Atmos Chem Phys 16 6319ndash6334 2016wwwatmos-chem-physnet1663192016doi105194acp-16-6319-2016copy Author(s) 2016 CC Attribution 30 License

Emission factors for gaseous and particulate pollutants fromoffshore diesel engine vessels in ChinaFan Zhang12478 Yingjun Chen1278 Chongguo Tian28 Diming Lou3 Jun Li5 Gan Zhang5 and Volker Matthias6

1Key Laboratory of Citiesrsquo Mitigation and Adaptation to Climate Change in Shanghai (China MeteorologicalAdministration) College of Environmental Science and Engineering Tongji University Shanghai 200092 PR China2Key Laboratory of Coastal Environmental Processes and Ecological Remediation Yantai Institute of Coastal Zone ResearchChinese Academy of Sciences Yantai Shandong 264003 PR China3School of Automobile Studies Tongji University Shanghai 201804 PR China4University of Chinese Academy of Sciences Beijing 100049 PR China5State key Laboratory of Organic Geochemistry Guangzhou Institute of Geochemistry Chinese Academy of SciencesGuangzhou Guangdong 510640 PR China6Helmholtz-Zentrum Geesthacht Institute of Coastal Research Max-Planck-Straszlige 1 21502 Geesthacht Germany7State Key Laboratory of Pollution Control and Resources Reuse College of Environmental Science and EngineeringTongji University Shanghai 200092 PR China8Shandong Provincial Key Laboratory of Coastal Environmental Processes YICCAS Yantai Shandong 264003 PR China

Correspondence to Yingjun Chen (yjchentjtongjieducn) and Chongguo Tian (cgtianyicaccn)

Received 7 July 2015 ndash Published in Atmos Chem Phys Discuss 1 September 2015Revised 20 April 2016 ndash Accepted 3 May 2016 ndash Published 24 May 2016

Abstract Shipping emissions have significant influence onatmospheric environment as well as human health espe-cially in coastal areas and the harbour districts Howeverthe contribution of shipping emissions on the environmentin China still need to be clarified especially based on mea-surement data with the large number ownership of vesselsand the rapid developments of ports international trade andshipbuilding industry Pollutants in the gaseous phase (car-bon monoxide sulfur dioxide nitrogen oxides total volatileorganic compounds) and particle phase (particulate mat-ter organic carbon elemental carbon sulfates nitrate am-monia metals) in the exhaust from three different diesel-engine-powered offshore vessels in China (350 600 and1600 kW) were measured in this study Concentrations fuel-based and power-based emission factors for various operat-ing modes as well as the impact of engine speed on emissionswere determined Observed concentrations and emission fac-tors for carbon monoxide nitrogen oxides total volatile or-ganic compounds and particulate matter were higher forthe low-engine-power vessel (HH) than for the two higher-engine-power vessels (XYH and DFH) for instance HHhad NOx EF (emission factor) of 258 g kWhminus1 compared

to 714 and 697 g kWhminus1 of DFH and XYH and PM EF of209 g kWhminus1 compared to 014 and 004 g kWhminus1 of DFHand XYH Average emission factors for all pollutants exceptsulfur dioxide in the low-engine-power engineering vessel(HH) were significantly higher than that of the previous stud-ies (such as 302 g kgminus1 fuel of CO EF compared to 217 to195 g kgminus1 fuel in previous studies 115 g kgminus1 fuel of NOx

EF compared to 223 to 87 g kgminus1 fuel in previous studies and940 g kgminus1 fuel of PM EF compared to 12 to 76 g kgminus1 fuelin previous studies) while for the two higher-engine-powervessels (DFH and XYH) most of the average emission fac-tors for pollutants were comparable to the results of the previ-ous studies engine type was one of the most important influ-ence factors for the differences Emission factors for all threevessels were significantly different during different operat-ing modes Organic carbon and elemental carbon were themain components of particulate matter while water-solubleions and elements were present in trace amounts The testinland ships and some test offshore vessels in China alwayshad higher EFs for CO NOx and PM than previous studiesBesides due to the significant influence of engine type onshipping emissions and that no accurate local EFs could be

Published by Copernicus Publications on behalf of the European Geosciences Union

6320 F Zhang et al Emission factors for gaseous and particulate pollutants

used in inventory calculation much more measurement datafor different vessels in China are still in urgent need Best-fit engine speeds during actual operation should be based onboth emission factors and economic costs

1 Introduction

Gaseous and particulate pollutants emitted from vessels oper-ating in the open ocean as well as in coastal areas and inlandwaterways have significant adverse impacts on human healthair quality and climate change (Cappa et al 2014 Righi etal 2011 Marmer and Langmann 2005 Winebrake et al2009) It has been estimated that 87 000 premature deaths oc-curred in 2012 due to burning of marine fuels with high sul-fur content Shipping-related particulate matter (PM) emis-sions have been reported to be responsible for approximately60 000 cardiopulmonary and lung cancer deaths annuallywith most cases occurring near coastlines in Europe (Vianaet al 2014) East Asia and South Asia (Corbett et al 2007)Approximately 9200 and 5200 t yearminus1 of PM are emittedfrom oceangoing and coastal ships respectively in the USA(Corbett and Fischbeck 2000) most of which are fine oreven ultrafine aerosols (Viana et al 2009 Saxe and Larsen2004) Globally about 15 of nitrogen oxides (NOx) and5ndash8 of sulfur oxides (SOx) emissions are attributable tooceangoing ships (Corbett and Fischbeck 2000) Shippingemissions affect acid deposition and ozone concentrationscontributing more than 200 mg S mminus2 yearminus1 over the south-western British Isles and Brittany as well as additional 6 ppbsurface ozone during the summer over Ireland (Derwent etal 2005) Moreover aerosol emissions from internationalshipping also greatly impact the Earthrsquos radiation budget di-rectly by scattering and absorbing solar radiation and indi-rectly by altering cloud properties (Righi et al 2011) Be-sides according to estimates from IMO (2014) total ship-ping emissions were approximately 938 million tonnes CO2and 961 million tonnes CO2e (CO2 equivalent) for GHGscombining CO2 CH4 and N2O for the year 2012 Interna-tional shipping emission accounts for approximately 22 and21 of global CO2 and GHG emissions on a CO2 equiva-lent (CO2e) basis respectively Because nearly 70 of shipemissions are estimated to occur within 400 km of land (En-dresen 2003) ships have the potential to contribute signif-icantly to air quality degradation in coastal areas In addi-tion ports are always the most concentrated areas for shipsto berth at emission reduction measures such as switchingfrom heavy fuels to cleaner fuels are required when ships areclose to ports or offshore areas but not all of them can obeythe regulations (De Meyer et al 2008) which results in sig-nificant influence on atmospheric environment of port citiesand regions

Rapid developments of ports international trade and theshipbuilding industry in China have negatively affected the

ambient air quality of the coastal zone due to shipping emis-sions It was estimated that 84 of SO2 and 113 ofNOx were emitted from ships in China in 2013 with portcities being the worst effect areas (httpnewsxinhuanetcompolitics2015-0608c_127890195htm) In 2013 therewere 018 million water transport vessels (Ministry of Trans-portation 2013) active in Chinese waters 8 ports in Chinawere listed among the worldrsquos largest 10 ports and 11 con-tainer ports were listed among the worldrsquos largest 20 con-tainer ports The number of ports with cargo handling capac-ity of more than 200 million t yearminus1 grew to 16 (Ministry ofTransportation 2010) Rapid development of ports in Chinahas resulted in increasingly serious pollution of ambient airparticularly in coastal zones and near ports Only a few stud-ies have focused on pollution from shipping emissions inChina Rough estimates of the influence of shipping emis-sions on ambient air in the port of Shanghai the largest portin China (Zhao et al 2013) and in the Bohai Rim (Zhang etal 2014) these estimates have been generated using empir-ical formulas One case study of real-world emissions of in-land vessels on the Grand Canal of China has been conducted(Fu et al 2013) and another study focused on inland shipsand several offshore vessels resulted in some rough emissiondata (Song 2015) Other studies also have developed to theinventories in large ports or delta regions (Zheng et al 2011Zheng et al 2009) by using EFs (emission factors) obtainedfrom other countries or areas However there are no system-atic studies of vessel emissions in the coastal zone or in portsnor accurate estimates of shipping emissions to ambient airbased on measured emission factors Conditions in China dif-fer substantially from those in other countries such as in ves-sel types (more small motor vessels and the type composi-tion of offshore vessels is shown in Table S1 in the Supple-ment more light-tonnage vessels e g with 793 less than3000 t in offshore area of Yangtze River Delta that is shownin Table S2) different fuel standards compared with othercountries (the GBT 17411-2012 standard with sulfur contentof less than 35 m m however the ISO 8217-2010 inter-national standard has the maximum sulfur content accordingto the relevant statutory requirements that always have lowervalues such as less than 01 in emission control areas be-sides a large percentage of diesel fuel was used in Chinaespecially in offshore areas seen in Table S3) The age ofvessels is also important (Chinese commercial vessels havean average age of 192 years compared with 80 years and89 years for Japan and Germany respectively) However alarge part of previous studies focused on emission of large-tonnage vessels such as cargo ships (Moldanova et al 2009Celo et al 2015) large marine ships (Khan et al 2013 Sip-pula et al 2014) tankers (Agrawal et al 2008b Winnes andFridell 2010) and so on whose fuel types were typically ofthe heavy oil variety and most of them had engines less than10 years Thus experimentally determined EFs for vesselsin other countries cannot be used directly to estimate ship-ping emissions and their contribution to ambient air quality

Atmos Chem Phys 16 6319ndash6334 2016 wwwatmos-chem-physnet1663192016

6320 F Zhang et al Emission factors for gaseous and particulate pollutants

used in inventory calculation much more measurement datafor different vessels in China are still in urgent need Best-fit engine speeds during actual operation should be based onboth emission factors and economic costs

1 Introduction

Gaseous and particulate pollutants emitted from vessels oper-ating in the open ocean as well as in coastal areas and inlandwaterways have significant adverse impacts on human healthair quality and climate change (Cappa et al 2014 Righi etal 2011 Marmer and Langmann 2005 Winebrake et al2009) It has been estimated that 87 000 premature deaths oc-curred in 2012 due to burning of marine fuels with high sul-fur content Shipping-related particulate matter (PM) emis-sions have been reported to be responsible for approximately60 000 cardiopulmonary and lung cancer deaths annuallywith most cases occurring near coastlines in Europe (Vianaet al 2014) East Asia and South Asia (Corbett et al 2007)Approximately 9200 and 5200 t yearminus1 of PM are emittedfrom oceangoing and coastal ships respectively in the USA(Corbett and Fischbeck 2000) most of which are fine oreven ultrafine aerosols (Viana et al 2009 Saxe and Larsen2004) Globally about 15 of nitrogen oxides (NOx) and5ndash8 of sulfur oxides (SOx) emissions are attributable tooceangoing ships (Corbett and Fischbeck 2000) Shippingemissions affect acid deposition and ozone concentrationscontributing more than 200 mg S mminus2 yearminus1 over the south-western British Isles and Brittany as well as additional 6 ppbsurface ozone during the summer over Ireland (Derwent etal 2005) Moreover aerosol emissions from internationalshipping also greatly impact the Earthrsquos radiation budget di-rectly by scattering and absorbing solar radiation and indi-rectly by altering cloud properties (Righi et al 2011) Be-sides according to estimates from IMO (2014) total ship-ping emissions were approximately 938 million tonnes CO2and 961 million tonnes CO2e (CO2 equivalent) for GHGscombining CO2 CH4 and N2O for the year 2012 Interna-tional shipping emission accounts for approximately 22 and21 of global CO2 and GHG emissions on a CO2 equiva-lent (CO2e) basis respectively Because nearly 70 of shipemissions are estimated to occur within 400 km of land (En-dresen 2003) ships have the potential to contribute signif-icantly to air quality degradation in coastal areas In addi-tion ports are always the most concentrated areas for shipsto berth at emission reduction measures such as switchingfrom heavy fuels to cleaner fuels are required when ships areclose to ports or offshore areas but not all of them can obeythe regulations (De Meyer et al 2008) which results in sig-nificant influence on atmospheric environment of port citiesand regions

Rapid developments of ports international trade and theshipbuilding industry in China have negatively affected the

ambient air quality of the coastal zone due to shipping emis-sions It was estimated that 84 of SO2 and 113 ofNOx were emitted from ships in China in 2013 with portcities being the worst effect areas (httpnewsxinhuanetcompolitics2015-0608c_127890195htm) In 2013 therewere 018 million water transport vessels (Ministry of Trans-portation 2013) active in Chinese waters 8 ports in Chinawere listed among the worldrsquos largest 10 ports and 11 con-tainer ports were listed among the worldrsquos largest 20 con-tainer ports The number of ports with cargo handling capac-ity of more than 200 million t yearminus1 grew to 16 (Ministry ofTransportation 2010) Rapid development of ports in Chinahas resulted in increasingly serious pollution of ambient airparticularly in coastal zones and near ports Only a few stud-ies have focused on pollution from shipping emissions inChina Rough estimates of the influence of shipping emis-sions on ambient air in the port of Shanghai the largest portin China (Zhao et al 2013) and in the Bohai Rim (Zhang etal 2014) these estimates have been generated using empir-ical formulas One case study of real-world emissions of in-land vessels on the Grand Canal of China has been conducted(Fu et al 2013) and another study focused on inland shipsand several offshore vessels resulted in some rough emissiondata (Song 2015) Other studies also have developed to theinventories in large ports or delta regions (Zheng et al 2011Zheng et al 2009) by using EFs (emission factors) obtainedfrom other countries or areas However there are no system-atic studies of vessel emissions in the coastal zone or in portsnor accurate estimates of shipping emissions to ambient airbased on measured emission factors Conditions in China dif-fer substantially from those in other countries such as in ves-sel types (more small motor vessels and the type composi-tion of offshore vessels is shown in Table S1 in the Supple-ment more light-tonnage vessels e g with 793 less than3000 t in offshore area of Yangtze River Delta that is shownin Table S2) different fuel standards compared with othercountries (the GBT 17411-2012 standard with sulfur contentof less than 35 m m however the ISO 8217-2010 inter-national standard has the maximum sulfur content accordingto the relevant statutory requirements that always have lowervalues such as less than 01 in emission control areas be-sides a large percentage of diesel fuel was used in Chinaespecially in offshore areas seen in Table S3) The age ofvessels is also important (Chinese commercial vessels havean average age of 192 years compared with 80 years and89 years for Japan and Germany respectively) However alarge part of previous studies focused on emission of large-tonnage vessels such as cargo ships (Moldanova et al 2009Celo et al 2015) large marine ships (Khan et al 2013 Sip-pula et al 2014) tankers (Agrawal et al 2008b Winnes andFridell 2010) and so on whose fuel types were typically ofthe heavy oil variety and most of them had engines less than10 years Thus experimentally determined EFs for vesselsin other countries cannot be used directly to estimate ship-ping emissions and their contribution to ambient air quality

Atmos Chem Phys 16 6319ndash6334 2016 wwwatmos-chem-physnet1663192016

F Zhang et al Emission factors for gaseous and particulate pollutants 6321

in China especially in offshore areas Systematical measure-ment EFs for different kinds of vessels in China is essential

Numerous studies of shipping emissions based on exper-imental measurements have been conducted since the In-ternational Maritime Organization (IMO) first began to ad-dress air pollution from vessels in 1996 particularly in de-veloped countries Most of these studies have been carriedout by performing tests on board the vessel from the exhaustpipe (Agrawal et al 2008b Murphy et al 2009 Fridell etal 2008 Juwono et al 2013 Moldanovaacute et al 2013) orby taking measurements within the exhaust plumes (Sinhaet al 2003 Chen et al 2005 Lack et al 2009 Murphyet al 2009 Berg et al 2012 Pirjola et al 2014 Petzoldet al 2008) NOx carbon monoxide (CO) sulfur dioxide(SO2) and PM are the main constituents of shipping emis-sions (Moldanova et al 2009 Williams et al 2009 Agrawalet al 2008a Poplawski et al 2011 Endresen 2003) thathave been quantified In addition black carbon (BC) (Lackand Corbett 2012 Sinha et al 2003 Moldanova et al 2009Corbett et al 2010) and cloud condensation nuclei (CCN)(Sinha et al 2003 Lack et al 2011) also have been re-ported in some studies Reported emission factors for COSO2 NOx PM and BC are in the ranges of 05ndash16 29ndash4422ndash109 03ndash76 and 013ndash018 g kgminus1 fuel respectively and02ndash62times 1016 particles kgminus1 fuel for CCN Besides charac-teristics of gaseous species and PM have attracted more at-tention recently (Anderson et al 2015 Celo et al 2015Mueller et al 2015 Reda et al 2015)

The IMO has set the emission limits for NOx and SOx inthe revised MARPOL (Maritime Agreement Regarding OilPollution) Annex VI rules (IMO 1998) Ships operating inthe emission control areas (ECAs) (the Baltic Sea the NorthSea the North America and the Caribbean) should use fu-els with a sulfur content of less than 01 m m since Jan-uary 2015 Even more stringent limits have been laid down insome national or regional regulations For example in someEU ports seagoing ships at berth are required to switch intofuels of under 01 m m sulfur since 2010 (The Councilof the European Union 1999) both marine gas oil and ma-rine diesel oil used in water area within 24 nautical miles ofcoastline in California should have a sulfur content of lessthan 01 m m since 2014 (California Code of RegulationTitles 13 and 17 2016) Emission standard of Tier II for NOx

set by MARPOL VI has been executed since January 2011 inECAs and more stringent rules of Tier III will be executedfrom January 2016 However in China no specific policy orlimit for shipping emissions has been implemented except inHong Kong which is making legislation about the limit of05 sulfur content fuel used when berth in the port from2015 But because of the serious air pollution currently inChina emission limits for the main sources such as vehicleexhaust coal combustion biomass combustion and fugitivedust have become more and more stringent A draft aimed atlimiting the emissions from marine engines set by Ministryof Environmental Protection is on soliciting opinions It has

set the limits of CO HC NOx and PM for different kinds ofvessels which are mainly based on the Directive 9768ECset by EU (European Union 2012) and 40 CFR part 1042 setby EPA (US Environmental Protection Agency 2009) In ad-dition an implementation plan has been released by the Min-istry of Transport of the Peoplersquos Republic of China in De-cember 2015 aiming to set shipping emission control areas toreduce SO2 emissions in China (Ministry of Transport of thePeoplersquos Republic of China 2015) All the regulations wereset mostly based on other directives and regulations Detailedmeasurement data will assist with further policy making itmore relevant to current situations of vessels

Average EFs are often used for shipping emissions inven-tories on large scales or in regional areas (Tzannatos 2010Eyring et al 2005) However to evaluate the effects of ship-ping emissions on air pollution in local areas such as nearports various ship speeds and operating modes should beconsidered including docking berthing and departing fromports etc Previous studies have confirmed that EFs are sig-nificantly different under various load conditions (Petzold etal 2010) or in different operating modes (Fu et al 2013Winnes and Fridell 2010) for individual vessels Thereforemore detailed measurements of EFs in different operatingmodes are necessary to better estimate the impacts of ship-ping emissions on the environment

In this study experimental data for three different diesel-engine-powered vessels were collected All pollutants weremeasured directly in the stack Gaseous emissions and PMfrom the diesel engines were the main targets including COcarbon dioxide (CO2) SO2 NOx total volatile organic com-pounds (TVOCs) and total suspended particulates (TSPs)Fuel-based EFs for the three vessels were calculated usingthe carbon balance method under different operating condi-tions In addition fuel-based average EFs as well as power-based average EFs to values reported in other studies and forother vessels were compared Finally the impacts of enginespeed on the EFs of NOx were evaluated

2 Experimental

21 Test vessels and fuel types

Initially it was hoped that the choice of measurement shipswould reflect the shipping fleet in general ie in terms ofengine type (engine speed and power output) fuel used en-gine age and mode of operation with more than 10 ves-sels planned to test However consideration was given to thepracticalities involved with the measurements ie installa-tion of sampling systems external conditions etc Besidestime and economic constraints weighed heavily and only sev-eral shipowners willing to participate in the project Thus thechosen vessels of different engine powers with diesel usedrepresent a compromise

wwwatmos-chem-physnet1663192016 Atmos Chem Phys 16 6319ndash6334 2016

6322 F Zhang et al Emission factors for gaseous and particulate pollutants

Table 1 Technical parameters of test vessels

Vessel Vessel Displacement Ship length Engine Vessel Rated FuelID type (ton) timeswidth power age speed consumption

(m) (kw) (year) (rpm) rate (g KWhminus1)

HH Engineering vessel 307 44times 13 350times 2 4 1200 200DFH Research vessel 3235 96times 15 1600times 2 18 900 200XYH Research vessel 602 55times 9 600 5 1000 200

Three different diesel-engine-powered offshore vesselsincluding one engineering vessel Haohai 0007 (HH) withlow-power and high-speed engine one large research ves-sel Dongfanghong 2 (DFH) with high-power and medium-speed engine and another research vessel Xiangyanghong08 (XYH) with medium-power and medium-speed enginewere selected for this study their technical parameters areshown in Table 1 High-speed and medium-speed engines arethe predominant engines used in vessels of offshore and in-land rivers in China which always use light diesel as fuelTwo of the test vessels were small motor light-tonnage ves-sels (HH and XYH) and another one was a medium-speedengine vessel with an 18 year-old engine Engineering ves-sels are designed for construction activities such as build-ing docks in port areas or waterways dredging etc Theyare common vessels in coastal areas of China because of theheavy demand for oilfield construction and port expansionThe maintenance of engineering vessels is typically poorerthan for other types of vessels and as a result they may haverelatively high emissions On the other hand research vesselsof DFH and XYH from universities and research institutesare generally well maintained and use high-quality diesel fuelbut with different engine powers which might have relativelylow emission factors for pollution Therefore these researchvessels can reflect the impact of engine power on emissionsand also can represent the lower end of expected EFs for Chi-nese vessels The test vessels in the present study could ac-count for 347 of the total vessels according to the distri-bution of vessels through gross tonnage in China (seen in Ta-ble S2) which could have a certain degree of representationEverything considered a general range of EFs for gaseousand PM pollutants emitted from different offshore vessels ofChina and their influence factors could be given through theon-board measurement

The fuels used in all test vessels were common diesel fuelsobtained from fuelling stations near the ports According tostatistical data the total oil consumption of vessels in Chinawas 2099 million tons in 2011 including 1099 million tonsbonded oil and 593 million tons domestic trade oil withlight fuel oil accounting for 40 of the domestic trade oiland 25 of the total consumption (shown in Table S3) (Zhu2013) The test vessels in the present study could reflect theemission condition of diesel vessels in China especially inoffshore areas where diesel oil always been used as fuel Re-

Table 2 Results from the fuel analysis (diesels)

Units HH DFH XYH

Total calorific value MJ kgminus1 4544 4540 4550Net calorific value MJ kgminus1 4251 4248 4255Ash content m 0001 lt 0001 lt 0001Sulfur (S) m 00798 00458 0130Carbon (C) m 8666 8640 8649Hydrogen (H) m 1332 1322 1344Nitrogen (N) m lt 02 lt 02 lt 02Oxygen (O) m lt 04 lt 04 lt 04

sults of fuel analyses are presented in Table 2 All of thesefuels had relatively low sulfur content (le 013 m) and lowmetals concentrations (V Al Si Pb Zn Mn etc)

22 Test operating modes

EFs are significantly different under differing load condi-tions and operating modes Vessel speed is also an impor-tant influence factor for emissions it was reported by Star-crest Consulting Group LLC (Starcrest Consulting Group2012) that 15ndash20 of fuel consumption could be reducedby 10 of the vessel speed In this study vessel operatingmodes were classified according to actual sailing conditionsThere were six modes of HH low speed (4 knots) mediumspeed (8 knots) high speed (11 knots) acceleration processmoderating process and idling four modes of DFH cruise(10 knots medium speed for DFH) acceleration processmoderating process and idling and five modes of XYH lowspeed (3 knots) high speed (10 knots) acceleration processmoderating process and idling Three to five sets of replicatesamples were collected for each operating mode

23 Emissions measurement system and chemicalanalysis of particulate matter

A combined on-board emissions test system (Fig 1) wasused to measure emissions from the coastal vessels underactual operating conditions There was no dilution in thistest system with all the species measured directly from theexhaust and there were four main components of the sys-tem a flue gas analyser three particulate samplers an eight-stage particulate sampler and a TVOCs analyser (see Sup-

Atmos Chem Phys 16 6319ndash6334 2016 wwwatmos-chem-physnet1663192016

F Zhang et al Emission factors for gaseous and particulate pollutants 6323

Flue gas

Flue gas analyzer

(PhotonⅡ)

O2 NO

2 NO N

2O CO

CO2 SO

2 HC

Particulate samplers

Quartz fiber filter (QFF)

OC EC soluble ions inorganic elements

Glass filter (GF)

Polycyclic aromatic hydrocarbon (PAHs)

Eight-stage particulate

sampler (MARK Ⅲ)

Particle size

Polytetra- fluoroethylene

filter (PTFE) PM weight

TVOCs analyzer (MiniRAE 2000)

TVOCs

Figure 1 On-board emissions test system and measured analytes

plement for more details) All analytes are also shown inFig 1 the flue gas analyser (Photon II) is aimed to test in-stantaneous emissions of gaseous pollution including O2NO2 NO N2O CO CO2 and SO2 (detection parametersfor the gaseous matter are shown in Table S4) Three par-ticulate samplers are installed to collect PM using differentfilters at the same time including a quartz fiber filter glassfilter and polytetrafluoroethylene filter to analyse differentchemical components of PM And the portable TVOCs anal-yser is used to monitor the concentration of total VOCs withisobutylene as correction coefficient gas Besides a temper-ature sensor is installed near the smoke outlet to test the fluegas temperature A total of 33 sets of samples for HH 20sets for DFH and 23 sets for XYH were collected with 3 to5 sets for each operating mode

The OC (organic carbon) and EC (elemental carbon) weremeasured on a 0544 cm2 quartz filter punched from each fil-ter by thermal optical reflectance (TOR) following the IM-PROVE protocol with a DRI Model 2001 ThermalOpticalCarbon Analyzer (Atmoslytic Inc Calabasas CA) Themeasuring range of TOR was from 005 to 750 microg C cmminus2

with an error of less than 10 Concentrations of water-soluble ions in PM25 such as Na+ NH+4 K+ Mg2+ Ca2+Clminus NOminus3 and SO2minus

4 were determined by ion chromatogra-phy (Dionex ICS3000 Dionex Ltd America) based on themeasurement method of Shahsavani et al (2012) The de-tection limit was 10 ng mLminus1 with an error of less than 5 and 1 mL RbBr with concentration of 200 ppm was put in thesolution as internal standard before sampling The concentra-tions of 33 inorganic elements in PM25 were estimated usinga inductively coupled plasma mass spectrometer (ICP-MS ofELAN DRC II type PerkinElmer Ltd Hong Kong) follow-ing the standard method (Wang et al 2006) The resolution

of ICP-MS ranged from 03 to 30 amu with a detection limitlower than 001 ng mLminus1 and the error was less than 5

24 Data analysis

Carbon balance formula was used to calculate the EFs for allexhaust gas components It was assumed that all carbon in thefuel was emitted as carbon-containing gases (CO CO2 andTVOC) and carbon-containing particulate matter So therewas a certain equilibrium relationship between the carbon inthe fuel and in the exhaust

CF = RFGtimes (c(CCO)+ c(CCO2)

+ c(CPM)+ c(CTVOC)) (1)

where CF represents the mass of C in per kg diesel fuel(g C kgminus1 fuel) RFG represents the flue gas emissionsrate (m3 kgminus1 fuel) and c(CCO) c(CCO2) c(CPM) andc(CTVOC) represent the mass concentrations of carbon asCO CO2 PM and TVOC (g C mminus3) in the flue gas respec-tively

The EF for CO2 was calculated as follows

EFCO2 = RFG middot c(CO2) middotMCO2 (2)

where EFCO2 is the EF for CO2 (g kgminus1 fuel) c(CO2) isthe molar concentration of CO2 (mol mminus3) and MCO2 is themolecular weight of CO2 (44 g molminus1)

The remaining EFs were calculated as follows

EFx =1X

1CO2middot

MX

MCO2

middotEFCO2 (3)

where EFx is the EF for species X (g kgminus1 fuel) 1X and1CO2 represent the concentrations of X and CO2 with thebackground concentrations subtracted (mol mminus3) and MX

represents the molecular weight of species X (g molminus1)In addition average EFs for each vessel were calculated

based on actual operating conditions as follows

EFXA =sum

XiEFi timesPi (4)

where EFXA is the average EF for species X EFi is the EFfor operating mode i for species X and Pi is the percentageof time spent in operating mode i during the shipping cycle

Power-based emission factors and fuel-based emissionfactors could be interconverted with the formula as follow-ing

EFXP = EFx middotFCR (5)

where EFXP is the power-based emission factor forspecies X (g kW hminus1) and FCR is fuel consumption rate foreach vessel (kg fuel (kW h)minus1)

wwwatmos-chem-physnet1663192016 Atmos Chem Phys 16 6319ndash6334 2016

6324 F Zhang et al Emission factors for gaseous and particulate pollutants

3 Results and discussion

31 Concentrations in shipping emissions

Concentrations of CO NOx SO2 TVOC and PM fromthe three vessels are shown in Fig S1 in the SupplementNearly all of the concentrations measured in the exhaustof low-engine-power vessel HH were higher than those ofthe two higher-engine-power vessels Concentrations of COSO2 and NOx from HH were 107ndash756 534ndash331 and878ndash1295 ppm respectively and 143ndash595 mg mminus3 PM Incontrast concentrations of CO SO2 NOx and PM were501ndash141 527ndash169 169ndash800 ppm and 706ndash218 mg mminus3respectively for DFH and 360ndash224 049ndash359 and 235ndash578 ppm and 056ndash631 mg mminus3 respectively for XYH

A previous study demonstrated that concentrations of COprimarily depend on engine power with higher CO emissionsresulting from vessel engines with lower power (Sinha et al2003) There was a similar trend in this study with gener-ally higher concentrations for HH and lower concentrationsfor DFH The CO concentrations in the present study weresimilar but slightly lower than those of inland vessels (Fuet al 2013) except in the idling mode of HH In differentoperating modes CO concentrations were significantly dif-ferent For example the maximum value was observed inidling mode and the minimum value in medium-speed modefor HH All three ships had the lowest CO concentrations attheir economic speeds (medium speed for HH cruise modefor DFH and high speed for XYH) demonstrating that theirengines are optimized for the most common operating mode

More than 80 of the NOx was NO in this study withNO2 and N2O accounting for lt 20 in all operating modes(Fig S1) Again nearly all of these concentrations werehigher in the exhaust gas of HH than in that of the two ves-sels In high-speed modes all of the vessels had high con-centrations of NOx NOx emissions mainly depend on thecombustion temperature of the engines More powerful com-bustion systems operate at higher temperatures thereby pro-ducing more NOx (Corbett et al 1999) However the NOx

emissions were much lower than for the inland vessels stud-ied by Fu et al (2013) particularly in cruise mode (NOx con-centrations of sim 1000 ppm)

SO2 concentrations in the exhaust gas depend on the sul-fur content of the fuel and the flow rate of the flue gas Therewere significant differences among the three vessels in theirflow rates which could account for the different concentra-tions of one vessel in different operating modes But becauseof the low-sulfur fuels used in these vessels the SO2 con-centrations were low compared with those in other studies(Williams et al 2009 Berg et al 2012)

Much lower concentrations of PM in the exhaust gas wereobserved in the present study compared to those of inlandships in China (Fu et al 2013) However they were similarto those from ships at berth reported by Cooper et al (Cooper2003) HH had higher PM concentrations than the two ves-

sels in the exhaust gas There were significant differencesamong the different operating modes because of changes inthe injection point of the engines (Sippula et al 2014 Li etal 2014)

32 Fuel-based emission factors

Fuel-based EFs for the gaseous species CO2 CO NO NO2N2O and TVOCs and for PM based on the carbon balancemethod were determined In addition SO2 was calculatedbased on the sulfur content of the fuels Fuel-based EFs forthe typical pollutants such as CO PM and nitrogen oxides indifferent operating modes are shown in Fig 2 (detailed EFsfor all the gaseous pollutants are shown in Table S5 and de-tailed EFs for PM and its chemical composition are shown inTable S6)

CO2 emissions from vessels primarily depend on the car-bon content of the fuel (Carlton et al 1995) Accordinglythe EFs for CO2 in the present study should theoreticallybe 3177 3168 and 3171 g kgminus1 fuel for complete combus-tion Under actual conditions CO2 emissions were 2940ndash3106 3121ndash3160 and 3102ndash3162 g kgminus1 fuel for HH DFHand XYH respectively which means they had combustionefficiencies with 925ndash978 985ndash997 and 978ndash997 interms of CO2 for these three vessels

CO emissions of HH were much higher than of XYH fol-lowed by DFH The power of their respective engines was350 600 and 1600 kW In addition there were large differ-ences in CO emissions among different modes All of thesethree vessels had relatively high EFs for CO while accelerat-ing compared with other modes but the highest EFs wereduring the idling modes of HH and DFH as well as dur-ing the low-speed mode of XYH Because CO emissionsin diesel engines primarily depend on the excess air ratio(which determines the fuelndashair mixture) combustion temper-ature and uniformity of the fuelndashair mixture in the combus-tion chamber (Doug 2004) ship engines with lower powergenerally have higher CO emissions (Carlton et al 1995)Localized hypoxia and incomplete combustion in the cylin-der were the main reasons for CO emission of diesel engineCO emissions always had positive relationships with the air-to-fuel ratio There was lower air-to-fuel ratio when in lowengine load which resulted in lower CO emission and viceversa (Ni 1999)

Much higher NOx EFs were observed for HH than forthe other two vessels These results were inconsistent withthose of Sinha et al (2003) in which emissions of NOx in-creased with the power of the ship engine With increasingvessel speed NOx EFs for HH first increased and then de-creased XYH had lower EFs when operating at high speedthan at low speed Nitrogen oxides included NO NO2 andN2O in the present study More than 70 of the NOx wasin the form of NO for all vessels because most of the NOx

emissions were generated through thermal NO formation(Haglind 2008) The primary reasons that slow diesel en-

Atmos Chem Phys 16 6319ndash6334 2016 wwwatmos-chem-physnet1663192016

F Zhang et al Emission factors for gaseous and particulate pollutants 6325

Figure 2 EFs for the typical pollutants in different operating modes

gines such as the one in HH have higher NOx emissions in-clude higher peak flame temperatures and the NO formationreactions being closer to their equilibrium state than in otherengines (Haglind 2008) NOx emissions from vessels aretemperature-dependent (Sinha et al 2003) and also are in-fluenced by the oxygen concentration in the engine cylinder(Ni 1999) In larger engines the running speed is generallyslower and the combustion process more adiabatic resultingin higher combustion temperatures and more NOx Besideswith the increasing of air-to-fuel ratios concentration of NOx

showed a tendency first to increase then to decrease whichalways had the maximum value in the operating mode thatclose to full load of engine because of the high temperatureand oxygen in the engine cylinder (Ni 1999) Furthermorethere were always higher EF values in acceleration processand lower in moderating process in this study When the en-gines were in transient operating conditions such as acceler-ation process or moderating process concentrations of NOx

always had corresponding changes in the cylinder Studiesabout diesel engines showed that when the rotational speedhad a sudden increase there would be a first increasing then

decreasing and last stable tendency for the NOx concentra-tions and vice versa (Tan et al 2012)

TVOCs emissions from HH were much higher than fromthe other two vessels the lowest emissions were observedfor DFH Previous studies (Sinha et al 2003) have reportedthat hydrocarbon emissions from vessels depend on enginepower with low-power engines emitting more hydrocarbonsThe present results were partially consistent with these pre-vious studies Besides hydrocarbon emissions also dependon the percentage utilization of engine power (Sinha et al2003) As for various operating modes TVOCs EFs had largedifferences For example HH had the highest TVOCs emis-sions in accelerating mode which was almost 3 times theheight of the lowest value in mediumndashspeed mode The EFsfor SO2 depended solely on the sulfur content of the fuelsand were 16 09 and 26 g kgminus1 fuel for HH DFH andXYH respectively in this study Hydrocarbon could be gen-erated because of the incomplete combustion For examplein diesel cylinders there will always be air present in wallregions and crevices this is also the case when scavenging

wwwatmos-chem-physnet1663192016 Atmos Chem Phys 16 6319ndash6334 2016

6326 F Zhang et al Emission factors for gaseous and particulate pollutants

occurred during the aeration which could cause the unevenmixing of air and fuel (Ni 1999)

Fuel-based EFs for PM and its chemical components wereshown in Table S6 OC and EC were the main componentsof PM followed by SO2minus

4 NH+4 and NOminus3 Metals such asV Ni Cr Fe As and Cd made up a proportionately smallpart of the total PM mass However other rare elements suchas Tb Er Yb and Lu had higher values than did some ofthe common elements PM was an in-process product duringthe combustion in a cylinder the forming process includedthe molecular cracking decomposition and polymerizationwhich resulted in lack of oxygen High temperature and oxy-gen deficiency were the main reasons for the formation indiesel engines which always had high concentration val-ues in high load operating modes (Ni 1999) HH had muchhigher PM emission factors than the other two vessels theengine type was considered to be the most significant influ-ence factor which had a good agreement with NOx emissionfactors

EFs for OC and EC and the ratios between them are shownin Fig 3 EFs for OC and EC for HH were higher than forthe other two vessels Organic matter (OM) is generally cal-culated as OCtimes 12 (Petzold et al 2008) to account for themass of elements other than carbon in the emitted moleculesOM EFs for individual vessels mainly depend on the enginetype and the amount of unburned fuel ie the efficiency ofcombustion (Moldanovaacute et al 2013) BC emissions also de-pend heavily on the engine type (Lack et al 2009) There-fore the different types of engines and their levels of mainte-nance could account for the large differences in OC and ECEFs observed among the three vessels in this study The ratiosof OC-to-EC in the present study were much lower than thosefor large diesel ships reported previously (OC EC= 12)(Moldanova et al 2009) and also lower than that reportedfor a medium-speed vessel (Petzold et al 2010) The usageof non-dilution sampling in this study was one possible rea-son for the lower OC to EC ratio Besides TOR was used tomeasure OC and EC in PM which always had a lower OCcontent compared with other methods (such as TOT) becauseof the different definitions of OC and EC (Khan et al 2012)Compared with other diesel engines the ratios of OC to ECin this study were higher than that of automobile diesel sootin which EC comprises 75ndash80 wt of the total PM (Clagueet al 1999) and also higher than heavy heavy-duty dieseltrucks (HHDDTs) with OC to EC ratios below unit for cruseand transient modes even though higher in cold-startidle andcreep modes (Shah et al 2004)

Studies have shown that SO2minus4 formed from vessel-emitted

SO2 is a major contributor to CCN and ship track forma-tion (Schreier et al 2006 Lauer et al 2007) Sulfate is alsoan important component of PM emitted from vessels In thepresent study EFs for SO2minus

4 were much lower than previ-ously reported (Petzold et al 2008 Agrawal et al 2008a)but similar to those detected by a high-resolution time-of-flight aerosol mass spectrometer in a previous study (Lack

Figure 3 EFs for OC and EC and the ratios between them

et al 2009) This may be because EFs for SO2minus4 are mainly

related to the sulfur content of the fuel SO2minus4 is not gen-

erally emitted directly from the engines but forms after re-lease from the stack (Lack et al 2009) Because PM was col-lected directly from engine emissions in the present study thesulfur-to-sulfate ratios were low (lt 06 for vessels) Otherions such as NOminus3 and NH+4 accounted for a small percent-age of the PM emitted from the vessels compared with SO2minus

4 consistent with previous studies (Lack et al 2009) SO2 ismore easily oxidized to SO3 in catalytic reaction cycles withmetals commonly present in the exhaust gas (V Ni) whilehydroxyl radicals are additional needed to convert NOx toNOminus3 (Moldanova et al 2009)

Na+ and Clminus were considered to originate from marineair Their concentrations were highly correlated (r2

= 078)the differing air demands of the engines under different con-ditions might have caused observed variations in the EFs rel-ative to the fuel demand

