UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The removal of smokes and mists Guthrie, David Alan 1955

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1955_A7 G9 R3.pdf [ 4.55MB ]
Metadata
JSON: 831-1.0059280.json
JSON-LD: 831-1.0059280-ld.json
RDF/XML (Pretty): 831-1.0059280-rdf.xml
RDF/JSON: 831-1.0059280-rdf.json
Turtle: 831-1.0059280-turtle.txt
N-Triples: 831-1.0059280-rdf-ntriples.txt
Original Record: 831-1.0059280-source.json
Full Text
831-1.0059280-fulltext.txt
Citation
831-1.0059280.ris

Full Text

THE REMOVAL OF SMOKES AND MISTS by DAVID ALAN GUTHRIE A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF APPLIED SCIENCE i n the Department of Chemical Engineering We accept this thesis as conforming to the standard required from candidates for the degree of MASTER OF APPLIED SCIENCE. Members of the Department of Chemical Engineering THE UNIVERSITT OF BRITISH CODJMBIA September, 1955 ABSTRACT A coloriiaetric quantitative analysis for di-n-octylphthalate and other aromatic esters has been developed which i s capable of determining as l i t t l e as 0.1 milligrams of an ester. This method i s based on the forma-tion of hydroxamic acid from esters using hydroxylamine hydrochloride i n an alkaline medium. On the addition of an acidified solution of f e r r i c per-chlorate, a red-colored complex of f e r r i c hydroxamate i s formed, proportion ate i n intensity to the weight of ester present. Mist composed of di«*i-octylphalate droplets of 0.869 microns average diameter was removed from a i r at substantially atmospheric temper-ature and pressure by passing the a i r up through a bed of 150/200 mesh s i l i c a gel fluidized i n a 2-1/U inch glass column. Removal efficiency, de-fined as the percent (by weight) removal of the mist was substantially i n -dependent of the entering concentration over the range 0.765 to 0.965 milligrams of ester per cubic foot of a i r . For a given bed, removal e f f i c i ency improved with decreasing superficial gas velocity. Two bed weights were used, 13.25 grams per square centimeter and 25.35 grams per square centimeter, and i t was found that the removal efficiency was practically i n dependent of the bed weights. The maximum removal efficiency was 88.8$ at a superficial bed velocity of 3.2 centimeters per second and a bed weight of 13.25 grams per square centimeter. The same mist was removed by passing the gas stream through various venturi nozzles with ports i n the throat through which fine s i l i c a gel (150/200 mesh) entered by gravity and aspiration into the gas stream. For the venturi nozzles the removal efficiency generally increased with increasing velocities; however, the maximum removal efficiency obtained was only about z that It i s shown/the behavior and collection efficiences obtained with the two devices can be sa t i s f a c t o r i l y explained i f the fluidized bed is assumed to collect the aerosol particles by diffusional processes only, and the venturi tube, by i n e r t i a l processes only, at least for aerosol particles of the size used i n this work. The problem of efficient removal of aerosol particles i n the range of 0,1 to 1.0 microns diameter has s t i l l not been solved i n an economical man-ner for many cases of industrial importance. The removal becomes even more d i f f i c u l t when the aerosol particles are f a i r l y uniform i n size. The purpose of the present work was to conduct a preliminary testing of new devices which might be more efficient for small particles than those now commonly used. TABLE OF CONTENTS ABSTRACT 1 INTRODUCTION AND THEORY 3 EXPERIMENTAL APPARATUS 9 PROCEDURE 13 RESULTS 16 DISCUSSION OF RESULTS k9 SUMMARY 55 NOMENCLATURE 57 BIBLIOGRAPHY 58 DATA AND CALCULATIONS 59 APPENDICES , 79 3 INTRODUCTION AND THEORY The problem of the removal of dispersoids from gas streams Is not a new problem, but because/relatively recent legialation more money and time are being spent to find economical methods of preventing atmospheric pollution. Many industries have made the sale of the products stripped from the stack gases pay for the removal costs. For example, the Consolidated Mining and Snelting Company at T r a i l , B. C #, has b u i l t a f e r t i l i z e r division to u t i l i z e the sulfates and sulfuric acid produced from the sulfur dioxide removed from exhaust gases from their metallurgical plants. Before any experimental work can be done i n dust removal, i t i s necessary to have a suitable source of smoke or mist and to have a method for determining the concentration of particles in the gas stream. To ob-tain the data for this report a homogenous aerosol was produced using apparatus similar to that reported by LaMer and Sinclair (6-7-18). The operation of the LaMer - Sinclair generator i s based on the principle of controlling droplet size by slow cooling of a vapour i n the presence of a r t i f i c i a l nuclei. Clean, dry a i r i s f i r s t passed over a heated nichrome wire, previously coated with sodium chloride, and then into a vaporizer. This air contains minute crystals of sodium chloride and these can serve as condensation nuclei. The generating substance (which i n this case was di-n-octylphthalate) i s heated i n the vaporizer at a constant temperature. The clean, dry a i r i s passed through the l i q u i d at a steady flow to saturate i t with vapor. This mixture of a i r , vapor and nuclei i s then passed into a reheater which i s kept at a higher temperature than the vaporizer i n order to ensure that a l l the generating material i s in the vapor state. The mixture i s next passed into a v e r t i c a l chimney where i t i s slowly cooled. Upon cooling, the vapor becomes sl i g h t l y supersaturated and drop-lets con-4-tinue to grow while r i s i n g i n the column by a process of diffusion. The theoretical aspect of the growth process has been discussed by Reiss and La Mer. (I4). The method for determining the concentration of aerosol i n the gas stream was modified from a method reported by H i l l (U,50 for the analysis of aliphatic esters. The qualitative spot tests for carboxylic acids and esters are the basis for this method. When an ester i s warmed i n an alkaline media with hydrox<pl4mine hydrochloride, hydroxaraic acid i s formed. Fiegl (1) reports the following equations: RCOOR • NHzOH • RCO(NHOH) * ROH Ferric iron forms a bright red or lavender complex with hydroxamic acids i n acid media according to the reaction: H RCO(NHOH) + F t " » R - C N " ? The red complex i s readily soluble i n aqueous ethanol. This method has not previously been reported for the determination of aromatic esters. This investigation shows i t to be a useful and sensitive method for the quantitative determination of esters such as benzoates and phthalates. Lapple (8) states that performance of a dust collector i s termed collection efficiency and i s generally expressed as a weight ratio of dust (or mist) collected to weight of dust entering the apparatus. According to this definition collector efficiency i s i n i t s e l f not a specific character-i s t i c of a given collector, but depends on operating conditions as well as on the physical properties of the particular dust treated. Perry ( l l ) classifies gas dispersoids as: Mechanical dispersoids Dust - particle diameter greater than 1 micron. Spray - particle diameter greater than 10 microns. s Condensed dispersoid Fume - particle diameter less than 1 micron. Mist - particle diameter less than 10 microns. For the sake of simplicity the.terms "dust" and "mist" will- arbitra-r i l y be used to describe any solid or liquid dispersoid respectively. Perry (11) and Lapple (8) l i s t the following forces or mechanisms utilized in the design of air purification equipments 1. Sonic 2. Thermal 3. Electrostatic l i . Gravitational £ • Physio chemical 6. Filtration 7.Inertia! Since the fi r s t four do not concern this report, i t is suggested that anyone interested in these refer to the references mentioned. Just exactly which mechanism governs dust or mist removed is dependent to a very great extent on the dispersoid particle size. Suitable equipment has been designed to remove particles of 5 microns diameter or larger; however, this equipment has been found to be unsuitable when used to remove particles of less than 1 micron diameter, particularly as the particle size becomes more uniform. To emphasize this point Wilson (18) has recently reported on a turbo-cyclone which reports efficiencies of over 90$ for heterogeneous aerosols of O J . to 10 microns diameter. When a homogeneous aerosol of 0.5 microns diameter was treated in the same apparatus, efficiencies reported were consider-ably less than $0%. The cyclone mentioned in the preceding paragraph is an example of one of the many pieces of equipment which have been designed on the basis that inertia! forces will effect the removal. The underlying principle of the 6 inertia! mechanism is that when a dust laden fluid flows towards a body, the fluid w i l l be deflected around the body whereas the dust particles by virtue of this greater inertia w i l l pass through the laminarr boundary layer arid impinge; on the collecting surface. Since both mass and velocity have an effect on the inertia of a body, both of these factors must be considered in the design of inertial type equipment. Collection of aerosol particles by inertial mechanisms has been discussed by Langmuir and Blodgett (9). They state that the collection efficiency is described by a target efficiency which represents the fraction of particles in the fluid volume swept by the body which w i l l impinge on the body. This target efficiency, ^ , is a function of the dimensionless group, Vf *^ , where U.