JP6323239B2 - Reducing agent addition device - Google Patents

Description

  The present invention relates to a reducing agent addition device for adding a reducing agent used for NOx reduction to an upstream side of a reduction catalyst in an exhaust passage.

  Regarding a combustion system in which a NOx purification device is provided in an exhaust passage of an internal combustion engine, Patent Document 1 discloses a reducing agent addition device that adds a reducing agent to an upstream side of a reduction catalyst in an exhaust passage. According to this, NOx in the exhaust is reduced on the reduction catalyst of the NOx purification device by the added reducing agent.

  Here, a problem of sulfur poisoning, in which SOx in the exhaust gas is adsorbed by the reduction catalyst and remains to reduce the NOx purification rate, has been conventionally known. With respect to this problem, Patent Document 1 discloses that a fuel used as a reducing agent is activated by plasma and an activated fuel is added. According to this, sulfur adsorbed by the reduction catalyst can be desorbed from the reduction catalyst, and recovery from sulfur poisoning can be achieved.

JP 2007-100572 A

  However, the phenomenon that sulfur is desorbed from the reduction catalyst by the activated fuel as described above requires that the combustion of the internal combustion engine be rich. For this reason, the above recovery method causes a significant deterioration in fuel consumption.

  In addition, sulfur is combined with hydrogen to generate hydrogen sulfide during rich combustion, but since the hydrogen sulfide is a malodorous component, it is not desirable to increase the amount of hydrogen sulfide generated in a rich state as described above.

  The present invention has been made in view of the above problems, and an object thereof is to provide a reducing agent addition device that suppresses generation of hydrogen sulfide and suppresses deterioration of fuel consumption.

  The invention disclosed herein employs the following technical means to achieve the above object. Note that the reference numerals in parentheses described in the claims and in this section indicate the correspondence with the specific means described in the embodiments described later, and do not limit the technical scope of the invention. .

  One of the disclosed inventions is provided in a combustion system in which an NOx purification device (15) for purifying NOx contained in exhaust gas of an internal combustion engine (10) on a reduction catalyst is provided in an exhaust passage (10ex). It is assumed that this is a reducing agent addition device for adding a reducing agent that is a hydrocarbon compound to the upstream side of the reduction catalyst. The reforming means (20, 30) that partially oxidizes the reducing agent and reforms, and the reducing agent that has been reformed by the reforming means at the time of a purification request that requires NOx purification on the reduction catalyst. The purification control means (81a) added to the exhaust passage and the recovery means required to recover the sulfur-poisoned reduction catalyst were reformed by the reforming means on the condition that the internal combustion engine was lean burning And a recovery control means (81b) for adding a reducing agent to the exhaust passage.

  Here, when a substance (partial oxide) obtained by partially oxidizing a hydrocarbon compound flows into the reduction catalyst, the partial oxide chemically reacts with sulfur adsorbed on the reduction catalyst and oxygen in the exhaust. Thus, it has been found that a phenomenon occurs in which sulfur is desorbed from the reduction catalyst and reduced.

  Based on this knowledge, in the above invention, the hydrocarbon compound (reducing agent) is partially oxidized and reformed, and the reformed reducing agent (partial oxide) is added during lean combustion. Therefore, the above-described phenomenon occurs in which the partial oxide chemically reacts with sulfur and oxygen, and sulfur is desorbed and reduced from the reduction catalyst. Therefore, since it is not necessary to make it rich as in the prior art, recovery from sulfur poisoning can be achieved while suppressing fuel consumption deterioration. Moreover, since it is not necessary to make it rich, generation of hydrogen sulfide, which is a malodorous component, can be suppressed.

1 is a schematic diagram showing a reducing agent addition apparatus according to an embodiment of the present invention and a combustion system to which the apparatus is applied. The graph explaining that an oxidation reaction arises in two steps, a cold flame reaction and a hot flame reaction. The figure explaining the reaction path | route of a cold flame reaction. The figure which shows the range which two-step oxidation reaction produces of atmospheric temperature and equivalent ratio. The flowchart which shows the procedure of the control which switches ozone production | generation and reformed fuel production | generation by the reducing agent addition apparatus shown in FIG. The flowchart which shows the process sequence of the reformed fuel production | generation by the strong partial oxidation mode of FIG. The flowchart which shows the procedure which controls the implementation period of S poison recovery process in the reducing agent addition apparatus shown in FIG. The flowchart which shows the procedure of the S poison recovery process of FIG. The graph of the test result showing the effect of S poison recovery by the reducing agent addition apparatus shown in FIG. The graph of the test result showing the time required for S poison recovery | restoration by the reducing agent addition apparatus shown in FIG. The graph of the test result showing that the effect of the S poison recovery differs depending on the reduction catalyst temperature in performing the S poison recovery process according to FIG. The figure explaining the consideration about the principle of S poison recovery by the reducing agent addition apparatus shown in FIG.

  The combustion system shown in FIG. 1 includes an internal combustion engine 10, a supercharger 11, a particulate collection device (DPF 14), a DPF regeneration device (regeneration DOC 14 a), a NOx purification device 15, a reducing agent purification device (purification). DOC16) and a reducing agent addition device. The combustion system is mounted on a vehicle, and the vehicle travels using the output of the internal combustion engine 10 as a drive source. The internal combustion engine 10 is a compression self-ignition type diesel engine, and light oil which is a hydrocarbon compound is used as a fuel used for combustion.

  The supercharger 11 includes a turbine 11a, a rotating shaft 11b, and a compressor 11c. The turbine 11a is disposed in the exhaust passage 10ex of the internal combustion engine 10 and rotates by the kinetic energy of the exhaust. The rotating shaft 11b couples the impellers of the turbine 11a and the compressor 11c to transmit the rotational force of the turbine 11a to the compressor 11c. The compressor 11c is disposed in the intake passage 10in of the internal combustion engine 10, compresses the intake air, and supercharges the internal combustion engine 10.

