EFFECTS OF BIODIESEL CHEMICAL COMPOSITION ON THE CHEMICAL AND PHYSICAL PROPERTIES OF THE PRIMARY AND SECONDARY DIESEL PARTICULATE MATTER

Ali Mohammad Pourkhesalian

Submitted in fulfilment of the requirements for the degree of Doctor of Philosophy

School of Chemistry, Physics and Engineering Science and Engineering Faculty Queensland University of Technology 2015

To “Sahar” whose unconditional love opened my eyes to an entirely new world…

Keywords

Atmospheric Aging, Diesel Particulate Matter (DPM), Particle Number Concentration (PNC), Particle Size Distribution (PSD), Biodiesel, Fatty Acid Methyl Ester (FAME), Biodiesel Feedstocks, Fuel Chemical Composition, Fuel Physical Properties, Oxygen Content, Viscosity, Surface Tension, Carbon Chain Length, Unsaturation, Scanning Mobility Particle Sizer (SMPS), Differential Mobility Spectrometer (DMS), Volatilization Tandem Differential Mobility Analyzer (V-TDMA), Volatility, Count Median Diameter (CMD), DustTrak, , Reactive Oxygen Species (ROS), Volatility, Ultra Violet Light (UV), Ozone, Semi- volatile, Partitioning, Atmospheric Simulation.

Effects of Biodiesel Chemical Composition on the Chemical and Physical Properties of the Primary and Secondary Diesel Particulate Matter i

Abstract

In recent years, there has been an increasing movement from conventional fossil diesel fuels to non-petroleum based fuels. Biodiesel is gradually finding its place as a reliable alternative to petrodiesel. However, a number questions remain unanswered in terms of the suitability of such fuels. This thesis investigated fresh and aged particulate matter emissions due to the combustion of biodiesel. Moreover, this study investigated the influence of the above mentioned parameters on the potential toxicity of diesel particulate matter, by measuring its reactive oxygenated species content. In order to achieve the above aim, a number of biodiesel fuels with controlled chemical composition were tested in a modern diesel engine. The biodiesels tested in this manuscript were chosen in such a way so that it was possible to assign a change in the physio-chemical properties of the emitted diesel particulate matter to a specific chemical characteristic of the biodiesel used. The engine ran on neat biodiesel, as well as blends of biodiesel and petrodiesel, and the diesel exhaust was sampled and diluted using a multi-stage diluter system. Then, the diluted and cooled diesel exhaust was guided to a flow-through reactor, where the diesel exhaust was mixed with high concentrations of ozone and irradiated with ultra-violet light for a short time. A variety of physical and chemical properties of the particles inside the chamber were continuously monitored using different instruments and techniques. Experimental measurements showed that the diesel particulate matter emitted by different biodiesels was different in terms of both its physical properties and chemical composition, depending upon its fatty acid methyl ester composition. For example, particle matter mass concentration increased with an increase in the carbon chain length and degree of unsaturation of the fuel. Short-chain saturated biodiesels caused the lowest particulate matter emissions; as well particles due to such fuels smaller in terms of median diameter. It was shown that NOx emissions due to biodiesel could be reduced by controlling the composition of the fatty acid methyl esters in biodiesel. Particle emissions decreased linearly with the oxygen content of the blends regardless of the type and percentage of biodiesel in the blends. However,

iiEffects of Biodiesel Chemical Composition on the Chemical and Physical Properties of the Primary and Secondary Diesel Particulate Matter

since the fuel oxygen content weight ratio increased with decreasing carbon chain length, it was not clear which of these factors was responsible for the reduction in PM emissions. Further, particle emissions decreased linearly with reductions in the viscosity and surface tension of the fuel, but only for specific blends; nevertheless significantly different particulate matter emissions were observed for various blends with the same viscosity and surface tension.

It was found that the volatility and oxidative potential of particles depended upon the fatty acid methyl ester carbon chain length of the biodiesel fuel. Particles emitted from the combustion of short-chain-saturated biodiesel were found to have a higher fraction of volatile organic substances adsorbed onto them, leading to higher reactive oxygen species emissions. The particles overall volatility and reactive oxygenated species correlated exponentially with the oxygen content of the blend, which could either be due to a reduction in soot onto which the volatiles condense, or an actual increase in the amount of volatile materials. However, the exponential relationship observed between overall volatility and reactive oxygen species indicated a combination of both effects. These results further highlighted the importance of fatty acid methyl ester composition on the toxicity of particles emitted from biodiesel combustion. In addition, they also challenge the effectiveness of current mass and number based emission standards in addressing the health effects of diesel particulate matter, while reinforcing the importance of more inclusive emission standards.

Analyzing the results from the experimental campaigns showed that the chemical composition of the fuel not only affected primary emissions, but it also had an effect on secondary emissions. Saturated fuels, which emitted less particulate matter, also emitted more volatile particulate matter and significantly more semi- volatiles in the gas phase. The volatile and semi-volatile fractions had the potential to partition between the gas phase and particle phase upon aging. Aged particles from more saturated fuels with higher oxygen content had a higher oxidative potential as expressed through the increase of reactive oxygenated species concentration.

It is apparent from the findings of this study that the chemical characteristics of different biodiesel fuels may be more important than its physical properties, in terms of particulate matter emissions and particulate toxicity. They also suggest the introduction of new standards, and alteration of existing regulations to categorize

Effects of Biodiesel Chemical Composition on the Chemical and Physical Properties of the Primary and Secondary Diesel Particulate Matter iii

biodiesel fuels based on their fatty acid methyl ester profile. To be able to exploit the benefit of biodiesel and avoiding potential harms, it is also recommended to take into account the excessive amount of volatiles and semi-volatiles released by the combustion of biodiesels, both in the gas and particulate phases.

ivEffects of Biodiesel Chemical Composition on the Chemical and Physical Properties of the Primary and Secondary Diesel Particulate Matter

Table of Contents

Keywords ...... i Abstract ...... ii Table of Contents...... v List of Figures ...... viii List of Tables ...... xi List of Abbreviations ...... xii Acknowledgements...... xv CHAPTER 1: INTRODUCTION ...... 1 1.1 The investigated research problem ...... 1 1.2 Study aim ...... 4 1.3 Specific objectives of the study ...... 5 1.4 Evidence of research progress linking the research papers ...... 6 1.5 References ...... 10 CHAPTER 2: LITERATURE REVIEW ...... 15 2.1 IntroducTion ...... 15 2.2 Background and definitions ...... 15 2.2.1 Aerosol and particulate matter ...... 15 2.2.2 Nucleation mode ...... 16 2.2.3 Accumulation mode ...... 16 2.2.4 Aitken mode...... 17 2.2.5 Coarse mode ...... 17 2.3 Aerosol science and theory ...... 17 2.3.1 Formation and growth ...... 17 2.3.2 Primary particles ...... 18 2.3.3 Secondary particles ...... 18 2.3.4 Vapor pressure ...... 18 2.3.5 Nucleation...... 19 2.3.6 Condensation ...... 19 2.3.7 Coagulation ...... 20 2.3.8 Particle size, mass concentration and number concentration ...... 20 2.3.9 Particle equivalent diameter ...... 20 2.3.10 Mass concentration ...... 21 2.3.11 Particle number concentration...... 22 2.3.12 Deposition mechanisms ...... 22 2.3.13 Diffusion ...... 22 2.3.14 Gravitational settling ...... 22 2.3.15 Electrostatic attraction ...... 23 2.3.16 Thermophoresis...... 23 2.3.17 Impaction ...... 23 2.4 Sources of particles ...... 23 2.4.1 Biogenic...... 23 2.4.2 Anthropogenic ...... 24 2.5 Diesel exhaust ...... 24 2.5.1 Diesel particulate matter ...... 25 2.5.2 Chemical properties...... 26

Effects of Biodiesel Chemical Composition on the Chemical and Physical Properties of the Primary and Secondary Diesel Particulate Matter v

2.5.3 Physical properties ...... 27 2.5.4 Effect of the fuel...... 28 2.5.4.1 Petrodiesel ...... 28 2.5.4.2 Synthetic diesel ...... 29 2.5.4.3 Biodiesel ...... 29 2.5.5 Effects of physical properties of the fuel ...... 29 2.5.6 Effects of chemical properties of the fuel ...... 30 2.5.7 Transformation and fate ...... 32 2.6 Atmospheric dilution and chemistry ...... 33 2.6.1 VOC reactions ...... 33 2.6.2 Dilution ...... 34 2.6.3 UV 35 2.6.4 Ozone ...... 36 2.7 Semi-volatile compounds ...... 37 2.8 Secondary organic aerosols ...... 39 2.9 Diesel exhaust atmospheric aging and simulations ...... 40 2.9.1 PAM chambers ...... 40 2.9.2 Smog chambers ...... 47 2.10 Health effects ...... 59 2.11 Instruments and techniques...... 61 2.11.1 Offline methods...... 61 2.11.2 CPC 62 2.11.3 DMA ...... 63 2.11.4 SMPS...... 64 2.11.5 TDMA ...... 64 2.11.6 V-TDMA ...... 65 2.11.7 AMS ...... 65 2.11.8 PM mass concentration ...... 65 2.11.9 ROS measurement ...... 66 2.12 Gaps in knowledge ...... 66 2.13 References ...... 66 CHAPTER 3: PARTICLE EMISSIONS FROM BIODIESELS WITH DIFFERENT PHYSICAL PROPERTIES AND CHEMICAL COMPOSITION ...... 79 3.1 Introduction ...... 83 3.2 Materials and methods...... 85 3.2.1 Engine and fuel specification ...... 85 3.2.2 Exhaust sampling and measurement system ...... 88 3.3 Results and discussion ...... 90 Specific PM emissions ...... 90 3.3.1 Specific PN emissions ...... 91 3.3.2 Particle number size distribution ...... 93 3.3.3 Particle median size ...... 93 3.3.4 Specific NOx emissions ...... 95 3.3.5 Influence of fuel physical properties and chemical composition on particle emissions ...... 96 3.3.6 Comparison of engine performance and particle emissions among used biodiesels ...... 98 3.4 Conclusions ...... 99 3.5 Acknowledgement ...... 99 3.6 References ...... 99 3.7 Appendix ...... 103

viEffects of Biodiesel Chemical Composition on the Chemical and Physical Properties of the Primary and Secondary Diesel Particulate Matter

CHAPTER 4: THE INFLUENCE OF FUEL MOLECULAR STRUCTURE ON THE VOLATILITY AND OXIDATIVE POTENTIAL OF BIODIESEL PARTICULATE MATTER113 4.1 Introduction ...... 118 4.2 Experimental setup and methodology ...... 120 4.3 Fuel Selection ...... 121 4.4 Results and discussion ...... 122 4.4.1 Volatility measurements ...... 122 4.4.2 BC Emissions ...... 124 4.4.3 ROS measurements ...... 125 4.4.4 The role of oxygen content ...... 127 4.5 Acknowledgments...... 132 4.6 Supporting Information Available ...... 132 4.7 Supporting Information ...... 133 4.7.1 Volatile Volumetric Fraction (VVF) ...... 134 4.7.2 Overall volatility ...... 134 4.8 References ...... 137 CHAPTER 5: EFFECT OF ATMOSPHERIC AGEING ON VOLATILITY AND ROS OF BIODIESEL EXHAUST NANOPARTICLES ...... 143 5.1 Introduction ...... 147 5.2 Experimental setup and methodology ...... 149 5.3 Fuel Selection ...... 152 5.4 Operation of the engine ...... 152 5.5 Results and discussion ...... 152 5.5.1 Volatility measurements ...... 152 5.5.2 ROS measurements ...... 157 5.6 Conclusion ...... 162 5.7 Acknowledgments...... 163 5.8 References ...... 163 CHAPTER 6: CONCLUSIONS ...... 169 6.1 Principal Significance of the Findings ...... 169 6.2 Directions for future research ...... 171

Effects of Biodiesel Chemical Composition on the Chemical and Physical Properties of the Primary and Secondary Diesel Particulate Matter vii

List of Figures

Figure 1.1 History of euro emissions standards for diesel cars [29] ...... 3 Figure 1.2 Representation of thesis flow, chapters’ content and consequent publications...... 9 Figure 2.1 Illustration of a DPM [17] ...... 26 Figure 2.2. Composition of particles from a heavy-duty diesel engine, tested in a transient cycle on an engine test bench [3] ...... 27 Figure 2.3 Typical diesel particles size and mass distribution (Kittelson 1998) ...... 28 Figure 2.4 Organic aerosols from different sources undergo chemical processing and mix with each other, as well as with inorganic aerosols [86]...... 33 Figure 2.5 Dependence of the organic fraction of DE on the dilution ratio [90]...... 35 Figure 2.6 Evolution of secondary organic matter in presence of primary particles [122]...... 43 Figure 2.7 Evolution of secondary particles in the absence of primary particles [122]...... 43 Figure 2.8 Evolution of size distribution while ageing [122]...... 44 Figure 2.9 Particle count in the presence and absence of primary particles [122]...... 44 Figure 2.10 Schematic of experimental setup [123]...... 45 Figure 2.11 SOA formation in the reactor from known precursors as a function of UV intensity [124] ...... 46 Figure 2.12 Total particle mass concentrations in the chamber (a) with α-pinene addition and (b) without α-pinene addition [4]...... 50 Figure 2.13 The change in size of (a) fresh, (b) 8–12h aged, and (c) 24h aged 100nm mono- disperse aerosol as a function of heating temperature [95]...... 52 Figure 2.14 Effect of ageing on volatility of particles for (a) idle, (b) 9kw load [95]...... 52 Figure 2.15 Temporal evolution of the organic matter, without after treatment, before wall loss correction [136]...... 54 Figure 2.16 Temporal evolution of the organic matter, without after treatment, after wall loss correction [136]...... 54 Figure 2.17 Time-series of (a) measured suspended mass, (b) median diameter, and (c) wall- loss corrected mass measured during photo-oxidation of DE. [96] ...... 56 Figure 2.18 Time series of (a) gas and (b) particle phase concentrations during photo-oxidation experiment at 4% load. Vertical grey bars are time periods when thermo-denuder was running [91]...... 58 Figure 2.19 Size distribution during ageing: (a) after filling, (b) before UV lights turned on, (c) shortly after UV was on, (d) after 3 hours of ageing [91]...... 58 Figure 2.20 Measured changes in PM mass caused by photo-oxidation in different engine loads: (a) 4%, (b) 7%, (c) 30% and (d) 85% [91] ...... 59 Figure 2.21 Schematic of TSI 3010 Butanol-based CPC [149]...... 63 Figure 2.22 Schematic diagram of cylindrical DMA by Knutson and Whitby [152]...... 64 Figure 3.1 Schematic diagram of experimental set up...... 90 Figure 3.2 Brake specific PM emission at (a) 1500 rpm 100% load and (b) 2000 rpm 100% load. The PM emissions shown are calculated based on DustTrak measurement...... 91 Figure 3.3 Brake specific PN emissions at (a) 1500 rpm 100% load and (b) 2000 rpm 100%...... 92

viiiEffects of Biodiesel Chemical Composition on the Chemical and Physical Properties of the Primary and Secondary Diesel Particulate Matter

Figure 3.4 Variations in particle median size among used fuels at (a)1500 rpm 100% load and (b) 2000 rpm 100% load , while (c) shows particle median size variation with total number concentration...... 95 Figure 3.5 Brake specific NOx emission at (a) 1500 rpm 100% load and (b) 2000 rpm 100% load...... 96 Figure 3.6 Variation in specific PM and PN emissions with used fuel surface tension, viscosity and oxygen content ((a), (b), (c) for PM and (d), (e), (f) for PN), Ordinate of (a), (b), (c) and (d), (e), (f) are same where abscissa of (a), (d) and (b), (e) and (c), (f) are same...... 97 Figure 3.7 Comparison of engine performance (power, BSFC) and particle emissions (PM, PN) among biodiesels and their blends where petrodiesel was used as a reference fuel...... 98 Figure 4.1 OV for all blends and modes tested. Rows present data measured at the same load, while columns data are measured at the same blend percentage. Different fuels are sorted along the x-axis according to their carbon chain length going from the shortest to the longest...... 123 Figure 4.2 BC Emission factors for all blends and modes tested. The emission factor for B0 at each mode is shown as a dashed line. Rows present data measured at the same load, while columns data are measured at the same blend percentage. Different fuels are sorted ...... 125 Figure 4.3 OP of PM for all four fuels, for all the blends and loads...... 127 Figure 4.4 Dependence of the BC emissions factor on fuel oxygen content for the four modes tested...... 128 Figure 4.5 Dependence of the a) Overall volatile content and b) ROS concentration on fuel oxygen content for the four modes tested. R2 shown in the graphs are McFadden's pseudo R2...... 129 Figure 4.6 Dependence of the ROS concentration on the overall volatile content (OV) for the 4 modes tested. R2 shown in the graphs are McFadden's pseudo R2...... 131 Figure 4.7 Experimental setup, HEPA: High-Efficiency Particulate Air Filter; EC: Electrostatic Classifier; TD: Thermo-Denuder; V-TDMA: Volatility Tandem Differential Mobility Analyzer; CPC: Condensation Particle Counter; SMPS: Scanning Mobility Particle Sizer...... 133 Figure 4.8 Dependence of the particle volume on the temperature of the thermodenuder for two particle sizes: circles 30nm and triangles 150 nm...... 133 Figure 4.9 Volumetric Volatile Fraction (VVF) for all pre-selected particle sizes and all fuel blends. Note that graphs in the same columns are for the same fuel blend percentages and graphs in the same rows for the same biodiesel fuel. All the graphs in the first column are the same as they are forB0 (pure petrodiesel) and are there only for comparison with other blends. Different engine loads are presented with different symbols. Error bars were omitted, as they would clutter the figures...... 134 Figure 5.1Experiment setup HEPA: High-Efficiency Particulate Air Filter; F T reactor: Flow- Through Reactor; O3 Gen: Ozone Generator; Ozone Analyzer: EC9810 Ecotech; EC: Electrostatic Classifier; TD: Thermo-Denuder; V-TDMA: Volatility Tandem Differential Mobility Analyzer; CPC: Condesation Particle Counter; SMPS: Scanning Mobility Particle Sizer; Impingers: used to collect DPM for ROS measurements...... 150 Figure 5.2 The volumetric volatile fraction (vvf) of particles versus particle pre-selection sizes for different biodiesels and different blends before and after aging. Each column is dedicated to one of the biodiesels and each row specifies one blend. Red markers represent the particles before ageing and blue markers show the particles after ageing. All the tests were carried out at quarter load, 1500rpm...... 153 Figure 5.3 Volatility of particulate matter versus different blends for petrodiesel and tested biodiesels before and after ageing in the flow through reactor. Blue points are due to aged particulate matter and red points represent fresh particles. Different tested

Effects of Biodiesel Chemical Composition on the Chemical and Physical Properties of the Primary and Secondary Diesel Particulate Matter ix

biodiesels can be seen at the top of the figure. B0 representing petrodiesel is repeated in all graphs for comparison...... 155 Figure 5.4 Change in volatility of particles before and after aging against oxygen content of the blends. Different blends are shown with colors and different shapes as can be seen in the legend show different fuels. The point corresponding to C1875, B100 is excluded from the model...... 157 Figure 5.5 ROS levels of particulate matter versus different blends for petrodiesel and tested biodiesels before aging (red circles) and after ageing (blue triangles). Different tested biodiesels can be seen at the top of the figure. B0 representing petrodiesel is repeated in all graphs for comparison ...... 158 Figure 5.6 The correlation between the change in ROS levels and oxygen content the fuels before and after ageing. The point corresponding to C1875 is excluded from the model...... 160 Figure 5.7 ROS levels in gas phase before and after ageing. Fresh and aged particulate matter are separated using blue color for aged PM and red for fresh PM...... 161 Figure 5.8 The correlation between ROS of particulate matter and the volatility of particulate matter. Fresh and aged particulate matter are separated using blue color for fresh PM and red for aged PM, also the shape of the markers represents different fuel...... 162

xEffects of Biodiesel Chemical Composition on the Chemical and Physical Properties of the Primary and Secondary Diesel Particulate Matter

List of Tables

Table 3-1 Test engine specification...... 86 Table 3-2 Fatty acid profile of used biodiesels...... 87 Table 3-3 Important physicochemical properties of tested fuels...... 88

Effects of Biodiesel Chemical Composition on the Chemical and Physical Properties of the Primary and Secondary Diesel Particulate Matter xi

List of Abbreviations

AMS: Aerosol Mass Spectrometer

CMD: Count Median Diameter

CO: Carbon Monoxides

CO2: Carbon Dioxides

CPC: Condensation Particle Counter

DMA: Differential Mobility Analyzer

DMS: Differential Mobility Spectrometer

DOC: Diesel Oxidation Catalyst

DPF: Diesel Particulate Filter

DPM: Diesel Particulate Matter

EC: Elemental Carbon

EGR: Exhaust Gas Recirculation

FAME: Fatty Acid Methyl Ester

HC: Hydrocarbon

NOX: Oxides of Nitrogen

OC: Organic Carbon

OV: Overall Volatility

PAH: Poly Aromatic Hydrocarbons

PAM: Potential Aerosol Mass

PM: Particulate Matter

PMC: Particulate Matter Concentration

PN: Particle Number

PN: Particle Number Concentration

POM: Particulate Organic Matter

xiiEffects of Biodiesel Chemical Composition on the Chemical and Physical Properties of the Primary and Secondary Diesel Particulate Matter

PSD: Particle Size Distribution

ROS: Reactive Oxygen Species

SMPS: Scanning Mobility Particle Sizer

SOA: Secondary Organic Aerosol

UV: Ultra Violet

V-TDMA: Volatilization Tandem Differential Mobility Analyzer

VOC: Volatile Organic Compounds

VVF: Volumetric Volatile Fraction

Effects of Biodiesel Chemical Composition on the Chemical and Physical Properties of the Primary and Secondary Diesel Particulate Matter xiii

Statement of Original Authorship

The work contained in this thesis has not been previously submitted to meet requirements for an award at this or any other higher education institution. To the best of my knowledge and belief, the thesis contains no material previously published or written by another person except where due reference is made.

Signature: QUT Verified Signature

Date: August 2015

xivEffects of Biodiesel Chemical Composition on the Chemical and Physical Properties of the Primary and Secondary Diesel Particulate Matter

Acknowledgements

I would like to acknowledge and sincerely thank my principal supervisor Prof. Zoran Ristovski for his care and support through out this course of PhD. I also would like to thank associate supervisors A/Prof. Richard Brown and Prof. Lidia Morawska for their valuable support.

I would like to thank Dr Ehsan Majd, Dr Farhad Salimi, Dr Svetlana Stevanovic, Dr Branka Miljevic and Dr Hao Wang for their help with instrumentation, study design, measurements, data analysis and manuscript preparation.

I also wish to praise the efforts of Mr. Noel Hartnett and Mr. Scott Abbett, technicians in Biofuel Engine Research Facilities (BERF), who have provided valuable assistance during all of the experimental measurements conducted at QUT.

I wish to extend my acknowledgement and thanks to all of my PhD colleagues at ILAQH and BERF, for their assistance, co-operation, friendship, social interaction and occasional humour. In particular, Bijan Yeganeh, Dr Mostafiz Rahman, Marc Mallet, Luke Cravigan, Jana Adamovska, Andjelija Milic, Farzaneh Hedayat and many others from ILAQH.

Effects of Biodiesel Chemical Composition on the Chemical and Physical Properties of the Primary and Secondary Diesel Particulate Matter xv

Chapter 1: Introduction

1.1 THE INVESTIGATED RESEARCH PROBLEM

It is now well understood that nanoparticles originating from diesel engines can cause a number of problems in both humans and the environment. This issue has become a great concern in recent years, as well as a source of public debate, and the topic of scientific research and publications. According to publications, particles smaller than 2.5 micron are a potential risk to living organisms and the environment [1-6].

In general, fine particles remain in the atmosphere much longer than their larger counterparts. For example, fine particles can last for days and they penetrate deeper into the human respiratory system [7], whereas larger particles usually deposit closer to their source, so their life span is not likely to exceed a few hours. The other issue is that, in comparison to larger particles with an equivalent mass dose, finer particles have a larger chemically active surface area which is capable of reacting with cells [8], and thus, finer particles are more chemically-reactive.

One of the most infamous sources of fine particles is compression ignition (CI) or diesel engines [9]. Diesel particulate matter (DPM) is well known because of its adverse effects on humans, animals and the environment [10, 11]. There are many cardiovascular and respiratory diseases related to long or short-term exposure to diesel originated particles [12, 13]. In recent years, the topic of human exposure to fine particles emitted from diesel engines has been drawing a great deal of attention. As a result, a significant amount of research has been carried out to date, with a number of publications pertaining to the issue [1-6]. However, due to the fact that diesel exhaust (DE) is now considered carcinogenic [14], never before has it been in the spotlight as much as it is nowadays. It would be neither an exaggeration nor unexpected that diesel engine manufacturers, as well as health and environmental organizations, will soon divert a massive portion of funds towards research on DE and therefore, it is predicted that diesel emissions will be the focus of many publications in the not-so-distant future.

Chapter 1: Introduction 1

Based on the fact that the highest number concentration of DPM occurs in the nanoparticle size range (<50nm), being capable of remaining suspended in the atmosphere for days and penetrating deep into the respiratory system [15], it is of great importance to investigate DPM in terms of both its physical properties and chemical composition.

To date, many authors have investigated the health effects of DPM, including its toxicological effects, epidemiological outcomes, effect on climate change etc, while at the same time, environmental protection agencies are also imposing more stringent regulations to force manufacturers produce cleaner engines. As can be seen in Figure 1 .1, there has been an increasingly downward trend in DE emissions from 1995 to 2011.

To meet recent stringent emission regulations, companies have developed and applied new technologies in engines [16]. The replacement of mechanical injection systems by a common rail system significantly reduced particulate matter (PM) emissions [17]. While using Exhaust Gas Recirculation (EGR) can lower nitrogen dioxides (NOx) emission, but the unwanted side-effect is an increase in PM emissions. The fact is that recent PM emission limitations cannot be attained merely by improving engine technologies and design, and therefore, exhaust after-treatment technologies, such as diesel particulate filters (DPMs), diesel oxidation catalysts (DOCs) and particle oxidation catalyst (POC) should be used [18-22]. Also, fuels and lubricant oils are undergoing constant modification, in an effort to reduce diesel emissions [5]. The elimination of sulphur from diesel fuel significantly reduced the formation of sulphur oxides in the combustion products, causing a notable decrease in PM emissions [23, 24]. Similarly, alternative diesel fuels are gradually finding their place as cleaner choices for petrodiesel. For instance, it has been confirmed that biodiesel can significantly reduce PM emissions, in conjunction with other exhaust emissions [25, 26] and there are a number of studies pointing out the benefits of blending a small amount of biodiesel with petrodiesel [27, 28].

2 Chapter 1: Introduction

Figure 1.1 Emission requirements for heavy diesel engine vehicles in Australia [29]

Biodiesel is currently one of the most promising alternatives to petrodiesel, and depending on its chemical and physical properties, may be considered a suitable product for blending with diesel fuel to help meet increasing fuel demands [30-32]. Biodiesel is a processed fuel which is derived from biological sources by trans- esterifying vegetable oils, animal fats or algae [32]. In terms of chemical composition, biodiesel fuels are mono-alkyl esters of fatty acids [32]. Recently, the PM produced by biodiesel combustion has become a popular research topic [4, 15, 17, 33-35]. This is understandable considering that emissions from diesel engines have been recently declared as carcinogenic [36]. It has been commonly found in the literature that biodiesel causes less PM mass concentration in the exhaust gas [33, 37, 38], however some studies have reported an increase in the particle number concentration (PNC) [34, 39, 40], as well as particle number per unit of particle mass [41]. An additional concern is the observation that non-petroleum diesel fuels emitted excessive amounts of volatile compounds upon combustion, which potentially result in more toxic emissions [15, 42, 43].

The challenge is getting more serious, as there is now concern that the particles are originating from volatile and semi-volatile compounds in the DE. It is believed that such particles are of greater interest, considering the fact that the semi-volatile fraction of DE is responsible for a majority of the redox activity of DPM [44-46]. It

Chapter 1: Introduction 3

is also likely that the redox activity of the semi-volatile fraction of DPM increases as the aerosol ages, following its exposure to atmospheric conditions [17, 47-49].

Based on the data provided in the literature so far, it remains unclear which chemical species contribute to the measured redox potential and overall toxicity of DPM. Several studies in this field [43, 44, 50] showed that the organic fraction, more precisely semi-volatile component of PM, correlated well with the measured oxidative potential (OP) of DPM.

The chemical composition of fuel is a key factor affecting all emission and by- products of an engine. Chemical properties such as carbon chain length, number of double bonds and the amount of oxygen borne by the fuel may directly influence the combustion process. They can also affect an engines output by affecting physical properties of the fuel and consequently the combustion process. There have been a growing number of publications that report on the correlation between fuel chemical composition and engine performance, as well as engine emissions [37, 51, 52].

The feedstock from which a biodiesel is derived determines the chemical composition and physical properties of the fatty acid esters in the fuel. Recent advances in science and technology have given us the ability to modify and control the fatty acid profile of oils and fats, which highlights the importance of searching for optimum fatty acid profiles, which can lower an engines emissions without sacrificing its performance [53, 54].

In summary, it is of great importance to obtain an understanding of how the physical properties and chemical composition of biodiesel influence particle emission characteristics. It is also vital to know the effect of such parameters on the transformation, dynamics and fate of DE and DPM.

1.2 STUDY AIM

The main aim of this study was to investigate the influence of the physical properties and chemical composition of biodiesel, as well as the effects of atmospheric conditions on engine exhaust particle emissions. This aim was achieved by carrying out several experimental measurement campaigns using biodiesel fuels with different physical properties and chemical compositions. The biodiesel fuels tested included conventional biodiesel feedstocks and biodiesel with a controlled chemical composition.

4 Chapter 1: Introduction

Changes in particle number and size due to fuel properties were characterized by using a Scanning Mobility Particle Sizer (SMPS). Information about the volatility of the particles was obtained by means of a Tandem Differential Mobility Analyzer (TDMA), which provided important data on the presence of toxic organic compounds adsorbed or condensed onto the particle surface. The atmospheric conditions were simulated using a flow-through chamber and relationships between the potential toxicity of DPM and chemical composition of the fuel and atmospheric aging were checked using a BPEA molecular probe (bis(phenylethynyl) anthracene- nitroxide).

1.3 SPECIFIC OBJECTIVES OF THE STUDY

The primary aim of this project was to investigate the effect of biodiesel chemical composition on the chemical and physical properties of the primary and secondary particles. This study investigated the advantages and disadvantages of using biodiesels in terms of PM emissions. The effect of chemical composition, together with the influence of different atmospheric conditions on the partitioning of semi-volatile compounds from the gas phase into the particle phase was also investigated.

The objectives of the presented study were to:

1. Investigate the effect of biodiesels, with a controlled chemical composition and physical properties, on engine exhaust particle emissions, as well as volatile and semi-volatile compounds in the DE.

By using controlled chemical composition biodiesel we are able to assign certain changes in chemical and physical properties of DPM to certain characteristics of the biodiesel. For instance, we can check which one of the chemical properties of fuel is more influential in terms of the amount of volatiles in the DE.

2. Examine the effect of dilution, aging, exposure to ozone and ultra-violate light (UV) on the partitioning of the semi-volatile compounds in fine diesel particles.

Chapter 1: Introduction 5

As the production of biodiesel is on the increase; it is of high importance to investigate the potential secondary PM emission as the result of DE caused by different biodiesels. It will be discussed in the Literature Review chapter that the effect of atmospheric aging on biodiesel exhaust is not well understood. So, the contribution of different biodiesels on the secondary organic aerosols needs to be studied in greater details.

3. Scrutinize the potential toxicity of primary and secondary PM as measured by the concentration of ROS and correlate it with specific chemical properties of the fuel.

The toxicity characteristics of DE and DPM change while being in the atmosphere due to the reactions with the oxidative agents in the atmosphere. So it is very beneficial to check the changes in of toxicity characteristics of DPM in the atmospheric conditions.

The importance of the above mentioned objectives will be elaborated later through the next chapter (Literature Review).

The results of this research can be used to:

1. Enhance our knowledge of DPM.

2. Give profound insight into the fundamental basis of nanoparticle toxicity.

3. Alter ongoing environmental regulations.

1.4 EVIDENCE OF RESEARCH PROGRESS LINKING THE RESEARCH PAPERS

Three papers frame the body of this thesis. These papers are from experimental campaigns in which different types of diesel fuels were tested in a modern common rail diesel engine. The three papers are interlinked and each successive paper can be considered as extension of the previous paper.

The first manuscript was published in the journal Fuel [55] in 2014. It explored the influence of the physical properties and chemical composition of biodiesel fuels and their blends (i.e. B20 and B50) on DPM. This paper showed that the oxygen content of the tested fuel or fuel mix was the main driving force behind the reported particle emission reductions, regardless of any variation in other physical properties (i.e. density viscosity and surface tension). Fuel oxygen content also affected the

6 Chapter 1: Introduction

particle median size, which was found to decrease with increasing fuel oxygen content. It was also found that particle emissions increased with an increase in fuel viscosity and surface tension, but only within specific blends. In other words, particle emissions from two different blends with the same viscosity and surface tension were found to be significantly different. However, it should be noted that the test engine for this campaign was equipped with a common rail injection system, which may have minimized the effect of small variations in fuel viscosity and surface tension, due to the high fuel injection pressure.

The second manuscript, which was published in Environment Science and Technology (ES&T) [56], investigated the effect of biodiesel FAME composition on the volatility and oxidative potential (OP) of the emitted particles. It was observed that the volatility and OP of particles depended upon the FAME carbon chain length of the biodiesel fuel. Particles emitted from short chain saturated biodiesel combustion were found to have a higher fraction of volatile organic substances adsorbed onto them, leading to higher reactive oxygen species (ROS) emissions. The particles overall volatility (OV) and ROS correlated exponentially with the oxygen content of the fuel or fuel mix, which could either be due to a reduction in soot onto which the volatiles condense, or an actual increase in the amount of volatile materials. After checking the emitted black carbon from the blend, it was concluded that the exponential relationship observed between OV and ROS was a combination of both causes: a reduction of elemental carbon mass and an increase of organics in the particulate phase. These results further highlighted the influence of FAME composition on the toxicity of particles emitted from biodiesel combustion. In addition, they also challenge the effectiveness of current mass and number based emission standards in addressing the health effects of DPM, and reinforce the importance of more inclusive emission standards.

The third manuscript was submitted to the Atmospheric Chemistry and physics (ACP) journal. In the latter manuscript, the effects of atmospheric aging on biodiesel exhaust were investigated. The objective of this research was to investigate the effect of the chemical composition of biodiesel and the potential of biodiesel exhaust to generate SOA, as well as to compare the potential toxicity of aged particles caused by biodiesel with that of petrodiesel. It was concluded that saturated fuels reduced PM emissions, however they caused more volatile PM and significantly more semi-

Chapter 1: Introduction 7

volatiles in the gas phase, with the potential to partition between the gas phase and particle phase upon aging. The potential toxicity of DPM from biodiesel exceeded that of petrodiesel. Exposure to oxidative agents caused the OP of particles to increase, meaning that the potential toxicity of DPM can increase while being aged in the atmosphere. It is worth mentioning that the different biodiesels tested during the experimental campaigns resulted in different fractions of volatility and different levels of ROS in both fresh and aged DPM. The results of this research challenge the findings of a number of previous studies claiming that use of biodiesel tends to decrease the level of toxicity of DE [57-59]. While current legislation treats all biodiesel fuels the same, the work presented in this paper shows that the exhaust gas emitted from their combustion is substantially different. Such emissions were found to depend on the chemical composition of biodiesels, which varied significantly according to the different feedstocks used. The exhaust gases from various biodiesel fuels also behaved differently when reacting with oxidative agents from the atmosphere, leading to different levels of organic contents and ROS. Moreover, the volatile and semi-volatile fraction of DE, the main source of secondary organic aerosol production, is not yet regulated. Perhaps new regulations should consider some of the most important characteristics of the biodiesel and classify them into groups based on the oxygen content or level of saturation etc [60]. This study encourages the application of PAM chambers, in order to check the maximum capacity of DE to cause SOA and harmful matter. It also points out the necessity of redefining current legislation and standards towards the use of biodiesel.

8 Chapter 1: Introduction

Thesis aim: Factors affecting physical and chemical properties and transformation of diesel particulate matter

Experimental study: Campaigns on FAME diesel fuels in 2012

Method: Comparisons of Method: Comparisons of Method: Examination engines performance, volatile organic fraction and comparisons of PM mass concentration, together with their aging of DPM in number concentration, oxidative potential of simulated atmospheric size distribution, NOx DPM emitted from conditions, comparisons emission while operation controlled chemical of volatility and ROS of on controlled chemical composition biodiesels aged DPM with fresh composition biodiesels DPM

Chapter 3: Published in Chapter 4: Published in Chapter 5: Under review Fuel Journal: Effects of ES&T: Effect of fuel in ACP: Effects of fuel chemical chemical composition on atmospheric dilution and composition on PNC, volatility and redox fuel composition on PM and size distribution activity of DPM DPM volatility OP of DPM

Chapter 6: Conclusion and suggestions for future research

Figure 1.2 Representation of thesis flow, chapters’ content and consequent publications.

Overall, this PhD thesis investigated the PM emitted by diesel engines, possible mitigation strategies (in terms of the use of alternative fuels) and the technologies available to reduce these emissions. More specifically, the effect of chemical composition and the physical properties of biodiesel fuels on diesel engine exhaust particle emissions were investigated.

Chapter 1: Introduction 9

1.5 REFERENCES

1. Mueller, N.C. and Nowack, B., Exposure Modeling of Engineered Nanoparticles in the Environment. Environmental Science & Technology, 2008. 42(12): p. 4447-4453. 2. Kittelson, D.B., Engines and Nanoparticles: A Review. Journal of Aerosol Science, 1998. 29(5-6): p. 575-588. 3. Burtscher, H., Physical Characterization of Particulate Emissions from Diesel Engines: A Review. Journal of Aerosol Science, 2005. 36(7): p. 896- 932. 4. Kumar, P., Robins, A., Vardoulakis, S., and Britter, R., A Review of the Characteristics of Nanoparticles in the Urban Atmosphere and the Prospects for Developing Regulatory Controls. Atmospheric Environment, 2010. 44(39): p. 5035-5052. 5. Maricq, M.M., Chemical Characterization of Particulate Emissions from Diesel Engines: A Review. Journal of Aerosol Science, 2007. 38(11): p. 1079- 1118. 6. McMurry, P.H., A Review of Atmospheric Aerosol Measurements. Atmospheric Environment, 2000. 34(12-14): p. 1959-1999. 7. Nemmar, A., Hoet, P.H.M., Vanquickenborne, B., Dinsdale, D., Thomeer, M., Hoylaerts, M.F., Vanbilloen, H., Mortelmans, L., and Nemery, B., Passage of Inhaled Particles into the Blood Circulation in Humans. Circulation, 2002. 105(4): p. 411-414. 8. Tetley, T.D., Health Effects of Nanomaterials. Biochemical Society Transactions, 2007. 35: p. 527-531. 9. Kim, S., Shen, S., Sioutas, C., Zhu, Y.F., and Hinds, W.C., Size Distribution and Diurnal and Seasonal Trends of Ultrafine Particles in Source and Receptor Sites of the Los Angeles Basin. Journal of the Air & Waste Management Association, 2002. 52(3): p. 297-307. 10. McClellan, R.O., Health-Effects of Exposure to Diesel Exhaust Particles. Annual Review of Pharmacology and Toxicology, 1987. 27: p. 279-300. 11. Giechaskiel, B., Alfoldy, B., and Drossinos, Y., A Metric for Health Effects Studies of Diesel Exhaust Particles. Journal of Aerosol Science, 2009. 40(8): p. 639-651. 12. Gerlofs-Nijland, M.E., Totlandsdal, A.I., Kilinc, E., Boere, A.J.F., Fokkens, P.H.B., Leseman, D., Sioutas, C., Schwarze, P.E., Spronk, H.M., Hadoke, P.W.F., Miller, M.R., and Cassee, F.R., Pulmonary and Cardiovascular Effects of Traffic-Related Particulate Matter: 4-Week Exposure of Rats to Roadside and Diesel Engine Exhaust Particles. Inhalation Toxicology, 2010. 22(14): p. 1162-1173. 13. Nemmar, A., Al-Maskari, S., Ali, B.H., and Al-Amri, I.S., Cardiovascular and Lung Inflammatory Effects Induced by Systemically Administered Diesel Exhaust Particles in Rats. American Journal of Physiology-Lung Cellular and Molecular Physiology, 2007. 292(3): p. L664-L670. 14. Lambe, A., Ahern, A., Williams, L., Slowik, J., Wong, J., Abbatt, J., Brune, W., Ng, N., Wright, J., and Croasdale, D., Characterization of Aerosol Photooxidation Flow Reactors: Heterogeneous Oxidation, Secondary Organic Aerosol Formation and Cloud Condensation Nuclei Activity Measurements. Atmos. Meas. Tech, 2011. 4(3): p. 445-461.

