Innovative Catalyst Substrate Components for Future Passenger Car Diesel Aftertreatment Systems

Transcription

1 26th Aachen Colloquium Automobile and Engine Technology Innovative Catalyst Substrate Components for Future Passenger Car Diesel Aftertreatment Systems Rolf Brück, Peter Hirth, Francois Jayat Continental Emitec GmbH, Lohmar, Germany Summary To keep an open minded and positive future for the Diesel engine it is a must that the engine, the engine management, the catalyst and filter systems fulfill their requirements not only in the certification test cycles but also during real world driving. Future catalyst and filter systems have to provide an efficiency which enables ultra clean vehicles. The additional potential of lowering CO2 emissions for the passenger car fleet will give the diesel powertrain also in electrified vehicles as 48 V Mild Hybrids and Plug-in Hybrids an opportunity for the next decades. In order to achieve this requirements each component of the aftertreatment system has to be optimized and adapted to its task. Metal substrates offer these opportunities. By using specially designed substrates tasks like uniform flow and ammonia distribution, passive and active thermal management and space optimized installation can be fulfilled. The influence of single catalyst components and the ideal combination was examined and tested on an engine test bench. As a result a guideline for future innovative exhaust systems can be given. 1 Introduction The discussion if the combustion engine and in particular the diesel engine will have a chance in the future, or if it will be replaced by electric drivetrains, is contrarily discussed. Most of today s studies show that even in 2035 the pure electric drivetrain for passenger cars will only represent a small market share. Independently from that many vehicles will be electrified. That means that the number of hybrids, from 48 V mild hybrids up to plug in hybrids, will rise. In order to make the electrified vehicles even more environmentally friendly, the requirements on the after-treatment of exhaust gases will rise.

2 th Aachen Colloquium Automobile and Engine Technology 2017 Until recently the Diesel engine was one of the key technologies to fulfill the CO2 fleet average. The discussion concerning ecology of the Diesel engine, in particular under real driving conditions, forced to consider this strategy. Already the 48 V hybrid technology in connection with a Gasoline engine shows clear advantages in fuel consumption [1, 2]. In principle all vehicles, independent from their powertrain, will have to meet the lowest emission limits under all driving conditions in the future. That also will lead to higher requirements on the gasoline engine and its catalyst systems. Lambda 1 operation up to wide open throttle driving and gasoline particulate filters are only a few examples. For that reason the question must be asked whether the diesel engine with its better thermodynamic efficiency will also make economically sense in combination with electrification in the future. Combinations of Diesel engines with CO2 neutral fuels like e.g. OME [3, 4] will affect the decision likewise. Basic prerequisite is surely that the future Diesel drivetrain concepts will show no disadvantage in emissions. The exhaust legislation in accordance with EU6d demands, for example the adherence of the nitrogen oxide (NOX) limit of 60 mg/km also under RDE-test conditions (Real Driving Emissions), is prescribed for passenger cars. In the US with the California LEV III legislation the NOX limits are continued to be reduced, whereby a further reduction of the NOX-emissions is demanded by around 50% between 2018 and 2025 down to a level of 30 mg/km. Also the legislation in China plans similar reduction rates. For the Diesel engine, besides an effective reduction of Hydrocarbons (HC) and Carbon Monoxid (CO), the Nitrogen Oxide (NOX) reduction is of special relevance thereby directly impacting also the NO2-imissions. Which of the two well-known technologies-lean NOX-Traps (LNT) or selective catalytic reduction (SCR) or both of them at the same time-will be used, will be influenced and decided in particular by their ability for real world driving, and here both, city as well as motorway. The SCR system by means of the aqueous urea solution (AdBlue ) seems to be set in each case. For compliance with the NOX-limits mentioned above and for the improvement of the environmental image of the diesel engine it is necessary that the effectiveness of the already existing SCR systems is continued to increase. New exhaust concepts are under development, fulfilling the requirements to inject high necessary AdBlue quantities required for the high load without creating deposits and in order to achieve operation temperature as quick as possible after engine cranking.

