EPA-650/2-74-088 OCTOBER 1974 Environmental Protection Technology Series ;:;:::::;:::;x;ii;;:::i:;:i:i:;^ ------- EPA-650/2-74-088 ASSESSMENT OF PARTICLE CONTROL TECHNOLOGY FOR ENCLOSED ASBESTOS SOURCES by C. F. Harwood, P. Siebert, and T. P. Blaszak IIT Research Institute 10 West 35th Street Chicago, Illinois 60616 Contract No. 68-02-1353 ROAP No. 21AFA-006 Program Element No. 1AB015 EPA Project Officer: D. K. Oestreich Control Systems Laboratory National Environmental Research Center Research Triangle Park, North Carolina 27711 Prepared for OFFICE OF RESEARCH AND DEVELOPMENT U.S . ENVIRONMENTAL PROTECTION AGENCY WASHINGTON, D.C. 20460 October 1974 ------- This report has been reviewed by the Environmental Protection Agency and approved for publication. Approval does not signify that the contents necessarily reflect the views and policies of the Agency, nor does mention of trade names or commercial products constitute endorsement or recommendation for use. ii ------- ABSTRACT The report gives results of a study to provide informa- tion, from both the literature and user contact, on the con- trol of asbestos emissions from enclosed sources. It assesses the state-of-the-art in asbestos emission control in terms of the devices or methods used and their efficiency. In addition, it gives results of a preliminary study to actually measure the effectiveness of baghouse control devices in con- trolling emissions from five asbestos plants, Baghouses are the predominant control device used in the asbestos industry. Cotton bags are used most frequently. Automatic shaking is used in most baghouses, with shake cycles of 1% to 4 hours most common. Most baghouses operate at two pressure drop ranges, 2.5-5 and 7.5-10 cm H^O. Air-to-cloth ratios range from 2 to 10:1. Published data on the removal efficiencies of the control devices was either non-existent, or quoted in general terms. Five baghouses were tested for removal efficiency in terms of mass and fiber number: although mass efficiency was very high, fiber concentrations exceeding 100 million fibers/cu meter, greater than about 0.05 ym long, are emitted. Using computer modeling, it was found that, even considering one source, asbestos concentrations of over 500 fibers/cu meter can be anticipated 5 km from the source. The exposure level at which asbestos in ambient air becomes a health hazard is not known. This report was submitted in fulfillment of IITRI Project Number C6291, Contract Number 68-02-1353, by the IIT Research Institute, under the sponsorship of the Environmental Protection Agency. Work was completed as of May 1974. iii ------- CONTENTS Page Abstract iii List of Figures v List of Tables vi Acknowledgements viii Sections 1 Conclusions 1 2 Recommendations 3 3 Introduction 5 4 Literature Survey 6 5 Compilation of Control Equipment User's Data 26 6 Emission Data Collection and Analysis 43 7 Estimate of Asbestos Dispersion 67 8 References 79 9 Appendices 82 IV ------- FIGURES Sampling Arrangement 45 Coordinate System Showing Gaussian Distributions in the Horizontal and Vertical 68 The Concentration of Asbestos Fibers with Distance from Source at Four Stability Conditions for Johns-Manvilie, Waukegan, Illinois 74 The Concentration of Asbestos Fibers with Distance from Source at Four Stability Conditions for Johns-Manvilie, Denison, Texas 75 The Concentration of Asbestos Fibers with Distance from Source at Four Stability Conditions for Raybestos, Marshville, North Carolina 76 The Concentration of Asbestos Fibers with Distance from Source at Four Stability Conditions for Johns-Manville, Asbestos, Canada 77 The Concentration of Asbestos Fibers with Distance from Source at Four Stability Conditions for GAF, Eden Mills, Vermont 78 v ------- TABLES No. 1 Filter Fabric Properties 13 2 Fabric Filter Applications -- Asbestos Mining and Manufacturing 19 3 Rule-of-Thumb Costs of Typical Collectors, Standard Mild Steel Construction 22 4 Typical Costs for the Canadian Asbestos Industry 23 5 Product Manufactured 28 6 Size of Asbestos Manufacturing Plants 29 7 Types of Asbestos Processed in Plants 30 8 Dust Control Devices 31 9 Method of Waste Disposal 32 10 Baghouse Manufacturers 35 11 Capacity of Dust Collection System 36 12 Air-to-Cloth Ratio 37 13 Bag Fabric 39 14 Bag Cleaning Mechanism 40 15 Bag Cleaning Cycle 41 16 Pressure Drop Across Bags 42 17 Samples and Sampling Conditions 60 18 Sample Weights and Mass Efficiency Data 61 19 Total Fiber Counts and Fiber Removal Efficiencies 62 20 Optical Microscope (500X) Size Distributions and Fractional Removal Efficiencies 63 21 Electron Microscope (16,364X) Size Distributions and Fractional Removal Efficiencies 64 VI ------- TABLES (cont.) No. Page 22 Key to Stability Categories 71 23 Dispersion Equation Source Terms 73 vii ------- ACKNOWLEDGEMENTS The help and cooperation of the asbestos industry is acknowledged. In particular, Johns-Manvilie, Raybestos, and the GAF Company, who gave the permission for their baghouses to be sampled, and assisted in the setting up of the sampling platforms. The help and guidance of the original Project Officer, Mr. Dennis Drehmel, and his successor, Mr. David Oestreich, is acknowledged. IITRI personnel who contributed to this program were David Becker, who produced the bibliography, Scott Preece, who obtained dispersion data through the COM computer model, Erdmann Luebcke, who undertook field sampling and optical counting, and Anant Samudra, who obtained electron microscope data. The Project Leader was Dr. Colin F. Harwood, assisted by Paul Siebert and Tom Blaszak. John Stockham had fiscal responsibility for the program. viii ------- SECTION 1 CONCLUSIONS A survey of the literature has revealed that there has been little reported on the specific problem of reducing as- bestos emissions to the atmosphere. A state-of-the-art review of dust collection has been made, and its application to asbestos has been stressed whenever possible. The litera- ture cites the baghouse as being the prime collector used in the industry. Information obtained from asbestos product manufacturers confirmed that baghouses are the predominant dust collection device used. Typically, baghouses in the asbestos industry use cotton fabric bags. The cleaning cycle is automated with 1% to 4 hours separating shake cycles. The usual operating capacity is 140 to 570 cubic meters per minute (5,000 to 20,000 cfm), with an air-to-cloth ratio of just under 3:1. Baghouses are normally free of operating problems. Normal installation costs range from $88.00 to $106.00 per cubic meter per minute ($2.50 to $3.10 per cfm) with operating costs ranging from $1.80 to $35.00 per cubic meter per minute per year ($0.05 to $1.00 per cfm per year). It is not common practice for the users of the baghouses to measure the efficiencies of their baghouses. No hard data on baghouse efficiency was obtained from users or from the literature. Preliminary field measurements undertaken during this study have confirmed the high efficiency of the baghouse in removing asbestos fiber from the air stream. ------- Both the mass efficiencies and measured fractional efficien- cies are 99+%. However, the dust load is high, resulting in large numbers of fibers being emitted despite high / C Q efficiencies. Emissions of about 10 -10 fibers/m of 78 3 >1.5 urn and 10 -10 fibers/m of <1.5 ym were found to be usual. The dispersion of asbestos fibers from a source was calculated using the Binomial Continuous Plume Dispersion Model (BCPDM) and the Climatological Dispersion Model (COM). The CDM model is designed to develop a long-term, quasi-stable pollutant concentration profile for a given area. Unreliable results were obtained using this model, which showed that it was not suited for the particular conditions of this study. The exact cause of the failure was not established. Results obtained with the BCPDM model were consistent. As expected, the estimated concentration values varied with the size of the emission source, the time of day, and the distance from the plant. At a distance of 1 kilometer from 37 the plant, peak values of about 10 to 10 fibers per cubic meter were calculated. A sharp decline was observed to about 10 kilometers distance, where the concentrations of about 2.5 x 10 to 5 x 10 fibers per cubic meter were cal- culated. Beyond 10 kilometers, the rate of decline of the fiber concentration was slow, at a distance of 30 kilometers, the values were calculated to be about 10 to 10 fibers per cubic meter. It is stressed that these values extend from a single source. Fiber emissions from other baghouses within the same plant and from adjacent plants would lead to increased ambient concnetrations. ------- SECTION 2 RECOMMENDATIONS This study has shown that current control devices emit very large numbers of small fibers. It is estimated that these fibers remain suspended and travel large distances from the source. It is recommended that methods to reduce these emissions should be investigated. Since baghouses are the accepted best method of reducing asbestos emissions, the most logical study would be to optimize the baghouse performance. A statistically designed experimental study would establish the efficiency controlling parameters. Baghouses could then be optimized to control emissions from specific asbestos waste types. While initial consideration should be given to optimizing baghouses, other types of devices should not be overlooked. It is recommended that both established and novel control devices be considered and tested either singly or in combina- tion with baghouses. For example, novel systems, such as foam injected cyclones, and established devices, such as elec- trostatic precipitators, have not been evaluated for asbestos emission control. It is also recommended that studies should be conducted to establish the fate of emitted asbestos fibers. The small size of the fibers, and their indestructable nature, implies that they may remain suspended indefinitely. The limited particle scavenging studies that have been conducted do not include fibrous particles. Studies on approximately spherical, ------- sub-micron particles indicate that scavenging by snow and rain is of ultra low efficiency. Implicit in the long-term sus- pension of these fine fibers, their chemical inertness, and the lack of an effective scavenging mechanism, is the possibility that the number of asbestos particles in the atmosphere is increasing. A test program is recommended to determine if this is true and, if so, its health signi- ficance. ------- SECTION 3 INTRODUCTION The control of hazardous emissions from operations involving asbestos products is mandatory under the applica- tion of the Clean Air Act. It is, therefore, necessary for the Environmental Protection Agency to evaluate the control technology applicable to enclosed sources of these hazardous emissions. In this way, it is possible to introduce legisla- tion which will require the application of operating practices capable of protecting the public health. The applicability and effectiveness of these practices will be supported by sound scientific procedures and experimental evidence. To develop the information required for this report, three methods were used: 1. The scientific literature was surveyed and a bibliography produced. Published literature per- tinent to the control of asbestos emissions from enclosed sources was reviewed. 2. Users and manufacturers of emission control equip- ment were contacted. Information on equipment applicabilities and the attainable degree of con- trol with various control approaches was solicited. 3. Sampling and analysis were conducted to obtain data on the emissions from five different enclosed sources. These manufacturing sources were selected because the control technology employed was typical of the industry. ------- SECTION 4 LITERATURE SURVEY INTRODUCTION Literaturfe Search The literature on emission control of asbestos has been surveyed by computer search. Very few references specific to asbestos have been found; for this reason, the search was broadened to include industries with related particulate emission problems. A bibliography has been produced; because of its large size, it has been bound under separate cover from the main report. It is some 231 pages in length and contains over 3,000 citations. References have been arranged by KWIC (key word in context), by author, and by citation. Abstracts of pertinent references are included in Appendix A. Types of Control Devices The major types of collection devices available are cyclones, wet scrubbers, electrostatic precipitators, and baghouses. Each of these devices has some applicability in the asbestos industry, although in the United States, baghouses are generally used as the final control device. Cyclones are inertial collectors, which collect particles by imparting a tangential velocity to the gas stream, so that the particles are moved by centrifugal forces to the cyclone wall, where they are collected. ------- Wet spray collectors, which are designed in a wide variety, operate by atomizing or breaking down liquid drop- lets to an extremely fine spray. The increased water surface area allows better contact with the particles and improves the chances that the contaminants will be captured. The waste, a wet slurry, is then removed from the gas stream by either centrifugal or mechanical means. Electrostatic precipitators are essentially a chamber with a series of high voltage electrodes between grounded plates. The corona around the discharge electrodes ionizes the gas, and, therefore, the dust which then migrates to the plates. The particles then fall, or are rapped loose or washed into hoppers. Micron and sub-micron material can only be collected by high efficiency bag filters, high efficiency scrubbers, or ESP's. Baghouses utilize fabric filters in a bag or tubular shape, which are mounted either in the plant itself or in separate modules. As particles are collected, they build up a filter cake which increases collection efficiency and the pressure drop across the filter. Occasionally, this filter cake must be removed, as the power required to overcome the pressure drop becomes too high. Use in the Asbestos Industry In asbestos mining, small fabric filters are used for control during drilling. Cyclones, bag collectors, and properly designed ducts are then used for dust control in the crushing operation. Cyclones, sometimes followed by baghouses, are used as control devices on the dryers. Asbestos milling involves crushing and separation from the dust by air aspiration, grading of the fibers by cyclones connected to the baghouse. Also connected to the baghouse are the screens, separators, recirculating systems, regrading ------- areas, pressure packers, and other dust control systems, i.e., vacuum systems. Generally, all conveyors are covered, and low velocity hoods are used for control of dust in other areas. (In many asbestos processing operations in which dust occurs, i.e., textiles and friction materials, cyclones and baghouses are also used.) Mill ventilation systems gen- erally receive 5070 of their load from the screens, 25% from dust control systems, and the remainder from various machines which open the fiber, separate dust from fibers, or segregate fiber lengths1. The milling process generally requires 9-10 tons air/ton ore processed, and approximately 10 ton air/ ton finished product2'3' * . The properties of asbestos which complicate its cleaning from ventilation air are5: 1) its fibrous nature, i.e., it tends to interlace and mat, thus creating pneumatic handling difficulties; 2) its friability, or ability to break down to smaller and smaller fibers so that a high efficiency is needed for sizes less than 10 ym; 3) as the size of the fibers decreases, they become increasingly affected by moisture so that an impenetrable cake is formed; 4) its low apparent specific gravity, which is one-fourth to one-fifth of its real specific gravity of 2.54-2.59, so that it is less easily collected by gravitational or inertial techniques. CYCLONES Cyclones are generally efficient collectors,, for particles which have diameters greater than 10-20 ym. Low efficiency cyclones can be used for 10-20 years in a high temperature, non-corrosive environment; however, their efficiencies are only 8070 for particles greater than 44 um6. High efficiency cyclones utilize a much smaller diameter and various other design refinements to achieve much higher efficiencies at the cost of a greater pressure drop. Due to their small diameters and correspondingly small flow 8 ------- capacities, high efficiency cyclones are usually used in multiple units. These multiple cyclones are capable of col- lection efficiencies of 100% for particles larger than 10 ym, and greater than 90% for particles larger than 2 ym6. Cyclones have the advantages of7: 1) simple design; 2) easy protection against wear; 3) high efficiency if scien- tifically designed; and 4) automatic locks to prevent leakage, Its major disadvantage is that for high efficiencies for micron-sized particles, very small diameters susceptible to clogging and wear are needed. For the large capacities needed for asbestos applications, small diameter, high efficiency cyclones are generally not practicable. In one asbestos mill in Quebec, twin cyclones of 2 m diameter were used from 1960-67 as the primary control device. The collec- tion efficiency achieved was only 70%, so that they had to be replaced when stricter emission regulations were effected8. Generally, low efficiency cyclones are used in asbestos milling to collect most of the fiber and a minimum of rock dust in the separation process1'1* . Cyclones are also used for the grading of asbestos into specific size ranges9. Also, they are frequently used as pre-cleaners before the final control device, which is usually a baghouse2. WET SPRAY COLLECTORS Wet spray collectors are available in a variety -of de- signs, including the following6'7 . 1. Water injection into fans modified with special fittings for ultrafine atomization of water. A single fan can clean up to 100,000 mj/hr (59,000 cfm) with a water flow of 0.1-0.3 1pm (0.026-0.079 gpm). ------- 2. Water sprayed filter beds containing several layers of specially shaped elements. The gas velocity is less than 2 m/sec previous to the injection of drop- lets from the spray wash at a flow of approximately 0.14 m , depending on the type and size of the dust. 3. Wet centrifugal scrubbers are more efficient than dry centrifugal devices such as cyclones; however, they have a greater corrosive action, tend to plug, and are more expensive to operate. 4. Water s pr ay s scrubber s are designed so that the air stream decelerates when the spray contacts it. 5. Baffle box scrubbers pull the dirty gas through several inches of turbulent water, sometimes using centrifugal action. 6. Venturi scrubbers atomize the water droplets by injection at high pressure into the high air velocity at the throat of the venturi section. As the venturi area increases, the larger droplets are still accelerating, while the gas and particles are decelerating, thus creating a second area of collection of particles on drops. Venturis operate at a liquid-to-gas ratio of approximately 0.2-0.9 &/m3 and are capable of achieving efficien- cies of 90-997o for particles less than 1 vim. All wet collectors have the disadvantages of requiring some water treatment for the wet slurry produced, and often for its removal from the gas stream. They can also require large amounts of water if the water in the slurry is not easily recycleable. The experience with wet collectors as the primary con- trol device in two mills showed the following difficulties: 1) blocking of the device; 2) corrosion; 3) difficulties in disposing of slurry products; and 4) overall efficiencies of 85-95%8. In wet processes used in some asbestos opera- tions, wet collectors are required, due to the nature of the operation. Wet processes are not generally used; they generate production difficulties because of the time required to dry the product. They also require additional handling after drying5. 