Dispersion of gas pollutant in a fan-filter-unit (FFU) cleanroom

Numerical simulation and experimental measurement of flow and concentration fields in a working fan-filter-unit (FFU) cleanroom have been conducted in this study. The purpose of the study is to find out the unsteady concentration distribution of a leaking gas pollutant. The standard K–ε model was used for the simulation of the flow field. To obtain the gas concentration field, SF₆ gas with a certain concentration was released as a simulated leaking source from a valve manifold box (VMB) for 5 or 10 min, respectively. Three Fourier transform infrared spectrometers (FTIRs) were simultaneously used to measure the spatial and temporal distributions of SF₆ concentrations. The measured data were then compared with the numerical results and the agreement is seen to be quite good. From the numerical results, the pollutant hot spots, peak pollutant concentration at the end of leaking, and time taken for the concentration to reduce to near background level are obtained.     

Experimental and numerical investigation of airflow and contaminant transport in an airliner cabin mockup

The study of airflow and contaminant transport in airliner cabins is very important for creating a comfortable and healthy environment. This paper shows the results of such a study by conducting experimental measurements and numerical simulations of airflow and contaminant transport in a section of half occupied, twin-aisle cabin mockup. The air velocity and air temperature were measured by ultrasonic and omni-directional anemometers. A gaseous contaminant was simulated by a tracer gas, sulfur hexafluoride or SF₆, and measured by a photo-acoustic multi-gas analyzer. A particulate contaminant was simulated by 0.7 μm di-ethyl-hexyl-sebacat (DEHS) particles and measured by an optical particle sizer. The numerical simulations used the Reynolds averaged Navier–Stokes equations based on the RNG k–ε model to solve the air velocity, air temperature, and gas contaminant concentration; and employed a Lagrangian method to model the particle transport. The numerical results quantitatively agreed with the experimental data while some remarkable differences exist in airflow distributions. Both the experimental measurements and computer simulations were not free from errors. A complete and accurate validation for a complicated cabin environment is challenging and difficult.     

Dispersion of Automotive Alternative Fuel Vapors within a Residence and Its Attached Garage

Abstract A study was undertaken to investigate the use of mass balance modeling techniques to predict air pollution concentrations in residential settings where the source is evaporative emissions of alternative fuels emitted in the attached garage. Field study measurements of the spatial and temporal distribution of evaporative emissions in an attached garage were used to conduct this investigation. In this field study, known quantities of methanol were allowed to evaporate in the garage. Methanol concentrations were then measured and simulated in the garage. Because chemical and physical properties of this contaminant are clearly understood, it was possible to simulate the concentrations inside the remainder of the house. A multi-zonal mass balance model (CONTAM88) was used to predict the across-residence spatial and temporal distribution of concentrations of evaporative emissions from a methanol source inside the attached garage. The model's input data include physical characteristics of the house; parameters characterizing the leakiness of the house; heat, ventilation and air-conditioning system characteristics; fixed airflow data from vent registers; on-site meteorological measurements; and information on the contaminant source. Before using the model to simulate methanol dispersions, the model's performance was tested. Sulfur hexafluoride (SF₆) measurements were taken throughout the residence and garage. The model was used to predict interzonal airflow rates and SF6 concentration distributions within the garage and the house, as well as to investigate the well-mixed zone assumption for the garage. Modeling results show that predicted garage SF₆ concentrations agreed reasonably well with the measured concentrations under mixed conditions, but the model underpredicted the SF₆ concentrations within rooms of the house where mixing was probably incomplete. Methanol simulation results showed that the model underpredicted by approximately 15 percent the garage methanol concentration after methanol emissions stopped. This study found that evaporative emissions in an attached garage have a tendency to infiltrate the house, with rooms adjacent to the garage showing the highest levels of methanol concentrations. Thus automotive evaporative emissions may represent a source of indoor pollution and human exposure in a residential attached garage and also in other critical locations of the residence.     

Short-term dispersion of indoor aerosols: can it be assumed the room is well mixed?

