pre- and after-treatment techniques for diesel engines in ... · the goal of this part is to...

CREATING//TR/WP 06/M06.03/ECN/03-08-05/version 3.0/ of 31

SIXTH FRAMEWORK PROGRAMME PRIORITY [1.6.2]

Sustainable Surface Transport

Pre- and after-treatment techniques for diesel engines in inland navigation

Technical report in the framework of EU project CREATING (M06.03, task II)

Project acronym: CREATING Project full title: Concepts to Reduce Environmental impact and Attain optimal

Transport performance by Inland Navigation Contract no.: FP6 - 506542


Document Title: Pre- and after-treatment techniques for diesel engines in inland navigation Technical report in the framework of EU project CREATING (M06.03, task II

(S)WP number: M06.03 task II

Document number: CREATING//TR/WP 06/M06.03/ECN/03-08-05/version 3.0/?

Document History

Version Comments Date Authorised by

1.0 draft June 24th 2.0 report 13-07-2005 3.0 ready for publication 03-08-2005 Classification Public Number of pages:

30

Number of annexes:

none

Responsible Organisation: ECN Contributing Organisation(s): CBRB

Principal Author(s): G.L.M.A. van Rens H.P.J. de Wilde Contributing Author(s):

WP leader Name: WP leader Organisation:

Petra Seiwerth, Alexander Kampfer via donau


Table of Contents 1. Abstract ..................................................................................................................................................... 8 2. Introduction ............................................................................................................................................. 10 3. Diesel oxidation catalysts ........................................................................................................................ 11

3.1. Introduction ..................................................................................................................................... 11 3.2. Working principle............................................................................................................................ 11

3.2.1. Carbon monoxide and hydrocarbon removal .......................................................................... 11 3.2.2. PM removal ............................................................................................................................. 11

3.3. Requirements ................................................................................................................................... 12 3.4. State of development ....................................................................................................................... 12 3.5. Cost-benefits.................................................................................................................................... 12

4. Wet scrubbers .......................................................................................................................................... 13 4.1. Introduction ..................................................................................................................................... 13 4.2. Working principle............................................................................................................................ 13

4.2.1. SOx removal ............................................................................................................................ 13 4.2.2. NOx removal ........................................................................................................................... 13 4.2.3. PM removal ............................................................................................................................. 14 4.2.4. Hydrocarbon removal .............................................................................................................. 14


5. Diesel particulate filters........................................................................................................................... 16 5.1. Introduction ..................................................................................................................................... 16 5.2. Filter types ....................................................................................................................................... 16

5.2.1. Wall-flow monolith ................................................................................................................. 16 5.2.2. Ceramic foam .......................................................................................................................... 16 5.2.3. Ceramic fibre trap.................................................................................................................... 17 5.2.4. Sintered metal filters................................................................................................................ 17

5.3. Regeneration.................................................................................................................................... 17 5.3.1. Introduction ............................................................................................................................. 17 5.3.2. Engine throttling ...................................................................................................................... 17 5.3.3. Fuel injection in the exhaust gas.............................................................................................. 17 5.3.4. Catalytic regeneration.............................................................................................................. 17 5.3.5. CRTTM ..................................................................................................................................... 18 5.3.6. Electrically heated ................................................................................................................... 18 5.3.7. Microwave irradiation ............................................................................................................. 18 5.3.8. Pressurised air.......................................................................................................................... 19


6. Selective catalytic reduction.................................................................................................................... 20 6.1. Introduction ..................................................................................................................................... 20 6.2. Working principle............................................................................................................................ 20 6.3. Requirements ................................................................................................................................... 20 6.4. State of development ....................................................................................................................... 21 6.5. Cost-benefits.................................................................................................................................... 21

7. Humidification of engine inlet air ........................................................................................................... 22 7.1. Introduction ..................................................................................................................................... 22 7.2. Working principle............................................................................................................................ 22

7.2.1. Humid air motor ...................................................................................................................... 22 7.2.2. CASS ....................................................................................................................................... 22


7.2.3. Swirl-flash ............................................................................................................................... 22 7.3. Requirements ................................................................................................................................... 22 7.4. State of development ....................................................................................................................... 22 7.5. Cost-benefits.................................................................................................................................... 22

8. ESP .......................................................................................................................................................... 23 8.1. Introduction ..................................................................................................................................... 23 8.2. Working principle............................................................................................................................ 23 8.3. Requirements ................................................................................................................................... 23 8.4. State of development ....................................................................................................................... 23 8.5. Cost-benefits.................................................................................................................................... 23

9. Emerging technologies ............................................................................................................................ 24 9.1. NOx adsorber................................................................................................................................... 24 9.2. Steam Jet Aerosol Collector ............................................................................................................ 24 9.3. Non-thermal plasma ........................................................................................................................ 24

10. Financial considerations ...................................................................................................................... 25 10.1. Introduction ................................................................................................................................. 25 10.2. NOx removal ............................................................................................................................... 26 10.3. SOx removal ................................................................................................................................ 26 10.4. PM removal ................................................................................................................................. 27

11. References ........................................................................................................................................... 28

Pre-and after-treatment techniques


List of Abbreviations CASS: Combustion Air Saturation System CO: Carbon monoxide CRTTM: Continuously Regenerating Technology, formerly known as, Continuously Regenerating Trap DI: Direct Injection ESP: ElectroStatic Precipitation HC: Hydrocarbons NOx: Nitrogen Oxides PM: Particulate matter SCR: Selective Catalytic Reduction SJAC: Steam Jet Aerosol Collector SOx: Sulphur Oxides VOC: Volatile Organic Compound Nm3/h: Flow m3/h under normal conditions (i.e. 298K and 1.013 bar) T: 1000 kg



1. ABSTRACT With emission legislation for trucks getting more and more stringent it can only be expected that emission limits for shipping will be tightened as well. Less emission from shipping can be achieved by better hull-shapes or new propulsion systems, as addressed in other work packages of CREATING. But at this moment tail pipe emissions are legislated, and engine measures or after-treatment techniques need to be used to reduce those emissions. In trucking industry large reductions in pollutant emissions have been achieved by engine modifications at the cost of a not-achieved decrease in fuel consumption. This will be treated in more detail in “improvement techniques” from sub work package M06.01 Task VI. The goal of this report is to describe pre- and after-treatment techniques, which are already used in industry, diesel powered trucks or ships. Requirements by these specific techniques and cost aspects are treated. Three promising techniques, which are still under development, are discussed as well. Techniques that are treated in detail in this report are, in order of appearance, diesel oxidation catalysts, wet scrubbers, diesel particulate filters, selective catalytic reduction, humidification of engine inlet air and electrostatic precipitation. For those readers, who are unfamiliar with the composition and dangers of emissions it is recommended to read M06.01 Task III “Environmental and health aspects” as certain background knowledge, especially regarding the composition of particulate matter is assumed. This report will form a basis for M06.03 Task III Retrofit, where, amongst other, the presented techniques will be screened for suitability for retrofit. This report is also a report on itself, which can be used to draw conclusions for the suitability to fit these techniques behind a new engine. Diesel oxidation catalysts are proven on cars and trucks. They require a fuel with sulphur content lower than present 0.2 or future 0.1% sulphur. They reduce CO and HC emissions significantly; the efficiency for PM removal depends on particle composition and fuel sulphur. Wet scrubbers are produced for sea shipping. They reduce SOx emissions significantly and are claimed to remove PM by 80%. A lot of different types of diesel particulate filters are on the market. The development is stimulated by the need for particulate removal devices for the automobile industry. It can reduce particulate matter significantly. The challenge in this type of filter is not so much capturing particles as well as cleaning the filter. Therefore several so-called regeneration techniques are developed, see chapter 5.3. They require pressure drop measurements over the particulate filter, as well as a control system. It seems likely that it is necessary to use a catalyst in the trap or added to the fuel. If this is the case low sulphur fuel is required. ESP is proven as an after-treatment for power plants. Wärtsilä did some experiments with an ESP to remove particulate matter from a stationary diesel engine. The main drawback is the large volume that is required to install an ESP. Selective catalytic reduction is a proven technology on inland ships, which is able to remove NOx by over 80%. Required is a reducing agent, i.e. ammonia or urea. Costs for the reducing agent could be reduced if an infrastructure is present. The biggest problem with SCR is the limited temperature window in which it can be used. Inlet air humidification is another way to reduce NOx emissions. Although it is less efficient than selective catalytic reduction, it does not require a special additive. Three emerging technologies, which are not yet manufactured on large scale, are NOx adsorber, steam jet aerosol collector and non-thermal plasma. Comparison of costs between the several techniques showed that the cheapest technique per year is generally not the cheapest technique if the cost per avoided emission is concerned. The cost comparison is based on landmark prices from manufacturers, who were so kind to share this information. The cheapest way to avoid sulphur oxide emissions is to use a fuel with lower sulphur content. This is also the cheapest way to reduce particulate matter emissions, although it is strongly recommended to use an oxidation catalyst in this case, as costs barely increase, and avoided PM emissions rise strongly. The cheapest techniques in a five-year write-off period per kg avoided particulate matter are an oxidation catalyst and a particulate trap. The latter reduces most particulate matter. Swirl-flash is the cheapest NOx abatement technique per year, although selective catalytic reduction is the cheapest NOx reduction technique per kg avoided NOx.