10 Chapter 1: Introduction

15. Sidhu, S., Graham, J., and Striebich, R., Semi-Volatile and Particulate Emissions from the Combustion of Alternative Diesel Fuels. Chemosphere, 2001. 42(5-7): p. 681-690. 16. Khair, M.K., Engine Design for Pm Control. DieselNet Technology Guide, 2002. 05a(http://www.dieselnet.com/tech/engine_design_pm.html). 17. Surawski, N.C., Miljevic, B., Ayoko, G.A., Roberts, B.A., Elbagir, S., Fairfull-Smith, K.E., Bottle, S.E., and Ristovski, Z.D., Physicochemical Characterization of Particulate Emissions from a Compression Ignition Engine Employing Two Injection Technologies and Three Fuels. Environmental Science & Technology, 2011. 45(13): p. 5498-5505. 18. Majewski, W.A., Diesel Oxidation Catalyst. DieselNet Technology Guide 2010(http://www.dieselnet.com/tech/cat_doc.html). 19. Majewski, W.A., Diesel Particulate Filters. DieselNet Technology Guide, 2011(http://www.dieselnet.com/tech/dpf.html). 20. Bugarski, A.D., Schnakenberg, G.H., Hummer, J.A., Cauda, E., Janisko, S.J., and Patts, L.D., Effects of Diesel Exhaust Aftertreatment Devices on Concentrations and Size Distribution of Aerosols in Underground Mine Air. Environmental science & technology, 2009. 43(17): p. 6737-6743. 21. Herner, J.D., Hu, S., Robertson, W.H., Huai, T., Collins, J.F., Dwyer, H., and Ayala, A., Effect of Advanced Aftertreatment for Pm and Nox Control on Heavy-Duty Diesel Truck Emissions. Environmental Science and Technology, 2009. 43(15): p. 5928-5933. 22. Yao, D., Lou, D.M., Tan, P.Q., Hu, Z.Y., and Feng, Q., Impact of after- Treatment Devices on Biodiesel Engine Particle Number Emission Characteristics, in Advanced Materials Research. 2014. p. 263-268. 23. Ristovski, Z.D., Jayaratne, E.R., Lim, M., Ayoko, G.A., and Morawska, L., Influence of Diesel Fuel Sulfur on Nanoparticle Emissions from City Buses. Environmental Science and Technology, 2006. 40(4): p. 1314-1320. 24. Zhang, J., He, K., Ge, Y., and Shi, X., Influence of Fuel Sulfur on the Characterization of Pm10 from a Diesel Engine. Fuel, 2009. 88(3): p. 504- 510. 25. Surawski, N.C., Ristovski, Z.D., Brown, R.J., and Situ, R., Gaseous and Particle Emissions from an Ethanol Fumigated Compression Ignition Engine. Energy Conversion and Management, 2012. 54(1): p. 145-151. 26. Xue, J., Grift, T.E., and Hansen, A.C., Effect of Biodiesel on Engine Performances and Emissions. Renewable and Sustainable Energy Reviews, 2011. 15(2): p. 1098-1116. 27. Makarevičiene, V., Lebedevas, S., Rapalis, P., Gumbyte, M., Skorupskaite, V., and Žaglinskis, J., Performance and Emission Characteristics of Diesel Fuel Containing Microalgae Oil Methyl Esters. Fuel, 2014. 120: p. 233-239. 28. Young, L.-H., Yi, J., Cheng, M.-T., Lu, J.-H., Yang, H.-H., Tsai, Y.I., Wang, L.-C., Chen, C.-B., and Lai, J.-S., Effects of Biodiesel, Engine Load and Diesel Particulate Filter on Nonvolatile Particle Number Size Distributions in Heavy-Duty Diesel Engineexhaust. Journal of Hazardous Materials, 2011(0). 29. State of the Environment. Emission Requirements for Heavy Diesel Engine Vehicles in Australia. 2009 [cited 2009; Available from: http://soer.justice.tas.gov.au/2009/image/1017/index.php. 30. Ma, F. and Hanna, M.A., Biodiesel Production: A Review. Bioresource Technology, 1999. 70(1): p. 1-15.

Chapter 1: Introduction 11

31. Li, Y., Tian, G., and Xu, H., Application of Biodiesel in Automotive Diesel Engines. 2013. 32. Demirbas, A., Biodiesel: A Realistic Fuel Alternative for Diesel Engines. 2008: Springer. 33. Cheng, C.H., Cheung, C.S., Chan, T.L., Lee, S.C., and Yao, C.D., Experimental Investigation on the Performance, Gaseous and Particulate Emissions of a Methanol Fumigated Diesel Engine. Science of the Total Environment, 2008. 389(1): p. 115-124. 34. Cheng, C.H., Cheung, C.S., Chan, T.L., Lee, S.C., Yao, C.D., and Tsang, K.S., Comparison of Emissions of a Direct Injection Diesel Engine Operating on Biodiesel with Emulsified and Fumigated Methanol. Fuel, 2008. 87(10- 11): p. 1870-1879. 35. Cheung, K.L., Polidori, A., Ntziachristos, L., Tzamkiozis, T., Samaras, Z., Cassee, F.R., Gerlofs, M., and Sioutas, C., Chemical Characteristics and Oxidative Potential of Particulate Matter Emissions from Gasoline, Diesel, and Biodiesel Cars. Environmental Science & Technology, 2009. 43(16): p. 6334-6340. 36. Benbrahim-Tallaa, L., Baan, R., Grosse, Y., Lauby-Secretan, B., El Ghissassi, F., Bouvard, V., Guha, N., Loomis, D., and Straif, K., International Agency for Research on Cancer Monograph Working Group. Carcinogenicity of Diesel-Engine and Gasoline-Engine Exhausts and Some Nitroarenes. Lancet Oncol, 2012. 13(7): p. 663-4. 37. Tinsdale, M., Price, P., and Chen, R., The Impact of Biodiesel on Particle Number, Size and Mass Emissions from a Euro4 Diesel Vehicle. 2010. 38. Kalligeros, S., Zannikos, F., Stournas, S., Lois, E., Anastopoulos, G., Teas, C., and Sakellaropoulos, F., An Investigation of Using Biodiesel/Marine Diesel Blends on the Performance of a Stationary Diesel Engine. Biomass and Bioenergy, 2003. 24(2): p. 141-149. 39. Fontaras, G., Karavalakis, G., Kousoulidou, M., Tzamkiozis, T., Ntziachristos, L., Bakeas, E., Stournas, S., and Samaras, Z., Effects of Biodiesel on Passenger Car Fuel Consumption, Regulated and Non- Regulated Pollutant Emissions over Legislated and Real-World Driving Cycles. Fuel, 2009. 88(9): p. 1608-1617. 40. Lee, H., Myung, C.L., and Park, S., Time-Resolved Particle Emission and Size Distribution Characteristics During Dynamic Engine Operation Conditions with Ethanol-Blended Fuels. Fuel, 2009. 88(9): p. 1680-1686. 41. Pham, P.X., Bodisco, T.A., Ristovski, Z.D., Brown, R.J., and Masri, A.R., The Influence of Fatty Acid Methyl Ester Profiles on Inter-Cycle Variability in a Heavy Duty Compression Ignition Engine. Fuel, 2014. 116(0): p. 140- 150. 42. Liu, Y.Y., Lin, T.C., Wang, Y.J., and Ho, W.L., Biological Toxicities of Emissions from an Unmodified Engine Fueled with Diesel and Biodiesel Blend. Journal of Environmental Science and Health Part a-Toxic/Hazardous Substances & Environmental Engineering, 2008. 43(14): p. 1735-1743. 43. Surawski, N.C., Miljevic, B., Ayoko, G.A., Elbagir, S., Stevanovic, S., Fairfull-Smith, K.E., Bottle, S.E., and Ristovski, Z.D., Physicochemical Characterization of Particulate Emissions from a Compression Ignition Engine: The Influence of Biodiesel Feedstock. Environmental Science & Technology, 2011. 45(24): p. 10337-43.

12 Chapter 1: Introduction

44. Biswas, S., Verma, V., Schauer, J.J., Cassee, F.R., Cho, A.K., and Sioutas, C., Oxidative Potential of Semi-Volatile and Non Volatile Particulate Matter (Pm) from Heavy-Duty Vehicles Retrofitted with Emission Control Technologies. Environmental Science & Technology, 2009. 43(10): p. 3905- 3912. 45. Verma, V., Ning, Z., Cho, A.K., Schauer, J.J., Shafer, M.M., and Sioutas, C., Redox Activity of Urban Quasi-Ultrafine Particles from Primary and Secondary Sources. Atmospheric Environment, 2009. 43(40): p. 6360-6368. 46. Miljevic, B., Heringa, M.F., Keller, A., Meyer, N.K., Good, J., Lauber, A., Decarlo, P.F., Fairfull-Smith, K.E., Nussbaumer, T., Burtscher, H., Prevot, A.S.H., Baltensperger, U., Bottle, S.E., and Ristovski, Z.D., Oxidative Potential of Logwood and Pellet Burning Particles Assessed by a Novel Profluorescent Nitroxide Probe. Environmental Science & Technology, 2010. 44(17): p. 6601-6607. 47. Shinyashiki, M., Eiguren-Fernandez, A., Schmitz, D.A., Di Stefano, E., Li, N., Linak, W.P., Cho, S.H., Froines, J.R., and Cho, A.K., Electrophilic and Redox Properties of Diesel Exhaust Particles. Environmental Research, 2009. 109(3): p. 239-244. 48. McWhinney, R.D., Gao, S.S., Zhou, S.M., and Abbatt, J.P.D., Evaluation of the Effects of Ozone Oxidation on Redox-Cycling Activity of Two-Stroke Engine Exhaust Particles. Environmental Science & Technology, 2011. 45(6): p. 2131-2136. 49. Miracolo, M.A., Hennigan, C.J., Ranjan, M., Nguyen, N.T., Gordon, T.D., Lipsky, E.M., Presto, A.A., Donahue, N.M., and Robinson, A.L., Secondary Aerosol Formation from Photochemical Aging of Aircraft Exhaust in a Smog Chamber. Atmospheric Chemistry and Physics, 2011. 11(9): p. 4135-4147. 50. Stevanovic, S., Miljevic, B., Surawski, N., Fairfull-Smith, K.E., Bottle, S.E., Brown, R., and Ristovski, Z.D., The Influence of Oxygenated Organic Aerosols (Ooa) on the Oxidative Potential of Diesel and Biodiesel Particulate Matter. Environmental Science & Technology, 2013. 47(14): p. 7655-7662. 51. Knothe, G., Sharp, C.A., and Ryan III, T.W., Exhaust Emissions of Biodiesel, Petrodiesel, Neat Methyl Esters, and Alkanes in a New Technology Engine. Energy & Fuels, 2006. 20(1): p. 403-408. 52. Pinzi, S., Rounce, P., Herreros, J.M., Tsolakis, A., and Pilar Dorado, M., The Effect of Biodiesel Fatty Acid Composition on Combustion and Diesel Engine Exhaust Emissions. Fuel, 2012. 53. Amaro, H.M., Guedes, A.C., and Malcata, F.X., Advances and Perspectives in Using Microalgae to Produce Biodiesel. Applied Energy, 2011. 88(10): p. 3402-3410. 54. Daroch, M., Shao, C., Liu, Y., Geng, S., and Cheng, J.J., Induction of Lipids and Resultant Fame Profiles of Microalgae from Coastal Waters of Pearl River Delta. Bioresource Technology, 2013. 146(0): p. 192-199. 55. Rahman, M.M., Pourkhesalian, A.M., Jahirul, M.I., Stevanovic, S., Pham, P.X., Wang, H., Masri, A.R., Brown, R.J., and Ristovski, Z.D., Particle Emissions from Biodiesels with Different Physical Properties and Chemical Composition. Fuel, 2014. 134(0): p. 201-208. 56. Pourkhesalian, A.M., Stevanovic, S., Salimi, F., Rahman, M., Wang, H., Pham, P.X., Bottle, S.E., Masri, A., Brown, R.J., and Ristovski, Z.D., Influence of Fuel Molecular Structure on the Volatility and Oxidative

Chapter 1: Introduction 13

Potential of Biodiesel Particulate Matter. Environmental Science & Technology, 2014. 48(21): p. 12577-12585. 57. Bagley, S.T., Gratz, L.D., Johnson, J.H., and McDonald, J.F., Effects of an Oxidation Catalytic Converter and a Biodiesel Fuel on the Chemical, Mutagenic, and Particle Size Characteristics of Emissions from a Diesel Engine. Environmental Science & Technology, 1998. 32(9): p. 1183-1191. 58. Jung, H., Kittelson, D.B., and Zachariah, M.R., Characteristics of Sme Biodiesel-Fueled Diesel Particle Emissions and the Kinetics of Oxidation. Environmental Science & Technology, 2006. 40(16): p. 4949-4955. 59. Jalava, P.I., Aakko-Saksa, P., Murtonen, T., Happo, M.S., Markkanen, A., Yli-Pirilä, P., Hakulinen, P., Hillamo, R., Mäki-Paakkanen, J., and Salonen, R.O., Toxicological Properties of Emission Particles from Heavy Duty Engines Powered by Conventional and Bio-Based Diesel Fuels and Compressed Natural Gas. Particle and fibre toxicology, 2012. 9(1): p. 37. 60. Giechaskiel, B., Mamakos, A., Andersson, J., Dilara, P., Martini, G., Schindler, W., and Bergmann, A., Measurement of Automotive Nonvolatile Particle Number Emissions within the European Legislative Framework: A Review. Aerosol Science and Technology, 2012. 46(7): p. 719-749.

14 Chapter 1: Introduction

Chapter 2: Literature Review

2.1 INTRODUCTION

This chapter provides background information and a basic definition of the terms used in aerosol science. Some useful theories and hypothesizes are introduced and elaborated upon, with relevant references provided for readers who want to gain a more in-depth understanding of the subject. The most common parameters measured when characterizing an aerosol, or the PM contained in an aerosol are also explained and a brief description of the origin, dynamics, transformation, deposition and fate of atmospheric aerosols is provided. As the topic of this research is intimately associated with diesel engines, diesel fuels and diesel emissions, an emphasis has been placed on the PM emitted by diesel engines. The most important parameters affecting DPM are explored, including the type of fuel, injection technology and after-treatment technologies. The transformation and fate of DE and DPM in the atmosphere, together with secondary organic aerosol production mechanisms are also investigated, as are the health effects of DPM and the factors influencing its potential toxicity. Finally, in accordance with the literature review, several gaps in knowledge are identified and the aims and objectives of this research are introduced.

2.2 BACKGROUND AND DEFINITIONS

2.2.1 Aerosol and particulate matter An aerosol is defined as a suspension of liquid droplets or solid particles in a gaseous medium; where the gas is usually air. Particulate matter is defined as the solid particles or liquid droplets capable of being suspended in a gaseous medium. Atmospheric aerosol particles have a broad size distribution which spans over more than four orders of magnitude, including freshly born nuclei particles composed of a few molecules (in the order of a few nano meters) to coarse particles composed of an agglomeration of smaller particles whose diameter can be more than 10 micro meters [1-3]. As a result, the chemical and physical properties of atmospheric particles can vary significantly within this wide size distribution.

Chapter 2: Literature Review 15

From a terminological point of view, different expressions have been used by researchers in different fields of science. For instance, chemists and toxicologists use terms such as coarse (particles with diameter more than 1000nm), fine (smaller than 1000 nm) and ultrafine particles (smaller than 100 nm) [4]. While environmental regulatory agencies prefer to use terms like PM2.5 or PM1; where the subscript indicates the cut off for particle diameter in m [2], aerosol scientists classify particles in terms of different modes, as such nucleation,𝜇𝜇 Aitken, accumulation and coarse modes; each of which has its own formation mechanisms, source, deposition mechanisms, chemical compounds and size range [5, 6]. This manuscript uses the latter terminology, which now will be explained below.

2.2.2 Nucleation mode Nucleation mode particles range in size from 0.003 to 0.03 µm and typically contain only 0.1-10% of the total PM mass, however it often includes more than 90% of the total particle count [3]. The nature and formation processes in relation to nucleation mode particles are still under investigation; however it is believed that nucleation mode particles are primarily volatile and consist mainly of hydrocarbon and hydrated sulphuric acid condensates. These are formed from gaseous precursors as the temperature decreases in the exhaust system, as well as after mixing with cold air in laboratory dilution tunnel or in the ambient air. Diesel particulate filters are not able to capture volatile and semi-volatile compounds (which are still in a gaseous state in the exhaust system), and it is only after cooling and dilution with ambient air that nucleation particles may start to form. Therefore, it is accepted that nucleation mode particles are formed outside of DE system [7]. Nucleation mode particles cannot exist in the atmosphere for a very long and they will quickly grow into larger particles, known as the accumulation mode [3].

2.2.3 Accumulation mode The accumulation mode is made up of sub-micron particles with a diameter in the range of 0.03µm to 0.5 µm. The accumulation mode extends through the fine, ultrafine, and the upper end of the nanoparticle range [8]. Most of the particle mass falls within the accumulation mode. The maximum concentration of accumulation particles occurs between 0.1 and 0.2 µm. This is where the carbonaceous agglomerates and associated adsorbed materials reside and constitutes up to 90% of

16 Chapter 2: Literature Review

total particle mass in DE [3, 9]. Particles in this mode are composed of solids, like carbon and metallic ash, together with condensates and adsorbed material, such as heavy hydrocarbons and sulphur species [10, 11].

2.2.4 Aitken mode The Aitken mode overlaps with the nucleation and accumulation modes, with a particle diameter range of 0.02 to 0.1 µm. Since this mode is not clearly visible in many figures, mode fitting statistical analysis should be undertaken to segregate this mode from others [2, 12]. The growth and coagulation of nucleation mode particles are sources of particles in the Aitken mode. They can also be produced in high numbers by primary combustion sources such as vehicles [13]. These particles are mainly composed of a soot/ash core with a readily absorbed outer layer of volatile material [12].

2.2.5 Coarse mode The coarse mode includes particles with an aerodynamic diameter above 1 µm, which contains 5 to 20 per cent of total PM mass and makes practically no contribution to particle count [8]. Constituents of coarse particles can include large solids or droplets [14]. They may be composed of resuspended dust, coal, oil fly ash, metal oxides of crustal elements, sea salt, mold spores, tire-wear debris and so on [2, 15]. They are mostly insoluble, non-hygroscopic and capable of travelling up to 10 kilometers from the source, with a life time of generally less than a few hours [14]. In vehicular appliances, coarse particles do not form during the combustion process, instead they start forming through the deposition and subsequent re-entrainment of PM from the walls of the engine cylinder, exhaust system or after-treatment devices [7, 16, 17].

2.3 AEROSOL SCIENCE AND THEORY

Aerosol science focuses on the generation, dynamics, transformation and removal of aerosols, and the influence of aerosols on the environment and human. In the following sections the basic concepts used in aerosol science will be explored.

2.3.1 Formation and growth The mechanisms by which particles are formed have a large influence on size distribution, as well as the physical and chemical properties of particles. Knowing

Chapter 2: Literature Review 17

the dominant particle formation mechanisms involved in a process allows us to make a good estimation of the generated PM properties [18]. The most important particle formation mechanisms in air pollution sources include physical attrition/mechanical dispersion, combustion particle burnout, homogeneous and heterogeneous nucleation, and droplet evaporation [5, 19].

2.3.2 Primary particles Primary particles are directly emitted from their source into the atmosphere [20] and their sizes generally range from 20 to 50 nm, while aggregates consist of chains of ten to hundreds of such primary particles [21]. The source from which primary particles are emitted can determine both the chemical and physical characteristics of PM [22, 23].

2.3.3 Secondary particles Secondary particles are formed in the atmosphere by gas to particle conversion resulting from the chemical reaction of gaseous components in the atmosphere [6]. It is now well understood that secondary particles have a significant effect on the atmospheric particulate burden [24].

They can be divided in two categories: inorganic and organic secondary aerosols. The processes that lead to the formation of sulphates, nitrates and ammonium, as secondary inorganic aerosols, are well understood [23]. However, due to the complexity of organic compounds and their dynamics in the atmosphere, secondary organic aerosols (SOAs) have not been fully characterized [25, 26].

2.3.4 Vapor pressure Imagine a sealed container, filled to half of its volume with liquid water. The partial pressure of water vapor in this system is the pressure that the vapor would exert on the containers wall if it was alone in the container. When there is a thermodynamic equilibrium between the water vapor and the liquid water at a given temperature, the water vapor exerts its saturation vapor pressure on the glass walls [23].

The ideal gas law can estimate the mass concentration in the vapor at the saturation vapor pressure and it is denoted as saturated vapor concentration, Csat. Consider compound X in an aerosol system where the liquid phase is replaced by a

18 Chapter 2: Literature Review

number of aerosol droplets. When the total mass concentration in the system is smaller than Csat, the mass will be in gas phase. If the mass concentration is higher than Csat, the vapor is saturated and the difference in mass concentration will be in droplet state [23].

2.3.5 Nucleation Nucleation is the process by which particles are formed from a supersaturated vapor. When the vapor pressure of a substance decreases, the conversion of vapor phase materials to a particulate form occurs. As mentioned previously, vapor pressure is defined as the pressure at which a liquid and a gas phase are in equilibrium [27] and a decrease in vapor pressure could be the result of dilution, cooling, chemical oxidization and photo-oxidization. There are four different kinds of nucleation processes:

- Homogeneous homo-molecular nucleation: a particle is formed when gas molecules from the same chemical composition merge together.

- Homogeneous hetero-molecular nucleation: a particle is formed when gas molecules from different kinds of chemical composition merge together.

- Heterogeneous homo-molecular nucleation: a particle is formed when gas molecules with the same chemical composition accumulate on particles with different chemical composition.

- Heterogeneous hetero-molecular nucleation: a particle is formed when gas molecules with different chemical composition accumulate on particles with different chemical composition [23].

2.3.6 Condensation Condensation describes the growth of a particle (droplet or dry solid particle) as a result of gas condensing onto it. The extent to which a particle grows depends on the particle size, saturation ratio and particle size relative to the gas mean free path. The mean free path is the average distance a gas molecule travels in the gas before it collides with another molecule. Depending on these parameters, different numbers of vapor molecules will collide with the aerosol particle surface and therefore, the growth will be different [5, 6].

Chapter 2: Literature Review 19

2.3.7 Coagulation Coagulation is when an aerosol particle collides with another aerosol particle and together they form a new one. The most common type of coagulation is thermal coagulation, which is a spontaneous reaction that occurs as a result of the Brownian motion of aerosol particles [6]. Another common cause of coagulation is when a small particle collides with a large particle. This is most likely to occur when there are a large number of small particles available to collide with large particles, due to their vast surface area. This results in a decrease in the number concentration of small particles and an often negligible increase in size for the larger particles. The coagulation rate depends strongly on particle number concentration. For example, when studying engine exhaust, which has a very high number concentration, size distribution changes in a matter of seconds. The opposite is true for ambient air, where this process can take several days. For aerosol concentrations greater than 105 particles/cm3, coagulation becomes an important factor [5, 6, 28]

2.3.8 Particle size, mass concentration and number concentration As previously mentioned, the characteristics of an aerosol are not merely determined by the physical and chemical properties of a sole particle; amd it is important to understand some of other properties of the PM affecting its transportation and transformation. Such properties are particle size distribution, particle mass concentration and particle number concentration [22]. Despite the fact that significant progress has been made in aerosol science, a great deal of questions remain regarding the dynamics and transformation of nanoparticles.

There are different techniques which give different quantities for each of the abovementioned properties, which are discussed below.

2.3.9 Particle equivalent diameter In an ideal aerosol, all particles would be spherical and with homogeneous physical and chemical properties. So it would be very easy to use the diameters of a collection of particles to estimate a size distribution and to assign it to another aerosol with the same particle sizes. However, it goes without saying that the situation in the real world is substantially different. Real-world aerosols are composed of particles with different and irregular shapes, as well as a variety of materials with different chemical compositions and different properties (e.g. density,

20 Chapter 2: Literature Review

diffusely, conductivity, and reflectivity). These differences have a variety of influences on the related aerosols causing them to behave differently. Hence, the concept of equivalent diameters has been introduced, because of the irregularities of particles within an aerosol. The equivalent diameter of the particle will be the diameter of a sphere with the same properties as the irregular shaped particle. Equivalent particle diameters are often defined by the measurement principle. The most frequently used equivalent diameters are as follows:

Aerodynamic equivalent diameter da: the aerodynamic diameter is defined as the diameter of a sphere with standard density that settles at the same terminal velocity as the particle of interest [29]

Vacuum aerodynamic equivalent diameter dva: the vacuum aerodynamic equivalent diameter is defined as the aerodynamic diameter in vacuum when the Knutson number is much smaller than 1. In such a regime, the air flow is no longer described as a continuum, but rather discrete collisions between the gas molecules and particles [29]. The latter equivalence diameter is frequently measured in instruments that use low-pressure (<1.5 Torr or 200 pa), with aerodynamic lens systems as inlets, such as many aerosol mass spectrometers, including the AMS [29].

Volume equivalent diameter, dve: the volume equivalent diameter is defined as the diameter of a sphere that has the same volume as the particle [30].

Electrical mobility equivalent diameter, dm: the electrical mobility diameter is the diameter of a sphere with the same migration velocity in a constant electric field as the particle of interest [31]. This particle size is used in Differential Mobility Analyzer (DMA) and Scanning Mobility Analyzer (SMPS).

2.3.10 Mass concentration Mass concentration of PM is perhaps one of the most frequently measured parameters of an aerosol and it is defined as the mass of particles in one unit volume of air.

Measurements of particle mass concentration are important for regulatory agencies, since air pollution and workplace exposure standards are mainly expressed in this way. Mass concentrations are also routinely measured in aerosol research studies, although research studies focus more on the speciation and size distribution of PM. The measurement of particle mass concentration is of high importance, in

Chapter 2: Literature Review 21

order to validate the sum of measured species or normalize each species. Mass concentration can either be measured directly by gravimetric methods or by the measurement of another property (e.g. light scattering), which can then be related to particle mass [17].

2.3.11 Particle number concentration The number concentration of particles within an aerosol (PNC) is another important characteristic emerging in aerosol science. PNC is defined as the number of particles in the unit of volume of air (number/cm3).

Recent health effects studies have suggested that health effects caused by particulate matter may be more sensitive to number concentration than to mass concentration [32, 33].

Particle counters can be categorized as direct and indirect detection instruments. Direct particle counters determine the number concentration by counting individual over-grown particles, while indirect instruments estimate number concentration by measuring some other qualities of the aerosol (e.g. light scattering) and relating them to the number concentration of particles [1].

2.3.12 Deposition mechanisms In aerosol science, deposition is the process by which particles within an aerosol are collected or deposited on a surface. Particles have different deposition mechanisms in accordance with their size [23], as is explained below.

2.3.13 Diffusion Brownian motion is the random movement of particles caused by collision between the particles and surrounding gas molecules. The movement and transport of particles caused by Brownian motion is called diffusion and is described by the Stokes-Einstein’s equation. Brownian diffusion is the most effective deposition mechanism for particles smaller than 0.1 μm [22, 30]. Hence, diffusion is the main deposition mechanism for the particles emitted in DE [34].

2.3.14 Gravitational settling Deposition by gravitational settling occurs as a result of the influence of the earth’s gravity on particles suspended in the air. As the terminal velocity of particles

22 Chapter 2: Literature Review

is directly proportional to the size of the particles, this deposition mechanism is more effective on larger coarse particles [35].

2.3.15 Electrostatic attraction When a charged particle is exposed to an electric field it will migrate at a velocity that is determined by a balance between the resisting electrostatic force and aerodynamic drag that resists its motion [30]. So larger particles deposit slower than smaller particles due to electrostatic forces [1]. The latter principle is used in many sizing instruments in order to classify the particles in a poly-disperse aerosol [30].

2.3.16 Thermophoresis Thermophoresis is a phenomenon observed in mixtures of mobile particles where the different particle types exhibit different responses to the force of a temperature gradient. Thermophoretic force results from a temperature gradient established in the gas medium. An aerosol in that temperature gradient experiences a force in the direction of decreasing temperature. So particles are directed and concentrated around the surfaces with lower temperature, and eventually deposited on them [22].

2.3.17 Impaction Larger particles have more inertia when in motion. So, they tend to maintain their velocity and not follow changes in the direction or speed of the flow, and thus, they may impact and stick to the wall of the exhaust or sampling system. It is clear that impaction is the dominant deposition mechanism for larger particles in a flow and occurs for coarse mode particles usually larger than 0.5 μm [1, 30, 36].

2.4 SOURCES OF PARTICLES

Sources of particle matter are classified into two broad categories: Natural, particles from natural sources; and anthropogenic particles generated from sources relating to human activity.

2.4.1 Biogenic While wind-blown dust, volcanoes and forest fires are significant sources of nanoparticles, the main natural sources of nanoparticles in many regions of the world are forests and oceans. Although number concentrations of nanoparticles in marine and forest environments are generally 100 to 1000 times less than those in urban areas, given the fact that the majority of the earths’ surface is comprised of

Chapter 2: Literature Review 23

environments such as these, their effect on the total amount of atmospheric particles is, indeed, non-negligible [37].

2.4.2 Anthropogenic Anthropogenic sources of PM include vehicular appliances, industry, mines, biomass combustion, chemical manufacturing, cooking, welding and so on [38]. It was shown in a recent source apportionment study that vehicular particles accounted for more than 65 percent of PM in urban areas [39]. Interestingly, almost all particles smaller than 1 µm either arise from combustion processes or are the result of secondary particle formation processes, whereas coarse particles, with diameters larger than 1 mm, are mainly primary in nature and result from the surface of the Earth (sea, land, vegetation) [40]. Among the different sources of combustion nanoparticles, diesel engines are the most significant in urban areas [3, 41, 42]

2.5 DIESEL EXHAUST

Diesel exhaust is a mixture of gases and particulates produced during the combustion of diesel fuel in compression ignition engines. In a complete combustion reaction, a compound reacts with an oxidizing element, such as oxygen, and the products are compounds of each element in the fuel together with the oxidizing element. For example, the complete combustion of any hydrocarbon fuel should yield only water (H2O) and carbon dioxide (CO2). However, complete combustion is almost impossible to achieve in the real world; and therefore, as actual combustion reactions move towards equilibrium, a wide variety of major and minor species will be present, such as carbon monoxide (CO) and pure carbon (soot or ash). Additionally, any combustion in atmospheric air, which is 78% nitrogen, will also create several forms of nitrogen oxides. Incomplete combustion also results in the formation of a number of different compounds, present both in the gaseous phase and as particulates [17, 43]. Workers exposed to DE are at risk of health effects ranging from irritation of the eyes and nose, headaches and nausea, to respiratory disease and lung cancer. In 1989, the International Agency for Research on Cancer (IARC) classified DE as a probable carcinogen group 2A based on sufficient experimental evidence and limited evidence of carcinogenicity in humans [44]. Later, in 2012, the IARC classified DE as carcinogenic Group 1 [45].

24 Chapter 2: Literature Review

2.5.1 Diesel particulate matter Diesel exhaust contains very small particles that are known as DPM, consisting primarily of solid elemental carbon (EC) cores with organic carbon (OC) compounds coating the surface [46, 47]. DPM is well known because of its adverse effects on humans, animals and the environment [34, 48]. DE is a complex mixture of the combustion products of diesel fuel; and its composition depends on the type of engine, operating conditions, and the type of fuel, additives and emission control systems used [17, 43]. There are many cardiovascular and respiratory diseases related to long- or short-term exposure to diesel particles [49, 50]. The primary emissions of diesel engines include gaseous precursors like H2SO4, SO2, SO3, H2O, low-volatile and semi-volatile organic compounds, agglomerated solid carbonaceous material (soot particles) of about 15 – 40 nm in diameter and metallic ash [7]. Figure 1 .1 illustrates the typical composition of diesel particles. DPM consists of soot with a diameter of 15-40 nm, which form agglomerates with a size of about 60-100 nm [7]. There is also metallic ash from the engine imbedded in the soot, as well as condensed organic compounds. DPM also consists of small nucleated particles that are formed when the warm engine exhaust passes through the colder exhaust system, before it leaves the exhaust pipe [17].

Chapter 2: Literature Review 25

Figure 2.1 Illustration of a DPM [17]

2.5.2 Chemical properties Information on the chemical composition of DPM is an important aspect of the study of diesel emissions. The chemical composition of DPM includes elemental carbon, organic material, sulphates, nitrates, ammonium, metal oxides, hydrogen ions and water [17]. It is believed that the chemical composition of DPM is mostly affected by fuel type, engine speed, load, fuel injection timing, turbocharging and EGR. The overall PM emissions tend to increase with increasing load, except when idling, which can have the highest emissions [51]. Figure 2 .2 shows the chemical composition of DPM from a heavy-duty diesel engine.

DPM emissions can be divided into a volatile (organic) mass fraction and a non-volatile mass fraction (mainly soot), a classification that may be of more biological significance [7]. The volatile mass fraction consists mainly of an organic fraction (e.g. unburned hydrocarbons, poly-aromatic hydrocarbons), nitrates and sulphates. The chemical composition of the volatile fraction determines its biological effect, as it can be dissolved in the lung fluid [34].

26 Chapter 2: Literature Review

Figure 2.2. Composition of particles from a heavy-duty diesel engine, tested in a transient cycle on an engine test bench [3]

2.5.3 Physical properties Diesel particles, as well as particles from other combustion sources, differ in size, composition and solubility, and therefore also in their toxicological properties. Typical diesel particles are agglomerates mainly consisting of spherical primary particles of about 15–40 nm in diameter [7]. However recent studies show that modern diesel engines that meet Euro IV and Euro V can cause significantly smaller primary particles [7, 52]. Basically, DPM is divided into three categories, named nucleation, accumulation and coarse modes. Particles in the nucleation mode range from 0.003 to 0.03 µm and typically contain only 0.1-10% of total PM mass, but it often includes more than 90% of the total particle count [3]. In contrast, most particle mass falls into the accumulation mode, in the 0.1 to 0.3 µm diameter range, and constitutes up to 90% of the total particle mass in DE. The coarse mode contains up to 20% of the particle mass and is formed through deposition and re-entrainment of particulate material from the walls of the exhaust or sampling system. Aerodynamic diameter of such particles is generally above 1 µm [3]. A schematic of diesel particle number and mass distribution versus particle diameter is shown in Figure 2 .3.

Chapter 2: Literature Review 27

Figure 2.3 Typical diesel particles size and mass distribution (Kittelson 1998)

2.5.4 Effect of the fuel Diesel engines are capable of operating on a variety of different fuels and thus, diesel fuel can be extracted from a variety of different sources, such as petroleum, animal fat and vegetable oil; and it can even be synthesized via the Fischer Tropsch process [53]. As a result, the chemical and physical properties of diesel fuels vary depending on the feedstock [54, 55]. The chemical composition of alternative diesel fuels are quite diverse and as a consequence, the physical properties of such fuels are different from those of petrodiesel [56], which in turn affects the combustion process, engine performance and engine emissions [56, 57]. The subsequent sections look into the effect of different diesel fuels on the operation and emissions of diesel engine.

2.5.4.1 Petrodiesel The most conventional and commercialized type of diesel is, without a doubt, petrodiesel, which comes from the fractional distillation of crude oil. Depending on the refinery process, as well as the composition of crude oil, the resulting petrodiesel can vary in terms of its quality. However, the average chemical formula of petrodiesel is considered C12H23 [53]. Cetane number is a metric to measure the quality of diesel fuel and is an inverse function of a fuels ignition delay [53, 57]. The cetane number of n-hexadecane is considered 100 as the reference; so the cetane numbers of other diesel fuels are estimated compared to that of n-hexadecane. In practice, petrodiesel contains various amounts of aromatics, organic sulphur and

28 Chapter 2: Literature Review

nitrogen compounds which decrease the quality and cetane number of the diesel fuel [58].

2.5.4.2 Synthetic diesel Diesel fuel can be synthesized through gasification of any carbonaceous material, including biomass, biogas, natural gas, coal etc. The gasified raw material is purified and converted by the Fischer–Tropsch process into synthetic diesel fuel. This process is usually called biomass-to-liquid (BTL) or gas-to-liquid (GTL) or coal-to-liquid (CTL), according to the raw material. Synthetic diesel generally has a very low content of sulphur and aromatics, which reduces the unregulated emissions, sulphur oxides, nitrous oxides and PM [57-59].

2.5.4.3 Biodiesel Biodiesel is currently finding its place as one of the most promising alternatives to fossil fuels, due to the vulnerability of fossil fuel resources and the price of oil, as well as a few environmental issues [55, 60, 61]. Biodiesel is a processed fuel which is derived from biological sources [61]. Most of the biodiesel fuels are composed of Fatty Acid Methyl Esters (FAMEs) [61]. Interestingly, Rudolf Diesel, inventor of the compression ignition engine, used peanut oil the first time he ran his engine [55]. Biodiesels are a suitable and practical fuel choice for numerous reasons:

• They can be derived from a number of feedstocks, such as rapeseed, soybean, palm, waste cooking oil, tallow, jatropha, algae and so on [55, 60, 61].

• Biodiesels are renewable and degradable [60, 62].

• Apparently diesel engines operating on biodiesel experience a decrease in emissions [63-66].

As mentioned before, the physical and chemical properties of fuels can affect all of the outputs of an engine and the following sections will investigate the latter in more detail.

2.5.5 Effects of physical properties of the fuel The quality of combustion is a key factor that affects all of the outputs of an engine. The physical characteristics of the fuel can play an important role in

Chapter 2: Literature Review 29

changing an engines output [56]. For example, viscosity, bulk modulus, surface tension, density, low-temperature properties (cloud point, pour point), heating value and lubricity are among the physical properties affecting the combustion process [60, 66, 67]. Physical properties of the fuel can have an impact on the quality of fuel spray and its droplet size (atomization process), as well as injection timing [67] and ignition delay [68]. The injection technology and consequently the pressure at which the fuel is injected into the combustion chamber, is the most important factor which determines to what extent the abovementioned parameters affect the performance and emissions of a diesel engine [53]. For direct injection diesel engines, where the injection pressure would not go any higher than 15 MPa, the mechanical pump and injectors are responsible for supplying fuel system pressure and acting as the timing and delivery device. When the pressure is not sufficient in the high-pressure surroundings of the combustion chamber, the effects of different physical properties become apparent. However, in modern common-rail fuel injection system, where the injection pressure is in excess of 100 MPa, the quality of the spray and timing of the injection can be assumed to be independent of the physical properties of the fuel [53].

2.5.6 Effects of chemical properties of the fuel Chemical properties, such as carbon chain length, number of double bonds and the amount of oxygen borne by the fuel may also affect the aforementioned physical properties and in turn, the combustion process.

There have been a growing number of publications that study the correlation between fuel composition and engine emissions, especially DPM [69-71]. The relationship between biodiesel chemical properties and the resulting emissions has also been investigated in numerous studies [41, 60, 63, 66, 69, 70]. A major difference between conventional commercial diesel fuel and alternative diesel is the oxygen content of the fuel. Where petrodiesel contains virtually no oxygen, most biodiesel fuels are quite rich in oxygen [60]. The oxygen content of diesel fuel plays an important role in reducing black carbon (BC) formation when local oxygen concentration decreases due to diffusion combustion [63-65]. It has been found that blends with a higher oxygen content produce less PM, but more volatile particles and increased levels of NOx [41, 63, 69]; while fuels with a longer carbon chain length led to more PM [66, 70]. A lack of polycyclic aromatic hydrocarbons (PAHs)

30 Chapter 2: Literature Review

reduces HC and CO concentrations in the exhaust [63]. Also, the chemical composition of a fuel can influence the cetane number, causing the engine to emit less PM and NOx.

The heavier fuels, with a higher density, are richer in terms of energy content, which can change the performance and emission characteristics of the engine [72].

Reducing the aromatics in diesel fuel can reduce NOx and unburned HC in the exhaust gas. The sulphur content of diesel fuel affects PM emissions and it has been found that reducing sulphur leads to a linear decrease in PM. For this reason, and for the sake ofsome exhaust after treatment devices, the European Union has limited diesel sulphur content to 10 ppm since 2009 [73].

It can be seen in the literature that oxygenated diesel fuels usually emit less PM [71, 74-76]. However, there are a few studies reporting that such fuels can cause an increase in the number concentration of particles, as well as particle number per unit of particle mass [77-80]. In a recent study it was concluded that oxygenated diesel fuels lead to less accumulation mode particles in comparison to petrodiesel, but they tend to increase nucleation mode particles [81]. Although oxygenated fuels produced less total carbon (TC), the organic carbon to elemental carbon ratio (OC/EC) of DPM was more than that of petrodiesel [81].

It is well established that a higher concentration of oxygen in intake gas increases the NOx concentration and decreases PM in the exhaust, however the effect of oxygen content of the fuel on the performance and emission characteristics of the engine is not well understood [60].

An additional concern was the observation that non-petroleum diesel fuels, being substantially different in terms of chemical composition, emit great amounts of volatile compounds upon combustion, which can result in more toxic emissions [76, 82-84]. It has also been reported that the combustion of alternative diesel fuels can generate smaller particles, which are capable of remaining suspended for longer periods of time in the atmosphere and adsorbing other semi-volatile particles, which changes its chemical activity and potential toxicity [82, 84-86]. Knowing that smaller particles can penetrate deeper into the lungs causing inflammation at the sites of deposition, underlines the significance of the fate and transformation of such particles in the atmosphere, as will be discussed later.

Chapter 2: Literature Review 31

Still there are questions which remained unanswered regarding the effect of chain length, degree of unsaturation and oxygen content on the physical and chemical composition of DPM. More importantly, the influences of such parameters on the potential toxicity of DPM are yet to be understood.

2.5.7 Transformation and fate A lot of effort is put into technologies to minimize the harmful outputs of internal combustion engines. Nowadays we can effectively and efficiently control the exhaust gases of a diesel engine without sacrificing performance characteristics of the engine. Technologies such as Exhaust Gas Recirculation (EGR), Variable Valve Timing (VVT), Diesel Particulate Filters (DPF) and Diesel Oxidative Catalysts (DOC) are being used to control both the gas and particulate phase of the exhaust gas. However, once the exhaust gas is released into the atmosphere from the tailpipe, there will be virtually zero control over its fate and transformation. The gas and particulate phase of DE will be mixed with the air, cooled down, diluted and will undergo photochemical reactions, as well as reacting with oxidative agents. The volatility of both the particulate and gas phases, which regulates whether a compound is present as a vapor or as condensed matter, can change and thus, there is a constant and complex mass exchange between the particulate and the gas phases [87]. For instance, a compound may partition into the gas phase, where it is exposed to oxidative agents like ozone or the hydroxyl radical (OH), followed by re- condensation of the less volatile oxidation product according to their saturation vapor pressure, and so on [88]. Knowing the wide range of initial compounds (both natural and anthropogenic) that are released into the atmosphere, and the complexity of reactions and possible paths in the gas phase, as well as condensed phases, thousands of compounds can be formed, with little hope of ever identifying each of them individually [87]. Figure 2 .4 illustrates how different aerosols from a variety of sources are mixed with each other and undergo chemical reactions as they age [87].

