Grantee Research Project Results
2011 Progress Report: Quantifying the Effects of the Mixing Process in Fabricated Dilution Systems on Particulate Emission Measurements via an Integrated Experimental and Modeling Approach
EPA Grant Number: R834561Title: Quantifying the Effects of the Mixing Process in Fabricated Dilution Systems on Particulate Emission Measurements via an Integrated Experimental and Modeling Approach
Investigators: Zhang, Ke Max
Institution: Cornell University
EPA Project Officer: Chung, Serena
Project Period: May 1, 2010 through April 30, 2013 (Extended to October 30, 2013)
Project Period Covered by this Report: May 1, 2011 through April 30,2012
Project Amount: $250,000
RFA: Novel Approaches to Improving Air Pollution Emissions Information (2009) RFA Text | Recipients Lists
Research Category: Air Quality and Air Toxics , Air
Objective:
The main objective of this study is to investigate a key uncertainty in PM emissions measurement by examining the following questions: How do the mixing processes in the current constant volume sampler (CVS) systems differ from those in the real-world conditions? How do the mixing processes in the different CVS systems differ from each other? How does the mixing process interact with aerosol dynamics that affect PM measurements in the CVS systems and PM transformation in the atmosphere?
Progress Summary:
The main objective of this project is to elucidate how turbulent mixing in fabricated dilution systems affects PM emission measurement. The overall progress of this project is excellent. During our first year, we developed a new turbulent reacting flow model, named the Comprehensive Turbulent Aerosol Dynamics and Gas Chemistry (CTAG), and evaluated CTAG against several field measurements on evolution of particle size distributions in dilution exhaust plumes. For the second year of this project, we completed the analysis of the turbulent mixing and aerosol dynamics inside two dilution tunnels. One of the most important findings is that we identified the dilution rate as a unifying parameter to compare different dilution tunnels. More detailed descriptions from this analysis are presented in the following sections. We have published one journal paper (Wang, et al., 2012), and another manuscript is under review. In addition, we have shared out results with U.S. EPA’s Office of Transportation and Air Quality and the California Air Resources Board (CARB), who are considering the long-term development of methods for measuring particle number emissions from vehicles. Our modeling activities have been following EPA's Guidance for Quality Assurance Project Plans (EPA QA/G-5M).
We employed the Comprehensive Turbulent Aerosol Dynamics and Gas Chemistry (CTAG) model to investigate the effects of those parameters on a set of particulate emission measurements comparing two dilution tunnels, i.e., a T-mixing lab dilution tunnel and a portable field dilution tunnel with a type of coaxial mixing. The turbulent flow fields and aerosol dynamics of particles were simulated inside two dilution tunnels. Particle size distributions under various dilution conditions predicted by CTAG were evaluated against the experimental data. It is found that in the area adjacent to the injection of exhaust, turbulence plays a crucial role in mixing the exhaust with the dilution air, and the strength of nucleation dominates the level of particle number concentrations. Further downstream, nucleation terminates and the growth of particles by condensation and coagulation continues. Sensitivity studies reveal that a potential unifying parameter for gas dynamics, i.e., the dilution rate of exhaust, plays an important role in new particle formation. The T-mixing lab tunnel tends to favor the nucleation due to a larger dilution rate of the exhaust than the coaxial mixing field tunnel. Our study indicates that numerical simulation tools can be potentially utilized to develop strategies to reduce the uncertainties associated with dilution samplings of emission sources.
The CTAG model was applied to simulate the inter-comparison experiments of two dilution tunnel designs by Lipsky and Robinson (2005). The objective is to identify the key parameters affecting the difference in PM measurement results from the two tunnels and quantify the impacts of those parameters. The two dilution tunnels are referred to as the Lab Tunnel and Field Tunnel in Figure 1. We studied two dilution ratios (DR), referred to as DR20 and DR120. For example, DR20 means that the dilution ratio will be 20 at the end of the dilution process.
Figure 1. The geometries of the lab Tunnel and the field Tunnel
employed in Lipsky and Robinson (2005)
In terms of mixing types, the lab tunnel employs a T-mixing configuration, while the field tunnel utilizes coaxial mixing. In addition, a fan-shaped mixing enhancer is installed in the field tunnel, but not in the lab tunnel. The two dilution tunnels also vary in size with the lab tunnel (2.3 m) longer than the field tunnel (0.9 m).
We first analyzed the turbulence flow fields inside the two dilution tunnels by applying the Large Eddy Simulation (LES) model. Figure 2 illustrates the contours of DR inside two tunnels. It can be seen that due to the effect of strong turbulence near the exhaust pipe, exhaust is mixing rapidly with the dilution air, illustrated by the fast transition in colors. The spatial inhomogeneity disappears under turbulent and molecular mixing. For both tunnels, the well-mixed status is achieved before the diluted exhaust reaches the end of the dilution tunnels. The diluted exhaust moves relatively parallel toward the downstream in the lab tunnel, while for the field tunnel, the diluted exhaust rotates forward due to the effect of the mixing enhancer.
