Field of the invention

The present invention relates to a method pertaining to treatment of exhaust flows arising from a combustion process by using a post-treatment system. The invention relates also to a system and a vehicle and to a computer programme and a computer programme product.

Background to the invention Growing official concern about pollution and air quality, primarily in urban areas, has led to the adoption of emission standards and rules in many jurisdictions.

Such emission standards often take the form of sets of

requirements which define acceptable limits for exhaust emissions from vehicles equipped with combustion engines. For example, these standards often regulate levels for emission of nitrogen oxides (N0 _X ) , hydrocarbons (HC) , carbon monoxide (CO) and particles from most types of vehicles.

In an endeavour to comply with such emission standards, the exhaust gases caused by combustion in combustion engines are post-treated (cleaned) . It is for example possible to employ a so-called catalytic cleaning process, so post-treatment systems, e.g. on vehicles and other means of transport, usually comprise at least one catalyst. Post-treatment systems may also comprise other components, e.g. particle filters, as alternatives to or in combination with one or more catalysts.

Combustion of fuel in an engine's combustion chambers, e.g. cylinders, results in the formation of soot particles.

Particle filters used to capture these particles work in such a way that the exhaust flow is led through a filter structure whereby soot particles are captured from the passing exhaust flow and are stored in the filter.

As it progressively fills with soot when the vehicle is travelling, the particle filter has sooner or later to be emptied of soot, which is usually achieved by so-called regeneration .

Regeneration involves the soot particles, consisting mainly of carbon particles, being converted to carbon dioxide and/or carbon monoxide in one or more chemical processes, and may be conducted in various ways.

It may for example take the form of so-called N0 ₂ -based

regeneration, often also called passive regeneration, which results in the formation of nitrogen monoxide and carbon monoxide by reaction between carbon and nitrogen dioxide. N0 ₂ - based regeneration affords the advantage that desired reaction rates, and hence the rate at which the filter is emptied, may be achieved at relatively low temperatures, which is

advantageous in particular where the post-treatment system comprises temperature-sensitive components.

N0 ₂ -based regeneration does however depend greatly on the supply of nitrogen dioxide. If the supply of nitrogen dioxide is reduced, the regeneration rate will also be reduced.

The supply of nitrogen dioxide may for example be reduced if its formation is hindered, which may for example be due to one or more components of the post-treatment system being poisoned by sulphur, which is normally present in certain kinds of fuel, e.g. diesel fuel.

Summary of the invention

An object of the present invention is to propose a method for desulphuration of at least one component of a post-treatment system. This object is achieved by a method according to the characterising part of claim 1.

The present invention relates to a method for desulphuration of a post-treatment system which is intended to treat an exhaust flow arising from combustion in a combustion engine and comprises at least one first component and one second component. If sulphur has accumulated in said first component of said post-treatment system, the method comprises

- raising a temperature of said first component by supplying said exhaust flow with fuel for oxidation in said post- treatment system,

- stopping said supply of fuel to said exhaust flow when said temperature of said first component is at a first level which exceeds a temperature tolerance of at least one second

component of said post-treatment system which is situated downstream of said first component, and

- resuming the supply to said post-treatment system of fuel for oxidation when said temperature of said first component has dropped to a second level which is lower than said first level.

If for example a vehicle is used for a lengthy period in such a way that the exhaust temperatures are kept relatively low, e.g. below 300-350°C, the sulphur which fuels such as diesel fuel normally contain will react chemically with the active coating, often consisting of noble or other metals, with which components of the post-treatment system are usually provided. In this reaction, sulphur molecules, e.g. in the form of sulphates, bind to metal atoms/ions which can then no longer take part in desired chemical reactions, i.e. the component becomes poisoned by sulphur deposition.

How long this poisoning takes depends for example on

prevailing temperatures in the post-treatment system, the volume of the exhaust flow and the cleanness of the fuel, i.e. in this case the presence of sulphur.

The effects of the sulphur deposition may be exacerbated if a regeneration is imminent. As previously mentioned, N0 ₂ -based regeneration depends on nitrogen dioxide NO ₂ , and sulphur poisoning has adverse effects on the characteristics of the post-treatment system with regard to N0 ₂ conversion, i.e.

conversion of nitrogen monoxide NO to nitrogen dioxide N0 ₂ . This means that the rate of the N0 ₂ -based regeneration will decrease. Where the situation with regard to sulphur

deposition is unfavourable, there will also be continuous further poisoning. In the worst case a passive regeneration will proceed so slowly that instead of being emptied the filter fills up with further soot so that the vehicle has eventually to come to a halt in order to carry out a so-called parked regeneration, resulting in unacceptable costs in terms of time and fuel consumption.