2. INTRODUCTION Now that transport by trucks is becoming increasingly environmentally friendly from Nitrogen Oxides (NOx) and Particulate Matter (PM) point of view, the shipping industry is forced to reduce its emissions. Gains from better hull-shapes or new propulsion systems, as addressed in other work packages of CREATING, will lead to a win-win situation for both environment and ship owner in the form of reduced costs. Emission legislation however is expressed in g/kWh engine power, which means that with tightening emission limits engines need to be adjusted in such a way that they produce less NOx and PM. Large reductions in pollutant emissions have been achieved by engine modifications at the cost of a not-achieved decrease in fuel consumption. At some point however it will become more cost-efficient to reduce these emissions by after-treatment techniques instead of in-engine modifications although some in-engine modifications will still be necessary. For truck manufacturers this transition point is Euro IV or V. The goal of this part is to describe pre- and after-treatment techniques, which are already used in industry, diesel powered trucks or ships, and to give the reader an insight in the requirements and cost aspects of these techniques. Three promising techniques, which are still under development, are discussed in a chapter called emerging technologies. That chapter mainly focuses on the strong points and the requirements of the techniques. In order to ease the comparison between the various techniques the chapters have the same structure. Each technique is first shortly introduced. This is followed by an explanation how the technique reduces the emissions. Requirements for implementation of this technique or other constraints, like low-sulphur fuel, additional fuel consumption or additional infrastructure are treated in the paragraph called requirements. The paragraph state-of development treats the state-of-development of the technique and, if reported, experiences considering durability. Each chapter ends with financial considerations regarding this emission reduction technology. Those findings are repeated in a conclusive paragraph on cost-benefits. Please note that the prices mentioned in this part are landmark prices only, which are susceptible to changes. For those readers, who are unfamiliar with the composition and dangers of emissions it is recommended to read M06.01 Task III “Environmental and health aspects”, as certain background knowledge, especially regarding the composition of particulate matter is assumed. This report is a report on itself, which can be used to draw conclusions for the suitability to fit these techniques behind a new engine. It will form a basis for M06.03 Task III “Retrofit”, where, amongst other, the presented techniques will be screened for suitability for retrofit. This report is also a report on itself, which can be used to draw conclusions for the suitability to fit these techniques behind a new engine.



3. DIESEL OXIDATION CATALYSTS 3.1. Introduction

Diesel oxidation catalysts are very effective in reducing the emissions of carbon monoxide (CO) and gaseous hydrocarbons (HC). The removal efficiency depends on catalyst composition, but up to 90% is possible for CO and HC [Clean air systems, 2005]. The effect of oxidation catalysts on particulate matter emissions is somewhat more complicated, however. This strongly depends on fuel and catalyst composition as will be explained below.

3.2. Working principle

3.2.1. Carbon monoxide and hydrocarbon removal The removal of carbon monoxide and hydrocarbons is treated in the same section, because the removal mechanism is the same for both emission reduction technologies. Carbon monoxide and hydrocarbons are oxidised by the excess oxygen in the exhaust gas. This reaction is catalysed by precious metals that are present in the catalyst, like platinum and palladium. The precious metals are coated on a substrate material, either metallic or ceramic. The influence of a catalyst on the soluble organic fraction, which consists of hydrocarbons, but also contributes to particulate matter emissions, is treated in the PM removal section. The oxidation of carbon monoxide and hydrocarbons is catalysed from a certain threshold temperature, called the light-off temperature of the catalyst. In fact several light-off temperatures exist; one for carbon monoxide and one for every different hydrocarbon species. Light-off temperature depends on catalyst formulation but roughly lies between 150 and 200°C [Eastwood, 2000], [Stein et al., 1995]. During a part of the driving cycle the exhaust gas temperature may lie beneath the light-off temperature, especially after a cold-start.

3.2.2. PM removal An oxidation catalyst may affect every component of particulate matter besides the ash. Due to the honeycomb structure, which is generally used for oxidation catalysts, the catalyst traps no particles. This means that oxidation catalysts have no risk of clogging, ensuring trouble-free operation. The downside is that the residence time of the particle is not long enough to oxidise the carbonaceous part of the particle by large amounts. It is in fact not entirely clear if a decrease in the carbonaceous part is achieved, which has everything to do with the measurement methods [Eastwood, 2000], yet a 5-15% reduction in the carbonaceous part seems realistic. The soluble organic fraction (SOF), which contains carcinogenic material like Poly-Aromatic Hydrocarbons (PAH), is removed from the exhaust gas in a different way than gaseous hydrocarbons. Four different stages can be identified, when the catalyst is warming-up [Eastwood, 2000]. First a storage period can be observed, where hydrocarbons adsorb or condense onto the catalyst surface. Secondly a temperature regime in which these adsorbed or condensed hydrocarbons are released is present. In the third regime the catalyst has finished releasing the hydrocarbons, but the light-off temperature has not yet been reached. The fourth regime is the actual oxidation regime, where the catalyst temperature exceeds the light-off temperature. These four phases do not need to occur consecutively. They might also occur simultaneously. In fact as SOF consists of several hydrocarbons every hydrocarbon might be in a different regime at the same temperature. The main problem of particulate removal by means of an oxidation catalyst is the oxidation of sulphur dioxide to sulphur trioxide, which in its turn forms sulphates, that contribute to particle mass. Sulphur oxide emissions consist approximately for 5% of SO3 and 95% of SO2. The oxidation of SO2 to SO3 is more favourable at higher temperatures. The formed sulphates with the associated water add significantly to particle mass. In fact an upper threshold temperature is present that represents the temperature at which the emitted particle mass equals the original particle mass again. Above this threshold temperature the oxidation catalyst creates additional particle mass. The key to a well functioning oxidation catalyst for PM removal is to find the catalyst composition that has a threshold temperature that is as high as possible. The sulphur level of the fuel influences this threshold temperature. The higher the fuel sulphur level the lower the threshold temperature.



3.3. Requirements Besides the influence of the sulphur level in fuel on the particulate matter emissions, fuel sulphur can poison a catalyst. Most oxidation catalysts are quite sulphur tolerant. It is claimed that fuel with more than 500 ppm sulphur will not damage the catalyst coating or the ceramic substrate [Clean air systems, 2005]. It will however not reduce the particulate matter emission significantly. For a comparison the presently used inland navigation fuel contains approximately 2000 ppm. If particulate matter reduction is one of the reasons to use an oxidation catalyst it is necessary to use a fuel with a lower sulphur level than presently required for inland shipping. For optimum performance the exhaust gas needs to be in the right temperature window.

3.4. State of development Diesel oxidation catalysts are used commercially in cars and some trucks. The performance of oxidation catalysts does decrease during time. Deterioration of each catalyst formulation will be different. An example from [Stein et al, 1995] gives a deterioration of 8% for PM, 13% for CO and 21% of HC after 2200 hours, meaning that the removal efficiency after 2200 hours is 92% of the original removal efficiency for PM etc. Oxidation catalysts that replace mufflers are designed in a way that they do not require any additional volume.

3.5. Cost-benefits The price of an oxidation catalyst is rather modest, and it is especially in combination with low-sulphur fuel a cheap and trustworthy solution to reduce particulate matter emissions. Investment costs for a diesel oxidation catalyst are below 1500 US Dollar. [EPA, 2004]



4. WET SCRUBBERS 4.1. Introduction

Wet scrubbers are capable of removing SOx, NOx, hydrocarbon and particulate matter emissions. Which emissions are reduced and how efficiently they are removed is strongly dependent on the configuration of the scrubber and the scrubbing liquid used. Wet scrubbers are widely used in the power industry, especially for the reduction of the emission of sulphur oxides. Usually water is used as the scrubbing liquid. An example of the development of a scrubber that uses a hydrocarbon scrubbing liquid is shown in the paragraph called hydrocarbon removal by means of a scrubber. More than 90% of sulphur oxides can be removed by means of a wet scrubber with water as a scrubbing liquid [Davies], [Haase&Koehne, 1999], [Trivett et al., 1999]. Approximately 15% of NOx emissions and 15-27% of particulate number emissions, with a potential up to 60%, are removed with pilot-scale set-ups [Haase&Koehne, 1999], [Van Rens, 2004]. The device presented in [Trivett et al., 1999] has been commercialised by Marine Exhaust Solutions under the trade name EcoSilencer [Lantz, 2005]. Up to 90% SOx, 7% NOx and up to 80% PM removal is claimed with this technique [MES, 2005a], with a sustainable SOx-removal level between 72 and 80% [MES, 2005b].