The elemental composition of PM in the present studydiffered from previous studies showing high elemental con-tent of S Ca V Fe Cu Ni and Al (Agrawal et al 2008aMoldanova et al 2009) V and Ni are typically associatedwith combustion of heavy fuel oil (Almeida et al 2005) Inthe present study the high-quality fuels resulted in low EFsfor V and Ni In our previous study PM from shipping emis-sions was estimated to account for 294 of the total PM25at Tuoji Island in China using V as a tracer of shipping emis-sions (Zhang et al 2014) Reconsidering the former resultsbased on the EFs obtained in the present study we deter-mined that the contribution of vessels near Tuoji Island hadbeen underestimated because the estimate should have in-cluded both heavy and other types of fuels However somerare elements such as Tb Er Yb and Lu had relatively highEFs compared with those of other elements in the presentstudy which may be related to the source of the fuels

Atmos Chem Phys 16 6319ndash6334 2016 wwwatmos-chem-physnet1663192016

F Zhang et al Emission factors for gaseous and particulate pollutants 6327

Tabl

e3

Fuel

-bas

edav

erag

eE

Fsin

the

pres

ents

tudy

and

prev

ious

stud

ies

(gkgminus

1fu

el)

Ves

selI

DC

O2

CO

NO

NO

2N

2ON

Ox

TV

OC

sPM

SO2

Sco

nten

t(

m)

HH

3071plusmn

1565

302plusmn

162

982plusmn

372

155plusmn

545

128plusmn

170

115plusmn

443

237plusmn

210

940plusmn

213

160

008

DFH

3153plusmn

176

693plusmn

100

302plusmn

160

509plusmn

042

038plusmn

018

357plusmn

220

124plusmn

004

072plusmn

033

092

005

XY

H31

51plusmn

175

920plusmn

295

266plusmn

163

471plusmn

042

030plusmn

015

316plusmn

220

418plusmn

015

016plusmn

007

260

013

Com

mer

cial

vess

el(W

illia

ms

etal

20

09)

3170

7ndash16

ndashndash

ndash60

ndash87

ndashndash

6ndash30

Car

gove

ssel

(Mol

dano

vaet

al

2009

)

3441

217

ndashndash

ndash73

4ndash

53

393

19

Die

sele

ngin

e(H

aglin

d20

08)

ndash7

4ndash

ndashndash

87ndash

76

542

7

Oce

an-g

oing

ship

s(S

inha

etal

20

03)

3135

195

ndashndash

ndash22

3ndash

ndash2

90

1

Oce

an-g

oing

ship

s(S

inha

etal

20

03)

3176

30

ndashndash

ndash65

5ndash

ndash52

22

4

Car

goan

dpa

ssen

ger

ship

s(E

ndre

sen

2003

)31

707

4ndash

ndash0

0857

ndash87

24

12ndash

76

10ndash5

40

5ndash2

7

Ship

sop

erat

ing

inndash

ndash42

ndash72

ndashndash

65ndash8

6ndash

ndash4

6ndash9

8N

ON

Eha

rbou

rare

as(P

irjo

laet

al

2014

)ndash

ndash16

ndash49

ndashndash

25ndash7

9ndash

ndash5

4ndash17

0SC

R

Ship

sop

erat

ing

inPo

rt(D

iesc

het

al

2013

)ndash

ndash16

37ndash

53ndash

ndash7

7

NO

NE=

No

trea

tmen

tofe

mis

sion

sSC

R=

sele

ctiv

eca

taly

ticre

duct

ion

wwwatmos-chem-physnet1663192016 Atmos Chem Phys 16 6319ndash6334 2016

6328 F Zhang et al Emission factors for gaseous and particulate pollutants

33 Fuel-based average emission factors

Based on actual operating conditions (Table S7) average EFsfor the three vessels in the present study (according to Eq 4)along with EFs from previous studies are shown in Table 3EFs for all of the pollutants except SO2 were significantlyhigher for HH than for the other two vessels potentially dueto poor combustion conditions Most of the EFs for DFH andXYH were within the range of emissions for other vesselsdue to having well maintained engines and the high qualityof the fuels used The EFs for NOx PM and SO2 were muchlower than reported in previous studies (other than NOx forocean-going vessels) All the sulfur of the fuels in the presentstudy were significantly below the emissions limit of 350 established by IMO in the revised MARPOL Annex VI rulesapplicable since 2012 (IMO 1998)

The IMO Tier I emissions limit for NOx is450times nminus02 g kWhminus1 (n rated speed 130 lt nlt 2000 rpm)The rated speed and fuel consumption rates for each vesselare shown in Table 1 Thus the emissions limits for HHDFH and XYH would be 545 575 and 565 g kgminus1 fuelrespectively calculated combined with Eq (5) The averagefuel-based EFs for NOx of ship HH was more than 100 above the IMO standard while those of the other two shipswere below the IMO standard (Table 3) PM emissions forHH were also higher than previously reported but thosefor the two research vessels were much lower (Table 3)Fuel type is one of the most important influence factors onpollutant emissions for example sulfur content in the fuelnot only influence the SO2 emission directly but also hadimpact on PM formation in the flue gas stack with low sulfurcontent in fuels reduces PM formation (Lack et al 2011)Vessels with higher sulfur content always had relativelyhigher PM emissions which were also shown in Table 3In addition different engines and levels of maintenancehave a significant impact on all combustion-dependentemissions Emission reduction measures have been used insome vessels For example NOx emissions can be reducedby measures such as selective catalytic reduction (SCR) anddirect water injection (DWI) which had been implementedon some vessels previously studied in a harbour in Finland(Pirjola et al 2014) The results showed that SCR effec-tively reduced NOx emissions while vessels with DWI hadhigh PM emissions The engine type might be an importantcause of the different emissions such as HH had muchhigher pollutants emissions with an engine produced inChina and yet DFHrsquos engine produced in Germany Besidesemission tests for a high-speed marine diesel engine withdifferent kinds of diesels showed that diesel type had limitedinfluence on emissions such as NOx CO and CH but asignificant impact on PM emission (289ndash415 ) becauseof the different sulfur content in fuel (Xu 2008)

34 Power-based emission factors

Based on the engine power and fuel consumption rates ofthe vessels power-based EFs were calculated (according toEq 5) and compared to results from previous studies (Ta-ble 4) The EFs for HH were much higher than those forthe other two vessels except for SO2 HH also had signif-icantly higher EFs for NOx than previously reported val-ues while EFs for NOx of DFH and XYH were within therange of previously reported results Engine type was consid-ered to be a significant influence factor for NOx emissionswith lower engine speed having higher NOx emission factors(Celo et al 2015) In addition compared to other vesselswith a similar engine type and diesel fuel HH still had rela-tively higher NOx EF (seen in Table 4) which could reflectthe impact of engine condition (engine quality and mainte-nance level) on shipping emissions CO EFs for the test ves-sels in the present study were higher than previous studieswhich produced similar results to those of inland ships andother test vessels in China (Fu et al 2013 Song 2015) Inspite of the influence of engine type on CO emissions thatwith the higher engine speed having higher CO EFs (Celoet al 2015) engine condition combined with fuel qualitymight have significant influence All of the EFs for SO2 inthe present study were lower than those in previous studiesbecause of the low sulfur content of the present fuels Gen-erally PM emissions from marine diesel fuels are dependenton the fuel (sulfate and metal oxide ash constituents) and oncombustion conditions (unburned hydrocarbons and carbonresidue constituents) (Cooper 2003) HH had the highest PMemissions among the test vessels although there were almostno differences among the fuels (Table S6) Besides HH hadeven higher PM EFs than previously reported vessels withHFO fuel and XYH had much lower PM EFs than all theother vessels with even lower sulfur content fuel Thereforecombustion conditions were likely the determining factor forthe differences It can be seen from Table 4 that most previ-ous studies focused on the heavy fuel oil of shipping emis-sions Compared with diesel fuels heavy fuel oil always hadrelatively low CO emission factors and high PM emissionfactors And among the heavy-fuel-oil-using vessels enginetype (engine speed and engine power level) always playedan important role on emissions such as NOx and CO whichwith lower engine speed having higher NOx EFs and lowerCO EFs

Combined with other emission data of test ships in China(Fu et al 2013 Song 2015) it could be seen that inlandand some test offshore ships in China always had higherNOx CO and PM emissions compared with other test ves-sels in previous studies And among the test vessels in Chinathere also were differences for different engine types andship types In addition emission factors that were used forcalculation of ship inventories in China always came fromother countries and areas However there seemed to be sig-nificant differences between the reference and test data such

Atmos Chem Phys 16 6319ndash6334 2016 wwwatmos-chem-physnet1663192016

F Zhang et al Emission factors for gaseous and particulate pollutants 6329

Tabl

e4

Pow

er-b

ased

EFs

inth

epr

esen

tstu

dyan

dpr

evio

usst

udie

s(g

kWhminus

1 )

Ves

sel

CO

2C

ON

ON

O2

N2O

NO

xT

VO

Cs

PMSO

2E

ngin

epo

wer

Eng

ine

Fuel

type

and

ID(k

W)

spee

dsu

lfur

cont

ent

(rm

p)(w

t)

HH

699plusmn

352

738plusmn

376

220plusmn

841

345plusmn

124

030plusmn

039

258plusmn

100

544plusmn

484

209plusmn

048

036

350

1200

DO

00

798

DFH

631plusmn

352

139plusmn

020

604plusmn

032

102plusmn

008

008plusmn

004

714plusmn

044

017plusmn

001

014plusmn

007

018

1600

900

DO

00

458

XY

H69

7plusmn

385

201plusmn

065

587plusmn

036

104plusmn

009

007plusmn

003

697plusmn

048

092plusmn

ndash0

04plusmn

001

057

600

1000

DO

01

30Ta

nker

(Win

nes

and

Frid

ell

2010

)ndash

161

ndashndash

ndash7

82ndash

058

ndash45

0060

0H

FO1

6

Ber

thed

ship

s(C

oope

r65

3ndash76

80

33ndash0

97

ndashndash

ndash14

2ndash2

02

ndash0

14ndash0

45

026

ndash53

AE

720

ndash127

072

0ndash18

00M

GO

00

6ndash1

220

03)

691ndash

803

077

ndash17

112

9ndash1

75

048

ndash06

72

5ndash9

612

70ndash2

675

720ndash

750

RO

05

3ndash2

269

1ndash69

40

92ndash0

98

96ndash

99

017

ndash01

91

014

8072

0M

DO

02

3C

rude

oilt

anke

r(A

graw

alet

al

2008

b)58

8ndash66

00

77ndash1

78

ndashndash

ndash15

8ndash2

10

ndash1

10ndash1

78

766

ndash86

015

750

90H

FO2

85

Cru

ise

ship

s(P

opla

wsk

iet

al

2011

)ndash

ndashndash

140

ndashndash

ndash2

914

20

US

EPA

621

14

ndashndash

ndash18

1ndash

131

103

Mar

ine

engi

ne(S

ippu

laet

ndash1

2ndash11

411

3ndash2

95

ndashndash

114

ndash30

90ndash

95

083

ndash63

6ndash

1500

HFO

27

al

2014

)0ndash

885

69ndash2

58

584

ndash33

90

83ndash1

97

015

ndash09

3D

FL

arge

mar

ine

ship

s(K

han

etal

20

13)

533ndash

612

035

ndash06

0ndash

ndashndash

166

ndash20

6ndash

091

ndash21

97

2ndash11

436

740ndash

6853

097

ndash102

HFO

21

5ndash3

14

Oce

ango

ing

cont

aine

rve

ssel

(Agr

awal

etal

20

08a)

588ndash

660

077

ndash18

1ndash

ndashndash

158

ndash21

0ndash

109

ndash17

67

66ndash8

60

5027

010

4H

FO2

05

Lar

geca

rgo

vess

el(M

olda

nova

etal

20

09)

667

042

ndashndash

ndash14

22

ndash1

0310

320

200

97H

FO1

9

Oce

ango

ing

carg

ove

ssel

614plusmn

10

83plusmn

001

ndashndash

ndash16

3plusmn

02

ndash1

51plusmn

007

87plusmn

01

128

IFO

180

(Cel

oet

al

2015

)62

6plusmn

70

26plusmn

001

114plusmn

01

081plusmn

002

58plusmn

007

525

IFO

180

628plusmn

90

30plusmn

001

113plusmn

01

094plusmn

002

87plusmn

01

450

IFO

180

628plusmn

10

81plusmn

003

122plusmn

001

083plusmn

001

64plusmn

01

450

IFO

180

609plusmn

11

31plusmn

002

ndashndash

ndash8

4plusmn

003

ndash0

37plusmn

001

47plusmn

001

440

IFO

6060

5plusmn

10

00ndash

ndashndash

167plusmn

01

ndash2

2plusmn

02

103plusmn

003

116

IFO

380

622plusmn

11

22plusmn

002

ndashndash

ndash10

7plusmn

004

ndash0

30plusmn

003

047plusmn

01

450

MD

O

AE

aux

iliar

yen

gine

DO

die

selo

ilH

FOh

eavy

fuel

oil

MG

Om

arin

ega

soi

lR

Or

esid

ualo

ilM

DO

mar

ine

dies

eloi

lD

Fdi

esel

fuel

IFO

int

erm

edia

tefu

eloi

l

wwwatmos-chem-physnet1663192016 Atmos Chem Phys 16 6319ndash6334 2016

6330 F Zhang et al Emission factors for gaseous and particulate pollutants

as 100 to 132 g kW hminus1 of NOx EF and 11 to 17 g kW hminus1

of CO EF used for inland ships for ship inventory calcula-tion (Zhu et al 2015) 100 to 181 g kW hminus1 of NOx EFand 11 to 15 g kW hminus1 of CO EF for harbour ships (Yanget al 2015) compared to 15 to 173 g kW hminus1 of NOx EFand 46 to 103 g kW hminus1 of CO EF from test inland ships(convert the fuel-based EF to power-based EF with a factorof 200 g kW hminus1) (Song 2015) and 697 to 258 g kW hminus1 ofNOx EF and 139 to 738 g kW hminus1 of CO EF in the presentstudy (Yang et al 2015) Besides whether there are obviousdifferences of EFs between other types of vessels in China(such as low-speed engine vessels with heavy fuel oil) andprevious studies is still unclear Therefore much more mea-surement data for different vessels in China are still in urgentneed for more accurate assessment of shipping emissions

35 Impact of engine speed on NOx emission factors

NOx is formed in the combustion chamber by a combinationof atmospheric nitrogen and oxygen under high-pressure andhigh-temperature conditions Many factors affect NOx for-mation including engine temperature injection point andfuel quality The IMO emissions limit for NOx is determinedby the rated speed of the engine however other factors mustalso be considered to reduce NOx emissions

The NOx EFs for the test vessels at various engine speedsare shown in Fig 4 The rated speeds of the vessels were1200 900 and 1000 rpm for HH DFH and XYH respec-tively The actual engine speeds of HH were much lower thanthe rated speed while the two larger-engine-power vesselsoperated close to their rated speeds except during one op-erating mode of DFH The NOx EFs for HH differed sig-nificantly in different operating modes ranging from 391to 143 g kgminus1 of fuel The NOx EF was highest when theengine speed reached sim 750 rpm (Fig 4) At lower enginespeeds the NOx EFs had fluctuating but lower values Athigher engine speeds closer to the rated speed of 1200 rpmthe NOx EFs were much lower The NOx EFs for the twolarger-engine-power vessels changed slightly with enginespeed but also had lowest values when their engine speedsapproached their rated speeds Combined with the dieselpropulsion characteristic curve there were large increasesin the fuel consumption rate when the engine speed in-creased Therefore a best-fit engine speed should be deter-mined based on both EFs and economic costs

Engineering approaches for reducing the NOx emissionsof marine engines may be applied before during or afterthe combustion process (Verschaeren et al 2014 Habib etal 2014) In the present study the NOx EFs of the two re-search vessels were below the IMO Tier I emissions limitsHowever for EMS measures should be taken to meet theIMO emissions limit including increasing the engine speedand applying engineering technologies during or after com-bustion such as exhaust gas recirculation (EGR) selectivenon-catalytic reduction (SNCR) or SCR

Figure 4 Emission factors for NOx at different engine speeds

4 Conclusions

Three offshore vessels with different engine power sourceswere chosen in this study to collect measured data of gaseousspecies and particulate matter including NO2 NO N2OCO CO2 TVOCs SO2 and the total suspended particulateBesides chemical composition of the PM were also analysedto give detailed EFs for OC EC water-soluble ions and metalelements Concentrations fuel-based EFs fuel-based aver-age EFs as well as power-based average EFs for species ofoffshore vessels in China were presented Furthermore im-pact of engine speed on NOx EFs was also discussed

There were higher concentrations of pollutants for low-engine-power vessel HH than for the other two vessels COconcentrations for offshore vessels were slightly lower thaninland vessels in China and all the three vessels had the low-est CO concentrations at their economic speeds (the speed ofthe least vessel operating expenditures during one voyagethey were high-speed mode cruise mode and high-speedmode for HH DFH and XYH respectively) More than 80 of the NOx was NO and all the offshore vessels had higherNOx concentrations in high-speed modes Because of thelow-sulfur fuels used in this study SO2 concentrations ofthese three offshore vessels were lower than that in the lit-eratures And the PM concentrations were much lower thaninland vessels while showing significant differences amongdifferent operating modes

Fuel-based EFs for gaseous species and PM were pre-sented based on the carbon balance method EFs for CO2were 2940ndash3106 3121ndash3160 and 3102ndash3162 g kgminus1 fuel forHH DFH and XYH Because of the combustion conditionssuch as excess air ratio combustion temperature and unifor-mity of the fuelndashair mixture EFs for CO showed high valuesin idling mode but low values in economic speed All theoffshore vessels had higher NOx EFs in low speed than inhigh speed but showed higher values when in acceleration

Atmos Chem Phys 16 6319ndash6334 2016 wwwatmos-chem-physnet1663192016

F Zhang et al Emission factors for gaseous and particulate pollutants 6331

process EFs for SO2 were 16 09 and 26 g kgminus1 fuel forHH DFH and XYH based on sulfur content of the fuels OCand EC were the main components of PM with low OC toEC ratios that were lower than 01 followed by SO2minus

4 NH+4 and NOminus3 Metals such as V Ni Cr Fe As and Cd made upa proportionately small part of the total PM mass

Fuel-based average EFs as well as power-based EFs for thethree different vessels of differing engine power were pre-sented EFs for most gaseous species and PM of HH weremuch higher compared with the other higher-engine-powervessels which was also gt 100 above the IMO standard forNOx Average PM EF of the low-engine-power vessel HHwas also much higher than that in the literatures Howeveraverage EFs for most species of the two larger-engine-powervessels were within the range of previously reported resultsEngine type was inferred as one of the most influence fac-tors for the differences of emission factors Inland and someoffshore ships in China always had higher NOx CO and PMemissions compared with other test vessels in previous stud-ies In addition emission factors that used for calculation ofship inventories in China always had lower values than testvessels

The impact of engine speed on EFs for NOx showed thatwhen the engine speed was close to the rated speed therewould be lower NOx EFs values However combined withthe high fuel consumption rate an optimal engine speedshould be determined based on both EFs and economic costsEmission reduction measures for NOx for some of the off-shore vessels in China are still essential to meet the IMOemission limit

Given the limits of vessel types and numbers this studysubstantially gives the EFs for gaseous species and PM ofthree different diesel-engine-powered offshore vessels How-ever as the development of ports in China emissions fromcargo ships and container ships with large engine power havebecome one of the most significant air pollution sources inport cities and regions Systematical measurement EFs of allkinds of offshore vessels in China are essential in order topresent the accurate emission inventory of ships

Information about the Supplement

Supplementary information includes the details of the real-world measurement system for vessels (Fig S1) the concen-trations of main gaseous matter and PM of shipping emis-sions (Fig S2) the types composition of offshore vessels inChina (Table S1) the distribution of vessels through grosstonnage in 2014 in offshore area of Yangtze River Delta (Ta-ble S2) the Chinese market consumption of marine oil in2011 (Table S3) the detection parameters for gaseous mat-ter (Table S4) the fuel-based EFs for the gaseous pollutants(Table S5) PM and the chemical composition in PM for dif-ferent operating (Table S6) modes and the actual operatingconditions of vessels (Table S7)

The Supplement related to this article is available onlineat doi105194acp-16-6319-2016-supplement

Acknowledgements This study was supported by the CAS Strate-gic Priority Research Program (No XDB05030303) the NaturalScientific Foundation of China (Nos 41273135 and 41473091)and the Shanghai science and Technology Commission project(No 15DZ1205401) The authors would like to express theirgratitude to the vessel owners Haohai Ocean Engineering LimitedLiability Company for vessel Haohai 0007 Ocean University ofChina for vessel Dongfanghong 02 and North China Sea Branchof the State Oceanic Administration for vessel Xiangyanghong08 for making their vessels available for the shipping emissionsstudy The views expressed in this document are solely those of theauthors and the funding agencies do not endorse any products orcommercial services mentioned in this publication

Edited by T Bond

References

Agrawal H Malloy Q G J Welch W A Miller J W andCocker D R In-use gaseous and particulate matter emissionsfrom a modern ocean going container vessel Atmos Environ42 5504ndash5510 doi101016jatmosenv200802053 2008a

Agrawal H Welch W A Miller J W and Cocker D R Emis-sion measurements from a crude oil tanker at sea Environ SciTechnol 42 7098ndash7103 doi101021es703102y 2008b

Almeida S M Pio C A Freitas M C Reis MA and Trancoso M A Source apportionment of fineand coarse particulate matter in a sub-urban area at theWestern European Coast Atmos Environ 39 3127ndash3138doi101016jatmosenv200501048 2005

Anderson M Salo K Hallquist A M and Fridell ECharacterization of particles from a marine engine op-erating at low loads Atmos Environ 101 65ndash71doi101016jatmosenv201411009 2015

Berg N Mellqvist J Jalkanen J-P and Balzani J Ship emis-sions of SO2 and NO2 DOAS measurements from airborne plat-forms Atmos Meas Tech 5 1085ndash1098 doi105194amt-5-1085-2012 2012

California Code of Regulation Titles 13 and 17 avail-able at httpsgovtwestlawcomcalregsBrowseHomeCaliforniaCaliforniaCodeofRegulationstransitionType=DefaultampcontextData=(scDefault) last access 20 April 2016

Cappa C D Williams E J Lack D A Buffaloe G M Coff-man D Hayden K L Herndon S C Lerner B M Li S-M Massoli P McLaren R Nuaaman I Onasch T B andQuinn P K A case study into the measurement of ship emis-sions from plume intercepts of the NOAA ship Miller FreemanAtmos Chem Phys 14 1337ndash1352 doi105194acp-14-1337-2014 2014

Carlton J Danton S Gawen R Lavender K Mathieson NNewell A Reynolds G Webster A Wills C and Wright AMarine exhaust emissions research programme Lloydrsquos RegisterEngineering Services London 63 1995

wwwatmos-chem-physnet1663192016 Atmos Chem Phys 16 6319ndash6334 2016

6322 F Zhang et al Emission factors for gaseous and particulate pollutants

Table 1 Technical parameters of test vessels

Vessel Vessel Displacement Ship length Engine Vessel Rated FuelID type (ton) timeswidth power age speed consumption

(m) (kw) (year) (rpm) rate (g KWhminus1)

HH Engineering vessel 307 44times 13 350times 2 4 1200 200DFH Research vessel 3235 96times 15 1600times 2 18 900 200XYH Research vessel 602 55times 9 600 5 1000 200

Three different diesel-engine-powered offshore vesselsincluding one engineering vessel Haohai 0007 (HH) withlow-power and high-speed engine one large research ves-sel Dongfanghong 2 (DFH) with high-power and medium-speed engine and another research vessel Xiangyanghong08 (XYH) with medium-power and medium-speed enginewere selected for this study their technical parameters areshown in Table 1 High-speed and medium-speed engines arethe predominant engines used in vessels of offshore and in-land rivers in China which always use light diesel as fuelTwo of the test vessels were small motor light-tonnage ves-sels (HH and XYH) and another one was a medium-speedengine vessel with an 18 year-old engine Engineering ves-sels are designed for construction activities such as build-ing docks in port areas or waterways dredging etc Theyare common vessels in coastal areas of China because of theheavy demand for oilfield construction and port expansionThe maintenance of engineering vessels is typically poorerthan for other types of vessels and as a result they may haverelatively high emissions On the other hand research vesselsof DFH and XYH from universities and research institutesare generally well maintained and use high-quality diesel fuelbut with different engine powers which might have relativelylow emission factors for pollution Therefore these researchvessels can reflect the impact of engine power on emissionsand also can represent the lower end of expected EFs for Chi-nese vessels The test vessels in the present study could ac-count for 347 of the total vessels according to the distri-bution of vessels through gross tonnage in China (seen in Ta-ble S2) which could have a certain degree of representationEverything considered a general range of EFs for gaseousand PM pollutants emitted from different offshore vessels ofChina and their influence factors could be given through theon-board measurement

The fuels used in all test vessels were common diesel fuelsobtained from fuelling stations near the ports According tostatistical data the total oil consumption of vessels in Chinawas 2099 million tons in 2011 including 1099 million tonsbonded oil and 593 million tons domestic trade oil withlight fuel oil accounting for 40 of the domestic trade oiland 25 of the total consumption (shown in Table S3) (Zhu2013) The test vessels in the present study could reflect theemission condition of diesel vessels in China especially inoffshore areas where diesel oil always been used as fuel Re-

Table 2 Results from the fuel analysis (diesels)

Units HH DFH XYH

Total calorific value MJ kgminus1 4544 4540 4550Net calorific value MJ kgminus1 4251 4248 4255Ash content m 0001 lt 0001 lt 0001Sulfur (S) m 00798 00458 0130Carbon (C) m 8666 8640 8649Hydrogen (H) m 1332 1322 1344Nitrogen (N) m lt 02 lt 02 lt 02Oxygen (O) m lt 04 lt 04 lt 04

sults of fuel analyses are presented in Table 2 All of thesefuels had relatively low sulfur content (le 013 m) and lowmetals concentrations (V Al Si Pb Zn Mn etc)

22 Test operating modes

EFs are significantly different under differing load condi-tions and operating modes Vessel speed is also an impor-tant influence factor for emissions it was reported by Star-crest Consulting Group LLC (Starcrest Consulting Group2012) that 15ndash20 of fuel consumption could be reducedby 10 of the vessel speed In this study vessel operatingmodes were classified according to actual sailing conditionsThere were six modes of HH low speed (4 knots) mediumspeed (8 knots) high speed (11 knots) acceleration processmoderating process and idling four modes of DFH cruise(10 knots medium speed for DFH) acceleration processmoderating process and idling and five modes of XYH lowspeed (3 knots) high speed (10 knots) acceleration processmoderating process and idling Three to five sets of replicatesamples were collected for each operating mode

23 Emissions measurement system and chemicalanalysis of particulate matter

A combined on-board emissions test system (Fig 1) wasused to measure emissions from the coastal vessels underactual operating conditions There was no dilution in thistest system with all the species measured directly from theexhaust and there were four main components of the sys-tem a flue gas analyser three particulate samplers an eight-stage particulate sampler and a TVOCs analyser (see Sup-

Atmos Chem Phys 16 6319ndash6334 2016 wwwatmos-chem-physnet1663192016

F Zhang et al Emission factors for gaseous and particulate pollutants 6323

Flue gas

Flue gas analyzer

(PhotonⅡ)

O2 NO

2 NO N

2O CO

CO2 SO

2 HC

Particulate samplers

Quartz fiber filter (QFF)

OC EC soluble ions inorganic elements

Glass filter (GF)

Polycyclic aromatic hydrocarbon (PAHs)

Eight-stage particulate

sampler (MARK Ⅲ)

Particle size

Polytetra- fluoroethylene

filter (PTFE) PM weight

TVOCs analyzer (MiniRAE 2000)

TVOCs

Figure 1 On-board emissions test system and measured analytes

plement for more details) All analytes are also shown inFig 1 the flue gas analyser (Photon II) is aimed to test in-stantaneous emissions of gaseous pollution including O2NO2 NO N2O CO CO2 and SO2 (detection parametersfor the gaseous matter are shown in Table S4) Three par-ticulate samplers are installed to collect PM using differentfilters at the same time including a quartz fiber filter glassfilter and polytetrafluoroethylene filter to analyse differentchemical components of PM And the portable TVOCs anal-yser is used to monitor the concentration of total VOCs withisobutylene as correction coefficient gas Besides a temper-ature sensor is installed near the smoke outlet to test the fluegas temperature A total of 33 sets of samples for HH 20sets for DFH and 23 sets for XYH were collected with 3 to5 sets for each operating mode

The OC (organic carbon) and EC (elemental carbon) weremeasured on a 0544 cm2 quartz filter punched from each fil-ter by thermal optical reflectance (TOR) following the IM-PROVE protocol with a DRI Model 2001 ThermalOpticalCarbon Analyzer (Atmoslytic Inc Calabasas CA) Themeasuring range of TOR was from 005 to 750 microg C cmminus2

with an error of less than 10 Concentrations of water-soluble ions in PM25 such as Na+ NH+4 K+ Mg2+ Ca2+Clminus NOminus3 and SO2minus

4 were determined by ion chromatogra-phy (Dionex ICS3000 Dionex Ltd America) based on themeasurement method of Shahsavani et al (2012) The de-tection limit was 10 ng mLminus1 with an error of less than 5 and 1 mL RbBr with concentration of 200 ppm was put in thesolution as internal standard before sampling The concentra-tions of 33 inorganic elements in PM25 were estimated usinga inductively coupled plasma mass spectrometer (ICP-MS ofELAN DRC II type PerkinElmer Ltd Hong Kong) follow-ing the standard method (Wang et al 2006) The resolution

of ICP-MS ranged from 03 to 30 amu with a detection limitlower than 001 ng mLminus1 and the error was less than 5

24 Data analysis

Carbon balance formula was used to calculate the EFs for allexhaust gas components It was assumed that all carbon in thefuel was emitted as carbon-containing gases (CO CO2 andTVOC) and carbon-containing particulate matter So therewas a certain equilibrium relationship between the carbon inthe fuel and in the exhaust

CF = RFGtimes (c(CCO)+ c(CCO2)

+ c(CPM)+ c(CTVOC)) (1)

where CF represents the mass of C in per kg diesel fuel(g C kgminus1 fuel) RFG represents the flue gas emissionsrate (m3 kgminus1 fuel) and c(CCO) c(CCO2) c(CPM) andc(CTVOC) represent the mass concentrations of carbon asCO CO2 PM and TVOC (g C mminus3) in the flue gas respec-tively

The EF for CO2 was calculated as follows

EFCO2 = RFG middot c(CO2) middotMCO2 (2)

where EFCO2 is the EF for CO2 (g kgminus1 fuel) c(CO2) isthe molar concentration of CO2 (mol mminus3) and MCO2 is themolecular weight of CO2 (44 g molminus1)

The remaining EFs were calculated as follows

EFx =1X

1CO2middot

MX

MCO2

middotEFCO2 (3)

where EFx is the EF for species X (g kgminus1 fuel) 1X and1CO2 represent the concentrations of X and CO2 with thebackground concentrations subtracted (mol mminus3) and MX

represents the molecular weight of species X (g molminus1)In addition average EFs for each vessel were calculated

based on actual operating conditions as follows

EFXA =sum

XiEFi timesPi (4)

where EFXA is the average EF for species X EFi is the EFfor operating mode i for species X and Pi is the percentageof time spent in operating mode i during the shipping cycle

Power-based emission factors and fuel-based emissionfactors could be interconverted with the formula as follow-ing

EFXP = EFx middotFCR (5)

where EFXP is the power-based emission factor forspecies X (g kW hminus1) and FCR is fuel consumption rate foreach vessel (kg fuel (kW h)minus1)

wwwatmos-chem-physnet1663192016 Atmos Chem Phys 16 6319ndash6334 2016

6324 F Zhang et al Emission factors for gaseous and particulate pollutants

3 Results and discussion

31 Concentrations in shipping emissions

Concentrations of CO NOx SO2 TVOC and PM fromthe three vessels are shown in Fig S1 in the SupplementNearly all of the concentrations measured in the exhaustof low-engine-power vessel HH were higher than those ofthe two higher-engine-power vessels Concentrations of COSO2 and NOx from HH were 107ndash756 534ndash331 and878ndash1295 ppm respectively and 143ndash595 mg mminus3 PM Incontrast concentrations of CO SO2 NOx and PM were501ndash141 527ndash169 169ndash800 ppm and 706ndash218 mg mminus3respectively for DFH and 360ndash224 049ndash359 and 235ndash578 ppm and 056ndash631 mg mminus3 respectively for XYH

A previous study demonstrated that concentrations of COprimarily depend on engine power with higher CO emissionsresulting from vessel engines with lower power (Sinha et al2003) There was a similar trend in this study with gener-ally higher concentrations for HH and lower concentrationsfor DFH The CO concentrations in the present study weresimilar but slightly lower than those of inland vessels (Fuet al 2013) except in the idling mode of HH In differentoperating modes CO concentrations were significantly dif-ferent For example the maximum value was observed inidling mode and the minimum value in medium-speed modefor HH All three ships had the lowest CO concentrations attheir economic speeds (medium speed for HH cruise modefor DFH and high speed for XYH) demonstrating that theirengines are optimized for the most common operating mode

More than 80 of the NOx was NO in this study withNO2 and N2O accounting for lt 20 in all operating modes(Fig S1) Again nearly all of these concentrations werehigher in the exhaust gas of HH than in that of the two ves-sels In high-speed modes all of the vessels had high con-centrations of NOx NOx emissions mainly depend on thecombustion temperature of the engines More powerful com-bustion systems operate at higher temperatures thereby pro-ducing more NOx (Corbett et al 1999) However the NOx

emissions were much lower than for the inland vessels stud-ied by Fu et al (2013) particularly in cruise mode (NOx con-centrations of sim 1000 ppm)

SO2 concentrations in the exhaust gas depend on the sul-fur content of the fuel and the flow rate of the flue gas Therewere significant differences among the three vessels in theirflow rates which could account for the different concentra-tions of one vessel in different operating modes But becauseof the low-sulfur fuels used in these vessels the SO2 con-centrations were low compared with those in other studies(Williams et al 2009 Berg et al 2012)

Much lower concentrations of PM in the exhaust gas wereobserved in the present study compared to those of inlandships in China (Fu et al 2013) However they were similarto those from ships at berth reported by Cooper et al (Cooper2003) HH had higher PM concentrations than the two ves-

sels in the exhaust gas There were significant differencesamong the different operating modes because of changes inthe injection point of the engines (Sippula et al 2014 Li etal 2014)

32 Fuel-based emission factors

Fuel-based EFs for the gaseous species CO2 CO NO NO2N2O and TVOCs and for PM based on the carbon balancemethod were determined In addition SO2 was calculatedbased on the sulfur content of the fuels Fuel-based EFs forthe typical pollutants such as CO PM and nitrogen oxides indifferent operating modes are shown in Fig 2 (detailed EFsfor all the gaseous pollutants are shown in Table S5 and de-tailed EFs for PM and its chemical composition are shown inTable S6)

CO2 emissions from vessels primarily depend on the car-bon content of the fuel (Carlton et al 1995) Accordinglythe EFs for CO2 in the present study should theoreticallybe 3177 3168 and 3171 g kgminus1 fuel for complete combus-tion Under actual conditions CO2 emissions were 2940ndash3106 3121ndash3160 and 3102ndash3162 g kgminus1 fuel for HH DFHand XYH respectively which means they had combustionefficiencies with 925ndash978 985ndash997 and 978ndash997 interms of CO2 for these three vessels

CO emissions of HH were much higher than of XYH fol-lowed by DFH The power of their respective engines was350 600 and 1600 kW In addition there were large differ-ences in CO emissions among different modes All of thesethree vessels had relatively high EFs for CO while accelerat-ing compared with other modes but the highest EFs wereduring the idling modes of HH and DFH as well as dur-ing the low-speed mode of XYH Because CO emissionsin diesel engines primarily depend on the excess air ratio(which determines the fuelndashair mixture) combustion temper-ature and uniformity of the fuelndashair mixture in the combus-tion chamber (Doug 2004) ship engines with lower powergenerally have higher CO emissions (Carlton et al 1995)Localized hypoxia and incomplete combustion in the cylin-der were the main reasons for CO emission of diesel engineCO emissions always had positive relationships with the air-to-fuel ratio There was lower air-to-fuel ratio when in lowengine load which resulted in lower CO emission and viceversa (Ni 1999)

Much higher NOx EFs were observed for HH than forthe other two vessels These results were inconsistent withthose of Sinha et al (2003) in which emissions of NOx in-creased with the power of the ship engine With increasingvessel speed NOx EFs for HH first increased and then de-creased XYH had lower EFs when operating at high speedthan at low speed Nitrogen oxides included NO NO2 andN2O in the present study More than 70 of the NOx wasin the form of NO for all vessels because most of the NOx

emissions were generated through thermal NO formation(Haglind 2008) The primary reasons that slow diesel en-

Atmos Chem Phys 16 6319ndash6334 2016 wwwatmos-chem-physnet1663192016

F Zhang et al Emission factors for gaseous and particulate pollutants 6325

Figure 2 EFs for the typical pollutants in different operating modes

gines such as the one in HH have higher NOx emissions in-clude higher peak flame temperatures and the NO formationreactions being closer to their equilibrium state than in otherengines (Haglind 2008) NOx emissions from vessels aretemperature-dependent (Sinha et al 2003) and also are in-fluenced by the oxygen concentration in the engine cylinder(Ni 1999) In larger engines the running speed is generallyslower and the combustion process more adiabatic resultingin higher combustion temperatures and more NOx Besideswith the increasing of air-to-fuel ratios concentration of NOx

showed a tendency first to increase then to decrease whichalways had the maximum value in the operating mode thatclose to full load of engine because of the high temperatureand oxygen in the engine cylinder (Ni 1999) Furthermorethere were always higher EF values in acceleration processand lower in moderating process in this study When the en-gines were in transient operating conditions such as acceler-ation process or moderating process concentrations of NOx

always had corresponding changes in the cylinder Studiesabout diesel engines showed that when the rotational speedhad a sudden increase there would be a first increasing then

decreasing and last stable tendency for the NOx concentra-tions and vice versa (Tan et al 2012)

TVOCs emissions from HH were much higher than fromthe other two vessels the lowest emissions were observedfor DFH Previous studies (Sinha et al 2003) have reportedthat hydrocarbon emissions from vessels depend on enginepower with low-power engines emitting more hydrocarbonsThe present results were partially consistent with these pre-vious studies Besides hydrocarbon emissions also dependon the percentage utilization of engine power (Sinha et al2003) As for various operating modes TVOCs EFs had largedifferences For example HH had the highest TVOCs emis-sions in accelerating mode which was almost 3 times theheight of the lowest value in mediumndashspeed mode The EFsfor SO2 depended solely on the sulfur content of the fuelsand were 16 09 and 26 g kgminus1 fuel for HH DFH andXYH respectively in this study Hydrocarbon could be gen-erated because of the incomplete combustion For examplein diesel cylinders there will always be air present in wallregions and crevices this is also the case when scavenging

wwwatmos-chem-physnet1663192016 Atmos Chem Phys 16 6319ndash6334 2016

6326 F Zhang et al Emission factors for gaseous and particulate pollutants

occurred during the aeration which could cause the unevenmixing of air and fuel (Ni 1999)

Fuel-based EFs for PM and its chemical components wereshown in Table S6 OC and EC were the main componentsof PM followed by SO2minus

4 NH+4 and NOminus3 Metals such asV Ni Cr Fe As and Cd made up a proportionately smallpart of the total PM mass However other rare elements suchas Tb Er Yb and Lu had higher values than did some ofthe common elements PM was an in-process product duringthe combustion in a cylinder the forming process includedthe molecular cracking decomposition and polymerizationwhich resulted in lack of oxygen High temperature and oxy-gen deficiency were the main reasons for the formation indiesel engines which always had high concentration val-ues in high load operating modes (Ni 1999) HH had muchhigher PM emission factors than the other two vessels theengine type was considered to be the most significant influ-ence factor which had a good agreement with NOx emissionfactors

