US20140048029A1

US20140048029A1 - Process for operating an engine supplied with a fuel containing a catalyst for regenerating a particulate filter

Info

Publication number: US20140048029A1
Application number: US14/004,832
Authority: US
Inventors: Virginie Harle; Michael Lallemand; Thierry Segue-Long
Original assignee: Rhodia Operations SAS
Current assignee: Rhodia Operations SAS
Priority date: 2011-03-17
Filing date: 2012-03-15
Publication date: 2014-02-20
Also published as: CN103502402B; BR112013023746A2; JP2016200149A; CN103502402A; WO2012123540A1; KR20130133867A; KR101605597B1; EP2686410A1; FR2972766A1; FR2972766B1; JP2014511960A

Abstract

The invention relates to a process for operating an internal combustion engine of a vehicle equipped with an exhaust system comprising a catalysed particulate filter in which the engine is supplied with a fuel containing a catalyst for regenerating the particulate filter. The process is characterized in that the concentration of catalyst in the fuel varies discontinuously.

Description

The present invention relates to a method for operating an internal combustion engine, notably a diesel engine, supplied with a fuel containing a catalyst for regenerating a particulate filter. This method applies to motor vehicles equipped with a catalyzed particulate filter for eliminating the black smoke from the exhaust gases of the engine.
In order to meet the new emissions control standards for vehicles, notably diesel vehicles, these vehicles are gradually being fitted with particulate filters (PFs). Such is already the case in Europe since the Euro 5 emission standard came into being. In most instances, a catalyst is used to assist with periodically burning off the soot captured on the filter and thus regenerates the PF.
The PF is regenerated by periodically increasing the temperature upstream of the PF to a temperature that is high enough to burn off the soot and thus regenerate the PF.
This temperature is typically above 650° C. and fuel is therefore generally burnt in the engine (post-injection) or on an oxydation catalytic converter upstream of the PF in order to achieve this level of heat. This is because the temperature of the exhaust gases for diesel engines is generally markedly lower, typically below 400° C. The temperature of the exhaust gases also tends to drop with new combustion technologies such as homogenous charge combustion of the HCCl type. It is also very low, often below 250° C., when the vehicle is used under certain conditions such as during use in town.
A second important parameter is the duration of the PF regeneration, i.e. the length of time for which the temperature upstream of the PF needs to be kept at a high level. Beyond the economic and environmental impact of greater additional fuel consumption, in some instances, such as for journeys around town which are often of short duration, it is not possible to maintain these conditions for longer enough to regenerate the PF.
It will be appreciated that there is some advantage to be had in being able to reduce the frequency and the duration of these regenerations and also in being able to perform them at a lower temperature. This is because it has the effect of reducing the fuel consumption of the vehicle because a lower quantity of fuel is used for the post-injection. As a result, the vehicle emissions of greenhouse gases (CO₂) are reduced.
This also allows use to be made in the PFs of materials that do not need to be as heat-resistant as, for example, silicon carbide, and therefore materials which are less expensive. Moreover, reducing the duration of the post-injections also proves profitable with regard to other criteria such as the life of the engine and of certain components such as the high-pressure fuel injectors or even the engine oil change intervals.
In order to meet these objectives, a catalyst which encourages this regeneration is generally used in two key ways:

- an oxydation catalyst is introduced into the porosity of the walls of the PF: the PF is then said to be a catalyzed particulate filter or alternatively is referred to as a Catalyst Soot Filter (CSF). The catalyst is generally made up of a precious metal, such as platinum and oxides of transition metal such as alumina or even reducible oxides such as oxides based on cerium, on cerium and zirconium or more generally on rare earths. This technology is currently widely used on the recent vehicles that meet the Euro 5 emission standard in Europe;
- use of a PF regeneration additive which is borne by the fuel supplied to the engine, alternatively known as a Fuel Borne Catalyst (FBC). Various FBC additives are known, notably those based on cerium and/or on iron. This technology is currently likewise fitted to diesel vehicles.

