Engine Exhaust Particulates PDF

This book provides a comparative analysis of both diesel and gasoline engine particulates, and also of the emissions resulting from the use of alternative fuels. Written by respected experts, it offers comprehensive insights into motor vehicle particulates, their formation, composition, location, measurement, characterisation and toxicology. It also addresses exhaust-gas treatment and legal, measurement-related and technological advancements concerning emissions. The book will serve as a valuable resource for academic researchers and professional automotive engineers alike.

113 downloads 3K Views 8MB Size

Report

Download pdf

Recommend Stories

Empty story

Idea Transcript

Energy, Environment, and Sustainability Series Editors: Avinash Kumar Agarwal · Ashok Pandey

Avinash Kumar Agarwal Atul Dhar Nikhil Sharma Pravesh Chandra Shukla Editors

Engine Exhaust Particulates

Energy, Environment, and Sustainability Series editors Avinash Kumar Agarwal, Department of Mechanical Engineering, Indian Institute of Technology Kanpur, Kanpur, Uttar Pradesh, India Ashok Pandey, Distinguished Scientist, CSIR-Indian Institute of Toxicology Research, Lucknow, Uttar Pradesh, India

This books series publishes cutting edge monographs and professional books focused on all aspects of energy and environmental sustainability, especially as it relates to energy concerns. The Series is published in partnership with the International Society for Energy, Environment, and Sustainability. The books in these series are editor or authored by top researchers and professional across the globe. The series aims at publishing state-of-the-art research and development in areas including, but not limited to: • • • • • • • • • •

Renewable Energy Alternative Fuels Engines and Locomotives Combustion and Propulsion Fossil Fuels Carbon Capture Control and Automation for Energy Environmental Pollution Waste Management Transportation Sustainability

More information about this series at http://www.springer.com/series/15901

Avinash Kumar Agarwal Atul Dhar Nikhil Sharma Pravesh Chandra Shukla •

Editors

Engine Exhaust Particulates

123

Editors Avinash Kumar Agarwal Department of Mechanical Engineering Indian Institute of Technology Kanpur Kanpur, Uttar Pradesh, India

Nikhil Sharma Department of Mechanical Engineering Indian Institute of Technology Kanpur Kanpur, Uttar Pradesh, India

Atul Dhar School of Engineering Indian Institute of Technology Mandi Mandi, Himachal Pradesh, India

Pravesh Chandra Shukla Department of Mechanical Engineering Indian Institute of Technology Bhilai Bhilai, Chhattisgarh, India

ISSN 2522-8366 ISSN 2522-8374 (electronic) Energy, Environment, and Sustainability ISBN 978-981-13-3298-2 ISBN 978-981-13-3299-9 (eBook) https://doi.org/10.1007/978-981-13-3299-9 Library of Congress Control Number: 2018961730 © Springer Nature Singapore Pte Ltd. 2019 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, speciﬁcally the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microﬁlms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a speciﬁc statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional afﬁliations. This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd. The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721, Singapore

Preface

Energy demand has been rising remarkably due to increasing population and urbanization. Global economy and society are signiﬁcantly dependent on the energy availability because it touches every facet of human life and its activities. Transportation and power generation are two major examples. Without the transportation by millions of personalized and mass transport vehicles and availability of 24 7 power, human civilization would not have reached contemporary living standards. The International Society for Energy, Environment and Sustainability (ISEES) was founded at Indian Institute of Technology Kanpur (IIT Kanpur), India, in January 2014 with the aim of spreading knowledge/awareness and catalysing research activities in the ﬁelds of energy, environment, sustainability and combustion. The society’s goal is to contribute to the development of clean, affordable and secure energy resources and a sustainable environment for the society and to spread knowledge in the above-mentioned areas and create awareness about the environmental challenges, which the world is facing today. The unique way adopted by the society was to break the conventional silos of specializations (engineering, science, environment, agriculture, biotechnology, materials, fuels, etc.) to tackle the problems related to energy, environment and sustainability in a holistic manner. This is quite evident by the participation of experts from all ﬁelds to resolve these issues. ISEES is involved in various activities such as conducting workshops, seminars and conferences in the domains of its interest. The society also recognizes the outstanding works done by the young scientists and engineers for their contributions in these ﬁelds by conferring them awards under various categories. The second international conference on “Sustainable Energy and Environmental Challenges” (SEEC-2018) was organized under the auspices of ISEES from 31 December 2017 to 3 January 2018 at J N Tata Auditorium, Indian Institute of Science Bangalore. This conference provided a platform for discussions between eminent scientists and engineers from various countries including India, USA, South Korea, Norway, Finland, Malaysia, Austria, Saudi Arabia and Australia. In this conference, eminent speakers from all over the world presented their views v

vi

Preface

related to different aspects of energy, combustion, emissions and alternative energy resources for sustainable development and a cleaner environment. The conference presented ﬁve high-voltage plenary talks from globally renowned experts on topical themes, namely “Is It Really the End of Combustion Engines and Petroleum?” by Prof. Gautam Kalghatgi, Saudi Aramco; “Energy Sustainability in India: Challenges and Opportunities” by Prof. Baldev Raj, NIAS Bangalore; “Methanol Economy: An Option for Sustainable Energy and Environmental Challenges” by Dr. Vijay Kumar Saraswat, Hon. Member (S&T), NITI Aayog, Government of India; “Supercritical Carbon Dioxide Brayton Cycle for Power Generation” by Prof. Pradip Dutta, IISc Bangalore; and “Role of Nuclear Fusion for Environmental Sustainability of Energy in Future” by Prof. J. S. Rao, Altair Engineering. The conference included 27 technical sessions on topics related to energy and environmental sustainability including 5 plenary talks, 40 keynote talks and 18 invited talks from prominent scientists, in addition to 142 contributed talks, and 74 poster presentations by students and researchers. The technical sessions in the conference included Advances in IC Engines: SI Engines, Solar Energy: Storage, Fundamentals of Combustion, Environmental Protection and Sustainability, Environmental Biotechnology, Coal and Biomass Combustion/Gasiﬁcation, Air Pollution and Control, Biomass to Fuels/Chemicals: Clean Fuels, Advances in IC Engines: CI Engines, Solar Energy: Performance, Biomass to Fuels/Chemicals: Production, Advances in IC Engines: Fuels, Energy Sustainability, Environmental Biotechnology, Atomization and Sprays, Combustion/Gas Turbines/Fluid Flow/Sprays, Biomass to Fuels/Chemicals, Advances in IC Engines: New Concepts, Energy Sustainability, Waste to Wealth, Conventional and Alternate Fuels, Solar Energy, Wastewater Remediation and Air Pollution. One of the highlights of the conference was the rapid-ﬁre poster sessions in (i) Energy Engineering, (ii) Environment and Sustainability and (iii) Biotechnology, where more than 75 students participated with great enthusiasm and won many prizes in a ﬁercely competitive environment. More than 200 participants and speakers attended this four-day conference, which also hosted Dr. Vijay Kumar Saraswat, Hon. Member (S&T), NITI Aayog, Government of India, as the chief guest for the book release ceremony, where 16 ISEES books published by Springer, Singapore, under a special dedicated series “Energy, Environment, and Sustainability” were released. This is the ﬁrst time that such signiﬁcant and high-quality outcome has been achieved by any society in India. The conference concluded with a panel discussion on “Challenges, Opportunities & Directions for Future Transportation Systems”, where the panellists were Prof. Gautam Kalghatgi, Saudi Aramco; Dr. Ravi Prashanth, Caterpillar Inc.; Dr. Shankar Venugopal, Mahindra and Mahindra; Dr. Bharat Bhargava, DG, ONGC Energy Center; and Dr. Umamaheshwar, GE Transportation, Bangalore. The panel discussion was moderated by Prof. Ashok Pandey, Chairman, ISEES. This conference laid out the road map for technology development, opportunities and challenges in energy, environment and sustainability domains. All these topics are very relevant for the country and the world in the present context. We acknowledge the support received from various funding agencies and organizations for the successful conduct of the second ISEES

Preface

vii

conference SEEC-2018, where these books germinated. We would therefore like to acknowledge SERB, Government of India (special thanks to Dr. Rajeev Sharma, Secretary); ONGC Energy Center (special thanks to Dr. Bharat Bhargava); TAFE (special thanks to Sh. Anadrao Patil); Caterpillar (special thanks to Dr. Ravi Prashanth); Progress Rail, TSI, India (special thanks to Dr. Deepak Sharma); Tesscorn, India (special thanks to Sh. Satyanarayana); GAIL, Volvo; and our publishing partner Springer (special thanks to Swati Mehershi). The editors would like to express their sincere gratitude to a large number of authors from all over the world for submitting their high-quality work in a timely manner and revising it appropriately at short notice. We would like to express our special thanks to Dr. Tapan Kumar Pradhan, Dr. Atul Dhar, Dr. Akhilendra Pratap Singh, Dr. Ludovica Luise, Dr. Joonsik Hwang, Dr. Chetan Patel, Dr. Pravesh Chandra Shukla, Dr. Sundeep Singh, Dr. Rohit Singla, Dr. Rajesh Prasad, Dr. Vikram Kumar, Dr. Dev Prakash Satsangi, Dr. Anoop Kumar Shukla, Mr. Maneesh Kumar, Mr. Neeraj Sharma, Mr. Sunil Kumar, Mr. Yeshudas Jiotode and Mr. Pawan Kumar, who reviewed various chapters of this book and provided very valuable suggestions to the authors to improve their manuscript. This book covers different aspects of both diesel and gasoline engine particulates. The ﬁrst half of this book is about diesel engine particulates, and the second half of this book is about gasoline engine particulates. This book provides a comprehensive insight into the motor vehicles’ particulates, its formation and composition, location of particulates, measurement, characterization and toxicology. This book also focuses on exhaust after-treatment devices and their comparison. Apart from this, the effect of engine design and operation variables, emission legislation and emission measurement are presented. Engine exhaust after-treatment concepts such as HC adsorbed systems, NO traps and advanced engines like RCCI, GDI and HCCI engines are covered in this book. The text in every chapter is complemented by illustrations and is written by ﬁeld expert. Kanpur, India

Avinash Kumar Agarwal Atul Dhar Nikhil Sharma Pravesh Chandra Shukla

Contents

Part I

General

1

Introduction to Engine Exhaust Particulates . . . . . . . . . . . . . . . . . . Avinash Kumar Agarwal, Atul Dhar, Nikhil Sharma and Pravesh Chandra Shukla

2

Ultraﬁne Particles in Concern of Vehicular Exhaust—An Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Shailendra Kumar Yadav, Rajeev Kumar Mishra and Bhola Ram Gurjar

Part II 3

4

3

7

Diesel Particulates

Image-Based Flame Temperature and Soot Analysis of Biofuel Spray Combustion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Joonsik Hwang, Felix Sebastian Hirner, Choongsik Bae, Chetankumar Patel, Tarun Gupta and Avinash Kumar Agarwal Characteristics and Fundamentals of Particulates in Diesel Engine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Khawar Mohiuddin and Sungwook Park

41

55 71

5

Numerical Modelling of Soot in Diesel Engines . . . . . . . . . . . . . . . . Pavan Prakash Duvvuri, Rajesh Kumar Shrivastava and Sheshadri Sreedhara

6

Physico-chemical Properties of Diesel Exhaust Particulates . . . . . . 121 Jianbing Gao and Guohong Tian

Part III 7

Alternate Fuel Origin Particulates

Oxygenated Fuel Additive Option for PM Emission Reduction from Diesel Engines—A Review . . . . . . . . . . . . . . . . . . . . . . . . . . . 141 Parameswaran Vijayashree and V. Ganesan

ix

x

Contents

8

Technological Evolution of Spark Ignition Direct Injection Engine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165 Nikhil Sharma and Avinash Kumar Agarwal

9

Alternative Fuels for Particulate Control in CI Engines . . . . . . . . . 181 Sam Shamun, Pablo Garcia and Erik Svensson

10 Particulate Emissions from Hydrogen Diesel Fuelled CI Engines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199 Priybrat Sharma and Atul Dhar Part IV

Gasoline Particulates

11 Particulate Emission from Gasoline Direct Injection Engine . . . . . . 215 Ludovica Luise 12 Nanoparticle Emissions in Reactivity-Controlled Compression Ignition Engine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239 Mohit Raj Saxena and Rakesh Kumar Maurya

Editors and Contributors

About the Editors Avinash Kumar Agarwal is Professor in the Department of Mechanical Engineering at Indian Institute of Technology Kanpur. His areas of interest are IC engines, combustion, alternative fuels, conventional fuels, optical diagnostics, laser ignition, HCCI, emission and particulate control, and large bore engines. He has published 24 books and more than 230 international journal and conference papers. He is Fellow of SAE (2012), ASME (2013), ISEES (2015) and INAE (2015). He has received several awards such as the prestigious Shanti Swarup Bhatnagar Award in engineering sciences (2016), Rajib Goyal Prize (2015) and NASI-Reliance Industries Platinum Jubilee Award (2012). Atul Dhar has been Assistant Professor at IIT Mandi since 2013. He received his M.Tech. and Ph.D. from the Department of Mechanical Engineering, IIT Kanpur, in 2006 and 2013, respectively. He was awarded the European Union Erasmus Mundus Fellowship in 2013 and the Young Scientist Award from the International Society for Energy, Environment and Sustainability in 2015. He has edited 3 books and published 12 chapters and more than 35 international peer-reviewed journal papers.

xi

xii

Editors and Contributors

Nikhil Sharma is Scientist at the Engine Research Laboratory, IIT Kanpur, India. He received his M.Tech. in mechanical engineering from NIT Hamirpur, India, in 2012, and his Ph.D. from IIT Kanpur in 2017. He was former Assistant Professor at Amity University’s Department of Mechanical and Automation Engineering, Noida. His areas of research include alternative fuels for IC engines (biodiesel, alcohols), emission control and particulate characterization.

Pravesh Chandra Shukla is Assistant Professor in the Department of Mechanical Engineering, IIT Bhilai. After completing his Ph.D. at IIT Kanpur, he worked as a postdoc at Lund University, Sweden. His research interests include internal combustion engines, alternative fuels (biodiesel, alcohols, HVO), diesel emissions and their control, unregulated emissions from diesel and gasoline engines, after-treatment devices and fuel spray and its characterization. He has authored 25 research publications and 2 chapters.

Contributors Avinash Kumar Agarwal Engine Research Laboratory, Department of Mechanical Engineering, Indian Institute of Technology Kanpur, Kanpur, India Choongsik Bae Department of Mechanical Engineering, Korea Advanced Institute of Science and Technology, Daejeon, Republic of Korea Atul Dhar School of Engineering, Indian Institute of Technology Mandi, Mandi, India Pavan Prakash Duvvuri Department of Energy Science and Engineering, Indian Institute of Technology Bombay, Mumbai, India; Combustion Research, Cummins Technical Center India, Pune, India V. Ganesan Department of Mechanical Engineering, Indian Institute of Technology Madras, Chennai, India

Editors and Contributors

xiii

Jianbing Gao University of Surrey, Surrey, UK Pablo Garcia Lund University, Lund, Sweden Tarun Gupta Department of Civil Engineering, Indian Institute of Technology Kanpur, Kanpur, India Bhola Ram Gurjar Centre for Transportation System, Indian Institute of Technology Roorkee, Roorkee, India Felix Sebastian Hirner Department of Mechanical Engineering, Korea Advanced Institute of Science and Technology, Daejeon, Republic of Korea Joonsik Hwang Department of Mechanical Engineering, Korea Advanced Institute of Science and Technology, Daejeon, Republic of Korea; Combustion Research Facility (CRF), Sandia National Laboratories, Albuquerque, USA Ludovica Luise Oxford Brookes University, Oxford, UK Rakesh Kumar Maurya Advanced Engine and Fuel Research Laboratory, Department of Mechanical Engineering, Indian Institute of Technology Ropar, Punjab, India Rajeev Mishra Department of Environmental Engineering, Delhi Technological University, New Delhi, India Khawar Mohiuddin Department of Mechanical Convergence Engineering, Graduate School of Hanyang University, Seoul, Republic of Korea Sungwook Park School of Mechanical Engineering, Hanyang University, Seoul, Republic of Korea Chetankumar Patel Department of Mechanical Engineering, Indian Institute of Technology Kanpur, Kanpur, India Mohit Raj Saxena Advanced Engine and Fuel Research Laboratory, Department of Mechanical Engineering, Indian Institute of Technology Ropar, Punjab, India Nikhil Sharma Engine Research Laboratory, Department of Mechanical Engineering, Indian Institute of Technology Kanpur, Kanpur, India Priybrat Sharma School of Engineering, Indian Institute of Technology Mandi, Mandi, India Rajesh Kumar Shrivastava Combustion Research, Cummins Technical Center India, Pune, India Pravesh Chandra Shukla Department of Mechanical Engineering, Indian Institute of Technology Bhilai, Bhilai, India

xiv

Editors and Contributors

Sam Shamun Lund University, Lund, Sweden Sheshadri Sreedhara Department of Mechanical Engineering, Indian Institute of Technology Bombay, Mumbai, India Erik Svensson Lund University, Lund, Sweden Guohong Tian University of Surrey, Surrey, UK Vijayashree Department of Mechanical Engineering, Indian Institute of Technology Madras, Chennai, India Shailendra Kumar Yadav Department of Environmental Engineering, Delhi Technological University, New Delhi, India

Part I

General

Chapter 1

Introduction to Engine Exhaust Particulates Avinash Kumar Agarwal, Atul Dhar, Nikhil Sharma and Pravesh Chandra Shukla

Abstract Emission legislations are getting stringent throughout the word and researchers/OEMs are working to meet these challenging emission norms. This book covers different aspects of diesel and gasoline engine particulates. The topics provide a comprehensive insight into motor vehicle origin particulates, its formation, and composition, measurement, characterization and toxicology. This book also focuses on exhaust gas after-treatment devices with emphasis on some basic aspects. Apart from this, particulate emissions from alternative fuels have also been included in this book. The text in every chapter is complemented by illustrations and is written by domain experts. Overall, this book covers a wide range of topic related to engine exhaust particles and will be of interest to established researchers in the ﬁeld as well as upcoming researchers. The topics are organized in three different sections, namely, (i) general, (ii) diesel particulates, (iii) alternate fuel origin particulates. Keywords Diesel engine particulates Alternative fuels RCCI

Gasoline engine particulates

A. K. Agarwal (&) N. Sharma Engine Research Laboratory, Department of Mechanical Engineering, Indian Institute of Technology Kanpur, Kanpur, India e-mail: [email protected] A. Dhar School of Engineering, Indian Institute of Technology Mandi, Mandi, India P. C. Shukla Department of Mechanical Engineering, Indian Institute of Technology Bhilai, Bhilai, India © Springer Nature Singapore Pte Ltd. 2019 A. K. Agarwal et al. (eds.), Engine Exhaust Particulates, Energy, Environment, and Sustainability, https://doi.org/10.1007/978-981-13-3299-9_1

3

4

1.1

A. K. Agarwal et al.

Introduction

There is a growing consensus amongst health experts all over the world that the particles in nano- and ultra-ﬁne range (20 nm diameter (nucleation mode) and later coagulate into the fraction of the ultraﬁne mode. The secondary fraction of UFPs (>100–200 nm) which are formed in the ambient air, including sulfates and sulfuric acid and organic compounds (the product of low volatility), is partially different from other directly human-made sources.

2.3.2

Sources and Characterisation

2.3.2.1

Source

The major emitters of UFPs are both natural and man-made. These UFPs are generally released through homogenous nucleation processes within the atmosphere in the form of primary or secondary or both. The direct sources of UFPs in the atmosphere are trafﬁc emissions, engine combustion, speciﬁc industrial processes (metallurgical process), etc. But during the estimation process of road vehicles emission, there are quite substantial uncertainties due to various reasons such as windblown dust, resuspension from road trafﬁc (Stanier 2003) (see Footnote 5). Although when compared to PM inventory (25%), UFPs emission inventory gives far greater relative importance to emission from road vehicles (60%) (Allen et al. 1999; Wang et al. 1999). Indeed, the observation of particle number count is reflecting the abundance of ultraﬁne particle primarily, but many researchers have justiﬁed it that, such particles (UFP) provide an excellent tracer of road vehicle trafﬁc emissions (Shi et al. 1999a, b). UFPs are discharged from practically every fuel ignition process, including diesel, gas jet engine, and also wood burning (external burning also). Thus, there is a developing worry that individuals living close to profoundly trafﬁcked roadways and other sources like airport terminals and rail yards contribute profoundly toward a variety of emissions generated from the burning of fuels which is generally present in high levels of UFPs and other air toxins (Junker et al. 2000; Zhu et al. 2002; Palmgren et al. 2003). UFPs are also generated through another process, technology in working place like printer, photocopy machine, etc. (Stephens et al. 2013; Viitanen et al. 2017).

2 Ultraﬁne Particles in Concern of Vehicular Exhaust …

2.3.2.2

17

Characterization of UFPs

Physical Characterization University of Minnesota in the late 1990s made an important and new development in the ﬁeld of particulate matter size range of diesel engine emission (Kittelson et al. 2006). According to that, UFPs formation falls under the following mode: Nucleation mode, Accumulation mode, and Coagulation mode. Nucleation mode Newly formed particle (>10 nm aerodynamic diameter) is present in the transient nuclei mode. This mode contains involatile materials from the new particles through a condensation process by homogeneous nucleation. There should be a substantial supersaturation of the vapor which is necessary for homogeneous nucleation. This process can occur both in hot combustion of gases and in metallurgical processes (Harrison et al. 1998). The oxidation of sulfur dioxide to sulfuric acid is well-known example in the atmosphere. It can undergo binary nucleation (with water) and tertiary nucleation (with water and ammonia) (Harrison et al. 1998; Kittelson 2015; Korhonen et al. 1999; Duffy and Nelson 1997). Nucleation mode starts with 1–2 nm but typically covers 20–30 nm in aerodynamic diameter. Further condensation of low volatile material coagulates and enters into the accumulation mode (100 nm to 2 µm) (Harrison et al. 1998; Kulmala et al. 2012) (Fig. 2.2). UFPs with *10 nm is known as narrow nucleation mode, which is developed by condensation of gaseous precursors. UFPs (large nucleation mode with *20– 30 nm) mostly contain sulfate particle and semi-organic compounds. The accumulation mode at around 60 nm is the outcomes of the combustion procedures which mostly incorporates soot and nonvolatile organic compounds but additionally sulfate and semi-nonvolatile organic compounds also. This mode is generally associated with diesel exhaust (Hämeri et al. 2004; Ahmed 2017) (see Footnote 5). Accumulation mode After transient nuclei mode, which contains growth by both condensation of low volatility materials and through coagulation, UFP enters into an accumulation mode with range of *100 nm to 2 µm. UFP in accumulation mode is subject to inefﬁcient loss from the atmosphere by wet and dry deposition processes. Due to the low number concentration, UFPs do not signiﬁcantly further grow through coagulation, and they can travel several days over long distances in the atmosphere due to their long life (Harrison et al. 1998; Stanier 2003; Kulmala et al. 2012; Hämeri et al. 2004; Rissler et al. 2012; Morawska et al. 2008).

18

S. K. Yadav et al.

Fig. 2.2 Particle formation in diesel combustion in different modes (Kittelson et al. 2016)

Coagulation mode Coarse mode particles are signiﬁcantly in mass concentration after crossing 2.5 µm aerodynamic diameter and do not have very long lives in the atmosphere. They are completely different from UFP particles in chemical composition and formation process. Particles from engine vehicle emissions can be separated into two general classes, contingent upon the area of their formation. The primary particles of combustion, which are generated in tailpipe or engine in the size range 30–500 nm, are generally sub-micrometer agglomerates of strong stage carbonaceous materials. It may also be made of metallic ash remains (from greasing up oil-added substances and from engine wear), adsorbed or consolidated hydrocarbons, and sulfur compounds. The hot, exhausted gases, from tail or near tailpipe, get quickly cool and nucleated or coagulated. The coagulation on the existing particles to form a larger number of very small particles in the ambient air is generally 30 nm or lesser than in diameter which are generally composed of hydrocarbons and hydrated sulfuric acid. Especially those where a large fraction of heavy-duty diesel vehicles are present. These UFPs are formed very quickly and are distinct from UFPs generated from photochemical nucleation processes occurring in the atmosphere further away from the source (Westerdahl et al. 2005; Thompson et al. 2004; Jayaratne et al. 2008; Moore et al. 2007; Karjalainen et al. 2010; Nøjgaard et al. 2012; Stanier and Pandis 2004; Stanier et al. 2004).

2 Ultraﬁne Particles in Concern of Vehicular Exhaust …

19

Chemical Characterization There is a lot of variation in UFPs composition because the relationship of elemental composition of it with its emission is an important way to source diagnostic. The presence of Si, Al, and Ti in low concentration in UFPs is an indicator of crustal minerals. There is a direct connection with the presence of NO, NO2, CO, Cu, and Zn in UFPs with engine vehicle movements. Elemental carbon may constitute a major fraction of UFPs, which is produced as aggregates in a major function of diesel motor combustion (Harrison et al. 1998; Cyrys et al. 2003; Kane and Johnston 2000; Meng et al. 2013). Formation variability of UFPs makes insufﬁcient known about chemical characterization of it in the urban and rural area. With the help of advance cascade impactor, the chemical composition can be done by collecting samples. Glen et al. (2000) have performed the studies in southern Californian, cities for UFPs chemical characteristics with particle range 0.056–0.1 µm diameter (average dp = 0.55– 1.16 µm/m3). He reported the concentration to be 50, 14, 8.7, 8.2, 6.8, and 3.7%, followed by organic compound, trace metal oxides, elemental carbon, sulfate nitrate, and ammonium ion. There was sodium 0.6 and 0.5% chloride as well. UFPs may contain Fe, Ti, Cr, Zn, and Cs in addition catalytic metals (Harrison et al. 1998). According to Seinfeld et al. (1998), the formation and growth of UFPs play a major role in reducing the trace gases (VOC, H2SO4,) from the atmosphere, as UFPs are constituent with the atmospheric aerosols (Report UP 2011; Seinfeld and Pandis 2006). New particle formation made of UFP from diesel aerosol having a principal component of black carbon (carbonaceous soot particles) with a blend of organic compounds (OC). Some of these polycyclic aromatic hydrocarbons (PAH) arise during the combustion process on the surface of particles (Rissler et al. 2012). It is interesting to note that as particle diameters become larger, particle number concentration decreases (Kulmala et al. 2012).6 Due to trace quantity of the UFP in mass concentration, its comprehensive knowledge of chemical composition in ambient air is still challenging for the research society. Moreover, there also exists no universal measurement protocols. However, it is well deﬁned that now trafﬁc site engine combustion includes SO2 and NOx. The nucleation of these gases into SO4 2 , and NO3 , is a signiﬁcant process for increasing particle formation near trafﬁc site (Harrison et al. 1998; Cyrys et al. 2003) (see Footnote 5). Throughout the most recent decade, generous endeavors have been improved to describe the physical and chemical properties of UFPs and their potential effect on individuals living in proximity to roadways and different outflow sources. UFP emission from engine vehicles is not static in the wake of leaving the tailpipe and experience physical change and substance responses in the air as they are

6

Overview E, Health ON, Of I (1999) 5. health impacts of ultraﬁne particles 5.1 g. 42–187.

20

S. K. Yadav et al.

transported far from the source. In a few cases, emission controls intended for PM mass have encouraged the development of a more noteworthy number of UFPs. Be that as it may, appropriately planned discharge control advancements can conﬁne the arrangement and outflow of the UFP and PM mass.

2.4

Transport of UFPs

When UFPs are released from the source into the ambient air, it is followed by dilution with ambient air or with surrounding environment. Dilution starts several chemicals and physical processes which include the secondary particle formation, evaporation, condensation, and coagulation process. These all get affected with distance from sources, and result of change is the chemical composition of UFPs. So there is marked difference between the characteristics of UFPs which is measured away from the emission source (roadways) and measured at the emission point (tailpipe emission). In aerosol science, wind speed and its direction, precipitation, humidity, and temperature are major factors affecting the transportation and transformation of particles, including UFPs. The concentration of UFPs is present in the strong diurnal variation and other seasonal variations which has been observed in various studies and have closely revealed to the amounts of trafﬁc volume. UFP’s concentration was mostly high on week working days (e.g., two peaks—one early morning and another in afternoon with trafﬁc rush hour). But there was a wider midday peak on weekends and in the afternoon of summer. It may increase the concentration of UFPs by the process of photochemical particle formation but due to unfavorable high temperature in summer condition that does not favor the nucleation process. There are several meteorological factors which affect the seasonal variability of particle number concentration (Lushnikov et al. 2010; Kumar 2013). The height of mixing layer and stability of the atmosphere has a great influence on the transportation of UFPs. For instance, low mixing height and strong stability may increase the particle number concentration due to the restriction of their vertical mixing. Lower temperature (generally rush hours of winter’s working days) leads to nucleation of volatile combustion products. The particular areas which have high meteorological differences between seasons have more pronounced particle number concentration and its transformation. According to Pirjola et al. (2006) Finland was having 2–3 times more number concentration in the winter months compared to summer with the highest observed value in February. In the same way, the number of concentration was observed in Pittsburgh (USA) in December (highest PNC) and July (lowest PNC), which was reported by Zhang et al. (2004) (Nøjgaard et al. 2012; Hussein et al. 2004; Pirjola et al. 2006). Furthermore, albeit the expanded photochemical movement can prompt new UFP arrangement in ambient air. UFPs emission from the vehicle’s tailpipe is not static; it undergoes a chemical reaction and physical transformation on or near roadways and in-vehicle concentration in the atmosphere as they are transported

2 Ultraﬁne Particles in Concern of Vehicular Exhaust …

21

Fig. 2.3 Concentration of UFPs at different locations (Morawska et al. 2008)

away from the source by a different mode. In Fig. 2.3, the UFPs concentration on different locations is shown (Morawska et al. 2008; Jayaratne et al. 2008; Sem et al. 1980; Hinds et al. 1982).

2.4.1

Measurement Methods

According to Harrison et al. (1998), atmospheric particulate is inherently variable in size and their chemical characteristics, which is difﬁcult to study than the gas-phase component of the atmosphere. Simple spectroscopic techniques are not effectively useful for quantitative and qualitative analyses in case of determination of aerosol composition in atmosphere (Harrison et al. 1998). Due to the size of UFPs, its measurement is highly challenging and results are dependent on the principle of measurement technique and method. There is a lack of “standard” for measurement technique or calibration standard by which the measurement of instrument can be evaluated and compared thoroughly (see Footnote 5). But many research groups have deﬁned the UFPs measurement protocols by using different measurement techniques after getting an UFP’s link with the penetration of the pulmonary alveolar, which causes breathing problems and induces the oxidative stress (Künzli 2015). Oxidative stress can cause cardiovascular and neurological problems (Wu et al. 2013; Lee et al. 2007; Sacks 2015). According to Nassbaumer et al. (2008) and Obaidullah et al. (2012), when particulate matter’s size range is a few micrometers to nanometer, then the choice of measurement and its technique become more difﬁcult and challenging (Nussbaumer et al. 2008; Obaidullah et al. 2012). Generally, particulate matter is measured in terms of mass concentration, number concentration of particles, and particle size distribution. But in case of UFPs, the majority of researchers do not give priority to

22

S. K. Yadav et al.

measure the mass concentration because of its low contribution to mass ratio, which is not so easy to maintain or analyze such as PM10 or PM2.5. Measurement of UFPs may be classiﬁed as “concentration” and “size distribution methods.” In the concentration method, UFP measurement can be mass concentration (for chemical characterization), particle number concentration (#/cm3), and surface area. All these may be based on different measurement principles such as gravimetric measurement, optical measurement, or microbalance measurement (Amaral et al. 2015).

2.4.1.1

Concentration of UFPS

Gravimetric method Gravimatric method is generally determined by weighing the ﬁlter before and after the sampling. Giechaskiel et al. (2014) explain that the ﬁlter paper collects the particulate matter in all granulometric fractions (nucleation, accumulation, and coarse) with a cyclone or impactor to remove the larger particles (Giechaskiel et al. 2014). The impactor is being used for size distribution also with the working principle of gravimetry with the arrangement of multiple impactor stages. In some cascade impactors, it may contain multiple oriﬁces (Amaral et al. 2015). Hinds describe that a cascade-type impactor (mostly in use) is operated based on the inertial classiﬁcation of particles (Hinds 1999). According to Vincent (2007), the aerosol sample passes through a sequence of stages. In each stage, an air jet containing the aerosol which reaches the impacting plate and particles larger than the cutoff diameter of the stage is collected. Smaller particles allow the gas flow that surrounds stage, in which the oriﬁces are smaller and have conditions for greater airspeed. This process continues until smaller particles are removed after ﬁltration (Vincent 2007). Nussbaumer et al. (2008) mentioned low-pressure cascade impactor (LPI), which is a frequently used impactor with the ability to collect particles in the range of 30 nm to 10 µm. It can be extended with appropriate ﬁlter to range of smaller particles (Vicente and Alves 2018). These charged particles pass through a low-pressure cascade impactor, and these low-pressure cascade impactors are composed of collection steps which are electrically isolated. When sampled particles impact on one speciﬁc stage, it produces an electric current which is registered in real time through an electrometer. Earlier Dekati mass monitor (DMM) was used for measuring vehicle gases, and DMM was estimating the particle density by the combination of aerodynamic size and mobility data. But later it got an adaptation to add one electrical mobility stage which is known as ELPI. Electrical low-pressure impactor (ELPI) is another type of advance impactor where principle is based on charging the aerosol electrically. ELPI can be used for measuring the concentration by particle distribution in real-time particle number (7–10 nm) and it has classiﬁed

2 Ultraﬁne Particles in Concern of Vehicular Exhaust …

23

the ultraﬁne particles according to its aerodynamic diameter. According to Vincent, as they aspirated, particles get electrically charged by the unipolar corona charger. These charged particles are collected at low-pressure cascade impactor which is composed of collection steps (electrically isolated) (Amaral et al. 2015; Venkataraman and Rao 2001). Micro-oriﬁce uniform deposit impactor (MOUDI) is another family of cascade which is the substitute of conventional cascade impactor. A conventional impactor is not able to collect and select smaller 0.4 µm particles at atmospheric pressure. With the combination of impacting stage and with a range of flow rate from 10 to 100 L/min, MOUDI covers a broad range of particle size (>37– 56 nm diameter on quartz ﬁber ﬁlter). Nano-MOUDI has limit *10–50 nm. It has integrated mass-based size distribution (size limit *56–10,000 nm) (Amaral et al. 2015; Vincent 2007; Venkataraman and Rao 2001). Optical method Optical detection methods are mostly used for the measurement of real-time particle concentration by scattering, absorption, and extinction of light beam. Particle number concentration of UFPs is mostly measured by a condensation particle counter (CPC). A condensation particle counter is a light-scattering counter in which there is a speciﬁc arrangement for UFPs, because they do not scatter the light sufﬁciently to get detected by conventional optical counters. In CPC, UFP passes through a condensation chamber before going to detection chamber. Condensation chamber is saturated with water or alcohol vapor (produced vapor) from the working fluid. When a particle enters into this chamber, its get coated with vapor. In the next subsection of chamber, temperature decreases enough to condense the vapor on a particle. Now the particle is enlarged enough to scatter the light which can be detected by the detector focal point of the laser beam with a flash of light. Through CPC, real-time number concentration can be produced with size limit *30–1000 nm. CPC can be used with other size distribution instrument for better measurement of UFPs like scanning mobility analyser (Amaral et al. 2015). Microbalance Methods Microbalance works with the principle of using the alternate frequency (for that particular particle which is collected over the surface of an oscillatory microbalance element) to determine the particle concentration. There are two types: tapered element oscillation microbalance (TEOM) which is working on the principle of tapered quartz wand’s altered resonance frequency by particle accumulation on sampling ﬁlter and quartz crystal microbalance (QCM) with piezoelectric property which changes the resonance frequency by small addition of particle mass (Amaral et al. 2015).

24

2.4.2

S. K. Yadav et al.

Size Distribution Measurement Methods

Size distribution methods quantify two parameters, namely, aerosol size (aerosol diameter) and aerosol concentration. On the basis of studies conducted by Giechaskiel et al. (2014), the aerosol particle size can be measured based on its properties like inertia, mobility, geometric size, electrical mobility, and optical properties. The measurement of particle size is carried out by an amalgamation of instruments involving corona charger, particle size classiﬁcations are done by impactors or mobility classiﬁers, and detection is done by optical counters or electrometers (Amaral et al. 2015; Giechaskiel et al. 2014). Microscopy As stated by Vincent, particle size studies using a microscopy involve the direct collection from ﬁlters, followed by the preparation of ﬁlter in order to improve visibility. Wentzel et al. (2003) underscored the plethora of information made available through image analysis such as rotation radius, fractal dimensions, number of primary particles per aggregate, size distribution of aggregates, and size distribution of primary particles. Giechaskiel delineated the examination of particle size, made morphology available through microscopy, while also proceeding to highlight this method’s major drawback, which is the time taken to analyze the number of particles which is statistically sufﬁcient (Künzli 2015; Lee et al. 2007; Nussbaumer et al. 2008). Diffusion Battery (EDB) Vincent hypothesized that diffusion determines particle movement. Hence, the equivalent diameter obtained in a diffusion battery is more appropriate in the case of nanometric particles. According to Fierz et al. (2011), recent gravimetric and optical methods are deemed inappropriate due to their low levels of sensitivity. Hinds (Hinds 1999) stated that the development of diffusion batteries can be attributed to the need for the determination of diffusion coefﬁcient of the particles (Obaidullah et al. 2012). Giechaskiel described that the particles are separated in the diffusion battery by their mobility. This led to the introduction of “Electrical Diffusion Battery (EDB).” Here, the particles are carried by a corona charger and subsequently entered the diffusion battery (can be either tube or screen). The EDB collection efﬁciency is a function of geometric properties of the tube or screen, the flow rate, and particle size, as theorized by Vincent et al. (2014) (Giechaskiel et al. 2014; Vincent 2007). Mobility Analyzer According to Giechaskiel et al. (2014), DMAs (the most recent model of Mobility Analyzer) make use of bipolar diffusion charging to provide a well-deﬁned distribution of charge in the aerosol. Loading is followed by the entry of particles into the classiﬁer permitting passage in a narrow range in electrical mobility, which are measured by an electrometer or CPC. Maruf Hossain et al. (2012) stated that the

2 Ultraﬁne Particles in Concern of Vehicular Exhaust …

25

volatility tandem differential mobility analyzer (VTDMA) consists of two nano-DMA, two long DMAs, an ultraﬁne condensation particle counter (UCPC), and a heating tube (Giechaskiel et al. 2014; Wentzel et al. 2003; Fierz et al. 2011; Maruf Hossain et al. 2012). Centrifugal Measurement of Particle Mass It is carried out via centrifugal particle mass analyzer (CPMA) or aerosol particle mass (APM). Giechaskiel et al. (2014) state that the CPMA consists of an external and an internal coaxial cylindrical electrode, wherein the latter turns faster, while passing in which the loaded particles come under the influence of an electrostatic and centrifugal ﬁeld experienced in opposite directions, whereas in the APM these electrodes spin at the same speed. The penetration of particles in the CPMA is proportional to the tension and rotation speed between the electrodes. Johnson et al. (2013) asserted that the CPMA classiﬁes an aerosol by the mass-to-charge ratio. The advantage of these instruments according to Giechaskiel et al. (2014) is their ability to register particles mass without weighing the particles (Giechaskiel et al. 2014; Johnson et al. 2013). Differential Mobility Spectrometers (DMS) DMS comprises of a particle loader, a classiﬁcation column, and the series of detectors as elaborated by Giechaskiel et al. (2014). He laid emphasis on the fact that SMPS is the most precise instrument for high-resolution size distributions of aerosols from vehicle exhaust. Nussbaumer et al. (2008) discussed the use of SMPS in the measurement of the number and size distribution of particles produced via combustion of biomass (Nussbaumer et al. 2008; Giechaskiel et al. 2014). According to Hosseini et al. (2010), FMPS consists of two concentric cylinders, a diffusion loader and 32 electrometers, that cover particle sizes from 5 to 560 nm. The sheath flow consists of a current of positively charged particles. Due to the large potential difference between the cylinders, particles are transported from one side to the other side. The particles with higher electrical mobility are collected next to the column top and vice versa. The difference between the two spectrometers is that instead of a CPC the FMPS uses multiple low noise electrometers for particle detection, giving real-time analysis (Hosseini et al. 2010). Fast Integrated Mobility Spectrometer (FIMS) Giechaskiel et al. (2014) describe the FIMS composition as a charger, a size classiﬁer, a condenser, and a detector. Kulkarni et al. (2011) and Olfert et al. (2008) state that the aerosol passes through a neutralizer thereby receiving a charge distribution of bipolar equilibrium. The aerosol passes through a mobility analyzer through which a butanol saturated gas flows. Based on their electrical mobility, particles are separated in different paths in the ﬁeld of the mobility analyzer. These classiﬁed particles then enter the ﬁeld-free condenser, where supersaturated butanol vapor increases their size by condensing over them. The drops are illuminated by the laser beam illuminated at the exit, and a 10 Hz camera captures these images, which in turn provides particle concentration and particle mobility diameter. SMPS,

26

S. K. Yadav et al.

FMPS, ELPI, MOUDI, and OPC are important UFP measurement instrument according to their application (Amaral et al. 2015; Giechaskiel et al. 2014; Kulkarni et al. 2011; Olfert et al. 2008).

