Control of particle dispersion from enclosed spaces

(1)

Dissertations in Forestry and Natural Sciences

DISSERTATIONS | ANNA KOKKONEN | CONTROL OF PARTICLE DISPERSION FROM ENCLOSED SPACES | No 328

ANNA KOKKONEN

CONTROL OF PARTICLE DISPERSION FROM ENCLOSED SPACES

PUBLICATIONS OF

THE UNIVERSITY OF EASTERN FINLAND

uef.fi

PUBLICATIONS OF

THE UNIVERSITY OF EASTERN FINLAND Dissertations in Forestry and Natural Sciences

ISBN 978-952-61-2984-6 ISSN 1798-5668

The thesis provides knowledge on effective practices to control particle dispersion from

enclosed workspaces to protect individuals working in or occupying adjacent areas from adverse particle exposure. The study was car- ried out in airborne infection isolation rooms of hospitals and renovation sites by investigat- ing air cleaning techniques within enclosures and factors promoting enclosure containment.

The findings of this thesis can be utilized as a starting point when setting consistent negative

pressure and ventilation design guidelines for contaminant containment.

ANNA KOKKONEN

UEF_Vaitoskirja_328_Anna_Kokkonen_LUMET_kansi_18_11_28.indd 1 28.11.2018 9:35:25

(2)

(3)

CONTROL OF PARTICLE DISPERSION FROM

ENCLOSED SPACES

(4)

(5)

Anna Kokkonen

CONTROL OF PARTICLE DISPERSION FROM ENCLOSED SPACES

Publications of the University of Eastern Finland Dissertations in Science and Forestry

No 328

University of Eastern Finland Kuopio

2018

Academic dissertation

To be presented by permission of the Faculty of Science and Forestry for public examination in the Auditorium SN201 in the Snellmania Building at the University

of Eastern Finland, Kuopio, on December, 14, 2018, at 12 o’clock noon

(6)

Grano Oy Jyväskylä, 2018 Editors: Pertti Pasanen

Distribution: University of Eastern Finland / Sales of publications www.uef.fi/kirjasto

ISBN: 978-952-61-2984-6 (nid.) ISBN: 978-952-61-2985-3 (PDF)

ISSNL: 1798-5668 ISSN: 1798-5668 ISSN: 1798-5676 (PDF)

(7)

Author’s address: Anna Kokkonen

University of Eastern Finland

Depart. of Environmental and Biological Sciences P.O. Box 1627

70211 KUOPIO, FINLAND email: anna.kokkonen@uef.fi

Supervisors: Research Director Pertti Pasanen, Ph.D.

70211 KUOPIO, FINLAND email: pertti.pasanen@uef.fi

Senior Researcher Marko Hyttinen, Ph.D.

70211 KUOPIO, FINLAND email: marko.hyttinen@uef.fi Reviewers: Professor Dhimiter Bello, Sc.D.

University of Massachusetts Lowell Depart. of Public Health

3 Solomont Way

LOWELL, MA 01854, USA email: Dhimiter_Bello@uml.edu Professor Jyrki Mäkelä, Ph.D.

Tampere University of Technology Depart. of Physics

P.O. Box 692

33101 TAMPERE, FINLAND email: jyrki.makela@tut.fi

Opponent: Adjunct Professor Emeritus Timo Tuomi, Ph.D.

Finnish Institute of Occupational Health P.O. Box 18

00032 HELSINKI, FINLAND email: timo.tuomi@ttl.fi

(8)

(9)

Kokkonen, Anna

Control of particle dispersion from enclosed spaces.

Kuopio: University of Eastern Finland, 2018 Publications of the University of Eastern Finland Dissertations in Science and Forestry 2018; 328 ISBN: 978-952-61-2984-6 (print)

ISSNL: 1798-5668 ISSN: 1798-5668

ISBN: 978-952-61-2985-3 (PDF) ISSN: 1798-5676 (PDF)

ABSTRACT

Inhaled airborne particles at work may cause adverse health effects, hence exposure to airborne particles needs to prevented. Moreover, the control of internal source released contaminants in the workplace is also emphasized for protecting the other persons working in or occupying adjacent areas from adverse biological and chemical exposure. The aim of the thesis was to find out practical and effective control methods for limiting the dispersion of particles from enclosed work environments: namely, the airborne infection isolation rooms (AIIRs) of hospitals and renovation sites.

The containment of AIIRs were studied by measuring air change rates of the patient room and anteroom, pressure differences, contaminant removal, and contaminant transmission during door openings and human movement. A tracer gas method was used to simulate the release of infectious agents from a patient. Dust containment of renovation sites was studied by measuring pressure differences and dust concentrations (filter samples and real-time monitoring) simultaneously from the renovation sites and adjacent areas. In addition, the effect of using on-tool local exhaust ventilation in limiting the dust dispersion was investigated. This thesis also explored the airborne particle removal efficiency of a short-term water misting method. The method was applied after the work has been completed to limit the dispersal of airborne dust that has not been captured in real-time by source controls.

Therefore, it is considered similar to a general ventilation method that dilutes the work-generated airborne dust concentrations. The airborne dust removal by short- term water misting was assessed through experimental studies of dust decay and dust removal in the controlled laboratory environment. In addition, the feasibility of the method at renovation sites was studied.

The findings of the thesis indicate that a high air change rate in AIIRs does not ensure efficient removal of infectious agents in the breathing zone of a healthcare worker (HCW). Instead, the local airflow patterns are more important. A high mean negative pressure between the AIIR and surroundings was shown to significantly limit particle transmission outside an AIIR. An anteroom aided in controlling the

(10)

particle transmission. In practice, however, the dilution is not effective enough according to the observed air change rates of anterooms studied. Generally, a HCW does not stay in the anteroom more than 2–3 minutes, thus to achieve at least 90%

contaminant removal, the minimum air change rate requirement of 40 1/h would be needed after entering the anteroom. This would enhance the total containment of an AIIR.

The results indicate that the maintenance of a continuous negative pressure over -5 Pa limits the dispersion of dust outside a renovation site. The enclosure containment was achieved by sealing the junctions between the temporary and existing partitioning structures to obtain better airtightness of an enclosure and building an airlock between the enclosure and adjacent areas. Although the renovation sites studied were at times under positive pressure, the dust dispersion outside an enclosure was not observed, when the mean target concentrations at the renovation site remained <4 mg/m³ for inhalable dust and <1 mg/m³ for respirable dust. The thesis indicates that the use of on-tool local exhaust ventilation (LEV) has an important role in dust containment if an enclosure fails to maintain negative pressure e.g., during door openings and human movement. Almost three times lower dust concentrations are expected outside an enclosure with LEV use during the dust- producing tasks.