The second way is generally more effective and allows the PF to be regenerated under all running conditions, notably during town driving, and more economically and in a way that is more environmentally friendly.
However, the major disadvantage with the FBC technology lies in the complexity of implementing it, notably in ensuring that the fuel has the most constant possible additive concentration as is currently implemented in vehicles fitted with this technology. Typically, the objective will be to maintain an additive concentration that does not change significantly in the fuel, i.e. an additive concentration that typically exhibits variations in concentration of under 20% or even of under 10%.
The systems that allow the introduction of FBC catalytic additives that assist with regenerating the PFs into the fuel generally rely on a large-sized reservoir with a minimum volume of 2 to 3 liters containing the reserve of additive and that needs to be installed in areas near the fuel tank.
The current methods for metering the additive also call for high-precision metering pumps that has to be controlled using an additional and dedicated electronic unit. This electronic unit is generally dependent on the vehicle electronic central control unit or ECU. This metering device needs to be managed very accurately in order to ensure that the additive content in the fuel is high enough to allow a good regeneration of the PF but no so high as to cause premature fouling of the PF with the inorganic residues from PF regeneration which remain trapped within it. In general, when the level of fuel in the tank increases following the addition of fuel, the ECU communicates this information to the computer and the computer informs the pump how much additive to inject into the tank in order to keep the additive concentration in the fuel constant at all times.
These metering pumps are extremely accurate and very expensive. The use of such methods also involves correctly slaving the metering system and checking its state of operation. The systems are therefore complex and, as a result, expensive.
It is an object of the invention to propose a method that is not as complicated and therefore not as expensive to implement as the known methods.
To this end, the method of the invention is a method for operating an internal combustion of a vehicle equipped with an exhaust system comprising a catalyzed particulate filter (CSF), in which the engine is supplied with a fuel containing a catalyst for regenerating the particulate filter, and it is characterized in that the catalyst concentration in the fuel varies discontinuously.
The method of the invention allows the CSF to be regenerated effectively, notably at low temperature, without requiring the complex systems of the prior art for keeping the concentration in the fuel at a constant value.

Other features, details and advantages of the invention will become even more fully apparent from reading the description which follows given with reference to the attached drawing and in which:

the single FIG. 1 gives the catalyst concentration of a fuel over time and as a function of the level of fill of the fuel tank.