^  = terminal velocity of the aerosol particle Vo average velocity of the fluid g - acceleration of gravity "Dp - diameter of the collecting body Curves showing the relationship between the target efficiency and the dimen-sionless group have been drawn by the above authors for targets of various shapes. High collection efficiences (90%) by inertial means can only be expected i f the value of the dimensionless group is 0.5 or more. Below this value collection efficiency decreases rapidly. A further discussion of inertial mechanisms is given by Wong et al (19,20); however, much of his work as reported is beyond the scope of this report. One of the most common methods of dust removal is filtration. Many industries have installed bag houses and use cloth filters (12). A great many of the so-called dust filters remove dust because of some other mechanism, for example, the common fibre f i l t e r used in many homes is actually a com-bined f i l t e r and inertial-type separator. Another common type of f i l t e r is 7 the coke f i l t e r used i n the sulfuric acid industry and i n this case particle collection i s attributed i n part to diffusional mechanisms;. The collection, of aerosol particles by diffusional mechanisms i s known to be of importance for particles of 1 micron diameter or less and i s the most important mechanism for droplets of less than 0.1 microns diameter. The s t a t i s t i c a l average linear displacement AS of a particle i n any given direction i s given by the equation y 4RTfc™t ~ where R Gas constant T Tamperature kjn Cunningham correction factor for Stoke »s Law t Time Vtl Fluid viscosity N Avagadro's number Dp Particle diameter For fixed experimental conditions using a homogeneous aerosol, the displace-ment of a particle because of Brownian motion should be a function of time only. Values for the Cunningham correction factor to Stoke»s Law, which allows the calculation of terminal velocities and diffusional displacements have been given by Lapple, (8). The coke f i l t e r , which has already been mentioned, i s an example of a fixed bed. A fixed bed i s one i n which each particle has a fixed position relative to every other particle and to the walls of the container; whereas a moving bed, which has been used for smoke removal (13), i s one i n which each particle remains fixed relative to the other particles but 8 moves; with respect to the walls. A third type of bed, a fluidized bed, i s one i n which each particle moves relative to every other particle and to the wall also. It i s important to note that the fluidized bed has a def-inite upper boundary and in this respect differs from a f l u i d - s o l i d transport stream. Although no industrial use of the fluidized bed for smoke removal has been reported, this type of bed has certain advantages. For example, the pressure drop through a i 1.„!..'r bed increases with velocity up to a certain point only, beyond which the pressure drop i s practically constant. More-over this pressure i s small compared to pressure drops through turbo-cyclones and venturi scrubbers. It i s possible with both the moving and fluidized beds to remove old solids and to add new solids continuously, but again be-cause of the turbulence within the fluidized bed this bed has the advantage of being able to retain more aerosol before needing replacing. The appar-atus for the fluidized bed i s much simpler to construct than are cyclones and yet the absorbed material can be removed and the absorbate reused, a factor which could prove to be e«onomically important. Meissner and Mickley (10) have reported that mists composed of sulfuric acid droplets 2 to l l | microns i n diameter were f i l t e r e d from air using fluidized beds of various non-porous and porous m a t e r i a l The non-porous bed had a relatively short l i f e , whereas the porous beds, l i k e s i l i c a gel, could absorb % t by weight, before sticking destroyed fl u i d i z a t i o n . They found that removal efficiency was substantially constant during the l i f e of •the beds and independent of the entering concentration over the range of from 20 to 120 pounds of acid (100$ basic) per 1,000,000 cubic feet. For a given fluidized bed, removal efficiency improved with increasing super-f i c i a l gas velocity and with increasing bed weight per unit area. Meissner s$A Mickley (10) proposed that the removal was affected by i n e r t i a l mechan-isms and thus explained the improvement of collection efficiency with i n -1 creased velocities. They proposed the equation where U 0 superficial gas velocity Ci i n l e t concentration C2 outlet concentration k constant W bed weight per unit area n constant EXPERIMENTAL APPARATUS: The equipment required for the analytical work consisted of t wo water baths complete with thermoregulators, mixers, heaters, and thermometers. A Klett - Summ§r-sgn photo-electric colorimeter was used to measure transmission i n the edi-octylphthalate determination. This imet'exd was connected to a con-stant voltage transformer. The construction of the homogenous aerosol generator was based on a design described by LaMer (6). A schematic sketch of the apparatus is shown on page it© . The vaporizer was a glass cylinder approximately 8 cms. i n d i -ameter and 30 cms. deep. The rehedferwas a round-bottomed two l i t e r flask. The chimney, a double-walled condensor, was f i t t e d into the reheater using ground glass joints. The enucleation chamber was a sphere of l£ cms. diameter equipped with a support rod and a grounl glass joint for the nichrome heating c o i l . This nichrome c o i l was made from 127 gage wire and had about 10 ohms total resistence. The c o i l was s i l v e r - soldered to two tungston contacts which passed through and were sealed i n the male end of the ground glass joint. The power was supplied from a constant voltage -transformer. The a i r baths suggested by LaMer (6) were iglaced with two stainless HOSE -SAMPLER MANOMETERS OL A S 8 COLUMN WET-TEST METER TUBINO DEPT. OF CHEM. ENO. U. B. C DOUBLE-WALL CHIMNEY E.M. F. FIGURE I SCHEMATIC SKETCH OF THE APPARATUS DAVID. A GUTHRIE JULY 12 1 9 5 3 II steel tanks each containing bath wax* These baths were regulated and were heated by two immersion heaters operating i n pa r a l l e l through a variac. Each bath was also equipped with a visible speed s t i r r e r and two thermometers located at separated points i n the bath. The a i r used was the building supply. It passed f i r s t through a pressure control valve and then through two drying tubes and two fibre-glass f i l t e r i n g tubes. At this point the a i r flow was divided and part of the flow passed through a meter into the mucleation chamber. The other part of the air was metered into the preheater, an 8 mm. glass c o i l immersed i n the bath, and then passed down through the long tube of the vaporizer. A l l a i r flow meters were capillary tubes which acted as ori'fiees. The pressure drop across these tubes was measured using manometers. Manometers were also installed to measure the absolute pressure on the inlet side of a l l cap i l l a r i e s . The rnucleation chamber, preheater, and reheater flask were a l l connected to the vaporizer using glass b a l l and socket joints. The a i r lines from the valve to «W» where the a i r enters the generator, the bleed-off l i n e , and the dilution flow line were a l l 8 mm. i n diameter. From the top of the chimney through to the wet-test meter 17 mm. tubing was used. The. line from the chimney was connected to the 2| inch glass column5used for the fluidized bed,by a 2 inch brass reducing coupling. The bed material was 1^0/200 Tyler mesh s i l i c a gel supported on a 200 mesh screen. Manometers were installed to measure the in l e t pressure to the bed and to measure the pressure drop across the bed. The a i r flow could be directed along three paths. I t could by-pass the column, pass through the column and by-pass the sampler, or pass through the column and through the sampler into the wet-test meter. The f i l t e r used to sample for aerosol concentrations i n a l l streams was a MF f i l t e r (3). I t i s sometimes referred to as a millipore f i l t e r . It i s not tru l y a paper, but i s actually a sheet of cellulose esters. 