xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 22 Ka 4500
controlfield tag 007 cr-bnu---uuuuu
008 s2010 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0003455
Filtration performance of a niosh-approved n95 filtering facepiece respirator with stapled head straps
h [electronic resource] /
by Daniel Medina.
[Tampa, Fla] :
b University of South Florida,
Title from PDF of title page.
Document formatted into pages; contains X pages.
Thesis (M.S.P.H.)--University of South Florida, 2010.
Includes bibliographical references.
Text (Electronic thesis) in PDF format.
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
ABSTRACT: Certain models of NIOSH-approved filtering facepiece air purifying respirators are manufactured with stapled head straps. Depending on the manufacturer, these head straps may be stapled to the filter media itself. This may cause leakage through the filter media of the respirator, potentially exposing the user to an unacceptable level of contaminant. In this study, monodisperse polystyrene latex (PSL) spheres were generated to challenge four replicates of a N95 single use respirator model made by the same manufacturer. Nominal particle sizes of the PSL spheres used to challenge the respirators were 0.5, 1, and 2 micrometers (μm) in diameter. All respirators were sealed onto a custom built testing assembly and tested in a sealed chamber. Particle sizes of interest were generated using a nebulizer, and passed through a diffusion dryer and a Krypton-85 radioactive source prior to entering the test chamber. The dryer reduces the humidity of the aerosol generating by the nebulizer, while the radioactive source neutralizes the charge of the aerosol cloud. The test chamber was constructed using a glass aquarium measuring 32 x 53 x 122 centimeters. Three stainless steel air diffusers were placed above the testing compartment to evenly distribute the aerosol in the chamber. An exhaust manifold was placed at the lower part of the chamber beneath another stainless steel diffuser below the area where test respirators were placed. The respirators were challenged as received from the manufacturer with 0.5, 1 and 2 μm sized (PSL) spheres. The same procedure was repeated for each respirator after sealing the areas where the head straps were stapled with silicon rubber. Testing was conducted at a flow rate of 85 liters per minute, as specified in the NIOSH respirator testing protocol. A laser particle counter was used to measure the concentration inside and outside of the respirator. The results showed unsealed efficiencies for particle sizes 0.5, 1, and 2 μm of 96.68%, 99.72%, 99.88% and sealed efficiencies of 97.35%, 99.82%, 99.93% respectively. There were no differences for particle size or sealing at 1.0 and 2.0 μm. A significant drop in efficiency was observed when testing with 0.5 μm PSL spheres. The drops in efficiency are not sufficient to reduce the integrity of the respirator for N95 certification. However, the leakages detected will have a cumulative effect when added to other sources of single use respirator leakage in the field.
Advisor: Yehia Y. Hammad, Sc.D.
x Environmental and Occupational Health
t USF Electronic Theses and Dissertations.
Filtration Performance of a NIOS H-Approved N95 Filtering Facepiece Respirator With Stapled Head Straps by Daniel E. Medina A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in Public Health Department of Environmental and Occupational Health College of Public Health University of South Florida Major Professor: Yehia Y. Hammad, Sc.D. Thomas E. Bernard, Ph.D. Steven P. Mlynarek, Ph.D. Date of Approval: December 11, 2009 Keywords: Single Use, Filter, Efficiency, Polystyrene Latex, Aerosol Copyright 2010, Daniel E. Medina
Acknowledgments First of all, I would like to thank NIOSH (T42-OH008438) for supporting future Industrial Hygienists by offering ERC scholarships and stipends. Without this economic help I would not be able to pursue a MSPH in Industrial Hygiene. I would like to thank my major professor and advisor, Dr. Yehia Y. Hammad, as well as my thesis committee, Dr. Steve Mlynarek, and Dr. Tom Bernard for their dedication to teaching and commitment towards their students. This project would not have been a success without their direction. I thank my parents for supporting my decision to become an Industrial Hygienist and their ongoing encouragement in times of difficulty. Their advice is an invaluable part of my life and has contributed greatly to the completion of the program. Lastly I thank the faculty and staff of the College of Public Health at the University of South Florida.
i Table of Contents List of Tables ii List of Figures iii List of Abbreviations v Abstract vi Introduction 1 Purpose of Study 1 Hypothesis 1 Literature Review 3 History of Air Filters 3 Mechanisms of Fibrous Filtration 5 NIOSH Methods for Certifying Respirators 10 Polystyrene Latex Spheres 13 Experimental Methods 15 Experimental Design 15 Steps for Sampling 26 Results & Statistical Analysis 27 Discussion 33 Conclusions & Recommendations for Future Work 35 References 37 Appendices 41 Appendix A: Equipment List 42 Appendix B: Relative Humidity and Temperature Data 45 Appendix C: Number Concentration Data 46 Appendix D: Prototype Respirator Assembly Data 54
ii List of Tables Table 1 Polystyrene Latex Sphere Characteristics 16 Table 2 Overall Average Collection Efficiency for Respirators Tested 27 Table 3 Collection Efficiencies for Respirators Tested by Trial 29
iii List of Figures Figure 1. Experimental Design 16 Figure 2. Nebulizer Components 17 Figure 3. Diffusion Dryer 17 Figure 4. Hygrometer 18 Figure 5. Temperature and Relative Humidity Inside Test Chamber (n = 42) 19 Figure 6. Krypton-85 Charge Neutralizer 20 Figure 7. Aerosol Generation Train 20 Figure 8. Aerosol Generation Train and Test Chamber 21 Figure 9. Test Chamber Components 22 Figure 10. Test Chamber 22 Figure 11. Prototype Testing Assembly 23 Figure 12. Testing Assembly Used in Study 24 Figure 13. Testing Assembly with Staples Unsealed 25 Figure 14. Close-up Inside the Testing Assembly 25 Figure 15. Respirator with Staples Sealed 26 Figure 16. Overall Average Collection Efficiency for Respirators Tested (n = 20) 28 Figure 17. Overall Collection Efficiency for Respirator 1 (n = 5) 30 Figure 18. Overall Collection Efficiency for Respirator 2 (n = 5) 30
iv Figure 19. Overall Collection Efficiency for Respirator 3 (n = 5) 31 Figure 20. Overall Collection Efficiency for Respirator 4 (n = 5) 31
v List of Abbreviations C Packing density CC Cunningham slip factor D Particle diffusion coefficient df Diameter of fiber dp Diameter of the particle E Filter efficiency ED Efficiency of diffusion f Dielectric constant of the fiber EF Fiber efficiency EG Efficiency of gravity EI Efficiency of impaction EQ Efficiency of electrostatic attraction ER Efficiency of interception J Dimensionless parameter used to calculate impaction efficiency KU Kuwabara hydrodynamic factor L Fiber length Nin Number concentration inside the filter Nout Number concentration outside the filter P Penetration Pe Peclet number PSL Polystyrene latex q Particle charge R Fiber radius r Dimensionless parameter (dp/df) STK Stokes number t Time Vin Velocity inside the filter Vout Face velocity outside the filter V0 Particle velocity VTS Particle terminal settling velocity y Distance from the center axis of the fiber cross section 0 Dielectric constant of a vacuum Viscosity p Density of particle
vi Filtration Performance of a NIOSH Approved N95 Filtering Facepiece Respirators With Stapled Head Straps Daniel E. Medina ABSTRACT Certain models of NIOSH-approved filte ring facepiece air purifying respirators are manufactured with stapled head straps. Depending on the manufacturer, these head straps may be stapled to the filter media itself. This may cause leakage through the filter media of the respirator, potentially exposing the user to an unacceptable level of contaminant. In this study, monodisperse polystyrene latex (PSL) spheres were generated to challenge four replicates of a N95 single use respirator model made by the same manufacturer. Nominal particle sizes of the PSL spheres used to challenge the respirators were 0.5, 1, and 2 micrometers ( m) in diameter. All respirators were sealed onto a custom built testing assembly and tested in a sealed chamber. Particle sizes of interest were generated using a nebulizer, and passed through a diffusion dryer and a Krypton-85 radioactive source prior to entering the test chamber. The dryer reduces the humidity of the aerosol generating by the nebulizer, while the radioactive source neutralizes the charge of the aerosol cloud. The test chamber was constructed using a glass aquarium measuring 32 x 53 x 122 centimeters. Three stainless steel air diffusers were placed
vii above the testing compartment to evenly distribute the aerosol in the chamber. An exhaust manifold was placed at the lower part of the chamber beneath another stainless steel diffuser below the area where test respirators were placed. The respirators were challenged as received from the manufacturer with 0.5, 1 and 2 m sized (PSL) spheres. The same procedure was repeated for each respirator after sealing the areas where the head straps were stapled with silicon rubber. Testing was conducted at a flow rate of 85 liters per minute, as specified in the NIOSH respirator testing protocol. A laser particle counter was used to measure the concentration inside and outside of the respirator. The results showed unsealed efficiencies for particle sizes 0.5, 1, and 2 m of 96.68%, 99.72%, 99.88% and sealed effi ciencies of 97.35%, 99.82%, 99.93% respectively. There were no differences for particle size or sealing at 1.0 and 2.0 m. A significant drop in efficiency was observed when testing with 0.5 m PSL spheres. The drops in efficiency are not sufficient to reduce the integrity of the respirator for N95 certification. However, the leakages detected will have a cumulative effect when added to other sources of single use respirator leakage in the field.