The essential feature of the method of the invention is that the catalyst concentration in the fuel varies discontinuously. What that means is that, unlike in the known methods, this concentration is not constant but can vary over time and what is more that it varies non-continuously. Thus, in a very short space of time or instantaneously it can adopt different values. It may be zero and vary within ranges which may, for example, vary by a factor 0 to 30, more particularly from 0 to 20. More particularly still, these ranges may vary from 0 to 15 and notably from 0 to 5. This concentration can thus remain constant at a certain value for a certain length of time and then change in a very short space of time or instantaneously to another value and then remain constant for another period of time.
The method of the invention can be implemented in various alternative ways.
According to a first alternative, the method is implemented under conditions such that during the CSF charging period, the catalyst concentration in the fuel varies just once, such that it increases. Thus, it changes from a value V₀which may be zero to a value V_nsuch that V_n>V₀.
The filter charging period means the period during which the exhaust gases flow through the CSF and during which the latter becomes progressively laden with soot. These are all the engine operating periods outside of the filter regeneration period.
According to a second alternative of the method of the invention, and still during the particulate filter charging period, the catalyst concentration in the fuel varies several times, such that it increases. Thus it passes from a value V₀which may be zero to a value V_nand then to another value V_n+1, these values being such that V_n+1>V_n>V₀.
According to another alternative, the method is implemented in such a way that during the particulate filter charging period, the catalyst concentration in the fuel varies once or several times such that it decreases. Thus, it may pass from a value V₀which is non-zero to a value V_nand then possibly to another value V_n+1, these values being such that V_n+1<V_n<V₀.
In the case of the second or third alternative, the number of times the variation occurs may be unlimited.
Finally, according to yet another alternative, the catalyst concentration in the fuel can be made to vary several times such that it increases or decreases during the CSF charging period, it being possible for this concentration to be zero over a period time.
The invention can be used with any type of CSF regeneration catalyst. These catalysts are well known. More particularly, and solely by way of example, this catalyst may take the form of a colloidal dispersion. The colloids of this colloidal dispersion may be based on a compound of a rare earth and/or of a metal which is chosen from groups IIA, IVA, VIIA, VIII, IB, IIB, IIIB and IVB of the periodic classification.
They may more particularly be based on compounds of cerium and/or iron.
It is also possible to use colloidal dispersions which contain detergent compositions.
The periodic classification of the elements to which reference is made here is the one published in the supplement to French Chemical Society Bulletin No. 1 (January 1966).
By way of example of colloidal dispersions mention may be made of those described in patent applications EP 671205, WO 97/19022, WO 01/10545 and WO 03/053560, the latter two notably describing dispersions based on compounds of cerium and of iron respectively, these dispersions also containing an amphiphilic agent.
Mention may also be made of application WO 2010/150040 which describes a colloidal dispersion based on a compound of iron, an amphiphilic agent and a detergent composition containing a quaternary ammonium salt.
The quaternary ammonium salt may be the product of reaction:
(i) of at least one compound which may contain:
(a) the condensation product of a hydrocarbyl-substituted acylating agent and of a compound having an oxygen or nitrogen atom capable of condensing the acylating agent, the condensation product having at least one tertiary amino group;
(b) a polyalkene-substituted amine having at least one tertiary amino group; and
(c) a Mannich reaction product having at least one tertiary amino group, the Mannich reaction product being derived from a hydrocarbyl-substituted phenol, an aldehyde and an amine; and
(ii) a quaternizing agent suitable for converting the tertiary amino group of compound (I) to a quaternary nitrogen.
The quaternizing agent may include dialkyl sulfates, benzyle halides, hydrocarbyl substituted carbonates; hydrocarbyl substituted epoxides in combination with an acid or mixtures thereof.
Catalyzed PFs are also well-known. They generally comprise a catalyst based on at least one metal selected from platinum or metals in the platinum group such as palladium for example. Combinations of platinum with these metals or even of these metals with one another are of course possible.
The metal of the catalyst can be incorporated into the filter or applied to the filter in a known way. It may for example be included in a coating (washcoat) which is itself applied to the filter. This washcoat may be selected from alumina, titanium oxide, silica, spinelles, zeolites, silicates, crystalline aluminium phosphates or mixtures thereof. Alumina may more particularly be used. The washcoat may also contain reducible materials capable of assisting directly or indirectly in the burning of the soot. By way of example mention may be made of material based on cerium oxides, such as cerine, mixed oxides based on cerium and zirconium, possibly doped, or even oxides of manganese.
If the PF catalyst is a catalyst to assist with burning off the soot, it is therefore present on the filter in relatively low quantity, namely in general in a quantity of at most 70 g/foot³(2.5 g/dm³). This quantity is expressed in terms of mass of elemental metal, for example mass of platinum, with respect to the volume of the PF. This quantity can more particularly be at most 60 g/foot³(2.1 g/dm³) and more particularly still, of at most 50 g/foot³(1.8 g/dm³).
