Projects

Temperature and Pollution Management in Urban Canyons

Large Eddy Simulation (LES) of airflow over a 2D urban surface using OpenFOAM. The bottom of the street is heated while other surfaces are adiabatic. Solution of normalized temperature is animated.

The influence of roof-edge roughness elements on airflow, heat transfer, and street-level pollutant transport inside and above a two-dimensional urban canyon is analyzed using an urban energy balance model coupled to a large-eddy simulation model. Simulations are performed for cold (early morning) and hot (mid afternoon) periods during the hottest month of the year (August) for the climate of Abu Dhabi, United Arab Emirates. The analysis suggests that early in the morning, and when the tallest roughness elements are implemented, the temperature above the street level increases on average by 0.5 K, while the pollutant concentration decreases by 2 % of the street-level concentration. For the same conditions in mid afternoon, the temperature decreases conservatively by 1 K, while the pollutant concentration increases by 7 % of the street-level concentration. As a passive or active architectural solution, the roof roughness element shows promise for improving thermal comfort and air quality in the canyon for specific times, but this should be further verified experimentally. The results also warrant a closer look at the effects of mid-range roughness elements in the urban morphology on atmospheric dynamics so as to improve parametrizations in mesoscale modelling.

Numerical domain for the urban-canyon CFD simulations; note that the passive scalar release occurs in all canyons.

Turbulence in the Arctic Lower Troposphere

Slant profiling with the Polar 6 aircraft; a conceptual monotonic ascent is shown with the height interval and leg length identified; the aspect ratio of the height interval vs. leg length is not to scale.

NETCARE 2014 research flight tracks and radiosonde launching platforms from Resolute Bay (Radiosondes S: 1-10) and the CCGS Amundsen ice-breaker (Radiosondes S: 11-14).

Aircraft measurements were used to characterize properties of clear-air turbulence in the lower Arctic troposphere. For typical vertical resolutions in general circulation models, there was evidence for both downgradient and countergradient vertical turbulent transport of momentum and heat in the mostly statically stable conditions within both the boundary layer and the free troposphere. Countergradient transport was enhanced in the free troposphere compared to the boundary layer. Three parametrizations were suggested to formulate the turbulent heat flux and were evaluated using the observations. The parametrization that accounted for the anisotropic nature of turbulence and buoyancy flux predicted both observed downgradient and countergradient transport of heat more accurately than those that did not. The inverse turbulent Prandtl number was found to only weakly decrease with increasing gradient Richardson number in a statistically significant way, but with large scatter in the data. The suggested parametrizations can potentially improve the performance of regional and global atmospheric models.

Ship Plume Intercept Using Aircraft in the Arctic

Amundsen Polar 6

A snapshot of the German Polar 6 aircraft while sampling the emitted plume of the Canadian Coast Guard Icebreaker Amundsen in Lancaster Sound near Resolute Bay, Nunavut, Canada, during the 2014 NETCARE campaign (Photo credit: Maurice Levasseur) - Data analysis and research leadership by AAA-Scientists

Decreasing sea-ice and increasing marine navigability in northern latitudes have changed Arctic ship traffic patterns in recent years and are predicted to increase annual ship traffic in the Arctic in the future. Development of effective regulations to manage environmental impacts of shipping requires an understanding of ship emissions and atmospheric processing in the Arctic environment. As part of the summer 2014 NETCARE (Network on Climate and Aerosols) campaign, the plume dispersion and gas and particle emission factors of effluents originating from the Canadian Coast Guard icebreaker Amundsen operating near Resolute Bay, NU, Canada, were investigated. The Amundsen burnt distillate fuel with 1.5 wt % sulfur. Emissions were studied via plume intercepts using the Polar 6 aircraft measurements, an analytical plume dispersion model, and using the FLEXPART-WRF Lagrangian particle dispersion model. The first plume intercept by the research aircraft was carried out on 19 July 2014 during the operation of the Amundsen in the open water. The second and third plume intercepts were carried out on 20 and 21 July 2014 when the Amundsen had reached the ice edge and operated under icebreaking conditions. Typical of Arctic marine navigation, the engine load was low compared to cruising conditions for all of the plume intercepts. The measured species included mixing ratios of CO2 , NOx , CO, SO2 , particle number concentration (CN), refractory Black Carbon (rBC), and Cloud Condensation Nuclei (CCN). The results were compared to similar experimental studies in mid latitudes.