1 Introduction Purpose of Study There are many models of N95 filtering facepiece respirators found on the market today that have the headstraps stapled onto the filter medium of the respirator. This has the potential to create leakage because the staple punctures the filter itself. When the head-straps are stretched by the wearer to put the respirator on, the holes created by the staples may enlarge and allow for more penetration into the respirator. This study compares the efficiency of a N95 single us e, filtering facepiece respirator model with stapled headstraps before and after sealing the stapled areas of the filter with silicon rubber. It is important to know if staple penetrations on the filter media allow leakage, and if it is great enough to reduce the efficiency of the filter below the requirment by NIOSH for certification. In the study, the effect of stretching the head straps to simulate putting on the respirator will not be considered and the focus will be on the leakage through the staple openings on the respirator as received from the manufacturer. Hypothesis There are three null hypotheses to be tested in this study. First, that there is no effect on the respirator performance due to the staples being sealed or unsealed. Second, that there is no effect on the respirator performance when tested with different
2 polystyrene latex (PSL) spheres of 0.5, 1, and 2 m diameters. The third null hypothesis is that there are no differences in respirator performance among respirators that were selected at random from the same box. The objectives to test the hypothesis are to create respirator testing conditions simulating NIOSH protocol and test the effi cienciy of each respirator with different sized, mono-disperse challenge aerosol clouds before and after sealing the area around the staples with silicon rubber.
3 Literature Review History of Air Filters The use of air filters has been noted for over two thousand years (Davies, 1973). Different adaptations and approaches have been made to protect workers in the mining industry and over dust generating processes throughout this time. The most common of these has been the use of cloths over the mouth and nose of workers in dusty environments to prevent lung disease. The Egyptians and Romans have written about the negative effects of dust in mining as far back as the first century A.D. and different literature has been written on the importance of adequate ventilation in dusty environments. The first recorded filtering respirator was invented by Bris Fradin in 1814 (Davies 1973). It was a simple box filled with cotton and a breathing tube that went into a persons mouth and the nostrils were covered with cotton. The first U.S. patent for a respirator was granted is 1849 to Lewis P. Hasslett. This patent worked on the same principles as Fradins wool box. In 1854, charcoal began to be used as an adsorbent of gases. John Tyndall took this concept and combined cotton, wool, lime, charcoal, and glycerin to create the firemans respirator in 1868. In the 1900s, great advances were made in respirator design. World War I created a huge drive to improve the current technology of respirators. Arsenic smoke was a weapon of choice which could not be pr operly filtered by charcoal. Generation of
4 these smokes using thermal energy created an aerosol of particles between 0.2 and 0.5 m which could easily penetrate fibrous filters. Unfortunately, respirators were not commonly used in the workplace in the early 1900s and many deaths and illnesses were recorded in the mining industry as well as in the construction industry. These workers were exposed to high concentrations of dust with little engineering, administrative or personal protective controls to prevent their exposures. One of the biggest breakthroughs in filtering technology occurred in the 1930s. The development of the Hansen filter by the Danish scientist, Nicolaij Hansen, incorporated the use of electrostatic charge to increase filtration efficiency. The filter was made of wool fibers sprinkled with particles of resin that were 1 m in diameter. When the filter is carded, or rubbed, friction between the resin and wool causes the filter to acquire an electrostatic charge. This dramatically increased the efficiency of an otherwise poor filter with large diameter wool fibers. Although the U.S. military adopted this technology for their equipment, the use of this technology was not seen in industry for a substantial amount of time. Todays dust filters are constructed of a variety of materials including wool, synthetic fibers, glass and cellulose. There are several different types of filters that depend on a variety of mechanisms of deposition. Fibrous filters, which are the main type of filters used for single use respirators, are made up of different sized fibers, varying in length and diameter, which are oriented to be perpendicular to the airflow when in use. Fibrous filters are by far the most efficient types of filters when dealing with low concentrations. They require a low pressure drop in comparison to other filtering techniques making it ideal for respirator use. To further understand the process
5 of filtration, the theories of mechanical filtr ation and evaluation of filter efficiency will be reviewed. Mechanisms of Fibrous Filtration The general equation for the penetration of particulates (P) of a fibrous filter is defined as the fraction of particles that penetrate the filter to the number of particles at the face of the filter P = Nout / Nin The amount of particulates which penetrate the respirator filter (Nin) and the amount of particulates that exists outside the respirator (Nout) can be either interpreted as a number count, number concentration, or a mass concentration. The filter efficiency (E) can be determined using the following equation: E = (Nout Nin) / Nout The efficiency of filters can also be determined theoretically through five different models that describe the forms of particle deposition onto the filter medium. The packing density (C) of the fibers in a fibrous filter is the ratio of the volume of all the fibers in a filter to the volume of the filter itself. The packing density is determined with the length of the fibers (L) and the radius of the fibers (R) used to construct the filter. C = R2L The packing density can also be determined for filters that are made up of multiple fibers that differ in length and radius.