EFs for OC and EC and the ratios between them are shownin Fig 3 EFs for OC and EC for HH were higher than forthe other two vessels Organic matter (OM) is generally cal-culated as OCtimes 12 (Petzold et al 2008) to account for themass of elements other than carbon in the emitted moleculesOM EFs for individual vessels mainly depend on the enginetype and the amount of unburned fuel ie the efficiency ofcombustion (Moldanovaacute et al 2013) BC emissions also de-pend heavily on the engine type (Lack et al 2009) There-fore the different types of engines and their levels of mainte-nance could account for the large differences in OC and ECEFs observed among the three vessels in this study The ratiosof OC-to-EC in the present study were much lower than thosefor large diesel ships reported previously (OC EC= 12)(Moldanova et al 2009) and also lower than that reportedfor a medium-speed vessel (Petzold et al 2010) The usageof non-dilution sampling in this study was one possible rea-son for the lower OC to EC ratio Besides TOR was used tomeasure OC and EC in PM which always had a lower OCcontent compared with other methods (such as TOT) becauseof the different definitions of OC and EC (Khan et al 2012)Compared with other diesel engines the ratios of OC to ECin this study were higher than that of automobile diesel sootin which EC comprises 75ndash80 wt of the total PM (Clagueet al 1999) and also higher than heavy heavy-duty dieseltrucks (HHDDTs) with OC to EC ratios below unit for cruseand transient modes even though higher in cold-startidle andcreep modes (Shah et al 2004)

Studies have shown that SO2minus4 formed from vessel-emitted

SO2 is a major contributor to CCN and ship track forma-tion (Schreier et al 2006 Lauer et al 2007) Sulfate is alsoan important component of PM emitted from vessels In thepresent study EFs for SO2minus

4 were much lower than previ-ously reported (Petzold et al 2008 Agrawal et al 2008a)but similar to those detected by a high-resolution time-of-flight aerosol mass spectrometer in a previous study (Lack

Figure 3 EFs for OC and EC and the ratios between them

et al 2009) This may be because EFs for SO2minus4 are mainly

related to the sulfur content of the fuel SO2minus4 is not gen-

erally emitted directly from the engines but forms after re-lease from the stack (Lack et al 2009) Because PM was col-lected directly from engine emissions in the present study thesulfur-to-sulfate ratios were low (lt 06 for vessels) Otherions such as NOminus3 and NH+4 accounted for a small percent-age of the PM emitted from the vessels compared with SO2minus

4 consistent with previous studies (Lack et al 2009) SO2 ismore easily oxidized to SO3 in catalytic reaction cycles withmetals commonly present in the exhaust gas (V Ni) whilehydroxyl radicals are additional needed to convert NOx toNOminus3 (Moldanova et al 2009)

Na+ and Clminus were considered to originate from marineair Their concentrations were highly correlated (r2

= 078)the differing air demands of the engines under different con-ditions might have caused observed variations in the EFs rel-ative to the fuel demand

The elemental composition of PM in the present studydiffered from previous studies showing high elemental con-tent of S Ca V Fe Cu Ni and Al (Agrawal et al 2008aMoldanova et al 2009) V and Ni are typically associatedwith combustion of heavy fuel oil (Almeida et al 2005) Inthe present study the high-quality fuels resulted in low EFsfor V and Ni In our previous study PM from shipping emis-sions was estimated to account for 294 of the total PM25at Tuoji Island in China using V as a tracer of shipping emis-sions (Zhang et al 2014) Reconsidering the former resultsbased on the EFs obtained in the present study we deter-mined that the contribution of vessels near Tuoji Island hadbeen underestimated because the estimate should have in-cluded both heavy and other types of fuels However somerare elements such as Tb Er Yb and Lu had relatively highEFs compared with those of other elements in the presentstudy which may be related to the source of the fuels

Atmos Chem Phys 16 6319ndash6334 2016 wwwatmos-chem-physnet1663192016

F Zhang et al Emission factors for gaseous and particulate pollutants 6327

Tabl

e3

Fuel

-bas

edav

erag

eE

Fsin

the

pres

ents

tudy

and

prev

ious

stud

ies

(gkgminus

1fu

el)

Ves

selI

DC

O2

CO

NO

NO

2N

2ON

Ox

TV

OC

sPM

SO2

Sco

nten

t(

m)

HH

3071plusmn

1565

302plusmn

162

982plusmn

372

155plusmn

545

128plusmn

170

115plusmn

443

237plusmn

210

940plusmn

213

160

008

DFH

3153plusmn

176

693plusmn

100

302plusmn

160

509plusmn

042

038plusmn

018

357plusmn

220

124plusmn

004

072plusmn

033

092

005

XY

H31

51plusmn

175

920plusmn

295

266plusmn

163

471plusmn

042

030plusmn

015

316plusmn

220

418plusmn

015

016plusmn

007

260

013

Com

mer

cial

vess

el(W

illia

ms

etal

20

09)

3170

7ndash16

ndashndash

ndash60

ndash87

ndashndash

6ndash30

Car

gove

ssel

(Mol

dano

vaet

al

2009

)

3441

217

ndashndash

ndash73

4ndash

53

393

19

Die

sele

ngin

e(H

aglin

d20

08)

ndash7

4ndash

ndashndash

87ndash

76

542

7

Oce

an-g

oing

ship

s(S

inha

etal

20

03)

3135

195

ndashndash

ndash22

3ndash

ndash2

90

1

Oce

an-g

oing

ship

s(S

inha

etal

20

03)

3176

30

ndashndash

ndash65

5ndash

ndash52

22

4

Car

goan

dpa

ssen

ger

ship

s(E

ndre

sen

2003

)31

707

4ndash

ndash0

0857

ndash87

24

12ndash

76

10ndash5

40

5ndash2

7

Ship

sop

erat

ing

inndash

ndash42

ndash72

ndashndash

65ndash8

6ndash

ndash4

6ndash9

8N

ON

Eha

rbou

rare

as(P

irjo

laet

al

2014

)ndash

ndash16

ndash49

ndashndash

25ndash7

9ndash

ndash5

4ndash17

0SC

R

Ship

sop

erat

ing

inPo

rt(D

iesc

het

al

2013

)ndash

ndash16

37ndash

53ndash

ndash7

7

NO

NE=

No

trea

tmen

tofe

mis

sion

sSC

R=

sele

ctiv

eca

taly

ticre

duct

ion

wwwatmos-chem-physnet1663192016 Atmos Chem Phys 16 6319ndash6334 2016

6328 F Zhang et al Emission factors for gaseous and particulate pollutants

33 Fuel-based average emission factors

Based on actual operating conditions (Table S7) average EFsfor the three vessels in the present study (according to Eq 4)along with EFs from previous studies are shown in Table 3EFs for all of the pollutants except SO2 were significantlyhigher for HH than for the other two vessels potentially dueto poor combustion conditions Most of the EFs for DFH andXYH were within the range of emissions for other vesselsdue to having well maintained engines and the high qualityof the fuels used The EFs for NOx PM and SO2 were muchlower than reported in previous studies (other than NOx forocean-going vessels) All the sulfur of the fuels in the presentstudy were significantly below the emissions limit of 350 established by IMO in the revised MARPOL Annex VI rulesapplicable since 2012 (IMO 1998)

The IMO Tier I emissions limit for NOx is450times nminus02 g kWhminus1 (n rated speed 130 lt nlt 2000 rpm)The rated speed and fuel consumption rates for each vesselare shown in Table 1 Thus the emissions limits for HHDFH and XYH would be 545 575 and 565 g kgminus1 fuelrespectively calculated combined with Eq (5) The averagefuel-based EFs for NOx of ship HH was more than 100 above the IMO standard while those of the other two shipswere below the IMO standard (Table 3) PM emissions forHH were also higher than previously reported but thosefor the two research vessels were much lower (Table 3)Fuel type is one of the most important influence factors onpollutant emissions for example sulfur content in the fuelnot only influence the SO2 emission directly but also hadimpact on PM formation in the flue gas stack with low sulfurcontent in fuels reduces PM formation (Lack et al 2011)Vessels with higher sulfur content always had relativelyhigher PM emissions which were also shown in Table 3In addition different engines and levels of maintenancehave a significant impact on all combustion-dependentemissions Emission reduction measures have been used insome vessels For example NOx emissions can be reducedby measures such as selective catalytic reduction (SCR) anddirect water injection (DWI) which had been implementedon some vessels previously studied in a harbour in Finland(Pirjola et al 2014) The results showed that SCR effec-tively reduced NOx emissions while vessels with DWI hadhigh PM emissions The engine type might be an importantcause of the different emissions such as HH had muchhigher pollutants emissions with an engine produced inChina and yet DFHrsquos engine produced in Germany Besidesemission tests for a high-speed marine diesel engine withdifferent kinds of diesels showed that diesel type had limitedinfluence on emissions such as NOx CO and CH but asignificant impact on PM emission (289ndash415 ) becauseof the different sulfur content in fuel (Xu 2008)

34 Power-based emission factors

Based on the engine power and fuel consumption rates ofthe vessels power-based EFs were calculated (according toEq 5) and compared to results from previous studies (Ta-ble 4) The EFs for HH were much higher than those forthe other two vessels except for SO2 HH also had signif-icantly higher EFs for NOx than previously reported val-ues while EFs for NOx of DFH and XYH were within therange of previously reported results Engine type was consid-ered to be a significant influence factor for NOx emissionswith lower engine speed having higher NOx emission factors(Celo et al 2015) In addition compared to other vesselswith a similar engine type and diesel fuel HH still had rela-tively higher NOx EF (seen in Table 4) which could reflectthe impact of engine condition (engine quality and mainte-nance level) on shipping emissions CO EFs for the test ves-sels in the present study were higher than previous studieswhich produced similar results to those of inland ships andother test vessels in China (Fu et al 2013 Song 2015) Inspite of the influence of engine type on CO emissions thatwith the higher engine speed having higher CO EFs (Celoet al 2015) engine condition combined with fuel qualitymight have significant influence All of the EFs for SO2 inthe present study were lower than those in previous studiesbecause of the low sulfur content of the present fuels Gen-erally PM emissions from marine diesel fuels are dependenton the fuel (sulfate and metal oxide ash constituents) and oncombustion conditions (unburned hydrocarbons and carbonresidue constituents) (Cooper 2003) HH had the highest PMemissions among the test vessels although there were almostno differences among the fuels (Table S6) Besides HH hadeven higher PM EFs than previously reported vessels withHFO fuel and XYH had much lower PM EFs than all theother vessels with even lower sulfur content fuel Thereforecombustion conditions were likely the determining factor forthe differences It can be seen from Table 4 that most previ-ous studies focused on the heavy fuel oil of shipping emis-sions Compared with diesel fuels heavy fuel oil always hadrelatively low CO emission factors and high PM emissionfactors And among the heavy-fuel-oil-using vessels enginetype (engine speed and engine power level) always playedan important role on emissions such as NOx and CO whichwith lower engine speed having higher NOx EFs and lowerCO EFs

Combined with other emission data of test ships in China(Fu et al 2013 Song 2015) it could be seen that inlandand some test offshore ships in China always had higherNOx CO and PM emissions compared with other test ves-sels in previous studies And among the test vessels in Chinathere also were differences for different engine types andship types In addition emission factors that were used forcalculation of ship inventories in China always came fromother countries and areas However there seemed to be sig-nificant differences between the reference and test data such

Atmos Chem Phys 16 6319ndash6334 2016 wwwatmos-chem-physnet1663192016

F Zhang et al Emission factors for gaseous and particulate pollutants 6329

Tabl

e4

Pow

er-b

ased

EFs

inth

epr

esen

tstu

dyan

dpr

evio

usst

udie

s(g

kWhminus

1 )

Ves

sel

CO

2C

ON

ON

O2

N2O

NO

xT

VO

Cs

PMSO

2E

ngin

epo

wer

Eng

ine

Fuel

type

and

ID(k

W)

spee

dsu

lfur

cont

ent

(rm

p)(w

t)

HH

699plusmn

352

738plusmn

376

220plusmn

841

345plusmn

124

030plusmn

039

258plusmn

100

544plusmn

484

209plusmn

048

036

350

1200

DO

00

798

DFH

631plusmn

352

139plusmn

020

604plusmn

032

102plusmn

008

008plusmn

004

714plusmn

044

017plusmn

001

014plusmn

007

018

1600

900

DO

00

458

XY

H69

7plusmn

385

201plusmn

065

587plusmn

036

104plusmn

009

007plusmn

003

697plusmn

048

092plusmn

ndash0

04plusmn

001

057

600

1000

DO

01

30Ta

nker

(Win

nes

and

Frid

ell

2010

)ndash

161

ndashndash

ndash7

82ndash

058

ndash45

0060

0H

FO1

6

Ber

thed

ship

s(C

oope

r65

3ndash76

80

33ndash0

97

ndashndash

ndash14

2ndash2

02

ndash0

14ndash0

45

026

ndash53

AE

720

ndash127

072

0ndash18

00M

GO

00

6ndash1

220

03)

691ndash

803

077

ndash17

112

9ndash1

75

048

ndash06

72

5ndash9

612

70ndash2

675

720ndash

750

RO

05

3ndash2

269

1ndash69

40

92ndash0

98

96ndash

99

017

ndash01

91

014

8072

0M

DO

02

3C

rude

oilt

anke

r(A

graw

alet

al

2008

b)58

8ndash66

00

77ndash1

78

ndashndash

ndash15

8ndash2

10

ndash1

10ndash1

78

766

ndash86

015

750

90H

FO2

85

Cru

ise

ship

s(P

opla

wsk

iet

al

2011

)ndash

ndashndash

140

ndashndash

ndash2

914

20

US

EPA

621

14

ndashndash

ndash18

1ndash

131

103

Mar

ine

engi

ne(S

ippu

laet

ndash1

2ndash11

411

3ndash2

95

ndashndash

114

ndash30

90ndash

95

083

ndash63

6ndash

1500

HFO

27

al

2014

)0ndash

885

69ndash2

58

584

ndash33

90

83ndash1

97

015

ndash09

3D

FL

arge

mar

ine

ship

s(K

han

etal

20

13)

533ndash

612

035

ndash06

0ndash

ndashndash

166

ndash20

6ndash

091

ndash21

97

2ndash11

436

740ndash

6853

097

ndash102

HFO

21

5ndash3

14

Oce

ango

ing

cont

aine

rve

ssel

(Agr

awal

etal

20

08a)

588ndash

660

077

ndash18

1ndash

ndashndash

158

ndash21

0ndash

109

ndash17

67

66ndash8

60

5027

010

4H

FO2

05

Lar

geca

rgo

vess

el(M

olda

nova

etal

20

09)

667

042

ndashndash

ndash14

22

ndash1

0310

320

200

97H

FO1

9

Oce

ango

ing

carg

ove

ssel

614plusmn

10

83plusmn

001

ndashndash

ndash16

3plusmn

02

ndash1

51plusmn

007

87plusmn

01

128

IFO

180

(Cel

oet

al

2015

)62

6plusmn

70

26plusmn

001

114plusmn

01

081plusmn

002

58plusmn

007

525

IFO

180

628plusmn

90

30plusmn

001

113plusmn

01

094plusmn

002

87plusmn

01

450

IFO

180

628plusmn

10

81plusmn

003

122plusmn

001

083plusmn

001

64plusmn

01

450

IFO

180

609plusmn

11

31plusmn

002

ndashndash

ndash8

4plusmn

003

ndash0

37plusmn

001

47plusmn

001

440

IFO

6060

5plusmn

10

00ndash

ndashndash

167plusmn

01

ndash2

2plusmn

02

103plusmn

003

116

IFO

380

622plusmn

11

22plusmn

002

ndashndash

ndash10

7plusmn

004

ndash0

30plusmn

003

047plusmn

01

450

MD

O

AE

aux

iliar

yen

gine

DO

die

selo

ilH

FOh

eavy

fuel

oil

MG

Om

arin

ega

soi

lR

Or

esid

ualo

ilM

DO

mar

ine

dies

eloi

lD

Fdi

esel

fuel

IFO

int

erm

edia

tefu

eloi

l

wwwatmos-chem-physnet1663192016 Atmos Chem Phys 16 6319ndash6334 2016

6330 F Zhang et al Emission factors for gaseous and particulate pollutants

as 100 to 132 g kW hminus1 of NOx EF and 11 to 17 g kW hminus1

of CO EF used for inland ships for ship inventory calcula-tion (Zhu et al 2015) 100 to 181 g kW hminus1 of NOx EFand 11 to 15 g kW hminus1 of CO EF for harbour ships (Yanget al 2015) compared to 15 to 173 g kW hminus1 of NOx EFand 46 to 103 g kW hminus1 of CO EF from test inland ships(convert the fuel-based EF to power-based EF with a factorof 200 g kW hminus1) (Song 2015) and 697 to 258 g kW hminus1 ofNOx EF and 139 to 738 g kW hminus1 of CO EF in the presentstudy (Yang et al 2015) Besides whether there are obviousdifferences of EFs between other types of vessels in China(such as low-speed engine vessels with heavy fuel oil) andprevious studies is still unclear Therefore much more mea-surement data for different vessels in China are still in urgentneed for more accurate assessment of shipping emissions

35 Impact of engine speed on NOx emission factors

NOx is formed in the combustion chamber by a combinationof atmospheric nitrogen and oxygen under high-pressure andhigh-temperature conditions Many factors affect NOx for-mation including engine temperature injection point andfuel quality The IMO emissions limit for NOx is determinedby the rated speed of the engine however other factors mustalso be considered to reduce NOx emissions

The NOx EFs for the test vessels at various engine speedsare shown in Fig 4 The rated speeds of the vessels were1200 900 and 1000 rpm for HH DFH and XYH respec-tively The actual engine speeds of HH were much lower thanthe rated speed while the two larger-engine-power vesselsoperated close to their rated speeds except during one op-erating mode of DFH The NOx EFs for HH differed sig-nificantly in different operating modes ranging from 391to 143 g kgminus1 of fuel The NOx EF was highest when theengine speed reached sim 750 rpm (Fig 4) At lower enginespeeds the NOx EFs had fluctuating but lower values Athigher engine speeds closer to the rated speed of 1200 rpmthe NOx EFs were much lower The NOx EFs for the twolarger-engine-power vessels changed slightly with enginespeed but also had lowest values when their engine speedsapproached their rated speeds Combined with the dieselpropulsion characteristic curve there were large increasesin the fuel consumption rate when the engine speed in-creased Therefore a best-fit engine speed should be deter-mined based on both EFs and economic costs

Engineering approaches for reducing the NOx emissionsof marine engines may be applied before during or afterthe combustion process (Verschaeren et al 2014 Habib etal 2014) In the present study the NOx EFs of the two re-search vessels were below the IMO Tier I emissions limitsHowever for EMS measures should be taken to meet theIMO emissions limit including increasing the engine speedand applying engineering technologies during or after com-bustion such as exhaust gas recirculation (EGR) selectivenon-catalytic reduction (SNCR) or SCR

Figure 4 Emission factors for NOx at different engine speeds

4 Conclusions

Three offshore vessels with different engine power sourceswere chosen in this study to collect measured data of gaseousspecies and particulate matter including NO2 NO N2OCO CO2 TVOCs SO2 and the total suspended particulateBesides chemical composition of the PM were also analysedto give detailed EFs for OC EC water-soluble ions and metalelements Concentrations fuel-based EFs fuel-based aver-age EFs as well as power-based average EFs for species ofoffshore vessels in China were presented Furthermore im-pact of engine speed on NOx EFs was also discussed

There were higher concentrations of pollutants for low-engine-power vessel HH than for the other two vessels COconcentrations for offshore vessels were slightly lower thaninland vessels in China and all the three vessels had the low-est CO concentrations at their economic speeds (the speed ofthe least vessel operating expenditures during one voyagethey were high-speed mode cruise mode and high-speedmode for HH DFH and XYH respectively) More than 80 of the NOx was NO and all the offshore vessels had higherNOx concentrations in high-speed modes Because of thelow-sulfur fuels used in this study SO2 concentrations ofthese three offshore vessels were lower than that in the lit-eratures And the PM concentrations were much lower thaninland vessels while showing significant differences amongdifferent operating modes

Fuel-based EFs for gaseous species and PM were pre-sented based on the carbon balance method EFs for CO2were 2940ndash3106 3121ndash3160 and 3102ndash3162 g kgminus1 fuel forHH DFH and XYH Because of the combustion conditionssuch as excess air ratio combustion temperature and unifor-mity of the fuelndashair mixture EFs for CO showed high valuesin idling mode but low values in economic speed All theoffshore vessels had higher NOx EFs in low speed than inhigh speed but showed higher values when in acceleration

Atmos Chem Phys 16 6319ndash6334 2016 wwwatmos-chem-physnet1663192016

F Zhang et al Emission factors for gaseous and particulate pollutants 6331

process EFs for SO2 were 16 09 and 26 g kgminus1 fuel forHH DFH and XYH based on sulfur content of the fuels OCand EC were the main components of PM with low OC toEC ratios that were lower than 01 followed by SO2minus

4 NH+4 and NOminus3 Metals such as V Ni Cr Fe As and Cd made upa proportionately small part of the total PM mass

Fuel-based average EFs as well as power-based EFs for thethree different vessels of differing engine power were pre-sented EFs for most gaseous species and PM of HH weremuch higher compared with the other higher-engine-powervessels which was also gt 100 above the IMO standard forNOx Average PM EF of the low-engine-power vessel HHwas also much higher than that in the literatures Howeveraverage EFs for most species of the two larger-engine-powervessels were within the range of previously reported resultsEngine type was inferred as one of the most influence fac-tors for the differences of emission factors Inland and someoffshore ships in China always had higher NOx CO and PMemissions compared with other test vessels in previous stud-ies In addition emission factors that used for calculation ofship inventories in China always had lower values than testvessels

The impact of engine speed on EFs for NOx showed thatwhen the engine speed was close to the rated speed therewould be lower NOx EFs values However combined withthe high fuel consumption rate an optimal engine speedshould be determined based on both EFs and economic costsEmission reduction measures for NOx for some of the off-shore vessels in China are still essential to meet the IMOemission limit

Given the limits of vessel types and numbers this studysubstantially gives the EFs for gaseous species and PM ofthree different diesel-engine-powered offshore vessels How-ever as the development of ports in China emissions fromcargo ships and container ships with large engine power havebecome one of the most significant air pollution sources inport cities and regions Systematical measurement EFs of allkinds of offshore vessels in China are essential in order topresent the accurate emission inventory of ships

Information about the Supplement

Supplementary information includes the details of the real-world measurement system for vessels (Fig S1) the concen-trations of main gaseous matter and PM of shipping emis-sions (Fig S2) the types composition of offshore vessels inChina (Table S1) the distribution of vessels through grosstonnage in 2014 in offshore area of Yangtze River Delta (Ta-ble S2) the Chinese market consumption of marine oil in2011 (Table S3) the detection parameters for gaseous mat-ter (Table S4) the fuel-based EFs for the gaseous pollutants(Table S5) PM and the chemical composition in PM for dif-ferent operating (Table S6) modes and the actual operatingconditions of vessels (Table S7)

The Supplement related to this article is available onlineat doi105194acp-16-6319-2016-supplement

Acknowledgements This study was supported by the CAS Strate-gic Priority Research Program (No XDB05030303) the NaturalScientific Foundation of China (Nos 41273135 and 41473091)and the Shanghai science and Technology Commission project(No 15DZ1205401) The authors would like to express theirgratitude to the vessel owners Haohai Ocean Engineering LimitedLiability Company for vessel Haohai 0007 Ocean University ofChina for vessel Dongfanghong 02 and North China Sea Branchof the State Oceanic Administration for vessel Xiangyanghong08 for making their vessels available for the shipping emissionsstudy The views expressed in this document are solely those of theauthors and the funding agencies do not endorse any products orcommercial services mentioned in this publication

Edited by T Bond

References

Agrawal H Malloy Q G J Welch W A Miller J W andCocker D R In-use gaseous and particulate matter emissionsfrom a modern ocean going container vessel Atmos Environ42 5504ndash5510 doi101016jatmosenv200802053 2008a

Agrawal H Welch W A Miller J W and Cocker D R Emis-sion measurements from a crude oil tanker at sea Environ SciTechnol 42 7098ndash7103 doi101021es703102y 2008b

Almeida S M Pio C A Freitas M C Reis MA and Trancoso M A Source apportionment of fineand coarse particulate matter in a sub-urban area at theWestern European Coast Atmos Environ 39 3127ndash3138doi101016jatmosenv200501048 2005

Anderson M Salo K Hallquist A M and Fridell ECharacterization of particles from a marine engine op-erating at low loads Atmos Environ 101 65ndash71doi101016jatmosenv201411009 2015

Berg N Mellqvist J Jalkanen J-P and Balzani J Ship emis-sions of SO2 and NO2 DOAS measurements from airborne plat-forms Atmos Meas Tech 5 1085ndash1098 doi105194amt-5-1085-2012 2012

California Code of Regulation Titles 13 and 17 avail-able at httpsgovtwestlawcomcalregsBrowseHomeCaliforniaCaliforniaCodeofRegulationstransitionType=DefaultampcontextData=(scDefault) last access 20 April 2016

Cappa C D Williams E J Lack D A Buffaloe G M Coff-man D Hayden K L Herndon S C Lerner B M Li S-M Massoli P McLaren R Nuaaman I Onasch T B andQuinn P K A case study into the measurement of ship emis-sions from plume intercepts of the NOAA ship Miller FreemanAtmos Chem Phys 14 1337ndash1352 doi105194acp-14-1337-2014 2014

Carlton J Danton S Gawen R Lavender K Mathieson NNewell A Reynolds G Webster A Wills C and Wright AMarine exhaust emissions research programme Lloydrsquos RegisterEngineering Services London 63 1995

wwwatmos-chem-physnet1663192016 Atmos Chem Phys 16 6319ndash6334 2016

F Zhang et al Emission factors for gaseous and particulate pollutants 6323

Flue gas

Flue gas analyzer

(PhotonⅡ)

O2 NO

2 NO N

2O CO

CO2 SO

2 HC

Particulate samplers

Quartz fiber filter (QFF)

OC EC soluble ions inorganic elements

Glass filter (GF)

Polycyclic aromatic hydrocarbon (PAHs)

Eight-stage particulate

sampler (MARK Ⅲ)

Particle size

Polytetra- fluoroethylene

filter (PTFE) PM weight

TVOCs analyzer (MiniRAE 2000)

TVOCs

Figure 1 On-board emissions test system and measured analytes

plement for more details) All analytes are also shown inFig 1 the flue gas analyser (Photon II) is aimed to test in-stantaneous emissions of gaseous pollution including O2NO2 NO N2O CO CO2 and SO2 (detection parametersfor the gaseous matter are shown in Table S4) Three par-ticulate samplers are installed to collect PM using differentfilters at the same time including a quartz fiber filter glassfilter and polytetrafluoroethylene filter to analyse differentchemical components of PM And the portable TVOCs anal-yser is used to monitor the concentration of total VOCs withisobutylene as correction coefficient gas Besides a temper-ature sensor is installed near the smoke outlet to test the fluegas temperature A total of 33 sets of samples for HH 20sets for DFH and 23 sets for XYH were collected with 3 to5 sets for each operating mode

The OC (organic carbon) and EC (elemental carbon) weremeasured on a 0544 cm2 quartz filter punched from each fil-ter by thermal optical reflectance (TOR) following the IM-PROVE protocol with a DRI Model 2001 ThermalOpticalCarbon Analyzer (Atmoslytic Inc Calabasas CA) Themeasuring range of TOR was from 005 to 750 microg C cmminus2

with an error of less than 10 Concentrations of water-soluble ions in PM25 such as Na+ NH+4 K+ Mg2+ Ca2+Clminus NOminus3 and SO2minus

4 were determined by ion chromatogra-phy (Dionex ICS3000 Dionex Ltd America) based on themeasurement method of Shahsavani et al (2012) The de-tection limit was 10 ng mLminus1 with an error of less than 5 and 1 mL RbBr with concentration of 200 ppm was put in thesolution as internal standard before sampling The concentra-tions of 33 inorganic elements in PM25 were estimated usinga inductively coupled plasma mass spectrometer (ICP-MS ofELAN DRC II type PerkinElmer Ltd Hong Kong) follow-ing the standard method (Wang et al 2006) The resolution

of ICP-MS ranged from 03 to 30 amu with a detection limitlower than 001 ng mLminus1 and the error was less than 5

24 Data analysis

Carbon balance formula was used to calculate the EFs for allexhaust gas components It was assumed that all carbon in thefuel was emitted as carbon-containing gases (CO CO2 andTVOC) and carbon-containing particulate matter So therewas a certain equilibrium relationship between the carbon inthe fuel and in the exhaust

CF = RFGtimes (c(CCO)+ c(CCO2)

+ c(CPM)+ c(CTVOC)) (1)

where CF represents the mass of C in per kg diesel fuel(g C kgminus1 fuel) RFG represents the flue gas emissionsrate (m3 kgminus1 fuel) and c(CCO) c(CCO2) c(CPM) andc(CTVOC) represent the mass concentrations of carbon asCO CO2 PM and TVOC (g C mminus3) in the flue gas respec-tively

The EF for CO2 was calculated as follows

EFCO2 = RFG middot c(CO2) middotMCO2 (2)

where EFCO2 is the EF for CO2 (g kgminus1 fuel) c(CO2) isthe molar concentration of CO2 (mol mminus3) and MCO2 is themolecular weight of CO2 (44 g molminus1)

The remaining EFs were calculated as follows

EFx =1X

1CO2middot

MX

MCO2

middotEFCO2 (3)

where EFx is the EF for species X (g kgminus1 fuel) 1X and1CO2 represent the concentrations of X and CO2 with thebackground concentrations subtracted (mol mminus3) and MX

represents the molecular weight of species X (g molminus1)In addition average EFs for each vessel were calculated

based on actual operating conditions as follows

EFXA =sum

XiEFi timesPi (4)

where EFXA is the average EF for species X EFi is the EFfor operating mode i for species X and Pi is the percentageof time spent in operating mode i during the shipping cycle

Power-based emission factors and fuel-based emissionfactors could be interconverted with the formula as follow-ing

EFXP = EFx middotFCR (5)

where EFXP is the power-based emission factor forspecies X (g kW hminus1) and FCR is fuel consumption rate foreach vessel (kg fuel (kW h)minus1)

wwwatmos-chem-physnet1663192016 Atmos Chem Phys 16 6319ndash6334 2016

6324 F Zhang et al Emission factors for gaseous and particulate pollutants

3 Results and discussion

31 Concentrations in shipping emissions

Concentrations of CO NOx SO2 TVOC and PM fromthe three vessels are shown in Fig S1 in the SupplementNearly all of the concentrations measured in the exhaustof low-engine-power vessel HH were higher than those ofthe two higher-engine-power vessels Concentrations of COSO2 and NOx from HH were 107ndash756 534ndash331 and878ndash1295 ppm respectively and 143ndash595 mg mminus3 PM Incontrast concentrations of CO SO2 NOx and PM were501ndash141 527ndash169 169ndash800 ppm and 706ndash218 mg mminus3respectively for DFH and 360ndash224 049ndash359 and 235ndash578 ppm and 056ndash631 mg mminus3 respectively for XYH

A previous study demonstrated that concentrations of COprimarily depend on engine power with higher CO emissionsresulting from vessel engines with lower power (Sinha et al2003) There was a similar trend in this study with gener-ally higher concentrations for HH and lower concentrationsfor DFH The CO concentrations in the present study weresimilar but slightly lower than those of inland vessels (Fuet al 2013) except in the idling mode of HH In differentoperating modes CO concentrations were significantly dif-ferent For example the maximum value was observed inidling mode and the minimum value in medium-speed modefor HH All three ships had the lowest CO concentrations attheir economic speeds (medium speed for HH cruise modefor DFH and high speed for XYH) demonstrating that theirengines are optimized for the most common operating mode

More than 80 of the NOx was NO in this study withNO2 and N2O accounting for lt 20 in all operating modes(Fig S1) Again nearly all of these concentrations werehigher in the exhaust gas of HH than in that of the two ves-sels In high-speed modes all of the vessels had high con-centrations of NOx NOx emissions mainly depend on thecombustion temperature of the engines More powerful com-bustion systems operate at higher temperatures thereby pro-ducing more NOx (Corbett et al 1999) However the NOx

emissions were much lower than for the inland vessels stud-ied by Fu et al (2013) particularly in cruise mode (NOx con-centrations of sim 1000 ppm)

SO2 concentrations in the exhaust gas depend on the sul-fur content of the fuel and the flow rate of the flue gas Therewere significant differences among the three vessels in theirflow rates which could account for the different concentra-tions of one vessel in different operating modes But becauseof the low-sulfur fuels used in these vessels the SO2 con-centrations were low compared with those in other studies(Williams et al 2009 Berg et al 2012)

Much lower concentrations of PM in the exhaust gas wereobserved in the present study compared to those of inlandships in China (Fu et al 2013) However they were similarto those from ships at berth reported by Cooper et al (Cooper2003) HH had higher PM concentrations than the two ves-

sels in the exhaust gas There were significant differencesamong the different operating modes because of changes inthe injection point of the engines (Sippula et al 2014 Li etal 2014)

32 Fuel-based emission factors

Fuel-based EFs for the gaseous species CO2 CO NO NO2N2O and TVOCs and for PM based on the carbon balancemethod were determined In addition SO2 was calculatedbased on the sulfur content of the fuels Fuel-based EFs forthe typical pollutants such as CO PM and nitrogen oxides indifferent operating modes are shown in Fig 2 (detailed EFsfor all the gaseous pollutants are shown in Table S5 and de-tailed EFs for PM and its chemical composition are shown inTable S6)

CO2 emissions from vessels primarily depend on the car-bon content of the fuel (Carlton et al 1995) Accordinglythe EFs for CO2 in the present study should theoreticallybe 3177 3168 and 3171 g kgminus1 fuel for complete combus-tion Under actual conditions CO2 emissions were 2940ndash3106 3121ndash3160 and 3102ndash3162 g kgminus1 fuel for HH DFHand XYH respectively which means they had combustionefficiencies with 925ndash978 985ndash997 and 978ndash997 interms of CO2 for these three vessels

CO emissions of HH were much higher than of XYH fol-lowed by DFH The power of their respective engines was350 600 and 1600 kW In addition there were large differ-ences in CO emissions among different modes All of thesethree vessels had relatively high EFs for CO while accelerat-ing compared with other modes but the highest EFs wereduring the idling modes of HH and DFH as well as dur-ing the low-speed mode of XYH Because CO emissionsin diesel engines primarily depend on the excess air ratio(which determines the fuelndashair mixture) combustion temper-ature and uniformity of the fuelndashair mixture in the combus-tion chamber (Doug 2004) ship engines with lower powergenerally have higher CO emissions (Carlton et al 1995)Localized hypoxia and incomplete combustion in the cylin-der were the main reasons for CO emission of diesel engineCO emissions always had positive relationships with the air-to-fuel ratio There was lower air-to-fuel ratio when in lowengine load which resulted in lower CO emission and viceversa (Ni 1999)

Much higher NOx EFs were observed for HH than forthe other two vessels These results were inconsistent withthose of Sinha et al (2003) in which emissions of NOx in-creased with the power of the ship engine With increasingvessel speed NOx EFs for HH first increased and then de-creased XYH had lower EFs when operating at high speedthan at low speed Nitrogen oxides included NO NO2 andN2O in the present study More than 70 of the NOx wasin the form of NO for all vessels because most of the NOx

emissions were generated through thermal NO formation(Haglind 2008) The primary reasons that slow diesel en-

Atmos Chem Phys 16 6319ndash6334 2016 wwwatmos-chem-physnet1663192016

F Zhang et al Emission factors for gaseous and particulate pollutants 6325

Figure 2 EFs for the typical pollutants in different operating modes

gines such as the one in HH have higher NOx emissions in-clude higher peak flame temperatures and the NO formationreactions being closer to their equilibrium state than in otherengines (Haglind 2008) NOx emissions from vessels aretemperature-dependent (Sinha et al 2003) and also are in-fluenced by the oxygen concentration in the engine cylinder(Ni 1999) In larger engines the running speed is generallyslower and the combustion process more adiabatic resultingin higher combustion temperatures and more NOx Besideswith the increasing of air-to-fuel ratios concentration of NOx

showed a tendency first to increase then to decrease whichalways had the maximum value in the operating mode thatclose to full load of engine because of the high temperatureand oxygen in the engine cylinder (Ni 1999) Furthermorethere were always higher EF values in acceleration processand lower in moderating process in this study When the en-gines were in transient operating conditions such as acceler-ation process or moderating process concentrations of NOx

always had corresponding changes in the cylinder Studiesabout diesel engines showed that when the rotational speedhad a sudden increase there would be a first increasing then

decreasing and last stable tendency for the NOx concentra-tions and vice versa (Tan et al 2012)

TVOCs emissions from HH were much higher than fromthe other two vessels the lowest emissions were observedfor DFH Previous studies (Sinha et al 2003) have reportedthat hydrocarbon emissions from vessels depend on enginepower with low-power engines emitting more hydrocarbonsThe present results were partially consistent with these pre-vious studies Besides hydrocarbon emissions also dependon the percentage utilization of engine power (Sinha et al2003) As for various operating modes TVOCs EFs had largedifferences For example HH had the highest TVOCs emis-sions in accelerating mode which was almost 3 times theheight of the lowest value in mediumndashspeed mode The EFsfor SO2 depended solely on the sulfur content of the fuelsand were 16 09 and 26 g kgminus1 fuel for HH DFH andXYH respectively in this study Hydrocarbon could be gen-erated because of the incomplete combustion For examplein diesel cylinders there will always be air present in wallregions and crevices this is also the case when scavenging

wwwatmos-chem-physnet1663192016 Atmos Chem Phys 16 6319ndash6334 2016

6326 F Zhang et al Emission factors for gaseous and particulate pollutants

occurred during the aeration which could cause the unevenmixing of air and fuel (Ni 1999)

Fuel-based EFs for PM and its chemical components wereshown in Table S6 OC and EC were the main componentsof PM followed by SO2minus

4 NH+4 and NOminus3 Metals such asV Ni Cr Fe As and Cd made up a proportionately smallpart of the total PM mass However other rare elements suchas Tb Er Yb and Lu had higher values than did some ofthe common elements PM was an in-process product duringthe combustion in a cylinder the forming process includedthe molecular cracking decomposition and polymerizationwhich resulted in lack of oxygen High temperature and oxy-gen deficiency were the main reasons for the formation indiesel engines which always had high concentration val-ues in high load operating modes (Ni 1999) HH had muchhigher PM emission factors than the other two vessels theengine type was considered to be the most significant influ-ence factor which had a good agreement with NOx emissionfactors

EFs for OC and EC and the ratios between them are shownin Fig 3 EFs for OC and EC for HH were higher than forthe other two vessels Organic matter (OM) is generally cal-culated as OCtimes 12 (Petzold et al 2008) to account for themass of elements other than carbon in the emitted moleculesOM EFs for individual vessels mainly depend on the enginetype and the amount of unburned fuel ie the efficiency ofcombustion (Moldanovaacute et al 2013) BC emissions also de-pend heavily on the engine type (Lack et al 2009) There-fore the different types of engines and their levels of mainte-nance could account for the large differences in OC and ECEFs observed among the three vessels in this study The ratiosof OC-to-EC in the present study were much lower than thosefor large diesel ships reported previously (OC EC= 12)(Moldanova et al 2009) and also lower than that reportedfor a medium-speed vessel (Petzold et al 2010) The usageof non-dilution sampling in this study was one possible rea-son for the lower OC to EC ratio Besides TOR was used tomeasure OC and EC in PM which always had a lower OCcontent compared with other methods (such as TOT) becauseof the different definitions of OC and EC (Khan et al 2012)Compared with other diesel engines the ratios of OC to ECin this study were higher than that of automobile diesel sootin which EC comprises 75ndash80 wt of the total PM (Clagueet al 1999) and also higher than heavy heavy-duty dieseltrucks (HHDDTs) with OC to EC ratios below unit for cruseand transient modes even though higher in cold-startidle andcreep modes (Shah et al 2004)

Studies have shown that SO2minus4 formed from vessel-emitted

SO2 is a major contributor to CCN and ship track forma-tion (Schreier et al 2006 Lauer et al 2007) Sulfate is alsoan important component of PM emitted from vessels In thepresent study EFs for SO2minus

4 were much lower than previ-ously reported (Petzold et al 2008 Agrawal et al 2008a)but similar to those detected by a high-resolution time-of-flight aerosol mass spectrometer in a previous study (Lack