The thesis shows, with respect to the premise of avoiding harmful wetting of surfaces, a short-term misting after a dust generating task is an effective measure to control the dispersal of airborne dust in enclosed workspaces. The method supplements on-tool source control measures since the misting substantially captures the airborne dust that has not been captured in real-time by the source controls. The results infer that the misting method could be applied in those environments where the effective air change for airborne dust removal is otherwise not feasible.

The thesis provides knowledge on effective practices to control the particle dispersion outside an enclosure, which can also be exploited in other enclosed workplaces. The findings of the present study can be utilized as a starting point when setting consistent negative pressure and ventilation design guidelines for contaminant containment.

Universal Decimal Classification: 628.511, 628.8, 697.98

CAB Thesaurus: Indoor air pollution; Air quality management; Particles; Dust; Dust control;

Work environment; Hospitals; Patient rooms; Buildings – Repair and reconstruction;

Ventilation; Air pressure; Air flow; Zoning; Spraying; Water

(11)

ACKNOWLEDGEMENTS

This thesis was carried out in the Department of Environmental and Biological Sciences, University of Eastern Finland (UEF), Kuopio in the Research group of Indoor Environment and Occupational Hygiene (IEOH). The study was funded by the Finnish Funding Agency for Technology and Innovation (grant 40239/10), Finnish Work Environment Fund (grants 112249, 118020), The Doctoral Programme in Environmental Physics, Health and Biology at UEF, and Kuopio Region Respiratory Foundation.

I am heartily grateful to my supervisors, Research Director Pertti Pasanen and Senior Researcher Marko Hyttinen for their guidance, critical comments, and support. I am deeply thankful to emeritus Professor Pentti Kalliokoski for his encouragement and advice in my early stage of this research work.

I wish to thank Professor Dhimiter Bello from University of Massachusetts Lowell and Professor Jyrki Mäkelä from Tampere University of Technology for providing their expertise and conducting the review of my thesis. I wish to thank Adjunct Professor Emeritus Timo Tuomi for accepting the invitation to act as opponent of my doctoral dissertation. I am also grateful to Joanne Enstone, M.Phil, from University of Nottingham School of Medicine for linguistic revision of my thesis.

I express my deepest gratitude to my collaborators, Senior Specialist Markku Linnainmaa and Senior Specialist Arto Säämänen for their guidance, challenging discussions, and critical comments during my study. I sincerely thank all the co- authors for their advice and contributions to the original articles. I also acknowledge all the team members of the past research projects that have been involved in my thesis. It has been a great opportunity to work with you both at the field and in the office.

I warmly thank you colleagues and friends of the Research group of IEOH. It has been my pleasure to share thoughts with you. You have made a trustful and supportative team to belong in. I also thank the colleagues of the Department of Environmental and Biological Sciences for your encouragement. A good laugh always helps.

I am grateful to my dear friends for being there and sharing life with me beyond work. I want to give my special thanks to my friend Anu for listening my bad moments with my thesis and for encouraging me to run with the project.

I wish to thank my parents for their care and support. I owe my gratitude to my dear parents-in-law for helping the everyday life when needed during this work.

Finally, as much as I enjoy my work, my family Mikko, Manu, and Unto is the best thing I could ever wish for. You are my super team.

Kuopio, October 2018 Anna Kokkonen

(12)

(13)

LIST OF ABBREVIATIONS AND SYMBOLS

AM Arithmetic mean

ACH Air changes per hour AII Airborne infection isolation AIIR Airborne infection isolation room

CDC Centers for Disease Control and Prevention CEN European Committee for Standardization CPWR Center to Protect Worker’s Rights

d^ae Aerodynamic diameter DH Department of Health, UK

EPA United States Environmental Protection Agency FTIR Fourier-transform infrared spectroscopy G Generation rate of internal contaminant source

GM Geometric mean

GSD Geometric standard deviation HCW Health care worker

HEPA High efficiency particulate air filter

HBIVCU Hospital bed integrated ventilation and cleaning unit HSE Health and Safety Executive, UK

HVAC Heating, ventilation, and air conditioning LEV Local exhaust ventilation

LOD Limit of quantification

MSAH Ministry of Social Affairs and Health, Finland NC Normalized concentration

NIOSH National Institute for Occupational Safety, USA NIST National Institute of Standards and Technology, USA NIPH Norwegian Institute of Public Health

NTP Normal pressure and temperature (1 atm, 20°C) OEL Occupational exposure limit

OSHA Occupational Safety & Health Administration, USA PAH Polycyclic aromatic hydrocarbons

PM10 Particulate matter, with aerodynamic cut size diameter smaller than 10 µm

PPE Personal protective equipment

p^pos The percentage of positive pressure periods Q^e Exhaust airflow

Qⁱ Replacement airflow RE Dust reduction efficiency RH Relative humidity of air

RRE Relative dust reduction efficiency SD Standard deviation

SFVH Swedish Association for Hospital Hygiene

(14)

SLIC Senior Labour Inspectors Committee SSI Statens Serum Institut, Denmark SVOC Semivolatile organic compound VM Ventilated mattress

WHO World Health Organization

∆p Pressure difference between contaminated and adjacent spaces

(15)

LIST OF ORIGINAL PUBLICATIONS

This thesis is based on data presented in the following articles, referred to by the Roman Numerals I-IV.

I Kokkonen A., Hyttinen M., Holopainen R., Salmi K. & Pasanen P. 2014.

Performance testing of engineering controls of airborne infection isolation rooms by tracer gas techniques. Indoor and Built Environment 23: 994−1001.

II Kokkonen A., Linnainmaa M., Säämänen A., Lappalainen V., Kolehmainen M. & Pasanen P. 2017. Evaluation of real-world implementation of

partitioning and negative pressurization for preventing the dispersion of dust from renovation sites. Annals of Work Exposures and Health 61: 681−691.

III Kokkonen A., Linnainmaa M., Säämänen A., Kanerva T., Sorvari J., Kolehmainen M., Lappalainen V. & Pasanen P. Control of dust dispersion from an enclosed renovation site into adjacent areas by using local exhaust ventilation. Submitted manuscipt.

IV Kokkonen A., Väänänen M., Säämänen A. & Pasanen P. The evaluation of short-term water misting to reduce airborne dust after renovation work.

Accepted manuscript. Annals of Work Exposures and Health, DOI:

10.1093/annweh/wxy096

(16)

(17)

AUTHOR’S CONTRIBUTION

I Tracer gas measurements were done with the guidance of senior researcher Marko Hyttinen, while the author analysed the data. Pressure difference measurements and data analyses were carried out by Rauno Holopainen and Kari Salmi. Interpreting the results and writing of the Paper were done by the author.

II The author planned the experiments together with her colleagues. The author was responsible for organising and carrying out the field work, laboratory analyses and data analysis. She did the satistical analyses for linear correlation.

Regression analyses were done with the guidance of professor Mikko Kolehmainen. Writing of the Paper was done by the author with the support of co-authors.

