USF Libraries
USF Digital Collections

Filtration efficiency of surgical masks

MISSING IMAGE

Material Information

Title:
Filtration efficiency of surgical masks
Physical Description:
Book
Language:
English
Creator:
Sanchez, Erin
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:

Subjects

Subjects / Keywords:
Polystyrene latex
Monodispersed aerosols
NIOSH certification tests
Chamber
Manikin
Particle counter
Dissertations, Academic -- Environmental and Occupational Health -- Masters -- USF   ( lcsh )
Genre:
non-fiction   ( marcgt )

Notes

Abstract:
ABSTRACT: Surgical masks are intended to be used to prevent transmission of disease from a health care worker to a patient. Often times, they are relied upon by health care workers for their own protection. In light of recent developments regarding preparation for health care worker response to global infectious diseases such as H1N1 Influenza, health care workers may experience a false sense of security when wearing surgical masks. The goal of this study was to evaluate the filtration efficiency of a double strap tie-on surgical mask. The manufacturer asserts a >95% efficiency with a 0.1 um challenge aerosol under FDA testing procedures. The NIOSH Title 42 CFR Part 84 certification criteria call for testing at a rate of 85 lpm representing a human moderate to heavy work load breathing rate. Three sizes of monodispersed aerosols (polystyrene latex beads: 0.5 um, 1.0 um, 2.0 um) were used. The specific aims were to measure the collection efficiencies of this mask for the various particle sizes. Two tests were performed. In the first, masks were affixed to a dummy head and the edges of the mask were not sealed. In the second, the edges of the masks were sealed to the head using silicone sealant, so all penetration was through the filtering material of the mask. Differences in upstream and downstream particle concentrations were measured. Thus, penetration by leakage around the mask and through the filtering material was measured. The experimental set up involved passing the aerosol from the nebulizer through a diffusion dryer and Kr-85 charge equilibrator ensuring a dry charge neutralized aerosol cloud for detection by a LASAIR particle counter. The analysis revealed that the filtration efficiency for 0.5 um particles ranged from 3% to 43% for the unsealed masks and 42% to 51% for the sealed. For 1.0 um particles, the efficiency was 58% to 75% for unsealed and 71% to 84% for sealed masks. For 2.0 um, the efficiency was 58% to 79% for unsealed masks and 69% to 85% for the sealed masks. The data were statistically significant and indicated that surgical masks were associated with very low filtration efficiency. This suggests that they may be inadequate against airborne viruses and bacteria.
Thesis:
Thesis (M.S.P.H.)--University of South Florida, 2010.
Bibliography:
Includes bibliographical references.
System Details:
Mode of access: World Wide Web.
System Details:
System requirements: World Wide Web browser and PDF reader.
Statement of Responsibility:
by Erin Sanchez.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains X pages.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
usfldc doi - E14-SFE0003323
usfldc handle - e14.3323
System ID:
SFS0027639:00001


This item is only available as the following downloads:


Full Text

PAGE 1

Filtration Efficiency of Surgical Masks by Erin Sanchez A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science Department of Environmental and Occupational Health College of Public Health University of South Florida Co-Major Professor: Yehia Y. Hammad, Sc.D. Co-Major Professor: Steve Mlynarek, Ph.D. Yangxin Huang, Ph. D. Date of Approval: February 18, 2010 Keywords: polystyrene latex, monodispersed aerosols, NIOSH certification tests, chamber, manikin, particle counter Copyright 2010, Erin Sanchez

PAGE 2

Dedication Firstly, I dedicate this thesis to God, for providing the opportunity to attend graduate school, giving me the strength to complete this program, and blessing my family with patience and understanding. Secondly, I dedicate this thesis to my wife Flor and my children, Alyssa, Destiny, and Elijah who supported and loved me throughout this process. Thirdly, I dedicate this thesis to my mom who has always encouraged me, loved me, and supported me. I love you all and thank you for being there for me.

PAGE 3

Acknowledgments I would like to thank my committee members for their time, patience, guidance and understanding during the process and completion of this research project. I would like to thank Dr. Mlynarek for providing me with a research topic after changing jobs. Dr. Mlynarek was instrumental in convincing me to enroll in the Industrial Hygiene program. I would like to thank Dr. Huang for providing guidance and expertise regarding statistical analysis. I would especially like to thank Dr. Hammad for the many late hours he spent with Daniel and me during the modification and construction of the chamber. I would like to thank him for his tremendous effort, the patience he exhibited and the wisdom he shared. I was honored to work with a committee of professionals that provided the care, guidance and wisdom to make this study successful. I would also like to thank the National Institute of Occupational Safety and Health for providing the funds that allowed me to attend graduate school. Without the funding I could not have attended graduate school. I would also like to thank Daniel Medina for the time and effort he provided throughout the process of modifying the chamber and running several tests. Working with the students and faculty at USF College of Public Health was a pleasure.

PAGE 4

i Table of Contents List of Tables ii List of Figures iii Abstract v Introduction 1 Background 1 Literature Review Studies Associated with Efficiency Testing 1 Study Purpose and Hypothesis 10 Research Methods 11 Materials and Methods 11 Protocol 26 Results 28 Discussion 34 Conclusion 42 References 44 Appendices 46 Appendix 1: Major Materials and Components of the Experiment 47 Appendix 2: NIOSH Procedures 48 Appendix 3: Respirator # 1 Concentration Levels 60 Appendix 4: Respirator # 2 Concentration Levels 62 Appendix 5: Respirator # 3 Concentration Levels 64 Appendix 6: Respirator # 4 Concentration Levels 66 Appendix 7: Respirator # 5 Concentration Levels 68 Appendix 8: Respirator # 6 Concentration Levels 70 Appendix 9: General Linear Model Statistics 72

PAGE 5

ii List of Tables Table 1 Summary of Similar Studies 9 Table 2 Polystyrene Latex Parameters 18 Table 3 Efficiency for Unsealed Mask at 0.5 um 29 Table 4 Efficiency for Sealed Mask at 0.5 um 29 Table 5 Efficiency for Unsealed Mask at 1.0 um 31 Table 6 Efficiency for Sealed Mask at 1.0 um 31 Table 7 Efficiency for Unsealed Mask at 2.0 um 32 Table 8 Efficiency for Sealed Mask at 2.0 um 32 Table 9 Average Efficiency Comparison of Unsealed and Sealed Mask at 37 0.5 um Table 10 Average Percent Difference for Unsealed vs Sealed Mask at 0.5 um 37 Table 11 Average Efficiency Comparison of Unsealed and Sealed Mask at 38 1.0 um Table 12 Average Percent Difference for Unsealed vs Sealed Mask at 1.0 um 39 Table 13 Average Efficiency Comparison of Unsealed and Sealed Mask at 39 2.0 um Table 14 Average Percent Difference for Unsealed vs Sealed Mask at 2.0 um 40 Table 15 Average Efficiency Compared to Particle Diameter 40

PAGE 6

iii List of Figures Figure 1. Aerosol chamber full view 12 Figure 2. Aerosol chamber door panels 12 Figure 3. Make up air entry w/ magnahelic gauge, aerosol entry port, weather Strip 13 Figure 4. Bottom panel 14 Figure 5. PVC bypass valves w/ “T” connection 14 Figure 6. Center diffusion baffles 15 Figure 7. Exhaust diffusion baffle 15 Figure 8. Bottom PVC exhaust system 15 Figure 9. Manikin head on mounting bracket 16 Figure 10. Manikin head inside chamber 16 Figure 11a. Pleated double strap tie on surgical mask 17 Figure 11b. Nose clip from surgical mask 17 Figure 12. Collison nebulizer 19 Figure 13. Nitrogen tank 19 Figure 14. Diffusion dryer 21 Figure 15. Diffusion dryer (yellow – unsaturated) 21 Figure 16. Diffusion dryer (green – saturated) 21 Figure 17. Kr-85 charge equilibrator 22 Figure 18. LASAIR Model 210 Particle Counter 23

PAGE 7

iv Figure 19. Aerosol sampling chamber flow diagram 24 Figure 20. Aerosol sampling chamber system setup 25 Figure 21. Efficiency comparison of 0.5 um particles by unsealed mask 30 Figure 22. Efficiency comparison of 0.5 um particles by sealed mask 30 Figure 23. Efficiency comparison of 1.0 um particles by unsealed mask 31 Figure 24. Efficiency comparison of 1.0 um particles by sealed mask 32 Figure 25. Efficiency comparison of 2.0 um particles by unsealed mask 33 Figure 26. Efficiency comparison of 2.0 um particles by sealed mask 33 Figure 27. Unsealed mask under normal condition 36 Figure 28. Sealed mask with aerosol crossed strap 36 Figure 29. Unsealed vs Sealed Mask Comparison at 0.5 um 37 Figure 30. Unsealed vs Sealed Mask Comparison at 1.0 um 39 Figure 31. Unsealed vs Sealed Mask Comparison at 2.0 um 40 Figure 32. Average Efficiency for Unsealed vs Sealed Masks by Particle Size 41

PAGE 8

v Filtration Efficiency of Surgical Masks Erin Sanchez ABSTRACT Surgical masks are intended to be used to prevent transmission of disease from a health care worker to a patient. Often times, they are relied upon by health care workers for their own protection. In light of recent developments regarding preparation for health care worker response to global infectious diseases such as H1N1 Influenza, health care workers may experience a false sense of security when wearing surgical masks. The goal of this study was to evaluate the filtration efficiency of a double strap tie-on surgical mask. The manufacturer asserts a >95% efficiency with a 0.1 um challenge aerosol under FDA testing procedures. The NIOSH Title 42 CFR Part 84 certification criteria call for testing at a rate of 85 lpm representing a human moderate to heavy work load breathing rate. Three sizes of monodispersed aerosols (polystyrene latex beads: 0.5 um, 1.0 um, 2.0 um) were used. The specific aims were to measure the collection efficiencies of this mask for the various particle sizes. Two tests were performed. In the first, masks were affixed to a dummy head and the edges of the mask were not sealed. In the second, the edges of the masks were sealed to the head using silicone sealant, so all penetration was through the filtering material of the mask. Differences in upstream and downstream particle concentrations were measured. Thus, penetration by leakage around the mask and through the filtering material was measured. The experimental set up involved passing

PAGE 9

vi the aerosol from the nebulizer through a diffusion dryer and Kr-85 charge equilibrator ensuring a dry charge neutralized aerosol cloud for detection by a LASAIR particle counter. The analysis revealed that the filtration efficiency for 0.5 um particles ranged from 3% to 43% for the unsealed masks and 42% to 51% for the sealed. For 1.0 um particles, the efficiency was 58% to 75% for unsealed and 71% to 84% for sealed masks. For 2.0 um, the efficiency was 58% to 79% for unsealed masks and 69% to 85% for the sealed masks. The data were statistically significant and indicated that surgical masks were associated with very low filtration efficiency. This suggests that they may be inadequate against airborne viruses and bacteria.