6 C = iRi 2Li The velocity inside the filter (Vin) can be determined if the face velocity of the gas approaching the filter (Vout) is known. Vin = Vout /(1 C) Although all the fibers in a filter have a combined effect on particle collection, theoretical calculations of efficiency are based on the single fiber theory. Single fiber theory looks at the efficiency of each mechanism of deposition on a single fiber to estimate the total efficiency of the filter. The efficiency of a single fiber (EF) can be calculated using the distance from the center axis of the fiber cross section (y) and radius of the fiber (R). EF = y/R Assuming that all fibers in the filter are the same length and diameter, the theoretical penetration of the filter can be calculated using the packing density (C), the single fiber efficiency (EF), time in seconds (t), and the radius of the fiber (R). P = exp[-4CEFt / 2 R] As stated above, fibrous filters are the main type of filter used in single use respirators. Collection of particulates into the filter media operates based on 5 mechanisms of collection. These mechanisms can be applied to a single fiber to predict the single fiber efficiency due to each specific mechanism. The mechanisms of deposition are inertial impaction, interception, diffusion, electrostatic attraction and gravitational settling. Inertial impaction, also known as inertial deposition, occurs when a particle following a streamline near the fiber is unable to continue on the streamline and, due to the inertia of the particle, hits the fiber and deposits. The efficiency of impaction (EI) is
7 dependent on Stokes number, the Kuwabara hydrodynamic factor (Ku), and the parameter J which is determined using the packing density and the dimensionless parameter (r). The parameter, r, is defined as the ratio between the diameter of the particle (dp) and the diameter of the fiber (df). The efficiency of impaction can be calculated as follows: EI = (Stk)J / 2Ku2 Where: J = (29.6 28C0.62)r2 27.5r2.8; for r < 0.4, J = 2; for r > 0.4, Ku = lnC/2 3/4 + C C2/4, Stk = pdp 2CCV0 / 18 df. For Stokes number (STK), p is the density of the particle, dp is the diameter of the particle, CC is the Cunningham correction factor, V0 is the particle velocity, is the viscosity of the air, and df is the diameter of the fiber. The Kuwabara factor describes the effect of fibers in proximity of each other. Inertial impaction is primarily responsible for deposition of particles larger than the most penetrating particle size. Interception occurs when a particle following a streamline is one particle radius away from the fiber. Interception assumes that the particle is not deviated from the streamline in any way. When particles are deposited by interception, the porosity of the filter is decreased and the efficiency of the filter is increased as well. Interception is the only mechanism of deposition that is not dependa nt of the face velocity of the flow. The single fiber efficiency of interception (ER) can be estimated according to: ER = (1 C)r2 / Ku(1 + r)
8 Deposition due to diffusion occurs with particles that are not necessarily on the streamline trajectory nearest to the fiber. Most particles that are affected by Brownian motion are very fine and are below 1 m in diameter. The efficiency due to diffusion is: ED = 2Pe-2/3 Where the Peclet number (Pe) is equal to dfVout / D, and D is the particle diffusion coefficient. Diffusion is the main mechanism of deposition for particles smaller than the most penetrating particle size. The single fi ber efficiency due to diffusion increases as the diameter of the particle decreases. Electrostatic attraction is one of the most important mechanisms of deposition. Increasing the charge on the fibers or particles, as well as reducing face velocity, increases the single fiber efficiency greatl y. Charged or neutral particles can both be attracted to a fiber with an electrostatic charge. The efficiency of the electrostatic attraction (EQ), based on studies with glass fibers, can be estimated by: EQ = 1.5[( f 1)/( f + 1) q2/12 2 V00dpdf 2] Where q is the charge of the particle, 0 is the dielectric constant of vacuum, and f is the dielectric constant of the fiber. Charged fibers can greatly increase the efficiency of the filter without causing a large pressure drop. These characteristics are very useful in filters that are used as respirators. For this reason most single use dust respirators rely on electrostatic charge to maintain a certain efficiency mandated by government regulations. They are typically made out of polypropylene and share some of the qualities of the resin and wool filter developed by Hansen mentioned earlier. Collection by gravitational settling (EG) is represented by G = VTS/V0. Where VTS is the terminal velocity of the particle and V0 is the initial velocity of the particle. This
9 type of mechanism accounts for very little deposition when compared to the other mechanisms of deposition. When the particles are large and the face velocity is less than 10 cm/s, the efficiency of gravitational settling may be comparable to other mechanisms of deposition. The efficiency due to gravitational settling depends on which way the air is flowing. It can be calculated as follows when V0 and VTS are both directed downwards: EG ~ G(1 + r) When V0 is directed upwards, or against gravitational pull, the equation is negative. As stated before, the overall mechanical efficiency of the filter itself can be estimated theoretically by adding all of the single fiber efficiencies for each of the mechanisms of deposition except electrostatic attraction. Although it is a mechanism of deposition, electrostatic attraction is not considered a mechanical mechanism as the other models discussed. However adding all the single fiber efficiencies does not take into consideration that the different mechanisms of deposition discussed happen simultaneously and a certain particle may fall into the range of deposition by more than one of the mechanisms. This may result in overestimating the collection efficiency of the filter. For the purposes of this study, the collection efficiency of the filter is determined by measuring the number concentration inside and outside of the respirator using a laser particle counter. This is a method that accommodates for the collective effects of the mechanisms descreibed above.
10 NIOSH Methods for Certifying Respirators The single use respirators of interest in this study are certified by the National Institute for Occupational Safety and Health (NIOSH). Title 30 CFR Part 11 was established in 1972 and required air purifying respirators to be certified by NIOSH and the Mine Safety and Health Administration (MSHA). The tests used to certify respirators were very similar to certification tests developed by the USBOM, dating back to the 1930s. The most recent publication by NIOSH to certify filters is Title 42 CFR Part 84. It is based on Title 30 CFR Part 11 but changes the specifications for air purifying respirators. The code of regulations was updated because of recent improvements in filter technology, specifically the use of electrostatic charge on filter mediums. Title 42 Part 84 was established in 1995 and allowed manufacturers of single use filtering facepiece respirators to sell respirators approved under Title 30 CFR Part 11 until 1998. The new code of regulations also takes into consideration filter efficiency degradation that can occur when oils or liquids are deposited on the fibers of the filter. It has been shown that certain oils can neutralize the electrostatic charge of the filters, thus reducing its collection efficiency (Yang & Lee, 2005). As a result, NIOSH created 3 different classes of air purifying particulate respirators: N, R, and P classes. The N series is meant for non-oil solid aerosols, while the R and P series are used for oil, liquid or solid aerosols. The R series is resistant to oil while the P series is deemed oil-proof. The difference between the R and P series is the amount of time they can be using in oily environments (3M, 1995). Each one of these
11 classes is also subdivided into three subcategories according to their filtration efficiency; in which respirators are labeled as the class followed by the rating. The number to the right of the letter class is representative of the percentage efficiency 95%, 99%, and 99.97%. NIOSH tests N filters with charge neutralized sodium chloride particles with a count median diameter of 0.075 0.02 m and a standard deviation of 1.86. The other two categories of filters, R and P, are tested using charge neutralized dioctyl phthalate (DOP) with a count median diameter of 0.185 0.02 m and a standard deviation of 1.60. All filter series filters are tested to a maximum loading of 200 mg and are required to maintain their designated efficiency to pass the certification. There is no time restriction to the use of N series filters. NIOSH recommends that R filters exposed to oil aerosols be dis posed of after 8 hours or after the filter has collected 200 mg of oil. P series filters are tested until the efficiency of the filter is no longer decreasing. NIOSH requires the manufacturer of P series filters to state a time use limitation. There is no time restriction recommendation on any of the filter classes if they are not exposed to oil aerosols. All filters that pass the NIOSH Title 42 CFR Part 84 tests are required to have NIOSH, the filter classification, and the manufacturer, and an approval number printed on them. Filters that were approved under Title 30 CFR Part 11 have both NIOSH and MSHA on the filter. It is important to note that not all single use dust masks sold for industry and home use are approved by NIOSH and do not necessarily provide the specified filtration efficiency required for NIOSH certification. Du st masks and surgical masks that are not certified under Title 30 CFR Part 11, may be certified by the Food and Drug
12 Administration (FDA) but can not be used wh ere respirators are required. The FDA does not conduct any testing of filters, but reviews test data supplied by the manufacturer. Surgical masks that are approved by the FDA may be less than 70% efficient when tested according to NIOSH methods and should not be used with the intention of preventing inhalation of particulates. Filters that are both approved by the FDA and certified by NIOSH are referred to as surgical masks with the NIOSH rating, such as a N95 surgical mask. In recent studies, it has been found that electrostatic charge has an effect on defining the most penetrating particle size. For mechanical means of filtration, the theoretical most penetrating particle size is 0.3 m. When electrostatic attraction is present, there is a shift in the most penetration particle size towards smaller particles in diameter. This size has been reported to be in the 40 to 50 nanometer range (Balazy et al, 2006). As electrostatic filters loose their charge, the most penetrating particle size shift upwards back to the 0.3 m range. Martin & Moyer (2000) have conducted experiments to test NIOSH approved N95, R95, and P95 filters that have been dipped in isopropanol which eliminates the electrostatic charge of the filter. The results of the study showed that penetration increased 30 to 40% for particulates in the 0.1 to 0.4 m range. It is important to note that NIOSH uses a challenge aerosol with a count median diameter of 0.075 m when testing electrostatically charged respirators. Theoretically, 0.3 m is the most penetrating particle size for purly mechanical filters. It is difficult to determine the performance of single use dust respirators in the field due to the large variability of environmental conditions. The dependency of electrostatic charge as the predominant mechanism of filtration in NIOSH approved
13 filters is of concern because there is no indication as to when the filter media has lost sufficient charge to allow greater than 5% pe netration of inhalable particulates. Although it is unlikely for a filter to loose its electrostatic charge in an 8 hour work day, there are no time restraints on the use of N95 respirators, and there are no guidelines for proper storage if the respirator is going to be reused. Polystyrene Latex Spheres PSL spheres have been used as challenge aerosols in numerous studies and is a common methods used to create monodisperse aerosols. One advantage of using monodisperse aerosols versus polydisperse aeros ols is that monodisperse data are easier to interpret than polydisperse data. As suggested by studies using monodisperse aerosols as challenge aerosols, there are several steps taken to create a successful aerosol using PSL spheres. The most common way to create a monodisperse aerosol is to nebulize a liquid, containing PSL spheres of a singl e known size. Evaporation-condensation methods of aerosolizing monodisperse aerosols is a good method for creating high concentrations, but there is an upper limit on the size and the aerosol is not near monodispersity. A vibrating orifice monodisp erse aerosol generator can create a highly monodisperse aerosol but at the expense of low concentrations. Since the filters used in this study are electrostatically charged, humidity and particle charges are important to control in order to obtain valid data. Most other studies have used a diffusion dryer or heated dryer to reduce the amount of water droplets that will enter the chamber. The same applies to neutralizing the aerosol cloud. The aerosol
14 must also pass through a radioactive charge neutralizer to create equilibrium between the negatively and positively charged particles. Once the aerosol cloud is dry and its charge neutral, it may be introduced to the test chamber for efficiency testing. There are several problems that may occur when using PSL spheres. Eevaporation, the formation of aggregates in the aerosol cloud and impurities in the monodisperse suspension may all affect the accuracy of the laser particle counter.
