xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EQE5YBPU9_TD55U3 INGEST_TIME 2014-05-13T17:59:13Z PACKAGE UFE0045820_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
1 DIFFUSIONAL RELEASE OF POLLUTANTS INTO AMBIENT AIR AND METHOD FOR ENHANCING DIFFUSIONAL COLLECTION IN AIR SAMPLING By LIN SHOU A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2013
2 2013 Lin Shou
3 To my beloved Mom
4 ACKNOWLEDGMENTS I am grateful for Dr. Chang Yu Wu (my supervisory committee chair) for his patient guidance and persistent encouragement in my study. I also sincerely thank Dr. Chang Yu Wu for giving me the opportunity to study in UF and teaching me to be more mature in my work and life. He is always a good advisor and a good f riend to me. I d like to thank the Florida Industrial and Phosphate Research Institute (FIPR) for funding the project of developing a personal sampler for sampling inorganic acids I would also lik e to thank Dr. Brian K. Birky, Director, for his valuable guidance, advice and comments. My deeply appreciation also goes to Alex Theodore for his invaluable support and inspiration in developing the personal sampler. Many thanks to Yu Mei Hsu and Danielle Hall for their kindness, continuous support and precious suggestions in my research Thanks are also due to Elizabeth Gomez and John Policandriotes, undergraduate students in the Department of Environmental Engineering Sciences, University of Florida, for their help in experiments. I m also grateful for the Florida Department of Transpo rtation (FDOT) Research Center for supporting my research to identify potential concerns of using ammoniated fly ash in the concretes My deep appreciation goes to Dr. Timothy Townsend for providing laboratory equipment and precious guidance in my researc h. I m also grateful for Tim Vinson and John Schert from t he Hinkley Center for Solid and Hazardous Waste Management for their kind support precious advice s and comments. Thanks also go to my colleagues, Joshua Hayes, Weizhi Cheng and Katheryn Campbell in the Department of Environmental Engineering Sciences, University of Florida, for their warmth, kindness and valuable help in the experiments.
5 Finally, I thank all my labmates, Danielle Hall, Nathan Topham, Myung Heui Woo, Jun Wang, Brian Damit Nima A Mo hajer and Matthew Tribby for their help in my PhD study. Many thanks also go to my good friends in both the United States and China, for their priceless love and friendship. Last but not least, I thank my parents and my little brother, for their invaluable love and spiritual support all these years.
6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 9 LIST OF FIGURE S ................................ ................................ ................................ ........ 1 0 LIST OF ABBREVIATIONS ................................ ................................ ........................... 12 ABSTRACT ................................ ................................ ................................ ................... 13 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 15 Assessment of Worker Exposure to Air Pollutants ................................ .................. 15 Personal Sam pling Technologies ................................ ................................ ........... 16 Personal Sampling Criteria ................................ ................................ ............... 16 Ae rosol Sampling Techniques ................................ ................................ .......... 18 Gas Sampling Techniques ................................ ................................ ............... 20 Co sampling of Gas and Aerosols ................................ ................................ .... 23 Exposure Models to Air Pollutants ................................ ................................ .......... 23 Occupational Exposure Problems Identified in the State of Florida ........................ 25 Sulfuric A cid M ist T hreat and I naccuracy of NIOSH S ampler 7903 .................. 25 Application of Ammoniated Fly Ash and Worker Exposure to Ammonia .......... 26 Research Objectives ................................ ................................ ............................... 31 2 DESIGN OF A PER S ON AL SAMPLER WITH COMPACT SIZE AND HIGH PERFORMANCE FOR OCCUPATIONAL SAMPLING ................................ ........... 33 Objective ................................ ................................ ................................ ................. 33 Design of the Sampler ................................ ................................ ............................ 33 Impact or Design ................................ ................................ ............................... 33 Denuder Design ................................ ................................ ............................... 35 Filter Pack Design ................................ ................................ ............................ 36 Prototypes Development ................................ ................................ ......................... 37 Summary ................................ ................................ ................................ ................ 38 3 THEORETICAL AND EXPERIMENTAL EVALUATION COLLECTION EFFICIENCY FOR SO 2 GAS ................................ .......................... 44 Objective ................................ ................................ ................................ ................. 44 Theory ................................ ................................ ................................ ..................... 44 Experimental ................................ ................................ ................................ ........... 45 Materials ................................ ................................ ................................ ........... 45
7 Coating ................................ ................................ ................................ ............. 45 Sampling and A nalysis ................................ ................................ ..................... 46 Denuder c ollection e fficiency ................................ ................................ ..... 46 Particle loss ................................ ................................ ................................ 48 Results and Discussion ................................ ................................ ........................... 49 Denuder Collection Efficiency ................................ ................................ ........... 49 Particle Loss ................................ ................................ ................................ ..... 51 Summary ................................ ................................ ................................ ................ 53 4 SAMPLING CO EXISTING GAS AND AEROSOLS ................................ ............... 62 Objective ................................ ................................ ................................ ................. 62 Experimental ................................ ................................ ................................ ........... 62 Aerosol Collecti on Efficiency of the Parallel Impactor ................................ ...... 62 Gas Collection Efficiency and Capacity of the Porous Membrane Denuder ..... 64 Co Sampling SO 2 Gas and H 2 SO 4 Mist ................................ ............................ 65 Results and Discussion ................................ ................................ ........................... 66 Aeroso l Collection Efficiency of the Impactor ................................ ................... 66 Gas collection Efficiency and Capacity of the Porous Membrane Denuder ...... 66 Co Sampling of SO 2 Gas and H 2 SO 4 Mist ................................ ........................ 67 Summary ................................ ................................ ................................ ................ 69 5 CHARACTERIZATION OF AMMONIA GAS RELEASE FROM CONCRETE WITH ADDED AMMONIATED FLY ASH ................................ ................................ 76 Objective ................................ ................................ ................................ ................. 76 Experimental ................................ ................................ ................................ ........... 76 Materials ................................ ................................ ................................ ........... 76 Experimental System & Conditions ................................ ................................ .. 77 Mathematical Model of NH 3 Diffusion in Concrete ................................ ............ 79 Results ................................ ................................ ................................ .................... 80 NH 3 Release from Concrete Mixing, Initial Settling and Curing Period ............. 80 Diffusivity of NH3 in Concrete ................................ ................................ ........... 83 Discussion ................................ ................................ ................................ .............. 85 Dissociation of NH3 in the Liquid Phase ................................ .......................... 85 Convective Mass Transfer on the Liquid Air Interphase ................................ ... 86 Summary ................................ ................................ ................................ ................ 88 6 CONCRETE MIXING SCENARIOS ................................ ................................ ........ 97 Objective ................................ ................................ ................................ ................. 97 Exposure Models ................................ ................................ ................................ .... 97 Scenario 1: Constructing an Outdoor Concrete Slab, Roadway or Bridge Deck ................................ ................................ ................................ .............. 97 Scenario 2 : Placement of Concrete in a Form with High Walls ...................... 100
8 Scenario 3: Ready Mix Concrete Truck ................................ .......................... 102 Scenario 4 : Placement of Concrete inside a Building ................................ ..... 103 Exposure Assessment Results ................................ ................................ ............. 104 Scenario 1 ................................ ................................ ................................ ...... 104 Scenario 2 ................................ ................................ ................................ ...... 106 Scenario 3 ................................ ................................ ................................ ...... 106 Scenario 4 ................................ ................................ ................................ ...... 107 Recommendations ................................ ................................ ................................ 108 7 CONCLUSIONS ................................ ................................ ................................ ... 123 LIST OF REFERENCES ................................ ................................ ............................. 130 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 139
9 LIST OF TABLES Table page 2 1 Specifications of impactor nozzle sizes ................................ .............................. 39 2 2 Propert ies of 4 different types of PMDs ................................ .............................. 39 3 1 Experimental results of Sets 1 4 of PMD I ................................ .......................... 55 3 2 Theoretical diffusional collection efficien cy of gases and particles by the PMDs ................................ ................................ ................................ .................. 55 5 1 Concrete composition and weight of concrete samples in the chamber experiment ................................ ................................ ................................ .......... 91 6 1 Assumed p arameters in Scenario 1 ................................ ................................ .. 110 6 2 Assumed parameters in Scenario 2 ................................ ................................ .. 111 6 3 Assumed parameters in Scenario 3 ................................ ................................ .. 111
10 LIST OF FIGURES Figure page 1 1 Sampling criteria for inhalable, thoracic, and respirable fractions ....................... 32 1 2 Schematic diagram of silica gel tube (NIOSH Method 7903) .............................. 32 2 1 Design of the personal sampler ................................ ................................ .......... 40 2 2 Streamlines and particle trajectories for a typical impactor ................................ 40 2 3 Schematics of the PMDs ................................ ................................ .................... 41 2 4 Photos of prototype PMDs ................................ ................................ .................. 42 2 5 The first prototype of the personal sampler ................................ ........................ 42 2 6 The second p ro to type of the personal sampler ................................ ................... 43 3 1 Experimental setup for testing aerosol collection efficiency and particle loss ..... 56 3 2 E xperimental setup for measuring particle c ollection efficiency of the impactor ................................ ................................ ................................ .............. 57 3 3 Collection efficiency of different types of denuders (predicted gas: SO 2 ) vs. denuder height ................................ ................................ ................................ .... 58 3 4 SO 2 removal efficiency of PMD I as a function of time ................................ ....... 59 3 5 SO 2 removal efficiency of GHD and PMDs as a function of time ........................ 60 3 6 Particle loss of PMDs in the size range of 1 to 10 m ................................ ........ 61 4 1 Experimental setup for measuring particle collection efficiency of the impactor ................................ ................................ ................................ .............. 71 4 2 Experimental setup for testing mixed gas and aerosol ................................ ....... 72 4 3 Particle penetration through the parallel impactor ................................ .............. 73 4 4 3 HCl and SO 2 gas (with a 10% coating) .................. 74 4 5 Experimental results of coexisting SO 2 gas and H 2 SO 4 aerosols ....................... 75 5 1 Experimental setup ................................ ................................ ............................. 92 5 2 Ammonia concentration at the surface of the concrete vs. time ......................... 93
11 5 3 RAP of the five sets of concrete in the first 8 hr mixing and curing perio ds ........ 94 5 4 Cumulative mass release of ammonia with time ................................ ................. 95 5 5 Diffusivity of NH 3 in concrete as a function of time ................................ ............. 96 6 1 Workers around a f reshly p laced c oncrete s lab in an o pen e nvironment ......... 113 6 2 Examples of Scenario 1 ................................ ................................ .................... 113 6 3 Dispersion model in Scenario 1 with wind ................................ ........................ 113 6 4 Placement of concrete in a form with high walls ................................ ............... 114 6 5 Example of Scenario 2 ................................ ................................ ..................... 114 6 6 Diffusion model for Scenario 2 ................................ ................................ .......... 115 6 7 A w orker n ear the d rum of the c oncrete m ixing t ruck ................................ ....... 115 6 8 Example of Scenario 3 ................................ ................................ ..................... 116 6 9 Workers in a r oom with an a ir f low r ate F ................................ ......................... 11 6 6 10 Example of scenario 4 ................................ ................................ ...................... 117 6 11 Initial ammonia concentration in fly ash vs. ammonia concentration in air ....... 118 6 12 Ammonia concentration downwind the concrete slab at various wind speeds 119 6 13 Ammonia concentration in the diffusion field at the end of 1 hr continuous placement with the maximum ammonia releasing rate ................................ ..... 120 6 14 Ammonia concentration as a function of time in the headspace of the ready mix truck ................................ ................................ ................................ ........... 120 6 15 Predicted ammonia concentration change with time in scenario 4 ................... 121 6 16 8 hr TWA a mmonia c oncentration as a f unction of i nitial a mmonia c oncentration in f ly a sh with v arious v entilation r ates ................................ ....... 122
12 LIST OF ABBREVIATIONS A CGIH American conference of governmental industrial hygienists A FA Ammoniated fly ash C AIR Clean a ir i nterstate r ule C EN Comit Europ en de normalisation F DOT Florida department of transportation F IPR Florida industrial and phosphate research institute G HD Glass honeycomb denuder I C Ion chromatography I OELV Indicative o ccupational exposure l imit v alue I SO International standards organization L NB Low NOx burners N IOSH National institute for occupational safety and health O SHA Occupational safety and health administration P EL Permissible exposure l evel P MD Porous membrane denuder R EL Recommended exposure limit S CR Selective catalytic reduction systems S NCR Selective non ca talytic reduction systems T WA Time weighted average T LV Threshold limit value
13 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy DIFFUSIONAL RELEASE OF POLLUTANTS INTO AMBIENT AIR AND METHOD FOR ENHANCING DIFFUSIONAL COLLECTION IN AIR SAMPLING By Lin Shou December 2013 Chair: Chang Yu Wu Major: Environmental Engineering Sciences A personal sampler relying on diffusional collection of acidic gases and aerosols was developed, and diffusional release of ammonia from concrete was characterized. The persona l sampler consist s of a parallel impactor a porous membrane denuder (PMD) and a filter pac k. T he parallel impactor was tested to have a collection efficiency curve that approximates the ACGIH respirable curve. The PMD s were proved to have collection efficiency for SO 2 HCl and HNO 3 h igher than 95% for four hours with a feed gas concentration twice their OSHA standards, respectively The sampler could collect more than 99% of aerosols within eight hours, and the interference between SO 2 gas and H 2 SO 4 aerosols was effectively minimized. The pe rsonal sampler could b e applied in occupational environment to a ss ess workers exposure t o hazardous gas es and particles accurately, which is critical in providing information for regulators and health and safety staff to make proper respiratory protection policies. Workers potential exposure to ammonia gas in using ammoniated fly ash is also important to current concrete construction industr ies and related regulators in setting a safety threshold Laboratory experiments were carried out with concrete added with fly ash cont aining ammonium loading ranging from 65 to 3200 ppm. The results showed
14 that at the end of the curing period, there was 20% to 70% of ammonium retained in the concrete. Of all the NH 3 gas released, over 80% was emitted in the first 8 hours after the concre te was mixed. According to the physical and chemical condition of the concrete, NH 3 release from concrete was divided into three periods: mixing, initial settling and curing. Mechanics of NH 3 release in the three periods were quantitatively analyzed for th e first time. Diffusivity of the mixing period and the initial settling period was 9.1 10 8 ( 1.25 10 7 ) m 2 /s and 3.93 10 9 ( 2.78 10 9 ) m 2 /s, respectively. In the curing period, NH 3 diffusivity in the concrete decreased from 5.36 10 9 m 2 /s to 0 in three weeks. To assess workers exposure to ammonia gas, the diffusion model was applied to four typical working scenarios of concrete construction and a safe ty threshold of 100 ppm of ammonia content in fly ash was recommended according to mode ling results
15 CHAPTER 1 INTRODUCTION Assessment of Worker Exposure to Air Pollutants In industrial hygiene, a primary goal is the assessment of worker exposure to hazardous materials to prevent occupational disease s and to improve occupational safety ( Carlton and Flynn 1997 ) In the early 1980s, Duan and Ott developed the basic concepts for exposure assessments ( Duan 1982 ; Ott 1982 ) According to their definition, human beings are the most important receptor of pollutants in th e environment. Ott ( 1982 ) defined the term exposure as an event that occurs when a person comes in contact with the p ollutant There are four routes to be considered in a comprehensive exposure assessment, including human contact with soil, water, food and air ( Lioy 1995 ) Air pollutants are dispersed ubiquitously and can enter human body through skin absorption or airway system. Personal exposure measurement is very important i n evaluating the health effects of an air pollutant According to the N ational R esearch C ouncil (NRC) a potential dose (the amount of material absorbed or deposited in the body for an interval of time) of a certain contaminant can be calculated by multiplying the integrated exposure with the volume of air exchanged in the lung in a specified time period ( Monn 2001 ) In addition, persona l exposure measurement also plays a significant role in industries where h ygienists typically choose appropriate control measures based on the results of breathing zone sampling ( Carlton and Flynn 1997 ) In the United States, the legal maximum allowable safe exposure of a worker to an air contaminant is dictated by federal standards which industries are required or recommended to maintain ( Tielemans et al. 1998 ) Commonly used standards in