Plume expansion rates were calculated using the analytical model and found to be 0.75+/-0.81, 0.93+/-0.37, and 1.19+/-0.39 for plumes 1, 2, and 3, respectively. These rates were smaller than prior studies conducted at mid latitudes, likely due to polar boundary layer dynamics, including reduced turbulent mixing compared to mid latitudes. All emission factors were in agreement with prior observations at low engine loads in mid latitudes. Icebreaking increased the NOx emission factor from EFNOx=43.1+/-15.2 to 71.6+/-9.68 and 71.4+/-4.14 g kg-diesel-1 for plumes 1, 2, and 3, likely due to change in combustion temperatures. The CO emission factor was EFCO=137+/-120, 12.5+/-3.70 and 8.13+/-1.34 g kg-diesel-1 for plumes 1, 2, and 3. The rBC emission factor was EFrBC=0.202+/-0.052 and 0.202+/-0.125 g kg-diesel-1 for plumes 1 and 2. The CN emission factor was reduced while icebreaking from EFCN=2.41+/-0.47 to 0.45+/-0.082 and 0.507+/- 0.037 1e16 kg-diesel-1 for plumes 1, 2, and 3. At 0.6 % supersaturation, the CCN emission factor was comparable to observations in mid latitudes at low engine loads with EFCCN=3.03+/-0.933, 1.39+/-0.319, and 0.650+/-0.136 1e14 kg-diesel-1 for plumes 1, 2, and 3.

Boundary Layer Mixing Height under Statically Stable Conditions

Domain of the Canadian Regional Weather Forecast Model (GEM15).

The atmospheric boundary layer mixing height (MH) is an important bulk parameter in air quality (AQ) modelling. Formulating this parameter under statically stable conditions, such as in the Arctic, has historically been difficult. In an effort to improve AQ modelling capacity in North America, MH is studied in two geographically distinct areas: the Arctic (Barrow, Alaska) and the southern Great Plains (Lamont, Oklahoma). Observational data from the Atmospheric Radiation Measurement program, Climate Research Facility and numerical weather forecasting data from Environment Canada's Regional Global Environmental Multiscale (GEM15) model have been used in order to examine the suitability of available parameterizations for MH under statically stable conditions and also to compare the level of agreement between observed and modelled MH. The analysis period is 1 October 2011 to 1 October 2012. The observations alone suggest that profile methods are preferred over surface methods in defining MH under statically stable conditions. Surface methods exhibit poorer comparison statistics with observations than profile methods. In addition, the fitted constants for surface methods are site-dependent, precluding their applicability for modelling under general conditions. The comparison of observations and GEM15 MH suggests that although the agreement is acceptable in Lamont, the default model surface method contributes to a consistent overprediction of MH in Barrow in all seasons. An alternative profile method for MH is suggested based on the bulk Richardson number. This method is shown to reduce the model bias in Barrow by a factor of two without affecting model performance in Lamont.

Classification of atmospheric stability conditions (ThetaV: virtual potential temperature, WS: Wind Speed).

Air Quality Monitoring in the Arctic

In an effort to characterize the effect of shipping on Arctic air quality during the 2013 shipping season, shipping movement has been visualized using real GPS coordinates from June 1st to November 1st.