PAGE 10

1 Introduction Background Surgical masks are not designed to protect health care workers from airborne particulates and will not provide as much protection as N-95 respirators. Smaller particulates are less effectively filtered by most surgical masks. In addition to relatively poor filtration efficiency, these masks permit leakage around the edges upon inhalation, and they cannot be fit tested. For healthcare workers dealing with patients ill with infectious agents like the Swine Flu (H1N1 in fluenza virus), surgical masks have been recommended by the Center for Disease Control and Prevention (CDC) as a last resort, when no National Institute of Occupational Safety and Health (NIOSH) approved respirator is available. Using surgical masks as a form of personal protective equipment (PPE) may lead to adverse health effects. Literature Review Studies Associated with Efficiency Testing Surgical masks have been used since the early 1900s in the health care setting to prevent transmission of infectious diseases, via large droplets, from the worker to the patient. The masks are also used to prevent splashes of blood and body fluids from the patient to the mucous membranes of the healthcare worker. In 2008, the Institute of Medicine reported that during an influenza pandemic, it may be necessary to protect more than 13 million health care workers from illness or from infecting their families or patients (Grinshpun, 2009).

PAGE 11

2 Early surgical masks were constructed from layers of cotton gauze and were designed to protect the mucous membranes of the nose, eyes and mouth where patient handling may have resulted in splashes or sprays of blood and body fluids. Health care workers have used and continue to use surgical masks as a form of personal protective equipment against airborne infectious diseases. In a Toronto hospital, all attending health care workers reported wearing “respirators” contracted severe acute respiratory syndrome (SARS) during patient intubation. Closer examination revealed that employees were wearing surgical masks (Oberg, 2008). While some surgical masks look similar to respirators, they are not, and do not offer th e same protection as respiratory protection devices. Respiratory protection devices are certified by NIOSH and are used to protect the wearer from inhaling contaminants suspe nded in the air. NIOSH approved respirators have a filtering medium capable of removing at least 95% of airborne particulates > 0.3 um in diameter. Respirators have been used in the health care setting when the workforce was concerned with the spread of tuberculosis. Surgical masks are not equipped with such filtering material to reduce particle penetration by 95%. An aerosol is a liquid droplet or solid particle dispersed in air. Bi oaerosols are aerosols of biological origin and include viruses, living organisms, such as bacteria and fungi. Bacteria are usually spherical or rod shaped, but may occur in clusters or chains. The adverse health affects of the biologic particles, particularly pathogenicity, depend not on the mass of the inhaled particles but on the number of particles. There are more than 17,000 species of bacteria and those that cause human disease are called human pathogens. Viral particles, called virions, are one of the smallest known bioaerosol agents, with a particle diameter ranging from 20 to 300 nm (Balazy, 2006a). Aerosol particles attach firmly to any surface they

PAGE 12

3 contact and this is what separates them from gas molecules and from millimeter size particles. When aerosol particles contact each other they adhere and form agglomerates. (Hinds, 1999) Filtration relies on the adhesion of the particles. Although surgical masks are not as efficient as air purifying NIOSH respirator s they too operate by mechanical filtration. A mechanical respirator traps the particulate matter that passes through the filter material. Surgical masks and respirator filters are c onstructed of flat, non-woven mats of fine fibers (NIOSH Science Blog, 2009). The fiber is laid so the long section of the fiber is perpendicular to the air crossing the path, therefore allowing several particles to be captured along the axis. The efficiency with which a fiber removes particles from an aerosol stream is called Single Fiber Efficiency. The assumption is that the particle adheres to the fiber and is permanently removed from the airflow. An examination of the Reynolds number (Ref) that characterizes the flow around a fiber having a diameter df reveals that, under most conditions, the flow inside a filter will be laminar. (Hinds, 1999). Flow is distorted and influenced by other fibers, even when they are several fiber diameters away. The efficiency is considered the number of particles collected on a unit length of fiber divided by the number of particles that would have passed by the fiber in one second (Hinds, 1999). There are five mechanisms for particles to be deposited on a filter and in the lungs: interception, inertial impaction, diffusion, gravitational settling, and electrostatic attraction. The first four mechanisms are called mechanical mechanisms. Interception and impaction are responsible for collecting the relatively larger particles while diffusion is responsible for capturing the smaller particles. Interception occurs when the particle follows a streamline and the particle comes within one radius particle

PAGE 13

4 of the fiber and adheres to it. The particle is assumed to follow its streamline perfectly and is not affected by inertia, settling, and Brownian motion. Inertial impaction occurs when the particle, due to inertia, is unable to adjust quickly enough to the change in the air stream near the fiber and collides into the fiber. Impaction is the most important mechanism for large particles. Diffusion is a mechanism in which the particles wander in a random motion (known as Brownian Motion) and leave the airflow streams and adhere to the collection surface and are effectively removed from the air. Diffusion is negligible for particles greater than 5 um; and is predominately important for particles less than 0.1 – 0.25 um (Fleeger, 2002). Gravitational settling is simply the particle settling due to gravitational forces and adhering to the filter material. Electrostatic deposition can be extremely important but difficult to quantify because it requires knowing the charge on the particles and on the fibers. Charged particles are attracted to oppositely charged fibers by Coulombic attraction (Hinds, 1999). Once the particles are adhered to the filter they are difficult to remove. The challenge aerosols in this test were 0.5 um, 1.0 um, and 2.0 um, and these sizes are generally captured through impaction and interception, but 0.5 um particles also diffuse to some degree by diffusion. Particles that are 0.3um, the most penetrating particle size (MPPS), are dominated by diffusion and interception, while particles below 0.1 um are affected only by diffusion. When the filter demonstrates high efficiency at 0.3 um, then the filter will be more effective against smaller and larger particle sizes. With the recent development of infectious diseases such as SARS, Avian influenza, and the threat posed by the H1N1 Influenza virus, the world has a renewed emphasis on infectious agents. The health care industry has an increased risk of

PAGE 14

5 occupational exposure based on the likelihood of encountering patients with the H1N1 virus. SARS developed in Asia and spread across more than 20 countries. Surgical masks became the staple image associated with respiratory protection for swine flu. Air, water and ground transportation have played a significant role in the spread of the diseases. People are capable of traveling from one country to another country half way across the world in less than 48 hrs. The CDC states the H1N1 virus was first detected in the United Stated in April 2009. The virus is spread in the same way as the seasonal flu. The flu is spread from person to person by inhalation of the large droplets spread though coughing and sneezing, and sometimes by contact with contaminated surfaces and touching their face and mouth. The symptoms of H1N1 and seasonal flu are very similar; therefore, infected persons continue to spread the disease without being diagnosed. The H1N1 virus has been associated with several deaths throughout the United States. Local Department of Public Health organizations, such as Florida, are tracking and posting confirmed cases and deaths, along with the county location on the internet. Viruses are intracellular parasites that can reproduce only inside a host cell. Infectious diseases vary in size with viruses at 0.02 to 0.3 um diameters, bacteria with 0.5 to 5.0 um diameters and droplets with 1 to 100 um in diameter (Grinshpun, 2009). The physical size of a SARS causing coronavirus was about 0.08 – 0.12 um (Lee, 2007). Surgical masks will provide a barrier protection against large droplets that are considered to be the primary route of SARS and H1N1 transmission; however, smaller particulates are less effectively filtered. Close contact, generally less than 3 ft, is required for transmission. Surgical masks may also be placed on patients with communicable diseases to contain respiratory droplets. Surgical masks cover the nose and mouth of the health

PAGE 15

6 care provider and are held in place by double straps. The masks are generally worn during medical procedures with the intent of reducing the spread of disease from the worker to the patient. The mask will provide a barrier for the worker against larger droplets, such as sneezes and coughs; however, it is not uncommon to find workers using surgical masks for protection against smaller airborne aerosols. Under 29 CFR 1910.134 the Occupational Safety and Health Admi nistration (OSHA) require use of NIOSH approved respirators for protection against air borne diseases such as Tuberculosis (TB) when engineering controls are not adequate. NIOSH respirators are at least 95% efficient for particles > 0.3 um. Surgical masks continue to be used as a form of respiratory protection. Surgical masks are not tested under the NIOSH certification however, the Food and Drug Administration (FDA) is respons ible for regulating medical devices and requires manufacturers to demonstrate efficiency with regards to fluid resistance, filter efficiency, differential pressure, and flammability. The manufacturer provides data and proposed claims to FDA for review and the FDA reviews the provided data and clears the mask for sale (3M, 2005). The two filter efficiency tests recommended include particulate filtration efficiency (PFE) usi ng a non-neutralized aerosol of 0.1 um latex spheres at a challenge velocity of 28 lpm. PFE is a quality indicator for surgical masks and is not an indicator of protection performance. It measures how well the mask filters out particles such as viruses and other submicron particles. The filter media of a surgical mask with a very high (> 95%) PFE may be less than 70% efficiency under NIOSH certification test methods. Bacterial filtration efficiency (BFE) testing uses a nonneutralized 3+/0.3 um staphylococcus aureus aerosol and a flow rate of 28.3 lpm. BFE measures how well the mask filters out bacteria when challenged with an aersosol