12 It i s about 15>0 microns thick and has a to t a l pore volume equal to 80% to 8$%. It i s supposedly insoluble i n petroleum ether; however the ether des-troys the c e l l structure. The greatest disadvantage of these f i l t e r s i s the cost,which i s about twenty cents per f i l t e r . The f i l t e r holder was made from a 2 inch brass cylinder. The up-per half was l j inches deep and threaded tightly into the 3§ inch bottom section. A 1/8 inch ring to support the f i l t e r was dsoldered to the inside of the bottom section so that as the top and bottom were screwed to-gether the f i l t e r was clamped into place. The ester used to produce the aerosol was practical grade di-n-octylphthalate purchased from the Eastman Chemical Company. The experimental set-up was similar*- when the venturi nozzles were being tested except that the 2 inch reducer coupling holding the screen was replaced by a rubber stopper to hold the venturi tube. The original/.' venturi, A, was designed by Forrest (2). Figure 2 is a f u l l scale diagram of this venturi nozzle set i n the rubber stopper. Two other ventaris B and C were designed similar^ to the f i r s t . A fourth venturi, D, was designed as two pieces. One piece was a 13 mm. glass tube necked down to 8mm. and then belled outwards. This piece was supported by a rod through the rubber stopper with which the tube could be raised or lowered relative tp the bottom piece, which was a piece of tubing passing through the stopper. A u.8 cm. diameter by 0.38 cm. thick plexiglass cap could be attached to the upper piece to serve as an impingement plate. The dimensions of the various venturi's are tabulated on the next pagee Venturi Throat Diameter Port Diameter A 5.69 2.87 B 3.91 2.6k C 7.00 2.61; D 7.93 open PROCEDURE; The percaent transmittancy data required for the quantitative analy-sis of esters was obtained using a Coleman spectrophotometer i n the Chemistry Department of the University of British Columbia. An analytical determina-tion was made on 10 ml. of pure petroleum ether (30 - 60° C) and a similiar run was made on 10 ml. of petroleum ether containing some di-n —octyphthalate and the light transmission measurements at various wave lengths of both these solutions were obtained. To obtain the calibration curve a single drop of the ester was weighed out i n a 100.,.mTL» volumetric flask. The flask was then f i l l e d up to the 100 m. mark with d i s t i l l e d petroleum ether. To obtain a point on the calibration curve a volume of this stock solution was pipetted into a 250 ml. wide-mouthed eriewieyer flask and then diluted to 10 ml. Then 0.3 ml. of 2.5 weight percent sodium hydroxide and 0.3 ml. of 2.5 weight percent hydroxyilamine hydrochloride were added to the ester solution i n the e r l e n — meyer. The petroleum ether was then evaporated i n a water bath maintained at 67° C. inarperiod of 12 minutes, after which the flask was placed i n another bath and cooled at 25° C. for 1 minute. Ten ml. of dilute modified solution A was pipetted into the flask over a period of 3/k minutes. The solution was swirled for \ minutes and them poured into a curvette and a balance was made dn the Klett - Summers on meter within \ minutes. This procedure was repeated for various weights of ester up to about one milligram. T • ^^BBJEJJ/^ ^ S T O P P E R ^ o FIGURE 2 DEPT. OF CHEMICAL ENG. D. A. OUTHRIE VENTURI NOZZLE A U. B. C. FULL S I Z E p 14-,;The zero point for the Klett - Summerson meter was 95% ethanol. The analysis for the mist samples was s i m i l a r e x c e p t that the millipore f i l t e r was treated with 10 ml, of petroleum ether for 5 minutes. Moreover i t was at times necessary to remove any s i l i c a gel that had been carried over onto the f i l t e r . This was done by f i l t e r i n g the ether solution after the 5 minute extraction period through a lw2f> cm. paper f i l t e r . The f i l t e r paper was then washed with 2 ml. of petroleum ether and then the analysis was carried out as for the calibration curve. Before any work could be done with the generator, i t was necessary to calibrate a l l the flowmeters• These were calibrated using clean, dry a i r metered through a wet-test meter. The volume of air metered, i t s pressure, and i t s temperature were a l l recorded as also was the absolute pressure on the inlet side of thecorifice. After heating the vaporizer bath to 132° C. and the reheater bath to 162° C., 25 ml. of fresh di-n-octylphthalate were placed i n the vaporization chamber and the nichrome c o i l , which had been previously coated with four layers of a saturated solution of sodium chloride, was fixed into position. The a i r flow was adjusted to pass the desired volumes into the rniucleation and vaporization chambers. The nichrome c o i l was then connected to the con-stant voltage transformer and the current was adjusted to 2.67 amperes. After the generator had been operating for an hour three microscope slides were coated wi th/^aqueous solution of gum arabic. This was done by putting a s ingle drop of solution on to a slide and then s preading i t with another sl i d e . Using this method the formation of bubbles can be avoided. After the glue had become tacky, the aerosol was allowed to jet into the slide through a 6 mm, glass tubing. No cover s l i p was used; instead the slides were placed i n a calcium chloride dissicator and allowed to dry for about two hours. At the end of this period the droplet diameter was 15" measured using the micro-hardness tester attachment for the Zeiss microscope i n the Metallurgical Department of the University of British Columbia. After a suitable homo;g3nous aerosol had been produced the gas stream was sampled by metering the gas througha^millipore f i l t e r . The amount of ester deposited on this f i l t e r was determined using the analytical method outlined. This measurement was made several times during a run and the av-erage was used as the in l e t concentration. The gas stream was then passed through the 2\ inch column contain-ing the s i l i c a gel and the aerosol concentration i n the effluent stream was measured. The superficial gas velocity through the bed was varied by bleed -ing -off some of the flow from the generator or by diluting the flow with clean dry a i r . The mist concentration i n the effluent gas stream was measured at each velocity. This procedure was repeated for two bed weights. The amount of aerosol removed by the empty column was also determined. The effect of inlet concentration was also studied. The venturi nozzles were tested in a similrar- fashion and the weight of s i l i c a gel entrained i n the gas stream was measured at several velocities. The beds were sampled periodically. The s i l i c a g e l l removed was mixed with petroleum ether, the resulting solution was f i l t e r e d and the f i l t r a t e was analysed for ester content. \6 RESULTS? I ANALYTICAL PROCEDURES. Table 1. Percent transmission of hydroxamate solution versus wave length i n microwavelengths. YJave length % transndttancy 350 100 375 80 liOO 66.6 U5o 62.5 5oo ; 5 ? a 550 61.8 600 68.5 625 73.2 650 78.U 675 81.U 700 81.7 Table 2. Times found most suitable for analytical steps. Step Time - minutes Evaporation -at 67° 12.0 Cooling at 25° 1.0 Addition of solution A @.75 Swirling . 0.50 Balancing of meter 0.25 Extraction of f i l t e r 5 These data are plotted in Figs. 3, and show clearly a maximum adt-sorption in the range of 500 mic. ^he timed procedure given i n Table 2 was found to allow accurate and reproducible analyses. \7 9 0 30 70 6 0 [60 ) c ) \ c 4 0 3 0 0 4 0 0 700 5 0 0 6 0 0 WAVE LENGTH m X FIGURE 3 TRANSMITTANCY VS. WAVE LENGTH OF FERRIC HYDROXAMATE IN 95 % ETHANOL Table 3. Calibration data for the Klett-Summerson meter for di-ni-octylphthalate, mgm. di-n*-octylphthalate Klett-Summerson Rdg. 0 8 0.1368 U$ o.nih 59 0.3U20 102 O.lilpU 115 0.5130 135 0.681t0 195 These results are shown on graph k» Since the K.S. reading i s actu-a l l y the logarithm of one minus the transmittancy value, the equation of the line can be written, S » 261.8C +• 9.6 which i s log ( 1-1 ) = 261.8C+ 9.6 or which again i s T T r 2 6 l . 8 c I 0 - I • c which i s a form of ^eer's Law. / where S = Klett-Summerson reading C = weight of ester - milligrams I B intensity of transmitted ligh t IQm intensity of incident light e - exponential e A l l straigt lines are f i t t e d by method of least squares except where otherwise stated. FIGURE 4 CONCENTRATION VS K. S. READING FOR Dl- OCTYLPHTHALATE Table i i . Rate of color fading of various hydroxamate solutions. Time elapsed Klett-Summerson mgm. sees. Meter Reading P.O.P. Number 1 25 117 .101 5o 115 . t o 90 112 .391 120 107 .372 Number 2 20 68 .223 kO 67 .219 60 67 .219 120 62 .208 Number 3 30 120 .1*22 60 118 , !o5 90 118 ,1O5 120 116 .1*07 300 111; .UOO 1300 110 .381; These valves are shown plotted in F i g . 5> and indicate that i f readings can be taken within 15 seconds, the error i n the naalysis due to a few seconds variation either way w i l l not be large. I t i s also apparent that the color fades steadily and there is no "preferred" time at which readings should be taken. 