III The author contributed to the planning and the implementation of the field experiments. Laboratory and data analyses of the filter samples were conducted by the Finnish Institute of Occupational Health. The author was responsible for the data interpretion. She carried out the statistical analyses with help of associate professor Jouni Sorvari and professor Mikko Kolehmainen. The author wrote the Paper with supportive co-operation of the co-authors.

IV The author planned the experiments together with supervisor and she organized the experiments. She collected the samples and carried out the laboratory analyses together with the co-author Maija Väänänen. The author did the data analyses. Writing of the Paper was done by the author with the support of co-authors.

(18)

(19)

1 INTRODUCTION

Inhaled airborne particles that are released inside a working space (Fig. 1) may pose a risk for adverse health effects. Adverse health effects may be minor impairment (irreversible), occupational disease and life-threatening effects (WHO 1999). Harmful particle exposure may have acute or chronic health effects. For example, short-term exposure to quartz might cause mechanical irritation (ICSC 2008). Long-term exposure to respirable quartz is associated with various respiratory diseases, such as silicosis (NIOSH 2002), lung cancer (IARC 2012b), chronic bronchitis and chronic obstructive pulmonary disease (NIOSH 2002). Diseases caused by asbestos, such as asbestosis and mesothelioma, occur after a long latency period (WHO 2010). Asbestos is classified as carcinogenic to humans (Group 1) (IARC 2012a). Lead (dust, mist or fume) can damage the central nervous system, reproductive system, and kidneys (IARC 2006, ICSC 2006). Inorganic lead is classified as probably carcinogenic to humans (Group 2A) (IARC 2006). Creosote, which mainly consists of polycyclic aromatic hydrocarbon (PAH) compounds, is classified as probably carcinogenic to humans (Group 2A) (IARC 1987). Inhalation of particles containing fungi, viral or bacterial pathogens may cause infectious diseases.

There is also increased risk of respiratory symptoms and infections and exacerbation of asthma among occupants of damp or mouldy buildings (WHO 2009b). There are many other airborne particle contaminants associated with various health risks.

Nevertheless, the brief examples mentioned here emphasize the importance of preventing airborne particle exposure.

Fig. 1. Generic illustration of an enclosed working space as a simple box, with a single internal source generating contaminants (G) (adapted from Burgess et al.

2004). Person inside the enclosure represents a worker. In this example, air is being removed (Q^e) from the enclosure with the general exhaust system and natural infiltration is providing the replacement air (Qi) from the adjacent area. Source control measures are not used in this illustrative example.

(22)

20

In reconstruction work, workers are exposed to dust (typically) during handling and processing of building materials. The exposure depends on the characteristics of a processed material, process method, tooling, engineering control techniques, location of the dust emission, and size of the working area, ventilation and airflow patterns. During reconstruction operations (such as cutting, sawing and drilling of concrete or brick and cleaning), workers have been reported to be exposed to high concentrations of respirable (particles with aerodynamic diameter ≤ 10 µm) and inhalable (aerodynamic diameter ≤ 100 µm) dust and respirable quartz (e.g., Akbar- Khanzadeh et al. 2002 & 2010, Croteau et al. 2002 & 2004, CPWR 2005, Shepherd et al. 2009, OSHA 2012, Garcia et al. 2014). Coworkers within the renovation site may also be exposed to high dust concentrations (Woskie et al. 2002). Moreover, the dismantling and removing of building materials may cause exposure to many other potentially present hazardous agents, such as asbestos, lead, creosote or microorganisms (e.g., Rautiala et al. 1996, OSHA 2004, SLIC 2006, Järnström et al.

2009).

According to general contaminant control principles, the primary goal is to prevent the generation of contaminants inside the working space. Next are actions to minimize the release of contaminants into the workplace. The final protection measure called on is the use of personal protective equipment (PPE) when contaminant exposure of workers cannot be reduced by engineering controls. Source control may be challenging since it is not always practical, or even possible, to substitute the hazardous substance, or change the particle generation process or work practices to minimize the contaminant exposure. An example of this is in isolation rooms of hospitals where a patient is the source of contamination. Infectious agents transmitted from a patient become airborne; hence, the controls are aimed at airborne particles. In general, this involves control measures through containment and use of general ventilation or local exhaust ventilation (LEV) to remove the airborne particles released from internal sources (WHO 1999). LEV removes dust, vapors, fumes and so forth, contaminants produced by work processes or activities, and it is located adjacent of the contaminant source. Containment involves preventing the spread of contaminants by enclosing the work area from adjacent areas, e.g., in hospital settings; Airborne Infection Isolation (AII) is the isolation of patients infected with organisms spread via airborne droplet nuclei <5 µm in diameter to minimize the potential for disease transmission (such as Mycobacterium tuberculosis, measles, severe acute respiratory syndrome (SARS) and influenza) (CDC 2003). Inside the enclosure, general control principles are applied.

The thesis focuses on the control of particle contaminants in the workplaces where the protection of the other persons working in or occupying adjacent areas is crucial (e.g., in hospitals). In this thesis, particle contaminant sources located inside the enclosed working spaces (Fig. 1). Effectiveness of the control measures in limiting the particle dispersion outside from an enclosure has not been studied sufficiently in airborne infection isolation rooms (AIIRs) and renovation sites. There are few case studies (Overberger et al. 1995, Rautiala et al. 1998) conducted in the renovation sites

(23)

21 to evaluate the enclosure and its containment. Studies on particle containment with negative pressure have mainly been evaluated in terms of the effectiveness of AIIRs (Hayden et al. 1998, Rydock 2002, Rydock & Eian 2004, Tang et al. 2005, Adams et al.

2011). Nevertheless, it is still unclear which pressure difference between the enclosure and adjacent area is adequate to prevent particle dispersion. Moreover, there is an insufficient evidence base for guidelines for pressure difference and minimum negative pressure requirements in AIIRs (Hyttinen et al. 2011) and renovation sites. In addition, evaluation of the effect on containment effectiveness by using general ventilation, LEV and a short-term water misting method within enclosures is needed to improve practices in controlling dispersion of the particles in the AIIRs and renovation sites. The purpose of this thesis is explore practical and effective control methods for limiting the dispersion of particles from enclosed spaces to adjacent areas. Further, the knowledge of particle controls can be exploited in other enclosed workplaces.

The purpose of an enclosure is to protect both worker and the other people working in or occupying the adjacent areas, from adverse biological and chemical exposure. The present thesis focuses on the enclosed spaces of airborne infection isolation rooms of hospitals (AIIRs) and renovation sites. Thus, particles in this thesis cover chemical and biological agents that are released inside the enclosure, such as different types of dusts related to reconstruction works and infectious agents (viruses, bacteria, and fungi) related to both reconstruction and the hospital environment.

