There is currently a widespread concern about the actual nitrogen oxides (NOx) and particulate matter (PM) emissions of gasoline direct injection (GDI) vehicles, which have dynamically penetrated into the US passenger vehicles market and are expected to eventually replace the less efficient port fuel injection (PFI) vehicles. Although GDI engines are knows to significantly improve fuel economy and lower greenhouse gas emissions (GHG) when compared to PFI engines, they produce higher PM emissions due to the direct spray of gasoline into the combustion chamber. This leads to locally rich, diffusion-governed liquid fuel combustion that creates more PM formation. Gasoline particulate filters (GPFs) are introduced as an effective strategy to reduce the tailpipe PM mass and the number of ultrafine particles under a range of driving conditions and at the same time meet California’s PM mass emission standards.

The difference between conditions of the type approval dyno test cycles defined by the vehicle emission regulations and the real driving can contribute to the differences between expected and actual pollution levels, average cycles are designed more than 20 years ago which cannot give full picture and necessarily accurate estimate of typical power demands on the vehicle’s engine. Furthermore, it is disclosed that some verified the engines and vehicles are meeting current stricter regulations on FTP75 and US06 dyno tests but emitted more PM and NOx pollutions out of limits, Volkswagen (VW) Scandal is a typical lesson for demand of real world driving emission test. This has led to the introduction of in-use vehicle emission monitoring and regulations by means of portable emissions measurement systems (PEMS). With recent developments in the US and the European Union (EU), PEMS is becoming an important regulatory device. Both the Environmental Protection Agency (EPA) and the California Air Resource Board (CARB) have begun testing the ability of PEMS to accurately measure real driving emissions.

The objective of this study is to examine the PM mass, PM soot, particle number (PN), and gaseous emissions of a current technology GDI vehicle during on-road testing with and without a catalyzed GPF. Testing was performed on three routes with different topological characteristics, representing urban, rural, highway and mountain driving conditions. The vehicle was tested in triplicated in downtown Los Angeles, Mt. Baldy, and in San Diego. The results of this work will be discussed in the context of the impact of GPF and driving patterns on tailpipe PM, PN, gaseous emissions and fuel economy.

The results showed statistic significant filtration rate up to 99% for PMsoot emissions by employing catalyst GPF instead of traditional TWC catalyst, relatively great reduction for total PM, PN and a trend of effective reduction for gaseous emissions were observed by analyzing the data, no measureable impact was found for fuel economy. With all these beneficial performances we can summarize the conclusion that GPF can be a reliable strategy for GDI vehicle emission control.

Anthropogenic air pollution consists of primary and secondary pollutants resulted from human activities. It is related to different environmental and health issues such as climate change, visibility and respiratory diseases.

The maritime transport is an important source of anthropogenic primary air pollution. Natural gas (NG) vessels have become more widely used due to the more stringent emission regulations; however, emission data from NG maritime operations is still limited. This thesis conducted a comprehensive analysis on the air quality, health effects and climate change impacts of switching from diesel to NG. Results showed that PM2.5, NOx, CO2 were reduced by about 93%, 92% and 18%, respectively. However, HCHO and CH4 increased several-fold. A health risk assessment showed the diesel plume increased long-term health risk and the NG plume increased short-term health risk. A global warming potential (GWP) analysis was performed and revealed that the average NG exhaust GWP was increased by 38%. Mitigation strategies for further reducing pollutants from NG exhaust are discussed and showed potential for reducing short-term health and climate impacts.