PAGE 16

7 containing bacteria. It assesses the ability of the mask to provide a barrier to large particles expelled by the wearer. The FDA does not have a minimum filtration efficiency (Oberg, 2008). The pressure differential is a measure of the air flow resistance of the mask and is an objective measure of breathability. The higher the pressure differential, the harder it is for the wearer to breathe. The fluid resistance test reflects the mask’s ability to minimize the amount of fluid that could transfer from the outer layers through to the inner layer as a result of splash or spray (Marusyk, R., 2009). The surgical masks tested in this study claimed 99% BFE and 95% PFE. Respirators are evaluated using the NIOSH certification testing method in accordance with Title 42 CFR Part 84. The new certification test was implemented in June 1995 outlining the procedures for testing and certifying air purifying and particulate respirators. The certification test identifies nine classes of filter with efficiencies of 95%, 99% and 99.97%. The filters also have a resistance to degradation and are labeled as N, R and P series. The rating for “N” series respirators is given when the filters are not oil resistant. The “R” rating is given when the filter is resistant to oil and “P” rating is given when the filter is oil proof. The testing parameters call for using NaCl particle sizes with a count median diameter in the range of 0.075 +/0.02 um (0.3 um Mass median diameter) and a geometric standard deviation not exceeding 1.86 at a challenge flow rate of 85 lpm (+/5%), which represents a moderately to high work rate. Sodium chloride (NaCl) particles are used when testing N-series filters, and dioctyl phthalate (DOP) oil are used for testing R and P series filters. The challenge aerosols are charge neutralized. Manikin based and live human studies have been conducted under various circumstances to determine filtration efficiency of masks. One study tested two chambers

PAGE 17

8 and determined that a small chamber (0.096 m3) was just as effective as a large walk in chamber (24.3 m3) for testing of masks, suggesting that laboratory based evaluations have a good potential to adequately represent the respirator field performance (Balazy, 2006b). Overall, surgical masks tests revealed penetration in the range of 4 – 90%. The aerosol concentration outside and inside the mask were measured to determine filtration efficiency. Tests concluded that penetration occurs mainly at the faceseal and the manufacturing of masks should focus on improving the faceseal efficiency instead of the filter medium. Several studies used aerosol generating jet nebulizers, charge neutralizers, aerosol sampling chambers and silicone to seal the mask. Electrostatic filter properties play a significant role in capture efficiency. Table 1 provides a summary of similar studies identifying the specifics parameters used and specific aims.

PAGE 18

9 Table 1. Summary of Similar Studies Study Study Description Study Specifics Results Comments AIJC Major Study; Anna Balazy; 2006a Efficiency test of N-95 and surgical masks Equipment used include: HEPA filter used to filter air, Kr-85 charge neutralizer, silicone sealant, aerosol chamber, flow rate at 85 lpm & 30 lpm, aerosol particle counter Penetration of virions exceeded 5%for N-95, Surgical masks 25 – 84.5% penetration 6-jet nebulizer, Silicone leak tested, MS2 virions used 0.01 to 0.08 um, Diffusion dryer not used Tara Oberg, 2008 Evaluate filter performance and facial fit of surgical masks Nine surgical masks were tested using monodispersed aerosols (0.895, 2.0, 3.1 um) – represent Bitrex size, Kr-85 charge neutralizer, HEPA filtered air, light scattering photometer, also used 0.075 um NaCl at 84 lpm Latex challenge: 0 84% penetration was 16% for 0.895 um, 15% for 2.0 um, 11% for 3.1 um; NaCl: 4 90% penetration Flow rate at 6 lpm (resting human breathing rate), mask sealed to metal plate; human subjects also used and fit tests conducted Sergey Grinshpun, 2009 Efficiency testing of N-95 and surgical masks using human subjects and manikins Test penetration under normal breathing conditions for N-95 and surgical masks under 0.3 – 1.0 um, 25 subjects used; breathing rate was recorded with breathing simulation system, masks sealed to manikin with glue, leak check, Kr-85 neutralizer, Dryer Surgical Penetration -Faceseal: 48%, Filter medium: 9% Electrical Low Pressure Impactor with an air diluter, leak check conducted 3M, 2005 N-95 and surgical mask comparison Compared N-95 and surgical masks, described PFE, BFE None None Anna Balazy, 2006b Manikin evaluation N-95 w/ challenge aerosols Aerosol concentration inside and outside at 85 lpm & 30 lpm, NaCl challenge aerosol 0.01 – 0.6 um aerosols, small and large test chamber used & showed no difference, Dryer, HEPA filter, Kr-85 neutralizer, particle counter, silicone sealants Penetration exceeded 5% for 9 of 10 masks at 85 lpm for N-95 respirators 6 –jet nebulizer, leak check JT Huang, 2007 Evaluation of Efficiency of masks Human subjects, masks were sealed to the face by using sticky tape to determine breathing resistance Greater resistance when sealed, observations indicated bacteria from cough was at least 1000 times more than generated by regular breathing or talking Idea for future human testing Byung Lee, 2005 Filtering Efficiency of N95 & R-95, surgical masks Room size indoor test chamber, real time aerosol size cascade impactor reports concentration and size every minute, mask sealed to face, Manikin, Bioaerosol target diameter of 0.04 – 1.3 um, neutralizer Surgical masks > 20% penetration for 0.04 um and < 15% for 1.3 um Neutralizer used after aerosols through filter, poly test aerosol, leak test Shu-an Lee, 2008 Respiratory Performance Offered by N95 Respirators and Surgical Masks Determine protection factor of N-95 and surgical masks against particles representing bacterial and viral sizes of 0.04 to 1.3 um, Walk in test chamber and human subjects performed OSHA fit testing exercises, Dryer, HEPA filter About 29% of N-95 and 100% of surgical masks had protection factor < 10, surgical average PF was 2.4 Human test subjects

PAGE 19

10 Study Purpose and Hypothesis The purpose of this study was to assess filtration efficiency resulting from leakage around the surgical mask and determine if the efficiency was different for sealed and unsealed masks using NIOSH certification met hods. The filtration efficiency was then compared to FDA methods. In this study monodispersed polystyrene latex (PSL) beads were used. These are aerosols composed of airborne particulates of a single size or a small size range as opposed to polydispersed particulates composed of airborne particulates of many different sizes. The first hypothesis was that the filtration efficiencies were not different between sealed and unsealed surgical masks. The second hypothesis was that surgical mask efficiencies for the particle sizes tested were greater than the 95% efficiency specified by NIOSH.

PAGE 20

11 Research Methods Materials and Methods The filtration efficiency of the double tie strap surgical masks was measured using a protocol that included using a maniki n head, in which the masks were affixed and tested with and without being sealed to the head. Sealing the masks prevented leakage between the mask edges and the face; therefore aerosol concentrations detected in the masks were those which passed through the filtering medium. The experiment was conducted in the USF College of Public Health Student lab where the average temperature was 74 F. An aerosol sampling chamber (see figure 1) was constructed by converting a 50 gallon aquarium into a tightly sealed testing chamber. The volume of the aerosol chamber is 190 liters. The chamber was used in a standing position at a height of 48”. Two wood door panels (see figure 2) were modified to enable testing within the chamber. A tight seal was created by applying weather stripping along the inside door edges. The top panel was designed to include an aerosol entry port at the top section and to allow for clean make up air through the middle section (see fi gure 3). A high efficiency particulate air (HEPA) filter capable of filtering 99.97% of pa rticles 0.3 um particles was installed and secured by wire mesh screen in between an 8” x 11” wood panel which was secured to the panel by metal screws and washers. Weather stripping was also placed along the edges of the wood frame and wood panel to reduce leakage. A magnahelic gauge (Dwyer

PAGE 21

12 Instruments, Inc, Michigan City, IN) was installed for indication of pressure inside the chamber and to reveal potential air leaks. Figure 1. Aerosol Chamber Figure 2. (left) top panel, (right) bottom panel

PAGE 22

13 Figure 3. Make up air entry with magnehelic gauge, aerosol entry port, weather stripping The bottom wood panel (see figure 4) was equipped with brass “T” entry ports to allow for the passage of Tygon tubing (see figure 5). Teflon tape was applied to the edges of the port openings to seal around the Tygon tubing. Two polyvinyl chloride (PVC) bypass valves and a PVC “T” connector were used to enable the operator to switch from inside to outside the mask and measure the aerosol concentration levels.

PAGE 23

14 Figure 4. Bottom Panel Figure 5. PVC Bypass Valves with “T” Connection The chamber contained three stainless steel baffles with 1/8 inch diameter holes spaced uniformly on 1/4 inch centers located in the middle of the chamber (see figure 6). Baffles were spaced three inches apart with the top baffle located 16” from the top of the chamber. The fourth baffle (see figure 7) was located 3 ” from the bottom of the chamber and was installed over the PVC plenum (see figure 8) used to exhaust the air out of the chamber.

PAGE 24

15 Figure 6. Center diffusion baffles Figure 7. Exhaust baffles Figure 8. PVC Exhaust The manikin head used for the experiment was an Airway Larry Management Trainer (Nasco: Life Form Products, Fort Atkinson, WI). The manikin was installed in the aerosol sampling chamber for every test (see figure 9 & 10). The head exhibited a nose and a mouth opening through which aerosols were passed. To achieve 85 lpm of air through the surgical mask two pieces of Tygon tubing were inserted through the mouth opening with one tube connected to the LASAIR and the other tube to an electric pump.

PAGE 25

16 Figure 9. Manikin head on mounting bracket Figure 10. Manikin inside chamber

PAGE 26

17 The masks used in this experiment were double strap tie on surgical masks (see figure 11a). They consisted of a pleated three layer filter medium. The pleated filter provided more surface area for ease of breathing through the filter. The mask also contained a metal forming nose clip (see figure 11b) which allows the user to adjust the nose clip according to the dimensions of the individual’s facial features. The nose clip was formed to the manikin’s head and nasal features. In respirator test #1 the straps were secured as it would be in real life and actual use. The lower strap was tied behind the neck and the top strap was tied on the top portion of the skull. Crossing the straps provided a tighter fit and this configuration was used for all tests thereafter. Figure 11a. Pleated Double Strap Tie on Surgical Mask Figure 11b. Nose Clip The experimental design called for generating monodispered PSL particles of three sizes: 0.5 um, 1.0 um, and 2.0 um. The PSL was received in 15 ml bottles. Before use, the bottles were slightly shaken to mix the particles and reduce clumping. For each trial, two drops of the PSL suspension were added to 40 ml of distilled water measured by a graduated cylinder. The suspension in the jar was swirled slightly to ensure mixing.