2.5

Health Impact of UFPs

Meng et al. (2013) studied the small particles (100 nm), the particles take an entry in the cell primarily (via clathrin and also caveolin-mediated endocytosis), and these particles are stored or found in the endosomes or hydrosomes in the cytoplasm (Chen et al. 2016).

2 Ultraﬁne Particles in Concern of Vehicular Exhaust …

27

After passing the blood–gas barrier, UFPs could enter the circulation and contact “extra-pulmonary tissue cell.” But the larger particles (>100 nm) may be degraded with minimized toxic consequences in the circulation. The UFPs > 100 nm particles may get deposited in the cell membrane or in the cell. This presence or deposition of UFPs may generate reactive oxygen species (ROS), which increased the transcription of pro-inflammatory mediator via generating oxidative stress. Thus, UFP circulation makes it more toxic and inflammatory as compared to ﬁne particles (Pirjola et al. 2006; Chen et al. 2016). Less information and lack of standard of UFPs either in the ﬁeld of environmental science or human health has not been able to attract more attention from governing authorities in the world (in developing and undeveloped countries). But as per current technology for measurement of nanosize particles, it is necessary to establish a pragmatic strategy to estimate the health impact of UFPs on human health with the integration of other ﬁelds (like nanotechnology, toxicology, epidemiology, and clinical medicines). There are three primary ways of UFP’s exposure (Cassee et al. 2002): Respiratory tract, Dermal exposure, and Ocular exposure. Respiratory Tract For the human body, inhalation is the main way to get UFPs exposure through respiration system. There is a hypothesis on UFPs having an efﬁciency to pulmonary deposition and it could also reach deeper inside of the lungs because UFPs cannot be stopped or ﬁltered by the nose and bronchioles. The human nose has a limitation for ﬁltering UFPs. It can ﬁlter more than 80% of particles larger than 1 µm, but only less than 5% of 100 nm particles during the resting period of breathing (Anjilvel and Asgharian 1995; Möller et al. 2008). Lungs are one of the most important parts of the respiratory system, and thus have a direct contact with the outer atmosphere, which has two parts airways and alveolar structure. Alveolar structure contains larger epithelial surface area. International commission and radiological protection (1994) had established a mathematical model for the fractional deposition of the particle (inhaled), in different parts of the human respiratory system through nose breathing during rest time or activity. All possible routes of the UFPs to enter the body have been shown in Fig. 2.4. There is quite an effective deposition of certain UFPs in the respiratory system (nasopharyngeal, tracheobronchial, and alveolar region). Alveolar region is having a high deposition efﬁciency of 20 nm with *50% and for particles range >100 nm to 2.5 µm) having 10–20% deposition efﬁciency. Due to high deposition of UFP, alveolar is facing more toxic effects to the lung by penetration of alveolar epithelium and UFP’s circulation into the bloodstream. With the help of multiple path particle dosimeter (MPPD), the exact amount of dosed of UFPs can be estimated which is deposited in the lung. MPPD is originally developed for dose estimation in the lower respiratory tract (Schmid et al. 2009; Yegambaram et al. 2015).

28

S. K. Yadav et al.

Fig. 2.4 The possible penetration routes of airborne UFPs (Chen et al. 2016)

According to Möller, in the lung periphery and airway, there is a less clearance of the UFP (inhaled > 100 nm) after respiration at a point of 24 h. Imbalance in mass balance of UFPs (high deposition and low clearance) may promote the accumulation in alveolar. Besides alveolar deposition, the UFPs deposition on the olfactory bulb surface is another principal issue for the health risk evaluation for uptake of UFPs through the respiratory system. This route of exposure (long-term exposure) of UFPs may produce the neurodegradation diseases (like Alzheimer’s disease, Parkinson’s disease) (Yegambaram et al. 2015; Liu et al. 2015; Knudsen et al. 2013; Bennat and Müller-Goymann 2000). Now, toxicological studies have clearly explained the deposition in nasal cavity, being absorbed by olfactory bulb and translocated into brain (via blood–brain barrier) (Boyes et al. 2012). Dermal Exposure There is no direct link or in vivo evidence supporting of penetration of UFPs through skin but we cannot fully reject this possibility. Because human skin has larger surface area and works as a direct barrier to outer environment, some research teams suggested that it may occur through two routes: intercellular trans-epidermal cell or diffusion through skin pores, penetrated or entered the human body (Torricelli et al. 2011; Versura et al. 1999). Ocular Exposure Ocular exposure is another way of UFPs exposure to human. It occurs in eyes by floating particles or by transferring during rubbing of eyes in high concentration area (high polluted area). UFPs could be retained in nasal cavity by the drainage from eye socket. Then, through nose to brain transport system, UFPs can reach to brain. Further, there are more possibilities of particle to enter the blood circulation

2 Ultraﬁne Particles in Concern of Vehicular Exhaust …

29

Fig. 2.5 The potential toxicity mechanisms of UFPs (Chen et al. 2016)

and travel throughout the body. It may cross the biological barrier after entering into blood circulation. Through this way, exposure may cause ocular diseases such as discomfort eye syndrome (DES) by internal associations. This was disclosed by linking to taxi drivers in potential exposure (Stanier et al. 2004; Torricelli et al. 2011; Versura et al. 1999; Novaes et al. 2010; Lee and Chang 2000; Knibbs et al. 2011). Figure 2.4 represents all expected route of UFPs to enter human body and its expected target tissues and organs (Fig. 2.5).

30

2.6 2.6.1

S. K. Yadav et al.

Policy for Fuel Grades for Emission Control Fuel Standards in Emission

The ﬁrst BS standards for regulation of air pollutants formulated by the central government of India in 2000 were inspired by CARB regulation norms of 1985. The BS standards were set up to curb the release of air pollutants from internal combustion engines of motor vehicles (Andersson et al. 2015; Act CA 2002; India Today 2016). European emission standards deﬁne the acceptable limits for exhaust emissions of new vehicles sold in EU and EEA member states. These emission standards are recorded in a European Union Directives with increased stringent standards over a progressive model (Bhatia 2017). In the European Union emissions of nitrogen oxides (NOx), total hydrocarbon (THC), non-methane hydrocarbons (NMHC), carbon monoxide (CO), and particulate matter (PM) are regulated for most vehicle types, including cars, trucks (lorries), locomotives, tractors, and similar machinery, barges, but excluding seagoing ships and aeroplanes. For each vehicle type, different standards apply. Compliance is determined by running the engine at a standardized test cycle. These standards are applicable only to new vehicles or vehicles not sold yet. Each vehicle must go through a series standardized test cycles before going on road. The standards do not put limitations on technology to be used for maintaining emission rates, but the existing technology is taken into consideration while formulating the norms. The standards provide for minor lifecycle model revisions for pre-compliant engines, but all new models must meet the planned standards before going on market (India Today 2016). Indian emission standards are modeled after euro emission standards, and are used to limit the emission of harmful exhaust gases. The new fuel, BS VI, will be introduced by April 2019.

2.6.2

Bharat Stage VI

The BS VI fuel will have approximately ﬁve times less sulfur content than BS IV fuel. The new fuel will have only 10 ppm sulfur content as compared to the 50 ppm content of BS IV. With lower sulfur content, petrol will emit less NOx, CO, and HC, while the advantage of low sulfur diesel is signiﬁcant reduction in particulate emission. Estimates suggest that a BS 4 compliant car running on BS 6 diesel could emit 50% less PM. Table 2.3 shows the PM concentration and particle number associated with fuel grades of different vehicle types.

2 Ultraﬁne Particles in Concern of Vehicular Exhaust …

31

Table 2.3 PM concentration and particle number associated with fuel grades of different vehicle types CI engine

LDVa

HDVa

Two-wheelerc

Three-wheelerc

PM 0.14–0.25 0.36 – – PN – – – BS 2 PM 0.08–0.17 0.15 0.10 0.10 PN – – – – BS 3 PM 0.05–0.10 0.10–0.221 0.05 0.05 PN – – – – BS 4 PM 0.025–0.06 0.02–0.03 – 0.0425 PN – – – – BS 5 PM – – – – PN – – – – BS 6 PM 0.0045 0.01 0.0045 0.025 6.67 1011 – – PN 6 1011 a India: Light-Duty: Emissions, Retrieved from: https://www.transportpolicy.net/standard/indialight-duty-emissions/?title=india:_light-duty:_emissions b India: Heavy-Duty: Emissions, Retrieved from: https://www.transportpolicy.net/standard/indiaheavy-duty-emissions/?title=india:_heavy-duty:_emissions c India: Motorcycles: Emissions, Retrieved from: https://www.transportpolicy.net/standard/indiamotorcycles-emissions/?title=india:_motorcycles:_emissions BS 1

2.6.3

Key Aspects of Bharat Stage VI-Vehicles with Gross Vehicle Weight (GVW) Less Than 3500 kg and Category M (Passenger Vehicle) and N (Goods Vehicle)—Light-Duty Vehicles

These vehicles are divided into two types, based on engine type.

2.6.3.1

Compressed Ignition Engine (CI)

There is 68% reduction seen in concentration level of NOx when using BS VI instead of BS IV. PM reduction level varies between 82 and 93%. Apart from tightening rules of PM, particulate numbers (PN) will also be used in this, whose emission is limited to 6 1011/km. With the stricter rule for PM and the new introduction of PN, the manufacturing companies will be forced to use diesel particulate ﬁlters (DPFs), which is the best available technology available for reduction of PM in diesel engine (International Council on Clean Transportation (ICCT) 2016).

32

2.6.3.2

S. K. Yadav et al.

Spark Ignition Engine (SI)

Limits for CO and HC remain unchanged from BS IV to BS VI, but there is 25% reduction in NOx concentration. The PN is a little relaxed in this category with upper limit of 6 1012/km. They will ensure that manufactures use gasoline particulate ﬁlters (GPFs) to control PM emission.

2.6.4

Vehicles with GVW More Than 3500 kg and Category M (Passenger Vehicle) and N (Goods Vehicle)— Heavy-Duty Vehicles

The PM emission is reduced by around 50–67% when transitioning from BS IV to BS VI. The PN is limited to 8 1011/kWh for steady-state cycle and 6 1011/kWh for transient cycle testing. Similarly like LDV, DSPs are expected to reduce the PM emission level (ICCT 2015).

2.6.4.1

Two-Wheeler

Motorcycle, scooty (Non-gear scooter), and other two-wheelers contribute for the majority trafﬁc on Indian roads, both in terms of current vehicle population and in terms of fraction of new vehicle sales (ICCT 2015). NOx emissions are reduced by around 70–85%. Rules with no flexibility in concern with emission limit will prove to be a great step; these step will help reduce hydrocarbons emissions greatly.

2.6.4.2

Three-Wheeler

Auto-rickshaws are the part of three-wheeler community in India. They use gasoline, CNG, and diesel as well. BS VI proposal limits the PM by doing a 44% reduction in mass, and in this regulation they are included to control the crankcase emission. The graph represents the reduction in PM emission when fuel is switched from BS IV to BS VI, and the data is segregated by vehicle types, as explained above (Fig. 2.6).

2 Ultraﬁne Particles in Concern of Vehicular Exhaust …

33

Fig. 2.6 BS IV and VI emission limits for compression ignition vehicles (International Council on Clean Transportation (ICCT) 2016)

2.7

Necessities for Future Research

There is a need of pragmatic approach to explore the physiochemical characteristic of UFPs and develop the UFPs exposure model and national standard of ambient air along with indoor air. The signiﬁcance of in-transit UFP exposure is highly dependent on personal (adult, child, nonsmoking ofﬁce worker than a smoker or someone who experiences a high occupational exposure), demographic and occupational context. More research is needed to better characterize the mechanisms that lead to UFP formation and fate of transport right after emission and in the atmosphere and standardized measurement methods and procedures (Allen et al. 1999). The terms UFPs and NPs are not clearly deﬁned and often used improperly; in future, there should be measurement to decide the “Hot Spot” (busy roads, intersections, railyards, and airport) for UFPs and new particle formation. Developed as well as developing countries do not have enough emission inventories of UFPs and NPs from vehicular emission, so that there are insufﬁcient air pollution models for it. There is quite enough gap about concentration response factor (CRF) or exposure–response relationship for health risk assessment of UFPs (Pirjola et al. 2006; Wang 2009). Acknowledgements We would like to express our appreciation to Abhishek Rai and Shivang Agarwal from the department of Environmental Engineering, Delhi Technological University for sharing their pearl wisdom with us during the course of literature survey and writing of the chapter. We want to extend our thanks to Mr. Veer Kumar, General Manager, Alfa Tech Services for providing insights on measurements of ultraﬁne particles by use of different instruments.

34

S. K. Yadav et al.

References Act CA (2002) Clean Air Act 2.1 Ahmed S (2017) PDX scholar modeling of ultraﬁne particle emissions and ambient levels for the near roadside environment Air Quality Science (2011) Ultraﬁne particles: why all the concern about something so small? Allen AG, Grenfell JL, Harrison RM et al (1999) Nanoparticle formation in marine airmasses: Contrasting behaviour of the open ocean and coastal environments. Atmos Res 51:1–14. https://doi.org/10.1016/S0169-8095(98)00111-2 Amaral SS, de Carvalho JA, Costa MAM, Pinheiro C (2015) An overview of particulate matter measurement instruments. Atmosphere (Basel) 6:1327–1345. https://doi.org/10.3390/ atmos6091327 Andersson J, Martini G, Mayer A (2015) History and future of particle number legislation in Europe. In: 19th ETH conference on combustion generated nanoparticles, pp 1–24 Anjilvel S, Asgharian B (1995) A multiple-path model of particle deposition in the rat lung. Fundam Appl Toxicol 28:41–50. https://doi.org/10.1006/FAAT.1995.1144 Baldauf RW, Devlin RB, Gehr P et al (2016) Ultraﬁne particle metrics and research considerations : review of the 2015 UFP workshop, pp 1–21. https://doi.org/10.3390/ijerph13111054 Bennat C, Müller-Goymann CC (2000) Skin penetration and stabilization of formulations containing microﬁne titanium dioxide as physical UV ﬁlter. Int J Cosmet Sci 22:271–283. https://doi.org/10.1046/j.1467-2494.2000.00009.x Bhatia N (2017) How the upcoming BS-VI grade fuel could affect you. Retrieved from https://www. autocarindia.com/auto-features/how-the-upcoming-bs-vi-grade-fuel-could-affect-you-406616 Boyes WK, Chen R, Chen C, Yokel RA (2012) The neurotoxic potential of engineered nanomaterials. Neurotoxicology 33:902–910. https://doi.org/10.1016/j.neuro.2011.12.013 Brimblecombe P, Sturges K, Maxwell R (2009) History of atmospheric environment. Atmos Environ 43:2–8. https://doi.org/10.1016/j.atmosenv.2008.09.065 Cassee F, Muijser H, Duistermaat E et al (2002) Particle size-dependent total mass deposition in lungs determines inhalation toxicity of cadmium chloride aerosols in rats. Application of a multiple path dosimetry model. Arch Toxicol 76:277–286. https://doi.org/10.1007/s00204002-0344-8 Chen R, Hu B, Liu Y, et al (2016) Beyond PM2.5: the role of ultraﬁne particles on adverse health effects of air pollution. Biochim Biophys Acta Gen Subj. https://doi.org/10.1016/j.bbagen. 2016.03.019 Coburn TC (2000) Statistical analysis of on-road particulate matter emissions from diesel vehicles. Inhal Toxicol 12:23–33 Cyrys J, Stölzel M, Heinrich J et al (2003) Elemental composition and sources of ﬁne and ultraﬁne ambient particles in Erfurt, Germany. Sci Total Environ 305:143–156. https://doi.org/10.1016/ S0048-9697(02)00494-1 Damanik N, Ong HC, Tong CW, Mahlia TM, Silitonga AS (2018) A review on the engine performance and exhaust emission characteristics of diesel engines fueled with biodiesel blends. Environ Sci Pollut Res. https://doi.org/10.1007/s11356-018-2098-8 Dawson KA, Salvati A, Lynch I (2009) Nanoparticles reconstruct lipids. Nat Nanotechnol 4:84– 85. https://doi.org/10.1038/nnano.2008.426 Dey S, Di L, Van Donkelaar A et al (2012) Remote sensing of environment variability of outdoor ﬁne particulate (PM2.5) concentration in the Indian Subcontinent: a remote sensing approach. Remote Sens Environ 127:153–161. https://doi.org/10.1016/j.rse.2012.08.021 Duffy BL, Nelson PF (1997) Exposure to emissions of 1,3-butadiene and benzene in the cabins of moving motor vehicles and buses in Sydney, Australia. Atmos Environ 31:3877–3885 Eastwood P (2008) Particulate emissions from vehicles. Wiley, New York ECMA. Emission norms for two and three wheelers. Retrieved from http://www.ecmaindia.in/ iestandards.aspx?mpgid=24&pgid1=25&pgidtrail=58#faq5

2 Ultraﬁne Particles in Concern of Vehicular Exhaust …

35

Fierz M, Houle C, Steigmeier P, Burtscher H (2011) Design, calibration, and ﬁeld performance of a miniature diffusion size classiﬁer. Aerosol Sci Technol 45:1–10. https://doi.org/10.1080/ 02786826.2010.516283 Fischer HJ, Zhang Q, Zhu Y, Weiss RE (2017) Functional time series models for ultraﬁne particle distributions. Ann Appl Stat 11:297–319. https://doi.org/10.1214/16-AOAS1004 Gamma Technologies (2018) Carbon canister and evaporative emission control system modeling. https://www.gtisoft.com/gamma_news/carbon-canister-and-evaporative-emission-controlsystem-modeling/. Accessed 8 Sep 2018 Giechaskiel B, Maricq M, Ntziachristos L et al (2014) Review of motor vehicle particulate emissions sampling and measurement: from smoke and ﬁlter mass to particle number. J Aerosol Sci 67:48–86. https://doi.org/10.1016/J.JAEROSCI.2013.09.003 Glen R. Cass, Lara A. Hughes, Prakash Bhave, Michael J. Kleeman, Jonathan O. Allen and Lynn G (Oct. 15, 2000) SalmonPhilosophical Transactions: Mathematical, Physical and Engineering Sciences, Ultraﬁne Particles in the Atmosphere 358(1775):2581–2592 Hämeri K, Hussein T, Kulmala M, Aalto P (2004) Measurements of ﬁne and ultraﬁne particles in Helsinki: connection between outdoor and indoor air quality. Boreal Environ Res 9:459–467 Harrison RM, Maynard D, Maynard RL (1998) UTM the Royal 8 3 M Society Hester RE, Harrison RM, Querol X (2016) Front matter. In: Airborne particulate matter sources, atmospheric process and health, pp 1–34 Hinds WC (1999) Aerosol technology : properties, behavior, and measurement of airborne particles. Wiley, London Hinds W, Macher J, First M (1982) Size distributions of test aerosols produced from materials other than DOP. Am Ind Hygiene Assoc J 25:20–28 Honey HF, McQuitty JB (1976), Dust in the animal environment. Department of Agricultural Engineering, University of Alberta Hosseini S, Li Q, Cocker D et al (2010) Particle size distributions from laboratory-scale biomass ﬁres using fast response instruments. Atmos Chem Phys 10:8065–8076. https://doi.org/10. 5194/acp-10-8065-2010 Hussein T, Puustinen A, Aalto PP et al (2004) Urban aerosol number size distributions. Atmos Chem Phys 4:391–411 ICCT (2015) DPFs are estimated to reduce PM mass emission by >90% and PN emissions by >99% relative to an uncontrolled diesel engine. Accelerating progress from Euro 4/IV to Euro 6/VI vehicle emission standards Iijima S (1985) Electron microscopy of small particles. J Electron Microsc (Tokyo) 34:249–265 India Today (2016) All you need to know about India’s BS-IV and BS-VI emission norms. Retrieved from https://www.indiatoday.in/auto/latest-auto-news/story/gadkari-plans-to-skipbsv-and-head-straight-to-bsvi-302434-2016-01-06 International Council on Clean Transportation (ICCT) (2016) India Bharat Stage VI Emission Standards. Retrieved from https://www.theicct.org/sites/default/ﬁles/publications/India%20BS %20VI%20Policy%20Update%20vF.pdf Jayaratne ER, He C, Ristovski ZD et al (2008) A comparative investigation of ultraﬁne particle number and mass emissions from a fleet of on-road diesel and CNG buses. Environ Sci Technol 42:6736–6742. https://doi.org/10.1021/es800394x Jiang X, Miclăuş T, Wang L et al (2015) Fast intracellular dissolution and persistent cellular uptake of silver nanoparticles in CHO-K1 cells: implication for cytotoxicity. Nanotoxicology 9:181–189. https://doi.org/10.3109/17435390.2014.907457 Johnson TJ, Symonds JPR, Olfert JS (2013) Mass-mobility measurements using a centrifugal particle mass analyzer and differential mobility spectrometer. Aerosol Sci Technol 47:1215– 1225. https://doi.org/10.1080/02786826.2013.830692 Junker C, Jennings SG, Kleefeld S et al (2000) PM10 concentration measurements in two urban centres and at two remote sites in Ireland. J Aerosol Sci 31:516–517 Kahl JDW, Martinez DA, Kuhns H et al (1997) Air mass trajectories to Summit, Greenland: a 44-year climatology and some episodic events. J Geophys Res Atmos 102:26861–26875. https://doi.org/10.1029/97jc00296

36

S. K. Yadav et al.

Kane DB, Johnston MV (2000) Size and composition biases on the detection of individual ultraﬁne particles by aerosol mass spectrometry. Environ Sci Technol 34:4887–4893. https:// doi.org/10.1021/es001323y Karjalainen P, Heikkila J, Ronkko T et al (2010) Effect of exhaust flow conditions and external cooling on the performance of the particle oxidation catalyst (POC). https://doi.org/10.4271/ 2010-01-2158 Kathuria V (2002) Vehicular pollution control in Delhi. Transp Res Trans Environ 7:373–387 Kittelson PD (2015) Ultraﬁne particles metrics and indicators. What are some physical metrics ? Kittelson DB, Hotel WG, Angeles L (2006) Ultraﬁne particle formation mechanisms. In: South Coast Air Quality Management District conference on ultraﬁne particles: the science, technology, and policy issues, p 37 Kittelson D, Graskow B, Wei Q et al (2016), pp 151–159 Knibbs LD, Cole-hunter T, Morawska L (2011) A review of commuter exposure to ultra ﬁne particles and its health effects. Atmos Environ 45:2611–2622. https://doi.org/10.1016/j. atmosenv.2011.02.065 Knudsen KB, Northeved H, Ek PK et al (2013) Differential toxicological response to positively and negatively charged nanoparticles in the rat brain. Nanotoxicology 8:1–33. https://doi.org/ 10.3109/17435390.2013.829589 Korhonen P, Kulmala M, Laaksonen A et al (1999) Ternary nucleation of H2SO4, NH3, and H2O in the atmosphere. J Geophys Res 104:26349–26353 Kulkarni P, Baron PA, Paul A, Willeke K (2011) Aerosol measurement : principles, techniques, and applications. Wiley, New York Kulmala M, Petäjä T, Nieminen T et al (2012) Measurement of the nucleation of atmospheric aerosol particles. Nat Protocol 7:1651–1667. https://doi.org/10.1038/nprot.2012.091 Kumar P (2013) Assessment of particle number concentration in different transportation modes along a route in Delhi. Int J Web Eng 3:1072–1077 Künzli PN (2015) Swiss perspective on particulate matter air quality standards Lee SC, Chang M (2000) Hong Kong 41:0–4 Lee J, Son J, Cho Y (2007) The adverse effects of ﬁne particle air pollution on respiratory function in the elderly. Sci Total Environ 385:28–36. https://doi.org/10.1016/j.scitotenv.2007.07.005 Liu F, Huang Y, Zhang F et al (2015) Macrophages treated with particulate matter PM2.5 induce selective neurotoxicity through glutaminase-mediated glutamate generation. J Neurochem 134:315–326. https://doi.org/10.1111/jnc.13135 Lushnikov AA, Zagaynov VA, Lyubovtseva YS (2010) Formation of aerosols in the atmosphere, pp 69–96. https://doi.org/10.1007/978-90-481-3212-6 Maruf Hossain AMM, Park S, Kim J-S, Park K (2012) Volatility and mixing states of ultraﬁne particles from biomass burning. J Hazard Mater 205–206:189–197. https://doi.org/10.1016/J. JHAZMAT.2011.12.061 Meng X, Ma Y, Chen R et al (2013) Size-fractionated particle number concentrations and daily mortality in a Chinese City. Environ Health Perspect 121:1174–1178. https://doi.org/10.1289/ ehp.1206398 Mishra UC, Lalit BY, Varma RK, Sadasivan S (1974) Studies on short lived fallout radioisotopes from Chinese and French nuclear weapon tests since 1964. J. Sci Ind Res 33:216–221 Möller W, Felten K, Sommerer K et al (2008) Deposition, retention, and translocation of ultraﬁne particles from the central airways and lung periphery. Am J Respir Crit Care Med 177:426– 432. https://doi.org/10.1164/rccm.200602-301OC Moore KF, Ning Z, Ntziachristos L et al (2007) Daily variation in the properties of urban ultraﬁne aerosol—part I: physical characterization and volatility. Atmos Environ 41:8633–8646. https:// doi.org/10.1016/j.atmosenv.2007.07.030 Morawska L, Ristovski Z, Jayaratne ER et al (2008) Ambient nano and ultraﬁne particles from motor vehicle emissions: characteristics, ambient processing and implications on human exposure. Atmos Environ 42:8113–8138. https://doi.org/10.1016/j.atmosenv.2008.07.050 Nøjgaard JK, Nguyen QT, Glasius M, Sørensen LL (2012) Nucleation and Aitken mode atmospheric particles in relation to O3 and NOX at semirural background in Denmark. Atmos Environ 49:275–283. https://doi.org/10.1016/j.atmosenv.2011.11.040

2 Ultraﬁne Particles in Concern of Vehicular Exhaust …

37

Novaes P, Saldiva PH, Matsuda M et al (2010) The effects of chronic exposure to trafﬁc derived air pollution on the ocular surface. Environ Res 110:372–374. https://doi.org/10.1016/j.envres. 2010.03.003 Nussbaumer T, Czasch C, Klippel N et al (2008) Particulate emissions from biomass combustion in IEA countries survey on measurements and emission factors Obaidullah M, Bram S, Verma VK, De Ruyck J (2012) A review on particle emissions from small scale biomass combustion. Int J Renew. Energy Res 2 Oberdörster G (2001) Pulmonary effects of inhaled ultraﬁne particles. Int Arch Occup Environ Health 74:1–8 Oberdörster G, Oberdörster E, Oberdörster J (2005) Nanotoxicology: an emerging discipline evolving from studies of ultraﬁne particles. Environ Health Perspect 113:823–839 ocregister.com (2008) Catalytic converter-started ﬁres are common Olfert JS, Kulkarni P, Wang J (2008) Measuring aerosol size distributions with the fast integrated mobility spectrometer. J Aerosol Sci 39:940–956. https://doi.org/10.1016/J.JAEROSCI.2008. 06.005 Palmgren F, Wåhlin P, Kildesø J et al (2003) Characterisation of particle emissions from the driving car fleet and the contribution to ambient and indoor particle concentrations. Phys Chem Earth 28:327–334. https://doi.org/10.1016/S1474-7065(03)00053-6 Pirjola L, Paasonen P, Pfeiffer D et al (2006) Dispersion of particles and trace gases nearby a city highway: mobile laboratory measurements in Finland. Atmos Environ 40:867–879. https://doi. org/10.1016/j.atmosenv.2005.10.018 Report NS (2011) Air quality monitoring, emission inventory and source apportionment study for Indian cities Rissler J, Swietlicki E, Bengtsson A et al (2012) Experimental determination of deposition of diesel exhaust particles in the human respiratory tract. J Aerosol Sci 48:18–33. https://doi.org/ 10.1016/j.jaerosci.2012.01.005 Rönkkö T, Virtanen A, Vaaraslahti K, Keskinen J, Pirjola L, and Lappi M (May 2006) Effect of dilution conditions and driving parameters on nucleation mode particles in diesel exhaust: Laboratory and on-road study, Atmos Environ 40(16):2893– 2901 Sacks J (2015) UFP health effects evidence that informed the 2012 PM NAAQS Review Sarnat JA, Demokritou P, Koutrakis P (2003) Measurement of ﬁne, coarse and ultraﬁne particles. Ann Ist Super Sanita 39:351–355 Schmid O, Möller W, Semmler-Behnke M et al (2009) Dosimetry and toxicology of inhaled ultraﬁne particles. Biomarkers 14:67–73. https://doi.org/10.1080/13547500902965617 Seinfeld JH, Pandis SN (1998) Atmospheric chemistry and physics: from air pollution to climate change. John Wiley and Sons, New York. Seinfeld JH, Pandis SN (2006) Atmospheric chemistry and physics: from air pollution to climate change, 2nd edn. Sem GJ, Whitby KT, Sverdrup GM (1980) design, instrumentation, and operation of a large mobile air pollution laboratory for ACHEX, pp 55–68 Shi JP, Khan AA, Harrison RM (1999a) Measurements of ultraﬁne particle concentration and size distribution in the urban atmosphere. Sci Total Environ 51–64 Shi JP, Harrison RM, Brear F (1999b) Particle size distribution from a modern heavy duty diesel engine. Sci Total Environ 305–317 Singh S, Nalwa HS (2007) Nanotechnology and health safety—toxicity and risk assessments of nanostructured materials on human health. J Nanosci Nanotechnol 7:3048–3070. https://doi. org/10.1166/jnn.2007.922 Solomon PA (2012) Ultraﬁne Particles in ambient air Stanier CO (2003) Ultraﬁne particles in the atmosphere : formation, emissions and growth , p 328 Stanier CO, Pandis SN (2004) Insight into secondary organic aerosol partitioning from temperature-ramped chamber experiments. J Aerosol Sci Tech 35 Stanier CO, Khlystov AY, Pandis SN (2004) Nucleation events during the Pittsburgh air quality study: description and relation to key meteorological, gas phase, and aerosol parameters. Aerosol Sci Tech 38:253–264. https://doi.org/10.1080/02786820390229570

38

S. K. Yadav et al.

Stephens B, Azimi P, El Orch Z, Ramos T (2013) Ultraﬁne particle emissions from desktop 3D printers. Atmos Environ 79:334–339. https://doi.org/10.1016/j.atmosenv.2013.06.050 Terzano C, Di Stefano F, Conti V et al (2010) Air pollution ultraﬁne particles: toxicity beyond the lung. Eur Rev Med Pharmacol Sci 14:809–821 Thompson N, Ntziachristos L, Samaras Z et al (2004) Overview of the European “particulates” project on the characterization of exhaust particulate emissions from road vehicles: results for heavy duty engines. https://doi.org/10.4271/2004-01-1986 Thorley AJ, Ruenraroengsak P, Potter TE, Tetley TD (2014) Critical determinants of uptake and translocation of nanoparticles by the human pulmonary alveolar epithelium. ACS Nano 8:11778–11789. https://doi.org/10.1021/nn505399e Torricelli AAM, Novaes P, Matsuda M et al (2011) Ocular surface adverse effects of ambient levels of air pollution. Arq Bras Oftalmol 74:377–381 Vattanasit U, Navasumrit P, Khadka MB et al (2013) Oxidative DNA damage and inflammatory responses in cultured human cells and in humans exposed to trafﬁc-related particles. Int J Hyg Environ Health 1–11. https://doi.org/10.1016/j.ijheh.2013.03.002 Venkataraman C, Rao GUM (2001) Emission factors of carbon monoxide and size-resolved aerosols from biofuel combustion. https://doi.org/10.1021/es001603d Versura P, Profazio V, Cellini M et al (1999) Eye discomfort and air pollution. Ophthalmologica 213:103–109. https://doi.org/10.1159/000027401 Vicente ED, Alves CA (2018) An overview of particulate emissions from residential biomass combustion. Atmos Res 199:159–185 Viitanen A, Uuksulainen S, Koivisto AJ, et al (2017) Workplace Measurements of Ultraﬁne Particles—a literature review. Annals Work Expos Health 61:1–10. https://doi.org/10.1093/ annweh/wxx049 Vincent JH (2007) Aerosol sampling : science, standards, instrumentation and applicatns. Wiley, London Wang LK (2009) Edited by Nazih K. Shammas, Yung-Tse Hung Wang Y, Allen A, Mark D, Harrison RM (1999) Development of a personal monitoring method for nitrogen dioxide and sulfur dioxide with Sep-Pak C18 cartridge sampling and ion chromatographic determination. J Environ Monit 1:423–426 Wentzel M, Gorzawski H, Naumann K-H et al (2003) Transmission electron microscopical and aerosol dynamical characterization of soot aerosols. J Aerosol Sci 34:1347–1370. https://doi. org/10.1016/S0021-8502(03)00360-4 Westerdahl D, Fruin S, Sax T et al (2005) Mobile platform measurements of ultraﬁne particles and associated pollutant concentrations on freeways and residential streets in Los Angeles. Atmos Environ 39:3597–3610. https://doi.org/10.1016/j.atmosenv.2005.02.034 Williams M (2004) Air pollution and policy—1952—2002. Sci Total Environ 335:15–20. https:// doi.org/10.1016/j.scitotenv.2004.04.026 Workshop UP (2015) Health effects of ambient ultraﬁne particles—new evidence on short-term exposures health effects of ambient ultraﬁne particles multicenter time-series studies. Evidence from the Beijing Olympics Health effects of personal exposure. Summary and research Wu S, Deng F, Liu Y et al (2013) Temperature, trafﬁc-related air pollution, and heart rate variability in a panel of healthy adults. Environ Res 120:82–89. https://doi.org/10.1016/j. envres.2012.08.008 www.epa.gov/pm-pollution/table-historical-particulate-matter-pm-national-ambient-air-qualitystandards-naaqs Yegambaram M, Manivannan B, Beach TG, Halden RU (2015) Role of environmental contaminants in the etiology of Alzheimer’s disease: a review. Curr Alzheimer Res 12:116–146 Zhang KM, Wexler AS, Zhu YF, Hinds WC, and Sioutas C (Dec. 2004) Evolution of particle number distribution near roadways. Part II: the ‘Road-to-Ambient’ process, Atmos Environ 38 (38):6655–6665 Zhu Y, Hinds WC, Kim S et al (2002) Study of ultraﬁne particles near a major highway with heavy-duty diesel trafﬁc. Atmos Environ 36:4323–4335. https://doi.org/10.1016/S1352-2310 (02)00354-0

Part II

Diesel Particulates

Chapter 3

Image-Based Flame Temperature and Soot Analysis of Biofuel Spray Combustion Joonsik Hwang, Felix Sebastian Hirner, Choongsik Bae, Chetankumar Patel, Tarun Gupta and Avinash Kumar Agarwal Abstract This chapter describes image-based flame temperature and soot characterization for biodiesel combustion in a constant volume combustion chamber. Flame temperature and qualitative soot emissions were derived from correlated color temperature (CCT) and hue–saturation–value calculations based on high-speed imaging. The direct imaging was carried out in a constant volume combustion chamber (CVCC) under simulated diesel engine conditions. Three different biodiesels produced from Jatropha oil, Karanja oil, and waste cooking oil (WCO) were utilized for the comparison. Conventional diesel fuel was tested as a baseline fuel. The fuels were injected by a common-rail injection system under ambient pressure of 5 MPa and ambient temperature of 978.15 K. Fuel injection pressure was varied from 40 to 120 MPa to investigate effects of injection pressure on spray flame. The experimental results showed that the flame intensity and flame

J. Hwang F. S. Hirner C. Bae (&) Department of Mechanical Engineering, Korea Advanced Institute of Science and Technology, 291 Daehak-Ro, Yuseong-Gu, Daejeon 305-701, Republic of Korea e-mail: [email protected] J. Hwang e-mail: [email protected] F. S. Hirner e-mail: [email protected] C. Patel A. K. Agarwal Engine Research Laboratory, Department of Mechanical Engineering, Indian Institute of Technology Kanpur, Kanpur 208016, India e-mail: [email protected] A. K. Agarwal e-mail: [email protected] T. Gupta Department of Civil Engineering, Indian Institute of Technology Kanpur, Kanpur, India e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2019 A. K. Agarwal et al. (eds.), Engine Exhaust Particulates, Energy, Environment, and Sustainability, https://doi.org/10.1007/978-981-13-3299-9_3

41

42

J. Hwang et al.

area got smaller as the injection pressure increased regardless of fuels. It implies that the soot formation slowed down due to enhanced air–fuel mixing. As the mixture quality was improved under high-pressure condition, the ignition delay got shortened. In terms of fuel, biodiesel fuels showed lower soot emissions than diesel due to inherent oxygen content in the fuel molecules. WCO biodiesel showed the largest soot emission reduction among the tested fuels. Meanwhile, Karanja biodiesel showed relatively higher soot emissions due to inferior fuel properties. Temperature analysis showed lower temperature ﬁeld in biodiesel spray flame as their soot emissions were lower.

Keywords Correlated color temperature (CCT) Hue–saturation–value (HSV) Spray, flame Constant volume combustion chamber (CVCC)

3.1

Introduction

In combustion science, characterization of radiation bands originating from combustion of hydrocarbon fuels has taken steady interests because it can provide detailed information about chemical process in flame. Understanding of chemical process and flame characteristics are essential for engine design and performance improvement. Flames in internal combustion engine can be divided into two different categories. The ﬁrst one is diffusion flame, which is represented by mixing-controlled combustion. Particulate matters give an intense, continuous radiation emission proﬁle. Diesel combustion is the example of diffusion flame where combustion process is dominated by air–fuel mixing. Diffusion flame in diesel engine presents yellowish-red flame due to soot particles. On the other hand, gasoline engine is governed by premixed flame which shows green-blue colors. This is especially predominant under lean premixed flames showing discrete radiation bands from excited intermediate radicals formed during the breakdown of reactant molecules returning to their ground state. The premixed flame normally does not produce much soot emissions, so yellowish-red color flame is not prominently visible. The knowledge of unique radiation characteristics of different types of combustion enabled the development of nonintrusive optical diagnostics using high-speed cameras and laser techniques. These optical diagnostics can provide flame characterization information, for example, flame temperature and soot emissions. As charge-coupled device (CCD) and complementary metal oxide semiconductor (CMOS) cameras developed and replaced conventional ﬁlm cameras, detailed analysis of flame could be accomplished. One could get color information with certain values which were not possible with ﬁlm cameras. Thus, many researchers are now trying to understand broadband radiation ﬁltering architecture with digital color cameras because it can present meaningful data for combustion analysis (Huang and Zhang 2010; Svensson et al. 2005; Cantrell et al. 2009). Different from laser diagnostics, which employ expensive and complicated

3 Image-Based Flame Temperature and Soot Analysis …

43

experimental setup, analysis of red, green, and blue (RGB) signals from direct imaging can provide much simpler and cost-effective solution for combustion diagnostics. It does not need speciﬁc ﬁlters and light sources but natural luminosity from flame. The main aim of this study is to exemplify flame analysis based on RGB digital color signals from high-speed camera. Correlated color temperature (CCT) and hue–saturation–value (HSV) were calculated based on the direct flame images in constant volume combustion chamber (CVCC) using three different biodiesels. The adequacy of the image analysis was conﬁrmed based on comparison between processed images and experimental results from previous researches.

3.2 3.2.1

Experimental Setup and Image Processing Test Fuel Properties

Three different biodiesels, Karanja biodiesel, Jatropha biodiesel, and waste cooking oil (WCO) biodiesel, were tested in this study. The Karanja biodiesel and Jatropha biodiesel were produced by a fuel production pilot plant in Indian Institute of Technology Kanpur. WCO biodiesel was produced by a Korean company called EMACBIO. Detailed process of fuel production can be found in previous publications (Hwang et al. 2017). The main fuel properties are summarized in Table 3.1. As it can be seen in the table, biodiesels showed higher fuel density, viscosity, and flash point. On the other hand, heating values were about 90% of the conventional diesel fuel. The biodiesel fuels showed lower volatility, so in terms of mixture preparation, biodiesels have disadvantages because of inferior fuel atomization and vaporization characteristics. However, this feature could be compensated by oxygen content in fuel molecules, which promotes complete combustion (Manin et al. 2014; Hwang et al. 2015; Agarwal et al. 2011).