Figure 3 EFs for OC and EC and the ratios between them

et al 2009) This may be because EFs for SO2minus4 are mainly

related to the sulfur content of the fuel SO2minus4 is not gen-

erally emitted directly from the engines but forms after re-lease from the stack (Lack et al 2009) Because PM was col-lected directly from engine emissions in the present study thesulfur-to-sulfate ratios were low (lt 06 for vessels) Otherions such as NOminus3 and NH+4 accounted for a small percent-age of the PM emitted from the vessels compared with SO2minus

4 consistent with previous studies (Lack et al 2009) SO2 ismore easily oxidized to SO3 in catalytic reaction cycles withmetals commonly present in the exhaust gas (V Ni) whilehydroxyl radicals are additional needed to convert NOx toNOminus3 (Moldanova et al 2009)

Na+ and Clminus were considered to originate from marineair Their concentrations were highly correlated (r2

= 078)the differing air demands of the engines under different con-ditions might have caused observed variations in the EFs rel-ative to the fuel demand

The elemental composition of PM in the present studydiffered from previous studies showing high elemental con-tent of S Ca V Fe Cu Ni and Al (Agrawal et al 2008aMoldanova et al 2009) V and Ni are typically associatedwith combustion of heavy fuel oil (Almeida et al 2005) Inthe present study the high-quality fuels resulted in low EFsfor V and Ni In our previous study PM from shipping emis-sions was estimated to account for 294 of the total PM25at Tuoji Island in China using V as a tracer of shipping emis-sions (Zhang et al 2014) Reconsidering the former resultsbased on the EFs obtained in the present study we deter-mined that the contribution of vessels near Tuoji Island hadbeen underestimated because the estimate should have in-cluded both heavy and other types of fuels However somerare elements such as Tb Er Yb and Lu had relatively highEFs compared with those of other elements in the presentstudy which may be related to the source of the fuels

Atmos Chem Phys 16 6319ndash6334 2016 wwwatmos-chem-physnet1663192016

F Zhang et al Emission factors for gaseous and particulate pollutants 6327

Tabl

e3

Fuel

-bas

edav

erag

eE

Fsin

the

pres

ents

tudy

and

prev

ious

stud

ies

(gkgminus

1fu

el)

Ves

selI

DC

O2

CO

NO

NO

2N

2ON

Ox

TV

OC

sPM

SO2

Sco

nten

t(

m)

HH

3071plusmn

1565

302plusmn

162

982plusmn

372

155plusmn

545

128plusmn

170

115plusmn

443

237plusmn

210

940plusmn

213

160

008

DFH

3153plusmn

176

693plusmn

100

302plusmn

160

509plusmn

042

038plusmn

018

357plusmn

220

124plusmn

004

072plusmn

033

092

005

XY

H31

51plusmn

175

920plusmn

295

266plusmn

163

471plusmn

042

030plusmn

015

316plusmn

220

418plusmn

015

016plusmn

007

260

013

Com

mer

cial

vess

el(W

illia

ms

etal

20

09)

3170

7ndash16

ndashndash

ndash60

ndash87

ndashndash

6ndash30

Car

gove

ssel

(Mol

dano

vaet

al

2009

)

3441

217

ndashndash

ndash73

4ndash

53

393

19

Die

sele

ngin

e(H

aglin

d20

08)

ndash7

4ndash

ndashndash

87ndash

76

542

7

Oce

an-g

oing

ship

s(S

inha

etal

20

03)

3135

195

ndashndash

ndash22

3ndash

ndash2

90

1

Oce

an-g

oing

ship

s(S

inha

etal

20

03)

3176

30

ndashndash

ndash65

5ndash

ndash52

22

4

Car

goan

dpa

ssen

ger

ship

s(E

ndre

sen

2003

)31

707

4ndash

ndash0

0857

ndash87

24

12ndash

76

10ndash5

40

5ndash2

7

Ship

sop

erat

ing

inndash

ndash42

ndash72

ndashndash

65ndash8

6ndash

ndash4

6ndash9

8N

ON

Eha

rbou

rare

as(P

irjo

laet

al

2014

)ndash

ndash16

ndash49

ndashndash

25ndash7

9ndash

ndash5

4ndash17

0SC

R

Ship

sop

erat

ing

inPo

rt(D

iesc

het

al

2013

)ndash

ndash16

37ndash

53ndash

ndash7

7

NO

NE=

No

trea

tmen

tofe

mis

sion

sSC

R=

sele

ctiv

eca

taly

ticre

duct

ion

wwwatmos-chem-physnet1663192016 Atmos Chem Phys 16 6319ndash6334 2016

6328 F Zhang et al Emission factors for gaseous and particulate pollutants

33 Fuel-based average emission factors

Based on actual operating conditions (Table S7) average EFsfor the three vessels in the present study (according to Eq 4)along with EFs from previous studies are shown in Table 3EFs for all of the pollutants except SO2 were significantlyhigher for HH than for the other two vessels potentially dueto poor combustion conditions Most of the EFs for DFH andXYH were within the range of emissions for other vesselsdue to having well maintained engines and the high qualityof the fuels used The EFs for NOx PM and SO2 were muchlower than reported in previous studies (other than NOx forocean-going vessels) All the sulfur of the fuels in the presentstudy were significantly below the emissions limit of 350 established by IMO in the revised MARPOL Annex VI rulesapplicable since 2012 (IMO 1998)

The IMO Tier I emissions limit for NOx is450times nminus02 g kWhminus1 (n rated speed 130 lt nlt 2000 rpm)The rated speed and fuel consumption rates for each vesselare shown in Table 1 Thus the emissions limits for HHDFH and XYH would be 545 575 and 565 g kgminus1 fuelrespectively calculated combined with Eq (5) The averagefuel-based EFs for NOx of ship HH was more than 100 above the IMO standard while those of the other two shipswere below the IMO standard (Table 3) PM emissions forHH were also higher than previously reported but thosefor the two research vessels were much lower (Table 3)Fuel type is one of the most important influence factors onpollutant emissions for example sulfur content in the fuelnot only influence the SO2 emission directly but also hadimpact on PM formation in the flue gas stack with low sulfurcontent in fuels reduces PM formation (Lack et al 2011)Vessels with higher sulfur content always had relativelyhigher PM emissions which were also shown in Table 3In addition different engines and levels of maintenancehave a significant impact on all combustion-dependentemissions Emission reduction measures have been used insome vessels For example NOx emissions can be reducedby measures such as selective catalytic reduction (SCR) anddirect water injection (DWI) which had been implementedon some vessels previously studied in a harbour in Finland(Pirjola et al 2014) The results showed that SCR effec-tively reduced NOx emissions while vessels with DWI hadhigh PM emissions The engine type might be an importantcause of the different emissions such as HH had muchhigher pollutants emissions with an engine produced inChina and yet DFHrsquos engine produced in Germany Besidesemission tests for a high-speed marine diesel engine withdifferent kinds of diesels showed that diesel type had limitedinfluence on emissions such as NOx CO and CH but asignificant impact on PM emission (289ndash415 ) becauseof the different sulfur content in fuel (Xu 2008)

34 Power-based emission factors

Based on the engine power and fuel consumption rates ofthe vessels power-based EFs were calculated (according toEq 5) and compared to results from previous studies (Ta-ble 4) The EFs for HH were much higher than those forthe other two vessels except for SO2 HH also had signif-icantly higher EFs for NOx than previously reported val-ues while EFs for NOx of DFH and XYH were within therange of previously reported results Engine type was consid-ered to be a significant influence factor for NOx emissionswith lower engine speed having higher NOx emission factors(Celo et al 2015) In addition compared to other vesselswith a similar engine type and diesel fuel HH still had rela-tively higher NOx EF (seen in Table 4) which could reflectthe impact of engine condition (engine quality and mainte-nance level) on shipping emissions CO EFs for the test ves-sels in the present study were higher than previous studieswhich produced similar results to those of inland ships andother test vessels in China (Fu et al 2013 Song 2015) Inspite of the influence of engine type on CO emissions thatwith the higher engine speed having higher CO EFs (Celoet al 2015) engine condition combined with fuel qualitymight have significant influence All of the EFs for SO2 inthe present study were lower than those in previous studiesbecause of the low sulfur content of the present fuels Gen-erally PM emissions from marine diesel fuels are dependenton the fuel (sulfate and metal oxide ash constituents) and oncombustion conditions (unburned hydrocarbons and carbonresidue constituents) (Cooper 2003) HH had the highest PMemissions among the test vessels although there were almostno differences among the fuels (Table S6) Besides HH hadeven higher PM EFs than previously reported vessels withHFO fuel and XYH had much lower PM EFs than all theother vessels with even lower sulfur content fuel Thereforecombustion conditions were likely the determining factor forthe differences It can be seen from Table 4 that most previ-ous studies focused on the heavy fuel oil of shipping emis-sions Compared with diesel fuels heavy fuel oil always hadrelatively low CO emission factors and high PM emissionfactors And among the heavy-fuel-oil-using vessels enginetype (engine speed and engine power level) always playedan important role on emissions such as NOx and CO whichwith lower engine speed having higher NOx EFs and lowerCO EFs

Combined with other emission data of test ships in China(Fu et al 2013 Song 2015) it could be seen that inlandand some test offshore ships in China always had higherNOx CO and PM emissions compared with other test ves-sels in previous studies And among the test vessels in Chinathere also were differences for different engine types andship types In addition emission factors that were used forcalculation of ship inventories in China always came fromother countries and areas However there seemed to be sig-nificant differences between the reference and test data such

Atmos Chem Phys 16 6319ndash6334 2016 wwwatmos-chem-physnet1663192016

F Zhang et al Emission factors for gaseous and particulate pollutants 6329

Tabl

e4

Pow

er-b

ased

EFs

inth

epr

esen

tstu

dyan

dpr

evio

usst

udie

s(g

kWhminus

1 )

Ves

sel

CO

2C

ON

ON

O2

N2O

NO

xT

VO

Cs

PMSO

2E

ngin

epo

wer

Eng

ine

Fuel

type

and

ID(k

W)

spee

dsu

lfur

cont

ent

(rm

p)(w

t)

HH

699plusmn

352

738plusmn

376

220plusmn

841

345plusmn

124

030plusmn

039

258plusmn

100

544plusmn

484

209plusmn

048

036

350

1200

DO

00

798

DFH

631plusmn

352

139plusmn

020

604plusmn

032

102plusmn

008

008plusmn

004

714plusmn

044

017plusmn

001

014plusmn

007

018

1600

900

DO

00

458

XY

H69

7plusmn

385

201plusmn

065

587plusmn

036

104plusmn

009

007plusmn

003

697plusmn

048

092plusmn

ndash0

04plusmn

001

057

600

1000

DO

01

30Ta

nker

(Win

nes

and

Frid

ell

2010

)ndash

161

ndashndash

ndash7

82ndash

058

ndash45

0060

0H

FO1

6

Ber

thed

ship

s(C

oope

r65

3ndash76

80

33ndash0

97

ndashndash

ndash14

2ndash2

02

ndash0

14ndash0

45

026

ndash53

AE

720

ndash127

072

0ndash18

00M

GO

00

6ndash1

220

03)

691ndash

803

077

ndash17

112

9ndash1

75

048

ndash06

72

5ndash9

612

70ndash2

675

720ndash

750

RO

05

3ndash2

269

1ndash69

40

92ndash0

98

96ndash

99

017

ndash01

91

014

8072

0M

DO

02

3C

rude

oilt

anke

r(A

graw

alet

al

2008

b)58

8ndash66

00

77ndash1

78

ndashndash

ndash15

8ndash2

10

ndash1

10ndash1

78

766

ndash86

015

750

90H

FO2

85

Cru

ise

ship

s(P

opla

wsk

iet

al

2011

)ndash

ndashndash

140

ndashndash

ndash2

914

20

US

EPA

621

14

ndashndash

ndash18

1ndash

131

103

Mar

ine

engi

ne(S

ippu

laet

ndash1

2ndash11

411

3ndash2

95

ndashndash

114

ndash30

90ndash

95

083

ndash63

6ndash

1500

HFO

27

al

2014

)0ndash

885

69ndash2

58

584

ndash33

90

83ndash1

97

015

ndash09

3D

FL

arge

mar

ine

ship

s(K

han

etal

20

13)

533ndash

612

035

ndash06

0ndash

ndashndash

166

ndash20

6ndash

091

ndash21

97

2ndash11

436

740ndash

6853

097

ndash102

HFO

21

5ndash3

14

Oce

ango

ing

cont

aine

rve

ssel

(Agr

awal

etal

20

08a)

588ndash

660

077

ndash18

1ndash

ndashndash

158

ndash21

0ndash

109

ndash17

67

66ndash8

60

5027

010

4H

FO2

05

Lar

geca

rgo

vess

el(M

olda

nova

etal

20

09)

667

042

ndashndash

ndash14

22

ndash1

0310

320

200

97H

FO1

9

Oce

ango

ing

carg

ove

ssel

614plusmn

10

83plusmn

001

ndashndash

ndash16

3plusmn

02

ndash1

51plusmn

007

87plusmn

01

128

IFO

180

(Cel

oet

al

2015

)62

6plusmn

70

26plusmn

001

114plusmn

01

081plusmn

002

58plusmn

007

525

IFO

180

628plusmn

90

30plusmn

001

113plusmn

01

094plusmn

002

87plusmn

01

450

IFO

180

628plusmn

10

81plusmn

003

122plusmn

001

083plusmn

001

64plusmn

01

450

IFO

180

609plusmn

11

31plusmn

002

ndashndash

ndash8

4plusmn

003

ndash0

37plusmn

001

47plusmn

001

440

IFO

6060

5plusmn

10

00ndash

ndashndash

167plusmn

01

ndash2

2plusmn

02

103plusmn

003

116

IFO

380

622plusmn

11

22plusmn

002

ndashndash

ndash10

7plusmn

004

ndash0

30plusmn

003

047plusmn

01

450

MD

O

AE

aux

iliar

yen

gine

DO

die

selo

ilH

FOh

eavy

fuel

oil

MG

Om

arin

ega

soi

lR

Or

esid

ualo

ilM

DO

mar

ine

dies

eloi

lD

Fdi

esel

fuel

IFO

int

erm

edia

tefu

eloi

l

wwwatmos-chem-physnet1663192016 Atmos Chem Phys 16 6319ndash6334 2016

6330 F Zhang et al Emission factors for gaseous and particulate pollutants

as 100 to 132 g kW hminus1 of NOx EF and 11 to 17 g kW hminus1

of CO EF used for inland ships for ship inventory calcula-tion (Zhu et al 2015) 100 to 181 g kW hminus1 of NOx EFand 11 to 15 g kW hminus1 of CO EF for harbour ships (Yanget al 2015) compared to 15 to 173 g kW hminus1 of NOx EFand 46 to 103 g kW hminus1 of CO EF from test inland ships(convert the fuel-based EF to power-based EF with a factorof 200 g kW hminus1) (Song 2015) and 697 to 258 g kW hminus1 ofNOx EF and 139 to 738 g kW hminus1 of CO EF in the presentstudy (Yang et al 2015) Besides whether there are obviousdifferences of EFs between other types of vessels in China(such as low-speed engine vessels with heavy fuel oil) andprevious studies is still unclear Therefore much more mea-surement data for different vessels in China are still in urgentneed for more accurate assessment of shipping emissions

35 Impact of engine speed on NOx emission factors

NOx is formed in the combustion chamber by a combinationof atmospheric nitrogen and oxygen under high-pressure andhigh-temperature conditions Many factors affect NOx for-mation including engine temperature injection point andfuel quality The IMO emissions limit for NOx is determinedby the rated speed of the engine however other factors mustalso be considered to reduce NOx emissions

The NOx EFs for the test vessels at various engine speedsare shown in Fig 4 The rated speeds of the vessels were1200 900 and 1000 rpm for HH DFH and XYH respec-tively The actual engine speeds of HH were much lower thanthe rated speed while the two larger-engine-power vesselsoperated close to their rated speeds except during one op-erating mode of DFH The NOx EFs for HH differed sig-nificantly in different operating modes ranging from 391to 143 g kgminus1 of fuel The NOx EF was highest when theengine speed reached sim 750 rpm (Fig 4) At lower enginespeeds the NOx EFs had fluctuating but lower values Athigher engine speeds closer to the rated speed of 1200 rpmthe NOx EFs were much lower The NOx EFs for the twolarger-engine-power vessels changed slightly with enginespeed but also had lowest values when their engine speedsapproached their rated speeds Combined with the dieselpropulsion characteristic curve there were large increasesin the fuel consumption rate when the engine speed in-creased Therefore a best-fit engine speed should be deter-mined based on both EFs and economic costs

Engineering approaches for reducing the NOx emissionsof marine engines may be applied before during or afterthe combustion process (Verschaeren et al 2014 Habib etal 2014) In the present study the NOx EFs of the two re-search vessels were below the IMO Tier I emissions limitsHowever for EMS measures should be taken to meet theIMO emissions limit including increasing the engine speedand applying engineering technologies during or after com-bustion such as exhaust gas recirculation (EGR) selectivenon-catalytic reduction (SNCR) or SCR

Figure 4 Emission factors for NOx at different engine speeds

4 Conclusions

Three offshore vessels with different engine power sourceswere chosen in this study to collect measured data of gaseousspecies and particulate matter including NO2 NO N2OCO CO2 TVOCs SO2 and the total suspended particulateBesides chemical composition of the PM were also analysedto give detailed EFs for OC EC water-soluble ions and metalelements Concentrations fuel-based EFs fuel-based aver-age EFs as well as power-based average EFs for species ofoffshore vessels in China were presented Furthermore im-pact of engine speed on NOx EFs was also discussed

There were higher concentrations of pollutants for low-engine-power vessel HH than for the other two vessels COconcentrations for offshore vessels were slightly lower thaninland vessels in China and all the three vessels had the low-est CO concentrations at their economic speeds (the speed ofthe least vessel operating expenditures during one voyagethey were high-speed mode cruise mode and high-speedmode for HH DFH and XYH respectively) More than 80 of the NOx was NO and all the offshore vessels had higherNOx concentrations in high-speed modes Because of thelow-sulfur fuels used in this study SO2 concentrations ofthese three offshore vessels were lower than that in the lit-eratures And the PM concentrations were much lower thaninland vessels while showing significant differences amongdifferent operating modes

Fuel-based EFs for gaseous species and PM were pre-sented based on the carbon balance method EFs for CO2were 2940ndash3106 3121ndash3160 and 3102ndash3162 g kgminus1 fuel forHH DFH and XYH Because of the combustion conditionssuch as excess air ratio combustion temperature and unifor-mity of the fuelndashair mixture EFs for CO showed high valuesin idling mode but low values in economic speed All theoffshore vessels had higher NOx EFs in low speed than inhigh speed but showed higher values when in acceleration

Atmos Chem Phys 16 6319ndash6334 2016 wwwatmos-chem-physnet1663192016

F Zhang et al Emission factors for gaseous and particulate pollutants 6331

process EFs for SO2 were 16 09 and 26 g kgminus1 fuel forHH DFH and XYH based on sulfur content of the fuels OCand EC were the main components of PM with low OC toEC ratios that were lower than 01 followed by SO2minus

4 NH+4 and NOminus3 Metals such as V Ni Cr Fe As and Cd made upa proportionately small part of the total PM mass

Fuel-based average EFs as well as power-based EFs for thethree different vessels of differing engine power were pre-sented EFs for most gaseous species and PM of HH weremuch higher compared with the other higher-engine-powervessels which was also gt 100 above the IMO standard forNOx Average PM EF of the low-engine-power vessel HHwas also much higher than that in the literatures Howeveraverage EFs for most species of the two larger-engine-powervessels were within the range of previously reported resultsEngine type was inferred as one of the most influence fac-tors for the differences of emission factors Inland and someoffshore ships in China always had higher NOx CO and PMemissions compared with other test vessels in previous stud-ies In addition emission factors that used for calculation ofship inventories in China always had lower values than testvessels

The impact of engine speed on EFs for NOx showed thatwhen the engine speed was close to the rated speed therewould be lower NOx EFs values However combined withthe high fuel consumption rate an optimal engine speedshould be determined based on both EFs and economic costsEmission reduction measures for NOx for some of the off-shore vessels in China are still essential to meet the IMOemission limit

Given the limits of vessel types and numbers this studysubstantially gives the EFs for gaseous species and PM ofthree different diesel-engine-powered offshore vessels How-ever as the development of ports in China emissions fromcargo ships and container ships with large engine power havebecome one of the most significant air pollution sources inport cities and regions Systematical measurement EFs of allkinds of offshore vessels in China are essential in order topresent the accurate emission inventory of ships

Information about the Supplement

Supplementary information includes the details of the real-world measurement system for vessels (Fig S1) the concen-trations of main gaseous matter and PM of shipping emis-sions (Fig S2) the types composition of offshore vessels inChina (Table S1) the distribution of vessels through grosstonnage in 2014 in offshore area of Yangtze River Delta (Ta-ble S2) the Chinese market consumption of marine oil in2011 (Table S3) the detection parameters for gaseous mat-ter (Table S4) the fuel-based EFs for the gaseous pollutants(Table S5) PM and the chemical composition in PM for dif-ferent operating (Table S6) modes and the actual operatingconditions of vessels (Table S7)

The Supplement related to this article is available onlineat doi105194acp-16-6319-2016-supplement

Acknowledgements This study was supported by the CAS Strate-gic Priority Research Program (No XDB05030303) the NaturalScientific Foundation of China (Nos 41273135 and 41473091)and the Shanghai science and Technology Commission project(No 15DZ1205401) The authors would like to express theirgratitude to the vessel owners Haohai Ocean Engineering LimitedLiability Company for vessel Haohai 0007 Ocean University ofChina for vessel Dongfanghong 02 and North China Sea Branchof the State Oceanic Administration for vessel Xiangyanghong08 for making their vessels available for the shipping emissionsstudy The views expressed in this document are solely those of theauthors and the funding agencies do not endorse any products orcommercial services mentioned in this publication

Edited by T Bond

References

Agrawal H Malloy Q G J Welch W A Miller J W andCocker D R In-use gaseous and particulate matter emissionsfrom a modern ocean going container vessel Atmos Environ42 5504ndash5510 doi101016jatmosenv200802053 2008a

Agrawal H Welch W A Miller J W and Cocker D R Emis-sion measurements from a crude oil tanker at sea Environ SciTechnol 42 7098ndash7103 doi101021es703102y 2008b

Almeida S M Pio C A Freitas M C Reis MA and Trancoso M A Source apportionment of fineand coarse particulate matter in a sub-urban area at theWestern European Coast Atmos Environ 39 3127ndash3138doi101016jatmosenv200501048 2005

Anderson M Salo K Hallquist A M and Fridell ECharacterization of particles from a marine engine op-erating at low loads Atmos Environ 101 65ndash71doi101016jatmosenv201411009 2015

Berg N Mellqvist J Jalkanen J-P and Balzani J Ship emis-sions of SO2 and NO2 DOAS measurements from airborne plat-forms Atmos Meas Tech 5 1085ndash1098 doi105194amt-5-1085-2012 2012

California Code of Regulation Titles 13 and 17 avail-able at httpsgovtwestlawcomcalregsBrowseHomeCaliforniaCaliforniaCodeofRegulationstransitionType=DefaultampcontextData=(scDefault) last access 20 April 2016

Cappa C D Williams E J Lack D A Buffaloe G M Coff-man D Hayden K L Herndon S C Lerner B M Li S-M Massoli P McLaren R Nuaaman I Onasch T B andQuinn P K A case study into the measurement of ship emis-sions from plume intercepts of the NOAA ship Miller FreemanAtmos Chem Phys 14 1337ndash1352 doi105194acp-14-1337-2014 2014

Carlton J Danton S Gawen R Lavender K Mathieson NNewell A Reynolds G Webster A Wills C and Wright AMarine exhaust emissions research programme Lloydrsquos RegisterEngineering Services London 63 1995

wwwatmos-chem-physnet1663192016 Atmos Chem Phys 16 6319ndash6334 2016

6324 F Zhang et al Emission factors for gaseous and particulate pollutants

3 Results and discussion

31 Concentrations in shipping emissions

Concentrations of CO NOx SO2 TVOC and PM fromthe three vessels are shown in Fig S1 in the SupplementNearly all of the concentrations measured in the exhaustof low-engine-power vessel HH were higher than those ofthe two higher-engine-power vessels Concentrations of COSO2 and NOx from HH were 107ndash756 534ndash331 and878ndash1295 ppm respectively and 143ndash595 mg mminus3 PM Incontrast concentrations of CO SO2 NOx and PM were501ndash141 527ndash169 169ndash800 ppm and 706ndash218 mg mminus3respectively for DFH and 360ndash224 049ndash359 and 235ndash578 ppm and 056ndash631 mg mminus3 respectively for XYH

A previous study demonstrated that concentrations of COprimarily depend on engine power with higher CO emissionsresulting from vessel engines with lower power (Sinha et al2003) There was a similar trend in this study with gener-ally higher concentrations for HH and lower concentrationsfor DFH The CO concentrations in the present study weresimilar but slightly lower than those of inland vessels (Fuet al 2013) except in the idling mode of HH In differentoperating modes CO concentrations were significantly dif-ferent For example the maximum value was observed inidling mode and the minimum value in medium-speed modefor HH All three ships had the lowest CO concentrations attheir economic speeds (medium speed for HH cruise modefor DFH and high speed for XYH) demonstrating that theirengines are optimized for the most common operating mode

More than 80 of the NOx was NO in this study withNO2 and N2O accounting for lt 20 in all operating modes(Fig S1) Again nearly all of these concentrations werehigher in the exhaust gas of HH than in that of the two ves-sels In high-speed modes all of the vessels had high con-centrations of NOx NOx emissions mainly depend on thecombustion temperature of the engines More powerful com-bustion systems operate at higher temperatures thereby pro-ducing more NOx (Corbett et al 1999) However the NOx

emissions were much lower than for the inland vessels stud-ied by Fu et al (2013) particularly in cruise mode (NOx con-centrations of sim 1000 ppm)

SO2 concentrations in the exhaust gas depend on the sul-fur content of the fuel and the flow rate of the flue gas Therewere significant differences among the three vessels in theirflow rates which could account for the different concentra-tions of one vessel in different operating modes But becauseof the low-sulfur fuels used in these vessels the SO2 con-centrations were low compared with those in other studies(Williams et al 2009 Berg et al 2012)

Much lower concentrations of PM in the exhaust gas wereobserved in the present study compared to those of inlandships in China (Fu et al 2013) However they were similarto those from ships at berth reported by Cooper et al (Cooper2003) HH had higher PM concentrations than the two ves-

sels in the exhaust gas There were significant differencesamong the different operating modes because of changes inthe injection point of the engines (Sippula et al 2014 Li etal 2014)

32 Fuel-based emission factors

Fuel-based EFs for the gaseous species CO2 CO NO NO2N2O and TVOCs and for PM based on the carbon balancemethod were determined In addition SO2 was calculatedbased on the sulfur content of the fuels Fuel-based EFs forthe typical pollutants such as CO PM and nitrogen oxides indifferent operating modes are shown in Fig 2 (detailed EFsfor all the gaseous pollutants are shown in Table S5 and de-tailed EFs for PM and its chemical composition are shown inTable S6)

CO2 emissions from vessels primarily depend on the car-bon content of the fuel (Carlton et al 1995) Accordinglythe EFs for CO2 in the present study should theoreticallybe 3177 3168 and 3171 g kgminus1 fuel for complete combus-tion Under actual conditions CO2 emissions were 2940ndash3106 3121ndash3160 and 3102ndash3162 g kgminus1 fuel for HH DFHand XYH respectively which means they had combustionefficiencies with 925ndash978 985ndash997 and 978ndash997 interms of CO2 for these three vessels

CO emissions of HH were much higher than of XYH fol-lowed by DFH The power of their respective engines was350 600 and 1600 kW In addition there were large differ-ences in CO emissions among different modes All of thesethree vessels had relatively high EFs for CO while accelerat-ing compared with other modes but the highest EFs wereduring the idling modes of HH and DFH as well as dur-ing the low-speed mode of XYH Because CO emissionsin diesel engines primarily depend on the excess air ratio(which determines the fuelndashair mixture) combustion temper-ature and uniformity of the fuelndashair mixture in the combus-tion chamber (Doug 2004) ship engines with lower powergenerally have higher CO emissions (Carlton et al 1995)Localized hypoxia and incomplete combustion in the cylin-der were the main reasons for CO emission of diesel engineCO emissions always had positive relationships with the air-to-fuel ratio There was lower air-to-fuel ratio when in lowengine load which resulted in lower CO emission and viceversa (Ni 1999)

Much higher NOx EFs were observed for HH than forthe other two vessels These results were inconsistent withthose of Sinha et al (2003) in which emissions of NOx in-creased with the power of the ship engine With increasingvessel speed NOx EFs for HH first increased and then de-creased XYH had lower EFs when operating at high speedthan at low speed Nitrogen oxides included NO NO2 andN2O in the present study More than 70 of the NOx wasin the form of NO for all vessels because most of the NOx

emissions were generated through thermal NO formation(Haglind 2008) The primary reasons that slow diesel en-

Atmos Chem Phys 16 6319ndash6334 2016 wwwatmos-chem-physnet1663192016

F Zhang et al Emission factors for gaseous and particulate pollutants 6325

Figure 2 EFs for the typical pollutants in different operating modes

gines such as the one in HH have higher NOx emissions in-clude higher peak flame temperatures and the NO formationreactions being closer to their equilibrium state than in otherengines (Haglind 2008) NOx emissions from vessels aretemperature-dependent (Sinha et al 2003) and also are in-fluenced by the oxygen concentration in the engine cylinder(Ni 1999) In larger engines the running speed is generallyslower and the combustion process more adiabatic resultingin higher combustion temperatures and more NOx Besideswith the increasing of air-to-fuel ratios concentration of NOx

showed a tendency first to increase then to decrease whichalways had the maximum value in the operating mode thatclose to full load of engine because of the high temperatureand oxygen in the engine cylinder (Ni 1999) Furthermorethere were always higher EF values in acceleration processand lower in moderating process in this study When the en-gines were in transient operating conditions such as acceler-ation process or moderating process concentrations of NOx

always had corresponding changes in the cylinder Studiesabout diesel engines showed that when the rotational speedhad a sudden increase there would be a first increasing then

decreasing and last stable tendency for the NOx concentra-tions and vice versa (Tan et al 2012)

TVOCs emissions from HH were much higher than fromthe other two vessels the lowest emissions were observedfor DFH Previous studies (Sinha et al 2003) have reportedthat hydrocarbon emissions from vessels depend on enginepower with low-power engines emitting more hydrocarbonsThe present results were partially consistent with these pre-vious studies Besides hydrocarbon emissions also dependon the percentage utilization of engine power (Sinha et al2003) As for various operating modes TVOCs EFs had largedifferences For example HH had the highest TVOCs emis-sions in accelerating mode which was almost 3 times theheight of the lowest value in mediumndashspeed mode The EFsfor SO2 depended solely on the sulfur content of the fuelsand were 16 09 and 26 g kgminus1 fuel for HH DFH andXYH respectively in this study Hydrocarbon could be gen-erated because of the incomplete combustion For examplein diesel cylinders there will always be air present in wallregions and crevices this is also the case when scavenging

wwwatmos-chem-physnet1663192016 Atmos Chem Phys 16 6319ndash6334 2016

6326 F Zhang et al Emission factors for gaseous and particulate pollutants

occurred during the aeration which could cause the unevenmixing of air and fuel (Ni 1999)

Fuel-based EFs for PM and its chemical components wereshown in Table S6 OC and EC were the main componentsof PM followed by SO2minus

4 NH+4 and NOminus3 Metals such asV Ni Cr Fe As and Cd made up a proportionately smallpart of the total PM mass However other rare elements suchas Tb Er Yb and Lu had higher values than did some ofthe common elements PM was an in-process product duringthe combustion in a cylinder the forming process includedthe molecular cracking decomposition and polymerizationwhich resulted in lack of oxygen High temperature and oxy-gen deficiency were the main reasons for the formation indiesel engines which always had high concentration val-ues in high load operating modes (Ni 1999) HH had muchhigher PM emission factors than the other two vessels theengine type was considered to be the most significant influ-ence factor which had a good agreement with NOx emissionfactors

EFs for OC and EC and the ratios between them are shownin Fig 3 EFs for OC and EC for HH were higher than forthe other two vessels Organic matter (OM) is generally cal-culated as OCtimes 12 (Petzold et al 2008) to account for themass of elements other than carbon in the emitted moleculesOM EFs for individual vessels mainly depend on the enginetype and the amount of unburned fuel ie the efficiency ofcombustion (Moldanovaacute et al 2013) BC emissions also de-pend heavily on the engine type (Lack et al 2009) There-fore the different types of engines and their levels of mainte-nance could account for the large differences in OC and ECEFs observed among the three vessels in this study The ratiosof OC-to-EC in the present study were much lower than thosefor large diesel ships reported previously (OC EC= 12)(Moldanova et al 2009) and also lower than that reportedfor a medium-speed vessel (Petzold et al 2010) The usageof non-dilution sampling in this study was one possible rea-son for the lower OC to EC ratio Besides TOR was used tomeasure OC and EC in PM which always had a lower OCcontent compared with other methods (such as TOT) becauseof the different definitions of OC and EC (Khan et al 2012)Compared with other diesel engines the ratios of OC to ECin this study were higher than that of automobile diesel sootin which EC comprises 75ndash80 wt of the total PM (Clagueet al 1999) and also higher than heavy heavy-duty dieseltrucks (HHDDTs) with OC to EC ratios below unit for cruseand transient modes even though higher in cold-startidle andcreep modes (Shah et al 2004)

Studies have shown that SO2minus4 formed from vessel-emitted

SO2 is a major contributor to CCN and ship track forma-tion (Schreier et al 2006 Lauer et al 2007) Sulfate is alsoan important component of PM emitted from vessels In thepresent study EFs for SO2minus

4 were much lower than previ-ously reported (Petzold et al 2008 Agrawal et al 2008a)but similar to those detected by a high-resolution time-of-flight aerosol mass spectrometer in a previous study (Lack

Figure 3 EFs for OC and EC and the ratios between them

et al 2009) This may be because EFs for SO2minus4 are mainly

related to the sulfur content of the fuel SO2minus4 is not gen-

erally emitted directly from the engines but forms after re-lease from the stack (Lack et al 2009) Because PM was col-lected directly from engine emissions in the present study thesulfur-to-sulfate ratios were low (lt 06 for vessels) Otherions such as NOminus3 and NH+4 accounted for a small percent-age of the PM emitted from the vessels compared with SO2minus

4 consistent with previous studies (Lack et al 2009) SO2 ismore easily oxidized to SO3 in catalytic reaction cycles withmetals commonly present in the exhaust gas (V Ni) whilehydroxyl radicals are additional needed to convert NOx toNOminus3 (Moldanova et al 2009)

Na+ and Clminus were considered to originate from marineair Their concentrations were highly correlated (r2

= 078)the differing air demands of the engines under different con-ditions might have caused observed variations in the EFs rel-ative to the fuel demand

The elemental composition of PM in the present studydiffered from previous studies showing high elemental con-tent of S Ca V Fe Cu Ni and Al (Agrawal et al 2008aMoldanova et al 2009) V and Ni are typically associatedwith combustion of heavy fuel oil (Almeida et al 2005) Inthe present study the high-quality fuels resulted in low EFsfor V and Ni In our previous study PM from shipping emis-sions was estimated to account for 294 of the total PM25at Tuoji Island in China using V as a tracer of shipping emis-sions (Zhang et al 2014) Reconsidering the former resultsbased on the EFs obtained in the present study we deter-mined that the contribution of vessels near Tuoji Island hadbeen underestimated because the estimate should have in-cluded both heavy and other types of fuels However somerare elements such as Tb Er Yb and Lu had relatively highEFs compared with those of other elements in the presentstudy which may be related to the source of the fuels

Atmos Chem Phys 16 6319ndash6334 2016 wwwatmos-chem-physnet1663192016

F Zhang et al Emission factors for gaseous and particulate pollutants 6327

Tabl

e3

Fuel

-bas

edav

erag

eE

Fsin

the

pres

ents

tudy

and

prev

ious

stud

ies

(gkgminus

1fu

el)

Ves

selI

DC

O2

CO

NO

NO

2N

2ON

Ox

TV

OC

sPM

SO2

Sco

nten

t(

m)

HH

3071plusmn

1565

302plusmn

162

982plusmn

372

155plusmn

545

128plusmn

170

115plusmn

443

237plusmn

210

940plusmn

213

160

008

DFH

3153plusmn

176

693plusmn

100

302plusmn

160

509plusmn

042

038plusmn

018

357plusmn

220

124plusmn

004

072plusmn

033

092

005

XY

H31

51plusmn

175

920plusmn

295

266plusmn

163

471plusmn

042

030plusmn

015

316plusmn

220

418plusmn

015

016plusmn

007

260

013

Com

mer

cial

vess

el(W

illia

ms

etal

20

09)

3170

7ndash16

ndashndash

ndash60

ndash87

ndashndash

6ndash30

Car

gove

ssel

(Mol

dano

vaet

al

2009

)

3441

217

ndashndash

ndash73

4ndash

53

393

19

Die

sele

ngin

e(H

aglin

d20

08)

ndash7

4ndash

ndashndash

87ndash

76

542

7

Oce

an-g

oing

ship

s(S

inha

etal

20

03)

3135

195

ndashndash

ndash22

3ndash

ndash2

90

1

Oce

an-g

oing

ship

s(S

inha

etal

20

03)

3176

30

ndashndash

ndash65

5ndash

ndash52

22

4

Car

goan

dpa

ssen

ger

ship

s(E

ndre

sen

2003

)31

707

4ndash

ndash0

0857

ndash87

24

12ndash

76

10ndash5

40

5ndash2

7

Ship

sop

erat

ing

inndash

ndash42

ndash72

ndashndash

65ndash8

6ndash

ndash4

6ndash9

8N

ON

Eha

rbou

rare

as(P

irjo

laet

al

2014

)ndash

ndash16

ndash49

ndashndash

25ndash7

9ndash

ndash5

4ndash17

0SC

R

Ship

sop

erat

ing

inPo

rt(D

iesc

het

al

2013

)ndash

ndash16

37ndash

53ndash

ndash7

7

NO

NE=

No

trea

tmen

tofe

mis

sion

sSC

R=

sele

ctiv

eca

taly

ticre

duct

ion

wwwatmos-chem-physnet1663192016 Atmos Chem Phys 16 6319ndash6334 2016

6328 F Zhang et al Emission factors for gaseous and particulate pollutants

33 Fuel-based average emission factors

Based on actual operating conditions (Table S7) average EFsfor the three vessels in the present study (according to Eq 4)along with EFs from previous studies are shown in Table 3EFs for all of the pollutants except SO2 were significantlyhigher for HH than for the other two vessels potentially dueto poor combustion conditions Most of the EFs for DFH andXYH were within the range of emissions for other vesselsdue to having well maintained engines and the high qualityof the fuels used The EFs for NOx PM and SO2 were muchlower than reported in previous studies (other than NOx forocean-going vessels) All the sulfur of the fuels in the presentstudy were significantly below the emissions limit of 350 established by IMO in the revised MARPOL Annex VI rulesapplicable since 2012 (IMO 1998)

The IMO Tier I emissions limit for NOx is450times nminus02 g kWhminus1 (n rated speed 130 lt nlt 2000 rpm)The rated speed and fuel consumption rates for each vesselare shown in Table 1 Thus the emissions limits for HHDFH and XYH would be 545 575 and 565 g kgminus1 fuelrespectively calculated combined with Eq (5) The averagefuel-based EFs for NOx of ship HH was more than 100 above the IMO standard while those of the other two shipswere below the IMO standard (Table 3) PM emissions forHH were also higher than previously reported but thosefor the two research vessels were much lower (Table 3)Fuel type is one of the most important influence factors onpollutant emissions for example sulfur content in the fuelnot only influence the SO2 emission directly but also hadimpact on PM formation in the flue gas stack with low sulfurcontent in fuels reduces PM formation (Lack et al 2011)Vessels with higher sulfur content always had relativelyhigher PM emissions which were also shown in Table 3In addition different engines and levels of maintenancehave a significant impact on all combustion-dependentemissions Emission reduction measures have been used insome vessels For example NOx emissions can be reducedby measures such as selective catalytic reduction (SCR) anddirect water injection (DWI) which had been implementedon some vessels previously studied in a harbour in Finland(Pirjola et al 2014) The results showed that SCR effec-tively reduced NOx emissions while vessels with DWI hadhigh PM emissions The engine type might be an importantcause of the different emissions such as HH had muchhigher pollutants emissions with an engine produced inChina and yet DFHrsquos engine produced in Germany Besidesemission tests for a high-speed marine diesel engine withdifferent kinds of diesels showed that diesel type had limitedinfluence on emissions such as NOx CO and CH but asignificant impact on PM emission (289ndash415 ) becauseof the different sulfur content in fuel (Xu 2008)