PAGE 27

18 Table 2. Polystyrene latex spheres parameters Nominal Size (um) Actual Size (um) Standard Deviation (um) Solids-Latex (%) 0.5 0.465 0.01 2.62 1.0 0.989 0.02 2.59 2.0 1.826 0.046 2.70 In this experiment, generation of monodispersed PSL was achieved by using a CN-24J 3-jet Collison Nebulizer (BGI, Inc, Waltham, MA) (see figure 12) with a 1.9” diameter glass jar. The nebulizer jet stem was placed inside the jar ensuring the bottom of the stem was in the water while keeping the jet ports above the liquid level. The house air supply was not used. Rather, nitrogen (see fi g 13) contained in an AIRGAS compressed gas cylinder was used to generate the aerosols. Nitrogen pressure was maintained at 20 PSI as directed in the manual. The nitrogen provided a steady, consistent gas which was controlled and monitored using a regulator pressure gauge. The nitrogen was relatively inexpensive. Maintaining a steady flow of compressed air when using house air is difficult due to the unpredictable pattern of use by other personnel and equipment in the facility. Therefore, pressure in the facility fluctuates considerably. Besides, house compressed air usually contains condensed water. The Collison nebulizer manual indicated that a 3jet unit running at 20 PSI resulted in a flow of 6 liters per minute of nitrogen. The nitrogen gas was filtered using fiber glass HEPA filter. The filter was placed in line prior to connecting to the nebulizer’s port. As the gas passed through the nebulizer the PSL aerosols were sprayed against the jar walls which acted as a barrier and allowed the aerosol particles to atomize at the appropriate particle diameter. The mist inside the jar exited the nebulizer where the connection port was fitted tightly into the diffusion dryer (ATI, Inc, Ownings Mills, MD).

PAGE 28

19 Figure 12. Collison Nebulizer Figure 13. Nitrogen Tank

PAGE 29

20 The utilization of the diffusion dryer (see fi gure 14) resulted in producing dry aerosol particles prior to entering the chamber. Reducing the amount of water reduced the relative humidity build up within the test chamber. The dryer allowed for the particles to enter the chamber dry and the particle counter appropriately determined the size, count, and concentration outside and inside the surgical mask. The silica gel beads were yellow (see figure 15) within the container and visible with the naked eye. The silica gel changed color from yellow to green when saturated (see figure 16). The silica gel did not change color instantaneously. Rather, the individual gel beads gradually changed color as the aerosols were generated and partial saturation occurred. The diffusion dryer was monitored continuously throughout the testing to prevent saturation. An additional dryer was available and as a result a dry chamber was used for each test. While one dryer was being used for testing the second dryer was placed in an oven set at 120 C indicated in the operator’s manual. The particles were dried as they passed through the silica gel chamber and allowed to enter the Kr-85 charge equilibrator. Partial saturation of the dryer was evident; however, the full saturation did not occur prior to completing the testing procedures. A test to determine relative humidity (RH) within the chamber was conducted and measured every 5 minutes for 3 hrs during aerosol generation. The RH in the chamber prior to testing was equal to the RH in the room which was 51.22% and the highest level achieved during testing was 51.06%. The results indicate that the particles entering the chamber are dry.

PAGE 30

21 Figure 14. Diffusion Dryer Figure 15. Unsaturated (yellow) Figure 16. Saturated (green) It should be noted that NIOSH certificati on tests were carried out using charge neutralized particles. The Kr-85 charge equilibrator (TSI Isotope Products Laboratories, Valencia, CA) (see figure 17) was the radioactive source used to neutralize the aerosol cloud prior to dispersion into the chamber. Kr-85 was a beta emitter. The aerosols naturally acquire electrostatic charge as they are released into the environment. The charged particles have a tendency to migrate to the Tygon tube walls, chamber walls, manikin head and to the surgical mask itself. The neutralization therefore permits the particles to provide for more dependable testing results.

PAGE 31

22 Figure 17. Kr-85 charge equilibrator Aerosol particles entering the chamber were measured using a LASAIR Particle Counter (Particle Measuring Systems, Inc, Boulder, CO) (see figure 18), Model 210, inside and outside the mask. The LASAIR sized and counted particles by measuring the amount of light scattered by each particle. The source of illumination is an internal 10 milliwatt HeNe laser. The instrument sampled air at 1 CFM (28.32 lpm). There were eight channels in the instrument which included the particle sizes of interest: 0.5 um, 1.0 um, and 2.0 um. The average outside and inside particle concentrations were displayed and recorded every 10 mins. The maximum concentration the instrument was capable of reading was 750,000 ft3. Prior to testing, the instrument was zeroed using manufacturer Ultipor N66 0.2 um rated zero calibration filters The LASAIR was configured to provide six 10 minute samples and the results were displayed in real time.

PAGE 32

23 Figure 18. LASAIR Model 210 Particle Counter The airflow into the chamber included: the nebulizer’s 6 lpm of nitrogen when operated at 20 PSI; the airflow through the surgical mask at 85 lpm as required by NIOSH, this flow is divided into two parts: 28.3 lpm for the particle counter and the balance, 56.7 lpm to the air pump and finally, 9 lpm in the plenum at the bottom of the chamber. Therefore, the total airflow in the chamber is 100 lpm (see figure 19). A TSI mass-flow meter was used to measure the airflow in the various system components. The system components were set up as depicted in figure 20.

PAGE 33

24 Figure 19. Aerosol Sampling Air Flow Diagram

PAGE 34

25 Figure 20. Aerosol Sampling Chamber System Set Up To determine the period of time required to reach the maximum (equilibrium) concentration in the chamber, the following equation was used: C = (G/Q) (1e (–Qt/V)), note: G/Q is equal to Cmax C/ Cmax = (1e (–Qt/V)), note: C/ Cmax = 0.99 = 1 e (–Qt/V) 0.01 = e (–Qt/V), this is 1% because the concentration can’t reach zero ln (0.01) = ln (e (–Qt/V)) 4.6 = Qt/V, desired Q = 100 lpm (assume Q = V), the volume of tank is 190 liters t = 4.6 V/Q t = 4.6 190 liters/ 100 lpm t = 8.74 mins 9 mins Six individual surgical masks were tested during the experiment. Three masks were unsealed and three masks were sealed. After the mask was secured on the manikin head and placed in the chamber, testing of the three different size aerosols was conducted until the filtration efficiency for each size was determined. An unsaturated diffusion dryer was used for each particle size test.

PAGE 35

26 Protocol Each trial was conducted using the following procedures: 1. Unsealed testing – the nose clip was formed to the nose. The surgical mask was secured to the manikin by tying the double tie straps behind the head. 2. Sealed testing – silicone sealant was applied to the inner edge of the surgical mask. The nose clip was formed to the nose and the mask was secured to the head by tying the double straps. A second layer of silicone was applied to the outside edge of the mask and to the face contact point to provide a complete seal. 3. Once the manikin was in the chamber the bottom door panel was installed (top panel in place). Eight clamps were used to tightly secure the panel in place, and they were sealed with tape along the edges to prevent leakage. 4. The magnahelic gauge was monitored throughout the experiment to ensure that there was no air leakage in the chamber. 5. The brass “T” connection ports with Tygon tubing running through were sealed with Teflon tape. The Tygon tube connecting the Kr-85 and chamber was also sealed with Teflon tape. 6. A 30 minute background check was conducted by operating the lower exhaust pump (9 lpm), the mask pump (56.7 lpm) and the LASAIR pump (28.32 lpm). Background readings were conducted with all components in place except the nebulizer. 7. After the 30 minutes were complete an additional 10 mins were monitored to determine and record background levels outside the mask.

PAGE 36

27 8. Once completed, the bypass valves were switched to conduct and record the inside concentration levels after an additional 10 minutes of monitoring. 9. The nebulizer was then turned on and allowed to generate PSL aerosols for 15 minutes to reach maximum concentration. Once maximum concentration was reached, 10 min samples were conducted to determine concentration levels outside the mask. 10. The bypass valves were switched to record inside concentration levels and the instrument was allowed to run for 1 minute to clean out residual particles in the line. 11. After 1 minute, a 10 minute sample was taken to determine inside concentration. 12. Measuring the inside and outside concentration levels continued in this fashion until five tests were completed for each trial. Alternating from inside to outside measurements provided a good and consistent concentration ratio throughout the experiment. 13. These procedures were repeated for each particle size and for each mask. If back to back particle size tests were run, the background levels were measured for one hour prior to testing. 14. At the conclusion of each trial the nebulizer was shut off, disassembled and cleansed using soapy water, distilled water, and a wire brush. The efficiency of the surgical mask was determined by first subtracting the background levels from the resulting concentrations inside and outside the mask. The following equation was used to calculate the efficiency:

PAGE 37

28 Efficiency = ((Concentration out – Concentration in)/ Concentration out) 100 The resulting value is the efficiency of the mask. An efficiency of 20% indicated that there was 80% penetration through the mask. The major materials and components of the experiment are presented in Appendix 1.

PAGE 38

29 Results The results of the six respirator tests are presented separately and the filtration efficiencies are analyzed for each individual mask, by aerosol particle size, and sealing status. The results of the individual tests are listed in Tables 3 8. Table 3 lists the efficiencies per trial for the unsealed surgical masks along with the standard deviation and average efficiencies. Table 4 lists the efficiencies per trial for the sealed surgical masks along with the standard deviation and average efficiencies. Figure 21 is a graph illustrating the efficiency per trial at 0.5 um unsealed and Figure 22 illustrates the efficiency with a sealed mask. Table 3. Efficiency (%) for Unsealed Mask at 0.5 um Respirator # Trial I Trial II Trial III Trial IV Trial V SD Avg 1 2.80 4.95 2.39 4.71 4.30 1.16 3.83 2 18.55 23.17 17.63 26.85 26.85 4.40 22.61 3 40.68 33.27 48.44 47.57 49.13 6.80 43.82 Table 4. Efficiency (%) for Sealed Mask at 0.5 um Respirator # Trial I Trial II Trial III Trial IV Trial V SD Avg 4 46.94 47.8045.6945.6952.83 2.96 47.79 5 43.47 43.2752.3034.4137.92 6.78 42.27 6 30.16 54.5051.0155.2064.47 12.70 51.07

PAGE 39

30 0 10 20 30 40 50 60 70 80 12345 Trial NumberPercent Efficiency Respirator 1 Respirator 2 Respirator 3 Figure 21. Efficiency comparison of 0.5 um particles by unsealed mask 0 10 20 30 40 50 60 70 80 12345 Trial NumberPercent Efficiency Respirator 4 Respirator 5 Respirator 6 Figure 22. Efficiency comparison of 0.5 um particles by sealed mask Tables 5 and 6 list the efficiencies per trial for the unsealed and sealed surgical masks along with the standard deviation and average efficiencies for 1.0 um. Figure 23 is a graph illustrating the efficiency per trial at 1.0 um unsealed and Figure 24 illustrates the efficiency with a sealed mask.