  A cooler 12 for cooling the intake air compressed by the compressor 11c is disposed on the downstream side of the compressor 11c in the intake passage 10in. The compressed intake air cooled by the cooler 12 is adjusted in flow rate by a throttle valve 13 and then distributed to a plurality of combustion chambers of the internal combustion engine 10 by an intake manifold.

  A regeneration DOC 14a (Diesel Oxidation Catalyst), a DPF 14 (Diesel Particulate Filter), a NOx purification device 15, and a purification DOC 16 are arranged in this order on the downstream side of the turbine 11a in the exhaust passage 10ex. The DPF 14 collects fine particles contained in the exhaust. The regeneration DOC 14a has a catalyst that oxidizes and burns unburned fuel in the exhaust. Due to this combustion, the fine particles collected by the DPF 14 are burned, and the DPF 14 is regenerated to maintain the collection ability. Note that combustion by supplying unburned fuel to the regeneration DOC 14a is not always performed, but temporarily performed at a time when regeneration is necessary.

  A supply pipe 26 of a reducing agent addition device is connected to the exhaust passage 10ex on the downstream side of the DPF 14 and the upstream side of the NOx purification device 15. The reformed fuel generated by the reducing agent addition device is added as a reducing agent from the supply pipe 26 to the exhaust passage 10ex. The reformed fuel is obtained by partially oxidizing a hydrocarbon compound (fuel) used as a reducing agent and reforming it into a partially oxidized hydrocarbon such as an aldehyde, which will be described in detail later with reference to FIG.

  The NOx purification device 15 includes a honeycomb-shaped carrier 15b that carries a reduction catalyst, and a housing 15a that houses the carrier 15b. The NOx purification device 15 purifies NOx contained in the exhaust by reacting NOx in the exhaust with the reformed fuel on the reduction catalyst and reducing it to N2. The exhaust gas contains O2 (oxygen) in addition to NOx, but the reformed fuel reacts selectively with NOx in the presence of O2.

  A reduction catalyst having a function of adsorbing NOx is used. Specifically, when the catalyst temperature is lower than the activation temperature at which the reduction reaction is possible, the reduction catalyst exhibits a function of adsorbing NOx in the exhaust. When the catalyst temperature is equal to or higher than the activation temperature, the adsorbed NOx is reduced by the reformed fuel and released from the reduction catalyst. For example, the NOx purification device 15 having a NOx adsorption function is provided by a reduction catalyst made of silver alumina supported on the carrier 15b.

  The purification DOC 16 is configured by accommodating a carrier carrying an oxidation catalyst in a housing. The purification DOC 16 oxidizes the reducing agent that has not been used for NOx reduction on the reduction catalyst and has flowed out of the NOx purification device 15 on the oxidation catalyst. This prevents the reducing agent from being released into the atmosphere from the outlet of the exhaust passage 10ex. Note that the activation temperature of the oxidation catalyst (eg, 200 ° C.) is lower than the activation temperature of the reduction catalyst (eg, 250 ° C.).

  Next, a reducing agent addition device that generates reformed fuel and adds it to the exhaust passage 10ex from the supply pipe 26 will be described. The reducing agent addition apparatus includes a discharge reactor 20, an air pump 20p, an air tube 23, a reaction vessel 25, a heater 30, an injection valve 40, and the like described in detail below. In particular, the discharge reactor 20 and the heater 30 provide “reforming means” for partially oxidizing and reducing the reducing agent. The discharge reactor 20 included in the reforming unit provides an “ozone generating device” that generates ozone, and the heater 30 included in the reforming unit provides a “heating unit” that heats the reducing agent.

  The discharge reactor 20 includes a housing 22 that forms a flow passage 22a therein, and a plurality of electrodes 21 are disposed in the flow passage 22a. These electrodes 21 have a flat plate shape arranged so as to face each other in parallel, and electrodes to which a high voltage is applied and electrodes having a ground voltage are alternately arranged. The voltage application to the electrode 21 is controlled by a microcomputer (microcomputer 81) included in the electronic control unit (ECU 80).

  The air blown by the air pump 20p flows into the housing 22 of the discharge reactor 20. The air pump 20p is driven by an electric motor, and the electric motor is controlled by the microcomputer 81. The air blown by the air pump 20 p flows into the flow passage 22 a in the housing 22 and flows through the interelectrode passage 21 a that is a passage between the electrodes 21.

  An air pipe 23 that internally forms an air passage 23b is connected to the downstream portion of the discharge reactor 20, and the air that has passed through the interelectrode passage 21a flows through the air passage 23b. A supply pipe 26 is connected to the downstream portion of the air pipe 23. A check valve 24 is attached to the air pipe 23. When the air pump 20p is driven, the check valve 24 is opened against the elastic force of a spring (not shown). When the air pump 20p is stopped, the check valve 24 is closed, and from the supply pipe 26 side. The reformed fuel is prevented from flowing backward to the discharge reactor 20 side.

  A reaction vessel 25 that forms a vaporization chamber 25 a is connected to the downstream side of the check valve 24 in the air pipe 23. The vaporization chamber 25a has a shape of a bag path that branches from the air passage 23b and communicates therewith, and communicates with the air passage 23b through an inflow port 23c formed in the air pipe 23.

  A heater 30 and an injection valve 40 are attached to the reaction vessel 25. The heater 30 has a heat generating portion 31 that generates heat when energized, and the power supply to the heat generating portion 31 is controlled by a microcomputer 81. Specifically, the amount of heat generated is controlled by the microcomputer 81 performing duty control on the amount of power supplied to the heat generating unit 31. The heat generating part 31 is disposed in the vaporizing chamber 25a and heats the fuel injected from the injection valve 40 to the vaporizing chamber 25a. The temperature of the vaporizing chamber 25 a is detected by the temperature sensor 27. The temperature sensor 27 outputs detected temperature information (detected temperature) to the ECU 80.