The Canadian Arctic has experienced decreasing sea ice extent and increasing shipping activity in recent decades. While there are economic incentives to develop resources in the north, there are environmental concerns that increasing marine traffic will contribute to declining air quality in northern communities. In an effort to characterize the relative impact of shipping on air quality in the north, two monitoring stations have been installed in Cape Dorset and Resolute, Nunavut, and have been operational since 1 June 2013. The impact of shipping and other sources of emissions on NOx, O3, SO2, BC, and PM2.5 pollution have been characterized for the 2013 shipping season from 1 June to 1 November. In addition, a high-resolution Air Quality Health Index (AQHI) for both sites was computed. Shipping consistently increased O3 mixing ratio and PM2.5 concentration. The 90 % confidence interval for mean difference in O3 mixing ratio between ship- and no ship-influenced air masses were up to 4.6-4.7 ppb and 2.5-2.7 ppb for Cape Dorset and Resolute, respectively. The same intervals for PM2.5 concentrations were up to 1.8-1.9 ug m-3 and 0.5-0.6 ug m-3. Ship-influenced air masses consistently exhibited an increase of 0.1 to 0.3 in the high-resolution AQHI compared to no ship-influenced air masses. Trajectory cluster analysis in combination with ship traffic tracking provided an estimated range for percent ship contribution to NOx , O3 , SO2, and PM2.5 that were 12.9-17.5 %, 16.2-18.1 %, 16.9-18.3 %, and 19.5-31.7 % for Cape Dorset and 1.0-7.2 %, 2.9-4.8 %, 5.5-10.0 %, and 6.5-7.2 % for Resolute during the 2013 shipping season. Additional measurements in Resolute suggested that percent ship contribution to black carbon was 4.3-9.8 % and that black carbon constituted 1.3-9.7 % of total PM2.5 mass in ship plumes. Continued air quality monitoring in the above sites for future shipping seasons will improve the statistics in our analysis and characterize repeating seasonal patterns in air quality due to shipping, local pollution, and long-range transport.

Air Quality Health Index (AQHI) consistently increases associated with ship-influenced air masses, showing that shipping pollution can be identified, albeit at very low levels with current levels of traffic.

Temperature Sensor Design for Space Exploration

Sensors to measure the atmospheric temperature on Mars at three elevations - sensors are installed on a light graphite mast that deploys in the vertical direction

Liquid water is an important indicator for the possibility of life on other planets. Mars, also known as the red planet, shows no trace of liquid water on its surface, but it seems that water ice exists below ground level in the Martian arctic. Earlier discoveries (e.g. Mars Odyssey Orbiter in 2002) show large amounts of subsurface water ice in the northern arctic plain. The Phoenix lander (2009) targeted this circumpolar region using a robotic arm to dig through the protective top soil layer to the water ice below and ultimately, to bring both soil and water ice to the lander platform for sophisticated scientific analysis.

The buffet of scientific instruments on Phoenix spacecraft were suited to uncover clues to the geologic history and biological potential of the Martian arctic. Phoenix was the first mission to return data from either polar region providing an important contribution to the overall Mars science strategy "Follow the Water" in NASA's long-term Mars Exploration Program.

The very cold temperatures of the Martian arctic are measured with thin wire thermocouples, a technology that has been used successfully on meteorological stations for both the Viking and Pathfinder missions. A Thermocouple is an electric circuit made of two dissimilar metals (Chromel and Constantan in the case of this sensor) attached in two junctions. The junctions, when exposed to two different temperatures, produce a small voltage, which is proportional to the temperature difference. Phoenix thermocouple sensors are designed by AAA Scientists. Three of these sensors are located on a 1.2 meter vertical mast to provide a profile of how the temperature changes with height near the surface.

Thermal Test Chamber Design

Computational domain for CFD analysis - air space within a freeze tube for the Ice Nucleation project

Computational Fluid Dynamics (CFD) has been used to study thermal instabilities within the freeze tube in the Ice Nucleation project. The freeze tube is made from brass that is cooled down to -50C. The temperature profile in the brass is usually linear, with the coldest point at the very bottom. For optical access, two windows are provisioned on the side. In order to prevent frosting on the windows, it is necessary to purge them with dry nitrogen. However, doing so causes the windows to be slightly warmer than the brass body. This initiates thermal plumes and turbulent instabilities within the freeze tube. Such instabilities are not desired since they disturb the falling droplet streak at the central axis of the tube.

Computational Fluid Dynamics (CFD) is used to model such instabilities using Large Eddy Simulation (LES) turbulence model and the Boussinesq approximation. Results for two temperature gradients are presented here: 10C and 2C difference between the window and the brass body.

Temperature profiles are shown in the figure below. The thermal plumes associated with warmer windows (left) are more noticeable. These plumes interact near the vertical axis of the freeze tube and are responsible for disturbing particle streak lines. This situation must be avoided by purging the windows with minimum amount of nitrogen so the thermal gradient between the windows and the brass body is minimized.