16 industries include those from the American Conference of Governmental Industrial Hygienists (ACGIH), the Occupational Safety and Health Administration (OSHA), and the National Institute for Occupational Safety and Health (NIOSH). In order to monitor the working environment for purposes of detecting exposures exceeding the federal standards, industries periodically send industrial hygienists out to check worker exposures. It is typical for the hygienist to randomly select some workers during the course of a work day, obtaining some short grab sample measureme nts of the concentration of a specified contaminant at randomly selected times ( Tielemans et al. 1998 ) Samples of these air pollutants are often collected by personal sampling system and are sent back to laboratories for analysis. A less frequently used method by hygienists for assessing worker s exposure and evaluating control options is exposure modeling ( Carlton and Flynn 1997 ) Models are abstract constructions and depictions of complex systems that permit easier comprehension for study and application. A goo d understanding of the kinetics of gas and aerosol movement is very important in predicting their behavior, optimizing air pollution control or sampling devices, and minimizing interference between various gases and particles ( Gordin and Amirav 2000 ) Personal Sampling Technologies Personal Sampling Criteria The hazard caused by inhaled particles depends on their chemical composition and on the site at which they deposit within the respiratory system (Hinds, 1999) ACGIH In ternational Standards Organization (ISO) and the Comit Europ en de Normalisation (CEN) defined the same particle sampling criteria for inhalable, thoracic, and respirable fractions ( Grner et al. 2001 ; Bartley 1986 ) The criteria curve defines
17 the desired sampling performance of an inhalable particulate matter sampler in terms of the fractional collection for particles from 0 to 100 m. The equation s for inhalable fraction and inhalable fraction sampling criterion IF ( d a ) are ( Hinds 1999 ) : for m/s ( 1 1 ) for m/s ( 1 2 ) w here IF is defined as the fraction of particles originally in the volume of air inhaled that enters the nose or mouth d a is the aerodynamic diameter in m and U 0 is the ambient air velocity in m/s. The thoracic fraction ( TF ) is that part of the inhalable fracti on which penetrates into the respiratory tract below the larynx which is defined by the ACGIH criterion as: (1 3) where (1 4) T he respirable fraction, RF, refers to the particles which reach the gas exchange region of the lung. RF is defined by ACGIH particle size selective sampling criteria as ( 1 5 ) w here F(x) is the cumulative fraction for a standardized normal var iable x; ( 1 6 ) The relationship between F(x) and x is as follows: for ( 1 7 ) for ( 1 8 )
18 Figure 1 1 give values of inhalable, thoracic, and respirable fractions. In the past two to three decades, researchers develop ed samplers that are intended for compliance with the ACGIH/ISO/CEN sampling conventions ( Baron 1998 ; Kenny 1996 ) Aerosol Sampling Techniques A variety of personal inhalable aerosol samplers are currently used in the United States and Europe. In 1997, Kenny et al. e valuated the performance of several commonly used samplers, including the IOM sampler (SKC In c., Eighty Four, PA), the 37 mm cassette (SKC Inc., Eighty Four, PA), the CIP 10 I sampler (Arelco, Fontenay Sous Bois, France), the GSP sampler (StroKhlein GmbH, Kaarst, Germany ), the PAS 6 sampler (produced by the Department of Air Quality, Wageningen Ag ricultural University), and the seven hole sampler (SKC Inc., Eighty Four, PA, and Casella Ltd., London, UK) ( Aizenberg et al. 2000 ; Kenny et al. 1997 ) The measured sampling efficiencies of all the personal inhalable samplers show ed strong dependence on external wind speed. Among all these samplers, the IOM sampler in particular agreed very well with measurements of aerosol inhalable convention in low winds ( Kenny et al. 1999 ) Monitoring of thoracic aerosol concentrations at workplaces is not typically done and few personal sampling instruments were developed ( Kenny 1996 ) The m ore c ommon situation is that particulate matter PM 10 is sampled to replace the thoracic fraction ( Breuer et al. 2012 ) At the end of 2009, an Indicative Occupational Exposure Limit Value (IOELV) of 0.05 mg/m 3 for sulfuric acid aerosols was regulated for the first time on the thoracic particle fraction in Europe ( Breuer et al. 2012 ) In Canada, the thoracic limit value for sulfuric acid i s 0.2 mg/m 3 and there is a recommendation to conform to this limit by the ACGIH. T he Institute for Occ upational Safety and Health of
19 the German Social Accident Insurances used a cyclone to achieve suitable collection characteristics (PM 10 ) at a flowrate of 5.34 L/min to measure the thoracic fraction of sulfuric acid ( Breuer et al. 2012 ) The cyclone was made of sulfuric acid resistant stainless steel and has a length of 17.5 cm. However, it was mentioned by the authors that the fabrication of this cyclone was extremely elaborate which makes it an unlikely candidate for a commercially available personal sampler. Fabris et al. ( 1998 ) designed a sampler named CIP 10 T and tested its penetration characteristic follow ing the thoracic convention ( Fabris et a l. 1998 ) However, the flow rate of CIP 10 T of 7 L/min, is too high for personal sampling. To approximate the respirable penetration curve, there are three types of size classification devices including : 1) horizontal elutriator ( Wright 1954 ) ; 2) cyclone ( Lippmann 1 976 ) ; and 3) inertial impactor. Among them, elutriators are not as convenient as cyclones and impactors in field measurement due to the large size and the necessity for horizontal operation ( Marple 1978 ) Cyclones have been commonly used for separating thoracic or respirable fractions of inhalable particles ( Kenny et al. 1999 ) ire ACGIH/CEN/ISO defined respirable or thoracic curve accurately except for a virtual cyclone ( Chen and Huang 1999 ) Inertial impactors are the most studied inertial particle separators Trakumas and Salter ( 2009 ) designed a parallel impactor which arranged several impactors with different nozzle sizes in parallel to overcome the sharp penetration characteristics of the impactor. With multiple nozzles of various sizes selected and arranged on a single stage impactor, the sa mpling characteristics of the impactor can approximate ACGIH/CEN/ISO defined respirable or thoracic sampling
20 convention accurately. The parallel impactor has a flow rate as low as 2 L/min, and it was compact and lightweight, which is suitable for personal sampling Gas Sampling Techniques Passive samplers are the most widely used device in personal gas sampling ( Monn 2001 ) The principle of such a device is the passive diffusion of gases and concentration of a gas in air can then be calculated according to Fick s law of diffusion ( Palmes et al. 1976 ) Passive samplers have been used to measure NO 2 CO, SO 2 O 3 VOC, ammonia and formaldehyde in the atmosphere ( Palmes et al. 1976 ; Lee et al. 1992 ; McConnaughey et al. 1985 ) However, the sampling time is long, which usually takes several days to one week. In addition, its accuracy is relatively low ( Monn and Hangartner 1990 ; Hangartner et al. 1996 ) Real time gas monitors are also available in the market. These devices deliver a small volume of an air sample to a detection element based on electricity, fluorescence affinity, or cell function ( Weis et al. 2005 ) The gas molecules are drawn into a gas sensor using sampling techniques such as headspace sampling, diffusion me thods, bubblers or pre concentrators ( Pearce et al. 2006 ) The gas sample draw n across the sensor array induces a reversible physical and/or chemical change in the sampling mat erial, which causes an associated change in electrical properties, such as conductivity ( Arshak et al. 2004 ) There are d ifferent types of gas sensors including solid state gas sensors ( Dubbe 2003 ) conducting polymer gas sensors using e.g. polyaniline ( Nicolas Debarnot and Poncin Epaillard 2003 ) mixed oxide gas sensors ( Zakrzewska 2001 ) am perometric gas sensors ( Martell et al. 2004 ) catalytic field effect devices ( Lundstrm et al. 1993 ) or gas sensor arrays used in electronic noses ( Strike et al. 1999 ) The se sensors have wide application s in different industries, such as exhaust
21 gas sensors for emission control in automotive applications monitoring of daily products for the food industry, high volume control of combustibles in the chemical industry and so on Most of them are easy to use and offer continuous monitoring of the concentration of an individual gas or gas mixtures ( Timmer et al. 2005 ) Reported m ajor disadvantages of gas sensors include: 1) sensitive to temperature and humidity; 2) suffer from baseline drift; 3) complex interface circuitry and 4) poor selectivi ty towards the target gas compared to other gases ( Timme r et al. 2005 ; Arshak et al. 2004 ) Diffusion denuders are well established devices to remove gases from ambient air ( Pathak and Chan 2005 ; Hayami 2005 ; Acker et al. 2005 ; Huang et al. 2004 ; Tsa i et al. 2004 ; Dasch et al. 1989 ; Cheng 2001 ) The inner surface of the flow path is coated with a material that absorbs a particular gas or vapor. When air passes through the denuder, gas molecules diffus e to the surface of the flow path and get absorbed on the coated surface. Because of the gas removal capability, denuders have been used to measure the concentrations of gases in sampled air. They have also been used to remove gases or vapors from an aeros ol stream to prevent their interference in aerosol measurement that can cause positive or negative sampling artifacts ( Keck and Wittmaack 2006 ; Hinds 1999 ) The original design of the denuder system is a straight tube ( Ferm 1979 ; Durham et al. 1986 ) which is 1 m long to allow for an operating flow rate of 10 L/ min. Denuders with various shapes and dimensions have been developed thereafter, e.g. annular denuder ( Possanzini et al. 1983 ; Sekiguchi et al. 2009 ) coiled denuder ( Pui et al. 1990 ) honeycomb denuder ( Koutrakis et al. 1993 ; Sioutas et al. 1996 ) parallel plate denuder ( Eatough et al. 1993 ) and s o on. Glass and stainless steel are commonly used for the
22 fabrication of denuders The honeycomb denuder made of glass (GHD) has been commercialized and integrated in a denuder 3500, Thermo Electron Co., Inc.). However, this commercial denuder filter system is quite bulky, heavy and relatively expensive, rendering it undesirable for personal sampling. In recent years, new types of denuders have been proposed. Continuously wetted denuders ( Simon and Dasgupta 1993 ; Dasgupta et al. 1997 ; Sekiguchi et al. 2009 ) use a continuously renewed liquid as an absorber and the effluent can be used to determine gas concentratio ns online. However, these denuders are connected to an Ion Chromatography (IC) system or other sizeable infrastructure which makes them infeasible for personal sampling. Proposed by Fitz and Motallebi ( 2000 ) the fabric denuder is ano ther type of new denuder. F abric filter paper is coated and loaded in a filter stack, and collected pollutants are extracted in the same way as common filters. The fabric denuder has a relatively small loading capacity and suffers from significant particle loss Tsai et al ( 2001 ) developed a personal porous metal denuder specifically for adsorbing acidic gases and ammonia gas in an occupational environment. Field test results showed that the porous metal denuder has very high collection efficiency; however, particle loss in the por ous metal disc is also high because particles are forced to penetrate the curved channel of the porous metal discs. As discussed there are a variety of denuders available for removing gases from an aerosol stream. However, m ost of them were designed for sampling in the ambient or indoor environment, where the pollutant concentration is lower than an occupational environment. Due to the low concentration, the sampling flow rate is much higher than
23 that used for personal sampling (10 Lpm vs 0.2 2 Lpm). The other issue with these denuders is the material glass. As fluoride reacts with glass, a glass based sampler is not suitable for sampling in an environment where flu oride species is an important component, such as phosphate fertilizer pl ants ( Hsu et al. 2007 ) and semiconductor and opt oelectronic industries ( Tsai et al. 2003 ) Therefore, i t is necessary to develop a new type of denuder wi th the advantage of compactness and a proper flowrate for personal sampling. It should also have a high gas adsorbing capacity to be applied in occupational environment s where gas concentrations are usually higher than that in the ambient environment Co sampling of Gas and Aerosols While most personal sampling devices are available to collect particles or gases separately there are only few devices available to sample both simultaneously. For example, Koutrakis et al. ( 1989 ) designed a portable sampler which includes a glass impacto r and a filter pack to collect particles and an annular denuder in between to absorb gases Demokritou et al. ( 2001 ) developed a portable multi pollutant sampler, which combined an elutriator and a passive sampler to sample aerosol and gas at the same time. However, they were designed for use indoor s or in t he ambient environment, where target gas concentrations are relatively low. The size of these devices was also too bulky to be applied as a personal sampler. To develop a personal sampler, effort is needed to further decrease the size and increase the capa city of the sampler in collecting gas Exposure Models to Air Pollutants There are two types of exposure models: theoretical and empirical ( Sexton and Ryan 1988 ) Theoretical models are based on mathematical equations, describing
24 known physical or chemical mechanisms in the atmosphere Howe ver, due to the difficulty in building representative mathematical equations and obtaining analytical solutions, theoretical models are difficult to reflect the complexity of most industrial scenarios Empirical models address the complexity of pollutant s movement by relating the exposure to various process parameters. Kromhout et al. ( 1994 ) built an exposure model for the rubber manufacturing industry by relating exposures to ventilation characteristics and production variables In another study, Woskie e t a l. ( 1994 ) developed an exposure model to metalworking fluids in automotive component manufacturing. The model related several factors, such as the fluid type, the presence of local exh aust and the atmospheric conditions to worker exposure by multiple linear regressions Carlton et al. ( 1997 ) developed an empirical model of exposure to mists generated during a spray painting task In developing the model laboratory wind tunnel experiments with a mannequin, flat plat, and spray nozzle were carried out to provide data to determine the mathematical relat ionship between these parameters The model indicated that worker s breathing zone concentration of the m annequin is a strong function of worker orientation to the freestream. These empirical models can identify factors important in controlling exposure, but are difficult to generalize for various situations In summary, current exposure models, no matter th eoretical or empirical, all have their own limitations and disadvantages. However, exposure models are a very important tool in identifying the whole picture of pollutants generat ion and transport process es In addition, modeling is necessary in some situations where field
25 measurements are limited. E xposure models can also be used to identify factors important in controlling exposure, although direct application of most empirical models to other industrial situations still remains difficult. Occupatio nal Exposure Problems Identified in the State of Florida Sulfuric A cid M ist T hreat and I naccuracy of NIOSH S ampler 7903 Strong inorganic acid mists containing sulfuric acid have been identified as a ( NTP 2005 ) The current OSHA 8 hour Time Weighted Average (TWA) of Permissible Exposure Level (PEL) of sulfuric acid mist is set at 1 mg/m 3 The newly recommended Threshold Limit Value Time Weighted Average (TLV TWA) of the thoracic particulate fraction by ACGIH is 0.2 mg/m 3 As stated above, Europe also gives an IOELV value as the thoracic fraction limit of sulfuric acid, which is 0.05 mg/m 3 ( Breuer et al. 2012 ) Phosphate fertilizer manufacture was listed in the NTP report as one of many occupational exposures to strong inorganic acids based on the data collected decades ago. Thus, characterization of the exposure level at modern facilities is a necessar y step to the establishment of the best policy for worker protection Currently, NIOSH Method 7903 ( Eller and Cassinelli 1994 ) is the approved method set by OSHA for measurin g the total concentration of acidic aerosols and gases. The sampler of NIOSH Method 7903 consists of one section of glass fiber filter plug followed by two sections of silica gel as shown in Figure 1 2 The glass fiber filter plug is designed to filter ou t the majority of aerosols while the silica gel sections are used mainly to adsorb acidic gases. The collected samples are desorbed in eluent and the aliquots are analyzed by IC. It should be noted that IC only identifies soluble ions, e.g. SO 4 2 Therefor e, using this analytical method, the sulfate ion from sulfuric acid
26 cannot be distinguished from those from other sulfates (e.g. ammonium sulfate and calcium sulfate). Thus, all the sulfates are conservatively assumed to be sulfuric acid. A prior FIPR proj ect conducted by Hsu et al. ( 2008 ) s howed that the presence of SO 2 gas in phosphate facilities could be absorbed by glass fiber and silica gel in the NIOSH sampler, which causes the formation of artifact sulfite and subsequent conversion to sulfate during the extraction procedures of NIOSH Method 7903 As such, it lead s to a much higher measured concentration of sulfuric acid mist than the actual level. Therefore, it is necessary to develop a new personal sampling system to replace the NIOSH 7903 sampler to accurately measure gas and aerosol exposure lev el s in occupation al settings To collect particles according to the human respirable or thoracic fraction, the parallel impactor is a good candidate. A device such as a denuder can be used to remove the interfering gases. However, the denuder should have a high capacity in absorbing acidic gases and be light enough to be incorporated in the personal sampling system. Furthermore, the residual sulfate present in silica gel and the instability of glass fiber plug demand the development of new samplers without using these mater ials. Application of Ammoniated Fly Ash and Worker Exposure to Ammonia As a skin, eye and respiratory tract irritant, ammonia can cause health problems to people who may be near the fresh concrete ( Leduc et al. 1992 ) Exposures to levels exceeding 50 ppm in air can result in immediate irritation to the nose and throat. Exposure to a concentration of 250 ppm is bearable for most people only for 30 to 60 minutes. Exposure to levels greater than 300 ppm can cause chemical burns to the eyes, skin, and respiratory tract ( USHHS 2004 ) OSHA sets a TWA PEL for ammonia as 50 ppm (35 mg/m 3 ) while the ACGIH TWA TLV and the NIOSH TWA REL is set at
27 25 ppm. Recently, the Florida Department of Transportation (FDOT) identified a mmonia exposure to be a problem when using ammoniated fly ash (AFA) in their concrete construction projects. In recent years, coal fired power plants are more frequently utilizing air pollution control technologies such as l ow NO x burners (LNB), s elective c atalytic r eduction systems (SCR) and s elective n on c atalytic r eduction systems (SNCR) to meet the NO x reduction required in the Clean Air Interstate Rule ( CAIR ) The process change has create d problems in the utilization of the fly ash ( Van der Brugghen et al. 1995 ) In these processes, ammonia is injected into hot flue gas to react with NO x to form N 2 and H 2 O. Ammonia is also injected in order to reduce fly ash resistivity, thus improving operation of electrostatic precipitators (i.e. flue gas conditioning) ( Trivedi and Phadke 2009 ) However, ammonia (NH 3 ) and sulfur trioxide (SO 3 ) in flue gas combine with water in the system to form ammonium bisulfate (NH 4 HSO 4 ) and ammonium sulfate ((NH 4 ) 2 SO 4 ) as shown in the following equations ( Tyra and Robl 2001 ) : (1 9 ) ( 1 10 ) Depending on the level of ammonia present in the flue gas, the collected fly ash may be heavily loaded with NH 4 HSO 4 and (NH 4 ) 2 SO 4 Unreacted ammonia (i.e. ammonia slip) in the flue gas may also get adsorbed on fly ash as flue gas cools down ( Turner et al 1994 ) Fly ash is a pozzolan, a siliceous material which in the presence of water react s with calcium hydroxide at room temperature to produce a cementitious compound ( Kosmatka et al. 2002 ) Because of its pozzolanic properties, fly ash is useful in cement
28 and concrete applications. However, elevated concentrations of ammonia in fly ash have raised concerns about the implications of AFA in concrete construction projects including: (1) off gassing of ammonia from AFA during various phases of handling and use an d from the resulting AFA concrete ; (2) worker health and safety due to the exposure to released ammonia, acute and chronic; and (3) nuisance odor. Concrete mixing, s ettling and curing is a complex process which involves chemical reactions and physical str uctural changes to the concrete matrix. The major chemical reaction causing the concrete to form its hardened state is hydration, which occurs between the P ortland cement and water. Portland cement is a mixture of many compounds, among which tricalcium silicate and dicalcium silicate are the two most important components constituting about 75% of Portland cement by weight. They react with water to form calcium hydrox ide (Ca(OH) 2 ) and calcium silicate hydrate (3CaO 2SiO 2 8H 2 O) following the chemical reactions shown below ( Kosmatka et al. 2002 ) : ( 1 11 ) ( 1 1 2 ) Secondary reaction occurs between calcium hydroxide, pozzolanic material, and water to form calcium silicate hydrate, which is shown in the following equation : ( 1 1 3 ) Calcium silicate hydrate, the newly formed compound in the chemical reactions, is the heart of the concrete. Various properties of the concrete, such as hardening, strength and dimensional stability, depend mainly on this hydrate. The hydration process occurs both during the mixing and curing perio ds. Pore spaces within the