So far, and because of the limited information available, only the start and end points of such aerosols are known, beginning with emissions from tailpipes or production from the oxidation of organic gases, and ending with deposition as dust or in rain [87] - the dynamics and reactions occurring in between these two points have not been explored sufficiently to date. Despite years of investigation and the application of numerous complicated techniques, only around 30% of the particulate

32 Chapter 2: Literature Review

organic matter (POM) has been identified as specific compounds [89]. Due to the existence of numerous types of organic molecules, including hydrocarbons, alcohols, aldehydes, carboxylic acids etc, the reactions which are occurring are extremely complex, producing a myriad of unidentified compounds, many of which remain unexplored [90]. The following sections will explore the works carried out on this subject in greater detail.

Figure 2.4 Organic aerosols from different sources undergo chemical processing and mix with each other, as well as with inorganic aerosols [87].

2.6 ATMOSPHERIC DILUTION AND CHEMISTRY

As mentioned previously, dilution conditions dramatically affect the physical and chemical characteristics of particles. Dilution changes the toxicity properties of particles and influences their impact on population exposures and public health. On the other hand, photochemical reactions and oxidization can affect a volatile or semi- volatile substance and cause gas to particle transformation. In many studies it has been concluded that dilution and photochemical reactions are the main processes through which volatile and semi-volatile compounds partition into the particle phase [87, 91, 92]. This section will further explain the transformation of DE in the atmosphere and the reactions involved, as well as the most important reactant agents.

2.6.1 VOC reactions High temperatures in the combustion chamber of internal combustion engines result in the reaction between N2 and O2 which exists in the ambient air. This reaction forms NOx (collective term for nitrogen oxides that originate from

Chapter 2: Literature Review 33

combustion) and depending on the amount of oxygen in the combustion chamber, the ratio of NOx varies. Ozone and ultraviolet (UV) light are two important components in reactions related to the production of NOx. NO reacts with ozone and forms NO2, which implies that the level of tropospheric ozone is low in regions with high NO emissions (urban regions). During daytime hours, the ambient NO2 reacts with UV light and forms NO and O. If free oxygen atoms are present in the air, NO can also be formed without UV light. With the oxygen molecule from the two latter reactions, it is also possible to form new ozone molecules [93].

Ozone is very reactive with alkenes due to its unstable double bonds. Alkenes, such as ethene and butane, are present in DE and thus, in the presence of ozone they undergo chemical reaction and cause OH radicals. This is the dominant process during night-time hours, but during the daytime UV light also contributes to the production of OH radicals which can react with combustion products [23].

The oxidation reactions between OH radicals and most volatile organic compounds (VOCs) lead to a lowering of the vapor pressure of the initial VOC. As the vapor pressure decreases the compounds are more likely to exit the gaseous state and enter the condensate phase by nucleation or condensation on pre-existing particles. When these gas to particle conversion processes occur in the atmosphere, they are typically denoted as secondary aerosol formation [23, 88].

2.6.2 Dilution Following their emission from mobile sources, particles disperse into the atmospheric background in two distinctive stages:

• tailpipe-to-road dilution by the strong turbulence generated by traffic, lasting about one to three seconds, resulting in a dilution factor up to 10000 [94].

• atmospheric turbulence-induced dilution caused by the wind and atmospheric instability, which lasts for 3-10 minutes, resulting in an additional dilution ratio of about 10 [61, 94].

Dilution and cooling conditions, whether in laboratories or the atmosphere, are key factors determining the degree to which these semi-volatile species are partitioned in the particulate phase [91, 95]. If the exhaust is highly diluted, the volatile organic compounds will be transferred from the particle to the gas phase to

34 Chapter 2: Literature Review

maintain equilibrium. If the exhaust is not sufficiently diluted (as is the case with the exhaust from vehicles in underground mines and tunnels) more volatile organic compounds will adsorb onto the elemental carbon (soot) particles. Robinson et al. [91] showed that the amount of organic compounds present (the emissions factor) is dilution dependent. Figure 2 .5 shows a set of emission factor data measured at different levels of dilution. Typical open air conditions correspond to dilution ratios of over 1000, while in low-dilution conditions, for example in underground mines, the ratios are below 100. Due to the high dilution ratio in open air roadside conditions, only a quarter of the organic molecules will condense on the particles, while in low-level dilution conditions, up to 80% of organics will condense on particles.

Figure 2.5 Dependence of the organic fraction of DE on the dilution ratio [91].

2.6.3 UV UV light is electromagnetic radiation with a wave length between that of visible light and X-rays, in the range of 10 nm to 400 nm. The Sun emits UV light, but a great deal of it (>97%) is blocked by the ozone layer. It can cause many substances to glow or fluoresce. However, in this research project, the ability of UV light to induce chemical reaction is of interest. When DE gases starts travelling into the atmosphere, it is exposed to UV light which makes DE oxidize. As noted previously, this mechanism would cause a vapor pressure reduction in the volatile and semi-volatile fraction of DE and lead to particle formation and condensation.

Chapter 2: Literature Review 35

In many papers, DE is exposed to different types of UV light form a variety of sources (e.g. sun light, black light, amalgam UV bulbs and so on) in order to investigate the effect of UV on secondary particles. Here, the most cited studies on this subject are reviewed.

Miracolo et al. [92] investigated the photo-oxidation of exhaust gases from a gas turbine by exposing it to UV and sunlight, and concluded that photo-oxidation created a substantial amount secondary PM, greatly exceeding the direct PM emissions at each engine load after an hour or less of aging during typical summertime conditions.

Leskinen et al. [96] aged DE in a chamber using UV exposure and the injection of ozone. They conducted several aging experiments with different mono-disperse particles during a 24 hour period of aging. Volatility and the ratio of organic carbon to elemental carbon (OC/EC) were monitored continuously during the experiments, as well as ozone concentration and UV radiation. By the end, it was concluded that after 24 hours of aging, 99% of the volumetric fraction of 100nm particles were volatile, suggesting that the particles were no longer primary, but they were formed in the chamber through nucleation.

Weitkamp et al. [97] exposed DE to black light and looked into the effect of UV light on the initiation of photo-oxidization of DE and claimed that after a few hours of exposure, photochemical aging produced a significant amount of secondary organic aerosols, almost twice as much as what was predicted by secondary organic aerosol models.

2.6.4 Ozone

Ozone, O3, is composed of three oxygen atoms. International Union of Pure and Applied Chemistry (IUPAC) called it tri-oxygen and it is considered to be an allotrope for oxygen. Ozone is much more active than di-oxygen (O2) with a half-life of an hour in the lower atmosphere. Ozone is formed from di-oxygen under the action of UV light and also electrical discharge in the atmosphere. The total concentration of ozone throughout the Earths’ atmosphere constitutes no more than 0.6 parts per million [98].

As mentioned, ozone is a powerful oxidant which can react with organic materials such as mucus, respiratory tissues, plant tissues and organic gases emitted

36 Chapter 2: Literature Review

from internal combustion engines [99]. A great deal of volatile and semi-volatile organics are being emitted to the atmosphere by diesel engines. As signified, diesel particulate filters are not able to remove there organics from DE. Once released into the atmosphere, ozone reacts with volatile and semi-volatile organics in the gas phase, reducing their vapor pressure and causing them to condense onto pre-existing particles or to form new particles [23, 26].

There are a number of studies investigating the influence of ozone on diesel emissions, all of which agreed to the fact that ozone plays a key role in aging and oxidizing DE. This causes transformation from a gas to a particle phase [100], which eventually results in an increase in PM toxicity [96, 101]. Sleiman et al. [102] examined ozone-initiated reactions leading to secondary organic aerosols and concluded that ozone induced the formation of ultrafine particles (<100 nm). In another study, atmospheric oxidation mechanisms for a wide range of volatile organic compounds were reviewed in detail by Atkinson and Arey [19]. Kroll and Seinfeld [103] noted that gas-phase oxidation, initiated by reactions with species such as the hydroxyl radical, nitrate radical and ozone, is the primary process by which the volatilities of organic species in the atmosphere evolve, and therefore, it is a likely cause of particle formation.

2.7 SEMI-VOLATILE COMPOUNDS

The United States Environmental Protection Agency (USEPA) has defined a volatile organic compound as any compound of carbon participating in atmospheric photochemical reactions, excluding a few that are listed as “Exempt VOC” [14]. There is no quantitative definition or lower-limit for “volatile”. According to the USEPA, semi-volatile organic compounds are composed primarily of carbon and hydrogen atoms, whose boiling points are in excess of 200°C [14]. Common semi- volatile organic compounds include phenols and phthalates [14]. There is no unanimous definition for semi-volatile compounds. For instance, the European Union applies a different definition: ‘a volatile organic compound is any organic compound whose boiling point is less than or equal to 250° C at an atmospheric pressure of 101.3 kPa’ [104], whereas the World Health Organization (WHO) defines semi- volatile compounds as ‘those with a boiling point range between 240-260 to 380- 400° C’ [105].

Chapter 2: Literature Review 37

One of the most important sources of semi-volatile compounds in urban areas are light and heavy duty diesel engines [2, 106]. Combustion of diesel produces a considerable amount of volatile and semi-volatile compounds, which are released into the atmosphere and exposed to ozone, UV light and other oxidative agents. After volatile compounds are oxidized, their vapor pressure is reduced and they will start changing state from the gas phase to the particle phase [96, 107]. There is a considerable amount of research that has been carried out on primary semi-volatile emissions produced by CI engines. Some of which performed toxicity assays in order to measure the extent of toxic compounds in DPM and the contribution of the semi- volatile fraction to the toxicity of particles [82, 86, 101, 107-110].

Semi-volatile compounds escape from after-treatment devices and after undergoing photochemical oxidization reactions, they condense on other pre-existing particles or form new particles by nucleation. Based on the fact that new after- treatment devices are able to efficiently remove solid particles, fewer surfaces are available for volatile or semi-volatile compounds to be adsorbed or absorbed. As a result, the occurrence of nucleation mode particle has increased significantly in new diesel engines [86, 108, 111].

Considering the move from conventional fossil diesel fuels to non-petroleum based fuels, a number of publications investigated emissions from alternative diesel fuels [2, 75, 79, 84-86]. In accordance with their experimental results, some authors have shown that non-petroleum diesel fuels produce more semi-volatile compounds and probably more toxic substances [82, 84]. It has been found that the mean diameter of DPM decreases while operating on alternative fuels such as biodiesels. Because smaller particles are more likely to allow for the adsorption of toxic exhaust components, and they facilitate the deep penetration of fine particles into the respiratory and circulatory system [82, 84-86], the reduced particles size and increased volatile content are two possible disadvantages of using alternative diesel fuels. A number of researchers showed that bio-fuels produce higher particle number concentrations [77-79], while in a few other articles, it was concluded that they produce less particle number concentration [75].

Despite the fact that only a small fraction of DPM is comprised of semi- volatile organic compounds, these compounds are thought to be responsible for the majority of DPM toxicity [112]. In recent studies, a direct relationship was found

38 Chapter 2: Literature Review

between the volatile fraction of DPM and its OP linking to DPM’s toxicity [113- 115], however the matter requires further investigation.

2.8 SECONDARY ORGANIC AEROSOLS

Secondary aerosols are formed in the atmosphere by the process of mass transfer from the gas phase to the particle phase [6]. Secondary aerosols comprise a large fraction of fine particles in urban areas. They can be divided in two categories: inorganic and organic secondary aerosols. The processes that lead to the formation of sulphates, nitrates and ammonium, as secondary inorganic aerosols, are well understood. However, due to the complexity of organic compounds and their dynamics in the atmosphere, secondary organic aerosols have not been fully characterized. As organic gases are oxidized by oxidization agents such as ozone, hydroxyl radical, nitrite radicals and UV light, their vapor pressure decrease and nucleation and condensation processes start. The origin and precursors of secondary aerosols are still uncertain and the mechanisms of secondary particle production also need more thorough investigation. Robinson discussed a theory that a large group of volatile organic compounds and other low-volatility gas-phase species contribute to secondary aerosol formation. To examine their theory, they exposed DE to UV light. The results of the experiment showed that the amount of particulate mass increased, almost instantly, upon UV exposure and the increase in secondary organic aerosol formation was almost linear. They used different models to calculate the expected increase in aerosol mass (i.e. the formation from previous known factors). The results of the experiment showed that a larger group of compounds were involved in secondary organic aerosol formation than previously thought, and that the mass increase upon UV exposure is considerably larger than model predictions [91]. Information on the formation and transport of secondary organic aerosol is important for making models that predict the spatial distribution of particles and their chemistry, based on knowledge of gaseous emissions, weather patterns and oxidant levels. However, predicted organic aerosol levels from most of these models have shown persistent discrepancies with measured data, generally underestimating secondary organic aerosol formation [86, 116]. Also, primary organic aerosols are commonly recognized as non-volatile in traditional inventories and air quality

Chapter 2: Literature Review 39

models. As is stated in previous sections and also supported by recent laboratory experiments [95, 117], increased dilution ratios to ambient conditions can make some semi-volatile fractions in the primary organic aerosol participate in the gas-particle partitioning process. Taking into account that traditional secondary organic aerosol formation models consider volatile organic compounds (such as monoterpenes and light aromatic compounds) as dominant gas precursors of photochemical reaction in the atmosphere [118], Robinson and his co-workers [117] proposed a different model using a modified secondary organic aerosol formation framework. Their model accounts for both the gas-particle partitioning of semi-volatile organic aerosols and the oxidation of all low volatility gas-phase organic vapors to simulate the formation of secondary organic aerosols in the atmosphere. Secondary organic aerosol fractions in the revised model showed a substantially higher contribution to organic aerosols than in the traditional model, indicating the important role of semi-volatile primary organic aerosols in the formation of secondary organic aerosols in the atmosphere [119].

2.9 DIESEL EXHAUST ATMOSPHERIC AGING AND SIMULATIONS

Diesel particulate filters are used to remove DPMs from DE gas. Volatile and semi-volatile compounds can pass through diesel particulate filters and be released into the atmosphere where they are cooled, diluted and exposed to ozone, UV light and other gaseous compounds. These processes can lead to the production of new secondary particles. Yet, our knowledge concerning the chemical transformations of DE is very limited. There is currently very little in the literature that addresses the impact of atmospheric transformation on the health effects of inhaled DE [120]. To study the transformation of DE in the atmosphere, scientists use smog chambers or flow through reactors in which atmospheric conditions are simulated.

Avery large number of PhD theses and journal articles can be found on the subject of flow-through reactors or smog chambers, many of which are out of the scope of this research. Therefore, this thesis will consider those reactor and chamber studies which are related to combustion particles, and specially DE particles.

2.9.1 PAM chambers Potential Aerosol Mass (PAM) chambers or flow-through reactors are enclosures in which the aerosol of interest is exposed to high concentration of

40 Chapter 2: Literature Review

reactants for a relatively short period of time (1 to 2 minutes). The amount of time the aerosol stays in the reactor is called the residence (treatment) time, which can be easily calculated as:

( ) ( ) = 3 𝑉𝑉(𝑚𝑚 ) 𝜏𝜏 𝑠𝑠 3 𝑚𝑚 𝑞𝑞 Where, is used as the variable for residence𝑠𝑠 time, is the capacity of the system, and 𝜏𝜏is the flow for the system. 𝑉𝑉 Employment𝑞𝑞 of PAM chambers is associated with high oxidant concentrations and short exposure times, mimicking a few hours to a few days of real-time atmospheric oxidation. The short residence time of PAM chambers further enables better control of the oxidant concentrations and minimizes wall losses, which could be significant in smog chambers [121]. Also, the response time in PAM chambers is far less than that of smog chambers, making it easier to observe the effect of changing a control variable. Moreover, experiments carried out using PAM chambers tend to be more repeatable and more reproducible [121]. Below, a few studies applying such chambers are reviewed.

In a recent study, McWhinney [101] studied the effect of ozone oxidation on the redox-cycling activity exhaust gases of a two-stroke engine in a flow reactor. Exhaust from a two-stroke Homelite gasoline engine was aged by passing it through two glass flow tubes (0.8 m in length and 0.04 m in diameter), where the exhaust was exposed to ozone. The residence time of the tubes was approximately 60 seconds in total, and the authors claimed that the exposure in the tubes was equivalent to an exposure of 60 ppb of ozone for two to eight hours in atmospheric conditions. After aging, and making use of a dithiothreitol (DDT) assay, it was concluded that the redox-cycling activity of particles reached higher levels, which provides experimental evidence to support the hypothesis that the toxicity of engine combustion particles due to redox-cycling may increase as they age.

In another study, fresh and aged exhaust from a vehicle was compared in terms of health effects [122]. Diluted DE was passed through a rectangular, 0.575 m3 photochemical reaction chamber equipped with forty eight UVA lamps. The residence time in the reactor varied from 20 to 60 minutes. Once the UV lamps were turned on, without the addition of any accelerant, photochemical reactions were

Chapter 2: Literature Review 41

initiated, resulting in a rapid NO depletion, NO2 formation and O3 accumulation. A stable SOA concentration of approximately 40 μg.m−3, at a chamber mean residence time of 30 min was achieved. Oxidative capacity tests were performed for particles sampled from fresh and aged exhaust, based on which a significant increase in oxidative capacity of aged particles was reported.

To develop and characterize an exposure generation system, Papapostolou et al. carried out some experiments on primary and secondary organic aerosols in a reactor [123]. They developed methods for photochemical oxidation of traffic primary emissions to produce mixtures of primary and/or secondary particles suitable for animal exposure. Their system utilized a 9 m3 reactor. Three different modes were tested:

• Diluted exhaust containing primary particles passing through the reactor without adding ozone and turning on the UV lamps;

• Diluted exhaust containing primary particles passing through the reactor in the presence of ozone and UV light;

• Filtered, diluted exhaust gas passing through the chamber with ozone and UV light.

In the end, the authors concluded that, in the presence of primary particles, a longer exposure resulted in higher secondary organic aerosol yield (Figure 2.6 and Figure 2 .7). While, in the second mode, higher baseline plenum primary particle mass concentration caused lower secondary organic aerosols, and in the third mode, higher plenum gas concentration resulted in higher secondary organic aerosol yield. Figure 2 .8 indicates initial nucleation with particle sizes around 50 nm, then particle growth to particle sizes around 200 nm. Note that particle formation mechanisms in the second and third modes are different. In the second mode, secondary PM results from the condensation of semi-volatile gases onto existing particle surfaces. Figure 2 .9 shows particle count concentration as a function of irradiation time, in experiments with both filtered and unfiltered primary particles. When primary particles are present, the number concentration remains fairly stable during the 24-h irradiation. In contrast, when primary particles are absent, particle count peaks rapidly then decays to a fairly stable level.

42 Chapter 2: Literature Review

Figure 2.6 Evolution of secondary organic matter in presence of primary particles [122].

Figure 2.7 Evolution of secondary particles in the absence of primary particles [123].

Chapter 2: Literature Review 43

Figure 2.8 Evolution of size distribution while aging [123].

Figure 2.9 Particle count in the presence and absence of primary particles [123].

Lambe et al. [124] characterized and compared two photo-oxidation flow reactors in terms of heterogeneous oxidation and secondary organic aerosol formation. A schematic of the experimental setup can be seen in Figure 2 .10. One of the reactors was the Toronto Photo-Oxidation Tube or TPOT and the other one was called the Potential Aerosol Mass reactor or PAM, with residence times of 110 and 240 seconds, respectively. In both reactors, a 254 nm UV light lamp was placed in

44 Chapter 2: Literature Review

the center of the tube for photo-oxidation. The main difference between the flow tubes was the degree to which gases and aerosols interacted with reactor wall surfaces. The TPOT was developed for aerosol heterogeneous oxidation studies where uniform OH exposure was more important than minimizing wall losses, while the PAM was developed with a separate flow to minimize wall interactions, as is important for SOA formation studies. Based on the presented data, they concluded that a high degree of oxidation can be achieved in both reactors; however they also showed that secondary organic aerosol yields from such reactors must be viewed with caution, given the important role that wall losses may be playing.

Figure 2.10 Schematic of experimental setup [124].

Keller and Burtscher presented a continuous photo-oxidation flow reactor as a treatment system for real time monitoring of the potential production of secondary organic aerosols from wood combustion [125]. The reactor was composed of three tubes. Gases from wood combustion were exposed to UVC and UVA in two first tubes, and the aged mixture was cooled down in the third tube to facilitate the condensation process. They showed that a residence time of only a few seconds is enough for secondary particle formation to start. In order to test the performance of the reactor, they used emission-relevant concentrations of known precursor molecules, with the goal of producing SOA using a short residence time. As can be seen in Figure 2 .11, the presence of ozone and UV light intensity plays an important role in secondary organic aerosol formation. They found that temperature, UV light wavelength and intensity, and relative humidity to be the most significant influential factors on secondary organic aerosol formation. They compared measurements from

Chapter 2: Literature Review 45

wood smoke aged in the reactor with previously published data from a smog chamber. A ratio was calculated to correlate secondary organic aerosol production to emitted CO2. The expressed ratio was found to be in close agreement with the results from the chamber. They showed that the reactor can be used as a screening tool for emission control measurements.

Figure 2.11 SOA formation in the reactor from known precursors as a function of UV intensity [124]

There are a few laboratories which use flow reactors to investigate the atmospheric aging process of particles from biogenic and anthropogenic sources. Some of the most cited ones are as follow:

QUAREC [126]: The Physical Chemistry Department of the University of Wuppertal has developed a reactor which is the largest of its kind. It consists of two cylinders, with a total length of 6.2 m and a diameter of 0.47 m. The reactor is temperature regulated and is equipped with UV lamps (with different wave-length). There are a number of publications investigating secondary organic aerosols which have used QUAREC, however none of them were related to combustion particles.

LOTASC [127]: This reactor is located at the University of Bayreuth. It is a cylindrical reactor with a diameter of 1 m and length of 4 m. Inner temperature and humidity are controllable in the range -25 to 25 degrees of Celsius and 2 to 80%, respectively.

46 Chapter 2: Literature Review

CRAC1 [128]: CRAC1 is a custom-built simulation cylindrical reactor designed for laboratory-based investigations of atmospheric processes, such as the oxidation of volatile organic compounds and secondary organic aerosol formation at ambient temperatures. Its diameter and length are 1.1 m and 4.1 m, respectively. Photo-oxidation reactions are induced by the injection of ozone and UV light exposure. Utilizing this reactor, there are several papers which study gas/particle partitioning and the photo-oxidization of volatile and semi-volatile organic compounds.

HIRAC [129]: Abbreviated from Highly Instrumented Reactor for Atmospheric Chemistry, HIRAC is a flow reactor in the School of Chemistry, University of Leeds, UK. It is made of stainless steel with dimensions of 1.2 m in diameter and 2.0 m in length. Several different types of UV lamps are installed inside the chamber allowing different photochemical exposure. The temperature of the chamber can be regulated within the range -70 to 70 degrees of Celsius. It has been used to study the effect of photo-oxidation on volatile organic compounds [130].

2.9.2 Smog chambers To study the transformation of DE in the atmosphere, scientists use smog chambers or flow through reactors in which atmospheric conditions are simulated. A smog chamber is defined as a large confined volume in which sunlight or simulated sunlight is allowed to irradiate air mixtures of atmospheric trace gases hydrocarbons, nitrogen oxides, ozone and so on, which undergo photochemical processing [131].

In theory, these chambers allow for the controlled study of complex reactions which occur in the atmosphere. Using these chambers allows researchers to investigate DE transformation in a controlled environment. In smog chambers, different chemicals such as volatile organics, nitrogen oxides (NOx), sulphur dioxide, ozone, OH radicals etc, are introduced and mixed with DE and aged over a certain period of time in proximity to sunlight or an artificial UV source. Then photo- oxidized particles are studied in terms of physical and chemical properties, such as volatility, hygroscopicity, size, number distribution and concentration etc. There is a limited amount of literature focusing on the aging and transformation of DPM in the atmosphere and even fewer studies on aged DPM from alternative fuelled diesel engines.

Chapter 2: Literature Review 47

Atkinson et al. used a smog chamber to study the effect of irradiation on mixtures of NOx, toluene, benzaldehyde, air and cresol back in 1980 [132]. Mixtures were irradiated at 303 K in a 5.8 m3 Teflon-lined smog chamber, with temperature, humidity, light intensity and spectral distribution being monitored as a function of time, as well as the concentrations of O3, NO, NO2, toluene, formaldehyde, benzaldehyde, o-cresol, m-nitro-toluene, and methyl nitrate. They observed the Cresol–NOx–air mixture to be less reactive than the toluene–NOx–air mixture, in terms of O3 formation and NO to NO2 conversion rates. They developed a model and concluded that it under-estimated nitro-toluene yields. Eventually they suggested that more experimental work should be done in order to validate current aromatic photo- oxidation mechanisms.

Izumi et al. [133] investigated photochemical aerosol formation by putting a mixture of aromatic hydrocarbons (AHCs) and NOx in a 4 m3 chamber. The evolution of aerosols was monitored by the measurement of number and volume concentrations. It was concluded that the yields for toluene were the largest, followed by ethyl-benzene, o- and m- ethyl-toluene. In their research, mixtures of volatile organic compounds, NOx and air were irradiated by natural sunlight in the presence and the absence of DE. They used exhaust gases emitted from five different diesel fuel formulations. Experiments were conducted under the same initial conditions for volatile organic compounds and NOx.

Geiger et al. investigated the effect of DE on photo-smog formation in the EUPHORE smog chamber [134]. EUPHORE, the European PHOto REactor, consists of two twin 200 m3 hemispherical Teflon outdoor atmospheric simulation chambers. The addition of DE to a well-defined simple VOC mixture caused a significant increase in ozone formation after irradiation, in comparison to smog experiments with a similar VOC/NOx ratio in the absence of exhaust gas. Based on a sensitivity test it was concluded that ozone formation was not dependent on the formulation of the diesel fuel. The results from the chamber tests were compared to the results of a chemical box model which showed good agreement.

In 2002, Kleindienst examined secondary organic aerosol formation from an automobile exhaust by irradiating exhaust gas in a chamber [135]. Their aim was to investigate the extent to which photochemical oxidation products of auto-mobile exhaust hydrocarbons contribute to secondary organic aerosol formation, as well as

48 Chapter 2: Literature Review

the accuracy of recently developed models for predicting secondary organic aerosol formation. They indicated that carbonyl groups dominated the aerosol and the hygroscopicity of aerosols was minor at relative humilities <70%. The results were compared to an aerosol model and it was concluded that 75-58% of fine PM was product of aromatic precursors.

Outdoor chamber experiments were conducted by Lee et al. in 2003, to investigate secondary organic aerosol production from the photo-oxidation of freshly emitted diesel soot and α-pinene [4]. Experiments were carried out in an outdoor dual Teflon film chamber system (135 m3 per chamber) at the University of North Carolina. All photochemical reactions of α-pinene in the presence of DE were carried out under natural sunlight. Diesel emissions were generated from a 1980 Mercedes Benz 300SD engine running under idling conditions. As shown in Figure 2 .12, photochemical reactions in the presence of DE without α-pinene increased particle mass by a factor of 2 compared to the initial diesel soot particle mass that was injected into the chamber. Following the introduction of α-pinene to the chamber, particle mass increased by a factor of 4 from the initial diesel soot particle mass. However, because of extra particle formation from the photo-oxidation of DE and cross photochemical reactions between DE component and α-pinene products, the yield from α-pinene may possibly be greater than just the observed increase when α- pinene was present. The partitioning coefficients of the major carbonyl products were much higher in diesel soot particles than predicted by theory. The high particle concentration of carbonyl products on diesel soot particles could be explained by heterogeneous acid-catalyzed reactions.

Chapter 2: Literature Review 49

Figure 2.12 Total particle mass concentrations in the chamber (a) with α-pinene addition and (b) without α-pinene addition [4].

Zielinska and Sammy used the EUPHORE 200 m3 environmental simulation chamber to investigate the transformation of DE in the atmosphere. In that research,

DE was aged in the chamber and different chemicals, such as N2O5 (to investigate NOx influence on DE), HCHO (to study reaction of OH radicals) and toluene, were introduced into the chamber. They concluded that OH radicals were of the most importance, in terms of chemical reactions with DE constituents, followed by ozone,

NO3 radicals and light [120, 136].

50 Chapter 2: Literature Review

Leskinen et al. studied the aging of wood chip combustion aerosols in a 143 m3 outdoor environmental smog chamber located at the University of Kuopio [15]. The chamber had a surface area to volume ratio to minimize wall losses, and the walls and ceiling was made of Teflon. To check the effect of aging on wood combustion particles, they used an SMPS, a V-TDMA and a thermal-optical carbon analyzer. The volatility and carbon analyzes were carried out before and after aging (17–24 hours), and at one intermediate point. Volatility tests showed that while the diameter reduction of 100 nm freshly emitted was between 6 and 10%, after being heated to 360 °C for particles aged for 24 h, a 74 to 86% decrease in particle size at 360 °C was observed (Figure 2 .13). Occasionally, new particle formation and growth were observed the following day. The new particles were found to be composed mainly of volatile material. They suggest the V-TDMA method to be an excellent tool for separating altered primary particles from secondary particles formed by nucleation, based on the differences in the volatilities of these two particle types.

Leskinen et al. studied aging of DE in the latter aforementioned smog chamber [96]. The same experimental procedure and instrumentation were used to check the change in volatility of diesel particles and new particle formation after aging in the chamber. During the experiment, particle size distribution, ozone and nitrogen oxide concentration, temperature, relative humidity, and total solar and total UV radiation in the chamber were monitored. As shown in Figure 2 .14, fresh particles from the diesel engine contained 37–45% semi-volatile compounds in idle mode, and 10–24% with a 9 kW load. After 24 h of aging in the chamber, the semi-volatile volumetric fraction at 360 °C for 50 nm particles increased to 57%, and the volatile volume fraction for 100 nm particles was 99%. They suggests that the particles in the 24h aged aerosol for this size class we no longer primary particles but particles that were formed in the chamber through nucleation and subsequent growth.

Chapter 2: Literature Review 51

Figure 2.13 The change in size of (a) fresh, (b) 8–12h aged, and (c) 24h aged 100nm mono-disperse aerosol as a function of heating temperature [96].

Figure 2.14 Effect of aging on volatility of particles for (a) idle, (b) 9kw load [96].

52 Chapter 2: Literature Review

Samy and Zielinska investigated secondary organic aerosol production from a diesel engine while aging in the EUPHORE simulation chamber [136]. They measured secondary organic aerosol production in the presence of toluene, HCHO and N2O5, both in daylight and at night. They observed high levels of secondary organic aerosol production for all of the abovementioned oxidative agents, however it was concluded that the greatest production was due to toluene, followed by HCHO.

An environmental chamber (27 m3 flexible bag made of 125 μm Teflon® fluorocarbon film (FEP)) was used by the PSI group to age DE [137]. The bag is suspended in a temperature controlled wooden enclosure, where the temperature can be stabilized to 1°C within the range of 17 to 25 °C. Four xenon arc lamps are used as the light source and purified air is supplied by a pure air generation system. An ozone generator can provide ozone levels from ppb to ppm concentrations in the chamber, which can also be humidified with ultrapure water up to a relative humidity of 80 %. A particle generation system is available for seed particle experiments.

In recent research conducted in the PSI chamber [137], the effect of different after-treatment devices on primary emissions and secondary organic aerosols was studied. DE was collected from three vehicles with three different after-treatment devices. Then, primary emissions were measured. Afterwards, exhaust gas was guided into the chamber where it was mixed with ozone and exposed to UV light. Figure 2 .15 illustrates temporal evolution of the organic matter and black carbon for the vehicle without after treatment devices. The figure is not corrected for wall losses; nevertheless, a substantial amount of secondary organic matter is differentiable. Figure 2 .16 shows the secondary organic yields for DE without an after-treatment device, after being corrected for wall losses. Eventually, it was concluded that while the presence of a diesel oxidation catalyst has no significant effect on the emission factor of primary organic matter, it caused a substantial reduction in secondary organic matter. After raw DE was aged for 6 hours, almost 80% of the aerosol was found to be secondary.

Chapter 2: Literature Review 53

Figure 2.15 Temporal evolution of the organic matter, without after treatment, before wall loss correction [137].

Figure 2.16 Temporal evolution of the organic matter, without after treatment, after wall loss correction [137].

54 Chapter 2: Literature Review

In other piece of research, Weitkamp et al. investigated secondary organic aerosol formation from DE in a smog chamber [97]. The chamber was a 10 m3 Teflon bag suspended in a temperature controlled room. Eighty eight black light bulbs were used as the source of UV light. DE was collected from a single cylinder diesel generator under idle, 30% and 60% loads. It was discussed that photo- oxidation was as a result of light-initiated oxidation reactions including the formation of oxidants such as ozone and the hydroxyl radical Figure 2.17 shows that the initial aerosol mass was 58 μg/m3, all of which was primary DPM. At time zero, the black lights were turned on causing a rapid and substantial production of new particle mass. For the first hour, Figure 2 .17a shows that the suspended mass actually increased, indicating that the production rate was greater than the wall-loss rate. After about an hour, the production rate slowed, and the suspended mass decreased due to wall-losses. Figure 2 .17b shows that the median particle diameter increased throughout the experiment, consistent with condensational growth. Based on the AMS data it was indicated that the secondary aerosol formed in the chamber was organic and there was no sign of sulphate or nitrite production during photo- oxidation. After comparing the measured results to a secondary organic aerosol model, it was found that less than 10% of secondary particulate mass was explainable. They concluded that oxidation of a wide array of volatile and semi- volatile precursors, which are not accounted for in existing models, contributed to the huge discrepancy between the model and measured results.

Chapter 2: Literature Review 55

Figure 2.17 Time-series of (a) measured suspended mass, (b) median diameter, and (c) wall-loss corrected mass measured during photo-oxidation of DE. [97]

Secondary aerosol formation from the photochemical aging of aircraft exhaust was studied in a smog chamber by Miracolo [92]. Exhaust gas from an in-use CFM56-2B gas turbine engine was sampled in a smog chamber, where it was exposed to sunlight and/or UV light to initiate photo-oxidation. The smog chamber was a 7 m3 Teflon bag suspended on an open metal frame. Black light UVA bulbs were used for UV exposure. An SMPS, V-TDMA, AMS and OC/EC analyzer were used to monitor the chemical and physical properties of particles. Separate tests were performed at different engine loads. After the UV lights were turned on, there was

56 Chapter 2: Literature Review

substantial secondary PM production. Figure 2 .18 shows a sharp increase in particle count due to nucleation, which happened for all other engine loads as well.

Chapter 2: Literature Review 57

Figure 2.18 Time series of (a) gas and (b) particle phase concentrations during photo-oxidation experiment at 4% load. Vertical grey bars are time periods when thermo-denuder was running [92].

Figure 2.19 Size distribution during aging: (a) after filling, (b) before UV lights turned on, (c) shortly after UV was on, (d) after 3 hours of aging [92].

58 Chapter 2: Literature Review

Figure 2.20 Measured changes in PM mass caused by photo-oxidation in different engine loads: (a) 4%, (b) 7%, (c) 30% and (d) 85% [92]

According to AMS and V-TDMA data, while at 4% load, secondary PM was dominated by organic compounds, while at higher loads, secondary particles were mainly composed of secondary sulphate (Figure 2 .20). Photo-oxidation created substantial secondary PM, greatly exceeding the primary aerosol emissions at each engine load after an hour or less of aging under typical summertime conditions. After several hours of photo-oxidation, the ratio of secondary-to-primary PM mass was on average 35, 17, 60, and 2.7 for the 4, 7, 30, and 85 % load experiments, respectively. Results from the experiments were compared against the results from a model. It was concluded that the model under-predicted the amount of secondary PM and the oxidation of traditional secondary particulate precursors could not explain the observed secondary particulate formation. They suggested that the under-prediction was due to the significant amount of low-volatility organics emitted by the engine, which was a significant source of secondary PM precursors, a fact that was overlooked in the model.

2.10 HEALTH EFFECTS

Over the last decade, study of the epidemiology of human exposure to ambient PM has clearly established a statistically significant correlation between levels of fine particle exposure and health effects [138, 139]. Studies have now been carried

Chapter 2: Literature Review 59

out in several countries and the results have consistently shown a significant impact on human health that is attributable to ambient particles [140-143]. Within such particles, the fine PM fraction (PM2.5 - smaller than 2.5 µm) has been linked to a range of respiratory and cardiovascular health problems, because of their long lifetime in the air (small particles have very long settling times) and respiratory deposition characteristics (they deposit deeper in the lungs). A number of epidemiological studies have shown that PM2.5 is correlated with severe health effects, including enhanced mortality [144].

As noted, the semi-volatile fraction of DPM has been shown to highly correlate with the overall toxicological potency of PM [34, 50, 85, 101, 114]. The adverse health effects of such compounds is mainly because of their ability to produce ROS, resulting in oxidative stress in the lungs [113]. ROS are chemically reactive molecules that contain oxygen. Increased levels of ROS in an environment can cause significant damage to cell structure, a process called oxidative stress. ROS are - - suspected to be responsible for the production of O2 and OH radicals, which are very reactive with organic tissue [145]. Measurement of OP, expressed through ROS concentration, can be used as a good estimate for its reactivity and toxicity. Based on the data provided in the literature so far, it remains unclear which chemical species contribute to the measured redox potential and overall toxicity. Several studies in this area [83, 108, 146] showed that the organic fraction, more precisely the semi-volatile component of PM, correlated well with the measured OP. Stevanovic et al. [146] further showed that the oxygenated fraction of the semi-volatile component of DPM was most responsible for the OP of DPM. Although semi-volatile organic compounds make up only a small part of DPM, they are believed to be accountable for the majority of DPM’s toxicity [112]. In recent publications, it was concluded that there is a direct relationship between the organic content of DPM and its ROS concentration, and hence its toxicity [114, 115]. Dilution and cooling conditions, whether in laboratories or the atmosphere, are key factors determining the degree to which these semi-volatile species are partitioned in the particulate phase and therefore, the degree to which they directly affect overall PM toxicity [91, 95]. The mechanisms of PM related adverse health effects are still not completely understood, but an hypothesis under investigation is that many of these effects

60 Chapter 2: Literature Review

originated from oxidative stress, initiated by the formation of ROS on the surface of and within target cells. Cumulative epidemiological and experimental data support the association of adverse health effects with cellular oxidative stress, including the ability of PM to induce pro-inflammatory effects in the nose, lung and cardiovascular system. This is because high levels of ROS cause a change in the redox status of the cell and its surrounding environment, thereby triggering a cascade of events associated with inflammation and, at higher concentrations, apoptosis. Further, it has also been found that the redox activity of diesel particles can be enhanced in aging process [101]. Since chemical reactions generally occur faster in the gaseous phase, the oxidation of semi-volatile organic compound vapors yields secondary organic aerosols, shifting them again into the particulate phase [91]. Thus, the cycling of semi-volatile organic compounds between the gas and particle phase ultimately leads to a net increase in oxygenated organic aerosols [147], with modified chemical and likely, toxicological, characteristics [109]. To investigate the atmospheric influence on DPM, as well as to scrutinize the potential adverse health effects related to such particles, researchers make use of a variety of instruments and techniques. Also, as mentioned earlier, transformation and dynamics of aerosols have been under extensive investigation for years. Only recently, due to the application of new technologies, have scientists been able to push and extend the boundaries of aerosol science [87]. In the following section, the most common instruments and techniques currently used to investigate the DPM and DE will be reviewed.

2.11 INSTRUMENTS AND TECHNIQUES

This section provides information on a variety of instruments and techniques, as well as related studies, which are currently in-use in the field of aerosol science. Especially, those instruments which are mostly important while studying DPM and DE and their transformations in the atmosphere.

2.11.1 Offline methods Offline measurement methods involve the collection of aerosol particles on substrates (e.g. filters, impactors and denuders) and the transfer of the collected samples to a laboratory for extraction and chemical analysis. Atomic absorption,

Chapter 2: Literature Review 61

scanning electron microscopy (SEM), high performance liquid chromatography (HPLC), gas chromatography (GC), ion chromatography (IC), proton nuclear magnetic resonance (PNMR), secondary ion mass spectrometry (SIMS), inductively coupled plasma mass spectrometry (ICP-MS), Inductively Coupled Plasma Atomic Emission Spectrometry (ICP-AES) and laser microprobe mass spectrometry are the main analytical techniques available for the offline chemical analysis of collected particles [148, 149].

2.11.2 Dilution system The dilution system is used to cool down and dilute exhaust gas so the aerosol is more analogous to diesel exhaust being released into the atmosphere. Also as many of the instruments used in the experiments had a limited operational range; the concentrations of diesel exhaust gaseous compounds as well as diesel particulate matter had to be reduced so that they can be handled by the instrumentations. The dilution system comprised two ejector Dekati diluters and a dilution tunnel. Dilution ratios of the Dekati systems were fixed at 10; while the dilution ratio of the dilution tunnel was variable between 0 and 50. Combining these three diluters, together with some modifications enabled us to provide variable dilution ratios between 10 and 1000.

2.11.3 CPC The Condensation Particle Counter (CPC) is an instrument which measures the number concentration of an aerosol of interest. The main challenge in the real-time counting of ultrafine particles is the fact that they are too small to be detected by conventional optics. Therefore, they need to be grown to a detectable size. Condensation techniques have been used for this purpose to increase a particles size by condensing vapor on its surface, leading to particle growth. These particles can then be counted using light scattering techniques. Conventional CPCs use water or Butanol as the working fluid to condense on the particles surface. In Butanol-based CPCs, the saturation chamber is heated and the growth chamber is cooler. On the contrary, water-based CPCs have a cool conditioner region and a hot condensation chamber, which is mainly due to the very high vapor diffusivity of water. As an example, Figure 2 .21 is a schematic of the different parts of a TSI 3010 Butanol- based CPC [1, 22, 150].

62 Chapter 2: Literature Review

Figure 2.21 Schematic of TSI 3010 Butanol-based CPC [150].

2.11.4 DMA Differential mobility analyzers (DMAs) are aerosol instruments widely used to measure size distributions and extract submicron mono-disperse particles from poly- disperse aerosols [151, 152]. The basic configuration of a DMA is illustrated in Figure 2 .22. DMAs are capable of segregating a mono-disperse aerosol flow out of a poly-disperse aerosol which would then be used in a variety of analytical instruments (as explained later).

