The California Air Resource Board (CARB) recently reported that due to exposure to particle emissions, 9,000 people die annually in California alone [11]. A method for quantifying the exposure during a daily commute as well as reducing the exposure for the passengers has been developed. A fractional recirculation of cabin air was proposed and studied to improve cabin air quality by reducing cabin particle concentrations. Vehicle tests were run with differing number of passengers (1, 2, 3, and 4), four fan speed settings and at 15, 40, and 70 mph. A manual control was installed for the recirculation flap door so different ratios of fresh air to recirculated air could be used. Full recirculation is the most efficient setting in terms of thermal management and particle concentration reduction, but this causes elevated CO₂ levels in the cabin [5]. The study demonstrated cabin CO₂ concentrations could be controlled below a target level of 2000ppm at various driving conditions and fan speeds with more than 85% of recirculation. Additionally, some energy saving is also expected with the air conditioning system. More recirculation means less energy is required to cool the cabin air, as opposed to cooling 100% outside air under hot weather conditions. The proposed fractional air recirculation method is a simple yet innovative way of improving cabin air quality.

The exposure to particulate matter (PM) has strong adverse health effects, such as aggravated asthma, decrease in lung function, and increase respiratory symptoms. PM is complex and dynamic in its chemical composition and physical mixing state. The properties of PM have not been thoroughly characterized, and the mechanism of PM causing adverse effect has not been well understood. This study focuses on characterizing physical properties of urban background air particles by analyzing the particle spectra, particle effective density, black carbon, and active surface area along with chemical characterization using AMS. The effective density profiles were used to estimate the respiratory system-deposited ambient particle mass according to lung deposition fraction curve reported by ICRP (1994).

The first study was to determine size-resolved effective densities of ambient aerosols in Riverside, CA for four 7-day periods during 2015 – 2016. The mass of size selected particles (50, 70, 101 and 152 nm) were measured to determine the effective density for particles. A catalytic stripper (CS) in alternating mode was used to remove volatile compounds on the aerosol before density measurements. Aerosol non-refractory composition measurement was conducted in June 2016 campaign to understand the effect of chemical composition on particle density. The average particle densities for all measurement campaigns with CS bypassing mode (BP) were 1.17 g/cm3 at 50 nm and 1.25 – 1.28 g/cm3 at 70, 101 and 152 nm. The average density after CS conditioning (CS mode) showed a decreasing trend from 1.22 g/cm3 to 1.04 g/cm3 withincreasing selected size, and a mass fractal dimension (Df) of 2.85. Both the BP and CS mode particles showed the lowest effective density from 6:00 AM to 9:00 AM and highest density from 11:00 AM to 3:00 PM. The diurnal variation of density became more profound as particle size increases for both sampling modes. The variation was found to be more intense for the CS mode compared than the BP mode. To confirm this observation, organic aerosol and ammonium nitrate mass in the similar size range were tested simultaneously with BP mode. The results showed effective density measurements correlated well with positively (R2=0.78) and negatively (R2=0.62). This study provided an update to the aerosol density profiles of a well-known receptor site (Riverside, CA) and investigated the transformation of particles in different seasons. The obtained effective density profiles were used in the proceeding study in order to effectively estimate the respiratory-deposited ambient aerosol mass.

In the second study, the size distributions and effective densities of ambient particles measured in September 2015 and June 2016 were used to calculate time-dependent lung deposited particulate matter (PM) mass using lung deposition fraction curve reported by ICRP (1994). The particle active surface area, black carbon (BC) mass, particle number (PN), solid particle number (SPN), and suspended PM mass (from particle size distributions and effective densities) were also obtained to investigate correlations with PM dose. Non-refractory organic and inorganic mass measured by an aerosol mass spectrometer (mAMS) provided additional information in relation to dose versus chemical nature of particles.

Ambient particle size distributions showed strong diurnal variations during the sampling period. While the lung deposited PM mass fraction (0.32-0.36) did not vary much with time, lung deposited and suspended PM mass in the ambient air showed similar trend. Deposition in the alveolar region was the highest (0.3-1.0 μg/m3), which was followed by the nasal region (0.16 – 0.83 μg/m3) and tracheobronchial region (0.05 – 0.27 μg/m3). All three regions peaked at 12:00 PM – 6:00 PM and reached the lowest value near midnight.