22. Table 5. Calibration Curves for Various Aromatic Esters. mgms. Cenco meter mgms, Cenco meter mgms. Cenco meter dioctylphtha- rdg. ' n^butylphtha- rdg. ethyl ben- rdg. late late zoate O.Ou98 87.O 0.0969 57.0 0.294O 38.0 0.1328 76.0 0.1610 55.5 o.mao 29.5 0.3320 57.0 0.3230 1*5.0 0.5880 2U.0 0.4980 46.O 0.1.850 35.0 0.8820 16.0 0.5976 52.0 O.646O 27.0 0.6972 3U.5 0.9690 •18.0 0.7968 28.0 O.8964 25.0 0.9960 22.5 The equation for the various lines of figure 6 are; 1. for di-n-octylphthalate 2. for n-butylphthalate 3. for ethyl benzoate _ T -0.625c I « I c e I - 1 e -0.600C I - I c e - - 6 1 2 C Table 6. Effect of Parent A c i d . Mg. of Ethyl mg. of Benzoic Acid Cenco Meter Ms. Benzoate _____________ 0.29U0 0 48.O 0.2940 0.1730 58.0 0.29UO 0.1730 59.0 0.2940 0.3u50 70.0 0.2940 0.3U50 68.0 The data given i n Table 5 are>plotteS-in Fig. 6 and show that t h e complex formed obeys Beer's Law s a t i s f a c t o r i l y f o r a v a r i e t y of e s t e r s over the range of concentrations i n v e s t i g a t e d . 0.1 0.2 0.3 0.4 0.3 0 . 6 0.7 0 .8 0.9 MILLIGRAMS OF ESTER 1.0 I.I 1.2 FIGURE 6 CONCENTRATION VS. CENCO METER READING FOR THREE E S T E R S a4 II. Generator Operating Conditions. Vapourizer Temperature s 132° C. Reheater temperature - 162° C. ^'Njucleation coil current z°2.67 amperes Air flow to preheater - 9.88 liters/min. Air flow to vnucleation -chamber r 8.78 liters/min. Air flows referred to 1 atmosphere and 70° F« A l l velocities reported are superficial velocities. For the fluid-ized bed and for venturi D, the velocity was calculated on the basis of the 2.25 inch column. For V e n t u r i s A, B, and C, the velocities were calculated on the basis of the various throat diameters listed. It has been found more convenient to w ork with P a 100 - E where E is the collection efficiency expressed as a percent; P then is percent not removed. The bed was 150/200 Tyler mesh s i l i c a gel. III. Mist Removal in the Fluidized Bed and in the Venturi Nozzles. Table 7. Effect of inlet concentration on collection efficiency. Bed weight 651 gms. = 25.35 gms* per sq. cm^  Bed height at rest 5 k2 cm. Run Velocity Inlet Concentration P % No. cm./sec. mgms./ft.3 Co. Not removed 8E 11.88 .690 22.2 9E 11.07 ,70k 25.2 6E 11.25 .728 28.6 5E 10.U8 .782 2U.6 2JE 11.82 .877 22.8 5 0 CO < 5 4 0 -e x < 3 0 a Ul AC © 2 0 o • o - •o -C) in 10 o K kl 0. 0 . 6 8 . 7 0 . 7 2 . 7 4 •76 •78 . 8 0 . 8 2 8 4 8 6 . 9 0 7 9 2 INLET E S T E R C O N C E N T R A T I O N " F I L T E R E D " MGMS /CU FT FIGURE 7 EFFECT OF INLET C0NC. ON MIST REMOVAL BED WEIGHT 35.3 5 GMS/SQ. CM. i Table 8 . Efficiency measurements i n empty column. Velocity % not cm./sec. removed 3.U1 72.3 6.90 70.1 11.82 82.6 Table 9. Results for the 34I gm. a 13.25 gms. per sq. cm. fluidized bed. Bed depth at rest - 23 cms. Run No. Vel. cm./sec. 1 v/Vei; P % not removed P % not re-moved- corre cted Condition of Bed 12A 3 . 0 8 0.568 13 .6 19.U -bed depth 25 cms. 5B 1.05 O.U97 16 .6 23.3 -not quite fluidized 8B 5 .71 O.4I8 m.i 19.4 -particles bouncing 11A 5.72 6.L.18 12 .0 16.5 on upper bounds^' of bed. 10A 6.79 0.383 1 6 . 0 21.7 -bed depth 28 cms. 7B 7.32 o .369 15.7 ,21.1 9A 8.75 0.338 19.5 26.3 -bed well fluidized 6B 10.32 0.311 19.4 24.7 8A 11.82 0.291 22.5 i 28.3 15A 11.82 0.291 2l i .8 31.2 UB 11.8.2 0.291 21.3 26.8 LIB 11.85 0.290 19.1 2U.1 -no jetting or lliA l j . , 1 0 0.277 25 .6 31 .5 slugging. 12B 13.51 0.273 21.2 26.1 -heavy carry over 13A 14.35 0.26U 29.1 35.0 of gel, excessive 9B LU.lt8 0.263 28.6 3 l u 8 jetting - d i f f i c u l t to measure depth. 3 0 2 8 < 2 6 5 2 4 < k l te 2 2 u o j ^ \ i • o CO 2 o 2 0 kl 0. 1 8 16 5 6 S U P E R F I C I A L 7 8 G A 8 V E L O C I T Y 9 10 C M S / 8 E C . II 12 13 FIGURE 8 E F F E C T OF VELOCITY ON MIST REMOVAL FOR EMPTY COLUMN 34 Figure 9 is a plot of superficial gas velocity, Vo, against the percent not removed, PJ both corrected and uncorrected values of P are plotted. Figure 10 is a plot of l / V Q versus the percent not removed. Table 10. Results for the 651 gms - 25.35 gms. per sq. cm. fluidized bed. Bed height at rest - kk cms. Run Nof Vel. cm./sec. ,1 /"Vo. . P % not Eemoved P % not re-moved-corrected Condition of bed 8C 3.20 0.558 11.2 16.0 7D k.kB O.U72 15.0 21.1 -not well fluidized 7C 1+.78 O.U57 11.2 15.6 6C 5.85 0.I4IU 12.0 16.1, -depth = kk cms. 5C 7.82 0.357 18.8 25.0 -well fluidized 8D 7.87 0.356 19.5 25.9 5E 10.1+8 0.309 2U.6 31.6 -depth r 52 cms. 9E 11.07 0.301 25.2 32.1 6E 11.25 0.299 28.6 36.3 IjE 11.82 0.291 22.8 28.7 UD 11.82 0.291 22.5 28.3 13C 11.82 0.291 21.1 26.6 9C 11.82 0.291 17.8 22.U 10E 11.82 0.291 18.8 23.7 8E 11.88 0.290 22.2 27.9 5D 12.55 0.283 20.2 25.2 6D 12.55 0.233 2U.8 30.9 10C 12.90 0.279 17.8 22.1 -jetting 11C 13.05 0.277 19.6 2U.3 -bed depth 51* cms. „ 12C 13.20 0.275 18.0 23.2 o -Q-x X 6 7 S u P E R M C I A L G A 8 8 9 V E L O C I T Y o o o X o o o 6 - O -o K E Y C O R R E C T E D X U N C O R R E C T E D C 10 C M / 8 E C ' I I 12 13 14 EFFECT OF VELOCITY ON MIST REMOVAL.' BED WT. 25.35 GMS/SQ CM. 34 < ,26 < Z 22 <0 18 u o cc Ul °- 14 10 O ( c O ) u o o FIGURE 12 6.0 I V/SUPERFICIAL GAS VELOCITY CMS/SEC. EFFECT OF VELOCITY ON MIST REMOVAL. BED WEIGHT 25.35 GM>5Q. CM. N 34 30 (0 < © 26 o z z < £ 2 2 CO 2 18 ui o OC Ul °- 14 10. KEY 13.25 0M8/8Q. CM. BED * 25.33 0M8/8Q. CM. BED — < > — OVERALL X o X ( X o ( O-X X X X o . o X 2.0 2.4 2.8 3.2 3.6 4 0 I 4-4 4.8 5.2 5.6 6-0 v/SUPERFICIAL OAS VELOCITY CMS/SEC. FIGURE 14 EFFECT OF VELOCITY ON MIST REMOVAL FOR FLUIDIZED BEDS Figures 11 and 12 are plots of the data on the preceding page only and figures 13 and Ik show the results for both fluidized beds. Table 11, Results for venturi A, Throat area - 2 5 .44 sq. ram. Port area r 12.9h sq. mm. Bed weight = iQO grams. Run No. Velocity cm./sec. 6F 1195 7F 1195 Table 12. Results for venturi B Throat area - 12.01 sq. mm. Port area s 10.92 sq. mm. Bed weight r 100 grams. Run No. Velocity cm./sec. P % not removed 61.9 60.U P % not removed 5P 6P 7P 6P 9P 10P UP 12P 17P! 18P 13P luF 15P 16P 888 888 1150 1150 1750 1750 2570 2570 2570 2570 2810 2810 3210 3210 64.2: 58.8 60.5 60.9 59.7 65.5 6U.9 58.8 59.6 61.0 5U.2 55.7 62.7 65.9 6 6 CO < 6 0 6 5 u OC z Z Ul o cc Ul o. O o o V E N T U R r * v / A / - ^ o ( J o CJ ! o i-i f I ( J\ t t -8 0 0 1 0 0 0 12 OO 1 4 0 0 1 6 0 0 It 1 0 O 2 0 0 0 21 2 0 0 2 4 0 0 2 6 0 0 2 8 0 0 9 0 0 0 3 2 0 0 43 4 0 8 U P E R M C I A L GA8 V E L O C I T Y C M 8 / 8 E C FIGURE 15 EFFECT OF VELOCITY ON MIST REMOVAL '. VENTURI B t 0* 3 1 Table 13. Results for venturi. C. Throat area s 38.^0 sq. mm. Port area r 10.92 sq. mm. Bed weight - 100 grams. Run No. Velocity cm./sec. P % not removed 19N 299 63.7 20N 299 57 .9 17N 379 6 2 . 3 18N 379 61+.9 15N 380 65 .8 16N 390 67 .8 7N 514 59A7 8 N 514 61.2 5N 80a 6 u.l 6N 80lt 55.6 13N 801+ 60 .5 lilN 80 k 63.2 9N 923 6 5 . 0 ION 923 59.2 11N 985 5U.2 12N 985 54.2 Graph 17 is a plot of the data for venturi's B and showing the increase i n removal e f f i c i e n c i e s observed >;ith i n c r e a s i n g v e l o c i t i e s . A p l o t of t h e o r e t i c a l t a r g e t e f f i c i e n c i e s f o r the s o l i d and a e r o s o l p a r t i c l e s used and based on the' values of Langmuir and. Bjod'gett (9), i s a l s o shorn f o r compari-son on the same graph. FIGURE 16 EFFECT OF VELOCITY ON MIST REMOVAL ! VENTURI C to 0 0 FIGURE 17 EFFECT OF VELOCITY ON MIST REMOVAL ! VENTURIS B S C Table l U . Results for Venturi D, with no cap. Throat area - U9.32 sq. mm. Open port gap = l.£9 cms. Bed weight = 100 gms. Run Velocity P % not No. cm./sec. removed 9J I1.48 62.2 10J U.U8 64.O 5K 6.29 71.6 6K 6.29 68.3 7J 6.48 65.4 8J 6.48 60.6 9K 8.16 75.7 10K 8.16 70.5 5J 8.27 72.4 6j 8.27 79.7 11J 12.0a 68.8 12J 12.0U 73.2 7K 12.04 73.5 8K 12.Oil 67.1 13J 13.00 80.3 I4J 13.00 78.9 5K 14.81 90.5 6K 1U.81 85.9 41 Table 15. Results for Venturi D with plexiglass cap. Bed weight = 100 gms. Open port gap - 1.59 cms. Run No. Velocity cm./sec. P £ not removed 10L 5.59 76.0 ILL 5.59 81.2 8L 8.23 79.2 9L 8.23 81 .5 8M 12 .Oli 72.9 5L 12. Ok 70.6 UL 12.0k 70.9 6L lk . 5 5 94.8 7L 14.55 74 .6 4M L U . 8® 78.2 5M 14.80 74.5 6M 14.80 65.8 The results tabulated i n Tables lk and 15 are plotted on figure 18 . Table 1$. Solids flow for venturi B. Velocity Wt. of gel collected cms./sec. gms. per min<> 723 2 .0 1151 2 .9 1418 8.6 1490 12.2 2110 17.8 2285 14.8 2880 15.3 6 0 0 1 0 0 0 1200 1400 SUPERFICIAL 1600 1800 2 0 0 0 2 2 0 0 2 4 0 0 2 6 0 0 2 8 0 0 3 0 0 0 OAS V E L O C I T Y CM8./SEC. FIGURE 19 EFFECT OF VELOCITY ON SOLIDS FLOW VENTURI B Table 17. Solids flow results for venturi C, Velocity Wt. of gel collected cms./sec. gms, per min. 275 3.0 455 3.8 588 4.7 719 5.8 902 11.0 i results for venturi D, Velocity ¥t. of gel collected cms./sec. gms. per min. 2.89 4.3 5.91 19.4 6.15 22.3 8.67 23.1 11.82 25.2 11.88 24.6 H . 7 2 22.U 10 1 0 0 0 1100 1 2 0 0 SUPERFICIAL 6AS VELOCITY CMS./SEC. FIGURE 20 EFFECT OF VELOCITY ON SOLIDS FLOW! VENTURI C 4<> 4*t LY. Results of Smoke Measurements* Average di-n-octylphthalate droplet diameter - 0.869 miserons. Table 19. Distribution data f o r aerosol measurements. Droplet Diameter Range No. of droplets i n Cumulative % range  Q)*50 or less 3 1.36 0.50 - 0.55 2 2.27 0.55 - 0.60 5 4.53 0.60 - 0.65 4 6.34 0.65 - 0.70 12 11.76 0.70 - 0.75 23 22.16 0.75 - 0.80 37 38.90 0.80 - 0.85 43 58.3U 0.85 - 0.90 22 68.38 0.90 - 0.95 41 86.91 0.95 - 1.00 14 93.23 1.00 - 1.05 6 95.93 1.05 - 1.10 3 97.29 1.10 - 1.15 1 97.74 1.15 - upwards 5 100.00 Figure 22 i s a plot of the above results. 