The mass concentration of regeneration catalyst in the fuel, notably when this is in the form of a colloidal dispersion, will advantageously be comprised between 0 and 30 ppm, this content being expressed in terms of elemental metal such as iron in the case of an iron-based colloidal dispersion. The catalyst content of the soot emitted by the engine, expressed in terms of mass of elemental metal, may be comprised between 0 and 8%, depending on the regeneration catalyst content of the fuel, on the fuel consumption of the vehicle and on its production of soot.
When implementing the method of the invention, the vehicle will run on a fuel containing a variable content of regeneration catalyst, it being possible for this content to be zero over certain periods. The soot produced by the engine will be more or less rich in elements that take an active role in regenerating the CSF, depending on the level of additive in the fuel.
Thus, the CSF will alternately become laden with soot containing no or variable concentrations of regeneration catalyst additive. The fuel used during the periodic regeneration of the CSF may or may not contain additive.
Regeneration is then performed in the conventional way under the control of the vehicle ECU using the technology chosen by the manufacturer.
The advantage of the invention is that the additive can be introduced into the fuel by simple systems which are less expensive than the known ones and the metering strategy of which is simpler and quicker to install on the vehicle. Notable preference is given to systems which require no interface with the central control system ECU of the vehicle, as this makes installing it on the vehicle simpler.
Simple embodiments allowing additive to be introduced in quantities that differ and vary over time will be given hereinbelow.
A first embodiment involves adding a dose of additive, generally liquid, by hand, this being poured into the vehicle fuel tank. The dose of additive is calculated so that the content of substance that is active in regenerating the CSF is high enough to promote combustion of the soot trapped in the CSF. By way of example, for an additive based on a colloidal suspension of iron particles such as dispersion C in example 3 of patent application WO 2010/150040, the elemental iron content of the fuel just after manual application of additive may advantageously be comprised between 2 and 30 ppm in terms of mass of metallic iron, more particularly between 5 and 20 ppm in terms of mass of metallic iron.
This simple means allows additive to be added to the fuel when necessary: in particular at a regular frequency when the vehicle is used predominantly in town—for example by adding additive every 1000 to 3000 km. This means may also be used when the indicator lamp on the vehicle instrument panel signals a fault with the pollution reduction means.
FIG. 1 illustrates one example of a curve of regeneration catalyst concentration in the fuel that can be obtained when a dose of additive is regularly added to the tank by hand, in this instance every 2200 km (or 44 hours of running). This example considers a fixed fuel consumption of 6 l/100 km, a fixed speed of 50 km/h so a fixed fuel consumption of 3 l/h. Typically, as soon as a dose of additive is added (event 1, denoted Ev1 in the FIGURE: a volume of regeneration catalyst making it possible to achieve a metallic iron content of 15 ppm in the 40 liters of fuel present in the tank), the iron content increases sharply, in this instance passing from 0 to 15 ppm. This iron content is constant over time until fuel is added to the tank, this leading to a dilution of the concentration of iron in the respective proportions of the remaining volume of additive-containing fuel and the volume of fuel (which therefore does not contain additive) added (event 2, denoted Ev2: 40 L of fuel (containing no additive) added to the 20 L of fuel remaining in the tank). This event is repeated four times in this example. On each addition, the iron concentration drops proportionally.
The FIGURE also indicates the periods of CSF regeneration (identified with stars in the FIGURE)—regenerations occurring at regular 700 km intervals, namely every 14 hours of operation. It will be noted for example that the soot loading of the CSF that corresponds to the first regeneration has come about with a vehicle operating 50% of the time on a fuel containing no additive and 50% of the time on a fuel to which 15 ppm by weight of iron has been added. Example 1 hereinbelow illustrates the benefit to be had through a CSF regeneration engine test carried out under these charging conditions.
It will thus be noted that the charging of the CSF with soot occurs with a fuel the additive concentration of which can vary and that the benefit is obtained by using a very simple system that involves pouring a dose of additive into the tank by hand in this instance every 2200 km (44 hours).
Another embodiment can be used by fitting the vehicle with a simple and autonomous means, i.e. a means not connected to the centrally ECU of the vehicle, for introducing the regeneration catalyst. This means may consist in adding a small FBC tank, typically of 1 l or less and a metering pump that allows a given quantity of additive to be injected into the fuel tank at regular intervals. By comparison with the existing system, the pump can be less complicated and therefore less expensive because the quantity injected will be fixed. No interface with the ECU is required because the pump can be programmed to inject for example at regular intervals (intervals in time such as every 5 to 10 hours and/or intervals in terms of distance such as every 1000 to 3000 km). Local devices located on the pump, such as the application of power or a GPS chip may inform the pump that the vehicle is driving along or may provide the distance that the vehicle has covered.
Examples will now be given.