  The injection valve 40 has a body in which an injection hole is formed, an electric actuator, and a valve body. When the electric actuator is energized, the valve body is opened and fuel is injected from the nozzle hole into the vaporizing chamber 25a. When the electric actuator is turned off, the valve body is closed and fuel injection is stopped. The microcomputer 81 controls the amount of fuel injected per unit time into the vaporizing chamber 25a by controlling energization to the electric actuator. The liquid fuel in the fuel tank 70t is supplied to the injection valve 40 by the pump 70p. The fuel in the fuel tank 70t is also used as the combustion fuel described above, and the fuel used for the combustion of the internal combustion engine 10 and the fuel used as the reducing agent are shared.

  The fuel injected from the injection valve 40 into the vaporizing chamber 25a collides with the heat generating portion 31, and is heated and vaporized. The vaporized fuel flows into the air passage 23b and is mixed with air. As a result, the gaseous fuel is partially oxidized by oxygen in the air and reformed into partially oxidized hydrocarbons such as aldehydes. The gas fuel thus reformed (reformed fuel) flows into the exhaust passage 10ex through the supply pipe 26.

  Now, when the discharge reactor 20 is energized, electrons emitted from the electrode 21 collide with oxygen molecules contained in the air in the interelectrode passage 21a. Then, ozone is generated from oxygen molecules. That is, the discharge reactor 20 generates oxygen as active oxygen by converting oxygen molecules into a plasma state by discharge. Therefore, ozone is contained in the air flowing through the air passage 23b when the discharge reactor 20 is energized.

  In the air passage 23b or the vaporization chamber 25a, a cold flame reaction occurs in which the gaseous fuel is partially oxidized by oxygen in the air. Specific examples of such partially oxidized fuel (reformed fuel) include partial oxides (for example, aldehydes) in which a part of the fuel (hydrocarbon compound) is oxidized to aldehyde groups (CHO). It is done.

  Here, the cold flame reaction will be described in detail with reference to FIGS.

  FIG. 2 is a simulation result simulating a phenomenon in which fuel (hexadecane) is sprayed on the heater 30 to be vaporized, and the vaporized fuel stays in the vicinity of the heater 30 and is reformed. Specifically, it shows changes in various physical quantities with respect to the elapsed time from the start of exposure when gaseous fuel (hexadecane) is exposed to 430 ° C. That is, (a) in the figure shows the change in ambient temperature. (B) shows the change in the molar concentration of fuel (hexadecane). (C) shows the change in the molar concentration of each of oxygen molecules consumed by oxidation, water molecules generated by oxidation, and carbon dioxide molecules. (D) shows the change in the molar concentration of acetaldehyde and propionaldehyde, which are reformed fuels produced by the cold flame reaction. Initial conditions at the start of fuel injection are 1 atm, hexadecane concentration 2200 ppm, oxygen concentration 20%, carbon dioxide concentration 9%, and water concentration 2%.

  As shown in FIG. 2, as soon as fuel is injected, the ambient temperature increases, the molar concentration of the fuel decreases, and the molar concentration of the reformed fuel increases. This phenomenon means that the fuel is oxidized to oxygen and generates heat, and reformed fuel is generated from the fuel. That is, a cold flame reaction is occurring. However, such a temperature rise and changes in various molar concentrations are temporary, and during the period of about 4 seconds from the start of fuel injection, no temperature rise and no change in molar concentration appear.

  When about 4 seconds elapse, the atmospheric temperature further rises, the reformed fuel molar concentration decreases, and the amount of carbon dioxide and water produced and the amount of oxygen consumed increase. This phenomenon means that the reformed fuel is oxidized to oxygen and generates heat, and the reformed fuel is completely burned to generate carbon dioxide and water. That is, a hot flame reaction is occurring. Note that the amount of temperature increase due to the cold flame reaction is smaller than the amount of temperature increase due to the hot flame reaction. Moreover, the oxygen consumption by a cold flame reaction is less than the oxygen consumption by a hot flame reaction.

  When the oxidation reaction occurs in two stages, the reformed fuel appears as an intermediate product during the period from the start of the cold flame reaction to the start of the hot flame reaction. Specific examples of the intermediate product include various hydrocarbon compounds such as aldehydes and ketones. FIG. 3 shows an example of a main reaction route through which aldehyde is generated.

  First, as shown in (1) in the figure, hydrocarbon (light oil) reacts with oxygen molecules to generate hydrocarbon peroxy radicals. This hydrocarbon peroxy radical is decomposed into an aldehyde and a hydrocarbon radical (see (2)). This hydrocarbon radical reacts with oxygen molecules to generate another hydrocarbon peroxy radical (see (3)). This hydrocarbon peroxy radical is decomposed into an aldehyde and a hydrocarbon radical (see (4)). This hydrocarbon radical reacts with oxygen molecules to generate another hydrocarbon peroxy radical (see (5)). In this way, hydrocarbon peroxy radicals are repeatedly generated while reducing the number of carbon atoms, and aldehydes are generated each time the carbon peroxy radical is generated. In the hot flame reaction, the fuel is completely burned to generate carbon dioxide and water, and no intermediate product appears. That is, the intermediate product produced by the cold flame reaction is oxidized to carbon dioxide and water.

  In the simulation shown in FIG. 2, the exposure temperature was 430 ° C. On the other hand, the present inventors further conducted analysis by simulation with different exposure temperatures. As a result, when the exposure temperature is 530 ° C., there is almost no period of staying in the cold flame reaction, and the oxidation reaction is completed in one stage. When the exposure temperature is set to 330 ° C., the start time of the cold flame reaction is delayed as compared with the case where the exposure temperature is set to 430 ° C. When the exposure temperature is 230 ° C. or lower, neither a cold flame reaction nor a hot flame reaction occurs, and no oxidation reaction occurs.