The DMA is composed of a cylinder, with a charged rod at the center, a particle-free-flow flows through the DMA and is called 'sheath' air. It is very important that this flow is laminar. The particle flow is injected along the outside edge of the DMA and particles with a positive charge move across the sheath flow towards the central rod, at a rate determined by their electrical mobility. Particles of a given mobility exit through the sample slit at the top of the DMA, while all other particles exit with the exhaust flow.

Chapter 2: Literature Review 63

Figure 2.22 Schematic diagram of cylindrical DMA by Knutson and Whitby [153].

2.11.5 SMPS The Scanning Mobility Particle Sizer (SMPS) spectrometer gives the size distribution of particles within an aerosol. The SMPS theory of operation is based on the principle of the mobility of a charged particle in an electric field. Particles entering the system are neutralized (using a radioactive source) such that they have a Fuchs equilibrium charge distribution. They then enter a Differential Mobility Analyzer (DMA) where the aerosol is classified according to electrical mobility, with only particles of a narrow range of mobility exiting through the output slit. This mono-disperse distribution then goes into a CPC which determines the particle concentration at that size [150].

2.11.6 TDMA The Tandem Differential Mobility Analyzer (TDMA) is useful for studying phenomena that lead to size changes in submicron aerosol particles. Evaporation, condensation and chemical reactions within airborne droplets are examples of such phenomena. With this method, aerosols of a known size are selected by a DMA. These mono-disperse aerosols then undergo processes that result in growth or shrinkage and the extent of growth or shrinkage is determined by a second DMA [154].

64 Chapter 2: Literature Review

2.11.7 V-TDMA Volatility Tandem Differential Mobility Analyzer (V-TDMA) is a TDMA technique where a thermo-denuder is placed between the two DMAs. The DMA selects a mono-disperse flow of the aerosol, which is directed through the thermo- denuder. In the thermo-denuder, the aerosol is heated and the volatile fraction of the particles is evaporated. Next, the aerosol is guided into the second DMA, where the size distribution of the heat treated aerosol is obtained. Based on ther two size distributions, the volatile fraction in particles of a certain size is determined. The same technique is used to check the hygroscopicity of particles and it is referred to as Hygroscopicity Tandem Differential Mobility Analyzer (H-TDMA).

2.11.8 AMS Aerosol mass spectrometers (AMS) are the most significant development in aerosol measurement techniques over the past 20 years. They have been shown to be capable of providing valuable information about the sources, chemical composition and transformation of aerosol particles [1, 87]. In the AMS, particles are sampled and focused into a narrow beam of light through a supersonic expansion inlet nozzle which then enters a particle sizing chamber. The sizing chamber also conducts time- of-flight particle velocity measurements that can be correlated with particle density and size. After passing through the chamber, particles are vaporized and ionized before being analyzed by a mass spectrometer.

2.11.9 PM mass concentration Mass concentration can either be measured directly by gravimetric methods (e.g. TEOM), or by the measurement of another property (e.g. light scattering in DustTrak), which can then be related to particle mass.

The TEOM is a real-time instrument used to measure particle mass concentration. It consists of a tapered element made of glass with a filter on the top. The element oscillates with its Eigen frequency and when mass is collected on the filter the weight of the element changes and so does the frequency. By measuring the difference in the oscillation frequency of the element the mass concentration can be determined.

Chapter 2: Literature Review 65

2.11.10 ROS measurement ROS levels can be measured using both online and offline methods. This study uses the BPEA molecular probe (bis(phenylethynyl) anthracene-nitroxide), which is applied in-situ to assess the OP of DE. Samples are collected by bubbling the aerosol through an impinger which contains a 4 μM BPEA solution (using an AR grade dimethylsulphoxide as a solvent). More details on the ROS sampling methodology, the theory behind its application and proof of concept, in the case of various combustion sources, can be found in studies by Miljevic et al. and Stevanovic et al. [114, 155-158]. Also, a variety of techniques for ROS measurement were reviewed in a study by Hedayat et al. [159].

2.12 GAPS IN KNOWLEDGE

The literature review identified the following gaps in knowledge:

• The effect of diesel fuel molecular structure on the performance and regulated emission characteristics of diesel engines.

• The effect of chemical composition and molecular structure of diesel fuel on the physical and chemical properties of DPM should be studied in more detail.

• The relationship between fuel chemical composition and organic content of DPM, as well as the relationship between the volatility of DPM and its toxicity need to be investigated.

• There is little information on the effect of dilution and exposure to oxidative agents on DE from alternative diesel fuels.

• There are a very few studies regarding the potential toxicity of biodiesel PM after being aged in the atmosphere.

• There is no data on the aging of DE in PAM chambers, in order to check the ultimate potential of DE in SOA generation.

2.13 REFERENCES

1. McMurry, P.H., A Review of Atmospheric Aerosol Measurements. Atmospheric Environment, 2000. 34(12-14): p. 1959-1999. 2. Kumar, P., Robins, A., Vardoulakis, S., and Britter, R., A Review of the Characteristics of Nanoparticles in the Urban Atmosphere and the Prospects

66 Chapter 2: Literature Review

for Developing Regulatory Controls. Atmospheric Environment, 2010. 44(39): p. 5035-5052. 3. Kittelson, D.B., Engines and Nanoparticles: A Review. Journal of Aerosol Science, 1998. 29(5-6): p. 575-588. 4. Lee, S., Jang, M., and Kamens, R.M., Soa Formation from the Photooxidation of Α-Pinene in the Presence of Freshly Emitted Diesel Soot Exhaust. Atmospheric Environment, 2004. 38(16): p. 2597-2605. 5. Hinds, W.C., Aerosol Technology: Properties, Behavior, and Measurement of Airborne Particles. 1982. 6. Seinfeld, J.H. and Pandis, S.N., Atmospheric Chemistry and Physics from Air Pollution to Climate Change. Atmospheric Chemistry and Physics, John Wiley & Sons, New York, 1998. 1326. 7. Burtscher, H., Physical Characterization of Particulate Emissions from Diesel Engines: A Review. Journal of Aerosol Science, 2005. 36(7): p. 896- 932. 8. Kittelson, D., Watts, W., and Johnson, J., Diesel Aerosol Sampling Methodology–Crc E-43. Final report, Coordinating Research Council, 2002. 9. Jung, H., Kittelson, D.B., and Zachariah, M.R., The Influence of Engine Lubricating Oil on Diesel Nanoparticle Emissions and Kinetics of Oxidation. 2003. 10. Bukowiecki, N., Kittelson, D., Watts, W., Burtscher, H., Weingartner, E., and Baltensperger, U., Real-Time Characterization of Ultrafine and Accumulation Mode Particles in Ambient Combustion Aerosols. Journal of Aerosol Science, 2002. 33(8): p. 1139-1154. 11. Bardouki, H., Liakakou, H., Economou, C., Sciare, J., Smolik, J., Zdimal, V., Eleftheriadis, K., Lazaridis, M., Dye, C., and Mihalopoulos, N., Chemical Composition of Size-Resolved Atmospheric Aerosols in the Eastern Mediterranean During Summer and Winter. Atmospheric Environment, 2003. 37(2): p. 195-208. 12. Lingard, J.J.N., Agus, E.L., Young, D.T., Andrews, G.E., and Tomlin, A.S., Observations of Urban Airborne Particle Number Concentrations During Rush-Hour Conditions: Analysis of the Number Based Size Distributions and Modal Parameters. Journal of Environmental Monitoring, 2006. 8(12): p. 1203-1218. 13. Kulmala, M., Vehkamaki, H., Petaja, T., Dal Maso, M., Lauri, A., Kerminen, V.M., Birmili, W., and McMurry, P.H., Formation and Growth Rates of Ultrafine Atmospheric Particles: A Review of Observations. Journal of Aerosol Science, 2004. 35(2): p. 143-176. 14. EPA. Epa's Terms of Environment Glossary, Abbreviations, and Acronyms. 2009 08/02/ 2009]; Available from: http://cfr.vlex.com/vid/19784887. 15. Leskinen, A.P., Jokiniemi, J.K., and Lehtinen, K.E.J., Characterization of Aging Wood Chip Combustion Aerosol in an Environmental Chamber. Atmospheric Environment, 2007. 41(17): p. 3713-3721. 16. D'Anna, A., Combustion-Formed Nanoparticles. Proceedings of the Combustion Institute, 2009. 32: p. 593-613. 17. Maricq, M.M., Chemical Characterization of Particulate Emissions from Diesel Engines: A Review. Journal of Aerosol Science, 2007. 38(11): p. 1079- 1118. 18. Shipman, D. Basic Concepts in Environmental Sciences 2010 30/01/2010]; Available from: http://www.epa.gov/apti/bces/index.htm.

Chapter 2: Literature Review 67

19. Atkinson, R. and Arey, J., Atmospheric Degradation of Volatile Organic Compounds. Chemical Reviews, 2003. 103(12): p. 4605-4638. 20. Chirico, R., DeCarlo, P.F., Heringa, M.F., Tritscher, T., Richter, R., Prevot, A.S.H., Dommen, J., Weingartner, E., Wehrle, G., Gysel, M., Laborde, M., and Baltensperger, U., Impact of Aftertreatment Devices on Primary Emissions and Secondary Organic Aerosol Formation Potential from in-Use Diesel Vehicles: Results from Smog Chamber Experiments. Atmospheric Chemistry and Physics, 2010. 10(23): p. 11545-11563. 21. Dobbins, R.A., Hydrocarbon Nanoparticles Formed in Flames and Diesel Engines. Aerosol Science and Technology, 2007. 41(5): p. 485-496. 22. Hinds, W.C., Aerosol Technology: Properties, Behavior, and Measurement of Airborne Particles. New York, Wiley-Interscience, 1982. 442 p., 1982. 1. 23. Seinfeld, J.H. and Pandis, S.N., Atmospheric Chemistry and Physics: From Air Pollution to Climate Change. 2012: John Wiley & Sons. 24. Odum, J.R., Hoffmann, T., Bowman, F., Collins, D., Flagan, R.C., and Seinfeld, J.H., Gas/Particle Partitioning and Secondary Organic Aerosol Yields. Environmental Science & Technology, 1996. 30(8): p. 2580-2585. 25. Vaden, T.D., Imre, D., Beránek, J., Shrivastava, M., and Zelenyuk, A., Evaporation Kinetics and Phase of Laboratory and Ambient Secondary Organic Aerosol. Proceedings of the National Academy of Sciences, 2011. 108(6): p. 2190-2195. 26. Donahue, N.M., Henry, K.M., Mentel, T.F., Kiendler-Scharr, A., Spindler, C., Bohn, B., Brauers, T., Dorn, H.P., Fuchs, H., Tillmann, R., Wahner, A., Saathoff, H., Naumann, K.-H., Möhler, O., Leisner, T., Müller, L., Reinnig, M.-C., Hoffmann, T., Salo, K., Hallquist, M., Frosch, M., Bilde, M., Tritscher, T., Barmet, P., Praplan, A.P., DeCarlo, P.F., Dommen, J., Prévôt, A.S.H., and Baltensperger, U., Aging of Biogenic Secondary Organic Aerosol Via Gas-Phase Oh Radical Reactions. Proceedings of the National Academy of Sciences, 2012. 109(34): p. 13503-13508. 27. Begat, P., Young, P.M., Edge, S., Kaerger, J.S., and Price, R., The Effect of Mechanical Processing on Surface Stability of Pharmaceutical Powders: Visualization by Atomic Force Microscopy. Journal of pharmaceutical sciences, 2003. 92(3): p. 611-620. 28. Akselsson, R., Bohgard, M., Gudmundsson, A., Hansson, H.C., Martinsson, B., and Svenningsson, B., Aerosoler. 1994, Lunds Tekniska Högskola and NOSA. 29. DeCarlo, P.F., Slowik, J.G., Worsnop, D.R., Davidovits, P., and Jimenez, J.L., Particle Morphology and Density Characterization by Combined Mobility and Aerodynamic Diameter Measurements. Part 1: Theory. Aerosol Science and Technology, 2004. 38(12): p. 1185-1205. 30. Kulkarni, P., Baron, P.A., and Willeke, K., Aerosol Measurement: Principles, Techniques, and Applications. 2011: John Wiley & Sons. 31. Flagan, R., Electrical Techniques. Aerosol Measurement: Principles, Techniques, and Applications, 2001: p. 537-568. 32. Oberdörster, G., Celein, R.M., Ferin, J., and Weiss, B., Association of Particulate Air Pollution and Acute Mortality: Involvement of Ultrafine Particles? Inhalation Toxicology, 1995. 7(1): p. 111-124. 33. Seaton, A., Godden, D., MacNee, W., and Donaldson, K., Particulate Air Pollution and Acute Health Effects. The Lancet, 1995. 345(8943): p. 176-178.

68 Chapter 2: Literature Review

34. Giechaskiel, B., Alfoldy, B., and Drossinos, Y., A Metric for Health Effects Studies of Diesel Exhaust Particles. Journal of Aerosol Science, 2009. 40(8): p. 639-651. 35. Rhodes, M., Front Matter. 2008: Wiley Online Library. 36. Hering, S.V., Impactors, Cyclones, and Other Inertial and Gravitational Collectors. Air sampling instruments for evaluation of atmospheric contaminants. 8th ed. Cohen and Hering, editors. Cincinnati (OH): ACGIH, 1995. 37. Odowd, C.D., Smith, M.H., Consterdine, I.E., and Lowe, J.A., Marine Aerosol, Sea-Salt, and the Marine Sulphur Cycle: A Short Review. Atmospheric Environment, 1997. 31(1): p. 73-80. 38. Remédios, C., Rosário, F., and Bastos, V., Environmental Nanoparticles Interactions with Plants: Morphological, Physiological, and Genotoxic Aspects. Journal of Botany, 2012. 2012. 39. Pey, J., Querol, X., Alastuey, A., Rodríguez, S., Putaud, J.P., and Van Dingenen, R., Source Apportionment of Urban Fine and Ultra-Fine Particle Number Concentration in a Western Mediterranean City. Atmospheric Environment, 2009. 43(29): p. 4407-4415. 40. Quinn, P., Anderson, T., and Bates, T., Physics and Chemistry of Aerosols. 2003. 41. Zhu, R., Cheung, C.S., and Huang, Z., Particulate Emission Characteristics of a Compression Ignition Engine Fueled with Diesel-Dmc Blends. Aerosol Science and Technology, 2011. 45(2): p. 137-147. 42. Zhu, Y., Hinds, W.C., Kim, S., Shen, S., and Sioutas, C., Study of Ultrafine Particles near a Major Highway with Heavy-Duty Diesel Traffic. Atmospheric Environment, 2002. 36(27): p. 4323-4335. 43. Grose, M., Sakurai, H., Savstrom, J., Stolzenburg, M.R., Watts, W.F., Morgan, C.G., Murray, I.P., Twigg, M.V., Kittelson, D.B., and McMurry, P.H., Chemical and Physical Properties of Ultrafine Diesel Exhaust Particles Sampled Downstream of a Catalytic Trap. Environmental Science & Technology, 2006. 40(17): p. 5502-5507. 44. Silverman, D.T., Samanic, C.M., Lubin, J.H., Blair, A.E., Stewart, P.A., Vermeulen, R., Coble, J.B., Rothman, N., Schleiff, P.L., and Travis, W.D., The Diesel Exhaust in Miners Study: A Nested Case–Control Study of Lung Cancer and Diesel Exhaust. Journal of the National Cancer Institute, 2012. 45. International Agency for Research on Cancer, Iarc: Diesel Engine Exhaust Carcinogenic. Press Release, 2012(213). 46. Majewski, W.A., Diesel Particulate Matter. DieselNet Technology Guide, 2000. 47. Wichmann, H.-E., Diesel Exhaust Particles. Inhalation Toxicology, 2007. 19(S1): p. 241-244. 48. McClellan, R.O., Health-Effects of Exposure to Diesel Exhaust Particles. Annual Review of Pharmacology and Toxicology, 1987. 27: p. 279-300. 49. Gerlofs-Nijland, M.E., Totlandsdal, A.I., Kilinc, E., Boere, A.J.F., Fokkens, P.H.B., Leseman, D., Sioutas, C., Schwarze, P.E., Spronk, H.M., Hadoke, P.W.F., Miller, M.R., and Cassee, F.R., Pulmonary and Cardiovascular Effects of Traffic-Related Particulate Matter: 4-Week Exposure of Rats to Roadside and Diesel Engine Exhaust Particles. Inhalation Toxicology, 2010. 22(14): p. 1162-1173.

Chapter 2: Literature Review 69

50. Nemmar, A., Al-Maskari, S., Ali, B.H., and Al-Amri, I.S., Cardiovascular and Lung Inflammatory Effects Induced by Systemically Administered Diesel Exhaust Particles in Rats. American Journal of Physiology-Lung Cellular and Molecular Physiology, 2007. 292(3): p. L664-L670. 51. Kweon, C.-B., Foster, D.E., Schauer, J.J., and Okada, S., Detailed Chemical Composition and Particle Size Assessment of Diesel Engine Exhaust. 2002, SAE Technical Paper. 52. Su, D.S., Müller, J.-O., Jentoft, R.E., Rothe, D., Jacob, E., and Schlögl, R., Fullerene-Like Soot from Euroiv Diesel Engine: Consequences for Catalytic Automotive Pollution Control. Topics in Catalysis, 2004. 30(1-4): p. 241-245. 53. Heywood, J.B., Internal Combustion Engine Fundamentals. Vol. 930. 1988: Mcgraw-hill New York. 54. Lapuerta, M., Armas, O., Hernandez, J.J., and Tsolakis, A., Potential for Reducing Emissions in a Diesel Engine by Fuelling with Conventional Biodiesel and Fischer-Tropsch Diesel. Fuel, 2010. 89(10): p. 3106-3113. 55. Ma, F. and Hanna, M.A., Biodiesel Production: A Review. Bioresource Technology, 1999. 70(1): p. 1-15. 56. Harrington, K.J., Chemical and Physical Properties of Vegetable Oil Esters and Their Effect on Diesel Fuel Performance. Biomass, 1986. 9(1): p. 1-17. 57. Durbin, T.D., Collins, J.R., Norbeck, J.M., and Smith, M.R., Effects of Biodiesel, Biodiesel Blends, and a Synthetic Diesel on Emissions from Light Heavy-Duty Diesel Vehicles. Environmental Science & Technology, 2000. 34(3): p. 349-355. 58. Dry, M.E., High Quality Diesel Via the Fischer–Tropsch Process–a Review. Journal of Chemical Technology and Biotechnology, 2002. 77(1): p. 43-50. 59. Alleman, T.L. and McCormick, R.L., Fischer-Tropsch Diesel Fuels- Properties and Exhaust Emissions: A Literature Review. 2003, SAE Technical Paper. 60. Li, Y., Tian, G., and Xu, H., Application of Biodiesel in Automotive Diesel Engines. 2013. 61. Demirbas, A., Biodiesel: A Realistic Fuel Alternative for Diesel Engines. 2008: Springer. 62. Zhang, X., Peterson, C., Reece, D., Haws, R., and Moller, G., Biodegradability of Biodiesel in the Aquatic Environment. Transactions of the ASAE, 1998. 41(5): p. 1423-1430. 63. Wang, J., Wu, F., Xiao, J., and Shuai, S., Oxygenated Blend Design and Its Effects on Reducing Diesel Particulate Emissions. Fuel, 2009. 88(10): p. 2037-2045. 64. Graboski, M.S. and McCormick, R.L., Combustion of Fat and Vegetable Oil Derived Fuels in Diesel Engines. Progress in Energy and Combustion Science, 1998. 24(2): p. 125-164. 65. Song, J. and Lee, K., Fuel Property Impacts on Diesel Particulate Morphology, Nanostructures, and Nox Emissions. SAE Technical Paper, 2007: p. 01-0129. 66. Knothe, G., Dependence of Biodiesel Fuel Properties on the Structure of Fatty Acid Alkyl Esters. Fuel processing technology, 2005. 86(10): p. 1059- 1070. 67. Salvi, A.A., Assanis, D., and Filipi, Z., Impact of Physical and Chemical Properties of Alternative Fuels on Combustion, Gaseous Emissions, and

70 Chapter 2: Literature Review

Particulate Matter During Steady and Transient Engine Operation. Energy & Fuels, 2012. 26(7): p. 4231-4241. 68. Ra, Y., Reitz, R.D., McFarlane, J., and Daw, C.S., Effects of Fuel Physical Properties on Diesel Engine Combustion Using Diesel and Bio-Diesel Fuels. 2008, SAE Technical Paper. 69. Knothe, G., Sharp, C.A., and Ryan III, T.W., Exhaust Emissions of Biodiesel, Petrodiesel, Neat Methyl Esters, and Alkanes in a New Technology Engine. Energy & Fuels, 2006. 20(1): p. 403-408. 70. Pinzi, S., Rounce, P., Herreros, J.M., Tsolakis, A., and Pilar Dorado, M., The Effect of Biodiesel Fatty Acid Composition on Combustion and Diesel Engine Exhaust Emissions. Fuel, 2012. 71. Tinsdale, M., Price, P., and Chen, R., The Impact of Biodiesel on Particle Number, Size and Mass Emissions from a Euro4 Diesel Vehicle. 2010. 72. Lee, R., Pedley, J., and Hobbs, C., Fuel Quality Impact on Heavy Duty Diesel Emissions:-a Literature Review. 1998, SAE Technical Paper. 73. Bacha, J., Freel, J., Gibbs, A., Gibbs, L., Hemighaus, G., Hoekman, K., Horn, J., Ingham, M., and Jossens, L., Diesel Fuels Technical Review. Chevron Global Marketing, 2007. 74. Kalligeros, S., Zannikos, F., Stournas, S., Lois, E., Anastopoulos, G., Teas, C., and Sakellaropoulos, F., An Investigation of Using Biodiesel/Marine Diesel Blends on the Performance of a Stationary Diesel Engine. Biomass and Bioenergy, 2003. 24(2): p. 141-149. 75. Cheng, C.H., Cheung, C.S., Chan, T.L., Lee, S.C., and Yao, C.D., Experimental Investigation on the Performance, Gaseous and Particulate Emissions of a Methanol Fumigated Diesel Engine. Science of the Total Environment, 2008. 389(1): p. 115-124. 76. Zhang, Z.-H. and Balasubramanian, R., Effect of Oxygenated Fuels on Physicochemical and Toxicological Characteristics of Diesel Particulate Emissions. Environmental Science & Technology, 2014. 77. Fontaras, G., Karavalakis, G., Kousoulidou, M., Tzamkiozis, T., Ntziachristos, L., Bakeas, E., Stournas, S., and Samaras, Z., Effects of Biodiesel on Passenger Car Fuel Consumption, Regulated and Non- Regulated Pollutant Emissions over Legislated and Real-World Driving Cycles. Fuel, 2009. 88(9): p. 1608-1617. 78. Lee, H., Myung, C.L., and Park, S., Time-Resolved Particle Emission and Size Distribution Characteristics During Dynamic Engine Operation Conditions with Ethanol-Blended Fuels. Fuel, 2009. 88(9): p. 1680-1686. 79. Cheng, C.H., Cheung, C.S., Chan, T.L., Lee, S.C., Yao, C.D., and Tsang, K.S., Comparison of Emissions of a Direct Injection Diesel Engine Operating on Biodiesel with Emulsified and Fumigated Methanol. Fuel, 2008. 87(10- 11): p. 1870-1879. 80. Pham, P.X., Bodisco, T.A., Ristovski, Z.D., Brown, R.J., and Masri, A.R., The Influence of Fatty Acid Methyl Ester Profiles on Inter-Cycle Variability in a Heavy Duty Compression Ignition Engine. Fuel, 2014. 116(0): p. 140- 150. 81. Xu, Z., Li, X., Guan, C., and Huang, Z., Characteristics of Exhaust Diesel Particles from Different Oxygenated Fuels. Energy & Fuels, 2013. 27: p. 7579–7586. 82. Liu, Y.Y., Lin, T.C., Wang, Y.J., and Ho, W.L., Biological Toxicities of Emissions from an Unmodified Engine Fueled with Diesel and Biodiesel

Chapter 2: Literature Review 71

Blend. Journal of Environmental Science and Health Part a-Toxic/Hazardous Substances & Environmental Engineering, 2008. 43(14): p. 1735-1743. 83. Surawski, N.C., Miljevic, B., Ayoko, G.A., Elbagir, S., Stevanovic, S., Fairfull-Smith, K.E., Bottle, S.E., and Ristovski, Z.D., Physicochemical Characterization of Particulate Emissions from a Compression Ignition Engine: The Influence of Biodiesel Feedstock. Environmental Science & Technology, 2011. 45(24): p. 10337-43. 84. Sidhu, S., Graham, J., and Striebich, R., Semi-Volatile and Particulate Emissions from the Combustion of Alternative Diesel Fuels. Chemosphere, 2001. 42(5-7): p. 681-690. 85. Surawski, N.C., Miljevic, B., Ayoko, G.A., Roberts, B.A., Elbagir, S., Fairfull-Smith, K.E., Bottle, S.E., and Ristovski, Z.D., Physicochemical Characterization of Particulate Emissions from a Compression Ignition Engine Employing Two Injection Technologies and Three Fuels. Environmental Science & Technology, 2011. 45(13): p. 5498-5505. 86. Cheung, K.L., Polidori, A., Ntziachristos, L., Tzamkiozis, T., Samaras, Z., Cassee, F.R., Gerlofs, M., and Sioutas, C., Chemical Characteristics and Oxidative Potential of Particulate Matter Emissions from Gasoline, Diesel, and Biodiesel Cars. Environmental Science & Technology, 2009. 43(16): p. 6334-6340. 87. Andreae, M.O., A New Look at Aging Aerosols. Science, 2009. 326(5959): p. 1493-1494. 88. Donahue, N.M., Robinson, A.L., Stanier, C.O., and Pandis, S.N., Coupled Partitioning, Dilution, and Chemical Aging of Semivolatile Organics. Environmental Science & Technology, 2006. 40(8): p. 2635-2643. 89. Simoneit, B.R., Kobayashi, M., Mochida, M., Kawamura, K., Lee, M., Lim, H.J., Turpin, B.J., and Komazaki, Y., Composition and Major Sources of Organic Compounds of Aerosol Particulate Matter Sampled During the Ace‐ Asia Campaign. Journal of Geophysical Research: Atmospheres (1984– 2012), 2004. 109(D19). 90. Jimenez, J.L., Canagaratna, M.R., Donahue, N.M., Prevot, A.S.H., Zhang, Q., Kroll, J.H., DeCarlo, P.F., Allan, J.D., Coe, H., Ng, N.L., Aiken, A.C., Docherty, K.S., Ulbrich, I.M., Grieshop, A.P., Robinson, A.L., Duplissy, J., Smith, J.D., Wilson, K.R., Lanz, V.A., Hueglin, C., Sun, Y.L., Tian, J., Laaksonen, A., Raatikainen, T., Rautiainen, J., Vaattovaara, P., Ehn, M., Kulmala, M., Tomlinson, J.M., Collins, D.R., Cubison, M.J., E., Dunlea, J., Huffman, J.A., Onasch, T.B., Alfarra, M.R., Williams, P.I., Bower, K., Kondo, Y., Schneider, J., Drewnick, F., Borrmann, S., Weimer, S., Demerjian, K., Salcedo, D., Cottrell, L., Griffin, R., Takami, A., Miyoshi, T., Hatakeyama, S., Shimono, A., Sun, J.Y., Zhang, Y.M., Dzepina, K., Kimmel, J.R., Sueper, D., Jayne, J.T., Herndon, S.C., Trimborn, A.M., Williams, L.R., Wood, E.C., Middlebrook, A.M., Kolb, C.E., Baltensperger, U., and Worsnop, D.R., Evolution of Organic Aerosols in the Atmosphere. Science, 2009. 326(5959): p. 1525-1529. 91. Robinson, A.L., Donahue, N.M., Shrivastava, M.K., Weitkamp, E.A., Sage, A.M., Grieshop, A.P., Lane, T.E., Pierce, J.R., and Pandis, S.N., Rethinking Organic Aerosols: Semivolatile Emissions and Photochemical Aging. Science, 2007. 315(5816): p. 1259-1262. 92. Miracolo, M.A., Hennigan, C.J., Ranjan, M., Nguyen, N.T., Gordon, T.D., Lipsky, E.M., Presto, A.A., Donahue, N.M., and Robinson, A.L., Secondary

72 Chapter 2: Literature Review

Aerosol Formation from Photochemical Aging of Aircraft Exhaust in a Smog Chamber. Atmospheric Chemistry and Physics, 2011. 11(9): p. 4135-4147. 93. Jacob, D., Introduction to Atmospheric Chemistry. 1999: Princeton University Press. 94. Chin, J.-Y., Batterman, S.A., Northrop, W.F., Bohac, S.V., and Assanis, D.N., Gaseous and Particulate Emissions from Diesel Engines at Idle and under Load: Comparison of Biodiesel Blend and Ultra Low Sulfur Diesel Fuels. Energy & Fuels, 2012. 95. Lipsky, E.M. and Robinson, A.L., Effects of Dilution on Fine Particle Mass and Partitioning of Semivolatile Organics in Diesel Exhaust and Wood Smoke. Environmental Science & Technology, 2006. 40(1): p. 155-162. 96. Leskinen, A.P., Jokinlemi, J.K., and Lehtinen, K.E.J., Transformation of Diesel Engine Exhaust in an Environmental Chamber. Atmospheric Environment, 2007. 41(39): p. 8865-8873. 97. Weitkamp, E.A., Sage, A.M., Pierce, J.R., Donahue, N.M., and Robinson, A.L., Organic Aerosol Formation from Photochemical Oxidation of Diesel Exhaust in a Smog Chamber. Environmental Science & Technology, 2007. 41(20): p. 6969-6975. 98. McNaught, A.D. and Wilkinson, A., Iupac Compendium of Chemical Terminology. Vol. 2. 1997: Blackwell Scientific Publications. 99. Finlayson-Pitts, B.J. and Pitts Jr, J.N., Tropospheric Air Pollution: Ozone, Airborne Toxics, Polycyclic Aromatic Hydrocarbons, and Particles. Science, 1997. 276(5315): p. 1045-1051. 100. Wang, J., Doussin, J.F., Perrier, S., Perraudin, E., Katrib, Y., Pangui, E., and Picquet-Varrault, B., Design of a New Multi-Phase Experimental Simulation Chamber for Atmospheric Photosmog, Aerosol and Cloud Chemistry Research. Atmos. Meas. Tech., 2011. 4(11): p. 2465-2494. 101. McWhinney, R.D., Gao, S.S., Zhou, S.M., and Abbatt, J.P.D., Evaluation of the Effects of Ozone Oxidation on Redox-Cycling Activity of Two-Stroke Engine Exhaust Particles. Environmental Science & Technology, 2011. 45(6): p. 2131-2136. 102. Sleiman, M., Destaillats, H., Smith, J.D., Liu, C.L., Ahmed, M., Wilson, K.R., and Gundel, L.A., Secondary Organic Aerosol Formation from Ozone- Initiated Reactions with Nicotine and Secondhand Tobacco Smoke. Atmospheric Environment, 2010. 44(34): p. 4191-4198. 103. Kroll, J.H. and Seinfeld, J.H., Chemistry of Secondary Organic Aerosol: Formation and Evolution of Low-Volatility Organics in the Atmosphere. Atmospheric Environment, 2008. 42(16): p. 3593-3624. 104. European Union, Directive 2004/42/Ce of the European Parliament and the Council. 2007, European Union Publications. 105. World Health Organization, Indoor Air Quality: Organic Pollutants. 1989: Copenhagen. p. World Health Organization Regional Office for Europe. 106. Mahbub, P., Ayoko, G.A., Goonetilleke, A., and Egodawatta, P., Analysis of the Build-up of Semi and Non Volatile Organic Compounds on Urban Roads. Water Research, 2011. 45(9): p. 2835-2844. 107. Li, Q., Wyatt, A., and Kamens, R.M., Oxidant Generation and Toxicity Enhancement of Aged-Diesel Exhaust. Atmospheric Environment, 2009. 43(5): p. 1037-1042. 108. Biswas, S., Verma, V., Schauer, J.J., Cassee, F.R., Cho, A.K., and Sioutas, C., Oxidative Potential of Semi-Volatile and Non Volatile Particulate Matter

Chapter 2: Literature Review 73

(Pm) from Heavy-Duty Vehicles Retrofitted with Emission Control Technologies. Environmental Science & Technology, 2009. 43(10): p. 3905- 3912. 109. Verma, V., Ning, Z., Cho, A.K., Schauer, J.J., Shafer, M.M., and Sioutas, C., Redox Activity of Urban Quasi-Ultrafine Particles from Primary and Secondary Sources. Atmospheric Environment, 2009. 43(40): p. 6360-6368. 110. Verma, V., Pakbin, P., Cheung, K.L., Cho, A.K., Schauer, J.J., Shafer, M.M., Kleinman, M.T., and Sioutas, C., Physicochemical and Oxidative Characteristics of Semi-Volatile Components of Quasi-Ultrafine Particles in an Urban Atmosphere. Atmospheric Environment, 2011. 45(4): p. 1025- 1033. 111. Portet-Koltalo, F., Preterre, D., and Dionnet, F., A New Analytical Methodology for a Fast Evaluation of Semi-Volatile Polycyclic Aromatic Hydrocarbons in the Vapor Phase Downstream of a Diesel Engine Particulate Filter. Journal of Chromatography A, 2011. 1218(7): p. 981-989. 112. Ayres, J.G., Borm, P., Cassee, F.R., Castranova, V., Donaldson, K., Ghio, A., Harrison, R.M., Hider, R., Kelly, F., Kooter, I.M., Marano, F., Maynard, R.L., Mudway, I., Nel, A., Sioutas, C., Smith, S., Baeza-Squiban, A., Cho, A., Duggan, S., and Froines, J., Evaluating the Toxicity of Airborne Particulate Matter and Nanoparticles by Measuring Oxidative Stress Potential - a Workshop Report and Consensus Statement. Inhalation Toxicology, 2008. 20(1): p. 75-99. 113. Mills, N.L., Tornqvist, H., Robinson, S.D., Gonzalez, M., Darnley, K., MacNee, W., Boon, N.A., Donaldson, K., Blomberg, A., Sandstrom, T., and Newby, D.E., Diesel Exhaust Inhalation Causes Vascular Dysfunction and Impaired Endogenous Fibrinolysis. Circulation, 2005. 112(25): p. 3930-3936. 114. Miljevic, B., Heringa, M.F., Keller, A., Meyer, N.K., Good, J., Lauber, A., Decarlo, P.F., Fairfull-Smith, K.E., Nussbaumer, T., Burtscher, H., Prevot, A.S.H., Baltensperger, U., Bottle, S.E., and Ristovski, Z.D., Oxidative Potential of Logwood and Pellet Burning Particles Assessed by a Novel Profluorescent Nitroxide Probe. Environmental Science & Technology, 2010. 44(17): p. 6601-6607. 115. Surawski, N.C., Miljevic, B., Roberts, B.A., Modini, R.L., Situ, R., Brown, R.J., Bottle, S.E., and Ristovski, Z.D., Particle Emissions, Volatility, and Toxicity from an Ethanol Fumigated Compression Ignition Engine. Environmental Science & Technology, 2010. 44(1): p. 229-235. 116. Swanson, J. and Kittelson, D., Evaluation of Thermal Denuder and Catalytic Stripper Methods for Solid Particle Measurements. Journal of Aerosol Science, 2010. 41(12): p. 1113-1122. 117. Robinson, A.L., Donahue, N.M., Shrivastava, M.K., Weitkamp, E.A., Sage, A.M., Grieshop, A.P., Lane, T.E., Pierce, J.R., and Pandis, S.N., Rethinking Organic Aerosols: Semivolatile Emissions and Photochemical Aging.(Reports). SCIENCE, 2007. 315(5816): p. 1259(4). 118. Koo, B., Ansari, A.S., and Pandis, S.N., Integrated Approaches to Modeling the Organic and Inorganic Atmospheric Aerosol Components. Atmospheric Environment, 2003. 37(34): p. 4757-4768. 119. Ning, Z. and Sioutas, C., Atmospheric Processes Influencing Aerosols Generated by Combustion and the Inference of Their Impact on Public Exposure: A Review. Aerosol and Air Quality Research, 2010. 10(1): p. 43- 58.

74 Chapter 2: Literature Review

120. Zielinska, B., Atmospheric Transformation of Diesel Emissions. Experimental and Toxicologic Pathology, 2005. 57: p. 31-42. 121. Kang, E., Root, M.J., Toohey, D.W., and Brune, W.H., Introducing the Concept of Potential Aerosol Mass (Pam). Atmos. Chem. Phys., 2007. 7(22): p. 5727-5744. 122. Papapostolou, V., Lawrence, J.E., Diaz, E.A., Wolfson, J.M., Ferguson, S.T., Long, M.S., Godleski, J.J., and Koutrakis, P., Laboratory Evaluation of a Prototype Photochemical Chamber Designed to Investigate the Health Effects of Fresh and Aged Vehicular Exhaust Emissions. Inhalation Toxicology, 2011. 23(8): p. 495-505. 123. Papapostolou, V., Lawrence, J.E., Ferguson, S.T., Wolfson, J.M., Diaz, E.A., Godleski, J.J., and Koutrakis, P., Development and Characterization of an Exposure Generation System to Investigate the Health Effects of Particles from Fresh and Aged Traffic Emissions. Air Quality, Atmosphere & Health, 2012: p. 1-11. 124. Lambe, A., Ahern, A., Williams, L., Slowik, J., Wong, J., Abbatt, J., Brune, W., Ng, N., Wright, J., and Croasdale, D., Characterization of Aerosol Photooxidation Flow Reactors: Heterogeneous Oxidation, Secondary Organic Aerosol Formation and Cloud Condensation Nuclei Activity Measurements. Atmos. Meas. Tech, 2011. 4(3): p. 445-461. 125. Keller, A. and Burtscher, H., A Continuous Photo-Oxidation Flow Reactor for a Defined Measurement of the Soa Formation Potential of Wood Burning Emissions. Journal of Aerosol Science, 2012. 49(0): p. 9-20. 126. Olariu, R.I., Barnes, I., Bejan, I., Arsene, C., Vione, D., Klotz, B., and Becker, K.H., Ft-Ir Product Study of the Reactions of No3 Radicals with Ortho-, Meta-, and Para-Cresol. Environmental Science & Technology, 2013. 47(14): p. 7729-7738. 127. Bartolomei, V., Gomez Alvarez, E., Glor, M., Gligorovski, S., Temime- Roussel, B., Quivet, E., Strekowski, R., Zetzsch, C., Held, A., and Wortham, H. Combustion Processes Indoors: A Source of High Oh Radical Concentrations through the Photolysis of Hono. in AGU Fall Meeting Abstracts. 2013. 128. Seakins, P. A Brief Review of the Use of Environmental Chambers for Gas Phase Studies of Kinetics, Chemical Mechanisms and Characterisation of Field Instruments. in EPJ Web of Conferences. 2010: EDP Sciences. 129. Anderson, F., Commane, R., Glowacki, D., Goddard, A., Hemavibool, K., Malkin, T.I.T., Heard, D., Pilling, M., and Seakins, P., Hirac-a Highly Instrumented Reactor for Atmospheric Chemistry. spectroscopy, 2007. 7: p. 191-216. 130. Malkin, T., Goddard, A., Heard, D., and Seakins, P., Measurements of Oh and Ho2 Yields from the Gas Phase Ozonolysis of Isoprene. Atmos. Chem. Phys, 2010. 10: p. 1441-1459. 131. Colin, E., Compendium of Chemical Terminology: Iupac Recommendations : Compiled by V. Gold, K.L. Loening, A.D. Mcnaught, and P. Sehmi, Blackwell, Oxford, Etc., 1987, Viii + 456 Pages. £45.00 (Hard Cover) Isbn 0- 632-01765-1; £29.50 (Soft Cover) Isbn 0-632-01767-3. Journal of Organometallic Chemistry, 1997. 356(2): p. C76-C77. 132. Atkinson, R., Carter, W.P.L., Darnall, K.R., Winer, A.M., and Pitts, J.N., A Smog Chamber and Modeling Study of the Gas-Phase Nox-Air Photo-

Chapter 2: Literature Review 75

Oxidation of Toluene and the Cresols. International Journal of Chemical Kinetics, 1980. 12(11): p. 779-836. 133. Izumi, K. and Fukuyama, T., Photochemical Aerosol Formation from Aromatic-Hydrocarbons in the Presence of Nox. Atmospheric Environment Part a-General Topics, 1990. 24(6): p. 1433-1441. 134. Geiger, H., Kleffmann, J., and Wiesen, P., Smog Chamber Studies on the Influence of Diesel Exhaust on Photosmog Formation. Atmospheric Environment, 2002. 36(11): p. 1737-1747. 135. Kleindienst, T.E., Corse, E.W., Li, W., McIver, C.D., Conver, T.S., Edney, E.O., Driscoll, D.J., Speer, R.E., Weathers, W.S., and Tejada, S.B., Secondary Organic Aerosol Formation from the Irradiation of Simulated Automobile Exhaust. Journal of the Air & Waste Management Association, 2002. 52(3): p. 259-272. 136. Samy, S. and Zielinska, B., Secondary Organic Aerosol Production from Modern Diesel Engine Emissions. Atmospheric Chemistry and Physics, 2010. 10(2): p. 609-625. 137. Chirico, R., DeCarlo, P., Heringa, M., Tritscher, T., Richter, R., Prévôt, A., Dommen, J., Weingartner, E., Wehrle, G., and Gysel, M., Impact of Aftertreatment Devices on Primary Emissions and Secondary Organic Aerosol Formation Potential from in-Use Diesel Vehicles: Results from Smog Chamber Experiments. Atmospheric Chemistry & Physics Discussions, 2010. 10: p. 16055-16109. 138. Dockery, D.W., Pope, C.A., Xu, X., Spengler, J.D., Ware, J.H., Fay, M.E., Ferris, B.G., and Speizer, F.E., An Association between Air Pollution and Mortality in Six U.S. Cities. New England Journal of Medicine, 1993. 329(24): p. 1753-1759. 139. Pope, C.A., Burnett, R.T., Thun, M.J., Calle, E.E., Krewski, D., Ito, K., and Thurston, G.D., Lung Cancer, Cardiopulmonary Mortality, and Long-Term Exposure to Fine Particulate Air Pollution. JAMA: The Journal of the American Medical Association, 2002. 287(9): p. 1132-1141. 140. Pope, C.A. and Dockery, D.W., Health Effects of Fine Particulate Air Pollution: Lines That Connect. Journal of the Air & Waste Management Association, 2006. 56(6): p. 709-742. 141. Pope, C.A., III, Muhlestein, J.B., May, H.T., Renlund, D.G., Anderson, J.L., and Horne, B.D., Ischemic Heart Disease Events Triggered by Short-Term Exposure to Fine Particulate Air Pollution. Circulation, 2006. 114(23): p. 2443-2448. 142. Pope, C.A., III, Schwartz, J., and Ransom, M.R., Daily Mortality and Pm10 Pollution in Utah Valley. (Particulate Pollution). Archives of Environmental Health, 1992. v47(n3): p. p211(7). 143. Schwartz, J., Air Pollution and Daily Mortality: A Review and Meta Analysis. Environmental Research, 1994. 64(1): p. 36-52. 144. Bernstein, J.A., Alexis, N., Barnes, C., Bernstein, I.L., Nel, A., Peden, D., Diaz-Sanchez, D., Tarlo, S.M., and Williams, P.B., Health Effects of Air Pollution. Journal of Allergy and Clinical Immunology, 2004. 114(5): p. 1116-1123. 145. Baulig, A., Garlatti, M., Bonvallot, V., Marchand, A., Barouki, R., Marano, F., and Baeza-Squiban, A., Involvement of Reactive Oxygen Species in the Metabolic Pathways Triggered by Diesel Exhaust Particles in Human Airway

76 Chapter 2: Literature Review

Epithelial Cells. American Journal of Physiology-Lung Cellular and Molecular Physiology, 2003. 285(3): p. L671-L679. 146. Stevanovic, S., Miljevic, B., Surawski, N., Fairfull-Smith, K.E., Bottle, S.E., Brown, R., and Ristovski, Z.D., The Influence of Oxygenated Organic Aerosols (Ooa) on the Oxidative Potential of Diesel and Biodiesel Particulate Matter. Environmental Science & Technology, 2013. 47(14): p. 7655-7662. 147. Miracolo, M.A., Presto, A.A., Lambe, A.T., Hennigan, C.J., Donahue, N.M., Kroll, J.H., Worsnop, D.R., and Robinson, A.L., Photo-Oxidation of Low- Volatility Organics Found in Motor Vehicle Emissions: Production and Chemical Evolution of Organic Aerosol Mass. Environmental Science & Technology, 2010. 44(5): p. 1638-1643. 148. Chow, J.C., Measurement Methods to Determine Compliance with Ambient Air Quality Standards for Suspended Particles. Journal of the Air & Waste Management Association, 1995. 45(5): p. 320-382. 149. Spurny, K.R., Analytical Chemistry of Aerosols: Science and Technology. 1999: CRC Press. 150. TSI, Model 3010 Condensation Particle Counter, Product Information. 1999. 151. Knutson, E. and Whitby, K., Aerosol Classification by Electric Mobility: Apparatus, Theory, and Applications. Journal of Aerosol Science, 1975. 6(6): p. 443-451. 152. Rosell-Llompart, J., Loscertales, I.G., Bingham, D., and Fernández de la Mora, J., Sizing Nanoparticles and Ions with a Short Differential Mobility Analyzer. Journal of Aerosol Science, 1996. 27(5): p. 695-719. 153. Zhang, J. and Chen, D., Differential Mobility Particle Sizers for Nanoparticle Characterization. Journal of Nanotechnology in Engineering and Medicine, 2014. 5(2): p. 020801-020801. 154. Rader, D.J. and McMurry, P.H., Application of the Tandem Differential Mobility Analyzer to Studies of Droplet Growth or Evaporation. Journal of Aerosol Science, 1986. 17(5): p. 771-787. 155. Stevanovic, S., Miljevic, B., Eaglesham, G.K., Bottle, S.E., Ristovski, Z.D., and Fairfull-Smith, K.E., The Use of a Nitroxide Probe in Dmso to Capture Free Radicals in Particulate Pollution. European Journal of Organic Chemistry, 2012. 2012(30): p. 5908-5912. 156. Stevanovic, S., Miljevic, B., Surawski, N.C., Fairfull-Smith, K.E., Bottle, S.E., Brown, R., and Ristovski, Z.D., Influence of Oxygenated Organic Aerosols (Ooas) on the Oxidative Potential of Diesel and Biodiesel Particulate Matter. Environmental Science & Technology, 2013. 47(14): p. 7655-7662. 157. Miljevic, B., Modini, R.L., Bottle, S.E., and Ristovski, Z.D., On the Efficiency of Impingers with Fritted Nozzle Tip for Collection of Ultrafine Particles. Atmospheric Environment, 2009. 43(6): p. 1372-1376. 158. Stevanovic, S., Ristovski, Z.D., Miljevic, B., Fairfull-Smith, K.E., and Bottle, S.E., Application of Profluorescent Nitroxides for Measurements of Oxidative Capacity of Combustion Generated Particles. Chemical Industry & Chemical Engineering Quarterly, 2012. 18(4): p. 653-659. 159. Hedayat F., S.S., Miljevic B., Bottle S., Ristovski Z.D., Review-Evaluating the Molecular Assays for Measuring the Oxidative Potential of Particulate Matter. Chemical Industry and Chemical Engineering Quarterly, 2014.