15 Experimental Methods Experimental Design One model of a N95 filtering facepiece respirator was tested using monodisperse PSL spheres of known sizes. The respirator was sealed to a plexiglass plate assembly and tested in a chamber at a flow rate of 85 LPM. Figure 1 shows the experimental desing for measuring the efficiency of the respirators. Compressed industrial grade nitrogen was used to operate the nebulizer at the required pressure, 20 psi, and was passed through a 47 mm HEPA filter, Figure 2. This filter was used to eliminate any particulate contaminants from the nitrogen before entering the aerosol generator. An MRE 3-Jet Collison nebulizer (BGI, Inc., Waltham, MA.) operating at 20 PSI and producing a flow rate of 6 LPM, was used to create the monodisperse challenge aerosol. The three sizes of PSL used in the study are nominally 0.5, 1, and 2 m in diameter. Figure 2 also shows the Collison nebulizer used in this study and a bottle of the PSL spheres. Table 1 shows the characteristics of the PSL spheres suspensions used in this study.
1 2 3 4 5 6 10 11 12 13 14 15 9 56.7 LPM 6 LPM 28.3 LPM 9 LPM 7 8 Figure 1: Experimental Design. 1, nitrogen tank; 2, HEPA filter; 3, aerosol generato r; 4, diffusion dryer; 5, charge neutralizer; 6, chamber; 7, HEPA filter; 8, pressure gauge; 9, air d iffusers; 10, N95 respirator; 11, pump; 12, in and out control valves; 13, particle count er; 14, exhaust manifold; 15, pump Table 1: Polystyrene Latex Sphere Characteristics Nominal Size ( m) Actual Diameter ( m) Standard Deviation Percent Solids 0.5 0.465 0.011 2.62% 1 0.989 0.020 2.59% 2 1.826 0.046 2.70% 16
Figure 2: Nebulizer Components. Compressed nitrogen tank and 47 mm HEPA filter (Left), Collison nebulizer and polystyrene latex container (Right) Since water vapor may condense into water droplets and onto the PSL spheres, potentially affecting the performance of the particle counter, the aerosol must go through a diffusion dryer (Hamilton Associates Inc., Owings Mills, MD.). A diffusion dryer is a general purpose aerosol dryer that uses silica gel to diffuse water vapor from the aerosol. When the silica gel is saturated it changes from a yellowish color to a dark green color. The silica gel is in cartridges that can be removed from the diffusion dryer tube and placed in an oven for regeneration. The dryer used in this study and the silica gel cartridges can be seen in Figure 3. Figure 3: Diffusion Dryer. Diffusion dryer (Left), silica gel cartridge (Right) 17
To determine if the diffusion dryer was absorbing all of the humidity created by the nebulization process, a simulation of a typical test run was performed while measuring the relative humidity and temperature in the chamber. This procedure was performed in a session simulating the entire sampling procedure which takes place over approximately 3 hours to complete for each particle size. Figure 4 shows the hygrometer (Fisher Scientific, Tampa, FL) used in the study. The sensor probe was placed as close to the respirator as possible. Measurements were recorded every five minutes over a period of three and a half hours. Figure 5 shows that there are no significant changes in temperature and relative humidity, RH, over this time period. The RH maintained at about 50% and the temperature at 73 F. The average RH and temperature in the test chamber were 50.59% and 73.81F. The RH and temperature outside of the chamber were 50.82 % and 73.68 F. A table with the data from the RH and temperature run can be found in appendix B. Figure 4: Hygrometer. RH/temperature display (Left), and hygrometer sensor probe (Right) 18
45 50 55 60 65 70 75 80 0 50 100 150 200 250 Time (min)Relative Humidity (%) / Temperature (F) Relative Humidity TemperatureFigure 5: Temperature and Relative Humidity Inside Test Chamber (n = 42). To evaluate the efficiency of the tested respirators, it is necessary to neutralize the charge of the challenge aerosol cloud. This means achieving the same number of positively and negatively charged particles in the aerosol cloud. To do this, the aerosol is passed through a charge neutralizer (TSI, Inc ., St. Paul, MN). A charge neutralizer ionizes the surrounding atmosphere into positive and negative ions. As the aerosol enters the neutralizer, charged particles aquire ions of the opposite charge and discharge. This resulting aerosol cloud has a bipolar distribution known as the Boltzmann equilibrium. The encapsulated radioactive beta source used in this study is Krypton-85. Figure 6 shows the charge neutralizer The generation train, including the nebulizer, dryer, and charge neutralizer can be seen in Figure 7. Figure 8 includes the chamber. 19
Figure 6: Krypton-85 Charge Neutralizer. Figure 7: Aerosol Generation Train. Nebulizer, dryer, and charge neutralizer (Left to Right) 20
Figure 8: Aerosol Generation Train and Test Chamber. Nebulizer, dryer, charge neutralizer, and test chamber (Left to Right) The efficiency of the N95 filtering facepiece respirators were tested inside a custom built test chamber with a volume of 190 liters. The chamber measured 32 x 53 x 122 cm. Figure 9 shows a diagram with the front and side view of the test chamber, with its components labeled. After passing through the K-85 radioactive source, the aerosol enters through an opening at the top of the chamber and mixes with clean air. The clean air is introduced into the chamber via a HEPA filter sheet measuring 19 x 24 cm. The pressure inside the chamber was monitored by a Magnahelic pressure gauge (Dwyer Instrument, Inc., Michigan City, IN) to ensure that there is a proper seal and a negative pressure maintained at 0.65 inches of water in the chamber. A series of three diffusers are placed between the upper part of the chamber and the respirator testing assembly in order to evenly distribute the aerosol throughout the chamber. Figure 10 shows the test chamber constructed for the experiment. 21
Aerosol Intake Outside Air Intake Pressure Gauge Additional Exhaust from Respirator Concentration Inside Respirator Concentration Outside Respirator Exhaust Manifold Air Diffusers Testing Apparatus Respirator Aerosol Intake Outside Air Intake Pressure Gauge Additional Exhaust from Respirator Concentration Inside Respirator Concentration Outside Respirator Exhaust Manifold Air Diffusers Testing Assembly Respirator Figure 9: Test Chamber Components. Figure 10: Test Chamber. Front view (Left), HEPA filter and pressure gauge (Center), and rear view (Right) 22
The testing assembly consisted of a N95 filtering facepiece sealed to a plexiglass plate using silicon rubber (GE Sealants and Adhesives, Huntersville, NC). Two different testing assemblies were used to test the respirators. The first, was a flat plexiglass plate with two exhaust ports. Figure 11 shows prototype respirator testing assembly. This method was discarded because it was suspected that the proximity of the staple openings to the plexiglass plate was causing the particles penetrating through the staple openings to impact on the plexiglass plate. The data recorded for this testing assembly are not discussed in this study but can be found in appendix D. Figure 11: Prototype Testing Assembly. In test chamber (Left) and rear view (Right) To create the new respirator testing assembly, the prototype testing assembly was used to trace out the edges of a sealed respirator onto a new sheet of plexiglass. The area traced was cut so that when sealing a new respirator to the assembly, the filter edges of the respirator lay over the edge of the cut plexiglass. A plastic chamber with two exhaust ports was sealed to the back of the plexiglass plate in order to move the air through the 23