  In the simulation shown in FIG. 2, the equivalence ratio, which is the ratio between the injected fuel and the supplied air, is 0.23. On the other hand, the present inventors further performed analysis by simulation with different equivalence ratios. If the equivalence ratio is strictly defined, it is a value obtained by dividing "the weight of fuel contained in the actual air-fuel mixture" by "the weight of fuel capable of complete combustion". When the equivalence ratio is 1.0, there is almost no period of staying in the cold flame reaction, and the oxidation reaction is completed in one stage. In addition, when the equivalent ratio is 0.37, the start time of the cold flame reaction is earlier than when the equivalent ratio is 0.23. In addition, the cold flame reaction rate is increased and the cold flame reaction period is shortened. In addition, the atmospheric temperature at the time when the cold flame reaction ends is increased.

  FIG. 4 summarizes these analysis results. That is, the relationship between the exposure temperature (atmosphere temperature) and the equivalence ratio and the presence / absence of the occurrence of a cold flame reaction is shown. The horizontal axis in FIG. A region with dots in the figure represents a region where a two-stage oxidation reaction occurs. As shown in the figure, in the region where the ambient temperature is lower than the lower limit value, it becomes a non-reactive region where no oxidation reaction occurs. Even if the atmospheric temperature is higher than the lower limit, if the equivalent ratio is in the region of 1.0 or more, it becomes a one-step oxidation reaction region where the oxidation reaction is completed in one step.

  Further, the boundary line between the two-stage oxidation reaction region and the first-stage oxidation reaction region varies depending on the ambient temperature and the equivalence ratio. That is, a two-stage oxidation reaction occurs when the ambient temperature is in a predetermined temperature range and the equivalent ratio is in a predetermined equivalent ratio range. These temperature range and equivalent ratio range correspond to the range of the region with dots in FIG. When the atmospheric temperature is adjusted to an optimum temperature (for example, 370 ° C.) within a predetermined temperature range, the equivalent ratio at the boundary line becomes the maximum value (for example, 1.0). Therefore, in order to cause the cold flame reaction at an early stage, the heater temperature is adjusted to the optimum temperature and the equivalence ratio is set to 1.0. However, since the cold flame reaction does not occur when the equivalent ratio exceeds 1.0, it is desirable to adjust the equivalent ratio to a value smaller than the margin by 1.0.

  In the simulation shown in FIG. 2, the ozone concentration in the air is zero. On the other hand, the present inventors further performed an analysis by simulation with different ozone concentrations in the air. The initial conditions in this simulation are 1 atm, hexadecane concentration 2200 ppm, and ambient temperature 330 ° C. As a result, it was confirmed that the start time of the cold flame reaction was earlier as the ozone concentration was higher. Such a phenomenon caused by ozone occurs for the following reason. That is, in (1), (3), and (5) in FIG. 3, hydrocarbon radicals and oxygen molecules react, but when ozone is contained in the air, this reaction is promoted, and aldehyde Will be generated in a short time.

  The microcomputer 81 provided in the ECU 80 includes a storage device that stores a program and a central processing unit that executes arithmetic processing according to the stored program. The ECU 80 controls the operation of the internal combustion engine 10 based on detection values of various sensors. Specific examples of the various sensors include an accelerator pedal sensor 91, an engine speed sensor 92, a throttle opening sensor 93, an intake pressure sensor 94, an intake air amount sensor 95, an exhaust temperature sensor 96, and the like.

  The accelerator pedal sensor 91 detects a user's accelerator pedal depression amount. The engine rotation speed sensor 92 detects the rotation speed (engine speed) of the output shaft 10 a of the internal combustion engine 10. The throttle opening sensor 93 detects the opening of the throttle valve 13. The intake pressure sensor 94 detects the pressure on the downstream side of the throttle valve 13 in the intake passage 10in. The intake air amount sensor 95 detects the mass flow rate of intake air.

  In general, the ECU 80 controls the injection amount and injection timing of combustion fuel injected from a fuel injection valve (not shown) according to the rotation speed of the output shaft 10a and the load of the internal combustion engine 10. Further, the ECU 80 controls the operation of the reducing agent addition device based on the exhaust temperature detected by the exhaust temperature sensor 96. That is, the microcomputer 81 performs control so as to switch between generation of reformed fuel and generation of ozone by repeatedly executing the program of the procedure shown in FIG. 5 at a predetermined period. The above program is triggered by the ignition switch being turned on, and is always executed during the operation period of the internal combustion engine 10.

  First, in step S5 of FIG. 5, it is determined whether or not the internal combustion engine 10 is in operation. If it is determined not to be in operation, the operation of the reducing agent addition device is stopped in step S18. Specifically, when the discharge reactor 20, the air pump 20p, the injection valve 40, and the heater 30 are energized, the energization thereof is stopped. Further, even when the NOx catalyst temperature is lower than the activation temperature and the NOx adsorption amount is in a saturated state, or when the NOx catalyst temperature is higher than the reducible range, step S18 is performed. Deactivate the equipment.

  On the other hand, if it is determined that the internal combustion engine 10 is in operation, it is determined in the next step S10 whether or not an S poisoning recovery process (see FIG. 8) described later is being performed. When it is determined that the S poisoning recovery process is not in progress, the reducing agent addition device is operated as follows according to the temperature of the reduction catalyst (NOx catalyst temperature) of the NOx purification device 15.

  That is, first, in step S11, it is determined whether or not the NOx catalyst temperature is higher than a predetermined temperature T1. If it is determined that the temperature is lower than the predetermined temperature T1, it is determined in subsequent step S12 whether or not the NOx catalyst temperature is higher than the second predetermined temperature T2. If it is determined that the temperature is lower than the second predetermined temperature, in the subsequent step S13, it is determined whether or not the NOx catalyst temperature is higher than the activation temperature T3.

  The predetermined temperature T1 and the second predetermined temperature T2 are set to be higher than the activation temperature T3. The predetermined temperature T1 is set to be higher than the second predetermined temperature T2. For example, when the activation temperature T3 is 250 ° C., the second predetermined temperature T2 is set to 350 ° C., and the predetermined temperature T1 is set to 400 ° C. The NOx catalyst temperature is estimated from the exhaust temperature detected by the exhaust temperature sensor 96. Here, the activation temperature T3 of the reduction catalyst is the lowest temperature at which NOx can be reduced and purified on the reduction catalyst.