Temperature profiles for gradient of 10C (left) and 2C (right) between windows and the brass body - Units [K]

Velocity magnitude profiles are shown in the figure below. Again there are significant velocities associated with the warmer windows (left) near the central axis of the freeze tube.

Velocity magnitude profiles for gradient of 10C (left) and 2C (right) between windows and the brass body - Units [m/s]

Vorticity is the best indicator for understanding turbulence. Vorticity describes the local spinning motion of a fluid near some point and is defined as curl of velocity. Figure below shows the vorticity evolution within the freeze tube. Boundaries and corners within the domain contribute to `seeding' voriticities in the flow that subsequently grow, advect, and interact with one another. Results here suggest that there will be turbulent instabilities in the central axis of the freeze tube due to large temperature gradients between the window and the brass body (left).

Vorticity magnitude profiles for gradient of 10C (left) and 2C (right) between windows and the brass body - Units [1/s]

Optical Instrument Design Based on Polarization

Clouds containing ice are important constituents of the Earth's climate system

Ice nucleation is an important aspect of climate change. Ice clouds play an important role in the Earth's thermal balance with space. They also affect precipitation patterns. Contrary to the common misconception, pure and airborne water droplets at thermal and dynamic equilibrium and atmospheric pressure do not freeze at 0C but -37C. It is believed that most ice nucleation in the atmosphere is not homogeneous but caused by heterogeneous freezing with dust, spores, bacteria and other agent acting as nucleation sites.

Ice nucleation chamber designed and built to measure water droplet freezing temperature as a function of various nucleant agents such as dust, spores, and bacteria

An apparatus is developed to detect the phase of falling water droplets (liquid or ice) as a function of temperature. The apparatus consists of five major sub-assemblies: a freeze tube, an optical assembly, a control unit, a fridge, and a purge system. The freeze tube and optical assembly are shown in the figure above. The freeze tube generates droplets of known diameter and frequency on top. As droplets fall, the optical assembly enables capturing images of the backscattered signal from the LASER that shines on to droplets. The optical assembly traverses in three directions, making it easy to image a large subset of interior space in the freeze tube, while preserving the alignment of the optics.

Figure below shows the principle of operation. A LASER beam is linearly polarized using a Glan-LASER polarizer and directed to a droplet. The back scattered signal from the droplet surface is then directed to two cameras using a beam splitter. The beam arriving at camera 1 is filtered by a Polaroid polarizer. Liquid droplets preserve linear polarization of light, so with correct setting of the Polaroid polarizer with respect to the Glan-LASER polarizer, no light intensity is transferred to camera 1. Ice, on the other hand, depolarizes light so some intensity appears on the filtered camera using the same polarizers' setting. This is the distinguishing behaviour for detecting ice versus liquid water. Camera 2 images the total intensity back scattered by the droplet.

Schematic for optical design of an instrument that detects ice using light depolarization

Figure below shows a sample set of images for liquid and ice particles. Notice that for liquid droplets the depolarization filtered image shows a very low intensity for a droplet streak. Ideally we should not see any droplet streaks in this setting, but there is slight depolarization by LASER transmission through the windows. For ice, however, the streak intensity for both polarization filtered and original images is high since there is depolarization due to surface roughness and birefringence of ice.

Total (left) and polarization filtered (right) back scattered signal from a liquid droplet

Total (left) and polarization filtered (right) back scattered signal from an ice droplet

Hybrid Ventilation Design

Computational domain for CFD analysis - A dining hall with displacement ventilation

Displacement ventilation (DV) has found a popular place in industrial ventilation design in the recent years. If designed and implemented correctly, it reduces building energy demand and provides better thermal comfort and air quality. The performance of DV is well understood and characterized for small rooms, with simple geometries, and with favorable boundary conditions. Complexities arise, however, when one tries to design and implement DV for larger and more complex industrial spaces under both cooling and heating modes, partial ventilation loads, with auxiliary heating and cooling devices, solar gain, thermal plumes due to occupants and heat generating appliances, and possibly with internal momentum sources such as ceiling fans, movement of people, and expiratory injections.

This research makes an effort to analyze performance of a hybrid ventilation system in the ventilation design process of a dining hall (above). The ventilation performance is studied under various scenarios such as heating versus cooling modes and full versus partial ventilation loads. Two additional ventilation strategies are investigated to alter air distribution within space. These are operating ceiling fans and a low level bypass exhaust. The performance indices are temperature and concentration profiles within the entire space.