34 Power-based emission factors

Based on the engine power and fuel consumption rates ofthe vessels power-based EFs were calculated (according toEq 5) and compared to results from previous studies (Ta-ble 4) The EFs for HH were much higher than those forthe other two vessels except for SO2 HH also had signif-icantly higher EFs for NOx than previously reported val-ues while EFs for NOx of DFH and XYH were within therange of previously reported results Engine type was consid-ered to be a significant influence factor for NOx emissionswith lower engine speed having higher NOx emission factors(Celo et al 2015) In addition compared to other vesselswith a similar engine type and diesel fuel HH still had rela-tively higher NOx EF (seen in Table 4) which could reflectthe impact of engine condition (engine quality and mainte-nance level) on shipping emissions CO EFs for the test ves-sels in the present study were higher than previous studieswhich produced similar results to those of inland ships andother test vessels in China (Fu et al 2013 Song 2015) Inspite of the influence of engine type on CO emissions thatwith the higher engine speed having higher CO EFs (Celoet al 2015) engine condition combined with fuel qualitymight have significant influence All of the EFs for SO2 inthe present study were lower than those in previous studiesbecause of the low sulfur content of the present fuels Gen-erally PM emissions from marine diesel fuels are dependenton the fuel (sulfate and metal oxide ash constituents) and oncombustion conditions (unburned hydrocarbons and carbonresidue constituents) (Cooper 2003) HH had the highest PMemissions among the test vessels although there were almostno differences among the fuels (Table S6) Besides HH hadeven higher PM EFs than previously reported vessels withHFO fuel and XYH had much lower PM EFs than all theother vessels with even lower sulfur content fuel Thereforecombustion conditions were likely the determining factor forthe differences It can be seen from Table 4 that most previ-ous studies focused on the heavy fuel oil of shipping emis-sions Compared with diesel fuels heavy fuel oil always hadrelatively low CO emission factors and high PM emissionfactors And among the heavy-fuel-oil-using vessels enginetype (engine speed and engine power level) always playedan important role on emissions such as NOx and CO whichwith lower engine speed having higher NOx EFs and lowerCO EFs

Combined with other emission data of test ships in China(Fu et al 2013 Song 2015) it could be seen that inlandand some test offshore ships in China always had higherNOx CO and PM emissions compared with other test ves-sels in previous studies And among the test vessels in Chinathere also were differences for different engine types andship types In addition emission factors that were used forcalculation of ship inventories in China always came fromother countries and areas However there seemed to be sig-nificant differences between the reference and test data such

Atmos Chem Phys 16 6319ndash6334 2016 wwwatmos-chem-physnet1663192016

F Zhang et al Emission factors for gaseous and particulate pollutants 6329

Tabl

e4

Pow

er-b

ased

EFs

inth

epr

esen

tstu

dyan

dpr

evio

usst

udie

s(g

kWhminus

1 )

Ves

sel

CO

2C

ON

ON

O2

N2O

NO

xT

VO

Cs

PMSO

2E

ngin

epo

wer

Eng

ine

Fuel

type

and

ID(k

W)

spee

dsu

lfur

cont

ent

(rm

p)(w

t)

HH

699plusmn

352

738plusmn

376

220plusmn

841

345plusmn

124

030plusmn

039

258plusmn

100

544plusmn

484

209plusmn

048

036

350

1200

DO

00

798

DFH

631plusmn

352

139plusmn

020

604plusmn

032

102plusmn

008

008plusmn

004

714plusmn

044

017plusmn

001

014plusmn

007

018

1600

900

DO

00

458

XY

H69

7plusmn

385

201plusmn

065

587plusmn

036

104plusmn

009

007plusmn

003

697plusmn

048

092plusmn

ndash0

04plusmn

001

057

600

1000

DO

01

30Ta

nker

(Win

nes

and

Frid

ell

2010

)ndash

161

ndashndash

ndash7

82ndash

058

ndash45

0060

0H

FO1

6

Ber

thed

ship

s(C

oope

r65

3ndash76

80

33ndash0

97

ndashndash

ndash14

2ndash2

02

ndash0

14ndash0

45

026

ndash53

AE

720

ndash127

072

0ndash18

00M

GO

00

6ndash1

220

03)

691ndash

803

077

ndash17

112

9ndash1

75

048

ndash06

72

5ndash9

612

70ndash2

675

720ndash

750

RO

05

3ndash2

269

1ndash69

40

92ndash0

98

96ndash

99

017

ndash01

91

014

8072

0M

DO

02

3C

rude

oilt

anke

r(A

graw

alet

al

2008

b)58

8ndash66

00

77ndash1

78

ndashndash

ndash15

8ndash2

10

ndash1

10ndash1

78

766

ndash86

015

750

90H

FO2

85

Cru

ise

ship

s(P

opla

wsk

iet

al

2011

)ndash

ndashndash

140

ndashndash

ndash2

914

20

US

EPA

621

14

ndashndash

ndash18

1ndash

131

103

Mar

ine

engi

ne(S

ippu

laet

ndash1

2ndash11

411

3ndash2

95

ndashndash

114

ndash30

90ndash

95

083

ndash63

6ndash

1500

HFO

27

al

2014

)0ndash

885

69ndash2

58

584

ndash33

90

83ndash1

97

015

ndash09

3D

FL

arge

mar

ine

ship

s(K

han

etal

20

13)

533ndash

612

035

ndash06

0ndash

ndashndash

166

ndash20

6ndash

091

ndash21

97

2ndash11

436

740ndash

6853

097

ndash102

HFO

21

5ndash3

14

Oce

ango

ing

cont

aine

rve

ssel

(Agr

awal

etal

20

08a)

588ndash

660

077

ndash18

1ndash

ndashndash

158

ndash21

0ndash

109

ndash17

67

66ndash8

60

5027

010

4H

FO2

05

Lar

geca

rgo

vess

el(M

olda

nova

etal

20

09)

667

042

ndashndash

ndash14

22

ndash1

0310

320

200

97H

FO1

9

Oce

ango

ing

carg

ove

ssel

614plusmn

10

83plusmn

001

ndashndash

ndash16

3plusmn

02

ndash1

51plusmn

007

87plusmn

01

128

IFO

180

(Cel

oet

al

2015

)62

6plusmn

70

26plusmn

001

114plusmn

01

081plusmn

002

58plusmn

007

525

IFO

180

628plusmn

90

30plusmn

001

113plusmn

01

094plusmn

002

87plusmn

01

450

IFO

180

628plusmn

10

81plusmn

003

122plusmn

001

083plusmn

001

64plusmn

01

450

IFO

180

609plusmn

11

31plusmn

002

ndashndash

ndash8

4plusmn

003

ndash0

37plusmn

001

47plusmn

001

440

IFO

6060

5plusmn

10

00ndash

ndashndash

167plusmn

01

ndash2

2plusmn

02

103plusmn

003

116

IFO

380

622plusmn

11

22plusmn

002

ndashndash

ndash10

7plusmn

004

ndash0

30plusmn

003

047plusmn

01

450

MD

O

AE

aux

iliar

yen

gine

DO

die

selo

ilH

FOh

eavy

fuel

oil

MG

Om

arin

ega

soi

lR

Or

esid

ualo

ilM

DO

mar

ine

dies

eloi

lD

Fdi

esel

fuel

IFO

int

erm

edia

tefu

eloi

l

wwwatmos-chem-physnet1663192016 Atmos Chem Phys 16 6319ndash6334 2016

6330 F Zhang et al Emission factors for gaseous and particulate pollutants

as 100 to 132 g kW hminus1 of NOx EF and 11 to 17 g kW hminus1

of CO EF used for inland ships for ship inventory calcula-tion (Zhu et al 2015) 100 to 181 g kW hminus1 of NOx EFand 11 to 15 g kW hminus1 of CO EF for harbour ships (Yanget al 2015) compared to 15 to 173 g kW hminus1 of NOx EFand 46 to 103 g kW hminus1 of CO EF from test inland ships(convert the fuel-based EF to power-based EF with a factorof 200 g kW hminus1) (Song 2015) and 697 to 258 g kW hminus1 ofNOx EF and 139 to 738 g kW hminus1 of CO EF in the presentstudy (Yang et al 2015) Besides whether there are obviousdifferences of EFs between other types of vessels in China(such as low-speed engine vessels with heavy fuel oil) andprevious studies is still unclear Therefore much more mea-surement data for different vessels in China are still in urgentneed for more accurate assessment of shipping emissions

35 Impact of engine speed on NOx emission factors

NOx is formed in the combustion chamber by a combinationof atmospheric nitrogen and oxygen under high-pressure andhigh-temperature conditions Many factors affect NOx for-mation including engine temperature injection point andfuel quality The IMO emissions limit for NOx is determinedby the rated speed of the engine however other factors mustalso be considered to reduce NOx emissions

The NOx EFs for the test vessels at various engine speedsare shown in Fig 4 The rated speeds of the vessels were1200 900 and 1000 rpm for HH DFH and XYH respec-tively The actual engine speeds of HH were much lower thanthe rated speed while the two larger-engine-power vesselsoperated close to their rated speeds except during one op-erating mode of DFH The NOx EFs for HH differed sig-nificantly in different operating modes ranging from 391to 143 g kgminus1 of fuel The NOx EF was highest when theengine speed reached sim 750 rpm (Fig 4) At lower enginespeeds the NOx EFs had fluctuating but lower values Athigher engine speeds closer to the rated speed of 1200 rpmthe NOx EFs were much lower The NOx EFs for the twolarger-engine-power vessels changed slightly with enginespeed but also had lowest values when their engine speedsapproached their rated speeds Combined with the dieselpropulsion characteristic curve there were large increasesin the fuel consumption rate when the engine speed in-creased Therefore a best-fit engine speed should be deter-mined based on both EFs and economic costs

Engineering approaches for reducing the NOx emissionsof marine engines may be applied before during or afterthe combustion process (Verschaeren et al 2014 Habib etal 2014) In the present study the NOx EFs of the two re-search vessels were below the IMO Tier I emissions limitsHowever for EMS measures should be taken to meet theIMO emissions limit including increasing the engine speedand applying engineering technologies during or after com-bustion such as exhaust gas recirculation (EGR) selectivenon-catalytic reduction (SNCR) or SCR

Figure 4 Emission factors for NOx at different engine speeds

4 Conclusions

Three offshore vessels with different engine power sourceswere chosen in this study to collect measured data of gaseousspecies and particulate matter including NO2 NO N2OCO CO2 TVOCs SO2 and the total suspended particulateBesides chemical composition of the PM were also analysedto give detailed EFs for OC EC water-soluble ions and metalelements Concentrations fuel-based EFs fuel-based aver-age EFs as well as power-based average EFs for species ofoffshore vessels in China were presented Furthermore im-pact of engine speed on NOx EFs was also discussed

There were higher concentrations of pollutants for low-engine-power vessel HH than for the other two vessels COconcentrations for offshore vessels were slightly lower thaninland vessels in China and all the three vessels had the low-est CO concentrations at their economic speeds (the speed ofthe least vessel operating expenditures during one voyagethey were high-speed mode cruise mode and high-speedmode for HH DFH and XYH respectively) More than 80 of the NOx was NO and all the offshore vessels had higherNOx concentrations in high-speed modes Because of thelow-sulfur fuels used in this study SO2 concentrations ofthese three offshore vessels were lower than that in the lit-eratures And the PM concentrations were much lower thaninland vessels while showing significant differences amongdifferent operating modes

Fuel-based EFs for gaseous species and PM were pre-sented based on the carbon balance method EFs for CO2were 2940ndash3106 3121ndash3160 and 3102ndash3162 g kgminus1 fuel forHH DFH and XYH Because of the combustion conditionssuch as excess air ratio combustion temperature and unifor-mity of the fuelndashair mixture EFs for CO showed high valuesin idling mode but low values in economic speed All theoffshore vessels had higher NOx EFs in low speed than inhigh speed but showed higher values when in acceleration

Atmos Chem Phys 16 6319ndash6334 2016 wwwatmos-chem-physnet1663192016

F Zhang et al Emission factors for gaseous and particulate pollutants 6331

process EFs for SO2 were 16 09 and 26 g kgminus1 fuel forHH DFH and XYH based on sulfur content of the fuels OCand EC were the main components of PM with low OC toEC ratios that were lower than 01 followed by SO2minus

4 NH+4 and NOminus3 Metals such as V Ni Cr Fe As and Cd made upa proportionately small part of the total PM mass

Fuel-based average EFs as well as power-based EFs for thethree different vessels of differing engine power were pre-sented EFs for most gaseous species and PM of HH weremuch higher compared with the other higher-engine-powervessels which was also gt 100 above the IMO standard forNOx Average PM EF of the low-engine-power vessel HHwas also much higher than that in the literatures Howeveraverage EFs for most species of the two larger-engine-powervessels were within the range of previously reported resultsEngine type was inferred as one of the most influence fac-tors for the differences of emission factors Inland and someoffshore ships in China always had higher NOx CO and PMemissions compared with other test vessels in previous stud-ies In addition emission factors that used for calculation ofship inventories in China always had lower values than testvessels

The impact of engine speed on EFs for NOx showed thatwhen the engine speed was close to the rated speed therewould be lower NOx EFs values However combined withthe high fuel consumption rate an optimal engine speedshould be determined based on both EFs and economic costsEmission reduction measures for NOx for some of the off-shore vessels in China are still essential to meet the IMOemission limit

Given the limits of vessel types and numbers this studysubstantially gives the EFs for gaseous species and PM ofthree different diesel-engine-powered offshore vessels How-ever as the development of ports in China emissions fromcargo ships and container ships with large engine power havebecome one of the most significant air pollution sources inport cities and regions Systematical measurement EFs of allkinds of offshore vessels in China are essential in order topresent the accurate emission inventory of ships

Information about the Supplement

Supplementary information includes the details of the real-world measurement system for vessels (Fig S1) the concen-trations of main gaseous matter and PM of shipping emis-sions (Fig S2) the types composition of offshore vessels inChina (Table S1) the distribution of vessels through grosstonnage in 2014 in offshore area of Yangtze River Delta (Ta-ble S2) the Chinese market consumption of marine oil in2011 (Table S3) the detection parameters for gaseous mat-ter (Table S4) the fuel-based EFs for the gaseous pollutants(Table S5) PM and the chemical composition in PM for dif-ferent operating (Table S6) modes and the actual operatingconditions of vessels (Table S7)

The Supplement related to this article is available onlineat doi105194acp-16-6319-2016-supplement

Acknowledgements This study was supported by the CAS Strate-gic Priority Research Program (No XDB05030303) the NaturalScientific Foundation of China (Nos 41273135 and 41473091)and the Shanghai science and Technology Commission project(No 15DZ1205401) The authors would like to express theirgratitude to the vessel owners Haohai Ocean Engineering LimitedLiability Company for vessel Haohai 0007 Ocean University ofChina for vessel Dongfanghong 02 and North China Sea Branchof the State Oceanic Administration for vessel Xiangyanghong08 for making their vessels available for the shipping emissionsstudy The views expressed in this document are solely those of theauthors and the funding agencies do not endorse any products orcommercial services mentioned in this publication

Edited by T Bond

References

Agrawal H Malloy Q G J Welch W A Miller J W andCocker D R In-use gaseous and particulate matter emissionsfrom a modern ocean going container vessel Atmos Environ42 5504ndash5510 doi101016jatmosenv200802053 2008a

Agrawal H Welch W A Miller J W and Cocker D R Emis-sion measurements from a crude oil tanker at sea Environ SciTechnol 42 7098ndash7103 doi101021es703102y 2008b

Almeida S M Pio C A Freitas M C Reis MA and Trancoso M A Source apportionment of fineand coarse particulate matter in a sub-urban area at theWestern European Coast Atmos Environ 39 3127ndash3138doi101016jatmosenv200501048 2005

Anderson M Salo K Hallquist A M and Fridell ECharacterization of particles from a marine engine op-erating at low loads Atmos Environ 101 65ndash71doi101016jatmosenv201411009 2015

Berg N Mellqvist J Jalkanen J-P and Balzani J Ship emis-sions of SO2 and NO2 DOAS measurements from airborne plat-forms Atmos Meas Tech 5 1085ndash1098 doi105194amt-5-1085-2012 2012

California Code of Regulation Titles 13 and 17 avail-able at httpsgovtwestlawcomcalregsBrowseHomeCaliforniaCaliforniaCodeofRegulationstransitionType=DefaultampcontextData=(scDefault) last access 20 April 2016

Cappa C D Williams E J Lack D A Buffaloe G M Coff-man D Hayden K L Herndon S C Lerner B M Li S-M Massoli P McLaren R Nuaaman I Onasch T B andQuinn P K A case study into the measurement of ship emis-sions from plume intercepts of the NOAA ship Miller FreemanAtmos Chem Phys 14 1337ndash1352 doi105194acp-14-1337-2014 2014

Carlton J Danton S Gawen R Lavender K Mathieson NNewell A Reynolds G Webster A Wills C and Wright AMarine exhaust emissions research programme Lloydrsquos RegisterEngineering Services London 63 1995

wwwatmos-chem-physnet1663192016 Atmos Chem Phys 16 6319ndash6334 2016

F Zhang et al Emission factors for gaseous and particulate pollutants 6325

Figure 2 EFs for the typical pollutants in different operating modes

gines such as the one in HH have higher NOx emissions in-clude higher peak flame temperatures and the NO formationreactions being closer to their equilibrium state than in otherengines (Haglind 2008) NOx emissions from vessels aretemperature-dependent (Sinha et al 2003) and also are in-fluenced by the oxygen concentration in the engine cylinder(Ni 1999) In larger engines the running speed is generallyslower and the combustion process more adiabatic resultingin higher combustion temperatures and more NOx Besideswith the increasing of air-to-fuel ratios concentration of NOx

showed a tendency first to increase then to decrease whichalways had the maximum value in the operating mode thatclose to full load of engine because of the high temperatureand oxygen in the engine cylinder (Ni 1999) Furthermorethere were always higher EF values in acceleration processand lower in moderating process in this study When the en-gines were in transient operating conditions such as acceler-ation process or moderating process concentrations of NOx

always had corresponding changes in the cylinder Studiesabout diesel engines showed that when the rotational speedhad a sudden increase there would be a first increasing then

decreasing and last stable tendency for the NOx concentra-tions and vice versa (Tan et al 2012)

TVOCs emissions from HH were much higher than fromthe other two vessels the lowest emissions were observedfor DFH Previous studies (Sinha et al 2003) have reportedthat hydrocarbon emissions from vessels depend on enginepower with low-power engines emitting more hydrocarbonsThe present results were partially consistent with these pre-vious studies Besides hydrocarbon emissions also dependon the percentage utilization of engine power (Sinha et al2003) As for various operating modes TVOCs EFs had largedifferences For example HH had the highest TVOCs emis-sions in accelerating mode which was almost 3 times theheight of the lowest value in mediumndashspeed mode The EFsfor SO2 depended solely on the sulfur content of the fuelsand were 16 09 and 26 g kgminus1 fuel for HH DFH andXYH respectively in this study Hydrocarbon could be gen-erated because of the incomplete combustion For examplein diesel cylinders there will always be air present in wallregions and crevices this is also the case when scavenging

wwwatmos-chem-physnet1663192016 Atmos Chem Phys 16 6319ndash6334 2016

6326 F Zhang et al Emission factors for gaseous and particulate pollutants

occurred during the aeration which could cause the unevenmixing of air and fuel (Ni 1999)

Fuel-based EFs for PM and its chemical components wereshown in Table S6 OC and EC were the main componentsof PM followed by SO2minus

4 NH+4 and NOminus3 Metals such asV Ni Cr Fe As and Cd made up a proportionately smallpart of the total PM mass However other rare elements suchas Tb Er Yb and Lu had higher values than did some ofthe common elements PM was an in-process product duringthe combustion in a cylinder the forming process includedthe molecular cracking decomposition and polymerizationwhich resulted in lack of oxygen High temperature and oxy-gen deficiency were the main reasons for the formation indiesel engines which always had high concentration val-ues in high load operating modes (Ni 1999) HH had muchhigher PM emission factors than the other two vessels theengine type was considered to be the most significant influ-ence factor which had a good agreement with NOx emissionfactors

EFs for OC and EC and the ratios between them are shownin Fig 3 EFs for OC and EC for HH were higher than forthe other two vessels Organic matter (OM) is generally cal-culated as OCtimes 12 (Petzold et al 2008) to account for themass of elements other than carbon in the emitted moleculesOM EFs for individual vessels mainly depend on the enginetype and the amount of unburned fuel ie the efficiency ofcombustion (Moldanovaacute et al 2013) BC emissions also de-pend heavily on the engine type (Lack et al 2009) There-fore the different types of engines and their levels of mainte-nance could account for the large differences in OC and ECEFs observed among the three vessels in this study The ratiosof OC-to-EC in the present study were much lower than thosefor large diesel ships reported previously (OC EC= 12)(Moldanova et al 2009) and also lower than that reportedfor a medium-speed vessel (Petzold et al 2010) The usageof non-dilution sampling in this study was one possible rea-son for the lower OC to EC ratio Besides TOR was used tomeasure OC and EC in PM which always had a lower OCcontent compared with other methods (such as TOT) becauseof the different definitions of OC and EC (Khan et al 2012)Compared with other diesel engines the ratios of OC to ECin this study were higher than that of automobile diesel sootin which EC comprises 75ndash80 wt of the total PM (Clagueet al 1999) and also higher than heavy heavy-duty dieseltrucks (HHDDTs) with OC to EC ratios below unit for cruseand transient modes even though higher in cold-startidle andcreep modes (Shah et al 2004)

Studies have shown that SO2minus4 formed from vessel-emitted

SO2 is a major contributor to CCN and ship track forma-tion (Schreier et al 2006 Lauer et al 2007) Sulfate is alsoan important component of PM emitted from vessels In thepresent study EFs for SO2minus

4 were much lower than previ-ously reported (Petzold et al 2008 Agrawal et al 2008a)but similar to those detected by a high-resolution time-of-flight aerosol mass spectrometer in a previous study (Lack

Figure 3 EFs for OC and EC and the ratios between them

et al 2009) This may be because EFs for SO2minus4 are mainly

related to the sulfur content of the fuel SO2minus4 is not gen-

erally emitted directly from the engines but forms after re-lease from the stack (Lack et al 2009) Because PM was col-lected directly from engine emissions in the present study thesulfur-to-sulfate ratios were low (lt 06 for vessels) Otherions such as NOminus3 and NH+4 accounted for a small percent-age of the PM emitted from the vessels compared with SO2minus

4 consistent with previous studies (Lack et al 2009) SO2 ismore easily oxidized to SO3 in catalytic reaction cycles withmetals commonly present in the exhaust gas (V Ni) whilehydroxyl radicals are additional needed to convert NOx toNOminus3 (Moldanova et al 2009)

Na+ and Clminus were considered to originate from marineair Their concentrations were highly correlated (r2

= 078)the differing air demands of the engines under different con-ditions might have caused observed variations in the EFs rel-ative to the fuel demand

The elemental composition of PM in the present studydiffered from previous studies showing high elemental con-tent of S Ca V Fe Cu Ni and Al (Agrawal et al 2008aMoldanova et al 2009) V and Ni are typically associatedwith combustion of heavy fuel oil (Almeida et al 2005) Inthe present study the high-quality fuels resulted in low EFsfor V and Ni In our previous study PM from shipping emis-sions was estimated to account for 294 of the total PM25at Tuoji Island in China using V as a tracer of shipping emis-sions (Zhang et al 2014) Reconsidering the former resultsbased on the EFs obtained in the present study we deter-mined that the contribution of vessels near Tuoji Island hadbeen underestimated because the estimate should have in-cluded both heavy and other types of fuels However somerare elements such as Tb Er Yb and Lu had relatively highEFs compared with those of other elements in the presentstudy which may be related to the source of the fuels

Atmos Chem Phys 16 6319ndash6334 2016 wwwatmos-chem-physnet1663192016

F Zhang et al Emission factors for gaseous and particulate pollutants 6327

Tabl

e3

Fuel

-bas

edav

erag

eE

Fsin

the

pres

ents

tudy

and

prev

ious

stud

ies

(gkgminus

1fu

el)

Ves

selI

DC

O2

CO

NO

NO

2N

2ON

Ox

TV

OC

sPM

SO2

Sco

nten

t(

m)

HH

3071plusmn

1565

302plusmn

162

982plusmn

372

155plusmn

545

128plusmn

170

115plusmn

443

237plusmn

210

940plusmn

213

160

008

DFH

3153plusmn

176

693plusmn

100

302plusmn

160

509plusmn

042

038plusmn

018

357plusmn

220

124plusmn

004

072plusmn

033

092

005

XY

H31

51plusmn

175

920plusmn

295

266plusmn

163

471plusmn

042

030plusmn

015

316plusmn

220

418plusmn

015

016plusmn

007

260

013

Com

mer

cial

vess

el(W

illia

ms

etal

20

09)

3170

7ndash16

ndashndash

ndash60

ndash87

ndashndash

6ndash30

Car

gove

ssel

(Mol

dano

vaet

al

2009

)

3441

217

ndashndash

ndash73

4ndash

53

393

19

Die

sele

ngin

e(H

aglin

d20

08)

ndash7

4ndash

ndashndash

87ndash

76

542

7

Oce

an-g

oing

ship

s(S

inha

etal

20

03)

3135

195

ndashndash

ndash22

3ndash

ndash2

90

1

Oce

an-g

oing

ship

s(S

inha

etal

20

03)

3176

30

ndashndash

ndash65

5ndash

ndash52

22

4

Car

goan

dpa

ssen

ger

ship

s(E

ndre

sen

2003

)31

707

4ndash

ndash0

0857

ndash87

24

12ndash

76

10ndash5

40

5ndash2

7

Ship

sop

erat

ing

inndash

ndash42

ndash72

ndashndash

65ndash8

6ndash

ndash4

6ndash9

8N

ON

Eha

rbou

rare

as(P

irjo

laet

al

2014

)ndash

ndash16

ndash49

ndashndash

25ndash7

9ndash

ndash5

4ndash17

0SC

R

Ship

sop

erat

ing

inPo

rt(D

iesc

het

al

2013

)ndash

ndash16

37ndash

53ndash

ndash7

7

NO

NE=

No

trea

tmen

tofe

mis

sion

sSC

R=

sele

ctiv

eca

taly

ticre

duct

ion

wwwatmos-chem-physnet1663192016 Atmos Chem Phys 16 6319ndash6334 2016

6328 F Zhang et al Emission factors for gaseous and particulate pollutants

33 Fuel-based average emission factors

Based on actual operating conditions (Table S7) average EFsfor the three vessels in the present study (according to Eq 4)along with EFs from previous studies are shown in Table 3EFs for all of the pollutants except SO2 were significantlyhigher for HH than for the other two vessels potentially dueto poor combustion conditions Most of the EFs for DFH andXYH were within the range of emissions for other vesselsdue to having well maintained engines and the high qualityof the fuels used The EFs for NOx PM and SO2 were muchlower than reported in previous studies (other than NOx forocean-going vessels) All the sulfur of the fuels in the presentstudy were significantly below the emissions limit of 350 established by IMO in the revised MARPOL Annex VI rulesapplicable since 2012 (IMO 1998)

The IMO Tier I emissions limit for NOx is450times nminus02 g kWhminus1 (n rated speed 130 lt nlt 2000 rpm)The rated speed and fuel consumption rates for each vesselare shown in Table 1 Thus the emissions limits for HHDFH and XYH would be 545 575 and 565 g kgminus1 fuelrespectively calculated combined with Eq (5) The averagefuel-based EFs for NOx of ship HH was more than 100 above the IMO standard while those of the other two shipswere below the IMO standard (Table 3) PM emissions forHH were also higher than previously reported but thosefor the two research vessels were much lower (Table 3)Fuel type is one of the most important influence factors onpollutant emissions for example sulfur content in the fuelnot only influence the SO2 emission directly but also hadimpact on PM formation in the flue gas stack with low sulfurcontent in fuels reduces PM formation (Lack et al 2011)Vessels with higher sulfur content always had relativelyhigher PM emissions which were also shown in Table 3In addition different engines and levels of maintenancehave a significant impact on all combustion-dependentemissions Emission reduction measures have been used insome vessels For example NOx emissions can be reducedby measures such as selective catalytic reduction (SCR) anddirect water injection (DWI) which had been implementedon some vessels previously studied in a harbour in Finland(Pirjola et al 2014) The results showed that SCR effec-tively reduced NOx emissions while vessels with DWI hadhigh PM emissions The engine type might be an importantcause of the different emissions such as HH had muchhigher pollutants emissions with an engine produced inChina and yet DFHrsquos engine produced in Germany Besidesemission tests for a high-speed marine diesel engine withdifferent kinds of diesels showed that diesel type had limitedinfluence on emissions such as NOx CO and CH but asignificant impact on PM emission (289ndash415 ) becauseof the different sulfur content in fuel (Xu 2008)

34 Power-based emission factors

Based on the engine power and fuel consumption rates ofthe vessels power-based EFs were calculated (according toEq 5) and compared to results from previous studies (Ta-ble 4) The EFs for HH were much higher than those forthe other two vessels except for SO2 HH also had signif-icantly higher EFs for NOx than previously reported val-ues while EFs for NOx of DFH and XYH were within therange of previously reported results Engine type was consid-ered to be a significant influence factor for NOx emissionswith lower engine speed having higher NOx emission factors(Celo et al 2015) In addition compared to other vesselswith a similar engine type and diesel fuel HH still had rela-tively higher NOx EF (seen in Table 4) which could reflectthe impact of engine condition (engine quality and mainte-nance level) on shipping emissions CO EFs for the test ves-sels in the present study were higher than previous studieswhich produced similar results to those of inland ships andother test vessels in China (Fu et al 2013 Song 2015) Inspite of the influence of engine type on CO emissions thatwith the higher engine speed having higher CO EFs (Celoet al 2015) engine condition combined with fuel qualitymight have significant influence All of the EFs for SO2 inthe present study were lower than those in previous studiesbecause of the low sulfur content of the present fuels Gen-erally PM emissions from marine diesel fuels are dependenton the fuel (sulfate and metal oxide ash constituents) and oncombustion conditions (unburned hydrocarbons and carbonresidue constituents) (Cooper 2003) HH had the highest PMemissions among the test vessels although there were almostno differences among the fuels (Table S6) Besides HH hadeven higher PM EFs than previously reported vessels withHFO fuel and XYH had much lower PM EFs than all theother vessels with even lower sulfur content fuel Thereforecombustion conditions were likely the determining factor forthe differences It can be seen from Table 4 that most previ-ous studies focused on the heavy fuel oil of shipping emis-sions Compared with diesel fuels heavy fuel oil always hadrelatively low CO emission factors and high PM emissionfactors And among the heavy-fuel-oil-using vessels enginetype (engine speed and engine power level) always playedan important role on emissions such as NOx and CO whichwith lower engine speed having higher NOx EFs and lowerCO EFs

Combined with other emission data of test ships in China(Fu et al 2013 Song 2015) it could be seen that inlandand some test offshore ships in China always had higherNOx CO and PM emissions compared with other test ves-sels in previous studies And among the test vessels in Chinathere also were differences for different engine types andship types In addition emission factors that were used forcalculation of ship inventories in China always came fromother countries and areas However there seemed to be sig-nificant differences between the reference and test data such

Atmos Chem Phys 16 6319ndash6334 2016 wwwatmos-chem-physnet1663192016

F Zhang et al Emission factors for gaseous and particulate pollutants 6329

Tabl

e4

Pow

er-b

ased

EFs

inth

epr

esen

tstu

dyan

dpr

evio

usst

udie

s(g

kWhminus

1 )

Ves

sel

CO

2C

ON

ON

O2

N2O

NO

xT

VO

Cs

PMSO

2E

ngin

epo

wer

Eng

ine

Fuel

type

and

ID(k

W)

spee

dsu

lfur

cont

ent

(rm

p)(w

t)

HH

699plusmn

352

738plusmn

376

220plusmn

841

345plusmn

124

030plusmn

039

258plusmn

100

544plusmn

484

209plusmn

048

036

350

1200

DO

00

798

DFH

631plusmn

352

139plusmn

020

604plusmn

032

102plusmn

008

008plusmn

004

714plusmn

044

017plusmn

001

014plusmn

007

018

1600

900

DO

00

458

XY

H69

7plusmn

385

201plusmn

065

587plusmn

036

104plusmn

009

007plusmn

003

697plusmn

048

092plusmn

ndash0

04plusmn

001

057

600

1000

DO

01

30Ta

nker

(Win

nes

and

Frid

ell

2010

)ndash

161

ndashndash

ndash7

82ndash

058

ndash45

0060

0H

FO1

6

Ber

thed

ship

s(C

oope

r65

3ndash76

80

33ndash0

97

ndashndash

ndash14

2ndash2

02

ndash0

14ndash0

45

026

ndash53

AE

720

ndash127

072

0ndash18

00M

GO

00

6ndash1

220

03)

691ndash

803

077

ndash17

112

9ndash1

75

048

ndash06

72

5ndash9

612

70ndash2

675

720ndash

750

RO

05

3ndash2

269

1ndash69

40

92ndash0

98

96ndash

99

017

ndash01

91

014

8072

0M

DO

02

3C

rude

oilt

anke

r(A

graw

alet

al

2008

b)58

8ndash66

00

77ndash1

78

ndashndash

ndash15

8ndash2

10

ndash1

10ndash1

78

766

ndash86

015

750

90H

FO2

85

Cru

ise

ship

s(P

opla

wsk

iet

al

2011

)ndash

ndashndash

140

ndashndash

ndash2

914

20

US

EPA

621

14

ndashndash

ndash18

1ndash

131

103

Mar

ine

engi

ne(S

ippu

laet

ndash1

2ndash11

411

3ndash2

95

ndashndash

114

ndash30

90ndash

95

083

ndash63

6ndash

1500

HFO

27

al

2014

)0ndash

885

69ndash2

58

584

ndash33

90

83ndash1

97

015

ndash09

3D

FL

arge

mar

ine

ship

s(K

han

etal

20

13)

533ndash

612

035

ndash06

0ndash

ndashndash

166

ndash20

6ndash

091

ndash21

97

2ndash11

436

740ndash

6853

097

ndash102

HFO

21

5ndash3

14

Oce

ango

ing

cont

aine

rve

ssel

(Agr

awal

etal

20

08a)

588ndash

660

077

ndash18

1ndash

ndashndash

158

ndash21

0ndash

109

ndash17

67

66ndash8

60

5027

010

4H

FO2

05

Lar

geca

rgo

vess

el(M

olda

nova

etal

20

09)

667

042

ndashndash

ndash14

22

ndash1

0310

320

200

97H

FO1

9

Oce

ango

ing

carg

ove

ssel

614plusmn

10

83plusmn

001

ndashndash

ndash16

3plusmn

02

ndash1

51plusmn

007

87plusmn

01

128

IFO

180

(Cel

oet

al

2015

)62

6plusmn

70

26plusmn

001

114plusmn

01

081plusmn

002

58plusmn

007

525

IFO

180

628plusmn

90

30plusmn

001

113plusmn

01

094plusmn

002

87plusmn

01

450

IFO

180

628plusmn

10

81plusmn

003

122plusmn

001

083plusmn

001

64plusmn

01

450

IFO

180

609plusmn

11

31plusmn

002

ndashndash

ndash8

4plusmn

003

ndash0

37plusmn

001

47plusmn

001

440

IFO

6060

5plusmn

10

00ndash

ndashndash

167plusmn

01

ndash2

2plusmn

02

103plusmn

003

116

IFO

380

622plusmn

11

22plusmn

002

ndashndash

ndash10

7plusmn

004

ndash0

30plusmn

003

047plusmn

01

450

MD

O

AE

aux

iliar

yen

gine

DO

die

selo

ilH

FOh

eavy

fuel

oil

MG

Om

arin

ega

soi

lR

Or

esid

ualo

ilM

DO

mar

ine

dies

eloi

lD

Fdi

esel

fuel

IFO

int

erm

edia

tefu

eloi

l

wwwatmos-chem-physnet1663192016 Atmos Chem Phys 16 6319ndash6334 2016

6330 F Zhang et al Emission factors for gaseous and particulate pollutants

as 100 to 132 g kW hminus1 of NOx EF and 11 to 17 g kW hminus1

of CO EF used for inland ships for ship inventory calcula-tion (Zhu et al 2015) 100 to 181 g kW hminus1 of NOx EFand 11 to 15 g kW hminus1 of CO EF for harbour ships (Yanget al 2015) compared to 15 to 173 g kW hminus1 of NOx EFand 46 to 103 g kW hminus1 of CO EF from test inland ships(convert the fuel-based EF to power-based EF with a factorof 200 g kW hminus1) (Song 2015) and 697 to 258 g kW hminus1 ofNOx EF and 139 to 738 g kW hminus1 of CO EF in the presentstudy (Yang et al 2015) Besides whether there are obviousdifferences of EFs between other types of vessels in China(such as low-speed engine vessels with heavy fuel oil) andprevious studies is still unclear Therefore much more mea-surement data for different vessels in China are still in urgentneed for more accurate assessment of shipping emissions

35 Impact of engine speed on NOx emission factors

NOx is formed in the combustion chamber by a combinationof atmospheric nitrogen and oxygen under high-pressure andhigh-temperature conditions Many factors affect NOx for-mation including engine temperature injection point andfuel quality The IMO emissions limit for NOx is determinedby the rated speed of the engine however other factors mustalso be considered to reduce NOx emissions

The NOx EFs for the test vessels at various engine speedsare shown in Fig 4 The rated speeds of the vessels were1200 900 and 1000 rpm for HH DFH and XYH respec-tively The actual engine speeds of HH were much lower thanthe rated speed while the two larger-engine-power vesselsoperated close to their rated speeds except during one op-erating mode of DFH The NOx EFs for HH differed sig-nificantly in different operating modes ranging from 391to 143 g kgminus1 of fuel The NOx EF was highest when theengine speed reached sim 750 rpm (Fig 4) At lower enginespeeds the NOx EFs had fluctuating but lower values Athigher engine speeds closer to the rated speed of 1200 rpmthe NOx EFs were much lower The NOx EFs for the twolarger-engine-power vessels changed slightly with enginespeed but also had lowest values when their engine speedsapproached their rated speeds Combined with the dieselpropulsion characteristic curve there were large increasesin the fuel consumption rate when the engine speed in-creased Therefore a best-fit engine speed should be deter-mined based on both EFs and economic costs

Engineering approaches for reducing the NOx emissionsof marine engines may be applied before during or afterthe combustion process (Verschaeren et al 2014 Habib etal 2014) In the present study the NOx EFs of the two re-search vessels were below the IMO Tier I emissions limitsHowever for EMS measures should be taken to meet theIMO emissions limit including increasing the engine speedand applying engineering technologies during or after com-bustion such as exhaust gas recirculation (EGR) selectivenon-catalytic reduction (SNCR) or SCR

Figure 4 Emission factors for NOx at different engine speeds

4 Conclusions

Three offshore vessels with different engine power sourceswere chosen in this study to collect measured data of gaseousspecies and particulate matter including NO2 NO N2OCO CO2 TVOCs SO2 and the total suspended particulateBesides chemical composition of the PM were also analysedto give detailed EFs for OC EC water-soluble ions and metalelements Concentrations fuel-based EFs fuel-based aver-age EFs as well as power-based average EFs for species ofoffshore vessels in China were presented Furthermore im-pact of engine speed on NOx EFs was also discussed