The decreased N0 ₂ conversion, i.e. the changed balance between NO and N0 ₂ in the exhaust flow, may also adversely affect the exhaust cleaning in other ways. The post-treatment system may for example comprise an SCR catalyst which is for example situated downstream of a particle filter and depends on N0 ₂ formation to achieve total NO _x conversion.

The present invention reduces problems of sulphur poisoning of components in the post-treatment systems by responding to the detection of sulphur poisoning by cyclically raising the temperature of the component where sulphur deposition has been found. Cyclic raising and lowering of the temperature causes "pulsing" of the temperature of components of the post- treatment system. The component primarily subject to the poisoning is, from the engine onwards, the first component coated with (noble) metal which the exhaust flow meets, e.g. an oxidation catalyst. The cyclic raising of the temperature affords the advantage that where the desulphuration is very temperature-dependent a high temperature may be achieved in the component primarily poisoned, and subsequently allowing the temperature to drop to a lower level before it is again raised will result in the temperature of downstream components of the post-treatment system not being raised to the same extent, owing to their thermal inertia. Thus the temperature of a sulphur-poisoned component may be raised to a

substantially higher temperature than would be tolerated by downstream components, which means that good desulphuration can be achieved without risk of damage to the downstream components .

Further characteristics of the present invention and

advantages thereof are indicated by the detailed description of embodiment examples set out below and the attached

drawings .

Brief description of drawings

Fig. 1A depicts schematically a vehicle with which the

present invention may be employed.

Fig. IB depicts a control unit in the control system for the vehicle depicted in Fig. 1.

Fig. 2 depicts the post-treatment system in more detail for the vehicle depicted in Fig. 1. Fig. 3 depicts an example of the regeneration (soot burnout) rate as a function of the amount of soot in the particle filter, and its temperature dependency.

Fig. 4 depicts the temperature dependency of the oxidation of nitrogen oxide to nitrogen dioxide in an oxidation catalyst, and the temperature dependency of the reaction rate when oxidising carbon by means of O ₂ . Fig. 5 depicts a method according to an embodiment example of the present invention.

Fig. 6 is a temperature diagram of a desulphuration process according to the present invention.

Detailed description of embodiments

Fig. 1A depicts schematically a power train of a vehicle 100 according to an embodiment of the present invention. The vehicle depicted has only one axle provided with tractive wheels 113, 114 but the invention is also applicable on vehicles in which more than one axle is provided with tractive wheels, and on vehicles with one or more further axles, e.g. one or more tag axles. The power train comprises a combustion engine 101 connected in a conventional way, via an output shaft of the engine, usually via a flywheel 102, to a gearbox 103 via a clutch 106.

The engine is controlled by the vehicle's control system via a control unit 115. The clutch 106, which may for example be automatically operated, and the gearbox 103 are also

controlled by the vehicle's control system by means of one or more suitable control units (not depicted). The vehicle's power train may of course also be of some other kind, e.g. a type with conventional automatic gearbox etc.

An output shaft 107 from the gearbox 103 drives the tractive wheels 113, 114 via a final gear 108, e.g. a conventional differential, and driveshafts 104, 105 which are connected to said final gear 108.

The vehicle 100 further comprises a post-treatment (exhaust cleaning) system 200 for treatment (cleaning) of exhaust emissions arising from combustion in the engine's combustion chambers (e.g. cylinders). The post-treatment system is depicted in more detail in Fig. 2, showing the vehicle's engine 101 from which the exhaust gases (the exhaust flow) generated by the combustion being led through a turbo unit 220. In turbo engines the exhaust flow arising from the combustion often drives a turbo unit which compresses the incoming air for the combustion in the

cylinders. Alternatively, the turbo unit may for example be of compound type. The function of various kinds of turbo unit is well-known and is therefore not described in more detail here. The exhaust flow is then led via a pipe 204 (indicated by arrows) to a diesel particle filter (diesel particulate filter, DPF) 202 via an oxidation catalyst (diesel oxidation catalyst, DOC) 205.