Anthropogenic secondary organic aerosol (SOA) is formed from the oxidation of volatile organic compounds (VOCs) such as aromatics and is critically impacted by NOx. Global transport models use the SOA parameters from the two major chemical pathways, RO2+NO and RO2+HO2 and the branching ratio of RO2+NO pathway (β), to predict SOA formation. This thesis attempted to improve the model prediction by experimentally investigating the SOA formation from those pathways with a novel approach of maintaining β constant throughout chamber experiments. At low-NOx conditions, multiple SOA yield curves were observed. The yield increased with HO2/RO¬2, indicating the contribution of RO2+RO2 pathway. The GEOS-Chem model showed that for the regions with high aromatic emissions but lower HO2/RO¬2, aromatic SOA was overestimated by up to 100%. At high-NOx conditions, SOA parameters were developed when controlling β at one during the chamber experiments to simulate the RO2+NO pathway without significant contribution from the other pathways. The global surface β was modelled using GEOS-Chem and four β scenarios were observed. SOA formation was investigated when simulating the daytime β profiles for those scenarios.

Air pollution and global climate change are important environmental issues that affect our society. Deeper understanding of atmospheric chemistry is required to understand these problems and to develop effective control strategies. Environmental chambers have been used for the past few decades to study atmospheric chemistry and investigate processes leading to secondary pollutant formation. This thesis work provides two different high sensitivity real time cavity enhance absorption spectroscopy (CEAS) instruments to detect NO3 radical and NO2/ glyoxal for environmental chamber study. These CEAS instruments can monitor these species with high sensitivity (in sub-ppb concentrations), with good selectivity, and with high spatial resolution in real time and in situ.

Environmental chamber studies using CEAS instruments include NO3 behavior and reaction kinetics between NO3 and amine; sources and sinks for oxides of nitrogen (NOx) in the atmospheric photooxidation reactions; glyoxal uptake by secondary organic aerosol (SOA) under dry and humid conditions and glyoxal formation from the photooxidation of aromatic and isoprene. These studies provide insightful data in atmospheric processes and lead to a better understanding of the atmospheric system, which will eventually extend to environmental policy.

This thesis is only a part of my research work. For more information about my other work, including cavity ring down spectroscopy studies for peroxy radical, aerosol optical extinction, transparent thin and time of flight mass spectrometry studies for the initial steps of ozone and alkenes reaactions, please refer to my PhD thesis in Chemistry department, UC-Riverside.

Indoor environmental chambers have been widely used to simulate the Secondary Organic Aerosol (SOA) generation processes in the atmosphere. Experimental data is used to optimize atmospheric models, contribute to scientific predictions, and policy making. However, the measured data, such as the quantified aerosol yield, can be biased due to the interactions between the gas-particle system and chamber walls. Minimizing uncertainties of particle- and vapor-wall deposition corrections thus become critical. This dissertation crystalizes the journey to review and explore the particle-wall interactions and to make responsible corrections in multi-generations of environmental chambers at UCR.Size-dependency of particle wall-loss varies between Teflon environmental chamber designs. The measured particle wall-loss rates in UCR previous-generation dual 90-m3 collapsible chambers were relatively insensitive to particle size with the rates exceeding those reported for some smaller volume chambers. Coagulation-free monodisperse seed injection experiments were performed and the particle wall-loss behaviors were experimentally observed to be dominated by electrostatic effect via charged chamber walls. Particles in the chamber were found to approach chamber-specific charge equilibrium, driven by size-dependent preferential loss of charged particles and bidirectional diffusion charging. The traditional particle wall-loss correction model has been modified into a new system to investigate the significance of particle-particle coagulation and dynamic change of particle wall-loss rates within experiments for previous-generation collapsible chambers. The modified model re-visited particle wall-loss corrections for thousands of historical runs over the past two decades, creating a massive data pool with increased correction accuracy. Dynamic change of particle wall-loss rate was observed within individual experiments. Final particle mass can be over-corrected due to coagulation, dependent on the particle number loadings present in the chamber. Mitigation strategies to minimize the electrostatic impact have been applied on UCR new-generation 120-m3 fixed-volume chamber. Size-dependent particle wall-loss patterns were characterized with monodisperse seed injection experiments with negligible coagulation. The observed trend was found to be more sensitive to particle size and much lower than previous-generation chambers. A three-component particle loss correction model has been developed to account for chamber dilution, particle coagulation and wall-loss. Performance evaluation shows good agreement between the new correction method with the modified traditional method.