4.2.1. SOx removal As mentioned above SOx removal efficiencies of over 90% are possible by a suitable type of wet scrubber. In an experimental study [Ives&Klokk, 1993] showed a removal efficiency of SOx from the exhaust gas of a marine diesel engine using heavy fuel oil of 71-73%. The main reaction involved is the reaction of SO2 with water after it is absorbed in the water: SO2+H2O↔H2SO3 The reaction product is in equilibrium with hydrogen sulphite and sulphite. Sulphite might react to sulphate by means of a redox reaction [Haase&Koehne, 1999]. [Götmalm, 1991] however says that a desulphurisation plant on every ship is hardly desirable from an environmental point of view, referring to the use of energy and natural resources to manufacture those devices. This is even more true when considering that [Götmalm, 1991] made his comment looking at sulphur oxide removal from engines using heavy fuel oil, which emit more sulphur oxides because of the higher sulphur content in the fuel in comparison with diesel fuel oil used in inland shipping. The effluent of the scrubber can be quite acidic; [Ives&Klokk, 1993] measured a pH of 2.7. Therefore it might be necessary to neutralise the effluent before discharge of the scrubbing water.

4.2.2. NOx removal NO2 can easily be absorbed in water [Davies]. [Haase&Koehne, 1999] show the reactions involved

2 NO2↔N2O4 3N2O4+2H2O↔4HNO3+2NO

When oxygen is present NO can oxidize to NO2: 2NO+O2↔2NO2

leading to the following overall reaction: 2N2O4+2H2O+O2↔4HNO3 Other reaction mechanisms are known to lead to the formation of NH3 en N2 [Davies]. The oxidation reaction of NO to NO2 is rather slow and governs the reaction velocity [Haase&Koehne, 1999]. An experimental investigation of [Haase&Koehne, 1999] showed that more than 15% removal of NO2 by means of a wet scrubber is not possible at “a moderate apparative expense”. Unfortunately over 90% of the NOx in exhaust gas is in the form of NO [Niven, 1993]. One of the options to enlarge the concentration of NO2 and to accelerate the above mentioned oxidation reaction is to put an oxidizing agent in the scrubbing water, which favours the oxidation of NO to NO2. This can either be KMnO4, O3, ClO2, H2O2 or MgO2. KMnO4 should be used only in closed systems [Haase&Koehne, 1999]. O3 and ClO2 are hard to handle. The latter two are more suitable for normal use. MgO2 is actually an indirect source of H2O2 because it reacts with water [Haase&Koehne, 1999]:



MgO2+H2O↔MgO+H2O2 Which means that MgO2 also neutralizes the water. H2O2 is slowly consumed during standstill. Because the reaction between MgO2 and water is slow, this problem does not occur with MgO2. Up to 16% of NO-absorption is shown when using H2O2 as an oxidizing agent [Haase&Koehne, 1999]. Another solution is to use so-called complexing agents (EDTA or NTA) that absorb NO directly. They are most commonly used in the form of Fe(II)EDTA [Davies]. To reduce NO subsequently it is necessary that SO3

2- is present in the scrubbing liquid. Luckily this is a reaction product of SO2 with water. The simplified main reaction mechanism is: 2NO+2SO3

2-↔N2+2SO42-

Where Fe(II)EDTA acts as a catalyst to dissolve NO. This system also needs to take place in a closed system, because EDTA can dissolve metal compounds in riverbeds, and is not biodecomposable [Haase&Koehne, 1999]. It is also possible to oxidize NO to NO2 before the wet scrubber by means of an oxidation catalyst. In itself this does not sound very attractive, because the removal efficiency of NO2 for a wet scrubber is still not high. However with this combination it is possible to bypass one of the problems of the oxidation catalyst, i.e. the formation of SO3 that will form sulphates when the exhaust gas cools down. The formed SO3 can easily be removed in the wet scrubber, especially if it is still in the gas phase, and oxidation selectivity is no longer an important issue for the choice of oxidation catalyst. It also facilitates the use of oxidation catalysts for heavy fuel oil, without increase in mass of particulate matter.

4.2.3. PM removal Wet scrubbers also remove particulate matter. This is sometimes referred to as a problem, because it might be necessary to treat the scrubbing liquid additionally, however it can also be seen as an opportunity to reduce the emission of particulate matter as well. Recall that particulate matter consists of a core of carbon and ash, in the form of metal oxides, with hydrocarbons and sulphates with water condensed on the particle. However this is the composition at 52°C. When entering the scrubber the associated sulphates might still be gaseous in the form of SO3 and the same goes for hydrocarbons. In this way a first quick win comes from the removal of SO3. However scrubbers can be optimised to remove the solid particles as well. Most suitable scrubber for this goal is a venturi scrubber. Unfortunately a pressure drop of 15 kPa is necessary to remove the small particles effectively. This pressure drop is higher than the pressure drop for which engines are designed. A high pressure drop also results in a higher fuel consumption. [Trivett et al., 1999] designed a scrubber that had a particulate matter removal efficiency of 70-80% for particles larger than 1 µm in diameter, which is only a fraction of particulate matter number emissions. [Van Rens, 2004] showed 15-27% particulate removal, measured with a smoke meter for a pilot plant wet scrubber behind a diesel engine. Scaled up and extrapolated, it is estimated that 60% of particulate emissions can be reduced.

4.2.4. Hydrocarbon removal A wet scrubber will always absorb some hydrocarbons when the exhaust gas is cooled below the dew point temperature of the hydrocarbons. This will result however in a mixture of water and hydrocarbons that requires additional wastewater treatment. To avoid a mixture of water and hydrocarbons, a scrubber with a hydrocarbon scrubbing liquid is being developed at the Energy Research Centre of the Netherlands to remove tar from the product gas of a biomass gasification plant. The device, called OLGA (Dutch acronym for oil-based gas washer), is a packed column scrubber, which uses an (undisclosed) oily substance as the scrubbing liquid. Hot gas above the tar dew point temperature enters OLGA. In OLGA the temperature decreases to just above the dew point temperature of water. Some of the tars will condense and will be collected in this fashion, while the volatile tar compounds are absorbed in the scrubbing liquid. A reduction of tar dew point to 4°C has been demonstrated [Boerrigter et al., 2005]. The present design of OLGA has a separate stage for liquid tar collection and absorption of volatile tars. The organic scrubbing liquid needs to be recycled, to limit the usage of scrubbing liquid. A special separator separates the liquid heavy tars from the scrubbing liquid, whilst the volatile hydrocarbons are removed by stripping the scrubbing liquid with hot inlet air for the engine. It might be possible to mix the bleed stream from the separator with diesel fuel, and combust it in the diesel engine. Special points of concern are the usage of scrubbing liquid and slip of the scrubbing liquid. From operational point of view it



would be ideal if diesel oil could be used as the scrubbing liquid. At this point there are no plans to develop the OLGA for maritime use. OLGA also removes particles, although it is not the goal of the apparatus. An additional particle separation step to clean the scrubbing liquid is necessary for high particle loadings. OLGA does not remove SOx and NOx, but it is possible to use a wet scrubber after OLGA, without having problems with condensing hydrocarbons.

4.3. Requirements Although a scrubber has a significant pressure drop and as a result gives a higher backpressure on the engine, pressure drops do not exceed engine manufacturers demands, if carefully looked after. However a small increase in pressure drop, even if it is within the limits of an engine manufacturer may lead to an increase in fuel consumption. No special requirements for the fuel are necessary, although for fuels with high sulphur content, it might be necessary to add a base to the scrubbing water in order to maintain an acceptable acidity. One of the problems of a scrubber is the possible necessity for a treatment of the scrubbing water. In that case recycling of the scrubbing water might be preferable. Marine exhaust solutions Inc. shows that treatment is possible to within the EPA water quality limits [MES, 2005b]. Approximately 0,6T soot sludge per week is produced during on-board use with 4800 kW installed electrical power. It was handled together with the regular onshore waste disposal. A problem with wet scrubbers might be that they are rather voluminous, especially when accounted for the additional space for wastewater treatment, but Marine exhaust solutions show for a 200-250 kW engine a package of approximately 2 m3 [MES, 2005c]. SOx removal will not be the major reason to implement a wet scrubber for inland shipping, due to the sulphur level in the used fuel. Particulate matter removal will be the governing reason to implement wet scrubbers in inland shipping.

4.4. State of development Wet scrubbing is frequently used in the power generation industry. Demonstrations of wet scrubbers on ships on pilot-scale are shown in for example [Trivett et al., 1999], [Ives and Klokk, 1993]. Marine Exhaust Solutions is able to deliver an entire system, which includes treatment of the scrubbing liquid, with the trade name EcoSilencer. It also acts as a muffler, as indicated by the name, reducing noise levels by 35dB. The first full-scale system has been build on a RoRo-ferry, running on a fuel with 2,5 % sulphur. A trial report [MES, 2005b] shows that wet scrubbers are reliable and durable. Regular maintenance consisted of boroscopic investigation of the silencer, level measuring probe cleaning and water circulation pump maintenance. The main problem from durability point of view seems to be the wastewater treatment, and more specifically the wastewater circulation pump, rather than the EcoSilencer itself.