29 concrete mixture are filled by calcium silicate hydrate as concrete cures and ages ( Rathbone et al. 2001 ) When using fly ash as a pozzolan in concrete, highly alkaline free lime is generated from hydration of the cement. Under a high alkalinity environment, the ammonium salts contained in the fly ash react with water and liberate ammonia as a g as as shown in the following equation ( USHHS 2004 ) : ( 1 1 4 ) High alkalinity shifts the equilibrium to the right. Ammonia gas is then released from the concrete and workers who are working around the fresh concrete are at risk for exposure to high ammonia concentration s Previous research studies investigated the production, characterization and control of AFA ( Rathbone et al. 2001 ; Gao et al. 2002 ; Giampa and Plaza 1999 ; Giampa 2003 ; K laots et al. 2001 ; Larrimore 2002 ; Rathbone and Robl 2001 ; Subramaniam et al. 2009 ) However, only two of them directly relate to the reuse of AFA in concrete. Van der Brugghen et al. ( 1995 ) conducted laboratory experiments with fly ash containing from 100 to 300 mg/kg of ammonium added into concrete. Ammonia concentrat ions were continuously measured in the ambient air during the preparation of the concrete and the pouring of the concrete to make floors. An u nsafe level of ammonia concentration was identified in the vicinity of the wet concrete. It was also noted that am monia odor increased in severity with higher concentrations of ammonium in the ash and in enclosed areas. Rathbone et al. ( 2001 ) carried out studies to measure the release of ammonia from mortar and concrete during mixing, placement and curing period s Ammonia concentration in the air above the mortar was measured as a
30 function of ventilation rate across the surface of th e concrete. It was recommended that with negligible ventilation, ammonia concentration in the concrete mix water should be less than 110 mg/L if the NIOSH exposure limit of 25 ppm in the air is not to be exceeded. However, the concrete mortar was first mix ed outside and then placed in a plastic chamber for ammonia gas measurement. Ammonia loss in the mixing period was not monitored. Since ammonium salts (NH 4 HSO 4 and (NH 4 ) 2 SO 4 ) can easily dissolve in water, it is likely that ammonia gas start s to release at the moment that AFA contacts with water, which might not have been effectively monitored in Rathbone s study. In addition, mass transfer rate s of ammonia during the concrete mixing, placing and curing period might be different due to complex physical moti on and chemical reactions in the process. In monitoring worker exposure to ammonia gas, real time ammonia monitors can be adopted to measure ammonia concentration continuously Different kinds of sensors relying on the conductivity change for effectively detecting ammonia have been reported ( Krutovertsev et al. 2001 ; Ballun et al. 2003 ) The accuracy of most monitors is around 1 ppm, which is enough in most situations compared with OSHA TWA PEL. However, to control the pollution problem related with the use of AFA, it is important to identify a safe threshold of ammonia loading in fly ash In previous research studies t here has been no mathematical or empirical model or quantitatively determined mass transfer rate ( or other parameters) developed to predict the behavior of ammonia release from concrete containing AFA. Therefore, it is beneficial to build a n exposure model to estimate ammonia release from liquid/solid phase concrete and thereby de cide a threshold based on modeling and experimental results. To ensure the building of a reasonable
31 model, experiments are necessary to identify exposure routes and factors that affect ammonia discharge patterns and rates. Research Objectives To overcome the interference issues related to the current standard personal sampling method NIOSH 7903 addressed above, t he first objective wa s to develop a personal sampler that can collect inorganic acidic gases and mist on different parts of the sampler to minimize the interferenc e. Second, the personal sampler should have a high capacity for occupational environment s where gas and aerosol concentrations are usually high. The performance of the personal sampler such as gas collection efficiency, aerosol collection efficiency and mi nimization of interference between the measurement of gas and aerosols were evaluated in the laboratory. Its performance was also compared with the existing NIOSH 7903 sampler to verify that it provides more reliable results To assess worker s exposure t o ammonia in a concrete construction environment the second objective was to characterize concentration of ammonia released during different phases of handling and use of concrete containing AFA D iffusion coefficient s between the concrete and its exterior surface were obtained by using a proper diffusion model as evaluated in different concrete mixing and hardening period s A safe threshold of ammonia content in fly ash was recommended based on w l wor k conditions using the model estimat ion
32 Figure 1 1. Sampling criteria for inhalable, thoracic, and respirable fractions Figure 1 2. Schematic diagram of silica gel tube (NIOSH Method 7903) ( Source: http://www.sigmaaldrich.com Last accessed July, 2013 )
33 CHAPTER 2 DESIGN OF A PER S ON AL SAMPLER WITH COMPACT SIZE AND HIGH PERFORMANCE FOR OCCUPATION AL SAMPLING Objective The objective of this chapter was to design a personal sampler to sample gas and aerosols simultaneously and accurately, which could be adopted to sample inorganic acids in occupational environments such as phosphate facilities. Interference between gas an d aerosols should be minimized during sampling. The design should consider the following criteria: (1) size and weight, (2) capacity, (3) ease of handling, (4) chemical compatibility, and (5) cost. Design of the Sampler Figure 2 1 a) shows the conceptual design of the entire sampling system. It consists of 3 main components: a parallel impactor at the front, a porous membrane denuder ( PMD ) in the middle and a filter pack at the end. When air flow passes through this personal sampler larger aerosols are re moved from the gas stream by the impactor, then target gases are removed by the denuder and finally the remaining fine aerosols are collected on filters. Impactor Design The parallel impactor is used to separate different sizes of aerosols and remove larger aerosols to prevent their deposition in the following denuder as shown in Figure 2 1 b). It consists of 4 parallel and separated impactors. The different nozzle sizes of each impactor have different cut sizes, that when combined, create collection efficiency curve s that follows different pattern s e.g., human respiratory pattern (respirable fraction or thoracic fraction) to satisfy OSHA regulations or air quality standards (PM 2.5, PM 10) satisfying EPA (Environmental Protection Agency) regulations ( Marple 1978 ;
34 Trakumas and Sal ter 2009 ) A magnetic X shape d gasket between the top and bottom plate help to seal the four separate impactors from each other. Filter substrate (Cellulose filter, Whatm an Grade 40) could be placed on the bottom plate to sample aerosols and removed for analysis after each experiment, as shown in Figure 2 1 a). To determine hazards of inhaled particles, the amount of particles penetrating into the alveola r regions of the lungs is of primary significance, which is defined as the respirable fraction of particles. The quantity of respirable particles is dependent on the particle size distribution ( Hinds 1999 ; Marple 1978 ) In this design, the size of each impactor was chosen to follow the ACGIH respirable curve. The flow rate through the sampler was designed to be 2 L/min. The specified cut sizes of each impactor and the corresponding nozzles sizes ( D in ) were calculated us ing the following equation: ( 2 1) where d 50 is nozzle cut sizes, C c is C unningham correction factor for each particle size, p is particle density, which is assumed to be 1 g/cm 3 Q is flow rate through the sampler, N is number of nozzles specified, which is 4 in this design, is air vis cosity (1.81 x 10 4 poise), and Stk 50 is Stokes number corresponding to 50% collection, which is 0.47. The flow rate through each impactor wa s controlled using an exit orifice. Because t he pressure drop across the impactor, P, is equal to the dynamic pressure in the nozzles and the pressure drop across each individual impactor is equal, the exit orifices can be sized using the following equation ( Trakumas and Salter 2009 ) :
35 ( 2 2) where Q is the flow rate through the individual impactor and A N,in and A N,out are the areas of the inlet nozzles and outlet orifices, respectively. Specification of the nozzle sizes of the impactor is shown in Table 2 1. T he study of the effect of the S/W ratio (where S is the jet to plate distance and W is the jet width or diamete r as shown in Figure 2 2 on the efficiency curve ) ( Marple and Liu 1974 ) showed that the 50% cutoff size was strongly dependent upon S/W only for S/W < 1/2 for round impactors. For S/W ratios larger than 1/2, and the shape of the efficiency curves are rela tive ly constant. The design criteri on for round impactors is to make the value of S/W equal to or slightly larger than 1.0. Among the four inlet nozzle sizes of the impactor the largest of which is 2.36 mm, as shown in Table 6. Therefore, the value of S s hould be bigger than 2.36 mm In this design, 3.5 mm was used to allow some space for the filter substrate Denuder Design The PMD in the middle of the personal sampler is used to collect gaseous compounds The PMD s are multi channel denuder s composed of a soft and porous membrane. This is the first time a porous membrane is utilized as the material to construct a denuder. The p orous membrane consists of fine fibers and ha s high porosities and a large surface area that can provide the basis fo r increas ing the gas absorbing capacity of the denuder This compact design is intended to ensure high gas collection efficiency of the denuder when applied in occupational settings such as phosphate fertilizer manufacturing facilities where acidic gas con centration s may be
36 much higher than the ambient leve l. Figure 2 1 c) show s the structure of the denuder. Between the parallel impactor and the PMD there is a transition section that allows the sample air to flow smoothly through the PMD channels as shown in Figure 2 1 a) In constructing the denuder prototype, ma terial strength and low cost were two i mportant factors considered. Cellulose filter (Whatman Grade 40), which has a suitable rigidity and reasonable price, was selected to build the multi chann el grid shape denuder. Cellulose filter papers were cut by a laser plotter (Epilog Zing 35W) into a grid of interconnecting parallel planes. Each interconnecting panel has tabs and inserts that help ensure straightness and uniformity of the grid structure. Within each grid cell is an insert made from the same material, and various densities of fold forms to accommodate different channel numbers and openings. The height and outside dimension of the assembly of cut pieces and inserts can be tailored to a pred etermined size and shape. However, as the number of grid divisions or insert folds increases, the difficulty of assembling the grid cells and inserts also increases. Four different types of PMDs, named as PMD I, II, III and IV respectively, were built with the cellulose medium. The different types correspond ed to different numbers of channels and different dimensions of channel openings. The schematics of PMD I, II, and III are shown in Fig ure 2 3 PMD IV has the same structure as PMD III, only with more zi gzag inserts. The prototypes of these 4 types of PMDs are shown in Fig ure 2 4 and their properties are listed in Table 2 2 Filter Pack Design The filter pack at the bottom of the personal sampler is to collect the penetrating fine aerosols. Figure 2 1 d) shows the structure of the filter pack. There are two filters in the filter pack. The first f ilter ( Teflo TM 1.0 m Pall Life Sciences Inc. ) collec ts fine
37 particles that pass through the impactor and the denuder. The second filter ( Glass fiber, 1.0 m, S KC Inc.) coated with Na 2 CO 3 collects acidic gases that evolve from collected aerosols on the first filter. A filter holder with a diameter of 37 mm is used to support the filters Prototypes D evelopment Two prototypes were developed in the project. In the first prototype, t he 3 components, impa ctor, denuder, and filter pack, were packaged as modules using acrylic for the casing material The length of the prototype was 10.8 cm and its outside diameter was 4.2 cm. The weight of the prototype was 6 5 g. Th e photos of the prototype and each part are shown in Figure 2 5 However, it was difficult to operate and had air leakage problem s if any part was not assembled coaxially. Due to this problem, the second prototype of the personal sampler was developed with Delrin as shown in Figure 2 6 In this prototype, key parameters such as the structure of the impactor, the inside diameter and length of the main body, were kept the same as those of the first prototype to ensure the same performance However, t he impac tion plate, denuder casing, and filter pack casing of the second prototype were redesigned and connected using threaded connections. The thickness of the wall was increased from 1 inch to 3 inches to improve the strength of the housing. This prototype was 12.2 cm long and 5.5 cm in outside diameter. Its weight was 320 g. As shown in Figu r e 2 6 b) and c), there are two top caps for the personal sampler: one is close d faced with a fitting for 1/4" tub ing, and the other is open faced. The t wo caps can be switched for different testing situations (e.g., chamber test vs. shape d gasket between the top and bottom piece of the impactor help s seal 4 different groups of nozzles from each othe r. Two uneven projections on the peripher y of the
38 gasket and two corresponding grooves on the same position of the top impactor piece fit with each other to help each inlet nozzle match the corresponding outlet nozzle. In addition it is very convenient to assemble and disassemble each part of the second prototype ; impactor, PMD and filters can be removed from the personal sampler easily to fulfill different experimental purposes Summary A personal sampling device has been designed and fabricated to measure the true level of inorganic acidic gases and aerosols simultaneously in workplaces where high acid aerosol mist and gas concentrations may be present The sampler consists of a parallel impactor upst ream, a PMD in the middle and a filter pack do wnstream. The parallel impactor, which has 4 parallel and separated impactors with different nozzle sizes (6.6, 4.6, 3.5 and 2.2 ) was designed to have a collection efficiency curve that approximates the ACGIH re spirable curve. The PMD utilize d the poro sity of membrane material and a configuration of multiple parallel flow channels to compact the size, decrease the weight and increase gas collection efficiency. In the filter pack, t he first filter was designed to collect all aerosols penetrating the para llel impactor and the denuder, while the second filter was designed to collect acidic gases resulting from the decomposition of the aerosols collected on the first filter. Two prototypes made of different materials were developed. The first prototype made of acrylic was rejected due to its air leakage problem. The second prototype, made of Delrin and stainless steel, was applied in the following laboratory test.
39 Table 2 1 Specifications of impactor nozzle sizes D 50 C c D in (mm) Vo (cm/s) Re D out (mm) S (mm) 6.6 1.03 2.36 190.31 297.99 1.15 3.5 mm 4.6 1.04 1.86 305.80 377.74 1.25 3.5 1.05 1.56 437.05 451.59 1.39 2.2 1.08 1.15 797.50 610.02 2.36 Table 2 2 Propert ies of 4 different types of PMDs Denuder Diameter (mm) Height (mm) Number of Channels Smallest Channel Opening Area (mm 2 ) Largest Channel Opening Area (mm 2 ) Weight (g) PMD Ia 47 35 80 17.26 39.83 6~9 PMD Ib 50 80 17.26 39.83 PMD II 50 112 10.51 39.83 PMD III 50 192 8.58 PMD IV 50 280 5.90
40 Figure 2 1 Design of the personal sampler: a) cross sectional view of the personal sampler, b) impactor, c) denuder, and d) filter pack Figure 2 2. Streamlines and particle trajectories for a typical impacto r
41 Fig ure 2 3 Schematics of the PMDs
42 a) b) c) d) Fig ure 2 4. Photos of prototype PMDs : a ) PMD I b) PMD II c ) PMD III and d) PMD IV Figure 2 5. The first prototype of the personal sampler
43 Figure 2 6. The second p ro to type of the personal sampler
44 CHAPTER 3 THEOR E TICAL AND EXPERIMENTAL EVALUATION OF THE PMDS COLLECTION EFFICIENCY FOR SO 2 GAS Objective Since t h e new PMD is aimed to be integrated in to the personal sampler and applied in occupational environments where the acidic gas concentration may be much higher than the ambient level t he objective of this chapter was to evaluate the four different types of PMDs performance in absorbing acidic gas of twice of OSHA sta ndards both theoretically and experimentally. SO 2 gas, which has a n OSHA PEL of 5 ppm was selected as a representative acidic gas. In addition, to minimize the interference between aerosols and gases, the particle loss of the four types of PMDs was also m easured Theory The collection efficiency for a hollow tube, which acts as a perfect sink for a certain gas, was given by Gormley and Kennedy (1949) : ( 3 1) ( 3 2) where is collection efficiency of the denuder; is the dimensionless deposition parameter; D is diffusion coefficient of the penetrating gas or particles; Q is volume flow rate through the entire denuder; and L is the length of the denuder. For particles, a simplified Gormley Kennedy Equation was also given by Hinds (1999): for ( 3 3) (3 4)
45 Experimental Materials The four types of PMDs were evaluated for their performance. To compare the out with the glass honeycomb denuder (GHD). The GHD is 47 mm in diameter and 38 mm long. It has an internal surface area of 508 cm 2 which is made possible by 212 hexagonal flow channels that are 2 mm on each side. The GHD has a weight of 106 g. Coating Both the PMDs and GHD were coated with sodium carbonate (Na 2 CO 3 )/glycerin to absorb SO 2 gas. The use of this coating material is well documented in previous resear ch ( Lin et al. 2010 ; Keck and Wittmaack 2006 ; Perrino et al. 1990 ; Perrino and Gherardi 1999 ) However, the coating procedure was different in this study due to the special characteristic of the PMD The strength of the p orous membrane was much less than that of glass or metal, and the upper operating temperature limitation of cellulose filter is generally in the range of 75 C to 80 C, which means the drying temperature of the denuder should not be higher than 75 C. Sor bent of known mass was added into deionized (DI) water to make the coating solution. In coating denuders, the first step was to fully immerse each clean denuder into the solution. Then, the container was sealed with aluminum foil and placed into a sonicati on bath for 30 min utes Thereafter, the denuders were taken out of the container using clean forceps and placed on a clean glass tray for drying in a pre heated furnace with a temperature of 50~60 C for 3 to 5 hours. Two concentrations, 5% (m/v) Na 2 CO 3 /gl ycerin and 10% (m/v) Na 2 CO 3 /glycerin in DI water were used in the experiments. In Experiment Sets 1 and 2, two PMD Ia
46 filters were coated with 5% and 10% solution respectively, to compare the effect of coating concentration on the capacity of the denuder. In Experiment Sets 3 to 6, all types of PMDs and the GHD were coated with 10% solution. Duplicated experiments were carried out for Experiment Sets 1 to 4 and triplicated experiments were carried out for Experiment S ets 5 and 6. Sampling and A nalysis Denuder c ollection e fficiency The experimental system for capacity testing is shown in Fig ure 3 1 In Experiment Set 1, target gas, SO 2 supplied as 10 ppm from a cylinder, was mixed with compressed air in a ratio of 1:9 to obtain a concentration of 1 ppm t performance in occupational environment, the feed SO 2 concentration used in Experiment Sets 2 to 6 was 10 ppm, which is twice the Permissible Exposure Level (PEL) of SO 2 set by U.S. Occupational Safety and Health Administration (OSHA). The flowrate of the gas stream, 2 L/min was controlled by a mass flow controller (OMEGA, Model FMA 5520). An SO 2 monitor (International Sensor Technology, Inc., Model IQ 350) which can monitor the SO 2 concentration in real time was connected upstream and downstream of the sampler to measure SO 2 gas concentrations. Two impingers in series were connected downstream of the sampler using 9 mM Na 2 CO 3 solution to absorb gas that penetrated the denuder. The solution in the two impingers was changed every 30 minutes. Most of the exhaust gas was collected by the first impinger, while the second impinger was used to check whether the first one remained fully functional. After sampling, hydrogen per oxide (H 2 O 2 ) was added to the sample solution to oxidize sulfite
47 to form sulfate. The TWA concentration of sulfate ions was determined by an IC system (Model ICS 1500, DIONEX Inc.). The impinger method was used instead of the SO 2 monitor when the feed SO 2 concentration was low (e.g. 1 ppm), because the detection limit of the SO 2 monitor is 0.1 ppm, which does not have the resolution needed for low SO 2 concentration. For real time gas monitoring using the SO 2 monitor, the collection efficiency ( Eff ) of SO 2 can be obtained by measuring the feed concentration ( C u ) upstream and the exit concentration ( C d ) downstream of the holders. Eff at any given time can be calculated by the following equation: ( 3 5 ) If the impinger method was used for measuring downstream gas concentration, the exit gas concentration in Eq uation ( 3 5) was the summation of the concentrations in impinger 1 and impinger 2. The consumption of Na 2 CO 3 was calculated according to the amount of SO 2 adsorbed. First, the TWA collection efficiency ( Eff TWA ) of PMDs was determined using the real time monitor measurements. Then, the total amount of SO 2 collected by the denuder can be determined accordingly. Since the exhaustion of Na 2 CO 3 follows the stoichiometric reaction betwee n Na 2 CO 3 and SO 2 ( 3 6) the amount of consumed Na 2 CO 3 can then be estimated accordingly.