The oxidation catalyst DOC 205 has various functions and is normally used primarily, as part of the post-treatment, to oxidise remaining hydrocarbons and carbon monoxide in the exhaust flow to carbon dioxide and water. The oxidation of hydrocarbons (i.e. oxidation of fuel) results also in the formation of heat which may be utilised to raise the

temperature of the particle filter during its emptying, its so-called regeneration.

The oxidation catalyst may also oxidise to nitrogen dioxide (N0 ₂ ) a large proportion of the nitrogen monoxides (NO) present in the exhaust flow. This nitrogen dioxide is for example utilised in N0 ₂ -based regeneration. Other reactions may also take place in the oxidation catalyst.

The post-treatment system further comprises an SCR (selective catalytic reduction) catalyst 201 situated downstream of the particle filter 202. SCR catalysts use ammonia (NH ₃ ) , or compounds from which ammonia may be generated/formed, as additive to reduce the amount of nitrogen oxides NO _x in the exhaust flow. The effectiveness of this reduction does however depend on the ratio between NO and NO2 in the exhaust flow, so this reaction too is adversely affected by lowered O2 conversion .

In the embodiment depicted, the components DOC 205, DPF 202 and the SCR catalyst 201 are integrated in a combined exhaust cleaning unit. It should however be noted that these

components need not be integrated in a single exhaust cleaning unit, as they might be arranged in some other way deemed appropriate, or one or more of them might for example be separate units. Fig. 2 shows also temperature sensors 210-212 and a differential pressure sensor 209.

Control systems in modern vehicles generally comprise a

communication bus system consisting of one or more

communication buses to connect a number of electronic control units (ECUs), e.g. the control units or controllers 115, 208, to various components on board the vehicle. Such a control system may comprise a large number of control units and the responsibility for a specific function may be shared by two or more of them. For the sake of simplicity, only control units 115 and 208 appear in Figs. 1A-B.

In the embodiment depicted, the present invention is

implemented in control unit 208, which in the embodiment depicted takes care as above of other functions in the post- treatment system, e.g. the regeneration (emptying) of the particle filter 202, although the invention might equally well be implemented in a control unit dedicated to it, or wholly or partly in one or more other control units with which the vehicle is already provided, e.g. the engine control unit 115. The function according to the present invention of control unit 208 (or the control unit or units in which the present invention is implemented) will depend not only on signals from one or more of the temperature sensors 110-112 but probably also on, for example, information received from, for example, the control unit or units which control engine functions, i.e. in the present example control unit 115. Control units of the type depicted are normally adapted to receiving sensor signals from various parts of the vehicle. Control unit 208 may for example receive sensor signals as above and also from other control units than the engine control unit 115. Such control units are also usually adapted to delivering control signals to various parts and components of the vehicle, e.g. control unit 208 may for example deliver signals to the engine control unit 115.

Control is often governed by programmed instructions,

typically in the form of a computer programme which, when executed in a computer or control unit, causes the

computer/control unit to effect desired forms of control action, e.g. method steps according to the present invention.

The computer programme usually forms part of a computer programme product which comprises a digital storage medium 121 (see Fig. IB) which has the computer programme 109 stored on it and may for example take the form of any from among ROM (read-only memory) , PROM (programmable read-only memory) , EPROM (erasable PROM) , flash memory, EEPROM (electrically erasable PROM), a hard disc unit etc., and be situated in or in communication with the control unit, in which case the computer programme is executed by the control unit. The vehicle' s behaviour in a specific situation may thus be modified by altering the computer programme's instructions.

A control unit example (control unit 208) depicted

schematically in Fig. IB may comprise a calculation unit 120 which may for example take the form of any suitable kind of processor or microcomputer, e.g. a circuit for digital signal processing (Digital Signal Processor, DSP) , or a circuit with a predetermined specific function (Application Specific

Integrated Circuit, ASIC) . The calculation unit is connected to a memory unit 121 which provides it with, for example, the stored programme code 109 and/or the stored data which the calculation unit needs for it to be able to perform

calculations. The calculation unit is also arranged to store partial or final results of calculations in the memory unit 121.