PAGE 40

31 Table 5. Efficiency (%) for Unsealed Mask at 1.0 um Respirator # Trial I Trial II Trial III Trial IV Trial V SD Avg 1 50.00 57.14 61.37 59.79 63.19 5.15 58.30 2 74.42 73.34 70.84 68.00 69.23 2.70 71.17 3 71.65 75.92 75.67 77.71 76.98 2.35 75.58 Table 6. Efficiency (%) for Sealed Mask at 1.0 um Respirator # Trial I Trial II Trial III Trial IV Trial V SD Avg 4 73.68 68.70 70.00 72.34 70.59 1.96 71.06 5 81.34 85.50 80.36 86.01 87.89 3.22 84.22 6 68.09 73.85 73.84 77.05 77.40 3.74 74.05 0 10 20 30 40 50 60 70 80 90 12345 Trial NumberPercent Efficiency Respirator 1 Respirator 2 Respirator 3 Figure 23. Efficiency comparison of 1.0 um particles by unsealed mask

PAGE 41

32 0 10 20 30 40 50 60 70 80 90 100 12345 Trial NumberPercent Efficiency Respirator 4 Respirator 5 Respirator 6 Figure 24. Efficiency comparison of 1.0 um particles by sealed mask Tables 7 and 8 list the efficiencies per trial for the unsealed and sealed surgical masks along with the standard deviation and average efficiencies. Figure 25 is a graph illustrating the efficiency per trial at 2.0 um unsealed and Figure 26 illustrates the efficiency with a sealed mask. Table 7. Efficiency (%) for Unsealed Mask at 2.0 um Respirator # Trial I Trial II Trial III Trial IV Trial V SD Avg 1 45.67 55.34 59.28 70.01 64.07 9.19 58.87 2 77.88 79.75 74.90 82.48 83.33 3.43 79.67 3 68.71 66.12 67.16 66.25 67.58 1.06 67.16 Table 8. Efficiency (%) for Sealed Mask at 2.0 um Respirator # Trial I Trial II Trial III Trial IV Trial V SD Avg 4 67.94 69.29 71.80 64.77 74.97 3.86 69.76 5 80.25 81.67 81.70 80.83 81.76 0.68 81.24 6 83.60 85.14 84.80 86.09 86.66 1.19 85.26

PAGE 42

33 0 10 20 30 40 50 60 70 80 90 12345 Trial NumberPercent Efficiency Respirator 1 Respirator 2 Respirator 3 Figure 25. Efficiency comparison of 2.0 um particles by unsealed mask 0 10 20 30 40 50 60 70 80 90 100 12345 Trial NumberPercent Efficiency Respirator 4 Respirator 5 Respirator 6 Figure 26. Efficiency comparison of 2.0 um by sealed mask

PAGE 43

34 Discussion The purpose of this experiment was to assess filtration efficiency resulting from leakage around the surgical mask and to determine if the efficiency was different for sealed and unsealed masks using 85 lpm from the NIOSH certification methods. The filtration efficiency was then compared to FDA methods. A JMP statistical software program was used to generate a General Linear Model that was used to analyze the data. The analysis evaluated the effects for the following: seal vs unsealed, particle size, and trials. The independent variable was efficiency. The fixed effects were the seal, particle size, and trial. The random effects were the masks themselves. We also examined the interaction between particle sizes and seal status. The fixed effect tests revealed that for sealed vs unsealed the results were statistically significant (p < 0.0001). Tests for particle size also were statistically significant (p < 0.0001). The random effects indicate that there was a statistically significant difference between mask #1 as compared to masks #2 and # 3. Masks #2 and #3 were not significantly different. The test for interacti on of seal status and particle size were statistically significant (p = 0.0006). The test for trials indicated that there was no statistical difference among trials (p = 0.2213). Tukey’s Honestly Significant Difference (HSD) test is a multiple comparison test and was conducted to compare each of the particle sizes to each other. The test revealed that efficiencies of the 1um and 2um particle sizes were not statistically different from each other. The test revealed that there is a statistical difference with the 0.5 um as

PAGE 44

35 compared to the 1 and 2 um sizes. A Tukey HSD test was also conducted to analyze the interaction of particle size and seal status. The results revealed that 1 and 2 um were similar when sealed and 1 and 2 um were similar when unsealed. For both sealed and unsealed conditions there was a significant difference for 0.5 um compared to 1 and 2 um particles. The manufacturer of the tested surgical mask claimed that the mask provided a PFE of >95% for 0.1 um particles sizes. The manufacturer indicated that the test was conducted using a particle challenge study based on filtration efficiency measured using the mass median aerodynamic diameter of particles and using the 28 lpm flow rate. This research experiment was conducted using 85 lpm air flow rate that is specified in the NIOSH certification testing method. It represents the breathing rate at moderate to heavy work load conditions. Trials were conducted with the surgical masks unsealed to the manikin head and tested using three monodispersed PSL particle sizes with diameters of 0.5 um, 1.0 um and 2.0 um. Trials were also conducted with the surgical masks secured to the manikin head and sealed with silicone along the edges of the mask and face. During each trial five tests were conducted and monitored to identify the concentration levels during the trial and to indicate the efficiency throughout the trial. The standard deviation and average concentrations for the trials were determined. As expected the results were quite consistent throughout the trials indicating that the sealed masks were 23% more efficient than unsealed masks at 0.5 um, 8% at 1.0 um and 10% at 2.0 um. The results of the 0.5 um unsealed masks tests were associated with the widest variability and the highest potential for leakage. The average efficiency ranged from 3.8% to 43.8%. However, the results were remarkably consistent when the sealed mask efficiency was

PAGE 45

36 evaluated at 0.5 um, where the efficiency ranged from 42.3% to 51%. The 0.5 um particles had the ability to follow the air flow patterns and enter the mask through gaps left by a non-tight fitting mask. The surgical masks had tie straps that were tightened based on an individuals comfort level as opposed to a person donning a NIOSH approved filtering facepiece device where the straps are elastic and self tightening. An evaluation of respirator test # 1 revealed the mask was tightened as it would be in real life and actual use. The lower strap was tied behind the neck and the top strap was tied on the top portion of the skull. The results under this configuration were 3% efficiency; that is, 97% penetration of 0.5 um diameter particles. The head was slightly smaller than an average sized head and this securing method provided a loose fit and there were visible gaps on the top section and under the chin. Crossing the straps provided a tighter fit and this configuration was used for all tests thereafter. Sealing the mask resulted in improvement of the efficiency by up to 40%. The faceseal edges were sealed and the aerosols were forced to enter the mask through the filter instead around the edges. Figure 27 was the configuration for respirator #1 and Figure 28 was the configuration for the other testing. Figure 27. Unsealed mask under normal use Figure 28. Sealed mask with crossed straps

PAGE 46

37 Table 9. Average Efficiency Comparison of Unsealed and Sealed Mask at 0.5 um Surgical Masks Unsealed Mask % Efficiency Sealed Mask % Efficiency Respirator # 1 vs 4 3.83 47.78 Respirator # 2 vs 5 22.6 42.27 Respirator # 3 vs 6 43.81 51 0 10 20 30 40 50 60 123 Respirator ComparisonPercent Efficiency (%) Unsealed Mask Avg Efficiency Sealed Mask Avg Efficiency Section 1: respirator 1 vs 4; Section 2: resp irator 2 vs 5; Section 3: respirator 3 vs 6 Figure 29. Unsealed vs. Sealed Mask Comparison at 0.5 um Table 10. Average Percent Difference for Unsealed vs. Sealed Mask at 0.5 um 0.5 um Comparison Percen t Avg % unsealed 23.41 Avg % sealed 47.02 Difference of efficiencies 23.61 Figures 21 through 26 plot the efficiencies when the masks were sealed and unsealed. The results clearly show the high variability in efficiency when the masks were unsealed and also indicated that the best efficiency through the filter medium was 51%. This efficiency is 44% less efficient than claimed by the manufacturer when using challenge particles that were 0.1 um under PFE testing methods. The smaller particles,

PAGE 47

38 0.5 um, were able to enter the breaks in the mask at a higher rate because these particles tend to follow the air movement very closely. They were too small for collection by impaction and too large for collection by diffusion. Figure 30 indicated that the sealed masks were 23% more efficient on average than unsealed masks. While the 0.5 um particles followed the airflow patterns the larger 1.0 um and 2.0 um particles were more affected by inertia. The particles impact on the filter more readily because they do not follow the air flow as easily and in turn are captured by the filter medium. The sealing of the mask allowed for determination of the actual efficiency of filtering material. The average efficiency increased approximately 8% from unsealed to sealed masks at 1.0 um and 10% at 2.0 um. Table 11. Average Efficiency Comparison of Unsealed and Sealed Mask at 1.0 um Surgical Masks Unsealed Mask % Efficiency Sealed Mask % Efficiency Respirator # 1 vs 4 58.29 71.06 Respirator # 2 vs 5 71.16 84.22 Respirator # 3 vs 6 75.58 74.04

PAGE 48

39 0 10 20 30 40 50 60 70 80 90 123 Respirator ComparisonPercent Efficiency (%) Unsealed Mask Avg Efficiency Sealed Mask Avg Efficiency Section 1: respirator 1 vs 4; Section 2: resp irator 2 vs 5; Section 3: respirator 3 vs 6 Figure 30. Unsealed vs. Sealed Mask Comparison at 1.0 um Table 12. Average Efficiency Comparison of Unsealed and Sealed Mask at 1.0 um 1.0 um Comparison Percent Avg % unsealed 68.34 Avg % sealed 76.44 Difference of efficiencies 8.10 Table 13. Average Efficiency Comparison of Unsealed and Sealed Mask at 2.0 um Surgical Masks Unsealed Mask % Efficiency Sealed Mask % Efficiency Respirator # 1 vs 4 58.87 69.75 Respirator # 2 vs 5 79.66 81.24 Respirator # 3 vs 6 67.16 85.26

