDNJO“ 11 12 xii LIST OF FIGURES Calculated and Observed Control of Alternaria Solani by dichloronapthoquinone. . . . . . . Terminal Velocity of Particles of Unit Density in Gravitational Field . . . . . . . Terminal Velocity of Particles in Centripetal Force Field for a Particle of Density 2, Traveling at 450 Centimeters per Second Through a Radius of Curvature of Five Centimeters . . . . . . . . . . . . . . Inertial Deposit of Cu804.5H20 on 7.08 Square Inches ovarea for Different Surfaces Held Perpendicular to the Air Blast . . . . . . . Temperature Gradient from Surface of Hot Vertical Plate as Found by Kennard (15). . . Dust Feed Chamber. . . . . . . . . . . . . . Dusting Chamber and Experimental Set-up. . . Interior of the Dusting Chamber. . . . . . . wet and Dry Bulb Temperature Device. . . . . Millipore Filter for Determining Dust Concentration. . . . . . . . . . . . . . . . Interior of Filter Holder. . . . . . . . . . ‘Aluminum.Disk and Sphere . . . . . . . . . . Page 27 34 39 43 56 61 63 64 67 69 70 73 Figures 13 14 15 16 17 18 19 2O 21 22 23 24 xiii Wooden Stand for Aluminum Spheres. . . . . . Experimental Dust Hopper . . . . . . . . . . Field Plots. . . . . . . . . . . . . . . . . Inertial Deposit of Five Dusts at Three Relative Humidities. . . . . . . . . . . . . [Average Gravitational Deposits of Dusts on an.A1uminum Disks under Conditions of Stirred Settling. . . . . . . . . . . . . . Average Inertial Deposits on Aluminum Disks. Inertial Deposit of Dust at Four.Air Velocities . . . . . . . . . . . . . . . . . The Effect of.Air Velocity on Inertial Deposit, Corrected for Incident Dust Quantity . . . . . . . . . . . . . . . . . . Average Deposits on One Inch Diameter Grounded Spheres at Center of Spherical Dust Cloud . . . . . . . . . . . . . . . . . The Effect of Relative Humidity on Deposition of Dust in Charged Cloud. . . . . Average Potential One Inch from Grounded Wire Sphere. . . . . . . . . . . . . . . . . The Effect of Relative Humidity on Cloud Potential One Inch from the Grounded Wire Sphere O O O O O O O O O O O O O O O O O O O Page 75 82 84 86 87 89 91 92 96 97 99 100 Figures 25 26 27 28 29 3O 31 xiv Page Average Filter Deposits of Dust for Electrical Deposition Tests. . . . . . . . . 102 Variation of Average Charge per Particle with Mean Particle Surface.Area for Six Dusts 104 Average Potential at Center of Charged Spherical Cloud Due to Friction Charging of Dust. . . . . . . . . . . . . . . . . . . 106 Charge per Particle Due to Friction Charging of Dust . . . . . . . . . . . . . . 108 Deposits Due to Charged and Uncharged Dusts on Onions. . . . . . . . . . . . . . . . . . 111 Dust Deposit on Celery . . . . . . . . . . . 112 Potential and Charge Distribution Along the Axis of a Charged Cylindrical Cloud Between Grounded Disks . . . . . . . . . . . . . . . 136 ACKNOWLEDGMENTS The work reported herein was possible only through the assistance and cooperation of many individuals. Professor A. W. Farrall, head, Agricultural Engineering Department, provided the research assistantship which was granted by the Rackham Foundation. Many hours of guidance were given by Dr. W; M. Carleton, who served as major professor for this work. Considerable assistance in the construction of equipment and taking of data was received from.Dr. H. D. Bowen, Mr. R. D. Brazee, Mr. R. W. Brittain and Mr. N. T. Ban. The Attasorb dust was furnished by the.Attapulgus Minerals and Chemicals Corporation. The field experiments were carried out with the c00per- ation of Mrs. M. Mooar and Mr. E. Lundberg of the Botany Department. Advice on the statistical analysis of the experimental results was given by Dr. R. J. Mbnroe, and aid in setting up the probability analysis was given by Dr. A. Grandage of the Institute of Statistics, North Carolina State College. Excellent facilities for the writing of the thesis were provided by Mrs. K. A. Edsall, D. H. Hill Library, North Carolina State College. xvi The author is grateful for this opportunity to express his sincere thanks to these individuals and to the many others who made this work possible. DEPOSITION OF AERIAL SUSPENSIONS OF PESTICIDES INTRODUCTION It is estimated by Shepard (27) that approximately $4,000,000,000 in damage is done each year by insects. He also estimates that 500,000,000 pounds of active in- gradients and 750,000,000 pounds of diluents such as lime, tale, and clays are used annually as insecticides to combat this attack and to prevent more extensive damage. The damage due to attack by plant diseases is estimated to be of a comparable magnitude. The method of application of these insecticides is principally through dusting and spraying, although other means such as dipping, poison bait, soil treatments, and seed treatments are sometimes used. In dusting, spraying, and fogging, the chemically active material is applied as a suspension of either solid particles or fine droplets in air. The individual parti- cles range down to one-half micron in diameter, approaching the colloidal range. The application of these materials as dusts, sprays, or fogs is quite inefficient. Bowen (3) estimates that only ten to twenty per cent of the material applied as dust actually is deposited on the plant surface. Sprays are somewhat more efficient, but f0gs are probably even less efficient. It is apparent, therefore, that any increase in efficiency of deposition of these materials would be of considerable economic importance. History of the Project Work was initiated by Bowen (3) in 1950 toward the increase of dust deposition on plants by utilization of electrostatic field forces. These field forces were produced by passing the dust through a charging nozzle, charging the dust to one sign. Bowen, Brazee, Hebblethwaite, Brittain, Ban, and the author have worked OOOperatively on various phases of the problem of in- creasing deposition of dusts and fogs and on evaluation techniques. The results of this work are as follows: Bowen (3) discusses the mechanism of particle charging in an ionized atmosphere, the effect of relative humidity on charging, the precipitation of the charged particles, discharge of charged particles in the atmos- phere, and the deposition of the dust on plant leaves under experhmental conditions. His results indicate the following: 1. Particles of dust can be charged either positive or negative by passing the dust-air suspension through an ionized electric field. 2. Charged particles will lose their charge in the atmosphere by collision with ions at a very slow rate. Experimental results showed the charge to remain for a distance of at least thirty-two feet in a cloud moving at two miles per hour in a cloud chamber (this would be approximately forty—four seconds), and indications were that the charge would remain even longer. An average in- crease in deposit of 400 per cent was noted. 3. There was a decrease in deposit at high relative humidities. 4. Deposit on bean plants showed a marked increase with charged dust. Fringing of the leaves (heavier deposit along the edges) was noted when the electric field due to space charges was supplemented by a field from shields held at high potential. Hebblethwaite (12) discusses construction of labo- ratory and field apparatus for charging fogs and dusts, the influence of charging current, relative humidity and temperature on dust deposition, field tests with electro- static dusting and the charging of fogs and smokes. His results were as follows: 1. There appeared to be a maximum.charging current for the cylindrical type nozzle beyond which there was a decrease in deposit. 2. This optimum.current appeared to be higher for smaller sized particles. 3. There appeared to be a decrease in deposit of charged dust with increase in relative humidity although the uncharged deposit remained fairly constant. The electrical deposit was greater than the uncharged deposit even at high humidities. 4. There appeared to be a very slight increase in deposit of both charged and uncharged dust with increase in temperature. 5. Although biological evaluations were inconclusive, there was a definite increase in deposit with charged dust in the field. 6. Limited tests on charging fogs and smokes indi- cated a considerable increase in deposit with charging. Brazee (5) discusses the colorimetric, light re- flection, microscopic, vapor state fluorescence, and volumetric titration methods of evaluation of dust deposit and the problem of reverse ionization for dusts of high resistivity. His results were as follows: 1. The lead content analysis of dust deposition can be quite accurate and the leaf printing technique for in- vestigation of coverage can be very reliable. 2. Light reflection and microscOpic examination are not flexible enough for extensive field usage. 3. The volumetric titration method of analysis appears to offer considerable promise as a method of field analysis of dust deposits. 4. The problem of reverse ionization of dusts of high resistivity can be very serious in the charging of the dust in tubular nozzles, especially at lower relative humidities. Bowen (4), discussed electric fields in charged dust clouds, inertial forces and gravitational forces. His results were: 1. Theoretical and measured potentials in a spherical charged cloud were very close. Therefore the conclusion can be drawn that electrical forces can be accurately calculated for any charged cloud configuration, provided the mathematics are available. 2. The hypothesis is advanced that increased turbu- lence in the dust cloud would, by decreasing the apparent viscosity of the air, increase inertial deposits on the plant surface. 3. In comparison of electrical, gravitational, and inertial forces, his calculations indicate at present levels of charging that inertial force is the most im- portant, followed by gravitational force and electrical forces. 4. Methods of analysis of electric and inertial forces are given. Brittain (6) discusses the experimental laboratory apparatus for running controlled experiments, the use of artificial leaf surfaces, methods of evaluation of leaf surface deposit, and results of experiments on actual leaf surfaces. His results indicate: 1. There was a considerable difference in inertial deposit for different artificial surfaces. 2. Refinements on the volumetric titration method of evaluation were develOped. 3. The use of the Polarograph as a means of evaluation indicates considerable promise as a means of determining leaf deposit at very low concentrations. 4. Tests of inertial deposit on bean, tomato, and lettuce leaves showed an increase in deposit on the lower surfaces of the leaves and an increase in deposit with increase in air velocity. The General Problem The basic problem to which this and the previous works have been devoted is the increase in efficiency of depo- sition and uniformity of coverage of plant surfaces by insecticides and fungicides as protection against attack. This problem is very complex because of the many factors introduced by virtue of its being concerned with biological systems. Not only is the process of particle deposition complicated because of the difficulty of de- fining, separating, and analyzing the various forces acting upon these small particles, but the problem of evaluation has evolved as one of the major obstacles in this study. The general attack to the problem.has been a theo- retical approach to the deposition process coupled with the development of evaluation techniques to test the hypotheses advanced and quantify the results. Of the various methods of evaluation investigated, only the volumetric titration and polarograph methods ‘ have exhibited the accuracy and flexibility needed for determination of deposits on plant surfaces. The Specific Problem The specific phases of this overall problem.of parti- cle deposition which are dealt with are, first, a con- sideration of the influence of particle sizes on their effectiveness as an insecticide or a fungicide; second, a theoretical analysis of the various forces affecting deposition and a comparison of their’magnitudes; and third, the results of laboratory and field work. INFLUENCE OF PARTICLE SIZE 0N EFFECTIVENESS 0F PESTICIDES Insecticides Since the deposition of insecticidal and fungicidal suspensions is concerned with biological systems, the agricultural engineer cannot limit the problem to simply a determination of the magnitudes and directions of the various physical forces acting upon these particles. In addition to the problem of increasing the efficiency of deposition upon plant surfaces there are the equally important problems of distribution of the deposited chemical over the leaf surfaces, both upper and lower, throughout the whole plant and the relative chemical effectiveness of the deposited particles. The following section deals with the effect of particle size on the effectiveness of pesticides. MOst insecticides which are applied as dusts are sold as 200 mesh or 325 mesh, the 200 mesh being pre- dominant. For a 200 mesh United States standard screen, the maximum.particle size is 74 microns. For a 325 mesh screen the maximum size is 44.microns. Therefore a 200 mesh dust means that the particle sizes are 74 microns, or less, the lower range for particle size being for the most part about one-half micron in diameter. For a 200 mesh dust, the number of these fine particles is relatively low, while for micronized dust, most of the particles will lie within the one-half to three micron range. From.a study of physical chemistry, colloids in par- ticular, it is shown that the surface activity of particles increases markedly with decrease in particle size. This is easily understood if one considers the division of a one inch cube. To begin with, it has six square inches of surface area. If cut into one-half micron cubes, it will have 184,000 square inches of surface area. Therefore if the surface area of a particle has any influence on its effectiveness as an insecticide or fungicide, the effect would be magnified with reduction in particle size. The major plant injuries caused by insects are those due to chewing or biting where a stomach poison is used, or piercing and sucking, where a contact poison is used. In either case, control is a matter of exposing the insect to a lethal dose of poison before too much damage can be done. For the insects which damage the plant by eating the leaves, there is probably little difference whether the poison is on the top or under side of the leaf and the uniformity of the microscopic distribution of the dust is probably not critical. For light infestation the dosage -10- could be quite low, while for heavy infestation the dosage would need to be increased to the point where the insects will eat a lethal dose before they consume large amounts of leaf area. For piercing and sucking type insects, the dosage per unit area of the insect would be the minimum application for control. For certain insects such as aphids, the distribution beneath the leaf would be of importance. For residual control of those insects which crawl about, the dust particles must adhere to the insect better than to the plant. McGovran et a1 (19) found a definite particle size effect with Paris Green used as a stomach poison on Mexican bean beetles. Their results are shown in Table I. Not only are the smaller particle sizes more efficient in killing, as shown by the percentage mortality after forty- eight hours, but the insect ingests far less leaf area before succumbing. In reducing the particle diameter from 22 microns to 1.1 microns the percentage of kill was doubled and yet only one-fifteenth of the leaf area was consumed. Furthermore the weight of the insecticide needed for kill was reduced by a factor of one-thirteenth. When Paris Green was applied as a spray the particle size effect showed a similar trend. Ho. OO«H " Inomog ma 0 ¢.Nw o Moose ma «H.0m Heme w.4< woa ease mm mm. «m m. m3 HA L Ho hm. wmm m.N med NH 1 . me we. see m.e odd mm ANS“: “many AR“ mason w< odaaapoomQH oeaaapoomnH .mE st< A EO\mSV Adv exam genus hpfiaspnoz mo comm ooomssm “won pammmon «seq oaoapnsm mpcomoH Np oopufi A a®HV G$H>OUOE SOL—uh; mmHBHHm Zdflm Z L 0 _ Q\ «MN» Quhxkkukhmu 35.. 