  If it is determined in steps S11, S12, and S13 that the catalyst temperature is lower than the activation temperature T3, the reducing agent adding device is operated in the ozone generation mode in step S14. When it is determined that the catalyst temperature is higher than the activation temperature T3 and lower than the second predetermined temperature T2, in step S15, the reducing agent addition device is operated in the strong partial oxidation mode. When it is determined that the catalyst temperature is higher than the second predetermined temperature T2 and lower than the predetermined temperature T1, in step S16, the reducing agent addition device is operated in the weak partial oxidation mode. When it is determined that the catalyst temperature is higher than the predetermined temperature T1, in step S17, the reducing agent adding device is operated in the oxidation stop mode.

  When the ozone generation mode is set in step S14 of FIG. 5, the air pump 20p is operated with a preset amount of power, and the electrode 21 of the discharge reactor 20 is energized and discharged with a preset amount of power. Give rise to Further, the energization to the heater 30 is stopped, and the energization to the injection valve 40 is stopped to stop the fuel injection. Therefore, according to the ozone generation mode, ozone is generated in the discharge reactor 20, and the generated ozone is added to the exhaust passage 10 ex through the air pipe 23 and the supply pipe 26.

  Here, when the heater 30 is energized, the ozone is heated and collapses. Moreover, if fuel injection is performed, ozone will react with fuel. In view of these points, since heating by the heater 30 is stopped and fuel injection is stopped in the ozone generation mode, it is possible to avoid ozone reacting with fuel and heating collapse. Therefore, the generated ozone is added to the exhaust passage 10ex as it is.

  When the strong partial oxidation mode is set in step S15 in FIG. 5, the subroutine processing in FIG. 6 is started.

  The outline of this process will be described according to the one-dot chain line in the drawing. First, in step S30, the temperature in the reaction vessel 25 is adjusted to a predetermined temperature range. Next, in step S40, the equivalence ratio, which is the ratio of fuel to air, is adjusted to a predetermined equivalence ratio range. These temperature range and equivalence ratio range are the range of the two-stage oxidation reaction region indicated by the dots in FIG. As a result, the above-described reformed fuel is generated by causing a cold flame reaction.

  The lower limit of the temperature range is set to 260 ° C., which is the boundary line between the first stage oxidation region and the second stage oxidation region and the non-reaction region. The upper limit of the temperature range is set to the maximum temperature on the boundary line between the first stage oxidation region and the second stage oxidation region. The upper limit of the equivalence ratio range is the maximum value of the boundary lines between the first stage oxidation region and the second stage oxidation region, and is set to an equivalent ratio corresponding to 370 ° C.

  Further, in step S50, the power supplied to the discharge reactor 20 is controlled according to the concentration of fuel in the reaction vessel 25. Thereby, ozone is generated, and the ozone is supplied into the reaction vessel 25, so that the start timing of the cold flame reaction is advanced and the cold flame reaction time is shortened. Therefore, even if the reaction vessel 25 is downsized so that the residence time of the fuel in the reaction vessel 25 is shortened, the cold flame reaction can be completed within the residence time. Therefore, the reaction vessel 25 can be downsized.

  The microcomputer 81 when executing the process of step S30 provides “temperature adjusting means” for adjusting the temperature of the vaporizing chamber 25a to a predetermined temperature range. The microcomputer 81 when executing the process of step S40 provides “equivalent ratio adjusting means” for adjusting the equivalent ratio, which is the ratio of the fuel and air supplied to the vaporizing chamber 25a, to a predetermined equivalent ratio range.

  Details of these steps S30, S40, and S50 will be described below with reference to FIG.

  First, the process of step S30 related to the temperature adjusting means will be described. First, in step S31, the temperature in the reducing agent addition apparatus, that is, the temperature in the reaction vessel 25 is acquired. Specifically, the temperature detected by the temperature sensor 27 is acquired. In subsequent step S32, it is determined whether or not the detected temperature Tact is higher than a preset target temperature Ttrg. Specifically, it is determined whether or not the difference value ΔT obtained by subtracting the target temperature Ttrg from the detected temperature Tact is greater than zero.

  If a negative determination is made that ΔT> 0 is not satisfied, the process proceeds to step S33, and the amount of heating by the heater 30 is increased. Specifically, the duty ratio for energizing the heater 30 is increased as the absolute value of the difference value ΔT is larger. On the other hand, if an affirmative determination is made that ΔT> 0, it is determined in subsequent step S34 whether or not the difference value ΔT exceeds an upper limit value (for example, 50 ° C.). If a negative determination is made that the upper limit value is not exceeded, the process proceeds to step S35, and the amount of heating by the heater 30 is reduced. Specifically, the energization duty ratio to the heater 30 is decreased as the absolute value of the difference value ΔT is larger. However, if an affirmative determination is made that the upper limit value has been exceeded, the process proceeds to step S36, and energization of the heater 30 is stopped. As a result, the ambient temperature is rapidly reduced.

  The target temperature Ttrg used in step S32 is set to an atmospheric temperature (for example, 370 ° C.) at which the equivalent ratio is maximum in the two-stage oxidation reaction region shown in FIG. Since the temperature of the vaporizing chamber 25a rises due to the cold flame reaction, the temperature of the heater 30 itself is controlled to a value lower than the target temperature Ttrg by the temperature rise due to the cold flame reaction.

  Next, the process of step S40 related to the equivalence ratio adjusting means will be described. In the process of step S40, if the difference value ΔT is 50 ° C. or less, the process proceeds to step S41, and the maximum value φmax of the equivalent ratio that causes the cold flame reaction corresponding to the detected temperature Tact is calculated. Specifically, the maximum value φmax of the equivalent ratio in the two-stage oxidation reaction region, which is the maximum value φmax of the equivalent ratio corresponding to the ambient temperature, or a value obtained by subtracting a predetermined margin from the maximum value φmax, The equivalent ratio φtrg is stored in the microcomputer 81. For example, the maximum value φmax of the equivalence ratio in the two-stage oxidation reaction region shown in FIG. 4 and the maximum value φmax of the equivalence ratio corresponding to the ambient temperature is mapped and stored in the microcomputer 81. Then, the maximum value φmax of the equivalence ratio corresponding to the detected temperature Tact is calculated with reference to the map.