10 ------- ELECTROSTATIC PRECIPITATORS Electrostatic precipitators are capable of cleaning a large volume of gas at high flow rates, temperatures, and humidities, A dry process gas requires either a very large precipitator, or a large humidity tower before cleaning10 . The advantages of electrostatic precipitators are7'5'8 : 1) no limits on theoretical efficiency achievable by design; 2) low pressure drop across the collector and, therefore, low fan power required; 3) exceptional reliability; 4) low operating cost for total power required of approximately 0.23 kW/100 m3/hr; and 5) simple construction. Their disadvantages are5: 1) a high installation cost, which in- creases exponentially with efficiency; 2) the design must be for a specific dust concentration; and 3) small particles may create a visible effluent, even though they are 98-99% efficient on a mass basis. Electrostatic precipitators are very rarely used for asbestos processes in the United States; however, they are more widely used in the United Kingdom. They have very high theoretical efficiencies, but for these efficiencies to be achieved, the process must be rigidly controlled with respect to particle velocity, moisture content, temperature, and dust concentration8. As asbestos milling produces a highly variable load, the design must be for a median operar tion, so that it will not achieve maximum design efficiency during normal operation5. BAGHOUSES Fabric Selection Baghouses (or fabric filters) are very efficient (95-99.9%) for smaller gas flows, if the temperature and humidity are not excessive. The choice of fabric used depends on: 1) temperatures of gas stream; 2) physical and 11 ------- chemical characteristics of the particles collected; 3) the chemical composition of the gas; and 4) the moisture content of the gas11 . A wide range of fabrics are available with varying characteristics and prices as shown in Table 1. In many applications, the temperature of the gas stream is the limiting factor; however, this is not a major problem in the case of asbestos. The determining fabric characteris- tics are12 the fiber used, whether it is woven or felted, its weight, permeability, tensile strength, abrasion resis- tance, collection efficiency, dimensional stability, whether or not it is napped, and its static charge. A woven fabric is suitable for lower air-to-cloth ratios of 0.46 to 1.24:1 m/min (m3/min of air/m2 of cloth, 1.5-4.0:1 cfm/ft2). With napping and blending, this can be increased to 3.1:1 m/min (10:1 ft/min). Felted cloths are usable at much higher air-to-cloth ratios of 1.24-3.72 m/min (4-12:1 ft/min); however, they are more expensive. Not all fabrics can be felted, and extremely fine particles can become embedded, making cleaning difficult. The weight of the 2 2 fabric generally ranges from 14-56 g/m (4-16 oz/yd ). The heavier fabrics have lower thread count of heavy, bulky yarns, include felted fabrics, and are less flexible. The weaves used for dust control range from one over one to four over one, and two or three over two. The higher the sum of the over and under threads, the higher the weight of the fabric, the collection efficiency, and the permeability. The permea- 2 bility is measured as cfm of air/ft of fabric at a pressure drop of 12.7 mm H«0 (0.5 in). The general practice is to use the most open fabric for the required collection efficiency. The tensile strength (or fiber tenacity in g/denier (fiber)) is recommended to be a minimum of 3.52-7.04 kg/cm2 (50-100 lb/in.2), while some fabrics, i.e., 2 nylon, have tensile strengths greater than 28.1 kg/cm 2 (400 lb/in. ). If all other factors are constant, increased 12 ------- Table 1. FILTER FABRIC PROPERTIES Fabric Cotton Wool Nylon Polypropylene Orion Acrylic Polyester Nomex Teflon Fiberglass Max. Temp. (°F) 180 190 225 225 260 260 275 450 475 500 (°C) 82.2 87.8 107.2 107.2 126.7 126.7 135.0 232.2 246.1 260 Abrasion Resistance Acid Alkali Resistance j Resistance i [ G | P , E G C i P E E G G E E F P F E E G P G G E F E F F F VG E G Tensile Strength G F E E G F G E G Moist Heat p 'P G f^ G G G E G E E Oxidizing Agents P P F G G G G G E G Cost Most Economical Economical Low-Moderate Moderate Moderate Moderate Moderate Expensive Very Expensive Expensive E VG G F P Excellent Very Good Good Fair Poor (Table abstracted from Reference 12) ------- tensile strength implies increased bag life. The abrasion resistance of filament fibers is usually greater than the staple form. The abrasion resistance is one of the critical factors in determining bag life, as it is decreased by either yarn failure by surface abrasion or intrayarn abrasion. The collection efficiency is determined along with the pressure drop by the filter cake, which builds up on a par- ticular fabric. Therefore, the fabric must be strong enough to support the increasing cake, and tight enough to "heal" quickly after cleaning. Dimensional stability is a problem, especially with synthetics, which have a tendency to stretch under high loads, or shrink at high temperature. Either type of change can change the porosity and permeability, i.e., make the fabric too loose or too tight. The most dimensionally stable fabrics are glass fibers and dacron. Napping of a fabric exposes more surface area for collection, and thus increases efficiency; however, it makes the fabric more difficult to clean. Napping is useful in light dust loads, low pressure drop, and high air-to-cloth ratio. Static electricity can be a factor in that particles can charge the fabric under certain conditions. This can improve efficiency and/or interfere with cleaning; however, it is generally considered that the detriments outweigh the advantage12 . The bag can be arranged in several ways12'13. Flat bags have the advantage of a high surface area-to-volume ratio, made possible by bag spacing. Collection is on the outside of the bag, so that if they are too close, the dust will bridge between two bags. Cleaning is usually by mechanical shaking or reverse air. Multiple pocket tube type bags collect dust on the inside, with the disadvantages that there are limitations on the average gas velocity at the seam, and dimensional instability. Single tubular bags are used for either top or bottom entry for inside collection. ------- Supported tube type bags are needed for collection on the outside of the bag. The bags are usually felt, capable of handling a high air-to-cloth ratio, and are cleaned by some sort of air pulse. Cleaning Mechanisms There are several major types of cleaning mechanisms: 1) mechanical shaking, 2) reverse air, 3) pulse air cleaning, or 4) sonic cleaning12'13. Mechanical shaking can either be by hand, which requires that section of the unit to be off-stream, or continuous or automatic shaking. This method uses a mechanical device and several preset, usually adjustable, timers to shut down one section of the baghouse at a time for cleaning. Reverse air methods are either of the atmospheric pressure, off-stream variety, which can be combined with mechanical shaking, or the high pressure type, which need 2 not be off-stream. Pulse air cleaning uses a 4.22-7.04 kg/cm (60-100 psig) pulse in supported tubular bags to create a shock wave, which moves the cloth off the supports and dis- lodges the filter cake. Its major advantage is that, like high pressure reverse cleaning, it can be done while the system is on-line. Sonic cleaning uses a low frequency air power horn or sonic generator operating at 250 cps. This creates a sympathetic vibration in the bags, which shakes loose the cake for cleaning by reverse air flow. Sonic cleaning and other methods, such as bubble cleaning, are very rarely used in practice. Design Characteristics The design of a baghouse5'12'13>llt is decided by the following characteristics of the dust to be collected and the gas stream. The type, size, acidity or alkalinity, and abrasiveness of the dust, along with the moisture and temperature of the air determine the filter fabric to be used. The size of the baghouse is determined by: 1) the 15 ------- dust loading, 2) the estimated air flow, 3) the pressure drop across the fabric, and 4) the gas velocity needed for the filtering characteristics for the particular dust. Bag sizing is determined by the type of cleaning method used with the limitation of a maximum length-to-diameter ratio of 30:1 for a uniform filtration profile from the botton to the top of the tube, and an acceptable entrance velocity at the inlet. Maintenance and inspection considerations require compartment walkways at cell plate and bag suspension level, and that bag grouping and suspension is suitable for effective accessibility. The critical characteristics1*'12 >13 of a baghouse are air-to-cloth ratio (theoretically the same as the linear velocity through the cloth, which actually varies with loca- tion), capacity, and pressure drop. The air-to-cloth ratio is generally expressed as the total gas flow in cfm (0.028 m3/min) divided by the square feet (0.9 m ) of cloth. It generally ranges between 1.5:1 and 4.0:1 for woven fabrics and ranges between 4.0:1 and 12:1 for felted fabrics for efficient collection. The capacity is determined by the amount of dust-laden air to be cleaned. In combination with the air-to-cloth ratio found to be necessary for the efficient removal of a specific dust, the capacity determines the total fabric area needed for the particular application. The pressure drop across the baghouse determines the fans required to push, or, as is more common to prevent erosion of fan blades from particulates, pull the air through the baghouse, The fan capacity will then require a certain level of power for operation of the baghouse. The need for preventive maintenance measures is deter- mined by consideration of the following15 : 1) pressure drop measured by manometer, 2) velocity pressure measured by 16 ------- manometer and pilot tube, with the; Dust Velocity = 4.005 x 10 x /Velocity Pressure where Dust Velocity is in ft/min (0.3 m/min) Velocity Pressure is in in. H20 (2.54 cm H20) 3) ammeter readings to measure the current to each fan motor, which is proportional to the air volume through that section of the baghouse, 4) the temperature of the gas, which will endanger the fabric life if it is too high. These measure- ments should be taken in conjunction with weekly visual inspections of the bags for leakage or breakages. The measurements should be compared with previous ones to determine any trends, such as an increase, decrease, increasing change, or stabilization. These trends can be interpreted as to the necessity or effectiveness of various maintenance procedures. Advantages and Disadvantages of Baghouses The advantages of baghouses for general dust cleaning operations include the following5'13 . 1. They are efficient down to sub-micron sizes. 2. They create a positive barrier to the dust laden air not dependent on supplementary devices or materials, changing direction, or electrical charging. 3. They present no secondary pollution problem., as does a wet collector. 4. Baghouses are extremely reliable. 5. Efficiency is fairly uniform over a wide range of particle sizes. The disadvantages of baghouses and their operation are either limitations on their application or maintenance difficulties produced by poor design. The temperature of the gas stream must be above dewpoint and less than 93.3° C 17 ------- (200° F) for most natural fabrics, while some synthetic fab- rics, especially Dupont "Nomex" and fiberglass can be used at temperatures up to 232.2-260° C (450-500° F). Baghouses also have the problem of fairly high pressure drops, depending on the fabric used and the design of the baghouse, thus necessitating considerable power consumption and operating expense. High moisture content of the gas is also a problem, as natural fabrics will rot at high humidities, and all fabrics tend to plug if the gas stream is very wet. Location of a broken bag can cause maintenance problems in some baghouse designs. Large space requirements can also be a limiting factor in some applications. Baghouses are the control device most commonly used in the asbestos industry. They are used so predominantly for asbestos control because of; "their extreme efficiency for asbestos and asbestos cement dust"1, "the toxicity of the dust, and the value of the product"13 . Collected raw asbestos is generally recycled directly to regrading screens; however, its resale value is $13.60-16.33/metric ton ($15-18/ton)3. The application of different types of bag- houses to various phases of the asbestos industry are shown in Table 2. Baghouses for asbestos mills are frequently purchased separately, and installed in the mill building. Therefore, they have the advantage of being able to recirculate warn, heated air in the winter, and to cool by the draft created in the summer. Recirculation of baghouse air can create an OSHA problem in that plant air can contain no more than 3 2 fibers greater than 5 ym in length per cm . There is a trend at present toward modular construction of baghouses, which decreases the cost and time involved in field instal- lation, but lose the advantages that a custom designed unit would have1*'12'17 18 ------- Table 2. FABRIC FILTER APPLICATIONS --ASBESTOS MINING AND MANUFACTURING Application Asbestos milling Asbestos ore dryers Asbestos cement raw material handling Asbestos cement finishing machines Textile carding Type of Collector Continuous Continuous Continuous Intermittent Intermittent Type of Cloth Cotton sateen Orion Cotton sateen Cotton sateen Cotton sateen Length of Bags cm ( in . ) 427 (168) 427 (1*8) 320 (126) 427 (168) 320-427(126-168) Diameter of Bags cm (in.) 12.7 (5) 12.7 (5) 12.7 (5) 12.7 (5) 20.4 (8) Air: Cloth Ratio m/min (ft/min) 0.75-0.91 (2.5-3.0) 0.75 (2.5) 0.75 (2.5) 0.62 (2.0) 1.55 (5.0) Expected Ap cm H70 (in. H20) 6.35-10.16 (2.5-4.0) 3.82-5.08 (1.5-2.0) 7.62 (3.0) 3.82-5.08 (1.5-2.0) 3.82-5.08 (1.5-2.0) Data based on several plants of Johns-Manvilie Corporation. Abstracted from Reference 2. ------- The only disadvantage of baghouses peculiar to asbestos is that the filter material and weave must be such that it is least possible for the fibers to interlace and interlock with the fabric5. The limitations on gas temperature and moisture content may also present problems in certain situa- tions; however, they can usually be overcome by the proper selection of fabric and the design of the baghouse. As an example, insulated baghouses using Nomex are used on ore dryers. An air-to-cloth ratio of 0.62-0.75:1 m/min (2.0-2.5:1 ft/min) is preferred for asbestos removal; however, a ratio of 0.91:1 m/min (3.0:1 ft/min) is more generally used, as it is more economical due to reduced space, blower, and filter bag requirements *. Other operating characteristics are determined by the size of the plant, the design of the particular baghouse, and the filter fabric used. COSTS OF OPERATING AND INSTALLING CONTROL DEVICES The economic and technical factors which determine the selection of a control device for a particular gas cleaning application are: 1) particle size, 2) pressure drop, 3) wet or dry state of gas and pollutant, 4) efficiency re- quired, 5) temperature of gas, 6) wet or dry effluent, 7) presence of dewpoint collectors, 8) cyclic variations of volume, temperature, and concentration, 9) availability of services, i.e., water, electricity, compressed air, etc., 10) pre-filtration conditioning requirements, 11) cost of initial dust and fume containment and effluent discharge sec- tions of the total installation, 12) corrosive conditions, 13) relative importance of capital and operating costs, 14) consequential costs from plant failure, 15) commodity values of the effluent, 16) disposal values of the effluent, 17) intermittent of continuous operation, 18) available supervisory and maintenance staff, 19) site restrictions, and 20) total gas volumetric flow19 . 20 ------- Graphs and tables of capital and operating costs ver- sus gas volume for various types of collectors are given in numerous publications19'20'21'22'23. The rule-of-thumb costs for typical collectors are given in Table 3. For asbestos opera- tions in particular, they are given in Table 4. In general, in the United Kingdom, ventilation aids add 20-37% to the cost of a new plant, the total dust extraction bill is 7% of the wages bill, and 2.7% of the total cost of production of asbestos fiber3. Fabric bags are responsible for 20-40% of the total equipment cost and the average bag life is 18-36 months, so that an installation operating for 10 years must be rebagged 3 to 7 times18 . For a baghouse, the installation-to-purchase cost ratio is typically 1.8 with an average low of 1.5, high of 2.0, and an extremely high value of 5.022 . The maintenance costs of baghouses are dependent on 1. type of bag suspension 2. internal walkways which determine bag accessibility 3. floor plan, i.e., number of rows between walkways, which determines the reach to the furthest bag 4. external access to upper and lower levels 5. dampers and operators 6. compartment isolation, i.e., whether or not each compartment's damper has separate controls 7. the type of cleaning mechanism (In this article11* , separate reverse air fans and capacities per com- partment of 0.12 for mechanical shaking, 0.39 for simple collapse, and 0.62 nr/min of reverse air per m2 of cloth (0.4, 1.25, and 2.0 cfm/ft2, respec- tively) in the compartment are recommended.) 8. bag cleaning controls, i.e., whether there are ad- justable program timers for the frequency and duration of bag cleaning and dust settling periods, pressure activated switches, or intermittent cleaning. 21 ------- Table 3. RULE-OF-THUMB COSTS OF TYPICAL COLLECTORS, STANDARD MILD STEEL CONSTRUCTION Type of Collector Equipment Cost $/lpm ($/cfm) Erection Cost $/lpm ($/cfm) Yearly Maintenance and Repair Cost $/lpm' ($/cfm) Mechanical collector Electrostatic precipitator Fabric filter Wet scrubber $0.002-0.009 ($0.07-0.25) $0.009-0.035 ($0.25-1.00) $0.012-0.044 ($0.35-1.25) $0.004-0.014 ($0.10-0.40) $0.001-0.004 ($0.02-0.12) $0.004-0.018 ($0.12-0.50) $0.009-0.018 ($0.25-0.50) $0.014-0.056 ($0.04-0.16) $0.0002-0.001 ($0.005-0.02) $0.0004-0.002 ($0.01-0.025) $0.001 -0.003 ($0.02 -0.08) $0.001 -0.002 ($0.02 -0.05) Data abstracted from Reference 18. ------- Table 4. TYPICAL COSTS FOR THE CANADIAN ASBESTOS INDUSTRY to Type Cyclone collector Multiple cyclone Wet collector Bag filter Electrostatic precipitator Typical Emission Rate kg/hr (lb/hr) (IT 135.0 (300) 45.0 (100) 3.2 (7) 0.5 (1) 2.7 (6) Cost per Process Rate cost/metric ton/hr (cost/ton/hr) $ 33 ($ 30) $ 66 ($ 60) $440 ($400) $880 ($800) $440 ($400) (1) Per ton per hour processed Data abstracted from Reference 8. Cost per Flow Rate cost/lpm (cost/cfm) $0.003 ($0.09) $0.006 ($0.18) $0.042 ($1.20) $0.085 ($2.40) $0.035 ($1.00) ------- 9. protection of bags by arrangement, suspension, and whether or not internal supports are required for the cleaning mechanism 10. hopper discharge valves 11. materials handling requirements11* Several examples of dust collection costs in the asbes- tos industry are cited in the literature. Johns-Manville Corporation is probably the largest asbestos manufacturer in the world. It has hundreds of fabric filters in over a hundred air and dust handling systems at a total cost of $18 million9. Their largest single unit at Asbestos, Canada processes 700-800 tons of ore per hour and handles approximate 127,000 m3/min (4.5 x 106 cfm) at a cost of $8 million. Johns-Manville has stated that total installation costs range from $70.50-$176.00/m3 ($2.00-5.00/ft3) at room temperature, though they are higher at higher temperatures2. Another o Quebec asbestos mill cost $1.9 million for a 19,800 m /min (700,000 cfm) installation8. An amosite mill in the United Kingdom, producing 21,768 metric ton/yr (24,000 ton/yr), has O a dust control system handling 2,550 m /min (90,000 cfm) operating at 99.5% efficiency for particles greater than 5 ym size which cost 27.5% of the total capital cost of the plant and operates for $195,000/yr3. Examples of other baghouse applications and their costs in other industries can also be found in the literature; however, their appli- cability to asbestos operations is only general, as the fibrous nature of asbestos creates special cleaning diffi- culties because of the ultra small particle size and the need for extremely high efficiencies. SUMMARY Baghouses are the most common control device in the asbestos industry. They are used both with and without cyclones as pre-cleaners. Baghouses are extremely efficient 24 ------- for the control of asbestos dust, achieving efficiencies over 99%, and have the additional advantage of recycling valuable material. An air-to-cloth ratio of 0.62-0.75:1 m/min (2.0-2.5:1 ft/min) is preferred, but a 0.91:1 m/min (3.0:1 ft/min) ratio is more common in the asbestos industry because of the saving in equipment costs. Baghouses may either be installed within the mill building with the advan- tage of recirculating heated air in the winter and cooling in the summer, or of separate construction. Temperature restrictions are not critical in most applications, so the fabric used is generally decided by the type of cleaning mechanism and the cost related to the bag life. Bag shape is also generally determined by baghouse design and cleaning mechanism. The costs of a baghouse installation in the asbestos industry are widely variant, depending on the size type of plant, baghouse capacity, type of cleaning mechanism, and filter bag used. Johns-Manvilie, the largest asbestos manu- facturer in the United States, suggests a cost of o 3 $70.50-$176.00/m ($2.00-5.00/ft ) at room temperature. Other sources suggest capital costs of $0.02-0.06/lpm ($0.60-1.75/cfm)18, and $0.08/lpm ($2.40/cfm)8. Annual operating costs are estimated as $0.07-0.28/lpm ($0.02-0.08/cfm) for maintenance and repair18 . Annual operating costs for blower operation are highly dependent on fan efficiency and local power costs. 