The control unit is further provided with respective devices 122, 123, 124, 125 for receiving and sending input and output signals. These signals may comprise waveforms, pulses or other attributes which the input signal receiving devices 122, 125 can detect as information for processing by the

calculation unit 120. The output signal sending devices 123, 124 are arranged to convert calculation results from the calculation unit 120 to output signals for conveying to other parts of the vehicle's control system and/or the

component/components for which the signals are intended. Each of the connections to the respective devices for receiving and sending input and output signals may take the form of one or more from among a cable, a data bus, e.g. a CAN (Controller Area Network) bus, a MOST (Media Oriented Systems Transport) bus or some other bus configuration, or a wireless connection. As previously mentioned, the engine's combustion results in the formation of soot particles which should not, and in many cases are not permitted to, be released into the vehicle's surroundings. Diesel particles consist of hydrocarbons, carbon (soot) and inorganic substances such as sulphur and ash. As mentioned above, these soot particles are captured by the particle filter 202, which works in such a way that the exhaust flow is led through a filter structure whereby soot particles are captured from the passing exhaust flow in order to be stored in the filter. The filter makes it possible for a very large proportion of the particles to be separated from the exhaust flow. The particles separated from the exhaust flow from the filter 202 progressively accumulate in the filter, which thus fills with soot over time. Depending on such factors as current running conditions, the driver's driving style and the

vehicle's load, a larger or smaller amount of soot particles will be generated, so the filling of the filter with

soot /particles will take place more or less quickly, but when the filter reaches a certain level of fullness it needs to be "emptied". If the filter is filled to too high a level, the vehicle's performance may be affected, while at the same time fire hazards may also arise from soot accumulation in

combination with high temperatures.

As above, the emptying of the particle filter 202 is effected by regeneration whereby soot particles, carbon particles, take part in a chemical process. Over time, the filter is

regenerated at more or less regular intervals, and the

determination of suitable times for regenerating it may for example be by means of a control unit 208 which may for example determine appropriate times at least partly on the basis of signals from a pressure sensor 209 which measures the differential pressure across the filter. The fuller the filter becomes, the higher the pressure difference across it.

Regeneration may be conducted in mainly two different ways. One is so-called oxygen (0 ₂ ) based regeneration, also called active regeneration, involving a chemical process

substantially according to C + 0 ₂ = C0 ₂ + heat. Oxygen-based regeneration thus converts carbon plus oxygen gas to carbon dioxide plus heat. This chemical reaction is highly temperature-dependent and requires high particle filter temperatures to enable appreciable rates of reaction to take place at all. At the same time, the temperature tolerance of the components of the exhaust system is often limited, which means that active regeneration may be subject to a maximum permissible temperature which is low relative to the

temperatures required for desired reaction rates to be

achieved. The temperatures required in this type of

regeneration for desired reaction rates may thus be too high relative to temperature tolerances of the components of the post-treatment system. The particle filter 202 and/or any downstream SCR catalyst are for example often subject to design limits with regard to the maximum temperature to which they may be allowed to be exposed.

For this reason, such systems often employ N0 ₂ -based

regeneration which primarily forms nitrogen oxide and carbon oxide by reaction between carbon and nitrogen dioxide

according to N0 ₂ + C = NO + CO. In other words, N0 ₂ -based regeneration is highly dependent specifically on 0 ₂ . N0 ₂ - based regeneration affords the advantage that desired reaction rates, and hence the rate at which the filter is emptied, may be achieved at substantially lower temperatures. N0 ₂ -based regeneration of particle filters typically takes place at temperatures within the range 200-500°C, although temperatures in the upper part of this range are normally preferable.

Compared with active regeneration, the result is a

substantially lower temperature range, the whole of which may be below the minimum temperature desired in active

regeneration. This is for example a great advantage where SCR catalysts are used, since the risk of such a high temperature level as to potentially cause damage to them can in principle be avoided. It is nevertheless still important to achieve a relatively high temperature.

Fig. 3 illustrates an example of regeneration (soot burn-out) rates in N0 ₂ -based regeneration as a function of the amount of soot in the particle filter in operating situations at two different temperatures (350°C and 450°C). The regeneration rate is also exemplified for respective low and high

concentrations of nitrogen dioxide. As may be seen in the diagram, the burn-out rate is low at low temperature (350°C) and low concentration of nitrogen dioxide. The temperature- dependency of the regeneration rate is indicated by the fact that the burn-out rate is relatively low even at high

concentration of nitrogen dioxide so long as the filter temperature is low. The burn-out rate is substantially higher at 450°C even in cases where a low concentration of nitrogen dioxide prevails, but a high temperature in combination with high contents of N0 ₂ is preferable.