3 26th Aachen Colloquium Automobile and Engine Technology Due to the increased urea dosing quantities, the danger of NH3-slip rises, in particular, if no ammonia slip catalysts are used. With the today s typical low pressure EGR systems then the risk exists, that exhaust gas with an ammonia content is entering the EGR line and thus the gas inlet side of the engine which might be objected to corrosion. For this reason the discussion is ongoing to use two SCR injectors which will increase the complexity and the cost of an SCR system. New innovative components like the low pressure EGR-filter with integrated NH3-slip catalyst can lead to a simplification. For reduction of cold start emissions a short SCR catalyst disks in front of the particulate filter (DPF) is of advantage and in order to evaporate large AdBlue quantities the electrically heated catalyst will be used. The following report describes the development and test results of the individual components and their combination to get a high efficient exhaust aftertreatment system. 2 Aftertreatment Components In order to develop a highly effective Diesel emission catalyst system it is necessary to fully understand the task for each individual component in the system and in particular their requirements. Figure 1 shows a typical Euro 6 Diesel aftertreatment exhaust system 2.1 Diesel Oxidation Catalyst (DOC) In modern catalyst systems the oxidation catalyst or the NOX-Adsorber (LNT) is arranged directly behind the turbocharger. In order to achieve highest possible HC and CO conversion, the DOC has to be heated up to temperatures above 220 C as quick as possible after engine cranking. However attention does not only have to be paid to the conversion of HC and CO, but ideally the oxidation catalyst should adjust a NO/NO2-ratio of 50% for the downstream SCR catalyst to ensure a good low temperature SCR efficiency by the so-called fast SCR reaction [5]. That means ideally temperatures above 250 C should be achieved, because then the engine NO2-emissions will not react with the hydrocarbons but in addition NO2 is formed.

4 th Aachen Colloquium Automobile and Engine Technology 2017 To clarify this, figure 2 shows a result of NO2 emissions in front of and behind oxidation catalyst for a cold and warm WLTC. Fig. 1: Typical Euro 6 Diesel after-treatment exhaust system Fig. 2: NO2 emissions in front of and behind oxidation catalyst for a cold and warm WLTC and with active heating of the DOC

5 26th Aachen Colloquium Automobile and Engine Technology Taking a closer look to the cold WLTC NO2-emissions a decrease can clearly be recognized for the WLTC cold during the first 800 s, whereas in the hot and active heated tests the NO2 emissions increases. For the layout of an ideal system the question must be asked how to heat the oxidation catalyst as fast as possible and which is the ideal size of it. Using engine catalyst heating measures on one hand the efficiency of the engine is worsened and on the other hand it is even more difficulty to find an operating strategy which compensates the driver influence. New innovative catalyst substrates with turbulent structures like the CS technology can contribute, by the use of additional thermal mass, to a passive temperature management [6]. The advantage of such a hybrid catalyst technology on passive temperature management was proven already several years ago [7]. Unfavourable at that time were the increased costs. The new CS (Cross Corrugation Structure) technology now makes it possible to achieve the same effect with an additional partly inserted flat foil (figure 3) with reduced costs. A typical behaviour during cold start or low load operation is the oscillation of the exhaust gas temperature around the catalytic light off temperature. In such a case the catalyst lights off in the acceleration phase to cool down below the light-off temperature needed for the catalytic reaction during the deceleration and idling. Due to the hybrid substrate technology the front area of the catalyst has a low thermal mass and the rear part has a high thermal mass. This causes on the one hand a fast light-off of the catalyst at the front side and on the other hand a decreased cooling down behaviour in the rear end. Fig. 3: CO-conversion of 400 cpsi catalyst compared to a CS Hybrid catalyst with the same catalytic surface Figure 3 shows as an example the CO-effectiveness of a standard catalyst with 400 cpsi and a catalyst with CS structure having the same geometrical surface area.