The hall consists of a lower floor, a mezzanine, and a top hat ceiling. The occupants sit at the lower floor and around the mezzanine. The total volume of the air in the hall is 5378.49 cubic meter. The heat sources for the domain are as follows: occupants (maximum 1160 people), lighting (8 hung on top of mezzanine level), uninsulated walls and windows under cooling mode, and radiant panels under heating mode (2 at lower level ceiling and 1 at mezzanine level ceiling). The heat sinks for the domain are as follows: uninsulated walls and windows under heating mode, and radiant panels under cooling mode. The hall roof and lower floor are considered thermally insulated. Ten tables at lower floor and a kitchen door are specified as sources for odor (food).

CutCell structured hexahedral mesh for the dining hall

The mesh developed for this project uses the CutCell method, which is structured and hexahedral, but refined close to the boundaries. This type of mesh is advantageous over the unstructured tetrahedral mesh scheme since it provides a better convergence and a more accurate solution, particularly when species transport is of concern.

Under cooling mode, it is recommended to use the displacement ventilation strategy with no ceiling fan operation and no bypass exhaust. Under heating mode, it is recommended to use ceiling fans, pushing warm air downwards, which results in higher temperatures. It is also recommended to use a low level bypass exhaust under heating mode when the ceiling fans are operating. This results in lower odor concentrations at the occupied zone due to the fact that more air exits through bypass exhaust when the fans are operating (also shown below).

Temperature contours within dining hall - Heating mode, bypass exhaust closed, no ceiling fan operation (left) - Temperature contours within dining hall - Heating mode, bypass exhaust closed, ceiling fans pushing air downwards (right)

Species (CO2) contours within dining hall - Heating mode, bypass exhaust, no ceiling fan operation (left) - Species (CO2) contours within dining hall - Heating mode, bypass exhaust, ceiling fans pushing air downwards (right)

Turbulence Modelling

Computational domain for CFD analysis - The structured mesh is generated using ANSYS 14.0 CutCell method. The mesh contains close to 1 million control volumes. Steady and parallel flow passes over the box from left to right. A solution to the flow field is desired.

Turbulence influences many aspect of our lives from small to vast scales. Turbulent motion associated with the gust of the wind on the streets, plume of cigarette smoke, convection currents in the core of the earth that maintain the earth's magnetic field, and the solar prominence, are just only a few examples to note. Turbulence can be crudely defined as a range of fluid vortex blobs (eddies) that move and interact with each other and the flow in a chaotic manner.

Airflow within buildings exhibits the very same phenomenon: turbulence. Flow within ducts, air handlers, around objects like furniture and human bodies, can be at times complex. Especially, when the Reynolds and Grashof numbers are lower than most other industrial applications, the flow may be characterized by transitional turbulence, exhibiting slow and large scale periodicity. Conventional Reynolds Averaged Navier Stokes (RANS) models do not work well in such situations, so that other alternatives like the Large Eddy Simulation (LES) may be attractive, albeit the choice of the latter requires immense computational memory and processing power.

Consider parallel and steady flow around a box, typical of flow within rooms, at a low Reynolds number (Re~6300). A solution to the flow field passed the object can be computed on a fine structured grid with close to 1 million control volumes (shown above).

CFD simulation using LES turbulence model - Contours of vorticity - Karman vortex shedding behind box immersed in parallel and steady flow (Re~6300)

CFD simulation using k-e RANS turbulence model - Contours of vorticity - Karman vortex shedding behind box immersed in parallel and steady flow (Re~6300)

The LES model resolves turbulent flow in greater detail. RANS model does not converge, due to slow and large scale transient instabilities. In addition, the RANS model packs large scale vorticity fluctuations towards that of the mean flow.