25 ------- SECTION 5 COMPILATION OF CONTROL EQUIPMENT USER'S DATA INTRODUCTION 'Based on our user's inquiries, baghouses are the predominate method of controlling asbestos emissions. Typically, these baghouses use cotton fabric and automatic 3 shakers. The usual capacity is 140-570 m /min (5,000-20,000 cfm) with an air-to-cloth ratio of less than 0.91 m3/min/m2 of cloth (3.0:1 cfm/ft2). The time between bag cleanings is less than 4 hr. Baghouses are relatively free of operating problems. Normal installation costs ranged from $0.088-0.106 per 1pm ($2.50-3.00/cfm) with operating costs ranging from $0.0018 to $0.035 per 1pm per year ($0.05-1.00/cfm/yr). SIZE AND PRODUCTS OF THE PLANTS The listing of asbestos users from all available sources totaled 249 locations. Of these, 56 could not be contacted, 36 refused or were unable to give any information, 31 no longer use asbestos, 35 use asbestos, but have no dust collectors, 12 did not reply to written inquiries, and 79 plants gave some information about their operations and dust collection. The companies or locations which use asbestos but have no control equipment were either involved in manufacturing finished products from pre-processed as- bestos of some form (88.6%), or in the sale of asbestos or processed asbestos (11.470). The companies which no longer 26 ------- use asbestos had generally used very little asbestos previously and had found substitute materials due to regula- tory pressures on the use or control of asbestos. The plants which did give information did not always reply to all the questions or did not have the information that was requested. Therefore, the total number of plants reported for any given inquiry may differ by a large degree from the total number which gave some information on all inquiries. The products manufactured by the 79 respondents are listed in Table 5. The largest number of plants manufactured either friction materials (19.0%), textiles (14.4%), floor tile (12.2%), or roof shingles (10.6%). The size of the respondent companies, both by total employees and those in handling asbestos are shown in Table 6. On a total employee basis, 54.1% have 200 or less, while on a basis of employees handling asbestos, 65.5% have 100 or less. The types and amounts of asbestos consumed are listed in Table 7. Chrysotile is used by 81.3% of the plants responding to this question; 50% use 907 metric tons/year (1,000 tons/yr) or less, and 31.3% use more than 4,535 metric tons/yr (5,000 tons/yr) of chrysotile. TYPES OF CONTROL DEVICES USED The number of plants using any type of control device and the total number of each device being used are presented in Table 8. Baghouses are used in the overwhelming majority of plants (80.07o), and also are the most popular in number of devices used (90.1%). Wet scrubbers are the second commonest used by plants and third by number; however, these devices comprise only 6.8% and 2.1%. of these groups, respec- tively. It was also found that the scrubbers used are wet 27 ------- Table 5. PRODUCT MANUFACTURED Asbestos-Containing Product Friction materials Textiles Floor tile Roofing shingles Cement pipe Gaskets Paper Wall board, siding, and/or plaster Insulation Raw asbestos Asbestos cement Paint, asphalt Liquid filter medium Industrial rubber Steel mills and foundary additives and insulation Chemical corrosive resistant materials Total Number of Plants No. of Plants Manufacturing Product 17 13 11 10 7 7 7 6 3 2 2 1 1 1 1 1 90 Percent 19.0 14.4 12.2 10.6 7.8 7.8 7.8 6.7 3.3 2.2 2.2 1.1 1.1 1.1 1.1 1.1 100.0 28 ------- Table 6. SIZE OF ASBESTOS MANUFACTURING PLANTS Size Category No. of Employees <100 101- 200 201- 300 301- 400 401- 500 501- 800 801-1,000 1,000-1,500 1,501-2,500 Total Plants with Total Employment in Indicated Size Category No. of Plants 14 19 11 4 3 4 2 1 3 61 7o of Plants 22.9 31.2 18.0 6.6 4.9 6.6 3.3 1.6 4.9 100.0 Plants with the Indicated Number of Asbestos Process Employees No. of1 Plants 36 12 4 1 1 1 55 r 70 of Plants 65.5 21.8 7.3 1.8 1.8 1.8 100.0 to VO ------- Table 7. TYPES OF ASBESTOS PROCESSED IN PLANTS Type of Asbestos Chrysotile Amosite Crocodolite Processed (2) Total Plants Usi No. 52 2 2 8' 64 ng Type Percent 81.3 3.1 3.1 12.5 100.0 Amount of Asbestos Used(l) <. 91 £. 100 1 (6.2) ___ «__ 92-453 101-500 4 (25.0) 1 (100.0) 454- 907 501-1,000 3 (18.0) 1 (100.0) •MM 908-1,361 1,001-1,500 1 (6.2) ____ -T_t_ - 1,362-4,513 1,501-5,000 2 (12.5) ^.^__ > 4,513 > 5,000 5 (31.3) metric tons/yr tons/yr u> o (1) For those 18 plants which, reported Quantity Consumed (2) Processed asbestos includes those industries which do not use raw asbestos, but rather some pre-processed form, such as insulation, cement, or cement pipe for fabrication of end products. ------- Table 8. DUST CONTROL DEVICES Control Device Baghouse Scrubber Cyclone-baghouse combination Cyclone Filter systems Scrubber-baghouse combination Total Plants Using Device Mo. 72 6 4 4 3 1 90 Percent 80.0 6.8 4.4 4.4 3.3 1.1 100.0 Total Devices Used No. 335 8 12 7 6 4 372 Percent 90.1 2.1 3.2 1.9 1.6 1.1 100.0 31 ------- centrifugal scrubbers designed and utilized for the control of wet processes in plants which also have baghouses for the control of dry processes. This type of wet centrifugal scrubber has an efficiency of 98% plus and is EPA approved for wet processes. Cyclone and cyclone-baghouse combinations are the third and fourth most common control systems. Both are used by 4.47o of the plants; however, the cyclone-baghouse combination comprises 3.2% of the total control devices, while cyclones alone comprise only 1.9%. Cyclones as the sole method of dust control in asbestos applications are rapidly declining in number. They are being replaced by baghouses, because their collection efficiencies cannot meet EPA regulations. Cyclone-baghouse or scrubber-baghouse combinations have the advantage of increased filter bag life, as the gas stream is precleaned of the larger, abrasive particles before it reaches the bags. Central vacuum filter systems are generally used in fabricating operations on asbestos products such as brake linings, asbestos cement pipe, or asbestos insulation. In design, these systems are often similar to a small-scale baghouse. WASTE DISPOSAL AND EFFICIENCY TESTING The methods of waste disposal and their relative usages are given in Table 9. The most prevalent methods of waste disposal are reuse (39.2%) and dumping (37.0%,). Reuse is most common in those industries which use raw asbestos as the raw material and use separate dust collection systems for it, e.g., cement pipe and insulation manufacturing. Of the plants which responded to whether or not they had conducted efficiency tests, 62.8% replied positively, i.e., that some kind of testing, including OSHA testing, had 32 ------- Table 9. METHOD OF WASTE DISPOSAL Method Reuse Dump Landfill Sell Store Wet slurry* Total No. of Plants Using Method 38 36 13 5 3 2 97 Percent of Total 39.2 37.0 13.4 5.2 3.1 2.1 100.0 * Form of Waste Ultimate Disposal not specified. 33 ------- been done, while 37.270 replied negatively. However, of those 37.27,, 7.77° of the total stated that they accepted the manufacturer's values. Very few (19) companies were willing and able to give numerical values for efficiency. Of those reporting, 73.7?0 had mass collection efficiencies greater than 99.0%. DETAILS OF BAGHOUSE USE As baghouses are the predominant means of asbestos con- trol, a more detailed inquiry was made as to the sizes and characteristics of those used. The major manufacturers of baghouses used in the asbestos industry are Wheelabrator-Frye the Pangborn Division of Carborundum Corporation, and W.W. Sly. Their products are used in 21.3, 13.9, and 10.67= of the plants reporting, and amount to 27.2, 13.8, and 14.27o of the total number of baghouses reported, respectively The complete listing of manufacturers is presented in Table 10. The gas flow capacities of both the plants and the baghouses can be found in Table 11. Of the plants replying, 61.9% had total flows of 2,830 m3/min (100,000 cfm) or greater; however, 87.670 of the baghouses in the plants are 577 m3/min (20,000 cfm) or less and 98.57o of the plants have baghouses within this range. The air-to-cloth ratios 3 in m /min of gas flow:square meter of cloth area (cfm of gas:ft of cloth) of 0.75:1 (2.5:1) or less and from 0.75:1 to 0.91:1 (2.5-3.0:1) are used by 28.670 of the plants. While 40.07o of the baghouses have ratios of 0.75:1 (2.5:1) or less, 20.97o have 0.75-0.91:1 (2.5-3.0:1) and 30.970 have 1.24-3.10:1 (4.0-10.0:1). Generally, the ratios of 0.91:1 (3.0:1) or less are on mechanically shaken units, while those with ratios of 1.24:1 (4.0:1) or greater are on reverse jet units. The air-to-cloth ratios used are given in Table 12. 34 ------- Table 10. BAGHOUSE MANUFACTURERS Manufacturer Wheelabrator-Frye Pangborn Div. , Carborundum Corp. W.W. Sly Industrial Clean Air (Rees Blowpipe) MikroPul Torit Flex Kleen, Sub. Research Cottrell American Air Filter Home /Custom Made J.A. Kleissler Dravo Amerjet Parsons Air Purification Methods Due on Johnson-March Corp, Western Precipitation, Sub. Joy Mfg. Cincinnati Mine Safety Appliance Buffalo Forge Northern Blower Tenner & Hans Aget Trybourns Walch Dynavane Farr Wm. J. Schmitt Porter Kurt & Bloom Mfg. Co. Hoffman Air & Filtration Ruemelin Carter Day Div. , Hart -Carter Co. Lorbrow John Wood Tek Air Wiedenmann Tongeren Fuller Total Plants Using Product No. 26 17 13 9 6 4 3 3 3 3 3 3 2 2 2 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 124 Percent 21.2 13.9 10.6 7.4 4.9 3.3 2.3 2.3 2.3 2.3 2.3 2.3 1.6 1.6 1.6 1.6 0.8 0.8 0.8 0.8 0.8 0.8 0.8 0.8 0.8 0.8 0.8 0.8 0.8 0.8 0.8 0.8 0.8 0.8 0.8 0.8 0.8 0.8 0.8 99.9 Baghouses in Operation No. 95 48 50 17 17 12 10 8 6 6 5 5 8 4 4 3 5 5 4 4 4 5 3 2 2 2 2 2 1 1 1 1 1 1 1 1 1 1 1 349 ^Percent 27.2 13.8 14.2 4.9 4.9 3.4 2.9 2.3 1.7 1.7 1.4 1.4 2.3 1.1 1.1 0.9 1.4 1.4 1.1 1.1 1.1 1.4 0.9 0.6 0.6 0.6 0.6 0.6 . 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 99.9 35 ------- Table 11. CAPACITY OF DUST COLLECTION SYSTEM Capacity m /min < 142 -T- 143- 283 284- 566 567- 849 850-1,416 1,417-2,832 2,833-5,663 > 5,663 cfm < 5,000 5,001- 10,001- 20,001- 30,001- 10,000 20,000 30,000 50,000 50,001-100,000 100,001-200,000 >200,000 Total Plants Having Stated Total Capacity No. 1 1 4 2 8 5 21 Percent 4.8 4.8 19.0 9.5 38.2 23.7 100.0 Baghouses in Plants Having Stated Total Capacity No. 19 27 31 4 2 4 1 88 Percent 21.6 30.7 35.3 4.5 2.3 4.5 1.1 100.0 Plants Having Baghouses Within Stated Range of Capacity No. 5 9 8 3 1 1 1 28 Percent 17.8 32.2 28.5 10.7 3.6 3.6 3.6 100.0 ------- Table 12. AIR-TO-CLOTH RATIO Air-to-Cloth Ratio m/min <. 0,62:1 0.63-0.75:1 0.76-0.91:1 0.92-1.24:1 1.25-3.10:1 ft/min < 2.0:1 2.1- 2.5:1 2.6- 3.0:1 3.1- 4.0:1 4.1-10.0:1 Total Plants Having Ratio No. 3 3 6 2 7 21 Percent 14,3 14.3 28.6 9.5 33.3 100.0 Baghouses Having Ratio Kfd. 22 22 23 9 34 110 Percent 20.0 20.0 20.9 8.2 30.9 100.0 37 ------- The fabric used (see Table 13) in the majority (72.2%) of baghouses in the asbestos industry is cotton. The auto- matic shaker type of cleaning mechanism is used by 59-0% of the plants and in 63.3% of the baghouses. The other less popular cleaning mechanisms used are listed in Table 14. The cleaning cycle used (see Table 15) is less than 15 min., from % to 1% hr, and from 1% to 4 hr in 25.0, 29.5, and 25% of the plants and 19.8, 31.2, and 30.2% of the baghouses, respectively. The pressure drop across the baghouses is 7.6 cm (3 in.) of water or less in 50% of the plants and 57.8% of the baghouses. The breakdown of pressure drop is given in Table 16. Seventy-two plants responded to the question on baghouse operating problems, and 63 said they had none. Shaker mechanisms and breaking bags were the major problem, each being reported by 4.2%, of the plants responding. Overloading, freezing bags, and steam were each reported by one plant (1.4%). COST OF EQUIPMENT The cost of installing or operating a baghouse is difficult to ascertain, as most companies do not give out such information or do not have it readily available. Manufacturers usually bid on a contract to supply control equipment. They claim that each unit is different with few standard parts. Therefore, they fabricate individual units to suit the particular plant. This explains to some extent their reluctance to quote a figure. In general, from user's data, the cost of purchasing a baghouse is in the range of 7c-10.5/lpm ($2.50-3.00/cfm). However, costs were reported as low as 3.5c/lpm ($1.00/cfm), and as high as 60c/lpm ($17.00/cfm). Total annual operating costs, which includes the costs for power and maintenance, ranged from 0.175C-3.5c/lpm/yr ($0.05-1.00/cfm/yr) with no trends within that range clear from the data available. 38 ------- Table 13. BAG FABRIC 10 Fabric Cotton Dacron Polyester Canvas Wool Nylon Orion Polyprolene felt Polyphrone felt Burlap Total Plant Using Fabric No. 36 8 5 2 2 1 1 1 1 1 58 Percent 62.3 13.8 8.6 3.4 3.4 1.7 1.7 1.7 1.7 1.7 100.0 Baghouses Using Fabric No. 164 31 15 4 2 4 3 3 1 227 Percent 72.2 13.7 6.6 1.8 0.9 1.8 1.3 1.3 0.4 100.0 Bag Cleaning Mechanism Used With Type of Fabric, No. (%) of Baghouses Hand Shaker 27 (16.4) Automatic Shaker 125 (76.8) 23 (74.2) _ __ 4 (100.0) 2 (100.0) 4 (100.0) 3 (100.0) __ — Reverse Jet 10 (6.7) 3 (9.7) 5 (33.3) • ~ _ ___ •_• — _-._ 1 (100.0) Pulse Jet 2 (1.2) 5 (16.1) 10 (66.7) _ _ _ ___ _ __ 3 (100.0) ------- Table 14. BAG CLEANING MECHANISM Cleaning Mechanism Automatic shaker Pulse jet Reverse jet Hand shaker Total Plants Using Mechanism No. 39 10 9 8 66 Percent 59.0 15.2 13.6 12.2 100.0 Baghouses Using Mechanism No. 160 28 33 32 253 Percent 63.3 11.0 13.1 12.6 100.0 40 ------- Table 15. BAG CLEANING CYCLE Cleaning Cycle At <_ 5 min 5 min < At £ 15 min 15 min < At <_ 45 min 45 min < At <_ 1 hr 1 hr < At 5 1% hr 1% hr < At <_ 4 hr 4 hr < At 5 1 day Intermit tent Total Plants Using Cycle No. 3 3 2 3 2 6 3 2 24 Percent 12.50 12.50 8.25 12.50 8.25 25.00 12.50 8.25 100.00 Baghouses Using Cycle No. 10 9 8 12 10 29 16 2 96 Percent 10.4 9.4 8.3 12.5 10.4 30.2 16.7 2.1 100.0 Bag Cleaning Mechanism Used No. (%) kand Shaker 8 (80.0) -- -- -- -- -- 12 (75.0) 2 (100.0) Automatic Shaker 2 (20.0) 2 (22.2) 6 (75.0) 12 (100.0) 10 (100.0) 29 (100.0) 4 (25.0) — Reverse Jet — — — Pulse Jet 7 (77.8) 2 (25.0) — — At = time between cleaning cycles ------- Table 16. PRESSURE DROP ACROSS BAGS Pressure Drop cm (in.) H20 Ap <_ 2. 2.54 (1) < Ap £ 5. 5.08 (2) < Ap £ 7. 54 (1) 08 (2) 62 (3) 7.62 (3) < Ap £ 10.2 (4) Total Plants Having Ap No. 2 3 5 10 20 Percent 10.0 15.0 25.0 50.0 100.0 Baghouses With Ap No. 2 40 17 43 102 Percent 2.0 39.2 16.6 42.2 100.0 Bag Cleaning Mechanism Used No. (%) Hand Shaker — 7 (17.5) — 7 (16.3) Automatic Shaker 2 (100.0) 16 (40.0) 11 (64.7) 26 (60.4) Reverse Jet 10 (25.0) 4 (23.5) 4 (9.3) Pulse Jet 7 (17.5) 2 (11.8) 6 (14.0) ------- SECTION 6 EMISSION DATA COLLECTION AND ANALYSIS INTRODUCTION Emissions data was collected from baghouses at five locations. They were chosen because of the nature of the asbestos processing undertaken. They include: two asbestos refining mills dealing with natural asbestos, two asbestos cement product plants where the asbestos was bound into the product, and an asbestos textile plant where the product is loosely bound. In all of the locations, chrysotile asbestos was used exclusively. Membrane filter samples were taken both upstream and downstream of the baghouse. Two types of samples were taken. Samples were taken over a four hour period to obtain .mass efficiency data. To obtain size efficiency data, samples were taken for short time periods of approximately 5 minutes on the inlet side, and 30 minutes on the outlet side. These filter samples were then examined by optical and electron microscope, and the number and size distribution of the fibers was noted. The purpose of the emission testing was to provide preliminary factual data on the extent of the emissions from baghouses. It was to provide a quick answer on the question of whether a study on improving the efficiency of baghouses was warranted, and to provide base data from which the ambient air exposure levels in the vicinity of the plant could be calculated. 43 ------- GENERAL SAMPLING PROCEDURES The sampling scheme used for this study followed the EPA method as detailed in the Federal Register 3£-247. The general outline is given here, specific details of the procedures at each site will be given separately in the follwoing text. Sampling points in the ductwork were selected in re- gions where the most stable flow patterns existed. The ports were located, where possible, eight to ten diameters down- stream and three to five diameters upstream from any bends, elbows, junctions, or other constrictions in the stack or duct . The velocity profile in the stack or duct was measured by means of an S-type pitot tube. The pitot tube was tra- versed across the stack or duct, the gas velocity was deter- mined at the center of each equal area zone. The average velocity was then determined by averaging the velocities in all of the zones from: (V ) _ v_ = s N where V = Average velocity S V. = Velocity at one point N = Number of points Two standard EPA Method 5 isokinetic sampling systems (as shown in Figure 1) were used to collect the air samples from the upstream and downstream ductworks of the baghouse. The sampling probes and nozzles were fabricated from stain- less steel. The probes were 1.27 cm (0.5 in.) I.D. and 107 cm (42 in.) long, and the nozzles were 0.63 cm (0.25 in.) I.D. 44 ------- Filter Holder Probe Reverse - Type Pitot Tube Figure 1. Sampling arrangement A5 ------- On the upstream side of the baghouse, samples were drawn through a cyclone followed by a 10.2 cm (4 in.) mem- brane filter (Millipore) of 0.8 ym pore size. On the down- stream side, no cyclone was used; the air stream was lead directly to the membrane filter. After sampling, the filters were placed into marked plastic folders. Material adhering to the inside of the probe and tube was washed into the cyclone sample collector using acetone. The collector was then marked and sealed. In some instances, it was not possible to collect samples by the isokinetic stack sampling method. This was because there was no suitable length of ducting or stack from which valid samples could be collected. Here, a high volume sampler was used with a 20.3 x 25.4 cm (8 x 10 in.) membrane filter. Some error would be introduced by using this method. However, because of the small size of the par- ticles passing through the baghouse, the error would be small. When this method was used, the sampler was located inside the baghouse, close to the exiting point. SPECIFIC SITE INFORMATION Johns-Manvilie, Asbestos Cement Pipe Plant, Waukegan, Illinois Plant Details - Dust Source: Lathes for cutting and trimming asbestos cement pipes containing asbestos, cement, and sand. Baghouse: Parsons unit, installed December, 1972 Four compartments 200 bags per compartment, total 800 bags Capacity: 283 nH/min (10,000 cfm) per compartment 1,132 m3/min (40,000 cfm) total Operating temperature: 20.5 C (69 F) Pressure drop: 15.2 cm (6 in.) H20 Air-to-cloth ratio; 2:1 46 ------- Bags : Cotton sateen 12.7 cm (5 in.) diameter 3.05 in. (10 ft) length ~ « Total area per bag: 1.22 m (13.1 ft ) Permeability: 4.57 + 1.52 m/min at 1.27 cm HoO (15 +~5 cfm/min at 0.5 in. H-OJ Bag life: 2-3 years' Baghouse Cleaning: Mechanical shake Each compartment is shaken for 2 minutes every 30 minutes A compartment is shut off while shaken; there are always 3 com- partments on stream. Collected waste is taken to dump. Baghouse Effluent: vented to the exterior through a short stack about 3 m (10 ft) tall and a total of 10 m (30 ft) above the ground Procedure - Upstream samples were collected from a 86 cm (34 in.) circular duct approximately 5.5 m (18 ft) from the floor. A Platform was erected to reach the duct at a point more than eight diameters from any obstruction or bend in the ductwork. •^0 sampling ports, one entering the duct vertically, the °ther horizontally, were cut into the duct; each was aPproximately 5 by 7 cm (2 by 3 in.). Linear gas velocities in the duct were measured using ^ S-type pitot tube. The duct was traversed both vertically atld horizontally. The pressure drop across the- manometer and the stack temperature were measured at distances of 10, 25, 41, 56, 71, and 87 cm (4, 10, 16, 22, 28, and 34 in.) inside the duct. The average duct velocity calculated from the vertical traverse was found to be 1,021 m/min (3,354 fpm) and from the horizontal traverse 1,032 m/min (3,386 fpm). Dust samples were collected isokinetically on Millipore Membrane filters of 0.8 ym pore size. To obtain fiber size 47 ------- distribution data, samples were collected with particle loadings such that the determination could be made by direct observation of the filter by either optical or electron microscopy. The time required to obtain ideal loading, that is, with the particles on the filter evenly distributed without touching, was not known. For this reason, samples were collected for times of 0.25, 15, and 30 minutes. The samples were removed isokinetically from a single sampling point 76 cm (30 in.) vertically inside the duct. At this point, the linear velocity was almost identical to the average velocity. Samples for mass efficiency measurement were collected at each of the traverse points for a period of 5 minutes. The entire sampling sequence was distributed over a 3% hour time period. Downstream samples were collected from a 81 by 107 cm (32 by 42 in.) rectangular stack. A platform was constructed inside the building about 8 m (25 ft) from the floor and 1.7 m (5 ft) from the roof. Three equally spaced sampling ports were cut into the 81 cm (32 in.) face of the duct at a dis- tance of 2.5 m (8 ft) from the baghouse fan, which was located in the stack between the baghouse and the louvred exit on the roof. This sampling point was unsatisfactory because of its close proximity to the fan, but it was the only practical point to erect a platform within the constraints of the plant layout. Linear gas velocities were measured by traversing the duct through each of the ports at depths of 10, 25, 41, 56, 71, 87, and 102 cm (4, 10, 16, 22, 28, 34, and 40 in.). The three ports gave average velocities of 1,083 m/min (3,555 fpm) , 1,113 m/min (3,654 fpm) and 1,175 m/min (3,856 fpm), respectively. 