6 th Aachen Colloquium Automobile and Engine Technology 2017 The CS substrate at the rear side has an increased thermal capacity by having inserted an additional flat foil. The test was a simulated start/stop driving in the city. The advantage of the hybrid concept on the catalytic efficiency can be noticed by the integral between the two efficiency curves. With the introduction of the 48 V power supply in mild hybrids and the possibility of recuperation electrical heating with heating powers of 1-4 kw are possible and make sense regarding CO2 emissions. On the other hand the 12 V electrically heated catalyst is a proven technology in several Diesel vehicles in the market today and can be adapted to 48 V by some minor modifications. Figure 4 shows the 48 V Emicat 4 variant. Also with the Emicat 4 the proven concept of a quite small, however fast heatable and catalytically active disk, mechanically supported in a metallic substrate, was maintained. In the classical arrangement the heated catalyst is mounted with the heated disk in the front (in direction of flow). As mentioned, in comparison to the known 12 V electrical system now heating powers up to 4 kw can be realized whereby the catalyst is heated within a few seconds (figure 5). Fig. 4: 48 V Emicat 4 The increase of the catalyst temperature in the center of the DOCs amounts controlled to up to 100 K with 0.9 kw and up to 180 K with 4 kw heating power. Due to this rise in temperature both the NO2 emissions (figure 2) and the HC and CO emissions are reduced (figure 6). Due to the higher temperatures, as expected, 33% (900 W) / 45% (4 kw) lower HC and 40% (900 W) / 58% (4 kw) lower CO emissions are achieved. Because in most cases these emissions are not the critical ones to achieve the limits, the precious metal loading of the catalyst can be reduced instead, which attributes to the cost compensation of the heated catalyst system. On the other hand the active heating means a safety margin to achieve the emission limits also in the extended temperature range of the RDE legislation (- 7 C).

7 26th Aachen Colloquium Automobile and Engine Technology Fig. 5: Catalyst temperature in front of catalyst, 50 mm behind gas inlet and outlet temperature of a Diesel oxidation catalyst with and without heating with 0.9 kw and 4 kw Fig. 6: HC emissions in the WLTC with and without heating; 0.9 kw and 4 kw

8 th Aachen Colloquium Automobile and Engine Technology 2017 Fig. 7: CO emissions in the WLTC with and without heating; 0.9 kw and 4 kw The result shows the advantage of the electrical heating, but also shows, that due to the thermal mass of the DOC the temperature behind the DOC rises to a temperature of 220 C only after 195 s (900 W) / 100 s (4 kw), which is the right temperature for a safe and deposit free injection of AdBlue. But, the goal is that the injection of AdBlue should start as early as possible after engine cranking in order to get a high SCR conversion efficiency and to run the engine in a more favourable way in terms of fuel consumption. Thus the question arises whether it is of advantage to place the heater disk more to the gas outlet side, in the center or even at the rear end of the DOC (figure 8). This would be also of advantage in case the first part of the catalyst is coated with a passive NOX adsorber coating because in such cases the SCR reaction could start before desorption temperature is reached.

9 26th Aachen Colloquium Automobile and Engine Technology Fig. 8: Theoretical possibilities of placing the heated disk within the DOC Positioning the heated disk at the gas outlet side would immediately increase the gas outlet temperature of the DOC and improve conditions thereby for the AdBlue dosage. However this variant would have no positive influence on the DOC effectiveness. A compromise would be the position in the catalyst center. Thus at least a part of the DOC could be still heated, which would have a positive influence on the HC, CO and NO2 emissions and also the temperature after the DOC would rise faster in the comparison to the position in front of the DOC. A further advantage of this position would be the possibility to heat the catalyst by short term, repeating "power on"-strategy to keep the catalyst hot during electrical driving conditions for plugin Hybrids. In order to determine the influence on the dosing release time a test was driven with the heater located at the gas outlet side and a heating power of 0.9 kw and 4 kw. Figure 9 shows the temperatures and the time for dosing release with the different strategies. With the heater positioned in the front, the dosing release time could be reduced from originally 812 s to 415 s with 900 W and to 185 s with 4 kw heating power. By placing the heater in the rear end of the DOC those times even could be reduced down to 313 s at 900 W and to 37 s at 4 kw.

10 th Aachen Colloquium Automobile and Engine Technology 2017 Fig. 9: Temperature in front of SCR catalyst and dosing release time unheated, heated disk at the inlet and outlet side with 0.9 kw and 4 kw electrical power In principle there is also the possibility to directly inject Urea onto the heated disk [8]. Due to the fact that the electrical heating energy is used to heat up the catalyst substrate wall directly and thus the Washcoat is heated from the inside, the wall temperature is higher than the gas temperature at the same position. Figure 10 shows the delta T with an exhaust gas temperature of around 200 C, a mass flow of kg/h and a heating energy of 2 kw.