48 In this study, the performance of the denuder was only tested in the laboratory. Since inlet SO 2 concentration is unknown in field, two denuders in series can be used with the second denuder used to check the collection efficiency of the first denuder. Particle l oss Fig ure 3 2 shows the experimental setup for measuring particle loss of the PMDs. A vibrating orifice aerosol generator (VOAG, Model 3450, TSI Inc.) was used to generate monodisperse particles. The droplet size depends on orifice size, solution feed rate, operating f requency and solution concentration. The aerosol solution was composed of oleic acid as the non volatile solute and isopropyl alcohol as the solvent. The particles thus generated were dispersed with compressed air and went through a Kr 85 neutralizer colum n to be dried and neutralized. The airflow, 50 L/min, then went through the hose and entered one end of the chamber and exited the other end of the chamber. The aerosol chamber was made of stainless steel, and the cylinder was approximately 40.64 cm in dia meter and 71.12 cm in height. The corresponding air velocity through the chamber was about 6 mm/s, i.e. calm air conditions were applied in the test chamber (Trakumas and Salter, 2009). The denuder sampler was vertically installed in the chamber and connec ted with a mass flow controller and a pump outside. Sampling flowrate of the sampler was controlled to be 2 L/min by the mass flow controller. An ultraviolet aerodynamic particle sizer (UV APS, TSI Model 3312) was connected upstream and downstream of the s ampler to measure the particle size distribution. To examine particle loss in the tubing and joints of the system, an empty sampler was also tested as a control, the result of which was used as a baseline for the PMDs.
49 Results and Discussion Denuder Colle ction Efficiency The collection efficiency of a d enuder can be predicted by Equation ( 3 1 ) with a known diffusion coefficient of the target gas. Fig ure 3 3 displays predicted collection efficiency versus denuder length of different denuder types for SO 2 ga s. It can be seen that when the denuder length is greater than 1.5 cm, the collection efficiency of all types of denuder for SO 2 is close to 100%. However, because the reagent on the channel wall will be consumed during sampling, the capacity of the denude r will decrease as sampling time increases, and the assumption that the tube surface serves as a perfect sink for Equation ( 3 1 ) becomes invalid. Other factors, such as incoming gas concentration and the amount of reagent coated on the denuder are also imp ortant For PMD I, as shown in Fig ure 3 4 when a feed SO 2 concentration of 1 ppm was used, analysis of all the samples obtained from impingers 1 and 2 downstream shows that the 30 minute average exhaust gas conc entration was below 0.01 ppm during the entire 8 hours. The low exhaust concentration means that the denuder has collection efficiency higher than 99.9% for the entire eight hours. According to U.S. Environmental Protection Agency (USEPA), the average ambi ent concentration of SO 2 was lower than 0.01 ppm from 1990 to 2009 in the United States ( chen et al. 2011 ) Our results demonstrate that the PMD has a capacity that is high enough to be used for ambient SO 2 sampling. Under a feed concentration of 10 ppm, the collection efficiency of PMDs decreases quickly with time. The decreasing rate is opposite to the number of channels As shown in Fig ure 3 4 i ncreasing the length of the denuder (Set 3 vs. Set 2) increases
50 the efficiency, as it increases the amount of reagent and allows a longer residence time for gas molecules to diffuse to the wall and react with the reagent Increasing the coating concentration (Set 2 vs. Set 1) also helps increase the overall collection efficiency The reason is due to the finite capacity of the sorbent. In g eneral, the more sorbents coated on the surface of the denuder (e.g. thicker coating), the more gas molecules can be trapped by the denuder. Meanwhile, the higher the incoming gas concentration, the quicker the sorbents become exhausted. Sorbents at the to p of the coating react with target gas to form a different substance which covers the top layer of the coating. In this study, sulfur dioxide reacts with sodium carbonate to form sulfite which slowly oxidizes further to sulfate. This top layer becomes a ba rrier for the target gas to reach unreacted sorbents inside. Therefore, even if some reagents have not Consequently a proper coating concentration should be selected according to designated capacity of the denuder. Currently, there is no theoretical model that can be applied directly to the PMDs to predict the relationship between gas adsorption capacity and coating thickness. The s urface condition of the porous membrane plays an important role in the process, but such information is not readily available. Research to develop such a model that can accurately describe the diffusion process and sorbent consumption is wa rranted. The 3 hr TWA collection efficiency of Set 1 to Set 4 was about 68.2%, 73.4%, 81.3% and 94.6%, respectively, as shown in Table 3 1 The corresponding 3 hr consumption of Na 2 CO 3 according to Equation 3 6 was in the range of 10.4 to 14.4 mg for these 4 sets of experiments. The consumption was about 0.2% of the total coating,
51 indicating the capacity had not been effectively utilized. It indicates that if a longer residence time is allowed for SO 2 gas to travel through the denuder and react with Na 2 CO 3 a higher efficiency and longer breakthrough time can be expected. Either increasing denuder length or decreasing gas flowrate can both achieve this goal, e.g. t wo denuders arranged in series as in Set 4. The result in comparing the collection efficiencies of the four different types of PMDs and the GHD are shown in Fig ure 3 5 It can be observed that as the number of channels increases and channel cross sectional area decreases, the capacity of the mance. For PMD I and II, the grid density at the edge was lower than that at the center, causing overall lower performance. Therefore, PMD III and IV were designed in a more uniform grid for the entire cross section to increase collection efficiency. 5 hou r TWA collection efficiencies of PMD I, II, III, and IV were 73.1%, 82.8%, 90.9% and 97.2%, respectively, while that of the GHD was 96.6%. Compared with the GHD, PMD IV has slightly higher collection efficiency. However, the weight of the GHD is about 10 t imes greater than that of the PMD IV. In addition, since the PMDs are made of filter paper, they are relatively cheap and can be disposable. Therefore, the PMDs have the potential to replace traditional denuders made of glass or metals for applications whe re light weight and low cost are important features. Particle Loss The mass based particle loss percentage of each type of PMD is shown in Fig ure 3 6 The particle size range selected for testing was from 1 to 10 m Particles larger than 10 m usually should be removed by an upstream impactor or cyclone to avoid large impaction loss on the denuder. For particles smaller than 1 m although the
52 diffusion effect will be enhanced, the total mass usually accounts for a very small percent age of all t he particle loss. As shown in Fig ure 3 6 as particle size increased particle loss on all four types of PMDs increased. For larger particle size s generated by the VOAG, particle number concentration was lower than that for smaller particle size s Accordin gly, relative concentration fluctuation was greater compared with that of smaller particles. Due to impaction, large oleic acid particles were broken into smaller particles of various sizes. Some of them were collected on the denuder while the rest of them penetrated, leading to a varied downstream particle size distribution. These two factors resulted in an increased overall variance of particle loss as particle size increased For the same particle size, particle loss increases as channel opening area dec reases. Since PMD II and PMD III have the same channel opening in the center field, their particle loss tendencies are similar. The average particle losses (1 10 m) of PMD I to PMD IV are 2.9%, 5.2%, 5.7% and 7.3%, respectively. Particle loss in the denud er system usually occurs in four ways: diffusion across the denuder, impaction at the tube inlet, gravitational sedimentation and interception. Sedimentation can be easily avoided by setting the denuder in the vertical position while interception usually c auses small particle loss (< 1 m) ( Hinds 1999 ) The loss due to diffusion is usually quite small, since gases usually have diffusion coefficients much higher than those of particles. The loss due to diffusion can also be calculated using Equation s ( 3 1) to ( 3 3). With a gas flowrate of 2 L/min, diffusional collection efficiency/loss on the PMDs of SO 2 gas and different sized particles are shown in Table 3 2 As shown, the loss is smaller than 0.28% for any particle between 1 10 m Impaction is the most important mechanism for large particles. The increasing particle
53 loss tendency with increasin g particle size in the experiments also proves that particle loss on the denuder is mainly due to inlet impaction, which is consistent with published research (Possanzini et al., 1983; Ferm, 1986; Koutrakis et al., 1988 ) Particle loss due to impaction can be reduced by optimizing the inlet geometry and fluid dynamic conditions. Alternatively, an e lutriator impactor or cyclone inlet with a mass median aerodynamic diameter (MMAD) of 2 .5 mm cut size is commonly added in fron t of a denuder system to overcome problems of impaction loss of large particles on the denuder ( Ianniello et al., 2007; Koutrakis et al., 1988 ) Summary The collection efficiency of f our different types of PMDs for SO 2 gas was evaluated Tested with a flow rate of 2 L/min and a feed SO 2 co ncentration of 1 ppm, PMD Ia ha d a collection efficiency higher than 99.9% for 8 hours, which demonstrates that the PMD has high enough capacity to be used in ambient SO 2 sampling. Further experiments of PMD I showed that coated reage nt amount and denuder length were both very important factors. In addition, by increasing the number of channels and decreasing the opening area of ch d accordingly. With a feed concentration of 10 ppm, PMD IV, whi ch has the highest number of channels and smallest channel opening areas, achieve d a 5 hour TWA collection efficiency of 97.2%; slightly higher than 96.6% of the glass honeycomb denuder of similar matrix. However, the PMD weighs less than 9 g, which is onl y one tenth of the glass honeycomb denuder. Particle losses of four types of PMDs with the same length and diameter were 2.9%, 5.2%, 5.7% and 7.3%, respectively for 1 10 m particles. Most particle losses in this siz e range were caused by impaction at th e inlet of the denuder. An impactor
54 arranged before the denuder can remove the majority of large particles and reduce particle loss in the denuders.
55 Table 3 1 Experimental results of Sets 1 4 of PMD I N o. 3 hr TWA Collection Efficiency Amount of Absorbed SO 2 (mg) Amount of Consumed Na 2 CO 3 (mg) S et 1 68.2 6.3 10.4 Set 2 73.4 6.7 11.2 Set 3 81.3 7.5 12.4 Set 4 94.6 8.7 14.4 Table 3 2 Theoretical diffusional collection efficiency of gases and particles by the PMDs Gases/Particles Diffusion Coefficient (cm 2 /s, 20 C 1 atm) Collection Efficiency (%) PMD I PMD II PMD III PMD IV SO 2 1.370 10 1 100 100 100 100 0.001 m 5.228 10 2 99.94 100 100 100 0.01 m 5.329 10 4 17.68 20.91 28.03 34.66 0.1 m 6.865 10 6 1.01 1.26 1.79 2.29 1 m 2.817 10 7 0.12 0.15 0.22 0.28 10 m 2.419 10 8 0.02 0.03 0.04 0.06
56 Figure 3 1 Experimental setup for testing aerosol collection efficiency and particle loss
57 Figure 3 2 E xperimental setup for measuring particle collection efficiency of the impactor
58 Figure 3 3 Collection efficiency of different types of denuders (predicted gas: SO 2 ) vs. denuder height
59 Figure 3 4 SO 2 removal efficiency of PMD I as a function of time
60 Figure 3 5 SO 2 removal efficiency of GHD and PMDs as a function of time
61 Figure 3 6 Particle loss of PMDs in the size range of 1 to 10 m
62 CHAPTER 4 EXISTING GAS AND AEROSOLS Objective The objective of this chapter was to measure collection efficie ncy was assessed to ensure it follows the designed pattern in Chapter 2. While their excellent performance as a gas denuder has been demonstrated in Ch apter 3, PMD III and IV were integrated into the personal sampler for testing its ability in adsorbing acidic gases other than SO 2 gas. The sampler was then evaluated for its ability to mitigate sampling artifact. Experimental Aerosol Collection Efficiency of the Parallel Impactor Figure 4 1 a) shows the experimental setup for measuring aerosol collection efficiency of the parallel impactor alone as a function of particle size. A vibrating orifice aerosol generator (VOAG, Model 3450, TSI Inc.) was used to generate monodisperse particles rangin acid as the non volatile solute and isopropyl alcohol as the solvent. The particle size produced by the VOAG was determined by the diameter of the orifice used in the VOAG and the ratio between the nonvolatile solute to solvent. A compressed air cylinder (AI B300, Airgas) was connected with the VOAG to provide air for dispersing the generated droplets and diluting the aerosol. The particles thus generated were further dispersed with dilu tion air with a flow rate of 50 L/min, went through the Kr 85 neutralizer column of the VOAG for charge neutralization, and then entered the test chamber from its bottom. The cylindrical test chamber was made of stainless steel approximately 16
63 inches in d iameter and 28 inches in height. The corresponding air velocity through the chamber ranged from 6 to 10 mm/s to achieve calm air conditions ( Trakumas and Salter 2009 ) A honey comb air flow straightener was installed close to the bottom of the chamber. At the top of the chamber, there were several sampling ports that can be connected with different samplers. Monodispersity of the particles generated in the chamber was verified b y an ultraviolet aerodynamic particle sizer (UV APS) (TSI Model 3312). The personal sampler drew air from the chamber with a connected pump outside. Sampling flow rate of the personal sampler was controlled to be 2 L/min by a rotameter. In this set of expe riments, the denuder and the second coated filter in the filter pack were not installed in the personal sampler to allow the measurement of solely After a set sampling time, the impaction substrate a nd the after filter were rehydrated in 0.1 N NH 4 OH aqueous solution in known volumes. The solution was agitated with an Analog V ortex Mixer (Fisher Scientific TM ) to detach the fluorescein from the filters for complete dissolution into the NH 4 OH solution. T he mass concentration of the fluorescein in each sampler was then measured by a fluorometer (Sequoia Turner 112 Digital Filter Fluorometer, G.K. Turner Associates). Since the mass concentration of the dye was proportional to the mass concentration of the c ollected particles, the penetration ( P ) percentage can be calculated according to the equation below: ( 4 1 ) where M f is the mass collected on the after filter, and M pi is the mass collected on the parallel impactor.
64 Gas Collection Efficiency and Capacity of the Porous Membrane Denuder 2 HCl and HNO 3 The experimental system is shown in Figure 4 2 SO 2 g as was from an SO 2 cylinder of 1% concentration (Certified standard, Airgas). For the HNO 3 and HCl gases, clean air from an air cylinder (Breathing Air, Grade D, Airgas) passed through two parallel glass bubblers containing HNO 3 and HCl solution, respectiv ely ( Tsai and W ang 1999 ) to carry acidic gases through forced convection. In each experiment, only one gas was tested. There were two other streams of air from the air cylinder. One stream went through a deionized (DI) water bubbler and the other went to the mixing ch amber directly. The humidity of the test atmosphere in the mixing chamber was controlled by adjusting the ratio of wet to dry dilution air. Concentration of the gases was varied by changing either the concentration in solution or the dilution flow from the air cylinder. A glass fiber filter (Type AE, pore size = diameter = 37 mm, SKC Inc.) was used downstream to prevent liquid droplets from entering the mixing chamber. The feed gas concentration of the three gases was adjusted to be twice of OSHA st andard of Permissible Exposure Level (PEL)s, 10 ppm, 10 ppm and 4 ppm for SO 2 HCl and HNO 3 respectively. The error of the target gas concentration was within 10%. After the mixing chamber, the gas was split into two streams. One stream of gas went throu gh impingers 1 and 2 in series to determine the incoming gas concentration. The other stream of gas went through the personal sampler (without impactor and after filters ) where the target gas was adsorbed by the coated PMD inside. The exhaust gas from the personal sampler went through impingers 3 and 4, which were replaced by fresh 9 mM Na 2 CO 3 solution every 30 minutes to measure the concentration of gas
65 penetrating the d enuder. Two impingers were placed in series to examine its gas penetration percentage. The samples were then analyzed by an Ion Chromatography (IC) system (ICS 1500, Dionex) to determine the upstream concentration (from impinger 1 and 2) and downstream con centration (from impinger 3 and 4). The collection efficiency of the denuder, Eff D was determined by: (4 2 ) where M i1 to M i4 are the mass of the acid collected by impingers 1 to 4, respectively. Co Sampling SO 2 Gas and H 2 SO 4 Mist In this set of experiment s co sampling of SO 2 gas and sulfuric acid mist by the personal sampler was conducted. NIOSH method 7903 sampler (silica gel tube, SKC Inc.) was also tested and its performance in collecting sul fate was compared with that of the personal sampler. PMD IV was adopted in the sampler to remove SO 2 gas. As shown in Figure 4 1 b), a stream of air (12 L/min) was fed through a Collison nebulizer (CN25, BGI Inc.) with 0.5% (M/M) H 2 SO 4 solution to produce aerosols. Mass median diameter (MMD) of particles produced by the nebulizer was 2.5 m and the geometric standard deviation (GSD) was 1.8 ( BGI 2013 ) Gas stream from the nebulizer was mixed with a stream of 10 ppm SO 2 gas (5 L/min) and a stream of dilution air (5 L/min) before it reached the chamber. Two NIOSH 7903 samplers were connec ted in series to sample the gas/aerosol from the chamber. Flow rate of the NIOSH samplers was controlled to be 0.5 L/min. The personal sampler was connected to another sampling port on top of the chamber and its flow rate was kept at 2 L/min. Any exhaust g as or particles were collected by the two impingers in series downstream of the personal sampler. The experiment lasted for 2 hours. After sampling, the impactor substrate,
66 denuder, after filters and the silica gel in the NIOSH sampler were extracted with DI water. These samples, together with the sample from the impinger, were analyzed by the IC system for sulfate ion concentration. This group of tests is called test group I. The results were compared with test group II with H 2 SO 4 aerosol only (SO 2 gas str eam replaced by a stream of compressed air of the same flow rate) to examine if the unit can function as designed to minimize interference. Sulfate concentration in the impinger was efficiency for particles and gas. Results and Discussion Aerosol Collection Efficiency of the Impactor The monodispersity of the test aerosol was first verified. The GSD of each APS, was between 1.0 to 1 .1, proving the monodispersity of the feed particles. Figure 4 3 shows the penetration characteristics of the impactor. The dotted curve represents the respirable convention. The solid curve is the theoretical characteristic of the impactor, and the open circles represent the experimental performance of the impactor at each specific particle size. As shown, the experimental data agree well with the respirable convention as designed, demonstrating the Gas collection Effic iency and Capacity of the Porous Membrane Denuder SO 2 HNO 3 and HCl gas are shown in Figure 4 4 Relative humidity in the tests was 27 7%. As shown in Figure 4 4 a), after 4 hours, collection efficiency of PMD III (10% Na 2 CO 3 coated) for SO 2 and HCl decreased to 83 1 % and 84 2 %, respectively; for
67 HNO 3 it decreased to 87 3 that of SO 2 while its collect ion efficiency for HNO 3 was higher. The difference was due to the lower feed concentration of HNO 3 than that of HCl and SO 2 For PMD IV (10% Na 2 CO 3 coated), since there are more channels and its channel opening area is smaller compared with that of PMD III the collection efficiency for all 3 types of gases were higher. The mean collection efficiency was above 95% in 4 hours, as shown in Figure 4 4 b). Co Sampling of SO 2 Gas and H 2 SO 4 Mist Relative humidity in these experiments was measured to be 25 5%. Experimental result s of sulfate collected on each part of the personal sampler and on the NIOSH 7903 sampler (per one cubic meter of air) are shown in Figure 4 5 In test group I, the amount of sulfate collected on the personal sampler, which consisted of sulfate from SO 2 and from sulfuric acid, was 7 .1 2 0.25 m g /m 3 The amount collected by the downstream impinger was 0.043 0.004 m g /m 3 In test group II, the amount of sulfate collected on the personal sampler, which consisted of only sulfate from sulfuric acid, was 4.05 0.23 m g /m 3 ; while the amount in the after impinger was 0.034 0.007 m g /m 3 The collection efficiencies of the perso nal sampler for test group s I and II were 99.40 0.05% and 99.19 0.14%, respectively. For the NIOSH samplers, it was found from test group II that sulfuric acid penetrated the glass fiber and reaches the silica gel part which is designed to adsorb gases In addition, the glass fiber of the NIOSH sampler, which is supposed to collect particles only, was shown to adsorb SO 2 and other acidic gases due to its high alkalinity ( Hsu et al. 2008 ; Chow 1995 ) Therefore, it is impossible to tell whether the sulfate in the NIOSH samplers was from sulf uric acid or
68 SO 2 gas. The overall amount of sulfate collected by the two NIOSH samplers on test groups I and II was 6.15 0.11 m g /m 3 and 3.36 0.33 m g /m 3 respectively. Collection efficiency of the first NIOSH 7903 sampler was 92.57 0.85 % and 94.71 1.97% for test group I and II, respectively, which is lower than that of the personal sampler. efficiency of the sulfuric acid mists was 3.7 0.84 %, 89.0 3.6% and 4.05 0.84%, respectively. Since there was no SO 2 gas in test group II, any sulfate ion collected by the denuder could be taken as particle loss, which was 3.27 0.49%. As shown in Figure 4 5 in test group I, the amount of sulfate collected on the impac tor and the after filters were similar to that of the test group II. However, for the denuder the amount of sulfate ion is much higher when samples were collected with SO 2 gas passing through the system than without SO 2 gas. In other words, the majority of SO 2 gas was collected by the denuder, instead of other parts of the personal sampler. After droplets are produced by nebulization, the volatile solvent (water) can evaporate quickly. The corresponding solid particle size, d p is determined by the droplet diameter created by the nebulizer and the sulfuric acid fraction in the solution, F v,sol as ( 4 3 ) For the 0.5% (M/M) sulfuric acid solution used in this study the corresponding F v ,sol is 0.0 0278 and the dry particle size is However, as a hygroscopic material, the aerosol size of sulfuric acid mist in the chamber would be decided by the relative humidity of the air in the chamber. According to the E AIM inorganic model I pr oposed by Clegg and Brimblecombe ( Carslaw et al. 1995 ; Massucci et al. 1999 ; Clegg and Brimblecombe 2005 2013 ) at 1 atm, 298 K and 25% relative humidity, the mole
69 fraction of water in a sulfuric acid aerosol is 0.675 and the particle density is 1.44 g/cm 3 Accordingly, the volume fraction of sulfuric acid in the aerosol droplets in the chamber F v ,aer is 0.0339 The equilibrium aerosol size under this condition would be ( 4 4 ) Combining these two relations together the resultant sulfuric acid aerosol diameter in the chamber d c from the nebulized droplet, d n is then Since the MMD of the droplet diameter produced by the nebulizer is 2.5 m, the MMD Assuming lognormal distribution, more than 97.7% of particles in the chamber have a diameter smaller than 3.5 most particles will penetrate the three impactors of 6.6, 4.6, 3.5 m For the fo urth impactor with a cutsize of 2.2 it can be expected that more than 84% of aerosols can penetrate the impactor. Overall, around 96% of the sulfuric acid particles should be able to penetrate the impactor and get collected on the after filters. The theoretical estimate agrees very well with the experimental value of 96.3%. A ccording to our previous measurement in Chapter 3 for particles in the size range of 1.1 to 3. 4 m, particle loss on PMD IV is in the range of 0.6% to 3.4% which also agrees wit h the experiment al result of 3.27 0.49% Summary A novel personal sampler designed in Chapter 2 which integrates a parallel impactor, a porous membrane denuder and a filter pack, was built and tested. The paral collection efficiency of par ticles ranging from 1 to 10 m was measured and the results were in very good agreement with the human respirable convention. The PMDs have the characteristics of high capacity, light weight and
70 disposability. PMD IV was proven to maintain gas collection e fficiency higher than 95% for 4 hours with gas feed concentration twice of OSHA standards for SO 2 HNO 3 and HCl gas. Experiments of co sampling of SO 2 gas and H 2 SO 4 aerosol also demonstrated that the personal sampler can effectively minimize interference between sulfuric acid particles and sulfur dioxide gas. The overall particle collection efficiency of the personal sampler was higher than 99%. As shown, the persona l sampler has the potential to be applied in an occupational environment to sample gas and aerosols simultaneously and minimize the interference between them. In addition, the compact size, lightweight, disposability and low cost of the personal sampler ar e favorable features for wide adoption.