There were higher concentrations of pollutants for low-engine-power vessel HH than for the other two vessels COconcentrations for offshore vessels were slightly lower thaninland vessels in China and all the three vessels had the low-est CO concentrations at their economic speeds (the speed ofthe least vessel operating expenditures during one voyagethey were high-speed mode cruise mode and high-speedmode for HH DFH and XYH respectively) More than 80 of the NOx was NO and all the offshore vessels had higherNOx concentrations in high-speed modes Because of thelow-sulfur fuels used in this study SO2 concentrations ofthese three offshore vessels were lower than that in the lit-eratures And the PM concentrations were much lower thaninland vessels while showing significant differences amongdifferent operating modes

Fuel-based EFs for gaseous species and PM were pre-sented based on the carbon balance method EFs for CO2were 2940ndash3106 3121ndash3160 and 3102ndash3162 g kgminus1 fuel forHH DFH and XYH Because of the combustion conditionssuch as excess air ratio combustion temperature and unifor-mity of the fuelndashair mixture EFs for CO showed high valuesin idling mode but low values in economic speed All theoffshore vessels had higher NOx EFs in low speed than inhigh speed but showed higher values when in acceleration

Atmos Chem Phys 16 6319ndash6334 2016 wwwatmos-chem-physnet1663192016

F Zhang et al Emission factors for gaseous and particulate pollutants 6331

process EFs for SO2 were 16 09 and 26 g kgminus1 fuel forHH DFH and XYH based on sulfur content of the fuels OCand EC were the main components of PM with low OC toEC ratios that were lower than 01 followed by SO2minus

4 NH+4 and NOminus3 Metals such as V Ni Cr Fe As and Cd made upa proportionately small part of the total PM mass

Fuel-based average EFs as well as power-based EFs for thethree different vessels of differing engine power were pre-sented EFs for most gaseous species and PM of HH weremuch higher compared with the other higher-engine-powervessels which was also gt 100 above the IMO standard forNOx Average PM EF of the low-engine-power vessel HHwas also much higher than that in the literatures Howeveraverage EFs for most species of the two larger-engine-powervessels were within the range of previously reported resultsEngine type was inferred as one of the most influence fac-tors for the differences of emission factors Inland and someoffshore ships in China always had higher NOx CO and PMemissions compared with other test vessels in previous stud-ies In addition emission factors that used for calculation ofship inventories in China always had lower values than testvessels

The impact of engine speed on EFs for NOx showed thatwhen the engine speed was close to the rated speed therewould be lower NOx EFs values However combined withthe high fuel consumption rate an optimal engine speedshould be determined based on both EFs and economic costsEmission reduction measures for NOx for some of the off-shore vessels in China are still essential to meet the IMOemission limit

Given the limits of vessel types and numbers this studysubstantially gives the EFs for gaseous species and PM ofthree different diesel-engine-powered offshore vessels How-ever as the development of ports in China emissions fromcargo ships and container ships with large engine power havebecome one of the most significant air pollution sources inport cities and regions Systematical measurement EFs of allkinds of offshore vessels in China are essential in order topresent the accurate emission inventory of ships

Information about the Supplement

Supplementary information includes the details of the real-world measurement system for vessels (Fig S1) the concen-trations of main gaseous matter and PM of shipping emis-sions (Fig S2) the types composition of offshore vessels inChina (Table S1) the distribution of vessels through grosstonnage in 2014 in offshore area of Yangtze River Delta (Ta-ble S2) the Chinese market consumption of marine oil in2011 (Table S3) the detection parameters for gaseous mat-ter (Table S4) the fuel-based EFs for the gaseous pollutants(Table S5) PM and the chemical composition in PM for dif-ferent operating (Table S6) modes and the actual operatingconditions of vessels (Table S7)

The Supplement related to this article is available onlineat doi105194acp-16-6319-2016-supplement

Acknowledgements This study was supported by the CAS Strate-gic Priority Research Program (No XDB05030303) the NaturalScientific Foundation of China (Nos 41273135 and 41473091)and the Shanghai science and Technology Commission project(No 15DZ1205401) The authors would like to express theirgratitude to the vessel owners Haohai Ocean Engineering LimitedLiability Company for vessel Haohai 0007 Ocean University ofChina for vessel Dongfanghong 02 and North China Sea Branchof the State Oceanic Administration for vessel Xiangyanghong08 for making their vessels available for the shipping emissionsstudy The views expressed in this document are solely those of theauthors and the funding agencies do not endorse any products orcommercial services mentioned in this publication

Edited by T Bond

References

Agrawal H Malloy Q G J Welch W A Miller J W andCocker D R In-use gaseous and particulate matter emissionsfrom a modern ocean going container vessel Atmos Environ42 5504ndash5510 doi101016jatmosenv200802053 2008a

Agrawal H Welch W A Miller J W and Cocker D R Emis-sion measurements from a crude oil tanker at sea Environ SciTechnol 42 7098ndash7103 doi101021es703102y 2008b

Almeida S M Pio C A Freitas M C Reis MA and Trancoso M A Source apportionment of fineand coarse particulate matter in a sub-urban area at theWestern European Coast Atmos Environ 39 3127ndash3138doi101016jatmosenv200501048 2005

Anderson M Salo K Hallquist A M and Fridell ECharacterization of particles from a marine engine op-erating at low loads Atmos Environ 101 65ndash71doi101016jatmosenv201411009 2015

Berg N Mellqvist J Jalkanen J-P and Balzani J Ship emis-sions of SO2 and NO2 DOAS measurements from airborne plat-forms Atmos Meas Tech 5 1085ndash1098 doi105194amt-5-1085-2012 2012

California Code of Regulation Titles 13 and 17 avail-able at httpsgovtwestlawcomcalregsBrowseHomeCaliforniaCaliforniaCodeofRegulationstransitionType=DefaultampcontextData=(scDefault) last access 20 April 2016

Cappa C D Williams E J Lack D A Buffaloe G M Coff-man D Hayden K L Herndon S C Lerner B M Li S-M Massoli P McLaren R Nuaaman I Onasch T B andQuinn P K A case study into the measurement of ship emis-sions from plume intercepts of the NOAA ship Miller FreemanAtmos Chem Phys 14 1337ndash1352 doi105194acp-14-1337-2014 2014

Carlton J Danton S Gawen R Lavender K Mathieson NNewell A Reynolds G Webster A Wills C and Wright AMarine exhaust emissions research programme Lloydrsquos RegisterEngineering Services London 63 1995

wwwatmos-chem-physnet1663192016 Atmos Chem Phys 16 6319ndash6334 2016

6326 F Zhang et al Emission factors for gaseous and particulate pollutants

occurred during the aeration which could cause the unevenmixing of air and fuel (Ni 1999)

Fuel-based EFs for PM and its chemical components wereshown in Table S6 OC and EC were the main componentsof PM followed by SO2minus

4 NH+4 and NOminus3 Metals such asV Ni Cr Fe As and Cd made up a proportionately smallpart of the total PM mass However other rare elements suchas Tb Er Yb and Lu had higher values than did some ofthe common elements PM was an in-process product duringthe combustion in a cylinder the forming process includedthe molecular cracking decomposition and polymerizationwhich resulted in lack of oxygen High temperature and oxy-gen deficiency were the main reasons for the formation indiesel engines which always had high concentration val-ues in high load operating modes (Ni 1999) HH had muchhigher PM emission factors than the other two vessels theengine type was considered to be the most significant influ-ence factor which had a good agreement with NOx emissionfactors

EFs for OC and EC and the ratios between them are shownin Fig 3 EFs for OC and EC for HH were higher than forthe other two vessels Organic matter (OM) is generally cal-culated as OCtimes 12 (Petzold et al 2008) to account for themass of elements other than carbon in the emitted moleculesOM EFs for individual vessels mainly depend on the enginetype and the amount of unburned fuel ie the efficiency ofcombustion (Moldanovaacute et al 2013) BC emissions also de-pend heavily on the engine type (Lack et al 2009) There-fore the different types of engines and their levels of mainte-nance could account for the large differences in OC and ECEFs observed among the three vessels in this study The ratiosof OC-to-EC in the present study were much lower than thosefor large diesel ships reported previously (OC EC= 12)(Moldanova et al 2009) and also lower than that reportedfor a medium-speed vessel (Petzold et al 2010) The usageof non-dilution sampling in this study was one possible rea-son for the lower OC to EC ratio Besides TOR was used tomeasure OC and EC in PM which always had a lower OCcontent compared with other methods (such as TOT) becauseof the different definitions of OC and EC (Khan et al 2012)Compared with other diesel engines the ratios of OC to ECin this study were higher than that of automobile diesel sootin which EC comprises 75ndash80 wt of the total PM (Clagueet al 1999) and also higher than heavy heavy-duty dieseltrucks (HHDDTs) with OC to EC ratios below unit for cruseand transient modes even though higher in cold-startidle andcreep modes (Shah et al 2004)

Studies have shown that SO2minus4 formed from vessel-emitted

SO2 is a major contributor to CCN and ship track forma-tion (Schreier et al 2006 Lauer et al 2007) Sulfate is alsoan important component of PM emitted from vessels In thepresent study EFs for SO2minus

4 were much lower than previ-ously reported (Petzold et al 2008 Agrawal et al 2008a)but similar to those detected by a high-resolution time-of-flight aerosol mass spectrometer in a previous study (Lack

Figure 3 EFs for OC and EC and the ratios between them

et al 2009) This may be because EFs for SO2minus4 are mainly

related to the sulfur content of the fuel SO2minus4 is not gen-

erally emitted directly from the engines but forms after re-lease from the stack (Lack et al 2009) Because PM was col-lected directly from engine emissions in the present study thesulfur-to-sulfate ratios were low (lt 06 for vessels) Otherions such as NOminus3 and NH+4 accounted for a small percent-age of the PM emitted from the vessels compared with SO2minus

4 consistent with previous studies (Lack et al 2009) SO2 ismore easily oxidized to SO3 in catalytic reaction cycles withmetals commonly present in the exhaust gas (V Ni) whilehydroxyl radicals are additional needed to convert NOx toNOminus3 (Moldanova et al 2009)

Na+ and Clminus were considered to originate from marineair Their concentrations were highly correlated (r2

= 078)the differing air demands of the engines under different con-ditions might have caused observed variations in the EFs rel-ative to the fuel demand

The elemental composition of PM in the present studydiffered from previous studies showing high elemental con-tent of S Ca V Fe Cu Ni and Al (Agrawal et al 2008aMoldanova et al 2009) V and Ni are typically associatedwith combustion of heavy fuel oil (Almeida et al 2005) Inthe present study the high-quality fuels resulted in low EFsfor V and Ni In our previous study PM from shipping emis-sions was estimated to account for 294 of the total PM25at Tuoji Island in China using V as a tracer of shipping emis-sions (Zhang et al 2014) Reconsidering the former resultsbased on the EFs obtained in the present study we deter-mined that the contribution of vessels near Tuoji Island hadbeen underestimated because the estimate should have in-cluded both heavy and other types of fuels However somerare elements such as Tb Er Yb and Lu had relatively highEFs compared with those of other elements in the presentstudy which may be related to the source of the fuels

Atmos Chem Phys 16 6319ndash6334 2016 wwwatmos-chem-physnet1663192016

F Zhang et al Emission factors for gaseous and particulate pollutants 6327

Tabl

e3

Fuel

-bas

edav

erag

eE

Fsin

the

pres

ents

tudy

and

prev

ious

stud

ies

(gkgminus

1fu

el)

Ves

selI

DC

O2

CO

NO

NO

2N

2ON

Ox

TV

OC

sPM

SO2

Sco

nten

t(

m)

HH

3071plusmn

1565

302plusmn

162

982plusmn

372

155plusmn

545

128plusmn

170

115plusmn

443

237plusmn

210

940plusmn

213

160

008

DFH

3153plusmn

176

693plusmn

100

302plusmn

160

509plusmn

042

038plusmn

018

357plusmn

220

124plusmn

004

072plusmn

033

092

005

XY

H31

51plusmn

175

920plusmn

295

266plusmn

163

471plusmn

042

030plusmn

015

316plusmn

220

418plusmn

015

016plusmn

007

260

013

Com

mer

cial

vess

el(W

illia

ms

etal

20

09)

3170

7ndash16

ndashndash

ndash60

ndash87

ndashndash

6ndash30

Car

gove

ssel

(Mol

dano

vaet

al

2009

)

3441

217

ndashndash

ndash73

4ndash

53

393

19

Die

sele

ngin

e(H

aglin

d20

08)

ndash7

4ndash

ndashndash

87ndash

76

542

7

Oce

an-g

oing

ship

s(S

inha

etal

20

03)

3135

195

ndashndash

ndash22

3ndash

ndash2

90

1

Oce

an-g

oing

ship

s(S

inha

etal

20

03)

3176

30

ndashndash

ndash65

5ndash

ndash52

22

4

Car

goan

dpa

ssen

ger

ship

s(E

ndre

sen

2003

)31

707

4ndash

ndash0

0857

ndash87

24

12ndash

76

10ndash5

40

5ndash2

7

Ship

sop

erat

ing

inndash

ndash42

ndash72

ndashndash

65ndash8

6ndash

ndash4

6ndash9

8N

ON

Eha

rbou

rare

as(P

irjo

laet

al

2014

)ndash

ndash16

ndash49

ndashndash

25ndash7

9ndash

ndash5

4ndash17

0SC

R

Ship

sop

erat

ing

inPo

rt(D

iesc

het

al

2013

)ndash

ndash16

37ndash

53ndash

ndash7

7

NO

NE=

No

trea

tmen

tofe

mis

sion

sSC

R=

sele

ctiv

eca

taly

ticre

duct

ion

wwwatmos-chem-physnet1663192016 Atmos Chem Phys 16 6319ndash6334 2016

6328 F Zhang et al Emission factors for gaseous and particulate pollutants

33 Fuel-based average emission factors

Based on actual operating conditions (Table S7) average EFsfor the three vessels in the present study (according to Eq 4)along with EFs from previous studies are shown in Table 3EFs for all of the pollutants except SO2 were significantlyhigher for HH than for the other two vessels potentially dueto poor combustion conditions Most of the EFs for DFH andXYH were within the range of emissions for other vesselsdue to having well maintained engines and the high qualityof the fuels used The EFs for NOx PM and SO2 were muchlower than reported in previous studies (other than NOx forocean-going vessels) All the sulfur of the fuels in the presentstudy were significantly below the emissions limit of 350 established by IMO in the revised MARPOL Annex VI rulesapplicable since 2012 (IMO 1998)

The IMO Tier I emissions limit for NOx is450times nminus02 g kWhminus1 (n rated speed 130 lt nlt 2000 rpm)The rated speed and fuel consumption rates for each vesselare shown in Table 1 Thus the emissions limits for HHDFH and XYH would be 545 575 and 565 g kgminus1 fuelrespectively calculated combined with Eq (5) The averagefuel-based EFs for NOx of ship HH was more than 100 above the IMO standard while those of the other two shipswere below the IMO standard (Table 3) PM emissions forHH were also higher than previously reported but thosefor the two research vessels were much lower (Table 3)Fuel type is one of the most important influence factors onpollutant emissions for example sulfur content in the fuelnot only influence the SO2 emission directly but also hadimpact on PM formation in the flue gas stack with low sulfurcontent in fuels reduces PM formation (Lack et al 2011)Vessels with higher sulfur content always had relativelyhigher PM emissions which were also shown in Table 3In addition different engines and levels of maintenancehave a significant impact on all combustion-dependentemissions Emission reduction measures have been used insome vessels For example NOx emissions can be reducedby measures such as selective catalytic reduction (SCR) anddirect water injection (DWI) which had been implementedon some vessels previously studied in a harbour in Finland(Pirjola et al 2014) The results showed that SCR effec-tively reduced NOx emissions while vessels with DWI hadhigh PM emissions The engine type might be an importantcause of the different emissions such as HH had muchhigher pollutants emissions with an engine produced inChina and yet DFHrsquos engine produced in Germany Besidesemission tests for a high-speed marine diesel engine withdifferent kinds of diesels showed that diesel type had limitedinfluence on emissions such as NOx CO and CH but asignificant impact on PM emission (289ndash415 ) becauseof the different sulfur content in fuel (Xu 2008)

34 Power-based emission factors

Based on the engine power and fuel consumption rates ofthe vessels power-based EFs were calculated (according toEq 5) and compared to results from previous studies (Ta-ble 4) The EFs for HH were much higher than those forthe other two vessels except for SO2 HH also had signif-icantly higher EFs for NOx than previously reported val-ues while EFs for NOx of DFH and XYH were within therange of previously reported results Engine type was consid-ered to be a significant influence factor for NOx emissionswith lower engine speed having higher NOx emission factors(Celo et al 2015) In addition compared to other vesselswith a similar engine type and diesel fuel HH still had rela-tively higher NOx EF (seen in Table 4) which could reflectthe impact of engine condition (engine quality and mainte-nance level) on shipping emissions CO EFs for the test ves-sels in the present study were higher than previous studieswhich produced similar results to those of inland ships andother test vessels in China (Fu et al 2013 Song 2015) Inspite of the influence of engine type on CO emissions thatwith the higher engine speed having higher CO EFs (Celoet al 2015) engine condition combined with fuel qualitymight have significant influence All of the EFs for SO2 inthe present study were lower than those in previous studiesbecause of the low sulfur content of the present fuels Gen-erally PM emissions from marine diesel fuels are dependenton the fuel (sulfate and metal oxide ash constituents) and oncombustion conditions (unburned hydrocarbons and carbonresidue constituents) (Cooper 2003) HH had the highest PMemissions among the test vessels although there were almostno differences among the fuels (Table S6) Besides HH hadeven higher PM EFs than previously reported vessels withHFO fuel and XYH had much lower PM EFs than all theother vessels with even lower sulfur content fuel Thereforecombustion conditions were likely the determining factor forthe differences It can be seen from Table 4 that most previ-ous studies focused on the heavy fuel oil of shipping emis-sions Compared with diesel fuels heavy fuel oil always hadrelatively low CO emission factors and high PM emissionfactors And among the heavy-fuel-oil-using vessels enginetype (engine speed and engine power level) always playedan important role on emissions such as NOx and CO whichwith lower engine speed having higher NOx EFs and lowerCO EFs

Combined with other emission data of test ships in China(Fu et al 2013 Song 2015) it could be seen that inlandand some test offshore ships in China always had higherNOx CO and PM emissions compared with other test ves-sels in previous studies And among the test vessels in Chinathere also were differences for different engine types andship types In addition emission factors that were used forcalculation of ship inventories in China always came fromother countries and areas However there seemed to be sig-nificant differences between the reference and test data such

Atmos Chem Phys 16 6319ndash6334 2016 wwwatmos-chem-physnet1663192016

F Zhang et al Emission factors for gaseous and particulate pollutants 6329

Tabl

e4

Pow

er-b

ased

EFs

inth

epr

esen

tstu

dyan

dpr

evio

usst

udie

s(g

kWhminus

1 )

Ves

sel

CO

2C

ON

ON

O2

N2O

NO

xT

VO

Cs

PMSO

2E

ngin

epo

wer

Eng

ine

Fuel

type

and

ID(k

W)

spee

dsu

lfur

cont

ent

(rm

p)(w

t)

HH

699plusmn

352

738plusmn

376

220plusmn

841

345plusmn

124

030plusmn

039

258plusmn

100

544plusmn

484

209plusmn

048

036

350

1200

DO

00

798

DFH

631plusmn

352

139plusmn

020

604plusmn

032

102plusmn

008

008plusmn

004

714plusmn

044

017plusmn

001

014plusmn

007

018

1600

900

DO

00

458

XY

H69

7plusmn

385

201plusmn

065

587plusmn

036

104plusmn

009

007plusmn

003

697plusmn

048

092plusmn

ndash0

04plusmn

001

057

600

1000

DO

01

30Ta

nker

(Win

nes

and

Frid

ell

2010

)ndash

161

ndashndash

ndash7

82ndash

058

ndash45

0060

0H

FO1

6

Ber

thed

ship

s(C

oope

r65

3ndash76

80

33ndash0

97

ndashndash

ndash14

2ndash2

02

ndash0

14ndash0

45

026

ndash53

AE

720

ndash127

072

0ndash18

00M

GO

00

6ndash1

220

03)

691ndash

803

077

ndash17

112

9ndash1

75

048

ndash06

72

5ndash9

612

70ndash2

675

720ndash

750

RO

05

3ndash2

269

1ndash69

40

92ndash0

98

96ndash

99

017

ndash01

91

014

8072

0M

DO

02

3C

rude

oilt

anke

r(A

graw

alet

al

2008

b)58

8ndash66

00

77ndash1

78

ndashndash

ndash15

8ndash2

10

ndash1

10ndash1

78

766

ndash86

015

750

90H

FO2

85

Cru

ise

ship

s(P

opla

wsk

iet

al

2011

)ndash

ndashndash

140

ndashndash

ndash2

914

20

US

EPA

621

14

ndashndash

ndash18

1ndash

131

103

Mar

ine

engi

ne(S

ippu

laet

ndash1

2ndash11

411

3ndash2

95

ndashndash

114

ndash30

90ndash

95

083

ndash63

6ndash

1500

HFO

27

al

2014

)0ndash

885

69ndash2

58

584

ndash33

90

83ndash1

97

015

ndash09

3D

FL

arge

mar

ine

ship

s(K

han

etal

20

13)

533ndash

612

035

ndash06

0ndash

ndashndash

166

ndash20

6ndash

091

ndash21

97

2ndash11

436

740ndash

6853

097

ndash102

HFO

21

5ndash3

14

Oce

ango

ing

cont

aine

rve

ssel

(Agr

awal

etal

20

08a)

588ndash

660

077

ndash18

1ndash

ndashndash

158

ndash21

0ndash

109

ndash17

67

66ndash8

60

5027

010

4H

FO2

05

Lar

geca

rgo

vess

el(M

olda

nova

etal

20

09)

667

042

ndashndash

ndash14

22

ndash1

0310

320

200

97H

FO1

9

Oce

ango

ing

carg

ove

ssel

614plusmn

10

83plusmn

001

ndashndash

ndash16

3plusmn

02

ndash1

51plusmn

007

87plusmn

01

128

IFO

180

(Cel

oet

al

2015

)62

6plusmn

70

26plusmn

001

114plusmn

01

081plusmn

002

58plusmn

007

525

IFO

180

628plusmn

90

30plusmn

001

113plusmn

01

094plusmn

002

87plusmn

01

450

IFO

180

628plusmn

10

81plusmn

003

122plusmn

001

083plusmn

001

64plusmn

01

450

IFO

180

609plusmn

11

31plusmn

002

ndashndash

ndash8

4plusmn

003

ndash0

37plusmn

001

47plusmn

001

440

IFO

6060

5plusmn

10

00ndash

ndashndash

167plusmn

01

ndash2

2plusmn

02

103plusmn

003

116

IFO

380

622plusmn

11

22plusmn

002

ndashndash

ndash10

7plusmn

004

ndash0

30plusmn

003

047plusmn

01

450

MD

O

AE

aux

iliar

yen

gine

DO

die

selo

ilH

FOh

eavy

fuel

oil

MG

Om

arin

ega

soi

lR

Or

esid

ualo

ilM

DO

mar

ine

dies

eloi

lD

Fdi

esel

fuel

IFO

int

erm

edia

tefu

eloi

l

wwwatmos-chem-physnet1663192016 Atmos Chem Phys 16 6319ndash6334 2016

6330 F Zhang et al Emission factors for gaseous and particulate pollutants

as 100 to 132 g kW hminus1 of NOx EF and 11 to 17 g kW hminus1

of CO EF used for inland ships for ship inventory calcula-tion (Zhu et al 2015) 100 to 181 g kW hminus1 of NOx EFand 11 to 15 g kW hminus1 of CO EF for harbour ships (Yanget al 2015) compared to 15 to 173 g kW hminus1 of NOx EFand 46 to 103 g kW hminus1 of CO EF from test inland ships(convert the fuel-based EF to power-based EF with a factorof 200 g kW hminus1) (Song 2015) and 697 to 258 g kW hminus1 ofNOx EF and 139 to 738 g kW hminus1 of CO EF in the presentstudy (Yang et al 2015) Besides whether there are obviousdifferences of EFs between other types of vessels in China(such as low-speed engine vessels with heavy fuel oil) andprevious studies is still unclear Therefore much more mea-surement data for different vessels in China are still in urgentneed for more accurate assessment of shipping emissions

35 Impact of engine speed on NOx emission factors

NOx is formed in the combustion chamber by a combinationof atmospheric nitrogen and oxygen under high-pressure andhigh-temperature conditions Many factors affect NOx for-mation including engine temperature injection point andfuel quality The IMO emissions limit for NOx is determinedby the rated speed of the engine however other factors mustalso be considered to reduce NOx emissions

The NOx EFs for the test vessels at various engine speedsare shown in Fig 4 The rated speeds of the vessels were1200 900 and 1000 rpm for HH DFH and XYH respec-tively The actual engine speeds of HH were much lower thanthe rated speed while the two larger-engine-power vesselsoperated close to their rated speeds except during one op-erating mode of DFH The NOx EFs for HH differed sig-nificantly in different operating modes ranging from 391to 143 g kgminus1 of fuel The NOx EF was highest when theengine speed reached sim 750 rpm (Fig 4) At lower enginespeeds the NOx EFs had fluctuating but lower values Athigher engine speeds closer to the rated speed of 1200 rpmthe NOx EFs were much lower The NOx EFs for the twolarger-engine-power vessels changed slightly with enginespeed but also had lowest values when their engine speedsapproached their rated speeds Combined with the dieselpropulsion characteristic curve there were large increasesin the fuel consumption rate when the engine speed in-creased Therefore a best-fit engine speed should be deter-mined based on both EFs and economic costs

Engineering approaches for reducing the NOx emissionsof marine engines may be applied before during or afterthe combustion process (Verschaeren et al 2014 Habib etal 2014) In the present study the NOx EFs of the two re-search vessels were below the IMO Tier I emissions limitsHowever for EMS measures should be taken to meet theIMO emissions limit including increasing the engine speedand applying engineering technologies during or after com-bustion such as exhaust gas recirculation (EGR) selectivenon-catalytic reduction (SNCR) or SCR

Figure 4 Emission factors for NOx at different engine speeds

4 Conclusions

Three offshore vessels with different engine power sourceswere chosen in this study to collect measured data of gaseousspecies and particulate matter including NO2 NO N2OCO CO2 TVOCs SO2 and the total suspended particulateBesides chemical composition of the PM were also analysedto give detailed EFs for OC EC water-soluble ions and metalelements Concentrations fuel-based EFs fuel-based aver-age EFs as well as power-based average EFs for species ofoffshore vessels in China were presented Furthermore im-pact of engine speed on NOx EFs was also discussed

There were higher concentrations of pollutants for low-engine-power vessel HH than for the other two vessels COconcentrations for offshore vessels were slightly lower thaninland vessels in China and all the three vessels had the low-est CO concentrations at their economic speeds (the speed ofthe least vessel operating expenditures during one voyagethey were high-speed mode cruise mode and high-speedmode for HH DFH and XYH respectively) More than 80 of the NOx was NO and all the offshore vessels had higherNOx concentrations in high-speed modes Because of thelow-sulfur fuels used in this study SO2 concentrations ofthese three offshore vessels were lower than that in the lit-eratures And the PM concentrations were much lower thaninland vessels while showing significant differences amongdifferent operating modes

Fuel-based EFs for gaseous species and PM were pre-sented based on the carbon balance method EFs for CO2were 2940ndash3106 3121ndash3160 and 3102ndash3162 g kgminus1 fuel forHH DFH and XYH Because of the combustion conditionssuch as excess air ratio combustion temperature and unifor-mity of the fuelndashair mixture EFs for CO showed high valuesin idling mode but low values in economic speed All theoffshore vessels had higher NOx EFs in low speed than inhigh speed but showed higher values when in acceleration

Atmos Chem Phys 16 6319ndash6334 2016 wwwatmos-chem-physnet1663192016

F Zhang et al Emission factors for gaseous and particulate pollutants 6331

process EFs for SO2 were 16 09 and 26 g kgminus1 fuel forHH DFH and XYH based on sulfur content of the fuels OCand EC were the main components of PM with low OC toEC ratios that were lower than 01 followed by SO2minus

4 NH+4 and NOminus3 Metals such as V Ni Cr Fe As and Cd made upa proportionately small part of the total PM mass

Fuel-based average EFs as well as power-based EFs for thethree different vessels of differing engine power were pre-sented EFs for most gaseous species and PM of HH weremuch higher compared with the other higher-engine-powervessels which was also gt 100 above the IMO standard forNOx Average PM EF of the low-engine-power vessel HHwas also much higher than that in the literatures Howeveraverage EFs for most species of the two larger-engine-powervessels were within the range of previously reported resultsEngine type was inferred as one of the most influence fac-tors for the differences of emission factors Inland and someoffshore ships in China always had higher NOx CO and PMemissions compared with other test vessels in previous stud-ies In addition emission factors that used for calculation ofship inventories in China always had lower values than testvessels

The impact of engine speed on EFs for NOx showed thatwhen the engine speed was close to the rated speed therewould be lower NOx EFs values However combined withthe high fuel consumption rate an optimal engine speedshould be determined based on both EFs and economic costsEmission reduction measures for NOx for some of the off-shore vessels in China are still essential to meet the IMOemission limit

Given the limits of vessel types and numbers this studysubstantially gives the EFs for gaseous species and PM ofthree different diesel-engine-powered offshore vessels How-ever as the development of ports in China emissions fromcargo ships and container ships with large engine power havebecome one of the most significant air pollution sources inport cities and regions Systematical measurement EFs of allkinds of offshore vessels in China are essential in order topresent the accurate emission inventory of ships

Information about the Supplement

Supplementary information includes the details of the real-world measurement system for vessels (Fig S1) the concen-trations of main gaseous matter and PM of shipping emis-sions (Fig S2) the types composition of offshore vessels inChina (Table S1) the distribution of vessels through grosstonnage in 2014 in offshore area of Yangtze River Delta (Ta-ble S2) the Chinese market consumption of marine oil in2011 (Table S3) the detection parameters for gaseous mat-ter (Table S4) the fuel-based EFs for the gaseous pollutants(Table S5) PM and the chemical composition in PM for dif-ferent operating (Table S6) modes and the actual operatingconditions of vessels (Table S7)

The Supplement related to this article is available onlineat doi105194acp-16-6319-2016-supplement

Acknowledgements This study was supported by the CAS Strate-gic Priority Research Program (No XDB05030303) the NaturalScientific Foundation of China (Nos 41273135 and 41473091)and the Shanghai science and Technology Commission project(No 15DZ1205401) The authors would like to express theirgratitude to the vessel owners Haohai Ocean Engineering LimitedLiability Company for vessel Haohai 0007 Ocean University ofChina for vessel Dongfanghong 02 and North China Sea Branchof the State Oceanic Administration for vessel Xiangyanghong08 for making their vessels available for the shipping emissionsstudy The views expressed in this document are solely those of theauthors and the funding agencies do not endorse any products orcommercial services mentioned in this publication

Edited by T Bond

References

Agrawal H Malloy Q G J Welch W A Miller J W andCocker D R In-use gaseous and particulate matter emissionsfrom a modern ocean going container vessel Atmos Environ42 5504ndash5510 doi101016jatmosenv200802053 2008a

Agrawal H Welch W A Miller J W and Cocker D R Emis-sion measurements from a crude oil tanker at sea Environ SciTechnol 42 7098ndash7103 doi101021es703102y 2008b

Almeida S M Pio C A Freitas M C Reis MA and Trancoso M A Source apportionment of fineand coarse particulate matter in a sub-urban area at theWestern European Coast Atmos Environ 39 3127ndash3138doi101016jatmosenv200501048 2005

Anderson M Salo K Hallquist A M and Fridell ECharacterization of particles from a marine engine op-erating at low loads Atmos Environ 101 65ndash71doi101016jatmosenv201411009 2015

Berg N Mellqvist J Jalkanen J-P and Balzani J Ship emis-sions of SO2 and NO2 DOAS measurements from airborne plat-forms Atmos Meas Tech 5 1085ndash1098 doi105194amt-5-1085-2012 2012

California Code of Regulation Titles 13 and 17 avail-able at httpsgovtwestlawcomcalregsBrowseHomeCaliforniaCaliforniaCodeofRegulationstransitionType=DefaultampcontextData=(scDefault) last access 20 April 2016

Cappa C D Williams E J Lack D A Buffaloe G M Coff-man D Hayden K L Herndon S C Lerner B M Li S-M Massoli P McLaren R Nuaaman I Onasch T B andQuinn P K A case study into the measurement of ship emis-sions from plume intercepts of the NOAA ship Miller FreemanAtmos Chem Phys 14 1337ndash1352 doi105194acp-14-1337-2014 2014

Carlton J Danton S Gawen R Lavender K Mathieson NNewell A Reynolds G Webster A Wills C and Wright AMarine exhaust emissions research programme Lloydrsquos RegisterEngineering Services London 63 1995

wwwatmos-chem-physnet1663192016 Atmos Chem Phys 16 6319ndash6334 2016

F Zhang et al Emission factors for gaseous and particulate pollutants 6327

Tabl

e3

Fuel

-bas

edav

erag

eE

Fsin

the

pres

ents

tudy

and

prev

ious

stud

ies

(gkgminus

1fu

el)

Ves

selI

DC

O2

CO

NO

NO

2N

2ON

Ox

TV

OC

sPM

SO2

Sco

nten

t(

m)

HH

3071plusmn

1565

302plusmn

162

982plusmn

372

155plusmn

545

128plusmn

170

115plusmn

443

237plusmn

210

940plusmn

213

160

008

DFH

3153plusmn

176

693plusmn

100

302plusmn

160

509plusmn

042

038plusmn

018

357plusmn

220

124plusmn

004

072plusmn

033

092

005

XY

H31

51plusmn

175

920plusmn

295

266plusmn

163

471plusmn

042

030plusmn

015

316plusmn

220

418plusmn

015

016plusmn

007

260

013

Com

mer

cial

vess

el(W

illia

ms

etal

20

09)

3170

7ndash16

ndashndash

ndash60

ndash87

ndashndash

6ndash30

Car

gove

ssel

(Mol

dano

vaet

al

2009

)

3441

217

ndashndash

ndash73

4ndash

53

393

19

Die

sele

ngin

e(H

aglin

d20

08)

ndash7

4ndash

ndashndash

87ndash

76

542

7

Oce

an-g

oing

ship

s(S

inha

etal

20

03)

3135

195

ndashndash

ndash22

3ndash

ndash2

90

1

Oce

an-g

oing

ship

s(S

inha

etal

20

03)

3176

30

ndashndash

ndash65

5ndash

ndash52

22

4

Car

goan

dpa

ssen

ger

ship

s(E

ndre

sen

2003

)31

707

4ndash

ndash0

0857

ndash87

24

12ndash

76

10ndash5

40

5ndash2

7

Ship

sop

erat

ing

inndash

ndash42

ndash72

ndashndash

65ndash8

6ndash

ndash4

6ndash9

8N

ON

Eha

rbou

rare

as(P

irjo

laet

al

2014

)ndash

ndash16

ndash49

ndashndash

25ndash7

9ndash

ndash5

4ndash17

0SC

R

Ship

sop

erat

ing

inPo

rt(D

iesc

het

al

2013

)ndash

ndash16

37ndash

53ndash

ndash7

7

NO

NE=

No

trea

tmen

tofe

mis

sion

sSC

R=

sele

ctiv

eca

taly

ticre

duct

ion

wwwatmos-chem-physnet1663192016 Atmos Chem Phys 16 6319ndash6334 2016

6328 F Zhang et al Emission factors for gaseous and particulate pollutants

33 Fuel-based average emission factors

Based on actual operating conditions (Table S7) average EFsfor the three vessels in the present study (according to Eq 4)along with EFs from previous studies are shown in Table 3EFs for all of the pollutants except SO2 were significantlyhigher for HH than for the other two vessels potentially dueto poor combustion conditions Most of the EFs for DFH andXYH were within the range of emissions for other vesselsdue to having well maintained engines and the high qualityof the fuels used The EFs for NOx PM and SO2 were muchlower than reported in previous studies (other than NOx forocean-going vessels) All the sulfur of the fuels in the presentstudy were significantly below the emissions limit of 350 established by IMO in the revised MARPOL Annex VI rulesapplicable since 2012 (IMO 1998)

The IMO Tier I emissions limit for NOx is450times nminus02 g kWhminus1 (n rated speed 130 lt nlt 2000 rpm)The rated speed and fuel consumption rates for each vesselare shown in Table 1 Thus the emissions limits for HHDFH and XYH would be 545 575 and 565 g kgminus1 fuelrespectively calculated combined with Eq (5) The averagefuel-based EFs for NOx of ship HH was more than 100 above the IMO standard while those of the other two shipswere below the IMO standard (Table 3) PM emissions forHH were also higher than previously reported but thosefor the two research vessels were much lower (Table 3)Fuel type is one of the most important influence factors onpollutant emissions for example sulfur content in the fuelnot only influence the SO2 emission directly but also hadimpact on PM formation in the flue gas stack with low sulfurcontent in fuels reduces PM formation (Lack et al 2011)Vessels with higher sulfur content always had relativelyhigher PM emissions which were also shown in Table 3In addition different engines and levels of maintenancehave a significant impact on all combustion-dependentemissions Emission reduction measures have been used insome vessels For example NOx emissions can be reducedby measures such as selective catalytic reduction (SCR) anddirect water injection (DWI) which had been implementedon some vessels previously studied in a harbour in Finland(Pirjola et al 2014) The results showed that SCR effec-tively reduced NOx emissions while vessels with DWI hadhigh PM emissions The engine type might be an importantcause of the different emissions such as HH had muchhigher pollutants emissions with an engine produced inChina and yet DFHrsquos engine produced in Germany Besidesemission tests for a high-speed marine diesel engine withdifferent kinds of diesels showed that diesel type had limitedinfluence on emissions such as NOx CO and CH but asignificant impact on PM emission (289ndash415 ) becauseof the different sulfur content in fuel (Xu 2008)

34 Power-based emission factors

Based on the engine power and fuel consumption rates ofthe vessels power-based EFs were calculated (according toEq 5) and compared to results from previous studies (Ta-ble 4) The EFs for HH were much higher than those forthe other two vessels except for SO2 HH also had signif-icantly higher EFs for NOx than previously reported val-ues while EFs for NOx of DFH and XYH were within therange of previously reported results Engine type was consid-ered to be a significant influence factor for NOx emissionswith lower engine speed having higher NOx emission factors(Celo et al 2015) In addition compared to other vesselswith a similar engine type and diesel fuel HH still had rela-tively higher NOx EF (seen in Table 4) which could reflectthe impact of engine condition (engine quality and mainte-nance level) on shipping emissions CO EFs for the test ves-sels in the present study were higher than previous studieswhich produced similar results to those of inland ships andother test vessels in China (Fu et al 2013 Song 2015) Inspite of the influence of engine type on CO emissions thatwith the higher engine speed having higher CO EFs (Celoet al 2015) engine condition combined with fuel qualitymight have significant influence All of the EFs for SO2 inthe present study were lower than those in previous studiesbecause of the low sulfur content of the present fuels Gen-erally PM emissions from marine diesel fuels are dependenton the fuel (sulfate and metal oxide ash constituents) and oncombustion conditions (unburned hydrocarbons and carbonresidue constituents) (Cooper 2003) HH had the highest PMemissions among the test vessels although there were almostno differences among the fuels (Table S6) Besides HH hadeven higher PM EFs than previously reported vessels withHFO fuel and XYH had much lower PM EFs than all theother vessels with even lower sulfur content fuel Thereforecombustion conditions were likely the determining factor forthe differences It can be seen from Table 4 that most previ-ous studies focused on the heavy fuel oil of shipping emis-sions Compared with diesel fuels heavy fuel oil always hadrelatively low CO emission factors and high PM emissionfactors And among the heavy-fuel-oil-using vessels enginetype (engine speed and engine power level) always playedan important role on emissions such as NOx and CO whichwith lower engine speed having higher NOx EFs and lowerCO EFs

Combined with other emission data of test ships in China(Fu et al 2013 Song 2015) it could be seen that inlandand some test offshore ships in China always had higherNOx CO and PM emissions compared with other test ves-sels in previous studies And among the test vessels in Chinathere also were differences for different engine types andship types In addition emission factors that were used forcalculation of ship inventories in China always came fromother countries and areas However there seemed to be sig-nificant differences between the reference and test data such