0.5S .60 . 65 .70 .75 . 80 . 85 .90 .96 1.00 1-0 6 110 1.15 AEROSOL DROPLET DIAMETER MICRONS FIGURE 22 AEROSOL DROPLET DIAMETER DISTRIBUTION DISCUSSION; The curve shown in figure 3> the plot of transmittancy versus wave length, agrees f a i r l y well with those obtained by H i l l (li) and Thompson (17). Since a minumum occurs i n the 5(XW(^ region a 520 f i l t e r was used i n the Klett - Summerson meter. H i l l (U) reported that the hydroxamate solution obeys Beer's Law up to concentration ofaimilligram of ester for fatty alipthatic compounds. This means that, since Beer's Law i s of form I : I D C ~ k c where I - transmitted light intensity IQ«» incident light intensity c z concentration k ; extinction coefficient a plot of log ( ~Y~ \ against the concentration should be a straight l i n e , o / As the scale on the Klett-Summerson meter was logarithmic a plot of K, S. reading against concentration yielded a straight l i n e . The calibration curve shown on figure h agrees very well with previous curves, proving that re-producible results can be obtained using the timed method described i n the procedure section. The several equations for the curve of figure h were l i s t e d t o show the relationship between the meter reading and the concentration of ester© The f i r s t equation i s the one used i n c a l c u l a t i n g , whereas the second and third equations are written to show that the hydroxamate solution obeys Beer's Law. It was noticed that the color of the hydroxamate solution faded with time, particularly i f there was any liquid petroleum ether remaining i n the easTenme^r flask. Figure 5 shows the rate of fading for various concentrations of di-n-octylphthalate. It should be noted that the i n i t i a l rate of fading i s more rapid than the rate after a minute or so. The data for the effect of the parent acid and the calibration curves of figure 6 were obtained work-ing i n conjunction with R. Rye (15). Rye used the exact method as proposed by H i l l (u), namely to evaporate the . dther solution to dryness and then for 5> seconds longer; however the reproducibility of this method was very de-pendent i n the s k i l l of the analyst. The parent acid affects an increase i n the amount of l i g h t /transmitted or rather inhibits the formation of the red-colored complex. The idea i n doing the work on the other esters was to determine i f the color was formed by the acid part of the ester alone; this would mean that the n-butylphthalate and the di-octylphthalate should have had the same calibration curves. From the results i t was decided that the rest of the compound affected the color formation. Since the curves of figure 6 are not used elsewhere i n the report the "best" line i s f i t t e d by eye. Before any work could be done on determining removal efficiency, i t was necessary to find the operating conditions required for producing a homogenous aerosol. It should be noted that the operating conditions decided upon are considerably, different from those suggested by LaMer (6). However i t seems that one difference must have balanced another because a reasonably homogenous aerosol was produced. The aerosol droplet size was measured using the Leitz microscope i n the Metallurgical Department of the University of British Columbia. The mist sample was collected by letting i t impinge onto a glass slide coated with gum arabic solution. Although some effort was made to determine just how tacky the gum arabic should be i n order that the drop-lets adhere to i t , no definite conclusions were reached. It i s suggested that the several slides be coated with the glue and l e f t for various lengths of time from 2 to £ minutes. In this way some, i f not a l l , of the slides w i l l be s u i t a b l e - % e n making the actual measurement of the droplet size, i t was found that i t was very advantageous to measure a single cluster of droplets. This meant that the microscope did not have to be refoeused. Figure 22 shows the droplet size distribution for the di-n-octylphtha-late mist. Considerable d i f f i c u l t y was found in getting the generator to produce a constant mass loading. It was thought that perhaps the millipore f i l t e r was not stopping a l l the droplets. This was disproved by using two f i l t e r s i n series and measuring the amount of ester collected on the second. In a l l cases this amount was so slight that i t can be considered neglible. Improvement i n steadying the generator mass loading was achieved by shortening the depth of the sampler, by placing some glass wool i n the line between the reheater and vaporizer to remove entrained.droplets, and by always coating the c o i l with the same amount of solium chloride. With these alterations, the mist concentration produced by the generator was constant after steady state conditions were reached. This usually required at least one hour. Several tests were made to determine whether or not the millipore f i l t e r was soluble i n the petroleum ether. It was found that they- were notj however, / ^ w l r e thought to be structurally deformed. The amount of ester l e f t on the paper f i l e r was about 10$ and so i t was necessary to adjust the i n l e t concentration when reporting the range of i n l e t concentration studied. The results obtained on the fluidized beds contradict those obtained by Meissner and Mickley (10). Whereas the results of this work show that the removal efficiency/- decreases as velocity increases, the results of Meissner and Mickley (10) show? that the collection efficiency increases as velocity sz increases. I t i s proposed that of the two mechanisms, i n e r t i a l and diffusional, that the more dominant one i s diffusional. The target efficiency to be expected i n this system of a dilute mist in a fluidized bed can be estimated from the results given by Langmuir and Blodgett (9), The numberical values obtained probably have a f a i r degree of uncertainty;.}; but the order of magnitude should be an indication of the re-moval mechanism operating i n these tests. The dime.nsionaless group U t Vo has been evaluated(see calculations section) for this system and i t i s obvious that f o r velocities over the range of 1 to 15 cms. per second that the target efficiencies are, for a l l practical purposes, zero over the entire range,, Using the equation for diffusional displacement given earlier, the average linear displacement of a particle of average diameter of 0,87 microns is about 10.7 microns i n 2.85 seconds. This may be compared to the average bed particle size of 90 microns and a bed void fraction of 0.i;3. I t is shown in the calculations that every aerosol particle must travel 7,9 microns, which means collection efficiency should be about 130$. This i s not v a l i d , of course, to a high degree of accuracy because the void space is not spheroidal as was assumed, and hence the average distance each droplet must travel w i l l be greater since a sphere represents the minimum dimensions for a given volume. The order of magnitude i s interesting since i t indicates the im-portance of the diffusion mechanism and i t would appear that there are sound reasons for assuming that the collecting action of a fluidized bed for the conditions used here is almost entirely by diffusion. The diffusion equations indicate that the diffusional displacement is proportional to the square root of the time and therefore to. 1 • The data was plotted this way on figures 10, 12, and lit, and results^ while no better than those depicted by figures 9, 11, and 13j are certainly no worse. The assumption here i s that the linear movement accomplished by diffusion i s dir e c t l y proportional to the collection efficiency. Some ; terror i s introduced i n this graph by the fact that the time is not quite proportional to the superficial velocity due largely to the varying bed expansion and therefore to bed void volume; however i t was not worth correcting for the varying void volume because of the spread of the data. The reason why Meissner's and Mickley 1s (10) results d i f f e r from these i s that the droplet size range used by them was from 2 to 10 microns, whereas the average mist size used here was 0.87 microns. Moreover the concentrationand velocities used by Miessner and Mickley were both greater than those used i n this work. That the removal efficiency of the empty column decreased with an increase i s velocity i s thought to have been caused i n part by the blowing over.. at'-'higher velocities, of some aerosol droplets which might have adhered to the walls at lower velocities. This effect of velocity on mist removal for the empty column could also be attributed to diffusional mechanisms. Figure 7 verifies that inlet concentration has no effect on collection efficiency over the range studied. This agrees with results reported by Meissner and Mickley (10); however no significant change i n collection efficiency by was obtained/increasing the bed weight. This fact i s shown by figure 13. It i s quite possible that the mist was -so dilute that i t was not possible to detect any improvement of the heavier bed over the lighter bed. Meissner and Mickley (10) reported that an increase of bed weight improved the collection efficiency, but they worked with a more concentrated mist. An attempt to obtain greater impingement efficiency was carried out using venturi contactors, in which fine s i l i c a gel entered at throat ports. At the moment of contact between the fluid and the solids in the venturi throat relative velocites were high-up to 3000 cm./sec. From the values given for target efficiency the following theoretical results can be predicted: Vo Ut Vo He 500 O.lu 0.13 1000 0.28 0.28 2000 0 . 