  In the subsequent step S42, the target equivalent ratio φtrg is set based on the maximum equivalent ratio value φmax calculated in step S41. Specifically, a value obtained by subtracting the maximum value φmax by a predetermined margin is set as the target equivalent ratio φtrg. Thereby, even if the actual equivalent ratio becomes larger than the target equivalent ratio φtrg, the possibility of exceeding the maximum value φmax can be reduced, and the possibility of reaching not only the cold flame reaction but also the hot flame reaction can be reduced.

  On the other hand, if the difference value ΔT is greater than 50 ° C. and the heater 30 is stopped in step S36, the process proceeds to step S43, and the target equivalent ratio φtrg is set to a preset value for air cooling. . The air cooling value is set to a value smaller than the maximum value φmax of the equivalence ratio corresponding to the target temperature Ttrg. In other words, the air flow rate is increased as compared with the case of step S42 to promote a rapid decrease in the ambient temperature.

  In the subsequent step S44, a value for supplying the fuel necessary for reducing all of the NOx flowing into the NOx purification device 15 to the NOx purification device 15 without excess or deficiency is set as the target fuel flow rate Ftrg. The target fuel flow rate Ftrg is the mass of fuel supplied to the NOx purification device 15 per unit time.

  Specifically, the target fuel flow rate Ftrg is set based on the NOx inflow rate and NOx catalyst temperature described below. The NOx inflow rate is the mass of NOx flowing into the NOx purification device 15 per unit time. For example, the NOx inflow rate can be estimated based on the operating state of the internal combustion engine 10. The NOx catalyst temperature is the temperature of the reduction catalyst that the NOx purification device 15 has. For example, the NOx catalyst temperature can be estimated based on the temperature detected by the exhaust temperature sensor 96.

  The target fuel flow rate Ftrg is increased as the NOx inflow rate increases. Further, since the amount (reducing power) in which NOx is reduced on the reduction catalyst varies depending on the NOx catalyst temperature, the target fuel flow rate Ftrg is set according to the difference in the reducing power depending on the NOx catalyst temperature.

  In the subsequent step S45, the target air flow rate Atrg is calculated based on the target equivalent ratio φtrg set in step S42 or step S43 and the target fuel flow rate Ftrg set in step S44. Specifically, the target air flow rate Atrg is calculated so that φtrg = Ftrg / Atrg.

  In the subsequent step S46, the operation of the air pump 20p is controlled based on the target air flow rate Atrg calculated in step S45. Specifically, the duty ratio for energizing the air pump 20p is increased as the target air flow rate Atrg is larger. In the subsequent step S47, fuel injection is performed by controlling the operation of the injection valve 40 based on the target fuel flow rate Ftrg set in step S44. Specifically, the valve opening time of the injection valve 40 is increased as the target fuel flow rate Ftrg is increased. Alternatively, the interval from the end of the current injection to the start of the next injection is shortened.

  Next, the process of step S50 related to the discharge power control means will be described. First, in step S51, the target ozone flow rate Otrg is calculated based on the target fuel flow rate Ftrg set in step S44. Specifically, the target ozone flow rate Otrg is calculated so that the ratio of the ozone concentration to the fuel concentration in the vaporizing chamber 25a becomes a predetermined value (for example, 0.2). For example, the ratio is set so that the cold flame reaction is completed within a predetermined time (for example, 0.02 seconds). Furthermore, in step S51, the target ozone flow rate Otrg is set to increase as the temperature of the reduction catalyst decreases.

  In the subsequent step S52, the target energization amount Ptrg to the discharge reactor 20 is calculated based on the target air flow rate Atrg calculated in step S45 and the target ozone flow rate Otrg calculated in step S51. Specifically, as the target air flow rate Atrg increases, the residence time of air in the interelectrode passage 21a is shortened, so that the target energization amount Ptrg is increased. Further, the target energization amount Ptrg is increased as the target ozone flow rate Otrg is increased. In subsequent step S53, the energization amount to the discharge reactor 20 is controlled based on the target energization amount Ptrg calculated in step S52. Specifically, the duty ratio for energizing the discharge reactor 20 is increased as the target energization amount Ptrg is larger. Alternatively, the interval from the end of current energization to the start of next energization is shortened.

  As described above, according to the strong partial oxidation mode shown in FIG. 6, the ozone generated in the discharge reactor 20, the oxygen in the air, and the fuel vaporized by the heater 30 are mixed with the oxygen in the air, and ozone is present. The fuel is partially oxidized in an environment where On the other hand, in the weak partial oxidation mode in step S16 of FIG. 5, the generation of ozone is stopped, the fuel vaporized by the heater 30 and the oxygen in the air are mixed, and the fuel is partially in an environment where there is no ozone. Oxidized.

  Specifically, in the weak partial oxidation mode, the processing in step S50 is abolished while the same processing as in steps S30 and S40 shown in FIG. 6 is performed. That is, while adjusting to a predetermined temperature range in step S30, it adjusts to a predetermined equivalence ratio range in step S40. However, the ozone generation in step S50 is not performed, and energization to the discharge reactor 20 is stopped.

  Moreover, in the oxidation stop mode by step S17 of FIG. 5, the process of FIG.6 S40 and S50 is abolished, implementing the process similar to step S30 shown in FIG. That is, the temperature is adjusted to a predetermined temperature range in step S30. However, the adjustment of the equivalence ratio in step S40 and the ozone generation in step S50 are not performed, and energization to the discharge reactor 20 is stopped. Further, the operation of the air pump 20p is stopped, and the supply of air (oxygen) onto the vaporizing chamber 25a and the guide surface 53a is stopped. However, the fuel is supplied to the vaporizing chamber 25a and vaporized by the heater 30. Therefore, gaseous fuel that is not partially oxidized without being oxidized by oxygen or ozone is added to the exhaust passage 10ex.