(24)

22

(25)

23

2 LITERATURE REVIEW

2.1 PARTICLES AND THEIR BEHAVIOUR IN THE AIR OF AN ENCLOSURE

The size and material characteristics of a particle determines its behavior in the air. For spherical particles, the diameter is usually the measure of size. For nonspherical particles, such as fibers, it is more complex to define a universal characteristic size (Kulkarni et al. 2011b). In occupational hygiene, the particle size is usually presented in terms of the aerodynamic diameter (dae) (WHO 1999). It is defined as the diameter of a hypothetical sphere of density 1 g/cm³ having the same terminal settling velocity when settling under gravity as the particle under consideration (Kulkarni et al. 2011b). Aerodynamic diameter is an appropriate property for characterizing the behavior of particles in the respiratory tract, filtration or performance of many types of engineering devices, where inertial behavior dominates (Hinds 1999, Kulkarni et al. 2011b).

Airborne particles can be classified as inhalable, thoracic and respirable fractions (Fig. 2) by their penetration of different regions of the respiratory tract under mean conditions according to European standard EN 481 (CEN 1993). Conventions are approximate to associations between aerodynamic diameter and the fractions to be collected or measured. The inhalable fraction is the mass fraction of total airborne particles that is inhaled through the nose and mouth. The thoracic fraction is that penetrating into the trachea-alveolar region, whereas the respirable fraction (or alveolar) penetrates the alveolar region of the lung. Relative to total airborne particles, 50% cut-size for the inhalable, thoracic, and respirable fractions are at an aerodynamic diameter of 100, 10, and 4.0 µm (Fig. 2, CEN 1993).

(26)

24

Fig. 2. Particle fraction conventions as percentages of total airborne particles by European standard EN 481 (CEN 1993).

Particles in workplace air are usually considered to polydisperse (Walter 2011).

Particle size distributions from many emission sources have been defined to fit the lognormal distribution (Walter 2011). The lognormal distribution has a tail for large values of the particle diameter. In occupational hygiene, mass concentration is traditionally measured to assess exposure to particles, and particles are studied in terms of mass distributions. Recently, more attention is paid to exposure to nano- sized particles, therefore the number and surface area distributions are also used to assess health effects.

In the context of occupational hygiene, an aerosol is described as particles suspended in air. Aerosols may be present in the form of airborne dust particles, bioaerosols (including e.g., viruses and bacteria), sprays, mists, smokes, and fumes (Fig. 3). Airborne particles may have interactions and reactions between gas molecules and liquid or solid particles, which then changes their size or chemical composition. For example, semivolatile organic compounds (SVOCs), such as polycyclic aromatic hydrocarbons (PAHs), may be absorbed on airborne particles.

The physical definition of dust is solid particles produced by the mechanical disintegration of material such as crushing, grinding and blasting (Seinfeld & Pandis 2006). The particles generally have irregular shapes (Kulkarni et al. 2011b) with a size range from below 1 µm up to at least 100 µm (WHO 1999). Mist is a suspension of

(27)

25 liquid droplets (>1 µm), which can be formed by condensation of supersaturated vapors or by spraying. A spray is a droplet aerosol formed by mechanical or electrostatic breakup of a liquid (Kulkarni et al. 2011b). Whereas the particle size of different dusts in construction and reconstruction workplaces have a wide range depending on the processed material and tooling (Fig. 3), infectious microorganisms in the hospital environment have smaller range variation. Microorganisms (Fig. 3) range between 0.02−0.30 µm for viruses and 0.3−10 µm for bacterial cells (Cole &

Cook 1998). The diameter of fungal spores have wider size range: 1−50 µm, most frequently 2–10 µm (Cole & Cook 1998, Elbert et al. 2007, Fröhlich-Nowoisky et al.

2009, Huffman et al. 2010).

Most infectious particles generated from human respiratory sources are primarily droplet nuclei, i.e. the airborne residuals from desiccation of suspended droplets (e.g., from coughing, sneezing, and speech) (CDC 2003). Droplet nuclei range in size from 1–5 µm (CDC 2003). Nicas et al. (2005) estimated most cough particles to be <10 µm, while Yang et al. (2007) showed three-modal size distribution of approximately 1 µm, 2 µm, and 8 µm for coughed droplets. Morawska et al. (2009) found similar modes, ≤0.8 µm and 1.8 µm, for breathing, coughing, and speech. In addition, modes of 3.5 µm and 5.5 µm were shown for speech (Morawska et al. 2009).

Frequently, in literature, particles are termed “small” and “large” with a typical cut- off size of about 1 µm considering particle properties, behavior, and measurement (Kulkarni et al. 2011b).

(28)

26

Fig. 3. Particle size ranges for aerosols particles (from Colbeck & Lazaridis 2013).

In addition to particle size, the particle motion depends on the kinetic properties of the gas (air) and external forces acting on the particles. External forces include gravitational, molecular, thermal, and electrical forces. A particle’s interaction with the surrounding gas molecules of air is based on the particle size relative to the mean free path of the gas molecules, i.e., the mean distance traveled by a molecule between successive collisions (Hinds 1999). When the probability of surrounding gas molecules striking the particle surface is high, the surrounding gas affects the behavior of the particles through a drag force (Tsuda et al. 2013). Drag force (air resistance) means the force that opposes the relative motion of a particle in air. It is a function of a particle’s velocity with respect to surrounding air. The Reynolds number is the ratio of inertial forces to viscous forces within a fluid, depending on the relative velocity between a particle and the surrounding gas. Newton’s drag (Reynolds number >1000) is important to particle motion for which the inertial effects are dominant (Hinds 1999). A particle’s inertia means its tendency to resist a change in its motion (Miller & Klonne 1997). At a high Reynolds number, the flow becomes turbulent (Hinds 1999, Kulkarni et al. 2011a). On the contrary, laminar flow around

(29)

27 a particle occurs at a low Reynolds number (<1) (Hinds 1999). In laminar flow, inertial forces are negligible compared to viscous forces, and Stokes’ law applies to particle motion. Stokes’ law combines the total resisting force that opposes the relative motion of a particle in the air on a spherical particle moving through the air (Hinds 1999).

Gravitational and inertial forces are dominant for particles of a few micrometers, while Brownian motion is significant for particles smaller than about 0.5 µm (Walter 2011). Brownian motion is the irregular, constantly changing random motion of a particle in the still air caused by collisions with surrounding gas molecules (Hinds 1999). Brownian diffusion occurs if there is a concentration gradient. The transport of particles is always from a region of higher concentration to a region of lower concentration (Hinds 1999). Diffusion transport favors short distances (Miller &

Klonne 1997). Large particles diffuse more slowly than small particles because of their large inertia and large surface area (Kulkarni et al. 2011a).

Mechanisms of particle behavior in the air often rely on the assumption that the particles are introduced in the air without an initial speed, although, in reality, particles have an initial speed depending on the particle generation process (Morawska 2006, HSE 2017). Moreover, airflow patterns within an enclosure affect the movement of particles in the air (Chen et al. 2006, Morawska 2006, Tang et al.