Chapter 2: Literature Review 77

78 Chapter 2: Literature Review Chapter 3: Particle emissions from biodiesels with different physical properties and chemical composition

Md Mostafizur Rahman1, A. M. Pourkhesalian1, M. I. Jahirul1, S. Stevanovic1, H. Wang1, A. R. Masri2, R. J. Brown1, Z. D. Ristovski*1

1 International Laboratory for Air Quality and Health, Queensland University of Technology, GPO Box 2434, Brisbane QLD, 4001, Australia

2Clean Combustion Research Group, The University of Sydney, NSW2006, Australia

*E-mail: [email protected]

Published in journal Fuel 2014

doi:10.1016/j.fuel.2014.05.053

Chapter 3: Particle emissions from biodiesels with different physical properties and chemical composition 79

Statement of Joint Authorship

Title: Particle emissions from biodiesels with different physical properties and chemical composition

Authors: Md Mostafizur Rahman, A. M. Pourkhesalian, M. I. Jahirul, S. Stevanovic, H. Wang, A. R. Masri, R. J. Brown, Z. D. Ristovski

A. M. Pourkhesalian: Contributed in design of the experimental program, assisted in data analysis and write up the manuscript.

Signature: Date:

Md Mostafizur Rahman: Performed the measurements and data analysis and wrote the manuscript.

M. I. Jahirul: Contributed in measurement, data analysis and assisted in write up the manuscript

S. Stevanovic: Contributed in emission measurement, and reviewed the manuscript

H. Wang: Contributed in emission measurement and reviewed the manuscript

A. R. Masri: Supervised the overall study design, set up and experiments, reviewed the manuscript

R. J. Brown: Supervised the overall study design, set up and experiments, reviewed the manuscript

Z. D. Ristovski: Supervised the overall study design, set up and experiments, reviewed the manuscript.

The authors listed above have certified that:

1. They meet the criteria for authorship in that they have participated in the conception, execution, or interpretation, of at least that part of the publication in their field of expertise;

80 Chapter 3: Particle emissions from biodiesels with different physical properties and chemical composition

2. They take public responsibility for their part of the publication, except for the responsible author who accepts overall responsibility for the publication;

3. There are no other authors of the publication according to these criteria;

4. Potential conflicts of interest have been disclosed to (a) granting bodies, (b) the editor or publisher of journals or other publications, and (c) the head of the responsible academic unit, and

5. They agree to the use of the publication in the student’s thesis and its publication on the Australasian Research Online database consistent with any limitations set by publisher requirements.

Principal Supervisor confirmation

I have sighted email or other correspondence from all Co-authors confirming their certifying authorship.

Name Zoran Ristovski

Signature Date

Chapter 3: Particle emissions from biodiesels with different physical properties and chemical composition 81

Title: Particle emissions from biodiesels with different physical properties and chemical composition

M. M. Rahman a, A. M. Pourkhesalian a, M. I. Jahirul a, S. Stevanovic a, P. X. Pham b, H. Wang a, A.R. Masri b, R. J. Brown a and *Z. D. Ristovski a

a ILAQH and BERF, Queensland University of Technology, Brisbane, QLD4001, Australia

b Clean Combustion Research Group, The University of Sydney, NSW2006, Australia

*E-mail: [email protected]

Abstract:

Biodiesels produced from different feedstocks usually have wide variations in their fatty acid methyl ester (FAME) so that their physical properties and chemical composition are also different. The aim of this study is to investigate the effect of the physical properties and chemical composition of biodiesels on engine exhaust particle emissions. Alongside with neat diesel, four biodiesels with variations in carbon chain length and degree of unsaturation have been used at three blending ratio (B100, B50, B20) in a common rail engine. It is found that particle emission increased with the increase of carbon chain length. However, for similar carbon chain length, particle emissions from biodiesel having relatively high average unsaturation are found to be slightly less than that of low average unsaturation. Particle size is also found to be dependent on fuel type. The fuel or fuel mix responsible for higher PM and PN emissions is also found responsible for larger particle median size. Particle emissions reduced consistently with fuel oxygen content regardless of the proportion of biodiesel in the blends, whereas it increased

82 Chapter 3: Particle emissions from biodiesels with different physical properties and chemical composition

with fuel viscosity and surface tension only for higher diesel-biodiesel blend percentages (B100, B50). However, since fuel oxygen content increases with the decreasing carbon chain length, it is not clear which of these factors drives the lower particle emission. Overall, it is evident from the results presented here that chemical composition of biodiesel is more important than its physical properties in controlling exhaust particle emissions.

Keywords: Biodiesel; particle emissions; fuel physical properties; fuel chemical composition.

3.1 INTRODUCTION Compression Ignition (CI) engines are increasing in popularity due to their higher thermal efficiency. They power a wide range of land and sea transport as well as provide electrical power, used in farming, construction and industrial applications. Tail pipe emissions of diesel engines, especially particulate matter (PM) are still a matter of concern due to its harmful effects both on human health and the environment [1, 2]. Exposure to diesel particulate matter (DPM) can cause pulmonary diseases such as asthma, bronchitis and lung cancer [1] and because of these adverse effects, the International Agency for Research on Cancer (IARC) included DPM as carcinogenic to human health.

The harmful effects caused by DPM are related to both the physical properties and chemical composition of the particles. The physical properties that influence respiratory health include particle mass, surface area, mixing status of particles, number and size distribution[3]. The particles deposit in different parts of the lung depending on their size. The smaller the particles the higher the deposition efficiency [4] and the greater the chance of them penetrating deep into the lung. The smaller particles stay suspended in the atmosphere for longer thus have a higher probability of being inhaled and consequently deposited deep in the alveolar region of the lung. Particle number governs the ability of particles to grow larger in size by coagulation while particle surface area determines the ability of the particles to carry toxic substances. Recent studies reveal that DPM surface area and organic compounds play a significant role in initiating various cellular and chemical processes responsible for respiratory disease [3, 5]. In addition to this, a large fraction of DPM

Chapter 3: Particle emissions from biodiesels with different physical properties and chemical composition 83

is black carbon, which is considered the second most potential greenhouse warming agent after carbon dioxide [2]. After treatment devices (ATD) like diesel particulate filters (DPF) and diesel oxidation catalysts (DOC) aid in reducing DPM [6]. Alternative fuels are another potential emission reducing source [7]. Of these fuels, biodiesel is considered one of the more promising for diesel engines [8, 9] as it produces less PM and other gaseous emissions [9-11]. Biodiesel in diesel engines also has the potential to neutralize carbon emissions as it comes from a renewable source of energy.

Biodiesel is a mixture of fatty acid esters with physicochemical properties that mostly depend on the structure of the ester molecule. They can be produced from a variety of feedstock sources such as vegetable oil, animal fat, municipal and industrial waste and some even from insects [12-15]. The vast majority of feedstocks actually used now days are derived from vegetable oils and animal fats. An extensive range of fatty acid profiles exist among these feedstocks [16], with fatty acid profile being even different within the same feedstocks. If plant sources are used, these variations can be controlled by manipulating stock growing conditions. Physical properties and chemical composition of biodiesel varies among different feedstocks, which can have a noticeable influence on engine performance and emissions [17]. McCormick et al.[18] reported constant PM emissions from different biodiesel feedstocks when the density was less than 0.89 g/cm3 or cetane number was greater than about 45, but increase of NOx emissions with the increase of biodiesel density and iodine number. In contradiction to these findings, a difference in particle emissions from biodiesel from different feedstocks has also been reported [19, 20]. Lapuerta et al.[10] reported a 10% increase of NOx and 20% decrease of particle emissions by unsaturated biodiesel. Benjumea et al.[21] found that the degree of unsaturation in biodiesel doesn’t significantly affect the engine performance but increases smoke opacity and THC emissions. Karavalakis et al. [22] reported noticeable influence of biodiesel origin on particle emissions, especially particle associated PAH and carbonyl emissions. Very recently Salamanaca et al.[23] reported increased PM and HC emissions from biodiesel that contains more unsaturated compounds that favor soot precursor formation. There is no distinction however in the literature, which indicates whether chemical composition of biodiesel, physical properties or a combination of these is responsible for this

84 Chapter 3: Particle emissions from biodiesels with different physical properties and chemical composition

variation in engine performance and emissions. This study therefore, aims to investigate the effect of biodiesel physical properties and chemical composition on engine exhaust particle emissions. It is an extension of the previous study [24] where results from the same experiments were presented for the engine performance characteristics and emission of pollutants including some preliminary results for the particle emission, particularly for pure biodiesel. It should be noted that the results for B100 are reproduced here for comparison purposes. Furthermore, the paper elaborates on these findings and presents new analysis in terms of the physical properties chemical composition of the fuels and their blends.

3.2 MATERIALS AND METHODS

3.2.1 Engine and fuel specification This experimental study was performed in a heavy duty 6 liters, six cylinders, turbocharged after cooled, common rail diesel engine typically used in medium size trucks. Test engine is the same as used in Pham et al. [24]. Table 3 -1 shows specification of the test engine. The engine was coupled to a water brake dynamometer, and both of them are connected to an electronic control unit (ECU). Engine was operated at 1500 rpm (maximum torque speed) and at 2000 rpm (intermediate speed), and four different loads including 25%, 50%, 75%, and 100% for each engine speed. Maximum load at any particular engine speed depends upon the type of fuel used. Therefore, for each fuel maximum load was measured at first when engine was in full throttle for a particular speed. This measured load is then considered as 100% load for that speed and other load conditions were determined based upon measured 100% load. Although PM emissions can be quiet different for transient testing, we had to conduct steady state tests as most of our measurement techniques required steady emission over longer sampling time.

Chapter 3: Particle emissions from biodiesels with different physical properties and chemical composition 85

Table 3-1 Test engine specification.

Model Cummins ISBe220 31 Cylinders 6 in-line Capacity (L) 5.9 Bore × Stroke (mm) 102 × 120 Maximum power (kW/rpm) 162/2500 Maximum torque (Nm/rpm) 820/1500 Compression ratio 17.3 Aspiration Turbocharged & after cooled Fuel Injection Common rail After treatment devices None Emissions certification Euro III

An ultra-low sulphur diesel (sulphur content 2.5mg/kg) and four biodiesels with different physical properties and chemical composition were used to run the engine. All four biodiesels originated from palm oil, and then were fractionated to separate its fatty acid ester components with specific composition. Since all of them have originated from the same feedstock, the given code names (C810, C1214 etc) are to identify them and to indicate their carbon chain length. For example C810 means biodiesel that is mainly composed of FAME’s with 8-10 carbon atoms. All four biodiesels were used at three blending ratios i.e. 100% biodiesel (B100), blends of 50% diesel and 50% biodiesel (B50), and blends of 80% diesel and 20% biodiesel (B20). Table 3 -2 shows the fatty acid profile of used biodiesels as found using gas chromatography mass spectrometry (GCMS) analysis. Biodiesel samples were analyzed using Perkin Elmer clarus 580GC-MS equipped with Elite 5MS 30m x 0.25mm x 0.25um column with a flow rate of 1mL/min. Before analyzing, each biodiesel was diluted with n-hexane (1:100 v/v). Initial temperature was 120 0C for 0.5 minutes, then raised to 310 0C for 2 minutes at 10 0C/min and kept at 310 0C for 2 minutes. The mass selective detector was optimized using calibrating standards with reference masses at m/z (40-350). Among four biodiesels, C810 is fully saturated and composed of 52% and 46% caprylic acid and capric acid ester respectively. C1214 is also dominated by saturated compounds but has comparatively longer carbon chain

86 Chapter 3: Particle emissions from biodiesels with different physical properties and chemical composition

length fatty acid ester i.e. 48% lauric, 19% % myristic, 10% palmitic and 18% oleic acid ester. On the other hand both C1618 and C1822 are dominated by long chain unsaturated fatty acid esters. C1618 is composed of 21% palmitic, 9% stearic, 58% cis-oleic and 10% linoleic acid ester where C1822 has 10% more oleic and linoleic acid ester. C1822 also has a small amount (4%) of trans-elaidic acid ester.

Table 3-2 Fatty acid profile of used biodiesels.

Biodiesels C810 C1214 C1618 C1822 Common Name Lipid Numbers Caprylic acid C8:0 52.16 0 0 0 Capric acid C10:0 46.38 0.17 0 0 Lauric acid C12:0 1.38 47.8 0.1 0 Myristic acid C14:0 0 18.89 0.06 0.03 Pentadecylic acid C15:0 0 0 0.03 0.02 Palmitic acid C16:0 0 10.19 21 4.45 Palmitoleic acid C16:1 0 0 0 0.12 Margaric acid C17:0 0 0 0.06 0 Stearic acid C18:0 0 2.55 9.47 2.53 Oleic acid C18:1cis 0 18.53 58.72 68.13 Elaidic acid C18:1trans 0 0 0 3.96 Linoleic acid C18:2 0 1.76 9.98 18.69 Arachidic acid C20:0 0 0.08 0.3 0.49 Gadoleic acid C20:1 0 0 0.24 1.03 Behenic acid C22:0 0 0.03 0.03 0.17 Glycerol -- 0.08 0 0 0 Some important properties of all used fuels related to combustion and emissions are shown in

Table 3-3 Among the used biodiesels physical properties varied with the variation in chemical composition i.e. carbon chain length and degree of saturation/unsaturation. Viscosity, heating value and iodine value increased with the increase of carbon chain length and degree of unsaturation, where saponification value decreased. Density of all four biodiesels was found higher than diesel, and no trend observed among biodiesels either with the carbon chain length or degree of unsaturation. Surface tension also increased with the carbon chain length in biodiesels, although no significant change was observed with the degree of unsaturation. For example, there is almost no difference in surface tension between C1618 and C1822, although C1822 contains much higher percentage of unsaturated compounds compared to C1618. Surface tension and cetane value of diesel were found to be lower than all four used biodiesels where calorific value was higher.

Chapter 3: Particle emissions from biodiesels with different physical properties and chemical composition 87

Viscosity of diesel was higher than C810 but lower than the rest of the three biodiesels.

Table 3-3 Important physicochemical properties of tested fuels.

Fuel type C810 C1214 C1618 C1822 Diesel Relevant properties * Average formula C9.5H19.7O2 C14.8H28.3O2 C18.3H35.3O2 C18.7H35.3O2 CxHy

Average unsaturation 0 0.22 0.7892 1.11 -- (AU) Oxygen content 18.72 13.25 10.74 10.83 0 * (wt%) Stoichiometric air 11.12 12.05 12.50 12.48 14.5 fuel ratio Relative density(kg/l) 0.877 0.871 0.873 0.879 0.8482 Viscosity (mm2/sec) 1.95 4.37 4.95 5.29 3.148 Surface tension 26.184 28.41 29.9 29.966 26 (mN/m) Cetane Number 62.96 65.57 61.06 53.65 48.5 Iodine number(g 1 max 8 65 105 I2/100g) Saponication value 330 233 195 185 (mg KOH/g) Acid value(mg 0.9 0.4 0.8 0.4 <0.05 KOH/g) Boiling point (0C) -- >150 165.6 >150 >180 * Gross Calorific 35.335 38.409 37.585 39.825 44.365 value(MJ/kg) Sulphur 0 0 0 0 2.5 content(mg/kg) values with * superscripts have taken from literature[25, 26]

3.2.2 Exhaust sampling and measurement system As shown in Figure 3.1 , the Dekati ejector diluter was used to partly sample raw exhaust from the engine exhaust pipe and then dilute it with particle free compressed air. A second Dekati diluter was connected in series with the first one to further increase the dilution ratio in order to further decrease concentration. A HEPA filter was used to provide particle free compressed air for the diluters. The purpose of the dilution was to bring down the temperature as well as the concentration of gases and PM within the measuring range of the instruments. Diluted exhaust was then sent to different gaseous and particle measuring instruments. A CAI 600 series CO2 analyzer was used to measure the CO2 concentration directly from the raw exhaust.

88 Chapter 3: Particle emissions from biodiesels with different physical properties and chemical composition

A second CO2 meter (SABLE, CA-10) connected via a three way valve between the two diluters was used to record the CO2 concentration from the diluted exhaust.

Background corrected CO2 was used as tracer gas to calculate the dilution ratio for each stage. After first stage dilution, CAI 600 series CLD NOx analyzer was used to measure the NOx. PM2.5 emissions were measured by a TSI DustTrak (Model 8530). DustTrak readings were converted into a gravimetric measurement by using the tapered element oscillating microbalance to DustTrak correlation for diesel particles published by Jamriska et al.[27]. It is worth noting this conversion can introduce significant uncertainties if the optical properties of particles significantly changes. The particle number size distribution for C810, C1214, C1618 and their blends was measured by a scanning mobility particle sizer (SMPS). This SMPS consisted of a TSI 3080 electrostatic classifier (EC) and a TSI 3025 butanol based condensation particle counter (CPC). Due to technical problems a new SMPS had to be used for the reference diesel and C1822. This SMPS system consisted of a 3085 classifier with a nano-DMA (differential mobility analyzer). As the measurement range of the two SMPS’s used was different, we have used a fitting procedure (see section 3.2) to recalculate the total PN and make the measurements comparable. A TSI 3089 nanometer aerosol sampler (NAS) was used in conjunction with a Tandem Differential Mobility Analyzer (TDMA) to collect preselected particles on Transmission Electron Microscopic (TEM) grids for morphological analysis. The EC in the TDMA preselected the size of the particles, which deposited on the TEM grid in the NAS. An Aethalometer (Magee Scientific, AE33, 7-Wavelength) was also connected after second stage dilution for black carbon (BC) measurement. Results from TEM analysis and BC data will be published in a separate paper.

Chapter 3: Particle emissions from biodiesels with different physical properties and chemical composition 89

Figure 3.1 Schematic diagram of experimental set up.

3.3 RESULTS AND DISCUSSION

Specific PM emissions All four biodiesels that were used, disregarding the variations in physical properties and chemical composition, reduced PM emissions in comparison to petrodiesel. Figure 3 .2 shows brake specific PM emissions at engine operating speeds of 1500 rpm and 2000 rpm, respectively. It was found that as the biodiesel percentage in the diesel-biodiesel blends increased, PM emissions decreased consistently. The maximum reduction in PM was observed for 100% biodiesel blends, an observation common in the literature [9, 28, 29]. Noticeable variations in PM emissions were also observed among the four biodiesels and their blends. In the case of using 100% biodiesel, a massive 98% reduction in PM was observed for biodiesel C810, where C1214, C1618 and C1822 reduced PM 83%, 70% and 76% respectively. Similar trends in PM emissions were also found for B50 and B20 blends although there was a difference of PM reduction proportion in these blends. For the B50 blend, PM reduction among biodiesels C810, C1214, C1618 and C1822 was 88%, 75%, 70% and 76% respectively. B20 was slightly lower, measuring 66%, 57%, 42% and 48% respectively. PM emissions from other tested engine loads are shown in the appendix (Figure S1). Similar trends in PM emissions were also observed for these loads at 2000 rpm engine speed, although at 1500 rpm 50% and 25% load, PM emissions from B20 (C1618) were found to be slightly higher than that from the diesel. These variations in PM emissions among biodiesels could be

90 Chapter 3: Particle emissions from biodiesels with different physical properties and chemical composition

due to either their chemical composition or their physical properties. Among the biodiesels, PM emissions increased consistently with biodiesel carbon chain length with the exception of C1822. This biodiesel carbon chain length was similar to C1618 but its degree of unsaturation was higher and its PM emissions were less. Pinzi et al. [30] also reported reduction of PM emission with the increase of degree of unsaturation but same carbon chain length in FAME. Opposing observations were also reported in the literature, which suggests that unsaturated compounds have a tendency to act as soot precursor [21, 23]. In addition, important physical properties of C1822 in regards to particle emissions i.e. viscosity and surface tension were also higher. This slight reduction in PM emissions from C1822 might be attributed to its high iodine or low cetane value. Fuels with low cetane value undergo prolonged premixed combustion phases that are responsible for less soot formation. In addition, NOx emissions from C1822 were highest among the fuels that were favorable for soot oxidation. This could also be responsible for comparatively low PM emissions from C1822.

Figure 3.2 Brake specific PM emission at (a) 1500 rpm 100% load and (b) 2000 rpm 100% load. The PM emissions shown are calculated based on DustTrak measurement.

3.3.1 Specific PN emissions Variations that were observed in PN emissions were similar to the PM emissions for the same fuels used. There were however, slight differences in the amount that PN emissions changed compared to PM emissions. All PN emissions were calculated for the size range from 10.2-514 nm. As the measurements for neat

Chapter 3: Particle emissions from biodiesels with different physical properties and chemical composition 91

diesel and C1822 were done using the nano-DMA, in the size range from 4.6nm- 156nm, a fitting procedure was used to recalculate the PN concentration to the same size range as used in the other measurements [31]. As shown in Figure 3.3 , PN emissions from B100 were found to be lower than diesel for all biodiesels. Among the biodiesels used, C810 reduced PN most and C1618 reduced PN the least compared to neat diesel, at 90% and 20% respectively. Reductions from C1214 and C1822 were measured at 60% and 35% respectively. For B50, in the case of C810, C1214 and C1822, the PN emissions remained lower than diesel although C1618 increased approximately 10%. Similar to B100, the lowest PN emissions were observed with C810 for B50 with C1214 and C1822 following the trend for B100. PN emissions from C1214 increased 15% with a large standard error at 2000 rpm, while at 1500 rpm it remained almost same to C1822. Apart from B100 and B50, PN emissions from B20 were found to be slightly less than diesel and almost the same among the biodiesels with the exception of around 15% increase from diesel at 1500 rpm. Brake specific PN emissions from other engine loads at both rpm are shown in appendix (Figure S2). PN emissions from all other loads and engine speeds showed a similar trend with the exception of 1500 rpm 50% load where PN emission from B20 appeared to have a different trend as compared to the rest of the results.

1.0x1014

13 (a) 8.0x10

6.0x1013

4.0x1013

2.0x1013

0.0

14 Diesel 1.2x10 (b) B20 1.0x1014 B50 B100 8.0x1013

6.0x1013 Brake specific PN(#/kW-hr)

4.0x1013

2.0x1013 0.0 C810 C1214 C1618 C1822 Biodiesel Type

Figure 3.3 Brake specific PN emissions at (a) 1500 rpm 100% load and (b) 2000 rpm 100%.

92 Chapter 3: Particle emissions from biodiesels with different physical properties and chemical composition

3.3.2 Particle number size distribution Particle size distribution (PSD) was always found to be unimodal with a single peak in the accumulation mode despite the variations in the fuel and the condition of engine operation (Appendix: Figure S3 to S5). Variations in PSD among the biodiesels were more prevalent for B100, followed by B50. Comparatively, PSD from B20 was found to be similar to petrodiesel regardless of the variations of biodiesel. Another important feature is that biodiesel reduced a higher proportion of large particles (mobility diameter >100 nm) compared to nanoparticles (mobility diameter <50 nm). Nanoparticle emissions from biodiesel however, did not exceed that of diesels, which have been reported in few studies [19, 32]. The presence of a nucleation mode peak in PSD can be responsible for increased nanoparticle emission. Schönborn et al. [33] found that biodiesel with a higher boiling point has more tendency to form nucleation mode particles, in an engine with a high pressure injection system (160 bar). In an engine with a mechanical injection system, Fisher et al. [34] reported that nucleation depends both on engine operating condition and biodiesel feedstocks. They concluded that a small variation in dilution and sampling system might be the main driving force that will trigger nucleation. Biodiesels used in our study have a noticeable variation in their boiling point. In addition we have used an unheated first stage dilution which is more prone to producing nucleation mode particles. In spite of that, we did not observe nucleation in any of our tests. Abundance of volatiles and semi-volatiles in the exhaust which can partition into particles upon cooling down are the primary contributor to the nucleation mode peak. Impurities in biodiesel, such as glycerol, don’t undergo complete combustion due to their high viscosity, poor atomization and mixing property. They can form partially oxidized volatiles and semi-volatiles that can also be a contributor to the formation of the nucleation mode peak. Biodiesels used in this study were free from glycerol and other impurities, which might facilitate the absence of nucleation mode peak in PSD. In consensus, the amount of volatile and semi-volatile materials produced by the biodiesels might not have been enough to trigger nucleation with the used dilution system.

3.3.3 Particle median size Particle size also varied among used fuels in a similar way to PM and PN. In case of diesel, the median size of the particles in the SMPS size distribution was 61

Chapter 3: Particle emissions from biodiesels with different physical properties and chemical composition 93

nm and 56 nm at 1500 and 2000 rpm respectively. For 100% biodiesel, particle median size was always found to be smaller than neat diesel and diesel-biodiesel blends (Figure 3.4 , a and b). Among four used biodiesels, C810 produced the smallest particle median size i.e. 40 nm and 43 nm at 1500 and 2000 rpm respectively, followed by C1822 which was 53 nm and 44 nm. C1214 and C1618 gave almost the same particle size as neat diesel with slight difference between two engines operating speeds. Particle size from B50 was found to be larger than B100 but smaller than B20 blends. Interestingly, particle emitted from all B20 blends were found to be larger than diesel with the largest particle median size observed for C1618 B20 blend. Similar trends in particle size were also observed at other engine loads which are shown in appendix (Figures S6 and S7). To gain some insight of the variation in particle sizes among different biodiesels and its blends, particle median size from all measurement was plotted against the particle number concentrations. As can be seen on Figure 3.4 c a moderate positive correlation, with a Pearson correlation coefficient of 0.61 was found between particle median size and total number concentration. This indicates that total PN number concentration through coagulation could be one of the key parameters influencing the overall particle size. Higher the particle number emissions, larger the particle size can be. The other factors may be the biodiesel viscosity, surface tension and especially oxygen contents which ensure the presence of more oxygen functional groups on the surface of particles responsible for enhanced particle oxidation and subsequent size reduction [35, 36].

94 Chapter 3: Particle emissions from biodiesels with different physical properties and chemical composition

70 (a) c 65 60 55 50 75 45 70

40 65

35 60 C810 C1214 C1618 C1822 65 55

(b) B20 50 60 B50 B100 45 40 Median(nm) Particle median size(nm) 55 Linear Fit Particle median size(nm)

35

50 5.0x106 1.0x107 1.5x107 2.0x107 2.5x107 Engine out particle concentration(#/cm3) 45

40 C810 C1214 C1618 C1822 Biodiesel type

Figure 3.4 Variations in particle median size among used fuels at (a)1500 rpm 100% load and (b) 2000 rpm 100% load , while (c) shows particle median size variation with total number concentration.

3.3.4 Specific NOx emissions NOx emissions were also found dependent on biodiesel carbon chain and degree of unsaturation (Figure 3.5). Biodiesels with higher degree of saturation and shorter carbon chain length emitted less NOx than biodiesels with relatively longer carbon chain and higher degree of unsaturation. Even NOx emissions from C810 and C1214 found less than diesel especially at higher blends. An interesting trend in NOx emissions also observed among blends of used biodiesels. The usual trend in most of the literature is to increase NOx emissions with the increase of biodiesel percent in the blend [7, 37]. Similar trend we observed here for long chained biodiesels with higher degree of unsaturation i.e. C1618 and C1822. However for saturated and short chained biodiesels i.e. C810 the opposite trend was also observed. NOx formation mostly depends on two phenomena, duration of premixed combustion phase and in cylinder temperature. Biodiesels with higher degree of unsaturation have low cetane number, and undergoes to prolonged premixed combustion favorable for thermal NOx formation. So the higher NOx emissions from C1618 and C1875 are expected

Chapter 3: Particle emissions from biodiesels with different physical properties and chemical composition 95

and due to their higher unsaturation as well as their higher heating value. On the other hand higher degree of saturation, higher cetane number and lower heating value of C810 and C1214 may cause shorter premixed combustion and lower in cylinder temperature responsible for less NOx emissions. Therefore the mystery of higher and lower [38] NOx emissions among different biodiesel studies in the literature may be underlying in biodiesels chemical composition, and the concept of NOx emissions increase in case of biodiesels is not always true. Rather, whether biodiesel will increase or reduce NOx emissions depends upon their chemical composition.

6.0 (a) 4.5

3.0

1.5

0.0 Diesel 6.4 B20 B50 B100 (b) 4.8

3.2 Brake specific NOx(g/kW-hr)

1.6

0.0 C810 C1214 C1618 C1822 Biodiesel type

Figure 3.5 Brake specific NOx emission at (a) 1500 rpm 100% load and (b) 2000 rpm 100% load.

3.3.5 Influence of fuel physical properties and chemical composition on particle emissions To understand what is the relative influence of fuel physical properties as well as chemical composition on particle emissions, PM and PN emissions for all used fuels were plotted against fuel viscosity, surface tension and oxygen contents. Variations in PM and PN emissions with fuel viscosity, surface tension and oxygen contents at 100% load are shown in Figure 3.6 , where the other loads are shown in appendix (Figures S8 to S10). As shown in Figure 3.6 , particle emissions increased with the increase of fuel viscosity and surface tension but only within a specific blend. For higher blend percentages (B50 and B100) there was almost a linear relationship between surface tension, viscosity and particle emissions. On the other hand, for the same viscosity and surface tension, particle emissions were also found to be significantly different among fuel/fuel mixes. It is evident from the literature, both viscosity and surface tensions have noticeable influence on fuel atomization

96 Chapter 3: Particle emissions from biodiesels with different physical properties and chemical composition

process [39], which is a key parameter relative to in-cylinder soot formation. Lower the viscosity and surface tension of fuel, more easily they evaporate, atomize and mix into in-cylinder air, and more complete their combustion are [40, 41]. In this case, the used engine was employed with a common rail injection system where fuel injection pressure was high (around 200 bars). Such a high injection pressure might minimize the effects of small variation in fuel viscosity and surface tension on fuel atomization and subsequent particle emissions. On the other hand, a more consistent negative relationship with R value more than 0.9 was observed between fuel oxygen content and particle emissions. This relationship did not depend on the blend percentage and went well with both PM and PN. Similar reduction in particle emissions with fuel oxygen content has also been reported in the literature [9, 42, 43]. Therefore, this is a clear indication that the fuel chemical composition, particularly the oxygen content, could be more important than its physical properties in terms of engine exhaust particle emissions.

0.0396 0.0352 a 0.0308 b c 0.0264

0.0220

0.0176

0.0132 B100 0.0088 B50 B20

Brake specific (g/kW-hr) PM 0.0044 0.0000

9.90E+013 8.80E+013 d e f 7.70E+013 6.60E+013 5.50E+013 4.40E+013

3.30E+013 B100 B50 2.20E+013 B20

Brake specific PN(#/kW-hr) 1.10E+013 0.00E+000 25 26 27 28 29 30 1 2 3 4 5 2 4 6 8 10 12 14 16 18 2 Fuel surface tension(mN/m) Fuel viscosity(mm /s) Fuel oxygen content(wt%)

Figure 3.6 Variation in specific PM and PN emissions with used fuel surface tension, viscosity and oxygen content ((a), (b), (c) for PM and (d), (e), (f) for PN), Ordinate of (a), (b), (c) and (d), (e), (f) are same where abscissa of (a), (d) and (b), (e) and (c), (f) are same.

Chapter 3: Particle emissions from biodiesels with different physical properties and chemical composition 97

3.3.6 Comparison of engine performance and particle emissions among used biodiesels A comparison between engine performance parameters and particle emissions is shown in Figure 3.7 . In this figure, the vertical axis represents the percentage change of engine power, brake specific fuel consumption and specific PM/PN emissions while the horizontal axis indicates biodiesel proportion in the blends. Neat diesel was used as a reference fuel to calculate the percentage changes. There is a significant difference among all four biodiesels used. For example, C810 provides the highest reduction in particle mass and number but the penalty for that is also highest, around 25% reduction in engine brake power and an additional 25% increase of specific fuel consumption due to that reduced engine power. This fuel and power penalty is lowest for C1618 but particle mass and number reduction is also the lowest in this fuel blend. Therefore it is necessary to make a trade-off between particle emission reduction, fuel and power penalty, ensuring maximum benefit; not just for emission levels but for engine power and fuel economy as well. Considering the aforementioned factors, C1822 seems to have advantage over the rest of the fuels, as it maintains the lowest power and fuel penalty regardless of the blending ratio to diesel and provide a reasonable reduction in engine exhaust particle emissions. The evidence suggests that biodiesels with a longer carbon chain length and higher degree of unsaturation might be a solution to reduce particle emissions to a certain extent with less fuel and engine power penalty.

100 90 80 70 C810 C1214 60 C1618 C1822 50 40 30 20

Reduced(%) 10 0 Power BSFC PM PN Power BSFC PM PN -10 Power BSFC PM PN -20 B100 B50 B20 -30 -40 Increased(%)

Figure 3.7 Comparison of engine performance (power, BSFC) and particle emissions (PM, PN) among biodiesels and their blends where petrodiesel was used as a reference fuel.

98 Chapter 3: Particle emissions from biodiesels with different physical properties and chemical composition

3.4 CONCLUSIONS In conclusion, biodiesel fuels with shorter carbon chain lengths and higher degrees of saturation have more potential to decrease engine exhaust particle emissions. With the increase of carbon chain length and degree of unsaturation, particle emissions also increase. Particle size also depends on type of fuel used. Fuel or fuel mix responsible for higher PM and PN emissions was also found to have a larger particle median size. This indicates that coagulation plays a role in overall engine exhaust particle size. Particle emissions increase linearly with fuel viscosity and surface tension but only for higher diesel-biodiesel blend percentages (B100, B50). On the other side PM emission reduces consistently with fuel oxygen content regardless of the proportion of biodiesel in the blends. High fuel injection pressure used by common rail injection systems might minimize the effects of small variation in fuel viscosity and surface tension on particle emissions. As the fuel oxygen content increases with the decrease of FAME carbon chain length it is not clear whether it is the FAME carbon chain length or the oxygen content that is the driving force which decreases particle emissions. The results support the view that chemical composition (i.e. carbon chain length, degree of unsaturation, oxygen content) of biodiesel is more important than its physical properties (i.e. viscosity, surface tension) in regards to reducing engine exhaust particle emissions.

3.5 ACKNOWLEDGEMENT Authors sincerely acknowledge Mr. Scott Abbett and Mr. Noel Hartnett technicians in the Biofuel Engine Research Facilities (BERF) at Queensland University of Technology (QUT) and PhD student Mohammad Aminul Islam for their valuable support during experimental work. This work was supported by the Australian Research Council under grants LP110200158, DP1097125 DP130104904 and DP120100126.

3.6 REFERENCES 1. Brito, J.M., Belotti, L., Toledo, A.C., Antonangelo, L., Silva, F.S., Alvim, D.S., Andre, P.A., Saldiva, P.H.N., and Rivero, D.H.R.F., Acute Cardiovascular and Inflammatory Toxicity Induced by Inhalation of Diesel and Biodiesel Exhaust Particles. Toxicological Sciences, 2010. 116(1): p. 67- 78. 2. Jacobson, M.Z., Strong Radiative Heating Due to the Mixing State of Black Carbon in Atmospheric Aerosols. Nature, 2001. 409(6821): p. 695-697.