4.5. Cost-benefits The trial system costs were 200 US Dollar per kW for 4x1200 kW [MES, 2005b]. Production costs are expected to be lower, but please note that one device cleaned the wastewater of all four EcoSilencers, thereby reducing costs of the trial system. As a rough estimate 200 Euro per kW will be used for a 1 MW engine on an inland shipping vessel.



5. DIESEL PARTICULATE FILTERS 5.1. Introduction

A diesel particulate filter (DPF), also referred to as a particulate trap, is a rather simple device. Particles are trapped in the device, so that almost no particles leave the filter. Over 80% of particulate matter mass can be removed this way, depending on filter type. Whether or not a particulate trap removes the soluble organic fraction of particulate matter depends on the exhaust gas temperature and the type of regeneration used, which will be treated in the next section. In principal a particulate trap does not reduce the emission of hydrocarbons that are still volatile at the temperature of the trap. The trapped particles will result in an increasing backpressure on the engine. As the backpressure influences fuel consumption, a penalty in the form of larger fuel consumption can be expected. Unacceptably high levels of backpressure might be obtained quite quickly. Therefore it is necessary to remove the trapped particles periodically or continuously by a so-called regeneration technique. The available techniques are presented below, preceded by different varieties of traps.

5.2. Filter types

5.2.1. Wall-flow monolith Wall-flow monoliths consist of channels with porous walls that are alternately plugged, as is illustrated in figure 1. The exhaust gas is forced to stream through these porous walls leaving the particles behind at the wall.

exhaust gas in

exhaust gas out

porouswall

plug

Figure 1: Schematic representation wall-flow monolith

This results in the formation of a particle cake that increases the backpressure on the engine. An important factor in the resulting backpressure on the engine is the pressure drop over the particle cake. Walls with small pore size achieve a better structure of the particle cake [Eastwood, 2000]. This leads to the somewhat unexpected conclusion that a high initial pressure drop over the wall-flow monolith leads to a lower pressure drop, when it has a higher particle loading. The removal efficiency of a wall-flow monolith is high, 90% is achievable [Davies], with efficiency depending on particle loading.

5.2.2. Ceramic foam A ceramic foam filter is a porous fixed bed with macroscopic holes, where gas is forced through. Particles are removed by diffusion, impaction and interception [Van Gulijk, 2002]. Therefore particulate matter removal is less efficient than a wall-flow monolith trap, with a removal efficiency ranging from 50 to 80% [Davies], [Karila et al, 2004]. Ceramic foams are generally made of aluminium oxide or silicon carbide [Karila et al., 2004]. Due to the fact that particulate deposition is less concentrated on single spots, less risk



of excessive temperatures during trap regeneration is present, compared to wall flow monoliths [Van Gulijk, 2002].

5.2.3. Ceramic fibre trap A ceramic fibre trap consists of a perforated metal tube, closed at one end, that acts as a support for layers of ceramic fibre that are wrapped around the tube. Gas generally enters the open side of the metal tube and is forced to stream through this filter bed. Approximately 70 to 80% of the particles are removed [Davies], [Karila et al., 2004]. Fibres are relatively thick with diameters of 10-12 mm [Karila et al., 2004]. Every fibre is composed of approximately 400 filaments with a diameter of 11µm [Davies].

5.2.4. Sintered metal filters Sintered metal filters consist of metal fibres that are baked together at a high temperature. Generally plates are made of them, although it is very well possible to manufacture them in different shapes. According to [Karila et al., 2004] it is a surface-type filter, meaning that particles are retained at the surface of the filter. A manufacturer of sintered metal filters claims that it is a deep-bed type filter [Bekaert, 2005], meaning that particles are trapped inside the filter. This discussion is mainly important for the storage capacity of the filter, and thus time between regenerations and lifetime. A deep-bed filter has a larger storage capacity than a surface-type filter. Due to the high porosity of 85% of the filter from [Bekeart, 2005] the claim seems plausible.

5.3. Regeneration

5.3.1. Introduction Typical temperatures for (non-catalysed) particulate matter combustion are 550 to 600°C [Eastwood, 2000], [Davies]. This temperature is almost never reached during typical driving cycles. Therefore it is necessary to apply a regeneration strategy. A large number of regeneration strategies are possible. Active and passive regeneration can be discerned. A particulate trap with active regeneration has the pressure drop over the particulate filter monitored continuously. When a certain threshold pressure drop is measured, regeneration will start. Active regeneration methods are engine throttling, fuel injection in the exhaust gas, electrically heated regeneration, regeneration by microwave irradiation and regeneration by pressurised air. Passive regeneration methods make use of a catalyst or oxidation by NO2. Regeneration by non-thermal plasma is also possible [Karila et al., 2004]. Non-thermal plasma can also be used without a particulate trap and is capable to remove amongst others, NOx, particulate matter and hydrocarbons. It is treated in Chapter 9 as an emerging technology.

5.3.2. Engine throttling One way to obtain regeneration is by increasing the exhaust gas temperature by periodically letting the engine run hotter. The moments for regeneration are determined by a ∆p-sensor that gives a signal to the engine management. With engines with flexible fuel injection timing this is achieved by post-injection of fuel [Karila et al., 2004]. A significant increase in fuel consumption occurs with post-injection.

5.3.3. Fuel injection in the exhaust gas Another means of increasing the exhaust gas temperature is by adding a fuel burner in the exhaust pipe, or injecting fuel in the exhaust gas, which is oxidised at a catalyst upstream of the trap, thereby releasing heat. Caterpillar showcased this last option on a 2007-prototype [Dieselnet, 2005a]. Filter temperature is monitored to minimise fuel consumption during regeneration.

5.3.4. Catalytic regeneration Catalyst-enhanced regeneration can be achieved by incorporating a catalyst in the trap material or by adding a catalyst in the fuel. The advantage of adding a catalyst in fuel is that it not only aids the regeneration of the trap; it will also reduce engine-out emissions, because it will catalyse soot oxidation in the engine as well. This will lead to less frequent regeneration, which saves energy. The drawbacks of adding a catalyst in the fuel are a constant need for an additive together with a small loss of catalyst, which will be emitted in air, and an additional ash build-up in the trap. The catalyst reduces the temperature at which soot will start to combust. The combustion of soot may now occur with an exhaust gas temperature, which is achieved during normal driving. If only short distances are



driven, meaning that the exhaust gas stays cool, a large build-up of particulate matter in the filter is unavoidable. This may result in engine damage if the backpressure becomes too high. If the combustion temperature of the soot is reached with high particle loading it might lead to an uncontrolled combustion, with temperatures as high as 1000°C. This might seriously damage the particulate trap. Therefore it is wise to control this with an active regeneration technique such as engine throttling. In fact this is the strategy PSA (Peugeot and Citroën) uses in their cars [Karila et al., 2005], [PSA, 2005]. Besides the effect the catalyst has on particulate matter emissions it also reduces the emissions of volatile organic compound and carbon monoxide

5.3.5. CRTTM CRT stands for continuously regenerating technology (it used to mean continuously regenerating trap) and is a registered trademark of Johnson Matthey. Instead of oxygen, nitrogen dioxide is used to oxidise soot, reducing nitrogen dioxide to nitrogen monoxide. Nitrogen dioxide is normally only a small portion of the nitrogen oxides emission. Therefore a very active platinum based catalyst is applied before the trap, which oxidises, besides carbon monoxide and volatile organic compound, nitrogen oxide to nitrogen dioxide [JM, 2005a]. Hereby the soot oxidation temperature is lowered to approximately 250°C, enabling continuous regeneration. Engines that will operate at low loads for long periods of time might need an additional means of active regeneration [Karila et al., 2004]. Low-sulphur fuel (<50 ppm sulphur) is required in order to convert enough NO to NO2 in the catalyst for soot oxidation [JM, 2005b], [Dieselnet, 2005b]. The very active catalyst will increase particle emissions for sulphur levels higher than approximately 150 ppm over the ESC cycle by converting sulphur dioxide into sulphates [Dieselnet, 2005b].

Figure 2: Drawing of a CRTTM filter [JM, 2005a]

The minimum NOx to PM ratio by weight is reported as 20:1-25:1 in recent literature [Dieselnet, 2005b]. This means that NOx reduction cannot take place upstream of the diesel particulate filter, although Johnson Matthey recently designed a combination of EGR with a CRT, called EGRT, which reduces NOx by 40% and PM by 90% [JM, 2005c]. For engines with low exhaust gas temperatures or low NOx/PM-ratio Johnson Matthey advises a Catalysed CRT, CCRTTM, where the trap itself is catalysed as well [JM, 2005b].

5.3.6. Electrically heated Particulate traps that are electrically conducting can easily be heated by applying an electrical current on them, when regeneration is required [Bekaert, 2005]. Electrically conducting traps are metal filters or filters from silicon carbide. Accumulated soot burns due to the elevated filter temperature. Exhaust gas temperature should be monitored (together with pressure drop over the particulate trap) in order to prevent overheating of the trap. It is possible to apply heating elements on the upstream face of ceramic wall-flow monoliths as well, but it relies on propagation of soot combustion through the trap [Davies]. Energy consumptions can be significant, with reported fuel efficiency penalties of 3 to 5% [Davies].