Thus passive regeneration depends not only on the temperature of the particle filter and the amount of soot as in Fig. 3 and as indicated by the chemical processes above, but also on the supply of nitrogen dioxide. However, the proportion of nitrogen dioxide N0 ₂ to the total amount of nitrogen oxides NO _x generated by the engine's combustion amounts to only 0 - 10% of the total amount of nitrogen oxides NO _x in the exhaust flow. When the engine is under heavy load, the proportion of N0 ₂ may even be as low as 2 - 4%. With the object of achieving rapid regeneration of the particle filter, it is therefore desirable that the proportion of nitrogen dioxide in the exhaust flow be as high as possible when the exhaust flow enters the particle filter.

It is therefore also desirable to increase the amount of nitrogen dioxide N0 ₂ in the exhaust flow arising from the engine's combustion. This conversion may be conducted in various different ways and may as above be achieved by means of the oxidation catalyst 205 whereby nitrogen monoxide may be oxidised to nitrogen dioxide. Oxidation of nitrogen monoxide to nitrogen dioxide in the oxidation catalyst is however also a highly temperature- dependent process, as exemplified in Fig. 4. As may be seen in the diagram, it is possible at favourable temperatures for the proportion of nitrogen dioxide to the total amount of nitrogen oxides in the exhaust flow (the continuous line) to be increased to nearly 60%. As the diagram also shows, a temperature of the order of 250-350 °C would therefore be optimum in passive regeneration for achieving as much

oxidation of nitrogen oxide to nitrogen dioxide as possible. As described above, however, a quite different temperature relationship applies for the actual burn-out process. This relationship is represented by the broken line in Fig. 4 and, as may be seen, the reaction rate is in principle non-existent at below a particle filter temperature of 200-250° (the temperature indications depicted are merely illustrative examples which actual values may deviate from, e.g. the way in which the temperatures are determined/calculated may affect the temperature limits. Some ways of determining the filter's temperature are exemplified below) .

The burn-out rate (the regeneration rate) thus increases with the amount of N0 _X in the exhaust flow, the temperature of the exhaust flow (the temperature of the particle filter) and the prevailing amount of soot in the filter. NC> ₂ -based

regeneration therefore requires a good supply of O ₂ . As indicated above, the NO ₂ content of the total amount of NO _x in the exhaust flow may be markedly increased by means of the oxidation catalyst 205, whereby the resulting NO ₂ content after the oxidation catalyst depends greatly on the temperature. However, the conversion of NO to NO ₂ by means of the oxidation catalyst depends not only on the temperature of the catalyst but also on whether it has been poisoned by undesirable coating.

Depending on how a vehicle is run, the temperature of the exhaust flow arising from the combustion will vary. If the engine is working hard, the exhaust flow will maintain a higher temperature, whereas when the vehicle is running with a relatively low load upon the engine the temperature of the exhaust flow will be substantially lower. If the vehicle is run for a lengthy period in such a way that the temperature of the exhaust flow stays relatively low, e.g. below 300-350°C, degradation of the function of the oxidation catalyst will occur because the sulphur in various forms which is usually present in the fuel reacts with the active coating, which usually comprises one or more noble or other suitable metals, e.g. aluminium. This has adverse effects upon the

characteristics of the oxidation catalyst with regard to N0 ₂ conversion. The sulphur may react with the coating and may for example form sulphates, e.g. aluminium sulphate, platinum sulphate and palladium sulphate, depending on which type of metal is present in the coating. These sulphates occupy the surface of the active coating and prevent desired reactions, e.g. oxidation of NO to N0 ₂ . Compared with the case of an oxidation catalyst which is not affected, N0 ₂ conversion in an oxidation catalyst with sulphur coating will therefore deliver a lower proportion of N0 ₂ in otherwise similar conditions.

This decrease in N0 ₂ conversion by the oxidation catalyst thus results in a smaller amount of N0 ₂ being available for N0 ₂ - based regeneration of the particle filter, thereby reducing the conversion of soot and hence the regeneration rate. In serious cases of sulphur coating in the oxidation catalyst, any NC>2-based regeneration conducted will in the worst case proceed so slowly that the soot replenishment is greater than the burn-out, potentially leading to the filter filling to such a high level that the vehicle has to come to a halt in order to carry out a so-called parked regeneration. This results, as above, in loss of time with its associated costs as well as increased fuel consumption.

The reduced NO2 conversion may also have further disadvantages. The post-treatment system may as above comprise an SCR

catalyst usually situated downstream of the oxidation catalyst and the particle filter. The SCR catalyst depends for its function on a good supply of O2 to enable the overall N0 ₂ conversion in the post-treatment system to fulfil stated requirements.