48 ------- Samples for microscopic examination of the fiber size dis- tribution were taken from the central port at a depth of 33 cm (3-3 in.). The samples were collected for time periods of 5> 15, and 30 minutes. To obtain mass efficiency data, samples were collected from each port at each traverse point for a period of 10 minutes. The upstream and downstream sampling was Performed within the same 3% hour total sampling time Period. E^ybestos. Asbestos Textile Plant. Marshville, North Carolina Slant Details - Dust Source: Carding machine which combs raw asbestos fiber into a roughly parallel array prior to spinning. Baghouse; Wheelabrator model 112, installed in 1953 Single compartment, 304 bags Capacity: 473 m3/min (16,700 cfm) Operating temperature: 26.7° C (80° F) Pressure drop: 5.59 cm (2.2 in.) 1^0 Air-to-cloth ratio: 3:1 Bags: Cotton sateen 20.3 cm (8 in.) diameter 2.82 m (9.25 ft) length 2 2 Total area per bag: 1.80 nT (19.4 ft ) Permeability: 4.57 + 1.52 m/min at 1.27 cm H£0 (15.0 + 5.0 cfm/min at 0.5 in. H20) Bag life: 5 years Baghouse Cleaning; Mechanical shake At the finish of each shift, the cake from the bags is shaken into an empty hopper - this material is put in polythene bags and dumped at a city site. Three times during each shift the material collecting in the bag- house hopper (without shaking) is collected and recycled. 49 ------- Baghouse Effluent: Effluent is either recirculated to conserve heat, or vented to the outside air, depending on ambient temperature conditions. A very short (1 m) stack is used to vent the effluent to the outside air. Sampling Procedures - Upstream samples were collected from a 76 cm (30 in.) diameter duct. The only convenient sampling point was about 3 duct diameters from the baghouse and 5 duct diameters from a bend. Because of the very thick duct walls and the inaccessibility of the duct, only one sampling port, which allowed a horizontal traverse to be made, was used. Linear gas velocities were measured at distances into the duct of: 2.5, 5.1, 7.6, 11.0, 15.0, 20.0, 29.0, 48.0, 56.0, 61.0, 69, 71, and 74 cm (1, 2, 3, 4,4, 6, 8, 11, 19, 22, 24, 25, 27, 28, and 29 in.). The average gas velocity was found to be 956 m/min (2,915 fpm). Dust samples were collected isokinetically on Millipore membrane filters of 0.8 ym pore size. Samples for the deter- mination of fiber size distribution were taken from a point 30.5 cm (12 in.) into the duct. Three sampling periods of 0.25, 1, and 5 minutes were utilized to obtain three filters with different loading levels. The filter with the most even loading was to be used for the fiber count. To deter- mine the mass loading, samples were removed isokinetically for five minutes at each traverse point. Two traverses were made over a total time period of about 6 hours. Downstream samples were taken from a 99 by 46 cm (39 by 18 in.) rectangular duct. A sampling port was cut into the 46 cm face of the duct well clear of any obstruction or bend. Only one port was cut because of the thick walls of the duct, which made it a difficult task. Linear gas velocities were measured at distances into the duct of 7.6, 50 ------- T-5, 23, 30, 38, 46, 53, 61, 69, 76, 84, and 91 cm (3, 6, 9, 12, 15, 18, 21, 24, 27, 30, 33, 36, and 39 in.). The average velocity was found to be 676 m/min (2,062 f pm) . Dust samples were collected isokinetically from the duct on 0.8 ym pore size Millipore membrane filters. Sam- Pies for fiber size distribution determination were collected at a point 61 cm (24 in.) into the duct for time periods of 5> 15, and 30 minutes. Samples for mass loading determination collected by traversing the duct and collecting a sample 10 minutes at each traverse point. The entire sampling Sequence was spread over a 6 hour time period. Both uPstream and downstream samples were collected over the same 6 hour time period. Johns -Manvi lie, Asbestos Cement Pipe Plant, Denison. Texas Details - Dust Source: Lathes for cutting and trimming asbestos cement pipes containing asbestos, cement, and sand. Baghouse: Wheelabrator Model 2b Single compartment, 640 bags Capacity: 736 m3/min (26,000 cfm) Operating temperature: 20.5° C (69° F) Pressure drop: 14.5 cm (5.7 in.) H90 Air-to-cloth ratio: 3:1 z Bags ; Cotton sateen 12.7 cm (5 in.) diameter 3.18 m (10.4 ft) long 2 9 Total area per bag: 1.27 m (13.6 ft ) Permeability: 4.57 + 1.52 m/min at 1.27 cm H90 (15.0 + 5.0 cfm/min at 0.5 in/H^O) Bag life: 1 year Baghouse Cleaning: Mechanical shake, manually started at shift breaks and changes. Waste is taken to plant dump. 51 ------- Sampling Procedure - Upstream samples were collected from a 80 cm (32 in.) diameter duct. A sampling location greater than eight diameters from any bend or flow disturbance was selected. A platform approximately 5 m (15 ft) high was erected to reach the two sampling ports which were cut into the duct to allow both vertical and horizontal duct traverses. Linear gas velocities were measured at distances into the duct of 1.7, 5.3, 9.7, 14.5, 20, 29, 43, 53, 67, 72, 76, and 80 cm (0.67, 2.1, 3.8, 5.7, 8.0, 11.4, 16.9, 20.8, 26.3, 28.2, 29.9, and 31.3 in.) in both a vertical and horizontal direction. The average gas velocities were found to be 1,510 m/min (4,605 fpm) from the vertical traverse and 1,539 m/min (4,694 fpm) from the horizontal traverse. Dust samples were collected isokinetically on Millipore membrane filters of 0.8 ym pore size. Samples for the determination of fiber size distribution were taken from a point 46 cm (18 in.) inside the duct vertically from the lower port. Samples were collected for 0.5, 1, and 5 minutes to obtain samples with different loading levels. To obtain information for the determination of the mass loading, samples were taken at each traverse point in both the verti- cal and horizontal ducts for a period of 5 minutes. Samples were collected over a total time period of 5 hr 7 min. Downstream samples were collected using a high volume sampler with an 0.8 ym pore size Millipore membrane filter, located within the baghouse close to the exit. The sampler was mounted on clamps such that the membrane face was presented face-on to the direction of the air flow. A high volume sampler was used because the very short stack between the fan and the louvred roof exhaust exit (about 1 m) presented no acceptable sampling point. Samples were collect6 for fiber size distribution determination for periods of 52 ------- 5, 15, and 30 minutes to obtain samples with different loading levels. Samples for mass loading determination were collected for 5 hr 7 min. Both the upstream and downstream samples were collected over the same 5 hr 7 min time period. Johns-Manville, Asbestos Ore Mill. Asbestos. Canada Plant Details - Dust Source; Fiber screening and air aspiration system. Baghouse: Wheelabrator, special design One compartment, 79,200 bags Capacity: 127,000 m3/min (4,500,000 cfm) Operating temperature? 21.1° C (70° F) Pressure drop: 7.62 cm (3 in.) 1^0 Air-to-cloth ratio: 3:1 Bags; Cotton sateen 12,7 cm (5.in.) diameter 4.27 m (14 ft) length 2 2 Total area per bag: 1.70 m (18.3 ft ) Permeability: 7.62 m/min at 1.27 cm H20 (25 cfm/min at 0.5 in. H20) Bag life: 6-10 years Baghouse Cleaning; Mechanical shake 2 minute shake cycle every 30 minutes Bags are shaken in groups of 7,200 bags Collected dust is recycled Cyclone: Cyclone is used as a pre-cleaner Unit is 2.44 m (8 ft) in diameter Capacity: 424 m^/min (15,000 cfm) Inlet velocity: 1,219 m/min (4,000 fpm) Exit velocity: 914 m/min (3,000 fpm) Claimed mass efficiency is 98.5% for 30-500 ym dust at 251 m3/min (9,100 cfm) Baghouse Effluent; recirculated to plant for most of the year, vented outside during warm weather. Sampling Procedure - Upatream samples were collected from a 38 cm (15 in.) duct leading to the cyclone* It would have been preferable to sample in the duct separating the cyclone and baghouse, 53 ------- but it was inaccessible and had a very short (less than 1 m) connecting duct. Two identical ducts lead to the cyclone from the plant. The ducts brought air from identical opera- tions in the plant and should have identical dust loadings. Therefore, only one duct was sampled. The dust loading within the duct was very high and quickly blocked the sampling probe and pitot tube. The average duct velocity was estimated to be about 665 m/min (2,618 ft/min). Single point samples were taken from a point 30 cm (12 in.) into the duct. Samples were collected under isokinetic conditions until signs of blocking occurred (as observed by sudden fluctuations on the manometer) Four samples of 1.4, 1.7, 2.8, and 3.5 liters were collected. Downstream samples were collected on a high volume sampler located within the baghouse between the sets of bags. The baghouse forms the complete top floor of the 14 story building and air is generally recirculated from the baghouse area to the plant by means of blowers. In exceptionally warm weather, the air is vented directly out through louvres in the side of the baghouse. There was no stack. Samples for fiber size distribution determination were collected for 5 and 30 minutes to obtain a loading range. Samples for mass loading determination were collected for 5 hr. GAF, Asbestos Ore Mill, Eden Mills, Vermont Plant Details - Dust Source: Fiber screening and air aspiration system. Baghouse; Wheelabrator, Model B-607, Series VIII -8 compartments 336 bags per compartment, 2,688 bags total Capacity: 8,490 m3/min (300,000 cfm) Operating temperature: 21.1° C (70° F) Pressure drop: 10.2 cm (4 in.) Air-to-cloth ratio: 3.2:1 54 ------- Bags: Cotton sateen 20.3 cm (8 in.) diameter 5.3 m (17.5 ft) length 2 2 Total area per bag: 3.22 m (35 ft ) Permeability: 4.88-6.10 m/min at 1.27 cm HoO (16-20 cfm/min at 0.5 in. H2&) Bag life: 5 years Baghouse Cleaning: Mechanical shake 2 minute shake cycle every 30 minutes , one compartment at a time Dust is recycled. Baghouse Effluent: Baghouse is not enclosed; therefore, air always diffuses back into the plant. Cyclone; Cyclone is used as a pre-cleaner. Unit is 3m (9.8 ft) in diameter. Procedure - Upstream samples were collected from a 76 cm (30 in.) duct leading from the cyclone to the baghouse. The duct Was short (about 3 m overall) and the only accessible section was about 1 m from the cyclone. Two sampling ports, at right angles to each other, were cut into the angled duct. Linear gas velocities were measured at distances into the duct of 2.5, 5.0, 10, 15, 20, 25, 30, 36, 41, 46, 51, 56, 51, 66, 71, and 76 cm (1, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, and 30 in.) from both sampling ports. The average gas velocity was found to be 1,330 m/min (4,364 fpm). Dust samples were collected isokinetically on Millipore filters of 0.8 urn pore size. Samples for the deter - of fiber size distribution were collected from a Point 30 cm (12 in.) into the duct for time periods of ^•» 5 and 30 seconds to obtain samples at different loading Bevels. A sample for the determination of mass loading was °btained by sampling at each traverse point for 5 minutes over a total time period of 6 hr. 55 ------- Downstream samples were collected using a high volume sampler located within the central part of the baghouse. The baghouse had no stack air being allowed to diffuse into the plant from the open baghouse. Samples for the determination of fiber size distribution were: collected for time periods of 10, 30, and 60 minutes to obtain samples with differing loading characteristics. The mass loading was determined by collecting a single sam- ple for a time period of 6 hr. Both upstream and downstream samples were collected over the same time period. SAMPLE ANALYSIS Introduction The mass efficiency and the efficiency in terms of fiber size removal were measured at each of the five sites which were sampled. The mass efficiency was found by weighing the quantity of material collected for a given volume of air, before and after the baghouse. Fractional efficiency was determined by comparing the size frequency of fibers be- fore and after the baghouse. In determining the mass efficiency of the baghouse, it should be noted that the measured efficiency includes non- asbestos as well as asbestos material. No attempt was made to analyze the filters to determine the asbestos content before and after the filter to see if there is a change in the composition. The fiber counts were made with the criteria that all particles having a greater than 3:1 aspect ratio were fibers. The assumption was made that all fibrous particles coming from an asbestos plant were asbestos. Spot checks on fibers were made, and the selected area diffraction pattern was found to be characteristic of chrysotile asbestos. Thus, the assumption was deemed valid. 56 ------- Mass Efficiency The mass efficiency was found by accurately weighing the samples collected from before and after the baghouse. The filters were weighed on electronic balances at a sensitivity °f + 0.1 mg. On the upstream side, the dust deposited in the probe, the sampling train, and the cyclone was washed with water into a weighed sampling bottle. It was then dried in vacuum oven at 110° C, cooled, and reweighed. From these weights, the mass efficiency was calculated. Fiber Counting Optical microscope analysis was performed using the Method described in the NIOSH criteria document on asbestos (HSM 72-10267). A portion of the membrane filter was mounted °n a slide. Using a 1:1 solution of dimethyl phthalate and diethyl oxalate, the filter was allowed to clear. A °lean cover slip was placed on top of the sample and the fiber concentration determined. The light microscope used was equipped with phase-contrast and polarized light. The objective lens of 4mm resulted in a total magnification °f 500X. From randomly chosen fields, the number of fields f°r a total count of 100 fibers was noted (with a minimum °f 20 fields observed), or, 100 fields were observed when the distribution was sparse. For the electron microscope analysis, a circle of the sample filter 3.5 mm in diameter was cut. This piece of filter was placed on top of a carbon-coated 100 mesh elec- tron microscope grid. The grid with the filter on it was Placed in a condensation washer using acetone as a solvent. The filter medium was dissolved away by the acetone, Depositing the fibers of the sample undisturbed on the car- bon substrate of the grid. The specimen was then counted on a Hitachi HU-11 transmission electron microscope at the Magnification of 16,364X. 57 ------- The optical microscope analysis enumerated the fibers greater than 5 ym in length and a minimum diameter of 0.5 ym. The electron microscope analysis counted fibers down to 0.06 ym in length and 0.020 ym in diameter. The NIOSH document on asbestos cited the above states that the fiber counts follow a Poisson distribution. The standard deviation for a count of 100 fields is /YOU, or 10 fibers or + 10%. To keep the statistical error at the 957o confidence level, approximately two standard deviations must be considered. The uncertainty in the counts is then + 207=. Calculation of Fiber Numbers To relate the number of fibers counted to the asbestos concentration in the air, the following equation was used: no. of fibers _ 3 m of air no. of fibers counted no. of fields effective filter area, area of microscope's field of view, cm 2 volume of air sampled, m" Effective filter area = 81.7 cm^ for 4 in. filter Effective filter area =63.2 cm2 for 4 in. hi-vol filter Effective filter area -1 Area of field of view = Area of field of view = 426.4 cm for 8 in. x 10 in. hi-vol filter / O 6.514 x 10~ cm for optical micro- scope (500X) -7 2 1.344 x 10 cm for electron micro- scope (16.364X) 58 ------- Efficiency Calculation To calculate the efficiencies, both by mass and by number, the following relationships were used: Hass Efficiency (%) = 100 , mass in a given volume of air after the baghouse mass in the same volume of air before the baghouse Number Efficiency (%) = no. of fibers per cubic meter after the baghouse 100 1 - no. of fibers per cubic meter berore the baghouse RESULTS OF THE ASBESTOS SAMPLING STUDIES The results of the sampling study are summarized in the following tables. Table 17 gives the details of the samples and the sampling conditions at each of the five locations. Ift Table 18, the mass efficiencies at each of the five kaghouses are given. The total number of fibers, in both the optical and electron microscope size ranges, for the uPstream and downstream sides of the baghouses are given in ^able 19. The distribution of the fibers in the optical Microscope size range (1.5-30.0 ym) is presented in Table 20. Ir* Table 21, the size distribution of the fibers in the electron microscope size range (0.06-1.50 ym) is given. DISCUSSION OF THE RESULTS The mass efficiency of the baghouses tested was shown to be extremely high and in all instances was greater than . A high mass efficiency had been anticipated, and figures are in agreement with the very limited data is available in the literature as reported earlier in text. The information gained from this study on the number of asbestos fibers emitted from a baghouse exhibiting a high mass 59 ------- Table 17. SAMPLES AND SAMPLING CONDITIONS Sampling Site Waukegan, Illinois Marshville, North Carolina Denison, Texas Asbestos, Canada Eden Hills, Vermont Sampling Location Upstream Upstream Upstream Downstream Downstream Downstream Upstream Downstream Upstream Upstream Upstream Downstream Downstream Downstream Upstream Downstream Upstream Hor. Upstream Vert. Downstream Downstream Downstream Downstream Ups tream Upstream Ups tream Upstream Downstream Downstream Downstream Upstream Ups tream Upstream Upstream Downstream Downstream Downstream Downstream Average Linear Velocity, m/min 1,026 1,026 1,026 1,098 1,098 1,098 1,026 1,098 3,410 3,410 3,410 3,400 3,400 3,400 3,410 3,400 4,560 4,560 665 665 665 665 1,330 1,330 1,330 1,330 i Sampling Time 4 min 25 sec 1 minute 15 seconds 5 minutes 15 minutes 27 minutes 1 hr (over 3^ hr) 3*5 hours 5 minutes 1 minute 15 seconds 5 minutes 15 minutes 30 minutes 2 hr 10 min f over the 1 4 hr 20 min ^arae 6 hr period] 60 minutes 60 minutes 16 minutes 5 minutes 30 minutes 5 hr 7 min = 1 sec (1.7 i) Z 1 sec (3.54 JO : 1 sec (1.4 fc) x I sec (2.83 i) 10 minutes 30 minutes 5 hours 1 second 5 seconds 30 seconds 1 hr 10 min 10 minutes 30 minutes 1 hour 6 hours Volume , Sampled (m ) 1.75 12.98 •I 2.83 x 10, 5.94 x 10, 4.42 x 10 2.76 6.404 4.18 4.18 4.30 1.40 8.49 195.00 1.7 x 10":? 