Impact of sampling conditions and procedure on particulate matter emissions from a marine diesel engine

Leonidas Ntziachristos, Erkka Saukko, Topi Rönkkö, Kati Lehtoranta, Hilkka Timonen, Risto Hillamo, Jorma Keskinen

    Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsProfessional


    Diesel engines are the most popular means of propulsion for ships due to their high fuel efficiency, robust operation, and long lifetime. Strict emission limits for NOx are being enforced for marine diesel engines as a function of their rated speed, in the framework of global efforts to reduce air pollution. Sulfur oxides emissions are also controlled by means of maximum sulfur limits in the fuel, in particular in sensitive areas for acidification around the world. Interestingly, PM emission limits from marine diesel engines are not in place yet, despite the recognized contribution of vessels to ambient PM concentrations. This is for various reasons, including the lack of established protocols to reliably determine PM emissions when high sulfur and residual fuels are being used. In this study we measured gaseous and PM emissions from a 1.6 MW medium speed turbocharged four-stroke marine diesel engine (Wärtsilä 4R32), installed on an engine test bed. The sampling protocol followed the guidelines of ISO 8178 and the engine was tested in two of the protocol's modes, at rated speed (750 rpm) and 25% load and rated speed and 75% load. The exhaust flow was split in two parts and PM samples were collected from a portion of the exhaust kept at constant temperature, following a primary dilution using an AVL Smart Sampler. This device allows for the dilution ratio to be varied at will, within a given range, while PM filter temperature remains within specific range (<52°C). In our tests, the dilution ratio was adjusted in three consecutive settings, nominally 5:1, 10:1, and 30:1 to check the impact of sampling conditions on the PM collected. PM filters were analyzed for sulfates, bound water and organic material. Tests were repeated with two fuels, a Heavy Fuel Oil (HFO) with a density of 1005 kg/m3 and ∼1% wt. sulfur content and a Light Fuel Oil (LFO) with a density of 820 kg/m3 and <20 ppm wt. sulfur content. The results showed that the same engine may produce PM emissions which vary by more than one order of magnitude at the same load by just changing fuel and sampling conditions. To put it in perspective, PM emissions per unit of energy produced (g/kWh) at the same load point varied from levels equivalent to early Euro I on-road diesel engines to levels as low as Euro IV engines. In general, PM emissions decreased when sampling at high dilution ratios and when shifting from HFO to LFO. The sensitivity of PM levels increased with decreasing dilution ratio therefore, in principle, higher dilution ratios lead to more repeatable PM results. Analysis of the filters showed that the majority of the PM collected consists of organic and inorganic volatile and semi-volatile material. Organic carbon is the most abundant species for LFO derived PM samples. Under low dilution ratio conditions, sulfates and associated water actually dominate the PM mass when HFO is used. In order to better understand the impact of fuels and sampling conditions on emissions, airborne particles were chemically analyzed using a Soot Particle Aerosol Mass Spectrometer (Aerodyne Research Inc.) that provides the speciation of particles distinguished in nitrate, ammonium and sulfate salts as well as various chemical groups for the organic load. This revealed the high contribution of inorganic, non-combustion produced, species with the HFO compared to the LFO fuel and confirmed that the majority of particles consists of volatile, mostly fuel derived components. Preliminary tests were also conducted utilizing a chamber that simulates the photo-oxidation of species to understand how the large volatile load of marine PM is transformed once emitted to the atmosphere. Measurements revealed that PM fast becomes significantly oxidized with a gain in total mass as more oxygen is added to the mixture. The results of this study show that control of PM emissions from marine engines will first require more strict control of the sampling conditions currently enforced by the ISO 8178 protocol, at least in terms of the dilution ratio range allowed. Also, if PM were to be included in the certification data of production engines, then fuel properties for the tests will have to be strictly controlled. Although this may be difficult to achieve in practice, current specifications for HFO certification fuels only determine maximum limits (for sulfur, ash content, density, etc.) and significant emission variance is expected using different fuels within the same specifications. In general, further work is required to determine which part of the emissions is an engine product and how much depends on fuel residues.
    Original languageEnglish
    Title of host publicationCIMAC Technical Paper Database
    Publication statusPublished - 2016
    MoE publication typeD3 Professional conference proceedings
    Event28th CIMAC World Congress - Helsinki, Finland
    Duration: 6 Jun 201610 Jun 2016


    Conference28th CIMAC World Congress


    Dive into the research topics of 'Impact of sampling conditions and procedure on particulate matter emissions from a marine diesel engine'. Together they form a unique fingerprint.

    Cite this