11 26th Aachen Colloquium Automobile and Engine Technology Fig. 10: Temperature difference between gas and catalyst wall temperatures with a heating power of 2 kw An additional effect is that the evaporating water is acting like a thermal insulation to the gas and thus the quantity of AdBlue that can be dosed deposit-free is independent from the exhaust mass flow. Figure 11 shows the quantity of AdBlue which can be dosed deposit free in addition (in comparison to the version without heating) dependent on the gas temperature and the exhaust mass flow with 1 kw heating power. Fig. 11: Maximum dosable amount of AdBlue as a function of mass flow and temperature at a heating power of 1 kw In the next step the effectiveness of the SCR of catalyst was examined.

12 th Aachen Colloquium Automobile and Engine Technology SCR catalyst/sdpf With most EU6 exhaust systems a so-called SDPF is used. It consists of a standard Diesel particulate filter (DPF) with a SCR coating. The disadvantage of this construction is that the amount of washcoat negatively affects the pressure loss compared to a standard catalyst, which leads to lower possible washcoat loadings compared to standard substrates. This negatively influences the SCR efficiency. In addition the thermal mass of a particulate filter compared to a thin wall metal substrate is higher whereby heating-up is delayed. Compared to a cordierite filter a thin wall metal substrate would have an approx. 30% lower thermal mass. By that it makes sense to place a short disk of a standard catalyst in front of the SDPF. The disk should be only that long that on the one hand the SCR light off effectiveness is improved and on the other hand the disadvantage for active DPF regeneration because of increased thermal mass is minimized. Lengths of mm have been proven here as favourable. Figure 12 shows the comparison of the SCR effectiveness with and without catalyst disk. Fig. 12: SCR effectiveness with and without catalyst disk and installation sketch

13 26th Aachen Colloquium Automobile and Engine Technology To ensure a high SCR effectiveness also under high load conditions, a separate SCR catalyst is installed behind the SDPF. A challenge with compact exhaust systems is the achievement of a good NH3 distribution over the entire catalyst cross section. Using a multi brick arrangement with gaps between the bricks (like in between the SDPF and the SCR catalyst), leads to some mixture and flow uniformity. The flow distribution improvement depends on the gap length between the bricks. Figure 13 shows the flow uniformity distribution as a function of the gap length. However an ideal gap size of mm means additional need of installation space. This space could be alternatively used upstream of the SDPF for application of an SCR catalyst disk with PE technology. The PE-technology is currently applied for high performance vehicles with gasoline engines both for the backpressure advantage compared to standard substrates and for equalization of cylinder-to-cylinder lambda variations. Fig. 13: Comparison of flow distribution as function of the gap length between two catalyst substrates The PE catalyst design offers similar remixing features also with SCR coating. In a direct comparison to a 400 cpsi standard catalyst, the Uniformity index could be improved from 0.89 up to a value of 0.93.

14 th Aachen Colloquium Automobile and Engine Technology 2017 Fig. 14: Gas flow velocity result downstream a PE-catalyst compared with a standard catalyst However, in many cases the packaging constraints in the engine compartment do not allow the installation of an additional SCR catalyst downstream of the SDPF hence leading to a separate SCR module in underfloor position. The advantage of the underfloor position is that exhaust downpipe acts as an additional mixing device. On the other hand, this means also that the close-coupled SCR will be partially overloaded with ammonia. Consequently, in an engine with low pressure EGR and extraction point downstream the SDPF, ammonia could arrive at engine intake potentially leading to corroding condensates in particular at low load conditions. For such cases, an interesting solution for ammonia abatement, could be an EGR filter with integrated catalytic function or even a small catalyst. 2.3 Low pressure EGR filter with ammonia oxidation function Continental Emitec already today produces low pressure EGR filters in very large numbers. Usually the EGR filter is placed directly behind DPF and fulfils in addition to the filtration the function of the OBD monitoring of the filter. The filter (figure 15) consists of a special wire mesh.

15 26th Aachen Colloquium Automobile and Engine Technology Fig. 15: Continental Emitec low pressure EGR filter with NH3 oxidation catalyst For the oxidation of ammonia 2 variants were checked. 1. A coated wire mesh 2. The use of a short catalyst disk The NH3 oxidation efficiency of a platinum coating has been determined on a synthetic gas test bench. NH3 concentrations ( ppm), exhaust gas temperatures (200 C 500 C) and exhaust mass flow (25 kg/h 100 kg/h) have been varied. Figure 16 shows the efficiency of the wire mesh coated with 1.13 mg Platinum.