Atmos Chem Phys 16 6319ndash6334 2016 wwwatmos-chem-physnet1663192016

F Zhang et al Emission factors for gaseous and particulate pollutants 6329

Tabl

e4

Pow

er-b

ased

EFs

inth

epr

esen

tstu

dyan

dpr

evio

usst

udie

s(g

kWhminus

1 )

Ves

sel

CO

2C

ON

ON

O2

N2O

NO

xT

VO

Cs

PMSO

2E

ngin

epo

wer

Eng

ine

Fuel

type

and

ID(k

W)

spee

dsu

lfur

cont

ent

(rm

p)(w

t)

HH

699plusmn

352

738plusmn

376

220plusmn

841

345plusmn

124

030plusmn

039

258plusmn

100

544plusmn

484

209plusmn

048

036

350

1200

DO

00

798

DFH

631plusmn

352

139plusmn

020

604plusmn

032

102plusmn

008

008plusmn

004

714plusmn

044

017plusmn

001

014plusmn

007

018

1600

900

DO

00

458

XY

H69

7plusmn

385

201plusmn

065

587plusmn

036

104plusmn

009

007plusmn

003

697plusmn

048

092plusmn

ndash0

04plusmn

001

057

600

1000

DO

01

30Ta

nker

(Win

nes

and

Frid

ell

2010

)ndash

161

ndashndash

ndash7

82ndash

058

ndash45

0060

0H

FO1

6

Ber

thed

ship

s(C

oope

r65

3ndash76

80

33ndash0

97

ndashndash

ndash14

2ndash2

02

ndash0

14ndash0

45

026

ndash53

AE

720

ndash127

072

0ndash18

00M

GO

00

6ndash1

220

03)

691ndash

803

077

ndash17

112

9ndash1

75

048

ndash06

72

5ndash9

612

70ndash2

675

720ndash

750

RO

05

3ndash2

269

1ndash69

40

92ndash0

98

96ndash

99

017

ndash01

91

014

8072

0M

DO

02

3C

rude

oilt

anke

r(A

graw

alet

al

2008

b)58

8ndash66

00

77ndash1

78

ndashndash

ndash15

8ndash2

10

ndash1

10ndash1

78

766

ndash86

015

750

90H

FO2

85

Cru

ise

ship

s(P

opla

wsk

iet

al

2011

)ndash

ndashndash

140

ndashndash

ndash2

914

20

US

EPA

621

14

ndashndash

ndash18

1ndash

131

103

Mar

ine

engi

ne(S

ippu

laet

ndash1

2ndash11

411

3ndash2

95

ndashndash

114

ndash30

90ndash

95

083

ndash63

6ndash

1500

HFO

27

al

2014

)0ndash

885

69ndash2

58

584

ndash33

90

83ndash1

97

015

ndash09

3D

FL

arge

mar

ine

ship

s(K

han

etal

20

13)

533ndash

612

035

ndash06

0ndash

ndashndash

166

ndash20

6ndash

091

ndash21

97

2ndash11

436

740ndash

6853

097

ndash102

HFO

21

5ndash3

14

Oce

ango

ing

cont

aine

rve

ssel

(Agr

awal

etal

20

08a)

588ndash

660

077

ndash18

1ndash

ndashndash

158

ndash21

0ndash

109

ndash17

67

66ndash8

60

5027

010

4H

FO2

05

Lar

geca

rgo

vess

el(M

olda

nova

etal

20

09)

667

042

ndashndash

ndash14

22

ndash1

0310

320

200

97H

FO1

9

Oce

ango

ing

carg

ove

ssel

614plusmn

10

83plusmn

001

ndashndash

ndash16

3plusmn

02

ndash1

51plusmn

007

87plusmn

01

128

IFO

180

(Cel

oet

al

2015

)62

6plusmn

70

26plusmn

001

114plusmn

01

081plusmn

002

58plusmn

007

525

IFO

180

628plusmn

90

30plusmn

001

113plusmn

01

094plusmn

002

87plusmn

01

450

IFO

180

628plusmn

10

81plusmn

003

122plusmn

001

083plusmn

001

64plusmn

01

450

IFO

180

609plusmn

11

31plusmn

002

ndashndash

ndash8

4plusmn

003

ndash0

37plusmn

001

47plusmn

001

440

IFO

6060

5plusmn

10

00ndash

ndashndash

167plusmn

01

ndash2

2plusmn

02

103plusmn

003

116

IFO

380

622plusmn

11

22plusmn

002

ndashndash

ndash10

7plusmn

004

ndash0

30plusmn

003

047plusmn

01

450

MD

O

AE

aux

iliar

yen

gine

DO

die

selo

ilH

FOh

eavy

fuel

oil

MG

Om

arin

ega

soi

lR

Or

esid

ualo

ilM

DO

mar

ine

dies

eloi

lD

Fdi

esel

fuel

IFO

int

erm

edia

tefu

eloi

l

wwwatmos-chem-physnet1663192016 Atmos Chem Phys 16 6319ndash6334 2016

6330 F Zhang et al Emission factors for gaseous and particulate pollutants

as 100 to 132 g kW hminus1 of NOx EF and 11 to 17 g kW hminus1

of CO EF used for inland ships for ship inventory calcula-tion (Zhu et al 2015) 100 to 181 g kW hminus1 of NOx EFand 11 to 15 g kW hminus1 of CO EF for harbour ships (Yanget al 2015) compared to 15 to 173 g kW hminus1 of NOx EFand 46 to 103 g kW hminus1 of CO EF from test inland ships(convert the fuel-based EF to power-based EF with a factorof 200 g kW hminus1) (Song 2015) and 697 to 258 g kW hminus1 ofNOx EF and 139 to 738 g kW hminus1 of CO EF in the presentstudy (Yang et al 2015) Besides whether there are obviousdifferences of EFs between other types of vessels in China(such as low-speed engine vessels with heavy fuel oil) andprevious studies is still unclear Therefore much more mea-surement data for different vessels in China are still in urgentneed for more accurate assessment of shipping emissions

35 Impact of engine speed on NOx emission factors

NOx is formed in the combustion chamber by a combinationof atmospheric nitrogen and oxygen under high-pressure andhigh-temperature conditions Many factors affect NOx for-mation including engine temperature injection point andfuel quality The IMO emissions limit for NOx is determinedby the rated speed of the engine however other factors mustalso be considered to reduce NOx emissions

The NOx EFs for the test vessels at various engine speedsare shown in Fig 4 The rated speeds of the vessels were1200 900 and 1000 rpm for HH DFH and XYH respec-tively The actual engine speeds of HH were much lower thanthe rated speed while the two larger-engine-power vesselsoperated close to their rated speeds except during one op-erating mode of DFH The NOx EFs for HH differed sig-nificantly in different operating modes ranging from 391to 143 g kgminus1 of fuel The NOx EF was highest when theengine speed reached sim 750 rpm (Fig 4) At lower enginespeeds the NOx EFs had fluctuating but lower values Athigher engine speeds closer to the rated speed of 1200 rpmthe NOx EFs were much lower The NOx EFs for the twolarger-engine-power vessels changed slightly with enginespeed but also had lowest values when their engine speedsapproached their rated speeds Combined with the dieselpropulsion characteristic curve there were large increasesin the fuel consumption rate when the engine speed in-creased Therefore a best-fit engine speed should be deter-mined based on both EFs and economic costs

Engineering approaches for reducing the NOx emissionsof marine engines may be applied before during or afterthe combustion process (Verschaeren et al 2014 Habib etal 2014) In the present study the NOx EFs of the two re-search vessels were below the IMO Tier I emissions limitsHowever for EMS measures should be taken to meet theIMO emissions limit including increasing the engine speedand applying engineering technologies during or after com-bustion such as exhaust gas recirculation (EGR) selectivenon-catalytic reduction (SNCR) or SCR

Figure 4 Emission factors for NOx at different engine speeds

4 Conclusions

Three offshore vessels with different engine power sourceswere chosen in this study to collect measured data of gaseousspecies and particulate matter including NO2 NO N2OCO CO2 TVOCs SO2 and the total suspended particulateBesides chemical composition of the PM were also analysedto give detailed EFs for OC EC water-soluble ions and metalelements Concentrations fuel-based EFs fuel-based aver-age EFs as well as power-based average EFs for species ofoffshore vessels in China were presented Furthermore im-pact of engine speed on NOx EFs was also discussed

There were higher concentrations of pollutants for low-engine-power vessel HH than for the other two vessels COconcentrations for offshore vessels were slightly lower thaninland vessels in China and all the three vessels had the low-est CO concentrations at their economic speeds (the speed ofthe least vessel operating expenditures during one voyagethey were high-speed mode cruise mode and high-speedmode for HH DFH and XYH respectively) More than 80 of the NOx was NO and all the offshore vessels had higherNOx concentrations in high-speed modes Because of thelow-sulfur fuels used in this study SO2 concentrations ofthese three offshore vessels were lower than that in the lit-eratures And the PM concentrations were much lower thaninland vessels while showing significant differences amongdifferent operating modes

Fuel-based EFs for gaseous species and PM were pre-sented based on the carbon balance method EFs for CO2were 2940ndash3106 3121ndash3160 and 3102ndash3162 g kgminus1 fuel forHH DFH and XYH Because of the combustion conditionssuch as excess air ratio combustion temperature and unifor-mity of the fuelndashair mixture EFs for CO showed high valuesin idling mode but low values in economic speed All theoffshore vessels had higher NOx EFs in low speed than inhigh speed but showed higher values when in acceleration

Atmos Chem Phys 16 6319ndash6334 2016 wwwatmos-chem-physnet1663192016

F Zhang et al Emission factors for gaseous and particulate pollutants 6331

process EFs for SO2 were 16 09 and 26 g kgminus1 fuel forHH DFH and XYH based on sulfur content of the fuels OCand EC were the main components of PM with low OC toEC ratios that were lower than 01 followed by SO2minus

4 NH+4 and NOminus3 Metals such as V Ni Cr Fe As and Cd made upa proportionately small part of the total PM mass

Fuel-based average EFs as well as power-based EFs for thethree different vessels of differing engine power were pre-sented EFs for most gaseous species and PM of HH weremuch higher compared with the other higher-engine-powervessels which was also gt 100 above the IMO standard forNOx Average PM EF of the low-engine-power vessel HHwas also much higher than that in the literatures Howeveraverage EFs for most species of the two larger-engine-powervessels were within the range of previously reported resultsEngine type was inferred as one of the most influence fac-tors for the differences of emission factors Inland and someoffshore ships in China always had higher NOx CO and PMemissions compared with other test vessels in previous stud-ies In addition emission factors that used for calculation ofship inventories in China always had lower values than testvessels

The impact of engine speed on EFs for NOx showed thatwhen the engine speed was close to the rated speed therewould be lower NOx EFs values However combined withthe high fuel consumption rate an optimal engine speedshould be determined based on both EFs and economic costsEmission reduction measures for NOx for some of the off-shore vessels in China are still essential to meet the IMOemission limit

Given the limits of vessel types and numbers this studysubstantially gives the EFs for gaseous species and PM ofthree different diesel-engine-powered offshore vessels How-ever as the development of ports in China emissions fromcargo ships and container ships with large engine power havebecome one of the most significant air pollution sources inport cities and regions Systematical measurement EFs of allkinds of offshore vessels in China are essential in order topresent the accurate emission inventory of ships

Information about the Supplement

Supplementary information includes the details of the real-world measurement system for vessels (Fig S1) the concen-trations of main gaseous matter and PM of shipping emis-sions (Fig S2) the types composition of offshore vessels inChina (Table S1) the distribution of vessels through grosstonnage in 2014 in offshore area of Yangtze River Delta (Ta-ble S2) the Chinese market consumption of marine oil in2011 (Table S3) the detection parameters for gaseous mat-ter (Table S4) the fuel-based EFs for the gaseous pollutants(Table S5) PM and the chemical composition in PM for dif-ferent operating (Table S6) modes and the actual operatingconditions of vessels (Table S7)

The Supplement related to this article is available onlineat doi105194acp-16-6319-2016-supplement

Acknowledgements This study was supported by the CAS Strate-gic Priority Research Program (No XDB05030303) the NaturalScientific Foundation of China (Nos 41273135 and 41473091)and the Shanghai science and Technology Commission project(No 15DZ1205401) The authors would like to express theirgratitude to the vessel owners Haohai Ocean Engineering LimitedLiability Company for vessel Haohai 0007 Ocean University ofChina for vessel Dongfanghong 02 and North China Sea Branchof the State Oceanic Administration for vessel Xiangyanghong08 for making their vessels available for the shipping emissionsstudy The views expressed in this document are solely those of theauthors and the funding agencies do not endorse any products orcommercial services mentioned in this publication

Edited by T Bond

References

Agrawal H Malloy Q G J Welch W A Miller J W andCocker D R In-use gaseous and particulate matter emissionsfrom a modern ocean going container vessel Atmos Environ42 5504ndash5510 doi101016jatmosenv200802053 2008a

Agrawal H Welch W A Miller J W and Cocker D R Emis-sion measurements from a crude oil tanker at sea Environ SciTechnol 42 7098ndash7103 doi101021es703102y 2008b

Almeida S M Pio C A Freitas M C Reis MA and Trancoso M A Source apportionment of fineand coarse particulate matter in a sub-urban area at theWestern European Coast Atmos Environ 39 3127ndash3138doi101016jatmosenv200501048 2005

Anderson M Salo K Hallquist A M and Fridell ECharacterization of particles from a marine engine op-erating at low loads Atmos Environ 101 65ndash71doi101016jatmosenv201411009 2015

Berg N Mellqvist J Jalkanen J-P and Balzani J Ship emis-sions of SO2 and NO2 DOAS measurements from airborne plat-forms Atmos Meas Tech 5 1085ndash1098 doi105194amt-5-1085-2012 2012

California Code of Regulation Titles 13 and 17 avail-able at httpsgovtwestlawcomcalregsBrowseHomeCaliforniaCaliforniaCodeofRegulationstransitionType=DefaultampcontextData=(scDefault) last access 20 April 2016

Cappa C D Williams E J Lack D A Buffaloe G M Coff-man D Hayden K L Herndon S C Lerner B M Li S-M Massoli P McLaren R Nuaaman I Onasch T B andQuinn P K A case study into the measurement of ship emis-sions from plume intercepts of the NOAA ship Miller FreemanAtmos Chem Phys 14 1337ndash1352 doi105194acp-14-1337-2014 2014

Carlton J Danton S Gawen R Lavender K Mathieson NNewell A Reynolds G Webster A Wills C and Wright AMarine exhaust emissions research programme Lloydrsquos RegisterEngineering Services London 63 1995

wwwatmos-chem-physnet1663192016 Atmos Chem Phys 16 6319ndash6334 2016

6328 F Zhang et al Emission factors for gaseous and particulate pollutants

33 Fuel-based average emission factors

Based on actual operating conditions (Table S7) average EFsfor the three vessels in the present study (according to Eq 4)along with EFs from previous studies are shown in Table 3EFs for all of the pollutants except SO2 were significantlyhigher for HH than for the other two vessels potentially dueto poor combustion conditions Most of the EFs for DFH andXYH were within the range of emissions for other vesselsdue to having well maintained engines and the high qualityof the fuels used The EFs for NOx PM and SO2 were muchlower than reported in previous studies (other than NOx forocean-going vessels) All the sulfur of the fuels in the presentstudy were significantly below the emissions limit of 350 established by IMO in the revised MARPOL Annex VI rulesapplicable since 2012 (IMO 1998)

The IMO Tier I emissions limit for NOx is450times nminus02 g kWhminus1 (n rated speed 130 lt nlt 2000 rpm)The rated speed and fuel consumption rates for each vesselare shown in Table 1 Thus the emissions limits for HHDFH and XYH would be 545 575 and 565 g kgminus1 fuelrespectively calculated combined with Eq (5) The averagefuel-based EFs for NOx of ship HH was more than 100 above the IMO standard while those of the other two shipswere below the IMO standard (Table 3) PM emissions forHH were also higher than previously reported but thosefor the two research vessels were much lower (Table 3)Fuel type is one of the most important influence factors onpollutant emissions for example sulfur content in the fuelnot only influence the SO2 emission directly but also hadimpact on PM formation in the flue gas stack with low sulfurcontent in fuels reduces PM formation (Lack et al 2011)Vessels with higher sulfur content always had relativelyhigher PM emissions which were also shown in Table 3In addition different engines and levels of maintenancehave a significant impact on all combustion-dependentemissions Emission reduction measures have been used insome vessels For example NOx emissions can be reducedby measures such as selective catalytic reduction (SCR) anddirect water injection (DWI) which had been implementedon some vessels previously studied in a harbour in Finland(Pirjola et al 2014) The results showed that SCR effec-tively reduced NOx emissions while vessels with DWI hadhigh PM emissions The engine type might be an importantcause of the different emissions such as HH had muchhigher pollutants emissions with an engine produced inChina and yet DFHrsquos engine produced in Germany Besidesemission tests for a high-speed marine diesel engine withdifferent kinds of diesels showed that diesel type had limitedinfluence on emissions such as NOx CO and CH but asignificant impact on PM emission (289ndash415 ) becauseof the different sulfur content in fuel (Xu 2008)

34 Power-based emission factors

Based on the engine power and fuel consumption rates ofthe vessels power-based EFs were calculated (according toEq 5) and compared to results from previous studies (Ta-ble 4) The EFs for HH were much higher than those forthe other two vessels except for SO2 HH also had signif-icantly higher EFs for NOx than previously reported val-ues while EFs for NOx of DFH and XYH were within therange of previously reported results Engine type was consid-ered to be a significant influence factor for NOx emissionswith lower engine speed having higher NOx emission factors(Celo et al 2015) In addition compared to other vesselswith a similar engine type and diesel fuel HH still had rela-tively higher NOx EF (seen in Table 4) which could reflectthe impact of engine condition (engine quality and mainte-nance level) on shipping emissions CO EFs for the test ves-sels in the present study were higher than previous studieswhich produced similar results to those of inland ships andother test vessels in China (Fu et al 2013 Song 2015) Inspite of the influence of engine type on CO emissions thatwith the higher engine speed having higher CO EFs (Celoet al 2015) engine condition combined with fuel qualitymight have significant influence All of the EFs for SO2 inthe present study were lower than those in previous studiesbecause of the low sulfur content of the present fuels Gen-erally PM emissions from marine diesel fuels are dependenton the fuel (sulfate and metal oxide ash constituents) and oncombustion conditions (unburned hydrocarbons and carbonresidue constituents) (Cooper 2003) HH had the highest PMemissions among the test vessels although there were almostno differences among the fuels (Table S6) Besides HH hadeven higher PM EFs than previously reported vessels withHFO fuel and XYH had much lower PM EFs than all theother vessels with even lower sulfur content fuel Thereforecombustion conditions were likely the determining factor forthe differences It can be seen from Table 4 that most previ-ous studies focused on the heavy fuel oil of shipping emis-sions Compared with diesel fuels heavy fuel oil always hadrelatively low CO emission factors and high PM emissionfactors And among the heavy-fuel-oil-using vessels enginetype (engine speed and engine power level) always playedan important role on emissions such as NOx and CO whichwith lower engine speed having higher NOx EFs and lowerCO EFs

Combined with other emission data of test ships in China(Fu et al 2013 Song 2015) it could be seen that inlandand some test offshore ships in China always had higherNOx CO and PM emissions compared with other test ves-sels in previous studies And among the test vessels in Chinathere also were differences for different engine types andship types In addition emission factors that were used forcalculation of ship inventories in China always came fromother countries and areas However there seemed to be sig-nificant differences between the reference and test data such

Atmos Chem Phys 16 6319ndash6334 2016 wwwatmos-chem-physnet1663192016

F Zhang et al Emission factors for gaseous and particulate pollutants 6329

Tabl

e4

Pow

er-b

ased

EFs

inth

epr

esen

tstu

dyan

dpr

evio

usst

udie

s(g

kWhminus

1 )

Ves

sel

CO

2C

ON

ON

O2

N2O

NO

xT

VO

Cs

PMSO

2E

ngin

epo

wer

Eng

ine

Fuel

type

and

ID(k

W)

spee

dsu

lfur

cont

ent

(rm

p)(w

t)

HH

699plusmn

352

738plusmn

376

220plusmn

841

345plusmn

124

030plusmn

039

258plusmn

100

544plusmn

484

209plusmn

048

036

350

1200

DO

00

798

DFH

631plusmn

352

139plusmn

020

604plusmn

032

102plusmn

008

008plusmn

004

714plusmn

044

017plusmn

001

014plusmn

007

018

1600

900

DO

00

458

XY

H69

7plusmn

385

201plusmn

065

587plusmn

036

104plusmn

009

007plusmn

003

697plusmn

048

092plusmn

ndash0

04plusmn

001

057

600

1000

DO

01

30Ta

nker

(Win

nes

and

Frid

ell

2010

)ndash

161

ndashndash

ndash7

82ndash

058

ndash45

0060

0H

FO1

6

Ber

thed

ship

s(C

oope

r65

3ndash76

80

33ndash0

97

ndashndash

ndash14

2ndash2

02

ndash0

14ndash0

45

026

ndash53

AE

720

ndash127

072

0ndash18

00M

GO

00

6ndash1

220

03)

691ndash

803

077

ndash17

112

9ndash1

75

048

ndash06

72

5ndash9

612

70ndash2

675

720ndash

750

RO

05

3ndash2

269

1ndash69

40

92ndash0

98

96ndash

99

017

ndash01

91

014

8072

0M

DO

02

3C

rude

oilt

anke

r(A

graw

alet

al

2008

b)58

8ndash66

00

77ndash1

78

ndashndash

ndash15

8ndash2

10

ndash1

10ndash1

78

766

ndash86

015

750

90H

FO2

85

Cru

ise

ship

s(P

opla

wsk

iet

al

2011

)ndash

ndashndash

140

ndashndash

ndash2

914

20

US

EPA

621

14

ndashndash

ndash18

1ndash

131

103

Mar

ine

engi

ne(S

ippu

laet

ndash1

2ndash11

411

3ndash2

95

ndashndash

114

ndash30

90ndash

95

083

ndash63

6ndash

1500

HFO

27

al

2014

)0ndash

885

69ndash2

58

584

ndash33

90

83ndash1

97

015

ndash09

3D

FL

arge

mar

ine

ship

s(K

han

etal

20

13)

533ndash

612

035

ndash06

0ndash

ndashndash

166

ndash20

6ndash

091

ndash21

97

2ndash11

436

740ndash

6853

097

ndash102

HFO

21

5ndash3

14

Oce

ango

ing

cont

aine

rve

ssel

(Agr

awal

etal

20

08a)

588ndash

660

077

ndash18

1ndash

ndashndash

158

ndash21

0ndash

109

ndash17

67

66ndash8

60

5027

010

4H

FO2

05

Lar

geca

rgo

vess

el(M

olda

nova

etal

20

09)

667

042

ndashndash

ndash14

22

ndash1

0310

320

200

97H

FO1

9

Oce

ango

ing

carg

ove

ssel

614plusmn

10

83plusmn

001

ndashndash

ndash16

3plusmn

02

ndash1

51plusmn

007

87plusmn

01

128

IFO

180

(Cel

oet

al

2015

)62

6plusmn

70

26plusmn

001

114plusmn

01

081plusmn

002

58plusmn

007

525

IFO

180

628plusmn

90

30plusmn

001

113plusmn

01

094plusmn

002

87plusmn

01

450

IFO

180

628plusmn

10

81plusmn

003

122plusmn

001

083plusmn

001

64plusmn

01

450

IFO

180

609plusmn

11

31plusmn

002

ndashndash

ndash8

4plusmn

003

ndash0

37plusmn

001

47plusmn

001

440

IFO

6060

5plusmn

10

00ndash

ndashndash

167plusmn

01

ndash2

2plusmn

02

103plusmn

003

116

IFO

380

622plusmn

11

22plusmn

002

ndashndash

ndash10

7plusmn

004

ndash0

30plusmn

003

047plusmn

01

450

MD

O

AE

aux

iliar

yen

gine

DO

die

selo

ilH

FOh

eavy

fuel

oil

MG

Om

arin

ega

soi

lR

Or

esid

ualo

ilM

DO

mar

ine

dies

eloi

lD

Fdi

esel

fuel

IFO

int

erm

edia

tefu

eloi

l

wwwatmos-chem-physnet1663192016 Atmos Chem Phys 16 6319ndash6334 2016

6330 F Zhang et al Emission factors for gaseous and particulate pollutants

as 100 to 132 g kW hminus1 of NOx EF and 11 to 17 g kW hminus1

of CO EF used for inland ships for ship inventory calcula-tion (Zhu et al 2015) 100 to 181 g kW hminus1 of NOx EFand 11 to 15 g kW hminus1 of CO EF for harbour ships (Yanget al 2015) compared to 15 to 173 g kW hminus1 of NOx EFand 46 to 103 g kW hminus1 of CO EF from test inland ships(convert the fuel-based EF to power-based EF with a factorof 200 g kW hminus1) (Song 2015) and 697 to 258 g kW hminus1 ofNOx EF and 139 to 738 g kW hminus1 of CO EF in the presentstudy (Yang et al 2015) Besides whether there are obviousdifferences of EFs between other types of vessels in China(such as low-speed engine vessels with heavy fuel oil) andprevious studies is still unclear Therefore much more mea-surement data for different vessels in China are still in urgentneed for more accurate assessment of shipping emissions

35 Impact of engine speed on NOx emission factors

NOx is formed in the combustion chamber by a combinationof atmospheric nitrogen and oxygen under high-pressure andhigh-temperature conditions Many factors affect NOx for-mation including engine temperature injection point andfuel quality The IMO emissions limit for NOx is determinedby the rated speed of the engine however other factors mustalso be considered to reduce NOx emissions

The NOx EFs for the test vessels at various engine speedsare shown in Fig 4 The rated speeds of the vessels were1200 900 and 1000 rpm for HH DFH and XYH respec-tively The actual engine speeds of HH were much lower thanthe rated speed while the two larger-engine-power vesselsoperated close to their rated speeds except during one op-erating mode of DFH The NOx EFs for HH differed sig-nificantly in different operating modes ranging from 391to 143 g kgminus1 of fuel The NOx EF was highest when theengine speed reached sim 750 rpm (Fig 4) At lower enginespeeds the NOx EFs had fluctuating but lower values Athigher engine speeds closer to the rated speed of 1200 rpmthe NOx EFs were much lower The NOx EFs for the twolarger-engine-power vessels changed slightly with enginespeed but also had lowest values when their engine speedsapproached their rated speeds Combined with the dieselpropulsion characteristic curve there were large increasesin the fuel consumption rate when the engine speed in-creased Therefore a best-fit engine speed should be deter-mined based on both EFs and economic costs

Engineering approaches for reducing the NOx emissionsof marine engines may be applied before during or afterthe combustion process (Verschaeren et al 2014 Habib etal 2014) In the present study the NOx EFs of the two re-search vessels were below the IMO Tier I emissions limitsHowever for EMS measures should be taken to meet theIMO emissions limit including increasing the engine speedand applying engineering technologies during or after com-bustion such as exhaust gas recirculation (EGR) selectivenon-catalytic reduction (SNCR) or SCR

Figure 4 Emission factors for NOx at different engine speeds

4 Conclusions

Three offshore vessels with different engine power sourceswere chosen in this study to collect measured data of gaseousspecies and particulate matter including NO2 NO N2OCO CO2 TVOCs SO2 and the total suspended particulateBesides chemical composition of the PM were also analysedto give detailed EFs for OC EC water-soluble ions and metalelements Concentrations fuel-based EFs fuel-based aver-age EFs as well as power-based average EFs for species ofoffshore vessels in China were presented Furthermore im-pact of engine speed on NOx EFs was also discussed

There were higher concentrations of pollutants for low-engine-power vessel HH than for the other two vessels COconcentrations for offshore vessels were slightly lower thaninland vessels in China and all the three vessels had the low-est CO concentrations at their economic speeds (the speed ofthe least vessel operating expenditures during one voyagethey were high-speed mode cruise mode and high-speedmode for HH DFH and XYH respectively) More than 80 of the NOx was NO and all the offshore vessels had higherNOx concentrations in high-speed modes Because of thelow-sulfur fuels used in this study SO2 concentrations ofthese three offshore vessels were lower than that in the lit-eratures And the PM concentrations were much lower thaninland vessels while showing significant differences amongdifferent operating modes

Fuel-based EFs for gaseous species and PM were pre-sented based on the carbon balance method EFs for CO2were 2940ndash3106 3121ndash3160 and 3102ndash3162 g kgminus1 fuel forHH DFH and XYH Because of the combustion conditionssuch as excess air ratio combustion temperature and unifor-mity of the fuelndashair mixture EFs for CO showed high valuesin idling mode but low values in economic speed All theoffshore vessels had higher NOx EFs in low speed than inhigh speed but showed higher values when in acceleration

Atmos Chem Phys 16 6319ndash6334 2016 wwwatmos-chem-physnet1663192016

F Zhang et al Emission factors for gaseous and particulate pollutants 6331

process EFs for SO2 were 16 09 and 26 g kgminus1 fuel forHH DFH and XYH based on sulfur content of the fuels OCand EC were the main components of PM with low OC toEC ratios that were lower than 01 followed by SO2minus

4 NH+4 and NOminus3 Metals such as V Ni Cr Fe As and Cd made upa proportionately small part of the total PM mass

Fuel-based average EFs as well as power-based EFs for thethree different vessels of differing engine power were pre-sented EFs for most gaseous species and PM of HH weremuch higher compared with the other higher-engine-powervessels which was also gt 100 above the IMO standard forNOx Average PM EF of the low-engine-power vessel HHwas also much higher than that in the literatures Howeveraverage EFs for most species of the two larger-engine-powervessels were within the range of previously reported resultsEngine type was inferred as one of the most influence fac-tors for the differences of emission factors Inland and someoffshore ships in China always had higher NOx CO and PMemissions compared with other test vessels in previous stud-ies In addition emission factors that used for calculation ofship inventories in China always had lower values than testvessels

The impact of engine speed on EFs for NOx showed thatwhen the engine speed was close to the rated speed therewould be lower NOx EFs values However combined withthe high fuel consumption rate an optimal engine speedshould be determined based on both EFs and economic costsEmission reduction measures for NOx for some of the off-shore vessels in China are still essential to meet the IMOemission limit

Given the limits of vessel types and numbers this studysubstantially gives the EFs for gaseous species and PM ofthree different diesel-engine-powered offshore vessels How-ever as the development of ports in China emissions fromcargo ships and container ships with large engine power havebecome one of the most significant air pollution sources inport cities and regions Systematical measurement EFs of allkinds of offshore vessels in China are essential in order topresent the accurate emission inventory of ships

Information about the Supplement

Supplementary information includes the details of the real-world measurement system for vessels (Fig S1) the concen-trations of main gaseous matter and PM of shipping emis-sions (Fig S2) the types composition of offshore vessels inChina (Table S1) the distribution of vessels through grosstonnage in 2014 in offshore area of Yangtze River Delta (Ta-ble S2) the Chinese market consumption of marine oil in2011 (Table S3) the detection parameters for gaseous mat-ter (Table S4) the fuel-based EFs for the gaseous pollutants(Table S5) PM and the chemical composition in PM for dif-ferent operating (Table S6) modes and the actual operatingconditions of vessels (Table S7)

The Supplement related to this article is available onlineat doi105194acp-16-6319-2016-supplement

Acknowledgements This study was supported by the CAS Strate-gic Priority Research Program (No XDB05030303) the NaturalScientific Foundation of China (Nos 41273135 and 41473091)and the Shanghai science and Technology Commission project(No 15DZ1205401) The authors would like to express theirgratitude to the vessel owners Haohai Ocean Engineering LimitedLiability Company for vessel Haohai 0007 Ocean University ofChina for vessel Dongfanghong 02 and North China Sea Branchof the State Oceanic Administration for vessel Xiangyanghong08 for making their vessels available for the shipping emissionsstudy The views expressed in this document are solely those of theauthors and the funding agencies do not endorse any products orcommercial services mentioned in this publication

Edited by T Bond

References

Agrawal H Malloy Q G J Welch W A Miller J W andCocker D R In-use gaseous and particulate matter emissionsfrom a modern ocean going container vessel Atmos Environ42 5504ndash5510 doi101016jatmosenv200802053 2008a

Agrawal H Welch W A Miller J W and Cocker D R Emis-sion measurements from a crude oil tanker at sea Environ SciTechnol 42 7098ndash7103 doi101021es703102y 2008b

Almeida S M Pio C A Freitas M C Reis MA and Trancoso M A Source apportionment of fineand coarse particulate matter in a sub-urban area at theWestern European Coast Atmos Environ 39 3127ndash3138doi101016jatmosenv200501048 2005

Anderson M Salo K Hallquist A M and Fridell ECharacterization of particles from a marine engine op-erating at low loads Atmos Environ 101 65ndash71doi101016jatmosenv201411009 2015

Berg N Mellqvist J Jalkanen J-P and Balzani J Ship emis-sions of SO2 and NO2 DOAS measurements from airborne plat-forms Atmos Meas Tech 5 1085ndash1098 doi105194amt-5-1085-2012 2012

California Code of Regulation Titles 13 and 17 avail-able at httpsgovtwestlawcomcalregsBrowseHomeCaliforniaCaliforniaCodeofRegulationstransitionType=DefaultampcontextData=(scDefault) last access 20 April 2016

Cappa C D Williams E J Lack D A Buffaloe G M Coff-man D Hayden K L Herndon S C Lerner B M Li S-M Massoli P McLaren R Nuaaman I Onasch T B andQuinn P K A case study into the measurement of ship emis-sions from plume intercepts of the NOAA ship Miller FreemanAtmos Chem Phys 14 1337ndash1352 doi105194acp-14-1337-2014 2014

Carlton J Danton S Gawen R Lavender K Mathieson NNewell A Reynolds G Webster A Wills C and Wright AMarine exhaust emissions research programme Lloydrsquos RegisterEngineering Services London 63 1995

wwwatmos-chem-physnet1663192016 Atmos Chem Phys 16 6319ndash6334 2016

F Zhang et al Emission factors for gaseous and particulate pollutants 6329

Tabl

e4

Pow

er-b

ased

EFs

inth

epr

esen

tstu

dyan

dpr

evio

usst

udie

s(g

kWhminus

1 )

Ves

sel

CO

2C

ON

ON

O2

N2O

NO

xT

VO

Cs

PMSO

2E

ngin

epo

wer

Eng

ine

Fuel

type

and

ID(k

W)

spee

dsu

lfur

cont

ent

(rm

p)(w

t)

HH

699plusmn

352

738plusmn

376

220plusmn

841

345plusmn

124

030plusmn

039

258plusmn

100

544plusmn

484

209plusmn

048

036

350

1200

DO

00

798

DFH

631plusmn

352

139plusmn

020

604plusmn

032

102plusmn

008

008plusmn

004

714plusmn

044

017plusmn

001

014plusmn

007

018

1600

900

DO

00

458

XY

H69

7plusmn

385

201plusmn

065

587plusmn

036

104plusmn

009

007plusmn

003

697plusmn

048

092plusmn

ndash0

04plusmn

001

057

600

1000

DO

01

30Ta

nker

(Win

nes

and

Frid

ell

2010

)ndash

161

ndashndash

ndash7

82ndash

058

ndash45

0060

0H

FO1

6

Ber

thed

ship

s(C

oope

r65

3ndash76

80

33ndash0

97

ndashndash

ndash14

2ndash2

02

ndash0

14ndash0

45

026

ndash53

AE

720

ndash127

072

0ndash18

00M

GO

00

6ndash1

220

03)

691ndash

803

077

ndash17

112

9ndash1

75

048

ndash06

72

5ndash9

612

70ndash2

675

720ndash

750

RO

05

3ndash2

269

1ndash69

40

92ndash0

98

96ndash

99

017

ndash01

91

014

8072

0M

DO

02

3C

rude

oilt

anke

r(A

graw

alet

al

2008

b)58

8ndash66

00

77ndash1

78

ndashndash

ndash15

8ndash2

10

ndash1

10ndash1

78

766

ndash86

015

750

90H

FO2

85

Cru

ise

ship

s(P

opla

wsk

iet

al

2011

)ndash

ndashndash

140

ndashndash

ndash2

914

20

US

EPA

621

14

ndashndash

ndash18

1ndash

131

103

Mar

ine

engi

ne(S

ippu

laet

ndash1

2ndash11

411

3ndash2

95

ndashndash

114

ndash30

90ndash

95

083

ndash63

6ndash

1500

HFO

27

al

2014

)0ndash

885

69ndash2

58

584

ndash33

90

83ndash1

97

015

ndash09

3D

FL

arge

mar

ine

ship

s(K

han

etal

20

13)

533ndash

612

035

ndash06

0ndash

ndashndash

166

ndash20

6ndash

091

ndash21

97

2ndash11

436

740ndash

6853

097

ndash102

HFO

21

5ndash3

14

Oce

ango

ing

cont

aine

rve

ssel

(Agr

awal

etal

20

08a)

588ndash

660

077

ndash18

1ndash

ndashndash

158

ndash21

0ndash

109

ndash17

67

66ndash8

60

5027

010

4H

FO2

05

Lar

geca

rgo

vess

el(M

olda

nova

etal

20

09)

667

042

ndashndash

ndash14

22

ndash1

0310

320

200

97H

FO1

9

Oce

ango

ing

carg

ove

ssel

614plusmn

10

83plusmn

001

ndashndash

ndash16

3plusmn

02

ndash1

51plusmn

007

87plusmn

01

128

IFO

180

(Cel

oet

al

2015

)62

6plusmn

70

26plusmn

001

114plusmn

01

081plusmn

002

58plusmn

007

525

IFO

180

628plusmn

90

30plusmn

001

113plusmn

01

094plusmn

002

87plusmn

01

450

IFO

180

628plusmn

10

81plusmn

003

122plusmn

001

083plusmn

001

64plusmn

01

450

IFO

180

609plusmn

11

31plusmn

002

ndashndash

ndash8

4plusmn

003

ndash0

37plusmn

001

47plusmn

001

440

IFO

6060

5plusmn

10

00ndash

ndashndash

167plusmn

01

ndash2

2plusmn

02

103plusmn

003

116

IFO

380

622plusmn

11

22plusmn

002

ndashndash

ndash10

7plusmn

004

ndash0

30plusmn

003

047plusmn

01

450

MD

O

AE

aux

iliar

yen

gine

DO

die

selo

ilH

FOh

eavy

fuel

oil

MG

Om

arin

ega

soi

lR

Or

esid

ualo

ilM

DO

mar

ine

dies

eloi

lD

Fdi

esel

fuel

IFO

int

erm

edia

tefu

eloi

l

wwwatmos-chem-physnet1663192016 Atmos Chem Phys 16 6319ndash6334 2016

6330 F Zhang et al Emission factors for gaseous and particulate pollutants

as 100 to 132 g kW hminus1 of NOx EF and 11 to 17 g kW hminus1

of CO EF used for inland ships for ship inventory calcula-tion (Zhu et al 2015) 100 to 181 g kW hminus1 of NOx EFand 11 to 15 g kW hminus1 of CO EF for harbour ships (Yanget al 2015) compared to 15 to 173 g kW hminus1 of NOx EFand 46 to 103 g kW hminus1 of CO EF from test inland ships(convert the fuel-based EF to power-based EF with a factorof 200 g kW hminus1) (Song 2015) and 697 to 258 g kW hminus1 ofNOx EF and 139 to 738 g kW hminus1 of CO EF in the presentstudy (Yang et al 2015) Besides whether there are obviousdifferences of EFs between other types of vessels in China(such as low-speed engine vessels with heavy fuel oil) andprevious studies is still unclear Therefore much more mea-surement data for different vessels in China are still in urgentneed for more accurate assessment of shipping emissions

35 Impact of engine speed on NOx emission factors

NOx is formed in the combustion chamber by a combinationof atmospheric nitrogen and oxygen under high-pressure andhigh-temperature conditions Many factors affect NOx for-mation including engine temperature injection point andfuel quality The IMO emissions limit for NOx is determinedby the rated speed of the engine however other factors mustalso be considered to reduce NOx emissions