EXAMPLE 1

A diesel engine supplied by the Volkswagen group (4-cylinder, 2-liters, turbocharged with air cooling, 81 kW) was used on an engine test bed. The exhaust line mounted downstream is a commercial line made up of an oxidation catalytic converter containing a washcoat based on platinum and on alumina followed by a commercial CSF containing a washcoat based on platinum and on alumina (total filter volume 3 L).
The fuel used is a commercial fuel in accordance with standard EN590 DIN 51628 containing less than 10 ppm of sulfur and containing 7% by volume of FAME or fatty acid methyl ester. When an FBC regeneration catalyst is used, the fuel has added to it the quantity of FBC additive that will make it possible to achieve various metallic iron contents expressed in the form of ppm by mass with respect to the mass of the fuel. The FBC additive used is an additive based on a colloidal dispersion of particles of iron such as dispersion C of example 3 of patent application WO 2010/150040, the elemental iron content of this additive being 4.3% by mass of metallic iron.
The iron content of the additive-containing fuel is monitored directly in the organic liquid using the X-ray fluorescence technique.
The test is performed in two successive steps: a CSF soot charging step, followed by a step of regenerating the CSF. The conditions for these two steps are strictly identical for the various tests, the exception being the kind of fuel used (whether or not it contains additive).
The charging phase is performed by running the engine at a speed of 3000 revs/min (rpm) and using a torque of 45 Nm for approximately 6 hours. This charging phase is halted when the CSF has become ladened with 12 g of particulates (or soot). During this phase, the temperature of the gases upstream of the CSF is 230 to 235° C. Under these conditions, particulate emissions are around 2 g/h.
After this charging phase, the CSF is removed and weighed in order to check the mass of particulates with which it has become laden during this phase.
The CSF is then refitted on the test bed and heated up by the engine which is returned for 30 minutes to the charging running conditions (3000 rpm/45 Nm). The engine conditions are then altered (torque 80 Nm/2200 rpm) and the engine central control unit (ECU) oversees a post-injection, allowing the temperature upstream of the CSF to be raised to 500° C. and regeneration thereof to begin. These conditions are maintained for 60 minutes, this time being measured from the start of the post-injection.
In all cases, the fuel used for regeneration corresponds to the last fuel used for the CSF charging phase.
The effectiveness of CSF regeneration is measured through two parameters:

- the mass of soot burnt off during regeneration, which is calculated from the weighing of the CSF before charging (Mo), after charging (Mc) and at the end of regeneration (Mr). The % of soot burnt off after the 60 minutes of regeneration is expressed as follows:

Total % of soot burnt=(Mc−Mr)/(Mc−Mo)*100

- the mass of soot burnt at any moment t in the regeneration calculated from the change in pressure drop across the CSF at each moment, DPt, considering that the pressure drop at the start of regeneration (DPc) corresponds to that of a CSF laden with the mass of soot (Mc−Mo) and the pressure drop after the 60 minutes (DPr) corresponds to that of the CSF laden with the soot that has not been burnt (Mr−Mo).

% of soot burnt(t)=((DPc−DPt)/(Dpc−Dpr))*total % of soot burnt
In general, the higher these parameters, the more effective the regeneration.
Various tests were carried out using different fuels during the charging of the CSF.
Three reference tests (not in accordance with the invention) were carried out either using a fuel containing no additive (test 1) or using a fuel to which additive is added throughout the charging and regeneration of the CSF (test 10 with a fuel additive content at 15 ppm of iron and test 11 with a fuel additive content at 3 ppm of iron).
8 tests (according to the invention) were carried out using a fuel containing no additive at the start of CSF charging (fuel No. 1) then a fuel containing additive (fuel No. 2) at the end of charging (tests 2 to 5 and 8 to 9) or, in the reverse order i.e. fuel containing additive at the start of charging followed by a fuel containing no additive (tests 6 to 7).
Each of the tests represents either a respective charging time with and without an additive-containing fuel, or a variation in the amount of FBC additive contained in the fuel.
Table 1 compares the results obtained during CSF regeneration expressing the % of soot burnt in total, i.e. at the end of the regeneration period (1 hour) or at the start of regeneration (20 minutes).
At each of the points in time a comparison is made against the theoretical effectiveness obtained by calculating the mean of the effectiveness of the CSF with a fuel containing no additive (test 1) and that of the CSF laden with an additive-containing fuel (test 10).