  The microcomputer 81 repeatedly performs the process shown in FIG. 7 at a predetermined cycle, separately from the processes in the various modes shown in FIG.

  In the process of FIG. 7, first, in step S61, it is determined whether a regeneration request for the DPF 14 has been made. The DPF regeneration request is a request to regenerate the particulate collection ability of the DPF 14 by burning the particulate component (soot) adhering to the DPF 14 and causing clogging. Note that the microcomputer 81 determines whether or not to request the regeneration of the DPF 14 by setting the DPF regeneration request flag to ON by executing a determination process different from FIG.

  In the determination of the DPF regeneration request, for example, a pressure difference between the upstream side and the downstream side of the DPF 14 in the exhaust passage 10ex is detected by a sensor, and the DPF regeneration request flag is turned on if the differential pressure exceeds a predetermined value. Set. Alternatively, the DPF regeneration request flag is set to ON when the travel distance of the vehicle reaches a predetermined distance or when the operation time of the internal combustion engine 10 reaches a predetermined time. When the differential pressure becomes less than a predetermined value or when the regeneration process of the DPF 14 is performed for a predetermined time, the DPF regeneration request flag is set to OFF to stop the DPF regeneration request.

  Returning to the description of FIG. 7, if it is determined that a DPF regeneration request has been made (S61: YES), in the next step S62, the temperature of the DPF 14 is the regeneration processing temperature at which soot burns (for example, 650 ° C.). ) Is determined. When it is determined that the regeneration processing temperature has not been reached (S62: NO), the temperature of the DPF 14 is increased in the subsequent step S63. Specifically, the temperature of the exhaust gas is raised by controlling the operation of a fuel injection valve (not shown) so as to delay the timing of injecting fuel into the combustion chamber. As a result of this temperature increase, when the temperature of the DPF 14 reaches the regeneration processing temperature, soot adhering to the DPF 14 is burned, and the regeneration processing of the DPF 14 is started.

  Further, since the internal combustion engine 10 according to the present embodiment is a compression self-ignition type, the internal combustion engine 10 is basically operated so as to burn in a lean state during normal operation in which the DPF regeneration process is not performed. Yes. In the DPF regeneration process, the combustion is performed in a lean state in the same manner as in the normal operation. That is, combustion is performed (lean combustion) in a state where the air-fuel ratio, which is the ratio between the fuel injected into the combustion chamber and the air sucked into the combustion chamber, is set to an excess of air.

  When it is determined that the temperature of the DPF 14 has reached the regeneration processing temperature (S62: YES), in the subsequent step S64, the fuel injection timing and the fuel injection amount are controlled so that the temperature of the DPF 14 is maintained at the regeneration processing temperature. .

  In a succeeding step S65, it is determined whether or not an S poison recovery request is made. The S poison recovery request is a request for recovering the reduction catalyst from the sulfur poisoning state in which the NOx purification rate is reduced by SOx in the exhaust gas being adsorbed and remaining on the reduction catalyst. Note that the microcomputer 81 executes a determination process different from that in FIG. 7 to determine whether or not to request the recovery of the reduction catalyst by setting the S poison recovery request flag to ON.

  In the determination of the S poisoning recovery request, for example, when the travel distance of the vehicle reaches a predetermined distance after the previous execution of the S poisoning recovery process, the S poisoning recovery request flag is set to ON. Alternatively, when the operation time of the internal combustion engine 10 has reached a predetermined time after completion of the previous S poisoning recovery process, or when the injection amount of fuel used for combustion of the internal combustion engine 10 has reached a predetermined amount, The S poison recovery request flag is set to ON. Then, when the S poisoning recovery process in step S65 is performed for a predetermined time or more, the S poisoning recovery request flag is set to OFF. Alternatively, when the fuel injection amount from the injection valve 40 reaches a predetermined amount after the start of the current S poison recovery process, the S poison recovery request flag is set to OFF.

  If it is determined that an S poisoning recovery request has been made (S65: YES), in the next step S66, the reducing agent addition device is operated so as to perform the S poisoning recovery process shown in FIG. . During the implementation period of the S poison recovery process, the NOx purification process shown in FIG. 5 is stopped. When the NOx purification process is being performed, the microcomputer 81 adds the reformed fuel to the exhaust passage 10ex at the time of a purification request that requires NOx purification on the reduction catalyst (see FIG. 1). I will provide a. Further, the microcomputer 81 when performing the S poisoning recovery process provides recovery control means 81b for adding the reformed fuel to the exhaust passage 10ex when the S poison recovery request is made. Further, the microcomputer 81 when performing the DPF regeneration processing in steps S62, S63, and S64 of FIG. 7 provides combustion removal control means 81c that burns and removes the particulates deposited on the DPF 14 by raising the temperature of the DPF 14. .

  In the S poisoning recovery process shown in FIG. 8, first, in step S71, the heater 30 is operated with a preset energization amount so that the temperature of the vaporizing chamber 25a becomes a predetermined temperature. In subsequent step S72, the air pump 20p is operated with a preset energization amount so that the flow rate of air blown into the air passage 23b becomes a predetermined flow rate.

  Note that the DPF regeneration process is performed by lean combustion as described above. Therefore, the S poison recovery process is performed in an environment in which exhaust from lean combustion flows into the reduction catalyst. Further, in the DPF regeneration process, as described above, the exhaust gas temperature is raised compared to the normal control of the internal combustion engine 10 to raise the DPF temperature to the regeneration process temperature (for example, 650 ° C.). Therefore, the S poison recovery process is performed in an environment in which exhaust gas having a temperature higher than that during normal control flows into the reduction catalyst.