71 Figure 4 1. Experimental setup for measuring particle collection efficiency of the impactor
72 Figure 4 2. Experimental setup for testing mixed gas and aerosol
73 Figure 4 3. Particle penetration through the parallel impactor
74 Figure 4 3 HCl and SO 2 gas (with a 10% coating): (a) PMD III, (b) PMD IV
75 Figure 4 5. Experimental results of coexisting SO 2 gas and H 2 SO 4 aerosols (PI parallel impactor; PMD porous membrane denuder; AF 1 after filter 1; AF 2 after filter 2; and N NIOSH sampler)
76 CHAPTER 5 CHARACTERIZATIO N OF AMMONIA GAS RELEASE FROM CONCRET E WITH ADDED AMMONIATED FLY ASH Objective The objectives of this chapter were: 1) to measure ammonia gas concentration at the surface of the concrete during its mixing and curing periods when AFA of different ammonium concentrations is added to the concret e; 2) to identify the mechanism of ammonia gas release from different states of the concrete (liquid/solid mixture to solid status); and 3) to mathematically model the behavior of ammonia gas release and obtain the diffusivity of NH 3 in the concrete. Exper imental Materials AFA samples of Class F were collected from two coal combustion units at Crystal River Power Complex, Units 4 and 5 (Crystal River, FL). At the time samples were received (10/28/2011), fly ash produced in Unit 4 was expected to have signif icantly higher ammonium concentrations than Unit 5 (~3200 and ~50 ppm, respectively). The two fly ash samples were mixed together in varying amounts to produce five sets of fly ash blends with different ammonium (NH 4 + ) concentrations. The mixtures were sha ken periodically for approximately 24 hours to guarantee homogeneity. A zero headspace extractor (ZHE) was used to make sequential extractions from the five sets of AFA samples using deionized (DI) water at a liquid/solid (LS) ratio of 10:1. Each ZHE was
77 m ixed in an end over end fashion for one hour between extractions. The extraction fluid was analyzed using ion chromatography (IC) (ICS 1500, DIONEX). Then, the five sets of AFA of known NH 4 + concentration were used to replace part of the portland cement fo r mixing with water and aggregates to make concrete in the experiments. The composition of the concrete mix for the samples followed the Standards for Specifications for Road and Bridge Construction published by Florida Department of Transportation ( FDOT 2010 ) as shown in Table 5 1. The AFA constituted 50% of the cementitious material. The ratio of water to cementitious material was kept constant at 0.53 for all sets of experiments ( Rathbone et al. 2001 ) Experimental System & Conditions To assess ammonia release during the mixing and curing of concrete made with AFA, the concrete mixing process was conducted in a controlled chamb er. The chamber, a s seen in Fig ure 5 1a), was equipped with multiple ports for gas exchange and gloves to allow access. Inside the chamber, two fans (Honeywell Table Top Air Circulator Fan HT 904) were installed at both sides to create sufficient wind flow of 3 to 4 m/s to allow the released ammonia to be evenly distributed throughout the chamber. Two real time ammonia gas monitors (PHD6TM, Honeywell Analytics, Connecticut) measured ammonia concentration in the air contained in the chamber: one probe was positioned 2 to 3 cm above the surface of concrete to measure the concentration near the concrete interface, and the other was placed at a corner of the chamber to measure
78 ambient ammonia concentrations of the chamber. Compressed air from an air cylinder (AI B300, Airgas, Rad nor, PA) was delivered to the chamber at a flow rate of 2.8 0.1 L/min. To control ambient pressure in the chamber, gas was pumped out by a vacuum pump from the opposite end of the chamber with the same flow rate. Ammonia in the exhaust gas was trapped by a 0.1 N H 2 SO 4 solution (150 mL), which was thereafter analyzed by IC. The experimental protocol had the following steps: 1) mixing of batch wet concrete 30 times with a rod to allow air voids to be released and smoothed the surface with a rubber spatula; and 4) leaving concrete in the mold for initial settling. The chamber experiment lasted 8 hours: the mixing and pouring process lasted approx curing period. Afte r the concrete specimen had hardened 8 hours later and ammonia gas concentration in the chamber had decreased to below 50 ppm, the concrete specimen was removed from the chamber and placed in a sealed container to allow monitoring of ammonia gas release fo r an e xtended period, as shown in Fig ure 5 1b). In the following 0.2
79 L/min) of ammonia free lab air was passed through the container to carry released ammonia gas. The container was connected to a trap with a 0.1 N H 2 SO 4 solution (150 mL) to collect ammonia exiting with the carrier gas. The solution in the beaker was analyzed by IC and replaced daily. Mathematical Model of NH 3 Diffusion in Concrete When NH 3 is released from the surface of the concrete, a concentration gradient is produced between the surface and inner layers of the concrete, which is the driving force of NH 3 diffusion inside the concrete. To calculate the diffusivity of NH 3 in the concrete, an analytical ( Cussler 1997 ) : (5 1) where, (5 2) Assuming initial concentration of NH3 in the concrete is uniform and C(0,t) and C(s,t) are constant during a finite time period ( t i to t i ), the flux at the interface can be calculated as: (5 3) Equation 5 3 can be re written as: (5 4)
80 where D ti is the diffusion coefficient of NH 3 in the concrete, which is assumed to be constant in a time period ( t i to t i ). Equation 5 4 reveals the following relationship: (5 5) where, (5 6) (5 7) The H H can be obtained with the following equation ( Ni 1998 ; Molen et al. 1990 ; Sommer et al. 1991 ) : (5 8) Results NH 3 Release from Concrete Mixing, Initial Settling and Curing Period Ammonia exists in the liquid in the form of ammonium ions (NH 4 + ) and free NH 3 4 + and NH 3 in the concrete. TAN concentrations in fly ash specimens Set 1 to Set 5 were 65, 500, 862, 1378 and 3211 mg/kg, respectively. Var iation of the readings of the replicates was within 5%. Mean NH3 gas concentrations measured by the real time ammonia monitors throughout the course of the experiment for all concrete specimens
81 are summarized in Fig ure 5 2 a) and b). Results were obtained from duplicated experiments and the variation was within 15%. As shown in Fig ure 5 2a), in all five sets of concrete specimens, the peak concentration of ammonia gas was observed approximately 20 minutes after the experiment started, 2 to 3 minutes afte r the concrete was poured into the mold. Ammonia concentrations at the surface of the concrete were always higher than that of the ambient air in the chamber, indicating that the concrete was continuously releasing ammonia during the 8 hour chamber experim ent. It should be noted that this is the first time ammonia concentration during the mixing period has been reported. In the beginning of the initial settling period, due to density difference, solid concrete settled rmed on top of the concrete specimen. As time goes on, bleed water was evaporated and the concrete got hardened. Due to fresh air continuously delivered to the chamber, NH 3 concentration in the initial settling period continuously decreased. In the curing period, NH 3 release also decreased consistently during da ily monitoring, as shown in Fig ure 5 2b). For concrete Set 1, ammonia gas concentration decreased to lower than 1 ppm after 5 days. For concrete Set 5, it took 18 days for ammonia concentration to dr op below 1 ppm. The majority of ammonia release occurred in the first 24 hours when the moisture content of the concrete specimen remained high. As shown, in all three periods, the higher the TAN concentration, the higher the gaseous NH 3 concentration is.
82 The released ammonia percentage (RAP) at different mixing/curing periods was calculated by normalizing the released NH 3 amount by the original NH 4 + content in the fly ash. As shown in Fig ure 5 3, in the mix ing period, about 8% to 15% of NH 3 was released from the concrete mixture. S ince the concrete was continuously stirred in this period, convectional mass transfer of NH 3 between liquid and air was enhanced. Later, in the initial settling period in the chamber, about 6 to 26% of NH 3 was rele ased from concrete. NH 3 period, about 50% to 12% of NH 3 was released. The overall RAP of the three periods varied from 83% to 31%. In other words, at the end of the curing period, there was still about 20% to 70% of ammonia retained in the concrete as ammonium salts. It should be noted that from Set 1 to Set 5, the tendency of RAP in the mixing pe riod was opposite from that in the later two periods. The reason was probably due to the increasing process resistance in the concrete hardening period. Portland cement is a mixture of many compounds, among which tricalcium silicate and dicalcium silicate are the two most important components constituting of about 75% of portland cement by weight. They react with water to form calcium hydroxide (Ca(OH) 2 ) and calcium silicate hydrate (3CaO.2SiO2.8H2O) ( Kosmatka et al. 2002 ) Secondary reaction also occurs between calci um hydroxide, pozzolanic material, and water to form calcium silicate hydrate. Calcium silicate hydrate fills the capillaries in
83 the concrete to increase the strength of the concrete. The hydration process occurs in the entire concrete curing periods until moisture in the concrete was dried out ( Rathbone et al. 2001 ) Therefore, as concrete cures, the resistance for NH 3 di ffusion from the concrete to the atmosphere got larger, which limited gas release to a large extent. In contrast, in the mixing period, due to enhanced convective mass transfer, the higher interphase concentration gradient, the more NH 3 was released. Diff usivity of NH3 in Concrete The cumulative mass release of NH 3 m ti is presented in Fig ure 5 4. As s hown, over 80% of the released NH 3 gas was emitted from the concrete in the first 8 hours. Diffusivity of NH 3 calculated according to the change of cumulative m ass release is presented in Fig ure 5 5. It clearly shows that for the five sets of experiments, the diffusion coefficient decreases with time until close to 0. However, the decreasing rate for each period is different. In the mixing period, diffusivity was 9.110 8 ( 1.2510 7 ) m 2 /s, which is the largest among the three periods. This is due to continuous stirring of the concrete. From the point of diffusion theory, continuous stirring represents a shorter di ffusional path for the gas molecules, which makes the diffusivity larger. At the same time, the manual stirring also caused large disparity of diffusivity. In the initial settling period, there was no more stirring to the concrete, and NH 3 was released fr om the liquid air interface. Diffusivity of this period was relatively
84 constant, which was around 3.9310 9 ( 2.7810 9 ) m 2 /s. This value is higher but close to other reported diffusivities of NH 3 in water at temperatures between 11.8 C and 25 C, which ra nged from 2.88 10 9 m 2 /s to 1.310 8 m 2 /s ( Ni 1998 ; Zhang et al. 1994 ; Muck and Steenhuis 1982 ; Welty et al. 2009 ; Molen et al. 1990 ; Olesen and Sommer 1993 ; Bouwmeester and Vlek 1981 ) In the curing period, the bleed water on the surface of the concrete has almost evaporated. At the same time, the concrete is f urther hardened compared with the initial settling period. Therefore, the diffusion rate is controlled by the solid phase mass transfer i n the concrete. As shown in Fig ure 5 5, as time goes on, diffusivity of NH 3 decreased very quickly from 5.36 10 9 m 2 /s to 0. In addition, since NH 3 is a reaction product of NH 4 + and OH the decreasing diffusivity is also related with the decreasing amount of OH ions as water evaporates from the concrete. The drying rate of concrete depends on environmental conditions, c oncrete dimensions and concrete properties. According to Kosmatka et al. ( 2002 ) the moisture content decreases 100% to 80% typically in several weeks to 2 months. Rathbone et al. ( 2001 ) reported that the loss rate of NH3 was slowed dramatically after three weeks, which agrees with the results of this study and both results echo such a phenomenon reported by Kosmatka.