Chapter 3: Particle emissions from biodiesels with different physical properties and chemical composition 99

3. Ristovski, Z.D., Miljevic, B., Surawski, N.C., Morawska, L., Fong, K.M., Goh, F., and Yang, I.A., Respiratory Health Effects of Diesel Particulate Matter. Respirology, 2012. 17(2): p. 201-212. 4. Broday, D.M. and Rosenzweig, R., Deposition of Fractal-Like Soot Aggregates in the Human Respiratory Tract. Journal of Aerosol Science, 2011. 42(6): p. 372-386. 5. Giechaskiel, B., Alfoldy, B., and Drossinos, Y., A Metric for Health Effects Studies of Diesel Exhaust Particles. Journal of Aerosol Science, 2009. 40(8): p. 639-651. 6. Herner, J.D., Hu, S.H., Robertson, W.H., Huai, T., Chang, M.C.O., Rieger, P., and Ayala, A., Effect of Advanced Aftertreatment for Pm and No(X) Reduction on Heavy-Duty Diesel Engine Ultrafine Particle Emissions. Environmental science & technology, 2011. 45(6): p. 2413-2419. 7. Bakeas, E., Karavalakis, G., and Stournas, S., Biodiesel Emissions Profile in Modern Diesel Vehicles. Part 1: Effect of Biodiesel Origin on the Criteria Emissions. Science of the Total Environment, 2011. 409(9): p. 1670-1676. 8. Varuvel, E.G., Mrad, N., Tazerout, M., and Aloui, F., Experimental Analysis of Biofuel as an Alternative Fuel for Diesel Engines. Applied Energy, 2012. 94(0): p. 224-231. 9. Xue, J., Grift, T.E., and Hansen, A.C., Effect of Biodiesel on Engine Performances and Emissions. Renewable and Sustainable Energy Reviews, 2011. 15(2): p. 1098-1116. 10. Lapuerta, M., Armas, O., and Rodríguez-Fernández, J., Effect of the Degree of Unsaturation of Biodiesel Fuels on Nox and Particulate Emissions. SAE Int. J. Fuels Lubr., 2008. 1(1): p. 1150-1158. 11. Surawski, N.C., Miljevic, B., Ayoko, G.A., Roberts, B.A., Elbagir, S., Fairfull-Smith, K.E., Bottle, S.E., and Ristovski, Z.D., Physicochemical Characterization of Particulate Emissions from a Compression Ignition Engine Employing Two Injection Technologies and Three Fuels. Environmental Science & Technology, 2011. 45(13): p. 5498-5505. 12. Salvi, B.L. and Panwar, N.L., Biodiesel Resources and Production Technologies – a Review. Renewable and Sustainable Energy Reviews, 2012. 16(6): p. 3680-3689. 13. Sharma, Y.C., Singh, B., and Upadhyay, S.N., Advancements in Development and Characterization of Biodiesel: A Review. Fuel, 2008. 87(12): p. 2355- 2373. 14. Morshed, M., Ferdous, K., Khan, M.R., Mazumder, M.S.I., Islam, M.A., and Uddin, M.T., Rubber Seed Oil as a Potential Source for Biodiesel Production in Bangladesh. Fuel, 2011. 90(10): p. 2981-2986. 15. Alptekin, E., Canakci, M., and Sanli, H., Evaluation of Leather Industry Wastes as a Feedstock for Biodiesel Production. Fuel, 2012. 95(0): p. 214- 220. 16. Moser and R., B., Impact of Fatty Ester Composition on Low Temperature Properties of Biodiesel–Petroleum Diesel Blends. Fuel, 2014. 115(0): p. 500- 506. 17. Hoekman, S.K., Broch, A., Robbins, C., Ceniceros, E., and Natarajan, M., Review of Biodiesel Composition, Properties, and Specifications. Renewable and Sustainable Energy Reviews, 2012. 16(1): p. 143-169. 18. McCormick, R.L., Graboski, M.S., Alleman, T.L., Herring, A.M., and Tyson, K.S., Impact of Biodiesel Source Material and Chemical Structure on

100 Chapter 3: Particle emissions from biodiesels with different physical properties and chemical composition

Emissions of Criteria Pollutants from a Heavy-Duty Engine. Environmental Science and Technology, 2001. 35(9): p. 1742-1747. 19. Surawski, N.C., Miljevic, B., Ayoko, G.A., Elbagir, S., Stevanovic, S., Fairfull-Smith, K.E., Bottle, S.E., and Ristovski, Z.D., Physicochemical Characterization of Particulate Emissions from a Compression Ignition Engine: The Influence of Biodiesel Feedstock. Environmental Science and Technology, 2011. 45(24): p. 10337-10343. 20. Allan, R., Williams, J., and Rogerson, J., Effect of the Molecular Structure of Individual Fatty Acid Alcohol Esters (Biodiesel) on the Formation of Nox and Particulate Matter in the Diesel Combustion Process. SAE Int. J. Fuels Lubr., 2008. 1(1): p. 849-872. 21. Benjumea, P., Agudelo, J.R., and Agudelo, A.F., Effect of the Degree of Unsaturation of Biodiesel Fuels on Engine Performance, Combustion Characteristics, and Emissions. Energy and Fuels, 2011. 25(1): p. 77-85. 22. Karavalakis, G., Boutsika, V., Stournas, S., and Bakeas, E., Biodiesel Emissions Profile in Modern Diesel Vehicles. Part 2: Effect of Biodiesel Origin on Carbonyl, Pah, Nitro-Pah and Oxy-Pah Emissions. Science of the Total Environment, 2011. 409(4): p. 738-747. 23. Salamanca, M., Mondragón, F., Agudelo, J.R., Benjumea, P., and Santamaría, A., Variations in the Chemical Composition and Morphology of Soot Induced by the Unsaturation Degree of Biodiesel and a Biodiesel Blend. Combustion and Flame, 2012. 159(3): p. 1100-1108. 24. Pham, P.X., Bodisco, T.A., Stevanovic, S., Rahman, M.D., Wang, H., Ristovski, Z.D., Brown, R.J., and Masri, A.R., Engine Performance Characteristics for Biodiesels of Different Degrees of Saturation and Carbon Chain Lengths. SAE Int. J. Fuels Lubr., 2013. 6(1): p. 188-198. 25. Liu, S., Shen, L., Bi, Y., and Lei, J., Effects of Altitude and Fuel Oxygen Content on the Performance of a High Pressure Common Rail Diesel Engine. Fuel, 2013(0). 26. Majewski, W.A. and Jääskeläinen, H., What Is Diesel Fuel. DieselNet Technology Guide Ecopoint Inc 2013. 06a(http://www.dieselnet.com/tech/fuel_diesel.php#properties). 27. Jamriska, M., Morawska, L., Thomas, S., and He, C., Diesel Bus Emissions Measured in a Tunnel Study. Environ. Sci. Technol., 2004. 38(24): p. 6701– 6709. 28. Lapuerta, M., Armas, O., and Rodríguez-Fernández, J., Effect of Biodiesel Fuels on Diesel Engine Emissions. Progress in Energy and Combustion Science, 2008. 34(2): p. 198-223. 29. Surawski, N.C., Miljevic, B., Ayoko, G.A., Elbagir, S., Stevanovic, S., Fairfull-Smith, K.E., Bottle, S.E., and Ristovski, Z.D., Physicochemical Characterization of Particulate Emissions from a Compression Ignition Engine: The Influence of Biodiesel Feedstock. Environmental Science & Technology, 2011. 45(24): p. 10337-43. 30. Pinzi, S., Rounce, P., Herreros, J.M., Tsolakis, A., and Pilar Dorado, M., The Effect of Biodiesel Fatty Acid Composition on Combustion and Diesel Engine Exhaust Emissions. Fuel, 2013. 104: p. 170-182. 31. Heintzenberg, J., Properties of the Log-Normal Particle Size Distribution. Aerosol Science and Technology, 1994. 21(1): p. 46-48. 32. Shi, X., He, K., Zhang, J., Ma, Y., Ge, Y., and Tan, J., A Comparative Study of Particle Size Distribution from Two Oxygenated Fuels and Diesel Fuel.

Chapter 3: Particle emissions from biodiesels with different physical properties and chemical composition 101

Frontiers of Environmental Science and Engineering in China, 2010. 4(1): p. 30-34. 33. Schönborn, A., Ladommatos, N., Williams, J., Allan, R., and Rogerson, J., The Influence of Molecular Structure of Fatty Acid Monoalkyl Esters on Diesel Combustion. Combustion and Flame, 2009. 156: p. 1396–1412. 34. Fisher, B.C., Marchese, A.J., Volckens, J., Lee, T., and Collett, J.L., Measurement of Gaseous and Particulate Emissionsfrom Algae-Based Fatty Acid Methyl Esters. SAE Int. J. Fuels Lubr., 2010. 03(02): p. 292. 35. Wang, J., Wu, F., Xiao, J., and Shuai, S., Oxygenated Blend Design and Its Effects on Reducing Diesel Particulate Emissions. Fuel, 2009. 88(10): p. 2037-2045. 36. Zhu, R., Cheung, C.S., and Huang, Z., Particulate Emission Characteristics of a Compression Ignition Engine Fueled with Diesel–Dmc Blends. Aerosol Science and Technology, 2011. 45(2): p. 137-147. 37. Hoekman, S.K. and Robbins, C., Review of the Effects of Biodiesel on Nox Emissions. Fuel Processing Technology, 2012. 96(0): p. 237-249. 38. Redel-Macías, M.D., Pinzi, S., Ruz, M.F., Cubero-Atienza, A.J., and Dorado, M.P., Biodiesel from Saturated and Monounsaturated Fatty Acid Methyl Esters and Their Influence over Noise and Air Pollution. Fuel, 2012. 97: p. 751-756. 39. Ejim, C.E., Fleck, B.A., and Amirfazli, A., Analytical Study for Atomization of Biodiesels and Their Blends in a Typical Injector: Surface Tension and Viscosity Effects. Fuel, 2007. 86(10–11): p. 1534-1544. 40. Lee, S.-w., Tanaka, D., Kusaka, J., and Daisho, Y., Effects of Diesel Fuel Characteristics on Spray and Combustion in a Diesel Engine. JSAE Review, 2002. 23(4): p. 407-414. 41. Chen, P.-C., Wang, W.-C., Roberts, W.L., and Fang, T., Spray and Atomization of Diesel Fuel and Its Alternatives from a Single-Hole Injector Using a Common Rail Fuel Injection System. Fuel, 2013. 103(0): p. 850-861. 42. S.S. Gilla, A.T., , , J.M. Herrerosb, A.P.E. York, Diesel Emissions Improvements through the Use of Biodiesel or Oxygenated Blending Components. Fuel, 2011(doi: 10.1016/j.fuel.2011.11.047). 43. Rahman, M.M., Stevanovic, S., Brown, R.J., and Ristovski, Z., Influence of Different Alternative Fuels on Particle Emission from a Turbocharged Common-Rail Diesel Engine. Procedia Engineering, 2013. 56(0): p. 381-386.

102 Chapter 3: Particle emissions from biodiesels with different physical properties and chemical composition

3.7 APPENDIX

Figure S1: Brake specific PM emissions at 75%, 50% and 25% loads respectively while the engine operated at 1500 and 2000 rpm.

Chapter 3: Particle emissions from biodiesels with different physical properties and chemical composition 103

Figure S2: Brake specific PN emissions at 75%, 50% and 25% loads respectively while the engine speed was 1500 and 2000 rpm.

104 Chapter 3: Particle emissions from biodiesels with different physical properties and chemical composition

Diesel ) 2.16x107 C810 -3 1.92x107 (a) C1214 (d) 7 cm 1.68x10 C1618 ( 7

p 1.44x10 C1822 1.20x107 9.60x106 7.20x106 4.80x106 dN/d logddN/d 2.40x106

10 20 30 40 50 60 70 8090100 200 300 10 20 30 40 50 60 70 8090100 200 300

7

) 2.16x10 7 -3 1.92x10 7

cm 1.68x10 (e) 7 ( 1.44x10 (b) p 1.20x107 9.60x106 7.20x106 4.80x106 dN/d logddN/d 2.40x106

10 20 30 40 50 60 70 8090100 200 300 10 20 30 40 50 60 70 8090 200 300

7

) 2.16x10 7

-3 1.92x10 1.68x107 (c) (f) cm 7 ( 1.44x10 p 1.20x107 9.60x106 7.20x106 4.80x106 6 dN/d logddN/d 2.40x10

10 20 30 40 50 60 70 8090100 200 300 10 20 30 40 50 60 708090100 200 300 Particle electrical mobility diameter(nm) Particle electrical mobility diameter(nm)

Figure S3: Particle number size distribution for B100, B50, and B20 at 1500 rpm 100% load (a, b, c respectively) and 2000 rpm 100% load (d, e, and f respectively).

Chapter 3: Particle emissions from biodiesels with different physical properties and chemical composition 105

3.0x107 3.0x107 7 7 ) 3 2.5x10 (a) 2.5x10 (d) 2.0x107 2.0x107

(#/cm 7 7 p 1.5x10 1.5x10 1.0x107 1.0x107 5.0x106 5.0x106

dN/d log d 0.0 0.0 10 20 40 60 80 100 200 10 20 40 60 80 100 200 7 7 3.0x10 3.0x10 Diesel 7 7 )

3 2.5x10 2.5x10 C810 7 7 C1214 (e) 2.0x10 (b) 2.0x10 C1618

(#/cm 7 7 p 1.5x10 1.5x10 C1822 1.0x107 1.0x107 5.0x106 5.0x106

dN/d log d 0.0 0.0 10 20 40 60 80 100 200 10 20 40 60 80 100 200 3.0x107 3.0x107 7 7 )

3 2.5x10 2.5x10 7 (c) 7 (f) 2.0x10 2.0x10

(#/cm 7 7 p 1.5x10 1.5x10 1.0x107 1.0x107 5.0x106 5.0x106

dN/d log d 0.0 0.0 10 20 40 60 80 100 200 10 20 40 60 80 100 200 Particle electrical mobility diameter(nm) Particle electrical mobility diameter(nm) Figure S4: Particle number size distribution for B100, B50, and B20 at 1500 rpm 50% load (a, b, c respectively) and 2000 rpm 50% load (d, e, and f respectively).

7 3.80x10 3.80x107

) 7 3 (a) 7 3.04x10 3.04x10 (d) 7 2.28x10 7

(#/cm 2.28x10 p 7 1.52x10 1.52x107 6 7.60x10 7.60x106

dN/d log d 0.00 10 20 40 60 80 100 200 0.00 10 20 40 60 80 100 200 7 7 3.80x10 3.80x10 Diesel ) 7 7 3 3.04x10 3.04x10 C810 (e) (b) C1214 2.28x107 2.28x107

(#/cm C1618 p 7 7 C1822 1.52x10 1.52x10 7.60x106 7.60x106

dN/d log d 0.00 0.00 10 20 40 60 80 100 200 10 20 40 60 80 100 200 3.80x107 3.80x107

) 7 7 3 3.04x10 (c) 3.04x10 (f) 2.28x107 2.28x107 (#/cm p 1.52x107 1.52x107 7.60x106 7.60x106

dN/d log d 0.00 0.00 10 20 40 60 80 100 200 10 20 40 60 80 100 200 Particle electrical mobility diameter(nm) Particle electrical mobility diameter(nm)

Figure S5: Particle number size distribution for B100, B50, and B20 at 1500 rpm 25% load (a, b, c respectively) and 2000 rpm 25% load (d, e, and f respectively).

106 Chapter 3: Particle emissions from biodiesels with different physical properties and chemical composition

Figure S6: Variations in particle median size among fuels at 1500 rpm 75%, 50% and 25% load.

Chapter 3: Particle emissions from biodiesels with different physical properties and chemical composition 107

Figure S7: Variations in particle median size among fuels at 2000 rpm 75%, 50% and 25% load.

108 Chapter 3: Particle emissions from biodiesels with different physical properties and chemical composition

Figure S8: Variation in specific PM emissions with fuel oxygen content at 75%, 50% and 25% load.

Chapter 3: Particle emissions from biodiesels with different physical properties and chemical composition 109

Figure S9: Variation in specific PM emissions with fuel surface tension at 75%, 50% and 25% load.

110 Chapter 3: Particle emissions from biodiesels with different physical properties and chemical composition

Figure S10: Variation in specific PM emissions with fuel viscosity at 75%, 50% and 25% load.

Chapter 3: Particle emissions from biodiesels with different physical properties and chemical composition 111

Chapter 4: The influence of fuel molecular structure on the volatility and oxidative potential of biodiesel particulate matter

A. M. Pourkhesalian1, S. Stevanovic1, F. Salimi1, M. M. Rahman1, H. Wang1, P.X. Pham2, A. R. Masri2, R. J. Brown1, Z. D. Ristovski*1

1 International Laboratory for Air Quality and Health, Queensland University of Technology, GPO Box 2434, Brisbane QLD, 4001, Australia

1Clean Combustion Research Group, The University of Sydney, NSW2006, Australia

*E-mail: [email protected]

Published in Environmental Science and Technology (ES & T) in 2014

Chapter 4: The influence of fuel molecular structure on the volatility and oxidative potential of biodiesel particulate matter 113

Statement of Joint Authorship

Title: The influence of fuel molecular structure on the volatility and OP of biodiesel particulate matter

Authors: A.M. Pourkhesalian, S. Stevanovic, F. Salimi, M.M. Rahman, H. Wang, P.X. Pham, S.E. Bottle, A.R. Masri, R.J. Brown, Z.D. Ristovski

A. M. Pourkhesalian: Performed the measurements and data analysis and wrote the manuscript.

Signature: Date: 05-03-2015

S. Stevanovic: Contributed in design of the experimental program, assisted in data analysis and write up the manuscript.

F. Salimi: contributed in data analysis.

Md Mostafizur Rahman: Contributed in measurement, data analysis and assisted in write up the manuscript.

H. Wang: Contributed in emission measurement and reviewed the manuscript

P. X. Pham: Contributed in emission measurement and reviewed the manuscript.

A. R. Maasri: Supervised the overall study design, set up and experiments, reviewed the manuscript

R. J. Brown: Supervised the overall study design, set up and experiments, reviewed the manuscript

Z. D. Ristovski: Supervised the overall study design, set up and experiments, reviewed the manuscript.

The authors listed above have certified that:

1. They meet the criteria for authorship in that they have participated in the conception, execution, or interpretation, of at least that part of the publication in their field of expertise;

114Chapter 4: The influence of fuel molecular structure on the volatility and oxidative potential of biodiesel particulate matter 115

2. They take public responsibility for their part of the publication, except for the responsible author who accepts overall responsibility for the publication;

3. There are no other authors of the publication according to these criteria;

4. Potential conflicts of interest have been disclosed to (a) granting bodies, (b) the editor or publisher of journals or other publications, and (c) the head of the responsible academic unit, and

5. They agree to the use of the publication in the student’s thesis and its publication on the Australasian Research Online database consistent with any limitations set by publisher requirements.

Principal Supervisor confirmation

I have sighted email or other correspondence from all Co-authors confirming their certifying authorship.

Name Zoran Ristovski

Signature Date

Chapter 4: The influence of fuel molecular structure on the volatility and oxidative potential of biodiesel particulate matter 115

Title: The influence of fuel molecular structure on the volatility and oxidative potential of biodiesel particulate matter

A.M. Pourkhesalian1, S. Stevanovic1, F. Salimi1, M.M. Rahman1, H. Wang1, P.X. Pham2, S.E. Bottle1, A.R. Masri2, R.J. Brown1, Z.D. Ristovski1*

1 ILAQH and BERF, Queensland University of Technology, Brisbane, QLD 4001, Australia

2 Clean Combustion Research Group, The University of Sydney, NSW 2006, Australia

*Corresponding Author: Z.D. Ristovski,

Email: [email protected]

Tel.: +(61)731381129.

116Chapter 4: The influence of fuel molecular structure on the volatility and oxidative potential of biodiesel particulate matter 117

TOC art

Abstract

We have studied the effect of chemical composition of biodiesel fuel on the physical (volatility) and chemical (reactive oxygenated species concentration) properties of nanoparticles emitted from a modern common-rail diesel engine. Particle emissions from the combustion of four biodiesels with controlled chemical compositions and different varying unsaturation degrees and carbon-chain lengths, together with a commercial diesel, were tested and compared in terms of volatility of particles and the amount of reactive oxygenated species carried by particles. Different blends of biodiesel and petrodiesel were tested at several engine loads and speeds. We have observed that more saturated fuels with shorter carbon chain lengths result in lower particle mass but produce particles that are more volatile and also have higher levels of Reactive Oxygen Species (ROS). This highlights the importance of taking into account metrics that are relevant from the health effects point of view when assessing emissions from new fuel types.

Keywords: Diesel exhaust nanoparticles, ROS, Volatility, conventional diesel fuel, methyl ester biodiesel, carbon chain length, level of saturation.

Chapter 4: The influence of fuel molecular structure on the volatility and oxidative potential of biodiesel particulate matter 117

4.1 INTRODUCTION

Alternative and renewable sources of fuels draw a substantial amount of attention motivated by the constant movement towards more stringent regulation against diesel emissions, the rising price of crude oil, and the vulnerability of fossil fuel resources. Biodiesel is currently one of the most promising alternatives to fossil fuels, and depending on its chemical and physical properties, may be considered as a suitable choice for blending with diesel fuel to account for increasing fuel demand [1-3]. Biodiesel is a processed fuel which is derived from biological sources by trans- esterifying vegetable oils, animal fats or algae [3]. In terms of chemical composition, biodiesel fuels are mono-alkyl esters of fatty acids [3]. Most of the biodiesel fuels are composed of Fatty Acid Methyl Esters (FAMEs) [3]. The reasons that make biodiesels a suitable and practical fuel choice are numerous. They can be derived from a number of feed stocks such as rapeseed, soybean, palm, waste cooking oil, tallow, jatropha, algae etc [1-3].Biodiesels are renewable and degradable [2, 4]. Lack of polycyclic aromatic hydrocarbons reduces Hydro carbons (HC) and Carbon monoxide (CO) concentrations in the exhaust [5]. In addition, the oxygen content of biodiesel plays an important role in reducing black carbon (BC) formation when local oxygen concentration decreases due to a diffusion combustion [5-7]. The higher cetane number of biodiesel leads to a more complete combustion [8] which in turn results in higher levels of NOx concentration in the exhaust gas [9].

Recently, particulate matter (PM) produced by biodiesel combustion has become a popular research topic [10-15]. This is understandable considering that emissions from diesel engines have been recently declared as carcinogenic [16]. It is commonly seen that biodiesel causes less PM mass concentration in the exhaust gas [10, 17, 18]. However, some studies have reported an increase in the particle number concentration [11, 19, 20] as well as particle number per unit of particle mass [21]. An additional concern was the observation that non-petroleum diesel fuels emit excessive amounts of volatile compounds upon combustion, which potentially results in more toxic emissions [14, 22, 23]. It is also reported that combustion of biodiesel generates smaller particles that can penetrate deeper in lungs causing inflammation at the sites of deposition [12-14, 22].

The quality of combustion is a key factor that affects all the outputs of an engine. Viscosity, surface tension, density, cetane number, low-temperature

118Chapter 4: The influence of fuel molecular structure on the volatility and oxidative potential of biodiesel particulate matter 119 properties (cloud point, pour point, etc. ) [2, 8], heating value and lubricity [8] are amongst the physical parameters of the fuel affecting the combustion process. On the other hand, chemical properties such as carbon chain length, number of double bonds and the amount of oxygen borne by the fuel may also affect the aforementioned physical properties and in turn, the combustion process. There have been a growing number of publications that study the correlation between fuel composition and engine emission, especially diesel particulate matter (DPM) [9, 17, 24]. The relationship between biodiesel chemical properties and resulting emissions were investigated [2, 5, 8, 9, 24, 25],and it has been found that blends with higher oxygen content produce less PM but more volatile particles and increased levels of NOx [5, 9, 25]; while biodiesel fuels with longer carbon chain length led to more PM [8, 24].

DPM is well known for its adverse effects on humans, animals and the environment [26-28]. Recently, based on the current body of knowledge, the International Agency for Research on Cancer (IARC) has labeled diesel exhaust as carcinogenic within class 1 [16]. While the mechanism by which DPM causes the adverse health effects are not known, a great deal of the harmful effects relate to its ability to cause oxidative stress [29-35]. Therefore, measurement of oxidative potential (OP), expressed through ROS concentration, can be used as a good estimate for its reactivity and toxicity. Based on the data provided in the literature so far, it remains unclear which chemical species contribute to the measured redox potential and the overall toxicity. Several studies in this area showed that [23, 29, 36] the organic fraction, more precisely semi-volatile component of PM, correlated well with the measured OP. Stevanovic et al. [29] further showed that the oxygenated fraction of the semi-volatile component of DPM was most responsible for the OP of DPM.

The primary objective of this work is to critically examine the influence of the carbon chain length and saturation levels of biodiesel fuel molecules on the overall volatility (OV) and OP of DPM. All the work has been undertaken by investigating particle emissions from a common-rail engine using four palm oil biodiesels and different blend percentages. All tested biodiesels were FAMEs with controlled fatty acid composition. This enabled us to assign the influence of a particular parameter more easily. This paper is an extension of a previous study[37] where the engine performance characteristics and emissions were investigated and presented, including some preliminary findings for the ROS emission, particularly for pure biodiesel. It

Chapter 4: The influence of fuel molecular structure on the volatility and oxidative potential of biodiesel particulate matter 119

should be noted that the results for B100 are reproduced here for comparison purposes. Furthermore, the paper elaborates on these findings and new analysis in terms of physio-chemical properties of the fuels and their blends.

4.2 EXPERIMENTAL SETUP AND METHODOLOGY

A schematic of the experimental setup is can be found in Figure 4 .7 in the supporting information. The test bed included a diesel engine coupled to a dynamometer and the accompanying instruments for PM and gas measurement. The specifications of the engine and dynamometer are shown in Supporting Information, Table 3 -1. Raw diesel exhaust was sampled from the exhaust line and passed through a dilution tunnel and a Dekati ejector diluter both of which were fed with HEPA- filtered air. A TSI DustTrak (Model 8530) measured the mass of particles. Gas analyzers, consisting of a NDIR (Non-dispersive infra-red) CAI 600 series CO2 and CO analyzer and a CAI 600 series CLD (Chemiluminescence detector) NOx analyzer, measured CO, CO2 and NOx concentrations before the dilution system.

Also, to determine the dilution ratio, a SABLE CA-10 CO2 analyzer was used to measure CO2 concentrations after each dilution step. Real-time measurements of black carbon concentrations were performed by an Aethalometer (AE33, 7- Wavelength). A Scanning Mobility Particle Sizer (SMPS TSI 3080, with a 3022 CPC) measured the size distribution of diesel exhaust. A Volatility Tandem Differential Mobility Analyzer (V-TDMA) consisting of an electrostatic classifier, a thermo-denuder and an SMPS (in-house designed column with a 3010 CPC) measured the volatile content of particles. The V-TDMA gave the change in particulate diameter for six pre-selected sizes: 30, 60, 90, 120, 150, 200 and 220 nm after they passed through the thermo-denuder with the temperature set at 300 °C. The flow rate through the thermo-denuder was kept constant at 1 lpm, which lead to a residence time of around 2 seconds. As the concentration of particles entering the thermo-denuder is rather small, there was no need to equip the thermo-denuder with a charcoal section. Absorption of the evaporated semi-volatiles on the walls of the thermo-denuder was sufficient to reduce the vapor pressure and prevent them re- condensing on to the existing non-volatile particles [38].

The volumetric volatile fraction (VVF) for each size was calculated from the difference in diameter of particles before and after the thermo-denuder[39].

120Chapter 4: The influence of fuel molecular structure on the volatility and oxidative potential of biodiesel particulate matter 121

This study approximates the fractal-like diesel aerosols by spherical particles with the corresponding electrical mobility diameter. Although the mentioned assumption is not accurate for larger agglomerates of diesel particles, it would not distract the results as this study aims at comparing various fuels and not at quantification of the amount of volatile material emitted by combustion of biodiesel.

The total amount of volatile matter is calculated for every sampling point and the parameter used for its quantification is the Overall Volatility (OV). This was previously introduced in the literature by Giechaskiel [40], where it was used for the calculation of the non-volatile fraction of PM. More details of the concept and the procedure used for calculation of OV can be found in the supplement material.

The BPEA molecular probe (bis(phenylethynyl) anthracene-nitroxide) was applied in-situ to assess the OP of different fuel stocks. Samples were collected by bubbling aerosol through an impinger which contains 20 mL of 4 μM BPEA solution (using an AR grade dimethylsulphoxide as a solvent). More details on the ROS sampling methodology, theory behind its application and proof of concept in the case of various combustion sources can be found in Miljevic et al. and Stevanovic et al. [41-45].

4.3 FUEL SELECTION

The present study investigates four biodiesels of controlled composition, differing in levels of unsaturation, carbon chain length and oxygen content. Commercial petrodiesel was used for blending. Some of the fuel properties were measured experimentally in a previous study [46], and others were estimated based on the chemical composition of methyl-esters which are shown in Supporting Information,

Table 3-3. We labeled the tested FAMEs as C810, C1214, C1618 and C1875, based on the number of carbon atoms in the most abundant fatty acid in that particular biodiesel stock. The Iodine values of C810 and C1214 are very small, implying that they are almost fully saturated; but the large difference in saponification values shows that the carbon chain length of their molecules is quite different and, so is the oxygen content. The Iodine value is used in the literature as the indication of the unsaturation while Saponification value presents a measure of the average molecular weight or chain length [46]. Conversely, C1618 and C1875

Chapter 4: The influence of fuel molecular structure on the volatility and oxidative potential of biodiesel particulate matter 121

have different iodine numbers but pretty close saponification values; which indicates that the carbon chain length of their molecules is close but with different levels of unsaturation. We chose these controlled fuels to be able to differentiate between the effect of unsaturation and carbon chain length.

The engine operated on blends of 100%, 50%, 20% and 0% of biodiesel and petrodiesel namely B100, B50 and B20 and B0, respectively. We conducted tests at idle, quarter (1500 and 2000 rpm) and full load (1500 rpm).

4.4 RESULTS AND DISCUSSION

4.4.1 Volatility measurements To estimate the thermo-denuder temperature at which the volatile content of DPM is removed, measurements of the temperature dependent volatility were conducted for 30 and 150 nm particles using B0 at idle (Figure 4 .8 in the Supporting Information). The temperature was ramped from room to 280°C in steps of 20°C. By 180°C the 30nm particles were almost completely evaporated and the larger 150nm particles reached a stable size. Therefore we can be confident that at the set temperature of 300°C all of the volatile material has been evaporated from the diesel particles and there will be no effect of the size and residence time of particles [47].

The VVF, calculated using Eq 4.1 (in Supporting Information section), for all the measured particle sizes and all the fuels and blends is shown in Figure 4 .9 in the Supporting Information. Out of the four loads studied, the highest volatility was observed for particles produced during idling. This was valid for all the measured blends and fuels. In addition to the idling conditions, higher volatility was only observed at full load, while particles produced at quarter load and at both speeds did not show significant volatility. For almost all of the fuels and blends tested and for most of the loads, higher volatility was observed for smaller particles. This is the most obvious for idling conditions where 30nm particles were mainly volatile with a VVF of over 75%. Similar dependence on the volatility as a function of particle size has been observed previously [48, 49].

To get a better insight into the influence of the fuel type and blend on the VVF, from the data presented in Figure 4 .9 (in Supporting Information), we calculated the OV for each of the 3 fuel blends (B20, B50 and B100) and for all four tested modes

122Chapter 4: The influence of fuel molecular structure on the volatility and oxidative potential of biodiesel particulate matter 123 being presented in Figure 4 .1. Unfortunately the volatility for C1214 at B100 was not measured.

Figure 4.1 OV for all blends and modes tested. Rows present data measured at the same load, while columns data are measured at the same blend percentage. Different fuels are sorted along the x-axis according to their carbon chain length going from the shortest to the longest.

What is obvious from all of the graphs is that for all of the higher blends (B50 and B100) the measured volatility is larger than that of B0. Furthermore an increase in the volatility with the increase of the biodiesel blend percentage for the same fuel type can also be observed. While previously [23] have observed an increase in the volatility with increase in the blend percentage, the results did not show a stock dependency. In these measurements we do see a dependence on the feedstock, with the fuel with the shortest carbon chain length (C810) producing the most volatile particles for all the tests except one, B20 at quarter load and 2000 rpm. Although C810 did not produce particles with the largest OV the observed values were all within the measurement error. Also, in the case of the higher blend percentages B50 and B100, a decrease in the OV is followed by an increase in the carbon chain length. This refers to all the loads tested but is most significant at idle and full load where a larger OV was measured. This is in contrast with a previous observation [50]

Chapter 4: The influence of fuel molecular structure on the volatility and oxidative potential of biodiesel particulate matter 123

where an increase in the volatile organic fraction (VOF) was observed for an increase in the carbon chain length. It is worth mentioning that Pinzi et al. [50] have used a mechanical direct-injection engines, as compared to the common rail engine used in our case, for which the fuel physical properties have a significant influence on the spray pattern and therefore emissions.

Some previous measurements conducted for petrodiesel point out that the volatile part of DPM is mainly due to the lubricating oil [51-53], and some recent measurements in the vicinity of busy roads also show that the main contributor to the primary organic aerosol (POA) is lubricating oil from both gasoline and diesel powered vehicles [54]. If that was the case, in our observations, the increase in the OV could be interpreted due to a decrease in the total non-volatile mass (volume) as the lubricating oil contribution would be the same for all of the different fuels and blends tested.

On the other side, measurements on biodiesel fuels show a large fraction of fuel derived organics contributing to the volatile part of DPM [55]. This contribution can be as large as 90% for B50 blends [56].

4.4.2 BC Emissions The change in the OV could be either due to the change in the emission of primary organic aerosol (POA) or could be due to the change in the emission of the non-volatile soot component of DPM with the change of the carbon chain length. A good measure of the nonvolatile mass emitted from a diesel engine is the BC concentration. Figure 4 .2 shows the BC emission factor (EF) for each of the three fuel blends (B20, B50 and B100) and for all 4 tested modes.

124Chapter 4: The influence of fuel molecular structure on the volatility and oxidative potential of biodiesel particulate matter 125

Figure 4.2 BC Emission factors for all blends and modes tested. The emission factor for B0 at each mode is shown as a dashed line. Rows present data measured at the same load, while columns data are measured at the same blend percentage. Different fuels are sorted

We can see that all blends and all fuels reduce the BC emissions as compared to B0. Higher blend percentages of biodiesel (B50 and B100) show a clearer trend where the fuel with the shortest carbon chain length has the smallest BC emissions. An increase in BC emission is observed with the increase in carbon chain length up to C1618, afterwards a decrease is observed for C1875. While there is not a big difference in the carbon chain length of the FAMEs that make C1618 and C1875, as both fuels contain large fraction of oleic acid of 53 and 63.5% respectively, the later fuel contains additional 28.1% of FAMEs that are fully saturated (linoleic and linolenic acids). A similar increase in soot emission with the increase in carbon chain length and a decrease with the level of saturation was also observed by Pinzi et al. [50], although the later was not as obvious.

4.4.3 ROS measurements OP measurement, expressed through ROS concentration of PM can be used as a good estimate for its reactivity and toxicity. An in-house developed profluorescent molecular probe BPEAnit was applied in an entirely novel, rapid and non-cell based

Chapter 4: The influence of fuel molecular structure on the volatility and oxidative potential of biodiesel particulate matter 125

way to assess particulate OP. Based on the data provided in the literature so far [45], there are some uncertainties related to the nature of chemical species responsible for the measured redox potential and overall toxicity. However, the majority of research in this field reported that organic fraction, more precisely semi-volatile organic content, is in a good correlation with ROS concentration [23, 36, 57].

Figure 4.3 illustrates the OP of particles for the same loads and blends as for the volatility measurements that were shown in Figure 4.1 (note logarithmic scale is used here). In idle mode, the ROS concentration is much higher than in the other cases. This may be a result of the additional combustion of lubricating oil, which would subsequently increase the overall organic content. Results do not show blend dependency in this mode, although it is clearly visible that combustion of C810 fuel generates the largest amount of ROS for all three blends and combustion of the fuel with the longest carbon chain length C1875 smallest amount of ROS. This could be a consequence of the way the OP is presented as the concentration of ROS per mass of emitted particles. As the fuels with the highest oxygen content have the lowest mass emission even for the same (or similar) ROS concentration they will exhibit the highest OP when calculated per unit mass of PM. For the other three loads and for most of the blends C810 produced higher ROS emissions. This is most apparent for B100, but it can also be observed even in 20 percent blends. While there is an increase in ROS emissions for 50 and 100 percent blends for all the tested fuels, C810 shows the largest increase of almost two orders of magnitude. This is a reasonable result as the C810 fuel molecules are of the shortest chain, fully saturated and with the largest oxygen content. These three parameters have synergetic effect in terms of prediction of resulting OP as have been seen previously as well [37].

Another important parameter that will influence resulting OP is the load on the engine. Generally, engine operating conditions significantly affect physicochemical properties of emitted PM and their subsequent toxicity that were demonstrated to change with varying engine loads and speeds. Also, OP tends to decrease with increasing engine load. This result can be explained by a more complete combustion that results from higher temperatures and lower air to fuel ratio at higher loads. In addition, at full load, the amount of lubricating oil that is available for combustion is minimal.

126Chapter 4: The influence of fuel molecular structure on the volatility and oxidative potential of biodiesel particulate matter 127

Figure 4.3 OP of PM for all four fuels, for all the blends and loads.

4.4.4 The role of oxygen content As seen previously [58], out of all the fuel physical and chemical properties, oxygen content has the most significant influence on particle mass and number emissions. To further investigate the influence of the fuel oxygen content on particle volatility and ROS concentration, we have presented the dependence of the BC emission (

Figure 4 .4), ROS emissions (

Figure 4 .5a) and OV (

Figure 4 .5b) as a function of fuel oxygen content. This is done for each of the four modes tested. Oxygen content is calculated for all the fuels and blends assuming that oxygen content of base-line diesel fuel is considered to be zero. As all the fuels are methyl esters with varying carbon chain lengths, oxygen percentage increases with the decreasing carbon chain length in respective molecules. Consequently, C810 is the fuel with the highest oxygen content, followed by C1214, C1618 and finally C1875, as the fuel with the longest carbon chain length as well as the most unsaturated fuel.

Chapter 4: The influence of fuel molecular structure on the volatility and oxidative potential of biodiesel particulate matter 127

The relationships between the variables were analyzed using the linear regression or generalized linear model with a log link function. The model assumptions were validated using the residuals versus fit values and QQ plots. The resulting fit functions, and their 95% condense intervals, indicate the relationship between the variables. Modelling and visualizations were performed using the ggplot2 package in R [59].

Figure 4.4 Dependence of the BC emissions factor on fuel oxygen content for the four modes tested.

Figure 4 .4 illustrates the decrease in the emitted BC mass with the increase of oxygen content in the fuels. The effect of oxygen content on BC formation and emissions was explained in the work of Gill [60][60][60][60][60][60][60]et al. [61]. It was suggested that oxygen present in the fuel molecule could promote the diffusion phase combustion, where soot is mainly produced, even though the diffusion phase is extended. Increased diffusion phase temperature will promote soot oxidation. Therefore the increase in oxygen content will result in a decrease in the

128Chapter 4: The influence of fuel molecular structure on the volatility and oxidative potential of biodiesel particulate matter 129 soot or BC emissions. It is important to note that there was a linear relationship between the oxygen content and BC. Apart from idle, all other three modes showed very similar trends; so on the average, BC is reduced by 270 (mg/kg fuel) per every 10 percent of increase in oxygen content in the fuel.

Figure 4.5 Dependence of the a) Overall volatile content and b) ROS concentration on fuel oxygen content for the four modes tested. R2 shown in the graphs are McFadden's pseudo R2.

As can be seen from

Figure 4 .5b) increasing the oxygen content leads to an increase in the ROS concentration, and therefore the OP of particles, for all of the modes tested. There is also an increase in the overall volatile content, or OV, as shown on

Figure 4 .5a). It is interesting to observe that both the OP and OV do not seem to have a linear dependence on the oxygen content, as with BC, but exhibit a more complex behavior with an exponential dependency. The data also shows much more scattering for OP than the BC data and is strongly biased for the two 1500 rpm modes (full and quarter) with the measurement of B100 C810 that had the highest oxygen content. Although not explicitly discussing the change in the volatile content with oxygen content, Pinzi et al. [50] saw an increase in the volatile content with an increase in the carbon chain length of the fuel. As fuels with longer carbon chain

Chapter 4: The influence of fuel molecular structure on the volatility and oxidative potential of biodiesel particulate matter 129

length have lower oxygen content they essentially observe an opposite effect. As mentioned previously, their trend could be due to the mechanical direct injection diesel engine that they have used whose emissions are more sensitive to the physical properties of the fuels as compared to a high-pressure common rail engine used in our case. In any case, as most of the new engines used today are common rail our test seems to be more relevant. The increase in the OV can be due to either the decrease in the available non-volatile soot onto which the volatiles condense or an actual increase in the amount of volatile material. If the trend displayed in

Figure 4 .5a) was linear then one of the above effects would have been dominant. As the trend displayed on

Figure 4 .5b) is nonlinear, in fact exponential, then one can assume that there is a combination of both effects: reduction in the mass of soot and increase in the fuel derived volatile (organic) material.

Similarly there are two processes that could lead to the increase in the OP. The first is the increase in the amount of ROS that is carried by the particles that could also be seen as the increase in the OV. The second is the decrease in the total mass of DPM as the OP is calculated per unit mass of particles. If the increase in the OP were due only to the decrease in the total mass then the dependence on oxygen content would be a reciprocal function and not an exponential. It is clear that the oxygen content plays a major role in the OP of DPM and most likely this is a combination of the decrease of the total DPM and an increase of the volatile organic fraction but also could be due to a change in the chemical composition of the organic fraction.

Figure 4 .6 shows the dependence of the OP expressed through the ROS concentration as a function of the OV for all of the 4 modes tested. If the chemical composition of the volatile organic material that contributes to the OV was similar for the same mode than one would expect a linear dependence between the ROS concentration and OV. While this has been observed for the same wood combustion conditions [41] and in some cases even for the organic carbon in diesel exhaust [36] in general, various fuels and loads produce a VVF with significantly different OP [29].

As can be seen from

130Chapter 4: The influence of fuel molecular structure on the volatility and oxidative potential of biodiesel particulate matter 131

Figure 4 .6, it seems that idle mode and quarter2000 mode can also be modeled with linear fit. So, a linear fit was also applied to those panels (i.e. idle1200 and quarter2000) to further explore the nature of this correlation. Calculated R2 values for linear fit were 0.45 and 0.27 for idle1200 and quarter2000 respectively. This indicates the existence of an unspecific pattern (neither exponential nor linear). The explanation for this may be that for idle 1200 most of the volatile species contribute to the measured OP. However these compounds are not equally reactive but still very volatile. In the case of quarter load at 2000 rpm, as the consequence of an incomplete combustion, a number of unburnt hydrocarbons are expected to be present in the exhaust. These hydrocarbons will contribute to the OV values and do not necessarily need to carry measurable oxidative potential. This phenomenon would lead to an unspecific pattern (neither exponential nor linear).

Figure 4.6 Dependence of the ROS concentration on the overall volatile content (OV) for the 4 modes tested. R2 shown in the graphs are McFadden's pseudo R2.

The conclusion that follows the investigation of the fuel type on the OV and related OP is that generally, the OP is ultimately coupled to the structure of fuel molecules, especially its oxygen content, chain length and level of unsaturation. It also highlights the fact that this relationship is dependent on the chemical

Chapter 4: The influence of fuel molecular structure on the volatility and oxidative potential of biodiesel particulate matter 131

composition of volatile matter. Even though complete combustion can yield volatile paraffins, their OP is very low compared to oxygenated organic aerosols.