PAGE 49

40 0 10 20 30 40 50 60 70 80 90 123 Respirator ComparisonPercent Efficiency (%) Unsealed Mask Avg Efficiency Sealed Mask Avg Efficiency Section 1: respirator 1 vs 4; Section 2: resp irator 2 vs 5; Section 3: respirator 3 vs 6 Figure 31. Unsealed vs. Sealed Mask Comparison at 2.0 um Table 14. Average Efficiency Comparison of Unsealed and Sealed Mask at 2.0 um 2.0 um Comparison Percent Avg % unsealed 68.56 Avg % sealed 78.75 Difference of efficiencies 10.19 The data presented in table 14 show the average efficiencies for all particle sizes and for sealed and unsealed configurations. For unsealed masks the data indicated the surgical masks were approximately 45% more efficient for particles with a diameter of 1.0 & 2.0 um as compared to 0.5 um diameter. For sealed masks the data indicated the surgical masks were approximately 30% more efficient for particles with a diameter of 1.0 um and 2.0 um as compared to particles of 5.0 um diameter. Table 15. Average efficiency compared to particle diameter Diameter Unsealed Sealed um Avg % Efficiency Avg % Efficiency 0.5 23.41 47.02 1 68.34 76.74 2 68.56 78.75

PAGE 50

41 Figure 32. Average Efficiency for Unsealed vs Sealed Masks Compared by Particle Size 0.5 1.0 2.0

PAGE 51

42 Conclusions Analysis of the data indicated that the first hypothesis, which stated that the filtration efficiencies were not different between sealed and unsealed surgical masks, is rejected. Analysis indicates that the second hypothesis, which stated that the filtration efficiencies of surgical masks were gr eater than those approved by NIOSH for N-95 respirators, is also rejected. Surgical masks are more appropriate for droplets of larger size such as droplets resulting from sneezing and coughing. Respirator s require a >95% filtration efficiency and the surgical mask maximum average efficiencies while sealed were 47% for 0.5um, 76% for 1.0 um and 78% for 2.0 um. The FDA PFE testing methods can not be compared to NIOSH testing methods. Based on the data healthcare workers should not use surgical masks as personal protective equipment, instead NIOSH approved respirators, such as the N-95 filtering face piece device, are more appropriate for protection against viruses as recommended by the CDC and OSHA. Providing a patient with a surgical mask to capture the larger droplets is a good practice. The limitations of this study include the fact that the air flow was constant instead of a pulsating flow rate simulating natural breathing rate. A constant air flow provides consistent results. Under NIOSH testing methods 20 respirators are tested. Systematic errors associated with this test include aerosol wall losses and instrument calibration. Based on the results of this study, recommendations for future research include: Conduct human testing of the surgical masks in the USF Breathing Lab.

PAGE 52

43 Study and compare the efficiencies of various manufactured surgical masks such as, but not limited to, ear loop masks, masks without formable nose clips, and different double strap tie-on surgical masks. Conduct a similar study with particles ranging from 0.1 um to 0.3 um PSL aerosol. These particle sizes are closer to the sizes of droplet nuclei containing viruses. Conduct studies using a manikin head of normal size and shape.

PAGE 53

44 References Balazy, A., Toivola, M., Adhikari, A., Siva subramani, S. K., Reponen, T., Grinshpun, S. A., (2006a). Do N95 respirators provide 95% protection against airborne viruses, and how adequate are surgical masks? Am. J. Infect. Control, 34: 51-7. Balazy, A., Toivola, M., Ropenen, T., Podgorski, A., Zimmer, A., Grinshpun, S. A., (2006b). Manikin based performance evaluation of N95 filtering facepiece respirators challenged with nanoparticles. Ann. Occup. Hyg., 50: 259-69. Brosseau, L. M., Evans, J. L., Ellenbecker, M. J. and Fildstein, M. L., (1989), Collection efficiency respirator filters challenged with monodispersed aerosols. Am. Ind. Hyg. Assoc. J. 50: 544-549. Center for Disease Control and Prevention: Interim guidance for the use of masks to control influenza transmission. Retrieved 09/24/09, from http://www.cdc.gov/flu/professionals/infectioncontrol/maskguidance.htm. Fleeger, A., Lillquist, D., (2002). Industrial Hygiene Reference & Study Guide. 1st Edition. Virginia: AIHA Press. Grinshpun, S. A., Haruta, H., Eninger, R. M., Reponen, T., McKay, R. T., Lee, S., (2009). Performance of an N95 filtering facepiece particulate respirator and a surgical mask during human breathing: two pathways for particle penetration. The Journal of Occupational and Environmental Hygiene; 6: 593-603. Hinds, W. C. (1999). Aerosol Technology: properties, behavior, and measurement of airborne particles. 2nd Edition. New York: Wiley –Interscience. Huang, J., Huang, V., (2007). Evaluation of the efficiency of medical masks and the creation of new medical masks. The Journal of International Medical Research; 35:213-223. Krypton, (2005). Human Health Fact Sheet, Argonne National Laboratory, EVS Lee, B. U., Yermakov, M. and Grinshpun, S.A. (2005). Filtering efficiency of N95 and R95 type facepiece respirators, dust-mist facepiece respirators, and surgical masks operating in unipolarly ionized indoor air environments. Aerosol and Air Quality Vol. 5, No.1, pp. 25-38. Lee, S. -A., Grinshpun, S. A., and Reponen, T ., (2007). Respiratory performance offered

PAGE 54

45 by N95 respirators and surgical masks: human subject evaluation with NaCl aerosol representing bacterial and viral particle size range. Ann. Occup. Hyg., 1-9. Marusyk, R., Questions and answers. Retrieved 10/19/09, from https:www.primed.ca. National Institute for Occupational Safety and Health (NIOSH). (1997) 42 CFR 84 respiratory protective devices: final rules and notice. Federal register 60: 110. US Centers for Disease Control and Prevention. NIOSH National Personal Protective Tec hnology Laboratory. (2007) Procedure No. TEB-APR-STP-0059 NIOSH Science Blog. Retrieved 10/19/09, from https://www.cdc.gov/niosh/blog. Oberg, T., Brosseau, L., (2008). Surgical mask filter and fit performance. Am. J. Infect. Control ; 36: 276-82. Occupational Safety and Health Administra tion (2007). Pandemic influenza preparedness and response guidance for healthcare workers and healthcare employers. Plog, B. A., (2002). Fundamentals of Industrial Hygiene 5th Edition. Chicago: National Safety Council. Rawson, D., US Department of Health a nd Human Services. (2003). The basics of surgical mask selection. Retrieved 10/20/09, from http://www/hhs.gov/pandamicflu/plan/sup4.html#modes. 3M Occupational Health and Environmental Safety Division: “Respirators and surgical masks: a comparison”, December 2005. 3M Occupational Health and Environmental Safety Division: “Regulation Update – 42 CFR 84”. Number 18, August 1995.

PAGE 55

46 Appendices

PAGE 56

47 Appendix 1: Major Materials and Components of the Experiment Component Manufacturer Specifications Comments Double Strap Tie-On Surgical Mask -----------------Three layer fabric, metal nose clip Six independent masks used Manikin Head Nasco: Life Form Products, Fort Atkinson, WI “Airway Larry” Airway Management Trainer LF03699U Contains Dry Natural Rubber Nitrogen Compressed Air AIRGAS (www.airgas.com) Operated at 20 PSI, 0.2 um fiberglass filter in 45 mm holder in line Equipped with Harris Regulator, Model 92-250 Collison Nebulizer BGI Inc., Waltham, MA 3-Jet stainless steel Q = 6 LPM; Operated at 20 PSI Polystyrene Latex Beads (0.05 um) Polyscientific, Inc Warrington, PA Geometric mean: 0.465 um, std deviation: 0.01 um Monodispersed Polystyrene Latex Beads (1.0 um) Polyscientific, Inc Warrington, PA Geometric mean: 0.989 um, std deviation: 0.01 um Monodispersed Polystyrene Latex Beads(2.0 um) Polyscientific, Inc Warrington, PA Geometric mean: 1.826 um, std deviation: 0.01 um Monodispersed Diffusion Dryer ATI, Inc Owings Mills, MD Length 11.1 in Diameter: 2.23 in Model DD250; Manufactured April 2008 Changed out for every test size Kr-85 (Krypton) TSI Isotope Products Laboratories Valencia, CA 10 mCi Activity: 370 mBq Source # 54-0018 Half life: 11 yrs Decay Mode: Beta Aerosol Sampling Chamber 50 gallon tank; Approx. 48” x 12.5” x 20.75” N/A Volume = 190 liters LASAIR Particle Measuring Systems, Inc; Size: 14” x 17” x 6.75” Boulder, CO Model 210 Serial #: 36071 Operates at 1 CFM Eight Channels with thresholds at: 0.2, 0.3, 0.5, 0.7, 1.0, 2.0, 3.0, 5.0 Bypass Valves Made of PVC Magnahelic Gauge Dwyer Instruments, Inc Michigan City, IN 0 – 2” H20 Exhaust Pump Environmental Monitoring Systems Model: 905CA23-097G Bottom Exhaust operated at 9 LPM Breathing Pump Emerson Electric Co. MFG # A007; Phase 1, HP 1/3; Pump #2 LR39793 Mouth port through mask operated at 57 LPM