Recent advances in air quality sensor technology have allowed for governments, academia, and communities to use sensors to measure air pollution at unprecedented spatial and temporal scales. While use of low-cost sensors in ambient air monitoring applications has dramatically increased in the past several years, there remain unanswered questions and challenges with designing, implementing, and deploying low-cost sensors. These challenges include quantifying the performance of sensors, developing defensible methods for deploying sensor networks, and communicating sensor data to the public in an understandable and meaningful way. This body of work addresses portions of these challenges by systematically evaluating the performance of sensors, deploying sensors with scientifically defensible methodology for a specific application, and developing methodologies and tools for disseminating and communicating community air monitoring data as information to the public. The main objective of this research is to provide clarity and vision on the appropriate use and applications of low-cost air quality sensors for ambient air monitoring. Systematically evaluating the performance of commercially available low-cost sensors with respect to regulatory grade instrumentation and understanding the measurement error associated with sensors is critically important when choosing a sensor for a specific monitoring application (i.e. fence-line monitoring, community monitoring, hot spot identification, mobile monitoring) or drawing conclusions from data collected by low-cost sensors. The first part of this research investigates the performance of 12 commercially available low-cost particulate matter (PM) sensors against regulatory-grade instrumentation. The next phase of this research includes the design of an ozone sensor node for a specific ambient air monitoring application - parallel monitoring to select a relocation site for a regulatory ambient air monitoring station. The sensor selection, sensor node development, and network deployment methodologies were designed to collect defensible data for determining a relocation site. In the last phase of this research, data tools are developed to store, process, analyze, and visualize large data sets generated by low-cost sensors to provide communities with information from their community monitoring networks.

Airborne particles are one of the most important and certainly the most visible aspects of air pollution. The effects span the areas of human health, visibility degradation, radiation and photochemistry and climate changes. Aerosols are small particles that suspended in the atmosphere which can be in the solid or liquid phase. Cooking is an element source of volatile organic compounds (VOCs) and a potential source of secondary organic aerosol (SOA) both indoors and outdoors. In this study, SOA emissions from oxidation reactions between canola oil, peanut oil and corn oil with O3 and OH radicals were characterized using a flow tube reactor and a potential aerosol mass(PAM) chamber.

SOA formed from emissions of atomizing cooking oils through reactions with excess ozone and OH radicals in a flow tube reactor or PAM chamber as well as primary organic aerosol (POA) emissions were characterized using a scanning mobility particle sizer (SMPS), a high-resolution time-of-flight aerosol mass spectrometer (HR-TOF-AMS) and an Aerosol Particle Mass Analyzer(APM).

The results show evidence for secondary organic aerosols(SOA) formation and explain the observed mass and composition changes, as well as the surface reaction, occurred during the oxidation reactions.

Heavy duty on-road vehicles represent one of the largest sources of NOx emissions and fuel consumption in North America. Heavy duty vehicles are predominantly fueled with diesel, with the recent interest in natural gas (NG) systems. As emissions and greenhouse gas regulations continue to tighten new opportunities for advanced fleet specific heavy duty vehicles are becoming available with improved fuel economy. NOx emissions have dropped 90% for heavy duty vehicles with the recent 2010 certification limit. Additional NOx reductions of another 90% are desired for the South Coast Air basin to meet its 2023 NOx inventory requirements and the California optional low NOx standard in 2015.

One of the difficulties in quantifying NOx emissions at the levels proposed in this research (90% of the 2010 certification level ~ 0.02 g/bhp-hr) is the measurement methods are approaching their detection limit to sufficiently quantify NOx emissions. Three upgraded NOx measurement methods were considered which include a raw NOx measurement integrated with real time exhaust flow, a real-time ambient correction approach, and a trace level ambient analyzer for accurate bag analysis. In summary the improved methods varied in their success where the raw sampling approach showed to be the most accurate and precise over the rage of conditions tested.