16 th Aachen Colloquium Automobile and Engine Technology 2017 Fig. 16: NH3 oxidation efficiency of the coated wire mesh (T = 500 C, NH3- concentration in the Feed = 50 ppm) If the extremely low geometrical surface of the EGR filter fleece is taken into account the measured oxidation efficiency of between 20 and 65%, depending on the boundary conditions appears to be comparably high, but not sufficient. Highest numbers could be found at high temperature of 500 C and lowest load of ammonia raw emissions (50 ppm) and lowest space velocity. Some potential is expected by increasing mass transfer coefficient by means of wire shape and dimension or by increasing the total GSA by adding additional fibre layers. Then, in the 2 nd step the efficiency of a catalyst disk installed in the low pressure EGR line has been determined (figure 17). The efficiency of the standard substrate catalyst disk is clearly better and exceeds in a wide operating range between 300 C and 500 C at all tested catalyst loads 90% by far. It could be however observed that NO2 and N2O also are produced in the temperature range C. The outcome of this is that an investigation program is necessary to determine to what extent the exhaust gas emissions of the engine are modified by the new nature of the recirculated exhaust gas. An alternative would be a 2 nd, additional injection point located downstream the low pressure EGR extraction point in the under floor.

17 26th Aachen Colloquium Automobile and Engine Technology Fig. 17: NH3 oxidation efficiency of a catalyst disk 40 mm long (T = 500 C, NH3- concentration in the Feed = 50 ppm) 2.4 Second AdBlue nozzle placed underfloor As briefly discussed, at high temperatures (higher than 550 C) NH3 oxidation is likely to happen, moreover this NH3 loss requires some overdosing of AdBlue and thus to an increased consumption. The advantage of a purely close coupled injection is however a very good NH3 distribution in front of the under floor SCR Catalyst. In the following the pros and cons of a second injection in the under floor position are examined Investigation of the exhaust gas temperature at high loads As a first step, temperature measurements at high engine load have been carried out with the goal to determine possible temperature advantages (temperature reduction). Figure 18 shows the temperatures in front of the close coupled and in front of the under floor injection position at 3000 rpm with full load. Due to the Diesel typical low temperature level and the high exhaust mass flow the temperature difference is only 38 K. The absolute temperature at this load point is still 570 C and thus in a range in which ammonia will be oxidized. A further temperature reduction would be only possible by positioning the 2 nd SCR injection point further downstream in the exhaust system or with an active cooling.

18 th Aachen Colloquium Automobile and Engine Technology 2017 Fig. 18: Temperature at the close coupled and under floor injection position with 3000 rpm with full load NH3 distribution As already mentioned, the challenge of the 2 nd SCR injection point is the realization of an Ammonia even distribution along a very short path with pressure loss as low as possible. Underfloor injection technology is already known from the EU 5 SCR exhaust systems. Mixers for both evaporation and mixing were used. Usually the SCR systems consisted however of 2-3 substrates with additional cones [9] in order to reach a good mixing. Figure 19 shows in principle the required NH3 Uniformity index of an exhaust system with one or two injection points. Also this graph explains again that the purely close coupled injection is advantageous. Innovative substrates, as for example the Ring Catalyst, allow to use smart exhaust flow paths and an active cooling in high load points without any disadvantage in low load points.

19 26th Aachen Colloquium Automobile and Engine Technology Fig. 19: In principle required NH3 Uniformity index of an exhaust system with one or two injection points Fig. 20: Ring Cat System with SCR injection point and active cooling for the high load point

20 th Aachen Colloquium Automobile and Engine Technology 2017 The decision, which system would be best, depends certainly on the absolute closecoupled temperature and on the question, if NH3 will be oxidized and if the resulting loss of NH3 (also the resulting higher AdBlue dosing quantity) can be tolerated. Other decisive factors are the efficiency of the system which defines the overdosing rate and some other influencing parameters. 3 Emission results For the emission measurements the following exhaust system was designed: Number of AdBlue dosing spots = 1 DOC (without LNT function) DOC as an electrically heated catalyst with a 4 kw heated disc at the gas outlet SCR-catalyst disc in front of SDPF PE-catalyst as a second close-coupled SCR-catalyst behind SDPF No underfloor catalyst The NOX results shown below represent just a first step since the AdBlue dosing strategy wasn t yet adapted to the used exhaust system (figure 21). Fig. 21: NOX results of the tested exhaust system in WLTC cold and warm compared to the standard EU6 exhaust system