PAGE 57

48 Appendix 2: NIOSH Title 42 CFR Part 84 Requirements

PAGE 58

49 Appendix 2: NIOSH Title 42 CFR Part 84 Requirements (Continued)

PAGE 59

50 Appendix 2: NIOSH Title 42 CFR Part 84 Requirements (Continued)

PAGE 60

51 Appendix 2: NIOSH Title 42 CFR Part 84 Requirements (Continued)

PAGE 61

52 Appendix 2: NIOSH Title 42 CFR Part 84 Requirements (Continued)

PAGE 62

53 Appendix 2: NIOSH Title 42 CFR Part 84 Requirements (Continued)

PAGE 63

54 Appendix 2: NIOSH Title 42 CFR Part 84 Requirements (Continued)

PAGE 64

55 Appendix 2: NIOSH Title 42 CFR Part 84 Requirements (Continued)

PAGE 65

56 Appendix 2: NIOSH Title 42 CFR Part 84 Requirements (Continued)

PAGE 66

57 Appendix 2: NIOSH Title 42 CFR Part 84 Requirements (Continued)

PAGE 67

58 Appendix 2: NIOSH Title 42 CFR Part 84 Requirements (Continued)

PAGE 68

59 Appendix 2: NIOSH Title 42 CFR Part 84 Requirements (Continued)

PAGE 69

60 Appendix 3: Respirator #1 Concentration Levels Background Air Efficiency: ((Conc outConc in)/Conc out) x 100 Size ( m) Out In Size ( m) I II III IV V SD Avg 0.5 379.6 158.8 0.5 2.80 4.95 2.39 4.71 4.30 1.16 3.83 Outside Concentration w/ 0.5 um PSL Mask not Sealed True Outside: PSL Concentration Background Size ( m) I II III IV V I II III IV V 0.5 350000 400000 410000 420000 460000 349620.4 399620.4 409620.4 419620.4 459620.4 Inside Concentration w/ 0.5 um PSL Mask not Sealed True Inside: PSL Concentration Background Size ( m) I II III IV V I II III IV V 0.5 340000 380000 400000 400000 440000 339841.2 379841.2 399841.2 399841.2 439841.2 Background Air Efficiency: ((Conc outConc in)/Conc out) x 100 Size ( m) Out In Size ( m) I II III IV V SD Avg 1 33.9 7.53333 1 50.00 57.14 61.37 59.80 63.19 5.15 58.30 Outside Concentration w/ 1.0 um PSL Mask not Sealed True Outside: PSL Concentration Background Size ( m) I II III IV V I II III IV V 1 240000 280000 250000 190000 150000 239966.1 279966.1 249966.1 189966.1 149966.1 Inside Concentration w/ 1.0 um PSL Mask not Sealed True Inside: PSL Concentration Background Size ( m) I II III IV V I II III IV V 1 120000 120000 96573.8 76383.9 55204.8 119992.47 119992.5 96566.27 76376.37 55197.27

PAGE 70

61 Appendix 3: Respirator #1 Concentration Levels (Continued) Background Air Efficiency: ((Conc outConc in)/Conc out) x 100 Size ( m) Out In Size ( m) I II III IV V SD Avg 2 1.3 0.6 2 45.67 55.34 59.28 70.00 64.07 9.19 58.87 Outside Concentration w/ 2.0 um PSL Mask not Sealed True Outside: PSL Concentration Background Size ( m) I II III IV V I II III IV V 2 5640 7909.7 9000.2 11046.6 11548.9 5638.7 7908.4 8998.9 11045.3 11547.6 Inside Concentration w/ 2.0 um PSL Mask not Sealed True Inside: PSL Concentration Background Size ( m) I II III IV V I II III IV V 2 3064 3532.4 3664.9 3313.6 4149.1 3063.4 3531.8 3664.3 3313 4148.5

PAGE 71

62 Appendix 4: Respirator #2 Concentration Levels Background Air Efficiency: ((Conc outConc in)/Conc out) x 100 Size ( m) Out In Size ( m) I II III IV V SD Avg 0.5 229.2 48.1 0.5 18.55 23.17 17.63 26.85 26.85 4.40 22.61 Outside Concentration w/ 0.5 um PSL Mask not Seal ed True Outside: PSL Concentration Background Size ( m) I II III IV V I II III IV V 0.5 700000 690000 680000 670000 670000 699770.8 689770.8 679770.8 669770.8 669770.8 Inside Concentration w/ 0.5 um PSL Mask not S ealed True Inside: PSL Concentration Background Size ( m) I II III IV V I II III IV V 0.5 570000 530000 560000 490000 490000 569951.9 529951.9 559951.9 489951.9 489951.9 Background Air Efficiency: ((Conc outConc in)/Conc out) x 100 Size ( m) Out In Size ( m) I II III IV V SD Avg 1 19.6 14.6 1 74.42 73.34 70.84 68.00 69.23 2.70 71.17 Outside Concentration w/ 1.0 um PSL Mask not Sealed True Outside: PSL Concentration Background Size ( m) I II III IV V I II III IV V 1 430000 450000 480000 500000 520000 429980.4 449980.4 479980.4 499980.4 519980.4 Inside Concentration w/ 1.0 um PSL Mask not Sealed True Inside: PSL Concentration Background Size ( m) I II III IV V I II III IV V 1 110000 120000 140000 160000 160000 109985.4 119985.4 139985.4 159985.4 159985.4

PAGE 72

63 Appendix 4: Respirator #2 Concentration Levels (Continued) Background Air Efficiency: ((Conc outConc in)/Conc out) x 100 Size ( m) Out In Size ( m) I II III IV V SD Avg 2 1.2 2.9 2 77.88 79.75 74.90 82.48 83.33 3.43 79.67 Outside Concentration w/ 2.0 um PSL Mask not Sealed True Outside: PSL Concentration Background Size ( m) I II III IV V I II III IV V 2 2090.9 3128.3 3290.3 3947.2 3831 2089.7 3127.1 3289.1 3946 3829.8 Inside Concentration w/ 2.0 um PSL Mask not Sealed True Inside: PSL Concentration Background Size ( m) I II III IV V I II III IV V 2 465.1 636 828.3 694.3 641.5 462.2 633.1 825.4 691.4 638.6

PAGE 73

64 Appendix 5: Respirator # 3 Concentration Levels Background Air Efficiency: ((Conc outConc in)/Conc out) x 100 Size ( m) Out In Size ( m) I II III IV V SD Avg 0.5 686.3 52.3 0.5 40.68 33.27 48.44 47.57 49.13 6.80 43.82 Outside Concentration w/ 0.5 um PSL Mask not Sealed True Outside: PSL Concentration Background Size ( m) I II III IV V I II III IV V 0.5 540000 600000 660000 630000 610000 539313.7 599313.7 659313.7 629313.7 609313.7 Inside Concentration w/ 0.5 um PSL Mask not Sealed True Inside: PSL Concentration Background Size ( m) I II III IV V I II III IV V 0.5 320000 400000 340000 330000 310000 319947.7 399947.7 339947.7 329947.7 309947.7 Background Air Efficiency: ((Conc outConc in)/Conc out) x 100 Size ( m) Out In Size ( m) I II III IV V SD Avg 1 28.4 9.8 1 71.65 75.92 75.67 77.71 76.98 2.35 75.59 Outside Concentration w/ 1.0 um PSL Mask not Sealed True Outside: PSL Concentration Background Size ( m) I II III IV V I II III IV V 1 99740.7 160000 180000 200000 160000 99712.3 159971.6 179971.6 199971.6 159971.6 Inside Concentration w/ 1.0 um PSL Mask not Sealed True Inside: PSL Concentration Background Size ( m) I II III IV V I II III IV V 1 28283.1 38531 43788.4 44588 36837.2 28273.3 38521.2 43778.6 44578.2 36827.4

PAGE 74

65 Appendix 5: Respirator # 3 Concentration Levels (Continued) Background Air Efficiency: ((Conc outConc in)/Conc out) x 100 Size ( m) Out In Size ( m) I II III IV V SD Avg 2 1.5 4.4 2 68.71 66.12 67.16 66.25 67.58 1.06 67.16 Outside Concentration w/ 2.0 um PSL Mask not Sealed True Outside: PSL Concentration Background Size ( m) I II III IV V I II III IV V 2 2383 3536.8 4491.1 5173.3 6197.8 2381.5 3535.3 4489.6 5171.8 6196.3 Inside Concentration w/ 2.0 um PSL Mask not Sealed True Inside: PSL Concentration Background Size ( m) I II III IV V I II III IV V 2 749.6 1202 1478.9 1749.8 2013.1 745.2 1197.6 1474.5 1745.4 2008.7

PAGE 75

66 Appendix 6: Respirator # 4 Concentration Levels Background Air Efficiency: ((Conc outConc in)/Conc out) x 100 Size ( m) Out In Size ( m) I II III IV V SD Avg 0.5 440.83 41.43 0.5 46.94 47.80 45.69 45.69 52.83 2.96 47.79 Outside Concentration w/ 0.5 um PSL Mask Sealed True Outside: PSL Concentration Background Size ( m) I II III IV V I II III IV V 0.5 660000 690000 700000 700000 700000 659559.17 689559.2 699559.2 699559.2 699559.2 Inside Concentration w/ 0.5 um PSL Mask Sealed True Inside: PSL Concentration Background Size ( m) I II III IV V I II III IV V 0.5 350000 360000 380000 380000 330000 349958.57 359958.6 379958.6 379958.6 329958.6 Background Air Efficiency: ((Conc outConc in)/Conc out) x 100 Size ( m) Out In Size ( m) I II III IV V SD Avg 1 11.2 2.4 1 73.68 68.70 70.00 72.34 70.59 1.96 71.06 Outside Concentration w/ 1.0 um PSL Mask Sealed True Outside: PSL Concentration Background Size ( m) I II III IV V I II III IV V 1 230000 310000 400000 470000 510000 229988.8 309988.8 399988.8 469988.8 509988.8 Inside Concentration w/ 1.0 um PSL Mask Sealed True Inside: PSL Concentration Background Size ( m) I II III IV V I II III IV V 1 60545.4 97014.7 120000 130000 150000 60543 97012.3 119997.6 129997.6 149997.6

PAGE 76

67 Appendix 6: Respirator # 4 Concentration Levels (Continued) Background Air Efficiency: ((Conc outConc in)/Conc out) x 100 Size ( m) Out In Size ( m) I II III IV V SD Avg 2 0.9 0 2 67.94 69.29 71.80 64.77 74.97 3.86 69.76 Outside Concentration w/ 2.0 um PSL Mask Sealed True Outside: PSL Concentration Background Size ( m) I II III IV V I II III IV V 2 3354.1 3947.8 4482.4 5488.6 7274.4 3353.2 3946.9 4481.5 5487.7 7273.5 Inside Concentration w/ 2.0 um PSL Mask Sealed True Inside: PSL Concentration Background Size ( m) I II III IV V I II III IV V 2 1075 1211.9 1263.8 1933.1 1820.3 1075 1211.9 1263.8 1933.1 1820.3