Monitoring of aerosols is typically performed over 3 h to diurnal time scales for outdoor concentration levels and 15 min to 8 h scales indoors. At these scales, concentration is assumed to be well mixed with little spatio-temporal variability around the sampler. Less attention has been given to the potential for acute exposure to contaminants during the initial minutes after a point-source release, where point-wise concentrations may greatly exceed the well-mixed conditions. Here, we seek to demonstrate that the commonly used well-mixed assumption is flawed in the first minutes after a contaminant is released because point-wise concentration levels are initially highly non-uniform and are influenced by turbulent structures caused by the presence of obstacles in the room. This assumption was examined by releasing 3 μm aerosols in a test room with HEPA filter ventilation and by varying controlled conditions of room furnishings (furnished vs. unfurnished) and contaminant release locations (at the inlet vent or under a desk). For each experiment, aerosol concentrations were measured simultaneously at seven locations by nephelometry. Complementary computational fluid dynamics simulations were performed to lend confidence to the experiments and to provide detailed pictures of the velocity and particle concentration profiles. The experimental and numerical results corroborated the hypothesis. For both release locations in the furnished room, a completely well-mixed condition did not occur 600 s after the release, and aerosol dispersion was dictated by the turbulent airflow pattern. For the empty room, there was significantly less spatial variability in the point-wise measured concentrations after 300 s than for the furnished room. This information may aid in evaluating the potential for occupant exposure to aerosolized hazardous substances and in supporting optimization of detector placement.     

Spatial concentration variation of cooking-emitted particles in a residential kitchen

Residential kitchen cooking generates significant amount of gases and particulate pollutants and is regarded as a major pollutant source in residential microenvironments. Existing risk assessments always assume well-mixed condition. We used a portable nephelometer to measure PM3.5 oil particle concentration in a kitchen under low air exchange rate condition. Two sampling locations were chosen: one near (0.3 m) and another far (2.8 m) from the cooking stove. Numerical simulation was also used to model the particle transport in a model kitchen. Particles passing through the pre-defined sampling areas were tracked and counted. The simulated results agree reasonably well with the measured concentration. It shows that concentration at the near-field is almost three times as high as that at the far-field. Exposure assessment results based on well-mixed assumption should be applied with caution.     

Vapor adsorption characteristics of toluene in an activated carbon adsorbent-loaded nonwoven fabric media for chemical filters applied to cleanrooms

Chemical filters are used extensively in the cleanrooms of the semiconductor factories to remove airborne molecular contamination (AMC). Adsorption by activated carbons (AC) as media within the chemical filter is one of the practical methods for removal of gaseous contamination in a cleanroom. The objective of this study is to evaluate coconut shell activated carbon adsorbent-loaded nonwoven fabric media performance by determining the breakthrough curves, the linear driving force (LDF), the intra-particle diffusion characteristics, the empty bed contact time (EBCT) and the bed depth service time (BDST), the mass-transfer zone (MTZ), and pressure drop. The testing conditions were maintained at 28 ± 1 °C, and relative humidity at 40 ± 2% with face velocities of 0.076, 0.114 and 0.152 m/s for removal efficiency and capacity determination. The challenge gas concentrations of toluene were fixed at 10, 31, 42 and 70 ppm to accelerate the breakthrough of media adsorption. The concentrations were measured by a real-time photoionization detector. Results showed that breakthrough curves correlate to the challenge vapor concentration and the face velocity. Saturated adsorption ratio was increased with raised challenge gas concentration and increased face velocity significantly.     

Evaluation of the COMIS Model by Comparing Simulation and Measurement of Airflow and Pollutant Concentration

Abstract This paper describes the measured and calculated results of airflow rates and pollutant concentration profiles in an airtight test house, the aim being to evaluate the calculation model COMIS for multizone air infiltration and pollutant transport. Firstly, the leakage areas of internal doors, exterior walls and windows were measured by the fan pressurization method. Secondly, two measurements were carried out, assuming that the test house consisted of ten zones. The concentrations and injection rate of SF₆ were measured in order to determine the airflow rates by a system identification method. The boundary conditions, such as indoor and outdoor temperatures, wind speed and direction, and wind pressures were also recorded in situ and saved simultaneously on diskettes, using a computerized data acquisition system. Thirdly, the measured boundary data and leakage characteristics were used as input in the simulation of airflow using COMIS; initial concentrations, injection rate, along with the previous data were used for simulating pollutant transport, assuming tracer gas SF6 as a pollutant. Lastly, the comparisons between measurement and simulation results of airflow rates and pollutant concentrations were carried out by linear regression analysis. The correlation coefficient between the measured and calculated air change rates was 0.72, and that for pollutant concentration was 0.94.     