3.54 x 10~, 1.4 x 10":; 2.83 x 10 0.57 1.70 101.8 4.25 x 10~4 3.08 4.10 11.90 23.80 143.00 Filter Size and Collection Method (all filters 0.8 urn membrane) , All 10 cm diameter. Isokinetic All 10 cm diameter. Isokinetic 10 cm-Isokinetic 20 cm x 25 cm Ki-Vol 10 cm-Isokinetic 10 cm~Hi-Vol 10 cm-Isokinetic 10 cm— Hi Vol Sampling Purpose Size Size Size Size Size Size Mass Mass Size Size Size Size Size Size Mass Mass Size & Mass Size & Mass Size Size Size Mass Size Size Size Mass Size Size Mass Size Size Size Mass Size Size Size Mass Filter No. 10 9 2 5 7 3 11 12 15 16 17 18 19 20 22 21 23 23 5 7 18 11 27 34 25 28 42 43 41 31 32 33 26 37 38 36 40 ------- Table 18. SAMPLE WEIGHTS AND MASS EFFICIENCY DATA Plant Location Waukegan , Illinois Marshville, North Carolina Denison, Texas Asbestos, Quebec Eden Mills. Vermont Sampling Location Upstream Downstream Upstream Downstream Upstream Downstream Upstream Downs treara Upstream Downstream Volume Sampled m3 (ft3) 1.755 (62) 12.984 (458.8) .2,765 (97.7) 6.404 (226.3) 4.163 (147.8) 196.0 (6,907.5) 1.40 (2) (0.05) 101.80 (3,600.) 3.085 (109) 143 (5,040) Filter Code Number (1) 11 12 . 22 21 23 11 (A) 28 41 (B) 26 40 (B) Sample Weight on Filter, R 0.2270 0.0000 0.0383 0.0002 0.8895 0.0000 0.0003 0.0103 1.6733 0.0084 Sample Weight in Cyclone and Probe, g 7.3822 0.0000 1.5511 0.0000 53.0758 0.1645 16.7.487 Total Sample Weight, £ 7.6092 0.0000 1.5894 0.0002 - 53.9653 0.0000 0 . 1648 0.0103 17.9220 0.0084 Dust Concentration R/m 4.336 0.000 0.580 0.000031 12.901 0.000 117.0 0.0001 5.804 0.000059 Mass Removal Efficiency 7. 100 with experimental limits >99.99 100 with experimental limits >99.99 >99.99 (1) All samples taken except: A) 25 x 20 cm Hi-Vol, B) 10 cm diameter Hi-Vol. (2) Heavy dust loading prevented sampling for more than a few seconds. ------- Table 19. TOTAL "FIBER COUNTS AND FIBER REMOVAL EFFICIENCIES Plant Location Waukegan, Illinois Marshville, North Carolina Denison, Texas Asbestos, Quebec Eden Mills, Vermont Sampling Location Upstream Downstream Upstream Downstream Downstream Upstream Downstream Upstream Downstream Upstream Downstream Volume of Air Sampled 00 1.75 13.0 2.83 x 10"3 4.42 x 10"1 5.94 x 10~2 4.18 8.49 1.69 x 10"3 1.70 4.25 x 10~4 23.8 Optical Microscope, 500X Total Fibers (No. ofj Fiber s/m ) > 1010 (1) 6.37 x 103 8.07 x 108 1.42 x 104 1.02 x 105 2.88 x 104 2.19 x 109 8.33 x 105 1.42 x 109 4.52 x 104 Efficiency CO > 99.99 > 99.99 97.18 99.96 > 99.99 Electron Microscope, 16,364X Total Fibers (No. ot Fibers/in ) > 1014 (1) 1.08 x 107 2.45 x 1011 3.28 x 109 5.04 x 109 3.20 x 107 1.38 x 107 1.24 x 1012 1.44 x 109 1.36 x 1013 1.29 x 108 Efficiency (%) > 99.99 98.69 97.94 57.90 99.88 > 99.99 Filter No. 11 12 17 20 18 23 18B 27 43B 31 36B (1) Samples too dense to count; sample diluted for size distribution. no' of fibers downstream) „,.,-. . Efficiency inni 100 11 uo. ol xibers upstream. ) 1 J ------- Table 20. OPTICAL MICROSCOPE (500X) SIZE DISTRIBUTIONS AND FRACTIONAL REMOVAL EFFICIENCIES Sampling Site Waukegan, Illinois Marshville, Jorth Carolina Denison, Texas Asbestos, Eden Mills, Sampling Location Jp stream 3ownstrean Upstream Downstream Upstream Downstream Upstream Downstream Upstream Downstream Total Fibers (No. of Fibers per m3) i> io10** 3. >. 37x10 J.07xl08 L.42xl04* L.02xl06 2.88x10 2.19xl09 C J. 33x10 L.42xl09 4 4.52x10 Size Distributions (by length) 1.5-10 Fiber Count (No. of Fibers per m3) >5.06xl09 3 2.70x10 5.46xl08 1.14xl04 7.32xl05 2.19xl04 1.31xl09 C 6.27x10 1.22xl09 4 3.42x10 % of Sample 50.6 42.4 67.6 80.0 71.8 76.2 59.7 75.3 86.8 75.7 Jm Efficiency m >99.99 >99.99 97.01 99.95 ____ >99.99. 10-20 um Fiber Count (No. of Fibers I % of Efficiency per m3)jSamplej (%) >2.10xl09 3, 2.31x10 1.78xl08 21.0 36.4 22.0 2.84xl03 20.0 1.92xl05 5.99xl03 6.61xl08 C 1.48x10 1.48xl08 q 8.14x10 18.8 20.8 30.2 17.8 10.4 18.0 >99.99 >99.99 96.88 99.98 >99.99 20-30 urn Fiber Count (No. of Fibers per m3) >1.67xl09 2 5.80x10 4. 84x1 O7 0.0 8.57xl04 5.76xl02 1.77xl08 4.00x10* 2.70xl07 2.03x10' % of Sample 16.7 9.1 6.0 0.0 7.4 2.0 8.1 4.8 1.9 4.5 Efficiency (%) >99.99 100 99.33 99.98 99.99 > 30 um Fiber Count (No. of Fibers per m^) >1.17xl09 n 7.71x10 3.55xl07 0.0 1.53xl04 2.88xl02 5.26xl07 "5 1.75x10^ 1.28xlO? j 8.13x10 ] i % of I Efficiency Sample^ (%) 11.7 i 1 12.1 4.4 0.0 1.5 1.0 2.4 2.1 0.9 1.8 >99.99 100 98.12 99.67 99.99 Filter No. 11 12 17 20 23 18B 27 43B 31 36B * Sample contained five fibers in 100 fields. ** Sample too dense to count; sample diluted for size distribution. Efficiency (Z) - 100 1 - no. of fibers downstream) no. of fibers upstream J ------- Table 21. ELECTRON MICROSCOPE (16.364X) SIZE DISTRIBUTION AND FRACTIONAL REMOVAL EFFICIENCIES Sampling Site Waukegan , Illinois Marshville, lorth Carolina Denison, Texas Asbestos f Quebec Eden Mills. Vermont Sampling Location Jpstream )owns tream Jpstream towns tream >owns trean Upstream Downstrean Jpstream Downstream Upstream Downstream Total (No of Fibers per m3) > 1014(1) l.OSxlO7 2.45X1011 }.04xl09 i.28x!09 J.20xl07 L.38xl07 1.24xl012 1 . 44xl09 1.36xl013 1.29xl08 - <0.06 ym Fiber Count (No of Fibers per m^) >8.7 xlO12 0.0 2.74xl010 1.28xl09 5.25xl08 0.0 0.0 0.0 3.48xl08 2.12xl012 1. 24x10 7 7. of Sample 8.7 0.0 11.2 25.3 16.0 0.0 0.0 0.0 24.2 15.6 9.6 Efficiency 100 95.33 98.08 (2) >99.99 0~T55-0.18 urn I Fiber Count (So. of Fibers per m^) >1.06xl013 0.0 l.lOxlO11 2.89xl09 1.88xl09 0.0 l.lOxlO6 2.36xl010 7.53xl08 6.17xl012 4.26xl07 % of Sample 10.6 0.0 44.9 57.3 57.3 0.0 8.0 1.9 52.3 45.4 33.0 Efficiency (7.) 0.0 97.37 98.29 (2) 96.81 >99.99 0.18-0.36 ym Fiber Count (No. of Fibers per mj) >1.25xl013 1.88x10 6.05xl010 6.90xl08 6.56x10 4.35xl06 1.66xl06 6. 94x10 10 1.38xl08 2.64xl012 1.86xl07 1 of Sample 12.5 17.4 24.7 13.7 20.0 13.6 12.0 5.6 9.6 19.4 14.4 Efficiency >99.99 99.89 98.92 61.80 99.80 >99.9« 0.36-0.54 Fiber Count (No. of Fibers per m3) >1.83xl013 1.40x10° 1.37xlOIU 1.06xl08 8.86x10 4.35xl06 3.86xl06 1.39X1011 1.73xl07 1.32xl01'' 7. of Sample 18.3 13.0 5.6 2.1 2.7 13.6 28.0 11.2 1.2 9.7 1.48xl07 11.5 Sfficiency (7.) >99.99 99.23 99.35 11.30 99.99 >99.99 [ 0.54-1.50 urn ' Fiber Count (No. of Fibers per m-*) >5.0 xlO11 7.52x10 3.31xl010 8.57xl07 7. of Sample 50.0 69.6 13.5 1.7 1.31xl08 2.33xl07 7.18x10 l.OlxlO13 1.83xl08 1.35xl012 4. 14x10 7 4.0 72.7 52.0 81.3 12.7 9.9 32.1 efficiency (7.) >99.99 99.74 99.60 69.20 >99.99 >99.99 Filter No. 11 12 17 20 18 23 18B 27 43B 31 36B (1) Samples too dense to count; total fiber count > 10 fibers/m . (2) Negative efficiency indicated within experimental limits. _„. . ._, 1nrif, no. of fibers downstream] Efficiency «) - 100[1 - no. of fibers upstream J ------- efficiency is of importance. No other comparable data was found in the literature. In general terms, it can be seen from Table 19 that the number of fibers emitted which were greater than 1.5 ym in length was of the order of 10 fibers Per cubic meter at the five locations. At the same time, the number of fibers emitted which were in the very small size group of 0.06 to 1.5 ym in length was of the order °f 10 fibers per cubic meter. Two reasons could be given to explain the above result. Either the collection efficiency for the very small particles could be very low or there could be very many more fibers challenging the fabric filter at the same collection efficiency. From Table 19, it is seen that, in general terms, the collection efficiency for fibers greater than 1.5 ym was in excess of 99.99%, while the collection efficiency for the fibers less than 1.5 ym in length was greater than 98%. (The one low result, 57.9%, for the collection efficiency at the Denison plant, could be in error, although it is noted that it also had the lowest efficiency in the greater than 1.5 ym size group at 97.18%.) Obviously.then, the reason for the very large number of fibers emitted in the 0.6 to 1.5 ym size range is the enormously numbers of fibers present in this range and not from significant change in the collection efficiency with Particle size. The form of the fractional efficiency curves for fabric filters have been studied by a number of workers21*'25. Peterson and Whitby26 reported that the fractional efficiency goes to a minimum at a particle size of about 0.3 ym, and increases in efficiency for particle sizes above and below this point. Their conclusions were based on the fractional efficiency measurements made using a monodisperse aerosol to challenge the filter. Colley27 has found that while the efficiency is at a minimum with a particle size of 0.1-0.2 ym, 65 ------- there is no evidence to show that the efficiency actually increases with decreasing particle size beyond this size. The explanation for the observed flattening of the curve is that larger particles have a comparatively high inertia and the path of the particles is not altered by Brownian movement. As the particles decrease in size, Brownian movement becomes more effective and will cause the particles to alter their path such that they are captured by the fabric. The present results show a pronounced flattening of the fractional efficiency curve, indicating that the smaller particles, which might be expected to penetrate the filter, are being captured by Brownian movement. For asbes- tos, which has a fibrous form, there will be a more compli- cated aerodynamic approach theory than for simple spheres. The fibrous form could aid capture efficiency if the par- ticles approach the fabric length-ways on. Alternatively, the capture efficiency could be reduced if the fibers line up in the direction of flow and present a small cross-section for capture. A highly significant point resulting from this study is the fact that, in very many instances, air is recirculated within an asbestos plant. Plant operators have considered this to be reasonable if the return air can be shown to contain asbestos at a level lower than the OSHA standard of less than 2 fibers per cubic centimeter, where a fiber is only counted if it is greater than 5 cm in length. It has been shown here that while the number of fibers greater than 5ym is low, there are very large numbers of fibers less than 5 ym. However, plant operators who recircu- late plant air are inadvertently exposing their workers to very high levels of very small fibers whose health significance has not been established. 66 ------- SECTION 7 ESTIMATE OF ASBESTOS DISPERSION INTRODUCTION Two methods were used to estimate the dispersion of the asbestos emissions from the source. Both were based on the well-known equations of Pasquill28 as restated by Gifford29 The first method involved the manual calculation of the Dispersion using the Binormal Continuous Plume Dispersion Model as detailed by Turner30 . This technique enables Predictions of the ambient pollutant concentration downwind from a known source under given metoerological conditions. The second method utilized the Climatological Dispersion Model (CDM). This model predicts the long-term (seasonal °r annual) quasi-stable pollutant concentrations in an area surrounding the source. The long-term wind rose data, obtainable from the "National Climactic Center, is required f°r this model, and, because of the very large number of computations, it is only suitable for use by computer. using the CDM were found to be inconsistent due to inherent limitations of the model. For this reason, those results obtained with the first model are included wi-th this report. THE BINORMAL CONTINUOUS PLUME DISPERSION MODEL The system considered in this model is shown in Figure 2. *he x-axis extends horizontally in the direction of the mean ^ind. The y-axis is in the horizontal plane perpendicular t° the x-axis, and the z-axis extends vertically. The plume 67 ------- travels along a parallel to the x-axis from point H, which represents the sum of the physical stack height, h, and the plume rise, AH. Origin (x,-y,Z) (x,-y,o) Figure 2. Coordinate system showing Gaussian distributions in the horizontal and vertical The concentration, X, of the aerosolized particles (i.e., less than about 20 ym diameter) at x,y,z from a continuous source with an effective emission height, H, is given by equation (1). X(x,y,z;H) = 2TTC U y z 1 2 (i) exp 1 I Z - H + •• exp Z + H 68 ------- Any consistent set of units may be used, for the present study: 3 X = concentration (fibers/m ) Q = emission rate (fibers/sec) u = mean wind speed (m/sec) H,x,y,z = effective stack height and coordinate (m) a , a = standard deviations of plume coordinate (m) In establishing this equation, a number of assumptions are made: The plume spread has a Gaussian distribution in both the horizontal and vertical planes. The wind is constant in speed and direction. The emission rate is uniform. Total reflection of the plume takes place at the earth's surface, i.e., there is no deposition or reaction at the surface. There is no diffusion in the direction of the plume travel. F°r concentrations calculated at ground level, i.e., z = 0, and downwind along the centerline of the plume, i.e., y = 0, equation simplifies to H 2 X(x,0,0;H) = (2) In most real circunstances, a stable layer exists above the unstable lower layer with the effect of restricting the vertical diffusion. The model can be modified to account for this situation by considering the height of the base of the stable layer from the ground level to be L. At a height above the plume centerline, the concentration r 69 ------- is one tenth of the concentration at the centerline at the same distance along x. When one tenth of the plume center- line concentration reaches the stable layer, height L, it is assumed that the distribution is affected by the stable layer. At this point, which is called x, , 2.15cr = L, or a = 0.47L. z At distance, x,. , the plume is assumed to have a Gaussian distribution in the vertical. It is further assumed that when the plume travels twice this distance, to 2x, , the plume has become uniformly distributed between the earth's surface and the stable layer at height L. Between these limits, the concentration does not change with height. At distances greater than 2x,, the downwind centerline con- centration is calculated from: X(x,0,z;H) = Q (3) /2rra LU y for any value of z from 0 to L. THE METHOD OF CALCULATION USING THE BINORMAL CONTINUOUS PLUME DISPERSION MODEL As input for the model, the following basic information is required. The mean wind speed, u, (m/sec) • The source emission rate, Q (fibers/sec) The effective source height, H (m) Stable layer height, L, for daytime and nighttime conditions (m) The stability class The values of mean wind speed, u, source emission rate, Q, and the effective source height, H, are measured values taken at a given site. The stable layer height, L, for 70 ------- daytime and nighttime conditions were assumed from the average conditions reported in the literature. Values of 800 m and 150 m were assumed for daytime and nighttime conditions, respectively. The stability class was taken from observa- tions of the meteorological conditions and reference to Table 22. Table 22. KEY TO STABILITY CATEGORIES Surface Wind Speed at 10 ra (m/sec) < 2 2-3 3-5 5-6 > 6 Day Incoming Solar Radiation Strong A A-B B C C Moderate A-B B B-C C-D D Slight B C C D D Night Thinly Overcast or > 4/8 Low Cloud E D D D >3/8 Cloud F E D D Note: The neutral class D should be assumed for overcast conditions during day or night. Ambient air concentrations downwind from the source were calculated using equation (2) for distances of x less than XT and from equation (3) for distances of x greater than 2x, . The distance x, is found from the assumed stable layer LI LI height, L, which gives the value of a (from a = 0.47L). Z Z* The distance x, can then be found directly from Figure 3.3 in the Atmospheric Dispersion Estimates Workbook30. The only remaining unknown is then a , which can be found for a given distance from the source directly from pigure 3.2 in Reference 30. 71 ------- RESULTS FROM THE BINORMAL CONTINUOUS PLUME DISPERSION MODEL Using the service terms for each site as shown in Table 23, values for the downwind concentration of asbestos fibers were estimated for both daytime and nighttime con- ditions using two stability classes in each instance. Detail6 results for distances up to 30 kilometers are tabulated in Appendix B. In these tables, separate values have been computed for those fibers observed by optical microscope (i.e., greater than 1.5 ym) and those observed by electron microscope (i.e., less than 1.5 ym). The results for the five locations at Waukegan, Illinois) Denison, Texas, Marshville, North Carolina, Asbestos, Canada, and Eden Mills, Vermont, are displayed graphically in Figures 3, 4, 5, 6, and 7. The fiber concentration, x, plotted on the graphs, represents the number of fibers observed by electron microscope, although this may be taken as the total number of fibers with very little error, since, in general, the number of fibers observed under the electron microscope exceeded those observed under the optical microscoP by about three orders of magnitude. Fiber concentrations are observed to increase initially with distance from the source due to the effect of plume height. However, after a short distance of about 0.1 to 0.3 kilometers, there is a rapid fall in concentration until about 10 kilometers distance after which the graph asymp- totically approaches ambient air fiber concentration. The effect of the stable layer height is pronounced; much higher values are estimated to occur during nighttime due to the lowering of the height of the stable layer. Typically, this difference is about an order of magnitude in the distance range of 10 to 30 kilometers from the source. 72 ------- Table 23. DISPERSION EQUATION SOURCE TERMS -4 UJ Plant Location Waukegan, 111. Marshville, N.C. Denison, Tex. Asbestos, Que. Eden Mills, Vt. Baghouse Air Throughput nr*/min 1.13 x 103 4.73 x 102 7.36 x 102 1.27 x 105 8.49 x 103 Emitted Fiber Concentration From Optical Microscope Analysis Number/in^ 6.4 x 103 1.4 x 104 2.9 x 104 8.3 x 105 4.5 x 104 From Electron Microscope Analysis Number/m^ 1.1 x 107 5.0 x 109 1.4 x 107 1.4 x 109 1.3 x 108 Source Term From Optical Microscope Analysis Fibers/ sec 1.2 x 105 1.1 x 105 3.5 x 105 7.6 x 108 6.4 x 106 From Electron Microscope Analysis Fibers/sec 2.0 x 108 4.0 x 1010 1.7 x 108 3.1 x 1012 1.8 x 1010 ------- JM Waukegan, III. Stability Classes o B a C D JDay JNight 10 4> 5 10 20 Distance From Source 30 Km. Figure 3. The concentration of asbestos fibers with distance from source at four stability conditions for Johns-Manvilie, Waukegan, Illinois 74 ------- 10 J-M Denison Texas Stability Classes Day Night 10' IO -vs VI k. <1> JD L 10 5 10 20 Distance From Source 30 Km. Figure 4. The concentration of asbestos fibers with distance from source at four stability conditions for Johns-Manville, Denison, Texas 75 ------- IO Raybestos N.C. Stability Classes 0 n Day Night 5 Distance 10 From Source Km. Figure 5. The concentration of asbestos fibers with distance from source at four stability conditions for Raybestos, Marshville, North Carolina 76 ------- Asbestos Canada Stability Classes o B a C * D v E /Day JNIght 10 M l_ tt) 5 10 Distance From Source 20 30 Km. Figure 6. The concentration of asbestos fibers with distance from source at four stability conditions for Johns-Manville, Asbestos, Canada 77 ------- GAP Vermont Stability Classes o B a C JDay j Night 10 X in k- 0) .o 5 10 Distance From Source 20 30 Km. Figure 7. The concentration of asbestos fibers with distance from source at four stability conditions for GAP, Eden Mills, Vermont 78 ------- SECTION 8 REFERENCES 1 Roper, G. W. Asbestos Mill Filters. IIT Research Institute. (Presented at IIT Research Institute Seminar on Asbestos, Chicago, Illinois. 1972). 2. Goldfield, Joseph. Fabric Filters in Asbestos Mining and Manufacturing. (Presented at APCO Fabric Filter Symposium. Charleston, Virginia. 1971). 3- Hills, D. W. Economics of Dust Control. Annals of the New York Academy of Science. 132:322-334, 1964. ^- Rozovsky, H. Air in Asbestos Milling. Canadian Mining Journal. 78:95-103, May 1957. **• Wieschhaus, L. J. Recovering Asbestos Floats with Dust Collectors. Rock Products. 50(8):104-105, 1947. 6- Dust Control Methods. Coal Age. 72(8):56-62, 1967. ?- Dust Collection Plants. Cement Technology. Pp. 229-232, November/December 1972. D • Quebec Asbestos Producers Join Forces in Fighting Dust. Engineering and Mining Journal. 