2mg}? x .. ow Segu- ‘A tQ eEEMZ-tMA-«N -30 MEG-530 .ov loin- qzoto 810$ um. $5.. «cm. Erwékuuut uo cots-(ob . .em Sex-mum mo 62... emu-‘80 #6 x sea H «Abba-lea «5 row «3% O r cu m. o «v .on I» a 4-. .. 1F . on 0 6» X o n3.~‘ n) n e veal/veg J/vzzg as'd -28.. The mechanism.of action of fungicides may be either through the vapor phase, where control will be pr0portiona1 to the surface area of fungicide per unit leaf area, or possibly through a spreading of the chemical either along the plant surface or systemically through the plant tissue, in which case control would be proportional to the square of the radius plus some effective distance from the parti- cle. Since the spores are immobile on the plant surface, the physical distribution of the particles on the surface must also be taken into account. The equation P s l-(l-fiE")n has been developed from theoretical considerations. Upon the determination of and the mechanism of action for a given fungicide the effects of variation of particle size and application rate can be predicted. Good agreement was found between theo- retical and experimental results for dichloronaptho- quinone. PHYSICAL FACTORS INFLUENCING DEPOSITION OF AERIAL SUSPENSIONS .Aerial suspensions are essentially unstable. Where certain colloidal suspensions can be quite stable because of the electric charge (zeta potential) or hydration, aerosols will eventually settle out. The basic problem in pesticidal work is to control the deposition of the fine particles such that (l) deposit is effected within a short time interval, thereby preventing excessive drifting, (2) coverage of plant surfaces is uniform throughout the plant and (3) the lower surfaces of the leaves are protected against the attack of fungi and certain insects. The physical factors influencing this deposition include viscosity, Brownian motion, and coagulation. The physical forces with which the particle suspension is concerned include gravitational, inertial, electrical, and thermal forces. The following discussion will deal with the manner in which these factors influence the deposition of aerial suspensions. Viscosity The effect of viscosity is essentially that of a parasitic force resisting all movement of the dust or -30- spray particles through the air. For any applied force the terminal velocity of small particles moving through air is given by Stokes' law: 2 v:2/9£‘7§P—§‘ --------- 8 Where v : the terminal velocity (cm./sec.) H II the particle radius (cm.) particle density (gr./ cm3) ‘b n acceleration (cm./sec.2) and m II coeficient of viscosity of air (gm./cm.sec.) “n (182.7,u poises at 180 C.(64O F.) to 195.8}1 poises at 400 C. (1040 F.) This equation has been found to be correct within five per cent for spherical particles between one and 50 microns radius falling in air. Particles less than one micron in radius will slip through the air molecules. Therefore a correction of -.04 micron must be subtracted from the Stokes radius for particles between .1 and .2 microns. Below .1 micron the rate of fall is so slow that the velocity is almost impossible to detect experimentally. Millikan has shown that Stokes' law also holds fairly well for non-spherical particles. For particles above 50 microns the turbulence created by their fall through the air tends to decrease the expected velocity. -31- ‘We note from Stokes' law that the terminal velocity for any particle in a given force field will increase with the square of the radius. Therefore any decrease in particle size will increase considerably the problem.of obtaining deposition within a short time interval. Brownian.Motion The random.movement of small particles, due to col- lisions with air molecules, causes collisions with other particles or droplets in the aerial suspension. Sinclair (28) states that indirect experiments indicate all dust particles adhere upon collision and small liquid particles coalesce. Brownian.motion will cause diffusion of the parti- cules due to concentration gradient or osmotic pressure potential. The average displacement of a spherical particle of radius r in time t is given by the Einstein liRTt I '-' 3Nr'lr """"" 9 ‘Where x : the average displacement equation as: = the gas constant (.082 liter atm./mole degree) = the absolute temperature (OK) the time (sec.) : the coefficient of viscosity H 43 (a #3 :n e = the particle radius (cmt) and Avogadro's number (6.02 x 1023 molecules/ 2 0| mole) For a particle one micron in diameter at a temperature of 2910K, x will be 5.25 x 10"3 cm. in one minute. There- fore the diffusion effect of Brownian motion may be neg- lected for all practical purposes. The main contribution of Brownian motion will be through coagulation. Coagulation Coagulation or coalescence of small particles will tend to increase fall out. Large clumps of particles would certainly cause a considerable decrease in effective- ness of fungicides and insecticides. Therefore coagu- 1ation is in general undesirable in pesticidal dusts. If the liquid medium Spreads readily over the plant surface coalescence of liquid particles would not be too critical from the standpoint of particle distribution on the leaf surface and the larger particles would deposit out faster. If the angle of liquid contact is large how- ever coalescence would result in poor distribution of control agent. According to Sinclair (28), for uniform particle sizes, the process of coagulation may be described by the differential equation: 2 29.2 g Air—ILL = an ------- 10 dt 39; Where n - the number of particles per cm. 3 t the time in seconds -33- K‘ the specific reaction rate constant and H a constant 3.0 x 10“10 chB/sec. at 2930K. 2 The rate of coagulation is independent of particle the coefficient of viscosity size. Under exceptional conditions of dusting where the application rate was fifty pounds of dust per acre of two mdcron diameter particles having a specific gravity of two, applied from four, one and one-half inch diameter nozzles at an air velocity of 60 miles per hour, at a speed of three miles per hour through the field covering a swath of six feet, the density of the dust cloud leaving the nozzle would be approximately 10 x 104 particles per cubic centi- meter. Discounting any diffusion or mixing with the air there would still be 9 x 104 particles per cubic centi- meter at the end of an hour. Since most values of n will be considerably below this, coagulation would not be of practical importance in dusting or spraying. The number of particles per cubic centimeter in commercial pesticidal fogs may be suf- ficiently high to warrant consideration of coagulation. Gravitational Force The force of gravity must be added vectorially to all other forces subsequently considered. Figure (2) shows the increase in rate of fall with increase of -34- .o oma an op goajmms we; epspmnoasop pw< .caowm HsCOHpmpfipmnm CH SpamCme pfls: no madcapewa mo Suwooac> amsfiunoe finesse-2» msxo-‘Q Heels-Rx MN on .m\ 6.x m. h b L J. C commas O _ 0 .V (veg/wag) TTVJ JO lea t O\ -35- particle radius. Calculations were based on a particle of unit density and air temperature of 18° C. (64° F.). The rate of fall of any particle of known radius and density may be determined by multiplying the rate of fall for that particle size obtained from the graph by the specific gravity. For a particle of sulfur five microns in radius (density : 2 gm./cm..3 ) we see that the rate of fall will be .6 cm./ sec. If a four mile per hour wind is blowing the horizontal component of travel will be 179 cm./sec. (about two yards.). Therefore if a particle of dust is initially one foot above the level of a plant surface it will be carried approximately 100 yards before settling to plant level, neglecting air turbulence. If dusting were to rely solely upon gravitational force, particles below five microns radius would be so subject to drift that they would be of little practical use in the field. Inertial Forces Particles carried in an air stream.tend to maintain their original direction of travel upon deflection of the air stream. The viscosity of the air resists this move- ment thereby subjecting the particle to a centripetal force. Where R is the radius of curvature of the path of the particle and V is the velocity of the air stream, the acceleration a to which the particle is subjected is given 2 bY: 82%" ---------- ll -36- The terminal velocity of the particle perpendicular to the air flow may be found from Stokes' law for small values of a. As v increases the particle will go through an inter- mediate region and then into turbulent motion. The regions of streamline, intermediate and turbulent motion are DI? v ‘7 diameter of the particle, F, the density of the air, v the defined by Reynolds' number, R = where D is the velocity, and 7 the coefficient of viscosity for air. The regions, values of Reynolds' number, and the equation describing the terminal velocity are shown in Table III as obtained from Dallavalle (9) and Sinclair (28). TABLE III VALUES OF REYNOLDS' NUMBER.AND TERMINAL VELOCITY EQUATIONS FOR PARTICLES IMotion Reynolds' Equation Number Zr' Streamline 10’4 37.5 the terminal velocity will be governed by Newton's law for bodies in turbulent motion. Values for flag—)2 (#1)} r were not given but if it can be assumed that at R = 2, v1 = v2 and at R = 500, v2 : v3 then a straight line relationship between these two values might give some insight into the behavior of the particles in the intermediate region. Bowen (4) has indicated that the air velocities in the major part of the plant region is 10 miles per hour or less. For an air velocity of 10 miles per hour (447 cm./ sec.) perpendicular to the leaf surface 10 cm. across, an assumed radius of curvature of five centimeters for the particle path and particle density of two, the particle size for which rv : .15 will be: Combining equations 11 and 12 2 mingcg ---------- 16 1.3 _(-15 <3II12‘LSecz.)(9)(182.7x10"6 cm.sec.)(5cm.) - 2(2gm/cm3)(447 cm./sec.)2 : 1.54 x 10’9 Cm.3 Therefore r : 11.5 x 10-4 cm. : 11.5 )1 Then -38- Therefore any particle less than 11.5 microns radius will move according to Stokes' law. Particles of radius greater than 11.5 microns will move in intermediate motion. Assuming an air velocity of 120 miles per hour (5370 cmt/sec.) moving perpendicular to a surface and for R : l and f’= 2 then for turbulent motion, combining equations 11 and 14, Therefore r ' 8 2 cm cm. sec. = 11.1 x 10"9 cm) Giving r : 2.2 x 10‘3 cm. : 22‘p Therefore turbulent motion would be found only for large particles at very high velocities. We may conclude that the major part of inertial deposit will be under streamline flow conditions and that the larger particles will deposit under conditions of intermediate motion at higher velocities. Figure (3) shows the variation in terminal velocity with particle size for a dust of density )9 Z 2 traveling along a circu- lar streamline around a leaf 10 cm. across (R = 5cm.) at an air velocity of 450 cmt/sec. (approximately 10 miles per hour). .Air temperature is assumed to be 18° C. For rv,>:.15 a linear relationship is assumed according with v2 in Table III and the assumption that at rv : .15, v1 . v2 and at rv = 37.5, v2 = v3. -39- .o oma so On Seasons m“ mossesoaSeu ofi< .sofipoa epeflecspepca :H msfi>os on ou Becomes one weaves mcopows m.HH seep acumenw modoapusa use ceases nesflaoa ca xcfi>os on Op Seasons one madden esopOHS m.da Scamp meHOHpneA .30 n no cusps>use mo uneven e ewsosno oom\Eo One pm mcaac>aup .m u Swansea no macapsea a you oaeflu ooeom Hmumgfispcoo CH moaOHpueQ ho zpfiooao> HecHSseB .m cssmflm Geese-kc “53 (Q New: Q VO- efi .mw ww m\ o\ H o u b o r 3.. 60\ r Qh< r8”- rgh {MM (veg/mu?) 4.1/70va 7VN/Wb’31 -40- 2 Upon the substitution of a - into the Stokes ”(4 equation we find, for laminar flow, v 2 rZFV2 =9 7R" Then for any given set of values for r,,o v? and R, assuming streamline motion, the terminal velocity of the particle will vary with the square of the air velocity. On the other hand, since the angular velocity about R will increase proportionally with V, the time allowable for deposition will be decreased in direct proportion to any increase in V. Therefore the increase in deposition through increase in air velocity should be linear. This should be true for a constant R (circular particle path) or for a varying R (as is actually the case in fluid flow about an object.). Therefore, for all other conditions being equal, D : KV where D is the inertial deposit on a surface and K’is a constant depending on particle size, density, surface configuration, erosion of dust from the surface and other factors. The deviation of a particle path from a circular streamline is discussed by Bowen (4). The variation of v with change in V for intermediate motion is not known. For turbulent motion of a particle, from equations ,_ 8rPV2 _ 8"5 " '7 ‘37s - V ma ------ 18 11 and 14 Then for turbulent motion, since the time for depo- sition varies inversely with V, the rate of deposit on a plant surface should be independent of air velocity. Therefore increase in air velocity in the turbulent region would not increase deposit. Brittain (6) investigated the effect of velocity on inertial deposits on various surfaces. Both artificial and plant surface were used. Two hydrated c0pper sulfate dusts were used, one of which was micronized. The mean particle diameter as calculated from his particle size distributions were 5.6 and 1.76 microns respectively. Brittain's results with the dust of larger particle size showed a linear decrease in deposit with increased air velocity. This is interpreted to indicate that the Reynolds number for the larger particles has exceeded two and that the particles are in turbulent motion, and, in addition, the eroding effect of the larger particles was evidently greater than the force of adhesion of the particles on the surface. The micronized dust exhibited a linear increase with velocity on both the artificial surfaces and plant surfaces. This is interpreted as indicating that deposition was taking place under laminar conditions for these small particles. Considerable variation in deposit was noted between the various artificial surfaces. Aluminum.and wax -42- surfaces showed but slight increase in deposit with in- creased velocity. The petroleum jelly surface deposit increased by the ratio of 1:3.8 and 1:5.3 for a velocity ratio increase of 1:3 and 1:4.7. This correspondence ap- pears to verify the preceding theoretical analysis within experimental error. The deposit of micronized dust on plant surfaces is' compared with that of the petroleum.jelly surface and the polished aluminum surface on an equal area basis in Figure (4). The values for deposit were corrected for velocity since, in the experimental procedure, the amount of dust blown at a surface varied directly with the air velocity. The results indicate that plant surfaces may be some- what less efficient in retaining dust than a petroleum jelly surface. The aluminum surface gave more erratic results, but of a magnitude similar to that of the plant surfaces. The increase in deposit was still linear however. This suggests that, upon the determination of some constant of pr0portionality between the petroleum jelly surface and various plant surfaces, this may be a means of evaluating inertial deposit in the field. Electrical Forces Little is known about the role of electrical forces in present-day methods of dusting or spraying. ‘Where the -43- Amv sweppfiam anu when .ncosOHs on.s mes noposcwc oaofipnmo some oflpossuwa< .pmmao aw: oz» 0p seasofiesoopoo sac: moosmssm economuflc sou oops Mo J.CH wo.o so pmdc 0 2m. om :0 mo pfimomoc HwapHmCH .q mndwwh o ~.