Influence of bathroom ventilation rates and toilet location on odor removal

This paper presents an experimental model of odor dispersion from a toilet in a mock-up bathroom. A bathtub, a sink, and a toilet were arranged in the bathroom. A hypothetical odor (SF₆ tracer gas) was emitted from a person sitting on the toilet in the bathroom. A new negative pressure wall-exhaust ventilation system, different from a traditional ceiling-exhaust system, was employed in the bathroom. The impacts of ventilation rates of 6.5, 8.5, 17, and 24 ACH (air changes per hour) and the location of the toilet on the concentration distribution of repugnant odors in the bathroom were evaluated. Local air quality index, QI, and odor removal efficiency, ORE, were introduced to evaluate the effectiveness of odor removal in the bathroom. Based on the results of our experiment, a quantitative assessment of odor removal in the bathroom is addressed in this paper.     

Cleanroom energy efficiency strategies: Modeling and simulation

To maintain ultra-low particle concentrations, cleanrooms can require several hundred air changes per hour. These ventilation rates make cleanrooms 30-50 times more energy intensive than the average U.S. commercial building. There are an estimated 12 million m² of cleanroom space in the U.S., consuming over 370 PJ of energy each year. This paper explores opportunities to improve the energy efficiency of cleanrooms while maintaining or improving operating conditions.This paper documents the modeling of a 1600 m² cleanroom in upstate New York. The TRNSYS model includes TMY2 weather data; building geometry and material properties; empirical data on occupancy, lighting and process equipment; and sophisticated HVAC systems. The model was validated based on metered steam, chilled water and electricity usage. Under 8% error was achieved in all fields.Four strategies were simulated: a heat recovery system for exhaust air, resulting in an 11.4% energy reduction with a 2.7-year simple payback; solar preheating of desiccant dehumidifier regeneration air (2.4% energy reduction, 11.5-year payback); improved lighting controls (0.3% energy reduction, 1.5-year payback); and demand-controlled filtration (4.4% energy reduction, 3.1-year payback). Implementation of recommended strategies is predicted to save 9 TJ, 862 tonnes of CO₂, and $164k annually.     

Aerodynamic effects of trees on pollutant concentration in street canyons

This paper deals with aerodynamic effects of avenue-like tree planting on flow and traffic-originated pollutant dispersion in urban street canyons by means of wind tunnel experiments and numerical simulations. Several parameters affecting pedestrian level concentration are investigated, namely plant morphology, positioning and arrangement. We extend our previous work in this novel aspect of research to new configurations which comprise tree planting of different crown porosity and stand density, planted in two rows within a canyon of street width to building height ratio W/H = 2 with perpendicular approaching wind. Sulfur hexafluoride was used as tracer gas to model the traffic emissions. Complementary to wind tunnel experiments, 3D numerical simulations were performed with the Computational Fluid Dynamics (CFD) code FLUENT™ using a Reynolds Stress turbulence closure for flow and the advection-diffusion method for concentration calculations. In the presence of trees, both measurements and simulations showed considerable larger pollutant concentrations near the leeward wall and slightly lower concentrations near the windward wall in comparison with the tree-less case. Tree stand density and crown porosity were found to be of minor importance in affecting pollutant concentration. On the other hand, the analysis indicated that W/H is a more crucial parameter. The larger the value of W/H the smaller is the effect of trees on pedestrian level concentration regardless of tree morphology and arrangement. A preliminary analysis of approaching flow velocities showed that at low wind speed the effect of trees on concentrations is worst than at higher speed. The investigations carried out in this work allowed us to set up an appropriate CFD modelling methodology for the study of the aerodynamic effects of tree planting in street canyons. The results obtained can be used by city planners for the design of tree planting in the urban environment with regard to air quality issues.     