21 26th Aachen Colloquium Automobile and Engine Technology The tested system, thanks to the combination of the different adopted measures, could reduce the NOX emissions in cold WLTC by 69% with 0.9 kw resp. 85% with 4 kw both heated at GE and by 75% with 0.9 kw resp. 86% with 4 kw both heated at GA. In the emissions measurements (without NH3 slip catalyst) no NH3 tailpipe emissions could be detected. The next step will be to adapt the dosing strategy to the exhaust system and to measure additional RDE cycles. 4 Conclusions The design of the DOC is of high importance for a proper function of a today s Diesel aftertreatment system. It could be shown that metallic substrates allow to use high efficient solutions. The CS structure in combination with an additional flat layer can be used to provide both very good light off behaviour in cold start due to low thermal mass as well as a high heat capacity after being heated up in order to keep the catalyst warm under low load phases. Heating the DOC at its gas inlet side improves HC and CO light off clearly. Moving the heated part of the DOC from gas inlet to the middle of the DOC, also helps to improve light off of the subsequent SCR catalyst in order to achieve earlier Urea dosing, without worsening the HC and CO light off behaviour too much. The EHC shows very high potential to be able to evaporate high amounts of Urea without creation of deposits. As SDPF Filters mounted directly behind the DOC show a high heat capacity and therefore allow comparably late Urea dosing, application of a short SCR catalyst disk in front of SDPF reduces the time to dosage begin significantly without disturbing filter regeneration too much. The distance between this short disk and the SDPF is a decisive factor for ammonia distribution on the front face of the SDPF. In order to improve SCR efficiency, a very uniform ammonia distribution is required, which can be achieved by application of radially open SCR catalyst structures like the PE structure. The SCR system can be simplified in order to avoid the effort for a second Urea injection point in the underfloor position by using a catalysed EGR filter for oxidation od excess ammonia behind SDPF. This can be designed by using a coated filter web without or in combination with short metal substrate disk. By this, high ammonia oxidation rates on the filter could be achieved. Application of Urea dosing versions with one and with two injection points does strongly depend on the temperature level at each injection point and the resulting tendency to oxidise ammonia, so this has to be decided application specifically. Under consideration of all of those design rules it is possible to set-up a very efficient Diesel exhaust after treatment system, both regarding emissions and costs.

22 th Aachen Colloquium Automobile and Engine Technology References [1] KNORR, T. et al. The Electric Heatable Catalyst An Efficient Measure for Emission Optimization in Mild Hybrid Vehicle Operation Strategies, 24. Aachener Colloquium Automobile and Engine Technology, 2015 [2] ZHANG et al. Die Zukunft der Benzinmotoren, 36. Internationales Wiener Motorensymposium, 2015 [3] MAUS, W.; JACOB, E. Synthetische Kraftstoffe OME1: Ein potenziell nachhaltig hergestellter Dieselkraftstoff, 35. Internationales Wiener Motorensymposium, 2014 [4] MAUS, W. et al. Sustainable Fuel-a Fantasy?, MTZ worldwide 6/2012, Springer-Verlag, 2012 [5] HIRTH, P. et al. Metallische SCR-Katalysatorsysteme; Kenngrößen für optimale Effektivität, 8. FAD-Konferenz Dresden, 2010 [6] BRÜCK, R. et al. Der Weg zur Erreichung der CARB post 2023" Emissionsgesetzgebung für Nutzfahrzeuge, 38. Internationales Wiener Motorensymposium, 2017 [7] SCHAPER, K.; KONIECZNY, R.; BRÜCK, R. Diesel-Katalysator mit Hybrid-Trägerstruktur, Abgasreinigung bei mageren Dieselmotoren, Haus der Technik, Essen, Juni 1999 [8] PFAHL, U.; SCHATZ, A.; KONIECZNY, R. Advanced Exhaust Gas Thermal Management for Lowest Tailpipe Emissions Combining Low Emission Engine and Electrically Heated Catalyst, SAE Technial Paper , Detroit, 2012 [9] SCHÜTTE, T. Ablagerungs-und Alterungsverhalten wässriger Harnstofflösung bei selektiver katalytischer Reduktion von Stickoxidemissionen, Dissertation Lüneburg, 2010