§.~§\.\.N\u kscuONw> Rex Obhi 000\ 00k. 0 F IF a O P.\ 0 .nSQ. w .0 \Q‘\\\ q 1‘Qo. \ u vQQO. Q vQQD. ero. . 0 e \eeu eeeuueflm rNxo. ‘38 £25K x c-Qefl (coma cue-tom. Eet.-E\-\¢~Q. .30. eucxxx-m. \m\\e\.. EeexeLueQAv e 36. (:wvdg) 1.160430 -44- relative humidity is below 60 per cent there may be con- siderable frictional charging of the dust particles. The exact magnitude of this charging and its importance in the overall picture in field applications have not been de- termined. Wilson, James and Campau (33) investigated the frictional charging of several dusts. They found that, in general, organic dusts charge positively and inorganic dusts negatively. Charging was measured by determining the charge left on the duster nozzle. They found a decrease in frictional charging with decrease in particle size. Particle distribution on plant leaves was improved by charging. macLeod and Smith (16) measured the deposition due to frictional charging by blowing the dust between two plates. The potential between the plates was 500 volts. They observed fractionation of sulfur and talc dusts (part of the dust settling on the positive plate and part on the negative plate). Therefore some dusts may have friction charges of both signs in the dust cloud. wampler and Hoskins (29) investigated the charging of spray particles during spraying. They tried to de- termine whether an increase in deposit of lead arsenate could be obtained by charging droplets. Deposit was measured on a plate on which the spray particles impinged. Charge was measured by catching the spray in a container and measuring the potential across a known resistance to ground. The charge caused by breakup of the dr0plets was found to increase with pressure. A D.C. potential applied to the Spray nozzle further increased the charge. Their increase in charge appeared to be directly proportional to applied voltage. Their conclusion was that no increase in deposit could be obtained by charging regardless of the applied voltage. This conclusion was due to faulty experimental technique. Since all the drOplets impinged on a plate they were simply measuring surface wetting upon which charge would have no effect. Through use of an ionizing field Bowen (3) developed a means for applying a charge of either sign to dust or fog particles. Some of the results obtained through charging were given in the Introduction. The ionizing field for the charging of particles for most of the studies of Bowen (3), (4) Hebblethwaite (12), Brazee (5), and this work was obtained through use of a cylindrical condenser. The outer cylinder was grounded and the wire or needle along the longitudinal axis of the cylinder was placed at a sufficiently high potential to give a corona discharge. The particles are charged to the sign of the electrode. The process of charging is discussed by Bowen (3). -46- .According to Bowen (3). the field intensity across the greater portion of the cylinder is given by Pauthenier to be: E0 : '5' --------- 19 Where Eo : the field intensity in esu in the nozzle 1 the current per cm. of wire length in esu, and u the mobility of the ions in esu (approximately 500). The maximum.charge that can be placed on any given particle is given by Ladenburg as: = - e-l 2 ----- q n3 .. Eof 1+2 (33.2)] r 20 Where 6’: the charge of an electron (4.8 x 10'10 esu) n : the number of charges 6 = the dielectric constant of the material, and r : the radius of the particle in mm. For most non-conducting dusts the value of l+2({%:;%) is between 1.5 and 2. For perfect conductors the value will be 3. The electrical field forces for various charged cloud configurations have been derived by Bowen (4). The potential distribution and electrical field forces in a spherical charged cloud with a grounded sphere at the center was found to be: 3 2 2 _ 2 2_477Ja _27J1b -al§ ---21 Which may Where‘V d': 8.: b: d : spheres to the The field Therefore be simplified to: 2 3- 3 2,4132 _ Lgai-jb a] ________ 22 the potential at any point. the charge density (assumed uniform) the radius of the inner sphere in cm, the radius of the outer sphere in cm. and the radial distance from.the center of the point at which V is being measured (cm.). intensity at any point is given as -E : 5'} 2 3 3 _ = ,2; 3b a-a “2‘1 ______ E 2 [- 3(12 J 23 Most plant systems will be a configuration of grounded plates in a charged cloud. Therefore, in addition to the spherical cloud distribution Bowen also considers, by the method of images and superposition, the potential distri- bution and field forces for a flat plate with a charged cloud on one side and for two plates with a charged cloud between the plates. .A specific example was worked out. .A derivation of the general equation for the potential distribution and field intensity between two plates enclosing a charged cloud is given in Appendix II. Neglecting edge effects the potential distribution is given by v : 27rd (ad-d2) -------- 24 -43- Where a : the distance between the plates in cm. and d the distance of a point P from one of the plates at which V is being determined. Again where -E :‘j2$ , then: -E : 21J (a-2d) --------- 25 The validity of these equations in a charged cloud has been well established experimentally by Bowen. Terminal velocity of charged_particles in electric lggggg. The force on any particle is given by the equation: F :IEq --------- 26 In the preceding discussion it was noted that, for any given experimental conditions, 4’ E : Kgd' ---------- 28 Where K, : Ec [1+2 (if-5%)) K, : some constant depending on the cloud Ky r‘‘ and -------- 27 configuration and efficiency of the charging mechanism and V = efficiency of the charging nozzle. Writing Fqu: (ml-"14036) ------- 29 But d', the charge per unit volume may be written Where Q : the total charge 2: :: a unit volume and n 2 the number of particles per unit volume. Therefore F : quzn = KzKlzl'l’an ..... 31 For any constant rate of application the weight of material per unit volume can be considered constant. Then W : wn = éirBI'n Where w : the weight of one particle Giving n : %P Therefore 2 2 F __ KgKl I-Wfl _ _ 32 - A” ..... Therefore the force on any charged particle in an electric field varies directly with the particle radius, and inversely with particle density for any constant rate of application. Using the distance moved by a particle per unit time as a criteria of deposit rate, then the terminal velocity of any particle, as given by Stokes' law (Equation 12), will be a relative measurement of deposition. F But from Fzma, azfi 3K2KJZI'WIIz Then a : 1,”. m ----------- 33 But m : 3771'31' 2 2 - 9K2K1 WV ___________ Then a - 161,”?— 34 Therefore, on substitution of (34) in (12): - 8%,, 872,59 -50- This equation indicates that the rate of deposit of a charged cloud will be independent of particle radius for a constant rate of application. The rate of deposition will increase linearly with increase in application rate, and will decrease with increase in particle density. It might also be noted that, since v varies with the square ofz/, considerable attention should be devoted to the perfection of these devices. Calculation of terminal velocity of charged particle under electrical field force. To determine what magnitude of v might be expected under field condition a calculation of v will be made under the following assumptions: Ec = the charging field in the nozzle : 7 statvolts/cm. 1+2(§1F%) : 2 for dust; I! a charging efficiency :‘é Dust rate : 50 pounds per acre applied through four 13” diameter nozzles in a six foot swath. fl : 182.7 0. poises Dust density : l Nozzle air velocity : 60 miles per hour Rate of travel of duster : 3 miles per hour The dust particle under consideration will be assumed to be just striking the lower leaf surface between two large leaves four inches apart at their center axis. -51- Then Ki statvolts/cm. flung—pk, : (§)<2H7): 9.34 K2 : 211(a-2d) where d: o 2i!(4in. x 2.54 cm/in.) : 65 cm. Wt/A (SOle/A)(453.6 gm./lb.) : 22,600 gm./A - ( 43, 5608.2/A)<6omn{hr) Time for one acre '(311111)(5280ftlfnile)(6ft.)=27-5mino/A Then 411(75) (60) (5280) (27.5) (2.83xlo'h (144) (so) : 2.02 x 108 cm.3/A Air vol. per acre W : 221600 gmLA : 11.2 x 10'5 gm/ cm3 2.02 x 10 cm /A v = K2 K12‘W112 ‘fiflrzrfla Then _(65cm)(9.343figiggg)2(11.2x10‘5gm/cm3) 8(182.7x10'° .91_ )7rz(1 gm/cm3) cm.sec. = 44 cm/sec. At a distance of one centimeter from the surface, v will be 35 centimeters per second. Therefore, under conditions which are now seen in the fields, electrical forces can be important in deposition. This analysis has not taken into account the mixing of the dust - air stream with other air. This would dilute the dust concentration and thereby decrease the electrical forces. Also any increase in density of material in the -52- dust particle or decrease in application rate would decrease the electrical forces for a given application rate. On the other hand, the electrical field forces will be far greater for those leaves exposed to the main dust cloud. Referring to Figures (2) and (3) it is noted that, for particles below two microns in radius, the terminal velocities acquired because of gravitational and inertial forces would be less than .1 and 5 cm./sec. respectively. Since the terminal velocity is independent of particle size for electrical forces, these forces will play a major role in the deposition of small particles. In very high field intensities there would be a further advantage in the use of smaller particles since these particles will have a smaller Reynold's number (R = 3%!5) at any given particle radius, thereby allowing higher terminal velocities for the smaller particles. Thermal Forces Potts (25) observed that ". . . a field of resistance surrounds all objects including plants and insects and repels most individual dust particles of small size as well as droplets less than 30 microns in diameter." He was probably observing thermal repulsion. The existence of a dust free space around a body of higher temperature than the surroundings has been known -53- for some time. The wedge of dust free air above a body was observed by Tyndall in 1870 and by Rayleigh in 1882. The complete layer of dust free air about a body was noted by .Aitken, Lodge, and Clark in 1884. The thickness of this dust free space was investigated by watson (31). For a vertical plane surface held at 23° C. above ambient the dust free space was observed to be .32 millimeters thick. He found that the thickness of the dust free space varied with the .52 power of the tempera- ture difference and was independent of the nature of the dust. There were some differences in thickness due to the shape of the hot body. For a region in which convection currents have been suppressed Paranjpe (24) has shown that the dark area can extend for distances of up to four:millimeters. Under natural conditions where convection can occur the dust free space will be reduced to one-half millimeter or less. The force of thermal repulsion is believed due to the difference in momentum imparted to the body by the molecules of air in a temperature gradient. Ramdas and Joglekar (26) have shown that the force of thermal re- pulsion is one thousand times the magnitude of radiation pressure. For small spheres they find that the magnitude of the thermal pressure is around 3 x 10'4 dynes/cm.2 per unit temperature gradient (degrees Centigrade) using oil droplets. Paranjpe (24) found the force to be around 6 x 10" dynes/cm.2 from observation of smoke particles. Results obtained by Ramdas and Joglekar also show that the increase in thermal pressure is directly proportional to increase in temperature gradient. For the smoke, Paranjpe found the velocity to vary as: y -.- 1.756 x 10-4 f}:— ------ 36 and for the oil droplets Ramdas and Joglekar found the velocity to vary as: v = .817 x 10-4 7‘39;— ------ 37 Where "3"} is the temperature gradient in oC/cm. Because of the low heat conductivity of air the thermal gradient next to a hot body is quite high. Kennard (15) investigated the temperature gradient from a vertical plate 21 centimeters high by 10 centimeters wide through the use of an interferometer. For plate temperatures ranging from l4.3° to 118° C. above ambient air temperature, he found the greater part of the tempera- ture drop to be within one centimeter of the plate. His results are shown in Figure (5). From this curve the temperature at any distance from the plate may be calcu- lated if plate temperature and ambient air temperature are known. Curtis (8) investigated the temperature of plant leaves. He found that on a clear day a leaf in direct -55- sunlight could be as much as 14° C. above air temperature. In a light breeze this temperature might be reduced to 8° C. above ambient or less. For a shaded leaf the temper— ature was found to be 1 to 2° C. below air temperature. This was believed due to radiation losses to the sky. On a cloudy day the leaf and air temperatures were essentially the same. Temperature measurements on tobacco leaves by Wilson (33) indicated that leaves in sunlight were from 12° to 20° F. above ambient air temperature. Leaves in the shade were from 3° to 10° F. above air temperature on the bottom and from 0° to 7° F. lower than air temperature on top. This temperature difference on a leaf in the shade is believed to be due to radiation to the bottom from the ground surface (which was 54° F. above air temperature) and radiation from the top to the sky. Therefore plant surfaces in the field can, under some conditions, be at a higher temperature than air tempera- ture and under other conditions they can be at a lower temperature. Recognizing that the values for temperature gradient are taken from data obtained from.a vertical plate whereas most plant leaves are nearly horizontal, the role played by thermal forces in field dusting or spraying will be investigated. -55- .:osa me So am ego? mcofiosofiofid opsam .Amav osscsow mo scoop mu..