filter. This was done to diminish the wall losses of particles penetrating the staple openings associated with impaction onto the plexiglass plate of the prototype fixture. Figure 12 and Figure 13 show the respirator testing assembly used in this study. The staples and the area of the cut plexiglass can be seen in Figure 14. One of the exhaust ports was connected to a control valve which could be opened or closed to either sample inside or outside of the respirator. The control valves were connected to a laser particle counter (Particle Measuring Systems, Inc., Boulder, CO) that operates at a flow rate of 28.3 LPM. The other port was used to draw 56.7 LPM in order to have a total of 85 LPM through the filter as specified in the NIOSH testing protocol. At the lower end of the chamber, there was a manifold used to exhaust the air from the chamber. The manifold was connected to a pump that operated at 9 LPM. Thus the total air flow rate in the test chamber was 100 LPM. A mass flow meter (TSI, Inc., Shoreview, MN) was used to calibrate all the pumps and to ensure the specified flow rate of 6 LPM through the nebulizer. Figure 12: Testing Assembly Used in Study. In test chamber (Left) and side view (Right) 24
Figure 13: Testing Assembly with Staples Unsealed. Figure 14: Close-up Inside the Testing Assembly. 25
Steps for Sampling After sealing the respirator to the plexiglass and sealing the chamber, all pumps are run until the background levels of particulates stabilize. An average number concentration, over 30 minutes, is taken as a background concentration inside and then outside of the respirator. Once background levels are established, the aerosol is generated over a period of 30 minutes or until concentration levels are stabilized. Then a ten minute sample measuring the concentration inside and a ten minute sample measuring the concentration outside of the respirator we re taken. The inside/outside concentration sampling was repeated five times for a respirator with the staples intact, or unsealed. This procedure was performed for 0.5, 1, and 2 m PSL spheres. The entire process was repeated for the same respirator with its staples sealed as shown in Figure 15. This procedure was followed for four respirators selected at random from the same box. Figure 15: Respirator with Staples Sealed. 26
27 Results & Statistical Analysis The average efficiencies of the four respirators, at each particle size tested, increased after sealing the staples with silicon rubber. Figure 14 shows the overall average efficiencies (n=20) of the four respirators and five trial runs at each PSL sphere size tested. Table 2 also shows the standard deviations at each particle size. The greatest increase in efficiency was observed at 0.5 m while the lowest increase was at 2 m. The standard deviation increased with smaller particle sizes and was greater by a magnitude of ten at 0.5 m when compared to 1 and 2 m. Table 2: Overall Average Collection Efficiency for Respirators Tested. Unsealed Sealed Particle Size ( m) Efficiency (%) SD (%) Particle Size ( m) Efficiency (%) SD (%) 0.5 96.681 0.879 0.5 97.358 0.729 1 99.729 0.083 1 99.823 0.077 2 99.884 0.076 2 99.936 0.045 The average efficiencies (n = 5) of each individual respirator are showed in Figures 15 through 18. Table 3 shows the data collected for each trial, as well as the standand deviation and average efficiency at each particle size. Respirator 3 showed the greatest drop in efficiency for all particle sizes, while respirator 2 showed no drop in efficiency at any of the particle sizes tested. Repirators 1, 3, and 4 all shared similar trends in efficiencies. At the smallest particle size tested, the drops in efficiency were the greatest. At the largest particle size, the drops in efficiency were the smallest.
95.000 96.000 97.000 98.000 99.000 100.000 00 511 522 5 Particle Size ( m)Efficiency (%) Unsealed Sealed Figure 16: Overall Average Collection Efficiency for Respirators Tested (n = 20). Respirators 3 and 4 had the lowest unsealed efficiencies, at 0.5 m, of 95.90% and 95.99% respectively. The highest unsealed efficiency, at 0.5 m, was 97.70% in respirator 1. The lowest unsealed efficiency observed at 1.0 m was in respirator 3 of 99.61%. Respirators 1 and 4 had the hi ghest unsealed efficiencies, at 1.0 m, of 99.79% and 99.77%. Respirator 3 had the lowest unsealed efficiency for 2.0 m of 99.78%. Respirators 1 and 2 had the highest unsealed efficiencies, at 2.0 m, of 99.94% for both respirators. Respirators 4 had the lowest sealed efficiencies, at 0.5 m, of 96.93%. The highest sealed efficiency, at 0.5 m, was 98.45% in respirator 1. The lowest sealed efficiency at 1.0 m was in respirator 2 of 99.73%. Respirators 1 had the highest sealed efficiency, at 1.0 m, of 99.92%. For 2.0 m, respirator 4 had the lowest sealed 28
29 efficiency of 99.87%. Respirators 1 and 2 ha d the highest sealed efficiencies, at 2.0 m, of 99.97% and 99.96% respectevly. Table 3: Collection Efficiencies for Respirators Tested by Trial. Respirator 1 Staples Particle Size ( m) Trial 1 (%) Trial 2 (%) Trial 3 (%) Trial 4 (%) Trial 5 (%) Efficiency (%) SD (%) Unsealed 0.5 97.97 97.69 97.62 97.65 97.59 97.70 0.15 Sealed 0.5 98.84 98.65 98.34 98.30 98.11 98.45 0.29 Unsealed 1 99.76 99.79 99.80 99.81 99.80 99.79 0.02 Sealed 1 99.92 99.91 99.92 99.92 99.92 99.92 0.00 Unsealed 2 99.95 99.94 99.98 99.87 99.94 99.94 0.04 Sealed 2 99.95 99.98 99.99 99.98 99.96 99.97 0.01 Respirator 2 Staples Particle Size ( m) Trial 1 (%) Trial 2 (%) Trial 3 (%) Trial 4 (%) Trial 5 (%) Efficiency (%) SD (%) Unsealed 0.5 97.50 96.91 97.11 97.14 96.95 97.12 0.23 Sealed 0.5 97.02 96.96 97.21 97.04 97.01 97.05 0.09 Unsealed 1 99.70 99.77 99.74 99.74 99.80 99.75 0.04 Sealed 1 99.69 99.75 99.75 99.74 99.72 99.73 0.03 Unsealed 2 99.96 99.95 99.93 99.94 99.94 99.94 0.01 Sealed 2 99.98 99.99 99.93 99.99 99.92 99.96 0.03 Respirator 3 Staples Particle Size ( m) Trial 1 (%) Trial 2 (%) Trial 3 (%) Trial 4 (%) Trial 5 (%) Efficiency (%) SD (%) Unsealed 0.5 96.31 96.03 95.75 95.78 95.66 95.90 0.26 Sealed 0.5 97.29 97.15 96.96 96.85 96.75 97.00 0.22 Unsealed 1 99.54 99.62 99.63 99.64 99.60 99.61 0.04 Sealed 1 99.78 99.82 99.86 99.84 99.84 99.83 0.03 Unsealed 2 99.80 99.73 99.77 99.77 99.83 99.78 0.04 Sealed 2 99.95 99.94 99.94 99.89 99.96 99.94 0.03 Respirator 4 Staples Particle Size ( m) Trial 1 (%) Trial 2 (%) Trial 3 (%) Trial 4 (%) Trial 5 (%) Efficiency (%) SD (%) Unsealed 0.5 95.29 96.10 95.94 96.18 96.45 95.99 0.43 Sealed 0.5 97.02 96.80 96.83 96.89 97.13 96.93 0.14 Unsealed 1 99.76 99.77 99.76 99.78 99.77 99.77 0.01 Sealed 1 99.81 99.82 99.81 99.80 99.84 99.82 0.01 Unsealed 2 99.87 99.86 99.90 99.86 99.87 99.87 0.02 Sealed 2 99.74 99.91 99.90 99.89 99.92 99.87 0.07
95.00 96.00 97.00 98.00 99.00 100.00 0 0.5 1 1.5 2 2.5 Particle Size ( m)Efficiency (%) Unsealed Sealed Figure 17: Overall Collection Efficiency for Respirator 1 (n = 5). 95.00 96.00 97.00 98.00 99.00 100.00 00 511 522 5 Particle Size ( m)Efficiency (%) Unselaed Sealed Figure 18: Overall Collection Efficiency for Respirator 2 (n = 5). 30
95.00 96.00 97.00 98.00 99.00 100.00 0 0.5 1 1.5 2 2.5 Particle Size ( m)Efficiency (%) Unsealed Sealed Figure 19: Overall Collection Efficiency for Respirator 3 (n = 5). 95.00 96.00 97.00 98.00 99.00 100.00 0 0.5 1 1.5 2 2.5 Particle Size ( m)Efficiency (%) Unsealed Sealed Figure 20: Overall Collection Efficiency for Respirator 4 (n = 5). 31