174(10):82-83, 1973. Q Control Techniques for Asbestos Emissions. Johns-Manvil] Research and Development Division internal report. Precipitators and Filters at 'Cleanest Ever1 Cement Works. Filtration and Separation. 10(1):40, 1973. Cross, F. L. Baghouse Filtration of Air Pollutants. Pollution Engineering. 6(2):25-34, 1974. b Dick, G. A. Fabric Filters. Canadian Mining Journal. 91(10):72-80, 1972. 79 ------- 13. Jensen, Kenneth E, Concepts of Fabric Filtration for Air Pollution Control. Filtration and Separation. 6(3):254,257, 1969. 14. Pring, R. T. Reducing Baghouse Maintenance by Design. Minerals Processing. 11(5):8-13, 1970. 15. Jones, A. H. How to Improve Maintenance of Fabric Dust Collectors. Minerals Processing, 10(5):21-23,35, 1969. 16. Dust and Fume Control Equipment in the Non-Ferrous Metals Industry. Filtration and Separation. 10(1):48, 1973. 17. Design and Operation of Asphalt Plant Bag Collectors. Pit and Quarry. 65(11):109, 1973. 18. Reigel, S. A., R. P. Bundy, and C. D. Doyle. Baghouses -' What to Know Before You Buy. Pollution Engineering. 5(5):32-34, 1973. 19. Cosby, W. T. and G. Punch. Dust Filters and Collectors, Cost and Performance of Filtration and Separation Equipment, Filtration and' Separation. 5(3):252-255,270, 1968. 20. Selikoff, I. J. , E. C. Hammond, and E. Heimann. Critical Evaluation of Disease Hazards Associated with Community Asbestos Air Pollution. Proceedings of the Second Inter- national Clean Air Congress. Pp. 165-171, 1970. 21. Minifie, F. G. and A. J. Moyes. Low Cost Electrostatic Precipitator. Filtration and Separation. 9(l):52-59, 1972. 22. Control Techniques for Asbestos Air Pollution. United States Environmental Protection Agency Report No. AP-11?* 1973. 23. Hailstone, R. E. Air Pollution in the Cement Industries- Minerals Processing. 10(5):11-15, 1969. 24. Stairmand, C. J. The Design and Performance of Modern Gas-Cleaning Equipment. Jnl. Inst. of Fuel. Pp. 58-81, February 1956. 25. Sommerlad, R. S. Fabric Filtration -- State of the Art. Livingston, New Jersey, Foster Wheeler Corporation, March 6, 1967. c 80 ------- 26. Petersen, C. M. and K. T. Whitby. Fractional Efficiency Characteristics of Unit Type Collectors. ASHRAE Journal. May 1965. 27. Colley, D. G. Stack Sampling Yields Fractional Efficien- cies for Dust Collectors. (Presented at Air Pollution Control Conference. Purdue University. October, 1970.) 28. Pasquill, F. The Estimation of the Dispersion of Windborne Material. Meteorol. Mag. 90(1063):33-49, 1961 29. Gifford, F. A. Uses of Routine Meteorological Observa- tions for Estimating Atmospheric Dispersion. Nuclear Safety. 2(4):47-51, 1961. 30. Turner, D. B. Workbook of Atmospheric Dispersion Estimates. PHS Publication No. 999-AP-26, 1970. 81 ------- SECTION 9 APPENDICES Page A. Selected Bibliography and Abstracts 83 B. Binomial Continuous Plume Dispersion Model Results 110 82 ------- Appendix A SELECTED BIBLIOGRAPHY AND ABSTRACTS 83 ------- SELECTED BIBLIOGRAPHY AND ABSTRACTS Addingley, C.G., "Dust Measurement and Monitoring in the Asbes- tos Industry," Annals of the New York Academy of Science, 132, pp 298-305, 1965.' All methods of dust counting of asbestos air concentra- tions require examination under a microscope. The Royco Particle Counter, using light scattering principles, is in- vestigated as an on-line monitor for asbestos concentrations in the air. Results show agreement with membrane filter techniques to within 25%. Addingley, C.G.. "Asbestos Dust and Its Measurements," Annals of Occupational Hygiene, 9, pp 73-82, 1966. The nature of asbestos dust and the testing requirements are discussed. Existing standard methods are briefly reviewed. The development of a membrane filter method of dust counting for asbestos is described in detail. It is thought to be an improvement on existing methods. Tyndallometric methods are considered, and a description of the application of the "Royco" Particle Counter, an instru- ment based on this principle, to factory testing is described. It is believed that this instrument represents a big advance in routine test methods. Aldred, Robert, "Dust Filtration Apparatus," British Patent No. 1293592, 1969. The patent discusses dust filtration in mines, avoiding the use of water, which gives further problems in cooling the mine, etc. Two stages are used, the first being a sort of cyclone, with deflectors, and then a band filter for retaining the finer dust. A fan is used downstream of the filter unit to pull the dirty air stream through the unit. 84 ------- Anonymous, "Bag Collects 60,000 Tons of Sinter Dust," Air Engineering, 10, No. 7, pp 8, 11, 1968. The dust collection system on the sinter line of the Bethlehem, Pennsylvania plant of the Bethlehem Steel Corpora- tion is described. The system operates at 99% efficiency and has reclaimed over 60,000 tons of dusts over five years. Anonymous, "Bag Life Extended at Copper Refinery," Air Engin- eering. 11, No. 1, pp 18-19, 1969. The baghouse at the Carteret, New Jersey plant of AMAX's United States Metals Refining Company is described. Adoption °f a new filter fabric made of Dupont's high temperature fiber, "Nomex" nylon, has resulted in an effective bag life ^ times greater than that of glass fiber bags. Anonymous, "Control Techniques for Asbestos Air Pollution," United States Environmental Protection Agency report No. AP-117, 1973. Asbestos is the generic name for a group of hydrated Mineral silicates that occur naturally in a fibrous form. ^he technological utility of asbestos derives from its physi- Cal strength, resistance to thermal degradation, resistance to chemical attack, and ability to be subdivided into fine fibers. The subdivision of asbestos into fine fibers produces Articulate matter that is readily dispersed into the atmosphere, Adverse effects of airborne asbestos on'human health have been Associated primarily with direct and indirect occupational e*posures, but a level of asbestos exposure below which there *s no detectable risk of adverse health effects to the gener- population has not yet been identified. Because of the of a practical technique of adequate sensitivity for small concentrations of airborne asbestos, neither 9ccurate emission factors nor emission-effect relationships are Bailable. 85 ------- Engineering appraisals9 based on limited data, indicate that the milling and basic processing of asbestos ore (crushing and screening the ore and aspirating the fiber to cyclones for grading) and the manufacture of asbestos-containing friction materials, asbestos-cement products, vinyl-asbestos tile, as- bestos textiles, and asbestos paper account for over 85 per- cent of total asbestos emissions. Other sources include: (1) the manufacture of other products containing asbestos, such as paints, coatings, adhesives, plastics, rubber materials* and molded insulating materials; (2) the use of spray on as- bestos products, such as those used for fireproofing or insula* ting; (3) the demolition of buildings or structures containing asbestos fireproofing or insulating materials; and (4) the sawing, grinding, or machining of materials that contain asbes- tos, such as brake linings and molded pipe insulation. In most of the manufacturing operations, the major emissions of asbestos occur when the dry asbestos is being handled, mixed with other dry materials, or dumped into the wet product mix, but the weaving of asbestos fibers into textiles and the machining or sanding of hard asbestos products also produce major emissions. Emissions are controlled in several ways: (1) by care- ful handling of dry materials to avoid generating dust; (2) by enclosing dusty operations; (3) by substituting wet processes for dry processes; (4) by wetting dry materials be- fore handling, sawing, or grinding; (5) by cleaning the dust-laden air by drawing it into ducts that lead to fabric filters; and (6) by reducing the amount of asbestos added to products the use of which leads to the generation of emission^' The last technique is particularly applicable to situations where the control of emissions by other methods is very diffi" cult, as with spray application of insulation or demolition of structures. The costs of needed emission control techniqueS 86 ------- can be estimated from those associated with existing practices. Anonymous, "Control Techniques for Asbestos Emissions," Johns- Manville Corporation Research and Engineering Division internal report, 1970, Johns-Manville is the largest producer of asbestos fiber and asbestos products in the free world. During the past 14 years, Johns-Manville has installed over 100 air and dust handling systems in its plants. The costs of these systems exceeded $18,000,000. The largest single installation was at the Jeffrey Mill No. 5 in Asbestos, Quebec. This bag- O £. house handles 1.27 x 10 lpm (4 x 10 cfm) and the total cost was $8,000,000. Anonymous, "Design and Operation of Asphalt Plant Bag Collec- tors," Pit and Quarry, 65, No. 11, p 109, 1973. Dust-laden exhaust gases from dryers in asphalt processing Plants are cleaned using baghouses. Operation and design of baghouse dust collection systems is described. Anonymous, "Dust and Fume Control Equipment in the Non-Ferrous Metals Industry," Filtration and Separation. 10, No. 1, p. 48, 1973. Bag filters, electrostatic precipitators and wet scrubbers are the equipment most used by the British non-ferrous metals industry to cope with fume, grit and dust arising from smel- ting and other operations. Emissions from the non-ferrous metals industry comprise dust and fine metallic particles, smoke from burning oil contaminating the raw materials, metallic oxide fumes, and, in the aluminum industry, chlorides and fluorides. Dust and metal particles are readily collected in settling cham- bers, cyclones, etc., but micron and submicron material (the Worst problem) can only be collected by high efficiency bag 87 ------- filters, high pressure drop scrubbers or electrostatic pre- cipitators. Anonymous, "Dust Collection Plants," Cement Technology, November-December, pp 229-232, The task imposed on dust collection plants as used in industry are many and various, resulting in the development of a range of dust collection plants based on a variety of working principles and designed to complement each other efficiently. On the basis of details supplied by the customer regarding his specific problem, the most suitable dust collecting sys- tem for his particular system can be worked out, which may, in fact, consist of a combination of two different dust collection methods. The types of equipment employed for dust collection come under the following general groups: (1) electrical dust precipitation; (2) cyclone dust collection; and (3) wet type dust collection. Anonymous, "Dust Control Methods," Coal Age. 72, No. 8, pp 56-62, 1967. The first step towards effective dust control is choosing equipment best suited to handle your dust problem. To do this requires an adequate understanding of dust control methods and dust collector characteristics. The different types of dust collectors are categorized in terms of their characteristics, including size, efficiency, and their applications. 88 ------- Anonymous, "Precipitators and Filters at 'Cleanest Ever' Cement Works," Filtration and Separation. 10, No. 1, p 40, 1973. The dust collection system at the new four million- tons Per acre cement factory at Northfleet, Kent is described. Fabric filters are preferred for smaller gas flows where temperatures and humidity are not excessive, e.g. at conveyor transfer points, cement packing and loading plants, and in some cases cement mill exhausts. But for treating very large gas volumes and flow rates electrostatic precipitators are used, particularly where high humidities and temperatures are encountered, as, for example, in the main kiln exhaust, the exhaust of excess air from moving grate clinker coolers, and, in some cases, the ventilating air from large cement mills. Anonymous, "Quebec Asbestos Producers Join Forces in Fighting Dust/' Engineering and Mining Journal, 174, No. 10, pp 82-83, 1973, The eastern townships of Quebec have been the center of the asbestos mining industry for nearly a century. In this area, approximately 115 miles east of Montreal, six mines produce approximately 29% of the world's asbestos. Because of close proximity of the towns to mine and mill operations, Quebec asbestos mining firms have long recognized the need to control the environment and have combined resources to do so. An Environmental Control Committee of the Quebec Asbes- tos Mining Association (QAMA) unites the competing companies in commitment of management, engineering, and environmental inspections to protect workers and communities. Substantial Deduction of dust emission points throughout the asbestos Plants has been achieved, often as a result of sharing hard-won knowledge among the member companies. 89 ------- Aureille, R. and Blanchot, "Experimental Investigation on the Effect of Different Parameters on the Separation Effic- iency of an Electrostatic Precipitator," Staub-Reinhalt. Luft, 31, No. 9, pp 23-28, 1971. To improve the efficiency of electrostatic precipitators for waste-gas cleaning of thermal power stations working with coal dust, Electricite de France, in its research and devel- opment center in Chatou, has erected a semi-industrial experi- mental electrostatic precipitator installation. With this installation, which is designed for horizontal flow of the waste gases, it is possible, during operation, not only to check the most important parameters likely to affect pre- cipitator efficiency, but also to vary each individual para- meter, while all other magnitudes remain unchanged. The experimental investigations extended from 1961 through 1969. Part of the results obtained are reported. Berlyand, M.E., "Investigations of Atmospheric Diffusion Providing a Meteorological Basis for Air Pollution Con- . trol," Atmospheric Environment, 6, No. 6, pp 379-388, 19/*' A summary of the principal lines of inquiry into the problem of atmospheric diffusion, including practical appli- cations, in the U.S.S.R. Topics summarized are Gaussian and K-theory diffusion models, plume and point-source diffusion methods, mulitple sources, and abnormal meteorological con- ditions. 'S, Charles E., and Wilder, John, "Handbook of Fabric Filter Technology," GCA Corporation internal report Billings, Fill No. GCA-TR-70-17-G, 1970. This report is the state-of-the-art of fabric filters. Major topics discussed are: (1) operating principles; (2) areas of application; (3) technology of fabric filtration processes; (4) types of fabric filters commercially available; (5) fabric selection; (6) engineering design of fabric 90 ------- systems; (7) performance of fabric filters; (8) economics; and (9) operation and maintenance of fabric filters. British Occupational Hygiene Society Committee on Hygiene Standards. "Hygiene Standards for Airborne Amosite Asbes- tos Dust. Annals of Occupational Hygiene. 16. pp 1-5, 1973. The sub-committee on asbestos has reviewed the informa- tion on the results of human exposure to airborne amosite dust and animal experiments. The subcommittee believes it has insufficient knowledge of the relationship between air- borne amosite dust exposure and the risk of asbestos to permit an accurate statement of the degree of protection afforded by a specified hygiene standard. Nevertheless, on the basis of comparisons between the effects of amosite and chrysotile dust on men and animals it is recommended that the standards for amosite should be no less stringent than those for chryso- tile. The sub-committee believes that a proper and reasonable objective would be to reduce the risk of contracting asbestosis to 1 percent of those who have a lifetime's exposure to the dust. By 'asbestosis1 this sub-committee means the earliest demonstrable effects on the lung due to asbestos. "The Hygiene Standards for Chrysotile Asbestos Dust" (British Occupational Hygiene Society, 1968) showed that the risk of being affected to the extent of having such early clinical signs will be less than 1 percent for an accumulated exposure of 100 fiber-year/cm . That is, for example, a concentration °f 2 fiber/cm for 50 years, 4 fiber/cm for 25 years, or 10 fiber/cm for 10 years. An accumulated exposure of 100 fiber-years per cm is therefore recommended for amosite. It is further recommended that exposures to amosite asbestos lie in certain ranges of dustiness be designated by 91 ------- categories according to the scheme recommended in the hygiene standards for chrysotile asbestos dust. Busse, Adrian D.; Zimmerman, John R., "User's Guide for the Climatological Dispersion Model," U. S. Environmental Protection Agency report No, EPA-R4-73-024, 1973. The Climatological Dispersion Model (CDM) determines long* term (seasonal or annual) quasi-stable pollutant concentra- tions at any ground-level receptor using average emission rates from point and area sources and a joint frequency distribution of wind direction, wind speed, and stability for the same period. This model differs from the Air Quality Display Model (AQDM) primarily in the way in which concentrations are determined from area sources and in the use in the CDM of Briggs1 plume rise formula and an assumed power law increase in wind speed with height that depends on stability. The material presented is directed toward the engineer familiar with computer techniques and will enable him to perform calculations with the CDM. Technical details of the computer programming are discussed; complete descriptions of input, output, and a test case are given. Flow diagrams and a source program listing are included. Companion papers by Calder (1971) on the technical details of the model and by Turner et al. (1972) on validation are included. Cheng, Lung, "Collection of Airborne Dust by Water Sprays," Industrial Engineering Chemical Process Design and Development, 12, No. 3. pp 221-225, 1973. A general theoretical equation was developed for the collection efficiency of airborne dust particles by spray drops. The model assumes an inertial impaction collection mechanism and is based upon mean interdrop length and mean 92 ------- interparticle area. Particular attention was given to the optimum drop size for collection with open and also confined sprays from a high-pressure nozzle. This general model was specifically adapted to the collection of dust particles in a horizontal tunnel by a solid-cone water spray by including the effects of duct size and spray configuration. The theoretical collection efficiencies agreed reasonably well with experimental data. The present results can be used to select a specific spray nozzle that will provide optimum col- lection efficiency for airborne dust based upon the available water flow rate and the line pressure. Cosby, W.T.; Punch, G., "Dust Filters and Collectors - Cost and Performance of Filtration and Separation Equipment," Filtration and Separation, 5, No. 3, pp 252-255, 270, 1968. In arriving at the 'best buy1 in gas cleaning plant, it is essential to consider the duties to be performed, methods of containing the dust or fume and its disposal after removal. In this paper, the authors try to show how the selection of equipment is influenced by factors other than the character- istics of particles and collectors. They base their cost comparison data on two hypothetical duties each demanding efficiencies of 98-99%. Annual costs are calculated for fabric filter, dry plate precipitators, and venturi scrubberse The effect on collector costs of high initial gas temperatures are considered and the provision of cooling equipment is taken into account. Cross, Frank L., Jr., "Baghouse Filtration of Air Pollutants," Pollution Engineering, 6, No. 2, pp 25-34, 1974. Filtration is one of the oldest methods of source control and is especially desirable for the removal of particulate mat- ter from a gas stream. Satisfactory efficiencies are obtained 93 ------- with this type of unit with only moderate power consumption and relatively low maintenance problems if the proper filter is selected. Dick, G.A., "Fabric Filters," Canadian Mining Journal. 