The present invention proposes a method for effectively reducing problems arising from sulphur coating in the

oxidation catalyst, without risk of damage to downstream components which are more temperature-sensitive. This is achieved by means of a method 500 illustrated in Fig. 5, beginning with a step 501 of determining whether unacceptable sulphur coating of the oxidation catalyst has occurred, which may be done in various different ways. It may for example be determined during the course of an N0 ₂ -based regeneration that the regeneration rate is unacceptably low. This determination may for example be by means of the differential pressure sensor 209 depicted in Fig. 2 which measures the differential pressure across the particle filter 202. During the

regeneration, the differential pressure across the particle filter will decrease progressively as the filter is emptied of soot particles and the throughflow resistance therefore decreases. If this differential pressure decrease takes place more slowly than expected during regeneration, which may for example be determined by comparing the rate of decrease relative to the prevailing regeneration conditions, it may be determined that the regeneration rate is unacceptably low despite prevailing temperature conditions, indicating sulphur coating in the oxidation catalyst.

Whether sulphur coating has occurred or not may also be determined in other ways. It is for example possible to use suitable models to estimate a degree of sulphur coating on the basis of the type of fuel used and a temperature history representing for example that of the oxidation catalyst. As mentioned above, the lower the temperature of the oxidation catalyst, the more quickly sulphur coating will take place. The sulphur coating rate depends also on the exhaust flow, and the greater the flow the more quickly will the active surfaces of the oxidation catalyst be coated with sulphur. Sulphur coating of the oxidation catalyst may also be determined in other suitable ways.

Where step 501 thus determines that sulphur coating in the oxidation catalyst has occurred, e.g. to some appropriate extent which may for example be controlled by regeneration rate, calculated coating or the like as above, the method moves on to step 502 for determination of a temperature T of the post-treatment system 200, e.g. a temperature representing that of the oxidation catalyst 205. This temperature is then compared at step 503 with a minimum temperature T3 at which oxidation of unburnt fuel to a desired extent in the oxidation catalyst is regarded as possible. If the post-treatment system temperature T (the oxidation catalyst temperature) is below said temperature T3, the method moves on to step 504 to try by engine control to achieve this oxidation catalyst temperature. It is in principle sufficient that the oxidation catalyst reaches a temperature of about 250°C. When engine control has begun, the method goes back to step 502 to see whether a desired temperature has been reached.

If conversely the oxidation catalyst temperature is above said minimum temperature T3, which may therefore be of the order of about 250°C, and oxidation of unburnt fuel in the oxidation catalyst to a desired extent is therefore regarded as

possible, the method moves on to step 505.

At step 505, a raising of the temperature of the oxidation catalyst commences, which may be achieved as described below. Its temperature needs to be over 400°C to enable the

accumulated sulphur to react with the passing exhaust flow. The binding of the sulphur to the metals of the coating may thus be broken, which may at least partly be by the oxidation of hydrocarbons (fuel) reacting with oxygen atoms in, for example, sulphates so that sulphur in new molecule form becomes detached from the catalyst coating and is then carried by the exhaust flow through the post-treatment system and/or becomes attached again downstream of the previous location. The chemical process in this respect may be conducted in various ways and is not described in more detail here, but when the bonding of the sulphur with the coating metal is broken the released/newly formed sulphur molecule, e.g.

sulphur dioxide, sulphuric acid or sulphite, is carried by the exhaust flow out of the oxidation catalyst or becomes attached again. The same sulphur atom may become attached and be released many times during its passage through the post- treatment system.

Heating may thus result in the oxidation catalyst being detoxified and recovering its original performance as regards NO2 conversion. The desulphuration process follows the Arrhenius equation, so the process rate increases with rising temperature. In other words, for best desulphuration effectiveness, as high a temperature as possible needs to be achieved in the oxidation catalyst. As mentioned above, however, there are often different temperature tolerances with respect to the

components of the post-treatment system.

The temperature rise is achieved by supplying unburnt fuel to the exhaust flow, in which, since the temperature T of the oxidation catalyst is above said minimum temperature T3, it then at least partly oxidises, with the associated release of heat. Step 505 therefore determines an appropriate amount of fuel for supply to the exhaust flow, and fuel injection then commences. The amount of fuel supplied may for example depend on oxidation catalyst temperature, current volume of exhaust flow, engine load, current vehicle speed etc., or may also be some predetermined constant amount, in which case the supply of fuel is controlled on the basis of the oxidation catalyst's temperature T with the object of reaching a first temperature Tl, the supply of fuel being halted when a desired first oxidation catalyst temperature Tl is or will be reached.