- opus; HSSopS0> eon mo cosmesm Sosa psofioosw osspssooaoe .m osswflm fiEEwNUc-kmzm ROT- zokk USS-Rme u- e- e e v w . F b F s fivo\emc\n- Maxi Kc exRV vLeKquKu Kw «vs \V exeuekektxmh It? x toxflvau cm TN. TM. 1?. '52 ’:-<; '2-3 Assuming an ambient air temperature of 20° 0., and a leaf temperature of 30° C., which is evidently quite justi- fiable, then from Figure (5) for a distance 1 millimeter from the leaf surface: t (.8)(ts-ta) + ta (.8)(10)+'20 : 28° C. Therefore the temperature gradient is 2° C. for a distance of one millimeter or 20° C. per centimeter. Assuming that the larger dust or Spray particles will behave similarly to oil droplets in a temperature gradient we find from equation (37) that: v z: (.817 x 10'4)(20) : 16 x 10": om./seo. For very fine particles we find from equation (36) that: v : (1.756 x 10‘4)20 : 35 x 10" cm./sec. These velocities are very low in comparison to those found for gravitational, inertial and electrical forces. Summary The physical system governing the deposition of small particles has been considered in this section. Since it is important that deposition occur within a short length of time the terminal velocities of the particles under the various forces have been determined as a measure of rate of deposit. Based on the information available the following general statements may be made: -53- Coagulation of dusts and sprays at present-day concen- tration may be neglected for all practical purposes. Dust and spray particles between one and fifty microns in radius will deposit out by gravitational forces ac- cording to Stokes' law. Particles above fifty microns will fall at a rate increasing linearly with radius. Particles below five microns in radius fall at such a slow rate that they would be carried considerable distances before settling out. Mbst deposition of particles by inertial forces will take place in a streamline manner. The rate of deposit will increase linearly with air velocity. At points directly under the applicator nozzle deposition may occur in the intermediate region where 2$R 5 500. Not many particles will experience a centripetal force great enough to cause them to move in turbulent motion. At high air velocities the rate of deposit will be independent of air velocity and, the actual deposit may decrease with increased air velocity because of erosion of the particles from the surface. .Although accurate information is not available concerning the magnitudes and direction of air velocities within the plant configuration, inertial forces can exceed gravitational forces many times. For particles in streamline motion the terminal velocity (therefore rate of deposit) of charged particles -59- in an electric field has been found to be independent of particle radius. The rate of deposit due to electrical forces is probably lower than that for inertial forces for large particles but for particles less than four microns in radius the electrical forces may cause a rate of deposit considerably higher. Even though plant surfaces may be as much as 14° C. above ambient air temperature thermal forces are of small magnitude when compared with the other forces. Although they may decrease the rate of deposit of small particles somewhat, they will be of little consequence to large particles. -50- EXPERE.‘IENTAL WORK A series of experiments were run under controlled laboratory conditions to quantify the effects of gravi- tational, inertial, and electrical forces on dusts of different density and particle size. The effects of these forces on the deposition of the dusts were measured at three relative humidities, and the effects of inertial forces were investigated at four air velocities. An additional experiment was conducted to determine the magnitude of friction charging of dust. Field tests were conducted to compare the effective- ness of plant disease control of commercial and micronized dusts at various rates per acre, both charged and un- charged 0 Laboratory Experiments Laboratory dust chambers. .A brief description of the laboratory dust chambers and their Operation will be given here. A.more complete discussion is given by Brittain (5). The laboratory test chambers consisted of two com- partments, a dust feed chamber and a dusting chamber. In the dust feed chamber, Figure (6), a conventional Niagara duster fan was driven by a variable speed three -61- Figure 6. Dust Feed Chamber The dust, after being evenly distributed in the V - shaped trough, was placed on the flat belt hopper. It was then fed into the conventional duster fan driven by a three horsepower motor and blown into the dusting chamber through the connecting tube. Relative humidity measurements were taken through the small window. -62- horsepower motor. The dust was fed into the fan from.a belt hepper driven by a small variable Speed electric motor. To prevent the falling of clumps of dust into the fan a rotating brush was driven at the feed end of the belt. Air entered the chamber from an eight inch diameter duct in which room air, outside air and steam were mixed to obtain the desired relative humidity. .A wet and a dry bulb thermometer were hung inside a window for observation of temperatures for determination of relative humidity. The dusting chamber in which the deposit data were taken is shown in Figures (7) and (8). The dust air stream from the feed chamber was introduced at a very high velocity to prevent clumping in the connecting tube. .A baffle deflected the air immedi- ately upon entering the dusting chamber to give a uniform dust cloud from which the electrical, inertial and gravi- tational deposits were determined. The dust was charged by connecting a charging nozzle of the type described by Bowen (3) to a source of D.C. voltage. A charging voltage of 10 KV was used on all experiments. Excess dust was removed by an exhaust fan. The intake baffle of the exhaust fan was adjusted so that the pressure in the dusting chamber was about one-fourth -63- r/‘I‘hiw :51 ‘ . 1 II “1 in tax ‘7 '5" I. :1 '8'. W +- . am: 7 . 1' .‘ iii: a" .1! Cfibe' I 0 e S“ '( Figure 7. Dusting Chamber and Experimental Set-up The three inch diameter aluminum disks were placed in clamps in the dusting chamber through the sliding glass door. Excess dust was removed from the chamber by means of the exhaust fan on top of the chamber. Deposit was measured on the analytical balances. Dust particle size distribution was measured by means of the microscope. Dust samples were weighed on the balances in the foreground. -64- Figurc 8. Interior of the Dusting Chamber A uniform dust cloud was obtained by blowing the air — dust suspension against a baffle. This suSpension was then directed against an aluminum disk by means of an auxiliary fan in which speed could be controlled. Dust concentration samples were drawn through the uillipore filter suspended above the ring stand. Relative humidity readings were taken from a wet and dry bulb device suspended in front of a small window. -65- to one—eight inches of water below atmospheric, as ob- served on a small manometer. The purpose of this slight depression was to keep dust from leaking into the labora- tory. Metering of dust. The dust was metered into the fan in the dust-feed chamber from a flat belt. The speed of the belt could be varied from .0927 to .191 inches per second. For the inertial and gravitational deposits ten grams of dust were placed on approximately four inches of belt length. For the electrical deposit tests, the dust was first placed in a‘V - shaped trough which was set on the belt. The small angle of the trough allowed the dust to be spread more uniformly. The handles of the trough were then depressed, releasing the dust onto the belt by opening the trough. Ten grams of dust were spread along twenty-five inches of belt length. Two additional grams were added at the first inch of belt length to raise potential in the dusting chamber quickly. The potential would then remain fairly constant for the remainder of the run. The belt was operated at .191 inches per second. The dust fan was operated at 3600 revolutions per minute for all tests. Auxiliary fan for inertial tests. Because of the length of the tube connecting the feed chamber to the -66- dusting chamber and because of irregularities in the feed rate due to catching of dust on the side of the inlet housing, the air-dust flow from the feed chamber was not considered uniform enough for accurate results. A.baffle was mounted in front of the tube nozzle and an auxiliary fan was used to obtain the required air velocities. The dust-air mixture was pulled from the chamber and blown at an aluminum disk held perpendicular to the air stream by a burette clamp. The auxiliary fan was powered by a series - wound motor, and speed could therefore be controlled by means of an auto-transformer. Air velocities were measured with a hot - wire anemometer. ‘Measurement and control of relative humidity. Rela- tive humidity was measured by means of two wet and dry bulb devices as shown in Figure (9). The constant level wet bulb system.is described by Henderson (13). The relative humidities used in the experiments were forty, sixty, and ninety per cent. Accuracy of control was within two per cent for the lower two relative humidities. Relative humidity for the ninety per cent relative humidity treatment could only be controlled within eighty-four per cent and ninety-one per cent for the first half of the inertial tests. After modification -67- Figure 9. Wet and Dry Bulb Temperature Device. -6 8.. of equipment the relative humidity could be maintained within two per cent for the remainder of the inertial tests and for the tests for electrical deposit. Relative humidity in both feed and dusting chambers were maintained the same. The relative humidity was controlled by injecting steam into the air stream, in the ducting ahead of the feed chamber. .Air could be drawn from either inside the room.or outside. At high relative humidities an auxiliary steam jet had to be added to the dusting chamber. Measurement of dust - air density. The number of grams of dust per cubic centimeter of air was determined by drawing a measured amount of air through a Millipore filter shown in Figures (10) and (11). The filter was weighed prior to and immediately following sampling. The rate of flow through the filter was 20.4 liters per minute. Samples were taken for one minute for the gravitational and inertial tests and for two minutes for the electrical deposition tests. Dust concentrations were determined for the second velocity run for each dust and relative humidity in the inertial and gravitational deposit tests and for each run for the electrical deposit tests. Measurement of cloud potential. Charged cloud po- tential was measured by means of a small probe coated -59- Figure 10. Millipore Filter for Determining Dust Concentration The weight of dust per cubic centimeter of air was determined by drawing a known volume of dust suspension through a membrane filter. A filter through which the dust suSpenSion has been drawn and an unused filter are Shown. -70- Figure 11. Interior of Filter Holder To prevent breakage of the membrane filter, it was supported by means of a porous sintered disk. -71- with radioactive Polonium.nitrate, an as emitter. The emission of the at particles ionized the air in the vicinity of the probe thereby supplying any charge necessary to bring the probe and measuring device up to cloud potential in a short period of time. The probe was lowered into the charged cloud at a given distance from the grounded screen and potential readings were taken on an electrostatic voltmeter or an electroscope. A.more detailed discussion of the measuring probes is given by Bowen (4). A potential traverse of the cloud verified the calcu- lated potential distribution except for a slight skewness. This skewness in the measured potential distribution was unaffected by a large baffle p1aced either ahead of or following the grounded Sphere and was not noticeably altered by any change in orientation of a small fan added to maintain turbulence. The skewness was also unaffected by changing the point on the sphere from.which it was suspended. This skewness, in which the potentials are slightly higher in the upper hemisphere than in the lower hemi- Sphere for any position of the probe with respect to the center, is believed due to differences in charge density. The charged particles above the center of the sphere will ‘have an upward component of electrical force and a down- ward component of gravitational force. Those particles -72- in the lower hemisphere will have a downward force com- ponent from both electrical and gravitational forces. Therefore the particles in the upper hemiSphere will deposit at a slower rate than those in the lower hemisphere, giving rise to unequal charge density. Gravitational dgposit. Deposits due to gravitational forces were obtained simultaneously with inertial deposits. A three inch diameter aluminum disk (Figure 12) was placed horizontally in an open box to prevent erosion by air currents. The deposit was obtained for stirred settling rather than tranquil settling as this was the condition found more commonly in the fields. The aluminum disk was carefully handled with tweezers and deposit was determined by weighing on the analytical chain balances with a magnetic damper, seen in Figure (7). The accuracy of measurement was within .0001 grams. Five dusts were used for the inertial and gravi- tational tests. These dusts, their mean particle diameter and density are shown in Table IV. The dusts used were micronized talc, red talc, standard attaclay, micronized attaclay and Attasorb. Inertial deposit. The deposit due to inertial forces was obtained by blowing a more or less homogeneous dust - air suspension from.the dust cloud in the chamber against a three inch diameter disk. The disk was placed -73- — _._ . ...