5 6 o.n 3900 0.81* 0.57 Up to 57$ collection efficiency based on an isolated target sphere and the space i t sweeps night be expected. This is a reasonably true picture 7 since, roughly there were about 10'' collecting spheres per minuter passing through the apparatus compared to 10^"0aerosol particles per minute. Collection by diffusion mechanisms might be expected to be at a minimum because of the relatively large separation of the collecting particles in the venturi and the extremely short collecting time. Percent removed in the venturi using tube B as an example, varies from 37$ at l,000cms. per second to l\2% at 3,000 cms# per second. The results obtained are of the expected order of magnitude, although the dependency of the efficiency on velocity is consider-ably less than was expected. Figure 17 shows the actual and the theoretical results• Figures l5> 16, 17, and 18 are plots of velocity versus percent not removed for the various venturi nozzles used. Although the removal ef-ficiencies obtained are not too good, they do improve with increased velocities, indicating that impingement is actually the collecting mechanism. The results shown by figure 18, for venturiD can be explained by the fact that the amount of air jetting up through the bed seemed greater for the uncapped nozzle than for the capped nozzle. Because of this the former would simply be a poorly f luidized bed and would be expected to behave as i t d id . I t should, however, be emphasized that the results for venturi D are not too useful since i t was impossible to determine throat velocities because only some of the gas passed up through the venturi . The graphs of the solids flow versus ve l -ocity, figures 19, 20 and 21, were determined to show how l i t t l e solids was picked up by the gas stream and to enable one to calculate the number of particles passing through the throat per minute. A s irni l iar curve for venturi A can be found i n the report by Forrest ( 2 ) . SUMMARY: The colorimetric analysis as reported was quite satisfactory and i t was used to measure as l i t t l e as 0.1 mgms. of di-n-octylphthalate. I t was found that the La^er-Sinclair generator, under suitable conditioi produced a reasonably homogenous aerosol of average droplet size 0.869 microns. The ester used was di-n-octylpmthalate. The results reported for the f luidized beds of 150/200 mesh s i l i c a gel show that the removal efficiency was independent of the in le t concentration over the range 0.765 to 0.965 mgms. of ester per cubic foot of a i r . They also show that the efficiencies were not affected by increasing the bed weight from 13.25 to 25.35 gros* per sq. cm. Theoretical calculations have been included to show the importance of the diffusional forces compared to i n e r t i a l forces as affecting this mist removal. The maximum removal efficiency'obtained was 88.8$ at a velocity of 3.20 cms. per second and a bed weight of 13.25 gms. per sq. cm. In order to increase the relative velocities of the aerosol droplets to the s i l i c a gel particles to study impingement mechanisms, venturi contactors were used. Although i n these cases the removal efficiencies were low - 35$ to U5$ - they did increase with an increase i n velocity indicating that the removal was, in fact, caused by impingement mechanisms. I t has been shown that the amount of s i l i c a gel entrained i n the gas stream was relatively small however^this fact i s not too important because theoretical calculations have shown that expected efficiencies would be about even for much greater solids flow. The actual efficiencies obtained are of the correct magnitude as predicted by the theoretical calculationsj however^the efficiency i s not as dependent on velocity as was expected. J"7 N O M E N C L A T U R E C 0 = inlet concentration of ester C 1 = inlet concentration of acid Cg = exit concentration of acid D^ = average diameter of s i l i c a particle Dp = average diameter of aerosol droplet E = percent removal efficiency P = 100 minus percent removal efficiency = target efficiency j ^ . = terminal velocity V 0 = average superficial velocity g = gravitational acceleration A S = average linear displacement R = gas constant T = absolute temperature t = time = Cunningham correction factor to Stoke*s law 2( = f l u i d viscosity N = Avagadro's number k = extinction coefficient W = bed weight per unit area n = empirical constant S = Klett, Summerson meter reading BIBLIOGRAPHY; 1. Fiegle, F. Laboratory Manual of Spot Tests p. 186-7 . New York, Academic Press (I9u3). 2. Forrest, D. B., Batchelor of Applied Thesis, Department of Chemical Engin-eering, University of British Columbia. 3 . Goetz, A., A m. j . of Public Health u3(2) 190-9 (1953). U. H i l l , U. T., Ind. E n g . Chem. Anal. Ed. 18 317-19 ( I 9 u 6 ) . 5. H i l l , U. T., Ind. Eng. Chem. Anal. Ed. 19 932 (19U7). 6 . LaMer, 7. K., and Gendron, P. R., Chem. i n canada, h UU(1952). 7. LaMer, V. K., Air Pollution Chap. 7U. McGraw H i l l Book Co., New York (1952). 8 . Lapple, C. E., Fluid and Particle Mechanics, Chapter 13 , and LU. University of Delaware, Newark, Delaware (195U). 9. Langmuir, I., and Blodgett, K. B., U.S.A.A.F. Tech. Report No. 5Ul8, Feb. 19U6. U.S. Dept. of Commerce, Office of Tech. Services P.B. 27565. 10. Meissner, H. P., and Mickley, H. S., Ind. E n g . Chem. Ul(6) 1238-U2 (19U9). 11. Perry, J. H., Chemical Engineer's Hand Book, p. 1013, McGraw H i l l Book co. New York (1950). 12. Pring, R. T., Air Pollution Chap. 35, McGraw H i l l Book Co., New York (1952). 1 3 . Product Engineering, p. 2lU (July 195U). lU. Reiss, H., and LaMer, 7. K., J. Chem. Phys. 1 8 , 1 (1950). 15. Rye, R. Department of Chemistry, University of Bri t i s h Columbia, Personal Communication. 16. Sinclair, D., Handbook on Aerosols, Chap. 6 , United States Atomic Energy Commission, Washington D. C. (1950). 17. Thompson, A. R., Australian J. S c i . Research 3A, 128-135 (1950). 18. Wilson, B. W., Australian J. Appl. S c i . 5 U7-57 (1953). 19. Wong, J. B., Rantz, W. E., and Johnstone, H. F., J. Appl. Phys. 26 2UU-9 (1955). 20. Wong, J. B. and Rantz W, E. Ind. Eng. Chem. 6 p.1371 (June 1952). DATA AND CALCULATIONS: Table A. For percent transmittancy of hydroxamate solution versus wave length 'Wave length B s B_ % transmittancy _ _ B s / B r  350 2 2 100 375 2 2 .5 80 u'OO 2 3 6 6 . 6 h$0 5 8 62.5 500 16 28 57 .1 550 29 kl 61.8 600 1*2 61.5 6 8 . 5 625 Uh 60 73.2 650 36 1*6 78.1; 675 22 27 8 l . i i 700 9 11 .0 81.7 Table B. Data for Calibration Curve for D, 0 . P. on the Klett-Summerson meter. KLett-Summerson milligms• meter rdg. di-octylphthalate 8 0 li5 0.1368 59 0.1710 102 0.31*20 115 O.I1IO4 135 0.5130 19k O.68I1O 6o C. Data f o r rate of color fading f o r typical hydroxamate solutions. Time sees* Run 1 K. S. rdg. mgrris.. ND.O.P., Time sees. Run 2 K . S . rdg. mgms. D.O.P. Time sees. Rune 3 K .S . rdg. mgms • D.O.P. a. 25 117 .101 20 68 .223 30 120 .U22 b. 50 115 .403 Uo 67 .219 60 118 . U i 5 c. 90 112 .391 60 67 .219 90' 118 .105 d. 270 107 .372 120 120 .208 120 116 .U07 e. 300 11U .Uoo f . 1500' 110 .38U Sample calculation for run l a using the calculated equation for the calibration curve. S = 261.8 C + 9.6 C r S - 9.6 261.B 117 - 9.6 = o.Ull mgms. 261.8 Table D. Calibration data f o r various esters. % transmission mgms. di-n-o ctylpththalate O.OU98 0.1328 0.3320 O.U980 0.5976 0.6972 0.7968 0.896U 0.9960 87 .0 76.0 57.0 U6.0 52.0 3U.5 28 .0 25.0 22.5 mgms. n-butylphthalate 0.0969 0.1610 0.3230 O.U850 0.6U60 0.9690 % transmission 57.0 55.5 U5.o 35.0 27 .0 18.0 mgms. ethyl benzoate 0.29UO O.UUlO 0.5880 0.8820 % transmission 38.0 29.5 2U.0 16 .0 -6/ Table E. Effect of parent acid on hydroxamate formation. mg. of ethyl benzoate mg. of benzoic acid Cenco meter rdg. 0.29U0 0 U8.0 0.291*0 0.1730 58.0 0.29l|0 0.1730 59.0 0.29uO 0.3u50 70.0 0.291*0 0.3li50 68.0 Table F. Operating data. Vapourizer temperature = 132° c. Reheater temperature = 162° c. Ionization coil current s 2.67 amps. Diameter of column - 2.25 in. Flow through the enucleation chamber. 1. Flowmeter reading _ 21.0 2. Inlet pressure = 17.0 Flow through the air preheater. 