32 A general linear mixed effects model with two fixed effects, staples and particle size, and one random effect, respirators, with an interaction term of staples and particle size was used. The model found that both fixed effects are statistically significant at an alpha value of 0.0001. The model also found interaction between the staples and the particle sizes to be statistically significant at an alpha value of 0.0002. Additional analysis included a Tukey honestly significant difference (HSD), a multiple comparison test, at an alpha value of 0.05. This model showed that the efficiency of the respirators when challenged with 0.5 m PSL spheres is significantly different from the efficiencies of the respirators when challenged with 1 and 2 m PSL spheres. Furthermore, the Tukey HSD demonstrated that the respirators had significantly different efficiencies (p < 0.05), when the staples were unsealed and sealed and challenged with 0.5 m spheres. A multiple Students t-test was performed to investigate the random effect of the respirators. This test showed that efficien cies for respirators 1 and 2 were significantly different and both were significantly different from that of respirators 3 and 4. Respirators 3 and 4 were not significantly different. This demonstrates an inherent variability in the performance of the respirators tested. However this is not of practical significance because all respirators had efficiencies greater than 95% at all particle sizes tested.
33 Discussion The efficiency of the single use respirators are determined by subtracting the background concentrations from the data and then using the following equation: E = (Nout Nin) / Nout where (Nin) is the number of particulates which penetrate the respirator filter medium and (Nout) is the number of particulates that remain outside the respirator. At all particle sizes, Respirators 1 and 3 had lower efficiencies when the staple openings were not sealed. Respirator 2 showed no difference in efficiency when the staple openings are sealed or unsealed. Respirator 4 showed no difference in efficiency when tested with particles of 1 and 2 m but had a lower efficiency for the unsealed staple openings when tested with particles of 0.5 m. It is unclear why respirator 2 did not s how any noticeable differences in collection efficiency before and after sealing the staple openings. One possibility is that the staples are closer to the edge of the filter and that this may reduce or eliminate leakage, however, all respirators showed a similar trend in collection efficiencies. At larger particle sizes (1 & 2 m), the efficiencies were higher because of inertial impaction. The smaller particles (0.5 m) have less inertia and deviate with the streamline to avoid impacting onto the filter but are too large to collect efficiently by diffusion. This particle size mainly deposits by interception and diffusion.
34 It is unclear why the standard deviation at 0.5 m was greater than those at 1.0 and 2.0 m by about one order of magnitude. It is possible that this variation is due to the change in predominant mechanism of collection. The 1.0 and 2.0 m particles are predominantly collected via impaction and very little contribution from the other mechanisms of deposition. At 0.5 m, interception and diffusion have greater contributions to collection efficiency. The change in collection mechanisms may have an affect on the consistency of particle collection, thus increasing the amount of variability. The difference in average penetration levels for the four respirators at 0.5 m was less than 1% for the respirators before and after sealing the staple openings. As the particle size increases, the penetration through to the staple openings diminishes. At 1 and 2 m the difference in average penetration is less than 0.2 %. This suggests that, at particles larger than 0.5 m, the efficiency of the respirators tested is above the certification criteria of NIOSH regardless of the leakage through the staple openings. It is unknown if the leakage through the staple openings is above 5% at particle sizes below 0.5 m, but this study provides evidence that smaller sized particles penetrate through the staple openings more efficiently.
35 Conclusions & Recommendations for Future Research The leakages observed in this study at all particle sizes did not exceed 5%. As stated earlier, NIOSH mandates filtration efficiency higher than 95% for certification of the respirator. This study suggests that the leakages through the staple openings are not a concern at particle sizes of 0.5 m and above, however, the leakages detected will have a cumulative effect when added to other sources of single use respirator leakages in the field. Statistical analysis showed that all three null hypotheses were rejected. The study suggests that the efficiencies of respirators with stapled head straps are less than that of respirators where the head straps are attached by other methods. It also suggests that leakages through the staple openings are greater when the respirator is challenged with particles in the most penetrative particle size range. More testing is necessary to determine if there is greater leakage through the staple openings when testing smaller particle sizes (< 0.5 m). Although the leakage through the staple openings may be insufficient to cause immediate concern, the study suggests that a monodisperse challenge aerosol of 0.3 m particles could reduce the efficiency of the respirators tested below 95%. Further tests should be conducted with challenge aerosols of smaller particle sizes. The particle counter that was used in this study has the capacity to resolve 0.2 and 0.3 m particle sizes. These respirators may show a greater leakage at particle sizes
36 smaller than 0.5 m. Research has showed that the most penetrative particle size range is 0.05 m to 0.500 m. More respirators from different manufacturers should be tested to determine if the results are similar. Some manufacturers of N95 respirators may place the staples of the head straps at different distances from the edge of the filter media. Staple thickness may also affect the leakage of the respirators. These parameters should be noted in further studies. As stated before, stretching the stapled head straps of a respirator prior to wearing may expand the size of the staple openings. Further investigation into the leakage caused by staples on the filter medium may involve stretching the headstraps at different distances. This can be achieved by using predetermined short and long stretching distances. Although the results of this study show significant leakage through staple openings on respirator filter medium, they can not be truly compared to NIOSH protocol because NIOSH tests respirators with sodium chloride polydisperse challenge aerosol 0.075 m in diameter with a standard deviation of 1.86 m. A better comparison can be obtained by further tests using the same protocol of this study but with a sodium chloride polydisperse aerosol in the same range as NIOSH testing protocol.
37 REFERENCES 3M (1995). Regulation Update 42 CFR 84: Occupational Health and Environmental Safety Division Number. Retreived from http://multimedia.3m.com/mws/mediawebserver 3M (2005). Respirators and Surgical Masks: A Comparison. Retreived from http://multimedia.3m.com/mws/mediawebserver Balazy, A., Toivola, M., Reponen, T., Podforski, A., Zimmer, A., & Grinshpun, S.A. (2006). Manakin-Based Performance Evaluation of N95 Filtering-Facepiece Respirators Challanged with Nanoparticles. Ann. Occup. Hyg, 50, 259-269. Chen, C.C., Chen W.Y., Huange S.H., Lin W.Y., Kuo Y.M., & Jeng F.T. (2001). Experimental Study on the Loading Characteristics of Needlefelt Filters with Micrometer-Sized Monodisperse Aerosols. Aerosol Science and Technology, 34, 262-273. 42 CFR Part 84. (1995). Respiratory Protective Devices; Final Rules and Notice. Federal Resgister, 60:110. Davies, C. N. (1973). Air Filtration. London & New York: Academic Press.