One of the goals of this study was to consider the most prominent ambient PM measurement method in consideration to health and lung deposition. Suspended PM mass had the highest correlation with respiratory doses, which was followed by organics measured using mAMS, active surface area, and BC mass. It is also found that particle hygroscopicity did not affect correlations between metrics and lung deposited PM mass. By taking into account the accessibility and cost of these measurements, we propose the particle active surface area and BC mass to be considered when evaluating/monitoring the health effects caused by PM.

As diesel PM regulation gets more stringent, the current gravimetric method which has been used for legal determination of PM mass for vehicle emissions will have difficulty accurately quantifying PM mass emissions. Progress in regulating diesel PM emissions by non-gravimetric means has been made in the Europe. The method, so called particle measurement program (PMP) protocol specifies counting solid particles larger than 23 nm. This method has already been adopted for Euro V and VI to regulate light-duty and heavy-duty vehicle emissions and it is also expected to be adopted to regulate emissions for aviation section. However, exclusion of sub-23 nm particles poses some potential issues. In this thesis, the PMP method was investigated with a focus on finding the nature of sub-23 nm particles which are excluded in the current PMP protocol.

A PMP-compliant system, an AVL advanced particle counter (APC) and an alternative volatile particle removal system, a catalytic stripper (CS) were evaluated and compared for measuring solid particle number emissions in chapters 2 and 3. The evaluations and comparisons were conducted using sulfuric acid and hydrocarbon particles as model volatile particles in laboratory tests, and diluted exhaust from a diesel particle filter (DPF)-equipped heavy-duty diesel vehicle operated on a heavy-duty chassis dynamometer under steady speed conditions at two different engine loads. This

study also compared particulate matter (PM) mass and particle number (PN) emissions from a heavy-duty diesel vehicle operating over the urban dynamometer driving schedule (UDDS) and actual on-road flow-of-traffic driving conditions, including two uphill (phases 1 and 2) and two downhill (phases 3 and 4) segments. The UDDS and on-road flow-of-traffic tests represent a broader engine operating conditions than those currently certified emissions testing cycles.

For the laboratory test, both the APC and CS removed more than 99% of the volatile particles in terms of PN when using aerosols composed of pure sulfuric acid or hydrocarbons. When using laboratory test aerosols consisting of mixtures of sulfuric acid

and hydrocarbons more than 99% of the particles were removed by the APC but the surviving particles were no longer entirely volatile, with 12-14% being solid.

For the chassis dynamometer test, PN emissions of particles between 3 and 10 nm downstream of the APC were 2 and 7 times higher than the PN emissions of particles above 10 nm at the 74 and 26% engine load, respectively. At the 26% engine load, PN

level of the 3 to 10 nm particles downstream of the APC were significantly higher than that in the dilution tunnel, demonstrating that the APC was making 3 to 10 nm particles. The PN emissions of 3 to 10 nm particles downstream of the APC were related to the

heating temperature of the APC evaporation tube, suggesting these particles are artifacts formed by renucleation of semivolatiles. Considerably fewer particles between 3 to 10 nm were seen downstream of the CS for both engine loads due mainly to removal of semivolatile material by the catalytic substrates, although some of this difference could be attributed to diffusion and thermophoretic losses.

Chapter 4 provides an evaluation of the nature of sub 23 nm particles downstream of the European PMP methodology with prescribed cycles and on-road flow-of-traffic driving conditions. Particle number concentrations and size distributions were measured using two PMP measurement systems in parallel. For this analysis, the focus is on the real-time results from multiple instruments. The results revealed that a significant fraction of particles downstream of both PMP systems for all tested cycles were

below 11 nm. The fraction of sub 11 nm particles observed downstream of the PMP system decreased when the overall dilution ratio of one PMP system was increased from 300 to 1500, suggesting those sub 11 nm particles were formed through re-nucleation of semivolatile precursors. When the evaporation tube temperature was increased from 300 to 500°C, no difference in particle number concentrations was observed, suggesting incomplete evaporation of semivolatile particles did not contribute to those sub 11 nm particles. Particle emissions were about one order of magnitude higher during flow-of-traffic driving along a highway with a steep grade than during the prescribed driving cycles. During the same flow-of-traffic condition, a sudden jump of PMP operationally defined solid particle concentration was observed, while the accumulation mode particle concentrations in the constant volume sampling (CVS) tunnel measured by engine exhaust particle sizer (EEPS) only showed a slight increase. This discrepancy was attributed to the extensive growth of the re-nucleated particles downstream of the PMP systems.