-—_.._._—-— uu Figure 12. Aluminum Disk and Sphere Gravitational and inertial deposits were determined from the three inch diameter aluminum disks. Electrical deposits were determined from the one inch diameter hollow aluminum sphere. -74- perpendicular to the air stream.in a vertical plane thereby eliminating any gravitational deposit. Air velocities of 350, 700, 2000 and 3000 feet per minute were used at three relative humidities. Deposit was measured by means of the analytical balances. The aluminum disks were hung on pegs on a wooden stand (see Figure (7)) prior to and following weighing. Electrical deposit. The deposit due to electrical field forces was measured through use of grounded hollow metal spheres made of aluminum foil shown in Figures (12) and (13). The Spheres were approximately .95 inches in diameter with the flange where the two hemispherical shells were joined being trimmed to one-sixteenth inch or less. These spheres were supported at the center of a grounded wire mesh Sphere 35.5 inches in diameter by means of a fine copper wire insulated with spaghetti tubing. The aluminum Spheres were inserted and removed from position through a door in the wire sphere. To keep the very light spheres properly aligned a weight was suspended from the bottom of the Sphere by means of a nylon thread. The spheres were carefully handled with tweezers. A.more detailed description of the Spheres is given by Bowen (3). .A radioactive probe mounted at the end of a piece of glass tubing was inserted above the aluminum.sphere. -75- Figure 13. Wooden Stand for Aluminum Spheres. -76- The height of the probe above the Sphere was adjusted by means of a calibrated pulley such that, by regulating probe height, the potential readings could be kept within the range of the electrostatic voltmeter. The potential at any point in the Sphere could be calculated from the potential at any known point. The electrical field forces were then calculated from the calculated potential distri- bution. Voltage readings were taken every fifteen seconds, the runs being 2.5 minutes in length. One millipore filter sample was taken for each run, the sample being drawn near the edge of the wire sphere. The six dusts used were standard talc, micronized talc, red talc, standard attaclay, micronized attaclay and Attasorb. Tests were made at three relative humidities. Dust particle sizes. The particle size distributions of the dusts were measured by counting the number of particles of given diameter ranges in randomly chosen areas of the field of a microscope. Size was determined by means‘of a calibrated stage micrometer in the eyepiece of the microscope. Counts were made by Ban. The size distributions of the various dusts are shown in Table XV in Appendix III. The mean particle diameters and densities are shown in Table IV. -77- The standard talc and attaclay (Fullers earth) dusts were 325 mesh dusts used as commercial fillers. Inert filler dusts were used because of possible danger from explosion of organic or combustible dusts. The micronized talc and attaclay dusts were obtained by running the 325 mesh dusts through a micronizer. The red talc was micro- nized talc which had been mixed with a red organic dye and then re-ground. This dust had been found to give severe reverse ionization upon charging in the field (see Brazee (4)). The Attasorb was a commercially available finely ground attaclay. TABLE IV AVERAGE DIAMETERS AND DENSITIES OF DUSTS Arithmetic Particle Dust mean Diameter Density Reference (microns) (gm./cm,3) Standard Talc 4.76 2.8 (29) Micronized Talc 3.44 2.8 (29) Red Talc 2.69 ~- Standard.Attac1ay 2.20 2.45 ( l) ‘Micronized Attaclay 1.36 2.45 (1.) Attasorb 1.41* 2.45 ( l) *Specifications published by the Attapalgus Minerals and Chemicals Corporation indicates a surface mean diameter of .4 to .6 microns as determined by air permeation.measurements. -73- Friction charging. A.series of tests were run to determine what magnitude of friction charging of dust one might expect from conventional dusting equipment. The cloud potential was determined in the grounded wire sphere. The deposit due to the frictional charging was not determined. Filter samples of dust cloud density were not taken as the same dust feed rates were used as had been used in the tests for determination of electrical deposits. All tests were run at sixty per cent relative humidity. Voltage readings were taken every fifteen seconds for a period of two and a half minutes for each run. The distance of the radioactive probe from the center of the sphere was adjusted to keep the potential readings within the scale of the electrostatic voltmeter. Experimental design. The test for inertial and gravitational deposits was a randomized block design with split plots or groups. Because of the time element for equilibrium to be reached in the change of relative humidi- ties these were designated blocks. Dusts were designated groups to minimize any carryover in the fan and tubing in changing from one dust to another. Air velocities were designated sub-groups or sub-plots. Each block was then replicated four times. The order of blocks, groups and subgroups was determined by drawing numbers from a box. Because of considerable population variance with some of the dusts, additional runs were made where it was felt advisable. The total number of runs was used rather than discarding questionable values as it was felt that there was no basis for determining which values were questionable in an unbiased manner. The tests for electrical deposits were also a random- ized block design with relative humidities as blocks and dusts as groups. Again four replications were run for each block. Three of the weights of filters were lost because of cracking of the filter membrane during the run. The missing values were determined in a manner suggested by Baten (2). ‘ The tests for potential due to frictional charging consisted of six dusts and four replications. Where readings were missed due to the indicator going off scale of the voltmeter a straight line relationship between the preceding and following values was assumed for the determi- nation of the potential at the missing value. The arithme- tic mean value of the ten voltage readings was used in the analysis of variance. One series of runs was discarded and a missing plot value determined for purposes of analysis. -80- Field Tests During the summer of 1952, field tests were run in an effort to determine whether control of plant diseases could be attained at application rates as low as two pounds per acre. Conventional control methods called for an application rate of fifty pounds per acre. Control of onion mildew was investigated on the Jack Kelly farm, near Parma, Michigan, and control of early and late celery blight was investigated on the Kramer farm near Comstock, Michigan. The plots were set up in cooper- ation with Dr. Nelson, Mrs. Mooar and Mr. Lundberg of the Botany Department. Experimental dust hopper. The conventional field duster used in cooperation with the Botany Department was found to be quite erratic in metering dust at rates below fifteen or twenty pounds per acre by Bowen during the 1951 season. In order to apply dust at rates below ten pounds per acre with a reasonable degree of accuracy a pneumatic metering device was developed. In this device air was pulled into a cylindrical container through holes in the periphery. The air was pulled out of the container through a tube at the center of the cylinder cover. The dust in the chamber was maintained in a fluffy state by means of an agitator. -81.. An unsteady state in the air flow resulted in an unbalance of forces causing the air to whirl about in the container. This high velocity air stream was utilized in picking up the dust from the fluffed dust mass. The heaviest particles and aggregates were thrown to the sides of the container by centrifugal force and the smaller particles were drawn off from the top. WOrking with Bowen, a plexiglass model was constructed and the effects of various hole positions were noted. A field model was then constructed using a metal h0pper and an agitator driven by a six volt D.C. motor using an air- craft antenna rotator gear system for speed reduction (Figure 14). Rates of application were then varied by closing off a given number of holes at the top of the container. Dust flow could be metered at from 1.6 to .2 pounds per minute. The rate of dust flow was found to be considerably affected by changes in relative humidity, indicating a certain hygroscopic nature of the dust. The duster was calibrated prior to each test in the field. The dust was charged by means of four charging nozzles. Approximately 12 KV D.C. was supplied to the charging nozzles from a tripler circuit connected to a 300 volt dynamotor which was powered from a six volt battery. -82.. Figure 14. Experimental Dust Hopper The dust-air suspension was drawn from the t0p of the container by the duster fan. The unit could be easily mounted or removed from the duster. -83- Experimental design. Randomized plots were establish- ed in the onion and celery fields as shown in Figure (15). Application rates of twenty five, ten, and two pounds per acre of sixty per cent sulfur and fourteen per cent Dithane micronized dust were compared with an application rate of fifty pounds per acre of 200 mesh thirty per cent sulfur, seven per cent Dithane dust in both the onion and the celery fields. In addition, Mangate, Copper-sulfur-zinc- manganese, copper-sulfur-manganese, and Dithane-sulfur- manganese dusts were also applied at a rate of fifty pounds per acre on the celery. The conventional hopper was used for application of the standard dust and the pneumatic hopper was used on all micronized dust applications. Four replications were run on the onion plots and two replications were run on the celery plots. A11 dusts were applied both charged and uncharged. One plot con- sisted of four rows for the onions, of which only the center two were sampled. Two rows per plot were used in the celery. -34- 1 1- :. Dithane-S ‘--- 1 ................. 1 Manzate ,, Charged 1 Uncharged ........ 1 Check 1 1 1 Cu—S-Zn-Mn L ’- 1. Cu—S-Mn ' 1 4_1 Dithane-S-Mn 1 1 1 2 /A u D-S 1 ....... LII-1911228941 ........ 1 ...... 91.182828 .......... 1 125441;; D1355 v v v D. - 1 . , 2#/A u D-S 1---__--QQ§!8§‘-1 1 Ugghggggg ________ 1 égfiéfi 3 3-: ' c , - x 1 1 Manzate 1 1 1 Cu-S-Zn-Mn 1 ....... 11 11911828911 . charged _________ , Cu-S-Mn 1 L 1 Dithane-S-Mn 1 1 1 Dithane-S ' . . Check Celery Plots--l952 I I I f ' ' 8 ' 7 ' 4 ' 1 ' 1. Check I I I I I . . , , , 2. Conventional 9 v . , , 3. u 25#/A Charged 0 L . 2 . l , 7 , 4. u 25#/A Uncharged 1 . , , , 5. u lO#/A Charged . , , , ' 6. u lol/A‘Uncharged . 6 . 6 , 5 , 1+ , 7. u 2 7.4 Charged 8. u 2 A Uncharged I I I I I I l I 1' I 8 I 3 ' I I I I I I I I I I I I I I I ' 7 ' 5 ' 7 ' 8 . I I I I I I I I I I ' 5 ' 3 ' 3 ' 6 ' I I I I I ' 3 z 1 I 2 z 2 z I I I I I I I 2 v 8 I 6 o 5 0 L. L L _L I Onion Plots--l952 Figure 15 -35- EZPERIMENTAL RESULTS Gravitational Deposits The results of the tests for gravitational deposits of dust under conditions of stirred settling are shown in Table XVI, in.Appendix III. Figure (16) shows the average deposits for the five dusts at three relative humidities. The analysis of variance, shown in Table V, indicates that there were significant differences at the one per cent level between gravitational deposits of the dusts. These differences are shown more clearly in Figure (17) where the method of presentation is a modification of that suggested by Duncan (10). Those values underlined in black are not significantly different at the one per cent level and those underlined in red are not significantly different at the five per 0 ANALYSIS OF VARIANCE OF GRAVITATIONAL DEPOSIT ent level. TABLE'V ONWHORIZONTALAALUMINUMIDISKS Source Degrees of Sum.of mean _F Freedom nggres Square Dusts 4 35353.2 8838.3 100.68** Relative Humidity1 2 369.26 184.63 2.10 Individuals 225 19,752.19 87.79zs2 Total 239 **Significant at the one per cent level *Significant at the five per cent level .mofipfloflusn.opflpmaoa omega pm memos o>Hm no pflmoooe HowpnosH FEE-231 u> CSUQ utmoqwq cm on ok or I..- Irv ).bl(rl.Lnl((l01luo|l(-Iil . -bJ OM. 0? L b «L E9 -86- \e\ee}\< weatexewt 0 eschews-T B «\Ooh Not X e\6.h \Uw.~teku..\< Q \I\quu‘ Q .as assess :0\ vQN non. v P... (1.4.55. __ a: x ) z Irodzd scream V ’87“ .ucec poo o>wh Lo ooCmLHmHCumm mo Ho>oa a pm pocsocewc nos oLm we; cw roowmsoccs sesame omega .Hobofi some poo oco one no haucocwuficafim usesoecfir ooc cam Enema :6 cosfiasoecs mosfieb emo:e .wswfippom reshape mo mcofiuwes00 Lees: mxmfie Escwfisao so memos co muwmoooe Hmcofiumpw>mpm ommpo>¢ 3.01m. V H \va TNNM 19¢..Q.m .aN .I ill.v(rlrII . 3'! II -‘e‘ i weEWuo; on On ow P L4 - A? he‘d-x .36. “flex 2:3st .3 3a $2.3. 0‘ 0 . .bH ossmwh u R OQHQ ~e< -33- There were no significant differences between deposits at various relative humidities at the five per cent level. The cause of the D x RH interaction can be seen in Figure (16), where there appears to be a trend on the part of standard attaclay deposit to decrease with increase in relative humidity. Inertial Deposits The results of tests for inertial dust deposit on aluminum disks in a vertical plane are shown in Table XVII in Appendix III. The analysis of variance of these results in shown in Table VI. Because of unequal sub- sample size the harmonic mean was calculated to determine the effective error for testing. The differences in deposits between dusts were statistically significant at the one per cent level. The overall average in inertial deposit is shown in Figure (18). The differences in deposit between the three attaclay dusts and between two talc dusts were not significant at either the one per cent or five per cent level. There was a significant difference between the two types of dusts at the one per cent level. ~89“ .HoboH use: Loo o>we exp um economefic mausocawwsmwm no: one can ow oocfiasocss mozao> moose .Ho>oH once Lee moo esp we pnesopefic hapsmaaewcmsm no: ope Remap ca rocwasoccs mosaop omens .mofiufiofissn obapoaoh oops» can moapwooaob saw each Lopo roamsobo one muHSmom .mxmfic ESSHESHS no muamoaoe HowusosH omopo>< .mH magmas . ae. MR. n .Q.m.w a .m u ... .3me ashteuoxk 00 0Q OK 0.0 Oh. 0V 0». o r? a . a p . ”7 fi p NEIL Ks‘t‘q Macaw-é- u\ IF 1 e3. ‘1 on... a «- £81m :\ 3315. a. motes-x v-m -90- TABLE VI ANALYSIS OF VARIANCE OF INERTIAL DEPOSITS Degrees Source of Sum of Mean F Freedom Squares Square Dusts 4 40103.100 10025.775 34.723** Relative Humidity 2 242.434 121.217 .4198 Air‘VelocitY 3 35117.11? 11705.705 40.54l** D x 1m 8 4908.4 613.550 2.125* D x V 12 14253.3 1187.775 4.114** RH x‘V 6 2892.633 482.106 1.670 D x V x RH 24 9107.200 379.467 1.314 Error term for testing 198 106624.184 288.739 **Significant at one per cent level *Significant at five per cent level The differences in inertial deposit due to air ve- locity were significant at the one per cent level. The variation in deposit with velocity is shown in Figure (19). There was a difference, at the one per cent level due to a D x‘V interaction. This indicates that the increase in deposit with velocity was statistically differ- ent for the different dusts. The deposit values corrected for variation in incident dust quantity (see Brittain (5)) are shown in Figure (20). ,‘9"' .moapwcfiesn o>wpoaou oonnp nebo cmwwuopw one mpasmom .moapaooaob has Agog pa ammo no ufimomoc HmwpnocH «csiekv \Eooflfi «3 COMM. 00.0N 0.0.0x 0 ’ .H\\\\\\\ 0 a 86133. chhtca. O «Leueuu< O hexucgu< 102530.45 D x , 33.x. \ooNNQOKuxz 0 3.9 3Q x .oa mummHm O O r .m. N w - oh H V 0 3 nu 1|. 3 MN d o w I. x; - 63 W 79x .zufipccsv undo useUHocH pom copooppoo .pflmoqmo HwfipnmcH :0 mowooam> pw< mo poemwx age .om magmas A.§\:\.VKV \AkxdmoNNA m\\¢\ 069m. QQQN 06a: 0 Ill..ll-l.lF|. I: lokll'l-!lu:.