The supply of unburnt fuel to the exhaust flow may take place in various different ways, e.g. the post-treatment system may comprise, in the exhaust system upstream of the oxidation catalyst, an injector (not depicted) which may be used to inject fuel into the exhaust flow.

Alternatively, fuel may be supplied to the exhaust flow by injection in the engine's combustion chambers (e.g. the engine's cylinders) so late during the combustion cycle that none or only some of the fuel intended for the regeneration burns in the cylinders, in which case fuel will accompany the exhaust flow to the post-treatment system. The present invention is suited to both types of fuel injection.

When fuel supply has commenced according to any suitable model as above, the method moves on step 506 to determine whether the oxidation catalyst's temperature T is above a first temperature Tl which is substantially higher than said minimum temperature T3. This first temperature Tl would result in the possibility of downstream components being damaged by reaching it. According to the present invention a cyclic raising of the temperature is effected for the component where sulphur storage has been found, i.e. in this case the oxidation catalyst 205, and the cyclic temperature rise means that the temperature of downstream components will not rise to

unacceptable levels. So long as the temperature Tl is not reached, the method goes back to step 505 to continue fuel supply. If Tl is reached, the supply of fuel to the exhaust flow is halted and the method moves on to step 507 to

determine whether the temperature T has dropped to a level T2 which is lower than Tl. So long as such is not the case, the method will stay at step 507 before moving on to step 508 when the temperature T has dropped to T2.

Said first temperature Tl and said second temperature T2 are determined/chosen such that the temperature at components situated downstream of the oxidation catalyst, e.g. the particle filter 202 and the SCR catalyst 201, will not rise above permissible levels.

Said first temperature Tl is preferably set to a level which is as high as possible but which, in combination with the lowering of the oxidation catalyst' s temperature T to the chosen second level T2, provides assurance that the

temperature at components situated downstream of the oxidation catalyst, e.g. the particle filter 202 and the SCR catalyst 201, will not rise above permissible levels. For example, the most temperature-sensitive component in the present example may be the SCR catalyst 201, and the method according to the invention makes it possible to ensure that the temperature after the particle filter/at the SCR catalyst's inlet 203 does not reach unacceptable levels.

Said second temperature T2 may be determined/chosen on the basis of calculation/modelling of a temperature pattern for the oxidation catalyst 205 and/or appropriate temperatures at components situated downstream of the oxidation catalyst, e.g. the particle filter 202 and the SCR catalyst 201.

Alternatively to, or in combination with, the above, said second temperature T2 may be determined/chosen on the basis of measurements of temperature pattern of the oxidation catalyst and/or appropriate temperatures at components situated

downstream of the oxidation catalyst, e.g. the particle filter and the SCR catalyst.

Said second temperature T2 is determined/chosen such as to provide assurance that when the method according to the invention is employed the temperature at components situated downstream of the oxidation catalyst, e.g. the particle filter and the SCR catalyst, will not rise above permissible levels.

Step 508 determines whether the desulphuration has been completed. If such is the case, the method ends at step 509. This determination may be done in any suitable way, e.g. by determining the amount of time t for which desulphuration has proceeded, and stopping it when it has proceeded for at least an appropriate first amount of time tl. Alternatively, this may for example be determined by a requirement that heating to the temperature Tl must take place a certain number of times. The determination may also be based on the differential pressure change across the particle filter during an attempted regeneration. In other words, in this case the differential pressure decrease which occurs during regeneration over for example a certain amount of time has to be of at least some appropriate magnitude, otherwise the regeneration rate is regarded as too slow and desulphuration is therefore not regarded as effected to the desired extent. The

desulphuration may also be ended on the basis of other

suitable parameters.

So long as desulphuration is to continue, the method goes back to step 502 from step 509.