2006, Li et al. 2007b, Pantelic & Tham 2013). Ventilation, temperature differences, human movement and door traffic generate airflow patterns, which are also influenced by the location and shape of furniture and other obstacles and the thermal plume of a human (Tang et al. 2005, 2006 & 2013; Eames et al. 2009). Generally, airflow in rooms is considered as turbulent. Smaller particles tend to distribute uniformly within an enclosure under well-mixed airflow pattern, whereas larger particles tend to settle (Holmberg & Li 1998). According to consensus in aerosol science, particles with diameter >50 µm do not usually remain airborne for any considerable time (WHO 1999). However, particle transport and dynamics is not well understood (Nazaroff 2004, Morawska 2006). Particles smaller than 2 µm have been simulated to act similarly as common dispersion and transport properties of air (Chen et al. 2006). Moreover, particles smaller than 1 µm have been found to disperse randomly and uniformly within an enclosure due to the airflow and kinetic particle movement (Mousavi & Grosskopft 2015). In the recent study by Bivolarova et al.

(2017), the air change rate was shown to have less influence on the transport behavior of fine and coarse particles (0.7 µm and 3.5 µm, respectively) than on smaller ultrafine particles (0.07 µm). Dust particles during construction, as well as the infectious agents in the hospital environments, may be smaller than <1 µm, having negligible settling due to gravity. Thus, for practical purposes, the particle movement within the air is more important for the small particles than their sedimentation through the air (WHO 1999). Small particles from their source of generation or release will move long distances with air currents especially when the process generates rapidly moving air streams (HSE 2017).

(30)

28

2.2 NATURAL MECHANISMS OF PARTICLE REMOVAL

Particles are removed from the air by the natural mechanisms of deposition onto surfaces. Deposition occurs through sedimentation, inertial deposition, interception mechanisms and diffusion. Deposition of particles on surfaces is size dependent (He et al. 2005). The larger the particle, the higher the gravitational deposition rate (Bivolarova et al. 2017). In addition, a higher room airflow can increase turbulent particle deposition (Thatcher et al. 2002, Bivolarova et al. 2017). Sedimentation has a major role in natural removal in particle clearance for larger particles. With a higher settling velocity, they tend to settle out more rapidly due to gravity, also in turbulent conditions (Kulkarni et al. 2011a). The settling velocity of a small spherical particle can be approximated from Stokes’ law. With the increase in sedimentation velocity or particle size, the inertial effects in the air become important (Owen & Ensor 1992).

Particle inertia is the dominant mechanism for deposition from turbulent flow for particles larger than 1 µm (Hinds 1999). Inertial deposition occurs when a particle in the fluid stream is forced to change direction, and the inertia carries the particle across the flow streamlines until its momentum in that direction is depleted by fluid drag. The particle deposits onto a surface by impaction. An increase in particle mass and velocity, and the sharpness of the change in stream direction increases the probability of particle impaction on the surface (Miller & Klonne 1997). The interception mechanism in deposition is especially important for fibers (Miller &

Klonne 1997). It is not dependent on particle motion across fluid streamlines. Particle deposition due to diffusion occurs when particle attachment onto surface generates a concentration gradient near the surface. The concentration gradient causes continuous particle diffusion to the surface, leading to a gradual decay in particle concentration (Hinds 1999).

Adhesion has an effect on attachment and detachment of the particles to surfaces.

Adhesion means particles contact one another, adhere to each other, and form agglomerates. The London – van der Waals forces are the main attractive adhesion forces, which act over short distances relative to the particle dimensions (Kulkarni et al. 2011b). In addition, the particle electrostatic charge can increase the adhesive force (Owen & Ensor 1992). Air humidity may affect particle adhesion. At high humidity, water molecules are adsorbed on the particle surface. They fill the capillary spaces at the contact point (Kulkarni et al. 2011b).

(31)

29

2.3 TECHNICAL APPLICATIONS FOR REMOVAL OF PARTICLES

The technical removal of airborne particles within enclosed spaces involves air cleaning technologies such as general ventilation and dilution, airflow direction, filtration, local exhaust ventilation and wet methods.

2.3.1 Ventilation

The purpose of general ventilation is to dilute airborne contaminants within the room by introducing a well-mixed air distribution of filtered air, and to remove the contaminants. General ventilation can comprise airflow into a room by natural forces, such as thermal buoyancy and wind, or by mechanical fan force (Li et al.

2007b). Ideally, general ventilation directs contaminants towards exhaust devices via non-mixed airflow patterns (CDC 2003). Location of supply and exhaust air devices should be arranged in a way that contaminants are directed away from the workers’

breathing zones. Breathing zone means a region of the body not more than 0.3 m from the mouth and nose. If air flows from the back to the front of a person, it can generate turbulent eddies introducing the contaminated air to the worker’s breathing zone (WHO 1999). When an enclosed space has many scattered contaminant sources, general ventilation can be effective in controlling the production of relatively low concentrations of low toxicity airborne contaminants (WHO 1999). For particle contaminant, however, general ventilation is not always a sufficient control method, because air velocities in general ventilation exhausts are usually low and large particles tend to settle instead of being efficiently transported to an exhaust (Burgess et al. 2004). In addition, turbulent mixing will disperse contaminants through a large part of the air within the room (WHO 1999).

The purpose of local exhaust ventilation (LEV) is to capture and remove contaminants at source to avoid spread through the room space. LEV also aims to result in controlled and directional airflow across an emission point into a hood connected to a ductwork or vacuum system (WHO 1999). In general, a hood needs to be positioned to surround the source of emission to ensure that extraction air is moving in the same direction as air moving through the process (WHO 1999).

Capture velocities of 0.5-1.0 m/s may control small particles, while larger particles require higher velocities, around 2.5-10 m/s (WHO 1999). Capture velocity is the air velocity required around an emission source to capture the contaminant cloud and draw it into the hood (HSE 2017). A hood should also have a minimum face velocity, i.e. the mean velocity of air at the open front face of a hood, to resist the effect of workroom draughts and general air turbulence (HSE 2017). The hood type and required airflow rate depend on the source characteristics, such as the position of the emission source, emission process and generation rate, particle size, and the speed of emission release and ambient air movement (WHO 1999). Hood types include enclosing hoods, capturing hoods, receiving hoods and jet assisted hoods. An

(32)

30

enclosing hood is the most effective mechanism for capturing contaminants. In partial enclosing hood (booths), the airflow dilutes and displaces the contaminant cloud. Commonly, a minimum face velocity of 0.4 m/s is shown to be effective for contaminant containment (HSE 2017). Capturing hoods draw contaminants into a hood, while receiving hoods utilize the motion of an air stream to carry contaminants from the source (e.g., a grinding wheel) into the collection hood. A jet assisted hood utilizes a clean air stream from a local supply to sweep the emission into the exhaust hood (WHO 1999).