38 DiNardi, S. R.(Ed.). (2003). The Occupational Environment: Its Evaluation, Control, and Management (2nd ed.). Fairfax, Virginia: AIHA Press. Herris, W. P., (2009). How Regulation a nd Innovation Have Shaped Respirator Protection. EHS Today Retrieved from http://ehstoday.com/ppe/respirators/regulation_innovation_shaped/ Hinds, W. C. (1999). Aerosol TEchonology: Properties, Behavior, and Measurment of Airborne Particles (2nd ed.). New York & Canada: John Wiley & Sons, Inc. Lim, K.S., Kwon S.B., Lee K.W., & Kim M.C. (2002). Simultaneous Use of Polystyrene Latex Particles of Different Sized to Evaluate Perfromance of a Cyclone and Impactor. Aerosol Science and Technology, 36, 1003-1011. Martin, Jr., S.B., & Moyer K.S. (2000). Electrostatic Respirator Filter Media: Filter Efficiency and Most Penetration Particle Size Effects. Applied Occupational and Environmental Hygiene., 15 609-617. Myojo, T., Ehara K., Koyama H., & Okuyama K. (2004). Size Measurement of Polystyrene Latex Particles Larger than 1 Micrometer using a Long Differential Mobility Analyzer. Aerosol Science and Technology., 38 1178-1184.
39 Moyer, E.S., & Bergman M.S. (2000). Electrostatic N-95 Respirator Filter Medina Efficiency Degradation Resulting form Intermittent Sodium Chloride Aersol Exposure. Applied Occupational and Enviromental Hygiene., 15 600-608. Ogawa, A. (1984). Separation of Particles from Air and Gases: Volume II (J.K. Beddow Ed.). Boca Raton, Florida: CRC Press, Inc. Plog, B. A., & Quinlan P. J. (Eds.). (2002). Fundamental of Industrial Hygiene (5th ed.). USA: National Safety Council. Qian, Y., Willeke K., Grinshpun S.A., Donnelly J. & Coffey C.C. (1998). Performance of N95 Respirators: Filtration Efficiency for Airborne Microbial and Inert Particles. American Industrial Hygiene Association Journal ., 59, 128-132. Rengasamy, S., Eimer B.C., & Shafer R.E. (2009). Comparison of Nanoparticle Filtration Performance of NIOSH-appproved and CE-Merked Particulate Filtering Facepiece Respirators. Ann. Occup. Hyg.,53, 117-128. Rosati, J.A., Leith D., & Kim C.S. (2003). Monodisperse and Ploydisperse Aerosol Deposition in a Packed Bed. Aerosol Science and Technology, 37, 528-535.
40 Song, C.B., Lee J.L., Park H.S., & Lee K.W. (2007). Effect of Solid Monodisperse Particles on the Pressure Drop of Fibrous Filters. Korean J. Chem Eng., 24 148153. Yang, S. & Lee W.M. (2005). Electrostatic Enhancement of Collection Efficiency of the Fibrous Filter Pretreated with Ionic Surfactants. Air and Waste Manag. Assic., 55 594-603.
42 Appendix A: Equipment List Charge Neutralizer: Krypton-85 Activity: 10 mCi Source # 54-0018 Isotopes Products Valencia, CA TSI, Inc. St. Paul, Minnesota Collison Nebulizer: MRE 3-Jet Nebulizer Operating pressure: 20 PSIG Operation flow rate: 6 LPM BGI, Inc. 58 Guinan St., Waltham, MA 02451 (781) 891-9380 FAX: (781) 891-8151 Compressed Nitrogen: Industrial Grade Nitrogen Airgas South 125 Townpark Dr NW Ste 400., Kennesaw, GA, 30144-5880 (770) 590-6200 Fax: (770) 590-6100 Laser Particle Counter: Lasair Model 210 Operating flow rate: 1 CFM Particle Sizing Channel Thresholds: 0.2, 0.3, 0.5, 0.7, 1, 2, 3, 5 m Particle Measuring Systems, Inc. 5475 Airport Blvd., Boulder, CO 80301 (303) 443-7100 1-800-238-1801 FAX: (303) 449-6870
43 Appendix A (Continued): Equipment List Mass Flow Meter: Mass Flow Meter Model 4146 TSI, Inc. 500 Cardinal Rd, Shoreview, MN 55126 Polystyrene Latex Spheres : Polybead polystyrene microspheres 2.62% Solids-Latex Polysciences, Inc. Warrington, PA 18976 (215) 343-6484 Pressure Gauge: Magnahelic Gage Dwyer Instrument, Inc. Michigan City, IN 46360 RH and Temperature Sensor: Digital Hygrometer, Thermometer, Dew Point Fisher Scientific 5904 Tampa Oaks Pkwy, Tampa, FL, 33610-9521 (813) 622-7554 Silica Gel Dryer: Diffusion Dryer Model DD250 Air Techniques International Division of Hamilton Assoc. Inc. 11403 Cronridge Dr., Owings Mills, MD 21117 (410) 363-9696 FAX: (410) 363-9695
44 Appendix A (Continued): Equipment List Silicon Rubber: Silicon II GE Sealants and Adhesives. Huntersville, NC
45 Appendix B: Relative Humidity and Temperature Data Time (min) RH Temp (F) 5 51.06 73.89 10 50.88 73.9 15 50.89 73.88 20 50.81 73.86 25 50.89 73.83 30 50.93 73.83 35 50.89 73.84 40 50.73 73.81 45 50.79 73.8 50 50.75 73.81 55 50.75 73.81 60 50.64 73.79 65 50.75 73.78 70 50.77 73.8 75 50.73 73.8 80 50.64 73.8 85 50.59 73.79 90 50.57 73.79 95 50.57 73.81 100 50.48 73.81 105 50.39 73.79 110 50.41 73.78 115 50.29 73.8 120 50.25 73.8 125 50.23 73.79 130 50.22 73.8 135 50.16 73.79 140 50.21 73.78 145 50.27 73.77 150 50.43 73.79 155 50.41 73.8 160 50.41 73.79 165 50.48 73.79 170 50.57 73.8 175 50.57 73.82 180 50.55 73.81 185 50.64 73.81 190 50.63 73.84 195 50.59 73.83 200 50.59 73.82 205 50.61 73.83 210 50.64 73.85
46 Appendix C: Number Concentration Data Respirator 1: Unsealed Background Size ( m) Out In 0.5 390.9 1.1 1 31.9 0 2 3.4 0.3 Outside Size ( m) I II III IV V 0.5 670000 680000 680000 690000 690000 1 180000 220000 220000 230000 230000 2 1335.9 1457.3 1539.6 1884.7 1516.3 Inside Size ( m) I II III IV V 0.5 13594.6 15701.3 16202.6 16175.1 16594.4 1 425.2 460.9 444.4 440.7 470.8 2 0.9 1.2 0.6 2.7 1.2 Outside Corrected for Background Size ( m) I II III IV V 0.5 669609.1 679609.1 679609.1 689609.1 689609.1 1 179968.1 219968.1 219968.1 229968.1 229968.1 2 1332.5 1453.9 1536.2 1881.3 1512.9 Inside Corrected for Background Size ( m) I II III IV V 0.5 13593.5 15700.2 16201.5 16174 16593.3 1 425.2 460.9 444.4 440.7 470.8 2 0.6 0.9 0.3 2.4 0.9