85 Discussion NH 3 r elease from concrete includes three processes: the chemical generation of NH 3 the diffusional mass transfer of NH 3 in the concrete, and the convective mass transfer of NH 3 gas from the concrete surface into free air stream. Dissociation of NH3 in the Liq uid Phase When water was mixed with cement and fly ash, ammoniated salts on the surface of fly ash was dissolved in water. The reaction between NH 4 + and NH 3 is effectively instantaneous so that they are at equilibrium at all points in the liquid ( Bouwmeester and Vlek 1981 ) The dissociatio n constant between NH 3 and NH 4 + could be written as follows: (5 9) The dissociation constant is a function of temperature. The fraction P of NH 3 released as gas is not only related to the dissociation constant, but also to the ion product for water and pH value of the solution ( Patoczka and Wilson 1984 ) : (5 10) Inclusion of the temperature dependence of the constants Kd and Kw yields: (5 11)
86 In the experiment, due to the existence of calcium hydroxide (Ca(OH) 2 ) in the concrete, pH value of the liquid remains above 12. The fraction of NH 3 released as gas at 298 K is 0.998, which means most TAN in water was tend to be released as gas. Convective Mass Transfer on the Liquid Air Interphase surface where NH 3 release occurs between the liquid air interphase. The mass transfer of NH3 gas from the concrete surface to free air stream is mainly dependent on air velocity and temperature ( Ni 1999 ) and it is largely by convection except for cases where there is almost no air movement on the interface. At 293 K and 1 atm, NH 3 gas diffusivity in air is around 0.253 cm 2 /s ( Poling et al. 2001 ) As a result of thermal motion, it would have moved a distance of (2 Dt ) 1/2 which is 0.71 cm in one second from its starting point ( Cussler 1997 ) Therefore, the effect of diffusion in air can be reasonably neglected compared with the wind velocity in this study or most open atmosphere cases ( Mohan and Siddiqui 1998 ) where wind speed is commonly in the range of 1 to 6 m/s. been proposed and widely applied in the previous studies of NH3 release from liquid to gas phase ( Haslam et al. 1924 ; Zhang et al. 1994 ; Arogo et al. 1999 ) In the two film mod el, the flux of NH3 can be written as ( Cussler 1997 ) : ( 5 12 )
87 where J is the flux of NH 3 release, g/s m 2 ; C g,i and C g,0 are the interfacial and bulk concentration of gaseous NH 3 respectively, g/m 3 ; C l,i and C i,0 are the interfacial and bulk concentration of NH 3 in the liquid phase, respectively, g/m 3 ; while k g and k l are mass transfer coefficient in the gas and liquid pha se, m/s. Haslam et al. ( 1924 ) carried out laboratory experiments of absorption of NH 3 in water based on the two film model. They found that k g was a function of air velocity and temperature, and k l was a function of temperature but was not appreciably affected by the air velocity. k l and k g can be calculated in the following equations: ( 5 13 ) ( 5 14 ) where v is air velocity, m/s; and T is temperature, K. With these two equations, k l for the initial settling phase can be estimated as 4131 cm/h; while kg is in the range of 11.26 to Zhang et al. ( 1994 ) and Arogo et al. ( 1999 ) also obtained k l and k g of similar equation by studying NH 3 release from liquid manure. They both accepted the assumption that NH 3 gas diffusion through the film is controlled by the gas phase tr ansfer. However, this assumption becomes invalid in this study as concrete cures. At the end of the curing period, there was still about 20% to 70% of TAN retained in the concrete as ammonium salts. Air movement over the surface of the concrete became irre levant. At the end of the initial settling period when most water on the concrete
88 surface has evaporated, free NH 3 needs to diffuse from inside of the concrete to the surface. As concrete cures, calcium silicate hydrate fills the capillaries in the concret e and the resistance for NH 3 diffusion in the concrete increases. Hence, NH 3 diffusion in the concrete became the rate limiting factor in the curing period. In the diffusion process, the rate of NH 3 depends on the combination of two kinetic factors: net d riving force and process resistance ( Glicksman 2000 ) In this process, the driving force is the concentration gradient of NH 3 while the process resistance is offered by the complicated physical structure of the concrete (a mixture of solid, liquid and capil laries with air). As discussed above the diffusivity of NH 3 in the concrete is different in different concrete mixing/curing period, which reflected different physical conditions of the concrete. Summary Ammonia gas diffusion from the concrete added with AFA can last several days to several wee ks. After the ammonia gas release rate gradually decreased to a non detectable level, there was still 20% to 70% of ammonium retained in the concrete. Of all the ammonia gas released during the entire concrete mixing and hardening period, over 80% was emit ted in the first 8 hour of concrete mixing and initial settling periods. According to different physical and chemical conditions of the concrete, the concrete mixing and curing process were separated into three periods: mixing, initial settling and curing. In the mixing period, ammonia gas release was enhanced by stirring in addition
89 to convective mass transfer. In the initial curing period, NH 3 release occurred on the transfer plays a dominant role in this process. Then, as water content further evaporated from the concrete surface and capillaries in the concrete were filled by calcium silicate hydrate during the curing period, resistance for NH 3 to diffuse from inside the conc rete to the surface got larger, which makes the diffusion inside the concrete the rate limiting factor of NH 3 release. The one dimensional diffusion mass transfer of NH 3 inside concrete can be ity was with the concrete mixing period 9.110 8 ( 1.2510 7 ) m 2 /s. In the initial settling period, the diffusivity was 3.9310 9 ( 2.7810 9 ) m 2 /s, which was higher but close to reported NH 3 diffusivity in water of similar temperature. In the curing period, the diffusivity quickly decreased to below detection limit due to decreased moisture content and filled pores of the concrete by calcium silicate hydrate. In this study, the physical and chemical processes of NH 3 release from concrete were quantitatively analyzed for the first time. In addition, a diffusion model derived from 3 in the concrete. With the diffusion model, NH 3 gas flux at the interface between the concrete and the atmo sphere at any time point from concrete mixing can be estimated. Such information is critical in assessing ammonia gas exposure of workers around the
90 concrete with given environmental parameter values, such as wind speed or air ventilation in a building.
91 Table 5 1 Concrete composition and weight of concrete samples in the chamber experiment Concrete Composition Weight Coarse Aggregate (Quikrete 550 g Fine Aggregate (Quikrete ASTM C33) 375 g Portland Cement (Quikrete ASTM C150 Type I) 188 g Water 199 g AFA 188 g Total 1.5 kg
92 Figure 5 1. Experimental setup : a) mixing and initial settling period, and b) curing period
93 Figure 5 2. Ammonia concentration at the surface of the concrete vs. time : a) mixing and initial settling period, and b) curing period
9 4 Fig ure 5 3 RAP of the five sets of concrete in the first 8 hr mixing and curing perio ds
95 Figure 5 4 Cumulative mass release of ammonia with time
96 Fig ure 5 5. Diffusivity of NH 3 in concrete as a function of time
97 CHAPTER 6 ASSESSMENT OF WORKERS EXPOSURE TO AMMONIA IN TYPICAL CONCRETE MIXING SCENARIOS Objective The objective of this chapter was to use the semi infinite diffusion model and the effective diffusion coefficients developed in Chapter 5 to estimate workers ammonia exposure in typical concrete mixing scenarios. In each scenario, working conditions such as the amount of concrete used, mixing time and wind velocity that would greatly affect were reasonably assumed. A safe thresh old of ammonia loading in fly ash based on the results of exposure assessment modeling for these scenarios was established Exposure Models I n practice, working conditions vary during concrete handling, mixing and finishing at job sites. T ypical cases in which worker exposure can be potentially high were described, including : 1) outdoor concrete construction, 2) placing concrete into a form with high walls, 3) concrete ready mix truck and 4) in a room with or without ventilation. For each scenario, in ord er to facilitate calculation, values were assumed for general parameters such as concrete density and mass of concrete to be used. Key environmental and operating parameters that can affect the exposure levels, such as wind speed and concrete mixing time, were then varied to obtain the relationship between ammonia concentration in air and ammonium concentration in concrete/fly ash. Scenario 1: Constructing an Outdoor Concrete Slab, Roadway or Bridge Deck In this scenario, workers place concrete into a slab form and finish the concrete slab in an open environment as shown in conceptually in Figure 6 1 Assumptions used in th is scenario are listed in Table 6 1
98 Ammonia quickly disperse s into the atmosphere during the construction of an outdoor slab, roadway or bridge deck. Even wind of a low velocity will enhance dispersion. Examples of this scenario are construction of a concrete slab, a roadway section and a bridge deck. Photos of the se working conditions are shown in Figures 6 2 a) and b), respectively. Zero wind speed is the worst case scenario for exposure of the workers to ammonia. The diffusion mechanics in the bridge deck placement were the same as that in the roadway pavement c ase, except that wind velocity is usually higher on a bridge and more air tends to move around the workers so that the ammonia exposure level should be relatively lower. Concrete is usually delivered in a ready mix truck that is driven on the road for up to an hour. During this period, workers are not exposed to the ammonia that is released c hr TWA exposure value. The transportation/mixing time was ammonia had been lost in transit, as shown in Table 6 1 In this scenario, a Gaussian dispersi on model can be applied to estimate ammonia concentration downwind of the concrete slab ( Turner et al. 1994 ) : ( 6 1 ) A Gaussian dispersion model was used to describe diffusion at any point downwind of the emission, as shown in Figure 6 3 Assumptions of this model are as follows: 1) m ass transfer due to bulk motion in the x direction far outweighs the
99 contribution due to diffusion; 2) c ontinuous emissions; 3) s teady state conditions; 4) c onstant wind velocity; and 5) c onservation of mass. According to the Gaussian dispersion model, the ammonia concentration immediately above the concrete slab (at the origin whe re x=0) approaches infinity, which is unrealistic. To estimate the ammonia concentration above the concrete sl ab, mass balance in a control volume ( V ) can be written as: ( 6 2 ) w here m in is the amount of ammonia generated by the concrete slab, and m out is the amount of the ammonia that goes out of the control volume. Since mass transfer due to bulk motion in the x direction far outweighs the contribution due to diffusion, the amount of ammonia diffused out of the control volume can be neglected. Assumin g th e control volume V = L (length) W (width) H (height), Equation 6 2 can be written as: ( 6 3 ) or with further simplification as: ( 6 4 ) Since J (flux of ammonia) changes over time, Equation 6 4 should be solved numerically with mid point method following Equation 6 5 : ( 6 5 ) Rearranging Equation 6 5 yields Equation 6 6:
100 ( 6 6 ) Initial ammonia concentration in the control volume C t=0 equals to 0. In order to compare modeling results with related regulations such as NIOSH R EL and OSHA PEL which are 8 hr TWA values, the time weighted average concentration C TWA in the control volume can also be calculated using the following equation: ( 6 7 ) The typical concrete slab thickness is 6 8 inches. Therefore, the total volume and surface area of the concrete slab can be calculate d. Based on this information and an assumed wind velocity, the mathematical relationship in Equation s 6 1 and 6 7 can predict the individual worker exposure. Scenario 2 : Placement of Concrete in a Form with High Walls In this scenari o, concrete is continuously pumped into a form with high walls as shown in Figure 6 4 Final placement of concrete in a footer is shown in Figure 6 5. Concrete consolidators working in a form with high walls often work right at the surface of the concrete as it is being placed in the form. Their job is to direct the placement of the concrete and vibrate the concrete to help the concrete flow around the rebar and get large air bubbles released from the concrete. If there is little or no wind blowing across the top of the form, ammonia can only exit from the top of the form to the ambient environment by diffusion. In this case, it is possible that relatively high concentrations of ammonia can develop inside the formed area.
101 Workers may be exposed to relativel y high ammonia concentrations during this period. In our model, this placement/consolidating period is assumed to be 1 hour. The concrete pumping rate and the open area at the top of the form will affect the ammonia concentration in the air inside the form to a large extent. Typical pumping rates of 20 to 75 cubic yards per hour were considered. W ind velocity was assumed to be zero in this case. After all of the concrete is placed in the footer, the ammonia release rate decreases The concrete finishers typ ically do not begin their work until 30 to 45 minutes after the concrete is placed. During the finishing period, it was assumed in the model that workers would be exposed to a lower ammonia release rate. As sumptions used in th is scenario are listed in Tabl e 6 2 The depth of a typical footer is usually 5 to 7 feet. Workers stand on the top mat of the reinforcing bar or on a construction bridge to finish most of the work. This case is much better than working at the bottom of the footer and is closer to the situation in Scenario 1. In this scenario, since the concrete is pumped into a form with high walls, it can be simplified as a line source at the bottom of the form that continuously releases ammonia in the upward direction (y), as shown in Figure 6 6. Ti me and d istance d ependent a mmonia c oncentration c an be o btained with the f ollowing Equation ( Glicksman 2000 ) : ( 6 8 ) where is mass release rate; y is the vertical distance from the emission source.
102 Scenario 3: Ready Mix Concrete Truck As shown in Figures 6 7 and 6 8 for Scenario 3, workers unload a ready mix concrete truck via the chute attached to the truck. In a worst case scenario the ready gas. After the truck arrives at the site, the dri ver or other workers around the truck could be exposed to a high concentration of ammonia if he/she begins to unload the drum. In other cases when concrete is delivered using a dump truck the ammonia concentration in the concrete is lower than when it is d elivered using a ready mix truck. In th is scenario a worst case situation was considered, i.e. there was no loss of ammonia from the truck. Ammonia diffused from the concrete was trapped in the headspace of the drum and the concentration in the headspace was calculated. The ratio of the total concrete volume to the volume of the drum is typically 1/2, which means that a typical ready mix truc k with a total volume of approximately 20 yd 3 is only filled to a maximum of 10 yd 3 of concrete in normal service. In practice, most ready mix trucks in Florida only deliver 9 yd 3 of concrete as the total weight of the vehicle and concrete would be close t o the maximum allowed weight (80,000 lbs) regulated by FDOT. T he total mixing time for the concrete in the truck should be no longer than 90 minutes. Assumptions used in the model calculation are listed in Table 6 3. Ammonia concentration in the headspace of the concrete drum can be calculated as: ( 6 9 )
103 w here is the cumulative release of ammonia mass; is the surface area of the interface between concrete and air in the drum; and is the volume of the headspace of the drum. Scenario 4 : Placement of Concrete inside a Building Th is scenario considers an enclosed building where wor kers would pump concrete into the building with little or no ventilation occurring as shown in Figures 6 9 and 6 10 Workers stay in the room for the placement and finishing of the concrete floor. In practice, concrete is mixed in a ready mix truck outsid e the room on the way from the ready mix plant to the site. W orkers are not exposed to ammonia in this period and it After the concrete truck arrive s at the site, the concrete is pumped into the ro om. A typical pumping period of 30 minute s for 9 cubic yards of concrete was assumed After the pumping period and until the end of the 8 th hour, workers would be exposed to ammonia at the lower emission rate. A constant fresh air flow rate, F was assumed in this case. Constant and varied working conditions of Scenario 4 are listed in Table 6 4 Ammonia concentration at time t in the room can be calculated with the following mass balance equation: ( 6 1 0 ) where M t A is the mass of ammonia released from the concrete inside the room, M in is the mass of ammonia entering the room from non concrete sources, which is assumed to be zero unless there is a specified source; M out is the mass of ammonia leaving the roo m ; and V R is volume of the room Assuming ammonia is instantaneously distributed throughout the room, the ammonia concentration of the air leaving the room is
104 equivalent to that of air inside the room. Then M out in a short time period can be estimated by the following relationship: ( 6 1 1 ) where F is wind flowrate through the room, L/min. Substituting Equation 6 12 into Equation 6 11 the ammonia concentration in the room at time t can be estimat ed by the following model which can be solved numerically : ( 6 1 2 ) Exposure Assessment Result s Scenario 1 In an open environment, even under stable weather, generally there is still wind going through the surface. Mass transfer due to wind in the downwind direction usually far overshadows the contribution by diffusion. The mean gas displacement along any axis at time t by Brownian motion is ( Hinds 1999 ) : ( 6 1 3 ) A s shown in the equation, the diffusing velocity caused by the concentration gradient decreases with time. In the first second, the mean displacement of ammonia caused by diffusion is 0.675 cm. According to the record from National Oceanic and Atmospheric A dministration (NOAA), monthly wind speed in Florida usually varies from 6 mph (2.68 m/s) to 14 mph (6.25m/s). Therefore, in an open field, gas displacement caused by the concentration gradient can be considered to be negligible in comparison with wind spee d. Most ammonia released from the concrete slab is diluted by wind immediately and does not accumulate above the concrete slab. To estimate ammonia
105 concentration just above the concrete slab, ammonia was assumed to be evenly distributed in a control volume of air just above the concrete slab. Depending on the assumed control volume, the airborne ammonia co ncentration would be different. As discussed above, even when the wind velocity is as low as 1 m/s, diffusion speed is still much smaller compared with wind speed. Unless there is turbulent flow in the y and z directions, which is too complicated to be inc luded in the simulation, it is unlikely that ammonia will diffuse a large distance from the surface of the concrete slab in these directions. Therefore, the control volume in this model was assumed to be the same length and width as the concrete slab (6.72 m 5 m) and a variable effective height of 0.1, 0.3, and 0.5 m was used to estimate the ammonia concentration above the concrete slab. The relationship between the initial ammonia concentration in fly ash and ammonia concentration at the surface of the c oncrete slab is shown in Figure 6 1 1 As shown in Figur e 6 11 a) (concrete mixed on site), since most ammonia is released during concrete mixing period, a threshold of ammonium in fly ash between 50 ppm and 80 ppm is recommended to meet NIOSH REL standard if concrete is mixed on site and workers are around the concrete. However, in practice, concrete is usually mixed in a ready mix truck running on the road, and workers are not exposed to ammonia release. In these cases, modeling result in Figure 6 11 b) an d c ) should be closer to the actual situations. Ammonium concentration in fly ash lower than 100 ppm should be adequate for meeting OSHA/ACGIH/NIOSH TWA standards. The estimated results of downwind ammonia concentration are shown in Figure 6 1 2 A s shown the higher above the concrete surface, the lower ammonia concentration was. In both y direction and z direction, the tendency was the same. The
106 highest ground level concentration always occurred along the centerline of the concrete slab. Even with a wind speed of 1 m/s, ammonia was quickly diluted by wind (even at the surface of the concrete slab). Ammonia concentration was lower than 1 ppm if the location was 200 m away from the emission source. Scenario 2 The modeling results are shown Figure 6 1 3 In this scenario, the assumption is ammonia can only diffuse in one direction and there is no wind in the form. Ammonia released from the concrete would be accumulated at the surface of the concrete slab and the concentration at the bottom of the form would b e higher and higher as time goes on. Therefore, if no dilution air is provided from the bottom of the form, concrete consolidators can be exposed to a considerably high ammonia concentration if they need to stay at the bottom of the form. In practice, usua lly there are pieces of equipment such as fans to create wind to dilute the air in the form, in which, if the created wind speed is high enough, the situation would be similar to Scenario 1 with wind. In addition, the depth of a typical footer is only 5 to 7 feet (1.524 m to 2.13 m), ammonia can be quickly diluted by wind at ground level instead of accumulating in the form, which will make the ammonia concentration closer to simulation results of scenario 1. For this complicated scenario, some field measure ment is strongly suggested. Scenario 3 In Scenario 3, the results of the worst case that all ammonia is trapped in the headspace of the ready mix truck are shown in Figure 6 1 4 Since concrete was constantly mixed in the truck, if the lid of the truck was closed, ammonia continuously diffused to the headspace of the truck would be trapped and the concentration would be increasing with time. In general, concrete is mixed in a tru ck for 30 min to 1 hour. In
107 addition, workers or drivers usually do not stay around the truck for a very long time. Hence, instant ammonia concentration instead of 8 hr TWA value was used to compare entration (IDLH). As shown, if the initial ammonium concentration in fly ash was higher than 50 ppm, instant high ammonia concentration in the headspace of the truck could cause a danger to human health. In practice, as long as the lid is left open and amm onia can diffuse to the ambient air or be quickly diluted by wind through the opening, ammonia concentrations can be much lower than the simulated results. Scenario 4 For Scenario 4, peak ammonia concentrations are closely related to wind velocity. Figu re 6 1 5 shows the predictions for an initial ammonia concentration in fly ash of 50 ppm with varying fresh air flow rates. As shown, peak concentration varied significantly under different fresh air flow rates. Under a higher ventilation rate, peak ammonia concentrations were much smaller. Different ventilation rates from 0 to 50 L/s per person (6 persons are involved) with three transportation/mixing times were then investigated. Figure 6 1 6 shows such predictions for Scenario 4 with comparison to 8 hr TW A of OSHA and NIOSH standards. If concrete is mixed inside the room, in order to follow both NIOSH and OSHA standards, ammonia concentration in fly ash should be less than 12.5 ppm if minimum ventilation rate of 7.5 L/s person can be guaranteed In a more common case, concrete usually is mixed in a mixing truck outside the room for at least 30 minutes. If so, ammonia concentration in fly ash of less than 55 ppm could satisfy the NIOSH standard.