TABLE 1

Results of the CSF regeneration engine tests using various fuels

	% of time
	during

			which fuel	% soot burnt in	% soot burnt
Test			No 2 was	total in 1 hour	after 20 min

No	Fuel No 1	Fuel No 2	used***	E**	Th**	E*	Th*

1*	Without	Without	N/A	60	N/A	39	N/A
	additive	additive
2	Without	15 ppm	50%	85	75	72	64
	additive	iron
3	Without	15 ppm	33%	84	70	68	55
	additive	iron
4	Without	15 ppm	25%	77	67	56	51
	additive	iron
5	Without	15 ppm	15%	78	64	54	46
	additive	iron
6	15 ppm	Without	50%	87	75	80	64
	iron	additive
7	15 ppm	Without	33%	83	70	69	55
	iron	additive
8	Without	7 ppm	33%	84	70	62	55
	additive	iron
9	Without	3 ppm	50%	86	75	68	64
	additive	iron
10*	15 ppm	15 ppm	N/A	90	N/A	90	N/A
	iron	iron
11*	3 ppm	3 ppm	N/A	88	N/A	88	N/A
	iron	iron

*comparative tests not in accordance with the invention
**E = experimental Th = theoretical
***% expressed with respect to the total filter charging time

It will first of all be noted that the addition of an FBC to the fuel throughout the CSF charging period (tests 10 and 11) allows a great increase in the effectiveness of regeneration because this regeneration is nearly complete (88 to 90% of soot burnt) after 20 minutes at 500° C., the iron concentration (3 to 15 ppm) has little impact on the regeneration.
Conversely, when a fuel containing no additive is used (test 1), regeneration is incomplete (60% after 1 hour) and also occurs far more slowly (39% regeneration after 20 minutes).
Charging the CSF using an alternation of fuel containing no additive and then fuel containing additive (or vice versa) allows a great increase in the effectiveness of CSF regeneration.
Charging 50% of the time with a fuel containing an additive of 15 ppm of iron (test 2 or 6) makes it possible to achieve near-total (85 to 87%) regeneration at the end of the test and regeneration which is very advanced after 20 minutes of regeneration (72 to 80%).
Unexpectedly, it is found that the values observed experimentally are significantly higher than the theoretical values when considering the contribution made by the FBC introduced and by the fuel containing no additive.
Test 2 represents the CSF charging conditions described for the charging of the CSF at the time of its first regeneration in FIG. 1.
These conclusions are valid whatever the proportion of the time that the CSF is charged with a fuel containing an additive (15% to 50%) and regardless of the order in which the fuel containing the additive is used (whether this is used at the beginning or at the end of the charging period).
Moreover, it may be seen that the beneficial synergistic effect can be observed with very small quantities of FBC in the fuel as illustrated in tests 8 and 9.

EXAMPLE 2

Another series of CSF regeneration engine tests was conducted using the same protocol and the same equipment as those described in example 1.
Here, the CSF was charged while varying the FBC (the same one as in example 1) concentration in the fuel more frequently.
The CSF chargings were therefore carried out using the sequence of fuels as described in table 2.

TABLE 2

Test No	CSF charging procedure

1*	Fuel containing no additive throughout the charging
12	1 hour with a fuel containing no additive
	1.5 hours with a fuel to which 1 ppm iron has been added
	1.5 hours with a fuel to which 3 ppm iron has been added
	1 hour with a fuel to which 5 ppm iron has been added
	Procedure equivalent to a fuel to which 2.2 ppm of iron added
	continuously has been added
13	1.5 hours with a fuel to which 5 ppm has been added
	1 hour with a fuel containing no additive
	1 hour with a fuel to which 3 ppm iron has been added
	1.5 hours with a fuel containing no additive
	Procedure equivalent to a fuel to which 2.1 ppm of iron added
	continuously has been added
11*	Fuel to which 3 ppm iron has been added throughout the
	charging

*comparative tests not in accordance with the invention

Table 3 compares the results obtained during CSF regeneration expressing the % of soot burnt in total, i.e. at the end of the regeneration period (1 hour) or at the start of regeneration (20 minutes).