PAGE 77

68 Appendix 7: Respirator # 5 Concentration Levels Background Air Efficiency: ((Conc outConc in)/Conc out) x 100 Size ( m) Out In Size ( m) I II III IV V SD Avg 0.5 160.8 18 0.5 43.47 43.27 52.30 34.41 37.92 6.78 42.27 Outside Concentration w/ 0.5 um PSL Mask Sealed True Outside: PSL Concentration Background Size ( m) I II III IV V I II III IV V 0.5 690000 670000 650000 610000 580000 689839.2 669839.2 649839.2 609839.2 579839.2 Inside Concentration w/ 0.5 um PSL Mask Sealed True Inside: PSL Concentration Background Size ( m) I II III IV V I II III IV V 0.5 390000 380000 310000 400000 360000 389982 379982 309982 399982 359982 Background Air Efficiency: ((Conc outConc in)/Conc out) x 100 Size ( m) Out In Size ( m) I II III IV V SD Avg 1 28.9 5.3 1 81.34 85.50 80.36 86.01 87.89 3.22 84.22 Outside Concentration w/ 1.0 um PSL Mask Sealed True Outside: PSL Concentration Background Size ( m) I II III IV V I II III IV V 1 99699.2 170000 190000 190000 200000 99670.3 169971.1 189971.1 189971.1 199971.1 Inside Concentration w/ 1.0 um PSL Mask Sealed True Inside: PSL Concentration Background Size ( m) I II III IV V I II III IV V 1 18607.3 24645.5 37308.8 26579.9 24217.5 18602 24640.2 37303.5 26574.6 24212.2

PAGE 78

69 Appendix 7: Respirator # 5 Concentration Levels (Continued) Background Air Efficiency: ((Conc outConc in)/Conc out) x 100 Size ( m) Out In Size ( m) I II III IV V SD Avg 2 0.7 0.4 2 80.25 81.67 81.70 80.83 81.76 0.68 81.24 Outside Concentration w/ 2.0 um PSL Mask Sealed True Outside: PSL Concentration Background Size ( m) I II III IV V I II III IV V 2 2696.3 3381.4 4684.8 4890.2 5860.4 2695.6 3380.7 4684.1 4889.5 5859.7 Inside Concentration w/ 2.0 um PSL Mask Sealed True Inside: PSL Concentration Background Size ( m) I II III IV V I II III IV V 2 532.8 620 857.5 937.8 1069 532.4 619.6 857.1 937.4 1068.6

PAGE 79

70 Appendix 8: Respirator # 6 Concentration Levels Background Air Efficiency: ((Conc outConc in)/Conc out) x 100 Size ( m) Out In Size ( m) I II III IV V SD Avg 1 38.8 11.4 1 68.09 73.85 73.84 77.05 77.40 3.74 74.05 Outside Concentration w/ 1.0 um PSL Mask Sealed True Outside: PSL Concentration Background Size ( m) I II III IV V I II III IV V 1 46319.4 69612 94226.6 100000 100000 46280.6 69573.2 94187.8 99961.2 99961.2 Inside Concentration w/ 1.0 um PSL Mask Sealed True Inside: PSL Concentration Background Size ( m) I II III IV V I II III IV V 1 14779.2 18201.8 24651.8 22949.3 22600.8 14767.8 18190.4 24640.4 22937.9 22589.4 Background Air Efficiency: ((Conc outConc in)/Conc out) x 100 Size ( m) Out In Size ( m) I II III IV V SD Avg 0.5 891 162.9 0.5 30.16 54.50 51.01 55.20 64.47 12.70 51.07 Outside Concentration w/ 0.5 um PSL Mask Sealed True Outside: PSL Concentration Background Size ( m) I II III IV V I II III IV V 0.5 330000 550000 470000 380000 310000 329109 549109 469109 379109 309109 Inside Concentration w/ 0.5 um PSL Mask Sealed True Inside: PSL Concentration Background Size ( m) I II III IV V I II III IV V 0.5 230000 250000 230000 170000 110000 229837.1 249837.1 229837.1 169837.1 109837.1

PAGE 80

71 Appendix 8: Respirator # 6 Concentration Levels (Continued) Background Air Efficiency: ((Conc outConc in)/Conc out) x 100 Size ( m) Out In Size ( m) I II III IV V SD Avg 2 4.4 3.2 2 83.60 85.14 84.80 86.09 86.67 1.19 85.26 Outside Concentration w/ 2.0 um PSL Mask Sealed True Outside: PSL Concentration Background Size ( m) I II III IV V I II III IV V 2 1995.7 2628 2987 3341.2 3091 1991.3 2623.6 2982.6 3336.8 3086.6 Inside Concentration w/ 2.0 um PSL Mask Sealed True Inside: PSL Concentration Background Size ( m) I II III IV V I II III IV V 2 329.7 393 456.5 467.5 414.8 326.5 389.8 453.3 464.3 411.6

PAGE 81

72 Appendix 9: General Linear Model Statistics Sheet 1: Fit Least Squares Response Efficiency Summary of Fit R Square 0.881843 R Square Adj 0.86855 Root Mean Square Error 8.07964 Mean of Response 60.42867 Observations (or Sum Wgts) 90 Fixed Effect Tests Source Nparm DF DFDen F Ratio Prob > F Sealed 1 1 78 67.2418 <0.0001* Particle Size 2 2 78 219.0141 <0.0001* Sealed*Particle Size 2 2 78 8.1600 0.0006* Trial 4 4 78 1.4637 0.2213 Effect Details Sealed Least Square Means Table Level Least Sq Mean Std Error N 53.444889 4.5587992 Y 67.412444 4.5587992 Particle Size Least Square Means Table Level Least Sq Mean Std Error 0.5 35.231333 4.6376708 1.0 72.395000 4.6376708 2.0 73.659667 4.6376708 LSMeans Differences Tukey HSD = 0.05 Level Least Sq Mean 2 A 73.659667 1 A 72.395000 0.5 B 35.231333 Levels not connected by same letter are significantly different

PAGE 82

73 Appendix 9: General Linear Model Statistics (Continued) Sheet 1: Fit Least Squares Response Efficiency Effect Details Sealed*Particle Size Least Square Means Table Level Least Sq Mean Std Error N, 0.5 23.419333 4.8666220 N, 1.0 68.347333 4.8666220 N, 2.0 68.568000 4.8666220 Y, 0.5 47.043333 4.8666220 Y, 1.0 76.442667 4.8666220 Y, 2.0 78.751333 4.8666220 LSMeans Differences Tukey HSD = 0.05 Level Least Sq Mean Y, 2.0 A 78.751333 Y, 1.0 A B 76.442667 N, 2.0 B 68.568000 N, 1.0 B 68.347333 Y, 0.5 C 47.043333 N, 0.5 D 23.419333 Levels not connected by same letter are significantly different Trial Least Square Means Table Level Least Sq Mean Std Error 1 56.990556 4.7915206 2 59.928889 4.7915206 3 60.509444 4.7915206 4 61.430556 4.7915206 5 63.283889 4.7915206 Sheet 1: Fit Least Squares Response Efficiency Effect Details Respirator Least Square Means Table Level Least Sq Mean Std Error 1 51.920238 1.4572433 2 63.416560 1.4572433 3 65.949202 1.4572433 LSMeans Differences Student’s t = 0.05 Level Least Sq Mean 3 A 65.949202 2 A 63.416560 1 B 51.920238 Levels not connected by same letter are significantly different


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-SFE0003323
035
(OCoLC)
040
FHM
c FHM
049
FHMM
090
XX9999 (Online)
1 100
Sanchez, Erin.
0 245
Filtration efficiency of surgical masks
h [electronic resource] /
by Erin Sanchez.
260
[Tampa, Fla] :
b University of South Florida,
2010.
500
Title from PDF of title page.
Document formatted into pages; contains X pages.
502
Thesis (M.S.P.H.)--University of South Florida, 2010.
504
Includes bibliographical references.
516
Text (Electronic thesis) in PDF format.
538
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
3 520
ABSTRACT: Surgical masks are intended to be used to prevent transmission of disease from a health care worker to a patient. Often times, they are relied upon by health care workers for their own protection. In light of recent developments regarding preparation for health care worker response to global infectious diseases such as H1N1 Influenza, health care workers may experience a false sense of security when wearing surgical masks. The goal of this study was to evaluate the filtration efficiency of a double strap tie-on surgical mask. The manufacturer asserts a >95% efficiency with a 0.1 um challenge aerosol under FDA testing procedures. The NIOSH Title 42 CFR Part 84 certification criteria call for testing at a rate of 85 lpm representing a human moderate to heavy work load breathing rate. Three sizes of monodispersed aerosols (polystyrene latex beads: 0.5 um, 1.0 um, 2.0 um) were used. The specific aims were to measure the collection efficiencies of this mask for the various particle sizes. Two tests were performed. In the first, masks were affixed to a dummy head and the edges of the mask were not sealed. In the second, the edges of the masks were sealed to the head using silicone sealant, so all penetration was through the filtering material of the mask. Differences in upstream and downstream particle concentrations were measured. Thus, penetration by leakage around the mask and through the filtering material was measured. The experimental set up involved passing the aerosol from the nebulizer through a diffusion dryer and Kr-85 charge equilibrator ensuring a dry charge neutralized aerosol cloud for detection by a LASAIR particle counter. The analysis revealed that the filtration efficiency for 0.5 um particles ranged from 3% to 43% for the unsealed masks and 42% to 51% for the sealed. For 1.0 um particles, the efficiency was 58% to 75% for unsealed and 71% to 84% for sealed masks. For 2.0 um, the efficiency was 58% to 79% for unsealed masks and 69% to 85% for the sealed masks. The data were statistically significant and indicated that surgical masks were associated with very low filtration efficiency. This suggests that they may be inadequate against airborne viruses and bacteria.
590
Advisor: Yehia Hammad, Sc.D.
653
Polystyrene latex
Monodispersed aerosols
NIOSH certification tests
Chamber
Manikin
Particle counter
690
Dissertations, Academic
z USF
x Environmental and Occupational Health
Masters.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.3323