It appears that chemical composition of volatile matter, which contributes to high levels of ROS, is different in partial and full load and therefore a more detailed chemical analysis of the organic content of DPM is required to shed more light on the toxicity of DPM. The findings of this research show that the more saturated fuels caused more potentially toxic substances in the exhaust; and even though the more oxygenated fuels decreased PM, they also led to higher levels of ROS in the particles. It was established that the majority of the redox activity is due to the amount of organics in the PM. This highlights the fact that “less PM” does not necessarily mean less harmful effect for the environment or less toxicity for humans and that new metrics (taking into account the potentially toxic parts of DPM) should be considered.

4.5 ACKNOWLEDGMENTS

The authors would like to acknowledge the support from the Australian Research Council under Grants LP110200158, DP1097125, DP130104904, DP110105535 and DP120100126. Laboratory assistance from Mr Noel Hartnett and Mr Scott Abbett were also appreciated.

4.6 SUPPORTING INFORMATION AVAILABLE

Engine specifications, fuels properties, schematic of experimental setup along with a V-TDMA scan and explanation of VVF and OV concept and estimation procedure can be found in Supporting Information. This information is available free of charge via the Internet at http://pubs.acs.org/

132Chapter 4: The influence of fuel molecular structure on the volatility and oxidative potential of biodiesel particulate matter 133

4.7 SUPPORTING INFORMATION

Figure 4.7 Experimental setup, HEPA: High-Efficiency Particulate Air Filter; EC: Electrostatic Classifier; TD: Thermo-Denuder; V-TDMA: Volatility Tandem Differential Mobility Analyzer; CPC: Condensation Particle Counter; SMPS: Scanning Mobility Particle Sizer.

Figure 4.8 Dependence of the particle volume on the temperature of the thermodenuder for two particle sizes: circles 30nm and triangles 150 nm.

Chapter 4: The influence of fuel molecular structure on the volatility and oxidative potential of biodiesel particulate matter 133

Figure 4.9 Volumetric Volatile Fraction (VVF) for all pre-selected particle sizes and all fuel blends. Note that graphs in the same columns are for the same fuel blend percentages and graphs in the same rows for the same biodiesel fuel. All the graphs in the first column are the same as they are forB0 (pure petrodiesel) and are there only for comparison with other blends. Different engine loads are presented with different symbols. Error bars were omitted, as they would clutter the figures.

4.7.1 Volatile Volumetric Fraction (VVF) The volumetric volatile fraction ( ) for each size was calculated from the difference in diameter of particles before𝑉𝑉𝑉𝑉𝑉𝑉 (d0) and after the thermo-denuder (d300). This difference was considered as the thickness of the volatile layer that coats the soot core of the particles. Substituting the measured diameters into Eq 4.1 gave the VVF of the particles as

(%) = × 100 = 1 3 × 100 4.1 𝑉𝑉0 − 𝑉𝑉300 𝑑𝑑300 𝑉𝑉𝑉𝑉𝑉𝑉 � 0 � � − � 0 � � 𝐸𝐸𝐸𝐸 4.7.2 Overall volatility𝑉𝑉 𝑑𝑑 To be able to compare the volatile fraction of diesel exhaust emitted by different fuels used, this paper uses a quantity called Overall Volatility (OV). This section elaborates the concept of this quantity and how to estimate it.

134Chapter 4: The influence of fuel molecular structure on the volatility and oxidative potential of biodiesel particulate matter 135

As it is well established, volatility data collected from V-TDMA can give a good estimation of how volatile certain particle sizes are. For instance, to know the volatile fraction of the particles with the diameter of 30 nm, a mono-dispersed stream of the aerosol containing only 30 nm particles is prepared by passing the poly dispersed aerosol through an Electrostatic Classifier (EC); then the mono-disperse stream passes a thermo-denuder for the volatile fraction to be stripped off the particles. Later, it goes to am Scanning Mobility Particle Sizer (SMPS) where the size distribution of the thermo-denuded aerosol is measured. Assuming the particles to be spherical and having the diameter of the particles before and after the thermo- denuder, the volatile fraction of the particles can be estimated. The process can be repeated for a range of diameters to cover the whole size distribution of the initial poly-disperse aerosol. So, in the end, by having the diameter before and after the thermo-denuder we will have the thickness of the organic layer covering the soot core of the particle, the volume of volatile matter in the particle and the volatile fraction of particles in each pre-selected sizes. For instance, in our research, we measured the volatility of 30, 60, 90, 120, 150, 200 and 220 nm particles. This range of particle diameters virtually covers the size distribution of diesel exhaust emitted from the combustion of petrodiesel and biodiesel. The V-TDMA results are appropriate to make comparisons between two sizes within an aerosol or to compare the volatile fraction of particles with the same size but from different aerosols. For example, it can be said that the 30 nm particles from petrodiesel are less volatile than the particles with the same size from bio-diesel. However, V-TDMA data is not enough to know whether or not biodiesel causes more volatile PM since it does not take into consideration the concentration of particles. For example it cannot be said that biodiesel caused more volatile particulate matter than petrodiesel merely based on V-TDMA results and estimation of VVF. In order to compare the total volatile matter in the particulate matter from two or more aerosols, we also need to take into account the number concentration of particles in the aerosol. Hence, to resolve this issue and make the comparisons easier, we estimated the total volume of organic matter which is emitted by the combustion of diesel fuels. The following is the procedure to estimate OV.

As it was defined, OV is the fraction of total organics measured in the particulate phase to the total measured volume of particulate matter; that is:

Chapter 4: The influence of fuel molecular structure on the volatility and oxidative potential of biodiesel particulate matter 135

(%) = , 𝑉𝑉𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜 𝑂𝑂𝑂𝑂 𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 = 𝑉𝑉 4.2

𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣 𝑜𝑜𝑜𝑜 𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜 𝑖𝑖𝑖𝑖 𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝 𝑝𝑝ℎ𝑎𝑎𝑎𝑎𝑎𝑎 𝐸𝐸𝐸𝐸 To determine𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 the𝑣𝑣𝑣𝑣𝑣𝑣 total𝑣𝑣𝑣𝑣 𝑣𝑣volume𝑜𝑜𝑜𝑜 𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝 of particulate𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚 matter:

= (

𝑉𝑉𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠× 𝑜𝑜𝑜𝑜 𝑡𝑡ℎ 𝑒𝑒 𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣 𝑜𝑜𝑜𝑜 𝑎𝑎 𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝 𝑤𝑤𝑤𝑤𝑤𝑤ℎ 𝑎𝑎 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 𝑠𝑠𝑠𝑠)𝑠𝑠 𝑠𝑠 So we have: 𝑛𝑛𝑛𝑛𝑛𝑛𝑛𝑛𝑛𝑛𝑛𝑛 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 𝑜𝑜𝑜𝑜 𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝 𝑜𝑜𝑓𝑓 𝑡𝑡ℎ𝑎𝑎𝑎𝑎 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠

= × 4.3

𝑉𝑉𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 ��𝑉𝑉𝑑𝑑𝑑𝑑=𝑖𝑖 𝑁𝑁𝑑𝑑𝑑𝑑=𝑖𝑖� 𝐸𝐸𝐸𝐸 Where = 30, 60, 90, 120𝑖𝑖 , 150, 200, 220.

To calculate𝑖𝑖 total organics measured in the particulate phase:

,

=𝑉𝑉𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜 ( × 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 𝑜𝑜𝑜𝑜 𝑡𝑡ℎ𝑒𝑒 𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣 𝑣𝑣 𝑜𝑜𝑜𝑜 𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜 𝑙𝑙𝑙𝑙𝑙𝑙 𝑙𝑙𝑙𝑙 𝑜𝑜𝑜𝑜 𝑎𝑎 𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝) 𝑤𝑤𝑤𝑤𝑤𝑤ℎ 𝑎𝑎 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 𝑛𝑛𝑛𝑛𝑛𝑛So𝑛𝑛 𝑛𝑛𝑛𝑛now𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 we have the above𝑜𝑜𝑜𝑜 𝑝𝑝 𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝expression𝑤𝑤 𝑤𝑤𝑤𝑤inℎ math:𝑡𝑡ℎ𝑎𝑎𝑎𝑎 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠

, = , × 4.4 𝑉𝑉𝑉𝑉𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜 ��𝑉𝑉𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜 𝑑𝑑𝑑𝑑=𝑖𝑖 𝑁𝑁𝑑𝑑𝑑𝑑=𝑖𝑖� 𝐸𝐸𝐸𝐸 Where = 30, 60, 90, 𝑖𝑖120, 150, 200, 220.

So, to estimate𝑖𝑖 OV, we need to know the thickness and the volume of organic layer covering the particle. This can be determined using the diameter of particles before passing through thermo-denuder and also the diameter of the same particles after passing the thermo-denuder, then we can estimate the initial volume of the particles along with the volume of organic layer covering the particles:

𝑣𝑣=𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣 𝑜𝑜𝑜𝑜 𝑜𝑜𝑜𝑜𝑜𝑜 𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜 𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 𝑎𝑎 𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝 𝑤𝑤𝑤𝑤𝑤𝑤 ℎ 𝑡𝑡ℎ𝑒𝑒 𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑 𝑜𝑜 𝑜𝑜 𝑖𝑖 𝑣𝑣 𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣𝑣 𝑜𝑜 𝑜𝑜 𝑡𝑡 ℎ𝑒𝑒 𝑝𝑝 𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝 𝑤𝑤 𝑤𝑤𝑤𝑤ℎ 𝑡𝑡 ℎ𝑒𝑒 𝑑𝑑 𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑 𝑜𝑜 𝑜𝑜 𝑖𝑖 𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏 𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝 𝑡𝑡ℎ𝑟𝑟𝑜𝑜𝑢𝑢𝑢𝑢ℎ 𝑎𝑎 𝑡𝑡ℎ𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒 − 𝑣𝑣𝑣𝑣𝑣𝑣Or:𝑣𝑣𝑣𝑣 𝑣𝑣 𝑜𝑜𝑜𝑜 𝑡𝑡ℎ𝑒𝑒 𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝 𝑤𝑤𝑤𝑤𝑤𝑤ℎ 𝑡𝑡ℎ𝑒𝑒 𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑 𝑜𝑜𝑜𝑜 𝑖𝑖 𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎 𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝 𝑡𝑡ℎ𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟ℎ 𝑎𝑎 𝑡𝑡ℎ𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒

, = , = , 4.5

𝑉𝑉𝑂𝑂 𝐷𝐷𝐷𝐷𝑖𝑖 𝑉𝑉𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜 𝑑𝑑𝑑𝑑=𝑖𝑖 𝑉𝑉𝑑𝑑𝑑𝑑=𝑖𝑖 − 𝑉𝑉𝑡𝑡ℎ𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒 𝑑𝑑𝑑𝑑=𝑖𝑖 𝐸𝐸𝐸𝐸

136Chapter 4: The influence of fuel molecular structure on the volatility and oxidative potential of biodiesel particulate matter 137

Knowing the diameter of particles before and after passing the thermo-denuder from V-TDMA:

= = × 4.6 6 𝜋𝜋 3 𝑉𝑉𝐷𝐷𝐷𝐷𝑖𝑖 𝑉𝑉𝑑𝑑𝑑𝑑=𝑖𝑖 𝑑𝑑𝑑𝑑 𝐸𝐸𝐸𝐸 = × ( ) 7 , 6 𝜋𝜋 3 𝑉𝑉𝑡𝑡ℎ𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒 𝑑𝑑𝑑𝑑=𝑖𝑖 𝑑𝑑𝑑𝑑𝑡𝑡ℎ𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒 𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸 𝑆𝑆 In the end, the OV can be estimated using the below equation and by substituting , , and from Eq 4.3, Eq 4.4 and SMPS data respectively.

𝑂𝑂 𝐷𝐷𝐷𝐷𝑖𝑖 𝐷𝐷𝐷𝐷𝑖𝑖 𝐷𝐷𝐷𝐷𝑖𝑖 𝑉𝑉 𝑉𝑉 (𝑁𝑁 , × ) (%) = × 100 4.8 ( × ) ∑ 𝑉𝑉𝑂𝑂 𝐷𝐷𝑝𝑝𝑖𝑖 𝑁𝑁𝐷𝐷𝐷𝐷𝑖𝑖 𝑂𝑂𝑂𝑂 𝑖𝑖 𝑖𝑖 𝐸𝐸𝐸𝐸 ∑ 𝐷𝐷𝐷𝐷 𝐷𝐷𝐷𝐷 As mentioned, , is𝑉𝑉 the volume𝑁𝑁 of volatile matter in particles with the diameter of ( ), 𝑉𝑉𝑂𝑂 𝐷𝐷is𝐷𝐷 𝑖𝑖 the number of particles with the diameter of ( ) and is the volume𝑖𝑖 𝑛𝑛𝑛𝑛 of𝑁𝑁 the𝐷𝐷𝐷𝐷𝑖𝑖 particles with the diameter of . 𝑖𝑖 𝑛𝑛𝑛𝑛 𝐷𝐷𝐷𝐷𝑖𝑖 𝑉𝑉 𝑖𝑖

4.8 REFERENCES

1. Ma, F. and Hanna, M.A., Biodiesel Production: A Review. Bioresource Technology, 1999. 70(1): p. 1-15. 2. Li, Y., Tian, G., and Xu, H., Application of Biodiesel in Automotive Diesel Engines. 2013. 3. Demirbas, A., Biodiesel: A Realistic Fuel Alternative for Diesel Engines. 2008: Springer. 4. Zhang, X., Peterson, C., Reece, D., Haws, R., and Moller, G., Biodegradability of Biodiesel in the Aquatic Environment. Transactions of the ASAE, 1998. 41(5): p. 1423-1430. 5. Wang, J., Wu, F., Xiao, J., and Shuai, S., Oxygenated Blend Design and Its Effects on Reducing Diesel Particulate Emissions. Fuel, 2009. 88(10): p. 2037-2045. 6. Graboski, M.S. and McCormick, R.L., Combustion of Fat and Vegetable Oil Derived Fuels in Diesel Engines. Progress in Energy and Combustion Science, 1998. 24(2): p. 125-164. 7. Song, J. and Lee, K., Fuel Property Impacts on Diesel Particulate Morphology, Nanostructures, and Nox Emissions. SAE Technical Paper, 2007: p. 01-0129. 8. Knothe, G., Dependence of Biodiesel Fuel Properties on the Structure of Fatty Acid Alkyl Esters. Fuel processing technology, 2005. 86(10): p. 1059- 1070. 9. Knothe, G., Sharp, C.A., and Ryan III, T.W., Exhaust Emissions of Biodiesel, Petrodiesel, Neat Methyl Esters, and Alkanes in a New Technology Engine. Energy & Fuels, 2006. 20(1): p. 403-408.

Chapter 4: The influence of fuel molecular structure on the volatility and oxidative potential of biodiesel particulate matter 137

10. Cheng, C.H., Cheung, C.S., Chan, T.L., Lee, S.C., and Yao, C.D., Experimental Investigation on the Performance, Gaseous and Particulate Emissions of a Methanol Fumigated Diesel Engine. Science of the Total Environment, 2008. 389(1): p. 115-124. 11. Cheng, C.H., Cheung, C.S., Chan, T.L., Lee, S.C., Yao, C.D., and Tsang, K.S., Comparison of Emissions of a Direct Injection Diesel Engine Operating on Biodiesel with Emulsified and Fumigated Methanol. Fuel, 2008. 87(10- 11): p. 1870-1879. 12. Cheung, K.L., Polidori, A., Ntziachristos, L., Tzamkiozis, T., Samaras, Z., Cassee, F.R., Gerlofs, M., and Sioutas, C., Chemical Characteristics and Oxidative Potential of Particulate Matter Emissions from Gasoline, Diesel, and Biodiesel Cars. Environmental Science & Technology, 2009. 43(16): p. 6334-6340. 13. Surawski, N.C., Miljevic, B., Ayoko, G.A., Roberts, B.A., Elbagir, S., Fairfull-Smith, K.E., Bottle, S.E., and Ristovski, Z.D., Physicochemical Characterization of Particulate Emissions from a Compression Ignition Engine Employing Two Injection Technologies and Three Fuels. Environmental Science & Technology, 2011. 45(13): p. 5498-5505. 14. Sidhu, S., Graham, J., and Striebich, R., Semi-Volatile and Particulate Emissions from the Combustion of Alternative Diesel Fuels. Chemosphere, 2001. 42(5-7): p. 681-690. 15. Kumar, P., Robins, A., Vardoulakis, S., and Britter, R., A Review of the Characteristics of Nanoparticles in the Urban Atmosphere and the Prospects for Developing Regulatory Controls. Atmospheric Environment, 2010. 44(39): p. 5035-5052. 16. Benbrahim-Tallaa, L., Baan, R., Grosse, Y., Lauby-Secretan, B., El Ghissassi, F., Bouvard, V., Guha, N., Loomis, D., and Straif, K., International Agency for Research on Cancer Monograph Working Group. Carcinogenicity of Diesel-Engine and Gasoline-Engine Exhausts and Some Nitroarenes. Lancet Oncol, 2012. 13(7): p. 663-4. 17. Tinsdale, M., Price, P., and Chen, R., The Impact of Biodiesel on Particle Number, Size and Mass Emissions from a Euro4 Diesel Vehicle. 2010. 18. Kalligeros, S., Zannikos, F., Stournas, S., Lois, E., Anastopoulos, G., Teas, C., and Sakellaropoulos, F., An Investigation of Using Biodiesel/Marine Diesel Blends on the Performance of a Stationary Diesel Engine. Biomass and Bioenergy, 2003. 24(2): p. 141-149. 19. Fontaras, G., Karavalakis, G., Kousoulidou, M., Tzamkiozis, T., Ntziachristos, L., Bakeas, E., Stournas, S., and Samaras, Z., Effects of Biodiesel on Passenger Car Fuel Consumption, Regulated and Non- Regulated Pollutant Emissions over Legislated and Real-World Driving Cycles. Fuel, 2009. 88(9): p. 1608-1617. 20. Lee, H., Myung, C.L., and Park, S., Time-Resolved Particle Emission and Size Distribution Characteristics During Dynamic Engine Operation Conditions with Ethanol-Blended Fuels. Fuel, 2009. 88(9): p. 1680-1686. 21. Pham, P.X., Bodisco, T.A., Ristovski, Z.D., Brown, R.J., and Masri, A.R., The Influence of Fatty Acid Methyl Ester Profiles on Inter-Cycle Variability in a Heavy Duty Compression Ignition Engine. Fuel, 2014. 116(0): p. 140- 150. 22. Liu, Y.Y., Lin, T.C., Wang, Y.J., and Ho, W.L., Biological Toxicities of Emissions from an Unmodified Engine Fueled with Diesel and Biodiesel

138Chapter 4: The influence of fuel molecular structure on the volatility and oxidative potential of biodiesel particulate matter 139

Blend. Journal of Environmental Science and Health Part a-Toxic/Hazardous Substances & Environmental Engineering, 2008. 43(14): p. 1735-1743. 23. Surawski, N.C., Miljevic, B., Ayoko, G.A., Elbagir, S., Stevanovic, S., Fairfull-Smith, K.E., Bottle, S.E., and Ristovski, Z.D., Physicochemical Characterization of Particulate Emissions from a Compression Ignition Engine: The Influence of Biodiesel Feedstock. Environmental Science & Technology, 2011. 45(24): p. 10337-43. 24. Pinzi, S., Rounce, P., Herreros, J.M., Tsolakis, A., and Pilar Dorado, M., The Effect of Biodiesel Fatty Acid Composition on Combustion and Diesel Engine Exhaust Emissions. Fuel, 2012. 25. Zhu, R., Cheung, C.S., and Huang, Z., Particulate Emission Characteristics of a Compression Ignition Engine Fueled with Diesel-Dmc Blends. Aerosol Science and Technology, 2011. 45(2): p. 137-147. 26. Kim, S., Shen, S., Sioutas, C., Zhu, Y.F., and Hinds, W.C., Size Distribution and Diurnal and Seasonal Trends of Ultrafine Particles in Source and Receptor Sites of the Los Angeles Basin. Journal of the Air & Waste Management Association, 2002. 52(3): p. 297-307. 27. McClellan, R.O., Health-Effects of Exposure to Diesel Exhaust Particles. Annual Review of Pharmacology and Toxicology, 1987. 27: p. 279-300. 28. Giechaskiel, B., Alfoldy, B., and Drossinos, Y., A Metric for Health Effects Studies of Diesel Exhaust Particles. Journal of Aerosol Science, 2009. 40(8): p. 639-651. 29. Stevanovic, S., Miljevic, B., Surawski, N., Fairfull-Smith, K.E., Bottle, S.E., Brown, R., and Ristovski, Z.D., The Influence of Oxygenated Organic Aerosols (Ooa) on the Oxidative Potential of Diesel and Biodiesel Particulate Matter. Environmental Science & Technology, 2013. 47(14): p. 7655-7662. 30. Ristovski, Z.D., Miljevic, B., Surawski, N.C., Morawska, L., Fong, K.M., Goh, F., and Yang, I.A., Respiratory Health Effects of Diesel Particulate Matter. Respirology, 2012. 17(2): p. 201-12. 31. Li, Q., Wyatt, A., and Kamens, R.M., Oxidant Generation and Toxicity Enhancement of Aged-Diesel Exhaust. Atmospheric Environment, 2009. 43(5): p. 1037-1042. 32. Shinyashiki, M., Eiguren-Fernandez, A., Schmitz, D.A., Di Stefano, E., Li, N., Linak, W.P., Cho, S.H., Froines, J.R., and Cho, A.K., Electrophilic and Redox Properties of Diesel Exhaust Particles. Environmental Research, 2009. 109(3): p. 239-244. 33. Jalava, P.I., Tapanainen, M., Kuuspalo, K., Markkanen, A., Hakulinen, P., Happo, M.S., Pennanen, A.S., Ihalainen, M., Yli-Pirila, P., Makkonen, U., Teinila, K., Maki-Paakkanen, J., Salonen, R.O., Jokiniemi, J., and Hirvonen, M.R., Toxicological Effects of Emission Particles from Fossil- and Biodiesel- Fueled Diesel Engine with and without Doc/Poc Catalytic Converter. Inhalation Toxicology, 2010. 22: p. 48-58. 34. Kipen, H.M., Gandhi, S., Rich, D.Q., Ohman-Strickland, P., Laumbach, R., Fan, Z.H., Chen, L., Laskin, D.L., Zhang, J.F., and Madura, K., Acute Decreases in Proteasome Pathway Activity after Inhalation of Fresh Diesel Exhaust or Secondary Organic Aerosol. Environmental Health Perspectives, 2011. 119(5): p. 658-663. 35. Kooter, I.M., van Vugt, M., Jedynska, A.D., Tromp, P.C., Houtzager, M.M.G., Verbeek, R.P., Kadijk, G., Mulderij, M., and Krul, C.A.M., Toxicological Characterization of Diesel Engine Emissions Using Biodiesel

Chapter 4: The influence of fuel molecular structure on the volatility and oxidative potential of biodiesel particulate matter 139

and a Closed Soot Filter. Atmospheric Environment, 2011. 45(8): p. 1574- 1580. 36. Biswas, S., Verma, V., Schauer, J.J., Cassee, F.R., Cho, A.K., and Sioutas, C., Oxidative Potential of Semi-Volatile and Non Volatile Particulate Matter (Pm) from Heavy-Duty Vehicles Retrofitted with Emission Control Technologies. Environmental Science & Technology, 2009. 43(10): p. 3905- 3912. 37. Pham, P., Bodisco, T., Stevanovic, S., Rahman, M., Wang, H., Ristovski, Z., Brown, R., and Masri, A., Engine Performance Characteristics for Biodiesels of Different Degrees of Saturation and Carbon Chain Lengths. 2013, SAE Technical Paper. 38. Orsini, D.A., Wiedensohler, A., Stratmann, F., and Covert, D.S., A New Volatility Tandem Differential Mobility Analyzer to Measure the Volatile Sulfuric Acid Aerosol Fraction. Journal of Atmospheric and Oceanic Technology, 1999. 16(6): p. 760-772. 39. Leskinen, A.P., Jokiniemi, J.K., and Lehtinen, K.E.J., Characterization of Aging Wood Chip Combustion Aerosol in an Environmental Chamber. Atmospheric Environment, 2007. 41(17): p. 3713-3721. 40. Giechaskiel, B., Alföldy, B., and Drossinos, Y., A Metric for Health Effects Studies of Diesel Exhaust Particles. Journal of Aerosol Science, 2009. 40(8): p. 639-651. 41. Miljevic, B., Heringa, M.F., Keller, A., Meyer, N.K., Good, J., Lauber, A., Decarlo, P.F., Fairfull-Smith, K.E., Nussbaumer, T., Burtscher, H., Prevot, A.S.H., Baltensperger, U., Bottle, S.E., and Ristovski, Z.D., Oxidative Potential of Logwood and Pellet Burning Particles Assessed by a Novel Profluorescent Nitroxide Probe. Environmental Science & Technology, 2010. 44(17): p. 6601-6607. 42. Stevanovic, S., Miljevic, B., Eaglesham, G.K., Bottle, S.E., Ristovski, Z.D., and Fairfull-Smith, K.E., The Use of a Nitroxide Probe in Dmso to Capture Free Radicals in Particulate Pollution. European Journal of Organic Chemistry, 2012. 2012(30): p. 5908-5912. 43. Stevanovic, S., Miljevic, B., Surawski, N.C., Fairfull-Smith, K.E., Bottle, S.E., Brown, R., and Ristovski, Z.D., Influence of Oxygenated Organic Aerosols (Ooas) on the Oxidative Potential of Diesel and Biodiesel Particulate Matter. Environmental Science & Technology, 2013. 47(14): p. 7655-7662. 44. Miljevic, B., Modini, R.L., Bottle, S.E., and Ristovski, Z.D., On the Efficiency of Impingers with Fritted Nozzle Tip for Collection of Ultrafine Particles. Atmospheric Environment, 2009. 43(6): p. 1372-1376. 45. Stevanovic, S., Ristovski, Z.D., Miljevic, B., Fairfull-Smith, K.E., and Bottle, S.E., Application of Profluorescent Nitroxides for Measurements of Oxidative Capacity of Combustion Generated Particles. Chemical Industry & Chemical Engineering Quarterly, 2012. 18(4): p. 653-659. 46. Islam, M., Magnusson, M., Brown, R., Ayoko, G., Nabi, M., and Heimann, K., Microalgal Species Selection for Biodiesel Production Based on Fuel Properties Derived from Fatty Acid Profiles. Energies, 2013. 6(11): p. 5676- 5702. 47. Riipinen, I., Pierce, J.R., Donahue, N.M., and Pandis, S.N., Equilibration Time Scales of Organic Aerosol inside Thermodenuders: Evaporation

140Chapter 4: The influence of fuel molecular structure on the volatility and oxidative potential of biodiesel particulate matter 141

Kinetics Versus Thermodynamics. Atmospheric Environment, 2010. 44(5): p. 597-607. 48. Kwon, S.B., Lee, K.W., Saito, K., and Shinozaki, O., Size-Dependent Volatility of Diesel Nanoparticles: Chassis Dynamometer Experiments - Environmental Science &Amp; Technology (Acs Publications). … science & technology, 2003. 49. Ristimäki, J., Vaaraslahti, K., Lappi, M., and Keskinen, J., Hydrocarbon Condensation in Heavy-Duty Diesel Exhaust. Environmental Science & Technology, 2007. 41(18): p. 6397-6402. 50. Pinzi, S., Rounce, P., Herreros, J.M., Tsolakis, A., and Pilar Dorado, M., The Effect of Biodiesel Fatty Acid Composition on Combustion and Diesel Engine Exhaust Emissions. Fuel, 2013. 104: p. 170-182. 51. Kittelson, D.B., Watts, W.F., and Johnson, J.P., On-Road and Laboratory Evaluation of Combustion Aerosols‚Äîpart1: Summary of Diesel Engine Results. Journal of Aerosol Science, 2006. 37(8): p. 913-930. 52. Sakurai, H., Park, K., McMurry, P.H., Zarling, D.D., Kittelson, D.B., and Ziemann, P.J., Size-Dependent Mixing Characteristics of Volatile and Nonvolatile Components in Diesel Exhaust Aerosols. Environmental Science & Technology, 2003. 37(24): p. 5487-5495. 53. Sakurai, H., Tobias, H.J., Park, K., Zarling, D., Docherty, K.S., Kittelson, D.B., McMurry, P.H., and Ziemann, P.J., On-Line Measurements of Diesel Nanoparticle Composition and Volatility. Atmospheric Environment, 2003. 37(9-10): p. 1199-1210. 54. Worton, D.R., Isaacman, G., Gentner, D.R., Dallmann, T.R., Chan, A.W.H., Ruehl, C., Kirchstetter, T.W., Wilson, K.R., Harley, R.A., and Goldstein, A.H., Lubricating Oil Dominates Primary Organic Aerosol Emissions from Motor Vehicles. Environmental Science & Technology, 2014. 48(7): p. 3698-3706. 55. Sadiktsis, I., Koegler, J.H., Benham, T., Bergvall, C., and Westerholm, R., Particulate Associated Polycyclic Aromatic Hydrocarbon Exhaust Emissions from a Portable Power Generator Fueled with Three Different Fuels–a Comparison between Petroleum Diesel and Two Biodiesels. Fuel, 2014. 115: p. 573-580. 56. Grigoratos, T., Fontaras, G., Kalogirou, M., Samara, C., Samaras, Z., and Rose, K., Effect of Rapeseed Methylester Blending on Diesel Passenger Car Emissions €“ Part 2: Unregulated Emissions and Oxidation Activity. Fuel, 2014. 128(C): p. 260-267. 57. Miljevic, B., Heringa, M.F., Keller, A., Meyer, N.K., Good, J., Lauber, A., Decarlo, P.F., Fairfull-Smith, K.E., Nussbaumer, T., Burtscher, H., Prevot, A.S., Baltensperger, U., Bottle, S.E., and Ristovski, Z.D., Oxidative Potential of Logwood and Pellet Burning Particles Assessed by a Novel Profluorescent Nitroxide Probe. Environmental Science & Technology, 2010. 44(17): p. 6601-7. 58. Rahman, M.M., Pourkhesalian, A.M., Jahirul, M.I., Stevanovic, S., Pham, P.X., Wang, H., Masri, A.R., Brown, R.J., and Ristovski, Z.D., Particle Emissions from Biodiesels with Different Physical Properties and Chemical Composition. Fuel, 2014. 134(0): p. 201-208. 59. Wickham, H., Ggplot2: Elegant Graphics for Data Analysis. 2009: Springer.

Chapter 4: The influence of fuel molecular structure on the volatility and oxidative potential of biodiesel particulate matter 141

60. Gill, S.S., Tsolakis, A., Herreros, J.M., and York, A.P.E., Diesel Emissions Improvements through the Use of Biodiesel or Oxygenated Blending Components. Fuel, 2012. 95: p. 578-586. 61. Gill, S., Tsolakis, A., Herreros, J., and York, A., Diesel Emissions Improvements through the Use of Biodiesel or Oxygenated Blending Components. Fuel, 2012. 95: p. 578-586.

142Chapter 4: The influence of fuel molecular structure on the volatility and oxidative potential of biodiesel particulate matter

Chapter 5: Effect of atmospheric aging on volatility and ROS of biodiesel exhaust nanoparticles

A. M. Pourkhesalian1, S. Stevanovic1, M. M. Rahman1, E. M. Faghihi1, S. E. Bottle1, A. R. Masri2, R. J. Brown1, Z. D. Ristovski*1

1 International Laboratory for Air Quality and Health, Queensland University of Technology, GPO Box 2434, Brisbane QLD, 4001, Australia

2Clean Combustion Research Group, The University of Sydney, NSW2006, Australia

*E-mail: [email protected]

Submitted for publication in Atmospheric Chemistry and Physics (ACP) in 2015

Chapter 5: Effect of atmospheric aging on volatility and ROS of biodiesel exhaust nanoparticles 143

Statement of Joint Authorship

Title: Effect of atmospheric aging on volatility and ROS of biodiesel exhaust nanoparticles

Authors: A.M. Pourkhesalian, S. Stevanovic, M. M. Rahman, E. M. Faghihi, S. E. Bottle, A. R. Masri, R. J. Brown, Z. D. Ristovski

A. M. Pourkhesalian: Performed the measurements and data analysis and wrote the manuscript.

Signature: Date: 05-03-2015

S. Stevanovic: Contributed in design of the experimental program, assisted in data analysis and write up the manuscript.

Md Mostafizur Rahman: Contributed in measurement, data analysis and assisted in write up the manuscript.

E. M. Faghihi: contributed in data analysis.

A. R. Masri: Supervised the overall study design, set up and experiments, reviewed the manuscript.

S.E. Bottle: Synthesis of chemicals required for ROS measurements.

R. J. Brown: Supervised the overall study design, set up and experiments, reviewed the manuscript.

Z. D. Ristovski: Supervised the overall study design, set up and experiments, reviewed the manuscript.

The authors listed above have certified that:

1. They meet the criteria for authorship in that they have participated in the conception, execution, or interpretation, of at least that part of the publication in their field of expertise;

2. They take public responsibility for their part of the publication, except for the responsible author who accepts overall responsibility for the publication;

144 Chapter 5: Effect of atmospheric aging on volatility and ROS of biodiesel exhaust nanoparticles 145

3. There are no other authors of the publication according to these criteria;

4. Potential conflicts of interest have been disclosed to (a) granting bodies, (b) the editor or publisher of journals or other publications, and (c) the head of the responsible academic unit, and

5. They agree to the use of the publication in the student’s thesis and its publication on the Australasian Research Online database consistent with any limitations set by publisher requirements.

Principal Supervisor Confirmation

I have sighted email or other correspondence from all Co-authors confirming their certifying authorship.

Name Zoran Ristovski

Signature Date

Chapter 5: Effect of atmospheric aging on volatility and ROS of biodiesel exhaust nanoparticles 145

Title: Effect of atmospheric aging on volatility and ROS of biodiesel exhaust nanoparticles

A. M. Pourkhesalian1, S. Stevanovic1, M. M. Rahman1, E. M. Faghihi 1, S. E. Bottle1, A. R. Masri2, R. J. Brown1, Z. D. Ristovski1*

1 ILAQH and BERF, Queensland University of Technology, Brisbane, QLD 4001, Australia

2 Clean Combustion Research Group, The University of Sydney, NSW 2006, Australia

*Corresponding Author: Z.D. Ristovski,

Email: [email protected]

Tel.: +(61)731381129.

Abstract

In the prospect of limited energy resources and climate change, effects of alternative biofuels on primary emissions are being extensively studied. Our Two recent studies have shown that biodiesel fuel composition has a significant impact on direct PM emissions, which were shown to be significantly different compared to petroleum diesel. [1, 2] Emissions appeared to have higher oxidative potential with the increase in oxygen content and decrease of carbon chain length and unsaturation levels of fuel molecules. Overall, both studies concluded that chemical composition of biodiesel is more important than its physical properties in controlling exhaust particle emissions. This suggests that the atmospheric aging processes, including secondary organic aerosol formation, of emissions from different fuels will be different as well. Measurements were conducted on a modern common-rail diesel engine. To get more information on realistic properties of tested biodiesel diesel

146 Chapter 5: Effect of atmospheric aging on volatility and ROS of biodiesel exhaust nanoparticles 147 particulate matter once they are released into the atmosphere, particulate matter was exposed to atmospheric oxidants, ozone and UV; and the change in their properties was monitored for different biodiesel blends. Upon the exposure to oxidative agents, the chemical composition of the exhaust changes. It triggers the cascade of photochemical reactions resulting in the partitioning of semi-volatile compounds between the gas and particulate phase. In most of the cases, aging lead to the increase in volatility and oxidative potential, and the increment of change was mainly dependent on the chemical composition of fuels as the leading cause for the amount and the type of semi-volatile compounds present in the exhaust.

Keywords: Diesel particulate matter, atmospheric aging, ROS, Volatility, methyl ester biodiesel, carbon chain length, level of saturation

5.1 INTRODUCTION

In the urban environments the most significant contributor to the overall PM burden are traffic emissions [3]. Harmful effects of DPM to humans and the environment have been extensively studied which resulted in the classification of Diesel Exhaust (DE) as carcinogenic by International Agency for Research on Cancer (IARC) [4] in 2013. Following this, environmental agencies will tend to implement more stringent regulations to meet air quality standards forcing engine manufacturers to further reduce engine emissions. As biodiesel produces significantly less particulate matter [5-7] along with other economical and environmental advantages [8-10]; it may be the available option of the fuel industry in reducing PM emissions.

There are numerous studies reporting that using biodiesel decreases the diesel primary emissions [5, 7, 9, 11, 12]. It is also well established that biodiesel decreases black carbon emissions and increases the Soluble Organic Fraction (SOF) [7, 13]. The decrease in particle emissions is also very often followed by the emission of smaller particles [5], while the increase in SOF results in excessive amount of volatile and semi-volatile matter in the exhaust [13-15]. Both the decrease in particle size and the increase in SOF are believed to be linked to the increase of OP of particles [15, 16].

People in urban and semi-urban environments are mainly exposed to a combination of fresh and aged primary emissions and SOA. The generation of SOA

Chapter 5: Effect of atmospheric aging on volatility and ROS of biodiesel exhaust nanoparticles 147

occurs during aging of primary emissions, mainly through oxidation of gas-phase organic compounds that can result either in the formation of new particles or condensation of these compounds onto pre-existing particles. [17, 18]. Consequently, atmospheric fate or aging of diesel exhaust has become of a great scientific interest and attracts significant attention from the researchers [19-27]. Although there are several studies on the aging of diesel exhaust, [18-20, 22, 23, 27-31], still a number of questions remained unanswered. Given the fact that biodiesel produces more of semi-volatile matter and thus the contribution of biodiesels to SOA formation in urban environments could be significant, the atmospheric aging of biodiesel exhaust is yet to be explored in more detail.

Smog chambers are extensively used to conduct aging experiments on anthropogenic aerosols and investigate their potential to form SOA [13, 32, 33]. Recently, flow-through reactors are drawing a great deal of attention for their advantages over smog chambers. While the use of smog-chambers to study the photochemical aging processes [19-27] has been shown to be a valuable tool, there are some significant disadvantages associated with their usage. Their size, complexity, high cost and operational time required (typically only one experiment per day) make them impractical when series of experiments need to be conducted. As an alternative to smog chambers, flow-through reactors or “Potential Aerosol Mass” (PAM) chambers are becoming very popular among researchers [34-36]. PAM chambers simulate atmospheric photo oxidation processes. Employment of PAM chambers is associated with high oxidant concentrations and short exposure times mimicking few hours to a few days of real-time atmospheric oxidation. The short residence time of PAM chambers further enables better control of the oxidant concentrations and minimizes wall losses, which could be significant in smog chambers [34].Also the response time in PAM chambers is by far less than that of smog chambers making it easier to observe the effect of changing a control variable. Moreover, experiments carried out on PAM chambers tend to be more repeatable and more reproducible [34].

So far, the PAM chambers have been used to investigate the potential of combustion generated aerosols to produce SOA including biomass burning [36] and 2-stroke engine exhaust [37]. The potential of biodiesel exhaust gases to cause SOA formation is yet to be studied.

148 Chapter 5: Effect of atmospheric aging on volatility and ROS of biodiesel exhaust nanoparticles 149

This research investigates the change in volatile organic content and the potential toxicity of biodiesel exhaust particulate matter after aging in a flow-through reactor. We explore the contribution of photochemical aging processes on the potential transfer of the semi-volatile fraction of biodiesel exhaust from the gas phase to the particle phase and the extent to which semi-volatile compounds are related to chemically active molecules that contain oxygen or Reactive Oxygen Species (ROS). The study, also aims to find a correlation between the chemical composition of biodiesel and the volatility and toxicity of biodiesel particulate matter while undergoing simulated aging processes. Specifically the main objective of this work is to critically examine the influence of the carbon chain length, saturation levels and oxygen content of biodiesel Fatty Acid Methyl Ester (FAME) fuel molecules. All the work has been undertaken by investigating particle emissions from a common-rail engine using four palm oil biodiesels at different blend percentages.

5.2 EXPERIMENTAL SETUP AND METHODOLOGY

Figure 5 .1 illustrates the experimental setup. The setup was composed of an existing EURO III diesel engine which was mounted on an engine dynamometer within the Biofuels Engine Research Facility (BERF) at Queensland University of Technology (QUT), and was used as the test bed. The engine was a 6-cylinder, turbo- charged, common-rail, Euro III, compression ignition engine with the displacement volume of 5.9 liters. More details of the experimental setup, engine specifications, dilution system along with the chemical and physical properties of fuels used can be found in our previous publication [7]. The main addition was that the diluted exhaust was passed through a flow-through reactor (PAM reactor) where the diluted exhaust is mixed with ozone and irradiated with UV light.

The flow-through-reactor was a 1-meter-long stainless steel cylinder which had four inlet ports and four outlet ports. There were also four lengthwise sampling points to enable measurement along the reactor length. In the centre of the reactor, there was space for a UV tube to irradiate the diesel exhaust. Considering the dimensions of the reactor, the UV tube as well as the flow rate through the reactor (7±0.5 lpm), the residence time (treatment time) of the reactor was estimated to be 60±10 seconds. The amount of time the exhaust gas was exposed to simulated

Chapter 5: Effect of atmospheric aging on volatility and ROS of biodiesel exhaust nanoparticles 149

atmospheric conditions was much less than in the real conditions therefore the concentration of ozone and exposure to UV had to be higher compared to real atmospheric conditions. Based on a previous literature, this residence time approximately correlated to an exposure of at least 2 to 8 hours in the atmosphere [37]. After aging in the reactor, diesel exhaust went to a number of different devices capable of measuring a variety of physical and chemical properties of gases and particles which will be explained later.

Figure 5.1Experiment setup HEPA: High-Efficiency Particulate Air Filter; F T reactor: Flow-Through Reactor; O3 Gen: Ozone Generator; Ozone Analyzer: EC9810 Ecotech; EC: Electrostatic Classifier; TD: Thermo-Denuder; V-TDMA: Volatility Tandem Differential Mobility Analyzer; CPC: Condesation Particle Counter; SMPS: Scanning Mobility Particle Sizer; Impingers: used to collect DPM for ROS measurements.