1. Flowmeter reading • 21.0 2. Inlet pressure - 16.0. The flowmeter calibration curves are listed in the Appendix C on page 80 « Table G. Data for removal by empty column* Run No. Through Bed inlet pressure i n Hg. p in Hg Bleed-off in l e t pressure in Hg. flowmeter Rdg. i n Hg. Vol. of Sample K.S. Rdg. DTO: ft.3 2T 0 .5 0.3 .502 116 .811 3T 0 .5 0.3 .700 159 .816 1*T o.5 0.3 .500 117 .822 6T — — .600 163 .978 7T — — .503 135 .952 8T — — ,5oo H*o .996 9 T 0 .5 0.3 .5oo 113 .790 10T o.5 0.3 .601 117 .681* LIT o.5 0 .5 .502 120 .81*0 12T 0 .5 0.3 .501 111* .796 i3T o.5 0.3 .501 116 .812 UlT o.5 0.3 .502 119 .833 15T — — . .600 170 1.022 16T — — .6oU 177 1.059 17T o.5 0.3 .5oo 119 .836 18T o.5 0.3 .503 117 .816 IX — — .5oo 11*8 1.058 2X — — .500 127 .896 3X o.5 0.3 1.0.Q1 221 .80 7 l*X o.5 0.3 .5oo 109 .760 5X o.5 0.3 1.1*0 10.20 .5oi 101* .723 6X o.5 0.3 i.l*o 10.20 .502 99 .679 7X o.5 0.3 3.10 2.80 .1*98 97 .671 8X 9X 10X 11X 12X 0 .5 o.5 o.5 0.3 0.3 0.3 3.10 2.80 .5oo .500 .5oo .502 .501 106 137 137 111 116 .736 .972 .972 .773 .831* * indicate'^ second f i l t r a t i o n step was carried out. !' indicates inches of carbon tetra-chloride. Method of Calculation For the T. series the average i n l e t mass loading was calculated and then from these, one value of collection efficiency was calculated. The calculations for the X series was similar except that i t was necessary to compute the exit concentrations at the several velocities used. These average points were plotted on figure 8 . THEORETICAL CALCULATIONS 1. To determine the value of the dimensionaless group Ut Vo g Dp Ut - 8(10"4) ft/sec - 2U.U (10"*) cm/sec ( l l ) g • 980 cm/sec/sec Dp - 89(10~4) cm (average of 150 to 200 mesh aperature) Vo - variable of 1 to 3000 cm/sec . . Ut Vo _ o.00028Vo g DP Table S - Theoretical target efficiencies for 150 to 200 mesh silica gel: 7? t Vo cms /sec Ut Vo g Dp fraction collected 500 0.1U 0.13 1000 0.28 0.28 2000 0.56 0.1+7 3000 0.81+ 0.57 Generator output - 1 milligram of ester per cubic foot. Approx. density of di-n-octylphthalate « 0.96 /. number of aerosol particles per cu. f t . • (0.01)(3) F 2 Ic 3 ; , 1 0 therefore number o f aerosol particles passing through column • 10^° per minute. A t r i a l calculation to check/o^aer of magnitude of results from the diffusion equations. Using the data for the 25.35 gsis per sqi.cm.bed weight and Vo, the superficial velocity, equal to 7.82 cm/sec (Run 5c) Prom figure 12 the amount of mist not removed " 18.6$. There-fore collection efficiency - 81.1$. The bed depth * 52 cms, therefore, volume of bed * (area)(height) • 1335 cc. Since the density o f silica gel » 0.85 gm/cc, and the bed weight * 651 gms then ^  , the void fraction, * 0,43. Then the true linear velocity - V6/0.U3 * 2.23 Vo. Contact time • bed depth - 52 • 2.85 seconds, true velocity 2.2317.82J From eq'n. for Brownian motion (pg. 286, Lapple) 4S - f URT l^ t J 3Tf a^NBp R - 8.31U (10 7 ) ergs/°C (gm. mole) T - 293° K t " 2.85 seconds - 1.81 (10" 4 ) poises N • 6.O64 (lO 2 3) molecules /gm mole Dp - 87 (lO - 6) cm k m - 1.202 AS > /(U)(8.31U)(107)(2930(l«2O2)(2.85) J (3)(3.LU)a(1.81)(10-4)(6.06U)(10»3)(87)(lO"**) AS • 10.7 microns Niow; since the void fraction € • 0.U3 then on the average every void (between particles is equal to 86$ of the volume of a particle. The volume of a particle » ^ D-,3 where Dp " 8 9 microns. Now assuming that as an approximation this void apace be considered similar to a sphere of (0.86)(89) microns diameter then in this volume half the aerosol particles lie further then (0.86)(89)(.793) - 30.4 microns: from 2 the center of void volume and half lie closer. On the average, then, every particle must travel from this median line to a solid surface, or 89(.86)_ -2 30.4 = 7.9 microns. The distance actually travelled was 10.7 microns. Therefore, collection efficiency should be 10.7 - 135.1$. Table H. Data f o r the 3 u l gram fluidized bed. Minimum depth r 23 cms. Maximum depth r 28 cms. Run Through Bed Bleed-off Dilution Vol. of K.S. No. Inlet pres. P Inlet pres. flowmeter Inlet pres. flowmeter Sample Rdg. in Hg. In Hg. i n Hg. rdg. i n Hg. i n Hg. rdg.in Hg. ft.3 Ik — — 0.500 2k — — 0.501 — 3A — — O.U98 lUo UA — o.5oo 150 5A* — — o.5oo 13U 6A* — — 0.501 129 8A# 0.30 .35 0.998 65 9A-* 0.80 .35 o.5o u.95" 1.003 58 10A* o.Uo .35 0.30 9.75" 1.000 U9 11A* o.Uo .35 0.30 18.75" 1.002 39 12A* o.Uo .35 0 . 4 0 2.65 mm-* 1.000 U3 13A* 0.90 .35 — — 2.90 9.95 0.999 69 lUA* 0.90 .35 — ~ 1.50 3.05 1.001 67 l£A* 0.80 .35 — 1.001 71 16A* — — — 0.500 132 17A* — — — 0.U99 138 IB — — o.5oo lUo 2B* — — — 0 . 5 o i 126 3B* — — — o.5oo 136 LB* 0.90 .35 — 1.001 60 $B* 0.75 .35 0.40 2.60 — 1.001 U9 6B* 0.50 .35 O.Uo 3.90" 0.998 55 7B* o.Uo .35 O.Uo 8.80" — 1.000 U7 8B* o.Uo .35 O.Uo 17.35" — 0.998 U3 9B* 0.80 .35 — — 1.25 11.15 0.991 75 10B# ov?5 .35 ~ — 0.8 U.U5 1.000 26.0 LIB* 0.80 .35 — 1.000 55 12B* 0 .75 ,35 - - 0.8 5-.15 1.065 60 13B» — — — o.5oo 132 li|B* — — — O.U98 132 l5B* — — — — --• O.U99 129 * indicates second f i l t r a t i o n step was carried out. "; indicates inchs of carbon tetra chloride. The method of calculation for the A series was similar. - to that of the T and X series. The concentration of ester was calculated from the reading for runs 6A, and 16A, 17A. The average of these was taken as the i n l e t mass loading and the collection efficiency calculation as the ratio of the exit mass loading over the average i n l e t concentration. The superficial velocity through the bed was calculated by correcting to suitable pressure. Sample calculation. Run No. mgms. D.O.P. per f t . 3 $k 0.952 6A 0.910 16A 0.93*1 17A 0.983 Avg. 0.9UU for run 8A K. So rdg. - 65 mass loading = .212 mgm./0.998 f t . 3 Percent not removed - 0.213 (11) - 22.5$ To calculate superficial velocity. Volume of air passing through bed - 16^ 66 l/m at 1 atmos. and 70° P. (from calibration curves) Avg. bed press. = 0.80 .55 = 0.7 Velocity - 29.9 (18.66) (1000) = 11.82 cm/sec, 3075 (25.66) J—WJ = (corr. v o l . i n lit./min.) ytec. ) (min.) cross ''section ( l i t r e ) (sec.) for Run 9A • i • ble.ed-off reading = k.95 in CCI4 a 5.78 lit./min. at 30.6 i n . c C Hg. ab. r 5.91 lit./min. at latm. Vol. through bed - 18.66 - 5.91 s 13.75 litres Velocity = 29^9 (13.75) )1000) = 8.75 cm/sec. 3 0 3 " (25.66) (~5o7 Percent not removed - .181* (100) =19 . 5 $ for Run 13A Vol. of dilution air added - 1*.06 l/m at 1 atmos. Velocity = 29.9 (22.22) (1000) = I4 .35 cm./eec. 3o77 25.66 T~5c~) Percent not removed = r : (0.227) (22.72) = 29.1$ (0.91*1*) (18.66) The corrected result was obtained by correcting the over-all collection efficiency for that amount removed by the empty column. Sample calculation Run number 12A Velocity - 3.08 cm./sec. Percent not removed s 13.6 % From figure 8 percent not removed by the empty column •1 - 70 % Therefore corrected value for percent nob removed by fluidized bed 3 13.6 * 0.70 - 1 9 #j£ 7o Table I. Data for the 651 gram fluidized bed. Minimum depth = kh cms. Maximum depth = 51* cms. Run Through Bed Bleed-off Dilution Vol. of K.S, No. Inlet pres. P Inlet pres. flowmeter inlet pres. flowmeter Sample Rdg, in Hg, in Hg. in Hg. rdg.in Hg. in Hg. rdg.in Hg. ft.3 IC - - — —_ — — —»«- 0 .500 192 2C* — — — O.U98 1U0 3C* — — — — 0.505 132 Uc* 1.0 0.7 — — 1.000 57 5c* .95 0.70 .80 6.U0" — 1.001 57 60* .70 0.70 .65 17.0" — .999 Uo 70* .65 0.70 .65 2.35 i — 1.000 38 8C* .65 0.70 .65 2.70 — 1.002 33 90* 1.0 0 .70 — — 1.005 55 10C* 1.8 0.70 5.20 2.85 1.000 50 11C* 2.7 0.70 8.30 U.Uo .998 52 12 C* 3.00 0.70 10. UO 5.35 l . o o U 50 13C* 1.0 0.70 — — 1.000 63 LUC* — — -- — o„5oo 152 15C* — — — — 0.501 137 ID — — — — 0.500 11*1 2D* — -- — — 0.502 121 3D* — — — — 0.500 123 1*D 1,0 0.70 — — 1.000 60 5D 2.60 0.70 8.00 1*.00 1.002 U9 6D 2,60 0.70 8.00 U.oo 1.005 58 7D 0.90 0.70 ,65 2.U0 — 1.000 U3 8D 0.80 0.70 .75 6.20 — .995 53 9D — — — — .5oo 136 IE — — — — .5oo 137 2E* — — — — .502 121 UE 0.9 .65 .30 2.50" u . 3 0 2*10 1.001 62 5E 0.9 .65 .30 2.50" 7.60 U.Uo 1.000 60 6E- 0.9 .65 .30 2.50" 5.62- 3.50 .998 6U 8E 0.9 .65 .30 2.50? 12.65 6.00 .998 60 9E 0.9 .65 .35 3.50" 985 5.oo 1.000 56 10E — — — . 0*U99 130 11E* — — -- O.U99 128 * indicates second filtration step was carried out. " indicates inches of carbon tetra chloride. Table J« Physical data on the Venturis Venturi A B C D Throat Diameter mm. 5.69 3.91 7.00 7.93 Throat Area sq. mm. 25.UU 12.01 38.50 U9.32 Port -^iameter mm. 2.87 2.k0 2.U0 open Port A r e a  sq. mm 12.9h 10.92 10.92 Table K. ^ata for Venturi A _ IQO gms. of gel in column. Run No. 6F 7F 8F 9F Through Bed inl e t pres. P in Hg. i n Hg. .15 .15 .10 .10 Vol.of Samole £t3 .502 .502 .5oo .SOU K . S . P-dg. 96 9h 139 159 7Z Table L. Data for Venturi B. C. 100 gms of s i l i c a gel i n column. Run Through Bed Bleed--off Dilution Vol.of No, inlet pres. P inlet pres. flowmeter in l e t pres. flowmeter Sample K.S in Hg in Hg. i n Hg. rdg.in H g . in Hg. rdg. ft.3 rdg IB mm mm 0.502 1U9 2N — — — — — 0.500 137 3N — — — — — — 0.502 153 UN — — — — — 0.506 1U2 5N neg. Wmmm — — .500 97 6N 0.25 it - - — — — .629 106 7N 0.25 tt 0.1 7.U0" — — ,5U9 99 8N 0.25 ti 0.1 7.U0" — ~ .500 93 9N 0,60 11 — — 1.2 5.U0 .502 87 ION 0,60 neg. — — 1.2 5.U0 .502 80 11N 0.60 it - - — 1.8 13.60 .500 69 12N 0.60 tt — — 1.8 13.60 .508 70 13N 0.35 " - - — — .500 92 I UN 0.35 it — — .500 96 15N 0.35 it 9*1 — .502 100 16N 0.35 ti 0.1 15.7'* — .500 102 17N .30 tt 0.2 2.00 — .500 95 18N .30 it 0.2 2.00 — .503 99 19N .30 tt 0.2 2.50 — .500 97 20N .30 ti o.2 2.50 — .500 89 2BI — —- — — .501 156 1M —_ — —— — .501 156 ai-i — — — — .500 150 3M — —— — — .50 2 139 UM — — — .500 1U6 5M <?