TABLE 3

	% of soot burnt in total
Test No	after 1 hour	% of soot burnt after 20 min

1	60	39
12	91	88
13	88	85
10	90	90

It will be noted that the addition of FBC additive to the fuel (tests 10, 12 and 13) allows an increase in the effectiveness (near complete regeneration and improved regeneration dynamics) by comparison with the regeneration of the CSF charged using a fuel containing no additive (test 1). Addition at different doses over time, including incorporating periods during which no additive is added to the fuel (tests 12 and 13) leads to the same result as use of a fuel, the additive content of which is perfectly controlled over time (test 10).

Claims

1. A method for operating an internal combustion of a vehicle equipped with an exhaust system comprising a catalyzed particulate filter, the method comprising supplying the engine with a fuel containing a catalyst for regenerating the particulate filter, characterized in that the catalyst concentration in the fuel varies discontinuously.

2. The method as claimed in claim 1, wherein the catalyst concentration in the fuel increases only once during a particulate filter charging period.

3. The method as claimed in claim 1, wherein the catalyst concentration in the fuel increases several times during a particulate filter charging period.

4. The method as claimed in claim 1, wherein the catalyst concentration in the fuel decreases once or several times during a particulate filter charging period.

5. The method as claimed in claim 1, wherein the catalyst concentration in the fuel increases or decreases several times during a particulate filter charging period.

6. The method as claimed in claim 1, characterized in that the catalyst that regenerates the particulate filter takes the form of a colloidal dispersion.

7. The method as claimed in claim 6, characterized in that the colloids of the colloidal dispersion are based on a compound of a rare earth; a compound of a metal which is chosen from groups IIA, IVA, VIIA, VIII, IB, IIB, IIIB and IVB of the periodic classification; or a mixture thereof.

8. The method as claimed in claim 7, characterized in that the colloids of the colloidal dispersion are based on a compound of cerium and/or iron.

9. The method as claimed in claim 6, characterized in that the colloidal dispersion contains a detergent composition.

10. The method as claimed in claim 8, characterized in that the colloidal dispersion is based on a compound of iron, an amphiphilic agent and a detergent composition containing a quaternary ammonium salt.

11. The method as claimed in claim 2, characterized in that the catalyst that regenerates the particulate filter takes the form of a colloidal dispersion, wherein the colloids of the colloidal dispersion are based on a compound of a rare earth; a compound of a metal which is chosen from groups IIA, IVA, VIIA, VIII, IB, IIB, IIIB and IVB of the periodic classification; or a mixture thereof.

12. The method as claimed in claim 11, characterized in that the colloids of the colloidal dispersion are based on a compound of cerium and/or iron.

13. The method as claimed in claim 3, characterized in that the catalyst that regenerates the particulate filter takes the form of a colloidal dispersion, wherein the colloids of the colloidal dispersion are based on a compound of a rare earth; a compound of a metal which is chosen from groups IIA, IVA, VIIA, VIII, IB, IIB, IIIB and IVB of the periodic classification; or a mixture thereof.

14. The method as claimed in claim 13, characterized in that the colloids of the colloidal dispersion are based on a compound of cerium and/or iron.

15. The method as claimed in claim 4, characterized in that the catalyst that regenerates the particulate filter takes the form of a colloidal dispersion, wherein the colloids of the colloidal dispersion are based on a compound of a rare earth; a compound of a metal which is chosen from groups IIA, IVA, VIIA, VIII, IB, IIB, IIIB and IVB of the periodic classification; or a mixture thereof.

16. The method as claimed in claim 15, characterized in that the colloids of the colloidal dispersion are based on a compound of cerium and/or iron.

17. The method as claimed in claim 5, characterized in that the catalyst that regenerates the particulate filter takes the form of a colloidal dispersion, wherein the colloids of the colloidal dispersion are based on a compound of a rare earth; a compound of a metal which is chosen from groups IIA, IVA, VIIA, VIII, IB, IIB, IIIB and IVB of the periodic classification; or a mixture thereof.

18. The method as claimed in claim 17, characterized in that the colloids of the colloidal dispersion are based on a compound of cerium and/or iron.