Purified air was drawn trough the ozone generator (Ozonizer HLO 800) and was injected into the reactor where it was mixed with diluted and cooled diesel exhaust. The flow rate through the ozone generator was restricted using a critical orifice and was measured to be 0.15 . To continuously monitor the concentration of ozone in the reactor an Echotech EC9810𝑙𝑙𝑙𝑙𝑙𝑙 Ozone Analyzer was used. The average concentration of ozone in the reactor was kept to be around 0.5±0.1 ppm during aging experiments. OH radical formation was not monitored and reported as this was

150 Chapter 5: Effect of atmospheric aging on volatility and ROS of biodiesel exhaust nanoparticles 151 a qualitative study showing the general effect without the quantification of SOA yields.

As mentioned before, to simulate sunlight and reactions induced by the irradiation of UV from the sun [34, 36], there was a space for a UV tube in the center of the reactor which was suitable to fit different types of UV light tubes (UV-A, UV- B and UV-C). The UV tube which was used in this study mostly emitted UV light at 253nm wave length (UVC).

A standard set of gas analyzer was used to continuously monitor the concentrations of CO, CO2 and NOx. More details on the gas analyzer set is described in a recent publication [7] .

A TSI DustTrak (Model 8530) measured the mass concentration of particles to be used to normalize the ROS levels of the diesel exhaust [38]. A Scanning Mobility Particle Sizer (SMPS TSI 3080, with a 3022 CPC) measured the size distribution of diesel exhaust. A Volatility Tandem Differential Mobility Analyzer (V-TDMA) consisting of an electrostatic classifier, a thermo-denuder and an SMPS (in-house designed column with a 3010 CPC) measured the amount of volatile matter in the diesel exhaust particles [39]. The V-TDMA yielded the change in the diameter of particles for six pre-selected sizes: 30, 60, 90, 120, 150, 200 and 220 nm after they have passed through the thermo-denuder with the temperature set at 300° C. Using the V-TDMA method, the volatility of particles was compared before and after aging in the reactor. The error range for the particle sizing instruments was less than 1 percent that led to a maximum 3% error for the volatility measurements[2].

To measure the ROS levels of diesel exhaust of different fuel stocks before and after aging, the BPEA molecular probe (bis(phenylethynyl) anthracene-nitroxide) was applied in-situ. Samples were collected by bubbling aerosol through an impinger containing 20 mL of 4 μM BPEA solution which used an AR grade dimethylsulphoxide as the solvent. More details on the ROS sampling methodology, theory behind its application and proof of concept in the case of various combustion sources can be found in previous publications [40-44].

Chapter 5: Effect of atmospheric aging on volatility and ROS of biodiesel exhaust nanoparticles 151

5.3 FUEL SELECTION

In this study four FAME fuels with controlled chemical compositions were tested and compared to petrodiesel. The fuels differed in terms of iodine value and saponification value which correspond to saturation degree and oxygen content of the fuels respectively. To differentiate the effects caused by change of saturation degree from those caused by oxygen content, two of the biodiesels had very close saturation levels but different carbon chain lengths, while the other two biodiesels were the same in terms of oxygen content but different in saturation levels. We used commercial petrodiesel to make 20% and 50% biodiesel blends. The FAMEs are labelled as C810, C1214, C1618 and C1875, based on the number of carbon atoms in the most abundant fatty acid in that particular biodiesel stock. Some of the most important chemical and physical prosperities of the fuels can be found in a previous study by Rahman et al [7].

5.4 OPERATION OF THE ENGINE

Blends of 100%, 50%, 20% and 0% of biodiesel, noted as B100 (pure biodiesel), B50 and B20 and B0 (pure petrodiesel) respectively, were used to run the engine. All the tests were conducted at quarter load and at the engine speed of 1500 rpm.

5.5 RESULTS AND DISCUSSION

5.5.1 Volatility measurements Figure 5 .2 shows the change in Volumetric Volatile Fraction (VVF) of particulate component before and after exposure to UV light and ozone. It is clearly seen in the figure that the volatility of particles increased after the exposure to oxidative agents. As the vapor pressure of the gases reduces due to photochemical reactions, some of the volatile substances partition form the gas phase into the particle phase resulting in more volatile particles [20, 45, 46]. It was observed that the exposure to oxidative agents did not lead to the formation of nucleation mode particles. It is to be noted that in cases when the dilution ratio is low and there is not enough surfaces available for the partitioning gas to condense on (e.g. application of diesel particulate filters), then formation of nucleation mode particles can occur [47], but at high levels of dilution, semi-volatile compounds tend to partition in the gas phase [30]. The reason why we did not observe any nucleation mode particles was

152 Chapter 5: Effect of atmospheric aging on volatility and ROS of biodiesel exhaust nanoparticles 153 that the dilution ratio was rather large at more than 400 and there was no diesel particulate filter in use, so there were already surfaces available for the condensation of gaseous matter. [48, 49]

Figure 5.2 The volumetric volatile fraction (vvf) of particles versus particle pre-selection sizes for different biodiesels and different blends before and after aging. Each column is dedicated to one of the biodiesels and each row specifies one blend. Red markers represent the particles before aging and blue markers show the particles after aging. All the tests were carried out at quarter load, 1500rpm. Error bars show the SEM for V-TDMA. All the tests were carried out at quarter load, 1500rpm.

In Figure 5 .2, it is clearly visible that the volatility of particles has increased with the increase of biodiesel percentage in the blends[2]. While for B20, the amount of volatile matter is quite close to that of B0; the volatility of particles has increased considerably with B50 and B100. These findings were in a good agreement with the previous observations, which showed a significant increase in volatility of diesel particulate matter with increasing biodiesel percentage in the blend [2, 15]. Particles

Chapter 5: Effect of atmospheric aging on volatility and ROS of biodiesel exhaust nanoparticles 153

produced by the combustion of C810 and C1214 were the most volatile; also the change in volatility of these particles was more significant with increasing percentage of biodiesel in the blend; whereas, the change in volatility was almost negligible for C1875 and all of its blends which was the most unsaturated biodiesel with the longest carbon chain length and lowest oxygen content.

The effect of oxygen content on the volatility of particles with gaseous and liquid diesel fuels was studied previously [2, 13, 50] and in our study, the increase of volatile matter in particles is more significant for more saturated fuels and with more oxygen content; where for C1875 the increase is just within the error range of the instrument.

The volatile fraction and the growth of volatile fraction for 30 nm particles were the largest in almost all cases which showed that more attention is to be paid to smaller particles; not only because of their ability to suspend longer and penetrate deeper in the lung [13-15, 51], but also because of their probability of carrying more oxidized volatile organics as they age.

In B0, apart from 30 nm particles, for all other particle sizes, the amount of volatile matter is within the error range of the differential mobility analyzer (DMA), implying that B0 emits particles that are mainly composed of soot. Also after aging experiments, the increase in volatile fraction in B0 particles was just small enough to be neglected showing that either the amount of volatile matter in the gas phase is very small or they do not undergo photochemical reactions.

To be able to compare cases more easily, we have estimated the total amount of volatile matter that is in the liquid phase for each case and we called it Overall Volatility (OV). To do so, particles are assumed to be spherical. The volume of volatile matter is estimated using V-TDMA data and the number concentration of each pre-selection size from SMPS data. The quantity, OV, is expressed through percentage and means that a certain proportion of the PM in the aerosol is composed of volatile matter while the rest is non-volatile (mainly elemental carbon)[52]. More details of the concept and the procedure of estimating OV can be found in our recent study [2].

Figure 5 .3 illustrates the OV of diesel exhaust versus blend for different biodiesel and compared to petrodiesel. As previously observed[15], the volatility of

154 Chapter 5: Effect of atmospheric aging on volatility and ROS of biodiesel exhaust nanoparticles 155 diesel particulate matter increases as the percentage of biodiesel goes higher in the blends. This fact is particularly seen in more saturated fuels with higher oxygen content. Blends of C810 and C1214, which have short carbon chain lengths and high saturation degree, have produced emissions with the most volatile PM; while C1618 and C1875, with longer carbon chain lengths and more carbon double bonds (more saturated fuels), have produced less and less volatile PM. This can be explained by assuming that the oxygen borne with the molecule of the fuel can cause more oxidized combustion products being of low volatility both in the liquid phase and in the gas phase. Those volatile substances having a lower vapor pressure would condense on soot particles just after the combustion while in the combustion chamber or in the exhaust system, causing the volatile fraction of particle become more significant. The volatile fraction of particles produced by more oxygenated fuels increases more after aging because there are more semi-volatile and volatile matter in the gas phase undergoing photo chemical reactions and transforming into the particle phase and condensing onto the primary particles.

Figure 5.3 Volatility of particulate matter versus different blends for petrodiesel and tested biodiesels before and after aging in the flow through reactor. Blue points are due to aged particulate matter and red points represent fresh particles. Different tested biodiesels can be seen at the top of the figure. B0 representing petrodiesel is repeated in all graphs for comparison.

The two heavier biodiesel fuels which were tested in the study (C1618 and C1875) are more unsaturated, less oxygenated and have a longer carbon chain length and thus are more comparable to petrodiesel. Both of them caused more volatile PM

Chapter 5: Effect of atmospheric aging on volatility and ROS of biodiesel exhaust nanoparticles 155

comparing to B0 and their increase in volatility was not as significant as the two other biodiesel fuels in this study

PM due to short-length hydrocarbons blends (C810 and C1214) was more volatile before, but also after the aging process it can be seen that the carbon chain length of the fuel was a more influential factor on the volatility of PM than saturation degree of the fuel. The reason would be the fact that when the carbon chain length is shorter, the weight ratio of oxygen borne by the fuel molecule will be higher and thus more oxidative agent is available for the organics, so the vapor pressure of the products, whether before or after the exposure, will be further lowered as a result of oxidization. To examine this supposition, the oxygen content of all the fuels and blends were either measured or calculated, and then the change in the volatility of PM emitted by each blend (after aging) was plotted against the oxygen content of the blends (Figure 5 .4). The effect of oxygen content of the blend on the increase of the volatile matter in the DPM is evident. Apart from two points (C1618, B100 and C1875, B100) that show low volatility with higher oxygen content; all other blends showed a good correlation between volatility and oxygen content.

156 Chapter 5: Effect of atmospheric aging on volatility and ROS of biodiesel exhaust nanoparticles 157

Figure 5.4 Change in volatility of particles before and after aging against oxygen content of the blends. Different shapes as can be seen in the legend show different fuels. The point corresponding to C1875, B100 is excluded from the model. The grey area shows the 95% confidence interval of the fit. R2= 0.66.

The correlations between the variables were analyzed using the linear regression or generalized linear model with a log link function. The validation of model assumptions was performed by the residuals versus fit values and QQ plots. Modelling and visualizations were carried out using the ggplot2 package in R [53]

5.5.2 ROS measurements Oxidative potential measurement, expressed through ROS concentration of PM can be used as a good estimate for its reactivity and toxicity. An in-house developed profluorescent molecular probe BPEAnit was applied in an entirely novel, rapid and non-cell based way to assess particulate oxidative potential. Based on the data provided in the literature so far [42], there are some uncertainties related to the nature of chemical species responsible for the measured redox potential and overall toxicity.

Chapter 5: Effect of atmospheric aging on volatility and ROS of biodiesel exhaust nanoparticles 157

However, the majority of research in this field reported that organic fraction, more precisely semi-volatile organic content, is in a good correlation with ROS concentration [15, 16, 40]

Figure 5 .5 illustrates the oxidative potential of particles for all three biodiesel blends at 25% load, 1500 rpm before and after the aging process. The points corresponding to B0 are repeatedly presented in all blends to ease comparison of petrodiesel particulate oxidative potential with that of other blends. It is evident that aging process increases potential toxicity of particles by increasing ROS concentration in the case of all four biodiesel as well as petrodiesel. This result is supported by volatility measurements. As stated above, particles become more volatile after aging as a result of oxidation of semi-volatile compounds and their subsequent condensation onto pre-existing particles.

Figure 5.5 ROS levels of particulate matter versus different blends for petrodiesel and tested biodiesels before aging (red circles) and after aging (blue triangles). Different tested biodiesels can be seen at the top of the figure. B0 representing petrodiesel is repeated in all graphs for comparison

In Figure 5 .5 it is clearly visible that the change in ROS content after exposure to UV and ozone is the most significant in the case of B100 and increase in oxidative potential decreases with the decreasing percentage of biodiesels in blends. To further explore this, the change in ROS concentration was plotted against oxygen content of the fuels. A clear trend between the fuel oxygen content and ROS concentration is

158 Chapter 5: Effect of atmospheric aging on volatility and ROS of biodiesel exhaust nanoparticles 159 evident from Figure 5 .6. The actual mechanism standing behind this aging transformation and subsequent partitioning remains unclear and is subject to further investigation. However, this graph is showing a clear indication that the amount of increased ROS concentration upon oxidative aging is more significant in the case of PM originating from the combustion of molecules with higher oxygen content. To get a further insight into the aging process of tested aerosols, the oxidative potential of the gas phase was measured as well. The results are summarized in Figure 5 .7 and they show interesting trends. When the compounds from the gas phase are oxidized they are transferred into the particle phase due to the decrease of their saturation vapor pressure. In this case one would expect a decrease in the concentration of the gaseous component after aging. This is the case with long chain, unsaturated biodiesels- C1875, C1618 and to some extent to C1214, as well as with petrodiesel that generally consists of around 75% paraffins and ~25% unsaturated compounds (In the literature the average formula for diesel is C12H23 [54], corresponding to C12 fuel but with no oxygen).

Chapter 5: Effect of atmospheric aging on volatility and ROS of biodiesel exhaust nanoparticles 159

Figure 5.6 The correlation between the change in ROS levels and oxygen content the fuels before and after aging. The point corresponding to C1875 is excluded from the model. The grey area shows the 95% confidence interval of the fit. R2= 0.52

On the contrary fuels with higher oxygen content show an increase in the oxidative potential of the gas phase. This could be due to the vapor pressure of the oxidized gas phase compounds not being low enough to lead to the condensation onto the particle phase. By looking at the volatility change in respect to aging, it is obvious that the volatility increases upon exposure to UV and ozone for all the fuels. This may mean that in the case of short-chain, saturated biodiesels, aging leads to the condensation of certain species from the gas phase, which further increases OV, but does not carry any ROS potency.

160 Chapter 5: Effect of atmospheric aging on volatility and ROS of biodiesel exhaust nanoparticles 161

Figure 5.7 ROS levels in gas phase before and after aging. Fresh and aged particulate matter are separated using blue color for aged PM and red for fresh PM.

The correlation between overall volatility of diesel particulate matter and the ROS can be seen in Figure 5 .8. This figure shows the measured ROS levels and volatility of particulate matter for all the modes and blends both before and after aging. Blue markers are due to fresh particles and red markers are due to aged particulate matter. Particulate matter produced by each fuel is specified using a symbol as can be seen in the legends. This categorization method allows us to easily spot clusters of a color or a symbol. The first observation from Figure 5 .7 is that the ROS correlates with OV. Note that the correlation is not linear but it is exponential. So a small increase in volatility of particulate matter can cause a significant increase in ROS levels. The symbols square and triangle are seen towards the righter and upper side of the figure which again shows that more saturated fuels with shorter carbon chain length can cause more volatile and particles with a higher oxidative potential. It is also evident that blue markers tend to move toward the righter and upper side of the figure underlining the exacerbating effect of atmospheric aging on the biodiesel particulate matter.

Chapter 5: Effect of atmospheric aging on volatility and ROS of biodiesel exhaust nanoparticles 161

Figure 5.8 The correlation between ROS of particulate matter and the volatility of particulate matter. Fresh and aged particulate matter are separated using blue color for fresh PM and red for aged PM, also the shape of the markers represents different fuel. The grey area shows the 95% confidence interval of the fit. R2= 0.86

5.6 CONCLUSION

The objective of this research was to shed some light on the potential of biodiesel exhaust to generate SOA and to compare the potential toxicity of aged particles caused by biodiesel with that of petrodiesel. The results presented here showed that chemical composition of the fuel can affect the primary emissions as well as secondary emissions. Saturated fuels causing less PM also caused more volatile PM and significantly more semi-volatiles in the gas phase with the potential to partition between the gas phase and particle phase upon aging. Moreover, aged particles from more saturated fuels with higher oxygen content have a higher oxidative potential as expressed through the increase of ROS concentration. Results of this research are in contrast with a few previous studies where it was claimed that

162 Chapter 5: Effect of atmospheric aging on volatility and ROS of biodiesel exhaust nanoparticles 163 use of biodiesel did not tend to increase the level of toxicity of diesel exhaust [11, 55, 56].

This study is not quantitative and cannot be used for SOA prediction within models. However, its qualitative analysis calls for the attention of regulating authorities and contributes to the overall body of knowledge on the effects of fuel composition on both engine primary as well as secondary emissions. Once emitted to the atmosphere, PM originating from combusted biodiesel will transform and become more toxic than its parent molecules and its overall toxicity will mainly depend on the chemical properties of the fuel. Therefore categorizing all biodiesel fuels in one group, as it is in current regulations, may not be the most appropriate way of classification in respect to their emissions. Various biodiesel fuels are very different in terms of both physical properties and chemical compositions which combust differently causing substantially diverse species in the exhaust gases, particularly in PM and the semi-volatile fraction. Moreover, the exhaust gases from various biodiesel fuels behave differently when reacting with oxidative agents from the atmosphere. Perhaps new regulation should take some of the most important characteristics of the biodiesel into account and classify them in groups based on the oxygen content or level of saturation etc. Furthermore, current legislations only regulate the nonvolatile fraction of primary particles. There are no regulations for VOCs, semi-volatile matter and the capability of the emissions to form SOA [57]. This study points out the necessity of redefining the legislations; and as experiments with flow-through reactors are more reproducible and less costly, they could be used for introduction of more stringent regulations.

5.7 ACKNOWLEDGMENTS

The authors would like to acknowledge the support from the Australian Research Council under Grants LP110200158, DP1097125, DP130104904, DP110105535 and DP120100126. Laboratory assistance from Mr Noel Hartnett and Mr Scott Abbett were also appreciated.

5.8 REFERENCES

1. Pham, P.X., Bodisco, T.A., Ristovski, Z.D., Brown, R.J., and Masri, A.R., The Influence of Fatty Acid Methyl Ester Profiles on Inter-Cycle Variability

Chapter 5: Effect of atmospheric aging on volatility and ROS of biodiesel exhaust nanoparticles 163

in a Heavy Duty Compression Ignition Engine. Fuel, 2014. 116(0): p. 140- 150. 2. Pourkhesalian, A.M., Stevanovic, S., Salimi, F., Rahman, M., Wang, H., Pham, P.X., Bottle, S.E., Masri, A., Brown, R.J., and Ristovski, Z.D., Influence of Fuel Molecular Structure on the Volatility and Oxidative Potential of Biodiesel Particulate Matter. Environmental Science & Technology, 2014. 48(21): p. 12577-12585. 3. Pey, J., Querol, X., Alastuey, A., Rodríguez, S., Putaud, J.P., and Van Dingenen, R., Source Apportionment of Urban Fine and Ultra-Fine Particle Number Concentration in a Western Mediterranean City. Atmospheric Environment, 2009. 43(29): p. 4407-4415. 4. World Health Organization, Outdoor Air Pollution a Leading Environmental Cause of Cancer Deaths. IARC scientific publications, 2013: p. 1-4. 5. Surawski, N.C., Miljevic, B., Bodisco, T.A., Brown, R.J., Ristovski, Z.D., and Ayoko, G.A., Application of Multicriteria Decision Making Methods to Compression Ignition Engine Efficiency and Gaseous, Particulate, and Greenhouse Gas Emissions. Environmental Science & Technology, 2013. 47(4): p. 1904-1912. 6. Fontaras, G., Karavalakis, G., Kousoulidou, M., Tzamkiozis, T., Ntziachristos, L., Bakeas, E., Stournas, S., and Samaras, Z., Effects of Biodiesel on Passenger Car Fuel Consumption, Regulated and Non- Regulated Pollutant Emissions over Legislated and Real-World Driving Cycles. Fuel, 2009. 88(9): p. 1608-1617. 7. Rahman, M.M., Pourkhesalian, A.M., Jahirul, M.I., Stevanovic, S., Pham, P.X., Wang, H., Masri, A.R., Brown, R.J., and Ristovski, Z.D., Particle Emissions from Biodiesels with Different Physical Properties and Chemical Composition. Fuel, 2014. 134(0): p. 201-208. 8. Tinsdale, M., Price, P., and Chen, R., The Impact of Biodiesel on Particle Number, Size and Mass Emissions from a Euro4 Diesel Vehicle. 2010. 9. Kalligeros, S., Zannikos, F., Stournas, S., Lois, E., Anastopoulos, G., Teas, C., and Sakellaropoulos, F., An Investigation of Using Biodiesel/Marine Diesel Blends on the Performance of a Stationary Diesel Engine. Biomass and Bioenergy, 2003. 24(2): p. 141-149. 10. Cheng, C.H., Cheung, C.S., Chan, T.L., Lee, S.C., and Yao, C.D., Experimental Investigation on the Performance, Gaseous and Particulate Emissions of a Methanol Fumigated Diesel Engine. Science of the Total Environment, 2008. 389(1): p. 115-124. 11. Bagley, S.T., Gratz, L.D., Johnson, J.H., and McDonald, J.F., Effects of an Oxidation Catalytic Converter and a Biodiesel Fuel on the Chemical, Mutagenic, and Particle Size Characteristics of Emissions from a Diesel Engine. Environmental Science & Technology, 1998. 32(9): p. 1183-1191. 12. Knothe, G., Sharp, C.A., and Ryan III, T.W., Exhaust Emissions of Biodiesel, Petrodiesel, Neat Methyl Esters, and Alkanes in a New Technology Engine. Energy & Fuels, 2006. 20(1): p. 403-408. 13. Sidhu, S., Graham, J., and Striebich, R., Semi-Volatile and Particulate Emissions from the Combustion of Alternative Diesel Fuels. Chemosphere, 2001. 42(5-7): p. 681-690. 14. Liu, Y.Y., Lin, T.C., Wang, Y.J., and Ho, W.L., Biological Toxicities of Emissions from an Unmodified Engine Fueled with Diesel and Biodiesel

164 Chapter 5: Effect of atmospheric aging on volatility and ROS of biodiesel exhaust nanoparticles 165

Blend. Journal of Environmental Science and Health Part a-Toxic/Hazardous Substances & Environmental Engineering, 2008. 43(14): p. 1735-1743. 15. Surawski, N.C., Miljevic, B., Ayoko, G.A., Roberts, B.A., Elbagir, S., Fairfull-Smith, K.E., Bottle, S.E., and Ristovski, Z.D., Physicochemical Characterization of Particulate Emissions from a Compression Ignition Engine Employing Two Injection Technologies and Three Fuels. Environmental Science & Technology, 2011. 45(13): p. 5498-5505. 16. Biswas, S., Verma, V., Schauer, J.J., Cassee, F.R., Cho, A.K., and Sioutas, C., Oxidative Potential of Semi-Volatile and Non Volatile Particulate Matter (Pm) from Heavy-Duty Vehicles Retrofitted with Emission Control Technologies. Environmental Science & Technology, 2009. 43(10): p. 3905- 3912. 17. Zielinska, B., Samy, S., McDonald, J.D., Seagrave, J., and Committee, H.H.R., Atmospheric Transformation of Diesel Emissions. Res Rep Health Eff Inst, 2010(147): p. 5-60. 18. Jathar, S.H., Miracolo, M.A., Tkacik, D.S., Donahue, N.M., Adams, P.J., and Robinson, A.L., Secondary Organic Aerosol Formation from Photo- Oxidation of Unburned Fuel: Experimental Results and Implications for Aerosol Formation from Combustion Emissions. Environmental Science & Technology, 2013. 47(22): p. 12886-12893. 19. Chirico, R., DeCarlo, P.F., Heringa, M.F., Tritscher, T., Richter, R., Prevot, A.S.H., Dommen, J., Weingartner, E., Wehrle, G., Gysel, M., Laborde, M., and Baltensperger, U., Impact of Aftertreatment Devices on Primary Emissions and Secondary Organic Aerosol Formation Potential from in-Use Diesel Vehicles: Results from Smog Chamber Experiments. Atmospheric Chemistry and Physics, 2010. 10(23): p. 11545-11563. 20. Leskinen, A.P., Jokinlemi, J.K., and Lehtinen, K.E.J., Transformation of Diesel Engine Exhaust in an Environmental Chamber. Atmospheric Environment, 2007. 41(39): p. 8865-8873. 21. Miracolo, M.A., Hennigan, C.J., Ranjan, M., Nguyen, N.T., Gordon, T.D., Lipsky, E.M., Presto, A.A., Donahue, N.M., and Robinson, A.L., Secondary Aerosol Formation from Photochemical Aging of Aircraft Exhaust in a Smog Chamber. Atmospheric Chemistry and Physics, 2011. 11(9): p. 4135-4147. 22. Miracolo, M.A., Presto, A.A., Lambe, A.T., Hennigan, C.J., Donahue, N.M., Kroll, J.H., Worsnop, D.R., and Robinson, A.L., Photo-Oxidation of Low- Volatility Organics Found in Motor Vehicle Emissions: Production and Chemical Evolution of Organic Aerosol Mass. Environmental Science & Technology, 2010. 44(5): p. 1638-1643. 23. Nakao, S., Shrivastava, M., Nguyen, A., Jung, H.J., and Cocker, D., Interpretation of Secondary Organic Aerosol Formation from Diesel Exhaust Photooxidation in an Environmental Chamber. Aerosol Science and Technology, 2011. 45(8): p. 964-974. 24. Samy, S. and Zielinska, B., Secondary Organic Aerosol Production from Modern Diesel Engine Emissions. Atmospheric Chemistry and Physics, 2010. 10(2): p. 609-625. 25. Verma, V., Ning, Z., Cho, A.K., Schauer, J.J., Shafer, M.M., and Sioutas, C., Redox Activity of Urban Quasi-Ultrafine Particles from Primary and Secondary Sources. Atmospheric Environment, 2009. 43(40): p. 6360-6368. 26. Wang, J., Doussin, J.F., Perrier, S., Perraudin, E., Katrib, Y., Pangui, E., and Picquet-Varrault, B., Design of a New Multi-Phase Experimental Simulation

Chapter 5: Effect of atmospheric aging on volatility and ROS of biodiesel exhaust nanoparticles 165

Chamber for Atmospheric Photosmog, Aerosol and Cloud Chemistry Research. Atmos. Meas. Tech., 2011. 4(11): p. 2465-2494. 27. Weitkamp, E.A., Sage, A.M., Pierce, J.R., Donahue, N.M., and Robinson, A.L., Organic Aerosol Formation from Photochemical Oxidation of Diesel Exhaust in a Smog Chamber. Environmental Science & Technology, 2007. 41(20): p. 6969-6975. 28. Geiger, H., Kleffmann, J., and Wiesen, P., Smog Chamber Studies on the Influence of Diesel Exhaust on Photosmog Formation. Atmospheric Environment, 2002. 36(11): p. 1737-1747. 29. Lee, S., Jang, M., and Kamens, R.M., Soa Formation from the Photooxidation of Α-Pinene in the Presence of Freshly Emitted Diesel Soot Exhaust. Atmospheric Environment, 2004. 38(16): p. 2597-2605. 30. Robinson, A.L., Donahue, N.M., Shrivastava, M.K., Weitkamp, E.A., Sage, A.M., Grieshop, A.P., Lane, T.E., Pierce, J.R., and Pandis, S.N., Rethinking Organic Aerosols: Semivolatile Emissions and Photochemical Aging. Science, 2007. 315(5816): p. 1259-1262. 31. Li, Q., Wyatt, A., and Kamens, R.M., Oxidant Generation and Toxicity Enhancement of Aged-Diesel Exhaust. Atmospheric Environment, 2009. 43(5): p. 1037-1042. 32. Izumi, K. and Fukuyama, T., Photochemical Aerosol Formation from Aromatic-Hydrocarbons in the Presence of Nox. Atmospheric Environment Part a-General Topics, 1990. 24(6): p. 1433-1441. 33. Atkinson, R., Carter, W.P.L., Darnall, K.R., Winer, A.M., and Pitts, J.N., A Smog Chamber and Modeling Study of the Gas-Phase Nox-Air Photo- Oxidation of Toluene and the Cresols. International Journal of Chemical Kinetics, 1980. 12(11): p. 779-836. 34. Kang, E., Root, M.J., Toohey, D.W., and Brune, W.H., Introducing the Concept of Potential Aerosol Mass (Pam). Atmos. Chem. Phys., 2007. 7(22): p. 5727-5744. 35. Lambe, A.T., Ahern, A.T., Williams, L.R., Slowik, J.G., Wong, J.P.S., Abbatt, J.P.D., Brune, W.H., Ng, N.L., Wright, J.P., Croasdale, D.R., Worsnop, D.R., Davidovits, P., and Onasch, T.B., Characterization of Aerosol Photooxidation Flow Reactors: Heterogeneous Oxidation, Secondary Organic Aerosol Formation and Cloud Condensation Nuclei Activity Measurements. Atmos. Meas. Tech., 2011. 4(3): p. 445-461. 36. Keller, A. and Burtscher, H., A Continuous Photo-Oxidation Flow Reactor for a Defined Measurement of the Soa Formation Potential of Wood Burning Emissions. Journal of Aerosol Science, 2012. 49(0): p. 9-20. 37. McWhinney, R.D., Gao, S.S., Zhou, S.M., and Abbatt, J.P.D., Evaluation of the Effects of Ozone Oxidation on Redox-Cycling Activity of Two-Stroke Engine Exhaust Particles. Environmental Science & Technology, 2011. 45(6): p. 2131-2136. 38. Jamriska, M., Morawska, L., Thomas, S., and He, C., Diesel Bus Emissions Measured in a Tunnel Study. Environmental Science & Technology, 2004. 38(24): p. 6701-6709. 39. Johnson, G.R., Fletcher, C., Meyer, N., Modini, R., and Ristovski, Z., A Robust, Portable H-Tdma for Field Use. Journal of Aerosol Science, 2008. 39(10): p. 850-861. 40. Miljevic, B., Heringa, M.F., Keller, A., Meyer, N.K., Good, J., Lauber, A., Decarlo, P.F., Fairfull-Smith, K.E., Nussbaumer, T., Burtscher, H., Prevot,

166 Chapter 5: Effect of atmospheric aging on volatility and ROS of biodiesel exhaust nanoparticles 167

A.S.H., Baltensperger, U., Bottle, S.E., and Ristovski, Z.D., Oxidative Potential of Logwood and Pellet Burning Particles Assessed by a Novel Profluorescent Nitroxide Probe. Environmental Science & Technology, 2010. 44(17): p. 6601-6607. 41. Stevanovic, S., Miljevic, B., Eaglesham, G.K., Bottle, S.E., Ristovski, Z.D., and Fairfull-Smith, K.E., The Use of a Nitroxide Probe in Dmso to Capture Free Radicals in Particulate Pollution. European Journal of Organic Chemistry, 2012. 2012(30): p. 5908-5912. 42. Stevanovic, S., Miljevic, B., Surawski, N.C., Fairfull-Smith, K.E., Bottle, S.E., Brown, R., and Ristovski, Z.D., Influence of Oxygenated Organic Aerosols (Ooas) on the Oxidative Potential of Diesel and Biodiesel Particulate Matter. Environmental Science & Technology, 2013. 47(14): p. 7655-7662. 43. Miljevic, B., Modini, R.L., Bottle, S.E., and Ristovski, Z.D., On the Efficiency of Impingers with Fritted Nozzle Tip for Collection of Ultrafine Particles. Atmospheric Environment, 2009. 43(6): p. 1372-1376. 44. Stevanovic, S., Ristovski, Z.D., Miljevic, B., Fairfull-Smith, K.E., and Bottle, S.E., Application of Profluorescent Nitroxides for Measurements of Oxidative Capacity of Combustion Generated Particles. Chemical Industry & Chemical Engineering Quarterly, 2012. 18(4): p. 653-659. 45. Donahue, N.M., Robinson, A.L., Stanier, C.O., and Pandis, S.N., Coupled Partitioning, Dilution, and Chemical Aging of Semivolatile Organics. Environmental Science & Technology, 2006. 40(8): p. 2635-2643. 46. Kroll, J.H. and Seinfeld, J.H., Chemistry of Secondary Organic Aerosol: Formation and Evolution of Low-Volatility Organics in the Atmosphere. Atmospheric Environment, 2008. 42(16): p. 3593-3624. 47. Lipsky, E.M. and Robinson, A.L., Effects of Dilution on Fine Particle Mass and Partitioning of Semivolatile Organics in Diesel Exhaust and Wood Smoke. Environmental Science & Technology, 2006. 40(1): p. 155-162. 48. Shi, J.P. and Harrison, R.M., Investigation of Ultrafine Particle Formation During Diesel Exhaust Dilution. Environmental Science & Technology, 1999. 33(21): p. 3730-3736. 49. Rönkkö, T., Virtanen, A., Vaaraslahti, K., Keskinen, J., Pirjola, L., and Lappi, M., Effect of Dilution Conditions and Driving Parameters on Nucleation Mode Particles in Diesel Exhaust: Laboratory and on-Road Study. Atmospheric Environment, 2006. 40(16): p. 2893-2901. 50. Xu, Z., Li, X., Guan, C., and Huang, Z., Characteristics of Exhaust Diesel Particles from Different Oxygenated Fuels. Energy & Fuels, 2013. 27: p. 7579–7586. 51. Cheung, K.L., Polidori, A., Ntziachristos, L., Tzamkiozis, T., Samaras, Z., Cassee, F.R., Gerlofs, M., and Sioutas, C., Chemical Characteristics and Oxidative Potential of Particulate Matter Emissions from Gasoline, Diesel, and Biodiesel Cars. Environmental Science & Technology, 2009. 43(16): p. 6334-6340. 52. Giechaskiel, B., Alfoldy, B., and Drossinos, Y., A Metric for Health Effects Studies of Diesel Exhaust Particles. Journal of Aerosol Science, 2009. 40(8): p. 639-651. 53. Wickham, H., Ggplot2: Elegant Graphics for Data Analysis. 2009: Springer. 54. Heywood, J.B., Internal Combustion Engine Fundamentals. Vol. 930. 1988: Mcgraw-hill New York.

Chapter 5: Effect of atmospheric aging on volatility and ROS of biodiesel exhaust nanoparticles 167

55. Jung, H., Kittelson, D.B., and Zachariah, M.R., Characteristics of Sme Biodiesel-Fueled Diesel Particle Emissions and the Kinetics of Oxidation. Environmental Science & Technology, 2006. 40(16): p. 4949-4955. 56. Jalava, P.I., Aakko-Saksa, P., Murtonen, T., Happo, M.S., Markkanen, A., Yli-Pirilä, P., Hakulinen, P., Hillamo, R., Mäki-Paakkanen, J., and Salonen, R.O., Toxicological Properties of Emission Particles from Heavy Duty Engines Powered by Conventional and Bio-Based Diesel Fuels and Compressed Natural Gas. Particle and fibre toxicology, 2012. 9(1): p. 37. 57. Giechaskiel, B., Mamakos, A., Andersson, J., Dilara, P., Martini, G., Schindler, W., and Bergmann, A., Measurement of Automotive Nonvolatile Particle Number Emissions within the European Legislative Framework: A Review. Aerosol Science and Technology, 2012. 46(7): p. 719-749.

168 Chapter 5: Effect of atmospheric aging on volatility and ROS of biodiesel exhaust nanoparticles

Chapter 6: Conclusions

Aerosols originating from diesel engines can have important adverse effects on the health of people and the environment. Understanding the physical and chemical dynamics and transformations of DPM helps us to comprehend their effects on climate and human health, find mitigation techniques and redefine regulations. The research program reported here has contributed significantly to scientific knowledge on the effect of fuel type and atmospheric conditions on DPM, as well as its dynamics and transformation.

The overall aim of this research was to investigate the effects of physical and chemical properties of fuel on DPM and also to examine the effect of atmospheric dilution conditions on the transformation of DPM. Firstly, this study investigated the effect of biodiesel fuel on the performance and regulated emissions of the engine. Then, it takes a closer look into the emission of particulate matter from different types of biodiesel fuels. The effect of fuel chemical characteristics on the organic fraction of DPM was investigated and the potential toxicity of such particles was also examined. Finally, the aging of DPM and diesel exhaust in the atmosphere, as well as in the proximity of oxidative agents, was also studied.

6.1 PRINCIPAL SIGNIFICANCE OF THE FINDINGS

This research has improved our insight into how the chemical characteristics and physical properties of biodiesel affect the emission and aging of DE and DPM. The results of the experimental campaigns were used in order to produce three manuscripts, which were/will be published in reputable scientific journals.

The 1st manuscript evaluated the effects of using different blends (20%, 50% and 100%) of biodiesel on the trends of emitted NOx, PMC and PNC. A linear relationship was detected between particle emissions and oxygen content of the fuel, which was independent from the quantity or type of biodiesel in the blend. However, since the weight ratio of oxygen in the molecule of the fuel increased with a decrease in carbon chain length, the aforementioned effects cannot be merely assigned to either carbon chain length or oxygen content. Particulate emissions also decreased linearly with a decrease in viscosity and surface tension of certain fuels and blends.

Error! No text of specified style in document. 169

Fuel type also affected particle size, whereby blends that emitted higher PMC and PNC were also found to be responsible for larger particle median diameters. Overall, this section of the study highlighted the greater importance of chemical composition of biodiesel over its physical properties in controlling exhaust PM emissions.

The 2nd manuscript looked further into the physical and chemical properties of particles emitted from the same biodiesel feedstocks used in the 1st manuscript. It aimed to reconfirm a direct relationship between the volatility and oxidative potential of particles, and also to understand correlations between the volatility of DPM and the unsaturation degree, oxygen content and carbon chain length of the fuel. Particles emitted from short chain saturated biodiesel combustion were found to carry a higher fraction of volatile organic matter and higher levels of ROS. It was found that the OV and ROS of PM had a positive exponential correlation with the oxygen content of the fuel used. This increase in OV and ROS could be due to either a decrease in the available non-volatile soot onto which the volatiles condense or an actual increase in the amount of volatile material. Following a comparison with black carbon emissions, it was concluded that the observed exponential relationship was likely to be a combined effect of both parameters. These results further highlight the role of biodiesel FAME composition in determining the possible health effects of particles emitted from biodiesel combustion in diesel engines. This study also challenges current mass and number based emission standards, and points to the fact that less PM does not necessarily mean less harm to human health and the environment.

The 3rd manuscript investigated the possible influence of atmospheric dilution and exposure to oxidants on the extent of partitioning of the semi-volatile fraction of diesel exhaust between the vapor and condensed phases. It also endeavored to determine the ROS of PM after being released into the atmosphere and undergoing photochemical reactions. The results of this study showed that the chemical composition of the fuel affected both the primary emissions, as well as secondary emissions. While it was confirmed that saturated fuels emitted less PM, they also led to more volatile PM and significantly more semi-volatiles in the gas phase, being capable of partitioning between the gas phase and particle phase upon aging. Aged particles from more saturated fuels with a higher oxygen content have a higher oxidative potential, meaning that they may be more toxic. The results of this research

170 Error! No text of specified style in document. 171 challenges the findings of a few previous studies claiming that use of biodiesel would not increase the level of toxicity of diesel exhaust.

Undoubtedly, the use of biodiesel has many advantages, in particular their ability to substantially reduce diesel particle mass emissions, and by regulating biodiesel FAME composition, the harmful aspects of biodiesel use could be minimized. By integrating the findings of previous and present studies, we can conclude that while biodiesel use can minimize PM emissions, it will not necessarily reduce the toxicity of those emissions. On the contrary, oxygenated, short-chained, saturated biodiesel emitted more toxic PM. Interestingly, the PM emitted from such biodiesels tended to become even more reactive after they were oxidized in the atmosphere.

Based on the findings of this thesis, it is suggested that current biodiesel standards should be redefined, in order to maximize the advantages associated with their use. Where current biodiesel standards, ASTM D6751-12 and EN 14214, focus on the physical properties of biodiesel, it is recommended that more attention be paid to the chemical composition of biodiesel fuel, since biodiesel FAME chemical composition is just as important as the physical properties of biodiesel.

6.2 DIRECTIONS FOR FUTURE RESEARCH

As mentioned previously, an increase in oxygen content and decrease in carbon chain length was found to correlate with reductions of PM in the exhaust gas. However, these two factors were also related to each other, and therefore, the ‘PM reduction’ effect cannot be solely assigned to each one of these factors. More in- depth experiments are necessary to be able to mark either carbon chain length or oxygen content as the main driver behind PM reductions in diesel exhaust. For instance, biodiesel fuels with the same oxygen content but different carbon chain length can be tested for PM concentration, volatile fraction and ROS levels, in order to distinguish the most influential factors.

The PAM chamber used in the third paper has its own drawbacks and cannot always be regarded as a good simulator of atmospheric conditions. In order to carry out quantitative experiments and gain greater insight into the matter, aging experiments could be repeated in an environmental chamber where the concentration of oxidative agents and exposure time are more realistic.

Error! No text of specified style in document. 171

In the aging experiments, there were cases where the volatility of PM increased, but it did not cause any increase in the ROS concentrations of the PM. It was hypothesized that some of the volatiles in the gas phase were not as reactive as others when oxidized. This assumption needs more detailed. While we did not have the opportunity to use it in this work, the Aerosol Mass Spectrometer (AMS) can provide additional data on the different organic compounds that have partitioned into the particle phase as a result of aging. Similarly, the use of an AMS in future campaigns is recommended to provide further information on the chemical composition of particulate matter, in order to more easily identify potential correlations.

Finally, a review of current literature highlights significant inconsistencies in the results reported by different laboratories, which emphasizes the need to develop a pervasive measurement technique. Hence, it would be beneficial to undertake comprehensive research into putting together a standard experimental procedure that was able to generate repeatable and reproducible data and experiments.

172 Error! No text of specified style in document.