«3 — 9,1 81a 60 e 5oo 97 6M 0.3 — 0.1 2.60 .501 90 7M 0.3 — 0.1 12.00" — — .500 92 8M 0.3 — 0.1 12.005»; — .502 93 9M 0.3 — 0.1 5.70" — .500 91 10M 0.3 — 0.1 5.70" — .500 99 I I 9 : 0.3 mm mm mm mm — .500 98 12M 0 . 3 — — — — .502 90 13M o.U — — — 0.9 6.10 .500 8U lUM — mm - - — — 15M 0.5 — - - 1B8 13.80 .500 69 16M 0.5 —— — 1.8 13.80 .501 73 17M 0.3 — — - - — .500 95 18M 0.3 — — — — — .502 100 19M — — — - - — .500 1U7 20M > — - - — .500 1U0 73 Table M. D ata for Venturi D. 1,59 cm. gap and no cap. 100 gms. of S i l i c a Gel in column. Through Bed Bleed-off Dilution Vol. of K.S. Run Inlet pres. P Inlet pres. flowmeter Inlet pres. flowmeter Sample rdg. No. i n Hg. in Hg. i n Hg. Rdg. in Hg. rdg.in o i l f t . 3 1J 2J 3J for runs with 5/8 i n . gap and no cap UJ — — 5 J * 0.25 neg. 6 J * 0.25 .05 7J 0.2 n 8J 0.2 it 9J 0.2 n 10J 0.2 » 11J* 0.2 it 12J* 0.2 tt 13J* 0.2 tt l U J * 0.2 it 15J* — — 16J* — — IK — — 2K — — 3K — — ilK — — 5K 0.3 fie-g. 6K 0.3 it ?K 0.3 ti 8K 0.3 tt 9K 0.3 II IOK 0.3 it 11K .05 $.60* .05 5.60U 12.90« 1.0 12J.90 12.90 1.0 12.90 .20 2.U5 .20 2.U5 0.2 1.80 0.2 1.80 0.1 12.it" 0.1 12.ii" 0.05 6.00" 0.05 6.00? 0.501 137 0.500 151 0.502 lU9 0.501 IhO 0.500 100 0.502- 109 0.503 98 0.500 91 O.U99 93 0.508 97 0.503 96 0.500 101 0.5U5 111 0.500 101 0.502 137 0.500 132 0.500 131 0.501 152 0.507 1U7 0.500 I42 0.502 107 0.500 102 0.500 107 0.508 100 0.500 110 0.500 103 0.500 LUO * indicates second f i l t r a t i o n step was carried out. "indicates inches of carbon tetra chloride. Table N. Data for lenturi D 1.59 cm. gap with plexiglass cap 100 gms of s i l i c a gel i n column. it un No. Through bed Inlet pres. P in Hg. i n Hg. Bleed-off Dilution Inlet pres, flowmeter Inlet pres. flowmeter i n Hg. rdg. i n Hg. rd. i n o i l Vol.of Sample ft3 K.S, rdg, 1L* — — — — .501 129 2L# — — — — .5oo lUo 3L* — — — — .5oo 142 liL* 0.2 neg. — — .500 100 5 L * 0.2 tt — — .503 100 6L* 0.25 re 1.10 10.2 .502 110 7L* 0.25 it — 1.10 10.2 .5oo 90 8L* 0.25 it o.5o 5.70" .5011 111 9L* 0.25 n o.5o 5.70" .503 l i U 10L* 0.25 it 1.25 2.00 .503 107 ILL* 0.25 tt 1.25 2.00 ,5oo . 113 IP* — .5oo 129 2P* — — .5oo 139 3P* — — ,5oo 130 U P * 0.25 neg. 1.2 12.h .501 85 5P* 0.25 n 1.2 12.U .U99 78 6P* 0.25 tt 1.2 12.U .509 90 7P* 0.20 it — — .500 109 •;8P* 0.20 n — — .500 101 9P* — — mm *- — .500 133 10P* 0.2 Neg. o.o5 6.10" . 5 o i 116 IIP* 0.20 1 0.05 6.10J? «502 120 12P* 0.20 f 0.05 6.10" .500 112 13P* 0,20 0.05 6.10" .5oo 119 * indicates second f i l t r a t i o n step was carried out-it indicates inches of carbon tetra chloride. 7J~ Table 0. s o l i d s flow data for venturi B. 100 gms. of s i l i c a gel i n column. Run No. 1 Manometer 1A 2 i n Hg. 2A Wt. ,of gelgms. Time seconds 1 16.0 16.2 16.0 12.9 9.5 30 2 16.0 1 ; . ^ 16.2 16.0 12.9 8.3 30 3 16.0 16.2 16.0 12.9 8.9 30 u 11.0 10.2 11.5 10.0 5.6 30 5. 11.0 10.2 11.5 10.0 6.3 30 6 11.0 10.2 11.5 10.0 6.3 30 7 10.0 8.3 9*5 8.0 2.5 30 8 10.0 8.3 9.5 8.0 1.9' 30 9 10.0 8.3 9.5 0.0 1*1 60 10 7.5 6.0 7.5 5.7 3.0 60 11 7.5 6.0 7.5 5.7 3's.O 60 12 7.5 6.0 7.5 5.7 2.8 60 13; 3.5 2.9 3.5 2.7 2.0 60 l i ; 3.5 2.9 3.5 2.7 1.7 60 15 3.5 2.9 3.5 2.7 2.U 60 16 17.0 17.8 16.5 17.0 7.1 60 17 17.0 17.8 16.5 • 17.0 8.0 60 18 17.0 17.8 16.5 17.0 7.2 60 19 same i as 18 plus 15.1 60 20 d i l ' n flow = 12.2 divisions 15.U 60 21 at 1.2 i n Hg. 15.3 60 Table P. SolxdSj flow data f o r venturi C. 100 gms of s i l i c a gel i n column. No. 1 Manometer 1A 2 i n Hg. 2A Wt. of gel gms. time seconds 1 5.5 i l . l 5.0 3.9 3.2 60 2 5.5 i l . l 5.0 3.9 2.8 60 3 5.5 4.1 5.o 3.9 3.0 60 U 11.0 8.9 10.0 8.6 ii .2 60 5 11.0 8.9 10.0 . 8.6 3.7 60 6 • 11.0 8.9 10.0 8.6 3.6 60 7 lii.O 13.ii lii.O 12.2 5.1 60 8 lii.O 13.ii lii.O 12.2 ii . 6 60 9 lii.O 13.U lii.O 12.2 ii.il 60 10 17.5 17.2 17.0 16.6 5.8 60 11 17.5 17.2 17.0 16.6 6.2 60 12 17.5 17.2 17.0 16.6 5.5 60 13 same as 12 plus 10.8 60 d i l ' n flow = 12,0 div. 11.2 60 15. at 1. 1 i n . Hg 10.9 60 77 Table Q. Solids flow data for venturi D. 100 gms. of s i l i c a gel i n column. Run No. 1 Manometers 14 2 2A Wt. of gel gms. ' Time seconds 1 2.5 1.9 2.5 1.7 U.o 15 2 2.5 2.5 1.7 U.8 15 3 2.5 1.9 2.S 1.7 U.2 15 U 8.5 5.7 8.0 5.2 19.5 15 5 8.5 8.0 5.2 19i7 15 6 8.5 5.7 8.0 5.2 19.0 15 7 11.0 7.3 i o . 5 7.0 22.3 15 8 15.5 13.7 15.5 n . 5 22.U 15 9 15.5 13.7 i 5 . o 11.5 23.2 15 10 15.5 13.7 i 5 . o n . 5 23.2 15 11 20.0 17.lt 20.0 15.8 25.2 15 12 20.0 17.4 20.0 15.8 25.5 15 13 20.0 17.4 20.0 15.8 2U.9 15 l U 22.0 IS 15 same as 13 21.8 15 16 plus d i l ' n flow of 13.7 i n o i l 23.8 15 17 at 1.1 i n Hg. 21.5 15 18 4.0 3.1 4.0 2.9 12.9 15 19 4.0 3.1 4.0 2.9 12.2 . 15 20 4.0 3.1 U.o 2.9 13.0 15 Table R. Aerosol Measurement data. 20 Drum Divisions = 1 micron. 1 16 17 18 16 19 2 18 19 18 14 17 3 18 19 17 19 17 k , 17 111 13 Iii 16 5 16 23 19 15 11 6 17 16 17 17 21 7 19 16 17 15 18 8 20 17 15 19 15 9 19 19 20 16 18 10 2h 20 16 16 19 11 Ik 12 21 18 17 12 17 20 14 18 16 12 18 18 18 15 19 111 19 19 19 19 15 15 16 17 15" 26 15 16 16 15 17 17 18 17 17 19 15 18 1$ 18 19 13 19 17 19 19 13 20 20 17 17 20 17 16 16 18 19 21 20 total3$8 3u9 378 360 3U1 over a l l total - 38/4O average particle size -19 Iii 17 16 15 10 20 20 17 21 15 17 17 25 19 19 21 16 17 16 15 19 17 20 16 19 16 111 11 20 19 lli 22 16 19 15 10 Iii 12 12 27 17 19 17 19 17 12 22 16 18 19 16 16 22 16 16 16 19 17 14 15 19 21 15 19 16 17 17 19 17 19 17 12 10 19 26 16 19 19 18 17 16 17 15 15 16 13 19 18 18 16 15 16 20 16 16 17 17 17 21 17 20 Iii 17 18 15 19 18 16 15 20 Iii 15 17 16 19 17 18 15 16 326 336 359 350 336 3hi 38ii0 - 0.869 microns 20(221) APPENDIX A. Solutions required for analysis: A. 2.5$ sodium hydroxide (4). Dissolve 2.5$ by weight of pure sodium hydroxide i n 95$ ethanol. Add approximately 1 milligram of sodium carbonate. Make solution fresh daily. B« 2.5$ hydroxylamine hydrochloride(4) Dissolve 2.5$ weight of pure hydroxylamine hydrochloride i n 95$ ethanol. Make fresh daily. Cc. Modified solution A. (5) . Dissolve O.U gms of iron i n 20 ml of 1 to 3 n i t r i c acid (1 acid to 3 water), add 15 ml of 70$ perchloric acid and heat to copious fumes of per-chloric acid. Cool and transfer to a 100 ml volumetric flask with the aid of 4O ml 5/water added from a pipette. -Add 10 ml of cone. n i t r i c acid and dilute to the mark with 70$ perchloric acid. Make a 1$ solution of this i n 95$ ethanol. The concentrated solution i s good indefinitely. The dilute solution i s good for a week. APPENDIX B: Formula for di-n-octylphthalates 0 CEj - CH3 t -C - 0 - CH2 - CH - CHj - CHa - CH3 -C - 0 - CEj - CH - CHjj - CHa - CH3 0 CHj - CH3 APPENDIX C. Flow Meter Calibration Data: Manometer 1 - meters flow to preheater Time Scale Rdg, Liters minutes ft3 1 div. = 1/L0W cfm " minute 3 0.312 5.5 0.101+ 2.95 U 0.k2h io.5 / O.lkl fc.Ol 3 0.773 17.5 0.191+ 5 . 5 0 3 0.765 2U.5 0 . 2 5 5 7.23 3 0.861 33.5 0.287 8 .13 3 0.996 U3.5 0.332 9.1+1 3 1.107 53.5 0.369 10.U5 3 1.180 61.5 0.393 11.15 3 1 .252 68.5 0.1+17 11.80 Temperature - 7 0 ° F . Pressure ~ 1 atmosphere Metering fluid - carbontetrachloride Metered fluid - air Manometer 2 - meters flow to m^ucleation chamber Time Scale Rdg. Liters minutes fta 1 div." 1/1°" cfm minute. 3 0.262 1+.5 0.087 2.1+7 3 0.1+20 10.5 0.11+8 3.86 3 0.1+79 13.5 0.1597 U.53 3 0.600 20.5 0.200 5.67 3 0.700 27.5 0.233 6.60 Temperature - 70° F . Pressure - 1 atmosphere Metering fluid - carbontetrachloride Metered fluid - air FIGURE 23 CALIBRATION CURVE FOR MANOMETER I : TO PREHEATER 2-5 0 ' ! I I 1 L 1 L 1 2 2,5 3 3.5 4 4.5 3 5.5 6. LIT E R S PER MINUTE OF AIR FIGURE 2 4 CALIBRATION CURVE FOR MANOMETER 2 '. TO NUCLEATION Manometer 3 - bleed o f f - manometer Time minutes f t 3 Scale Rdg. 1 div.» l/ 1 0 r t cfm Liters minute Inlet press i n . Hg. 3' .188 0.6 + .063 1.78 0.10 3 .319 1.4 + .106 2.9U 0.20 3 .31+5 1.7 + .115 3.25 0.25 3 .U23 2.3 + .ll+l 3.99 0.1+0 3 .530 4 .0 + .177 5.01 0.60 2 .U32 5.8 + .216 6.12 0.80 2 .572 10.2 + .286 8.09 1.1+0 2 .672 11+.1 + .336 9.50 1.90 2 .422 • 0.65 t .211 5.97 .80 2 .668 1.65 A .331+ 9.1+5 1.95 2 .811 2.25 A .406 11.1*5 2.60 2: .918 2.85 A 1-459 13.1 3.20 Temperature - 70° F. Metering f l u i d + carbontetrachloride Metered f l u i d - air A mercury Manometer 1+ • dilution manometer Time f t 3 Scale Rdg. cfm Liters minutes in.Oil minute 3 0.106 1.10 0.0353: 1.00 3 .199 2.80 0.0662 1.88 3 .259 1+.55 0.0865 2.1+5 3 .311+ 6.55 0.101+8 2.97 3 .361 8.35 0.1202 3.1+1 3 .1+11 11.90 0.1369 3.88 3" .1+50 13.1+0 0.150 1+.25 Temperature « 70° F. Pressure - 1 atmosphere Metering f l u i d - draft gage o i l Metered f l u i d - air S.G- - .826 L I T E R S PER MINUTE OF AIR I FIGURE 26 CALIBRATION CURVE FOR MANOMETER 4 DILUTION FLOW 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0059280/manifest

Comment

Related Items