Design of a Cost-effective Solar Thermal Power Plant

Artist impression of a conceptual solar thermal power plant in the Mojave desert

A design study was conducted to evaluate the cost-effectiveness of solar thermal power generation in a 50 kWe power plant that could be used in a remote location. The system combines a solar collector-thermal storage system utilizing a heat transfer fluid and a simple Rankine cycle power generator utilizing R123 refrigerant. Evacuated tube solar collectors heat mineral oil and supply it to a thermal storage tank. A mineral oil to refrigerant heat exchanger generates superheated refrigerant vapor, which drives a radial turbogenerator. Supplemental natural gas firing maintains a constant thermal storage temperature irregardless of solar conditions enabling the system to produce a constant 50 kWe output. A simulation was carried out to predict the performance of the system in the hottest summer day and the coldest winter day for southern California solar conditions. A rigorous economic analysis was conducted. The system offers advantages over advanced solar thermal power plants by implementing simple fixed evacuated tube collectors, which are less prone to damage in harsh desert environment. Also, backed up by fossil fuel power generation, it is possible to obtain continued operation even during low insolation sky conditions and at night, a feature that stand-alone PV systems do not offer.

Solar thermal power plant thermodynamic cycle

Using life cycle costing analysis, the cost of producing electricity from a plant of this design is estimated at $0.33 /kWh. Compared to advanced solar thermal and photovoltaic power plants, the proposed system offers cost savings and reliability. The proposed plant utilizes evacuated tube collectors that remain efficient even in slightly cloudy sky conditions. These collectors require less maintenance and operate more reliably than tracking concentrators in the harsh desert environment. The proposed system is base loaded by natural gas, so that it can provide power 24 hours a day, regardless of solar insolation availability.

Pipe Flow Simulations of Miscible Fluids of Different Density

The present numerical algorithm is implemented in C++ as an application of PELICANS open source software. In this simulation, the flow of higher density fluid is imposed from the left. Simultaneous fluid transport with different densities occur in many instances related to buildings and infrastructure (e.g. sewage).

CFD simulation: concentration development for mixing of two miscible fluids of different densities in a pipe

CFD simulation: velocity vector field development for mixing of two miscible fluids of different densities in a pipe

Stirling Engine Cogeneration System Performance

Whispergen Stirling engine based combined heat and power system for residential applications

There is renewed interest in Micro Cogeneration of Heat and Power (Micro CHP) in small residential scales. The increasing electricity prices motivate use of variable energy sources such as biomass, natural gas, gasoline and diesel to produce electricity in the first place and then recover any excess heat for thermal applications (hot water and space heating). Various technologies such as internal (Otto and Diesel cycles) and external (Stirling cycle) combustion engines and PEM fuel cell systems have been manufactured in Micro CHP units to achieve cogeneration of electricity and heat for small residences.

Our research shows that when high quality energy sources (natural gas and other petroleum products) are available, reciprocating internal combustion engines such as Otto and Diesel and PEM fuel cells provide the highest power efficiencies (~20%, ~25% and ~30% respectively). On the other hand, when low quality sources (biomass) of energy is available, only external combustion engines such as Stirling are feasible but can nevertheless produce electricity at acceptable power efficiencies (~10%). The overall energy efficiencies for these systems including power and heat is high (>~70%) and may justify their use over grid electricity if high marginal efficiencies can be achieved. For such technologies, however, maintenance, flexibility and reliability must be taken into account. Also, the payback time for some systems are high since they have high initial installation costs (e.g. PEM fuel cells).

Schematic for Whispergen Stirling engined based combined heat and power system installation

For the first time in industry we have tested performance of a Stirling engine based Micro CHP system operated on animal waste Rothsay biodiesel (figure above). Our results show that comparable power and heat efficiencies can be obtained using biodiesel versus No. 2 diesel. Both fuels result in a power efficiency of about 11% and a heat efficiency of about 75% for a total energy efficiency of about 86%. When run on biodiesel, the micro CHP unit requires more warm up time to reach steady state performance. Biodiesel produces less NOx, but diesel produces less Unburned Hydrocarbon (UHC) and particulate emissions. Other emissions including CO, CH2O, C2H4O, CH4, and C2H2 are comparable between the two fuels. Particulate matter emissions were characterized by filter deposition. Images below shows Scanning Electron Microscopy (SEM) of the filters.

SEM image for particulate emissions using No. 2 diesel

SEM image for particulate emissions using biodiesel

Optical Instrument Design Based on Laser Imaging

Shadowgraphy and Particle Tracking Velocimetry (PTV) set up

Significant sources of contagious air indoor are droplets produced by expiratory actions (coughing and sneezing). If they are small enough, these droplets remain airborne for extended periods of time and can pose infection risk. An artificial air-assist atomizer is built to produce fine (less than a micrometer) to coarse (larger than a millimeter) droplets using transient injections for the purpose of indoor air quality testing (figure above).