Figure 2. Time-averaged distributions of DR in the cross-section through the exhaust pipe within
the two dilution tunnels. Blue represents the exhaust and red represents the dilution air, colors
betwen blue and red represents the degree of dilution.
the two dilution tunnels. Blue represents the exhaust and red represents the dilution air, colors
betwen blue and red represents the degree of dilution.
Then we applied CTAG to simulate the experimental conditions with results shown in Figure 3. The goal was to validate CTAG against measured particle size distributions (PSD). The comparisons between the measured and simulated PSDs are shown in Figure 4 for both dilution tunnels under DR20 and DR120 conditions. A large nucleation mode is observed for the lab tunnel under DR20 due to the strong nucleation, while it is rarely observed for the other three cases. The weighted deviation of PSD for the lab tunnel under DR20 between simulation and measurement is 12.6%, while the weighted deviation for the field tunnel under DR20 is 8.5%. The differences between simulation and experiment for both tunnels under DR120 are less than 3% because only the accumulation mode exists.
Figure 3. Dilution-corrected PSDs at the end of two dilution tunnels under (a) DR20 (b) DR120.
Once CTAG was validated against the measurement results, we conducted sensitivity analysis to investigate the impacts of individual parameters on PM emission measurements. Those parameters include relative humidity of dilution air, mixing enhancer, and resident time. Given one of the objectives of our project is to develop a practical guideline on how to quantify how turbulent mixing affects PM emission measurement, it is important to reduce the number of parameters necessary to include in the guideline.
Based on the results of the sensitivity analysis, we proposed the dilution rate of the exhaust as a parameter unifying the effects of mixing types, mixing enhancer, dilution ratio and residence time. Dilution rate is represented by the scalar dissipation rate of exhaust:
Where k and ε are the turbulent kinetic energy and turbulent dissipation rate, respectively. 
is the mixture variance and Cφ is a constant.
is the mixture variance and Cφ is a constant.The contours of dilution rate for various cases are shown in Figure 4. It can be seen that the level of dilution rate varies significantly with the mixing type, the mixing enhancer, and RT. T-mixing in lab tunnel has the advantage of allowing a higher degree of penetration into the cross-stream than the coaxial mixing in the field tunnel (Case 1 vs. Case 6). The corresponding maximum dilution rate in the lab tunnel (83.2 s-1) is higher than that in the field tunnel (31.9 s-1). The mixing enhancer also tends to increase the dilution rate, especially in areas directly behind the mixing enhancer (Case 1 vs. Case 7), which prompts the formation of UFPs. Furthermore, RT can affect the level of dilution rate significantly. For the lab tunnel, the maximum dilution rate increases 92.4% under the half RT and decreases 56.1% with the double RT. A similar trend also can be found for the field tunnel. In general, the formation of new particles increases with the increasing dilution rate of exhaust.
Figure 4. TIme-averaged dilution rate of exhaust inside the dilution tunnels.
With mechanistic understandings of the coupled turbulence and aerosol dynamic processes, we can design and operate the dilution systems to achieve the different goals, e.g., retaining tailpipe-level emissions or capturing the effects of plume processing on exhaust particles. This is important for the ongoing effort to define a standardized dilution sampling methodology for characterizing emissions from stationary combustion sources.
Future Activities:
We will simulate the evolutions of diesel exhaust plumes inside a full-scale wind tunnel, during on-road chasing, and in CVS emission measurement systems. The objective is to study how the mixing process in the CVS compares to that in the atmosphere, and how the difference affects PM emission measurement. In addition, we will model an ejector dilutor, which has been used in Portable Emission Measurement Systems (PEMS).
Journal Articles on this Report : 2 Displayed | Download in RIS Format
| Other project views: | All 5 publications | 2 publications in selected types | All 2 journal articles |
|---|
| Type | Citation | ||
|---|---|---|---|
|
|
Wang YJ, Zhang KM. Coupled turbulence and aerosol dynamics modeling of vehicle exhaust plumes using the CTAG model. Atmospheric Environment 2012;59:284-293. |
R834561 (2011) R834561 (2012) R834561 (Final) |
Exit Exit |
|
|
Wang YJ, Yang B, Lipsky EM, Robinson AL, Zhang KM. Analyses of turbulent flow fields and aerosol dynamics of diesel engine exhaust inside two dilution sampling tunnels using the CTAG model. Environmental Science & Technology 2013;47(2):889-898. |
R834561 (2011) R834561 (2012) R834561 (Final) R834554 (Final) |
Exit Exit Exit |
Supplemental Keywords:
PM2.5, combustion, on-road, non-road, stationary sources, ultrafine, public health, human exposure, climate changeRelevant Websites:
Energy and the Environment Exit
Progress and Final Reports:
Original AbstractThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.