In capturing hoods, the capture velocity needs to be sufficiently large to overcome other air velocity patterns in the emission source in a way such that the resultant airflow is from the source into exhaust hood (Burgess et al. 2004). A capturing hood may be suitable for a contaminant cloud that does not have strong initial speed and direction (HSE 2017). The velocity of air being drawn towards the hood opening rapidly decreases with the distance for the same exhaust volume (WHO 1999). In practice, the air velocity will fall to approximately 10% of the face velocity at one hood diameter out from the face of a capturing hood (HSE 2017). Required capture velocities depend on the condition of dispersion of contaminants: e.g., contaminant release at low velocity into moderately still air may need a capture velocity of 0.5−1.0 m/s, while a release at high initial velocity into turbulent air (such as grinding) may require a velocity of 2.5−10 m/s (Burgess et al. 2004, HSE 2017). A specific capturing hood application is a low-volume/high-velocity capture systems. They are used as extractor hoods fitted on-tool (e.g., saws, masonry saws, grinders), or they can be positioned very close to the operating point (Burgess et al. 2004). They provide high capture velocities, greater than 50 m/s with airflows less than 0.02 m³/s by applying a small hood face area (Burgess et al. 2004).

2.3.2 Filtration

Air cleaners equipped with particulate air filters are used to remove airborne particles in enclosed spaces. Removal mechanisms in the filter are diffusion, interception, inertial impaction (Fig. 4), and electrostatic attraction. Filtration efficiency can vary considerably with the particle size and characteristics of a filter (Fig. 4). Particles of approximately 0.3 µm are the most difficult to capture (Nazaroff 2004). An increase in particle size enhances filtration by the mechanisms of interception and inertial impaction, whereas a decrease in particle size increases particle collection by Brownian diffusion (Raynor et al. 2011). In a diffusion mechanism, a small particle under Brownian motion collides with a filter fibre and adheres to it. Interception occurs when a particle follows the air streamline within one particle radius of the fibre surface. Interception is not dependent on flow velocity (Hinds 1999). Interception and diffusion are the only important mechanisms near the minimum efficiency (Hinds 1999, Fig. 4). An inertial impaction mechanism occurs when a particle deviates from the air stream due to its inertia. The importance of

(33)

31 inertial impaction mechanism increases when the particle size and air velocity increases (Raynor et al. 2011). In the event of a sufficiently large settling velocity of a particle, it may deviate from the air stream. However, this gravitational settling mechanism is significant only for large particles at low filtration velocities (Raynor et al. 2011). Electrostatic deposition is useful for intermediate sized particles, where interception is the main mechanical deposition mechanism (Dunnett 2013). In this region, particles in the size range 0.05−0.5 µm (Hinds 1999), having weak Brownian motion and inertia, result in minimum filtration efficiency (Raynor et al. 2011, Dunnett 2013). With an increase in charge on either particles or fibers, the collection efficiency can substantially be enhanced (Hinds 1999). In addition, electrostatic filters are more efficient at low filtration velocities (Dunnett 2013). Enhancing the filter efficiency by charging is important for air-cleaning applications that require high efficiency and low pressure drops (Hinds 1999). The pressure drop is the resistance to the air flowing through a filter, which is directly proportional to the thickness and packing density of the filter at a given filter face velocity. Both collection efficiency and pressure drop increase in fibrous filters with particle loading (Hinds 1999).

Fig. 4. Theoretical collection efficiencies of aerosol particle collection mechanisms (diffusion, interception, impaction, sedimentation) calculated from the single fiber efficiency theory for a fibrous filter (horizontal filter surface) with a fiber diameter of 2 µm, operating at an airflow velocity of 0.10 m/s. Diffusion-interception interaction (DR) is the particle collection due to an enhancement of interception by particle diffusion. Total filter efficiency shows the collection efficiency of the filter due to all mechanisms combined (from Hinds 1999).

(34)

32

2.3.3 Water suppression

The removal effect of water suppression is based on coagulation. Coagulation is the particle growth resulting from the collision of particles with each other because of their velocity difference, which adhere to form larger particles. Coagulation leads a decrease in particle number concentration, but an increase in mean particle size (Hinds 1999). If any removal mechanisms are present, there is a change in the mass concentration (Hinds 2011). According to an indoor air quality model, the coagulation effect for ultrafine particles (<0.1 µm) has shown to be significant when the total concentration is higher than 10⁴ cm^-3 (Hussein et al. 2009).

Large water droplets settle through small airborne dust particles by the mechanisms of inertia and interception capturing dust particles with high collision efficiency (Hinds 1999, Hinds 2011). Water droplets, however, do not come to contact with dust particles of all different sizes (Charinpanitkul & Tanthapanichakoon 2011).

Smaller dust particles are able to escape from interactions with water droplets since they are too small to be collected by either inertia of interception mechanisms (Seinfeld & Pandis 2006, Charinpanitkul & Tanthapanichakoon 2011). If the aerosol distribution is such that most of the aerosol mass lies in the range of particles larger than 5µm, then removal of dust particles by water droplets will be rapid (Seinfeld &

Pandis 2006). Removal rates for large dust particles (>5µm) are roughly independent of the water droplet size distributions (Seinfeld & Pandis 2006). For very large particles (>20µm), the collection efficiency of a water droplet approaches unity, whereas removal of particles in the 0.1−1 µm size range, i.e. minimum collection efficiency regime, is expected to be relatively slow (Seinfeld & Pandis 2006). On the other hand, water spray droplets induced by spray nozzles have velocities much greater than their settling velocities. Thus, they can have a higher airborne dust capture efficiencies for smaller particles (Hinds 1999).

(35)

33

2.4 CONTROL TECHNIQUES

Figure 5 shows control measures in two working environments studied in the thesis. Airborne infection isolation rooms (AIIRs) and renovation sites have the same enclosure and negative pressure principles for containment, but the source control and air cleaning systems within an enclosure have their own special features in these environments (Fig. 5). Identification of hazards is the basis for selecting appropriate control techniques. In this section, control practices and their performance regarding environments studied are reviewed.

Fig. 5. Airborne particle control in AIIRs and reconstruction sites. Control methods studied in the thesis are marked with bold.

(36)

34

2.4.1 Enclosure and negative pressurization

An enclosure is separated from other spaces and has its own ventilation system.

An enclosure is constructed in an airtight manner by using impervious partitioning structures. The leakage of contaminated air to outside the enclosure is prevented by sealing off all lead-throughs, ducts, vents, and windows. This allows maintaining of pressure difference between the spaces. Assuming the adjacent area as a reference, the enclosure can be adjusted under negative pressure by a vacuuming fan. The negative pressure prevents spreading of contaminants to adjacent spaces (WHO 1999).