47 Appendix C (Continued): Number Concentration Data Respirator 1: Sealed Background Size ( m) Out In 0.5 681.8 0.9 1 24.1 0.2 2 2.6 0 Outside Size ( m) I II III IV V 0.5 670000 700000 700000 690000 680000 1 150000 180000 260000 280000 300000 2 2705.8 3990.4 4716.7 4813.6 4940 Inside Size ( m) I II III IV V 0.5 7751.2 9445.3 11624.1 11704 12814.6 1 118.1 153.9 204.8 231.4 237.9 2 1.3 0.7 0.7 0.9 1.9 Outside Corrected for Background Size ( m) I II III IV V 0.5 669318.2 699318.2 699318.2 689318.2 679318.2 1 149975.9 179975.9 259975.9 279975.9 299975.9 2 2703.2 3987.8 4714.1 4811 4937.4 Inside Corrected for Background Size ( m) I II III IV V 0.5 7750.3 9444.4 11623.2 11703.1 12813.7 1 117.9 153.7 204.6 231.2 237.7 2 1.3 0.7 0.7 0.9 1.9
48 Appendix C (Continued): Number Concentration Data Respirator 2: Unsealed Background Size ( m) Out In 0.5 454.3 0.6 1 20 1.8 2 2.7 0 Outside Size ( m) I II III IV V 0.5 550000 610000 630000 610000 620000 1 140000 190000 160000 180000 180000 2 4691.9 4364.2 4863 4652.2 4650.2 Inside Size ( m) I II III IV V 0.5 13752.4 18861.9 18203.6 17419.9 18888.9 1 423.3 445.8 412.4 467.9 363 2 1.9 2.3 3.2 2.9 2.6 Outside Corrected for Background Size ( m) I II III IV V 0.5 549545.7 609545.7 629545.7 609545.7 619545.7 1 139980 189980 159980 179980 179980 2 4689.2 4361.5 4860.3 4649.5 4647.5 Inside Corrected for Background Size ( m) I II III IV V 0.5 13751.8 18861.3 18203 17419.3 18888.3 1 421.5 444 410.6 466.1 361.2 2 1.9 2.3 3.2 2.9 2.6
49 Appendix C (Continued): Number Concentration Data Respirator 2: Sealed Background Size ( m) Out In 0.5 698.8 5.2 1 18.2 0 2 1.7 0.8 Outside Size ( m) I II III IV V 0.5 620000 600000 590000 600000 600000 1 190000 240000 240000 240000 240000 2 1477.1 1636.1 1653.8 1603.9 1592.9 Inside Size ( m) I II III IV V 0.5 18463.8 18236 16470.7 17725.9 17936.6 1 589.3 590.2 598.7 632.9 676 2 1.1 0.9 1.9 1 2 Outside Corrected for Background Size ( m) I II III IV V 0.5 619301.2 599301.2 589301.2 599301.2 599301.2 1 189981.8 239981.8 239981.8 239981.8 239981.8 2 1475.4 1634.4 1652.1 1602.2 1591.2 Inside Corrected for Background Size ( m) I II III IV V 0.5 18458.6 18230.8 16465.5 17720.7 17931.4 1 589.3 590.2 598.7 632.9 676 2 0.3 0.1 1.1 0.2 1.2
50 Appendix C (Continued): Number Concentration Data Respirator 3: Unsealed Background Size ( m) Out In 0.5 1747.1 8.7 1 23.1 0 2 2.2 0 Outside Size ( m) I II III IV V 0.5 670000 790000 790000 790000 800000 1 130000 170000 170000 170000 160000 2 4784.4 5781 4952.4 5629.6 5601.2 Inside Size ( m) I II III IV V 0.5 24673.7 31337.3 33510.3 33270.6 34657.7 1 599.4 639.4 623.4 613.6 636 2 9.8 15.4 11.3 12.8 9.5 Outside Corrected for Background Size ( m) I II III IV V 0.5 668252.9 788252.9 788252.9 788252.9 798252.9 1 129976.9 169976.9 169976.9 169976.9 159976.9 2 4782.2 5778.8 4950.2 5627.4 5599 Inside Corrected for Background Size ( m) I II III IV V 0.5 24665 31328.6 33501.6 33261.9 34649 1 599.4 639.4 623.4 613.6 636 2 9.8 15.4 11.3 12.8 9.5
51 Appendix C (Continued): Number Concentration Data Respirator 3: Sealed Background Size ( m) Out In 0.5 424.9 0.5 1 10.7 0 2 5.6 0.4 Outside Size ( m) I II III IV V 0.5 670000 770000 780000 780000 780000 1 120000 210000 260000 240000 220000 2 2475.4 4176 4718.9 4738 5017.4 Inside Size ( m) I II III IV V 0.5 18126.8 21904.4 23676 24554.5 25372.4 1 268 383.8 370.9 378.8 347.8 2 1.7 3.1 3.1 5.6 2.4 Outside Corrected for Background Size ( m) I II III IV V 0.5 669575.1 769575.1 779575.1 779575.1 779575.1 1 119989.3 209989.3 259989.3 239989.3 219989.3 2 2469.8 4170.4 4713.3 4732.4 5011.8 Inside Corrected for Background Size ( m) I II III IV V 0.5 18126.3 21903.9 23675.5 24554 25371.9 1 268 383.8 370.9 378.8 347.8 2 1.3 2.7 2.7 5.2 2
52 Appendix C (Continued): Number Concentration Data Respirator 4: Unsealed Background Size ( m) Out In 0.5 293.8 0.9 1 15.1 0.2 2 1 0 Outside Size ( m) I II III IV V 0.5 700000 790000 770000 760000 760000 1 270000 300000 300000 290000 290000 2 2171.8 2214 1852.7 2186.3 2408.7 Inside Size ( m) I II III IV V 0.5 32953.5 30791.1 31287.7 29038.8 26952.4 1 651.1 688.7 728.2 634.7 655.3 2 2.9 3 1.8 3.1 3.1 Outside Corrected for Background Size ( m) I II III IV V 0.5 699706.2 789706.2 769706.2 759706.2 759706.2 1 269984.9 299984.9 299984.9 289984.9 289984.9 2 2170.8 2213 1851.7 2185.3 2407.7 Inside Corrected for Background Size ( m) I II III IV V 0.5 32952.6 30790.2 31286.8 29037.9 26951.5 1 650.9 688.5 728 634.5 655.1 2 2.9 3 1.8 3.1 3.1
53 Appendix C (Continued): Number Concentration Data Respirator 4: Sealed Background Size ( m) Out In 0.5 335.2 1 1 68.1 3.2 2 2.5 0 Outside Size ( m) I II III IV V 0.5 650000 660000 680000 680000 670000 1 170000 190000 190000 180000 170000 2 1801.7 2636.7 3389.2 2824.7 3460.4 Inside Size ( m) I II III IV V 0.5 19356 21132.9 21556.1 21130.4 19226.1 1 321.8 339.1 355.6 358.9 279.7 2 4.6 2.5 3.3 3 2.7 Outside Corrected for Background Size ( m) I II III IV V 0.5 649664.8 659664.8 679664.8 679664.8 669664.8 1 169931.9 189931.9 189931.9 179931.9 169931.9 2 1799.2 2634.2 3386.7 2822.2 3457.9 Inside Corrected for Background Size ( m) I II III IV V 0.5 19355 21131.9 21555.1 21129.4 19225.1 1 318.6 335.9 352.4 355.7 276.5 2 4.6 2.5 3.3 3 2.7
Appendix D: Prototype Respirator Assembly Data Prototype Testing Assembly Staples Particle Size ( m) Trial 1 (%) Trial 2 (%) Trial 3 (%) Trial 4 (%) Trial 5 (%) Efficiency (%) SD (%) Unsealed 0.5 98.08 98.07 98.42 98.61 98.97 98.43 0.38 Sealed 0.5 99.37 99.40 99.34 99.33 99.38 99.36 0.03 Unsealed 1 99.28 99.35 99.34 99.59 99.30 99.37 0.12 Sealed 1 99.66 99.78 99.79 99.77 99.83 99.77 0.06 Unsealed 2 98.78 99.43 99.66 99.52 99.43 99.37 0.34 Sealed 2 99.61 99.77 99.72 99.76 99.75 99.72 0.07 Prototype Testing Assembly Efficiency95 96 97 98 99 100 00 511 522 5 Particle Size ( m)Efficiency (%) Unsealed Sealed 54