108 Recommendations According to the simulation results, for Scena rio 1, ammonia usually can quickly diffuse to the ambient air and the concentration around the concrete slab is not expected to exceed OSHA, NIOSH or ACGIH exposure limits if ammonium concentration in fly ash can be kept lower than 100 ppm. In Scenario 2, if there is no dilution air and walls on both sides are very high, ammonia can be accumulated in the range of 3 m above the fresh concrete, which can be dangerous for workers who stay at the bottom of the form. However, in most common cases where the form is 5 to 7 feet deep, ammonia can be quickly diluted by wind at ground level and the situation will be similar to that in S cenario 1. In Scenario 3, to avoid immediate high exposure to ammonia that can cause some health problems to workers, it is recommende d that ammonium concentration in fly ash not exceed 50 ppm (5 ppm in concrete). If the lid is left open during driving, the ammonium concentration would be much lower. Otherwise, the driver/worker should absolutely avoid sticking his/her head into the head space of the truck. In Scenario 4, if the minimum ventilation rate specified by ASHRAE of 7.5 L/s per person can be guaranteed and concrete is mixed outside the room for at least 30 minutes (the most common case), a threshold of ammonium concentration in fly ash of 55 ppm (5.5 ppm in concrete) is recommended in order to satisfy ACGIH TLV and NIOSH REL standards (25 ppm). All above recommendations are based on laboratory observed ammonia release rates and application of mass balance model. It is possible that ammonia concentration can vary significantly in practice due to weather and the specific working environment. Therefore, field testing in scenarios where ammonia concentration is potentially high (e.g. at the bottom of a form, at the opening of a read y mix truck, or just above the
109 surface area of fresh concrete in the chute) can provide useful data in establishing the threshold for ammonium concentration in fly ash or concrete.
110 Table 6 1 Assumed p arameters in Scenario 1 Scenario Parameters Assumed Values Concrete density 2400 kg/m 3 Volume of Concrete 6.88 m 3 (9 cubic yards) Concrete thickness 20.32 cm (8 inches) Dimensions of the concrete slab 1 6.72 m (L) 5 m (W) M ass fraction of cement in concrete 20% M ass fraction of fly ash in cement 50% Placement time 30 min Ammonia concentration in fly ash 0 200 ppm Transportation/mixing time 0 1 hr (0 min, 30 min and 1 hr will be used for calculations) Wind velocity 2 1 6 m/s (1 m/s, 3m/s and 6 m/s will be used for the calculations) 1 Dimensions of the concrete slab is decided according to the volume of the concrete and its thickness. 2 Wind velocities of 1 6 m/s were selected from the Pasquill Stability Class of Weather Tables.
111 Table 6 2 Assumed p arameters in Scenario 2 Scenario Parameters Assumed Values Concrete density 2400 kg/m 3 Mass fraction of cement in concrete 20% Mass fraction of fly ash in cement 50% Transportation/mixing period 0 1 hr Placing/consolidating period 1 hr Ammonia concentration in fly ash 0 200 ppm Pump rate 20 75 cubic yards/hr Table 6 3 Assumed p arameters in Scenario 3 Scenario Parameters Assumed Values Concrete density 2400 kg/m 3 Concrete mixed 6.88 m 3 (9 cubic yards) Concrete / Ready Mix concrete truck volume ratio 0.5 ( Truck Volume: 20 cubic yards; 3.7 m in length and 2 m in diameter) M ass fraction of cement in concrete 20% M ass fraction of fly ash in cement 50% Ammonia concentration in fly ash 0 200 ppm Concrete mixing time of the truck 0 1 hr
112 Table 6 4 Assumed w orking c onditions in Scenario 4 Scenario Parameters Assumed Values Concrete density 2400 kg/m 3 Concrete mixed 6.88 m 3 (9 cubic yards) Concrete thickness 20.32 cm (8 inches) Size of the room 2 6.72 m (L) 5 m (W) 3 m (H) M ass fraction of cement in concrete 20% M ass fraction of fly ash in cement 50% Workers involved 6 people Minimum ventilation rate 1 7.5 L/s per person Placing time 30 min Ammonia concentration in fly ash 0 200 ppm Transportation/mixing period 0 1 hr Ventilation rate 0 200 L/s per person 1 Ventilation rate of 7.5 L/s per person is selected according to the minimum house ventilation rate specified in ASHRAE 62. 2 Height (H) of the room is selected according to the typical ceiling height in practice; length (L) and width (W) of the room are decided according to the overall volume of concrete and concrete thickness.
113 Figure 6 1 Workers around a f reshly p lac ed c oncrete s lab in an o pen e nvironment a) b) Figure 6 2 Examples of Scenario 1 : a) Roadway Placement and b) Bridge Deck Placement Figure 6 3 Dispersion m odel in Scenario 1 with w ind
114 Figure 6 4 Placement of c oncrete in a f orm with h igh w alls Figure 6 5 E xample of Scenario 2
115 Figure 6 6 Diffusion m odel for Scenario 2 Figure 6 7. A w orker n ear the d rum of the c oncrete m ixing t ruck
116 Figure 6 8. Example of Scenario 3 Figure 6 9. Workers in a r oom with an a ir f low r ate F
117 Figure 6 10. E xample of s cenario 4
118 Figure 6 11. Initial a mmonia c oncentration in f ly a sh vs. a mmonia c oncentration in a ir (Scenario 1 with w ind v elocity of 1 m/s)
119 Figure 6 12. Ammonia c oncentration d ownwind the c oncrete s lab at v arious w ind s peeds
120 Figure 6 13. Ammonia c oncentration in the d iffusion f ield at the e nd of 1 hr c ontinuous p lacement with the m aximum a mmonia r eleasing r ate Figure 6 14. A mmonia c oncentration as a f unction of t ime in the h eadspace of the r eady mix t ruck ( i nitial a mmonium c oncentration in f ly a sh v aries from 5 ppm to 50 ppm)
121 Figure 6 1 5. P redicted a mmonia c oncentration c hange with t ime in s cenario 4
122 Figure 6 16. 8 hr TWA a mmonia c oncentration as a f unction of i nitial a mmonia c oncentration in f ly a sh with v arious v entilation r ates for t hree t ransportation/ m ixing t imes
123 CHAPTER 7 CONCLUSIONS To estimate worker s exposure to air pollutants in an occupational environment a personal sampler was designed and fabricated to measure inorganic acidic gases and mist concentrations simultaneously and accurately. In the design, t here were several specific considerations. The first consideration was that the hazard caused by inhaled particles is closely related to their sizes. Therefore, a parallel impactor consist ing of 4 impactors of different cut sizes (6.6, 4.6, 3.5 and 2.2 ) to follow human respirable particle convention was selected to be the first part of the sampler. Then, to remove gases without causing interferences with fine particles, a new type of denuder PMD which utilizes porous membrane as the material to increase surface area was developed Cellulose filter, chosen for its proper strength and low cost w as cut by a laser plotter and assembled into multi grid shapes Four types of PMDs (PMD I, II, III and IV) with increasing numbers of channels and decreasing channel opening area s were constructed The weight of the PMDs was 6 to 9 g. The last stage of the s ampler, the filter pack, was designed to collect all small particles that penetrate the first and the second part s There were two filters in the filter pack. The first filter was designed to collect aerosols that pass through the denuder. The second filte r was coated with sodium carbona te to collect acidic gases evolv ing from collected aerosols on the first filter. Two lab testing prototypes were fabricated according to the design. T he design of t he second prototype addressed the leaking problem of the first prototype The casing of the second prototype was made of Delrin, while the parallel impactor was
124 made of stainless steel. This prototype was 12.2 cm long and 5.5 cm in outside diameter. Its weight was 320 g. ampling inorganic acid was measured in the laboratory. The parallel impactor was prove n to follow closely with the respirable penetration charact eristics for particles ranging from Gas collection efficiency of the f gas collec tion efficiency for SO 2 gas was evaluated both theoretically and experimentally. Based on the Gormley and Kennedy equation, when the denuder length is greater than 1.5 cm, the collection efficiency of the four types of PMDs for SO 2 should be close to 100%. However, the assumption of the Gormley and Kennedy equation that the tube surface serves as a perfect sink was i mpractical Hence, the PMDs performance deviated from the theoretical prediction In the experiment, u sing 10% sodium carbonate coating and a feed concentration of 1 ppm, the collection efficiency of PMD I version (a) for sulfur dioxide over 8 hours was higher than 99.9%. For a feed concentration of 10 ppm, 5 hr TWA collection efficiency for sulfur dioxide was 73.1%, 82.8%, 90.9%, 97.2% for PMD I, II, III and IV, respectively, compared with 96.6% of the Glass Honeycomb Denuder (GHD) which has similar structure to PMD IV. However, the weight of PMD IV is only one tenth of the lar capacity to traditional glass honeycomb denuders, yet are much lighter and less expensive. This study demonstrates the great potential of this new type of denuder for many applications in the field of environmental and industrial hygiene monitoring. Pa rticle loss
125 of the four types of PMDs in the particle was also measured and was 2.9%, 5.2%, 5.7% and 7.3%, respectively. The PMD IV s gas adsorbing capacity for HNO 3 and HCl was also evaluated The experimental results showed that the PMD IV achieved collection efficiency for HNO 3 and HCl gas higher than 95% for 4 hours with gas feed concentrations twice their OSHA standard s Experiments proved that the entire personal sampler could collect mor e than 99% of aerosols for 8 hours. Collection efficiency of the NIOSH 7903 sampler was 95.27 3.84%, which is lower than that of the personal sampler. In testing co existence of SO 2 gas and H 2 SO 4 mist, e xperimental results showed that the amount of sulfa te ion collected on the impactor and the after filters with SO 2 was similar to that without SO 2 However, on the denuder, it was clearly shown that the amount of sulfate ion collected was much higher when sampling air contained SO 2 gas than without SO 2 gas In other words, the majority of SO 2 gas was collected by the denuder instead of other parts of the personal sampler. Therefore, by sampling gas es and particles using different components of the sampler, interference between particles and gases can be eff ectively minimized. By sampling gas and aerosol on different parts, the personal sampler has overcome the sampling artifact issue of the current NIOSH sampler In addition, t he sampler has demonstrated its high capacity and compact size. It can be applied in an occup ational environment to better a ss dense inorganic acid mists, which is a potential carcinogenic threat that may be present in various industries such as fertilizer and l ead acid batteries manufacturing copper re fining and electroplating industries. Accurate assessment of workers risk of exposure could
126 provide information to health and safety staff s of these industries in developing cost effective respiratory protection programs In addition, it is also necessary for regulators in developing science based limits and protection policies for workers in these industries. Ammonia gas diffusion from the concrete added with AFA can last several days to several weeks. After the ammonia gas release rate gradually decrease d to a non detectable level, there was still 20% to 70% of ammonium retained in the concrete. Of all the ammonia gas released during the entire concrete mixing and hardening period, over 80% was emitted in the first 8 hour of concrete mixing and initial se ttling periods. According to different physical and chemical conditions of the concrete, the concrete mixing and curing process were separated into three periods: mixing, initial settling and curing. In the mixing period, ammonia gas release was enhanced b y stirring in addition to convective mass transfer. In the initial curing period, NH 3 release occurred on the transfer plays a dominant role in this process. Then, as water co ntent further evaporated from the concrete surface and capillaries in the concrete were filled by calcium silicate hydrate during the curing period, resistance for NH 3 to diffuse from inside the concrete to the surface got larger, which makes the diffusion inside the concrete the rate limiting factor of NH 3 release. The one dimensional diffusion mass transfer of NH 3 inside concrete can be mixing period 9.1 10 8 ( 1.25 10 7 ) m 2 /s. In the initial settling period, the diffusivity was 3.93 10 9 ( 2.78 10 9 ) m 2 /s, which was higher but close to reported NH 3 diffusivity in water of similar temperature. In the curing period, the diffusivity quickly decreased to
127 below detection lim it due to decreased moisture content and filled pores of the concrete by calcium silicate hydrate. In this study, the physical and chemical processes of NH 3 release from concrete were quantitatively analyzed for the first time. In addition, a diffusion mod el derived from 3 in the concrete. With the diffusion model, NH 3 gas flux at the interface between the concrete and the atmosphere at any time point from concrete mixing can be esti mated. Such information is critical in assessing ammonia gas exposure of workers around the concrete with given environmental parameter values, such as wind speed or air ventilation in a building. Four typical scenarios, including outdoor concrete constru ction, placing concrete into a form with high walls, concrete ready mix truck and in a r oom with or without ventilation were described and important environmental parameter values for each scenario such as wind speed or air ventilation in a building were reasonably assumed. The effective diffusion coefficient s were then applied in the well mixed mass balance model for each scenario to estimate workers possible exposure to ammonia gas. Although the results varied as a function of input parameter, exposure assessment modeling for scenarios where concrete was poured in outdoor applications found that the NIOSH recommended exposure level in air of 25 ppm would not be exceeded under normal working conditions when ammonia concentrations in fly ash were less than 100 ppm. Exposure assessment modeling for scenarios where concrete was poured in an indoor setting revealed that if a minimum ventilation rate was maintained, ammonia concentrations of 50 ppm in fly ash would not exceed the NIOSH recommended
128 expo sure level in air of 25 ppm. However, the exposure assessment modeling also found that under some scenarios (e.g., poorly ventilated environments such as one that might occur inside the drum of a mixing truck and at the bottom of a high walled form with a continuous pour of concrete), ammonia concentrations to which a worker would be exposed could exceed safe levels even if ammonia concentrations in fly ash were less than 100 ppm. Based on the modeling results, it is recommend ed that if FDOT chooses to s pecify an upper limit to the ammonia concentration in fly ash used for concrete production, 100 ppm would be protective of human health for most situations and would be achievable by vendors in the State of Florida using current ash treatment practices. H o wever, it is recommended that under conditions where poor ventilation might occur, appropriate safety protocols should be evaluated by a trained professional (e.g., certified industrial hygienist) prior to job commencement so necessary safety measures and/ or job specific ammonia concentration limits are used. Job managers should also be aware that even with the utilization of a 100 ppm ammonia limit in fly ash, the occurrence of ammonia odors may not be eliminated. Fly ash, as a by product of the generation of electricity from coal, is widely applied in road construction and brick making industries. It is very important that a proper threshold of ammonia concentration in fly ash be adopted by regulators to these industries. The reasonable threshold not only matters in protecting workers health and safety in these industries, but also plays a significant role for companies that process fly ash into consistent high quality product for the concrete industry.
129 T o better a ss future research activities should consider the following: f irst of all, efforts should be made to explore commercialization of the personal sampler. Since workers in the field desire to have a dispos able unit to operate, the sampler should be made so For example, some material of the lab prototype (Delrin and stainless steel) can be replaced with more cost saving and acid resistant plastic, e.g., th e a crylonitrile b utadiene s tyrene (ABS) plastic The size and weight of the sampler can further be reduced while keeping the important parameters (e.g., impactor nozzle sizes, channel number and length of the PMDs) the same. Second, some possible improvements can be made on the personal sampler for more app lications. For instance, a new impactor following the thoracic fraction curve can be designed and adopted into the sampler to follow recommendation for sulfuric acid. The PMDs can be coated with c itric acid to adsorb ammonia gas and lab testing should be carried out to ensure its capacity for ammonia Third, field testing in phosphate facilities should be carried out to demonstrate the advantages of the new sampler to industrial users in addition to lab testing data. Once the sam pler demonstrates its ability in adsorbing ammonia, it can also be used to test validating the results of the mathematical model and optimizing the assumptions and boundary co nditions use d in the equations of the model In addition, data from field testing are also complementary to modeling results for regulators and industrial administrators in deciding the exposure limit value or proper air pollution control techniques.