91, No. 10, pp 72-80, 1972. A fabric filter is a device used for freeing process gases and liquids from suspended impurities, and both wet and dry filtration is practiced in the mining industry for ticle separation. This paper deals solely with dry filtra- tion and gas cleaning techniques for the recovery of parti- culate matter to meet air quality control standards. The mining industry has long recognized the need for effective control of airborne contaminants and has used some form of "cloth filter" or "baghouse" for dust control work since the turn of the century. This method of gas cleaning has repre- sented one of the best means of obtaining the highest con- sistent recovery of gas-borne particulate matter and fabric filters have demonstrated that, when properly operated and maintained, collection efficiencies of airborne particulates can be exceptionally high. Edmisten, Norman G.; Bunyard, Francis L., "A Systematic Pro- cedure for Determining the Cost of Controlling Particular Emissions from Industrial Sources," Journal of the Air ^n Pollution Control Association, 20, No. 7, pp 446-452, 19/U' The increasing concern about the air quality of our na- tion has created a similar concern in the costs associated with reducing emissions to desirable levels. The purpose of this paper is to present a methodology for assessing the cost of controlling particulate emissions from industrial sources. Basic equipment costs were collected and evaluated for dry centrifugal and wet collectors, fabric filters, elec- trostatic precipitators, and afterburners. Manufacturers, 94 ------- installers, users, and operators of air pollution control equipment were contacted to obtain the necessary cost data for the year 1968. A basic premise of the procedure devel- oped is that the most meaningful approach to the evaluation and comparison of air pollution control costs is based on the total cost of control annualized over the expected economic life of the equipment. Items such as capital charges and expenditures for operation, maintenance, and collected waste disposal are generally more significant in cost accounting than the depreciated value of the initial investment. By defining the size and efficiency of collection required; the Degree of difficulty in installing the equipment; and know- ledge of some of the characteristics of the involved process, gas stream, and pollutant characteristics, the cost of control can be estimated with the assistance of cost factors and guidelines presented. Gifford, Franklin A., Jr., "Atmospheric Dispersion," Nuclear Safety, 1, No. 3, pp 56-68, 1960. One of the chief sources of uncertainty in estimating the hazard associated with accidental or planned release to the atmosphere of fission-product activity has been the lack °f reliable measured values of atmospheric dispersion coef- ficients. In the absence of any obvious alternative, Button's ^ell-known mathematical dispersion model has been used in reactor hazards analyses for evaluating effects far beyond limits for which the model can confidently be expected to ke reliable, e.g., distances of the order of 1 km and near adiabatic (neutral) conditions of atmospheric stability. Consequently, the appearance, in several recent papers, of a sizable quantity of new atmospheric dispersion observations ^•s of considerable interest in connection with the meteorology °f nuclear safety problems. Furthermore, the calculation °f atmospheric dispersion by the method of moving averages, 95 ------- as has been proposed recently, seems to provide an improved means of calculating dispersion, not only because the tech- nique has less restrictive boundary conditions but also be- cause it is well adapted to the interpretation of continuously monitored atmospheric data. Gifford, Franklin A., Jr., "Atmospheric Dispersion Calculations Using the Generalized Gaussian Plume Model," Nuclear Safety, 2, No. 2, pp 56-59, 67-68, 1960. A number of formulas for dealing with various practical dispersion problems that arise in reactor hazard analyses are based on the widely used dispersion model formulated by Sutton. However, results of recent dispersion experiments have more and more often been presented in terms of the simple Gaussian interpolation formula. Gifford, Franklin A., Jr., "Use of Routine Meteorological Observations for Estimating Atmospheric Dispersion," Nuclear Safety, 2, No. 4, pp 47-51, 1961. Estimates of atmospheric dispersion are essential infor- mation in the selection of a reactor site and in the evaluation of the hazards of reactor operation. In selecting a site, the dispersion characteristics of the atmosphere at the various sites under consideration are important because most reactors, if not all, generate or induce some atmospheric radioactivity during routine operation and because there is the possibility of accidental release of radioactivity to the atmosphere. Only a few forecasters are familiar with low-level dispersion problems, and consequently it is desirable that simple, easily applied methods of estimating atmospheric dispersion, preferable those employing routine meteorological observations, be dis- cussed. 96 ------- Gifford, Franklin A., Jr., "The Area Within Ground-Level Dosage Isopleths," Nuclear Safety, 4, No.2, pp 91-92, 97, 1962. The total radioactive dosage to a population has frequent- ly been identified as an important aspect of the potential hazard associated with reactor accidents. The total population dosage is equal to the product of people times radioactive dosage, summed over the population, with appropriate high- and low-dosage cutoffs taken into account. To expedite computation of this quantity, it is evidently necessary to be able to calculate the area inside ground-level isodose contours, i.e., the intersection between the surface formed by a given air concentration or dosage Value and the ground. Based on ground-level air-concentration isopleths compu- ted by means of the generalized Gaussian dispersion model is described. Goldfield, Joseph, "Fabric Filters in Asbestos Mining and Asbestos Manufacturing," APCO Fabric Filter Symposium, Charleston, Virginia, 1971. Johns-Manvilie Corporation has hundreds of fabric filters In use in its plants. The advantages of baghouses are their high efficiency and reliability. The largest single baghouse is that installed at the Jeffrey Mill No. 5 in Asbestos, Quebec, Canada. This installation is described in detail. Its overall mass efficiency is 99.992%. Hailstone, R.E., "Air Pollution Control in the Cement Industry," Minerals Processing, 10, No. 5, pp 11-15, 1969. There are many technical difficulties to adapt presently available emission control devices to this complex manufacturing Process of cement making. Controlling emissions within the limits of recently enacted or pending air pollution control 97 ------- regulations are great in magnitude and cost. Neglect of any one of a multitude of design parameters, or inadequate, impro- per design of control devices can make a continuous high level operating efficiency essentially impossible to attain. Increased technology may permit further emission control improvements but at high cost. Proper emphasis should now be placed on the "technically feasible, economically reason- able, practically enforceable" air pollution control regula- tion, and logical priorities. The portland cement industry recognizes that only through the cooperative efforts of the control agency, the public and industry will it be able to achieve the goals of desirable air quality levels. Hills, D.W., "Economics of Dust Control," Annals of the New York Academy of Science, 132, pp 322-334, 1964. A brief historical introduction is given about conditions in the U.K. asbestos textile industry in the late 1920's and early 1930*s; reference is made to the 1931 conference between employers and the Home Office that resulted in a code of practice being established for dust suppression in asbestos textile factories. Details are given of the various methods used for dust control in the Company's factories together with the cost of these measures. Some figures are also given for the cost of dust control at the Cape Asbestos Company, Ltd.'s new amosite mill at Penge. Current work on improving dust control is discussed together with the part now played in this by the Asbestosis Research Council. Although asbestos textiles form only a part of the as- bestos industry, their particular significance for this monograph is that, at least in the United Kingdom, the hazards associated with the processing of asbestos were recognized sooner on the textile side than almost anywhere else. This 98 ------- was because most of the processes were dry, and hence dust was readily formed, and also because the workers in some of the largest factories formed a close-knit population whose employment records went back for many years. Thus, when the hazard was first investigated, an excellent sample of people with varying periods of exposure to dust was available. Hussey. A.M.W., "Dust Collection in Industry," Filtration and Separation, 10, No. 2, pp 181-188, 1973. The author describes the four basic types of dust col- lection equipment: cyclone collectors, wet scrubbers, bag filters, and electrical precipitators, and discusses some of their major applications. He emphasizes that a successful installation requires complete understanding between supplier and user and to assist towards this he summarizes equipment selection data which should be considered when installations are being planned. Jensen, Kenneth, E., "Concepts of Fabric Filtration for Air Pollution Control," Filtration and Separation. 6, No. 3, pp 254,257, 1969. A brief review of other accepted methods of air and gas cleaning equipment would be in order to relate them to cloth filtration. Where conditions permit, the filtration of gases through a fabric media offers a number of advantages: efficiencies down to the sub-micron range is inherent; there is a positive barrier in place that does not depend on supplementary de- -ices or materials such as water, changes in direction, or electrical charges; and there is no secondary pollution Problem. The material collected in a dry state is re-usable if this is desired and in some cases can be sold to help offset collection costs. 99 ------- The application of a fabric collector to a dust collection problem is not, however, completely dependent upon economic and efficiency factors. The physical conditions that pre- vail have a great bearing on whether or not cloth collection is the proper answer. Jones, Allen H., "How to Improve Maintenance of Fabric Dust Collectors," Minerals Processing, 10, No. 5, pp 254,257, 1969. The key to low maintenance of a cloth dust collection system is early detection of problems. Dust collection sys- tems are self destructive once they begin to malfunction and small problems becomes disasters very rapidly. Trouble in one part of a system can affect another in a short period of time. Weekly visual inspection of a collection system combined with proper measurements is mandatory to maintain low cost maintenance, free from shut-downs and crises. The only way to accurately evaluate dust filter perfor- mance and to anticipate maintenance trouble is to take mano- meter, pilot tube, ammeter, and. tachometer readings. If these readings are taken on a regular basis and compared with former readings it will not be necessary to wait for disaster to hit. Lundgren, D.A.; Greene, V.W., "Filtration and Dust Control Equipment in the Production of Controlled Air Environ- ments," Filtration and Separation, 5, No. 5, pp 405-412, 1968. After summarizing information which is basic to the de- sign and operational problems of controlling the contamination level of air within an enclosure, the authors discuss per- formance and cost criteria for selecting air cleaning devices. They deal in turn with the characteristics of seven types of 100 ------- air filters and other dust control devices and conclude by considering the problems of controlling microbes in air. Minifie, F.G.; Moyes, A.J., "Low Cost Electrostatic Precipita- tion," Filtration and Separation, 9, No. 1, pp 52-59, 1972. Three well-known types of dust collector are compared and contrasted. The phenomena occurring in an electrostatic precipitator are described to show why the various components are needed. The factors which lead to heavy costs in elec- trostatic precipitator construction are described. Means of reducing these costs, using unit precipitators of pre-estab- lished design, are indicated. Costs of installing three types of dust collector on two different duties are compared, and the fallacy that electrostatic precipitators are expensive is dispelled. Minnick, L. John, "Control of Particulate Emissions from Lime Plants - A Survey," Journal of the Air Pollution Control Association, 21, No. 4, pp 195-200, 1971. This paper describes the achievements of the lime indus- try in developing methods of handling and controlling the various finely divided products which they produce. An ex- tensive survey provides useful data on the availability and performance of many of the control devices that are currently In use, and an analysis is made of the operating efficiencies and costs of this equipment. The environmental control pro- grams which are currently underway in this industry are des- cribed, and an evaluation is made of these programs. The ultimate goals that are believed to be attainable are pre- sented from the standpoint of emission control from individual processes as well as from operating plant complexes. While the paper deals primarily with practical operating and engin- eering aspects of the subject, some information is also inclu- ded on methods of tests and the monitoring systems that are in use. 101 ------- Morrison, Joseph N., Jr., "Controlling Dust Emissions at Belt Conveyor Transfer Points," Transactions of the Society of Mining Engineers, 250, No. 1, pp 47-53, 1971. A comprehensive solution is offered to the problem of dust emissions at belt conveyor transfer points. Details of enclosure design are discussed and a straightforward procedure for calculating required dust control exhaust volume is pre- sented. Many design variables are taken into account which heretofore have been commonly ignored or inadequately consi- dered. These include belt widths, belt speeds, enclosure openings, material flow rate, material bulk densities, material lump sizes, height of material fall, material temperature, and ambient air temperature. All of these questions are handled by means of a "fill-in-the-blanks" type of calculation form, permitting quick, reliable solutions by relative "non-experts. Pasquill, F., "The Estimation of the Dispersion of Windborne Material," The Meteorological Magazine, 90, No. 1063, 33-49, The theoretical estimation of the concentrations arising from sources of gaseous or finely divided particulate material has for long been based on treatments of atmospheric diffusion developed by Sir Graham Sutton. These formulae are reliable for specifying the average distribution, over a few hundred meters downwind of a source operating for a few minutes on level unobstructed terrain, with a steady wind direction and neutral conditions of atmospheric stability. Extension to other circumstances has depended on empirical and often specu- lative adjustments of the diffusion parameters. During the last few years, investigations have shown that a fairly rational allowance can now be made for the effects of much of the wide variation in atmospheric turbulence which occurs in reality. This progress includes some extension to longer distances of travel. 102 ------- The purpose of this article is to review the recent background of theoretical and experimental results, and to give details of the proposed system of calculating the distribution of concentration downwind of a source. These details are set out in two appendices, the first giving complete instruc- tions for carrying out the calculations, the second presenting an example. Popa, Bazil; Jancau, Vasile, "The Probability of Certain Con- centrations in the Dispersion of Solid Dust Particles in Industrial Regions," Staub-Reinhaltung der Luft. 33, No. 1, pp 20-24, 1973. The measurement of particulate components in the air is necessary to the health of a community. The prediction of concentrations of particulates is necessary for the planning of controls on present sources and the introduction of new industry, i.e., new sources in the region. The influence of the wind is of prime importance; not only its speed, but its direction, is discussed. Wind is a random variable and mathematical presentation using statistics are given. Various distribution curves (exponential, logistic, Fisher-Tippett type II (Frechent) and type I (Gumbel), Cauchy, normal Laplace-Gauss) are compared. The town of Chy, Romania is used as an example. Pring, R.T., "Reducing Baghouse Maintenance by Design," Minerals Processing, 11, No. 5, pp 8-13, 1970. Structural baghouses comprise one branch of the fabric collector family and are characterized by: 1. Large volume capacity 2. Normally high-temperature service 3. Custom design to fill specific needs 4. Use of large diameter filter bags 103 ------- 5. Absence of internal moving parts 6. Suitability for continuous heavy duty service Because of design flexibility, baghouses can be furnished in almost infinite configurations and, within reason, at almost any price the purchaser is willing to pay - ranging from the "Model T" to the "Cadillac". Further, the choice of accessories can affect cost materially. Because the very substantial investment in air pollution control equipment returns no profit, the industry-wide ten- dency is, understandably, to keep capital costs to an absolute minimum. It is suggested that familiarity with the many available options in design and accessories and how they affect maintenance and operation will help prevent problems after startup. Rajhans, Cyan S., "Fibrous Dust - Its Measurement and Control, The Canadian Mining and Metallurgical Bulletin, 63, No. 8, pp 900-910, 1970. The strategy of fibrous dust sampling is discussed, sampling methods are critically reviewed and their application to coal dust is demonstrated. Fiber counting is described in detail. An attempt is made to explain the basis of determining the threshold limit value of asbestos and other dusts. The paper also discusses such dust control methods as enclosure of the process, effective local exhaust ventilation, segregation, substitution, wet processing, and continuous monitoring of the return air for recirculation. Reigel, S.A.; Bundy, R.P.: Doyle, C.D., "Baghouses - What to Know Before You Buy, Pollution Engineering, 5, No. 5, pp 32-34, 1973. With the advent of stricter emission standards designed o to produce the national quality standard of 75 |jg/m of 104 ------- particulate, only the most efficient removal devices will be suitable. The baghouse traditionally yields high removal efficiencies (99.9+%). Reitze, William B.; Haladay, D.A.; Romer, Harold; Fenner, E.M. , "Control of Asbestos Fiber Emissions from Industrial and Commercial Sources," Proceedings of the Second Interna- tional Clean Air Congress, pp 100-103, 1970. There are five major sources from which asbestos fiber enters the air: (1) raining; (2) milling; (3) manufacturing; (4) certain segments of the construction industry; and (5) naturally occurring sources. The first four are created by modern man's technology and the last by normally occurring changes in our environment. The operations of each source with controls that are now in use are listed. Roper, G.W., "Asbestos Mill Filters," IIT Research Institute Seminar on Asbestos, 1972. The asbestos milling industry is the largest user of cloth filters of all industries, since the separation of the fibers from the ore is done pneumatically. Dust collection systems installed by the Wheelabrator Corporation at various asbestos mills are described. Rozovsky. H., "Air in Asbestos Milling," Canadian Mining Journal, 78, pp 95-103, 1957. The production of asbestos fiber in modern times depends almost completely on air movement. Almost every part of the operation employs air. Whether it is for ventilation and dust control or for drying, separation or conveying, air is required. Nearly 400 tons of air per minute are employed in all phases of this industry in Canada. It is readily apparent 105 ------- that to move this 10,000,000 cfm a considerable amount of power and equipment is required. Nearly 90% of this air is used directly in drying and milling processes. The purpose of this paper is to discuss the importance of air and its use in asbestos milling plants. Rushton, A.; Griffiths, P.V.R., "Role of Cloth in Filtration," Filtration and Separation, 9, No. 1, pp 81-89, 106, 1972. The influence of cloth weave patterns on the permeability of clean monofilament and multifilament cloths is presented and a successful correlation of pressure drop-flow data is reported for several cloth types. The effect of cloth pore structure on the mechanisms of cake deposition and the con- ditions required for bridging the pore is discussed,, Schoek, V.E., "In Mechanical Dust Collectors, It's the Fabric that Really Counts," Engineering and Mining Journal, 173, No. 1, pp 98-99, 1972. Baghouse filters for reducing particulate emissions and recovering metal values are becoming more efficient, thanks to better filter fabrics. Here are some tips on fabric selec* tion and installation. Selikoff, I.J.; Hammond, E.G.; Heimann, E., "Critical Evalua- tion of Disease Hazards Associated with Community Asbes- i tos Air Pollution," Proceedings of the Second Internatio*1 Clean Air Congress, pp 165-171, 1970. The results of 3,000 consecutive autopsies in New York City is correlated with asbestos bodies. The results of sampling the ambient air of New York City show an asbestos —93 air level of 11-60 x 10 g/m . Types of exposure as well as sources and control methods are discussed. 