3.2.2

Constant Volume Combustion Chamber (CVCC) Setup

A schematic diagram of the CVCC is shown in Fig. 3.1. The CVCC was fabricated by carbon steel (S45C) material to endure high internal pressure and temperature conditions. The maximum expected in-chamber pressure and temperature were 15 MPa and 2000 K, respectively. The volume of the CVCC was approximately 1.4 l. Among the six sides of the chamber, one side opposite to injector was ﬁtted with a 9.6-cm-diameter quartz window for direct high-speed imaging. Four dummy metal windows were installed in remaining four sides of the chamber. The test fuels were injected using a common-rail injection system. The high-pressure fuel pump

44

J. Hwang et al.

Table 3.1 Fuel properties of WCO biodiesel, Karanja biodiesel, and Jatropha biodiesel Property

Diesel

Waste cooking oil biodiesel

Karanja biodiesel

Jatropha biodiesel

ASTM method

Lower heating value (MJ/kg) Density (@ 288 K) (kg/m3) Kinematic viscosity (@ 313 K) (mm2/s) Flash point (K) Cetane number

42.98

38.85

39.89

41.5

D240

820

878

886

857

D4052

2.2

3.32

5.66

4.66

D445

329

448

417

423

D3828

>52

51.34

59–60

60–63

https://www.transportpolicy.net/ standard/south-korea-fuelsdiesel-and-gasoline/; Patel and Sankhavara (2017)

was driven by an electric motor through a belt. The injection pressure and injection quantity were controlled by a common-rail engine controller (Zenobalti, ZB-9013). The injection pressure varied from 40 to 120 MPa. The injection quantity was 30 mg. A seven-hole solenoid injector (Bosch) with a spray hole diameter of 0.130 mm and an injection angle of 150° was used for the experiments. A high-speed digital video camera (Vision Research Inc, Phantom V.7.0) equipped with a prime lens (Nikkon, 60 mm f/2.8D) was used to capture spray flame images for further processing. The exposure time and sampling rate were set as 10 µs and 20,000 frames per second, respectively. A signal from the injection system was utilized to trigger the high-speed camera. The chamber temperature was kept at 473 K by embedded electric heaters in the chamber wall. For the experiments, the chamber was vacuumed to 0.0005 MPa and ﬁlled with mixture of acetylene (C2H2), hydrogen (H2), nitrogen (N2), and oxygen (O2) gases for preignition. The composition was 3% C2H2, 0.5% H2, 28.38% O2, and 68.12% N2 up to an initial pressure of 2.376 MPa. The mixture was ignited by two spark plugs, so the in-chamber pressure and temperature were increased after the preignition. The fuel was injected 1.74 s after the preignition started. The in-chamber pressure and temperature are 5 MPa and 978.15 K at that timing.

3.2.3

Image Process Procedure

The recorded high-speed images were analyzed to get qualitative flame temperature and soot emissions. The flame temperature could be obtained by calculating

3 Image-Based Flame Temperature and Soot Analysis …

45

Fig. 3.1 Schematic of CVCC system

correlated color temperature (CCT). The chromaticity diagram in Fig. 3.2 includes Planckian Locus (Wikipedia 2017). The Planckian Locus is a pathway, which a theoretical blackbody color will take as the blackbody temperature changes. As shown in Fig. 3.2, if the black body temperature is getting hotter, the surface color will be changed from red to orange, yellow, white and ﬁnally to blue. The corresponding temperature is described in the diagram. The lines crossing pathway indicate a constant correlated color temperature. The RGB signals recorded by high-speed camera can be converted into the chromaticity map. Any color presented in the form of RGB can be expressed by the value of chromaticity parameters x and y which are presented as X-axis and Y-axis in Fig. 3.2. In the previous research of Hernández-Andrés et al., they suggested to use Eq. (3.1) to convert the chromaticity values into CCT (Hernández-Andrés et al. 1999). The RBG color information should be converted into chromaticity coordinates X and Y to apply Eq. 3.1. The detailed conversion method is described in Agarwal (2012). n n n x xe CCT ¼ A0 þ A1 exp ð3:1Þ þ A2 exp þ A3 exp ;n ¼ t1 t2 t3 y ye The constant xe is 0.3366, ye is 0.1735, A0 is −949.86315, A1 is 6253.80338, A2 is 28.70599, A3 is 0.00004, t1 is 0.92159, t2 is 0.20039, and t3 is 0.07125. The above equation is only valid for a CCT range from 3000 to 50,000 K. The flame temperature analysis cannot provide exact precise temperature because it is indirect measurement of flame temperature. A certain amount of error is introduced

46

J. Hwang et al. 0.9

Fig. 3.2 Chromaticity space and Planckian locus (Wikipedia 2017)

520

0.8

540

0.7

560 0.6

500

580

0.5

Tc(K)

Y

4000

3000 2500

6000

0.4

2000 1500

10000

0.3

490

600 620 700

∞

0.2 0.1

480

0.0 0.0

470 460 0.1

380

0.2

0.3

0.4

0.5

0.6

0.7

0.8

X

according to flame temperature range; however, it enables a qualitative comparison of flame temperature among the fuels. Flames in internal combustion engine are mostly composed of yellowish-red diffusion flame from the glowing soot particles or greenish-blue premixed flame from the excited gaseous species such as CH* and C2* radicals (Huang and Zhang 2010). In this study, soot incandescence was dominant in diesel and biodiesel fuel spray combustions. Similar to the temperature analysis, RBG signals were converted into hue–saturation–value (HSV) model for qualitative soot measurement. The hue (H) value corresponds uniquely to each RGB combination. Using a correlation suggested by Smith, RGB signals can be converted into unique H value as the following equations (Eq. 3.2a to 3.2e) (Smith 1978): H00 ¼

GB if R ¼ Max MaxðR; G; BÞ MinðR; G; BÞ

ð3:2aÞ

¼ 2þ

BR if G ¼ Max MaxðR; G; BÞ MinðR; G; BÞ

ð3:2bÞ

¼ 4þ

RG if B ¼ Max MaxðR; G; BÞ MinðR; G; BÞ

ð3:2cÞ

H00 360 6

ð3:2dÞ

H0 ¼

3 Image-Based Flame Temperature and Soot Analysis …

47

Fig. 3.3 Correlation between H and flame types

H ¼ H0 if H0 [ 0

ð3:2eÞ

¼ H0 þ 360 if H0 \0 From a previous study, it was revealed that an H value between 0 and 80 indicates the presence of diffusion flames (Huang and Zhang 2010). Meanwhile, an H value lying between 180 and 300 indicates the presence of premixed flames. The coloration is attributed to chemiluminescence emissions from excited radicals. Figure 3.3 shows correlation between H value and flame types. The theory also suggests that the red signal in the diffusion flame indicates the soot induced digital coloration; on the other hand, the blue and green signals in the premixed flame represent the chemiluminescence induced coloration of the flame from CH* or C2* radicals (Huang and Zhang 2011). Therefore, the signal intensity from red color where H value lies between 0 and 80 can be considered as soot concentration.

3.3

Results and Discussion

Figure 3.4 shows direct images of spray flame at 40, 80, and 120 MPa injection conditions. The brownish black-colored region in the center of spray flame represents soot emissions from fuel-rich combustion. It can be seen that sooting tendency was decreased as the injection pressure was increased. This is due to the smaller

48

J. Hwang et al.

fuel-rich zone formed as mixture quality was enhanced with high injection pressure. The another interesting feature is that the overall flame intensity was decreased as the injection pressure was increased. This implies that the amount of soot emissions got lowered with higher injection pressure. The visible flame duration was also got shortened under higher injection pressure. Quantitative flame area result is shown in Fig. 3.5. The flame here presents number of pixels which are greater than certain threshold light intensity. The real scale was not considered so it does not have unit. It can be seen that the peak of flame area got lowered and ignition delay also got shortened as the injection pressure increased. The reason is that more homogenous air–fuel mixture was formed at higher injection pressure. The atomization of injected fuel droplets was more enhanced by higher momentum at 120 MPa than 40 MPa. Thus, chemical reaction was accelerated and showed shorter ignition delay and higher flame area increasing rate. It is noticeable that the flame lift-off length (LOL) got longer as fuel injection pressure got higher. The flame was started adjacent to injector under 40 MPa; on the other hand, the flame was started from a certain distance from injector at 120 MPa. Longer lift-off length is ﬁnally resulted in smaller flame area. In terms of fuels, biodiesels showed narrower spray flame than diesel fuel because of inferior atomization process. The smaller spray angle also could be found in previous research (Hwang et al. 2016). The spray flame morphology was not distinguishable by direct imaging; however, biodiesel fuels tended to have shorter visible flame duration than diesel fuel. It is attributed to two possible reasons. The ﬁrst one is that the amount of soot generated was lower with biodiesel fuel so that they showed lower flame intensity and shorter visible flame duration. The second one is that the soot oxidation process was accelerated by inherent oxygen content in the fuel. In Fig. 3.5, it can be seen that the diesel showed the largest flame area and the WCO biodiesel has the smallest flame area. The Karanja biodiesel showed relatively higher flame area and smaller enhancement according to injection pressure. The main reason is that the fuel atomization could not be improved much because of higher fuel density and viscosity than other fuels. Figure 3.6 shows the estimation of soot emission in the spray flame. The qualitative soot emission comparison was performed based on the theories described in the previous section. It can be seen that overall soot emissions got lower with higher injection pressure as the mixture quality was enhanced. One obvious thing is that the soot cloud in the center of the spray flame could be considered as lower soot region because it did not have flame luminosity. The calculation is based on the flame luminosity so it makes much error in that region. The longer LOL was also conﬁrmed with soot analysis showing lower luminosity around injector area with higher injection pressure. The soot images showed the bright white area at the beginning of the spray flame; however, the overall intensity got lowered as the soot oxidation was progressed. In terms of fuels, as it was conﬁrmed that the WCO took much advantages of injection pressure, the soot emissions sharply decreased with higher injection pressure case. This trend was also found in the previous research (Hwang et al. 2014).

3 Image-Based Flame Temperature and Soot Analysis … Fig. 3.4 Direct spray flame high-speed images at injection pressures of a 40 MPa, b 80 MPa, and c 120 MPa

49

50

J. Hwang et al.

Fig. 3.5 Flame area calculated from processed spray flame image (based on pixel numbers)

Figure 3.7 shows calculated temperature ﬁeld in the spray flame under different injection pressures. The high-speed camera was not calibrated, so the quantitative temperature assessment was not possible. However, qualitative comparison in temperature ﬁeld could be carried out because they were taken under identical experimental conditions. The red color in the image represents higher temperature region; meanwhile, the blue color means lower temperature region. It can be seen that under 40 MPa, diesel fuel showed the highest temperature ﬁeld among the fuels. The diesel fuel also showed higher temperature from the beginning of the ignition process and the duration of high temperature was longer than other fuels. The main reason is that the energy from exothermic reaction was greater with diesel fuel due to higher heating value. The injection quantity in this experiment was ﬁxed at 30 mg, so the total input energy by fuel injection could be varied. The biodiesel fuels showed relatively lower temperature ﬁeld than diesel fuel; however, the differences got smaller as the fuel injection pressure got increased. As the injection pressure increased, the initial high-temperature region in diesel was decreased.

3.4

Conclusions

In this chapter, an image-based temperature and soot analysis were introduced and applied into spray flame of biodiesel fuel and diesel combustion in a constant volume combustion chamber. The correlated color temperature (CCT) and hue– saturation–value (HSV) were calculated and successfully presented under different injection pressures. The results showed that the estimation had consistent overall tendency compared to experimental data. Biodiesel fuels showed lower soot luminosity because of inherent oxygen content and faster soot oxidation process

3 Image-Based Flame Temperature and Soot Analysis … Fig. 3.6 Soot emission estimation based on HSV theory at injection pressures of a 40 MPa, b 80 MPa, and c 120 MPa

(a)

51

t = 1.80 ms

t = 2.85 ms

t = 3.90 ms

t = 1.35ms

t = 1.95 ms

t = 2.55 ms

Diesel

WCO

Karanja

Jatropha

(b) Diesel

WCO

Karanja

Jatropha

(c)

52 Fig. 3.7 Qualitative temperature ﬁeld comparison between fuels at injection pressures of a 40 MPa, b 80 MPa, and c 120 MPa

J. Hwang et al.

(a)

(b) Diesel

WCO

Karanja

Jatropha

(c)

t = 1.35 ms

t = 1.95 ms

t = 2.55 ms

3 Image-Based Flame Temperature and Soot Analysis …

53

than diesel fuel. However, the image processing had some possible error sources because it is based on the image intensity. The soot cloud region shown as brownish black region was presented as sootless area due to lack of the flame intensity. This implies that we need to be careful to select region of interest to apply the theory to the images. Acknowledgements The work undertaken in Korea was ﬁnancially supported by the National Research Foundation (NRF) of Korea (2014K1A3A1A19067560) and that undertaken in India was ﬁnancially supported by the Department of Science and Technology of India (INT/KOREA/ P-23 dated 06-07-2015) under the Indo-Korea Joint International Cooperation Project. The support is gratefully acknowledged. This funding enabled researchers to participate in exchange visits to conduct experiments and develop consumables.

References Agarwal A (2012) Endoscopic combustion diagnostics of biodiesel fueled compression ignition engine. IIT Kanpur, July 2012 Agarwal AK, Gupta T, Kothari A (2011) Particulate emissions from biodiesel vs diesel fueled compression ignition engine. Renew Sust Energ Rev 15:3278–3300 Cantrell K, Erenas MM, de Orbe-Payá I, Capitán-Vallvey LF (2009) Use of the hue parameter of the hue, saturation, value color space as a quantitative analytical parameter for bitonal optical sensors. Anal Chem 82:531–542 Hernández-Andrés J, Lee RL, Romero J (1999) Calculating correlated color temperatures across the entire gamut of daylight and skylight chromaticities. Appl Opt 38:5703–5709 Huang HW, Zhang Y (2010a) Dynamic application of digital image and colour processing in characterizing flame radiation features. Meas Sci Technol 21:1–9 Huang H, Zhang Y (2010b) Dynamic application of digital image and color processing in characterizing flame radiation features. Meas Sci Technol 21:1–9 Huang H, Zhang Y (2011) Digital color image processing based measurement of premixed CH4 + air and C2H4+ air flame chemiluminescence. Fuel 90:48–53 Hwang J, Qi D, Jung Y, Bae C (2014) Effect of injection parameters on the combustion and emission characteristics in a common-rail direct injection diesel engine fueled with waste cooking oil biodiesel. Renew Energy 63:9–17 Hwang J, Jung Y, Bae C (2015) Comprehensive assessment of soot particles from waste cooking oil biodiesel and diesel in a compression ignition engine. SAE Int J Fuels Lubr 8:290–297. https://doi.org/10.4271/2015-01-0809 Hwang J, Bae C, Gupta T (2016) Application of waste cooking oil (WCO) biodiesel in a compression ignition engine. Fuel 176:20–31 Hwang J, Bae C, Patel C, Agarwal RA, Gupta T, Agarwal AK (2017) Investigations on air-fuel mixing and flame characteristics of biodiesel fuels for diesel engine application. Appl Energy 206:1203–1213 Manin J, Skeen S, Pickett L (2014) Effects of oxygenated fuels on combustion and soot formation/ oxidation processes. SAE Int J Fuels Lubr 7:704–717 Patel RL, Sankhavara CD (2017) Biodiesel production from Karanja oil and its use in diesel engine: a review. Renew Sustain Energy Rev 71:464–474

54

J. Hwang et al.

Smith A (1978) Color gamut transform pairs. In: Proceeding of 5th annual conference on computer graphics and interactive techniques, 12–19, 1978 South Korea: Fuels: diesel and gasoline | Transport policy (2018). Available from: https://www. transportpolicy.net/standard/south-korea-fuels-diesel-and-gasoline/ Svensson K, Mackrory A, Richards M, Tree D (2005) Calibration of an RGB, CCD camera and interpretation of its two-color images for KL and temperature. SAE Tech Pap 2005-01-0648. https://doi.org/10.4271/2005-01-0648 Wikipedia, “Color temperature,” Wikipedia Foundation, Inc., 10 08 2017 [Online]. Available: https://en.wikipedia.org/wiki/Color_temperature. Accessed 23 July 2018

Chapter 4

Characteristics and Fundamentals of Particulates in Diesel Engine Khawar Mohiuddin and Sungwook Park

Abstract Diesel engine is preferred over its gasoline counterpart due to high torque output and better fuel efﬁciency. However, these beneﬁts come at the cost of higher emissions. Engine exhaust particulates from a compression ignition (CI) engine, principally consist of combustion-generated carbonaceous soot. Production of this soot is mainly attributed to incomplete combustion of fuel. A complete understanding of diesel particulate composition is mandatory so that requisite measures can be taken to reduce the engine-out emissions of a CI engine. Information about soot composition helps in chemical and biological characterization of soluble/insoluble organic compounds. This chapter covers the basics of soot emission from CI engines and composition of soot along with its structure. Fundamentals of particle formation, oxidation, adsorption, and condensation of soot have also been discussed in the later part of the chapter. Keywords Diesel engine

4.1

Emission Particulates Soot formation

Introduction

Particulate matter (PM) emissions of a vehicle are deﬁned as the total mass of solids and other accompanied volatile or soluble constituents. PM emitted from any source is a big health hazard since it carries cancer-causing polyaromatic hydrocarbons (PAHs). PM emitted from a compression ignition engine is predominantly carbonaceous (soot) combined with absorbed hydrocarbons and inorganic compounds (sulfates and water, etc.). The magnitude of different constituents of particulates K. Mohiuddin Department of Mechanical Convergence Engineering, Graduate School of Hanyang University, Seoul 04763, Republic of Korea S. Park (&) School of Mechanical Engineering, Hanyang University, Seoul 04763 Republic of Korea e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2019 A. K. Agarwal et al. (eds.), Engine Exhaust Particulates, Energy, Environment, and Sustainability, https://doi.org/10.1007/978-981-13-3299-9_4

55

56

K. Mohiuddin and S. Park Unburnt Fuel 7% Unburnt Oil 25%

Sulfate % Water 14%

Carbon 41%

Ash & Other 13%

Fig. 4.1 Composition of PM from a heavy-duty CI engine

emissions can be signiﬁcantly different depending upon the engine design, emission control strategy employed, quality of fuel, use of appropriate cylinder bore smoothness, and piston ring pack for good oil control (Singal and Gandhi 1996). A very typical composition of particulates from a heavy-duty diesel engine has been illustrated in Fig. 4.1. Major portion of the emitted particulates is soot (elementary carbon). Second largest fraction is of the organic compounds resulting from unburnt hydrocarbons from fuel or lubricating oil. Fraction of sulfates in the particulates is determined by sulfur content of fuel and lubricating oil. During combustion process, sulfur gets oxidized and is converted into different oxides (SO2 and SO3, etc.) depending on temperature at different stages. Table 4.1 shows chemical composition of PM in % for fuels with varying sulfur content. The PM percentage decreases with increasing sulfur content except for the highest sulfur content case. Emission rate for light-duty diesel engines ranges from 0.2 to 0.6 g/km and 0.5– 1.5 g/brake kW h for larger direct-injection engines (Heywood 1988). Simultaneous reduction of NOx and particulate matters has always been a challenging task while dealing with diesel engines since any combustion strategy designed to reduce NOx results in higher PM emissions and vice versa. Soot being a primary constituent of diesel particulates requires a deep understanding of its characteristics and formation process to reduce these emissions to the allowable limit as per emission regulations.

4.2

Particulate Formation in Diesel Engines

Fuel injection in a diesel engine starts toward the end of compression cycle and combustion starts immediately after that. Therefore, the fuel does not get sufﬁcient time to get uniformly distributed. This nonuniform fuel distribution in a diesel

4 Characteristics and Fundamentals of Particulates in Diesel Engine Table 4.1 Chemical composition of PM, % (Singal and Gandhi 1996) Fuel-derived HCs Lube-derived HCs Carbon Sulfur

57

Fuel sulfur content (wt%) 0.055 0.12 0.22

0.33

25 12.7 37.7 1.9

17.9 10.6 28.5 5.2

18 12.3 30.3 3.3

16.9 10.3 27.2 4.1

engine is one of the most important factors responsible for pollutant formation. Figure 4.2 shows two phases of combustion and pollutant formation in a diesel engine with direct fuel injection. Soot is formed inside the flame region generally in air-deﬁcient zones (in the core of fuel sprays, which is rich in unburnt fuel). Because of mixing with hot gases, the temperature of fuel vapors gets higher. Soot oxidation takes place inside the flame when it reacts with unburnt oxygen giving a characteristic yellow color to the flame. Excess of air dilution prohibits the combustion from getting initiated or the combustion process getting completed where it has already started and eventually results in flame quenching. Therefore, the fuel vapors injected later during the same cycle also become a source of particulate formation.

Fig. 4.2 Premixed and mixing controlled combustion in a diesel engine (Heywood 1988)

58

4.3

K. Mohiuddin and S. Park

Structure of Diesel Particulates

Structure of particulates from a compression ignition (CI) engine is almost similar to Fig. 4.3. It consists of primary particles (often referred to as spherules) agglomerated into aggregates (will be referred to as particles). These spherules can form clusters (containing as many as 4000 spherules) as well as long chains. The spherules vary in diameter with most of them in the range of 7.5–1 µm (Hunter and Undem 1999). Typical distribution of spherule size is shown in Fig. 4.4.

Fig. 4.3 Schematic representation of diesel particle agglomeration (Mohankumar and Senthilkumar 2017) Fig. 4.4 Typical size distribution curve of spherules (Amann and Siegla 1982)

4 Characteristics and Fundamentals of Particulates in Diesel Engine

59

Fig. 4.5 Substructure of a carbon particle (Smith 1981)

Going further into the substructural details of diesel particulate, it is suggested that the concentric lamellas of diesel particulates are arranged in the same pattern as the ones in structure of carbon black (Lahaye and Prado 1981). The reason behind this fact is that both soot (in CI engines) and carbon black (in oil furnace) are produced in similar environment. The carbon atoms are bonded together in hexagonal face-centered arrays in planes called platelets. The layered arrangement of these platelets forms crystallites, and the average spacing between these layers is 0.355 nm (Smith 1981). Planes of these crystallites are arranged parallel to surface of the particle. This unordered layer structure is known as turbostratic. The substructure of a carbon particle is shown in Fig. 4.5.

4.4

Fundamentals of Soot Formation

Soot particles from a CI engine originate from carbon which is a constituent of diesel fuel; therefore, their inception initiates from a molecule of fuel having 12–22 atoms of carbon and ends up in particles composed of spherules containing approximately 105 atoms of carbon. The most convenient temperature zone for soot formation to take place in a diesel engine ranges from 1000 to 2800 K with pressure value ranging from 50 to 100 atm. Conversion of liquid fuel into soot particles and ﬁnally into gas phase takes place in six stages, namely, pyrolysis, nucleation, surface growth, coalescence, agglomeration, and oxidation. This process has been illustrated schematically in Fig. 4.6. An important point to note here is that the soot formation process is dependent on various factors like fuel injection pressure, fuel quality, and most importantly reaction pressure and temperature (Hauser et al. 2001).

4.4.1

Pyrolysis

Pyrolysis is deﬁned as a process in which hydrocarbons undergo a lot of decomposition and atomic realignment of fuel molecules before the start of soot formation. Throughout the pyrolysis process, there is usually inadequate oxygen supply. These

60

K. Mohiuddin and S. Park

Fig. 4.6 Schematic of particle formation (Mohankumar and Senthilkumar 2017)

reactions are endothermic in nature and that is why they have high dependency on temperature (Smith 1981). Fuel pyrolysis paves the way for the formation of precursors of soot. Pyrolysis and oxidation rates vary depending upon the type of flame produced. Less soot is produced in premixed flames due to sufﬁcient oxygen content, whereas soot production rate is relatively higher in case of diffusion flame due to lower oxygen content. Oxygen rate is also directly proportional to the temperature. Hence, two factors, i.e., oxygen content and temperature, directly influence the soot formation rate. Since the pyrolysis reaction involves “free radical mechanism”, the presence of free radicals like O, O2, or OH results in acceleration of pyrolysis process by facilitating radical formation due to branching reactions. Acetylene (C2H2), ethylene (C2H4), methane (CH4), propene (C3H6), and benzene (C6H6) are the main products of pyrolysis process in laminar diffusion flames.

4.4.2

Nucleation

Nucleation is deﬁned as the transformation from molecular system to a particulate system, or formation of embryonic species which grow faster as compared to their decomposition (Calcote 1981). Researchers have proposed many theories to explain the process of nucleation and subsequently surface growth of soot particles. Most prominent among those are Palmer and Cullis (1965), Gaydon and Wolfhard (1970), and Lahaye and Prado (1978). There are three commonly accepted approaches which deﬁne soot production. These approaches are characterized by the formation temperature. At low temperature, the most effective species in soot formation through pyrolysis is aromatics or unsaturated aliphatics having a high molecular weight. Diffusion flames (which involve intermediate temperature ranges) produce soot for almost all the fuels provided they are burnt at high stoichiometric ratio. And at extremely high temperatures (which is out of the range of diesel combustion temperature), another nucleation process which involves carbon vapor takes place (Amann and Siegla 1982). A very simple nucleation model applicable in low and intermediate temperature zone was proposed by Graham et al. (1975). Two paths, i.e., direct (fast) and indirect (slow) for production of soot, were proposed. Direct path involved production of soot from aromatic hydrocarbons at low temperatures by condensing the aromatic rings and changing their structure similar to graphite.

4 Characteristics and Fundamentals of Particulates in Diesel Engine

61

Fig. 4.7 Formation of soot from aromatic and aliphatic hydrocarbons (Amann and Siegla 1982)

Indirect path involved breakup of the aromatic rings into smaller hydrocarbon species. Later, these individual species would polymerize and produce large unsaturated molecules resulting in soot nucleation. This process would, however, take place at temperature above 1800 K. Aliphatic molecules can produce soot only by following latter path according to this model. This model has been shown in Fig. 4.7.

4.4.3

Soot Surface Growth

Next step in soot formation process is the surface growth. After culmination of the nucleation process, tiny particles produced during the nucleation process (with negligible soot loading) start gathering mass due to surface growth. Surface growth is the most important factor for increase in soot mass (Wu et al. 2001). This process involves deposition of hydrocarbon intermediates on the surface of spherules which originate from nuclei. The soot fraction FV, in units of soot volume per unit volume of gas, is related to the number density N and the volume-mean diameter of a soot particle by FV ¼

p 6

ND3

ð4:1Þ

N is deﬁned as the number of particles per unit volume of gas and D is the spherule diameter. The rate of change of particle number density with time t is expressed as

62

K. Mohiuddin and S. Park

dN ¼ N_ u N_ a dt

ð4:2Þ

where N_ u is deﬁned as the rate of appearance of new nuclei and N_ a is the rate at which the spherules agglomerate. At the highest point on N curve shown in Fig. 4.8, these two values, i.e., N_ u and N_ a , are equal. N_ u is greater than N_ a toward the left of this point since diameter of the particle remains unchanged at lower values and rise of this curve is essentially governed by nucleation process. Toward the other side of this highest point, N_ a is greater than N_ u . On this side of the curve, the number density falls due to agglomeration and results in increase in particle diameter. Reason behind the formation of “distorted shaped” concentric shells is the surface growth which takes place on spherules and nuclei. They make the outer portions of the spherules look different from unorganized spherules. Empty spaces between adjacent spherules can partially be ﬁlled with the surface growth process taking place on these agglomerated particles and resultantly form a peanut-like conﬁguration.

4.4.4

Agglomeration

Agglomeration is the next step in soot formation process. Once the soot particles are formed, all the available mobile particles and spherules start inter-particulate collisions resulting in an increase in particle size and consequently a decrease in Fig. 4.8 Variation in soot volume fraction FV, particle number density N, particle size d, and soot H/C ratio (Amann and Siegla 1982)

4 Characteristics and Fundamentals of Particulates in Diesel Engine

63

particle number. This process is called agglomeration. The probability of this process to occur is given by Smoluchowski equation. dN ¼ KN 2 dt

ð4:3Þ

where K is coagulation coefﬁcient and N is deﬁned as the number density. Particle density is the number of particles per unit volume. The above relation shows that the decrease in particle number density is directly proportional to the number density squared. Three different types are generally involved in the process of soot formation. The ﬁrst one takes place during initial stages of soot formation where spherules coagulate together due to collision and form a single spheroid. This process can easily be visualized in hydrocarbon pyrolysis where in the beginning, soot particle may be highly viscous like a tarry liquid (Amann and Siegla 1982). Surface growth restores the original spherical shape quickly due to small size of particles. This process only affects the particles up to diameter of 10 nm. Second type of agglomeration takes place when the spherules get solidiﬁed before collisions and surface growth has ceased. In this case, the resulting structure resembles a cluster where initial spherules have retained much of their individual identity. Third type of agglomeration takes place due to continuous coalescence of soot particles and results in the formation of chain-like structure of discrete spherules. This growth of the chain-like structure has been attributed to the presence of positive charge in this structure which has been measured by Wersborg et al. (1973). This type of coalescence is termed as aggregation.

4.4.5

Hydrogenation

As the pyrolysis of fuel proceeds during soot formation, there is a continuous decrease in the H/C ratio in a flame. This dehydrogenation process continues as the soot formation process propagates resulting in decrease in value of H/C ration from a normal of 2 to almost 0.2 at the culmination of agglomeration process. Addition of mass to soot particles due to this process is very minute and this happens due to the reaction of gas-phase molecules. These gas-phase hydrocarbons which are reacting together are basically acetylenes having larger polymers. The smaller polyacetylenes undergo polymerization by the same process leading to the nucleation resulting in increase in larger polymers. Increase in larger polymers results in further decrease of H/C ratio until it reaches a steady-state value.

64

K. Mohiuddin and S. Park

Fig. 4.9 Schematic of a conceptual model showing fuel-rich premixed flame, soot formation, soot oxidation, and NO formation zones (Dec 1997)

4.4.6

Soot Oxidation

Oxidation in soot formation process is the conversion of carbon and other hydrocarbon molecules into combustion products due to oxidation. Soot oxidation process in a diesel engine is very important because it results in reduced soot mass. There is no speciﬁc time during whole formation process for soot to get oxidized. This process can take place during any of the stages discussed so far starting from pyrolysis till agglomeration. Rate of the oxidation reaction depends upon the process during which it takes place and the air–fuel mixture formation at that time. Talking about diesel engines, it is a well-established fact that most of the soot produced during the combustion process gets oxidized inside the engine before it could get exhausted to the atmosphere. There are a lot of species in close vicinity of diesel flame that can oxidize the soot, i.e., O2, O, CO2, H2O, and OH (Lee et al. 1962). Out of these, OH radical is the most effective oxidant for soot under stoichiometric and fuel-rich conditions at atmospheric pressure. Collisions of soot with OH radicals contribute up to 20% reduction in its oxidation process (Haynes and Wagner 1981). However, OH and O2 radicals can oxidize soot under fuel lean conditions as well (Haynes et al. 1991). A conceptual model by Dec (1997) shown in Fig. 4.9 clearly highlights the soot oxidation zone in a diesel flame. Soot particle oxidation generally takes place when the reaction temperature exceeds 1300 K (Glassman 1996). This high resistance to oxidation is attributed to the graphite-like structure of the soot (Hauser et al. 2001). Oxidation of soot takes place in two stages. Initially, oxygen gets absorbed on the surface of soot and in the second stage, it gets desorbed from the surface of soot (Glassman 1996).

4 Characteristics and Fundamentals of Particulates in Diesel Engine

4.5

65

Adsorption and Condensation

Adsorption and condensation of hydrocarbons is the last process of particulate formation. This process does not take place inside the engine. Rather, it takes place when the gases from cylinder get exhausted out from the engine and get diluted with atmospheric air. Experimental setups for PM measurement use dilution tunnel which simulates exact atmospheric conditions. A sample of diluted exhaust is continuously collected using a very ﬁne ﬁlter and the engine keeps running a predeﬁned proﬁle during this process. This collection ﬁlter is equilibrated to maintain the required level of humidity by removing the excess water content. Mass of the ﬁlter is measured after collecting the sample and particulate mass is obtained by calculating the increase in ﬁlter mass. Particulate mass is obtained by measuring the mass gain by ﬁlter. As per EPA procedure, the ﬁlter temperature must be below 52 °C. The ﬁlter temperature can be controlled by tunnel dilution ratio as shown in Fig. 4.10. This study has also been carried out by Black and High (1979), Najt (1978). The same experimental procedure has also been used by Frisch et al. (1979). Ratio between flow rates of diluted stream and engine exhaust is deﬁned as “Tunnel Dilution Ratio.” Dependence of ﬁlter temperature on the dilution ratio for a given exhaust gas temperature can be seen clearly in the ﬁgure below. Total sample collected from the ﬁlter is further divided into extractable and non-extractable fractions. The mass of collected particulates is greatly affected by variation of dilution ratio and this effect has been shown in Fig. 4.11 where extractable and non-extractable fractions are highlighted. The non-extractable fraction comprises carbonaceous soot which is generated during combustion and dilution does not affect it. At dilution ratio of 1 (no dilution), there is a very minute difference between the total and non-extractable fraction. Extractable fraction comprises the species which can be extracted from the exhaust using a chemical Fig. 4.10 Variation in sample temperature as a function of exhaust temperature Te due to change in tunnel dilution ratio (Amann and Siegla 1982)

66

K. Mohiuddin and S. Park

process. The extractable fraction increases with increase in dilution ratio till a certain value after which this fraction starts decreasing. This phenomenon was experimentally veriﬁed by MacDonaid et al. (1980). They experienced 56% reduction in extractable fraction when the temperature of ﬁlter was increased from 35 to 100 °C keeping the dilution ratio ﬁxed at 10.8. Also, when the dilution ratio was decreased from 100 to 5 at constant ﬁlter temperature of 52 °C, the particulate mass was found to increase by 36%. Results from this study were used by Plee and MacDonald (1980) and they concluded that both adsorption and condensation phenomena occur during the dilution process. Intermolecular forces are responsible for adhesion of unburned hydrocarbon molecules to the surface of soot particles. This adhesion phenomenon is called adsorption. These forces are called van der Waals forces. There are two main driving factors for this process. One is the available fraction surface area of soot particle which has already been adhered by hydrocarbon molecules and second one is the partial pressure of gaseous hydrocarbons. As it can be seen in Fig. 4.11, an increase in the dilution ratio from 1 results in increased extractable fraction because the effect of declining temperature on number of active sites dominates. At higher values of dilution ratio, the sample temperature no more responds to the dilution ratio and extractable fraction decreases due to decrease in hydrocarbon partial pressure. Condensation of soot takes place due to increased vapor pressure of gaseous hydrocarbons. This process starts when this pressure becomes higher than the saturated vapor pressure. If the dilution ratio is increased, it results in reduction of

Fig. 4.11 Effect of dilution ratio on extractable and non-extractable fractions of particulate mass (Amann and Siegla 1982)

4 Characteristics and Fundamentals of Particulates in Diesel Engine

67

hydrocarbon mole fraction which in turn decreases the vapor pressure. Resultant reduction in temperature lowers the saturation pressure. The most favorable conditions for condensation are high concentration of hydrocarbons (with low volatility). Hydrocarbons with low volatility are lubricating oil and unburned hydrocarbons which have undergone pyrolysis but not yet undergone combustion process. To investigate the role of engine lubricating oil in production of particulates in diesel engine, (Mayer et al. 1980) carried out experiments on a passenger car at different speeds and engine load conﬁgurations. It was concluded that the lubricating oil contributed 1.35–25% of the particulate mass with highest contribution at higher speeds. The trend was same in case of low load conditions also but the contribution was found to be greater than the higher load condition at same engine speed. Particulate mass originating from lubricating oil was found to be in extractable zone. This was a very clear indication that lubricating oil has no contribution in combustion process.

4.6

Summary

Soot in engines is produced due to fuel-rich combustion. Diesel engines always have a lean equivalence ratio but the combustion in a CI engine still results in production of soot. Reason behind this characteristic is the result of heterogeneous combustion in diesel engines. Large fraction of soot is produced during slow diffusive burning process. During this process, recently injected fuel stays oxygen deprived because major portion of the available oxygen has already been consumed by the fuel injected earlier in the same cycle. Another important factor in this context is the fuel and oxygen mixing and that is the reason why diesel engines are specially designed to facilitate proper fuel and air mixing. Carbonaceous soot generated during combustion is the main constituent of diesel particulate matter. Soot is collected on a ﬁlter in a dilution tunnel and this collected sample is further characterized into extractable and non-extractable fraction. During the dilution process, 10–30% of the collected sample mass comprises extractable fraction. This extractable fraction gets adsorbed on surface of the soot particles. Condensation of gaseous hydrocarbons can also result in adsorption. Usually, condensation takes place in the exhaust region where there is a high concentration of hydrocarbons with low volatility. High-boiling-point components are also a favorable condition for condensation to take place. Researchers have proved that lubricating oil contributes signiﬁcantly to the production of PM. Along with the main constituents, i.e., carbon and hydrogen, minor traces of nitrogen, oxygen, and few other elements are also a part of particulates of a diesel engine. Soot formation in a CI engine begins nucleation process where embryonic species are formed due to chemical changes in the fuel. Growth of these nuclei leads to the formation of soot spherules which are up to 30 nm in diameter. Agglomeration overtakes the nucleation process as this cycle proceeds further.

68

K. Mohiuddin and S. Park

Chains and clusters are formed due to collision and agglomeration process. Surface growth takes place on these agglomerated particles as the agglomeration process is completed. Oxidation process reduces the mass of soot inside the cylinder. Out of all the available radicals, OH radical is the principal oxidant in the stoichiometric flame zone. As the mixture becomes leaner, oxygen radical starts contributing the most in the oxidation process. With the start of expansion stroke, soot oxidation ceases due to decreasing temperature. Decrease in soot temperature is a consequence of increasing cylinder volume and fresh cooler air entering the cylinder.