Distributions of respiratory contaminants from a patient with different postures and exhaling modes in a single-bed inpatient room

This study investigated contaminant transport and evaluated the ventilation performance in a single-bed inpatient room. The study performed comparative experimental analysis on the distributions of respiratory contaminants breathed out and coughed out by a patient in a full-scale chamber, which simulated a single-bed inpatient room. The contaminant exhaled by the patient was simulated by an SF₆ tracer gas and 3-μm particles at steady-state conditions. The differences in the contaminant distribution between the coughing and breathing cases were insignificant for the mixing ventilation case, while for the displacement ventilation, the contaminant concentrations in the upper part of the room were higher for the coughing case. The contaminant concentrations in the inpatient room for the case with the patient sitting on the bed were lower than those for the patient supine on the bed for the displacement ventilation under the same supply airflow rate. The SF₆ tracer gas and 3-μm particles released at a notable initial velocity for simulating a cough could give similar contaminant distributions in the inpatient room. Therefore, the experimental data can be used to validate a CFD model, and the validated CFD model can be used to investigate transient coughing and breathing processes.     

Mixing of a Point-Source Indoor Pollutant by Forced Convection

The mixing rate of pollutants emitted from indoor sources influences the effectiveness of pollutant removal by building ventilation and the potential variability of exposure for a given release scenario. Quantitative information is scant on the mixing rate and the factors that govern it. We present mixing data for an instantaneously released tracer gas, carbon monoxide, in a sealed, unoccupied room under a range of forced airflow conditions, in which the flow is induced by blowers. The resulting mixing times, from 2 to 42 minutes, are related to the mechanical power of the air jets produced by the blowers. Mixing times are found to correlate well with the inverse of the cube root of power, in accordance with theoretical predictions and experimental observations for mixing in chemical reactors. The exposure index, defined as the time-averaged concentration at a point relative to the time-averaged concentration for the room as a whole, is presented for three experimental conditions, yielding quantitative information on the appropriateness of the well-mixed hypothesis under various flow conditions. In general, the exposure period following instantaneous release of a point-source pollutant must be much greater than the mixing time for the assumption of uniform mixing to hold. The correlation between mixing time and power input is used to predict the mixing time from the mixing action of a supply air jet for a typical ventilation scenario. The predicted mixing time, τ_mix～7 min, is substantially lower than the time scale for removal by ventilation, τ_vent～48 min. Under these conditions, complete mixing of an instantaneous release, point-source pollutant would be approximately attained within the interior space well before the pollutant would be thoroughly removed by ventilation.     

A study on the relationship between biodegradability enhancement and oxidation of 1,4-dioxane using ozone and hydrogen peroxide

Advanced oxidation involving O₃/H₂O₂ was used to eliminate 1,4-dioxane and to enhance the biodegradability of dioxane-contaminated water. Oxidation experiments were carried out in a bubble column reactor operating in fed-batch. The rate of dioxane removal and enhancement in biodegradability was investigated at hydrogen peroxide to ozone ratios between 0 and 0.6 mol:mol and pH between 5 and 11. A theoretical model was also applied to predict the experimental data and to investigate the effects of dioxane concentration, pH, and H₂O₂ concentration. The model predictions fit the experimental data well and there was a linear correlation between dioxane oxidation and BOD enhancement. At low dioxane concentrations, the oxidation rate was first order and it gradually approached zero order with increasing dioxane concentration. Also, the biodegradability of the solution increased with pH up to about 9 and it stayed constant with further pH increase. Hydrogen peroxide initially enhanced dioxane removal and biodegradability enhancement of the solution. However, at H₂O₂:O₃ ratios greater than about 0.4-0.45 mol:mol, i.e. about 2.90 mM for H₂O₂ concentration, H₂O₂ had negative impacts and resulted in reduced dioxane removal and biodegradability increase.     

Assessment of a model development for window glass breakage due to fire exposure in a field model

This paper presents a model development for the prediction of window glass breakage and fallout in a field model. Glass breakage is based on the temperature difference and the allowable glass breaking stress; and glass fallout is determined by a preset number of successive breakages. As a validation, generally good agreements are obtained between the numerical predictions and the data from a compartment fire experiment. The predicted glass surface temperature and the adjacent gas temperature are within 10–25% of the test data. For fire sizes of 170, 280 and 390 kW, the time of initial occurrence of glass breakage are shown within reasonable range of the experimental results. For the 680 kW fire case, the model shows an earlier glass fallout time, however, the predicted glass temperature at fallout is around 450 °C and is consistent with previous experiments. Further research to improve the model is discussed such as on radiation modeling and the criteria of glass breakage and fallout.     