The ISL G NZ 8.9 liter NG engine met and exceeded the target NOx emissions of 0.02 g/bhp-hr. This engine significantly reduced 97%-100% of NOx emissions compared with previous ISL G 8.9 engines. The NOx emissions decreased as the duty cycle was decreased which was the opposite trend for the diesel vehicles. It is expected NG vehicles could play a role in the reduction of the south coast NOx inventory problem given their near zero emission factors demonstrated.

The overall goal of this research was an in depth investigation of emissions from a number of sources, with emphasis on two under-explored areas. One study delved into the characterization of emissions from wildland fires an increasingly important source due to climate change, and the other undertook a deeper examination of the in-use emissions from heavy duty trucks operating in the port regions of Southern California.

An in-depth characterization of emissions from lab scale wildland fires was carried out, including: polyaromatic hydrocarbons (PAHs), levoglucosan, anions and cations, elemental and organic carbon. The unique aspect of this research was the comparison of data from two independent measurement methods; one method used real-time analytical methods and the other collected a sample over time and was analyzed off-line with reference methods. A side-by-side comparison of the two approaches showed significant differences and inaccuracies for many of the analyses by the Aerosol Mass Spectrometer. Further analysis enabled transformation of real time analysis to useful data.

Usually marine ports have the highest emissions in a region as many of the sources have weak emission standards. In this research the emissions of criteria and toxic pollutants were measured from the engines in heavy duty trucks representative of those operating in the port regions. An important finding was the emissions measured during real world drive cycles more accurately represented the emission characteristics for the vehicles than the certification cycle or values determined from in-use compliance testing. Further, emissions from NOx control technologies were important to understand as emissions were multiple times higher for near port activities as compared with regional activity. High emissions reduction benefits from the SCR were observed but largely depended on catalyst temperature. Another important analysis completed during this thesis was the emissions factors from yard tractors in the port region for a period of over ten years. Both case studies provide additional insight into the emissions from port sources.

Emissions from marine diesel engines are mainly uncontrolled and affect regional air quality and health of people living near ports. Many emission control strategies are evolving to reduce these emissions and their impacts. This dissertation characterizes the effectiveness of new technologies for reducing NO_x and PM_2.5 emissions from a range of marine diesel engines. Researchers, regulators and policy makers require these characterizations to develop emission inventories and suitable mitigation strategies.

Three NO_x control technologies were analyzed: injection timing retard, water in fuel emulsion, and selective catalytic reduction (SCR). Each significantly reduced NO_x emissions. The SCR, however, increased PM_2.5 emissions by 150-380% indicating a need for technology modification before implementation. Additionally, two fuel control strategies for PM_2.5 based on cleaner burning fuels were evaluated: the effects of switching from high-sulfur heavy fuel oil to lower-sulfur marine distillate oil and switching from diesel to biodiesel blends were tested. Results showed significant PM_2.5 reductions with minimal change in NO_x; however, the biodiesel fuel increased formation of nucleation mode particles.

In-use emission benefits of a diesel-electric hybrid tug were characterized. Activity data showed that the average load factors of tug boat engines were up to 83% lower than that specified in the certification cycles typically used for developing emission inventories. Reductions of 73% for PM_2.5, 51% for NO_x and 27% for CO₂ were seen in comparison to a similar conventional tug. The majority of these reductions were attributed to the hybrid tug's energy management system, which directs use of auxiliary power for propulsion. Additional in-use harbor-craft measurements showed significant ocean current effects with a three to six fold increase gaseous and PM_2.5 emissions.

Overall this research showed that 1) new control technologies should be evaluated in the pilot stage to ensure that they do not increase emissions, 2) use of certification cycle load factors can significantly overestimate emissions from marine applications, and 3) actual in-use measurements are needed for accurate localized inventories. Finally, a new activity and emissions based protocol was developed to establish emission benefits of a multi-powered diesel-electric hybrid system.