5.3.7. Microwave irradiation Another way to remove particles capture by a particle trap is by means of microwave irradiation. Microwave irradiation is directly absorbed by the particle, thereby heating the particle only, when ceramic filter materials are used. Metals reflect microwaves and can this way be used as wave-guides for the microwaves [Karila et al., 2005].



A complication in microwave regeneration is the non-homogenous and therefore not complete combustion of the particles [Karila et al., 2005]. To overcome this problem special coatings are being developed that convert microwave energy to heat energy at the surface of the filter, thereby creating a more uniform temperature distribution [Elzinga et al., 2004]. Care should be taken not to exceed electromagnetic radiation limits.

5.3.8. Pressurised air The most basic approach is cleaning the filter with pressurised air. Pulses of pressurised air dislodge the trapped particles. The pulses are directed opposite to the exhaust gas flow. The big advantage of this system is that it can be operated at cool exhaust gas temperatures, meaning that it is possible to trap a part of the soluble organic fraction and that all components of particulate matter are removed from the trap. Problem with pressurised air regeneration is that air seeks the way of least resistance thereby cleaning only a part of the trap surface. The dislodged particulate matter needs to be separated in some way from the pressurised air and exhaust gas stream. One strategy is to bypass the particle filter temporarily and let the particles settle in a “dead-flow” system. Another strategy is to burn the particles on an electrical heater, which requires less energy due to the fact that the entire trap does not have to be heated.

5.4. Requirements The requirements when using a particle trap differ for each regeneration technology, but every active technique requires a control, which generally consists of a delta p sensor and a temperature sensor. Only pressurised air regeneration does not need a temperature sensor. The only passive technology for which it seems feasible to operate without the aid of an active regeneration technology is the CRT-filter. This filter however needs a low-sulphur fuel. This is compensated by the fact that other regeneration technologies require additional fuel consumption, which equals approximately 1-2% of the fuel consumption.

5.5. State of development Most publications and most experience are achieved with the particulate trap system of PSA and the CRTTM. In 2000 PSA gradually started to equip its diesel-powered cars with particle filter systems. Service-intervals for the filters increased from 80.000 km to 120.000 km for newer models up to 2004. Over 1 million cars equipped with this system have been produced. Dutch and German governments use financial incentives to support implementation of particulate traps on cars. The CRT system is often used in retrofit programs for buses in the United States of America and Canada. It is becoming more interesting with the availability of low-sulphur fuel. Other systems are commercially available as well, mainly used for retrofitting heavy-duty diesel engines.

5.6. Cost-benefits According to EPA the price for catalytic diesel particulate filter is expressed by the following formulae [EPA, 2004]: Table 1: Cost of CDPF in 2002 US Dollars [EPA, 2004] particulate filter regeneration (DI) near-term 146x+75 10x+147 long-term 112x+57 8x+111 with x the engine displacement in litres, and the cost of regeneration equipment specified for direct injected diesel engines. Due to the large depreciation of the US Dollar in comparison with the Euro over the last three years, an average conversion factor of 1 is used to convert Dollars to Euros. Maintenance needs to take place every 4500 hours, which costs 260 dollar. The costs for maintenance do not incorporate loss of availability.



6. SELECTIVE CATALYTIC REDUCTION 6.1. Introduction

Selective catalytic reduction (SCR) is a technique to remove NOx emissions by means of injecting a reducing agent. Two different types of SCR exist. One type uses ammonia to reduce nitrogen monoxide and nitrogen dioxide to nitrogen and water. It is either injected as ammonia or as urea. Urea is favourable, because it can be handled more easily. The second type uses a hydrocarbon to reduce NOx emissions. It is most convenient to use diesel fuel as the source for hydrocarbons, as it is already on board, but other hydrocarbons are possible. Whereas SCR on ammonia basis can reduce up to 95% NOx emissions, NOx reduction by hydrocarbon injection is limited to approximately 40% [Eastwood, 2000]. Due to the limited reduction achieved by injecting hydrocarbons, only the reduction of NOx by ammonia/urea injection is treated in this chapter.

6.2. Working principle The reactions with injection of ammonia are [Niven, 1993]:

4 NO+4 NH3+O2 → 4 N2+6 H2O 6 NO2+8 NH3 → 7 N2+ 12 H2O

Urea combined with water will react to: CO(NH2)2 + H2O → CO2 +2 NH3

This reaction only takes place above 275°C [Götmalm, 1991]. This is not a significant objection because the temperature window of the entire system is narrower. Reduced catalyst activity due to the formation of sulphates is observed at 320°C (depending on fuel sulphur content). These sulphates are capable of plugging the channels in the catalyst. The catalyst activity also deteriorates with overheating above 400 to 450°C [Sher, 1998]. Argillon claims a temperature window between 250 and 520°C [Argillon, 2005]. A problem of a SCR is the release of ammonia in the air, known as ammonia slip. The higher the desired removal efficiency of nitrogen oxides the higher the resulting ammonia slip will be, as illustrated in figure 3.

Figure 3: Ammonia (NH3) slip as function of the percentage NOx removed as obtained from [Sher, 1998]

Ammonia is a strong greenhouse gas and more environmental unfriendly and more detrimental to health than NOx. A trade-off between ammonia and nitrogen oxides emissions needs to be established. Application of an oxidation catalyst after the SCR can reduce ammonia emissions together with emissions of carbon monoxide and volatile organic compound and is applied in most commercially available systems.

6.3. Requirements A SCR that includes an oxidation catalyst is approximately as big as a silencer (+/-20% [Argillon, 2005]) and weighs 30 to 60% more than a silencer [Argillon, 2005]. As mentioned urea or ammonia are necessary for successful operation. Several urea-producing companies made a trademark, called AdBlue to introduce a standard urea-solution for automotive purposes. It contains approximately 32,5% (between 31,8-33,3%) urea



per weight in demineralised water [Volvo, 2005a], [Dureal, 2005]. In shipping business another urea-solution is used, containing 40% urea. 15-20 l/h urea-solution is consumed per MW engine power [Argillon, 2005]. An infrastructure in order to provide ships with the urea-solution is necessary. The amount of injected urea should be carefully controlled, in order to avoid excessive ammonia slip. This could be done by engine mapping or by measuring NOx emissions on-line. No special fuel is required.

6.4. State of development SCR is demonstrated on ships on sea and for inland shipping [Eschenbacher, 2005]. Expected lifetime ranges from 10000 hours to 40000 hours [Argillon, 2005], which is probably highly dependant on fuel sulphur content. Measurements on a SCR system on an inland shipping vessel showed that 85% NOx reduction is possible for all loads [De Wilde and Kos, 2004]. SCR is going to be implemented on some truck engines for compliance with NOx emission levels for Euro IV and V [Volvo, 2005b], [Scania, 2005]. SCR is also used for retrofit purposes.

6.5. Cost-benefits As a landmark costs for an SCR-installation for retrofit range between 20 and 65 euro/kW. This depends on the available space, and thereby complexity of the system, and installed power. Costs may drop significantly when several identical ships are equipped with this system [Eschenbacher, 2005]. Installation costs are estimated as the cost of the SCR installation. The used urea solution with this technique is a 40% solution, which costs approximately 300 to 400 euro per metric ton when an infrastructure is not present. Prices will drop when an infrastructure is present [Schoonebeek, 2005]. This means that the costs of urea consumption ranges between 4.5 and 8 €/MWh.



7. HUMIDIFICATION OF ENGINE INLET AIR 7.1. Introduction

Several ways to humidify the engine inlet air are available. The most promising systems have one thing in common; they cool the compressed air by adding water to the inlet air. Cooling the compressed air will result in more power, less fuel consumption and less NOx emissions. That is why compressed air cooling (a.k.a. charge air cooling) is already applied in many modern engines. Normally the air is cooled by a cooler and not by the injection of water. The injection of water leads to combustion air that contains more water vapour. This will result in lower peak temperatures, thereby reducing NOx-formation. Three techniques will be distinguished. The humid air motor, swirl-flash and CASS.


7.2.1. Humid air motor The humid air motor brings warm water in contact with the compressed inlet air, often called charge air, in a humidification tower. Water evaporates thereby increasing air humidity to near saturation and decreasing charge air temperature. Water is heated by the engine cooling system. Additional work is not required. Approximately 50% of NOx emissions are reduced over the ECE-R49 cycle [Nord et al., 2000]. The source of water can be seawater that is led through an intake strainer [Björsell, 2005].

7.2.2. CASS Wärtsilä developed a system called CASS (i.e. Combustion Air Saturation System). Water, which is not warmed, is injected under high pressure (approximately 130 bar) in the charged inlet air, forming a very fine mist. Water vaporises due to the heat in the charge air. Saturated air is achieved. Additional charge air cooling is possible. A separate stage, called WMC, removes the remaining water droplets from the inlet air. Efficiencies up to 50% are achieved. Water consumption is one to two times the fuel consumption [Dogliani & Paro, 2004]. Due to the high pressure during injection it seems probable that this system requires demineralised water, this is however never stated.