In chapter 5, the PMP system was evaluated over a standard laboratory testing cycle and uphill and downhill on-road, flow-of-traffic driving conditions. The PM mass emissions for the UDDS and on-road tests were more than 6 times lower than the U.S. 2007 heavy-duty PM mass standard. The PM mass emissions for the UDDS fell between those for the uphill and downhill driving on-road driving conditions. The PN emissions of particles larger than 23 nm for the UDDS and downhill on-road driving conditions were 3 times lower than the Euro VI heavy-duty PN limit for transient cycles, while the PN emissions from the uphill on-road driving conditions were 4 to 5 times higher than the Euro VI PN limit. The PN emissions of particles larger than 23 nm for the UDDS (with an average engine load of 38%) were comparable in magnitude to those for the phase 3 downhill segment (with an aveage engine load of 40%) of the on-road test, and were 25% lower than those for the phase 4 downhill segment (with an avearage engine load of 18%) of the on-road test. The variability of the PN emissions of particles larger than 23 nm

ranged from 10 to 40% for the UDDS and on-road tests, comparable to that found in the European PMP study.

The Environmental Protection Agency and European Commission are requiring in-situ on-board diagnostic systems to increase accuracy in monitoring nitrous oxides and particulate matter vehicle emissions. Particulate Matter (PM) sensors are a derivative of this regulation and have been implemented to assess diesel particulate filter failure, as well as solid particle number. The design of PM sensors is varied by company (e.g., Honeywell, Bosch, Emisense, and Delphi) in addition to the applied method. The method of investigation in this study is electrical charging and the capacity for corona discharge generation.

Boltzmann’s Law for charged particles is generally used to predict electrical charging (i.e., particle charging). However, the PM sensor design and method in this study generates a signal that is amplified higher than can be theoretically explained by this law. Signal amplification can only be attributed to ionization due to the presence of corona discharge, besides particle charging in an electric field. A fundamental study probing into the effects of corona discharge on PM sensor signal amplification is necessary to understand the underlying operational principles of the sensor. This study will provide qualitative insights into the electrical properties of PM and the conditions in which electrical properties are influenced by varying physical conditions. The investigation is performed by correlating sensor response to varying physical conditions of PM. The results show that the sensor response is linearly dependent on particle concentration and that signal strength is limited on the sample flow rate. Understanding the operational principles of this design could lead to the development of a relatively inexpensive PM sensor with unprecedented high signal sensitivity.

Measurement of particulate matter (PM) emission of light-duty vehicles (LDVs) is a complex issue with many stakeholders, including engine manufacturers, health effect scientists, climatologists and regulatory agencies. With particulate emission regulatory limits in California and United States decreased by more than two orders of magnitude since the 80s, efforts continue to investigate alternatives to the gravimetric method for quantification of LDVs PM emission.

Several alternative metrics including total particle number (TPN), solid particle number (SPN), black carbon (BC), and particle surface (PS) area were measured along with gravimetric PM mass to study the correlation between different methods, using a variety of LDVs chosen to represent different emission levels and a range of current technologies in the modern fleet. It was found that gravimetric PM mass is strongly dependent on chemical nature of the PM. Thus correlations of gravimetrically determined PM mass with alternative metrics were different between two different testing cycles (e.g. FTP vs SFPT-US06). Alternative metrics are free from such cycle dependency which is due to adsorption artifact and therefore considered as a good future metric or supplemental metric such as SPN for EU regulation. Particle surface (PS) area turned out to be very sensitive with wide range and current instrumentation has a room to improve in terms of measuring tunnel blank and reducing electrometer drift.