-hl -1- 3 I- b Q “If n EIF . m! w rhx W U .8. fl 0 m r l \CxoOVVv‘ \k‘VtGQmQ MN N0 fleeu0u«( 0 o m hd\u¢uu< Vu~w£0Lumz m 10% W... 07K \ooQCOLuxrx 0 36k poms. x x 1mm 0.9! -93- There appears to be a general decrease in deposit with increased air velocity perpendicular to the surface. Relative humidity had no significant effect on the inertial deposit of dust. The D x RH interaction was significant at the five per cent level indicating possible differences in trends between dusts. During the tests observations of the deposit on aluminum disks were recorded. These observations were made in a general manner and the percentage of deposit due to clumping was not determined. It was noted that the greater part of the inertial deposits was sometimes in the form of small clumps. The degree of clumping seemed to increase with relative humidity. In general, the micronized talc dust gave a fine deposit with some clumping at a relative humidity of ninety per cent. The red talc deposit tended to have more clumps. The standard attaclay dust tended to clump to a greater degree than the talc dusts at all relative humidities. The frequency of clumping of the micronized attaclay appeared to increase with lower relative humidities while.Attasorb appeared to have the Opposite trend. In general, the tales appeared to give a finer deposit than the attaclay dusts. The mean values of the filter deposits taken during the gravitational and inertial deposit tests are shown in -94- Table VII. It was felt that not enough samples had been taken to warrant a statistical analysis. TABLE'VII AVERAGE FILTER DEPOSITS FOR GRAVITATIONAL AND INERTIAL TESTS Red Std. Dust n Talc Talc .Attaclay ,p.Attaclay Attasorb Deposit 110 158 150 114 157 (xlO'Lgms.) Electrical Deposits Results of the tests to determine the magnitude of deposit due to electrical field forces are shown in Table XXII in Appendix III. The analysis of variance of the sphere deposit is given in Table VIII. Differences between deposits of the various dusts are significant at the one per cent level. The average deposits for the six dusts are shown in Figure (21). In this analysis the multiple range test suggested by Duncan (10), is used. In this modification of the determination of an L.S.D. value, the shortest significant range R is determined for any sample size P from R : Smr where Sm is the standard error of the mean and r is taken from.his tables of special significant student ranges. TABLE'VIII ANALYSIS OF VARIANCE OF SPHERE DEPOSIT DUE TO ELECTRICAL FORCES Source Degrees of Sum of Mean F Freedom Squares Square Dusts 5 1186.45 237.29 8.55** , Relative Humidity 2 1677.78 838.89 30.23** D x RH 10 376.55 37.66 1.36 Individuals 54 1498.50 27.75 Total 71 **Significant at the one per cent level The Attasorb deposit is significantly lower than all other deposits at the one per cent level. The electrical deposit of red talc was significantly greater than all other dusts except the micronized talc at the one per cent level, and greater than the micronized talc deposit at the five per cent level. The effect of relative humidity on electrical deposit was significant at the one per cent level. The decrease in deposit with increase in relative humidity appears to be linear (Figure (22)). The decrease in deposit from forty to ninety per cent relative humidity was forty-three per cent. There was no significant D x RH interaction. The charged cloud potentials are shown in Tables XIX and XX,.Appendix III. The analysis of variance of the ~96 ’ .AOHV cwocsc anm Coxmu neon o>m£ n no mozamb 0:5 .HmbmH acme Loo mac on» um pamsmuoac hapnmewmficmwm nos ope xcwap CH ccafiapmccs cmoss .HeboH ammo Lea m>wp 6:» an pampeuefic haquonchHm pom ohm nee CH omcaaponcd mosflw> cmons .cSOHc umsc Hmofiponam no pounce um moposaw ESCHESH< pocnsoau pouofiaaa gocH one so muamoaoc ommho>< .Hm madman 0 “T. 3.56 mes-$6. m s b. .\u 3w med .\ \ufl/ &¢\u6«hv. $$m hi\~6uhv\ s‘ Ovmn.w mmbm.¢ Om.v Hm.n mNmN.w ommb.¢ vH.¢ 0H.n NHbH.® momm.¢ mo.¢ 00.0 O¢OO.® mv¢m.v $0.0 mm.m womb.m moan.¢ mh.& #m.m \ 0N 50% so...\ hNK mm...” H @333ka NH NH Em b b DaOJtOVH-QQD .czoao compare cfi pmso mo soapfiwocea so zpflwwajm m>prHmm no pocmmm one .mm onswfim rho-:31 Maison-0. azmo 660. Om 0% 0h 00 0.“. Q? o P p P L r “I. O ION M... 'nl (ta/v.4 g) JJS‘OdJO 39V83/‘V ION Oxx fin -93- cloud potential one inch from the grounded spherical screen (Table IX) indicates significant differences between po- tentials of the dusts at the one per cent level. The very low potentials at ninety per cent relative humidity were not analyzed because it was felt that part of this drop in potential may have been due to leakage of current along the glass rod at this high relative humidity. From Figure (23) we see that the potentials for the Attasorb and micronized attaclay were significantly greater than for all other dusts at the one per cent level. There were no significant differences between.Attasorb and micro- nized attaclay, red talc and standard attaclay, and standard talc and micronized talc at the five per cent level. The standard talc potential was significantly lower than all other dust potentials at the one per cent level and the micronized talc dust potential was significantly lower than the red talc or standard talc potentials at the five per cent level. There was a significant difference between dust potentials at different humidities at the one per cent level. In general the cloud potential tended to decrease with increased relative humidity, although the standard talc cloud showed an increase in potential at sixty per cent relative humidity (Figure (24)). The significant D x RH interaction would be due to the difference in slope between the micronized talc and standard talc and the other dusts. -99 i .HopoH ucoo Lea one esu aw unopemwan hHuCaefiewcme no: one xcaan Cw reseflpopc: among was once you o>Hm no Hopoa occwcfikficmfim a pm unopoeuflc uoc chm cop CH cocfiapovzz menaw> omoze .AOHV naccsc FOLM cone» cams p go mozawm .esczam on“? weTCSOAu 80pm nocH one Hmwpnouom cwmpm>< .wm ossmfih who) opus b 0mm ope.» 03 con 08. o .16L1 cle. L, .11Y( . AYI. . “vlust>afiodxhcxmu>( 5.3» (xx‘ \ h0\h6>«< .VVW U\MN w‘ ‘eethvx esu.“ \CQ 0\nwh .me. ombn.H¢H vmmo.®OH om.v ¢m.n vmmn.mna vmbn.bOH #H.¢ 0H.n mmmw.wnH wmmw.¢oa wo.¢ HH.n Obmm.mnH momm.HOH no.0 $0.». #Hbm.bmd movm.®m mb.n mm.m a 9.“ htw , .0...‘ be ww.nn u no.mnHH n m u Em IL 8.38 \K D-aNIOdH-DCO .mnozqm mpwk vewcsopc esp 80pm nosH ego Hafipscpoa cacao no spaeaaam mpapaaem a6 poeacm are .sm .masmfia 25.6 ...-no ate-rs: ...-2:..qu on on cm Ob oh 0? o (P P b b P b “I 0 .oo\ IQQN - . room 0 - O . x l . - c -00? .06» 0 e +00% £1... aux-“553% D I 8.6.x. hers-{9422 x .8h oxek not 0 . hut-2.2x VLQVC-fim D Senna-TI». 0 Ion-Q xcxuchi \GN-QQKMRK q - voom ("W/I) 1VIJN3-lad anon NVJN -101- TABLE IX .ANALYSIS OF VARIANCE OF CLOUD POTENTIALS Source Degrees of Sum of Mean F Freedom Squares Square Dusts 5 981,672.20 196,334.44 21.67** Relative humidity l 720,055.03 720,055.30 79.47** D x RR 5 554,276.34 110,855.27 l2.23** Individuals 36 ‘326,l98.75 9061.07 Total 47 **Significant at one per cent level Filter deposits for the tests are shown in Table XXI, Appendix III. the Millipore Filter Deposits. Table I shows the analysis of Variance of There was a significant difference between dust deposits at the one per cent level. There was no significant difference between deposits at various relative humidities at the five per cent level. There was also no significant D x RH interaction. The average filter deposits are shown in Figure (25). The filter deposit of standard attaclay was considerably below that of all the other dusts. The average charge density and charge per particle for the dusts are shown in Table XIII. These values were calculated from the average voltage one inch from.the .Ho>oH ucoo Lon mco on» um unopmkufiv thCmcfikficmHm no: ohm xewan a“ nocfiapocc: emonu cam unoo pea o>«% no Ho>oH cosmofiuficwam a pm psopougap no: one pep CH tocfiapoccs madam» omonb .pouaam oceppfiofi m gauchnp soncoamam has n past on» mo oESHob csocx a Mawzwpc hp oocHELopev macs anamoaou was .mumoa coauamoaoc amoappcoam pom umsa no mpfimoaoc houaam owmpo>< .mm manmfih a/OZ ‘ uttb oxk 0.2 0.3 o! 03 o~< 02 e L . . O LHT .2...o\HQ N0< \III New. a\ 2.1:. acts-$.35. wet-5.x .\ o. «a. I“. e 0.- not owmm.bw mHmH.HN Om.v HN.n NH¢N.bm ObNb.ON vH.¢ mH.n mwab.mm Nmmm.om mo.¢ mo.n OHmm.mm va@.ofl no.0 mm.m Nmnm.vm mec.mH mb.n ¢m.m am. ...e. .w. ..u and u 3.2.x. u 3 u 5w mmdmn (LNV’DV‘LOCC -lO3- grounded screen, the average dust deposition in the filter, and the mean particle weight as outlined in.Appendix IV. There is evidently a straight line increase in charge with particle surface area (Figure (26)). The charges on the particles, although from air ions, are the equivalents of from 280 to 7680 electrons in magnitude. TABLE X ANALYSIS OF VARIANCE OF MILLIPORE FILTER DEPOSIT Source Degrees of Sum of mean F Freedom Squares Square Dusts 5 33862.11 6772.42 13.02** Relative Humidity 2 1906.03 953.02 1.83 D x RH 10 4578.64 457.86 .88 Individuals 51 26531.00 570.22 Total 71 -3:68 **Significant at one per cent level Friction Charging of Dust The results of tests to determine the magnitude of the frictional charging of the dusts used are shown in Tables XXIII and XXIV in Appendix III. The analysis of variance of the results (Table XI) indicates a difference between dust potentials significant at the one per cent level. The significant difference between replications is believed due to increases in magnitude of the frictional .mpmsa wa no“ ecn< commusm oHOpraa coma spas oaowppmp pea cmuwno emwuc>< no :oapwfipw> «when» Sue-V moth-mam. Mao-LSQ 9.0; 6.? co. oh. hr h L b P b b b b iP b b P -104- .6m chem-a Tm . .hao>fipwmon newsmSO poems» spare HH< .pmso mo maamhmnc coHaOHpa on 05$ csoac Hwoapcfiam nempmno no season as Hwfiuoouom ommho>< .bm opswah 3. . nd~Yu .Qfifl . ho. v 3m u a: at.) 00km one». .33 ooo\ con“ 6 u . . . ) 0 . OrOIOIIIIIIL 43.2%.- 0K ~o<3§< MI\MC£&‘3\ 0‘05 .Vfi.“ £0h8$£( “SR :\ 8!. tot sets-.1 .Im -lO7~ from around one-tenth to one-half of that placed on the particle by the charging nozzle. The charge per particle appeared to be proportional to mean particle weight (Figure (28)). .All frictional charges were found to be negative in sign. Field Experiments The results of field experiments in dusting are pre- sented in Tables XXV and XXVI in.Appendix III. There was no appreciable attack of fungi on either the celery or the onions. In the case of the celery, there was evidence of a slight attack of both early and late blight, but the infestation was very light and uniformly scattered through the plots. In view of this lack of infestation, no analysis of variance on the plots for evaluation of fungus control was run. However, the lack of infestation did allow for an evaluation of the effect of dusts upon plant growth. Examination of the results indicated that there might be some response to the dusts. An analysis of variance was run on the results to determine just what effects on plant growth was caused by the dusts. The field of celery was known to have a shallower depth of muck on the east side. Comparison of the over- all results indicated a ten per cent reduction in yield ~108- .cmm pmumsc accoflpco>coo a swsopnp cacao was scamccamsm nflm n pmsu one .pmsm Mo mafimncno :owpofipa 0» egg omofippcm pea mmmwzo .wm mpzwflm mascxxyagauz$$:UqurbméR\\<§hN\ o °\IQ\*nM-\ p b b b “fi\ b n n P b“ P b L b a Q 0 r\ m- -N (17'5“?) 393mg .7791;qu NVJW -lO9- on the east plots. There was also approximately ten per cent reduction in yield in the late celery as compared to the early celery. TABLE XII ANALYSIS or VARIANCE, CHARGED VERSUS UNCHARGED DUSTS Source Degrees of Sum.of Mean Square F Freedom Squares Dusts 1 309.8 309.8 8.04** Individuals 254 9786.5 38.5 **Significant at one per cent level From the analysis of variance, it was evident that there was a definite reduction in yield due to charging of the dust, significant at the one per cent level by the F test. It was also evident from the results that the various dusts increase yields to varying degrees, but as this particular phase of the work is not of concern to this project, further consideration of this variation will not be made. There was no significant increase in yield for the micronized dusts when compared with the plots of standard dithane sulfur. There was also no significant difference in yield between any of the micronized treatments, charged or uncharged, or between micronized treatments and the check. 4 -110- The plots of late celery showed a very pronounced physiological effect due to the micronized dusts, whether charged or uncharged, with the exception of the two pound per acre uncharged plot. The plants were considerably larger and lighter in color than the neighboring plants dusted with standard dusts. The results from the work on onions gave no signifi- cant differences between any treatments. Figures (29) and (30) show the visible results of charging dusts in the field. -111- Figure 29. Deposits Due to Charged and Uncharged Dusts on Onions. The area between the arrows has been dusted with the same amount of dust that was used in the other four rows, but the dust was not charged. The high loss of dust into the air behind the dust can also be seen. -112- Figure 30. Dust Deposit on Celery -113- INTERPRETATION OF RESULTS For purposes of discussion, a summary of the experi— mental results from the laboratory tests is shown in Table XIII. In addition to the information presented in the preceding section, the particle size analysis is further broken down to show the percentages of particles equal to or greater than 11 microns diameter, the percentage equal to or less than three microns diameter, the percentage of particles one-half micron in diameter, the percentage of particle weight contributed by particles equal to or greater than 11 microns diameter, the percentage weight contributed by particles equal to or less than three microns diameter and the percentage of particle weight contributed by the particles one-half micron average diameter. Gravitational Deposit Examination of the percentage of particle weights of those particles above 11 microns in diameter may serve to explain the difference in gravitational deposits between the various dusts. Recalling the rapid increase in terminal velocity with diameter of particles under a gravitational force field as shown in Figure (2), one can readily see ‘why the standard attaclay, with 72.6 per cent of its weight due to particles above 11 microns diameter, has a higher -ll4- .ao\mpao> o.mmm o.mom m.oma m.m>a m.mma m.woa omnwaOJNv mpao> «.mm o.eem ~.Hm H.Ha n.56H “.mefl 1.ea HI ownpaop nonpofinh maaop mn.6me mm.moo mm.mme ma.aee n.0mm “.AAN 1.aa H omupao>. no .maw 4-0Hn o.oH ~<.Hm mm.mm mm.¢m ne.e~ mm.am pHmoaoe ohonmm .mam euoax «4.0ma mo.maa wo.aoa o.mma «A.moa mm.ooH pammaoe opafla .mam euoau A.mm o.ee 4.4m H.46 4.06 --- pamoaoe ~32qu .mam Suede mm.ma o~.ma 46.44 mw.oa eo.a~ --- samenee.>eue maa\aw me.m me.m me.~ --- m.~ w.m spanned oaofipusm m.ae NH-oHH ~m.mm mm.m~ m¢.mua mm.mem ma.amm «H.054H semen; oaoapnem N.ae wuoan AH.HH H~.HH H4.He o~.~m so.~o mo.eea umnu.unsm oaoaanam .ae euoau «4.4 mm.H ~.~ mo.~ ee.m eu.¢ Renegade oHoHpnem «pane anomapp< .p<.a .p<.e¢m case com case a case .epm page Amesaab decay mafibmmm ho MM o.m« «.63 «.o« 5.5« 0.3 4.3 8335 mm 336 6.8633. .64 a ...