7.2.3. Swirl-flash With the swirl-flash technique pre-heated water (150-250°C) is injected under high pressure in the exhaust gas. In the charge air the experienced pressure is lower, which results in the sudden vaporisation of the water, called flashing. The temperature of the charge air provides the required energy for the phase change. Injection of warm water will lead to a decrease of the charge air temperature. The water is pre-heated by cooling the exhaust gas. NOx-reduction up to 20% is demonstrated on an inland shipping vessel [De Wilde & Kos, 2004]. It seems that in this set-up the inlet air was not entirely saturated. Otherwise NOx-reduction efficiencies in the order of 50% were to be expected.

7.3. Requirements Swirl-flash requires demineralised water, whereas this is not a necessity for the concept of the humid air motor. Swirl-flash also needs to cool the exhaust gas to obtain energy to warm the water. Swirl-flash and CASS also need some sort of control to know how much water needs to be injected.

7.4. State of development All three technologies have been proved on board of a ship. Humid air motor has the longest experience on board of a ship, followed by CASS and swirl-flash. It seems that there is room for further improvement of the latter system in comparison to the demonstration

7.5. Cost-benefits Investment costs for the swirl-flash system were approximately 20000 euro for the demonstration project on a 550 kW engine [Van Liere, 2005]. This excluded a device that creates demineralised water. An additional 10000 to 30000 euro is used as a first rough estimate for this device, resulting in 30000 to 50000 euro investment costs, or 55 to 91 €/kW. The humid air motor has an investment and installation cost of approximately 150-250 €/kW [Björsell, 2005]. As engine manufacturers need to do a special engine design, these costs could be drastically reduced when engine manufactures design their new engines with the use of this technique in their mind.



8. ESP 8.1. Introduction

Electrostatic precipitation (ESP) is often used to clean the exhaust gas of power plants efficiently from fine particles. It is also used to reduce particle emissions from a large stationary diesel engine. With unlimited size, over 99% particle removal is achievable. Therefore it seems a possible solution to reduce the particle emissions of diesel engines on ships by ESP.

8.2. Working principle A high voltage creates a corona discharge, which creates a lot of ions. Particles are charged by these ions. The applied voltage can be negative as well as positive. The overall-charge of the gas remains neutral. The charged particles are subsequently attracted towards the electrode of opposite polarity. If an ESP is operated dry the particles will form a cake, which has to be dislodged periodically. This is done by hammers or by ultrasound. Another option is to operate the ESP with additionally supplied water. The big advantage is that the electrodes are cleaned continuously. This will also lead to lower power consumption, because the resistance of the particle cake is absent.

8.3. Requirements The most important requirement for an ESP is the required space. Commercially available ESP’s are rather large. A 1 MW engine produces approximately 8100 kg/h exhaust gas, assuming a typical 8.1 kg/kWh. This equals approximately 6200 Nm3/h. An example for a suitable ESP, in this case a wet ESP, has the following dimensions 7.62*1.80*1.80 metres [CGS, 2005]. Data from [Van Paasen et al., 2004] show that an upper-boundary for power consumption is given by equalling the volume flow in m3/h to the required energy in Watts. This leads to the conclusion that the required energy is rather modest, with less than 1% of engine power. It is required to dispose the collected dust in a safe and environmentally acceptable way. No special fuel is required.

8.4. State of development ESP is demonstrated behind a stationary diesel engine, where size is not important. It is frequently used in power plants. No publications of demonstration on board of a ship were found, but a removal efficiency of 90% seems reasonable.

8.5. Cost-benefits The investment costs for an ESP is not so sensitive to the gas flow. An ESP for an engine with 6200 m3/h costs approximately €160.000. Installation costs and start-up are 25% of the investment costs [Van Paasen et al., 2004]. This means an initial cost of 200 €/kW installed engine power.



9. EMERGING TECHNOLOGIES 9.1. NOx adsorber

A NOx adsorber adsorbs NOx and stores it as nitrates. Often a catalyst is added before the trap to oxidise NO to NO2 in order to speed-up the storage of NOx. The problem with NOx adsorbers is that materials that store nitrates are very susceptible to sulphation, meaning that sulphates decrease NOx-adsorption efficiency. The NOx is released from the adsorber under rich conditions, which can be achieved by for example late-ignition in the engine or fuel injection in the exhaust, and is subsequently oxidised by a (catalysed) reaction with carbon monoxide [Bosteels and Searles, 2002]. The system is in production to remove NOx from exhaust gas of some direct injection gasoline engines and some light-duty diesel engines [Karila et al, 2004]. The performance of the NOx adsorber, however, deteriorates severely with fuel sulphur levels as low as 15 ppm [Bardasz et al., 2004]. Before application of this technique on ships the sulphation problem should be satisfactorily solved.

9.2. Steam Jet Aerosol Collector The Steam Jet Aerosol Collector (SJAC)-technology is a novel technology, which is being developed at the Energy research Centre of the Netherlands (ECN). The technology facilitates physical removal of particulate matter by applying a water coating on them, thus enlarging the particle size. Tests show that 57% of the number of particles can be removed on a part stream of a marine diesel engine under favourable conditions. Another demonstration shows up to 50% mass-reduction with the full exhaust gas stream of a 15 kW generator-set running on automotive fuel. The technique is however still under development and higher removal rates are to be expected. The principle of the SJAC is the injection of steam in cooled exhaust gas. By the injection of steam the exhaust gas becomes supersaturated and condensation will occur. Because kernels stimulate condensation, the particles will grow and physical removal will be possible. Particulate matter and heavier hydrocarbons will be removed this way. Theoretically one might also expect removal of SOx-emissions. Treatment of the resulting wastewater will however be necessary. This is not expected to pose serious difficulties, because the wastewater quality will not differ much from the wastewater with wet scrubbing. Space requirement will depend mainly on residence time and the chosen droplet removal device. No special fuel will be required.

9.3. Non-thermal plasma Non-thermal plasma is a novel technique that is able to reduce both particulate matter and NOx-emissions. Non-thermal plasma is a gas that has been ionised into a mixture of highly reactive molecules. Plasma is created by high-voltage discharges. The generated electric field generates and produces free electrons, which travel through the gas creating O and OH-radicals. They are very effective in oxidising exhaust gas emissions, thereby reducing soot-emissions. In some applications NOx is reduced by adding hydrocarbons into the exhaust gas, rather similar to selective catalytic reduction with hydrocarbons. NO is oxidised to NO2 by the oxygen-radicals. The injected hydrocarbons are converted to hydrocarbon radicals, which react with NO2, reducing nitrogen dioxide to nitrogen. It is theoretically possible that harmful substances are created by non-thermal plasma. Non-thermal plasma is in one application combined with a diesel particulate filter with the trademark Electrocat. In that application it is solely used to start the regeneration of the particulate filter. 90% particulate matter removal was shown with successful regeneration [McAdams et al., 2003]. A pilot-scale set-up showed NOx reduction potential of 30 to 40%, whereas a set-up with a particle filter in front of the non-thermal plasma showed approximately 50-70% removal efficiency. Power consumption was assumed to be 5% of engine power [McAdams et al., 2003].



10. FINANCIAL CONSIDERATIONS 10.1. Introduction

This chapter will give a comparison of costs based on data from manufacturers and supply companies. The absolute decrease in emissions will be calculated by assuming that the emissions of harmful substances equal the legislation issued by the Central commission for navigation on the Rhine (CCR), shown in table x. This assumption will lead for ships sailing on the river Rhine to an overestimation of the amount of reduced emissions, because engines that are compliant with this legislation will not be tuned to the maximum of every emission limit. The emission values stem from measurements over ISO 8178 mode E3 [CBRB, 2005]. Table 2: Emission limits issued by CCR [CCR, 2003]

Nominal power (kW)

n (rev/min) CO (g/kWh) HC (g/kWh) NOx (g/kWh) Particulate Matter (g/kWh)

37-75 All 6.5 1.3 9.2 0.85 75-130 All 5.0 1.3 9.2 0.70 >130 500-2800 5.0 1.3 45*n-0.2 0.54 ≥ 2800 5.0 1.3 9.2 0.54

For main propulsion normally medium and high-speed diesel engines are used, being engines with a nominal power higher than 130 kW and an engine speed between 500 and 2800 revolutions per minute. In this range the maximum NOx emission in accordance with the legislation issued by the CCR is related to the rated engine speed. In order to specify costs by additional fuel consumption an assumption of the actual fuel consumption will be needed, as most increases in fuel consumptions due to the use of an after-treatment device are specified relative to the fuel consumption without this device. Therefore an engine has been chosen that is compliant with CCR emission levels for which the necessary engine data is available. As an example for these calculations Volvo Penta’s marine diesel type D49A MT with rating 1 is used. It has a rated power of 940 kW at 1600 rpm and is compliant with the CCR emission levels. Specific fuel consumption for this engine at rated power is 206 g/kWh [Volvo, 2004]. From the data of the manufacturer it is possible to determine the specific fuel consumption over ISO 8178 mode E3, which is approximately 208 g/kWh. The emissions of sulphur oxides can be determined from fuel consumption by assuming maximum fuel sulphur level of 0.2% by weight and that 5 vol-% of the sulphur oxides is emitted in the form of sulphur trioxide, while the rest is emitted in the form of sulphur dioxides. This leads to a total emission of SOx over ISO 8178 mode E3 of 0.84 g/kWh. In further analysis it will be assumed that the cycle used to calculate emissions of the engines is representative for normal ship operation, meaning that in order to get an emission per year per kW the data is simply multiplied by the hours in a year the ship is operating. The emission values used in the calculation in this chapter are summarised in table 3. Table 3: emission values used for cost calculations Emission type Emission value [g/kWh] NOx 10.29 CO 5.0 HC 1.3 PM 0.54 SOx 0.84 bsfc 208 The fuel price will play an important role in cost determination. The product price for gasoil at July 7th 2005 in the Netherlands for more than 4000 litres has been used for this calculation, as shown in table 4 [Warner, 2005]. The price of gasoil with 50 ppm is based on the additional fuel cost of low-sulphur gas oil in comparison to gas oil from [Oliecentrale, 2005]. The price per kg has been calculated with the average density from product information of a supplier [Shell, 2002], [Shell, 2003].