Exposure to particulate matter (PM) and pollutant gas (NOx) is associated with increased cardiopulmonary morbidity and mortality. Mobile source emissions contribute to PM and NOx emissions significantly in urban areas. Hybrid Electric Vehicles (HEVs) plays an important role to reduce emissions. While sales of electric vehicles has been increasing, electric vehicles have to overcome issues with charging time, driving range, and production/sales cost for more widespread market penetration. Hybrid electric vehicles have a potential to serve as a bridge technology between current internal combustion engine powered vehicles and zero emissions vehicles such as electric vehicles and fuel cell vehicles. Current regulations require emissions to be tested on a chassis dynamometer. However, it is known that on-road emissions can be quite different from that those measured on regulatory driving cycles in the lab. In this study, emissions from two HEVs with different combustion technologies (gasoline direction injection vs port fuel injection) were compared using PEMS (portable emissions measurement system) and tailpipe sensors under cold weather conditions. The study has found the frequency and duration of re-ignition events vary depending on the type of HEV. Prius (PFI HEV) showed more frequent re-ignition events compared to Sonata (GDI HEV) for both city and highway driving conditions. Prius re-ignited almost every one minute while Sonata re-ignited every two minutes on average during the city driving condition. Reignition events affected emissions profile significantly during the city driving condition. As a result, Prius showed higher NOx emissions during the city driving condition while Sonata showed higher NOx emissions during the cold-cold start and highway driving condition. For PM emissions, PFI technology is known to make minimal amount of soot which is shown in the cold-cold start result while GDI technology is more prone to generating soot. This gap is reduced in city and highway driving condition due to more frequent re-ignition events of the PFI HEV (referring to Prius). Future studies should include more vehicles to understand whether the re-ignition events are vehicle specific or technology specific.

View the NCST Project Webpage

Biomass burning is a dynamic combustion process during which large concentrations of particles and trace gases are released into the atmosphere. Trace gases emitted by fires include significant amounts of carbon monoxide, nitrogen oxides, volatile organic compounds, and long-lived greenhouse gases such as carbon dioxide and methane. This work surveys 18 different vegetation types of southeast (SE) and southwest (SW) U.S. commonly present in prescribed burns. The work provides key factors for prediction of Particulate Matter (PM) and gas phase emissions, and serves as a basis for atmospheric models used for land management. It was found that EC/TC ratio is the best surrogate for prediction of hydrocarbon emissions, performing better than the commonly used Modified Combustion Efficiency (MCE). The emissions factors were observed to decrease exponentially with elemental carbon to total carbon ratio (EC/TC ratio) used as an indirect measure of fire intensity. No regional/vegetation type dependency was observed for hydrocarbon emissions; the emissions tend to change with MCE and fire intensity. PM emission factors were found to correlate with EC/TC ratio or MCE. Metals, anion, and cation emissions were affected most directly by fuel composition.

We study the PM2.5 and ultrafine exhaust emissions from a new natural gas-fired turbine power facility to better understand air pollution in California. To characterize the emissions from new natural gas turbines, a series of tests were performed on a GE LMS100 gas turbine. These tests included PM2.5 and wet chemical tests for SO2/SO3 and NH3, as well as ultrafine (less than 100 nm in diameter) particulate matter measurements. The turbine exhaust had an average particle number concentration that was 2.3x103 times higher than ambient air. The majority of these particles were nanoparticles; at the 100nm size, stack particle concentrations were about 20 times higher than ambient, and increased to 3.9x104 times higher on average in the 2.5 - 3 nm particle size range. This study also found that ammonia emissions were higher than expected, but in compliance with permit conditions. This was possibly due to an ammonia imbalance entering the catalyst, some flue gas bypassing the catalyst, or not enough catalyst volume. SO3 accounted for an average of 23% of the total sulfur oxides emissions measured. Some of the SO3 is formed in the combustion process, it is likely that the majority formed as the SO2 in the combustion products passed across the oxidizing CO catalyst and SCR catalyst. The 100MW turbine sampled in this study emitted particle loadings similar to those previously measured from turbines in the SCAQMD area, however, the turbine exhaust contained far more particles than ambient air.

The power consumed by an air conditioner accounts for a significant fraction of the total power used by hybrid and electric vehicles especially during summer. This study examined the effect of recirculation of cabin air on power consumption of mobile air conditioners both in-lab and on-road. Real time power consumption and vehicle mileage were recorded by an On Board Diagnostic monitor and carbon balance method. Vehicle mileage improved with increased cabin air recirculation. The recirculation of cabin air also significantly reduced in-cabin particle concentrations. Recirculation of cabin air is an excellent and immediate solution to increase vehicle mileage and improve cabin air quality.