Impact of land use on urban mobility patterns, emissions and air quality in a Portuguese medium-sized city

The main objective of this work was to evaluate the impact of urban development trends in mobility patterns of a medium sized Portuguese city and air quality consequences, using a sequential modeling process, comprising i) land use and transportation, TRANUS model; ii) road traffic air pollutants emissions, TREM model and; iii) air quality, TAPM model. This integrated methodology was applied to a medium sized Portuguese city. In order to evaluate the implementation of the methodology, a preliminary study was performed, which consisted on the comparison of modeled mobility patterns and CO and PM₁₀ concentrations with measured data used in the definition of the current scenario. The comparison between modeled and monitored mobility patterns at the morning peak hour for a weekday showed an RMSE of 31%. Regarding CO concentrations, an underestimation of the modeled results was observed. Nevertheless, the modeled PM₁₀ concentrations were consistent with the monitored data. Overall, the results showed a reasonable consistency of the modeled data, which allowed the use of the integrated modeling system for the study scenarios.The future scenarios consisted on the definition of different mobility patterns and vehicle technology characteristics, according to two main developing trends: (1) “car pooling” scenario, which imposes a mean occupancy rate of 3 passengers by vehicle and (2) the “Euro 6” scenario, which establishes that all vehicles accomplish at least the Euro 6 standard technology. Reductions of 54% and 83% for CO, 44% and 95% for PM₁₀, 44% and 87% for VOC and 44% and 79% for NO_x emissions were observed in scenarios 1 and 2, respectively. Concerning air quality, a reduction of about 100 μg m⁻³ of CO annual average concentration was observed in both scenarios. The results of PM₁₀ annual concentrations showed a reduction of 1.35 μg m⁻³ and 2.7 μg m⁻³ for scenarios 1 and 2 respectively.     

Thermal comfort and indoor air quality in the lecture room with 4-way cassette air-conditioner and mixing ventilation system

We performed the experimental and the numerical studies on thermal comfort (TC) and indoor air quality (IAQ) in the lecture room with cooling loads when the operating conditions are changed. Predicted mean vote (PMV) value and CO₂ concentration of the lecture room were measured and compared to the numerical results. Both of them showed a reasonable agreement with each other and then we applied the numerical model to analyze TC and IAQ for a couple of different operating conditions. From the results we found that the increment of the discharge angle of 4-way cassette air-conditioner makes uniformity of TC worse, but rarely affects IAQ. It turned out that TC and IAQ are hardly affected by the variation of the discharge airflow. Finally TC was merely affected by the increment of the ventilation rate, but when the ventilation rate is more than 800 m³/h, the average CO₂ concentration can be satisfied with the standard limits of Japanese in our case studies.     

Traffic-related air pollution and cardiovascular mortality in central Taiwan

In this study, the relationship between cardiovascular mortality and traffic-related air pollutants (NO₂, CO, PM₁₀, and six volatile organic compounds (VOCs), propane, iso-butane, propylene, benzene, meta-, para-, and ortho-xylenes) was investigated. The concentrations of NO₂, PM₁₀ and CO from 1993 to 2006 were measured at a fixed-site air monitoring station, and VOC data from 2003 to 2006 were obtained from a photochemical assessment monitoring site in an urban area in central Taiwan. Outcome variables were data on mortality due to cardiovascular diseases (ICD-9-CM 410-411, 414, 430-437) from 1993 to 2006. Cardiovascular mortality averaged 1.5 cases, ranging between 0 and 9 cases per day. Daily air pollution levels ranged from 0.5 to 80.5 ppb for NO₂ and from 0.1 to 3.8 ppm for CO. From the subset of data from 2003 to 2006, daily average values ranged from 0.6 to 17.5 ppb for propane, 0.3 to 6.7 ppb for iso-butane, 0.3 to 6.7 ppb for propylene, 0.2 to 3.8 ppb for benzene, 0.3 to 26.0 ppb for m,p-xylene, and 0.02 to 7.6 ppb for o-xylene. Poisson generalized additive model was used to estimate the effects of elevated air pollutant levels on daily mortality, adjusting for meteorological conditions and temporal trends. Single-pollutant model showed that cardiovascular mortality was significantly associated with NO₂ lagged 2 days, and with propane, iso-butane, and benzene lagged 0 day. The relative risk for an interquartile range increase in air pollutant levels was 1.053 for NO₂, 1.064 for propane, 1.055 for iso-butane, and 1.055 for benzene. In conclusion, daily cardiovascular mortality showed association with data on acute exposure to traffic air pollutants in Taichung, which is an important factor to consider in studying cardiovascular mortality in urban environments.     