106 ------- Sullivan, Ralph J.; Athanassiadis, Yanis C., Air Pollution Aspects of Asbestos, National Air Pollution Control Administration, No. PH-22-68-25, Washington, B.C., 1969. Inhalation of asbestos may cause asbestosis, pleural or peritoneal mesothelioma, or lung cancer. Mesothelioma is a rare form of cancer which occurs frequently in asbestos workers All three of these diseases are fatal once they become es- tablished. The dose necessary to produce asbestosis has been estimated to be 50 to 60 million particles per cubic foot- years. No information is available on the dose necessary to induce cancer. Random autopsies of lungs have shown "asbes- tos bodies" in the lungs of one-fourth to one-half fo samples from urban populations. Thus, the apparent air pollution by asbestos reaches a large number of people. Animals have been shown to develop asbestosis and cancer after exposure to asbestos. No information has been found on the effects of asbestos air pollution on plants or materials. The likely sources of asbestos air pollution are uses of the asbestos products in the consturction industry and asbes- tos mines and factories. Observations in Finland and Russia indicate that asbestos does pollute air near mines and fac- tories. However, no measurements were reported of the con- centration of asbestos near likely sources in the United States, A concentration in urban air of 600 to 6,000 particles per cubic meter has been estimated. Bag filters have been used in factories to control asbes- tos emissions; the cost of this type of control in a British factory was approximately 27.5 percent of the total capital cost and about 7 percent of the operating cost. No informa- tion has been found on the costs of damage resulting from Asbestos air pollution. 107 ------- No satisfactory analytical method is available to deter- mine asbestos in the atmosphere. Turner, D.Bruce, Wgrkbopk of Atmospheric Dispersion Estimates, National Air Pollution Control Administration, Cincinnati, 1970. This workbook presents methods of practical application of the binormal continuous plume dispersion model to estimate concentrations of air pollutants. Estimates of dispersion are those of Pasquill as restated by Gifford0 Emphasis is on the estimation of concentrations from continuous sources for sampling times up to 1 hour. Some of the topics discussed are determination of effective height of emission, extension of concentration estimates to longer sampling intervals9 inver- sion break-up fumigation concentrations, and concentrations from area, line, and multiple sources. Twenty-six example problems and their solutions are given. Some graphical aids to computation are included. Werle, Donald K., "Fabric Filters in Pollution Control/' IIT Research Institute internal report No. I.ITRI«C8196~T49 1972. Fibrous filters are commonly used in the cleaning of air for ventilation purposes and for industrial gas cleaning„ The latter application, of primary interest in this report9 can involve the cleaning of gases with very high dust loadings, and many filter fabrics are available for use where corrosive gases or moderate temperatures are involved. The intent of the report is to cover the operating principles, design methodology, economics, and application of fabric filters in air pollution control. 108 ------- Wieschhaus, L.J., "Recovering Asbestos Floats with Dust Col- lectors," Rock Products, 50, No. 8, pp 104-105, 1947. Dust clouds created in the milling of asbestos have long defied efficient and economical collection and it is only recently that this problem has been successfully solved. Be- cause of the vast quantities of asbestos lost in the form of dust and because of the high market value of these "floats" at the present time, it may be well to consider some of the factors involved in the collection of asbestos. 109 ------- Appendix B BINORMAL CONTINUOUS PLUME DISPERSION MODEL RESULTS 110 ------- SYMBOLS USED IN THE FOLLOWING TABLES x = distance from source downwind in the direction of the mean wind (kilometers) a - the standard deviation in the crosswind direction of y the plume concentration distribution (meters) a = the standard deviation in the vertical of the plume concentration distribution (meters) X = fiber concentration (fibers per cubic meter) OM = number of fibers observed by optical microscope EM = number of fibers observed by electron microscope 111 ------- Calculation of Concentrations for Various Distances JOHNS-MANVILLE CORPORATION ASBESTOS PRODUCTS PLANT WAUKEGAN, ILLINOIS Mean Wind Speed: 4.47 m/sec Source Term -- Based on Optical Microscopy: 1.20 x 10 fibers/sec « Based on Electron Microscope: 2.03 x 10 fibers/sec Source Height: 10 m Daytime Mixing Height: 800 m Nighttime Mixing Height: 150 m Stability Class B; DAYTIME X (km) 0.1 0.3 0.5 0.8 1.0 2.0 3.0 3.1 6.2 7.0 10.0 15.0 20.0 30.0 (m) 19.0 52.0 83.0 130 158 290 420 440 800 880 1,200 1,700 2,150 3,050 a z (m) 11.0 31.0 51.0 85.0 109 233 360 370 800 800 800 800 800 800 X 3 (fibers/in ) OM 2.71 x 101 5.03 1.98 T 7.68 x 10"t 4.94 x 10"} 1.26 x 10, 5.65 x 10, 5.25 x 10, 2.37 x 10, 2.15 x 10", 1.58 x 10~, 1.11 x 10"; 8.81 x 10"o 6.21 x 10"J EM 4.58 x 10o 8.51 x 10^ 3.35 x 10o 1.30 x 10, 8.36 x 10, 2.13 x 107 9.56 x 10} 8.88 x 107 4.01 x 10} 3.64 x 10} 2.67 x 10} 1.88 x 10} 1.49 x 10} 1.05 x 10L 112 ------- Stability Class C; DAYTIME X (km) 0.1 0.3 0.5 0.8 1.0 2.0 3.0 5.0 6.5 7.3 14.6 17.0 20.0 25.0 30.0 a (m) 12.5 34.0 55.0 86.0 105 200 285 450 570 630 1,150 1,330 1,520 1,850 2,170 a z (m) 7.5 20.5 30.2 49.0 61.0 115 170 265 340 380 800 800 800 800 800 X 3 (fibers /in ) OM 3.76 x 10} 1.09 x 101 4.87 1.99 1.32 1 2.70 x 10~t 1.76 x 10"; 7.16 x 10"« 4.41 x 10", 3.57 x 10, 1.65 x 10" « 1.42 x 10", 1.25 x 10", 1.02 x 10"o 8.72 x 10~J EM 6.36 x lo£ 1.84 x 10, 8.24 x 10, 3.37 x 10, 2.23 x 10, 6.26 x 10, 2.98 x 10, 1.21 x 107 7.46 x 10t 6.04 x lot 2.79 x 107 2.40 x I0t 2.11 x 107 1.73 x lot 1.48 x 104- Stability Class D: NIGHTTIME X (km) 0.1 0.3 0.5 0.8 1.0 2.0 3.0 3.5 7.0 10.0 15.0 20.0 30.0 (m7) 8.1 22.0 36.0 55.0 67.0 130 188 215 400 550 790 990 1,430 a z (m) 4.6 12.0 17.5 26.5 31.5 49.0 64.0 71.0 150 150 150 150 150 X 0 (fibers/m ) OM 2.18 x IQ} 2.29 x I0t 1.15 x 101 5.46 2.85 1.31 . 7.02 x 10~t 5.54 x 10"t 2.52 x 10~t 1.84 x 10"7 1.28 x 10~t 1.02 x 10", 7.06 x 10"z EM 3.69 x 10? 3.87 x 107 1.95 x 10, 9.24 x 10, 6.51 x 10, 2.22 x 10, 1.19 x 10, 9.37 x 10, 4.26 x 10, 3.11 x 10, 2.17 x 10, 1.73 x 10, 1.19 x 10^ 113 ------- Stability Class E: NIGHTTIME X (km) 0.1 0.3 0.5 0.8 1.0 2.0 3.0 5.0 8.0 16.0 20.0 25.0 30.0 (m) 6.0 16.5 26.5 40.0 50.0 95.0 140 220 335 620 750 920 1,070 On) 3.5 8.8 13.0 18.0 21.3 33.5 43.0 55.0 71.0 150 150 150 150 X 3 (fibers/nT) OM 6.80 . 3,07 x 107 1.85 x 10t 1.02 x 10X 6.97 2.57 1.38 , 6.95 x 10 "t 3.56 x 10"v 1.63 x 10": 1.35 x 10"J 1.10 x 10"i 9.44 x 10 EM 1.15 x 10? 5.19 x 10? 3.13 x 10? 1.73 x 10? 1.18 x 10o 4.35 x 10o 2.33 x 10n 1.18 x 10« 6.02 x 10- 2.76 x 10^ 2.28 x 10, 1.86 x 10, 1.60 x 10 114 ------- Calculation of Concentrations for Various Distances JOHNS-MANVILLE CORPORATION ASBESTOS CEMENT PIPE PLANT DENISON, TEXAS Mean Wind Speed: 4.07 m/sec Source Term -- Based on Optical Microscopy: 3.53 x 10 fibers/sec Based on Electron Microscopy: 1.69 x 10 fibers/sec Source Height: 10 m Daytime Mixing Height: 800 m Nighttime Mixing Height: 150 m Stability Class B; DAYTIME 8 (tan) 0.1 0.3 0.5 0.8 1.0 2.0 3.0 3.1 6.2 7.0 10.0 15.0 20.0 30.0 a o5 19.0 52.0 83.0 130 158 290 420 440 800 880 1,200 1,700 2,150 3,050 az (m) 11.0 31.0 51.0 85.0 109 233 360 370 800 800 800 800 800 800 X 3 (fibers /m ) OM 8.74 x lo} 1.70 x 101 6.40 2.48 1.60 - 4.08 x 10"} 1,83 x 10"} 1.70 x 10, 7.65 x 10", 6.95 x 10", 5.10 x 10", 3.60 x 10", 2.84 x 10", 2.01 x 10"z EM 4.18 x 10o 8.14 x 10o 3.06 x 10. 1.19 x 10, 7.66 x 10, 1.95 x 107 8.76 x 107 8.14 x ID} 3.66 x 107 3.33 x 10} 2.44 x 10} 1.72 x 107 1.36 x 101 9.62 115 ------- Stability Class C: DAYTIME (km) 0.1 0.3 0.5 0.8 1.0 2.0 3.0 5.0 6.5 7.3 14.6 17.0 20.0 25.0 30.0 ay (m) 12.5 34.0 55.0 86.0 105 200 285 450 570 630 1,150 1,330 1,520 1,850 2,170 vz (m) 7.5 20.5 30.2 49.0 61.0 115 170 265 340 380 800 800 800 800 800 X . (fibers/m ) OM 1.22 x 10? 3.52 x lot 1.57 x 101 6.41 4.25 1.20 -, 5.69 x 10~t 2.31 x 10"! 1.42 x 10 "I 1.15 x 10"; 5.32 x 10~« 4.60 x 10~~ 4.02 x 10, 3,31 x 10~« 2.82 x 10"z EM 5.84 x 10 J 1.69 x 10, 7.52 x 10, 3.07 x 10, 2.03 x 10« 5.75 x 10« 2.72 x 10« 1.11 x 107 6.80 x 107 5.51 x lot 2.55 x lOt 2.20 x lot 1.92 x lot 1.58 x lot 1.35 x 101 Stability Class D: NIGHTTIME (km) 0.1 0.3 0.5 0.8 1.0 2.0 3.0 3.5 7.0 10.0 15.0 20.0 30.0 a (m) 8.1 22.0 36.0 55.0 67.0 130 188 215 400 550 790 990 1,430 vz (m) 4.6 12.0 17.5 26,5 31.5 49.0 64.0 71.0 150 150 150 150 150 X 3 (fibers/in ) OM 6.96 x lot 7.39 x lot 3.72 x I0t 1.76 x 107 1.24 x 101 4.24 2.27 1.79 , 8.16 x 10"t 5.93 x 10"t 4.13 x 10"t 3.30 x 10 "I 2.28 x 10"1 EM 3.33 x lo£ 3.54 x 10? 1.78 x 10, 8.43 x 10, 5.94 x 10, 2.03 x 10, 1.09 x 10« 8.57 x 10« 3.91 x 10« 2.84 x 10« 1.98 x 10, 1.58 x 10« 1.09 x 10Z 116 ------- Stability Class E: NIGHTTIME X (km) 0.1 0.3 0.5 0.8 1.0 2.0 3.0 5.0 8.0 16.0 20.0 25.0 30.0 (m5) 6.0 16.5 26.5 40.0 50.0 95.0 140 220 335 620 750 920 1,070 az Cm) 3.5 x 3 (fibers/nr) OM 2.20 x 10} 8.8 9.93 x 10t 13.0 18.0 21.3 33.5 43.0 55.0 71.0 150 150 150 150 5.96 x 10t 3.29 x 10t 2.25 x 101 8.29 4.46 2.25 1.15 , 5.26 x 10"! 4.35 x 10"! 3.55 x 10"! 3.05 x 10"1 EM 1.05 x 10? 4.75 x 107 2.85 x 107 1.58 x 107 1.08 x 10? 3.97 x 10^ 2.14 x 10. 1.08 x 10- 5.51 x 10« 2.52 x 10- 2.08 x 10« 1.70 x 10o 1.46 x 10^ 117 ------- Calculation of Concentrations for Various Distances RAYBESTOS - MANHATTAN ASBESTOS TEXTILE MANUFACTURING PLANT MARSHVILLE, NORTH CAROLINA Mean Wind Speed: 3.35 m/sec Source Term -- Based on Optical Microscopy: 1.12 x 10 fibers/sec Based on Electron Microscopy: 3.97 x 10 fibers /sec Source Height: 10 m Daytime Mixing Height: 800 m Nighttime Mixing Height: 150 m Stability Class B; DAYTIME X (km) 0.1 0.3 0.5 0.8 1.0 2.0 3.0 3.1 6.2 7.0 10.0 15.0 20.0 30.0 (m) 19.0 52.0 83.0 130 158 290 420 440 800 880 1,200 1,700 2,150 3,050 (mZ) 11.0 31.0 51.0 85.0 109 233 360 370 800 800 800 800 800 800 X 3 (fibers/nr) OM 3.37 x 101 6.27 2.47 n 9.56 x 10 T 6.15 x 10"| 1.57 x 10, 7.04 x 10, 6.54 x 10, 2.95 x 10", 2.68 x 10"; 1.96 x 10", 1.39 x 10", 1.10 x 10". 7.73 x 10"J EM 1.19 x 10? 2,22 x 10? 8.76 x 10? 3.39 x 10? 2.28 x 10? 5.57 x 10? 2.50 x 10? 2.32 x 10? 1.05 x 107 9.50 x 10o 6.95 x 10o 4.92 x 10o 3.90 x 10o 2.74 x 10J 118 ------- Stability Class C: DAYTIME X (km) 0.1 0.3 0.5 0.8 1.0 2.0 3.0 5.0 6.5 7.3 14.6 17.0 20.0 25,0 30.0 (/) 12.5 34.0 55.0 86.0 105 200 285 450 570 630 1,150 1,330 1,520 1,850 2,170 az (m) 7.5 20.5 30.2 49.0 61.0 115 170 265 340 380 800 800 800 800 800 X 3 (fibers/nr) OM 4.69 x loJ- 1.36 x 101 6.07 2.47 1.64 , 4.61 x 10"t 2.19 x 10", 8.92 x 10", 5.49 x 10", 4.45 x 10"; 2.05 x 10"; 1.77 x 10"; 1.55 x 10"; 1.27 x 10", 1.09 x 10"z EM 1.66 x 10? 4.82 x 10? 2.15 x 10? 8.76 x 10? 5.81 x 10? 1.63 x 10? 7.76 x 10? 3.16 x 10? 1.95 x 10? 1.58 x 10o 7.27 x 10o 6.27 x 10o 5.49 x 10^ 4.50 x 10o 8.86 x 10J Stability Class D; NIGHTTIME X (km) 0.1 0.3 0.5 0.8 1.0 2.0 3.0 3.5 7.0 10.0 15.0 20.0 30.0 CT (m) 8.1 22.0 36.0 55.0 67.0 130 188 215 400 550 790 990 1,430 az (m) 4.6 12.0 17.5 26.5 31.5 49.0 64.0 71.0 150 150 150 150 150 X 3 (fibers/m ) OM 2.71 x 10} 2.85 x 107 1.44 x 101 6.80 4.80 1.64 , 8.74 x 10 ~t 6.90 x 10 ~t 3.14 x 10 ~t 2.29 x 10 ~t 1.59 x 10 ~t 1.27 x 10, 8.79 x 10"z EM 9.61 x lo5 1.01 x loi 5.10 x 10? 2.41 x 10? 1.70 x 10? 5.81 x 10? 3.10 x 10? 2.45 x 10? 1.11 x 10? 8.12 x 10? 5.64 x 10? 4.50 x 10? 3.12 x 10^ 119 ------- Stability Class E; NIGHTTIME X (km) 0.1 0.3 0.5 0.8 1.0 2.0 3.0 5.0 8.0 16.0 20.0 25.0 30.0 a y (m) 6.0 16.5 26.5 40.0 50.0 95.0 140 220 335 620 750 920 1,070 a z (m) 3.5 8.8 13.0 18.0 21.3 33.5 43.0 55.0 71.0 150 150 150 150 X 3 (fibers/m ) OM 8.46 T 3.83 x I0t 2.30 x lo! 1.27 x 101 8.68 3.20 1.72 -, 8.65 x 10~t 4.43 x 10"| 2.03 x 10"t 1.68 x 10"r 1.37 x 10 i 1.18 x 10"1 EM 3.00 x 10$ 1.36 x 10£ 8.15 x 10? 4.50 x 10? 3.08 x 10? 1.13 x 10? 6.10 x 10? 3.07 x 10c 1.57 x 10? 7.20 x 107 5.95 x 10? 4.86 x 10? 4.18 x 10^ 120 ------- Calculation of Concentrations for Various Distances JOHNS-MANVILLE CORPORATION MILL NO. 5 ASBESTOS, QUEBEC, CANADA 8 Mean Wind Speed: 3.89 m/sec Source Term -- Based on Optical Microscopy: 7.63 x 10 fibers/sec Based on Electron Microscopy: 3.05 x 10 fibers/sec Source Height: 40 m Daytime Mixing Height: 800 m Nighttime Mixing Height: 150 m 12 Stability Class B: DAYTIME X (km) 0.1 0.3 0.5 0.8 1.0 2.0 3.0 3.1 6.2 7.0 10.0 15.0 20.0 30.0 (m) 19.0 52.0 83.0 130 158 290 420 440 800 880 1,200 1,700 2,150 3,050 °z (m) 11.0 31.0 51.0 85.0 109 233 360 370 800 800 800 800 800 800 x 3 (fibers/nr) OM 3.97 x 10? 1.68 x 10? 1.08 x 10.J 5.05 x 10:? 3.39 x 10X 9.10 x 10, 4.10 x 10, 3.82 x 10, 1.73 x 10, 1.57 x 10, 1.15 x lOf 8.14 x 10| 6.44 x 107 4.53 x 101 EM 1.58 x 10? 6.72 x 104 4.32 x 10^ 2.02 x 104 1.36 x 10; 3.64 x 10? 1.64 x 10? 1.53 x 10c 6.92 x 10? 6.28 x 10? 4.60 x 10c 3.25 x 10? 2.57 x 10c 1.81 x 10D 121 ------- Stability Class C: DAYTIME X (km) 0.1 0.3 0.5 0.8 1.0 2.0 3.0 5.0 6.5 7.3 14.6 17.0 20.0 25.0 30.0 ay (m) 12.5 34.0 55.0 86.0 105 200 285 450 570 630 1,150 1,330 1,520 1,850 2,170 CTZ (m) 7.5 20.5 30.2 49.0 61.0 115 170 265 340 380 800 800 800 800 800 X 3 (fibers/in ) OM 4.52 x 1071 1.33 x 107 1.57 x 10? 1.06 x 10o 7.86 x 10. 2.55 x 10- 1.25 x 10, 5.18 x 10, 3.20 x 10, 2.59 x 10, 1.20 x 10, 1.04 x 107 9.10 x lOT 7.48 x lot 6.37 x 101 EM 1.81 x 10? 5.31 x 107 6.28 x 107 4.24 x 107 3.14 x 107 1.02 x 10; 5.00 x 10? 2.07 x 10? 1.28 x 10^ 1.04 x 10c 4.80 x 10c 4.16: x 10? 3.64 x 10r 2.99 x 10? 2.55 x 103 Stability Class D: NIGHTTIME X (km) 0.1 0.3 0.5 0.8 1.0 2.0 3.0 3.5 7.0 10.0 15.0 20.0 30.0 a (m) 8.1 22.0 36.0 55.0 67.0 130 188 215 400 550 790 990 1,430 (m2) 4.6 12.0 17.5 26.5 31.5 49.0 64.0 71.0 150 150 150 150 150 X 3 (fibers/in ) OM 6.15 x lo:11 9.25 x 10o 7.20 x 10, 1.37 x 10? 1.32 x 10? 7.03 x 10n 4.27 x 10- 3.49 x 10^ 1.84 X 10q 1.34 x 10, 9.34 x 10, 7.45 x 10, 5.16 x 10 EM 2.46 x 10ft7 3.70 x 10? 2.88 x 107 5.48 x 107 5.28 x 107 2.81 x 107 1.71 x 107 1.40 x 10ft 7.36 x 10? 5.36 x 10? 3.73 x 10? 2.98 x 10? 2.06 x 10° 122 ------- Stability Class E: NIGHTTIME (km) 0.1 0.3 0.5 0.8 1.0 2.0 3.0 5.0 8.0 16.0 20.0 25.0 30.0 a (m) 6.0 16.5 26.5 40.0 50.0 95.0 140 220 335 620 750 920 1,070 a z (m) 3.5 X 3 (fibers/in ) OM 1.79 x io;22 8.8 1.38 x 10o 13.0 18.0 21.3 33.5 43.0 55.0 71.0 150 150 150 150 1.58 x 10- 7.38 x lOf 1.00 x 10o 7.67 x 10^ 6.73 x 10o 3.96 x 10^ 2.24 x 10^ 1.19 x 10« 9.8 x 10« 8.02 x 10o 6.89 x 10^ EM 7.16 x 10"i9 5.52 x 10^ 6.32 x 10$ 2.95 x I0i 4.00 x 10^ 3.87 x 10^ 2.69 x 104 1.58 x 10^ 8.95 x 10? 4.76 x 105 3.93 x 10^ 3.21 x 10? 2.75 x 10b 123 ------- Calculations of Concentrations for Various Distances GAF, INCORPORATED ASBESTOS MILL EDEN MILLS, VERMONT Mean Wind Speed: 3.89 m/sec Source Term -- Based on Optical Microscopy: 6.40 x 10 fibers/sec Based on Electron Microscopy: 1.83 x 10 fibers/sec Source Height: 20 m Daytime Mixing Height: 800 m Nighttime Mixing Height: 150 m 10 Stability Class B: DAYTIME X (km) 0.1 0.3 0.5 0.8 1.0 2.0 3.0 3.1 6.8 7.0 10.0 15.0 20.0 30.0 a y (m) 19.0 52.0 83.0 130 158 290 420 440 800 880 1,200 1,700 2,150 3,050 a z (m) 11.0 31.0 51.0 85.0 109 233 360 370 800 800 800 800 800 800 X 3 (fibers/ni ) OM 4.79 x 10? 2.64 x 10« 1.15 x lOf 4.61 x lOt 2.99 x 101 7.72 3.46 3.21 1.45 1.32 , 9.67 x 10~r 6.83 x 10"t 5.40 x 10" t 3.80 x 10"1 EM 1.32 x 10? 7.55 x 10c 3.29 x 10c 1.32 x 10? 8.55 x 10? 2.21 x 10.J 9.89 x 10^ 9.18 x 10- 4.15 x 10^ 3.77 x 10^ 2.77 x 10o 1.95 x ID. 1.54 x 10o 1.09 x 10J 124 ------- Stability Class C; DAYTIME X (km) 0.1 0.3 0.5 0.8 1.0 2.0 3.0 5.0 6.5 7.3 14.6 17.0 20.0 25.0 30.0 a (m) 12.5 34.0 55.0 86.0 105 200 285 450 570 630 1,150 1,330 1,520 1,850 2,170 a z (m) 7.5 20.5 30.2 49.0 61.0 115 170 265 340 380 800 800 800 800 800 X 3 (fibers/m ) OM 1.58 x 10? 4.67 x 10, 2.53 x 10, 1.14 x 107 7.55 x 107 2.24 x 10J 1.07 x 101 4.38 2.70 2.19 1.01 , 8.72 x 10~r 7.63 x 10 ~t 6.27 x 10 i 5.35 x 10"1 EM 4.52 x 10? 1.34 x 10c 7.23 x 10c 3.26 x 10c 2.22 x 10? 6.40 x 10? 3.06 x 10? 1.25 x 10? 7.72 x 10- 6.26 x 10o 2.89 x 10- 2.49 x 10. 2.18 x 10o 1.79 x 10o 1.53 x 10J Stability Class D: NIGHTTIME (km) 0.1 0.3 0.5 0.8 1.0 2.0 3.0 3.5 7.0 10.0 15.0 20.0 30.0 a (m) 8.1 22.0 36.0 55.0 67.0 130 188 215 400 550 790 990 1,430 az (m) 4.6 12.0 17.5 26.5 31.5 49.0 64.0 71.0 150 150 150 150 150 X 3 (fibers/inj) OM 1.09 , 4.92 x 10, 4.34 x 10, 2.70 x 10, 2.03 x 107 7.56 x 107 4.14 x lot 3.30 x 10t 1.55 x 107 1.13 x 101 7.83 6.25 4.33 EM 3.12 x 10? 1.41 x 10£ 1.26 x 10c 7.72 x 10? 5.80 x 10c 2.16 x 10^ 1.18 x 10? 9.49 x 107 4.43 x 10? 3.23 x 10? 2.24 x 107 1.79 x 10? 1.24 x 10^ 125 ------- Stability Class E: NIGHTTIME X (km) 0.1 0.3 0.5 0.8 1.0 2.0 3.0 5.0 8.0 16.0 20.0 25.0 30.0 a (m5) 6.0 16.5 26.5 40.0 50.0 95.0 140 220 335 620 750 920 1,070 a z (m) 3.5 X 3 (fibers /m ) OM 2.07 x 10:3 8.8 2.74 x 10o 13.0 18.0 21.3 33.5 43.0 55.0 71.0 150 150 150 150 4.65 x 10, 3.93 x 10- 3.16 x 10^ 1.38 x lOf 7.81 x 10,1 4.04 x lOt 2.12 x KT 9.98 8.25 6.73 5.78 EM 5.91 c 7.83 x 10? 1.33 x 10? 1.12 x 10? 9.04 x 10c 3.95 x 10^ 2.23 x 10c 1.16 x 10? 6.06 x 10? 2.85 x 10? 2.36 x 10? 1.92 x 10? 1.65 x 10^ 126 ------- TECHNICAL REPORT DATA (Please read Instructions on the reverse before completing) 1. REPORT NO. EPA-65f)./2-74-Q88 2. 3. RECIPIENT'S ACCESSION-NO. 4. TITLE AND SUBTITLE Assessment of Particle Control Technology for Enclosed Asbestos Sources 6. REPORT DATE October 1974 6. PERFORMING ORGANIZATION CODE Colin F Thomas P. Blaszak PaulSiebert, and 8. PERFORMING ORGANIZATION REPORT NO C6291-11 i PERFORMING ORG '\NIZATION NAME AND ADDRESS IIT Research Institute 10 West 35th Street Chicago, Illinois 60616 10. PROGRAM ELEMENT NO. 1AB015; ROAP 21AFA-006 11. CONTRACT/GRANT NO. 68-02-1353 12. SPONSORING AGENCY NAME AND ADDRESS EPA, Office of Research and Development NERC-RTP, Control Systems Laboratory Research Triangle Park, NC 27711 13. TYPE OF REPORT AND PERIOD COVERED Final; 6/73-5/74 14. SPONSORING AGENCY CODE 15. SUPPLEMENTARY NOTES B. ABSTRACT -phc report gives results of a study to provide information, from both the literature and user contact, on the control of asbestos emissions from enclosed sou- rces. It assesses the state-of-the-art in asbestos emission control in terms of the devices or methods used and their efficiency. In addition, it gives results of a pre- liminary study to actually measure the effectiveness of baghouse control devices in controlling emissions from five asbestos plants. Baghouses are the predominant con- trol device used in the asbestos industry. Cotton bags are used most frequently. Automatic shaking is used in most baghouses, with shake cycles of 1-1/2 to 4 hours most common. Most baghouses operate at two pressure drop ranges, 2. 5-5 and 7. 5-10 cm H2O. Air-to-cloth ratios range from 2 to 10:1. Published data on the re- moval efficiencies of the control devices was either non-existent, or quoted in gen- eral terms. Five baghouses were tested for removal efficiency in terms of mass and fiber number: although mass efficiency was very high, fiber concentrations exceed- ing 100 million fibers/cu meter, greater than about 0.05 urn long, are emitted. Using computer modeling, it was found that, even considering one source, asbestos con- centrations of 500 f/cu meter can be anticipated 5 km from the source. The exposure level at which asbestos in ambient air becomes a health hazard is not known. KEY WORDS AND DOCUMENT ANALYSIS DESCRIPTORS b.lDENTIFIERS/OPEN ENDED TERMS c. COSATI Field/Group Air Pollution Asbestos Fibers Oust Dust Collectors Measurement Cyclone Separators Scrubbers Efficiency Air Pollution Control Stationary Sources Enclosed Sources Particulate Baghouses Fabric Filters 13B , 07A 08G HE 11G 13A 14B 8- DISTRIBUTION STATEMENT Unlimited 19. SECURITY CLASS (ThisReport) Unclassified 21. NO. OF PAGES 135 20. SECURITY CLASS (This page) [Unclassified 22. PRICE PA Form 2220-1 (9-73) 127 ------- |