References Amann CA, Siegla DC (1982) Diesel particulates: what they are and why. Aerosol Sci Technol 1 (1):73–101. https://doi.org/10.1080/02786828208958580 Black F, High L (1979) Methodology for determining particulate and gaseous diesel hydrocarbon emissions SAE Technical Paper 790422. https://doi.org/10.4271/790422 Calcote HF (1981) Mechanisms of soot nucleation in flames—a critical review Combustion and flame 42:215–242 Dec J (1997) A conceptual model of di diesel combustion based on laser-sheet imaging. SAE Technical Paper 970873. https://doi.org/10.4271/970873 Frisch LE, Johnson JH, Leddy DG (1979) Effect of fuels and dilution ratio on diesel particulate emissions. SAE Technical Paper 790417. https://doi.org/10.4271/790417 Gaydon AG, Wolfhard HG (1970) Flames: their Structure, radiation and temperature, 3rd edn Glassman I (1996) Combustion. Academic Press, San Diego Graham SC, Homer JB, Rosenfeld JLJ (1975) The formation and coagulation of soot aerosols generated by the pyrolysis of aromatic hydrocarbons. Proc R Soc Math Phys Eng Sci 344:259– 285. https://doi.org/10.1098/rspa.1975.0101 Hauser R, Eisen EA, Pothier L, Christiani DC (2001) A prospective study of lung function among boilermaler construction workers exposed to combustion particulates. Am J Ind Med 39:454–462 Haynes BS, Bartok W, Saroﬁm AF (1991) Fossil fuel combustion. Wiley, New York, pp 261–326 Haynes BS, Wagner HG (1981) Soot formation progress in energy and combustion. Science 7:229–273. https://doi.org/10.1016/0360-1285(81)90001-0 Heywood JB (1988) Internal combustion engine fundamentals. McGraw-Hill, New York Hunter DD, Undem BJ (1999) Identiﬁcation and substance p content of vagal afferent neurons innervating the epithelium of the guinea pig trachea. Am J Resp Crit Care Med 159:1943–1948 Lahaye J, Prado G (1978) Mechanisms of carbon black formation. Chem Phys Carbon 14:167–294 Lahaye J, Prado G (1981) Morphology and internal structure of soot and carbon blacks. Part Carbon Form During Combust 33–51. doi:https://doi.org/10.1007/978-1-4757-6137-5 Lee KB, Thring MW, Beer JM (1962) On the rate of combustion of soot in a laminar soot flame. Combust Flame 6:137–145 MacDonaid JS, Plee SL, D’Arcy JB, Schreck RM (1980) Experimental measurements of the independent effects of dilution ratio and ﬁlter temperature on diesel exhaust particulate samples. SAE Technical Paper 800185. https://doi.org/10.4271/800185 Mayer WJ, Lechman DC, Hilden DL (1980) The contribution of engine oil to diesel exhaust particulate emissions. SAE Technical Paper 800256. https://doi.org/10.4271/800256 Mohankumar S, Senthilkumar P (2017) Particulate matter formation and its control methodologies for diesel engine: a comprehensive review. Renew Sustain Energy Rev 80:1227–1238. https:// doi.org/10.1016/j.rser.2017.05.133

4 Characteristics and Fundamentals of Particulates in Diesel Engine

69

Najt P (1978) Light-duty particulate measurements using a CVS tunnel. In: Symposium on diesel particulate emissions, measurement, and characterization, Ann Arbor, MI Palmer HB, Cullis CF (1965) The formation of carbon from gases. Chem Phys Carbon 1:265 Plee SL, MacDonald JS (1980) Some mechanisms affecting the mass of diesel exhaust particulate collected following a dilution process. SAE Technical Paper 800186. https://doi.org/10.4271/ 800186 Singal SK, Gandhi KK (1996) Diesel fuel quality and particulate emissions. SAE Technical Paper 962480. https://doi.org/10.4271/962480 Smith OI (1981) Fundamentals of soot formation in flames with application to diesel engine particulate emissions. Prog Energy Combust Sci 7:275–291 Wersborg BL, Howard JB, Williams GC (1973) Physical mechanisms in carbon formation in flames. Symp (Int ) Combust 14:929–940. https://doi.org/10.1016/S0082-0784(73)80085-2 Wu W, Samet JM, Ghio AJ, Devlin RB (2001) Activation of the EGF receptor signaling pathway in airway epithelial cells exposed to Utah Valley PM. Am J Physiol Lung Cell Mol Physiol 281:L483–L489. https://doi.org/10.1152/ajplung.2001.281.2.L483

Chapter 5

Numerical Modelling of Soot in Diesel Engines Pavan Prakash Duvvuri, Rajesh Kumar Shrivastava and Sheshadri Sreedhara

Abstract Numerical modelling of soot in diesel engines has evolved over four decades from simple empirical correlations to complex aerosol dynamics and detailed kinetics. These modelling approaches assist in cost-effective diesel engine combustion chamber design to meet the requisite emission legislations. This chapter presents a brief overview on modelling soot in diesel engines. The chapter starts with a description of the physical and chemical processes involved in soot formation, namely gas phase kinetics, nucleation, surface reactions, and coagulation. A brief literature review of the existing modelling techniques for soot formation in diesel engines till date has been presented. The models have been categorized as empirical, phenomenological, and statistical depending on the details of physics and chemistry represented by the models. The uncertainties and the model constants involved in most of these models and the possible effect on resulting soot have been discussed briefly. Considering the inclusion of control on soot particle number in the recent emission legislations, special emphasis has been given to soot models accounting for particle size and number predictions. Some models, both empirical and detailed, have been applied to closed cycle 3D CFD combustion simulations by the authors. The results are compared against published experimental data for crank angle history of soot at varying operating conditions. As part of the detailed soot models, soot particle dynamics has been modeled and coupled with gas phase kinetics. Simulated results have been compared against experimental data for soot particle size distribution at the exhaust of a diesel engine. Keywords Soot models

Diesel engines Soot particle size distribution

P. P. Duvvuri (&) Department of Energy Science and Engineering, Indian Institute of Technology Bombay, Mumbai, India e-mail: [email protected] P. P. Duvvuri R. K. Shrivastava Combustion Research, Cummins Technical Center India, Pune, India S. Sreedhara Department of Mechanical Engineering, Indian Institute of Technology Bombay, Mumbai, India © Springer Nature Singapore Pte Ltd. 2019 A. K. Agarwal et al. (eds.), Engine Exhaust Particulates, Energy, Environment, and Sustainability, https://doi.org/10.1007/978-981-13-3299-9_5

71

72

5.1

P. P. Duvvuri et al.

Introduction

Combustion has been an integral part of human civilization since ages and so has been its unwanted byproduct: Soot. Soot is formed during combustion of hydrocarbons or any carbonaceous matter. It indicates inefﬁcient combustion and is a cause of environmental and health concerns thereby making it a topic of research for diverse ﬁelds like environmental studies, chemical and mechanical engineers, etc. Soot has many adverse health effects and results in a variety of cardiopulmonary issues. Of special concern are the ﬁner particles as these can penetrate cell membranes, enter the blood flow, and even reach the brain (Oberdörster et al. 2004). A very high content of soot in the atmosphere can lead to visibility issues and regional warming due to light scattering. The major sources of soot are industries and transportation, diesel engines being an inherent part of these. Rising numbers of carbon-based fuel usage have resulted in higher production of soot. It can be argued that this can be controlled by the use of renewable and low-carbon alternatives. However, a major part of transportation is powered by crude oil and it will be difﬁcult to expand renewable energy supply to the scale of energy supply from oil. With renewable energy constituting about 1% of the total energy supply, to de-carbonize energy supply, efforts should be focused on improving combustion processes from fossil fuels. Cullen and Allwood (2010) have ranked the conversion devices based on percentage of primary energy usage and diesel engines with 12% top the chart. The signiﬁcant contribution of diesel engines to soot resulted in government intervention and the creation of emission legislations for diesel engines. Figure 5.1 shows the legislative limit history of particulate matter (PM) emissions in the European Union for truck and bus engines and for non-road diesel engines (note the logarithmic y-axis). Almost an exponential decrease in the permissible PM mass can be observed in the last three decades. Until recently, the limits were only set on a mass basis to control the PM concentration. However, soot emitted from diesel engines has a particle distribution with diameters ranging from a few nanometers to a few microns and the mass is constituted mainly by the larger particles which have low concentration numbers. Fig. 5.1 Legislative limit history of diesel engine emissions in the European Union

5 Numerical Modelling of Soot in Diesel Engines

73

The smaller ultraﬁne particles which do not account for much in terms of mass but are large in numbers have a high negative impact on health. Due to this, the upcoming legislative limits concentrate on limiting the particle number (PN) as well. The ﬁrst such legislation is already in place in some countries. In light of these norms, predicting the soot particle number density emitted from diesel engines becomes as important as predicting the mass concentration of soot. As the experimental analysis of soot in diesel engines results in considerably higher lead times and cost, simulation and modelling present an alternative solution. Results from modelling also offer a deep insight into the combustion processes in terms of the spatial and temporal evolution of each chemical species, temperature, turbulent ﬁelds, etc. However, experiments are needed to validate these models before using them. A typical diesel engine combustion chamber optimization using simulations involves calibrating the models with available baseline data and then using these calibrated models to tweak the parameters like chamber geometry, injector geometry, operating parameters, etc. Based on the effect of these parameters, optimization is done either mathematically or by taking educated guesses. Soot modelling in diesel engines has been a topic of research since more than 40 years and advances in this topic have led to considerable improvement in the predictive capability of soot mass. However, very few studies have concentrated on modelling particle statistics and validating the models for soot particle number density and particle size distribution (PSD). The current chapter aims at providing a brief overview of the various numerical models available to predict the in-cylinder soot from diesel engines along with their applicability and limitations. The models have been categorized as empirical, phenomenological, and statistical models. Before deep diving into the soot models, it is important to understand the physical and chemical processes of soot formation namely nucleation, surface growth, oxidation, and coagulation. Hence, an overview of the processes involved in soot formation in diesel engines has also been provided. Selected numerical models (empirical and statistical) have been applied to some experimental cases and the results from these case studies have been discussed for a better understanding. In addition to the in-cylinder soot mass evolution, the results discussed also include soot particle size distributions.

5.2

Modelling the Processes Involved in Soot Formation

Soot is not a very clearly deﬁned entity. Soot from the combustion of hydrocarbons can be considered to be a number of agglomerates, each of them containing a number of primary particles whose core is made up of ﬁne carbonaceous matter. Typical primary particle size varies between 15 and 40 nm in diameter and agglomerate size varies in the range of 60–100 nm (Burtscher 2005). Soot from diesel exhaust is mostly coated with condensed soluble organic fractions and some inorganic content. Although soot and PM are often used interchangeably, soot is a subset of particulate matter. Soot, soluble organic fractions, and inorganic content

74

P. P. Duvvuri et al.

Fig. 5.2 Some processes involved in soot formation. Source Tree and Svensson (2007); reprinted with permission from Elsevier

together constitute particulate matter. Physically, a soot particle can be visualized as a spheroid with layers of structured carbon. Both graphitic (Lee et al. 2003) and fullerene-like microstructures (Su et al. 2004) have been reported for diesel engines. The formation of this structure and the chemical composition depends very much on the ambient environment the soot particle has to go through over a period of time. The material density of soot reported by most of the authors falls around 1.8 g/cc (Burtscher 2005; Tree and Svensson 2007). Soot from diesel engines has a fractal-like structure and a range of fractal dimensions from 1.2 to 3 (Li et al. 2011; Skillas et al. 1998) have been reported by various authors. Extensive experimental methods have been used to characterize the physics and chemistry of these particles, details of which can be found in some review articles (Burtscher 2005; Maricq 2007; Giechaskiel et al. 2014; Myung et al. 2014; Choi et al. 2014). Soot formation involves complex physical and chemical processes which can be explained using chemical kinetics and aerosol dynamics to be occurring in the processes mentioned below, some of which have been represented in Fig. 5.2. Each of these processes and the modelling techniques have been discussed in brief. • • • • •

Precursor formation (Sect. 2.1) Nucleation and condensation (Sect. 2.2) Surface growth (Sect. 2.3) Coagulation (Sect. 2.4) Oxidation (Sect. 2.5)

5.2.1

Precursor Formation

Soot precursors are the chemical species which are the building blocks for soot. During fuel pyrolysis, smaller hydrocarbons are formed which in turn form the required soot precursors. The chemical reaction kinetic scheme taking place during combustion forms an inherent part in soot modelling as it serves as a prelude to the formation of the required soot precursor. The combustion chemistry for any hydrocarbon is devised after a lot of experiments in which the formation and consumption of each chemical species are identiﬁed. The inclusion of the relatively lesser produced soot precursors as compared to major species like OH, O2, CO, etc.,

5 Numerical Modelling of Soot in Diesel Engines

75

increases the stiffness of the numerical problem. This kinetic scheme is hence reduced to decrease the computational cost, the reduction strategies for which can be found in the review by Lu and Law (2009). The kinetic scheme is reduced considering the importance of certain species and reactions by validating against experimentally determined ignition delay, flame velocity and species mass fraction. With the increasing importance of soot and unburnt hydrocarbons, and with the advent of higher computational power, more number of reactions are being employed in diesel engine combustion simulations to ensure a good representation of soot precursors. Among the various chemical species like polyynes, neutral radicals, ionic species, polycyclic aromatic hydrocarbons (PAH), etc. that have been proposed to be possible precursors for soot formation, PAH are accepted widely to be the most probable soot precursors (Karataş and Gülder 2012; Mansurov 2005; Richter and Howard 2000). The formation of the smaller aromatics (naphthalene and benzene) is believed to be the rate controlling step for soot formation (McEnally et al. 2006; Glassman 1988). The formation of the ﬁrst aromatic ring has been a topic of debate, the history of which can be found in the review by Richter and Howard (2000). Depending on the fuel structures, certain reactions might contribute more than the rest during the ﬁrst ring formation. Reactions (5.1) to (5.3) show some important pathways leading to the ﬁrst aromatic ring. Acetylene exists in high concentrations in rich mixtures and favors the ﬁrst two reactions. As intermediate radical species are highly reactive, their concentration is generally much lower compared to that of the major reactants. Due to this, reactions involving two radical reactants are generally insigniﬁcant. Although radical–radical reactions are generally insigniﬁcant due to their inherent instability, reaction (5.3) is favored because the propargyl radicals are stabilized due to resonance. ð5:1Þ ð5:2Þ ð5:3Þ The higher ring formation reactions can happen either due to the famous hydrogen abstraction carbon addition (HACA) reaction (5.4) (Frenklach and Wang 1994) or due to direct aromatic combination (5.5). The direct combination path has been found to be important for aromatic fuels (Frenklach et al. 1988) and relaxes to the HACA mechanism aided by the benzene molecule decomposition to acetylene. However, in the case of diesel engines, the combustion duration is so low that the latter reaction might become important. Also important is the fact that diesel has

76

P. P. Duvvuri et al.

some inherent aromatic content and the aromatic combination path becomes a signiﬁcant contributor for aromatic fuels.

ð5:4Þ

ð5:5Þ Some other reactions which might lead to second ring formation have been shown in reactions (5.6) and (5.7). Both the reactions are again resonant stabilized radical-radical reactions like reaction (5.3). ð5:6Þ ð5:7Þ Based on these reactions, it can be understood that a number of possible pathways can exist for PAH formation and it is still an area of active research. Most of the PAH and fuel oxidation kinetic schemes used in diesel engines originate from the research of laboratory flames. Including so many reactions in a kinetic scheme for diesel engine combustion simulations is taxing on the computational time. Hence, either a smaller hydrocarbon is used to represent the soot precursor or the kinetic schemes are reduced to have lower number of species and reactions. The earliest soot models for diesel engines considered liquid and vapor fuel to be the soot precursor (Hiroyasu and Kadota 1976). In later studies, based on the suggestion of Tesner et al. (1971), a generic intermediate species was used by some authors for diesel engines (Mehta et al. 1988; Zellat et al. 2005). After that, acetylene has been used as a soot precursor and till date remains the most widely used soot precursor in diesel engines (Belardini et al. 1996; Liu et al. 2005; Bolla et al. 2014). This is because the kinetic scheme required to produce acetylene is significantly smaller as compared to that required to produce PAH species. Using acetylene as a soot precursor is a good assumption as acetylene plays a central role in all the higher PAH formation reactions and in surface reactions. With the advent of increased computational power, there has been a rise in soot models considering PAH to be the precursors in diesel combustion. A variety of PAH species like benzene (Pang et al. 2011), naphthalene (Tao et al. 2004), pyrene (Vishwanathan and Reitz 2010), acenapthylene (Golovitchev et al. 2000), acepyrene (Karlsson et al. 1998), coronene (Puduppakkam et al. 2017), and corannulene (Durán et al. 2004) (shown in Fig. 5.3) have been used by different authors.

5 Numerical Modelling of Soot in Diesel Engines

77

Fig. 5.3 Typical PAH soot precursors

In general, the PAH kinetic schemes for diesel engine simulations are reduced from detailed PAH chemistry developed based on smaller hydrocarbons in laboratory flames like those in references (Frenklach and Warnatz 1987; Appel et al. 2000; Wang and Frenklach 1997; Slavinskaya et al. 2012). Pitsch et al. (1995) and Karlsson et al. (1998) simulated soot for diesel engine conditions using a detailed PAH kinetic scheme developed by Mauss (1997) following the approach of Frenklach and Warnatz (1987). The PAH chemistry included the formation of the ﬁrst ring through the reactions (5.1) to (5.3). Using the HACA mechanism for ring growth shown in reaction (5.4), Golovitchev et al. (2000) developed a PAH chemistry with acenapthylene (A2R5) as the precursor for use in diesel combustion simulations. The reaction sequence has been later used by Kikusato et al. (2014) to develop a soot model. This kinetic scheme was further reduced by Chen et al. (2009) for easier implementation in diesel engine simulations. Tao et al. (2004) developed a soot model with naphthalene as the precursor, for which the PAH chemistry was taken from Wang and Frenklach (1997). This kinetic scheme was further reduced by Pang et al. (2011) and only the ﬁrst ring formation was considered. Vishwanathan and Reitz (2010) used a reduced kinetic scheme for diesel combustion containing PAH till pyrene developed by Xi and Zhong (2006) based on the PAH chemistry of Wang and Frenklach (1997). In a later work, Vishwanathan and Reitz (2015) reduced the PAH kinetic scheme of the famous Appel Bockhorn Frenklach (ABF) mechanism (Appel et al. 2000) and found that soot is more sensitive to lower hydrocarbons than the PAH species. Wang et al. (2013) reduced the detailed PAH chemistry proposed by Slavinskaya et al. (2012) and applied to diesel and oxygenated fuel soot simulations.

78

5.2.2

P. P. Duvvuri et al.

Nucleation and Condensation

Nucleation is the process by which the ﬁrst soot particles form from gas phase precursors. Nucleation remains the least understood process among all the soot processes. During nucleation, the two-dimensional planar gas phase PAH form three-dimensional particles. The condensed phase carbon forms an aerosol. The following theories exist for the formation of the ﬁrst soot particles during combustion, a schematic of which has been shown in Fig. 5.4 (Wang 2011). A. Chemical growth to curved structures like fullerene B. Dimerization of precursors to form stacks C. Chemical reactions between precursors to form large cross-linked threedimensional structures During PAH formation by combustion reactions, the concentration of PAH species produced reduces by orders of magnitude with an increase in the number of aromatic rings. For PAH larger than pyrene, the aromatic concentrations are close to the soot concentration itself. This leads to the belief that a soot nucleus is formed by moderately sized PAH species like pyrene or coronene. Happold et al. (2007) carried out mass spectroscopy for ethylene flames and observed periodic peaks in the spectrum, indicating the presence of condensed PAH stacks. The observation of soot being caused by moderately sized PAH and by building blocks negates the theory of nucleation by fullerene-like structures. Violi (2004) used molecular dynamics and kinetic Monte Carlo methods for the evolution of nanoparticles based on chemical reactions. However, this approach cannot account for particle nucleation during low temperatures and low H atom concentrations as the reactions to higher PAH due to HACA would be absent in this case. The dimerization theory on the other hand still holds and can explain the observation of moderately sized PAH

Fig. 5.4 Some existing theories for nucleation of soot. Source Wang (2011); reprinted with permission from Elsevier

5 Numerical Modelling of Soot in Diesel Engines

79

forming soot nucleus, periodic peaks in mass spectroscopy, and nucleation even at a comparatively lower temperature and H atom concentrations. Dimerization theory is also strongly supported by an observation that bimodal soot PSD is better obtained with second-order nucleation rates whereas ﬁrst-order rates only produce a unimodal distribution (Wang 2011). Molecular dynamic simulations by Schuetz and Frenklach (2002) show that pyrene dimer (one of the most widely used soot precursor) has a signiﬁcant lifetime and collision frequency thereby proving the possibility of soot nucleation through pyrene dimerization. The dimer stabilization was attributed to the pattern of very fast energy transfer because of the formation of internal rotors due to the conversion of translational energy to rotational energy when two pyrene molecules approach during collision. Figure 5.5a shows a series of snapshots indicating different orientations of the dimer. Figure 5.5b shows the translational and rotational energies of the pyrene dimer. It can be observed that a peak of one form is closely followed by a peak on another which suggests a consistent energy transformation pattern. Later work (Wong et al. 2009) proved the stability of other PAH dimers ranging from phenanthrene to coronene based on a similar theory. The dimer hence formed creates further three-dimensional clusters. The small size of PAH molecules and the presence of aliphatic bonds during the formation processes can cause rearrangements of atoms resulting in liquid like spherical particles.

Fig. 5.5 Proof for the existence of a stable pyrene dimer in soot. a Snapshots in time showing internal rotations of each monomer; b Translational (dashed line) and rotational (solid line) kinetic energy of both the pyrene molecules. Source Schuetz and Frenklach (2002); reprinted with permission from Elsevier

80

P. P. Duvvuri et al.

Fig. 5.6 Comparison of experimental TEM images of diesel soot (center and right images; Source Mosbach et al. (2009); reprinted with permission from Elsevier) with a computational cluster of 50 coronene molecules (left image; Source Totton et al. (2010); reprinted with permission from Elsevier)

The number of PAH dimers in a cluster can be found out based on experimental investigations of the number of carbon atoms in small soot clusters. Based on molecular dynamic simulations and experimental techniques like spectrometry and electron microscopy, Kubo (2009) found out that for a diesel engine, the PAH molecules grow up to 100 carbon atoms, form stable dimers, and agglomerate to result in nucleating particles that are a few nanometers in diameter. This strongly supports the dimerization theory of nucleation for diesel engines. Totton et al. (2010) simulated the internal structure of a nascent soot particle by minimizing the potential energy content of molecular clusters. Figure 5.6 shows a comparison of their computational result for a cluster of 50 coronene molecules with transmission electron microscopy (TEM) measurements (Mosbach et al. 2009). Nucleation of soot in diesel engines has been traditionally modeled as a graphitization reaction (Bolla et al. 2014; Tao et al. 2004; Vishwanathan and Reitz 2010; Golovitchev et al. 2000) in which the soot precursor is assumed to form solid carbon directly. There is a lack of clear consensus about the number of carbon atoms in the nascent soot particle formed by diesel combustion and hence models use a wide range of nucleating carbon atoms. For example, Hong et al. (2005) used 32 carbon atoms for nucleation from acetylene whereas Bolla et al. (2014) used 100 carbon atoms as proposed by Leung et al. (1991). The nascent soot particles serve as a host for some gas phase species to condense over these particles. Condensation can happen due to the deposition of soot precursor gaseous species or other hydrocarbon species. Detailed soot models (Aubagnac-Karkar et al. 2015; Balthasar et al. 2009) generally represent nucleation through PAH dimerization as a collision between two gas phase precursors and condensation as a collision between a gas phase precursor and a soot particle. Figure 5.7 shows a pictorial depiction of the nucleation and condensation in detailed models. The collision processes of aerosols can be represented by the Smoluchowski equation (Smoluchowski 1917)

5 Numerical Modelling of Soot in Diesel Engines

81

Fig. 5.7 Nucleation and condensation in detailed soot models

Zi

1 Z

0

0

Ni ¼ 0:5 bj;ij Nij Nj dj

bi;j Ni Nj dj

ð5:8Þ

where Ni number of particles of size i bi;j collision frequency for collision between particles of size i and j The ﬁrst term on the right-hand side accounts for the formation of particles of size i by collision between particles of size i − j and j. The rate is halved because the same collision between two particles is accounted for twice in the integral. The second term represents the loss of particles of size i. The collision frequency is calculated from the aerosol theory (Friedlander 1977) depending on the aerosol regime being continuum or free molecular. This, in turn, is deﬁned by the value of Knudsen number given by Kn ¼ k=dp

ð5:9Þ

where k mean free path dp particle diameter The aerosol regime is said to be free molecular for Kn > 10, continuum for Kn < 0.1 and in a transition regime for Kn values in between 0.1 and 10. The collision frequency for these regimes is given by bfm i;j ¼

ﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ 1 sﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃs ﬃ 2 3 6 6kb T 1 1 1=3 1=3 mi þ mj þ 4p qs mi mj

ð5:10Þ

82

P. P. Duvvuri et al.

! Cu kb T 1=3 Cuj i 1=3 ¼2 þ 1=3 mi þ mj 1=3 3l vi vj

bco i;j

ð5:11Þ

where kb T ps vi l Cui fm and co

Boltzmann constant temperature of gas density of soot particle volume of particle i dynamic viscosity of gas Cunningham correction factor for particle i represent free molecular and continuum regimes respectively

Nucleation and condensation are generally modeled as free molecular collisions as the participating particle sizes are very small. In Eq. (5.8), for nucleation, both j and i-j are in the gas phase whereas for condensation, j is in the solid phase and i-j is in the gas phase.

5.2.3

Surface Growth

During surface growth, the particles formed by nucleation undergo surface reactions and keep growing. It should be noted that the number of particles does not change during surface growth. Surface growth occurs at slightly lower temperatures as compared to nucleation as it happens slightly away from the primary reaction zone. The most widely used surface growth mechanism is the celebrated HACA mechanism proposed by Frenklach and Wang (1991) as shown in reactions (5.12) to (5.16). Csoot represents a carbon atom on the edge of the surface and Csoot represents a radical site. In this, acetylene (from pyrolysis reactions) attacks a radical site formed by the hydrogen abstraction and results in cyclization and an aromatic ring addition to the reactive site. Csoot H þ H $ Csoot þ H2

ð5:12Þ

Csoot þ H $ Csoot H

ð5:13Þ

Csoot þ C2 H2 $ Csoot H þ H

ð5:14Þ

Csoot þ O2 $ products

ð5:15Þ

Csoot H þ OH $ products

ð5:16Þ

5 Numerical Modelling of Soot in Diesel Engines

83

The rate of each reaction is given by Rgrowth ¼ kg;s Cg avs Si Ni

ð5:17Þ

where kg,s Cg a vs Si Ni g,s

per site rate coefﬁcient concentration of the gaseous species g fraction of surface sites available for reaction number density of surface sites surface area of particle i number density of particle i represent gas phase and soot surface respectively

The number density of soot sites is estimated based on a quasi-stationary assumption for the sites, i.e., the gas phase reactions happen immediately after the site is formed. Mauss et al. (1994) argued that the third step should be represented by a reverse reaction as at high temperatures, and the surface growth is limited. They added reversibility to these reactions along with an additional step for an intermediate species aiding in ring closure. This mechanism is referred to as the HACA ring closure (HACARC) mechanism. Similar reversible reactions have been suggested by Colket and Hall (1994) to avoid the use of a tunable steric factor (a). a is generally considered a tunable constant and has been subjected to much debate. It is believed to change with temperature and age of the particle. Many co-relations have been proposed for a, some basic forms of which have been shown in Table 5.1. Most of these co-relations are based on simpler laboratory flames. The last correlation in Table 5.1 has been proposed for diesel engines based on the hypothesis that mixture inhomogeneity plays an important role along with aging in determining a and hence an equivalence ratio term has been added. Zhao et al. (2016) also argue that in cases where H atom concentration is low, the hydrocarbon radicals play a decisive role in hydrogen abstraction during surface reactions and hence added a reaction with CH3 radical as in the place of a hydrogen atom in the ﬁrst reaction of the HACA scheme. The molecular orientation plays a very important role in the surface reactions. The availability of a site for reaction is very much dependent on how crowded it is Table 5.1 Co-relations for a Co-relation

Authors

Parameters

tan h (a/log l1 + b)

Appel et al. (2000)

0.004 exp(10,800/T) a + b exp(−CAp)

Xu and Faeth (2001) Singh et al. (2005)

(6974.6/T2a ) exp(−88.06/Ta) a + b exp(−CAp) + dU

Veshkini et al. (2015) Zhao et al. (2016)

a, b = constants l1= ﬁrst moment of soot PSD T = Temperature C = decay time inverse Ap= particle age Ta= thermal age d = constant U=Equivalence ratio

84

P. P. Duvvuri et al.

Fig. 5.8 Types of functional sites on a PAH molecule. Source Mosbach et al. (2009); reprinted with permission from Elsevier

or what kind of forces act nearby it. Figure 5.8 shows a large PAH molecule with different sites. Among these sites, as can be seen the bay sites would be the least reactive due to crowding. Based on kinetic Monte Carlo simulations, Frenklach (1996) found that instead of the expected arm-chair sites from HACA, more zig-zag sites were produced. He resolved this by proposing the formation of a ﬁve-membered ring that acts as a surface nuclei for further surface growth. After the last step in the HACA mechanism, a lot of stabilization reactions might happen (Blanquart and Pitsch 2009). In simpler soot models in diesel engines, surface growth is modeled as a global reaction as compared to a number of reactions constituting the HACA scheme. Most of the simpler models (Bolla et al. 2014; Vishwanathan and Reitz 2010) follow the global reaction proposed by Leung et al. (1991) in which the rate is considered proportional to the square root of the area of soot particles to account for the aging of particles. Detailed models (Hong et al. 2005; Aubagnac-Karkar et al. 2015; Balthasar et al. 2009) account for surface reactions using mechanisms similar to that of HACA (Mauss et al. 1994; Colket and Hall 1994; D’Anna 2008).

5.2.4

Coagulation

Coagulation can be due to coalescence in which particles collide to form another particle or due to agglomeration in which similar particles group together as shown in Fig. 5.2. At a free molecular level, coagulation is expected to result in coalescence and the surface area and volume of the resulting coagulated particle are obtained by conserving the total volume of the resulting sphere. For continuum regime, coagulation is expected to cause agglomeration and the volume and surface area of the resulting particle are obtained by conserving surface area. Coagulation decreases the number of particles and increases the particle size and hence results in a large variation in the PSD over time. In general, due to the large variation in particle number density and size, logarithmic axes are used for the PSD. Traditionally, a unimodal lognormal distribution is used to quantify the PSD. However, it should be noted that nucleation and agglomeration if present simultaneously can cause a bimodal distribution. In fact, it is highly possible that an

5 Numerical Modelling of Soot in Diesel Engines

85

actual peak for small particles of less than 5 nm size is not captured due to experimental limitations of measuring small particles. Coagulation can be modeled using the Smoluchowski equation shown in Eq. (5.8) (Smoluchowski 1917). For coagulation, both i and i − j in Eq. (5.8) represent particle phase. In general, for diesel soot models, it is difﬁcult to model the difference between coalescence and agglomeration and hence most of the models are based on a coalescence approximation, i.e., two particles add up to form a single particle keeping the volume conserved. The particle size range of soot formed typically lies between 1 nm to a few microns. This range falls in the transition regime between free molecular and continuum for most laboratory flames at low pressures. The transition collision frequencies are difﬁcult to determine from fundamentals and hence empirical relations are used to calculate this. The most used relation is an assumption that the transition collision frequency (btr) is the harmonic mean of the collision frequencies of free molecular (bfm) and continuum (bco) regimes. This was proposed by Pratsinis (1988) for generic aerosol and was applied to soot model for laminar premixed flames by Kazakov and Frenklach (1998). btr ¼

bfm bco bfm þ bco

ð5:18Þ

As a proof that this is a valid assumption, the collision frequency for all the regimes (Eq. (5.10) for free molecular, Eq. (5.11) for continuum and Eq. (5.18) for transition) is presented and a close agreement is typically cited as shown in Fig. 5.9. However, the plot only shows collision between particles of same size. If collision between different sizes are considered, the difference between the harmonic assumption and other curves becomes wider as shown in Fig. 5.10. The harmonic assumption might be valid for laboratory flames but becomes arguable in the case of diesel engines, where due to higher in-cylinder pressures Fig. 5.9 Collision frequency for equal sized soot particles at 100 bar, 1800 K

86

P. P. Duvvuri et al.

Fig. 5.10 Collision frequency for a 25 nm particle colliding with different particle sizes at 100 bar, 1800 K

(of up to 180 bar), most of the particles lie in the continuum regime as can be seen in Fig. 5.10. In the continuum regime (Kn < 0.1), the harmonic approximation (solid line) differs signiﬁcantly with the continuum collision frequency (dashed line), the latter being an actual representation as the collision is in the continuum regime. Based on this argument, Duvvuri et al. (2018) enhanced the transition collision frequency by a multiplication factor of 100 and obtained a good agreement between experimental and simulated size distributions. Simpler models for coagulation of soot in diesel engine account for the coagulation rate (Rcoag) as a global reaction nCðsÞ ! Cn ðsÞ

ð5:19Þ

with the rate being given as a simpliﬁed version of Eq. (5.8) by assuming a monodisperse population of particles. Rcoag ¼ 1=2bN 2

ð5:20Þ

Detailed soot models solve for the particle interactions in Eq. (5.8). Earlier soot models (Karlsson et al. 1998; Pitsch et al. 1995; Fusco et al. 1994) considered free molecular collision for coagulation while the later ones improved this to a transition regime for which the collision frequency was given by the harmonic assumption (Hong et al. 2005; Aubagnac-Karkar et al. 2015; Kazakov and Foster 1998).

5 Numerical Modelling of Soot in Diesel Engines

5.2.5

87

Oxidation

While undergoing all the aforementioned processes, the soot precursors, particles and agglomerates are subjected to oxidation reactions simultaneously. For diesel engines, oxidation becomes important especially during late cycle (i.e., after combustion and before the exhaust valve opening) as during this time, the plume spreads out due to mixing and the temperatures are favorable for oxidation. Oxidation is a surface reaction and mostly happens due to molecular oxygen; OH and O radicals. Oxidation is enhanced by turbulent mixing. A detailed description about oxidation of soot can be found in the review by Stanmore et al. (2001). One of the earliest oxidation models proposed is the Nagle Strickland-Constable (NSC) model (Nagle and Strickland-Constable 1962) named after the authors. The NSC model accounts for soot oxidation with molecular oxygen and involves a set of reactions for two types of reaction sites, the more reactive sites (A) and the lesser reactive ones (B). A þ O2 ! A þ 2CO

ð5:21Þ

B þ O2 ! A þ 2CO

ð5:22Þ

A!B

ð5:23Þ

A steady-state assumption is made to calculate the more reactive surface fraction and the overall reaction rate is given by Rox ¼

k A pO 2 x þ k B pO 2 ð 1 x Þ 1 þ k Z pO 2

ð5:24Þ

where kA, kB, kZ rate constants pO 2 partial pressure of oxygen x fraction of surface occupied by site A The NSC oxidation model was derived based on graphite oxidation which involves fewer edge site carbon atoms as compared to that of a more amorphous and heterogeneous soot produced from combustion of carbonaceous fuels. Due to this, correction factors for the NSC oxidation rates have been proposed by some authors based on experimentally observed diesel soot nanostructures (Ladommatos et al. 2002; Song et al. 2012). A simpliﬁed oxidation formulation was given by Lee et al. (1962) with a global reaction for soot oxidation (5.25) which was later employed by Leung et al. (1991) for non-premixed flames.

88

P. P. Duvvuri et al.

CðsÞ þ 0:5O2 ! CO

ð5:25Þ

Fenimore and Jones (1967) found that about 10% of the collision of soot particles with the OH radicals could remove a carbon atom in flames (26). Neoh et al.(1981) found that at higher temperatures and for rich mixtures, the oxidation reaction with the OH radicals is more important than that with the oxygen molecules and proposed collision efﬁciency values for the reaction. CðsÞ þ OH ! CO þ 0:5H2

ð5:26Þ

The radical O has also been found to play a role in oxidation but is an order of magnitude lesser than that of oxygen and the hydroxyl radicals. As mixing plays an important role in oxidation, simpler computational models for oxidation which depend more on kinetics and in which turbulent flow or flame structure is not well resolved, need to account for mixing also. Garo et al. (1994) compared different timescales for diesel combustion and concluded that the oxidation will be influenced by turbulence whereas nucleation and coalescence might not be influenced. Magnussen and Hjertager (1977) proposed a model which accounts for mixing in turbulent combustion. The oxidation rate, given by Eqs. (5.27) to (5.29), is derived from turbulence theory and takes into account the dissipation and kinetic energy of a turbulent flow. The ﬁrst equation represents a condition when the soot concentration is lower than oxygen concentration and the second equation represents a condition when the soot concentration is higher than the oxygen concentration. In the second case, soot has to compete with unburnt fuel for oxygen availability and hence the denominator has terms corresponding to the oxidation of both. Rox;1 ¼ A cs ð=kÞ kg=m3 =s cO 2 cs r s Rox;2 ¼ A k cs r s þ cf r f rs Rox ¼ min Rox;1 ; Rox2

ð5:27Þ ð5:28Þ ð5:29Þ

where A c R k f, s, O2

constant local mean concentration stoichiometric oxygen requirement dissipation rate turbulent kinetic energy represent fuel, soot and oxygen respectively

Although some earlier soot models proposed for diesel engines (Gorokhovski and Borghi 1993) use the formulation for oxidation by Lee et al. (1962), the NSC oxidation model (Nagle and Strickland-Constable 1962) and the OH oxidation

5 Numerical Modelling of Soot in Diesel Engines

89

model by Neoh et al. (1981) have been the most widely used simpliﬁed oxidation models. In general, both of these are used together to account for soot oxidation in diesel engine combustion simulations (Vishwanathan and Reitz 2010; Kikusato et al. 2014; Tao et al. 2009). The mixing model (Magnussen and Hjertager 1977), although used in the past for diesel engines, has seen a decreased application recently as the turbulent structures are better resolved by detailed flow solvers and kinetic soot oxidation models are preferred. In detailed soot models for diesel engines, oxidation is modeled along with surface growth, both of which form a set of kinetic reactions for surface reactions as shown in reactions (5.12) to (5.16).

5.3

Soot Models for Diesel Engines

Soot modelling in diesel engines is a complex procedure as compared to simpler laboratory flames due to the presence of transient nature, spray, turbulence, moving grids, asymmetry, etc. Numerical modelling of soot in diesel engines has evolved from simple empirical relations to detailed kinetics coupled with particle dynamics. Although there are techniques available which use detailed chemistry and particle dynamics to model soot in simpler laboratory flames, the problem that a diesel engine combustion analyst faces while adapting these techniques is to minimize the numerical effort required for such models and still have reasonable accuracy. This section discusses in brief the soot models that have been proposed till date for application to combustion simulations in diesel engines. These models have been categorized as empirical, phenomenological, and statistical models. The statistical models have been further classiﬁed as those based on method of moments (MOM), sectional methods and Monte Carlo methods. Some particular models of each of these types have been discussed in detail and signiﬁcant contribution by others has been mentioned in brief. A review about soot models in generic flames can be found in (Kennedy 1997) and for diesel engines can be found in (Xi and Zhong 2006; Omidvarborna et al. 2015). As this chapter details the soot models applicable in combustion simulations, data-based soot models have not been included. However, a brief mention is needed for these due to a recent rise in their usage (Bocci and Rambaldi 2011; Finesso et al. 2014; Grahn et al. 2012; Ghazikhani and Mirzaii 2011; Tauzia et al. 2017). Data-based soot models rely on co-relation of soot produced with other easily measurable parameters like load, speed, air fuel ratio, intake temperature, intake pressure, etc. Although the data-based models provide a very quick estimate of soot, the data is generally engine speciﬁc and so the models are in general not applicable to other engines. As they do not incorporate the inherent soot processes, data-based models should be avoided in multidimensional combustion simulations. These models can be used for real-time control and onboard diagnostics or for time-saving zero-dimensional or multizone models.

90

5.3.1

P. P. Duvvuri et al.

Empirical Models

Diesel engine soot models which are based on empirical rate forms for some of the soot processes explained in Sect. 5.2 have been characterized as empirical models. The earliest soot models for diesel engines were proposed by Khan et al. (1973) and Hiroyasu et al. (1976). Khan et al. (1973) expressed soot formation in an Arrhenius rate form depending on the equivalence ratio and neglected soot oxidation. Hiroyasu’s soot model (Hiroyasu and Kadota 1976) considered both formation and oxidation in the form of Arrhenius rates. The net soot rate at a particular time was obtained by subtracting the rate of soot oxidation from that of soot formation as shown in Eq. (5.30). Although the initial formulation had a soot formation rate based on both liquid and vapor fuel as precursors, this was changed to vapor fuel only in a later work (Nishida and Hiroyasu 1989). dms dmsf dmso ¼ dt dt dt dmsf Esf 0:5 ¼ Asf mfg P exp dt RT dmso Eso 1:8 ¼ Aso ms P exp dt RT

ð5:30Þ ð5:31Þ ð5:32Þ

where A E m P T s, sf, so, fg

tunable constant activation energy mass gas pressure gas temperature represent net soot, soot formed, soot oxidized, and gaseous fuel, respectively

Until recently, modiﬁed versions of this model have been the most widely used soot models due to their simplicity and low computational cost. Modiﬁcations include using NSC oxidation rates (Patterson et al. 1994; Kong et al. 1995), or mixing based oxidation rates (Hou and Abraham 1995; Chan and Cheng 2007), replacing fuel vapor precursor with acetylene (Sun and Reitz 2006; Kong et al. 2007; Vishwanathan and Reitz 2008). Another empirical model based on the formulation of Tesner et al. (1971) saw a rise in diesel engine application during the early years of combustion simulations in diesel engines (Mehta et al. 1988; Zellat et al. 2005; Nakakita et al. 1990). The model is slightly more detailed than that of Hiroyasu’s model and is based on depicting soot formation as a chain type kinetic process with radical nuclei which are formed due to branching and termination. The formation rate for soot nuclei and number density is given by

5 Numerical Modelling of Soot in Diesel Engines

91

dn ¼ n þ ðf gÞn g Nn dt

ð5:33Þ

dN ¼ ða bN Þn dt

ð5:34Þ

where n n f g g N a, b

net concentration of nuclei forming radicals rate of production of radicals linear chain branching coefﬁcient linear chain breaking coefﬁcient quadratic chain breaking coefﬁcient Number density of soot particles constants

The radicals are produced by a ﬁrst-order decomposition of fuel. This approach coupled with the mixing model for oxidation (Magnussen and Hjertager 1977) was applied to diesel engine soot simulation by some authors (Mehta et al. 1988; Zellat et al. 2005; Nakakita et al. 1990). The model being elementary in nature assumes a ﬁxed particle size which is not valid for diesel engines and hence could not be trusted for particle number density predictions. Boulanger et al. (2007) later added a coagulation term to the model thereby eliminating the need for a constant size assumption. After the early applications, this approach did not receive a good enough following in later years. Although the empirical models give directional trends with respect to change in parameters, the models need tuning of multipliers for each operating point. Also, these soot models do not involve all the inherent soot processes and particle dynamics which become important for particle number density predictions. Being one of the earliest and most used models, the Hiroyasu model (Nishida and Hiroyasu 1989) is often used as a reference to quantify the improvement in better predictable models.