-4.3m 32.. com Sea a 35.3w page L.) ILIIIIIIII. vegan-Hoe HHHN age ~116- deposition rate than the micronized attaclay or.Attasorb, of which only 33.4 and 40.3 per cent of the dust weight respectively was contributed by particles greater than 11 microns diameter. The same trend is noted in the talc dusts where the micronized talc, with 7.76 per cent of its weight being contributed by particles above 11 microns diameter, had a higher deposition rate than the red talc which had 3.73 per cent of its weight contributed by the large particles of dust. The greater deposit of standard attaclay compared with the tale dusts may be due to the more excessive clumping of the attaclay particles or to errors in particle size determination. The trend toward decrease in gravitational deposit of standard attaclay with increase in relative humidity may have been due to impingement and adherence of part of this dust on the baffle plate or duster fan housing resulting in a lower cloud density. Examination of the filter deposits from the inertial and electrical tests at higher relative humidities indicates a similar trend for this dust. The rather constant gravitational deposits regardless of relative humidity would indicate little change in dust density or particle cohesion. -117- Inertial Deposits The tests for inertial deposits of dust indicate, as did the results of Brittain, a decrease in deposit with increase in air velocity. The initial increase in deposit for the talc dusts might be interpreted to mean that the erosion of these dusts from the surface was not so severe at the lower air velocities. These results would tend to corroborate the theory that air velocities high enough to cause turbulent motion of the particles would result in no increase in deposit with further increase in air velocity. Some of the reduction in deposit may also have been due to a "sand blasting” effect of the larger particles although results of Brittain's work showed a similar decrease in deposit for plates coated with petroleum.jelly. The higher inertial deposits of the talc dusts may have been due to the greater density of these dusts since particle size analysis would indicate the deposit should have been somewhat similar to that of standard attaclay. If erosion of the dust was taking place, these dusts may have been.more adherent to the aluminum surface. The relatively high inertial deposits of the micro- nized attaclay and Attasorb dusts would indicate that either the smaller particles have not yet reached a high Reynold's number in the inertial force field, thereby allowing them to deposit out faster, or that these smaller particles are less subject to erosion. -118- The absence of any change in inertial deposit with relative humidity would tend to validate the experimental method. Filter Deposit The filter deposits were somewhat reversed in trend to the gravitational deposits for the dusts. Where the standard attaclay dust had been highest for gravitational deposits it was found to give the lowest filter deposit. The Opposite was true of the micronized attaclay dust. The various individual differences are probably due to the differences in particle size distributions and experi- mental error. The filter deposit, because of the fairly rapid air flow through the membrane filter, probably gives a fairly accurate description of actual particle density in the air. Although all dusts were blown into the chamber at the same number of grams per:minute the filter deposits indi- cate actual differences in cloud density. These differ- ences may have been due to deposit of dust in the duster fan, along the tube walls or on the baffle. Cloud Potentials Due to Charging The cloud potential due to a charged dust suSpension in air was found to be a direct function of the particle surface area times the number of particles per unit volume. It was expected from theory that the charge per particle ~119- should vary in direct pr0portion to the particle surface area. The verification of this was seen in Figure (26) where the charge per particle, plotted against the mean surface area per particle, gave a straight line relation- ship. Sphere Deposit With the exception of the.Attasorb and red talc the Sphere deposits were not significantly different in weight. Attasorb, while appearing very similar to micronized attaclay in almost every other reSpect, gave a considerably smaller deposit on the spheres. Red talc on the other hand, gave a somewhat higher deposit than the other dusts. A.possible explanation of these results may be in the resistivity of the dusts. Brazee (5) found that the resistivity of red talc was quite high but values of resistivity for the other dusts used were not determined. If the resistivity of the red talc was lower than that of the other dusts, then this dust might be expected to discharge to the grounded sphere more quickly than dusts of higher resistivity. The dusts of high resistivity would, by their holding of charges on their surface, set up a charged field about the surface, somewhat similar to that of the zeta potential on colloidal particles. This charge would then resist further deposit of dust. -120- If this be the case, then the Attasorb may be a dust of still higher resistivity, thereby further decreasing the electrical deposit. The similarity in magnitude of deposit for the other dusts, even though being of quite different particle sizes and particle size ranges, is interpreted as possibly a verification of the theoretical analysis in which it was found that rate of electrical deposition would be independent of particle size. Friction Charging of Dusts The charge placed on dusts through collision with other particles and the dust fan and tube was found to be of a smaller magnitude, opposite in sign but somewhat comparable with that placed on the particles by means of a charging nozzle. The charge per particle appeared to be somewhat correlated with the weight per particle as shown in Figure (28). Since the larger dust particles will move further with respect to the air stream in the eddies and whorls of the air blast, they may therefore suffer a greater number of collisions, thereby receiving a greater friction charge. This would indicate that, all other conditions being equal, if there was friction charging, dusts of larger particle size would have a higher deposition rate than dusts of smaller particle size. -121- The charge per particle for the micronized attaclay was found to be roughly five and a half times that of Attasorb even though the particles were found similar with respect to particle size. There is evidently a difference in particle surface characteristics due to the different methods of grinding these fine dusts. Charging of Dust in the Field Since there was no attack of the field plots by plant diseases no information can be obtained on the merits of charging dusts or the use of finely ground dusts in the field. There was a visible physiological effect on the celery plants due to the dust. The nature of this effect was not determined. Comparison of Terminal Velocities of the Particles in Various Force Fields The question of which of the force fields considered is most important in application of dusts, sprays, and fogs is almost impossible to determine because of the many plant configurations encountered. Gravitational forces would cause deposition on the top surface of leaves, proportional to the projected horizontal area. If this projected area and rate of particle fall-out (determined from the size distribution of the particles in question) are known a fairly accurate prediction of deposit and drifting may be determined. -122- Inertial forces act only when the particle susPension is deflected. Since the plant leaves are arranged such that some leaves will be parallel to the air stream, some perpendicular to the air strewn and the majority at some angle to the air stream.the calculation of what deposit might be expected is somewhat more difficult than the deposit due to gravitational forces. Electrical forces are highly dependent on the configu- ration of the grounded plant system. Higher potential gradients could be found for onions than for the interior of closely arranged celery leaves for the same charged cloud density. For a leaf with no grounded objects in the immediate vicinity, the gradient may be of such a magnitude as to cause a fringed deposit of the leaf edges, while the interior leaves of a bushy plant would allow for but small potential gradients. Thermal forces can cause resistance to deposition on leaves which are exposed to solar radiation. They will be unimportant on shaded leaves or on cloudy days. Large particles will probably be little affected by a thermal force field. Comparisons are further complicated since all forces must be added vectorially. Gravitational forces will always be downward, inertial forces will be toward the surface, electrical forces will be toward the surface and perpendicular to equipotential lines, and thermal forces -123... will be normal to the plant surface, being either toward or away from the surface depending on the temperature gradient. As a rough means of comparison of the relative imp portance of the forces in deposition, Table XIV shows particle terminal velocities caused by any one force acting independently, which might be expected under field conditions for three particle sizes. summary of those calculated in previous sections. ParticlelParticle TABLE XIV TERMINAL VELOCITY OF PARTICLES IN FORCE FIELD (CM. /SEC .) A1: rt The values are a Diameter Density Gravitationa Inertgzrcgiectrical Thermal 2.p. 1 0.012 0.495 44 0.0035 10 p. l 0.30 12.4 44 0.0016 20 p. l 1.20 49.5 44 0.0016 2 p. 2 0.024 0.99 22 0.0035 10 p 2 0.60 24.8 22 0.0016 20 p. 2 2.40 99.0 22 0.0016 The values shown.must not be construed as a direct comparison of what deposit one might expect since any one particle will be acted upon by all of the forces at all times and the actual deposit will be due to the vector sum ~124- of these forces. They do indicate however, that electrical forces will be very important in the deposition of small particles. Thermal forces will be relatively unimportant and gravitational and inertial forces will be very important in the deposition of the larger particles. -125- CONCLUSIONS A.review of literature has indicated that, in general, reduction in particle size of pesticidal dusts increases their effectiveness against insects and fungi. The smaller particles were also found less subject to erosion by rain. Analysis has indicated that, where the degree of control of insects is a function of particle surface area, the dosage rate of insecticide may be reduced in direct proportion to reduction in particle size, if the efficiency of application remains constant. The equation P = 1 - (l - fiLln has been derived with which it is believed the effect of changes in application rate and particle size on control of plant diseases can be calculated. Upon the determination of the effective ’area of control at of any specified particle size the probability of control P can be determined for any arbitrary area A.and number of particles n. If the mechanism.by which the particle acts to effect control is determined, changes in control through changes in particle radius may be determined. Where this mechanism is a particle surface area phenomenon, any change in particle size will allow for a proportional change in application rate at any given percentage of control. Where the mechanism.is through a diffusing of the chemical along the plant surface or ~126- systemically through the plant, at will be prOportional to 1r(r+a)2 where r is the particle radius and a is the distance of protection of the particle from the particle surface. It has been found analytically that the effects of Brownian motion and coagulation may be neglected in the consideration of conventional dust and Spray applications. It is doubtful whether they would be of importance in conventional fogging applications. If the terminal velocity of a particle moving through air is used as the criteria of deposition rate of these particles, analysis shows that the rate of fall of particles below five microns in radius is so small that drifting and eddying of air currents almost eliminate gravitational forces as a means of obtaining a deposit of these particles without excessive drifting. Analysis has shown that most inertial deposition of dust will take place under conditions of laminar flow for present field conditions. The rate of inertial deposit, as found theoretically and experimentally, will increase linearly with velocity for Reynolds' numbers below two. For high air velocities causing turbulent particle motion relative to the air stream.it has been found that there will be no increase in deposit by further increase in velocity. Experimental results indicated a decrease in -127- deposit with increased air velocities. This was believed to be due in part to erosion of the particles from the surface. , The magnitude of the inertial deposit of fine dusts was found highly dependent on the adherence of the dust to other surfaces. The inertial deposit ranged from over five times that due to gravity to twenty per cent less than that of gravity depending on the nature of the dust. Air veloci- ties of from four to thirty-four miles per hour perpen- dicular to the surface were used. .Although inertial forces will cause considerable deposit of large particles they will not be of particular value for particles below two microns in radius. Relative humidity had no significant effect on either gravitational or inertial deposition. Theoretical analysis indicates that the terminal velocity (therefore deposition rate) of small particles is independent of particle size for charged particles moving in an electric field.~ Experimental results from four dusts appeared to confirm this analysis although two dusts showed considerable deviation. It was thought this deviation may have been due to dust resistivity. The charge per particle applied by the charging nozzle was found to be directly proportional to particle surface area. -128— A linear decrease in deposit was found for increasing relative humidity. A.corresponding decrease in cloud potential indicated this decrease may be due to decrease in efficiency of the charging nozzle. Charging of dust particles by friction was found to vary from one-half to one-tenth of that placed on the particle in the charging nozzle. The charge per particle was found to vary linearly with mean particle weight. .A difference in charge per particle of nearly 81 per cent was noted on two dusts of similar particle size distri- bution and density. This difference was believed due to changes in surface characteristics brought about by the different grinding processes employed for the two dusts. Thermal forces were found to be of a relatively small magnitude even though the plant surfaces have been found to be 20° 0. above ambient air temperature in direct sun- light. These forces would probably have little effect on the deposition of large particles but they may be of importance on deposition of particles of one micron diameter or less. .Although gravitational and inertial forces may play a large part in the deposition of particles above ten microns diameter electrical forces must be relied upon to obtain deposition of the more effective smaller particles within a short period of time under field conditions. -129- Experimental evidence indicates that the deposit rate of particles charged by friction would be greater for larger particle sizes. In view of the magnitude of frictional charge observed, it is believed that under present day methods most of the deposition of the smaller dust fractions may be through electrical field forces. Frictional charging is very erratic and is subject to relative humidity and the chemical and surface properties of the dusts. Two dusts may charge Opposite in sign and, upon mixing, the resulting frictional charge would cause coagulation and nullification of the electrical field forces. ' Theoretically there should be no decrease in ef- ficiency of deposit with decrease in particle size for particles charged in an ionized field. Controlled charging of the dust should result in greatly improved protection of plants from insect and fungicide attack, while allowing considerable reduction in pounds of material per acre for effective control. APPENDIX I CALCULATION OF 0‘ -131- APPENDIX I CALCULATION OF 0‘ The value of at , the area of control of any one particle, may be determined from experimental results in the following manner: Since P : 1 - (1 - 31-)“ Where P the probability of control (lOOP : per- centage control) and n z the number of particles in an arbitrary area A, then 1 - P 3 (l -€i¥1’ or “fit—531'?- Transposing, f—zl- l-P ora:A(l-nl/l-P). """ 1'38 From the results of McNew and Burchfield in the application of dichloronaphoquinone of .81 microns average particle radius, a spray concentration of 64 parts per million gave .35‘pg/cm2 of ingredient, or 95,600 particles per square centimeter. -132- Therefore, 1 ppm.