Ventilation measurement using spot sampling of sulphur hexafluoride on a solid adsorbent

A technique for measuring the ventilation rate using spot sampling of sulphur hexafluoride (SF₆) on a solid adsorbent for subsequent thermal desorption and gas chromatographic analysis has been developed. This paper describes the mathematical model used, the analytical details of the technique, as well as results from a validation test using an experimental chamber. The mathematical model assumes an exponential decay of the tracer gas concentration. The precision of the analytical procedure was estimated to be better than 9% whilst the error of the measured ventilation rate of the test chamber was 5%.     

Application of field model and two-zone model to flashover fires in a full-scale multi-room single level building

This paper presents a comparison of the results from a computational fluid dynamics (CFD) model and a two-zone model against a comprehensive set of data obtained from one flashover fire experiment. The experimental results were obtained from a full-scale prototype apartment building under flashover conditions. Three polyurethane mattresses were used as fuel. The CFAST two-zone model (version 2.0) was also used to predict results for this flashover fire test. The mass release rate, gas temperature, radiation heat flux and gas compositions (O₂, CO₂ and CO) were measured. A CFD program, CESARE-CFD Fire Model, has been developed and was used also to predict results for polyurethane-slab fire. A simple flame spread model was incorporated into the CFD program to predict the mass release rate and heat release rate during the fire instead of providing it as an input as is required for most zone and CFD models. It was found that the CFD model provided reasonable predictions of the magnitude and the trends for the temperatures in the burn room and the species concentrations, but over-predicted the temperatures in the adjacent enclosures. From a life safety perspective, the CFD model conservatively predicted the concentrations of CO and CO₂. The predicted temperatures from the CFAST fire model agreed well with the experimental results in most areas. However, the CFAST model under predicted the temperature in the lower layer of the room of fire origin and the concentration of CO in most areas.     

Gas-particle partitioning and seasonal variation of polycyclic aromatic hydrocarbons in the atmosphere of Zonguldak, Turkey

Atmospheric concentrations and gas-particle partition coefficients were determined for polycyclic aromatic hydrocarbons (PAHs) in the atmosphere of Zonguldak, Turkey between May 2007 and April 2008. Total concentrations of PAHs ranged from 0.52 ng m^− 3 to 636 ng m^− 3 in the particle phase and from 5.60 ng m^− 3 to 725 ng m^− 3 in the gas phase. The annual mean concentrations of PAHs in the particle and gas phase were found to be 114 ng m^− 3 and 184 ng m^− 3, respectively. Significant seasonal variations of particle and gas phase PAH concentrations were observed with higher levels during cold period. The distribution of PAHs between the particle and gas phase was investigated and it was found that three ring PAHs were associated primarily with the gas phase, four ring PAHs were distributed almost equally between the two phases and five and six ring PAHs were mainly associated with the particle phase. Gas-particle partition coefficients (K_p) of PAHs have been calculated and correlated with their subcooled liquid vapor pressures (P_Lº). The slopes (m_r) varied from − 0.63 to − 0.23 were far from the theoretical value (−1) due to the short distance between the sampling point and the emission sources. The relationships between temperature and gas phase partial pressures of PAHs were examined using the Clausius-Clapeyron equation and the obtained positive slopes indicated that PAH concentrations increased with decreasing air temperature as a result of high dominance of local emissions.