5.3.2

Phenomenological Models

Soot models which account for most of the soot processes explained in Sect. 5.2 (nucleation, surface growth, oxidation and coagulation) with the help of simpliﬁed models have been classiﬁed as phenomenological models. These models generally solve for two equations, one for soot mass and the other for number density as shown below dYs ¼ Rnuc þ Rgrowth Rox dt

ð5:35Þ

92

P. P. Duvvuri et al.

dN ¼ Rnuc Rcoag dt

ð5:36Þ

where Ys N R nuc, growth, ox, coag

soot mass fraction Number density of soot particles rate of soot process represent nucleation, growth, oxidation, and coagulation, respectively

Soot mass increases due to nucleation and surface growth and decreases due to oxidation. Soot number density increases due to nucleation and reduces due to coagulation. The reduction in particle number density due to oxidation of nascent particles is generally ignored in phenomenological models. These models have a monodisperse assumption in a computational cell, i.e., the average particle size for all the particles in a computational cell is assumed to be same. Most of the phenomenological models follow the approach formulated by Leung et al. (1991) for soot formation in non-premixed flames. The model was based on global reaction rates for soot nucleation and surface growth by acetylene; oxidation by oxygen; coagulation by square dependence on number density. The global reactions are given by C2 H2 ! 2CðsÞ þ H2

ð5:37Þ

C2 H2 þ nCðsÞ ! ðn þ 2ÞCðsÞ þ H2

ð5:38Þ

CðsÞ þ 1=2O2 ! CO

ð5:39Þ

nCðsÞ ! Cn ðsÞ

ð5:40Þ

A detailed kinetic scheme involving 31 species and 85 reactions was also proposed, which, however, at that time was computationally very expensive for diesel combustion simulations. Belardini et al. (1996) used a global reaction for producing acetylene directly from fuel (along with a global reaction for its oxidation to CO2) and used the acetylene hence produced to apply the formulation of Leung et al. (1991). They compared the soot evolution w.r.t crank angle obtained through simulations and experiments and a good agreement has been reported. Fusco et al. (1994) applied a similar concept to diesel engines by considering nucleation by a generic precursor; surface growth by acetylene, soot oxidation by NSC model (Nagle and Strickland-Constable 1962) and assuming a mono-dispersed particle distribution. The generic precursor was produced by the Shell ignition model (Halstead et al. 1977), a widely used reduced kinetic scheme in diesel combustion simulations. Kazakov and Foster (1998) modiﬁed this model by considering a generic species for surface oxidation as well. Tao et al. (2009) further improved the

5 Numerical Modelling of Soot in Diesel Engines

93

Fig. 5.11 Schematic of the nine-step phenomenological soot model of Tao et al. (2009). Courtesy of Dr. Feng Tao

model by including oxidation of soot by OH and nucleation by acetylene as shown in Fig. 5.11. OH concentration was estimated from a H2–O2–CO2 system by a quasi-steady-state assumption. These changes made the soot model suitable with complex chemical kinetic systems as well as for simpler chemistries like the Shell ignition model. To account for turbulence-chemistry interactions, the soot process rates were multiplied by a factor based on turbulent and chemical timescales. Good experimental validation has been reported for varying operating parameters. Because of the inclusion of details of the soot processes, a number of models based on the work of Leung et al. (1991) and Tao et al. (2009) have been reported in literature (Liu et al. 2005; Bolla et al. 2014; Tao et al. 2004; Vishwanathan and Reitz 2010, 2015; Kikusato et al. 2014; Pang et al. 2015; Sukumaran et al. 2013; Zhao et al. 2017; Cheng et al. 2013; Cai et al. 2016). Detailed kinetic schemes with fuel oxidation, pyrolysis, and soot precursor production reactions are being coupled with phenomenological soot models in the recent years due to the advent of higher computational power. When coupled with a detailed gas phase chemistry, the processes 1, 2, 8, and 9 shown in Fig. 5.11 (acetylene formation from fuel oxidation and pyrolysis, soot precursor formation from acetylene, acetylene oxidation, and soot precursor oxidation respectively) become an integral part of the gas phase kinetic scheme and hence the soot model only has to account for the other processes (nucleation, coagulation and surface reactions). Recent phenomenological soot models have started including better details in the soot models like nucleation through PAH (Vishwanathan and Reitz 2010; Pang et al. 2015) and surface reactions through HACA (Zhao et al. 2017; Cheng et al. 2013).

94

P. P. Duvvuri et al.

Fig. 5.12 In-cylinder soot evolution from a phenomenological soot model. Source Bolla et al. (2014); reprinted with permission from Elsevier

Good qualitative results are generally reported for phenomenological models. One such result comparing simulations and experiments for in-cylinder soot evolution as reported by Bolla et al. (2014) is shown in Fig. 5.12. Although the phenomenological models account for soot particle dynamics and growth in a much better way than the simpler empirical models, they have an underlying monodisperse assumption in each computational cell, due to which they could not be relied upon to represent the intricate statistical nature of soot PSD in diesel engines.

5.3.3

Statistical Models

Soot models including a statistical representation of the soot PSD have been classiﬁed as statistical models. A detailed review about the existing statistical models for particles dispersed in reactive flows can be found in the review by Rigopoulos (2010). Statistical models are generally coupled with detailed chemical kinetics and are the most detailed soot models applied to diesel engines till date. Along with the mass of soot emitted, recent emission legislations also limit the particle number. Due to this, predicting the PSD has become even more important. Statistical techniques are used to obtain the PSD either in an approximate way or by solving for a mean value. The following statistical methods have been used till date for soot modelling in diesel engines

5 Numerical Modelling of Soot in Diesel Engines

95

• Method of moments (MOM) • Sectional method • Monte Carlo methods Although Lagrangian methods can be used as a workaround so that each particle is tracked separately by solving its equations of motion, it is highly unrealistic due to high computational requirements. With the current computational power, this is realizable only when the system being considered is very small such that it has very low number of particles.

5.3.3.1

Method of Moments

In many engineering systems, knowing the entire PSD is not necessary and the knowledge of some average parameters is sufﬁcient enough. For example, for a soot PSD, knowing the soot mass and particle number density might be enough from an emission legislation point of view. The moments of a statistical distribution are such averaged entities given by Mr ¼

1 X

mri Ni

ð5:41Þ

i¼1

where Mr rth moment of the distribution mi mass of particles of size i Ni number density of particles of size i If the curve represents the distribution of particle number density versus mass, the zeroth moment (M0) would represent the total soot number density, the ﬁrst moment (M1) would represent the soot mass, and the second moment (M2) represents the deviation from the average mass. The moments can be negative or fractional as well. For example, the surface area can be represented by M2/3. To understand the solution procedure, let us transform Eq. (5.8) into the moment space. This could be represented by an inﬁnite number of moments given by 1 X 1 dM0 1 X ¼ b Ni Nj 2 i¼1 j¼1 i:j dt

ð5:42Þ

dM1 ¼0 dt

ð5:43Þ

1 X 1 dM2 X ¼ ijbi:j Ni Nj dt i¼1 j¼1

ð5:44Þ

96

P. P. Duvvuri et al. 1 X 1 X dM3 ¼3 ij2 bi:j Ni Nj dt i¼1 j¼1

ð5:45Þ

M1 remains constant as the total mass of the particle phase does not change during coagulation but just gets transferred between particles during their interactions. This transformation can be viewed as a single higher dimensional equation versus inﬁnite lower dimensional equations which are later reduced to three or four equations using certain assumptions. The mathematical difﬁculty lies in the fact that for closure, the right-hand side of each of these equations has to be expressed as a function of integral moments appearing in the left-hand side. This creates a problem as it is difﬁcult to express the source terms directly in terms of the moments. To overcome this, certain assumptions are made for each source term to ensure that they can be expressed in terms of the other moments. Sometimes these approximations lead to the inclusion of fractional or negative moments which have to be calculated even if they might not be required as an end result. For theoretical understanding, only coagulation process has been shown in the above equations. If particle growth is included, the set of equations then become dM0 ¼ N0 C0 dt

ð5:46Þ

dM1 ¼ N1 þ C1 dt

ð5:47Þ

dM2 ¼ N2 þ C2 þ G2 dt

ð5:48Þ

dMr ¼ Nr þ Cr þ Gr dt

ð5:49Þ

where N, C, and G denote nucleation, coagulation, and surface growth terms respectively. It should be noted that the growth term can also render closure impossible if the growth dependence on particle size is anything higher than a linear dependence. For this reason, all the studies using MOM assume either a size independent growth (for free molecular regime) or a linearly dependent growth (for continuous regime). The latter is a valid assumption if the chemical reactions causing particle growth are of the ﬁrst order, which is generally the case with reactions involving radical species. So with the use of detailed chemistry involving the necessary growth species, the modelling framework can still be relied upon. Another issue encountered by the use of MOM is that recovering PSD from the moments is difﬁcult as a continuous function with inﬁnite information cannot be exactly speciﬁed by ﬁnite information contained in a ﬁnite number of moments. A close inspection of the problem suggests that both the issues (closure of equation and difﬁculty in recovering PSD) are related to loss of information and can

5 Numerical Modelling of Soot in Diesel Engines

97

be solved if a parametric estimate of the PSD function is possible. This paved way to the earliest solution method for MOM in which a presumed parametric shape is taken for the size distribution function and the moments are expressed in terms of these parameters. The resulting equations are solved for these parameters thereby obtaining the moments and hence the size distribution. For aerosols, traditionally a lognormal distribution of the form of Eq. (5.50) has been used to deﬁne the size distribution function (Pratsinis 1988). ! ln2 v=vavg 1 1 Nv ¼ pﬃﬃﬃﬃﬃﬃ exp v 18 ln2 r 3 2p ln r

ð5:50Þ

where Nv Number density of particles of volume v vavg average particle volume r standard deviation of the volume distribution The moments for this can be derived to be Mr ¼

M0 vravg exp

9 2 2 r ln r 2

ð5:51Þ

This makes the closure much easier as each moment can be expressed in the form of known parameters. Hong et al. (2005) have used this approach along with detailed chemistry (for the precursor and growth species acetylene) to simulate soot in diesel engines. The ﬁrst three moments have been considered and soot transport is represented by moments as transport scalars as shown in Eq. (5.52). The source term for each process (nucleation, surface reactions, etc.) is calculated based on the theory presented in Sect. 5.2 of this chapter. dMr þ r ðu Mr Þ ¼ rðDs rMr Þ þ S_ r dt

ð5:52Þ

_ r;nuc þ M _ r;coag þ M _ r;growth þ M _ r;ox S_ r ¼ M

ð5:53Þ

where u gas velocity Ds diffusion coefﬁcient of soot S_ r source term for rth moment Although presuming a size distribution function leads towards an easier solution for MOM, in cases without any prior knowledge of the expected shape of the PSD, use of this method is questionable. A unimodal lognormal distribution has been typically used for atmospheric aerosols. Traditional diesel combustion has been known to evolve from a nucleating mode into an agglomeration mode which means during

98

P. P. Duvvuri et al.

this evolution or at the exhaust, it will contain a bimodal distribution. In such a case, either an improved presumption method or an alternative closure technique is more promising. A lot of techniques like Taylor expansion (Frenklach and Harris 1987), quadrature MOM (McGraw 1997), and hybrid MOM (Mueller et al. 2009) have been proposed for univariate (mass based or volume based) or multivariate moments [volume and surface area based (Mueller et al. 2009) or volume, surface area, and reactive hydrogen sites (Blanquart and Pitsch 2009)] for soot modelling of laboratory flames. Till date, the most used solution method for MOM in diesel engines has been the MOM by interpolative closure (MOMIC) (Frenklach and Harris 1987; Frenklach 2002) and hence only this has been discussed in detail here. A detailed explanation about MOMIC can be found in the paper by Frenklach (2002). MOMIC is based on the idea that the fractional moments which render closure difﬁcult can be expressed as a function of integral moments using logarithmic Lagrange interpolation. For example, for the positive fractional moments, the interpolation is represented by log lr ¼ Lr log l0 ; log l1 ; . . .; log lrmax

ð5:54Þ

lr ¼ Mr =M0

ð5:55Þ

where lr reduced rth moment Lr Lagrange interpolation with respect to r Rearranging volume terms in the free molecular collision frequency given by Eq. (5.10) into mass terms and substituting in Eq. (5.44), the following expression can be obtained 1 X 1 X

mi þ mj

1=2

1=2

1=2

mi mj

1=3

mi

1=3

þ mj

2 Ni Nj

ð5:56Þ

i¼1 j¼1

This could have been expressed in independent moments had it not been for the term in square root. If a grid function Fl with integral values for l is deﬁned such that Fl ¼

1 X 1 2 X l 1=2 1=2 1=3 1=3 mi þ mj mi mj mi þ mj Ni Nj

ð5:57Þ

i¼1 j¼1

The Lagrange interpolation based upon the ﬁrst three grid functions produces 3=8

3=4

1=8

F1=2 ¼ F0 F1 F2

ð5:58Þ

The interpolations are done directly in the moment space rather than doing it in the size distribution space and then reverse transforming as this could lead to further

5 Numerical Modelling of Soot in Diesel Engines

99

loss in information. Each moment is generally solved as a transport scalar. Till date, this method has been the most extensively used as a closure method for MOM in diesel engines (Karlsson et al. 1998; Pitsch et al. 1995; Balthasar et al. 2009; Maly et al. 1998; Zhong et al. 2012). Almost all the models based on MOMIC consider nucleation through PAH and surface reactions by mechanisms similar to HACA (Frenklach and Wang 1991). The method has been used coupled with flamelet based combustion models (Karlsson et al. 1998; Pitsch et al. 1995; Balthasar et al. 2009; Nakov et al. 2009; Dederichs et al. 1999), with detailed kinetic solver based combustion models (Puduppakkam et al. 2017; Zhong et al. 2012; An et al. 2016) and even with stochastic reactor models (Yoshihara et al. 1994; Pasternak et al. 2014; Matrisciano et al. 2015). MOM is computationally the least expensive method among the statistical models as only the ﬁrst few moments give most of the required information. The computational advantage comes from the fact that very few transport equations are to be solved (one for each moment) thereby rendering it useful for coupling with large set of equations involved in diesel combustion. MOMIC however has an inherent limitation of loss of information related to the PSD. The size distribution has to be reconstructed based on moment reversal by some postprocessing generally by presuming a shape for the size distribution function. Also, although MOM has been present for about 25 years in diesel engines and good qualitative results have been reported for soot at exhaust, very few validations for in-cylinder evolution of soot particle or number density have been reported for diesel engines. This is due to the fact that even though the least expensive among the statistical methods, it is still expensive and has seen only limited application as compared to the empirical or phenomenological models. Figure 5.13 shows one such experimental validation by Zhong et al. (2012) for the in-cylinder evolution of soot particle size. The presence of a qualitative trend and a quantitative underprediction for simulated particle size as compared to experimental data can be observed.

5.3.3.2

Sectional Method

In sectional methods, the PSD is discretized into a number of bins or sections each representing either a piecewise function similar to the ﬁnite element method or an average size. In the case of an average size assumption, the major challenge is to

Fig. 5.13 Validation of an MOM-based soot model. Source Zhong et al. (2012); reprinted with permission from Taylor and Francis Ltd.

100

P. P. Duvvuri et al.

account for the averaged quantities as the size increase by growth or by coagulation can be anywhere between two neighboring nodes. Another important issue arises in deﬁning these bins or the selection of grid as the PSD has a large range of sizes differing in several orders of magnitude and is also dynamically changing. A uniform grid is hence out of the discussion as this would result in very high computational times. While selecting a binning strategy, care needs to be taken to minimize the discontinuities in the neighboring bins wherever a sudden change happens like in the case of nucleating particles or particle disappearance due to oxidization. One of the ﬁrst methods used by Bleck (1970) for grid selection was to select a grid in geometric progression. Once the grid is deﬁned, a transport equation is solved for soot in each section with the necessary source terms for nucleation, surface reactions, and coagulation. Accuracy is enhanced by higher number of sections and by having a smaller range of bin size such that the volume difference is that of only a few carbon atoms. Particle size change results in the transfer of a particle from one section to the other. Similar to a flow in CFD, the numerical schemes are to be selected based on the problem. For example, to calculate the growth kernel for surface growth as the particles move from a lower bin to higher bin, an up-winding scheme shall be preferred. In cases where oxidization is also to be considered, a central differencing scheme involving two neighboring bins should be preferred. Netzell et al. (2007) used a sectional model for soot modelling in turbulent diffusion flames and deﬁned the sections in a geometrical progression of volume as vi;max ¼ vMIN

vMAX VMIN

i=imax ð5:59Þ

v1;min ¼ vMIN

ð5:60Þ

v1;max ¼ vMIN þ vC2

ð5:61Þ

where imax vi,max vi,min vMIN vMAX vc2

number of sections upper boundary of section i lower boundary of section i volume of the ﬁrst nucleating particle volume of the largest particle volume of two carbon atoms

Netzell et al. (2007) used 100 sections and improved the accuracy by deﬁning a very narrow ﬁrst section involving only a two carbon atom volume so that the nucleation is captured better. Linear assumption is made for the distribution of the conserved scalar (soot volume fraction) in each bin with the slope being proportional to the mean volume fraction of neighboring bins. A transport equation is solved for the soot volume fraction in each section with the source term consisting of an addition of the independent source terms for nucleation, condensation, surface

5 Numerical Modelling of Soot in Diesel Engines

101

reactions, and coagulation. Nucleation, condensation, and coagulation have been modeled as collisional source terms based on the Smoluchowski equation (Smoluchowski 1917) as explained in Sect. 5.2. Surface reactions are modeled by HACARC (Mauss et al. 1994) and the change in soot due to surface reactions is distributed in neighboring sections (i, i + 1 for growth and condensation; i, i − 1 for oxidation). The sectional method for soot by Netzell et al. (2007) has been implemented in diesel engines by Marchal (2008; Marchal et al. 2009), followed by Vervisch Kljakic (2012) and Aubagnac Karkar et al. (2014, 2015). Most of these models are based on flamelet combustion models and tabulated chemistry to reduce computational time. However, tabulation results in a lack of proper coupling of the soot in particle phase and the gaseous species. Till date, the sectional soot model has seen very few applications coupled with detailed kinetics combustion models (Duvvuri et al. 2018; Ravet et al. 2016; Luo et al. 2014) as the computational cost is very high. These models solve for transport of the soot mass fraction in each section with the source term being a summation of source terms from independent soot processes similar to that for moments as in Eq. (5.52). As a detailed chemistry involving PAH species is used for nucleation in most of these models, the number of chemical species to be solved is very high. Adding the number of sections to these is further taxing on the computational power. Vervisch Kljakic (2012) studied the effect of number of sections and found that using 30 sections gives reasonably accurate size distributions for diesel engines. As the sectional method solves for the transport equation of different particle sizes, accounting for size-based transport properties becomes easier. Good agreement for qualitative trends for soot PSD at exhaust, as shown in Fig. 5.14, have been reported (Aubagnac-Karkar et al. 2015) using this method [in the ﬁgure, a phenomenological soot kinetics (PSK) model has also been compared by Aubagnac-Karkar et al. (2015)]. Another sectional method based approach is to model the processes involved in soot formation (nucleation, growth, etc.) using gas phase kinetics and represent particles as a class of molecular mass as in the work of D’Anna and Kent (2008). Fig. 5.14 Experimental vs numerical PSD obtained using a sectional method. Source Aubagnac-Karkar et al. (2015); reprinted with permission from Elsevier

102

P. P. Duvvuri et al.

Due to the large number of species and reactions involved, this approach is rendered computationally very expensive for diesel engine combustion simulations. Till date, such an application to diesel engine has been reported only once (Fraioli et al. 2011). Although some qualitative trends were reproduced in simulations, the high computational times prohibited Fraioli et al. (2011) to run even a sensitivity analysis. Sectional methods involve high computational costs when coupled with multidimensional reactive flows and due to this, their use has been restricted in the case of diesel combustion systems. Chowdhury and Yu (2014) applied a sectional method for diesel engines using a simpler multizone model for combustion to reduce the computational times. However, better results call for detailed simulation of flow and chemistry. Another possible shortcoming in near future might be the further massive increase in computational times with the introduction of other variables in the PSD. A univariate PSD cannot account for all the surface properties or growth chemistry and inclusion of other variables would mean further discretization of the PSD thereby resulting in more equations to be solved.

5.3.3.3

Monte Carlo Methods

Monte Carlo methods can be employed to predict the PSD accurately but at high computational costs. These are stochastic methods relying on random sampling and so for better accuracies, higher number of samples needs to be taken. Not until very recently have been these applied to diesel engines due to their computational costs. The computational modelling group at Cambridge (Mosbach et al. 2009; Etheridge et al. 2011) have used a detailed chemistry stochastic reactor model along with a population balance solver called SWEEP (2006) to simulate soot in combustion engines. The simulations are also capable of predicting the physical properties of primary particles and the atomic composition of each soot particle aggregate along with the types of functional sites available on the PAH molecules. Monte Carlo simulations provide the ability to introduce a number of variables without much increase in computational time. Due to this, properties like soot morphology and number of reactive sites of each type have been simulated in detail for homogeneously charged combustion engines. Lee and Huh (2012) have also used SWEEP along with conditional moment closure to model soot. Figure 5.15 shows the similarity in the trends between experiments and simulations for the in-cylinder evolution of PSD obtained by Mosbach et al. (2009). Monte Carlo based models, till date, have only been applied to homogeneous combustion systems either in laboratory flames or in combustion engines and are yet to see an application for nonhomogeneous combustion systems and hence in conventional diesel combustion.

5 Numerical Modelling of Soot in Diesel Engines

103

Fig. 5.15 Experimental and simulated in-cylinder PSD evolution. Source Mosbach et al. (2009); reprinted with permission from Elsevier

5.4

Case Studies

Some of the soot models presented in the Sect. 5.3 have been applied to practical diesel engine conditions for better understanding. Experimental data from the published literature for an optical engine, a heavy-duty engine and a light-duty engine has been chosen for model validation. Varying operating regimes like high-temperature and low-temperature combustion and partially premixed combustion modes have been chosen for a better validation of the soot models. A similar modelling procedure has been followed for all the cases for the ease of comparison. Closed cycle combustion simulations have been performed on a sector grid using CONVERGE (Richards et al. 2016), a CFD code for reacting multiphase flows. The current work uses adaptive mesh reﬁnement (AMR) based on velocity and temperature gradients across a computational cell. Based on grid sensitivity studies performed by the authors (Duvvuri et al. 2018) and based on the recommendations from literature for similar modelling approach (Senecal et al. 2014), a near nozzle grid size of 0.35 mm has been used. Injection rates have been generated based on in-house one-dimensional codes coupling hydraulic and mechanical linkages for fuel systems. Spray breakup is modeled using a Kelvin Helmholtz (KH) instability based primary breakup and a secondary breakup based on a competition between KH and Rayleigh Taylor (RT) instabilities (Patterson and Reitz 1998). A Weber number based rebound and slide mechanism is used for spray wall interactions (Naber and Reitz 1988). Details of the nonreactive modelling can be found in previous work by the authors (Duvvuri et al. 2018). Combustion has been modeled using the SAGE kinetic solver (Senecal et al. 2003) along with a multizone approach (Babajimopoulos et al. 2005). In the multizone approach, computational cells within a bin size of 5 K temperature and 0.05 equivalence ratio have been clustered together. The use of AMR along with a multizone approach reduces the computational time signiﬁcantly and facilitates the use of detailed chemistry coupled with statistical soot models.

104

P. P. Duvvuri et al.

Table 5.2 Details of experimental engine and operating conditions Details

Optical engine (Singh et al. 2005)

Heavy duty (Shen 2016)

Light duty (Zhang et al. 2014)

Engine Bore/Stroke (mm) Compression ratio Spray Angle (deg) Number of holes Nozzle dia (mm) IMEP (bar) Intake O2 % Speed (rpm) Inj. pressure (bar)

Cummins N14 139.7/152.4 11.2:1 152 8 0.196 *4 12.6 1200 1600

Scania D13 130/160 15 120 8 0.179 6 14 1200 1200

GM 1.9 82/90.4 16.7 144 6 0.154 8 21 2300 950

A kinetic scheme consisting of 121 species and 593 reactions has been used (Luo et al. 2014). The kinetic scheme besides having fuel oxidation and pyrolysis reactions also has PAH kinetic scheme by Mauss (1997) embedded in it. In this, the ﬁrst aromatic ring forms due to the reactions (5.1) to (5.3) (i.e., through both C3 and C4 routes) and the higher aromatics are formed by the HACA reaction (Frenklach and Wang 1994) as in reaction (5.4). Three soot models namely a modiﬁed Hiroyasu soot model, an MOM-based model and a sectional soot model have been applied for the experimental cases. As all these models have been explained in detail in Sect. 5.3, only a brief overview is being provided here. The modiﬁed Hiroyasu soot model considers acetylene as the soot precursor and NSC oxidation rates similar to references (Sun and Reitz 2006; Kong et al. 2007; Vishwanathan and Reitz 2008). MOM is based on interpolative closure and is similar to the approach of Karlsson et al. (1998). The sectional method is based on the soot model by Marchal (2008). For the detailed soot models, nucleation happens through acepyrene precursor (A4R5), surface growth, and oxidation through the HACARC mechanism (Mauss et al. 1994). Smoluchowski equation [Eq. (5.8)] (Smoluchowski 1917) is used for coagulation with the harmonic assumption for transition regime collision as in Eq. (5.18). Table 5.2 shows some details of the experimental engines and the operating conditions. More details can be found in the original references (Singh et al. 2005; Shen 2016; Zhang et al. 2014).

5.4.1

Optical Engine

Experimental data from the Cummins N14 optical engine at Sandia national laboratories (Singh et al. 2005, 2006; Genzale et al. 2009) has been used for model validations in the current study. Singh et al. (2005) measured the in-cylinder spatial and temporal distribution of soot for varying operating conditions among which a high-temperature and a low-temperature conditions have been taken for the current

5 Numerical Modelling of Soot in Diesel Engines

105

Fig. 5.16 Computational grid during spray (CA = 7 deg ATDC)

study. The high-temperature condition represents a typical diesel combustion scenario with a short ignition delay and the low-temperature condition represents a high exhaust gas recirculation (EGR), high ignition delay, late injection [0 degree after top dead center (ATDC)] scenario. Figure 5.16 shows the computational grid during spray at three perpendicular axes. The boundaries have been reﬁned and the spray region is reﬁned due to the velocity gradients resulting in cell addition by AMR. In a previous work by the authors (Duvvuri et al. 2018), a good agreement was obtained between experiments and simulations for liquid, vapor spray and for mixture distribution for the low-temperature case. The mixture distribution for this engine at various crank angles and for different spray angles has been measured by Genzale et al. (2009) for a non-combusting case using planar laser-induced fluorescence (PLIF) of toluene tracer in fuel. Figure 5.17 shows a comparison of the mixture distribution obtained from simulations with the PLIF images from Genzale et al. (2009). A good agreement for the nonreactive modelling part proved to be a good starting point for modelling the reactive part. Figure 5.18 shows the in-cylinder pressure and heat release rate (HRR) obtained for the two cases considered. A good agreement can be observed which provides conﬁdence in the fuel oxidation and pyrolysis reactions, i.e., till the starting point of soot simulations. Soot for these two cases has been simulated using two approaches: with a modiﬁed Hiroyasu model and with an MOM-based model. The important model constants involved in both the models have been varied to check the sensitivity. For the Hiroyasu model, formation and oxidation rate multipliers [Asf in Eq. (5.31) and Aso in Eq. (5.32)] were changed between acceptable limits. For MOM, the fraction of surface sites, a has been changed. The role of a in soot modelling and its proposed co-relations have already been discussed in Sect. 2.3. Figure 5.19 shows the sensitivity of in-cylinder soot evolution towards these model parameters. It can be observed that the Hiroyasu model is very sensitive with respect to the formation and oxidation rate multipliers. This can be expected as the model is empirical in nature and has been made for the ease of tuning based on speciﬁc

106

P. P. Duvvuri et al.

Fig. 5.17 Mixture distribution from experiment (Genzale et al. 2009) and simulation (Duvvuri et al. 2018) (CA = 12 deg ATDC)

Fig. 5.18 Experimental (Singh et al. 2005) and simulated in-cylinder pressure and HRR

Fig. 5.19 Sensitivity of model constants for the Hiroyasu model (a) and MOM (b)

5 Numerical Modelling of Soot in Diesel Engines

107

Fig. 5.20 Experimental (Singh et al. 2005) and simulated in-cylinder soot using Hiroyasu model and MOM

operating conditions. Although this might seem like a welcoming feature as it provides a freedom of choice of model parameters to get an exceptional agreement with the experimental data, in cases with a lack of experimental data for in-cylinder evolution, same output at the exhaust can be obtained with multiple pairs of formation and oxidation multipliers. This becomes important because generally the soot model parameters are tuned based on an agreement of soot mass obtained at the exhaust valve opening (EVO) from simulations with the soot mass at exhaust obtained from experiments. Soot obtained from MOM is not as sensitive as that from the Hiroyasu soot model. As a is a multiplier in surface reactions which play a signiﬁcant role during the late cycle, the soot mass is sensitive to a only during the late cycle. Although a is a multiplier for both growth and oxidation reactions, in this case, soot mass reduction by oxidation seems to be more prominent than soot mass addition due to surface growth. A lower sensitivity to a shows that the soot mass is probably less dependent on a and more dependent on the gas phase concentrations obtained from the kinetic scheme. This is a welcome feature as the MOM-based soot model is more reliable than the Hiroyasu soot model for in-cylinder evolution in cases where the experimental data is scarce to establish the formation and oxidation multipliers for the latter. Based on the sensitivity study, the best suited model constants have been used to compare the soot mass obtained from both the models with experimental data as shown in Fig. 5.20. A good agreement can be observed between experimental data and simulations. However, it should be noted that for such an agreement, the Hiroyasu model needs considerable tuning of the formation and oxidation rate multipliers as compared to the minimal tuning required for an MOM-based soot model. A good prediction of the temporal evolution of soot generally follows a good prediction of the in-cylinder spatial distribution. Figure 5.21 shows a comparison of the spatial distribution of soot obtained at different crank angles from simulations using MOM (Duvvuri et al. 2018) and from experiments using two-color

108

P. P. Duvvuri et al.

Fig. 5.21 Experimental (Singh et al. 2006) and simulated (Duvvuri et al. 2018) spatial distribution of soot using MOM

thermometry by Singh et al. (2005, 2006). The high-temperature case forms soot nearby the spray similar to that in typical diesel spray diffusion flames whereas the low-temperature case forms soot nearby the walls due to a premixed charge and a start of combustion after the end of injection. The simulations are able to capture the soot structures at each crank angle.

5.4.2

Heavy-Duty and Light-Duty Engine

A similar modelling strategy as for the optical engine has been applied to the heavy-duty and the light-duty engines shown in Table 5.2. The heavy-duty engine operating condition represents a partially premixed combustion whereas that of the light-duty engine represents a typical high-temperature combustion. For these engines, as experimental measurements for PSD at exhaust was available, a sectional method has been applied for validation. A good agreement for in-cylinder pressure and HRR was obtained similar to that for the optical engine. Based on a sensitivity study performed by the authors (Duvvuri et al. 2019) and based on similar observations by Vervisch Kljakic (2012), 30 sections have been used in the current simulations. Figure 5.22 shows the in-cylinder source term evolution from each soot process obtained from simulations for the light-duty engine. Nucleation and condensation happen as long as the suitable gas phase PAH species are present along with high temperatures, i.e., during the main combustion event. Surface reactions exist till late in the cycle as they can happen at lower temperatures than that during the main combustion event and due to the presence of the requisite gas phase species, i.e., acetylene for growth and oxygen and OH for oxidation. As oxygen is present in abundance compared to acetylene, oxidation source term has a higher signiﬁcance during a much later part of the cycle compared to surface growth. Note that the source term for oxidation is negative as the soot

5 Numerical Modelling of Soot in Diesel Engines

109

Fig. 5.22 Soot process source term evolution for the light-duty engine

Fig. 5.23 Evolution of PSD for the light-duty engine

mass reduces due to oxidation. As soot volume fraction of each section is solved as a transport scalar in the sectional model, the evolution of in-cylinder soot PSD can be monitored. Figure 5.23 shows the same obtained from simulations of the light-duty engine at different crank angle values. The PSD starts evolving from nucleation (CA = 0) to growth in size due to condensation and surface growth (up to CA = 15) and then the number of particles reduces due to coagulation (up to CA = 110) and the average particle size reduces due to oxidation (up to CA = 110). The PSD at EVO obtained from simulations for the light-duty engine and the heavy-duty engine has been shown against experimental data (Shen 2016; Zhang et al. 2014) in Fig. 5.24. Simulated PSD has been normalized for a better comparison. Good qualitative trends can be observed. The simulated data mimics the spread out distribution for the light-duty engine with two peaks as compared to a unimodal distribution for the heavy-duty engine. This establishes conﬁdence in the predictive capability of the sectional method for soot modelling.

110

P. P. Duvvuri et al.

Fig. 5.24 Comparison of PSD at exhaust from experiments (Shen 2016; Zhang et al. 2014) and simulations

5.5

Summary

In the past 40 years, diesel engine soot models have evolved from simple empirical co-relations to detailed statistical models coupled with precursor chemistry. The understanding of the soot processes has improved over time thanks to the extensive studies on laboratory flames. However, much remains to be known still, the most important among them being the formation of the ﬁrst soot particle. A review about the existing modeling approaches for soot processes in general and diesel engines in particular has been provided with special emphasis on models capable of predicting the particle number density and size distribution considering the recent emission norms. The models have been classiﬁed into empirical, phenomenological and statistical soot models. Empirical models are computationally lucrative due to their simplicity along with an ability to provide directional trends for soot mass and hence have been the most widely used for over thirty years. Phenomenological models account for most of the soot processes and can be used for predicting qualitative trends for soot mass and in some cases, directional trends for particle number density. Among the statistical models, the method of moments is computationally least expensive and is capable of predicting soot mass and particle number density. The sectional method is the most detailed soot model applied to diesel combustion till date and can predict particle size distribution. Monte Carlo based methods, although applied to homogeneous combustion cases, are yet to see an application in conventional diesel combustion. Some case studies have been presented to validate the soot models in diesel engines. Three soot models have been applied to varying conditions in three engines. Good qualitative trends have been obtained for soot mass versus crank angle. The Hiroyasu model was found to be highly sensitive to the model constants as compared to the method of moments. The sectional soot model was able to predict the particle size distribution at exhaust.

5 Numerical Modelling of Soot in Diesel Engines

5.5.1

111

Way Ahead

Methods need to be developed to reduce the computational cost for the detailed soot models. As the phenomenological models are a step closer towards lower computational costs, good validations are needed for their predictability of number density as most of these models have been validated only for soot mass. For statistical soot models, lower computational costs can be accommodated by reducing the large kinetic schemes while ensuring they represent varying combustion regimes. Another important area which needs to be investigated in detail is the validity of assumptions of the soot models for diesel engines as most of these are borrowed from laboratory flames which operate at lower ambient pressure and are comparatively steady in nature. For example, the difference originating from a change in aerosol regime, mixing time scale, wall interaction can play a major role in soot formation. Finally, more detailed validations comparing simulations with experimental data (in the form of in-cylinder soot mass, number density, and particle size distribution evolution; spatial distribution of in-cylinder soot) are needed for the statistical soot models to establish conﬁdence in their predictability. Acknowledgements The authors have beneﬁted from the support provided from Cummins Inc. especially from Dr. Sujith Sukumaran, combustion research, Cummins Inc.; Dr. Anuradda Ganesh, director, research, innovation, and compliance, Cummins India Limited and Dr. John Deur, director, combustion research, Cummins Inc. The authors would like to thank Elsevier (Fig. 2, 4, 5, 6, 8, 12, 14 and 15 in the chapter) and Taylor and Francis Ltd. (Figure 13 in the chapter) for providing permissions to use ﬁgures. The authors are grateful to Dr. Feng Tao, combustion research, Cummins Inc. for providing his nine-step phenomenological soot model ﬁgure (Fig. 11 in this chapter). The authors are also thankful to Dr. Mengqin Shen and Prof. Martin Tuner, Department of Energy Sciences, Lund University and Dr. Yizhou Zhang, Prof. Jaal Ghandhi and Prof. David Rothamer, direct-injection engine research consortium, university of Wisconsin for sharing experimental data.