The NOx EFs for the test vessels at various engine speedsare shown in Fig 4 The rated speeds of the vessels were1200 900 and 1000 rpm for HH DFH and XYH respec-tively The actual engine speeds of HH were much lower thanthe rated speed while the two larger-engine-power vesselsoperated close to their rated speeds except during one op-erating mode of DFH The NOx EFs for HH differed sig-nificantly in different operating modes ranging from 391to 143 g kgminus1 of fuel The NOx EF was highest when theengine speed reached sim 750 rpm (Fig 4) At lower enginespeeds the NOx EFs had fluctuating but lower values Athigher engine speeds closer to the rated speed of 1200 rpmthe NOx EFs were much lower The NOx EFs for the twolarger-engine-power vessels changed slightly with enginespeed but also had lowest values when their engine speedsapproached their rated speeds Combined with the dieselpropulsion characteristic curve there were large increasesin the fuel consumption rate when the engine speed in-creased Therefore a best-fit engine speed should be deter-mined based on both EFs and economic costs

Engineering approaches for reducing the NOx emissionsof marine engines may be applied before during or afterthe combustion process (Verschaeren et al 2014 Habib etal 2014) In the present study the NOx EFs of the two re-search vessels were below the IMO Tier I emissions limitsHowever for EMS measures should be taken to meet theIMO emissions limit including increasing the engine speedand applying engineering technologies during or after com-bustion such as exhaust gas recirculation (EGR) selectivenon-catalytic reduction (SNCR) or SCR

Figure 4 Emission factors for NOx at different engine speeds

4 Conclusions

Three offshore vessels with different engine power sourceswere chosen in this study to collect measured data of gaseousspecies and particulate matter including NO2 NO N2OCO CO2 TVOCs SO2 and the total suspended particulateBesides chemical composition of the PM were also analysedto give detailed EFs for OC EC water-soluble ions and metalelements Concentrations fuel-based EFs fuel-based aver-age EFs as well as power-based average EFs for species ofoffshore vessels in China were presented Furthermore im-pact of engine speed on NOx EFs was also discussed

There were higher concentrations of pollutants for low-engine-power vessel HH than for the other two vessels COconcentrations for offshore vessels were slightly lower thaninland vessels in China and all the three vessels had the low-est CO concentrations at their economic speeds (the speed ofthe least vessel operating expenditures during one voyagethey were high-speed mode cruise mode and high-speedmode for HH DFH and XYH respectively) More than 80 of the NOx was NO and all the offshore vessels had higherNOx concentrations in high-speed modes Because of thelow-sulfur fuels used in this study SO2 concentrations ofthese three offshore vessels were lower than that in the lit-eratures And the PM concentrations were much lower thaninland vessels while showing significant differences amongdifferent operating modes

Fuel-based EFs for gaseous species and PM were pre-sented based on the carbon balance method EFs for CO2were 2940ndash3106 3121ndash3160 and 3102ndash3162 g kgminus1 fuel forHH DFH and XYH Because of the combustion conditionssuch as excess air ratio combustion temperature and unifor-mity of the fuelndashair mixture EFs for CO showed high valuesin idling mode but low values in economic speed All theoffshore vessels had higher NOx EFs in low speed than inhigh speed but showed higher values when in acceleration

Atmos Chem Phys 16 6319ndash6334 2016 wwwatmos-chem-physnet1663192016

F Zhang et al Emission factors for gaseous and particulate pollutants 6331

process EFs for SO2 were 16 09 and 26 g kgminus1 fuel forHH DFH and XYH based on sulfur content of the fuels OCand EC were the main components of PM with low OC toEC ratios that were lower than 01 followed by SO2minus

4 NH+4 and NOminus3 Metals such as V Ni Cr Fe As and Cd made upa proportionately small part of the total PM mass

Fuel-based average EFs as well as power-based EFs for thethree different vessels of differing engine power were pre-sented EFs for most gaseous species and PM of HH weremuch higher compared with the other higher-engine-powervessels which was also gt 100 above the IMO standard forNOx Average PM EF of the low-engine-power vessel HHwas also much higher than that in the literatures Howeveraverage EFs for most species of the two larger-engine-powervessels were within the range of previously reported resultsEngine type was inferred as one of the most influence fac-tors for the differences of emission factors Inland and someoffshore ships in China always had higher NOx CO and PMemissions compared with other test vessels in previous stud-ies In addition emission factors that used for calculation ofship inventories in China always had lower values than testvessels

The impact of engine speed on EFs for NOx showed thatwhen the engine speed was close to the rated speed therewould be lower NOx EFs values However combined withthe high fuel consumption rate an optimal engine speedshould be determined based on both EFs and economic costsEmission reduction measures for NOx for some of the off-shore vessels in China are still essential to meet the IMOemission limit

Given the limits of vessel types and numbers this studysubstantially gives the EFs for gaseous species and PM ofthree different diesel-engine-powered offshore vessels How-ever as the development of ports in China emissions fromcargo ships and container ships with large engine power havebecome one of the most significant air pollution sources inport cities and regions Systematical measurement EFs of allkinds of offshore vessels in China are essential in order topresent the accurate emission inventory of ships

Information about the Supplement

Supplementary information includes the details of the real-world measurement system for vessels (Fig S1) the concen-trations of main gaseous matter and PM of shipping emis-sions (Fig S2) the types composition of offshore vessels inChina (Table S1) the distribution of vessels through grosstonnage in 2014 in offshore area of Yangtze River Delta (Ta-ble S2) the Chinese market consumption of marine oil in2011 (Table S3) the detection parameters for gaseous mat-ter (Table S4) the fuel-based EFs for the gaseous pollutants(Table S5) PM and the chemical composition in PM for dif-ferent operating (Table S6) modes and the actual operatingconditions of vessels (Table S7)

The Supplement related to this article is available onlineat doi105194acp-16-6319-2016-supplement

Acknowledgements This study was supported by the CAS Strate-gic Priority Research Program (No XDB05030303) the NaturalScientific Foundation of China (Nos 41273135 and 41473091)and the Shanghai science and Technology Commission project(No 15DZ1205401) The authors would like to express theirgratitude to the vessel owners Haohai Ocean Engineering LimitedLiability Company for vessel Haohai 0007 Ocean University ofChina for vessel Dongfanghong 02 and North China Sea Branchof the State Oceanic Administration for vessel Xiangyanghong08 for making their vessels available for the shipping emissionsstudy The views expressed in this document are solely those of theauthors and the funding agencies do not endorse any products orcommercial services mentioned in this publication

Edited by T Bond

References

Agrawal H Malloy Q G J Welch W A Miller J W andCocker D R In-use gaseous and particulate matter emissionsfrom a modern ocean going container vessel Atmos Environ42 5504ndash5510 doi101016jatmosenv200802053 2008a

Agrawal H Welch W A Miller J W and Cocker D R Emis-sion measurements from a crude oil tanker at sea Environ SciTechnol 42 7098ndash7103 doi101021es703102y 2008b

Almeida S M Pio C A Freitas M C Reis MA and Trancoso M A Source apportionment of fineand coarse particulate matter in a sub-urban area at theWestern European Coast Atmos Environ 39 3127ndash3138doi101016jatmosenv200501048 2005

Anderson M Salo K Hallquist A M and Fridell ECharacterization of particles from a marine engine op-erating at low loads Atmos Environ 101 65ndash71doi101016jatmosenv201411009 2015

Berg N Mellqvist J Jalkanen J-P and Balzani J Ship emis-sions of SO2 and NO2 DOAS measurements from airborne plat-forms Atmos Meas Tech 5 1085ndash1098 doi105194amt-5-1085-2012 2012

California Code of Regulation Titles 13 and 17 avail-able at httpsgovtwestlawcomcalregsBrowseHomeCaliforniaCaliforniaCodeofRegulationstransitionType=DefaultampcontextData=(scDefault) last access 20 April 2016

Cappa C D Williams E J Lack D A Buffaloe G M Coff-man D Hayden K L Herndon S C Lerner B M Li S-M Massoli P McLaren R Nuaaman I Onasch T B andQuinn P K A case study into the measurement of ship emis-sions from plume intercepts of the NOAA ship Miller FreemanAtmos Chem Phys 14 1337ndash1352 doi105194acp-14-1337-2014 2014

Carlton J Danton S Gawen R Lavender K Mathieson NNewell A Reynolds G Webster A Wills C and Wright AMarine exhaust emissions research programme Lloydrsquos RegisterEngineering Services London 63 1995

wwwatmos-chem-physnet1663192016 Atmos Chem Phys 16 6319ndash6334 2016

6330 F Zhang et al Emission factors for gaseous and particulate pollutants

as 100 to 132 g kW hminus1 of NOx EF and 11 to 17 g kW hminus1

of CO EF used for inland ships for ship inventory calcula-tion (Zhu et al 2015) 100 to 181 g kW hminus1 of NOx EFand 11 to 15 g kW hminus1 of CO EF for harbour ships (Yanget al 2015) compared to 15 to 173 g kW hminus1 of NOx EFand 46 to 103 g kW hminus1 of CO EF from test inland ships(convert the fuel-based EF to power-based EF with a factorof 200 g kW hminus1) (Song 2015) and 697 to 258 g kW hminus1 ofNOx EF and 139 to 738 g kW hminus1 of CO EF in the presentstudy (Yang et al 2015) Besides whether there are obviousdifferences of EFs between other types of vessels in China(such as low-speed engine vessels with heavy fuel oil) andprevious studies is still unclear Therefore much more mea-surement data for different vessels in China are still in urgentneed for more accurate assessment of shipping emissions

35 Impact of engine speed on NOx emission factors

NOx is formed in the combustion chamber by a combinationof atmospheric nitrogen and oxygen under high-pressure andhigh-temperature conditions Many factors affect NOx for-mation including engine temperature injection point andfuel quality The IMO emissions limit for NOx is determinedby the rated speed of the engine however other factors mustalso be considered to reduce NOx emissions

The NOx EFs for the test vessels at various engine speedsare shown in Fig 4 The rated speeds of the vessels were1200 900 and 1000 rpm for HH DFH and XYH respec-tively The actual engine speeds of HH were much lower thanthe rated speed while the two larger-engine-power vesselsoperated close to their rated speeds except during one op-erating mode of DFH The NOx EFs for HH differed sig-nificantly in different operating modes ranging from 391to 143 g kgminus1 of fuel The NOx EF was highest when theengine speed reached sim 750 rpm (Fig 4) At lower enginespeeds the NOx EFs had fluctuating but lower values Athigher engine speeds closer to the rated speed of 1200 rpmthe NOx EFs were much lower The NOx EFs for the twolarger-engine-power vessels changed slightly with enginespeed but also had lowest values when their engine speedsapproached their rated speeds Combined with the dieselpropulsion characteristic curve there were large increasesin the fuel consumption rate when the engine speed in-creased Therefore a best-fit engine speed should be deter-mined based on both EFs and economic costs

Engineering approaches for reducing the NOx emissionsof marine engines may be applied before during or afterthe combustion process (Verschaeren et al 2014 Habib etal 2014) In the present study the NOx EFs of the two re-search vessels were below the IMO Tier I emissions limitsHowever for EMS measures should be taken to meet theIMO emissions limit including increasing the engine speedand applying engineering technologies during or after com-bustion such as exhaust gas recirculation (EGR) selectivenon-catalytic reduction (SNCR) or SCR

Figure 4 Emission factors for NOx at different engine speeds

4 Conclusions

Three offshore vessels with different engine power sourceswere chosen in this study to collect measured data of gaseousspecies and particulate matter including NO2 NO N2OCO CO2 TVOCs SO2 and the total suspended particulateBesides chemical composition of the PM were also analysedto give detailed EFs for OC EC water-soluble ions and metalelements Concentrations fuel-based EFs fuel-based aver-age EFs as well as power-based average EFs for species ofoffshore vessels in China were presented Furthermore im-pact of engine speed on NOx EFs was also discussed

There were higher concentrations of pollutants for low-engine-power vessel HH than for the other two vessels COconcentrations for offshore vessels were slightly lower thaninland vessels in China and all the three vessels had the low-est CO concentrations at their economic speeds (the speed ofthe least vessel operating expenditures during one voyagethey were high-speed mode cruise mode and high-speedmode for HH DFH and XYH respectively) More than 80 of the NOx was NO and all the offshore vessels had higherNOx concentrations in high-speed modes Because of thelow-sulfur fuels used in this study SO2 concentrations ofthese three offshore vessels were lower than that in the lit-eratures And the PM concentrations were much lower thaninland vessels while showing significant differences amongdifferent operating modes

Fuel-based EFs for gaseous species and PM were pre-sented based on the carbon balance method EFs for CO2were 2940ndash3106 3121ndash3160 and 3102ndash3162 g kgminus1 fuel forHH DFH and XYH Because of the combustion conditionssuch as excess air ratio combustion temperature and unifor-mity of the fuelndashair mixture EFs for CO showed high valuesin idling mode but low values in economic speed All theoffshore vessels had higher NOx EFs in low speed than inhigh speed but showed higher values when in acceleration

Atmos Chem Phys 16 6319ndash6334 2016 wwwatmos-chem-physnet1663192016

F Zhang et al Emission factors for gaseous and particulate pollutants 6331

process EFs for SO2 were 16 09 and 26 g kgminus1 fuel forHH DFH and XYH based on sulfur content of the fuels OCand EC were the main components of PM with low OC toEC ratios that were lower than 01 followed by SO2minus

4 NH+4 and NOminus3 Metals such as V Ni Cr Fe As and Cd made upa proportionately small part of the total PM mass

Fuel-based average EFs as well as power-based EFs for thethree different vessels of differing engine power were pre-sented EFs for most gaseous species and PM of HH weremuch higher compared with the other higher-engine-powervessels which was also gt 100 above the IMO standard forNOx Average PM EF of the low-engine-power vessel HHwas also much higher than that in the literatures Howeveraverage EFs for most species of the two larger-engine-powervessels were within the range of previously reported resultsEngine type was inferred as one of the most influence fac-tors for the differences of emission factors Inland and someoffshore ships in China always had higher NOx CO and PMemissions compared with other test vessels in previous stud-ies In addition emission factors that used for calculation ofship inventories in China always had lower values than testvessels

The impact of engine speed on EFs for NOx showed thatwhen the engine speed was close to the rated speed therewould be lower NOx EFs values However combined withthe high fuel consumption rate an optimal engine speedshould be determined based on both EFs and economic costsEmission reduction measures for NOx for some of the off-shore vessels in China are still essential to meet the IMOemission limit

Given the limits of vessel types and numbers this studysubstantially gives the EFs for gaseous species and PM ofthree different diesel-engine-powered offshore vessels How-ever as the development of ports in China emissions fromcargo ships and container ships with large engine power havebecome one of the most significant air pollution sources inport cities and regions Systematical measurement EFs of allkinds of offshore vessels in China are essential in order topresent the accurate emission inventory of ships

Information about the Supplement

Supplementary information includes the details of the real-world measurement system for vessels (Fig S1) the concen-trations of main gaseous matter and PM of shipping emis-sions (Fig S2) the types composition of offshore vessels inChina (Table S1) the distribution of vessels through grosstonnage in 2014 in offshore area of Yangtze River Delta (Ta-ble S2) the Chinese market consumption of marine oil in2011 (Table S3) the detection parameters for gaseous mat-ter (Table S4) the fuel-based EFs for the gaseous pollutants(Table S5) PM and the chemical composition in PM for dif-ferent operating (Table S6) modes and the actual operatingconditions of vessels (Table S7)

The Supplement related to this article is available onlineat doi105194acp-16-6319-2016-supplement

Acknowledgements This study was supported by the CAS Strate-gic Priority Research Program (No XDB05030303) the NaturalScientific Foundation of China (Nos 41273135 and 41473091)and the Shanghai science and Technology Commission project(No 15DZ1205401) The authors would like to express theirgratitude to the vessel owners Haohai Ocean Engineering LimitedLiability Company for vessel Haohai 0007 Ocean University ofChina for vessel Dongfanghong 02 and North China Sea Branchof the State Oceanic Administration for vessel Xiangyanghong08 for making their vessels available for the shipping emissionsstudy The views expressed in this document are solely those of theauthors and the funding agencies do not endorse any products orcommercial services mentioned in this publication

Edited by T Bond

References

Agrawal H Malloy Q G J Welch W A Miller J W andCocker D R In-use gaseous and particulate matter emissionsfrom a modern ocean going container vessel Atmos Environ42 5504ndash5510 doi101016jatmosenv200802053 2008a

Agrawal H Welch W A Miller J W and Cocker D R Emis-sion measurements from a crude oil tanker at sea Environ SciTechnol 42 7098ndash7103 doi101021es703102y 2008b

Almeida S M Pio C A Freitas M C Reis MA and Trancoso M A Source apportionment of fineand coarse particulate matter in a sub-urban area at theWestern European Coast Atmos Environ 39 3127ndash3138doi101016jatmosenv200501048 2005

Anderson M Salo K Hallquist A M and Fridell ECharacterization of particles from a marine engine op-erating at low loads Atmos Environ 101 65ndash71doi101016jatmosenv201411009 2015

Berg N Mellqvist J Jalkanen J-P and Balzani J Ship emis-sions of SO2 and NO2 DOAS measurements from airborne plat-forms Atmos Meas Tech 5 1085ndash1098 doi105194amt-5-1085-2012 2012

California Code of Regulation Titles 13 and 17 avail-able at httpsgovtwestlawcomcalregsBrowseHomeCaliforniaCaliforniaCodeofRegulationstransitionType=DefaultampcontextData=(scDefault) last access 20 April 2016

Cappa C D Williams E J Lack D A Buffaloe G M Coff-man D Hayden K L Herndon S C Lerner B M Li S-M Massoli P McLaren R Nuaaman I Onasch T B andQuinn P K A case study into the measurement of ship emis-sions from plume intercepts of the NOAA ship Miller FreemanAtmos Chem Phys 14 1337ndash1352 doi105194acp-14-1337-2014 2014

Carlton J Danton S Gawen R Lavender K Mathieson NNewell A Reynolds G Webster A Wills C and Wright AMarine exhaust emissions research programme Lloydrsquos RegisterEngineering Services London 63 1995

wwwatmos-chem-physnet1663192016 Atmos Chem Phys 16 6319ndash6334 2016

F Zhang et al Emission factors for gaseous and particulate pollutants 6331

process EFs for SO2 were 16 09 and 26 g kgminus1 fuel forHH DFH and XYH based on sulfur content of the fuels OCand EC were the main components of PM with low OC toEC ratios that were lower than 01 followed by SO2minus

4 NH+4 and NOminus3 Metals such as V Ni Cr Fe As and Cd made upa proportionately small part of the total PM mass

Fuel-based average EFs as well as power-based EFs for thethree different vessels of differing engine power were pre-sented EFs for most gaseous species and PM of HH weremuch higher compared with the other higher-engine-powervessels which was also gt 100 above the IMO standard forNOx Average PM EF of the low-engine-power vessel HHwas also much higher than that in the literatures Howeveraverage EFs for most species of the two larger-engine-powervessels were within the range of previously reported resultsEngine type was inferred as one of the most influence fac-tors for the differences of emission factors Inland and someoffshore ships in China always had higher NOx CO and PMemissions compared with other test vessels in previous stud-ies In addition emission factors that used for calculation ofship inventories in China always had lower values than testvessels

The impact of engine speed on EFs for NOx showed thatwhen the engine speed was close to the rated speed therewould be lower NOx EFs values However combined withthe high fuel consumption rate an optimal engine speedshould be determined based on both EFs and economic costsEmission reduction measures for NOx for some of the off-shore vessels in China are still essential to meet the IMOemission limit

Given the limits of vessel types and numbers this studysubstantially gives the EFs for gaseous species and PM ofthree different diesel-engine-powered offshore vessels How-ever as the development of ports in China emissions fromcargo ships and container ships with large engine power havebecome one of the most significant air pollution sources inport cities and regions Systematical measurement EFs of allkinds of offshore vessels in China are essential in order topresent the accurate emission inventory of ships

Information about the Supplement

Supplementary information includes the details of the real-world measurement system for vessels (Fig S1) the concen-trations of main gaseous matter and PM of shipping emis-sions (Fig S2) the types composition of offshore vessels inChina (Table S1) the distribution of vessels through grosstonnage in 2014 in offshore area of Yangtze River Delta (Ta-ble S2) the Chinese market consumption of marine oil in2011 (Table S3) the detection parameters for gaseous mat-ter (Table S4) the fuel-based EFs for the gaseous pollutants(Table S5) PM and the chemical composition in PM for dif-ferent operating (Table S6) modes and the actual operatingconditions of vessels (Table S7)

The Supplement related to this article is available onlineat doi105194acp-16-6319-2016-supplement

Acknowledgements This study was supported by the CAS Strate-gic Priority Research Program (No XDB05030303) the NaturalScientific Foundation of China (Nos 41273135 and 41473091)and the Shanghai science and Technology Commission project(No 15DZ1205401) The authors would like to express theirgratitude to the vessel owners Haohai Ocean Engineering LimitedLiability Company for vessel Haohai 0007 Ocean University ofChina for vessel Dongfanghong 02 and North China Sea Branchof the State Oceanic Administration for vessel Xiangyanghong08 for making their vessels available for the shipping emissionsstudy The views expressed in this document are solely those of theauthors and the funding agencies do not endorse any products orcommercial services mentioned in this publication

Edited by T Bond

References

Agrawal H Malloy Q G J Welch W A Miller J W andCocker D R In-use gaseous and particulate matter emissionsfrom a modern ocean going container vessel Atmos Environ42 5504ndash5510 doi101016jatmosenv200802053 2008a

Agrawal H Welch W A Miller J W and Cocker D R Emis-sion measurements from a crude oil tanker at sea Environ SciTechnol 42 7098ndash7103 doi101021es703102y 2008b

Almeida S M Pio C A Freitas M C Reis MA and Trancoso M A Source apportionment of fineand coarse particulate matter in a sub-urban area at theWestern European Coast Atmos Environ 39 3127ndash3138doi101016jatmosenv200501048 2005

Anderson M Salo K Hallquist A M and Fridell ECharacterization of particles from a marine engine op-erating at low loads Atmos Environ 101 65ndash71doi101016jatmosenv201411009 2015

Berg N Mellqvist J Jalkanen J-P and Balzani J Ship emis-sions of SO2 and NO2 DOAS measurements from airborne plat-forms Atmos Meas Tech 5 1085ndash1098 doi105194amt-5-1085-2012 2012

California Code of Regulation Titles 13 and 17 avail-able at httpsgovtwestlawcomcalregsBrowseHomeCaliforniaCaliforniaCodeofRegulationstransitionType=DefaultampcontextData=(scDefault) last access 20 April 2016

Cappa C D Williams E J Lack D A Buffaloe G M Coff-man D Hayden K L Herndon S C Lerner B M Li S-M Massoli P McLaren R Nuaaman I Onasch T B andQuinn P K A case study into the measurement of ship emis-sions from plume intercepts of the NOAA ship Miller FreemanAtmos Chem Phys 14 1337ndash1352 doi105194acp-14-1337-2014 2014

Carlton J Danton S Gawen R Lavender K Mathieson NNewell A Reynolds G Webster A Wills C and Wright AMarine exhaust emissions research programme Lloydrsquos RegisterEngineering Services London 63 1995

wwwatmos-chem-physnet1663192016 Atmos Chem Phys 16 6319ndash6334 2016

6332 F Zhang et al Emission factors for gaseous and particulate pollutants

Celo V Dabek-Zlotorzynska E and McCurdy M Chem-ical Characterization of Exhaust Emissions from SelectedCanadian Marine Vessels The Case of Trace Metalsand Lanthanoids Environ Sci Technol 49 5220ndash5226doi101021acsest5b00127 2015

Chen G Huey L G Trainer M Nicks D Corbett J RyersonT Parrish D Neuman J A Nowak J Tanner D HollowayJ Brock C Crawford J Olson J R Sullivan A WeberR Schauffler S Donnelly S Atlas E Roberts J Flocke FHubler G and Fehsenfeld F An investigation of the chemistryof ship emission plumes during ITCT 2002 J Geophys Res-Atmos 110 D10S90 doi1010292004jd005236 2005

Clague A D H Donnet J Wang T K and Peng J C M Acomparison of diesel engine soot with carbon black Carbon 371553ndash1565 doi101016s0008-6223(99)00035-4 1999

Cooper D A Exhaust emissions from ships at berth AtmosEnviron 37 3817ndash3830 doi101016s1352-2310(03)00446-12003

Corbett J J and Fischbeck P S Emissions from waterborne com-merce vessels in United States continental and inland waterwaysEnviron Sci Technol 3254ndash3260 2000

Corbett J J Fischbeck P S and Pandis S N Global nitrogenand sulfur inventories for oceangoing ships J Geophys Res-Atmos 104 3457ndash3470 1999

Corbett J J Winebrake J J Green E H Kasibhatla PEyring V and Lauer A Mortality from ship emissionsA global assessment Environ Sci Technol 41 8512ndash8518doi101021es071686z 2007

Corbett J J Lack D A Winebrake J J Harder S Silber-man J A and Gold M Arctic shipping emissions invento-ries and future scenarios Atmos Chem Phys 10 9689ndash9704doi105194acp-10-9689-2010 2010

The Council of the European Union Council Direc-tive 199932EC of 26 April 1999 Official Journal ofthe European Communities 13ndash18 available at httppublicationseuropaeuenpublication-detail-publicationa093e44c-bf78-434b-92c7-d726dc5ad012language-enformat-PDFsource-5133251 (last access 20 April 2016)1999

Doug W Pounderrsquos marine diesel engines and gas turbines Else-vier Ltd 8th Edn Oxford UK 2004

De Meyer P Maes F and Volckaert A Emissions from in-ternational shipping in the Belgian part of the North Seaand the Belgian seaports Atmos Environ 42 196ndash206doi101016jatmosenv200706059 2008

Derwent R G Stevenson D S Doherty R M CollinsW J Sanderson M G Johnson C E Cofala J Mech-ler R Amann M and Dentener F J The contribu-tion from shipping emissions to air quality and acid de-position in Europe Ambio 34 54ndash59 doi1016390044-7447(2005)034[0054tcfset]20co2 2005

Diesch J-M Drewnick F Klimach T and Borrmann S Investi-gation of gaseous and particulate emissions from various marinevessel types measured on the banks of the Elbe in Northern Ger-many Atmos Chem Phys 13 3603ndash3618 doi105194acp-13-3603-2013 2013

Endresen Oslash Emission from international sea transportationand environmental impact J Geophys Res 108 4560doi1010292002jd002898 2003

European Union DIRECTIVE 201233EU available athttpeur-lexeuropaeuLexUriServLexUriServdouri=OJL201232700010013ENPDF (last access 20 April 2016)2012

Eyring V Kohler H W van Aardenne J and Lauer A Emissionfrom international shipping 1 the last 50 years Geophys Res110 D17305 doi1010292004JD005619 2005

Fridell E Steen E and Peterson K Primary particles in shipemissions Atmos Environ 42 1160ndash1168 2008

Fu M Ding Y Ge Y Yu L Yin H Ye W andLiang B Real-world emissions of inland ships on theGrand Canal China Atmos Environ 81 222ndash229doi101016jatmosenv201308046 2013

Habib H A Basner R Brandenburg R Armbruster U andMartin A Selective Catalytic Reduction of NOx of Ship DieselEngine Exhaust Gas with C3H6 over CuY Zeolite ACS Cataly-sis 4 2479ndash2491 2014

Haglind F A review on the use of gas and steam turbine com-bined cycles as prime movers for large ships Part III Fu-els and emissions Energ Convers Manage 49 3476ndash3482doi101016jenconman200808003 2008

IMO International Maritime Organization Regulations for the pre-vention of air pollution from ships and NOx technical code An-nex VI of the MARPOL convention 7378 London 1998

IMO available at httpwwwimoorgenOurWorkEnvironmentPollutionPreventionAirPollutionPagesGreenhouse-Gas-Studies-2014aspx (last access 20 April 2016)2014

Juwono A M Johnson G Mazaheri M Morawska L RouxF and Kitchen B Investigation of the airborne submicrome-ter particles emitted by dredging vessels using a plume capturemethod Atmos Environ 73 112ndash123 2013

Khan B Hays M D Geron C and Jetter J Differences in theOCEC Ratios that Characterize Ambient and Source Aerosolsdue to Thermal-Optical Analysis Aerosol Sci Tech 46 127ndash137 doi101080027868262011609194 2012

Khan M Y Ranganathan S Agrawal H Welch W ALaroo C Miller J W and Cocker III D R Mea-suring in-use ship emissions with international and USfederal methods J Air Waste Manage 63 284ndash291doi101080109622472012744370 2013

Lack D A and Corbett J J Black carbon from ships a review ofthe effects of ship speed fuel quality and exhaust gas scrubbingAtmos Chem Phys 12 3985ndash4000 doi105194acp-12-3985-2012 2012

Lack D A Corbett J J Onasch T Lerner B Massoli PQuinn P K Bates T S Covert D S Coffman D Sierau BHerndon S Allan J Baynard T Lovejoy E RavishankaraA R and Williams E Particulate emissions from commercialshipping Chemical physical and optical properties J GeophysRes-Atmos 114 D00f04 doi1010292008jd011300 2009

Lack D A Cappa C D Langridge J Bahreini R Buffaloe GBrock C Cerully K Coffman D Hayden K Holloway JLerner B Massoli P Li S-M McLaren R Middlebrook AM Moore R Nenes A Nuaaman I Onasch T B PeischlJ Perring A Quinn P K Ryerson T Schwartz J P Spack-man R Wofsy S C Worsnop D Xiang B and WilliamsE Impact of Fuel Quality Regulation and Speed Reductions onShipping Emissions Implications for Climate and Air Quality

Atmos Chem Phys 16 6319ndash6334 2016 wwwatmos-chem-physnet1663192016

F Zhang et al Emission factors for gaseous and particulate pollutants 6333

Environ Sci Technol 45 9052ndash9060 doi101021es20134242011

Lauer A Eyring V Hendricks J Joumlckel P and Lohmann UGlobal model simulations of the impact of ocean-going ships onaerosols clouds and the radiation budget Atmos Chem Phys7 5061ndash5079 doi105194acp-7-5061-2007 2007

Li X Xu Z Guan C and Huang Z Effect of injection timingon particle size distribution from a diesel engine Fuel 134 189ndash195 2014

Marmer E and Langmann B Impact of ship emissionson the Mediterranean summertime pollution and climateA regional model study Atmos Environ 39 4659ndash4669doi101016jatmosenv200504014 2005

Ministry of Transportation Report on China Shipping Developmentof 2010 Part III Port Service China Communications Press2010

Ministry of Transportation China Statistical Yearbook of 2013 PartXVI Transport Post and Telecommunication Services ChinaCommunications Press 2013

Ministry of Transport of the Peoplersquos Republic of ChinaImplementation plan of shipping emission control areasof the Pearl River Delta Yangtze River Delta and Bo-hai Sea Region (BeijingndashTianjinndashHebei) waters availableat httpwwwmotgovcnxiazaizhongxinziliaoxiazai201512P020151231359179618188doc (last access 20 April 2016)2015

Moldanova J Fridell E Popovicheva O Demirdjian BTishkova V Faccinetto A and Focsa C Characterisa-tion of particulate matter and gaseous emissions from alarge ship diesel engine Atmos Environ 43 2632ndash2641doi101016jatmosenv200902008 2009

Moldanovaacute J Fridell E Winnes H Holmin-Fridell S BomanJ Jedynska A Tishkova V Demirdjian B Joulie S BladtH Ivleva N P and Niessner R Physical and chemical char-acterisation of PM emissions from two ships operating in Eu-ropean Emission Control Areas Atmos Meas Tech 6 3577ndash3596 doi105194amt-6-3577-2013 2013

Mueller L Jakobi G Czech H Stengel B Orasche J Arteaga-Salas J M Karg E Elsasser M Sippula O StreibelT Slowik J G Prevot A S H Jokiniemi J Rabe RHarndorf H Michalke B Schnelle-Kreis J and Zimmer-mann R Characteristics and temporal evolution of particu-late emissions from a ship diesel engine ApEn 155 204ndash217doi101016japenergy201505115 2015

Murphy S M Agrawal H Sorooshian A Padroacute L T Gates HHersey S Welch W Jung H Miller J and Cocker III D RComprehensive simultaneous shipboard and airborne characteri-zation of exhaust from a modern container ship at sea EnvironSci Technol 43 4626ndash4640 2009

Ni J Principles of Automotive Internal Combustion EngineShanghai Tongji University Press 1999 (in Chinese)

Petzold A Hasselbach J Lauer P Baumann R Franke KGurk C Schlager H and Weingartner E Experimental stud-ies on particle emissions from cruising ship their characteris-tic properties transformation and atmospheric lifetime in themarine boundary layer Atmos Chem Phys 8 2387ndash2403doi105194acp-8-2387-2008 2008

Petzold A Weingartner E Hasselbach I Lauer P Kurok Cand Fleischer F Physical Properties Chemical Composition

and Cloud Forming Potential of Particulate Emissions from aMarine Diesel Engine at Various Load Conditions Environ SciTechnol 44 3800ndash3805 doi101021es903681z 2010

Pirjola L Pajunoja A Walden J Jalkanen J-P Roumlnkkouml TKousa A and Koskentalo T Mobile measurements of shipemissions in two harbour areas in Finland Atmos Meas Tech7 149ndash161 doi105194amt-7-149-2014 2014

Poplawski K Setton E McEwen B Hrebenyk D Gra-ham M and Keller P Impact of cruise ship emissionsin Victoria BC Canada Atmos Environ 45 824ndash833doi101016jatmosenv201011029 2011

Reda A A Schnelle-Kreis J Orasche J Abbaszade G Lintel-mann J Arteaga-Salas J M Stengel B Rabe R HarndorfH Sippula O Streibel T and Zimmermann R Gas phase car-bonyl compounds in ship emissions Differences between dieselfuel and heavy fuel oil operation (vol 94 pg 467 2014) AtmosEnviron 112 369ndash380 doi101016jatmosenv2015030582015

Righi M Klinger C Eyring V Hendricks J Lauer A and Pet-zold A Climate Impact of Biofuels in Shipping Global ModelStudies of the Aerosol Indirect Effect Environ Sci Technol 453519ndash3525 doi101021es1036157 2011

Saxe H and Larsen T Air pollution from ships in three Danishports Atmos Environ 38 4057ndash4067 2004

Schreier M Kokhanovsky A A Eyring V Bugliaro LMannstein H Mayer B Bovensmann H and Burrows J PImpact of ship emissions on the microphysical optical and ra-diative properties of marine stratus a case study Atmos ChemPhys 6 4925ndash4942 doi105194acp-6-4925-2006 2006

Shah S D Cocker D R Miller J W and Norbeck J MEmission Rates of Particulate Matter and Elemental and OrganicCarbon from In-Use Diesel Engines Environ Sci Technol 382544ndash2550 doi101021es0350583 2004

Shahsavani A Naddafi K Haghighifard N J MesdaghiniaA Yunesian M Nabizadeh R Arhami M Yarahmadi MSowlat M H Ghani M Jafari A J Alimohamadi M Mote-valian S A and Soleimani Z Characterization of ionic com-position of TSP and PM10 during the Middle Eastern Dust(MED) storms in Ahvaz Iran Environ Monit Assess 1846683ndash6692 doi101007s10661-011-2451-6 2012

Sinha P Hobbs P V Yokelson R J Christian T J KirchstetterT W and Bruintjes R Emissions of trace gases and particlesfrom two ships in the southern Atlantic Ocean Atmos Environ37 2139ndash2148 doi101016s1352-2310(03)00080-3 2003

Sippula O Stengel B Sklorz M Streibel T Rabe R OrascheJ Lintelmann J Michalke B Abbaszade G Radischat CGroeger T Schnelle-Kreis J Harndorf H and ZimmermannR Particle Emissions from a Marine Engine Chemical Com-position and Aromatic Emission Profiles under Various Op-erating Conditions Environ Sci Technol 48 11721ndash11729doi101021es502484z 2014

Song Y Research of Emission Inventory and Emission Characterof Inland and Offshore Ships Beijing Institute of Technology2015 (in Chinese)

Starcrest Consulting Group Developing Port Clean Air Pro-grams available at httptheicctorgsitesdefaultfilesICCT_SCG_Developing-Clean-Air-Programs_June2012pdf (last ac-cess 20 April 2016) 2012

wwwatmos-chem-physnet1663192016 Atmos Chem Phys 16 6319ndash6334 2016

6334 F Zhang et al Emission factors for gaseous and particulate pollutants

Tan P Feng Q Hu Z and Lou D Emissions From a VehicleDiesel Engine During Transient OPerating Conditions J EngThermophys 33 1819ndash1822 2012 (in Chinese)

Tzannatos E Ship emissions and their externali-ties for Greece Atmos Environ 44 2194ndash2202doi101016jatmosenv201003018 2010

US Environmental Protection Agency Current Methodologiesin Preparing Mobile Source Port-Related Emission Invento-ries Final Report April 2009 available at httparchiveepagovsectorswebpdfports-emission-inv-april09pdf (last access20 April 2016) 2009

Verschaeren R Schaepdryver W Serruys T Bastiaen M Ver-vaeke L and Verhelst S Experimental study of NOx reductionon a medium speed heavy duty diesel engine by the applicationof EGR (exhaust gas recirculation) and Miller timing Energy76 614ndash621 doi101016jenergy201408059 2014

Viana M Amato F Alastuey A Querol X Moreno TGarcia Dos Santos S Dolores Herce M and Fernandez-Patier R Chemical Tracers of Particulate Emissions fromCommercial Shipping Environ Sci Technol 43 7472-7477doi101021es901558t 2009

Viana M Hammingh P Colette A Querol X Degraeuwe Bde Vlieger I and van Aardenne J Impact of maritime transportemissions on coastal air quality in Europe Atmos Environ 9096ndash105 doi101016jatmosenv201403046 2014

Wang X Bi X Sheng G and Fu H Hospital in-door PM10PM25 and associated trace elements inGuangzhou China Sci Total Environ 366 124ndash135doi101016jscitotenv200509004 2006

Williams E J Lerner B M Murphy P C Herndon SC and Zahniser M S Emissions of NOx SO2 COand HCHO from commercial marine shipping during TexasAir Quality Study (TexAQS) 2006 J Geophys Res 114doi1010292009jd012094 2009

Winebrake J J Corbett J J Green E H Lauer A and EyringV Mitigating the Health Impacts of Pollution from Oceango-ing Shipping An Assessment of Low-Sulfur Fuel MandatesEnviron Sci Technol 43 4776ndash4782 doi101021es803224q2009

Winnes H and Fridell E Emissions of NOX and particlesfrom manoeuvring ships Transport Res D-Tr E 15 204ndash211doi101016jtrd201002003 2010

Xu J Study on emission of high-speed marine diesel engineWuhan University of Technology 2008

Yang J Yin P-L Ye S-Q Wang S-S Zheng J-Y and OuJ-M Marine Emission Inventory and Its Temporal and SpatialCharacteristics in the City of Shenzhen Environ Sci 36 1217ndash1226 2015 (in Chinese)

Zhang F Chen Y Tian C Wang X Huang G Fang Y andZong Z Identification and quantification of shipping emissionsin Bohai Rim China Sci Total Environ 497ndash498 570ndash577doi101016jscitotenv201408016 2014

Zhao M Zhang Y Ma W Fu Q Yang X Li C Zhou B YuQ and Chen L Characteristics and ship traffic source identifi-cation of air pollutants in Chinarsquos largest port Atmos Environ64 277ndash286 doi101016jatmosenv201210007 2013

Zheng J Zhang L Che W Zheng Z and Yin S Ahighly resolved temporal and spatial air pollutant emis-sion inventory for the Pearl River Delta region China andits uncertainty assessment Atmos Environ 43 5112ndash5122doi101016jatmosenv200904060 2009

Zheng M Cheng Y Zeng L and Zhang Y Developing chem-ical signatures of particulate air pollution in the Pearl RiverDelta region China J Environ Sci-China 23 1143ndash1149doi101016s1001-0742(10)60526-8 2011

Zhu H Sinopec marine fuel oil business development planreaearch Beijing University of Chemical Technology 2013

Zhu Q Liao C Liu J and Zhang Y Research on the air pollu-tant emission in the typical river ports and harbors GuangdongJournal of Safety and Environment 15 191ndash195 2015 (in Chi-nese)

Atmos Chem Phys 16 6319ndash6334 2016 wwwatmos-chem-physnet1663192016