Table 4: fuel prices July 7th 2005 [Warner, 2005] Type of fuel €/100 litre €/kg Gasoil 0.2% sulphur 51.23 0.610 Gasoil 50 ppm sulphur 52.83 0.635 For the following calculations it is assumed that a ships sails approximately 33% of the hours in a year with a load profile according to ISO 8178 mode E3, meaning that the ship sails on 68.75% of full power for a third of a year. This means that each kW installed engine power generates 1990 kWh in a year. All investments are paid off in 5 years with 7% interest. The rest value is assumed to be equal to zero.

10.2. NOx removal NOx emissions can be substantially removed by pre- and after-treatment techniques, like SCR, humid air motor and swirl-flash. For SCR a reduction rate of 80% will be assumed, for the humid air motor a 50% reduction and for swirl-flash a reduction rate of 30% Table 5: Financial aspects NOx reduction investment cost

(€/kW) annual cost of operation (€/kW)

annual cost (investment and operation) (€/kW)

€/kg NOx

SCR 40-130 9-16 19-48 1.2-2.9 Humid air motor 150-250 0 37-61 3.6-6.0 Swirl-flash 55-91 0* 13-22 2.1-3.6 *additional power consumption for the fresh water maker has been neglected Note that swirl-flash has the lowest annual cost. SCR is the cheapest per kg NOx reduced. Humid air motor seems rather expensive in this comparison, but this is due to the fact that it has not been proven on the relative small scale of inland shipping and custom designs need to be made. It is expected that prices will drop, when production on a larger scale is possible.

10.3. SOx removal SOx emissions can be significantly lowered by using a low sulphur fuel or by using a wet scrubber. A change to low sulphur fuel with a sulphur content of 0.1% by weight is scheduled and it will reduce SOx emissions with 50%. A decrease in fuel sulphur level to 15 ppm will lead to a subsequent reduction of 98.5%, which is a 99.25% decrease in SOx emissions from the situation with 0.2% sulphur in fuel. Fuel with 50 ppm sulphur, which is already available, results in a 97.5% reduction. For this cost consideration the shift to 0.1% sulphur in fuel is not taken into account. A shift from a fuel with 0.2% sulphur to a fuel with 50 ppm sulphur results in additional costs of 0.0052 €/kWh, which equals €6.35/kg SOx not emitted. The EcoSilencer from marine exhaust solutions decreases the emission by 80%. The investment cost is 200 euro per kW, which equals €49 euro per kW installed engine power each year. Additional fuel consumption is estimated to equal 1% of installed engine power, due to a higher backpressure on the engine. Table 6: costs for SOx reduction investment costs

(€/kW) annual cost of operation (€/kW)


€/kg SOx

Low-sulphur fuel (50 ppm)

0 10.35 10.35 6.35

EcoSilencer 200 2.52 51.30 38



If the only goal is SOx reduction it is financially and environmentally, due to higher reduction of SOx emissions, better to use low-sulphur fuel. The EcoSilencer is more valuable for use in seagoing ships, for which it is in fact designed.

10.4. PM removal Certain techniques require a reduction in fuel sulphur level. Lowering the fuel sulphur level will already result in a decrease in particulate matter mass, because the sulphates will be formed to a lesser extent. Lets assume that 5-vol% of the sulphur present will be converted to SO3 and that all SO3 will be converted to sulphates. These sulphates add up to the particle mass. An amount of water, which is most of the time assumed to be equal to the mass of sulphates is also added to the particle mass. This means that a fuel with 0.2% sulphur creates 0.062g/kWh sulphates, and thus 0.12 g/kWh additional particle mass in comparison to a fuel without sulphur. A fuel with 50 ppm sulphur only produces approximately 0.003 g/kWh sulphates with associated water. This means that PM emissions can be reduced by 22% by switching to low-sulphur fuel. An oxidation catalyst is on its most efficient when it is combined with low-sulphur fuel. This is also the case for catalytic particulate traps. With particulate traps an additional 2% fuel consumption is assumed for increased backpressures and regeneration. Particle removal efficiency, including regeneration, is estimated as 85%. To estimate the efficiency of an oxidation catalyst it is assumed that the soluble organic fraction is 50% of the mass without sulphates and is totally oxidised, and 70% of sulphur oxides will form sulphates. This leads to an overall efficiency for the oxidation catalyst of 53%. A SCR contains an oxidation catalyst as well, and combined with low-sulphur fuel it will give the same removal efficiency as an oxidation catalyst. For the efficiency of the EcoSilencer 75% is used. For an ESP 90% particulate removal is used. Table 7: Cost-comparison particle reduction investment costs

(€/kW) annual costs of operation (€/kW)


€/kg PM

Low-sulphur fuel (50 ppm)

0 10.35 10 43

EcoSilencer 200 2.52 51 64 Particulate trap 8.4 15.6 18 19 Oxidation Catalyst

1.5 10.35 11 19

Selective Catalytic Reduction with low-sulphur fuel

40-130 19-26 29-58 51-102

ESP 200 2.52 51 53 In this comparison the distinction between low annual costs and cost per kg reduced particulate matter becomes clear. The cheapest solution to reduce some particulate matter is the use of low-sulphur fuel, this is however a rather expensive way to reduce a kilogram particulate matter. Use of an oxidation catalyst in combination with low-sulphur fuel has approximately the same annual costs, but reduces particulate matter much more efficiently, which results in a lower price per kilogram particulate matter reduced than only the use of low-sulphur fuel. Use of a particulate trap is per kg as cheap as an oxidation catalyst over a 5-year period, but a particulate trap reduces more particulate matter. Selective catalytic reduction is not a cheap option to reduce just particulate matter emissions. However if selective catalytic reduction is chosen for reduction of NOx emissions, the particulate matter reduction is “for free”.



11. REFERENCES [Argillon, 2005] nomen nescio, Sinox SCR Figures, www.argillon.com, accessed 2005

[Bardasz et al., 2004] E.A. Bardasz, F.A. Antoon, E.A. Schiferl, J.C. Wang and W. Totten, The impact of lubricant and fuel derived sulfur species on efficiency and durability of diesel NOx adsorbers, Oils, rheology, tribology and driveline fluids, SP 1894, 2004-01-301, 2004, SAE, Warrendale USA [Bekaert, 2005] nomen nescio, Diesel particulate filtration, http://www.bekaert.com/bft/Products/Filter Elements _ Systems/Diesel Particulate Filtration.htm, Last reviewed: April 5th 2005 [Björsell, 2005] Personal communication with Mr. A. Björsell from Munters Europe AB [Boerrigter et al., 2005] H. Boerrigter, S.V.B. van Paasen, P.C.A. Bergman, J.W. Könemann, R. Emmen, A. Wijnands, “OLGA” tar removal technology, proof-of concept (PoC) for application in integrated biomass gasification combined heat and power (CHP) systems, January 2005, ECN-C—05-009, available in digital format at www.ecn.nl [Bosteels and Searles, 2002] D. Bosteels and R.A. Searles, Exhaust emission catalyst technology, new challenges and opportunities in Europe, Platinum metals review, 2002, volume 46, number 1, pp 27-36, print ISSN: 0032-1400, electronic ISSN: 1471-0676 [CBRB, 2005] nomen nescio, Richtlijn nr 16, overeenkomstig §1.07 RSOR, Emissie van schadelijke gassen en luchtverontreinigende deeltjes –testprocedure, wetstekst beschikbaar via http://www.cbrb.nl/downloads/reader/richtlijn16nl.pdf, accessed May 10th 2005 [CCR, 2003] nomen nescio, Exhaust and pollutant particle emissions from diesel engines, issued by the central commission for navigation on the Rhine, www.ccr-zkr.org, this is an amendment to the Rhine vessel inspection regulations, acquired from Internet in 2003 [CGS, 2005] nomen nescio, Wet electrostatic precipitator, dimension chart no. CG-WESP-18 http://www.cgscgs.com/pdf/pg12.pdf, accessed 2005