References An YZ, Li X, Teng SP, Wang K, Pei YQ, Qin J, Zhao H (2016) Development of a soot particle model with PAHs as precursors through simulations and experiments. Fuel 179:246–257. https://doi.org/10.1016/j.fuel.2016.03.100 Appel J, Bockhorn H, Frenklach M (2000) Kinetic modeling of soot formation with detailed chemistry and physics: Laminar premixed flames of C2 hydrocarbons. Combust Flame 121:122–136. https://doi.org/10.1016/S0010-2180(99)00135-2 Aubagnac-Karkar D, Michel JB, Colin O, Vervisch-Kljakic PE, Darabiha N (2015) Sectional soot model coupled to tabulated chemistry for diesel RANS simulations. Combust Flame 162:3081– 3099. https://doi.org/10.1016/j.combustﬂame.2015.03.005 Aubagnac-karkar D, Michel J, Colin O, Noël L, Darabiha N (2014) A sectional soot model for RANS simulation of diesel engines. SAE Tech Pap 2014–01–1590. https://doi.org/10.4271/ 2014-01-1590

112

P. P. Duvvuri et al.

Babajimopoulos A, Assanis DN, Flowers DL, Aceves SM, Hessel RP (2005) A fully coupled computational fluid dynamics and multi-zone model with detailed chemical kinetics for the simulation of premixed charge compression ignition engines. Int J Engine Res 6:497–512. https://doi.org/10.1243/146808705X30503 Balthasar M, Eismark J, Magnusson I (2009) Soot modelling in heavy-duty diesel engines. In: Bockhorn H, D’Anna A, Saroﬁm AF, Wang H (eds) Combustion generated ﬁne carbonaceous particles, Proceedings of an international workshop held in Villa Orlandi, Anacapri, 13–16 May 2007, pp. 663–688. KIT scientiﬁc publishing, Karlsruhe Belardini P, Bertoli C, Beatrice C, Giacomo ND (1996) Application of a reduced kinetic model for soot formation and burnout in three-dimensional diesel combustion computations. Symp Combust 26:2517–2524. https://doi.org/10.1016/S0082-0784(96)80083-X Blanquart G, Pitsch H (2009) A joint volume-surface-hydrogen multi-variate model for soot formation. In: Bockhorn H, D’Anna A., Saroﬁm AF, Wang H (eds) Combustion generated ﬁne carbonaceous particles, Proceedings of an international workshop held in Villa Orlandi, Anacapri, 13–16 May 2007. KIT scientiﬁc publishing, Karlsruhe, pp 437–463 Bleck R (1970) A fast, approximative method for integrating the stochastic coalescence equation. J Geophys Res 75:5165–5171. https://doi.org/10.1029/JC075i027p05165 Bocci E, Rambaldi L (2011) Soot emission modelization of a diesel engine from experimental data. In: Murgante B, Gervasi O, Iglesias A, Taniar D, Apduhan BO (eds) Computational science and its applications—ICCSA 2011. Springer, Berlin, pp 316–327 Bolla M, Farrace D, Wright YM, Boulouchos K (2014) Modelling of soot formation in a heavy-duty diesel engine with conditional moment closure. Fuel 117:309–325. https://doi.org/ 10.1016/j.fuel.2013.09.041 Boulanger J, Liu F, Neill WS, Smallwood GJ (2007) An improved soot formation model for 3D diesel engine simulations. J Eng Gas Turbines Power 129:877–884. https://doi.org/10.1115/1. 2718234 Burtscher H (2005) Physical characterization of particulate emissions from diesel engines: a review. J Aerosol Sci 36:896–932. https://doi.org/10.1016/j.jaerosci.2004.12.001 Cai G, Yen M, Abraham J (2016) On formulating a simpliﬁed soot model for diesel and biodiesel combustion. Chem Eng Sci 144:249–259. https://doi.org/10.1016/j.ces.2016.01.039 Chan TL, Cheng XB (2007) Numerical modeling and experimental study of combustion and soot formation in a direct injection diesel engine. Energy Fuels 21:1483–1492. https://doi.org/10. 1021/ef0605201 Chen W, Shuai S, Wang J (2009) A soot formation embedded reduced reaction mechanism for diesel surrogate fuel. Fuel 88:1927–1936. https://doi.org/10.1016/j.fuel.2009.03.039 Cheng X, Chen L, Yan F, Dong S (2013) Study on soot formation characteristics in the diesel combustion process based on an improved detailed soot model. Energy Convers Manag 75: 1–10. https://doi.org/10.1016/j.enconman.2013.05.033 Choi S, Myung CL, Park S (2014) Review on characterization of nano-particle emissions and PM morphology from internal combustion engine: part 2. Int J Automot Technol 15:219–227. https://doi.org/10.1007/s12239-014-0023-9 Chowdhury GM, Yu F (2014) Modeling soot size distribution evolution and pollutant formation inside diesel engine using a 0D multi-zone gas parcel model with detailed chemistry and soot microphysics. SAE Tech Pap 2014–01–1603. https://doi.org/10.4271/2014-01-1603 Colket MB, Hall RJ (1994) Successes and uncertainties in modeling soot formation in laminar, premixed flames. In: Bockhorn H (ed) Soot formation in combustion. Mechanisms and models. Springer-Verlag, pp 442–468 Cullen JM, Allwood JM (2010) The efﬁcient use of energy: tracing the global flow of energy from fuel to service. Energy Policy 38:75–81. https://doi.org/10.1016/j.enpol.2009.08.054 Dederichs AS, Balthasar M, Mauss F (1999) Modeling of NOx and soot formation in diesel combustion. Oil Gas Sci Technol 54:245–249. https://doi.org/10.2516/ogst:1999021 Durán A, Carmona M, Monteagudo JM (2004) Modelling soot and SOF emissions from a diesel engine. Chemosphere 56:209–225. https://doi.org/10.1016/j.chemosphere.2004.03.008

5 Numerical Modelling of Soot in Diesel Engines

113

Duvvuri PP, Shrivastava RK, Sreedhara S (2018) Non-reactive flow in-cylinder flow simulation for a diesel engine. In: Massarotti N, Nithiarasu P, Dutta P, Ranganyakalu C (eds) 5th international conference on computational methods for thermal problems, THERMACOMP2018, Bangalore Duvvuri PP, Shrivastava RK, Sreedhara S (2018) Diesel engine soot simulations using method of moments. In: Proceedings of the 7th international and 45th national conference on fluid mechanics and fluid power (FMFP), Mumbai, India Duvvuri PP, Sukumaran S, Shrivastava RK, Sreedhara S (2019) Modeling soot particle size distribution in diesel engines Duvvuri PP, Sukumaran S, Shrivastava RK, Sreedhara S (2018) Modeling the effect of parametric variations on soot particle size distribution in a diesel engine. In: Proceedings of the ASME 2018 internal combustion fall technical conference, San Diego D’Anna A (2008) Detailed kinetic modeling of particulate formation in rich premixed flames of ethylene. Energy Fuels 22:1610–1619. https://doi.org/10.1021/ef700641u D’Anna A, Kent JH (2008) A model of particulate and species formation applied to laminar, nonpremixed flames for three aliphatic-hydrocarbon fuels. Combust Flame 152:573–587. https://doi.org/10.1016/j.combustﬂame.2007.08.003 Etheridge J, Mosbach S, Kraft M, Wu H, Collings N (2011) Modelling soot formation in a DISI engine. Proc Combust Inst 33:3159–3167. https://doi.org/10.1016/j.proci.2010.07.039 Fenimore CP, Jones GW (1967) Oxidation of soot by hydroxyl radicals. J Phys Chem 71:593–597. https://doi.org/10.1021/j100862a021 Finesso R, Misul D, Spessa E (2014) Development and validation of a semi-empirical model for the estimation of particulate matter in diesel engines. Energy Convers Manag 84:374–389. https://doi.org/10.1016/j.enconman.2014.04.053 Fraioli V, Beatrice C, Lazzaro M (2011) Soot particle size modelling in 3D simulations of diesel engine combustion. Combust Theory Model 15:863–892. https://doi.org/10.1080/13647830. 2011.578662 Frenklach M (1996) On surface growth mechanism of soot particles. Symp Combust 26:2285– 2293. https://doi.org/10.1016/S0082-0784(96)80056-7 Frenklach M (2002) Method of moments with interpolative closure. Chem Eng Sci 57:2229–2239. https://doi.org/10.1016/S0009-2509(02)00113-6 Frenklach M, Clary DW, Gardiner WC, Stein SE (1988) Effect of fuel structure on pathways to soot. Symp Combust 21:1067–1076. https://doi.org/10.1016/S0082-0784(88)80337-0 Frenklach M, Harris SJ (1987) Aerosol dynamics modeling using the method of moments. J Colloid Interface Sci 118:252–261. https://doi.org/10.1016/0021-9797(87)90454-1 Frenklach M, Wang H (1991) Detailed modeling of soot particle nucleation and growth. Symp. Combust. 23:1559–1566. https://doi.org/10.1016/S0082-0784(06)80426-1 Frenklach M, Warnatz J (1987) Detailed modeling of PAH proﬁles in a sooting low-pressure acetylene flame. Combust Sci Technol 51:265–283. https://doi.org/10.1080/ 00102208708960325 Frenklach M, Wang H (1994) Detailed mechanism and modeling of soot particle formation. In: Bockhorn H (ed) Soot formation in combustion, mechanisms and models. Springer-Verlag, pp 165–192 Friedlander SK (1977) Smoke, dust and haze: fundamentals of aerosol behavior. Wiley, New York Fusco A, Knox-Kelecy AL, Foster DE (1994) Application of a phenomenological soot model for diesel engine combustion. In: The international symposium on diagnostics and modeling of combustion in internal combustion engines (COMODIA), pp 571–576 Garo A, Said R, Borghi R (1994) Model of soot formation: coupling of turbulence and soot chemistry. In: Bockhorn H (ed) Soot formation in combustion, mechanisms and models. Springer-Verlag, pp 527–548 Genzale CL, Reitz RD, Musculus MPB (2009) Optical diagnostics and multi-dimensional modeling of spray targeting effects in late-injection low-temperature diesel combustion. SAE Int J Engines 2:150–172. https://doi.org/10.4271/2009-01-2699

114

P. P. Duvvuri et al.

Ghazikhani M, Mirzaii I (2011) Soot emission prediction of a waste-gated turbo-charged DI diesel engine using artiﬁcial neural network. Neural Comput Appl 20:303–308. https://doi.org/10. 1007/s00521-010-0500-7 Giechaskiel B, Maricq M, Ntziachristos L, Dardiotis C, Wang X, Axmann H, Bergmann A, Schindler W (2014) Review of motor vehicle particulate emissions sampling and measurement: from smoke and ﬁlter mass to particle number. J Aerosol Sci 67:48–86. https://doi.org/10. 1016/j.jaerosci.2013.09.003 Glassman I (1988) Soot formation in combustion processes. Symp Combust 22:295–311. https:// doi.org/10.1016/S0082-0784(89)80036-0 Golovitchev VI, Nordin N, Jarnicki R, Chomiak J (2000) 3-D diesel spray simulations using a new detailed chemistry turbulent combustion model. SAE Tech Pap 2000–01–1891. https://doi.org/ 10.4271/2000-01-1891 Gorokhovski M, Borghi R (1993) Numerical simulation of soot formation and oxidation in diesel engines. SAE Tech Pap 930075. https://doi.org/10.4271/930075 Grahn M, Johansson K, Vartia C, McKelvey T (2012) A structure and calibration method for data-driven modeling of NOX and soot emissions from a diesel engine. SAE Tech Pap 2012– 01–0355. https://doi.org/10.4271/2012-01-0355 Halstead MP, Kirsch LJ, Quinn CP (1977) The autoignition of hydrocarbon fuels at high tempertures and pressures—ﬁtting of a mathematical model. Combust Flame 30. https://doi. org/10.1016/0010-2180(77)90050-5 Happold J, Grotheer H-H, Aigner M (2007) Soot precursors consisting of stacked pericondensed PAHs. In: Bockhorn H, D’Anna A, Saroﬁm AF, Wang H (eds) Combustion generated ﬁne carbonaceous particles, Proceedings of an international workshop held in Villa Orlandi, Anacapri, 13–16 May 2007, pp. 277–288. KIT scientiﬁc publishing, Karlsruhe Hiroyasu H, Kadota T (1976) Models for combustion and formation of nitric oxide and soot in direct injection diesel engines. SAE Tech Pap 760129. https://doi.org/10.4271/760129 Hong S, Wooldridge MS, Im HG, Assanis DN, Pitsch H (2005) Development and application of a comprehensive soot model for 3D CFD reacting flow studies in a diesel engine. Combust Flame 143:11–26. https://doi.org/10.1016/j.combustﬂame.2005.04.007 Hou Z, Abraham J (1995) Three-dimensional modeling of soot and NO in a direct-injection diesel engine. SAE Tech Pap 950608. https://doi.org/10.4271/950608 Karataş AE, Gülder ÖL (2012) Soot formation in high pressure laminar diffusion flames. Prog Energy Combust Sci 38:818–845. https://doi.org/10.1016/j.pecs.2012.04.003 Karlsson A, Magnusson I, Balthasar M, Mauss F (1998) Simulation of soot formation under diesel engine conditions using a detailed kinetic soot model. SAE Tech Pap 981022. https://doi.org/ 10.4271/981022 Kazakov A, Frenklach M (1998) Dynamic modeling of soot particle coagulation and aggregation: implementation with the method of moments and application to high-pressure laminar premixed flames. Combust Flame 114:484–501. https://doi.org/10.1016/S0010-2180(97) 00322-2 Kazakov A, Foster DE (1998) Modeling of soot formation during DI diesel combustion using a multi-step phenomenological model. SAE Tech Pap 982463. https://doi.org/10.4271/982463 Kennedy IM (1997) Models of soot formation and oxidation. Prog Energy Combust Sci 23:95– 132. https://doi.org/10.1016/S0360-1285(97)00007-5 Khan IM, Greeves G, Wang CHT (1973) Factors affecting smoke and gaseous emissions from direct injection engines and a method of calculation. SAE Tech Pap 730169:687–709. https:// doi.org/10.4271/730169 Kikusato A, Kogo K, Zhou B, Jin K, Daisho Y, Sato K, Fujimoto H, Terashima H, Morii Y (2014) Numerical simulation on soot formation in diesel combustion by using a CFD code combined with a parallelized explicit ODE solver. SAE Tech Pap 2014–01–2567. https://doi.org/10. 4271/2014-01-2567 Kong S-C, Sun Y, Rietz RD (2007) Modeling diesel spray flame liftoff, sooting tendency, and NOx emissions using detailed chemistry with phenomenological soot model. J Eng Gas Turbines Power 129:245–251. https://doi.org/10.1115/1.2181596

5 Numerical Modelling of Soot in Diesel Engines

115

Kong S-C, Han Z, Reitz RD (1995) The development and application of a diesel ignition and combustion model for multidimensional engine simulation. SAE Tech Pap 950278. https://doi. org/10.4271/950278 Kubo S (2009) In-cylinder soot nanoparticle formation mechanism. In: Bockhorn H, D’Anna A, Saroﬁm AF, Wang H (eds) Combustion generated ﬁne carbonaceous particles, Proceedings of an international workshop held in Villa Orlandi, Anacapri, 13–16 May 2007, pp. 651–662. KIT scientiﬁc publishing, Karlsruhe Ladommatos N, Song H, Zhao H (2002) Measurements and predictions of diesel soot oxidation rates. In: Proceedings of the Institution of Mechanical Engineers, Part D: J. Automobile Eng., pp 677–689 Lee KB, Thring MW, Beér JM (1962) On the rate of combustion of soot in a laminar soot flame. Combust Flame 6:137–145. https://doi.org/10.1016/0010-2180(62)90082-2 Lee J, Huh KY (2012) Prediction of number density, volume fraction and size distribution of soot particles in a diesel engine under various operating conditions. SAE Tech Pap (2012) https:// doi.org/10.4271/2012-01-0136 Lee KO, Zhu J, Ciatti S, Yozgatligil A, Choi MY (2003) Sizes, graphitic structures and fractal geometry of light-duty diesel engine particulates. SAE Tech Pap. https://doi.org/10.4271/200301-3169 Leung KM, Lindstedt RP, Jones WP (1991) A simpliﬁed reaction mechanism for soot formation in nonpremixed flames. Combust Flame 87:289–305. https://doi.org/10.1016/0010-2180(91) 90114-Q Li Z, Song C, Song J, Lv G, Dong S, Zhao Z (2011) Evolution of the nanostructure, fractal dimension and size of in-cylinder soot during diesel combustion process. Combust Flame 158:1624–1630. https://doi.org/10.1016/j.combustﬂame.2010.12.006 Liu Y, Tao F, Foster DE, Reitz RD (2005) Application of A multiple-step phenomenological soot model to HSDI diesel multiple injection modeling. SAE Tech Pap 2005-01-09. https://doi.org/ 10.4271/2005-01-0924 Lu T, Law CK (2009) Toward accommodating realistic fuel chemistry in large-scale computations. Prog Energy Combust Sci 35:192–215. https://doi.org/10.1016/j.pecs.2008.10.002 Luo Z, Raju M, Senecal PK, Lehtiniemi H, Perlman C, Mauss F, Shethaji T, Li Y (2014) Efﬁcient implementation of detailed soot models in internal combustion engine simulations. In: 2014 Spring technical meeting of the Central states section of the Combustion Institute (2014) Magnussen BF, Hjertager BH (1977) On mathematical modeling of turbulent combustion with special emphasis on soot formation and combustion. Symp Combust 16:719–729. https://doi. org/10.1016/S0082-0784(77)80366-4 Maly RR, Stapf P, Konig G (1998) Progress in soot modelling for engines. In: The international symposium on diagnostics and modeling of combustion in internal combustion engines (COMODIA), pp 25–34, Kyoto, Japan Mansurov ZA (2005) Soot formation in combustion processes (review). Combust Explos Shock Waves 41:727–744. https://doi.org/10.1007/s10573-005-0083-2 Marchal C (2008) Modelisation de la formation et de l’oxydation des suies dans un moteur automobile. Ph.D. thesis, Université d’Orléans Marchal C, Moréac G, Vovelle C, Mounaïm-Rousselle C, Mauss F (2009) Soot modelling in automotive engines. In: Proceedings of ECM 2009, 4th European combustion meeting, Vienna, Austria Maricq M (2007) Chemical characterization of particulate emissions from diesel engines: a review. J Aerosol Sci 38:1079–1118. https://doi.org/10.1016/j.jaerosci.2007.08.001 Matrisciano A, Borg A, Perlman C, Lehtiniemi H, Pasternak M, Mauss F (2015) Soot source term tabulation strategy for diesel engine simulations with SRM. SAE Tech Pap 2015–24–2400 Published (2015). https://doi.org/10.4271/2015-24-2400 Mauss F (1997) Entwicklung eines kinetischen Modells der Russbildung mit schneller polymerisation. Ph.D. thesis, RWTH Aachen

116

P. P. Duvvuri et al.

Mauss F, Trilken B, Breitbach H, Peters N (1994) Soot formation in partially premixed diffusion flames at atmospheric pressure. In: Bockhorn H (ed) Soot formation in combustion, mechanisms and models. Springer-Verlag, pp 325–349 McEnally CS, Pfefferle LD, Atakan B, Kohse-Höinghaus K (2006) Studies of aromatic hydrocarbon formation mechanisms in flames: Progress towards closing the fuel gap. Prog Energy Combust Sci 32:247–294. https://doi.org/10.1016/j.pecs.2005.11.003 McGraw R (1997) Description of aerosol dynamics by the quadrature method of moments. Aerosol Sci Technol 27:255–265. https://doi.org/10.1080/02786829708965471 Mehta PS, Gupta AK, Gupta CP (1988) Model for prediction of incylinder and exhaust soot emissions from direct injection diesel engines. SAE Tech Pap 881251. https://doi.org/10.4271/ 881251 Mosbach S, Celnik MS, Raj A, Kraft M, Zhang HR, Kubo S, Kim KO (2009) Towards a detailed soot model for internal combustion engines. Combust Flame 156:1156–1165. https://doi.org/ 10.1016/j.combustﬂame.2009.01.003 Mueller ME, Blanquart G, Pitsch H (2009a) Hybrid method of moments for modeling soot formation and growth. Combust Flame 156:1143–1155. https://doi.org/10.1016/j. combustﬂame.2009.01.025 Mueller ME, Blanquart G, Pitsch H (2009) A joint volume-surface model of soot aggregation with the method of moments. Proc Combust Inst. https://doi.org/10.1016/j.proci.2008.06.207 Myung CL, Ko A, Park S (2014) Review on characterization of nano-particle emissions and PM morphology from internal combustion engine: part 1. Int J Automot Technol 15:203–218. https://doi.org/10.1007/s12239-014-0022-x Naber JD, Reitz RD (1988) Modeling Engine spray/wall impingement. SAE Tech Pap 880107. https://doi.org/10.4271/880107 Nagle J, Strickland-Constable RF (1962) Oxidation of carbon between 1000–2000 C. In: Proceedings of the ﬁfth carbon conference, pp 154–164 Nakakita K, Nagaoka M, Fujikawa T, Ohsawa K, Yamaguchi S (1990) Photographic and three dimensional numerical studies of diesel soot formation process. SAE Tech Pap 902081. https:// doi.org/10.4271/902081 Nakov G, Mauss F, Wenzel P, Steiner R, Krüger C, Zhang Y, Rawat R, Borg A, Perlman C, Fröjd K, Lehtiniemi H (2009) Soot simulation under diesel engine conditions using a flamelet approach. SAE Tech Pap 2, 2009–01–2679. https://doi.org/10.4271/2009-01-2679 Neoh KG, Howard J, Saroﬁm AF (1981) Soot oxidation in flames. In: Siegla DC, Smith GW (eds) Particulate carbon, formation during combustion. Plenum press, New York, pp 261–277 Netzell K, Lehtiniemi H, Mauss, F (2007) Calculating the soot particle size distribution function in turbulent diffusion flames using a sectional method. Proc Combust Inst 31(I):667–674. https:// doi.org/10.1016/j.proci.2006.08.081 Nishida K, Hiroyasu H (1989) Simpliﬁed three-dimensional modeling of mixture formation and combustion in a D.I. diesel engine. SAE Tech Pap 890269. https://doi.org/10.4271/890269 Oberdörster G, Sharp Z, Atudorei V, Elder A, Gelein R, Kreyling W, Cox C (2004) Translocation of inhaled ultraﬁne particles to the brain. Inhal Toxicol 16:437–445. https://doi.org/10.1080/ 08958370490439597 Omidvarborna H, Kumar A, Kim DS (2015) Recent studies on soot modeling for diesel combustion. Renew Sustain Energy Rev 48:635–647. https://doi.org/10.1016/j.rser.2015.04. 019 Pang KM, Jangi M, Bai XS, Schramm J (2015) Evaluation and optimisation of phenomenological multi-step soot model for spray combustion under diesel engine-like operating conditions. Combust Theory Model 19:279–308. https://doi.org/10.1080/13647830.2015.1019929 Pang KM, Ng HK, Gan S (2011) Development of an integrated reduced fuel oxidation and soot precursor formation mechanism for CFD simulations of diesel combustion. Fuel 90:2902– 2914. https://doi.org/10.1016/j.fuel.2011.04.027 Pasternak M, Mauss F, Perlman C, Lehtiniemi H (2014) Aspects of 0D and 3D modeling of soot formation for diesel engines. Combust Sci Technol 186:1517–1535. https://doi.org/10.1080/ 00102202.2014.935213

5 Numerical Modelling of Soot in Diesel Engines

117

Patterson MA, Reitz RD (1998) Modeling the effects of fuel spray characteristics on diesel engine combustion and emission. SAE Tech Pap 980131. https://doi.org/10.4271/980131 Patterson MA, Kong S-C, Hampson GJ, Reitz RD (1994) Modeling the effects of fuel injection characteristics on diesel engine soot and NOx emissions. SAE Tech Pap 940523. https://doi. org/10.4271/940523 Pitsch H, Wan YP, Peters N (1995) Numerical investigation of soot formation and oxidation under diesel engine conditions. SAE Tech Pap 952357. https://doi.org/10.4271/952357 Pratsinis SE (1988) Simultaneous nucleation, condensation, and coagulation in aerosol reactors. J Colloid Interface Sci 124:416–427. https://doi.org/10.1016/0021-9797(88)90180-4 Puduppakkam K, Modak A, Naik C (2017) Uniﬁed soot model for predicting soot particle numbers in CFD simulations of GDI and diesel engines. In: International multidimensional engine modeling user’s group meeting, Detroit Ravet F, Dutfoy L, Rathinam B, Lehtiniemi H, Seidel L, Netzer C, Mauss F (2016) Soot modeling with Particle sectional model (PSM) in diesel engine. Results and discussion. In: Proceedings of THIESEL 2016: conference on thermo- and fluid dynamic processes in direct injection engines Richards K, Senecal P, Pomraning E (2016) CONVERGE (v2.3). Convergent Science Inc., Madison, WI Richter H, Howard J (2000) Formation of polycyclic aromatic hydrocarbons and their growth to soot—a review of chemical reaction pathways. Prog Energy Combust Sci 26:565–608. https:// doi.org/10.1016/S0360-1285(00)00009-5 Rigopoulos S (2010) Population balance modelling of polydispersed particles in reactive flows. Prog Energy Combust Sci 36:412–443. https://doi.org/10.1016/j.pecs.2009.12.001 SWEEP2: Cambridge soot simulator. Cambridge (2006) Schuetz CA, Frenklach M (2002) Nucleation of soot: molecular dynamics simulations of pyrene dimerization. Proc Combust Inst 29:2307–2314. https://doi.org/10.1016/S1540-7489(02) 80281-4 Senecal PK, Pomraning E, Anders JW, Weber MR, Gehrke CR, Polonowski CJ, Mueller CJ (2014) Predictions of transient flame lift-off length with comparison to single-cylinder optical engine experiments. J Eng Gas Turbines Power 136:111505. https://doi.org/10.1115/1. 4027653 Senecal PK, Pomraning E, Richards KJ, Briggs TE, Choi CY, Mcdavid RM. Patterson MA (2003) Multi-dimensional modeling of direct-injection diesel spray liquid length and flame lift-off length using CFD and parallel detailed chemistry. SAE Tech Pap 2003–01–1043. https://doi. org/10.4271/2003-01-1043 Shen M (2016) Particulate matter emissions from partially premixed combustion with diesel, gasoline and ethanol. Ph.D. thesis, Lund University Singh J, Balthasar M, Kraft M, Wagner W (2005a) Stochastic modeling of soot particle size and age distributions in laminar premixed flames. Proc Combust Inst 30:1457–1464. https://doi. org/10.1016/j.proci.2004.08.120 Singh S, Reitz RD, Musculus MPB (2005) 2-color thermometry experiments and high-speed imaging of multi-mode diesel engine combustion. SAE Tech Pap 2005–01–3842. https://doi. org/10.4271/2005-01-3842 Singh S, Reitz RD, Musculus MPB (2006) Comparison of the Characteristic time (CTC), Representative interactive flamelet (RIF), and Direct integration with detailed chemistry combustion models against optical diagnostic data for multi-mode combustion in a heavy-duty DI diesel engine. SAE Tech Pap 2006-01-00. https://doi.org/10.4271/2006-01-0055 Skillas G, Künzel S, Burtscher H, Baltensperger U, Siegmann K (1998) High fractal-like dimension of diesel soot agglomerates. J Aerosol Sci 29:411–419. https://doi.org/10.1016/ S0021-8502(97)00448-5 Slavinskaya NA, Riedel U, Dworkin SB, Thomson MJ (2012) Detailed numerical modeling of PAH formation and growth in non-premixed ethylene and ethane flames. Combust Flame 159:979–995. https://doi.org/10.1016/j.combustﬂame.2011.10.005

118

P. P. Duvvuri et al.

Smoluchowski (1917) Versuch einer mathematischen Theorie der Koagulationskinetik kolloiderLösnngen. Kolloid-Zeitschrift. 21, 98–104 (1917). https://doi.org/10.1007/ bf01427232 Song J, Song C, Lv G, Wang L, Bin F (2012) Modiﬁcation to Nagle/Strickland-Constable model with consideration of soot nanostructure effects. Combust Theory Model 16:639–649. https:// doi.org/10.1080/13647830.2011.645883 Stanmore BR, Brilhac JF, Gilot P (2001) The oxidation of soot: a review of experiments, mechanisms and models. Carbon 39:2247–2268. https://doi.org/10.1016/S0008-6223(01) 00109-9 Su DS, Müller J-O, Jentoft RE, Rothe D, Jacob E, Schlögl R (2004) Fullerene-like soot from EuroIV diesel engine: consequences for catalytic automotive pollution control. Top Catal 30 (31):241–245. https://doi.org/10.1023/B:TOCA.0000029756.50941.02 Sukumaran S, Van Huynh C, Kong SC (2013) Numerical modeling of soot emissions in diesel sprays based on detailed fuel and PAH chemistry. Combust Sci Technol 185:1696–1714. https://doi.org/10.1080/00102202.2013.831085 Sun Y, Reitz RD (2006) Modeling diesel engine NOx and soot reduction with optimized two-stage combustion. SAE Tech Pap 2006-01-0027. https://doi.org/10.4271/2006-01-0027 Tao F, Golovitchev VI, Chomiak J (2004) A phenomenological model for the prediction of soot formation in diesel spray combustion. Combust Flame 136:270–282. https://doi.org/10.1016/j. combustﬂame.2003.11.001 Tao F, Reitz RD, Foster DE, Liu Y (2009) Nine-step phenomenological diesel soot model validated over a wide range of engine conditions. Int J Therm Sci 48:1223–1234. https://doi. org/10.1016/j.ijthermalsci.2008.08.014 Tauzia X, Maiboom A, Karaky H (2017) Semi-physical models to assess the influence of CI engine calibration parameters on NOx and soot emissions. Appl Energy 208:1505–1518. https://doi.org/10.1016/j.apenergy.2017.08.232 Tesner PA, Smegiriova TD, Knorre VG (1971) Kinetics of dispersed carbon formation. Combust Flame 17:253–260. https://doi.org/10.1016/S0010-2180(71)80168-2 Totton TS, Chakrabarti D, Misquitta AJ, Sander M, Wales DJ, Kraft M (2010) Modelling the internal structure of nascent soot particles. Combust Flame 157:909–914. https://doi.org/10. 1016/j.combustﬂame.2009.11.013 Tree DR, Svensson KI (2007) Soot processes in compression ignition engines. Prog Energy Combust Sci 33:272–309. https://doi.org/10.1016/j.pecs.2006.03.002 Vervisch Kljakic P (2012) Modélisation des oxydes d’azote et des suies dans les moteurs Diesel. Ph.D. thesis, IFP Energies Nouvelles/ CNRS Veshkini A, Dworkin SB, Thomson MJ (2015) A soot particle surface reactivity model applied to a wide range of laminar ethylene/air flames. Combust Flame 161:3191–3200. https://doi.org/ 10.1016/j.combustﬂame.2014.05.024 Violi A (2004) Modeling of soot particle inception in aromatic and aliphatic premixed flames. Combust Flame 139:279–287. https://doi.org/10.1016/j.combustﬂame.2004.08.013 Vishwanathan G, Reitz RD (2010) Development of a practical soot modeling approach and its application to low-temperature diesel combustion. Combust Sci Technol 182:1050–1082. https://doi.org/10.1080/00102200903548124 Vishwanathan G, Reitz RD (2015) Application of a semi-detailed soot modeling approach for conventional and low temperature diesel combustion—part II: model sensitivity. Fuel 139:771–779. https://doi.org/10.1016/j.fuel.2014.08.025 Vishwanathan G, Reitz RD (2008) Numerical predictions of diesel flame lift-off length and soot distributions under low temperature combustion conditions. SAE Tech Pap 2008-01-13, 776– 790. https://doi.org/10.4271/2008-01-1331 Wang H (2011) Formation of nascent soot and other condensed-phase materials in flames. Proc Combust Inst 33:41–67. https://doi.org/10.1016/j.proci.2010.09.009 Wang H, Deneys Reitz R, Yao M, Yang B, Jiao Q, Qiu L (2013) Development of an n-heptane-n-butanol-PAH mechanism and its application for combustion and soot prediction. Combust Flame 160:504–519. https://doi.org/10.1016/j.combustﬂame.2012.11.017

5 Numerical Modelling of Soot in Diesel Engines

119

Wang H, Frenklach M (1997) A detailed kinetic modeling study of aromatics formation in laminar premixed acetylene and ethylene flames. Combust Flame 110:173–221. https://doi.org/10. 1016/S0010-2180(97)00068-0 Wong D, Schuetz CA, Frenklach M (2009) Molecular dynamics of PAH dimerization. In: Bockhorn H, D’Anna A, Saroﬁm AF, Wang H (eds) Combustion generated ﬁne carbonaceous particles, Proceedings of an international workshop held in Villa Orlandi, Anacapri, 13–16 May 2007, pp. 247–257. KIT scientiﬁc publishing, Karlsruhe Xi J, Zhong B (2006a) Reduced kinetic mechanism of n-Heptane oxidation in modeling polycyclic aromatic hydrocarbon formation in diesel combustion. Chem Eng Technol 29:1461–1468. https://doi.org/10.1002/ceat.200600190 Xi J, Zhong B-J (2006b) Soot in diesel combustion systems. Chem Eng Technol 29:665–673. https://doi.org/10.1002/ceat.200600016 Xu F, Faeth GM (2001) Soot formation in laminar acetylene/air diffusion flames at atmospheric pressure. Combust Flame 125:804–819. https://doi.org/10.1016/S0010-2180(01)00221-8 Yoshihara Y, Kazakov A, Wang H, Frenklach M (1994) Reduced mechanism of soot formation— application to natural gas-fueled diesel combustion. Symp Combust 25:941–948. https://doi. org/10.1016/S0082-0784(06)80730-7 Zellat M, Rolland T, Poplow F (2005) Three dimensional modeling of combustion and soot formation in an indirect injection diesel engine. SAE Tech Pap 900254. https://doi.org/10. 4271/900254 Zhang Y, Ghandhi J, Rothamer D (2014) Comparison of particulate size distributions from advanced and conventional combustion—part I: CDC, HCCI, and RCCI. SAE Tech Pap 2014– 01–1296. https://doi.org/10.4271/2014-01-1296 Zhao F, Yang W, Zhou D, Yu W, Li J, Tay KL (2017) Numerical modelling of soot formation and oxidation using phenomenological soot modelling approach in a dual-fueled compression ignition engine. Fuel 188:382–389. https://doi.org/10.1016/j.fuel.2016.10.054 Zhao F, Yu W, Su W (2016) Sensitivity study of engine soot forming using detailed soot modelling oriented in soot surface growth dynamic. Fuel 168:81–90. https://doi.org/10.1016/j. fuel.2015.10.092 Zhong B-J, Dang S, Song Y-N, Gong J-S (2012) 3-D simulation of soot formation in a direct-injection diesel engine based on a comprehensive chemical mechanism and method of moments. Combust Theory Model 16:143–171. https://doi.org/10.1080/13647830.2011. 598567

Chapter 6

Physico-chemical Properties of Diesel Exhaust Particulates Jianbing Gao and Guohong Tian

Abstract Diesel particulate matter (PM) has brought about huge environmental problems. The diameter of diesel PM is smaller than 1 lm that could be easily inhaled into the respiratory system. The PM diameter is smaller if turbocharger and common rail are used. Restrict exhaust emission legislations were carried out to decrease diesel PM emissions both in mass and number level. The formation mechanism of diesel PM is complex, and the cylinder combustion conditions have a huge effect on PM physico-chemical properties. In this chapter, the detailed analysis of diesel PM physico-chemical properties is made such as particulate ingredients, diameter and mass distributions, microstructures and oxidation behaviours. Diesel PM mainly contains soot, soluble organic fraction and ash whose percentages change with PM formation conditions such as engine load and speed, engine type and fuel type. Diameter distributions of diesel PM show a peak value around 100 nm that is caused by nucleation mode, while the mass distributions present double peaks around 40 and 300 nm, where the nucleation and accumulation mode particulates dominate, respectively. And, the researches of microstructures and oxidation behaviour make the foundation of decreasing PM emission and optimizing PM after-treatment device. Keywords Diesel engine

6.1

Particulate matter Physico-chemical properties

Introduction

Diesel engines are gaining more and more share in the vehicle market due to their excellent fuel economy and durability, especially in Europe where the share is around 44.8% in 2017. However, diesel exhaust emissions have brought about negative effects on human health and environment (Ma et al. 2016). Compared with spark ignition engines, the most evident difference in exhaust emissions is J. Gao G. Tian (&) University of Surrey, Surrey GU2 7XH, UK e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2019 A. K. Agarwal et al. (eds.), Engine Exhaust Particulates, Energy, Environment, and Sustainability, https://doi.org/10.1007/978-981-13-3299-9_6

121

122

J. Gao and G. Tian

particulate matter. Ammann’s results (Ammann et al. 1998) showed that the heterogeneous production of HNO2 from NO2-containing suspended soot particles was 105–107 faster than previously reported surface suspending reactions, which implied that the soot particles also promoted the formation of photochemical smog. The diameters of diesel particulates are in the nanoscale level, and number concentration and mass concentration research 108/cm3 (Agarwal et al. 2014) and 17.6 mg/m3 (Xing et al. 2013) level, respectively. The applications of advanced technologies such as high-pressure common rail and turbocharger cause higher PM number concentration and smaller diameter. The smaller size PM is easier to be sucked into the respiratory tract, which may cause serious respiratory diseases. Due to the diffusion combustion characteristics of diesel engines, PM formation is inevitable in the combustion process. The PM formation process experiences the pyrolysis of diesel fuel, nucleation, surface growth, coalescence, agglomeration and oxidation (Mohankumar and Senthilkumar 2017). Large amounts of PAHs are formed during the pyrolysis process and PAHs are PM precursors, which causes much more PM emission for higher PAHs content fuels. The nuclei have an initial diameter of 1.5–2 nm that they are formed from gas phase reactants. The PM mass increase mainly happens in the surface growth process that gas phase hydrocarbon deposition occurs on the surface of spherules. Primary particles are formed in the process with diameter in the range of 20–70 nm. The primary particles form PM aggregations which are in the diameter of micrometre level with the actions of collision and adsorption. Due to the high-temperature atmosphere for PM formation, oxidation happens for the PAHs, precursors and primary particles and PM aggregation in the whole process. PM formation process has a huge influence on PM physico-chemical properties. Deeply understanding the physico-chemical properties of diesel, PM contributes to decrease the PM emission and to optimize the PM capture device.

6.2 6.2.1

Physico-Chemical Properties PM Ingredients

As complex ingredients, diesel particulate matter mainly contains soot, organic compound and ash, as indicated in Fig. 6.1. The PM ingredients are closely related to cylinder combustion and after-treatment technologies. As can be seen, large amounts of particulate are composed of unburnt oil so that PM emission is high for the engines whose oil consumption is high, which is one of the main reasons which cause high PM emission for old engines. The application of diesel oxidation catalyst (DOC) is also useful to decrease PM emission that DOC could effectively decrease the unburnt fuel adhering on PM surfaces to some extent. PM generated at high engine load and speed conditions tend to have low percentages of organic compound (unburnt fuel and oil) and high soot content. Due to the complex process

6 Physico-chemical Properties of Diesel …

123

Fig. 6.1 Ingredients of diesel PM (Gao et al. 2018a)

7%

Ash Soot Unburnt fuel Unburnt oil Sulfate/water

25%

41% 13%

Fig. 6.2 FTIR spectra of diesel particulate (Chien et al. 2008)

14%

Absorbance

Biodiesel soot Diesel soot Anhydride COOH Anhydride

Anhydride COOH

600

1200

1800

2400

3000

3600

-1

Wavenumber / cm

of cylinder combustion, particulate ingredients change greatly with engine operation conditions, diesel fuel, combustion technology, exhaust gas recirculation, valve timing and fuel injection timing and so on (Mohankumar and Senthilkumar 2017; Ruiz et al. 2015; Tsai et al. 2018). Organic compounds contained in diesel particles are mainly from the unburnt fuel and lubricating oil. Figure 6.2 shows FTIR spectra of diesel particulate when engine is fuelled with diesel and biodiesel. As can be seen, diesel PM generated from biodiesel combustion presented high content of organic compounds compared with that of diesel fuel, especially the oxygen-containing functional groups. Due to higher oxygen content of biodiesel fuel, peak temperature of cylinder combustion is higher than that of diesel combustion and the fuel is also more completely combusted, which also leads to lower HC, CO emission and higher NOx emission. Condensation and aggregation of hydrocarbon happen in the process of PM formation so that sampling temperature has a huge influence on HC content. The organic compounds contained in diesel PM are mainly composed of aldehydes, alkanes, alkenes, aliphatic hydrocarbons, polyaromatic hydrocarbon (PAHs) and their derivatives (Mohankumar and Senthilkumar 2017). Gao et al. (2018b) detected methyne, anhydride, alkenes, methyl, methylene, carbonyl and hydroxyl in diesel PM, and showed that the organic compounds decreased greatly after partly oxidation. About 40 kinds of organic compounds was also detected in diesel PM using gas chromatography–mass spectrometer (GC-MS) that the number of carbon atoms contained in the organic compounds were in the range of C4–C26 (Zhiyuan and Diming 2012). The mass percentage of alkane and ethers were 20.2 and 69.09%,

124

J. Gao and G. Tian 500

Fig. 6.3 EDX spectra of ash PM (Liati et al. 2012)

Intensity

400

C

300

O Si

200

0

Zn, Mg and Al Fe

Fe

100

Zn 0

200

400

600

800 1000 1200

Energy loss / keV

respectively, and the use of biodiesel decreased the ethers content and increased the alkane content. As mentioned in reference (Gao et al. 2018a), ash mainly consisted of metallic species such as Al, Na, Mg, Zn and Ca which were stemmed from engine wear, lubricating oil additives and catalytic converter. Figure 6.3 shows the EDX spectra which were used to test the ingredients contained in ash. As indicated in Liati et al. (2012), much metallic species were from lubricating oil that higher lubricating oil consumption caused higher ash content. And, during engine cold start and warm-up conditions, friction is more serious due to high viscosity caused by low lubricating oil temperature. So that, frequently operating at cold start conditions causes high ash content of particulate. The use of biodiesel fuel also increases the ash content. Particulate oxidation catalyst (POC) could partially oxidize soot that causes the decrease of the soot content with the results of higher ash content. Similar to POC, non-thermal plasma (NTP) can oxidize PM which leads to more ash remaining in PM. And, most of the metallic species showed catalyst effect on PM oxidation, as mentioned in reference [88] that Ca, Mg and Zn presented high catalytic activities on PM oxidation, while Na showed good catalytic activity on hydroxide. The metal species contained in PM ash were observed using scanning electron microscope (SEM) (Liati et al. 2012) that the Fe–Cr metal species and Fe-oxide flakes were mainly from engine wear and catalyst. Also, there were hundreds of micrometres-large spherical or drop-shaped glassy particles with Al-silicate composition. During the typical transient driving cycle, engines experience the acceleration, deceleration and idle conditions frequently, which causes the continuous changes of PM concentration, diameter and surface area. Figure 6.4 shows the surface diameter and surface concentration (particle concentration and electrical aerosol detector signal) as a function of time. The particle diameter is smaller than 40 nm at the ﬁrst 20 s of the transient driving cycle. The diameter is mainly focused on the ranges of 400–500 nm at most of the engine operating conditions. The shapes of diameter and particle concentration show the similar tendency that the time corresponding to the peaks and the platform is similar. Different from the diameter history, the particle concentration changes enormously that it covers 103–108/cm−3 which is greatly dependant on engine operating conditions.