130 LIST OF REFERENCES Acker K, Mller D, Auel R, Wieprecht W, Kalass D (2005) Concentrations of nitrous acid, nitric acid, nitrite and nitrate in the gas and aerosol phase at a site in the emission zone during ESCOMPTE 2001 experiment. Atmospher ic Research 74 (1):507 524 Aizenberg V, Grinshpun SA, Willeke K, Smith J, Baron PA (2000) Measurement of the sampling efficiency of personal inhalable aerosol samplers using a simplified protocol. J Aerosol Sci 31 (2):169 179. doi: http://dx.doi.org/10.1016/S0021 8502(99)00037 3 Arogo J, Zhang R, Riskowski G, Christianson L, Day D (1999) Mass transfer coefficient of ammonia in liquid swine manure and aqueous solutions. Jou rnal of Agricultural Engineering Research 73 (1):77 86 Arshak K, Moore E, Lyons G, Harris J, Clifford S (2004) A review of gas sensors employed in electronic nose applications. Sensor Review 24 (2):181 198 Ballun G, Hajdu F, Harsanyi G Highly sensitive amm onia sensor. In: Electronics Technology: Integrated Management of Electronic Materials Production, 2003. 26th International Spring Seminar on, 2003. IEEE, pp 471 475 Baron PA (1998) Personal aerosol sampler design: a review. Applied Occupational and Enviro nmental Hygiene 13 (5):313 320 Bartley DL, Doemeny, L. J. (1986) Critique of 1985 ACGIH report on particle size selective sampling in the workplace. The American Industrial Hygiene Association Journal 47 (8):443 447 BGI (2013) Collison nebulizer droplet nu mber output. http://www.bgiusa.com/agc/droplet_number_output.htm Accessed Aug. 20 2013 Bouwmeester RJ, Vlek PL (1981) Rate control of ammonia volatilization from rice paddies. Atmospheric Environment (1967) 15 (2):131 140 Breuer D, Heckmann P, Gusbeth K, Schwab G, Blaskowitz M, Moritz A (2012) Sulfuric acid at workplaces applicability of the new Indicative Occupational Exposure Limit Value (IOELV) to thoracic particles. Journal of Environmental Monitoring Carlton GN, Flynn MR (1997) A model to estimate worker exposure to spray paint mists. Applied Occupational and Environmental Hygiene 12 (5):375 382 Carslaw KS, Clegg SL, Brimblecombe P (1995) A Thermodynamic Model of the System HCl HNO 3 H 2 SO 4 H 2 O, Including Solubilities of HBr, from< 200 to 328 K. Journal of Physical Chemistry 99 (29):11557 11574
131 Chen CC, Huang S H (1999) Shift of aerosol penetration in respirable cyclone samplers. American Industrial Hygiene Association journal 60 (6):720 729 chen WH, Chen YY, Huang C (2011) A Simplified Model of Predicting SO2 Absorption by Single Atmospheric Raindrops with Chemical Dissociation and Internal Circulation. Aerosol and Air Quality Research 11 (7):13 Cheng YS (2001) Condensation Detect ion and Diffusion Size Separation Techniques. In: Baron PA, Will eke K (eds) Aerosol Measurement : Principles, Techniques, and Applications vol Ch. 19. 2nd edn. Wiley, New York, pp 569 601 Chow JC (1995) Measurement methods to determine compliance with ambie nt air quality standards for suspended particles. Journal of the Air & Waste Management Association 45 (5):320 382 constant of the bisulfate ion from Raman and ion interaction model ing studies of 109 (11):2703 2706 Clegg SL, Brimblecombe P (2013) Extended AIM Aerosol Thermodynamics Model. http://www .aim.env.uea.ac.uk/aim/aim.php Accessed August 14 2013 Cussler EL (1997) Diffusion: Mass transfer in fluid systems. Cambridge U niversity P ress Dasch JM, Cadle SH, Kennedy KG, Mulawa PA (1989) Comparison of Annular Denuders and Filter Packs for Atmospheri c Sampling. Atmos phric Environ ment 23 (12):2775 2782. Dasgupta PK, Ni L, Poruthoor SK, Hindes DC (1997) A multiple parallel plate wetted screen diffusion denuder for high flow air sampling applications. Analytical Chemistry 69 (24):5018 5023 Demokritou P, Kavouras IG, Ferguson ST, Koutrakis P (2001) Development and laboratory performance evaluation of a personal multipollutant sampler for simultaneous measurements of particulate and gaseous pollutants. Aerosol Science & Technology 35 (3):741 752 Duan N (198 2) Models for human exposure to air pollution. RAN D Corp., Santa Monica, CA (USA) Dubbe A (2003) Fundamentals of solid state ionic micro gas sensors. Sensors and Actuators B: Chemical 88 (2):138 148 Durham J, Ellestad T, Stockburger L, Knapp K, Spiller L (1986) A transition flow reactor tube for measuring trace gas concentrations. Journal of the Air Pollution Control Association 36 (11):1228 1232
132 Eatough DJ, Wadsworth A, Eatough DA, Crawford JW, Hansen LD, Lewis EA (1993) A multiple system, multi channel d iffusion denuder sampler for the determination of fine particulate organic material in the atmosphere. Atmospheric Environment Part A General Topics 27 (8):1213 1219 Eller PM, Cassinelli ME (1994) NIOSH manual of analy tical methods. DIANE Publishing Fabri s J, Grner P, Kauffer E, Wrobel R, Vigneron J (1998) Personal Thoracic CIP10 T Sampler and Static Version CATHIA T. Annals of Occupational Hygiene 42 (7):453 465 FDOT (2010) Standards for specifications for road and bridge construction. Florida Departmen t of Transportation, Tallahassee, FL Ferm M (1979) Method for determination of atmospheric ammonia. Atmospheric Environment (1967) 13 (10):1385 1393 Fitz DR, Motallebi N (2000) A fabric denuder for sampling semi volatile species. Journal of the Air & Waste Management Association 50 (6):981 992 Gao Y, Chen X, Fujisaki G, Mehta A, Suuberg E, Hurt R (2002) Dry and semi dry methods for removal of ammonia from pulverized fuel combustion fly ash. Energy & Fuels 16 (6):1398 1404 Giampa VM The fate of ammonia and m ercury in the carbon burn In: Fifth International Ash Utilization Symposium, University of Kentucky, Lexington, 2003. pp 621 629 Giampa VM, Plaza OP Ammonia removal from coal fly ash by carbon burn Out. In: Proceedings: 14th Internation al Symposium on Management and Use of Coal Combustion Products (CCPs), 1999. Electric Power Research Institute, pp 1 4 Glicksman ME (2000) Diffusion in solids: field theory, solid state principles, and applications. vol v. 1. Wiley, Gordin A, Amirav A (20 00) SnifProbe: new method and device for vapor and gas sampling. Journal of Chromatography A 903 (1 2):155 172. doi:10.1016/s0021 9673(00)00877 3 F (2001) Study of fifteen respirable aerosol samplers used in occupational hygiene. Annals of Occupational Hygiene 45 (1):43 54 Hangartner M, Kirchner M, Werner H (1996) Evaluation of passive methods for measuring ozone in the European Alps. Analyst 121 (9):1269 1272 Haslam R, Hershey R, Kean R (1924) Effect of Gas Velocity and Temperature on Rate of Absorption. Industrial & Engineering Chemistry 16 (12):1224 1230
133 Hayami H (2005) Behavior of secondary inorganic species in gaseous and aerosol phases measured in Fukue Island, Japan, in dust season. Atmospheric Environm ent 39 (12):2243 2248 Hinds WC (ed) (1999) Aerosol Technology: properties, behavior, and measurement of airborne particles. 2nd e d. edn. John Wiley & Sons, Inc. Hsu YM, Wu CY, Lundgren D, Nall JW, Birky BK (2007) Chemical Characteristics of Aerosol Mists in Phosphate Fertilizer Manufacturing Facilities. J ournal of Occup ational and Env ironmental Hyg iene 4:17 25 Hsu YM, Wu CY, Lundgren DA, Birky BK (2008) Minimization of Artifacts in Sulfuric Acid Mist Measurement Using NIOSH Method 7903. Environmental Scien ce & Technology 42 (15):5694 5699 Huang Z, Harrison RM, Allen AG, James JD, Tilling RM, Yin JX (2004) Field intercomparison of filter pack and impactor sampling for aerosol nitrate, ammonium, and sulphate at coastal and inland sites. Atmospheric Research 7 1 (3):215 232 Keck L, Wittmaack K (2006) Miniature parallel plate denuder for the collection of inorganic trace gases and their removal from aerosol laden air. J ournal of Aerosol Sci ence 37 (10):1165 1173. doi:10.1016/j.jaerosci.2005.11.005 Kenny L, Aitken R, Baldwin P, Beaumont G, Maynard A (1999) The sampling efficiency of personal inhalable aerosol samplers in low air movement environments. J ournal of Aerosol Sci ence 30 (5):627 638 Kenny LC (1996) Developments in workplace aerosol sampling. A review. Ana lyst 121 (9):1233 1239 Kenny LC, Aitken R, Chalmers C, Fabris JF, Gonzalez Fernandez E, Kromhout H, Lidn G, Mark D, Riediger G, Prodi V (1997) A collaborative european study of personal inhalable aerosol sampler performance. The Annals of Occupational Hy giene 41 (2):135 153. doi:10.1016/s0003 4878(96)00034 8 Kosmatka SH, Kerkhoff B, Panarese WC (2002) Design and control of concrete mixtures. 14 edn. Portland Cement Association, Skokie, Illinois Koutrakis P, Fasano AM, Slater JL, Spengler JD, McCarthy JF, Leaderer BP (1989) Design of a personal annular denuder sampler to measure atmospheric aerosols and gases. Atmospheric Environment (1967) 23 (12):2767 2773. doi:10.1016/0004 6981(89)90557 x Koutrakis P, Sioutas C, Ferguson ST, Wolfson JM, Mulik JD, Burton RM (1993) Development and evaluation of a glass honeycomb denuder/filter pack system to collect atmospheric gases and particles. Environmental Science & Technology 27 (12):2497 2501. doi:10.1021/es00048a029
134 KROMHOUT H, SWUSTE P, BOLEIJ JS (1994) Empirical modelling of chemical exposure in the rubber manufacturing industry. Annals of Occupational Hygiene 38 (1):3 22 Krutovertsev S, Ivanova O, Sorokin S (2001) Sensing properties of polyaniline films doped with Dawson heteropoly compounds. Journal of Analytica l Chemistry 56 (11):1057 1060 Klaots I, Gao YM, Hurt RH, Suuberg EM Adsorption of ammonia on coal fly ash. In: International Ash Utilization Symposium, 2001. pp 1 59 Larrimore L (2002) Effects of ammonia from post combustion NOx control on ash handling an d use. Fuel Chemistry Division Reprints 47 (2):2 Leduc D, Gris P, Lheureux P, Gevenois P A, De Vuyst P, Yernault JC (1992) Acute and long term respiratory damage following inhalation of ammonia. Thorax 47 (9):755 757 Lee K, Yanagisawa Y, Hishinuma M, Spengler JD, Billick IH (1992) A passive sampler for measurement of carbon monoxide using a solid adsorbent. Environmental Science & Technology 26 (4):697 702 Lin IK, Bai HL, Wu BJ (2010) Analysis of Relationship between Inorganic Gases and Fine Particles in Cleanroom Environment. Aerosol and Air Quality Research 10 (3):10. doi:10.4209/aaqr.2009.10.0065 Lioy PJ (1995) Measurement methods for human exposure analysis. Environmental health perspectives 103 (Suppl 3):35 Lippmann M (1976) Fine Particles: Aer osol Generation, Measurement, Sampling, and Analysis. Academic Press, New York Lundstrm I, Svensson C, Spetz A, Sundgren H, Winquist F (1993) From hydrogen sensors to olfactory images twenty years with catalytic field effect devices. Sensors and Actuators B: Chemical 13 (1):16 23 Marple VA (1978) Simulation of respirable penetration characteristics by inertial impaction. J Aerosol Sci 9 (2):125 134. doi:10.1016/0021 8502(78)90072 1 Marple VA, Liu BY (1974) Characteristics of laminar jet impactors. Environm ental Science & Technology 8 (7):648 654 Martell D, Warburton RG, Lindner LA, Lindner J (2004) Amperometric gas sensor. EP Patent 0,762,116,
135 Massucci M, Clegg SL, Brimblecombe P (1999) Equilibrium partial pressures, thermodynamic properties of aqueous and solid phases, and Cl 2 production from aqueous HCl and HNO 3 and their mixtures. The Journal of Physical Chemistry A 103 (21):4209 4226 McConnaughey PW, McKee ES, Pritts IM (1985) Passive colorimetric dosimeter tubes for ammonia, carbon monoxide, carbon dio xide, hydrogen sulfide, nitrogen dioxide and sulfur dioxide. American Industrial Hygiene Association journal 46 (7):357 362 Mohan M, Siddiqui TA (1998) Analysis of various schemes for the estimation of atmospheric stability classification. Atmospheric Envi ronment 32 (21):3775 3781. doi: http://dx.doi.org/10.1016/S1352 2310(98)00109 5 Molen Jvd, Beljaars A, Chardon W, Jury W, Faassen Hv (1990) Ammonia volatilization from arable land after application of cattle slurry. 2. Derivation of a transfer model. Netherlands journal of agricultural science 38 (3A):239 254 Monn C (2001) Exposure assessment of air pollutants: a review on spatial heterogeneity and indoor/outdoor/personal exposure to susp ended particulate matter, nitrogen dioxide and ozone. Atmospheric Environment 35 (1):1 32 Monn C, Hangartner M (1990) Passive sampling for ozone. Journal of the Air & Waste Management Association 40 (3):357 358 Muck R, Steenhuis T (1982) Nitrogen losses fr om manure storages. Agricultural wastes 4 (1):41 54 Ni J (1998) Emission of carbon dioxide and ammonia from mechanically ventilated pig house. Dissertationes de Agricultura Ni J (1999) Mechanistic models of ammonia release from liquid manure: a review. Jou rnal of Agricultural Engineering Research 72 (1):1 17 Nicolas Debarnot D, Poncin Epaillard F (2003) Polyaniline as a new sensitive layer for gas sensors. Analytica Chimica Acta 475 (1):1 15 NTP N (2005) 11th Report on Carcinogens. Rep Carcinog:1 A32 Olesen JE, Sommer SG (1993) Modelling effects of wind speed and surface cover on ammonia volatilization from stored pig slurry. Atmospheric Environment Part A General Topics 27 (16):2567 2574. doi: ht tp://dx.doi.org/10.1016/0960 1686(93)90030 3 Ott WR (1982) Concepts of human exposure to air pollution. Environment International 7 (3):179 196
136 Palmes E, Gunnison A, DiMattio J, Tomczyk C (1976) Personal sampler for nitrogen dioxide. The American Industria l Hygiene Association Journal 37 (10):570 577 Pathak RK, Chan CK (2005) Inter particle and gas particle interactions in sampling artifacts of PM2.5 in filter based samplers. Atmospheric Environment 39 (9):1597 1607 Patoczka J, Wilson DJ (1984) Kinetics of the desorption of ammonia from water by diffused aeration. Separation Science and Technology 19 (1):77 93 Pearce TC, Schiffman SS, Nagle HT, Gardner JW (2006) Handbook of machine olfaction. Wiley. com, Perrino C, De Santis F, Febo A (1990) Criteria for th e choice of a denuder sampling technique devoted to the measurement of atmospheric nirous and nitric acids. Atmospheric Environment Part A General Topics 24 (3):617 626. doi:10.1016/0960 1686(90)90017 h Perrino C, Gherardi M (1999) Optimization of the coat ing layer for the measurement of ammonia by diffusion denuders. Atmospheric Environment 33 (28):4579 4587. doi:10.1016/s1352 2310(99)00273 3 Poling BE, Prausnitz JM, John Paul OC, Reid RC (2001) The properties of gases and liquids, vol 5. McGraw Hill New Y ork, Possanzini M, Febo A, Liberti A (1983) New design of a high performance denuder for the sampling of atmospheric pollutants. Atmospheric Environment (1967) 17 (12):2605 2610 Pui DYH, Lewis CW, Tsai CJ, Liu BYH (1990) A compact coiled denuder for atmos pheric sampling. Environ Sci Technol 24 (3):5 Rathbone RF, Robl TL (2001) A study of the effects of post combustion ammonia injection on fly ash quality: characterization of ammonia release from concrete and mortars containing fly ash as a pozzolanic admix ture. University of Ke ntucky Research Foundation (US) Rathbone RF, Tyra MA, Harper L (2001) Rates of ammonia loss from mortar. Paper presented at the International Ash Utilization Symposium Sekiguchi K, Kim K, Kudo S, Sakamoto K, Otani Y, Seto T, Furuuch i M, Yoshikawa F, Ehara K, Hirasawa M (2009) Evaluation of Multichannel Annular Denuders for a Newly Developed Ultrafine Particle Sampling System. Aerosol Air Qual Res 9:50 64
137 Sexton K, Ryan PB (1988) Assessment of human exposure to air pollution: methods, measurements, and models. Air Pollution, the Automobile, and Public Health, Ann Y Watson, Richard R Bates, and Donald Kennedy, eds, National Academy Press: New York:207 238 Sigma Aldrich Supelco ORBO 53 activated silica gel tube http://www.sigmaaldrich.com/Graphics/Supelco/objects/6800/6734.pdf Sigma Aldrich. doi: http://www.sigmaaldrich.com/G raphics/Supelco/objects/6800/6734.pdf Simon PK, Dasgupta PK (1993) Wet effluent denuder coupled liquid/ion chromatography systems: annular and parallel plate denuders. Analytical Chemistry 65 (9):1134 1139 Sioutas C, Wang P, Ferguson S, Koutrakis P, Mulik J (1996) Laboratory and field evaluation of an improved glass honeycomb denuder/filter pack sampler. Atmospheric Environment 30:10 Sommer SG, Olesen JE, Christensen BT (1991) Effects of temperature, wind speed and air humidity on ammonia volatilization fro m surface applied cattle slurry. J Agric Sci 117 (1):91 100 Strike D, Meijerink M, Koudelka Hep M (1999) Electronic noses A mini review. Fresenius' journal of analytical chemistry 364 (6):499 505 Subramaniam R, Bagchi S, Jadhav R (2009) Reduction in partic ulate matter concentration by ammonia injection in flue gas a case study. Asian Journal of Water, Environment and Pollution 6 (1):109 112 Tielemans E, Kupper LL, Kromhout H, Heederik D, Houba R (1998) Individual based and group based occupational exposure assessment: some equations to evaluate different strategies. Annals of Occupational Hygiene 42 (2):115 119 Timmer B, Olthuis W, Berg Avd (2005) Ammonia sensors and their applications a review. Sensors and Actuators B: Chemical 107 (2):666 677 Trak umas S, Salter E Parallel particle impactor novel size selective particle sampler for accurate fractioning of inhalable particles. In: Journal of Physics: Conference Series, 2009. IOP Publishing, p 012060 Trivedi S, Phadke R (2009) Flue gas conditioning. I n: Electrostatic Precipitation. Springer, pp 389 394 Tsai CJ, Wang SH (1999) Testing the porous metal denuder using high concentration acidic aerosols. J Aerosol Sci 30 (Supplement 1):S741 S742. doi:10.1016/s0021 8502(99)80381 4
138 Tsai CJ, Huang CH, Lin YC, Shih TS, Shih BH (2003) Field Test of a Porous Metal Denuder Sampler. Aerosol Sci Tech 37:967 974 Tsai CJ, Huang CH, Lu HH (2004) Adsorption capacity of a nylon filter of filter pack system for HCl and HNO3 gases. Separation Science and Technology 39 (3): 629 643 Tsai CJ, Huang CH, Wang SH, Shih TS (2001) Design and Testing of a personal porous metal denuder. Aerosol Science & Technology 35 (1):611 616 impact on flue gas treatment technologies. Chemical Engineering Science 49 (24):4315 4325 Tyra M, Robl T (2001) Research in ammonia diffusivity in Portland cement based mixes. Paper presented at the International Ash Utilization Symposium, USHHS (2004) Toxicological Profile for Ammo nia. Center for Diease Control, Washington, D. C. Van der Brugghen F, Gast C, Van den Berg J, Kuiper W, Visser R Problems encountered during the use of ammonium contaminated fly ash. In: Proceedings EPRI/EPA 1995 Joint Symposium on Stationary Combustion NO x Control., 1995. Weis BK, Balshaw D, Barr JR, Brown D, Ellisman M, Lioy P, Omenn G, Potter JD, Smith MT, Sohn L (2005) Personalized exposure assessment: promising approaches for human environmental health research. Environmental health perspectives 113 ( 7):840 Welty JR, Wicks CE, Rorrer G, Wilson RE (2009) Fundamentals of momentum, heat, and mass transfer. John Wiley & Sons, Woskie SR, Smith TJ, Hammond SK, Hallock MH (1994) Factors affecting worker exposures to metal working fluids during automotive com ponent manufacturing. Applied Occupational and Environmental Hygiene 9 (9):612 621 Wright BM (1954) A size selecting sampler for airborne dust. Brit J industr Med 11:284 288 Zakrzewska K (2001) Mixed oxides as gas sensors. Thin Solid Films 391 (2):229 238 Zhang R, Day D, Christianson L, Jepson W (1994) A computer model for predicting ammonia release rates from swine manure pits. Journal of Agricultural Engineering Research 58 (4):223 229
139 BIOGRAPHICAL SKETCH Lin Shou was born in 1983 in Zhejiang, China. She was granted with a Bachelor of Engineering degree in T hermal E nergy & P ower E ngineering in June 2005 at Zhejiang University, China. Her over all GPA rank ed top 5 % among 112 students of the same grade In her four year undergraduate study, she won scholarship and the honor of excellent student leader and outstanding student of Zhejiang University every year. After granted with a bachelor s degree, Lin Shou was admitted to the two year m aster s p rogram in the s pecialty of R efrigeration and C ryogenics E ngineering in Zhejiang University exempted from the usually mandatory examinations for graduate studies. During her m aster s study, Lin Shou was awarded G ranted H ope S pecial s cholarship, a scholarship onl y given to the top two students in her specialty and t he f irst c lass s cholarship Also she was the only student in her specialty that got the honor of e xcellent graduate of Zhejiang Province China in April 2007. She joined Dr. Chang Yu Wu s research group at the University of Florida in 2009 and started pursuing her Ph.D. degree in the D epartment of E nvironmental E ngineering Science s Her research focused on diffusional release of pollutants into ambient air and method for enhancing diffusi onal collection in air sampling. She was awarded the Third Place in the Air & Waste Management Association (A&WMA) s Annual Poster Competition in 2010, and also was awarded Honorable Mention in the 2012 Doctoral Level Platform Paper Competition of A&WMA.