Enclosures in AIIRs are single-patient rooms. AIIRs have their own special HVAC system, by which air is exhausted directly outdoors or recirculated through HEPA filtration before return (CDC 2007). As for renovation sites, enclosures can be created by exploiting existing walls and room divisions or by assembling temporary wall structures, such as plastic films, plywood, or gypsum boards. Heating, ventilation and air conditioning (HVAC) systems operating at the renovation site must be shut down, if possible, to avoid the contamination of the systems (SLIC 2006, OSHA2007).

Installing one or more portable exhaust fan units into the renovation site achieves negative pressure. In addition, the exhaust air should be filtered with a fine filter or high-efficiency particulate air filter (HEPA) (SLIC 2006).

There are only a few studies of the performance of renovation site partitioning and negative pressure method in preventing the spread of contaminants into adjacent areas (Overberger et al. 1995, Rautiala et al. 1998). Overberger et al. (1995) studied a renovation site separated from the adjacent area by utilizing the existing interior walls and using slabs of plasterboard between the false and true ceiling. An airlock under negative pressure was located between the construction zone and adjacent area. However, pressure differences were not confirmed. During reconstruction activities, the total suspended particulate concentration in the adjacent area was close to the concentration before reconstruction work started. In the renovation site, the concentration was tenfold compared to prior work. In another study (Rautiala et al. 1998), a negatively pressurized enclosure (∆p not monitored) was constructed using a plastic barrier from the floor to the ceiling, for mouldy building remediation. Concentrations of microorganisms were 100 times higher in the renovation site during dismantling than before it, while concentrations in the adjacent area were at a similar level prior to and during the reconstruction.

Several studies have showed that many AIIRs cannot always maintain the negative pressure and 9−49% were found to be even positively pressurized relative to surrounding areas (Sutton et al. 1998, Pavelchak et al. 2000, Rice et al. 2001, NIST 2006, Saravia et al. 2007). Moreover, despite of negative pressure, openings of doors may result in spread of contaminants outside an AIIR (Hayden et al. 1998, Pavelchak et al. 2000, Rydock & Eian 2004, Johnson et al. 2009, Adams et al. 2011) since the negative pressure changes to a positive for a period during door openings (Hayden

(37)

35 et al. 1998, Tang et al. 2005). Turbulence and bidirectional airflow have been shown to occur during door openings (Li et al. 2007a) due to temperature differences between an enclosure and adjacent space, air density, time and angle of door opening (Tang et al. 2005, 2006) and door type (Tang et al. 2013, Kalliomäki et al. 2016).

Use of an anteroom or airlock can limit the transmission of contaminants into the surroundings by controlling the airflow patterns through doorways (CDC 2007, OSHA 2007). The purpose of an anteroom between an AIIR and an adjacent space (usually corridor) is to dilute and exhaust contaminants before they may spread outside an AIIR. In reconstruction sites, airlocks act as a barrier between the enclosure and adjacent spaces, since they do not have (usually) their own ventilation system. A three-partial airlock system is required for demolition of hazardous substances, such as asbestos, to supply the make-up air through the airlock doorways into the enclosure (Government Degree 798/2015/MSAH). Make-up air through an airlock, however, may not be enough to ensure sufficient make-up air for very large enclosures. Thus, Pocock et al. (2013) tested changes in pressure differences (∆p) by introducing additional make-up air into a ventilated enclosure with two three-partial airlocks (total volume of the enclosure 60 m³) constructed of wooden frames and plastic films. The airflow of an exhaust fan unit was 1800 m³/h achieving an air change rate of 38 1/h. The pressure difference was -9.6 Pa. From one to four additional inlet filters were installed in the wall of the enclosure, which resulted in an additional flow rate through inlet filters from 265 m³/h (one filter) to 691 m³/h (four filters). The increase in controlled make-up air reduced ∆p approximately 1 Pa, varying between -8.7 Pa to -5.6 Pa depending on the number of filters installed. However, researchers noted that adding inlet-filters (not HEPA) to enclosure walls might worsen the containment by allowing a potential pathway from the enclosure (Pocock et al. 2013).

Adams et al. (2011) assessed the containment efficiency of an AIIR with anteroom at different negative pressures with and without door openings and person movement. Target pressure differences were -2.5 Pa (mean -2.1 Pa), -11 Pa (mean - 10.7 Pa) and -20 Pa (mean -20.7 Pa). During ingress or egress of a person, the AIIR’s containment improved when the ∆p was ≤ -2.5 Pa. However, no statistically significant differences in particle escape from the AIIR to the corridor due to neither

∆p nor person movement-pressure interactions were observed. Instead, the traffic condition, ∆p, and person movement-pressure interactions were all significantly related to particle transmission from the AIIR to the anteroom. When the pressure difference was -2.5 Pa, the increase in particle dispersion from the patient room to the anteroom during person movement was 50-fold relative to absence of person movement. It led to a 2-fold increase in net particle escape to the surroundings from the AIIR. Anterooms were found to be effectively limiting the particle dispersion outside the AIIR (Adams et al. 2011). Conversely, Johnson et al. (2009) found that presence of the anteroom did not have a significant effect on the particle dispersion outside the temporary enclosure due to door traffic, which was most likely explained by the unrefined construction of the anteroom (Johnson et al. 2009). The anteroom without separate ventilation was constructed from plastic films with plastic curtains

(38)

36

at doorways. Temporary enclosures were constructed from PVC tubing and fittings and plastic films. They were exhausted by using a portable fan (Johnson et al. 2009).

A pressure differential ≤ -2.5 Pa is suggested to be more protective in limiting the dispersion of particles from the enclosure (Adams et al. 2011). However, a negative pressure as great as -15 Pa did not completely prevent the escape of particles from the enclosure (with anteroom) to the adjacent area during provider traffic (Rydock &

Eian 2004). Fluctuation in pressure differences due to door openings and other external factors, such as wind speed and elevator movements can be controlled with a novel pressure adjustment technique applied in reconstruction processes: the target

∆p is achieved and maintained with an active airflow adjustment between the exhaust air and circulated air (Arpomaa & Ahtola 2017).

2.4.2 General ventilation

General ventilation is effective in diluting work-produced contaminant concentrations in work environments: e.g., in renovation sites, general ventilation (62 1/h) reduced respirable quartz concentration by 66% and respirable dust by 70%

during concrete grinding (Akbar-Khanzadeh et al. 2010). This resulted in a significantly lower dust exposure compared to no general ventilation (Akbar- Khanzadeh et al. 2010). In another study (Akbar-Khanzadeh et al. 2007), general ventilation of 40 1/h resulted in respirable dust and respirable quartz reduction efficiencies by 73% and 70%, although there was no significant difference between general ventilation and no ventilation in respirable dust and quartz concentrations.

According to Jankowski (2011), general ventilation (air change rate not known) combined with on-tool local exhaust ventilation increased dust removal by 2.5−14%

compared to LEV only.