Droplet breakup and dispersion behavior is studied with this atomizer. The shadowgraphy technique has been used to measure the size of droplets. This technique requires backlighting droplets so that a high contrast image results when droplets impede the reception of LASER light by the camera. This image can be analyzed by a computer program to properly size droplets that are in focus (figures below).

Schematic for shadowgraphy and PTV setup

Droplets produced by the artificial cough/sneeze atomizer

Droplet sizing in Spray by Shadowgraphy (units in millimeter)

Four major breakup mechanisms are observed (dumbbell, filament, stripping, and catastrophic). Droplet axial dispersion is characterized by considering various time and length scales present in the flow. At the onset of injection, small droplets succeed in overpassing larger droplets since they are faster to accelerate. However, at larger distances, larger droplets gain higher momentum and overtake smaller droplets. As far as radial turbulent dispersion is concerned, smaller droplets diffuse towards the periphery of the spray faster than larger droplets. This is true since their relaxation time is small enough to respond to turbulent eddies in the flow. On the other hand, larger droplets experience a weaker redial dispersion.

Numerical/Experimental Analysis of Building Ventilation Systems

CFD simulation of cough droplets dispersion in a hospital inpatient room with displacement ventilation

Ventilation design for healthcare facilities is a sensitive industry. On the one hand, designers intend to provide a safe environment with reduced risk of nosocomial infection. On the other hand, it is desired to reduce energy consumption and ecological footprint of healthcare facilities.

Most Hospitals in North America use conventional mixing type mechanical ventilation systems with excessive Air Exchange Rates (ACH). Although effective in overall dilution of contagious air, such ventilation strategies impose high energy costs. There is keen interest to use alternative ventilation strategies such as displacement, natural, or hybrid systems to achieve lower energy consumption and infection risk simultaneously.

Success of such ventilation design strategies depend on various building design and occupancy parameters. Locations of the diffuser, exhaust, objects, and occupants in the room, and also the thermal (wall temperatures) and velocity (air inlet and expiratory injections) boundary and initial conditions highly affect the success or failure of such ventilation systems.

Our research carefully considers these design parameters by simulations and experiments so that we can predict contagious agent dispersion and the associated infection risk while recommending optimum ventilation design strategies and room use by the occupants to achieve lower energy consumption and reduced infection risk at the same time.

The above picture shows a CFD simulation for the dispersion of droplets released by a horizontal patient cough in a displacement ventilated room. Generally horizontal injections, as opposed to vertical injections, increase mixing in the room and pose a higher risk of infection for the occupants in the room.

Ventilation experiments for parametric testing of a stratified air system under various boundary and initial conditions

Schematic for ventilation experiments for parametric testing of a stratified air system under various boundary and initial conditions

The above figures show the actual and schematic for an experiment that reveals sensitivity of a low energy stratified air system in mitigating infection risk to subtle changes in boundary and initial conditions. Thermal manikins are used in place of a patient and a subject. Droplets are released using an atomizer that simulates patient coughs and sneezes. Exposure of an occupant, either a nurse or a visitor, is measured by an aerosol collection pole (1) that delivers droplets to an Aerodynamic Particle Sizer (APS). This pole measures droplets at lower, breathing, and higher elevations. 32 thermocouples and 4 thermoelectric anemometers measure interior/boundary temperatures and air velocities on poles/walls.

Case 1 injects droplets with a moderate velocity at inclined angle. Cases 2 and 3 inject droplets at higher and lower velocities. Case 4 injects droplets to the side wall. Case 5 injects droplets horizontally, and case 6 injects droplets vertically. Case 7 reduces the thermal plume strength associated with the subject manikin. Cases 8 and 9 place the subject manikin at locations away from the injection. Case 10 increases the air change rate. In general it is observed that higher injection velocities, inclined injections, lower thermal plume strengths, and lower air change rates increase contaminant exposure, while lower injection velocities, injections towards the walls, placement of the subject away from the injection, and higher ventilation rates decrease contaminant exposure.