If fuel supply is to take place continuously in order to keep the oxidation catalyst at a constant temperature Tl, the temperature of the downstream components would also rise to corresponding high levels. Thus keeping the oxidation

catalyst's temperature T constant at the level Tl would cause the temperature of downstream components to become higher than permissible. The method according to the present invention affords the advantage that a higher desulphuration rate may be achieved by raising the oxidation catalyst's temperature T to a first level Tl which is above a temperature tolerance of at least one of the components situated downstream of the

oxidation catalyst, e.g. the particle filter 202 and the SCR catalyst 201, in the post-treatment system 200. Stopping this temperature rise when the temperature Tl is reached, i.e.

stopping the fuel supply, and allowing the oxidation

catalyst's temperature T to drop to said second level T2 will result in the temperature of components situated downstream of the oxidation catalyst not being able to rise to unacceptably high levels before it again begins to drop. Pulsing the oxidation catalyst's temperature T thus makes it possible locally to allow substantially higher temperatures, with associated increase in reaction rate, while at the same time this heat wave will be damped by subsequent portions of the oxidation catalyst and, for example, a particle filter 202, so that for example an SCR catalyst 201 situated downstream of the filter is exposed to a substantially lower temperature than the oxidation catalyst. The method according to the present invention may thus be used to raise the maximum temperature in the oxidation catalyst by a relatively large number of degrees without neglecting other requirements, while at the same time also creating potential for desorbing as much stored sulphur as possible.

Fig. 6 is an example of a temperature diagram for part of a desulphuration process according to the invention in a system of the type depicted in Fig. 2, where a fuel supply according to step 505 begins at t=t0. The continuous curve 601

represents the oxidation catalyst temperature, which is pulsed between levels Tl and T2 as above, and the curve 602

represents the temperature after the particle filter 202 and hence at the inlet to the SCR catalyst. The fuel supply stops at t=tl before resuming at t=t2, stopping again at t=t3, and so on. As may be seen, the SCR catalyst 201 will be exposed to a substantially lower temperature than the oxidation catalyst because of the damping to which the heat wave is subjected by the thermal inertia of the components.

The method referred to may also be further improved. During the pulsing of the temperature T above, the engine 101 may also be adjusted to run at different lambda values. As is well known, lambda=l means a fuel/air ratio which results in stoichiometric combustion, and larger and smaller lambda values respectively represent air surplus and air shortage during the combustion. The composition of the exhaust flow may be directed towards nearly stoichiometric ratios (i.e. lambda=l) when fuel is added. In other words, in this case the engine is run at the least possible lambda, in which case fuel is then supplied as far as possible without exceeding temperature requirements for arriving at nearly stoichiometric ratios. This may entail keeping the oxygen level at the inlet to the oxidation catalyst to a minimum, in which case

hydrocarbons will to the greatest possible extent react with oxygen atoms in sulphur/metal combinations. In one embodiment said engine and/or fuel supply are controlled in such a way as to achieve a lambda value below 1.5 at the inlet to the oxidation catalyst.

When thereafter the fuel supply is shut off and the

temperature drops back, the engine's settings may be altered so that a considerably higher lambda is employed and a

definite oxygen surplus instead occurs in the system. In this way it is possible to vary the conditions in the post- treatment system in order to optimise the desorption of sulphur which is combined in various compounds. The advantage of varying the conditions in this way is that different sulphur compounds such as base metal/sulphur and noble

metal/sulphur disassociate and are desorbed at different rates .

The rate at which the temperature T is raised, which may be varied by the rate at which fuel is supplied, and lambda values may be adapted according to the metals present in the post-treatment system in order to achieve as optimum

desulphuration as possible.

The invention is exemplified above in relation to the system depicted in Fig. 2. The post-treatment system set-up depicted in Fig. 2 is commonly employed in heavy vehicles, at least in jurisdictions where stricter emission requirements prevail.

In one embodiment, the particle filter comprises instead noble metal coatings so that the chemical processes taking place in the oxidation catalyst take place instead in the particle filter, in which case the post-treatment system will therefore not have a DOC. The invention is nevertheless applicable here too, with for example the temperature Tl adapted to suit this system instead.

The post-treatment system 200 may also comprise more

components than exemplified above, e.g. it may, in addition to, or instead of, said DOC 205 and/or SCR 201, comprise an ASC (ammonia slip catalyst) (not depicted) , in which case appropriate temperature adaptation may here too be effected.

The present invention is exemplified above in relation to vehicles. The invention is nevertheless also applicable to any other means of transport /processes in which particle filter systems as above are applicable, e.g. watercraft or aircraft with combustion processes as above.

Further embodiments of the method and the system according to the invention are referred to in the attached claims. It should also be noted that the system may be modified according to different embodiments of the method according to the invention (and vice versa) and that the present invention is in no way restricted to the embodiments described above of the method according to the invention, but relates to and

comprises all embodiments within the protective scope of the attached independent claims.