= 1450 particles/cm2 : .145 particles/100 micron square. Their results indicate that at a spray concentration of 50 ppm.(7.25 particles/ 100 micron square) the control was 92 per cent. Substituting into equation (35), It! (100 x 10")2 (1'2h292) (1 x lo") (1 - 7125/7655) (1 x 10-4) (1 - .705) s 2.95 x 10'5 cmZ/particle. APPENDIX II DERIVATION OF POTENTIAL DISTRIBUTION IN CHARGED CLOUD BETWEEN TWO PARALLEL PLATES -134- APPENDIX II DERIVATION 0F POTENTIAL DISTRIBUTION IN CHARGED CLOUD BETWEEN TWO PARALLEL PLATES A plant system is essentially an arrangement of grounded surfaces. The system.may range from a series of grounded tubes or fingers (onions) to a somewhat ordered series of parallel leaves as in beans or tobacco. Very few of the leaves are exactly parallel as they are ar- ranged to take maximum advantage of the light. Some insight toward what potentials and electrical field forces might be expected in a plant configuration may be gained by considering the leaves to be grounded parallel disks. In this idealized system it will be assumed that the disks are of such size as to shield all exterior electrical forces from the region between the disks. Only the field forces due to the charged cloud between the plates will be considered. It is recognized that in field conditions there would be considerable electrical field effects from.the charge distribution radially outward from the axis of the disks. The charge distribution between the disks satisfies Poisson's equation where: VZV : 4.717" where‘V : the potential and f’= the charge density. -l$5- Considering the system to be a cylindrical charge distribution then along the z axis (Figure (51)), by symmetry, 2 - a'y : -4710 v - ____ ‘7 .92' If (’13 assumed constant the following boundary conditions must be satisfied: .2 = a, V : O and at 2: 41.” 2 2 a2. Since VZV” 42,79: 0 is an ordinary differential O. equation of second order independent of x and y, then a’v : -4 7”" Integration with respect to z gives at! : 41/7/02 f C” where C, is an arbitrary constant of 32? integration. But at/ : O at 2: 2, therefore 32- 2 f -477/ong': O, or, Cl: 4”,; : 2,,fl, Then av : -4.m'3f277/°a. a: Integrating again with respect to z yields 2 V : ‘47/{5 {277,029.wa where C2 is an arbitrary constant But V : O at Z = 0, therefore CZ : 00 Therefore V - -21fl£2 2mgza .-.- 27.4(a2 - 32) gives the potential at any point along the axis between the disks. The electrical field intensity E is then E - - 421 = 477/‘2-21’Ia. .mxmfim gownsOpo cooapom csoac HaOHpccfiazo cowpmzo m no mnx< was waoaa acapspfiepmac omaaao one aefipcopoa .Hm cheese ..SszukoQ oz< 3.6K 3.3.x adrfioud _ 3 mkmx on 2. ON -136... APPENDIX III EXPERDIENTAL MSULTS DUST PARTICLE SIZE ANALYSIS -l38- TABLE XV (Number of particles) Dust Micronized Talc Micronized Attaclay Red Talc Standard Attaclay Attasorb Standard Attaclay waH bwww #WNH mrwNH waH waHRqflRMflms Particle Size Range (Microns) 0 cu o 0 <3 279799773?” r# N ~¢ ~o u: :3 31, g: 75 83 77 36 22 6 7 12 l 68 80 61 37 14 9 7 4 1 57 51 29 28 19 14 9 7 7o 53 33 17 16 5 5 16 2 172 140 28 8 19 2 3 3 209 223 42 7 4 l 2 146 144 41 13 6 3 266 223 41 15 1 2 88 145 76 46 17 10 4 l 148 117 29 20 14 1 4 2 132 96 49 37 18 18 6 13 2 4O 69 44 21 s 9 5 6 39 79 37 22 13 7 3 5 116 73 34 21 8 8 3 4 1 75 78 16 10 8 6 6 3 98 88 37 9 5 4 2 6 92 77 24 15 5 4 3 2 139 143 30 12 4 1 129 146 32 7 5 1 1 179 171 39 9 2 3 180 156 40 11 3 2 l 21 59 53 26 18 15 6 l4 2 1 36 57 55 19 14 9 9 9 1 1 38 30 27 22 10 ll 8 l4 5 3 43 46 47 18 15 10 5 13 4 2 -139- TABLE XVI GRAVITATIONAL DEPOSIT 0N ALUMINUM DISK (110’4 grams) Dust Replica- Relative Humidity (Per cent) ti°n8 40 60 80 Micronized Talc 1 ll 25 62 2 20 21 30 3 20 21 17 4 19 21 22 5 22 21 23 6 21 20 29 7 22 16 26 8 24 19 21 9 18 19 22 10 18 25 18 ll l9 19 20 12 15 22 22 13 21 21 l4 l4 16 23 23 15 13 11 l6 16 20 ' 18 24 Av. 18.7 20.1 24.3 Micronized Attaclay l 42 12 8 2 13 15 ll 3 12 17 7 4 11 10 ll 5 7 14 47 6 8 10 8 7 8 21 23 8 6 9 l5 9 9 l2 14 10 ll 8 4 11 10 9 5 12 11 10 10 13 10 13 7 l4 9 10 12 15 10 10 8 l6 l9 13 11 Av. 12.3 12.1 12.6 Red Talc 1 30 32 ll -l40- TABLE XVI Continued Dust Replica- Relative Humidity (Per cent) tions 40 60 80 Red Talc 6 16 22 7 (cont.) 7 16 27 12 8 15 26 13 9 15 20 13 10 l6 16 15 11 13 18 11 12 13 12 13 13 15 23 18 14 16 l9 17 15 23 18 13 16 18 20 19 Av. 18.3 19.9 12.4 .Attaclay 1 45 55 56 2 29 54 69 3 50 60 43 4 59 96 45 5 94 47 17 6 85 58 53 7 50 45 60 8 47 51 15 9 31 19 29 10 43 29 32 ll 46 38 31 12 54 44 31 13 34 48 45 14 49 34 26 15 37 33 22 16 39 36 44 AV. [+905 4607 3807 Attasorb 1 l7 16 10 2 l6 9 15 3 19 22 11 4 13 17 8 5 15 10 8 6 ll 14 ll 7 15 13 9 8 15 l3 l2 9 9 14 ll 10 ll 15 ll 11 12 11 10 12 7 13 15 13 14 11 10 l4 18 12 11 15 11 ll 13 16 10 12 10 -141- om mm mm «H cod es em a on mm mm SH .>< mm o co mm m Ham Hm aw ca and on an «H «A on on ma 4 am am am m Had mm 4H AH 62 mod mm mm m Hm mm m ma mm mm mm 4 we mm mm ma m cm mma on ma wad em mm A mas on «n ma a seaoepp< GONHGOHOHE and «as am an end em med we med «8 as an .>4 am A 04H 25 e on mu m Hma mad cod on mod on «a do mm on as ea 4 Add «0H moa om mm mm mm mm add so me Hm m nod mod mo AH we and «ma on can aNH em an m maa and we om omd we wma cm mwa “ma mm mm a cane W ucNHGOHOHE 000m ooom con 0mm ooom ooom och can 000m ooom cos c m. paooaos use spfiooae> na< ,pfiooam> and n om 8 3 m. p25 -pqoo away epfieaasm opfipeaom m Amsdnw «noa NV MmHQ ZDZHEDH< Zo BHmomHQ HdHBmQZH HH>N Hfimda -142- Ns No Ho oN on as os so No mm on oN .64 Hm oN S N on as ms mm oo mm no No s so ss No no so Hm Ns Hm so on NN om m os oN on sN no no on Hs mm on on No N oN sN NN oN ms Hs so 5N sN on on HH H anemones Ns on on MN os ms on so on oN NN sH .52 NN oN «N on on oN oN oN mN sH sH HH s oH Ho os oH mN so oH MN sN oN oN o m Hm on oN oH so os ms os ms Ns om oH N so oo no so on oo oo Hs ms oH oH oH H ooHoooos onoooopm NNH Ho Ho sN NNH No HHH Ns NsN oHH no HN .ss ms oH mm o mo o no ooH N on so o Ho so ooH o oo oo ooH oN ooH ooH so on moH oo oo sN s ssH so NHH nH NHH sMH oo oH sNH oo oNH mN m osH oo NH oH HoH oo ooH on ooN oHH No HH N mo Ho HoH os «NH Ho sHH on oos soH ooH «N H oHoe sum ooom oooN ooo oom oooo oooN ooo own ooom oooN ooo on mm opHooHoo NH< epHoOHo> HHH soHooHoo 2H4 m P oo oo os m omen Loose Homo soHoHaom osHpoHom m 60.3.3. pdoo HHBn MAM-we TABLE XVIII FILTER DEPOSITS FOR INERTIAL TESTS (x10’4 grams) Dust Replica- Relative Humidity (Per cent); tions 40 6o 90 Micronized Talc 1 104 45 68 2 144 180 106 3 120 Ave. 122.7 112.5 87 Micronized Attaclay 1 158 104 93 2 219 232 145 3 157 Av. 178.0 162 119 Red Talc 1 145 118 116 2 189 193 166 3 126 Av. 153.3 155.5 141 Standard Attaclay 1 129 72 71 2 185 69 . 160 3 110 Av. 141.3 70.5 115.5 — i TABLE XVIII Continued -144- Dust Replica- Relative Humidity (Per cent) tions 40 60 90 Attasorb 1 118 72 241 2 170 100 168 3 156 155 150 4 172 157 150 5 166 Av. 154 142 177.3 Micronized Attaclay l 43 20 12 2 26 14 15 3 30 20 16 4 23 24 14 Av. 31 2O 14 Attasorb 1 15 16 12 2 21 17 9 3 21 16 15 4 16 16 18 Av. 18 16 14 -145- ONN com mom oom mom mom bsm Hmm Non msN s osN mom mom Nmm omm msm Nsm st mom osN m OHN 0mm msm Hmm omm mom 0mm mmm on moN N ooN wNm ONM HNm HNm NHm oom omN mwN omN H 00 o oNN sNN oNN HsN HmN OmN NsN OMN CON **s ooN own Nom omm Nos ham mom 0mm owN OmH m ooN mNm ONm OHM NNm mNm osm mNm «Hm omN N OHN ooN sum mom ooN mwN moN ooN NoN «ON H 0s OHNB ooNquHOHE omH mON OHN oON HHN HHN OHN mON mmH omH s ooN NoN owN mmN moN mom omN msN mNN omH m «ON HNN CNN mNN mmN mHN NHN ONN NHN ONN N 0mm mwN th omN moN NoN omN msN osN osN H **om OHN smm ssm msm com Hom mom msm Nom osN s OMN mom mom sum mom own Hum oom smm ooN m ooN Nom mom Hem oom oom mom mom mmm ooN N ooH wwN HoN com com mom mom omN CNN mmN H 00 00H ooH osH mmH smH soH ooH ooH OmH osH s 00H ooN 0mm msN ooN NNN mmN ooN omN mNN m o mom ONM on osm 0mm mom 0mm mNm N o msN mON mNN NNN NMN mmN OMN mmN mmN H as oHoB opooompm OmH mmH ONH mOH om on oo ms om «H Assoc Romy mooHp opHoHsom Amooooomv QaHe nooHqum oproHom pmsn ompon omHsnonpo mmoHc: noouom oooodonw Sony nouH ooovononm HdHBZMBom QDoHo BHmomma H¢0Hmeomam MHM mumdfi -146- ooN mHm on sHm on mom Nom oom ooN oNN s ooN oom on mNm omm omm Hmm me ooN oHN m ooN msm omm oom Nmm oom smm son omm ooN N ooN st Nom ooN HHH on HNm oHH oom ooN H oo oNH ooN NoN ooN mom mHm oHH HHm ooN osN *s ooH Nom HoN ooN ooN omN ooN som NmN oHN 2m osN HHm NHm mHm HHH oHH on NHm ooN ooN *N ooN oos oNs oos oms oos oos oos oom oos H os soHooopo deUGwpm omH omH mmH mmH NmH oNH moH oHN ooN mNN s NoN mom mHm oNH NNm HNm on mom ooN ooN m on own Hmm Nmm oom mom mom mom omm ooN N ooN ooN ooN ooN soN moN ooN ooN ooN osN H 2*oo oHN mmm mmm msm ssm Nsm mom sHH ooN oNN s moH HNm oNH mom Non HNm me ooN ooN oHN m ooN Non mom Non mom Non mom «on com ooN N ooH ooN soN oom ooN HoN ooN ooN ooN osN H oo omH ooN osN osN ooN NoN ooN oHN oHN omH *s ooN on HoN ooN ooN ooN ooN ooN ooN ooN *o *mNN *mmm *mom *NHM *Nmm *mom *0N4 0&4 00¢ *ONN N ooN oos oos sms sos mos oos mos sms mos H os oHoa oom ooN NoN ooN soN ooN ooN ooN ooN osN oHN s NNN ooN ooN ooN ooN ooN ooN ooN ooN NsN m mHm osm omm mom HNH mom Non mom Non ooN N ..ooooo osN ooN ooN NoN ooN osN NsN osN NmN oHN H 2*oo oHea UONHGOHOHE omH mmH oNH moH oo oo oo ms on, mH Assoc mom. mooHo ooHoHaom Amonooomv 259 nooHHmom oproHom poem condeoo NHN mam¢a -147- oom st oHs mHs sHs NNs mom 0mm Cmm MNm *s Com oos mss mos HNs CHs mom com on Com *m *Oom *Nm4 *O©¢ *de *N©¢ *NBQ *mOm *MNQ *mOm 0w¢ N Csm Cmo CNC CNC CNC 0mm 0mm omm Com CCm H 04 nnomopp< NsN mom CNN moN NoN ooN HCN CCN osN CMN s NCN ooN moN CCN mwN NmN HoN CNN mmN mNN m ooN NmN u mbN NmN CNN ooN NmN mmN CNN N CNm omm mom omm Nmm omm Nmm Hsm Non CNm H 2*oo Com Cos mos Nos oss Cms sCs mom CCN CNN s CHs oos Cos mos oss Css Css CHs Com CCN m Cms NCm sms mos Nom CCm CCo mos mms Cos N CNm Nos oos mos, mos mos ass CNs oom mNm H oo CNm mHs oHs Cms HHs mos Nwm mom Com ooN *s Com oos oos Cms mos mss HNs osm msm CNN *m *me *mHm *hom *mOm *mOm *OHm *00# .l I 0w< N Cos oso CCC Coo Coo coo Cwo Cso omm Cms H Cs OHoB . ooNHnoHon NmN CNN ooN ooN moN ooN CmN ssN mNN CCN **s CHN CmN CmN CNN CNN NNN CNN HNN mHN CoH **m CmN NoN mmN osN mMN msN mmN CNN mNN oCN **N A.pnoov ooN «Hm mHm on moo oos oos oos mom ooN ***H oo soHoooos unwooopm CmH mmH CNH moH Co on CC ms Cm mH Apnoo Mom- 11. oooHo soHoHsom Amooooomv oaHa nooHHmom oproHom amen CozaHpnoo MHN Mflmdfi ~148- .ooonom Coonsonw on» song nooH moo owmpHoo on» mopr Cm.o hp ooHHmeHda omopHop mHnB .noonom song monqu NH ooonm 59H; omen mosHo> HH<*** .oooeom Coonsonm on» Souk noaH moo ommpHop on» mopHm oo.s he CNHHCHNHCE owproo mHsB moo owopHop esp .CNNHom Bonn nonH .QNNHom Bonn mooooH pszo ononm anB coon modHob Hd<** .ooonom esp song oHHa mopHm mom.N so oooHpHo ooooHos mHoo uHonnooo ononm an3 Comp mosHo> HH4* mmN soN uoN ooN NoN ooN son NHm NNm NoN s oom mNm on on HoN ooN ooN NoN osN oHN m osN ooN HoN ooN ooN ooN HoN ooN NsN oHN N ooN on on som ooN oom «Hm mom oom NoN H **oo osm oss Nss sms oNs mHs Non mom on ooN s mom oos oos oos Nos oos oos woo woo ooN m 0mm Hss wms oss oss oss Cos mHs omm «Hm N oom mss Nos oos mos oNs oHs oHs mom oHH H oo L.poooo Dnomwppd. omH omH oNH moH oo oo oo ms cm mH “some Nomv mooHp .opHoHeom Amooooomv oaHa -ooHHoom opHooHom ‘ pogo Coanpnoo NHN mflmda —l49- TABLE XX AVERAGE POTENTIAL ONE INCH FROM GROUNDED SPHERE ----- Charged Dust* Dust Replica- Relative Humidity (Per cent) tions 40 6O 90 Standard Talc l 211 297 52.8 2 297 350 44.1 3 253 342 51.8 4 150 320 40.5 Micronized - Talc 1 264 302 63.6 2 314 323 53.6 3 338 323 52.2 4 500 328 49.2 Av. 354 319 54.7 Red Talc 1 437 274 49.6 2 678 345 72.7 3 650 291 61.0 4 547 307 39.1 Av. 578 304 58.1 Standard Attaclay 1 416 304 72.5 2 703 345 48.2 3 655 298 44.2 4 660 285 51.0 Av. 609 308 54.0 Attasorb l 681 423 67.7 2 1101 489 52.8 3 955 400 55.0 4 886 388 52.2 Av. 906 425 56.9 Micronized Attaclay 1 630 408 61.3 2 1038 412 49.6 3 917 395 57.0 4 931 384 59.7 Av. 879 400 56.9 -150- TABLE XXI FILTER DEPOSIT FOR.CHARGED TESTS Dust Replica- Relative Humidityg(Per cent) tions 40 60 90 Standard Talc l 152 151 158 2 159 172 118 3 166 187 182 4 160 162 157 Av. 159 168 154 Micronized Talc 1 164 145 182 2 177 163 163 3 182.8* 156 166 4 185 169 136 Av. 177 158 162 Red Talc 1 78 155 154 2 137 148 145 3 113 150 168 4 132 145 131 Av. 115 150 150 Standard Attaclay l 71 126 103 2 92 130 98 3 34 105 110 4 167 91 108 Av. 104 113 105 Attasorb l 127 173 180 2 145 179 173 3 172 181 160 4 215 201 159 Micronized Attaclay l 86 151 130.8* 2 143 181 143 3 172 182 145 4 183 140 147.6* Av. 146 164 142 *Missing values calculated by method suggested by Baten (2). -151- TABLE XXII ELECTRICAL DEPOSIT ON ALUMINUM SPHERES (x10‘4 grams) Dust Replica- Relative Humidity (Per cent) tions 40 60 90 Standard Talc l 16 28 23 2 26 19 12 3 37 22 17 4 14 29 19 Av. 23 24 18 Micronized Talc 1 22 31 22 2 30 18 14 3 32 28 15 4 37 28 19 Av. 30 26 18 Red Talc 1 33 35 26 2 37 19 18 3 36 37 19 4 34 39 22 Av. 35 33 21 Standard .Attaclay l 30 22 15 2 35 20 9 3 32 28 14 4 28 28 19 ‘Av. 31 25 14 -152- osN mom NNm mom osm Non oHs mom oos oomm s oNN on NHm mom osm mom oom men oom ooN m omm oos mos on on mos oNs oss oHs oom N - - u - - ooH omH ooN ooN osm H NoHoooos dewapm oHs oos mos mos mos Non oos oos oNs oHon s mos Nos oos oss oos oos mos NNs oom oHooN m on oom mom can own ooss oHooo oNs oom ooN N oos mos oss oos oHoos . oos mom on ooH H OHoe oom ooN owm mom son mom Non Non oom on NooH s ooN Nms oHs oos oNs oNs NooH - on smom m omm mos oss oms oms oHs oos oos mos smoN N ooN oHN Noom oom oos mos smss - . on H 6Hoa UCNHHOHOHE mom mom mom oom mom own «on NmNs mss smsm s mom oos oos oms oos mHs mom mom mom sooN m oom Hos oos oos Nos mos mos oos mms scom N osN mom mom mom mom omm oom oom oom oom H oHoa s onooaopm ooH ooH oNH ooH oo no oo ms on, mH Amoqooomv oEHB meHp some NoN oo .. spHoHaom opHooHom -ooHHoom pogo ZOHBOHmh Mm nucm4mo QDOHO BmDD ho HdHBZHBom HHHNN MHm4B “H 4“! .oopon omHsnonpo mmdes someom oooqoonw no nopnmo pm omen modHNb HHd -153- .Hopomo on» pm HoHpoopom esp no ous> esp mo>Hw CN.H moaHp osHo> mHne .noonom oooadonw on» Sony mononH cop ooxop mquomoH peosdomnom HH on» mo>Hm NN.H mmaHp ousb mHnB .omosom oooosonm onp Sony monooH HHm oozes mmoHooon poosuomnsm HH on» mowa ms.N moaHp osHo> mng .noonom oooodohw onp Scum monooH poo“ some» mwsHooon pcoooomnsm HH can mmon mN.s moaHp oon> mHnB .ooonom Coonsonw esp Sony monooH osp more» mquowon voodoomnsm HddN .Hopooo map no HoHpoopom 029 no osHo> one moon no.5 mosHp osHo> mHnB .soonom Coonsonm map Sony nocH moo coxop mmcHoomn voodoompom HHd H ooH osN ooN ooN ooN on mNm oom msm sooN s oNH oHN ooN ooN ooN ooN «Hm oom soos oos o oo ooH oOH oNH omH ooH oHN ooN NoN ooN N ooN ooN soN oom ooN ooN oHN oNH ooH . H onomopps omm NNs oNs mos oNs NHs mom oom on ooN s ooN st oHs oNs mos oms HNs oHs osm HmoN m ooH omN on Nsm own son Non mom own HmsN N ooN oNN on omm osN Nom oom Hoes - oos H aoHoooos . ooNHoonon ooH oHH oNH ooH oo no oo ms on, «H Amooooomv osHB mQOHp some mom oo .. soHoHsom opHooHom nooHHoom soon '1 I, I I llllIl Codqfipdou HHHNN mHmda pOHm wdemHE* N.Hss w.HHoN 5.0mm o.mmm m.moNH m.oooH .>< H.oCo H.ommN o.mHo m.Hom CmsH m.owsH s “6., sooo s52 mom oos“ HéooH Hoso m 1.. :5 THEN oxss Nsom 03o EH: N s.NHN o.ooNN *mnH o.mos H.msoH m.NnN H hdHospp¢ CHMB anomoppd ooNHqOAOHE hoHoopp< .opm oHoB Com ooNHsCHOHE oHoa .opm quprHqum page pdoo Mom 00 II hgfidfiadm 0>dedom GZHUm4 >HNN mqm