6 Physico-chemical Properties of Diesel …

125

Fig. 6.4 Surface diameter and surface concentration as a function of time (Biswas et al. 2008): triangle; surface diameter (Ds); back dashed line, EDA current; grey line, total particle concentration

6.2.2

PM Microstructure

Figure 6.5 shows the morphology of PM aggregation which was peeled off from the collection plate of NTP reactor. The PM aggregation shows the branch-like shape with many embossments on the surface. Because, the PM aggregation was sampled from NTP reactor that the shape was partly affected by the action of electric ﬁeld force and plasma. The morphology showed closely related to speciﬁc surface area and porosity, which signiﬁcantly affected PM oxidation behaviours. The surface

Fig. 6.5 Morphology of PM aggregation

126

J. Gao and G. Tian

area and volume of microstructure pore increase with C/H ratio (Rockne et al. 2000) which implies that the surface area and volume mainly depended on element carbon rather than organic carbon. Figure 6.6 presented the TEM and HRTEM of diesel PM. Diesel particles overlapped with each other, forming the snake-like shape. Nanostructure of diesel PM shows onion-like structures with crystallite randomly arranged or core– shell-like structure with void cores and densely arranged shells. The nanostructure of diesel PM greatly depends on the cylinder combustions that higher combustion temperature leads to more densely arranged crystallites. During the oxidation process of diesel PM, the nanostructure changes from onion-like structure to the core–shell-like structure, eventually, the core–shell-like structure completely break up. In the PM formation process, it is easier to become the void cores if exhaust gas recirculation (EGR) is applied (Al-Qurashi and Boehman 2008), and the internal and surface oxidation happen simultaneously at the initial stage of oxidation. Also, the oxygen content decreases with the proceeding of PM oxidation. For the PM

Fig. 6.6 Morphology of diesel particulate: upper panel, TEM; bottom panel, HRTEM (Ma et al. 2011; Wang et al. 2014)

6 Physico-chemical Properties of Diesel …

127

formed at high engine load and speed conditions, the crystallites tend to be more densely arranged and to be with more void cores. In order to avoid the influence of volatilized organic compound on PM oxidation behaviour, diesel PM is often pretreated at high-temperature and non-oxidized atmosphere. Gao et al. (2017) investigated the nanostructure changes after being pretreated at high-temperature conditions that the nanostructure changes from the onion-like structures to the void cores, which implies that the oxidation activity decreases after pretreatment. In the oxidation process, topological form and nanostructure of diesel PM change greatly which could be observed by transmission electron microscopy (TEM) and high-resolution TEM (HRTEM). Figure 6.7 show the TEM ﬁgures, diameter distributions and HRTEM ﬁgures during PM oxidation process. The diameter distribution is the statics results of the TEM ﬁgures. It seems that the

Fig. 6.7 TEM ﬁgures (upper panel), primary diameter distribution (middle panel) and HRTEM (bottom panel) ﬁgures during oxidation process (Song et al. 2006)

128

J. Gao and G. Tian

primary particles have become hollow that could be deduced from the outer boundary of the PM aggregations (TEM ﬁgure). The primary particle diameter decreased by *25% after 75% mass loss which also indicates that the porosity increases during the oxidation process. The nanostructure after 75% mass loss is completely different from the initial state. It changes from the onion-like structures to the capsule structures with densely arranged long fringes. In order to quantify the characteristics of the PM nanostructure, the fringe separation distance, fringe length and curvature are measured. PM formation conditions have a huge influence on these parameters. High PM formation temperature conditions tend to cause smaller fringe separation distance and tortuosity, and longer fringe length. Figure 6.8 shows the fringe length and tortuosity distribution at different start of fuel injection (SOI). The fringe length and tortuosity show smaller value for advancing SOI conditions than that of retarding SOI conditions. The median value is around 0.91 and 0.76 nm for fringe length and it is 1.19 and 1.50 for fringe tortuosity. For the diesel PM sampled at high engine load and speed conditions, it also shows small fringe tortuosity and huge fringe length. During the

Fig. 6.8 Fringe length and tortuosity distribution: (a, b), advanced start of fuel injection; (c, d), retard start of fuel injection (Yehliu et al. 2013)

6 Physico-chemical Properties of Diesel …

129

oxidation process, the fringe length increases and fringe tortuosity decreases gradually which partly causes the oxidation reactivity to decrease.

6.2.3

PM Diameter Distribution

Figure 6.9 shows the diameter distribution with EGR rate ranging from 0 to 37%. The particle diameter corresponding to the peak value varies between 50 and 100 nm. In order to decrease NOx emission, the exhaust gas is often introduced into the cylinders to decrease cylinder temperature to suppress the NOx formation. With the increase of EGR rate, particle number increases dramatically, especially when EGR is higher than 30%. And high EGR rate leads to the peak shift of particle diameters to huger value. Due to low combustion temperature, more HC is formed which causes more HC to be condensed into the liquid phase PM. More small particles aggregated into huge particles due to high HC content in PM. The particle number difference was mainly in the regions of aggregation particles. Different from the diameter distributions in Harris and Maricq (2001), the diameter distributions show evidently two peaks around 50 and 100 nm where the particles are in the nucleation mode and aggregation mode, respectively, as indicated in Fig. 6.10. The nucleation particles are mainly composed of inorganic salt and HC in liquid phase. As shown in the ﬁgure that the particle number concentration is the least for ultra-low sulphur diesel (ULSD), especially for the nucleation mode particles, which is caused by less sulphate formation in combustion process. The nucleation particle concentration decreases by 90% when ULSD is used. Biodiesel also contributes to the decrease of particle formation and the phenomenon is more evident at low engine load conditions, which is consistent with the results in Xing et al. (2013). Decreasing the sulphur content in diesel is one of the main methods in the past decades to decrease PM emission. Compared with diesel, Fischer–Tropsch (FT) contains less polycyclic aromatic hydrocarbons (PAHs) that are precursors in the process of PM formation. Less PAHs content could suppress Fig. 6.9 Diameter distributions of diesel particulate at different EGR rate (Zhu et al. 2010)

130

J. Gao and G. Tian

Fig. 6.10 The influence of biodiesel fuel on particle size distribution (Zhu et al. 2010)

the nucleation process of diesel particles, which partly leads to the FT diesel to be widely used in diesel engines. Completely different from the particle size distribution, the mass distribution presents two peaks around 300 nm and 7 lm, and the engine speed shows little influence on the ﬁrst peak (Xing et al. 2013). The number concentration may be high although the mass emission is low for the modern diesel engines. Gao et al. (2018) also made the statistics of PM diameter distribution based on TEM ﬁgures, which indicated the PM geometry diameter, however, the values obtained using electronic low-pressure impacter (ELPI) were aerodynamic diameter. During combustion process, sulphate plays a vital role in the PM formation that it promotes the PM formation and it exists mainly in the forms of nuclear mode. The use of ultra-low sulphur diesel (ULSD) could effectively decrease the PM formation (Wåhlin et al. 2001). Chuepeng et al. (2011) compared the effect of ULSD and biodiesel on PM diameter and mass distribution that the use of biodiesel

6 Physico-chemical Properties of Diesel …

131

is more helpful in terms of decreasing PM emission. The sulphur content is 9 times higher for ULSD than that of biodiesel and it is more than 200 times for total aromatics content. Less aromatics and sulphur effectively decrease the precursor during PM formation process. The same phenomenon was also demonstrated on a 2.4 L common rail direct injection diesel engine (Hwanam and Byungchul 2008).

6.2.4

PM Oxidation Behaviours

The relations of PM oxidation behaviours with engine operation conditions (Al-Qurashi and Boehman 2008; Seong and Boehman 2013), biodiesel fuel (Song et al. 2006; Agudelo et al. 2014) and physico-chemical properties (Seong and Boehman 2013; Mühlbauer et al. 2016; Zhao et al. 2015) were widely investigated to resolve the PM regeneration problem. After long time operation of DPF, more and more diesel particles are accumulated on the ﬁlter which causes high engine backpressure with the results of poor engine performance. Researches on the oxidation behaviours of diesel particles contribute to resolve the DPF regeneration problem and optimize the DPF regeneration device. The commonly used method to achieve DPF regeneration is to incinerate the particles accumulating on the ﬁlter. The ignition temperature of diesel particles is around 450 °C and burn out temperature is around 600 °C at the conditions of without catalyst, as indicated in Fig. 6.11. Diesel particle oxidation temperature shows greatly depend on particle formation conditions such as EGR conditions, engine load and speed, fuel types and engine technologies. Different particle formation conditions cause the differences in particle ingredients and soot nanostructures. As mentioned in Gao et al. (2018a), ingredients and soot nanostructures were fundamental factors leading to different diesel oxidation behaviours. Diesel ash has catalytic actions on soot oxidation, and the catalytic actions are enhanced with the presence of SOF. With the proceeding of the periodic DPF regeneration, the ash adhered on DPF ﬁlter wall increases, which leads to higher catalytic performance. Oxygen-containing functional groups provide

100

Fig. 6.11 Oxidation behaviours of diesel PM (Gao et al. 2018b)

Percentage / %

80 60 40

Sample 1 Sample 2 Sample 3 Sample 4

20 0 100

200

300

400

500

Temperature / °C

600

132

J. Gao and G. Tian 100

Fig. 6.12 DSC-based oxidation proﬁles (Gao et al. 2017) 1-H/H / %

80 2.5 °C/min 5.0 °C/min 7.5 °C/min

60 40 20 0 150

200

250

300

350

400

450

500

550

Temperature / °C

active sites and surfaces conducing to soot oxidation. Also, the oxygen-containing organic compounds contained in PM cause amorphous carbon that can be observed by HRTEM ﬁgures and Raman spectra. Microstructures are vital factors influencing soot oxidation. Higher porosity and smaller primary diameter cause higher speciﬁc surface area contributing to oxygen chemisorption during oxidation process. During the temperature increasing process, the volatilization and oxidation of organic compound happen simultaneously when the temperature is low. However, the TGA experiment could not effectively distinguish the organic compounds volatilization and oxidation. Figure 6.12 shows the DSC-based oxidation proﬁles that were converted from the heat release during PM oxidation process. The heat release reflects the chemical reaction (oxidation) rather than the volatilization. The volatilization of organic compounds dominates the mass loss when the temperature is lower than 250 °C (Gao et al. 2017), and much heat is released after temperature researches 300 °C.

6.2.5

PM Activation Energy

During PM oxidation process, activation energy is one of the indexes that could reflect PM oxidation activity, and it shows closely relate to the energy to oxidize PM. The commonly used method to calculate PM activation energy is using TAG proﬁles based on Flynn–Wall–Ozawa (FWO) method, the Friedman–Reich–Levi (FRL) methods (López-Fonseca et al. 2005). As indicated above, the mass loss is mainly caused by the volatilization at the initial stage of TGA experiment so that there are huge errors if the calculation is based on TGA proﬁles. In order to decrease the errors, the PM is pretreated in many references to remove the organic compounds (Sharma et al. 2012). However, the pretreatment at high temperature neglects the influence of organic compounds on the activation energy. Oxygencontaining organic compounds provide active sites for PM oxidation. The activation energy calculation using DSC-based oxidation proﬁles is with higher precision

6 Physico-chemical Properties of Diesel …

133

compared with that based on TGA proﬁles when the temperature is low (before the volatilized organic compounds are completely reacted) (Gao et al. 2017). During the oxidation process, the activation energy increases generally, however, the phenomenon is also observed that the value decreased slightly at the end of the oxidation. As the temperature increases during the oxidation process, the oxygen content decreases and the graphitization aggravates, which causes the oxidation activity to decrease. As the oxidation proceeding, the ash content increases gradually with more serious catalytic effect on PM oxidation which contributes to decrease activation energy. The joint effect of the physico-chemical property changes causes the different activation energy tendency as the function of mass loss. The relations of activation energy and the average values of fringe length and fringe tortuosity are indicated in Fig. 6.13. Both the activation energy and fringe parameters (fringe length and tortuosity) could indicate PM oxidation activity. The activation energy increases generally with increasing fringe length and decreasing tortuosity. The equations about the activation energy and fringe parameters (fringe length, fringe separation distance, fringe tortuosity) are explored during the PM oxidation process that the fringe length has the hugest influence on PM activation energy. Raman spectral analysis is complementary to HRTEM investigations for carbon-containing materials (Mühlbauer et al. 2016), and it can be used to obtain the detailed information about PM oxidation reactivity by analysing the nanostructures (Zhao et al. 2015). Diesel PM shows two peaks at positions corresponding to the Raman shift at 1360 cm−1 (D peak) and 1590 cm−1 (G peak) for the ﬁrst order Raman spectra. The G peak is caused by the stretching mode E2g symmetry at sp2 sites, while the origin of the D peak is still under debate (Gao et al. 2017). Additional peaks at around 1200 cm−1 (D4), 1500 cm−1 (D3) and 1620 cm−1 (D2) Raman shift positions were also observed in the Raman spectra of disordered and amorphous carbon materials. Correlations between the Raman parameters and nanostructures have been discussed by many researchers (Rockne et al. 2000; Mühlbauer et al. 2016; López-Fonseca et al. 2005; Sharma et al. 2012) that

210

Activation energy / kJ mol

Activation energy / kJ mol

220 2

R =0.98

200 180 160 140 120

195

2

R =0.86

180 165 150 135 120 105

0.8

1.0

1.2

1.4

1.6

Average fringe length / nm

1.8

1.10

1.15

1.20

1.25

1.30

Average fringe tortuosity

Fig. 6.13 Average fringe length and fringe tortuosity versus activation energy: left panel, fringe length; right panel, right panel (Gao et al. 2018a)

134

J. Gao and G. Tian

Fig. 6.14 Correlations between soot activity (Tmax: temperature corresponding to maximum mass loss rate) and Raman parameters (Mühlbauer et al. 2016)

high amorphous carbon content and D peak intensity contribute PM oxidation. Figure 6.14 shows the correlations between soot oxidation reactivity and particle structure parameters from Raman spectroscopy. For the diesel PM sampled at the ﬁve given operation points, the soot activity have the same tendency with the Raman parameters (D1 of FWHM and ID/IG). Sharma (Ma et al. 2011) compared the intensity ratio (IG/ID1, IG/ID3, IG/ID4, IG/I(D1+D3+D4)), FWHM and peak position differences of different soot. The literature (Wang et al. 2014) shows that D1 FWHM and ID1/IG changes in soot samples taken at different engine operating parameters and no trend is observed, however, the tendency of the corresponding temperature at maximum oxidation rate is similar to that of ID1/IG. Figure 6.15 clearly indicated the correlations between soot activity and physical parameters. The changes of geometric mean diameter (GMD), total particle number concentration (TPNC) and average primary particle diameter (APPD) from HRTEM show the similar tendency with soot activity. And, the single parameter may present inaccurate tendency with soot oxidation activity because the soot activity is the joint results of PM physico-chemical properties, such as PM composition, speciﬁc surface area, porosity distribution and crystallite arrangement.

6.3

Conclusions

PM physico-chemical properties are closely related to PM formation conditions such the engine types, fuel types, engine conditions and engine technologies. Diesel PM mainly contains ash, soot, organic compounds, sulphate and water with soot and organic compounds dominated. Large amounts of oxygen-containing compounds are contained in diesel PM, and they are closely related to PM

6 Physico-chemical Properties of Diesel …

135

Fig. 6.15 Correlations between soot reactivity (Tmax: temperature corresponding to maximum mass loss rate) and GMD, TPNC APPD and PM composition (Mühlbauer et al. 2016)

oxidation. Metallic species in ash are mainly from engine wear, lubricating oil additives and catalytic converter, and the metallic species show catalytic action on PM oxidation. PM diameter distributions show two peaks at *50 nm and *100 nm corresponding to nucleation mode and aggregation mode particles. PM nanostructures show onion-like and core–shell-like structures which have a huge influence on PM oxidation. Huge fringe length causes low oxidation activity while huge fringe tortuosity is corresponding to high oxidation activity. And, the DSC-based oxidation proﬁles present PM oxidation with high precision at low-temperature conditions. Deep understanding of the PM physico-chemical properties contributes to decrease PM emission and optimize the after-treatments.

136

J. Gao and G. Tian

References Agarwal AK, Dhar A, Gupta JG, Kim WI, Lee CS, Park S (2014) Effect of fuel injection pressure and injection timing on spray characteristics and particulate size–number distribution in a biodiesel fuelled common rail direct injection diesel engine. Appl Energy 130:212–221 Agudelo JR, Álvarez A, Armas O (2014) Impact of crude vegetable oils on the oxidation reactivity and nanostructure of diesel particulate matter. Combust Flame 161(11):2904–2915 Al-Qurashi K, Boehman AL (2008) Impact of exhaust gas recirculation (EGR) on the oxidative reactivity of diesel engine soot. Combust Flame 155(4):675–695 Ammann M, Kalberer M, Jost D, Tobler L, Rössler E, Piguet D et al (1998) Heterogeneous production of nitrous acid on soot in polluted air masses. Nature 395(6698):157 Biswas S, Hu S, Verma V, Herner JD, Robertson WH, Ayala A et al (2008) Physical properties of particulate matter (PM) from late model heavy-duty diesel vehicles operating with advanced PM and NOx emission control technologies. Atmos Environ 42(22):5622–5634 Chien Y-C, Lu M, Chai M, Boreo FJ (2008) Characterization of biodiesel and biodiesel particulate matter by TG, TG–MS, and FTIR. Energy Fuels 23(1):202–206 Chuepeng S, Xu H, Tsolakis A, Wyszynski M, Price P (2011) Particulate matter size distribution in the exhaust gas of a modern diesel engine fuelled with a biodiesel blend. Biomass Bioenerg 35(10):4280–4289 Gao JB, Chao-Chen MA, Xing SK, Sun LW, Huang LY (2017a) Oxidation reactivity changes of diesel particulate matter after being pre-treated. Trans Beijing Inst Technol 37(9):913–918 Gao J, Ma C, Xing S, Sun L (2017b) Oxidation behaviours of particulate matter emitted by a diesel engine equipped with a NTP device. Appl Therm Eng 119:593–602 Gao J, Ma C, Xing S, Sun L, Huang L (2018a) A review of fundamental factors affecting diesel PM oxidation behaviors. Sci China Technol Sci 61(3):330–345 Gao J, Tian G, Ma C, Chen J, Huang L (2018b) Physicochemical property changes during oxidation process for diesel PM sampled at different tailpipe positions. Fuel 219:62–68 Gao J, Ma C, Tian G, Chen J, Xing S, Huang L (2018) Oxidation activity restoration of diesel particulate matter by aging in air. Energy Fuels. https://doi.org/10.1021/acs.energyfuels. 7b03404 Harris SJ, Maricq MM (2001) Signature size distributions for diesel and gasoline engine exhaust particulate matter. J Aerosol Sci 32(6):749–764 Hwanam K, Byungchul C (2008) Effect of ethanol-diesel blend fuels on emission and particle size distribution in a common-rail direct injection diesel engine with warm-up catalytic converter. Renew Energy 33(10):2222–2228 Liati A, Eggenschwiler PD, Gubler EM, Schreiber D, Aguirre M (2012) Investigation of diesel ash particulate matter: a scanning electron microscope and transmission electron microscope study. Atmos Environ 49:391–402 López-Fonseca R, Landa I, Gutiérrez-Ortiz M, González-Velasco J (2005) Non-isothermal analysis of the kinetics of the combustion of carbonaceous materials. J Therm Anal Calorim 80 (1):65–69 Ma C, Zhong L, Yu S, Xing S, Tian J (2011) Conﬁguration and thermal gravimetric analysis of diesel Particulate Matter after plasma. In: 2011 international conference on IEEE, electric information and control engineering (ICEICE), 2011, pp. 5762–5765 Ma C, Gao J, Zhong L, Xing S (2016) Experimental investigation of the oxidation behaviour and thermal kinetics of diesel particulate matter with non-thermal plasma. Appl Therm Eng 99:1110–1118 Mohankumar S, Senthilkumar P (2017) Particulate matter formation and its control methodologies for diesel engine: a comprehensive review. Renew Sustain Energy Rev 80:1227–1238 Mühlbauer W, Zöllner C, Lehmann S, Lorenz S, Brüggemann D (2016) Correlations between physicochemical properties of emitted diesel particulate matter and its reactivity. Combust Flame 167:39–51

6 Physico-chemical Properties of Diesel …

137

Rockne KJ, Taghon GL, Kosson DS (2000) Pore structure of soot deposits from several combustion sources. Chemosphere 41(8):1125–1135 Ruiz FA, Cadrazco M, López AF, Sanchez-Valdepeñas J, Agudelo JR (2015) Impact of dual-fuel combustion with n-butanol or hydrous ethanol on the oxidation reactivity and nanostructure of diesel particulate matter. Fuel 161:18–25 Seong HJ, Boehman AL (2013) Evaluation of raman parameters using visible Raman microscopy for soot oxidative reactivity. Energy Fuels 27(3):1613–1624 Sharma HN, Pahalagedara L, Joshi A, Suib SL, Mhadeshwar AB (2012) Experimental study of carbon black and diesel engine soot oxidation kinetics using thermogravimetric analysis. Energy Fuels 26(9):5613–5625 Song J, Alam M, Boehman AL, Kim U (2006) Examination of the oxidation behavior of biodiesel soot. Combust Flame 146(4):589–604 Tsai J-H, Lin S-L, Chen S-J, Guo C-J, Huang K-L, Lee J-T et al (2018) Emission characteristics of particulate matter and particle-bound metals from a diesel engine generator fueled with waste cooking oil-based biodiesel blended with n-butanol and acetone. Aerosol Air Qual Res 18:1246–1254 Wåhlin P, Palmgren F, Dingenen RV, Raes F (2001) Pronounced decrease of ambient particle number emissions from diesel trafﬁc in Denmark after reduction of the sulphur content in diesel fuel. Atmos Environ 35(21):3549–3552 Wang Y, Liang X, Shu G, Wang X, Sun X, Liu C (2014) Effect of lubricant oil additive on size distribution, morphology, and nanostructure of diesel particulate matter. Appl Energy 130: 33–40 Xing S-K, Ma C-C, Ma S (2013) Experimental study of effects of non-thermal plasma on exhaust PM quantity and mass in diesel engine. Neiranji Gongcheng (Chinese Internal Combustion Engine Engineering) 34(1):8–12 Yehliu K, Armas O, Vander Wal RL, Boehman AL (2013) Impact of engine operating modes and combustion phasing on the reactivity of diesel soot. Combust Flame 160(3):682–691 Zhao Y, Wang Z, Liu S, Li RN, Li MD (2015) Experimental study on the oxidation reaction parameters of different carbon structure particles. Environ Prog Sustain Energy 34(4):1063– 1071 Zhiyuan TPRSH, Diming L (2012) Soluble organic fraction and polycyclic aromatic hydrocarbons in particulate matter emissions from diesel engine with biodiesel fuel. J Mech Eng 8:020 Zhu L, Zhang W, Liu W, Huang Z (2010) Experimental study on particulate and NOx emissions of a diesel engine fueled with ultra low sulfur diesel, RME-diesel blends and PME-diesel blends. Sci Total Environ 408(5):1050–1058

Part III

Alternate Fuel Origin Particulates

Chapter 7

Oxygenated Fuel Additive Option for PM Emission Reduction from Diesel Engines—A Review Parameswaran Vijayashree and V. Ganesan

Abstract As on today, of the two types of engine used in mobility sector, diesel engines offer superior fuel and thermal efﬁciencies, better durability, greater torque, and higher power output compared to the gasoline engines. However, the diesel engines are a major source of both regulated and unregulated emissions which is responsible for the deteriorating air quality. The problem can be tackled either by using alternate power plant such as hybrid and electrical vehicles or by using alternate source of energy like biodiesel and oxygenated additives to diesel. The second option seems to be more attractive, as it does not need any major modiﬁcation to the millions of existing engines. In this direction, there is an urgent need to ﬁnd sustainable and environmentally friendly fuel types for the diesel engine application. In this chapter, the authors embark on the analysis and review of the application of oxygenated alternative fuels such as biodiesel, acetone–butanol– ethanol (ABE) solution and water emulsion as oxygenated fuel reformulation strategies. These strategies are aimed at achieving reduction of engine particulate emissions without much compromise on energy efﬁciency of the diesel engine. After the careful review of around 115 published literature, it is found that, still more research under controlled conditions is a must on these oxygenated fuels to gain more insight on their effects. This is especially true for the emissions of particulate matter (PM). Further, it is brought out through this review, a combination of factors such as higher oxygen content, more complete combustion and cooling effect to reduce this pollutant. If employed appropriately by having a diesel blend which contains proper amount of biodiesel, ABE solution and a small amount of water (0.5%), the regulated PM emissions can be reduced considerably. What it means is that such oxygenated fuels exhibit excellent performance in both brake

P. Vijayashree V. Ganesan (&) Department of Mechanical Engineering, Indian Institute of Technology Madras, 600 036 Chennai, India e-mail: [email protected] P. Vijayashree e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2019 A. K. Agarwal et al. (eds.), Engine Exhaust Particulates, Energy, Environment, and Sustainability, https://doi.org/10.1007/978-981-13-3299-9_7

141

142

P. Vijayashree and V. Ganesan

thermal efﬁciency (BTE) and NOx–PM trade-off. This chapter proposes an oxygenated diesel fuel blend not only for scientiﬁc study but also for the future practical application. Nomenclature ABE CO DOC EGR HC NOx OH PAHs PBDD/Fs PBDEs PCBs PCDD/Fs PM POPs SOx VOCs

7.1

Acetone–butanol–ethanol Carbon monoxide Diesel oxidation catalyst Exhaust gas recirculation Unburnt hydrocarbons Nitrogen oxides Hydroxyl radical Polycyclic aromatic hydrocarbons Polybrominated dibenzodioxins and furans Polybrominated diphenyl ethers Polychlorinated biphenyls Polychlorinated dibenzodioxins and furans Particulate matter Persistent organic compounds Oxides of sulfur Volatile organic compounds

Introduction

Diesel engines have become the engine of choice for on-road and off-road operations due to its better fuel economy. They operate on fossil diesel fuel, which is a mixture of hydrocarbons with carbon number C6–C20 aliphatic, alkanes such as tetra-, penta-, and hexa-decane as major components. It has small quantities of branched alkanes and aromatic alkanes (Guarieiro et al. 2008; Pang et al. 2006). The major disadvantage of diesel engines is that both the complete and incomplete combustion of diesel fuel result in emissions of hundreds (Correa and Arbilla 2006) of gaseous and particulate matters (PM) (Doğan 2011; Dutta and Radner 2009) COx, SOx, NOx, (Nelson et al. 2008; Zielinska et al. 2010) PAHs, VOCs, dioxins, and dioxin-like compounds. These emissions are the real threat to both the atmospheric and ecological environments (Borrás et al. 2009; Lin and Huang 2003). Over the last 50 years, there is a real threat of sea level increase due to global warming and this should be tackled on war footing (Baruch 2008; Islam et al. 2015; Viola et al. 2010; Wallington et al. 2013). The average temperature of the earth is on the rise (Viola et al. 2010; IPCC 2007) causing climate change and the energy security is in peril (Baruch 2008; Wallington et al. 2013). The energy demand is estimated to double by 2060 and would cause great strain in the coming years on the reserve of traditional fossil fuels (Nel and Cooper 2009). The increased

7 Oxygenated Fuel Additive Option for PM Emission Reduction …

143

extraction of the fossil fuel will lead to higher carbon emissions, which again contributes to global warming. Hence, there is an urgent need to focus attention on the alternate source of energy, viz., green energy (Winter 2014). It is a fact that mobility sector, especially road transport is a major contributor (Agarwal 2007; Hernández et al. 2012; Uherek et al. 2010) to the global warming. They are the major consumers of fossil fuels, namely diesel, gasoline, natural gas, petroleum gas, and other kinds of fuels (Demirbas 2007, 2008, 2009). The direct emissions of greenhouse gases such as carbon dioxide (CO2) into the environment and indirect greenhouse gases such as carbon monoxide (CO), oxides of nitrogen (NOx), and volatile organic compounds (VOCs) are the main cause for global warming. In addition, the emissions of aerosols and particulate matter (PM), affect the oxidation–reduction mechanism and capacity of the atmosphere (Wade et al. 1994). There is unprecedented deterioration of air quality in the urban areas due to these emissions. Recent studies (Nelson et al. 2008) have established a correlation between adverse human health problems and exposure to vehicle-related air pollutant emissions. Stringent regulations to reduce pollution have been enacted throughout the world to curb diesel engine emissions. Keeping in view the emission standards imposed worldwide over the years (Doğan 2011; McClellan et al. 2012; Rakopoulos et al. 2010; Wang et al. 2012), a number of research studies have been carried out to achieve a striking balance between energy efﬁciency and maximum power output (Lin et al. 2012). It should be clear that there is an urgent need for the alternate fuels to abate pollution from diesel engines especially the PM. This chapter addresses this point and the following sections will discuss the possible alternatives and their impact.

7.2

Oxygenated Fuel Alternatives and Reformation of Conventional Diesel Fuel

In the following sections, possible oxygenated fuel alternatives and possible reformation of conventional diesel fuel for reducing particulate matter is reviewed. The oxygenated fuels and fossil fuels differ by the oxygen bonding in the chemical structure. To use the oxygenated fuels, it must form as homogeneous mixture when added together with diesel fuel and must have minimum volatility and good cetane number (Patil and Taji 2013). Alcohols, biodiesel, ethers, and water emulsions are some of the oxygenated fuels. Ethanol (alcohol) has 35% (Wang et al. 2012), biodiesel has about 10%, acetone (ketone) 28% (Tsai et al. 2014), and diglyme (ether) has 36% of oxygen contents. Similarly, other promising oxygenated additives are diethyl adipate, diethyl succinate, dimethoxymethane, ethyl glycol monoacetate (Lin and Huang 2003), dimethyl carbonate, etc. (Patil and Taji 2013; Ren et al. 2008). The higher oxygen content in these fuel additives, play an important part in the reductions of particulate matters and soot. The oxygen

144

P. Vijayashree and V. Ganesan

enrichment enhances premixed combustion and improved diffusive combustion and thereby reduce unregulated carbonyls emissions, PAHs, CO, HC, and other POPs (Wang et al. 2012; Campos-Fernandez et al. 2013; Satyanarayanamurthy 2012). However, these reductions increase NOx emissions considerably.

7.2.1

Alcohols and Bio-alcohols

Bioethanol (Song 2014) is a bio-alcohol, which is one of the most researched alternate fuels. It is produced from fermentation of sugars obtained from agricultural waste, barley, cassava, corn, sorghum, sugar beets and sugarcane, etc. (Rakopoulos et al. 2010; Song 2014). Initially, the sucrose is converted to fructose and glucose by hydrolysis with the help of enzyme invertase found in yeast and then they are converted to ethanol by zymase (Demirbas 2007). Higher octane numbers found in bio-alcohols reduces the knocking tendencies in gasoline engines. Also, the higher oxygen content enhances cleaner burning and less pollutant emissions (Zhou et al. 2014). Lower combustion temperatures due to higher latent heat of vaporization of alcohols reduce NOx emission to a greater extent (Agarwal 2007). Further, brake power, brake thermal efﬁciency, and speciﬁc fuel consumption are improved (Agarwal 2007). Throughout the world, a number of research studies are going on in using ethanol (renewable) and methanol as alternate fuels. The difference between ethanol and methanol is that the former is biomass–derived, whereas the latter is fossil-based. Solubility in diesel fuel of methanol is lesser compared to ethanol (Rakopoulos et al. 2010). The main difﬁculty in using alcohols as the alternate fuel is due to their higher auto-ignition temperatures, higher vapor pressure, lower caloriﬁc values, and lower cetane number. Further, they are highly hygroscopic and their lubricity is low. These two factors have greater impact on engine performance. This impedes to use ethanol and methanol in diesel fuels (Rakopoulos et al. 2010) Dual/multi-injection systems, emulsiﬁed blends, and ethanol fumigation to the intake air are being attempted to overcome the above problems related to ethanol use (CamposFernández et al. 2012; Fraioli et al. 2014). Energy efﬁciency of alcohol blends is improved by employing cetane number enhancers to the blends (Doğan 2011). Higher alcohols, though they have lower oxygen content, namely, butanol and pentanol, provide increased cetane number, better miscibility, higher density, and energy content (Lin et al. 2012; Zhou et al. 2014; Campos-Fernández et al. 2012; Yilmaz et al. 2014). They can be attempted to overcome the shortcomings in using lower alcohols, viz., ethanol and methanol. However, there are not many research studies being carried out in this area (Uherek et al. 2010).

7 Oxygenated Fuel Additive Option for PM Emission Reduction …

7.2.2

145

Biodiesel

Biodiesel are transesteriﬁed vegetable oil (Demirbas 2007) which is comprised of fatty acid methyl/ethyl esters (Wang et al. 2013). They are biodegradable and nontoxic (Ramadhas et al. 2004; Ma and Hanna 1999). Biodiesel can be produced from vegetable oil by alcoholysis/transesteriﬁcation and thermal cracking/pyrolysis or by hydrodeoxygenation and microemulsion (Doğan 2011; Ma and Hanna 1999; No 2011). Feedstock for biodiesel can be classiﬁed under three groups, viz., ﬁrst-, second-, and third generation (No 2011). The ﬁrst-generation biodiesel is produced from edible sources such as rapeseed, soybean, coconut, sunflower, and olive and the second generation is from nonedible sources, namely cottonseed, castor, jatropha, corn, jojoba, karanja, linseed, palm ﬁber and rice husks, etc. The third-generation sources are microalgae, water emulsions, animal fats, and waste cooking oil (Chisti and Yan 2011; Sims et al. 2010; Wen et al. 2010). Though these sources are promising, they face number of problems while processing due to the requirement of materials and also very high intensive energy requirement during transesteriﬁcation and pyrolysis. Large quantities of biodiesel production, in future, will threaten food security, oxidation problems, and high cost of feedstock and storage. Transesteriﬁcation is a complex process involving the reaction of triglycerides in vegetable oils with alcohol especially excess methanol. The result of the process is dependent on the reactive conditions of a catalyst (acid or base) (Wen et al. 2010), temperature, reaction time, and molar ratio of alcohol to vegetable oil (Hwang et al. 2013). Further, catalyst, either KOH or NaOH must be involved to form glycerol and esters (Wen et al. 2013). Large amounts of glycerol (10% by mass) are produced by this process (Izquierdo et al. 2012). As of now, it is not economically viable because of its disposal problems and low market demand (Rakopoulos et al. 2010). In addition, saponiﬁcation is another problem during transesteriﬁcation (Ma and Hanna 1999). Secondary pollutants produced from the transesteriﬁcation processes, cost of pretreatment, and puriﬁcation processes are the other issues involved in this process. Emissions, namely carbon monoxide, hydrocarbons, particulate matters, PAHs, and POPs get reduced due to its improved combustion process (Wang et al. 2013; Chang et al. 2014; Lin et al. 2010, 2011). The combustion process gets improved (Campos-Fernández et al. 2012; Johnson 2009) because of it being an oxygenated fuel and its higher cetane number. It is evident from the literature (No 2011; Lapuerta et al. 2008) higher NOx emission is caused from neat biodiesel used in diesel engine compared to the conventional diesel engines. Though, biodiesel can be used in its neat form or as blends of biodiesel and fossil diesel in diesel engines. Some of the shortcomings faced compared to conventional diesel (Agarwal 2007; Agarwal et al. 2008; Demirbas 2007) are: (i) higher cloud and pour points, (ii) high viscosity, (iii) injector choking, (iv) cold start problems, and (v) lower energy content.

146

P. Vijayashree and V. Ganesan

For converting biomass to fuel by pyrolysis process, the animal fats and vegetable oils are heated to higher temperatures with or without a catalyst, to break them into simpler molecules (Demirbas 2008; Ramadhas et al. 2004; Ma and Hanna 1999). Linear and cyclic parafﬁns, namely alkanes, alkenes, and aromatics and the undesirable oxygenated compounds like aldehydes, carboxylic acids, and ketones are the other products obtained during pyrolysis (Agarwal 2007). These chemical compounds are separated by fractionation. The required fraction is blend with fossil diesel and used in the diesel engines (DeOliveira et al. 2006). Biodiesel produced from microalgae oil is from third-generation source. It is the latest emerging research on biodiesel production. Biological photocatalyst process of microalgae gives the beneﬁt of reducing the CO2 levels in the atmosphere (Ahmed et al. 2010). It gives a solution for retarding climate change and is not a threat for agricultural lands. Hence, microalgae can be considered as one of the promising and sustainable sources of fuel (Al-lwayzy and Yusaf 2013; Haik et al. 2011). Mwangi et al. (Campos-Fernández et al. 2012) reported that B2 (2%-microalgae biodiesel + 98% fossil diesel) blends, with butanol and with water addition showed signiﬁcant reductions in pollutant emission. Similar studies have also reported that compared to conventional fossil diesel, use of microalgae biodiesel reduced CO, HC, and PM pollutant emission, except NOx and increase in fuel consumption, (Islam et al. 2015; Yilmaz et al. 2014; Hariram and Kumar 2013; Topare et al. 2011).

7.2.3

Biomass-Derived Fuels

Development of more sustainable sources of fuels is an immediate and urgent need of the day (Rakopoulos et al. 2010; Wang et al. 2013). Sustainable and economically viable CO2, bio-char (solid), bio-alcohols, biodiesel, bio-oil, vegetable oil (liquid) and biogas, bio-hydrogen, syngas (gaseous) fuels referred as biofuels, can be produced from abundantly found biomass (Demirbas 2007, 2008; Izquierdo et al. 2012). They have reasonably good oxygen content compared to petro fuels. Biofuels are promising alternative source of energy for the mobility sector. It is estimated that on an average it can meet about 50% of the world’s energy demand by 2050 (Beatrice et al. 2014). Among them, biodiesel and acetone, butanol, ethanol, methanol, propanol (ABEMP) are highly promising fuels that can be used with minimum modiﬁcations on the existing diesel engines (Misra and Murthy 2010). These fuels have great potential to reduce greenhouse gases. Therefore, there is a good possibility to replace the petroleum-based fuels. Other promising biofuels are bio-syngas from biomass gasiﬁcation, bio-hydrogen from catalytic thermal cracking of bio-oil and bio-oil from pyrolysis. Moreover, production of fuel from biomass is more environmentally friendly compared to petroleum processing. Dermirbas states the advantages of use of biofuels as, they are environmentally friendly, easily available and exploitable from

7 Oxygenated Fuel Additive Option for PM Emission Reduction …

147

biomass, sustainable potential and offer great economic to oil importing countries (Demirbas 2007, 2009; Rakopoulos et al. 2010; Demirbas 2007; Rakopoulos et al. 2010). Over the past two decades, focus and interest on use of alternative fuels from biodiesel or vegetable-derived fuels have increased. It is mainly due to the abovementioned advantages and reduction of HC, CO, PAHs, and PM (Tsai et al. 2010). However, there will be a slight increase in NOx emission compared to conventional diesel.

7.2.4

Blends of Acetone, Butanol, and Ethanol

To minimize the problems faced with straight biodiesel, additives and solvents like ethanol, butanol, and acetone are added to the blends (Chang et al. 2014; Öztürk 2015; Tüccar et al. 2014). Further, the emissions from internal combustion engines are found to decrease by use of these blends (Doğan 2011). Acetone–butanol– ethanol (ABE) fermentation process is used to produce acetone, ethanol and butanol. Chaim Weizmann developed this process during the First World War by (Ni and Sun 2009). In this method, two-stage processes are involved, viz. acidogenesis and solventogenesis. Acetic and butyric acids are produced during acidogenesis and acetone, butanol, and ethanol during solventogenesis process (Chang 2010). During fermentation process, acetone, butanol, and ethanol with volumetric percentage range between 20 and 30%, 60 and 75%, and 1 and 5%, respectively, are obtained (Zhou et al. 2014). Ethanol though attempted on and off in diesel engines has low cetane number problem. Further, its blends have reduced viscosity and caloriﬁc value (Rakopoulos et al. 2011). Butanol is the main product of the ABE process (Zhou et al. 2014). It is a better alternative for blending with diesel for its properties, namely less hydrophilicity, easier miscibility, higher cetane, lower volatility, and heating values, which are better than other blends. Also, it is to be noted that the butanol has similar energy density as gasoline (Wen et al. 2014). Further, butanol is used in the food and plastic industries (Wen et al. 2014). When acetone is considered as an alternative, it is a good oxygenated additive to diesel blends since it is better soluble and has higher oxygen content (Tsai et al. 2014). The ABE fermentation process depends on cellulosic sources as substrates, viz., carbohydrate-rich feedstock. Potatoes, rice straw, cassava, bagasse, maize, beets, domestic waste, etc., are some of the substrates (Wen et al. 2014; Tsai et al. 2014; Claassen et al. 2000; Ponthein and Cheirsilp 2011). In the ABE fermentation process, the solventogenic activity is driven by anaerobic Clostridium bacteria strains (Fraioli et al. 2014; Maddox et al. 1995). The negative point in using ABE is that production costs are quite high. Further, energy demands for removing the water content and separating the additives from the fermentation mixture are prohibitive (Chang et al. 2014; Wu et al. 2014). Hence, to ensure its applicability and viability in engines research, it is appropriate to use

148

P. Vijayashree and V. Ganesan

ABE directly without separation, which would lead to saving of energy and cost (Zhou et al. 2014). However, only a limited study is available in the open sources on ABE (Nithyanandan et al. 2014; Wu et al. 2014). Therefore, it is worth conducting systematic studies to establish the potential of ABE.

7.2.5

Diesohol Blends

Diesohols or biodiesohols are oxygenated fuel blends consisting of biodiesel, alcohol (mostly ethanol), and diesel. Mixing of alcohol with diesel results in lowering cetane number. Biodiesels enhance the cetane number, viscosity, and heating value (Subbaiah et al. 2010). Ethanol is a solvent and can reduce the viscosity of biodiesels (Wang et al. 2013). Hence, diesohols or biodiesohols are being produced to use in diesel engines. Though there is a good improvement in emission reduction and fuel combustion, the production cost from crop source is very high (Hernández et al. 2012; Lin et al. 2012; Tsai et al. 2014).