[Clean air systems, 2005] nomen nescio, CleanAir ASSURETMDOC, reduces particulate matter, carbon monoxide and hydrocarbon emissions. Brochure acquired at internet 26-04-2005 from http://www.cleanairsys.com/materials/pdf/ASSURE%20Brochure.pdf [Davies] M.J. Davies, Diesel Exhaust Aftertreatment-Options and advances for the next decade, year and place of publication unknown. [De Wilde and Kos, 2004] H.P.J. de Wilde, G.P.A. Kos, Beperking NOx-uitstoot bij een bestaande scheepsdieselmotor, effectiviteit SwirlFlash en SCR-techniek, 2004, ECN-C—04-083, available in digital format at www.ecn.nl [Dieselnet, 2005a] nomen nescio, Caterpillar introducing exhaust gas recirculation in 2007 engines, http://www.dieselnet.com/news/0503cat.html, Dieselnet, March 31 2005 [Dieselnet, 2005b] nomen nescio, CRT Filter Revision 2005.01a acquired from www.dieselnet.com/tech/dpf_crt.html April 28th 2005 (subscription required) [Dogliani & Paro, 2004] M. Dogliani, D. Paro, 2004 Leadership forum technical discussion, Recent developments to increase the environmental performance of new and existing engines, from www.rina.org/UploadedFiles/RINA presentation ICCL Miami.pps, accessed at May 3rd 2005



[Dureal, 2005] nomen nescio, AdBlue Specifications, Dureal, from www.dureal.com/img/AdBlue_Specs_web_lay-out.pdf, accessed May 3rd 2005 [Eastwood, 2000] P. Eastwood, Critical topics in exhaust gas aftertreatment, Exeter 2000, ISBN 0 86380 242 7 [Elzinga et al., 2004] G.D. Elzinga, R.W. van den Brink, F.P.F. van Berkel, Microgolf regeneratie van roetfilters voor dieselmotoren, February 2004, ECN-C—04-025

[EPA, 2004] nomen nescio, Final regulatory analysis: control of emissions from non-road diesel engines, EPA420-R-04-007 chapter 6: estimated engine and equipment costs, accessed at http://www.epa.gov/nonroad-diesel/2004fr/420r04007g.pdf, 2005 [Eschenbacher, 2005] Personal communication with J. Eschenbacher April 8th 2005 [Götmalm, 1991] Ö.A. Götmalm, Diesel exhaust control, the institute of marine engineers, marine managemet (holdings) LTD, 1991 [Haase&Koehne, 1999] F. Haase, H. Koehne, Design of scrubbers for condensing boilers, progress in energy and combustion science, volume 25, number 3, 1999, ISSN 0360-1285. [Ives&Klokk, 1993] R. Ives, S.N. Klokk, Exhaust gas sulphur removal by sea water washing, marine diesel engines, CIMAC, 1993, D58 [JM, 2005a] nomen nescio, Continuously Regenerating Technology (CRTTM), Johnson Matthey from http://www.jmcsd.com/CRT.pdf, accessed at April 28th 2005 [JM, 2005b] nomen nescio, Open the Johnson Matthey emission control toolbox..., Johnson Matthey from http://www.jmcsd.com/JM_Toolbox_Summary_Brochure.pdf, accessed at April 28th 2005 [JM, 2005c] nomen nescio, EGRTTM system upgrade, Upgrade your CRT to EGRT for PM and NOx control or phase-in your emission requirements for more flexible budgeting, Johnson Matthey, http://www.jmcsd.com/EGRT_from_CRT_Datasheet_040916.pdf, accessed at April 28th 2005 [Karila et al., 2004] K. Karila, T. Kärkkäinen, m. Larmi, S. Niemi, C-E Sandström, J. Tamminen and J. Tiainen, Reduction of particulate emissions in compression ignition engines, Publication of the internal combustion engine laboratory Helsinki Unviersity of Technology 78, Espoo 2004. [Lantz, 2005] Personal communication with Ms K. Lantz from Marine Exhaust Solutions, 2005 [McAdams et al, 2003] R. McAdams, P Beech, R. Gillespie, C. Guy, S Jones, T. Liddell, R. Morgan, J. Shawcross, D. Weeks, D. Hughes, J. Oesterle, Non-thermal plasma based technologies for the after-treatment of automotive exhaust particulates and marine diesel exhaust NOx, DEER 2003, August 2003, Newport, Rhode Island [MES, 2005a] nomen nescio, Ecosilencer®, Your sulphur emission solution, Marine Exhaust Solutions Inc., received after request, 2005 [MES, 2005b] nomen nescio, SO2 exhaust gas abatement Ecosilencer®, System trial results august 2003 to december 2004, Marine Exhaust Solutions Inc. (MES,) January 2005, received after request.



[MES, 2005c] information on a system for yachts, http://www.marineexhaustsolutions.com/products_superyacht.asp, Marine Exhaust Solutions Inc., (MES) accessed 2005 [Nord et al., 2000] K. Nord, G. Zurita, B. Tingvall and D. Haupt, Particles and emission from a diesel engine with a humid air motor system, (Draft 00-11-15) Rapport no PRO-02/3, 2000 [Niven, 1993] H. Niven, Impact of emission legislation on large engine design and operation, transactions of the institution of diesel and gas turbine engineers, publication 475, 1993, ISSN 0261-0345 [Oliecentrale, 2005] website of a company that shows their commercial list prices (in Dutch), www.oliecentrale.nl [PSA, 2005] PSA Peugeot Citroën, the particle filter system, http://www.psa-peugeot-citroen.com/modules/fap/fap_en.pdf accessed May 1st 2005 [Scania, 2005] nomen nescio, Technologies for Euro 4 and Euro 5, edition 04.09 enXX1595167, from http://www.scania.com/Images/R32480_EngineTech_ENG_tcm10-62331.pdf accessed in 2005 [Schoonebeek, 2005] Personal communication with Mr. Schoonebeek from Hydro gas and chemicals Benelux, 2005 [Shell, 2002] Shell Gasolie scheepvaart, productinformatie 1145NL01, 20-11-2002 accessed May 13th 2005 at www.shell-tic.com [Shell, 2003] Shell Gasolie basis ultra laag zwavel, productinformatie 1861NL01, 02-12-2003 accessed May 13th 2005 at www.shell-tic.com [Sher, 1998] edited by E.Sher, Pollutant formation and control, handbook of air pollution from internal combustion engines, Academic press, Boston, 1998, ISBN 0-12-639855-0 [Stein et al., 1995] H. J. Stein, G. Hütwohl and G. Lepperhoff, Performance of oxidation catalysts for heavy-duty diesel engines, Catalysis and automotive pollution control III, Proceedings International Symposium (Capoc3) on Catalysis and Automotive Pollution Control, 3rd, Brussels, Belgium, 1994, edited by A. Frennet and J.-M Bastin, pp 517-528, studies in surface science and catalysis, vol 96, 1995 [Trivett et al., 1999] A. Trivett, G. MacDonald, G. Trivett, A report on field trials of a novel wet scrubber for marine diesel engines, Advanced Marine Machinery systems with low pollution and high efficiency, conference proceedings part I, marpower 99, ISBN 1-90 25 36-13-4 [Van Gulijk, 2002] C. van Gulijk, Rational design of a robust diesel particulate filter, Wageningen 2002, ISBN: 90 6464 449-7 [Van Liere, 2005] J. van Liere, Personal communication with Mr. Van Liere of Alpha Power Systems, 2005 [Van Paasen et al., 2004] S.V.B. van Paasen, L. Rabou, R. Bär, Tar Removal with a wet electrostatic precipitator (ESP) a parametric study, presented at “2nd World Conference on Biomass for Energy, Industry and Climate protection”, proceedings p.781-784, 10-14 May 2004, Rome Italy [Van Rens, 2004] G.L.M.A. van Rens, Conceptual design of an integrated, scrubber based, bas cleaning system for marine diesel engines, april 2004, TUDelft, DMS 04/02



[Volvo, 2004] nomen, nescio, Volvo Penta marine diesel D49A MT, obtained from http://www.volvo.com/NR/rdonlyres/3FFC56E6-1E68-4B61-8734-EFC0ACD117B8/0/D49A_MT.pdf, English 03-2004 [Volvo, 2005a] nomen nescio, Veilig te verwerken, makkelijk op te slaan, http://www.volvo.com/trucks/netherlands-market/nl-nl/aboutus/Environment/euro4_euro5/practicalities/safe_to_handle.htm, accessed 2005 [Volvo, 2005b] nomen nescio, Volvo’s oplossingen voor Euro 4 en Euro 5, http://www.volvo.com/trucks/netherlands-market/nl-nl/aboutus/Environment/euro4_euro5/intro.htm, accessed 2005 [Warner, 2005] personal communication with Ms. Warner, from CBRB Rotterdam.

pre- and after-treatment techniques for diesel engines in ... · the goal of this part is to...

Documents