Many studies regarding the role of general ventilation in particle control have been conducted in isolation rooms of hospitals. Tung et al. (2009) studied the contaminant dispersion (tracer gas) within a mock-up AIIR with pressure differences of -2.5, -5, -8 and -15 Pa and air change rates of 12 and 24 1/h. The best ventilation efficiency was achieved with ∆p of -15 Pa and air change rate of 24 1/h. Higher air change rate showed less dispersal of contaminants within the enclosure, whereas the highest negative pressure related to the most effective contaminant removal.

Nevertheless, increase in air change rate was more beneficial to decrease the exposure of a health care worker (HCW) than increasing the ∆p. On the other hand, higher air exchanges alone have not shown to reduce the risk of airborne infection within an enclosure, yet higher air change rate results in a better dilution of contaminants (Melikov et al. 2010b, Memarzadeh & Xu 2011, Pantelic & Tham 2013).

Increasing air change rates from 3 1/h to 6 or 12 1/h (Melikov et al. 2010b), from 4 1/h to 12 1/h (Memarzadeh & Xu 2011) or from 2.5 1/h to 5.5 1/h (Mousavi & Grosskopf 2015) were not proportionally effective in reducing particle concentrations within the enclosure. In some cases, instead, a higher air change rate has shown to increase the

(39)

37 breathing zone exposure in turbulent airflow (Pantelic & Tham 2013, Mousavi &

Grosskopf 2015).

Within an enclosure, air flows ideally from a clean area towards the contaminated area. In AIIRs, this means clean supply air should flow first to the area where an HCW operates, and then across the patient into the exhaust (CDC 2005). However, Sutton et al. (1998) reported that only one of the tested 27 AIIRs had an airflow pattern such that the HCW was not positioned between the source and exhaust.

Another study’s results (NIST 2006) indicated that two of the ten tested AIIRs had airflow towards the anteroom door, which most likely increased the potential of contaminants entering the breathing zone of an HCW or transferring outside of the AIIR during door openings (NIST 2006).

Air distribution and local airflow patterns within enclosures are proposed to be the most important factors in contaminant transmission (Memarzadeh & Xu 2011, Pantelic & Tham 2013). Different airflow patterns impact differently on containment (Kao & Yang 2006). Positioning of the exhaust devices is diverse: some studies suggest placing exhaust devices near a patient’s head (Cheong & Phua 2006, Kao &

Yang 2006), while others suggest ceiling level (Beggs et al. 2008, Qian & Li 2010, Thatiparti et al. 2017). Nonetheless, contaminants are better controlled when the pathway between the source and exhaust is uninterrupted (Memarzadeh & Xu 2011).

Cheong & Phua (2006) found low-level extraction more effective in removing contaminants from the HCW’s breathing zone compared to ceiling level. The most effective contaminant removal efficiency was achieved when the supply air grilles induced a laminar flow to the HCW with minimal air mixing in the enclosure (Cheong & Phua 2006). Kao & Yang (2006) simulated that the parallel-directional airflow by positioning the supply air devices in the wall opposite the patient and exhaust devices behind the patient’s head, indicated a uniform flow path with a steady airflow direction. Floor level extraction showed a poor contaminant control efficiency with an up-draft effect when the supply air was entering from the ceiling level (Kao & Yang 2006). Conversely, ceiling level exhaust was found to be more effective in removing fine particles compared to floor level exhaust: e.g., particles of 1−10 µm were removed by 90 % with ceiling level extraction while only 19−29% were removed by floor level exhaust (Qian & Li 2010). Removal of large particles (50 µm) was less dependent on the location of exhaust since their major removal mechanism is by deposition on horizontal surfaces (Qian & Li 2010). Beggs et al. (2008) also showed that ceiling level air supply and extraction resulted in a four to five times lower bioaerosol concentration in an empty-room simulation compared to two other ventilation strategies which had high-level air supply with low-level extraction on the wall or vice versa. In the study by Thatiparti et al. (2017), the original ceiling- mounted ventilation arrangement used traditionally in AIIRs was altered by altering the exhaust near the patient’s head on the ceiling to assess the possible flow pathway of infectious particles within an AIIR (air change rate 12 1/h). This resulted in a better outcome compared to the original ceiling level extraction; however, not all the

(40)

38

airborne particles were removed. Remaining aerosols (38%) were able to flow towards the HCW’s breathing zone (Thatiparti et al. 2017).

2.4.3 Local exhaust ventilation

LEV on different reconstruction activities is usually on-tool extraction or ventilation shroud configuration, in which a tool is connected to a vacuum source to provide ventilation airflow for the tool. Vacuum sources used are e.g., portable construction site vacuums with either a HEPA filter or a disposable filter. Portable LEV systems recirculate filtered exhaust air back into the workspace.

Personal exposures to respirable dust and respirable quartz during reconstruction tasks are substantially lower when on-tool extraction based LEV systems are used (Table 1). Task-specific personal exposures may vary substantially due to factors such as the task, practices, material, and tooling used (Croteau et al. 2004, Tjoe Nij et al.

2004, Flanagan et al. 2006, Beaudry et al. 2013), amongst others. In spite of using highly effective LEVs, respirable dust and quartz have not always been reduced below the occupational exposure limit values (OELs) (Croteau et al. 2004, Akbar- Khanzadeh et al. 2007, Shepherd et al. 2009, Akbar-Khanzadeh et al. 2010).

Overexposures may be explained by the ventilation parameters, such as insufficient airflow, overloaded filters, hood entry or other losses in the LEV systems impairing particle removal efficiency (Flynn & Susi 2003, Shepherd et al. 2009). Hood design and fitting to different tools, and sufficient fan capacity to maintain adequate transport velocity of the dust particles, influence the dust capture efficiency (Flynn

& Susi 2003). Moreover, tasks such as surface grinding generate dust at much higher initial velocity and produce more extensive dust emission than, for example, drilling, thus, those activities require the higher airflow rates of the LEVs used to achieve effective dust capture (Shepherd et al. 2009).

Control of particle dispersion from enclosed spaces

Dissertations in Forestry and Natural Sciences

ANNA KOKKONEN

CONTROL OF PARTICLE DISPERSION FROM ENCLOSED SPACES

uef.fi

ANNA KOKKONEN

CONTROL OF PARTICLE DISPERSION FROM

ENCLOSED SPACES

Anna Kokkonen

CONTROL OF PARTICLE DISPERSION FROM ENCLOSED SPACES

ABSTRACT

ACKNOWLEDGEMENTS

LIST OF ABBREVIATIONS AND SYMBOLS

LIST OF ORIGINAL PUBLICATIONS

AUTHOR’S CONTRIBUTION

CONTENTS

1 INTRODUCTION

2 LITERATURE REVIEW

2.1 PARTICLES AND THEIR BEHAVIOUR IN THE AIR OF AN ENCLOSURE

2.2 NATURAL MECHANISMS OF PARTICLE REMOVAL

2.3 TECHNICAL APPLICATIONS FOR REMOVAL OF PARTICLES

2.4 CONTROL TECHNIQUES