7.2.6

Vegetable Oil

Vegetable oils have been tried ever since the invention of diesel engines (Agarwal 2007; Misra and Murthy 2010; Altın et al. 2001; Lin et al. 2011; Sidibé et al. 2010), especially during critical situation like the Second World War when there was a scarcity for diesel (Ramadhas et al. 2004). Lower aromatic and sulfur contents compared to petro-diesel, and the urgency for renewable and alternative fuel for diesel and its availability during the oil crisis in 1970s gained large interest in this area (Agarwal et al. 2008). Vegetable oils are extracted from oil-bearing agricultural seeds such as castor, coconut, cotton flax, groundnut, jatropha, karanja, linseed, olive, palm, rapeseed, soybean, and sunflower, etc. (Lin et al. 2011) through cold/hot straining or pressing and decanting or ﬁltering. The unﬁltered oil contains free fatty acids, phospholipids, triglycerides, and waxes. However, their physiochemical properties depend really on the biomass source (Agarwal et al. 2008). They are biodegradable and have the capacity to reduce CO2 emissions. They have properties close to those of diesel and are being used in conventional diesel engines without any major modiﬁcations (Yilmaz et al. 2014). However, vegetable oils do have some negative characteristics such as clogging of fuel nozzles and of particulates deposition in the combustion chamber walls. Further, there are shortcomings like the cost of production, operational difﬁculty, storage problems for a long time, and durability and engine wear (Ma and Hanna 1999; Lin et al. 2011). Due to lower caloriﬁc values, the power output from the vegetable oil run engines will be slightly lower. As vegetable oils are less volatile and more viscous, speciﬁc fuel consumption will be more. Thereby, the engine will

7 Oxygenated Fuel Additive Option for PM Emission Reduction …

149

operate with lower thermal efﬁciency and will have worse combustion characteristics because of longer ignition delays. Further, CO and HC emissions will be higher, but will have lower smoke and NOx (Misra and Murthy 2010; Lin et al. 2011; Sidibé et al. 2010). Vegetable oils can be preheated or pretreated to blend easily with diesel oil before using in the engines. The problems caused due to this fuel can be eliminated by attempting dual fueling, exhaust gas recirculation, modifying the engine injection system, and combustion strategies. Solvents such as acetone, butanol, and ethanol (ABE), methanol form microemulsions reduce the poor atomization and subsequent incomplete combustion (Agarwal 2007; Zhou et al. 2014; Ramadhas et al. 2004; Ma and Hanna 1999; Misra and Murthy 2010). These solvents can be researched upon and also, vegetable oils can be blended easily with diesel that reduces the viscosity problems. Another favorable point to consider vegetable oil as an alternate fuel is that it can be converted to biodiesel, which is easily adaptable in the diesel engines as fuel.

7.2.7

Water–Diesel Emulsion

Since 1970s, water–diesel emulsions in engines studies are being carried out (Lif and Holmberg 2006) to understand the effect of addition of a small amount of water in biodiesel blends. It is reported that there is reduction in NOx compared to using straight biodiesel, which increases NOx emissions and there can be a trade-off between reductions of CO and PM (Chang et al. 2013, 2014; Liu et al. 2012). Alahmer reported that there is a signiﬁcant effect on the physical and chemical kinetics of combustion when small amount of water is present in the fuel (Alahmer 2013). He states that this reduces the combustion temperatures and OH radicals are released in the combustion environment, which can control NOx formation and further oxidation of soot. Therefore, there is at advantage of reducing both NOx and PM emissions. Further, the possibilities for micro-explosions of water droplets might result in higher turbulence that enhances the mixing of fuel and oxidants can be envisaged. This can cause better and complete combustion, and thereby reduce pollutants emissions very signiﬁcantly (Chang et al. 2013, 2014; Liu et al. 2012; Alahmer 2013). Also, the addition of surfactants to the emulsions reduces the surface tension between the diesel and water phases (Ghannam and Selim 2009), stabilization of water–diesel mixtures that minimize the coalescence mechanism of the water phase and are imperative for water–diesel emulsion stability. Ghannam and Selim (2009) have reported in their study on stability of water–diesel emulsions that the stability of water–diesel emulsions is inversely proportional to water fractions, surfactants and water–diesel emulsions are directly proportional to water fractions. As already stated, one of the diesel engine emission control techniques is water addition. Various methods have been tried to introduce water into the engine. Some of them are: (a) simultaneous injection of water and diesel, (b) steam induction at inlet to mix water with air, and (c) stabilized or non-stabilized diesel water

150

P. Vijayashree and V. Ganesan

emulsions (Lin et al. 2008). Subramanian (2011), carried out a detailed study on water–diesel emulsion and timed water injection into the intake manifold and have evaluated the performance in controlling NOx and smoke emissions. NOx and smoke control were more effective with emulsiﬁcation whereas injection method seems to work well due to better diffusion combustion phase for CO and HC emissions. Water–diesel emulsions will be concentrated more in this review. As mentioned earlier, water emulsions result in micro-explosion phenomena, where diesel, which has less volatility, plays a controlling role in the overall boiling temperature during combustion (Huo et al. 2014). Secondary atomization occurs during micro-explosions, which enhances the air–fuel mixing. This is due to the disintegration of bigger droplets into smaller ones during micro-explosions, since vapor expansion occurs as water evaporates faster and reaches its superheated stage much earlier than diesel, thereby an advantage in emission reduction (Kadota and Yamasaki 2002). Ithnin et al. (Ithnin et al. 2014) have established in their study on water in diesel emulsions that there was improvement in combustion efﬁciency without engine modiﬁcations and considerable reduction in NOx and PM emissions. The important point to note is that presence of water increases the amount of OH radicals and oxidizes the soot precursors, which reduces the soot formation. Attia and Kulchitskiy (2014) have concluded that for higher HC and CO reductions, the emulsion droplets should be as ﬁne as possible, whereas for higher reduction in NOx emissions, it is better to have larger droplets.

7.2.8

Particulate Matters

The following three factors in a diesel engine, namely the completeness of combustion, the quality of the fuel, and the lubricating oil determine the nature and composition of particulate matter. Particulate matter is formed due to incomplete combustion of hydrocarbon fuels which is due to poor air–fuel mixing (Subramanian 2011). It may be in the form of soot or cenospheres. Soot is nothing but carbonaceous particles generated during gas-phase combustion process of gaseous hydrocarbon fuels and the carbonaceous particles formed due to direct pyrolysis of liquid hydrocarbon fuels are cenospheres. Also, PM can be classiﬁed as: (i) soluble organic fractions (SOF), (ii) solid particles from unburned fuels, (iii) sulﬁdes, and (iv) other fuel additives (Wang et al. 2012; Gill et al. 2012; Lin et al. 2006). Particle formation takes place via the following processes, namely, nucleation, accumulation, and condensation. The ﬁnal process is agglomeration of the soot nuclei formed by partially burned or unburned fuels (Gill et al. 2012; Lin et al. 2006). The mass increase through nucleation and condensation mechanisms, which happens within both the cylinder and exhaust pipe, increases the particle sizes. Many studies have revealed that in urban areas, major source of emissions is from diesel engines and ultraﬁne particle pollution are from the mobility sectors (Knibbs

7 Oxygenated Fuel Additive Option for PM Emission Reduction …

151

et al. 2011; Papastefanou 2008; Shi et al. 2001; Zhu et al. 2002; Zwack et al. 2011). Sulfur content in the diesel and oxygen from air react to produce gaseous SO2 and a few sulﬁde particles, which are very harmful. Hence, proper approach should be thought of, to aim at ultra-low sulfur in the fuel oil to considerably reduce both PM and the formation of sulfur oxides (Lin and Huang 2003). Walsh et al. (Walsh 2011) report that commonly studied sizes of PM2.5 and PM10 of particulate matters are mainly responsible for respiratory and lung problems, lower cardiac activities, and high mortality in people and stated that the size of the PM can be condensed to reduce above problems. Initially, attempts of particle abatement technologies were done on stationary engines and then extended to mobile diesel engines. One of the leading aftertreatment techniques for the abatement of PM is by employing diesel particle ﬁlter (DPF). These ﬁlters are almost 98% efﬁcient (Johnson 2009; Mayer et al. 2008) and have high removal efﬁciency on PAHs (Heeb et al. 2005; Ratcliff et al. 2010). These ﬁlters are widely used in light-duty diesel vehicles. However, they require regeneration processes to remove the soot (Gill et al. 2012) and also affects the fuel economy (Lapuerta et al. 2012). Another point to note is that ﬁtting of DPF into the exhaust system may enhance the formation of dibenzofurans (PCDD/Fs), polychloride dibenzo-p-dioxins, and polychlorinated biphenyls (PCBs) via de novo synthesis (Heeb et al. 2007, 2008, 2013; Hsieh et al. 2011). Normally, diesel oxidation catalyst (DOC) is used to treat both carbon monoxide (CO) and volatile organic compounds (VOC) of the exhaust. But, it has minimum effect on the soot fraction. So, diesel particulate ﬁlters (DPF) in combination with DOC are employed to reduce soot in the exhaust (Vaaraslahti et al. 2006). The DPF is similar to an oxidation catalyst, which consists of semipermeable channels and walls. Only gases are allowed to pass through these channels, where soot particles are trapped. The catalyst on the channel walls, oxidize the trapped soot particles into CO2 and also reduce HC and CO by oxidation. In general, the use of oxygenated fuels causes reduction of particulate matters and enhances combustion in the presence of oxygen. Alcohol Addition and Its Effect on PM Emission When large amount of lower alcohol is added to butanol–diesel blends, PM emissions were found to reduce (Rakopoulos et al. 2010; Rakopoulos et al. 2010). Tsai et al. (2014) have reported similar results from their studies for isopropyl alcohol. The higher oxygen content of the alcohols helps oxidation of carbon to CO2, enhances complete combustion, and inhibits the formation of soot (Wang et al. 2013). Alcohol fractions are found to reduce the carbon content (Doğan 2011) and the sulfur content of the overall fuel blend (Tsai et al. 2014), which are the main source for PM formation. Similarly, biodiesel combined with water-containing ethanol (Lee et al. 2011) reduces the sulfur content, thereby, SOF formation and nucleation mode PM emissions. Biodiesel Blends and Its Effect on PM Emission According to Lin et al. (2006, 2008) PM emission is found to increase with increase in biodiesel in fuel blends. This may be due to the increase in soluble organic fraction in the exhaust. Tsai et al. (2010) have also reported similar results that there was an increase of 15% mean

152

P. Vijayashree and V. Ganesan

PM emission for a blend of 50% biodiesel-50% diesel. This is due to higher cetane number causing nebulization and higher biodiesel fractions causing higher viscosity. In another study (Tsai et al. 2014), using biodiesohols containing waste cooking oil, biodiesel, and acetone reduced the PM emission by about 8–40%. They attribute it to the lower hydrogen and carbon content and also the higher oxygen content in the blend (Hwang et al. 2013; Lin et al. 2006). ABE blends and Their Effect on PM Emission Zhou et al. (2014) have used 50 and 80% ABE solution in diesel blends and with no water content during their research. Their approach was to study the combustion characteristics of ABE-diesel blends at low-temperature conditions in a conventional constant volume chamber. They found that the higher oxygen content in the ABE-diesel blends produced better oxidation and thereby less soot formation when compared to conventional diesel. However, a recent study (Chang et al. 2014) reports that around 11–22% PM reductions could be achieved by the addition of different amount of water-containing ABE-diesel blends as given in Appendix. Other Oxygenated Fuel Blends and Their Effect on PM Emission A study by Wang et al. (2012) state that glycol ethers can also reduce PM emissions. However, with much lower efﬁciency compared to biodiesel blends and alcohols. Ren et al. (2008) are of the opinion that PM reduction is directly related to the oxygen content in the fuel blends. They report that for every 1% increase in oxygen content in the fuel mass, there was about 3.5% reduction in PM. Soot formation precursors were reduced by using oxygenated additives. This is due to the generation of radicals that limit aromatic rings formation and oxidize soot to CO2, which in turn the overall PM formation is reduced (Wang et al. 2012). In diesel engines, there are two phases of combustion, viz., (i) premixed phase and (ii) diffusive phase. Diffusive combustion leads to more complete combustion, which can be achieved by the addition of oxygenated additives that increases oxygen content and lowers PM emissions (Ren et al. 2008). Water Addition and Its Effect on PM Emission According to Lin et al. (2008, 2012), there is considerable reduction of PM emission when water is added to diesel blends. The OH radicals from water in the fuel/air mixture help to oxidize the soot being formed during the combustion process and thereby reduce total PM emission. Another point to note is that lower cylinder temperatures cause less pyrolysis reactions that form cenospheres and the dilution effect of the added water fraction reduces the sulfur content. It is to be noted that sulfur particles form the nuclei for particulate formation (Lin et al. 2012). It is found that water content of 0.5% is more appropriate than 1% by volume in reducing PM emission. It is due to the fact that the combustion temperature lowers due to the cooling effect of the higher water content causing large PM formations. Similarly, Wang et al. (2012) report from their studies state that 10% water content reduced PM emissions close to 20%, while 15% water content reduced PM by approximately 18% indicating that there is a limit for water addition. Further results are given in Appendix.

7 Oxygenated Fuel Additive Option for PM Emission Reduction …

7.2.9

153

NOx–PM Trade-off

Reduction of NOx and PM emissions simultaneously is a challenging job. It requires a reducing mechanism for NOx, and an oxidizing mechanism for the reduction of PM emission. Better combustion efﬁciency causes higher in-cylinder temperatures, which encourage the formation of thermal NOx but reduces PM formation. Therefore, the traditional methods may not work, to simultaneously reduce NOx and PM in diesel engines (Wang et al. 2012; Ren et al. 2008). PM emission reduces when ignition delay is reduced by better premixed combustion phase with higher mixing rate of fuel with air. However, NOx emission increases due to high in-cylinder temperatures (Subramanian 2011). Similarly, EGR can also reduce the NOx but PM emissions will increase. This envisages the employment of aftertreatment measures, which are again costly. Therefore, to achieve NOx–PM trade-offs attempts are being made in fuel reformations which are less costly. According to Chang et al. (2014) and Huo et al. (2014) NOx–PM trade-off can be achieved by fuel emulsiﬁcation, which has high potential. Subramanian (2011) from his studies reported using of emulsiﬁed water– diesel mixture for simultaneous reduction of NOx and PM. Further, alcohols blended with diesel (Yilmaz et al. 2014) and oxygenated additives such as diglyme (Gill et al. 2012) could also achieve NOx–PM trade-offs. As already mentioned, lower local adiabatic temperatures due to water addition can bring down NOx emission. The resulting OH radicals due to water addition causes micro-explosion which helps in better air–fuel mixing and combustion, thereby, there will be less PM emissions (Subramanian 2011; Huo et al. 2014). Moreover, the OH radicals act as oxidizer and this causes lesser PM emission.

7.3

Summary and Conclusions

It should be clearly understood that the diesel engines are the major source of carcinogenic and mutagenic persistent organic pollutants. Therefore, it is utmost important to reduce these emissions also from the diesel engines keeping in mind their excellent energy efﬁciency. Reduction in emissions from such an engine can be addressed in two ways (i) from the point of view of design change and (ii) from the point of view of fuel additives. The second one is more attractive since there will be minimum modiﬁcations to the existing engines. In this connection, biodiesel has become more attractive for its environmental beneﬁts and it is produced from renewable vegetation resources. Because of its higher viscosity and low caloriﬁc value compared to fossil diesel there is a requirement for advanced injection timing for biodiesel combustion. Because of the above factors, it emits higher amount of NOx. These shortcomings can be eliminated by using solvents such as water in biodiesel–diesel blends and alcohols.

154

P. Vijayashree and V. Ganesan

After carefully reviewing more than 100 published literatures, the following points are summarized along with the ﬁndings and recommendations regarding the reduction of emission and oxygenated fuel strategy for diesel engines by using diesel–biodiesel–alcohols–water blends with additives: From the present review, it is evident that the use of biodiesel in diesel engines is a viable proposition and is also a very promising renewable fuel alternative. Engines can be operated with blends of biodiesel with fossil diesel or even as neat biodiesel without any major engine modiﬁcations. One of the more attractive biodiesel sources is microalgae. Since it has a third-generation source, it does not compete with food crops compared to other crop-based sources. However, microalgae biodiesel also emits higher amount of NOx with reduced amount of PM, and PAHs as other biodiesels. Simultaneous reduction of NOx and PM is an important challenge, which is being addressed by many researchers throughout the world. Different techniques have been proposed to achieve this NOx–PM trade-off, by aiming at low-temperature combustion to control the formation of thermal NOx and particulate matter especially soot through fuel additives. PM emission reduction can be achieved by blending acetone–butanol–ethanol (ABE) in diesel. PM emission is reduced by oxidation from the higher oxygen content in ABE–diesel blends and the NOx formation are reduced by higher latent heat of these blends that lowers the combustion temperatures. ABE solution without separation process can signiﬁcantly reduce costs in the fermentation process. Further, microalgae can be considered as an important source of biomass in the fermentation process. This will further increase the importance of microalgae as a source of future biofuels. The authors of this chapter are of the opinion that water emulsion technique can be a better fuel for NOx–PM trade-off. OH radicals generated by micro-explosions caused by combustion of water particles in the fuel, oxidize the soot and reduce NOx formation. Also, it causes longer ignition delay by which combustion temperatures are reduced, lowering the NOx emission. From the present review, the authors would like to point out the gap that exists in the importance of using different fuel additives. For example, the results obtained by various studies using ABE-diesel blends and water addition techniques are not conclusive. Thus, there is an urgent need for further investigations on their emission from diesel engines. From the literature review, it is seen that a number of studies have been carried out on oxygenated fuels in diesel engines on PM reduction from diesel engines. All studies have reported different percentages of reduction in PM emission depending on engine speed and blend ratios. However, the general agreement from the results discussed seems to decrease in PM emissions. Finally, the authors would like to conclude that, more research under controlled conditions should be encouraged, on the use of oxygenated additives such as water emulsion and ABE blends to accurately arrive at on the various emissions, especially the PM. Further, the various fuel properties of these kinds of fuels should be evaluated accurately and should be made available to researchers.

7 Oxygenated Fuel Additive Option for PM Emission Reduction …

155

Appendix: Influence of Various Oxygenated Fuels on NOx and PM Emissions Engine details

Test details

Blend details

NOx

PM

References

DI diesel gen. set 4 cylinders, water-cooled engine

Tests carried out under steady-state condition at 75% of maximum load Tests carried out under steady-state condition at 80% max. power output of 3.2 kW Tests carried out under steady-state condition with varying BMEP (bar) Tests carried out under steady-state condition at Idling, 1.6 and 3.2 kW

Water–diesel

–

Lin et al. (2006)

Water with NOE-7F-natural organic enzyme-7F + diesel

–

22.2– 44.0% # 31.1– 61.4% #

Butanol (10–25%)– diesel Butanol (10–25%)– water (0.5%)–diesel Butanol (10–25%)– water (1%)–diesel

9–21% #

4– 28.4% # 13–25% # 14–36% #

Lin et al. (2012)

Yilmaz et al. (2014)

Yanmar S. P. Co, Ltd, Thailand DI diesel, single cylinder, water-cooled engine

Kirloskar India DI diesel single cylinder. water-cooled engine

Yanmar S. P. Co, Ltd, Thailand DI diesel engine, single cylinder, water-cooled engine

13–25% # 14–36% #

Algae methyl esters (10%) + diesel

25% "

–

Algae methyl esters (15%) + diesel Algae methyl esters (20%) + diesel

27% "

–

34% "

–

Neat soybean oil (1%) + diesel Soybean Biodiesel + diesel Water-containing, acetone + isopropyl, alcohol + soybean oil Water-containing Acetone + isopropyl alcohol + soybean biodiesel Pure acetone + isopropyl alcohol + soybean oil Pure acetone isopropyl alcohol + soybean biodiesel

3.7–4.5% "

7.1– 22.7% # 7.9– 27.3% # 1.7– 31.8% #

1.8–2.5% " 1.9–13.9% #

9.1–15.7% #

14.3– 36.4% #

5.79–12.0% #

11.1– 31.8% #

8.26–13.9% #

14.3– 36.4% #

Zhou et al. (2014)

(continued)

156

P. Vijayashree and V. Ganesan

(continued) Engine details

Test details

Blend details

NOx

PM

References

Mitsubishi DI diesel, four cylinders, water-cooled engine

Tests carried out under steady-state condition at 50% load (rated power 12 kW and Torque 50 Nm) (i) Tests carried out under steady-state condition

2% microalgae biodiesel

2.0% "

22.0% #

Fraioli et al. (2014)

2% microalgae biodiesel + 20% butanol 2% microalgae biodiesel + 20% butanol + 0.5% water

25.0% #

57.2% #

28.2% #

59.5% #

B25A25

15.6–22.7% #

11.6– 15.8% #

B50A25

10.1–21.3% # 9.50–19.3% #

16.2– 22.7% # 10.9– 18.5% #

(i) Mitsubishi DI diesel, four cylinders, water-cooled engine

(ii) Cummins DI diesel turbocharged water-cooled engine, Diesel generator set, Anhui Quan Chai Group Corp. China): DI diesel, water-cooled engine Yanmar S. P. Co, Ltd, Thailand DI Diesel, single cylinder, water-cooled engine

B75A25 (ii) US-HDD transient cycle with Cummins B5.9-160 Tests carried out under steady-state condition: 125 Nm, (75% of maximum load)

Tests carried out under steady-state condition three modes of operations: Idle, 1.6, 3.2 kW loads

Chang et al. (2014)

B-biodiesel, A-Water-containing ABE solution

Soy biodiesel (10– 30%) Water with surfactant (10%) + soy biodiesel (10–20%) NOE-7F water (10%) + soy biodiesel (10–20%)

22.4– 57.5% " 57.5– 58.8% #

Lin et al. (2008)

87.7– 89.6% #

Biodiesel–diesel blends

0.8–14.1% "

Ethanol–biodiesel– diesel blends

0.9–10.9% #

5.0– 47.9% #; 3.7– 65.6% #

Lee et al. (2011)

(continued)

7 Oxygenated Fuel Additive Option for PM Emission Reduction …

157

(continued) Engine details

Test details

Blend details

Yanmar S. P. Co, Ltd, Thailand DI diesel engine single cylinder, water-cooled engine

Five modes of operation: 1500 rpm, 1800 rpm, 40 Nm, 80 Nm

ABE10 ABE20 ABE30 ABE10W0.5

0.8–14.1%

ABE20W0.5

0.9–10.9%

ABE30W0.5

Yanmar S. P. Co, Ltd, Thailand DI diesel, single cylinder, water-cooled engine

Not speciﬁed

Yanmar S. P. Co, Ltd, Thailand DI diesel, single cylinder, water-cooled engine

Not speciﬁed

NOx

ABE-acetone– butanol–ethanol W-water content Waste cooking oil biodiesel (1–70%) + diesel (99–30%) Biodiesohols waste cooking oil biodiesel (1%) + acetone (1–3%) + 1% isopropanol + diesel Biodiesohols waste cooking oil biodiesel (3%) + acetone (1–3%) + 1% isopropanol + diesel Biodiesohols waste cooking oil biodiesel (5%) + acetone (1–3%) + 1% isopropanol + diesel Biodiesohols waste cooking oil biodiesel (10%) + acetone (1–3%) + 1% isopropanol + diesel Biodiesohols waste cooking oil biodiesel (20%) + acetone (1–3%) + 1% isopropanol + diesel

PM

References

20.8– 27.0% 24.7– 43.5% 28.7– 48.8% 27.9– 39.0% 37.5– 61.1% 47.3– 62.5%

Lin et al. (2008)

# # # # # #

2.3– 35.5% #

Lee et al. (2011)

3.6– 15.3% #

10.5– 18.4% #

Lee et al. (2011)

11.4– 19.1% #

19.8– 27.5% #

35.4– 37.5% #

(continued)

158

P. Vijayashree and V. Ganesan

(continued) Engine details

Test details

Blend details

Diesel engine generator (QC495, Anhui QuanChai Group Corp. China): direct injection; water-cooled

Tests carried out under steady-state condition at 125 Nm— [75% of maximum torque]

Palm biodiesel blends (10–30%) + Diesel 16% BioSolutions + palm biodiesel (10– 30%) + diesel

NOx

PM

References

7.0– 45.5% # 88.6– 89.4% #

Chen et al. (2010)

" Increase # Decrease

References Agarwal AK (2007) Biofuels (alcohols and biodiesel) applications as fuels for internal combustion engines. Prog Energy Combust Sci 33:233–271 Agarwal D, Kumar L, Agarwal AK (2008) Performance evaluation of a vegetable oil fuelled compression ignition engine. Renewable Energy 33:1147–1156 Ahmed AS, Khan S, Hamdan S, Rahman R, Kalam A, Masjuki HH, et al (2010) Biodiesel production from macro algae as a green fuel for diesel engine. J Energy Environ 2(1):1–5 Alahmer A (2013) Influence of using emulsiﬁed diesel fuel on the performance and pollutants emitted from diesel engine. Energy Convers Management 73:361–369 Al-lwayzy SH, Yusaf T (2013) Chlorella protothecoides microalgae as an alternative fuel for tractor diesel engines. Energies, 6:766–783 Altın R, Çetinkaya S, Yücesu HS (2001) The potential of using vegetable oil fuels as fuel for diesel engines. Energy Convers Manage 42:529–538 Attia AMA, Kulchitskiy AR (2014) Influence of the structure of water-in-fuel emulsion on diesel engine performance. Fuel 116:703–708 Baruch JJ (2008) Combating global warming while enhancing the future. Technol Soc 30:111–121 Beatrice C, Di Blasio G, Guido C, Cannilla C, Bonura G, Frusteri F (2014) Mixture of glycerol ethers as diesel bio-derivable oxy-fuel: impact on combustion and emissions of an automotive engine combustion system. Appl Energy 132:236–247 Borrás E, Tortajada-Genaro LA, Vázquez M, Zielinska B (2009) Polycyclic aromatic hydrocarbon exhaust emissions from different reformulated diesel fuels and engine operating conditions. Atmos Environ 43:5944–5952 Campos-Fernandez J, Arnal JM, Gomez J, Lacalle N, Dorado MP (2013) Performance tests of a diesel engine fueled with pentanol/diesel fuel blends. Fuel 107:866–872 Campos-Fernández J, Arnal JM, Gómez J, Dorado MP (2012) A comparison of performance of higher alcohols/diesel fuel blends in a diesel engine. Appl Energy 95:267–275 Chang W-L (2010) Acetone–butanol–ethanol fermentation by engineered Clostridium beijerinckii and Clostridium tyrobutyricum. The Ohio State University, Columbus Chang Y-C, Lee W-J, Lin S-L, Wang L-C (2013) Green energy: water-containing acetone– butanol–ethanol diesel blends fueled in diesel engines. Appl Energy 109:182–191

7 Oxygenated Fuel Additive Option for PM Emission Reduction …

159

Chang Y-C, Lee W-J, Wang L-C, Yang H-H, Cheng M-T, Lu J-H et al (2014a) Effects of waste cooking oil-based biodiesel on the toxic organic pollutant emissions from a diesel engine. Appl Energy 113:631–638 Chang Y-C, Lee W-J, Wu TS, Wu C-Y, Chen S-J (2014b) Use of water containing acetone– butanol–ethanol for NOx–PM (nitrogen oxide–particulate matter) trade-off in the diesel engine fueled with biodiesel. Energy 64:678–687 Chang YC, Lee WJ, Yang HH, Wang LC, Lu JH, Tsai YI et al (2014c) Reducing emissions of persistent organic pollutants from a diesel engine by fueling with watercontaining butanol diesel blends. Environ Sci Technol 48:6010–6018 Chen K-S, Lin Y-C, Hsieh L-T, Lin L-F, Wu C-C (2010) Saving energy and reducing pollution by use of emulsiﬁed palm-biodiesel blends with bio-solution additive. Energy 35:2043–2048 Chisti Y, Yan J (2011) Energy from algae: current status and future trends: algal biofuels—a status report. Appl Energy 88:3277–3279 Claassen PA, Budde MA, López-Contreras AM (2000) Acetone, butanol and ethanol production from domestic organic waste by solventogenic clostridia. J Mol Microbiol Biotechnol 2:39–44 Correa SM, Arbilla G (2006) Aromatic hydrocarbons emissions in diesel and biodiesel exhaust. Atmos Environ 40:6821–6826 Demirbas A (2007a) Progress and recent trends in biofuels. Prog Energy Combust Sci 33:1–18 Demirbas A (2007b) Importance of biodiesel as transportation fuel. Energy Policy 35:4661–4670 Demirbas A (2008) Biofuels sources, biofuel policy, biofuel economy and global biofuel projections. Energy Convers Manage 49:2106–2116 Demirbas A (2009) Political, economic and environmental impacts of biofuels: a review. Appl Energy 86(Suppl 1):S108–S117 DeOliveira E, Quirino RL, Suarez PAZ, Prado AGS (2006) Heats of combustion of biofuels obtained by pyrolysis and by transesteriﬁcation and of biofuel/diesel blends. Thermochim Acta 450:87–90 Doğan, O (2011) The influence of n-butanol/diesel fuel blends utilization on a small diesel engine performance and emissions. Fuel 90:2467–2472 Dutta PK, Radner R (2009) A strategic analysis of global warming: theory and some numbers. J Econ Behav Organ 71:187–209 Fraioli V, Mancaruso E, Migliaccio M, Vaglieco BM (2014) Ethanol effect as premixed fuel in dual-fuel CI engines: experimental and numerical investigations. Appl Energy 119:394–404 Ghannam MT, Selim MYE (2009) Stability behavior of water-in-diesel fuel emulsion. Petrol Sci Technol 27:396–411 Gill SS, Tsolakis A, Herreros JM, York APE (2012) Diesel emissions improvements through the use of biodiesel or oxygenated blending components. Fuel 95:578–586 Guarieiro LLN, Pereira PADP, Torres EA, da Rocha GO, de Andrade JB (2008) Carbonyl compounds emitted by a diesel engine fuelled with diesel and biodiesel–diesel blends: sampling optimization and emissions proﬁle. Atmos Environ 42:8211–8218 Haik Y, Selim MYE, Abdulrehman T (2011) Combustion of algae oil methyl ester in an indirect injection diesel engine. Energy 36:1827–1835 Hariram V, Kumar MC (2013) Combustion analysis of algal oil methyl ester in a direct injection compression ignition engine JEng. Sci Technol 8:77–92 Heeb NV, Ulrich A, Emmenegger L, Czerwinski J, Mayer A, Wyser M (2005) Secondary emissions risk assessment of diesel particulate traps for heavy duty applications. The Automotive Research Association of India Heeb NV, Zennegg M, Gujer E, Honegger P, Zeyer K, Gfeller U et al (2007) Secondary effects of catalytic diesel particulate ﬁlters: copper-induced formation of PCDD/Fs. Environ Sci Technol 41:5789–5794 Heeb NV, Schmid P, Kohler M, Gujer E, Zennegg M, Wenger D et al (2008) Secondary effects of catalytic diesel particulate ﬁlters: conversion of PAHs versus formation of Nitro-PAHs. Environ Sci Technol 42:3773–3779 Heeb NV, Zennegg M, Haag R, Wichser A, Schmid P, Seiler C, et al (2013) PCDD/F formation in an iron/potassium-catalyzed diesel particle ﬁlter. Environ Sci Technol 47:6510–6517

160

P. Vijayashree and V. Ganesan

Hernández D, Fernández JJ, Mondragón F, López D (2012) Production and utilization performance of a glycerol derived additive for diesel engines. Fuel 92:130–136 Hsieh L-T, Wu EM-Y, Wang L-C, Chang-Chien G-P, Yeh Y-F (2011) Reduction of toxic pollutants emitted from heavy-duty diesel vehicles by deploying diesel particulate ﬁlters. Aerosol Air Qual Res 11:709–715 Huo M, Lin S, Liu H, Lee C-fF (2014) Study on the spray and combustion characteristics of water-emulsiﬁed diesel. Fuel 123:218–229 Hwang J, Qi D, Jung Y, Bae C (2013) Effect of injection parameters on the combustion and emission characteristics in a compression ignition engine fuelled with waste cooking oil biodiesel. SAE technical paper, 2013 IPCC (2007) Climate change 2007: synthesis report. In: Pachauri RK, Reisinger A (eds) Team TCW, Geneva, Switzerland: IPCC Islam MA, Rahman MM, Heimann K, Nabi MN, Ristovski ZD, Dowell A et al (2015) Combustion analysis of microalgae methyl ester in a common rail direct injection diesel engine. Fuel 143:351–360 Ithnin AM, Noge H, Abdul Kadir H, Jazair W (2014) An overview of utilizing water-in-diesel emulsion fuel in diesel engine and its potential research study. J Energy Inst 87:273–288 Izquierdo JF, Montiel M, Palés I, Outón PR, Galán M, Jutglar L et al (2012) Fuel additives from glycerol etheriﬁcation with light oleﬁns: state of the art. Renew Sustain Energy Rev 16:6717– 6724 Johnson TV (2009) Review of diesel emissions and control. Int J Eng Res 10:275–285 Kadota T, Yamasaki H (2002) Recent advances in the combustion of water fuel emulsion. Prog Energy Combust Sci 28:385–404 Knibbs LD, Cole-Hunter T, Morawska L (2011) A review of commuter exposure to ultraﬁne particles and its health effects. Atmos Environ 45:2611–2622 Lapuerta M, Armas O, Rodríguez-Fernández J (2008) Effect of biodiesel fuels on diesel engine emissions. Prog Energy Combust Sci 34:198–223 Lapuerta M, Rodríguez-Fernández J, Oliva F (2012) Effect of soot accumulation in a diesel particle ﬁlter on the combustion process and gaseous emissions. Energy 47:543–552 Lee W-J, Liu Y-C, Mwangi FK, Chen W-H, Lin S-L, Fukushima Y et al (2011) Assessment of energy performance and air pollutant emissions in a diesel engine generator fueled with water-containing ethanol–biodiesel–diesel blend of fuels. Energy 36:5591–5599 Lif A, Holmberg K (2006) Water-in-diesel emulsions and related systems. Adv Colloid Interface 123:231–239 Lin C-Y, Huang J-C (2003) An oxygenating additive for improving the performance and emission characteristics of marine diesel engines. Ocean Eng 30:1699–1715 Lin Y-C, Lee W-J, Chen C-C, Chen C-B (2006a) Saving energy and reducing emissions of both polycyclic aromatic hydrocarbons and particulate matter by adding bio-solution to emulsiﬁed diesel. Environ Sci Technol 40:5553–5559 Lin Y-C, Lee W-J, Hou H-C (2006b) PAH emissions and energy efﬁciency of palmbiodiesel blends fueled on diesel generator. Atmos Environ 40:3930–3940 Lin Y-C, Lee W-J, Chao H-R, Wang S-L, Tsou T-C, Chang-Chien G-P et al (2008) Approach for energy saving and pollution reducing by fueling diesel engines with emulsiﬁed biosolution/ biodiesel/diesel blends. Environ Sci Technol 42:3849–3855 Lin YC, Hsieh LT, Wu CC, Wu TY (2010) Characteristics of polychlorinated dibenzo-pdioxin/ dibenzofuran emissions from a biodiesel-fuelled diesel engine. Sustain Environ Res 20:409– 416 Lin Y-C, Liu S-H, Chen Y-M, Wu T-Y (2011a) A new alternative parafﬁnic–palmbiodiesel fuel for reducing polychlorinated dibenzo-p-dioxin/ dibenzofuran emissions from heavy-duty diesel engines. J Hazard Mater 185:1–7 Lin L, Cunshan Z, Vittayapadung S, Xiangqian S, Mingdong D (2011b) Opportunities and challenges for biodiesel fuel. Appl Energy 88:1020–1031

7 Oxygenated Fuel Additive Option for PM Emission Reduction …

161

Lin S-L, Lee W-J, Lee C-fF, Wu Y-p (2012) Reduction in emissions of nitrogen oxides, particulate matter, and polycyclic aromatic hydrocarbon by adding watercontaining butanol into a diesel-fueled engine generator. Fuel 93:364–372 Liu S-H, Lin Y-C, Hsu K-H (2012) Emissions of regulated pollutants and PAHs from waste-cooking-oil biodiesel-fuelled heavy-duty diesel engine with catalyzer. Aerosol Air Qual Res, 12:218–227 Ma F, Hanna MA (1999) Biodiesel production: a review. Bioresour Technol 70:1–15 Maddox I, Qureshi N, Roberts-Thomson K (1995) Production of acetone–butanol– ethanol from concentrated substrate using clostridium acetobutylicum in an integrated fermentation-product removal process. Process Biochem 30:209–215 Mayer A, Ulrich A, Heeb NV, Czerwinski J, Neubert T (2008) Particle ﬁlter properties after 2000 h real world operation. SAE international McClellan RO, Hesterberg TW, Wall JC (2012) Evaluation of carcinogenic hazard of diesel engine exhaust needs to consider revolutionary changes in diesel technology. Regul Toxicol Pharmacol: RTP 63:225–258 Misra RD, Murthy MS (2010) Straight vegetable oils usage in a compression ignition engine—a review. Renew Sustain Energy Rev 14:3005–3013 Nel WP, Cooper CJ (2009) Implications of fossil fuel constraints on economic growth and global warming. Energy Policy 37:166–180 Nelson PF, Tibbett AR, Day SJ (2008) Effects of vehicle type and fuel quality on real world toxic emissions from diesel vehicles. Atmos Environ 42:5291–5303 Ni Y, Sun Z (2009) Recent progress on industrial fermentative production of acetone–butanol– ethanol by Clostridium acetobutylicum in China. Appl Microbiol Biotechnol 83:415–423 Nithyanandan K, Wu H, Huo M, Lee C-F (2014) A preliminary investigation of the performance and emissions of a port-fuel injected SI engine fueled with Acetone–Butanol–Ethanol (ABE) and gasoline. SAE Technical Paper No S-Y (2011) Inedible vegetable oils and their derivatives for alternative diesel fuels in CI engines: a review. Renew Sustain Energy Rev 15:131–149 Öztürk E (2015) Performance, emissions, combustion and injection characteristics of a diesel engine fuelled with canola oil–hazelnut soapstock biodiesel mixture. Fuel Process Technol 129:183–191 Pang X, Shi X, Mu Y, He H, Shuai S, Chen H et al (2006) Characteristics of carbonyl compounds emission from a diesel-engine using biodiesel–ethanol–diesel as fuel. Atmos Environ 40:7057– 7065 Papastefanou C (2008) Chapter 1 atmospheric aerosol particles. In: Constantin P (ed) Radioactivity in the environment. Elsevier, p 3–9 Patil A, Taji S (2013) Effect of oxygenated fuel additive on diesel engine performance and emission: a review. IOSR J Mech Civ Eng (IOSR-JMCE), 30–35 Ponthein W, Cheirsilp B (2011) Development of Acetone Butanol Ethanol (ABE) production from palm pressed ﬁber by mixed culture of Clostridium sp. and Bacillus sp. Energy Procedia 9:459–467 Rakopoulos DC, Rakopoulos CD, Giakoumis EG, Dimaratos AM, Kyritsis DC (2010a) Effects of butanol–diesel fuel blends on the performance and emissions of a high-speed DI diesel engine. Energy Convers Manage 51:1989–1997 Rakopoulos CD, Dimaratos AM, Giakoumis EG, Rakopoulos DC (2010b) Investigating the emissions during acceleration of a turbocharged diesel engine operating with bio-diesel or n-butanol diesel fuel blends. Energy 35:5173–5184 Rakopoulos D, Rakopoulos C, Papagiannakis R, Kyritsis D (2011) Combustion heat release analysis of ethanol or n-butanol diesel fuel blends in heavy-duty DI diesel engine. Fuel 90:1855–1867 Ramadhas AS, Jayaraj S, Muraleedharan C (2004) Use of vegetable oils as I.C. engine fuels—a review. Renewable Energy 29:727–742

162

P. Vijayashree and V. Ganesan

Ratcliff MA, Dane AJ, Williams A, Ireland J, Luecke J, McCormick RL et al (2010) Diesel particle ﬁlter and fuel effects on heavy-duty diesel engine emissions. Environ Sci Technol 44:8343– 8349 Ren Y, Huang Z, Miao H, Di Y, Jiang D, Zeng K et al (2008) Combustion and emissions of a DI diesel engine fuelled with diesel-oxygenate blends. Fuel 87:2691–2697 Satyanarayanamurthy Y (2012) Experimental investigations of real time secondary co-injection of water-diethyl ether solution in di-diesel engine fuelled with palm kernel methyl ester. J Eng Sci Technol 7:711–721 Shi JP, Evans DE, Khan AA, Harrison RM (2001) Sources and concentration of nanoparticles (

Engine Exhaust Particulates

Recommend Stories

Idea Transcript

Helpful Links

Smile Life

Get in touch