  In subsequent step S73, the injection valve 40 is operated at a preset valve opening time and injection interval so that the fuel injection amount per unit time into the vaporizing chamber 25a becomes a predetermined amount. In the following step S74, the discharge reactor 20 is operated with a preset energization amount so that the amount of ozone contained in the air becomes a predetermined amount. In short, in the S poison recovery process of FIG. 8, the fuel is partially oxidized in an environment in which ozone is present, as in the strong partial oxidation mode in step S15 of FIG.

  However, in the S poisoning recovery process, the heater 30, the air pump 20p, the injection valve 40, and the discharge reactor 20 are operated with preset control amounts. Further, in the S poison recovery process, the strong partial oxidation mode, the weak partial oxidation mode, and the oxidation stop mode are not switched, and the fuel partially oxidized in the ozone environment is added to the exhaust passage 10ex regardless of the catalyst temperature. To do.

  Returning to the description of FIG. 7, in step S67, it is determined whether or not the DPF regeneration process is completed. Specifically, if the above-described DPF regeneration request flag is switched off, it is determined that the DPF regeneration process has been completed. Note that when the DPF regeneration request flag is switched from on to off, the operation of the fuel injection valve is controlled to increase the exhaust gas temperature and maintain the DPF regeneration processing temperature, and the DPF regeneration processing is terminated. . As a result, the temperature of the DPF 14 falls and becomes lower than the regeneration processing temperature.

  If it is determined that the DPF regeneration process has not ended (S67: NO), the process returns to step S64, and the DPF regeneration process is continued. If it is determined that the DPF regeneration process is completed (S67: YES), the DPF regeneration process is terminated regardless of whether or not the S poison recovery request is made. Therefore, the S poison recovery process is continued during the DPF regeneration process (S67: NO) and during the S poison recovery request (S65: YES).

  It is assumed that the S poison recovery process ends during the implementation period of the DPF regeneration process (S67: NO), but if the DPF regeneration process ends with the S poison recovery request being made. In this case, the S poisoning recovery process is forcibly terminated. In addition, when the S poisoning recovery process is forcibly terminated in this way, when performing the next S poisoning recovery process, the S poisoning recovery request flag is turned on / off in consideration of the forced termination. Set.

  Next, the effect of the S poison recovery process of FIG. 8 will be described.

  FIG. 9 shows the NOx purification rate when the test conditions are varied as follows. That is, when a test 1 for measuring the NOx purification rate using a new reduction catalyst not poisoned with S was performed, the NOx purification rate was 91%. After that, when this new reduction catalyst was used to poison the S until the vehicle mileage reached 80000 km and the NOx purification rate was measured in that state, the NOx purification rate was reduced to 65%. It was confirmed that After that, the S poison recovery process shown in FIG. 8 was performed on the S poisoned reduction catalyst, and the test 3 for measuring the NOx purification rate in that state was performed. As a result, the NOx purification rate was improved to 90%. It was confirmed that Therefore, according to the S poisoning recovery process, it was confirmed that the NOx purification rate can be substantially recovered.

The implementation time of the S poisoning recovery process according to Test 3 is 15 minutes, and FIG. 10 shows the test results of measuring the concentration of SO ₂ desorbed from the reduction catalyst in this 15 minutes. As shown in the figure, immediately after the execution of the S poisoning recovery process, the concentration of desorbed SO ₂ rises rapidly, and the concentration of desorbed SO ₂ reaches its maximum in about 1 minute after the execution. Thereafter, the concentration of the desorbed SO ₂ is plummeted, the concentration of the desorption SO ₂ becomes less than 1ppm at elapsed time 15 minutes. Therefore, it was confirmed that the remaining SO _{2 of the} reduction catalyst can be sufficiently reduced by performing the S poisoning recovery process for 15 minutes.

  In the tests according to FIGS. 9 and 10, the temperature of the reduction catalyst is an optimum temperature (for example, 550 ° C.) described later. Specifically, when the DPF temperature is increased in accordance with the DPF regeneration process, the heat of the DPF 14 is transmitted to the reduction catalyst through the exhaust and the exhaust pipe, and the temperature of the reduction catalyst is increased. Therefore, the reduction catalyst temperature during the DPF regeneration process varies depending on the separation distance L (see FIG. 1) between the NOx purification device 15 and the DPF 14 in the exhaust passage 10ex. In view of this point, the separation distance L is set so that the reduction catalyst has an optimum temperature (for example, 550 ° C.) when the DPF 14 is maintained at the regeneration processing temperature (for example, 650 ° C.).

FIG. 11 is a test result showing that the optimum temperature exists. In this test, the S poison recovery process was performed for 15 minutes with different exhaust temperatures, and the amount of SO ₂ desorption was measured for each test with different exhaust temperatures. That is, the test result of FIG. 11 represents the relationship between the reduction catalyst temperature and the NOx purification rate. Further, the SO ₂ desorption amount shown on the rightmost side in FIG. 11 is tested by injecting fuel from the injection valve 40 while stopping the heater 30 and the discharge reactor 20. That is, the SO ₂ desorption amount is measured for the case where the injected fuel is supplied to the reduction catalyst without being partially oxidized and reformed.

  The test results in FIG. 11 show that when the fuel is not partially oxidized to be reformed to aldehyde, the reduction catalyst temperature cannot be sufficiently recovered at 550 ° C., whereas when the partial reforming is performed, the reduction catalyst temperature is reduced. It shows that S poisoning can be sufficiently recovered in a region of 500 ° C. or higher. In addition, the higher the temperature is in the region of 500 ° C or higher, the more the S poison recovery is not promoted, and the optimal temperature (550 ° C) exists, and the NOx purification rate decreases when the temperature is higher than the optimal temperature. Represents. That is, it is desirable that the reduction catalyst temperature be in the range of 500 ° C. to 600 ° C. in order to sufficiently recover S poisoning, and it is more desirable that the reduction catalyst temperature be in the range of 525 ° C. to 600 ° C.

  The present inventors consider the reason why the NOx purification rate decreases when the temperature is higher than the optimum temperature as follows. That is, when the temperature is higher than the optimum temperature, the inside of the exhaust passage 10ex becomes excessively hot, and as a result, the reducing agent added to the exhaust passage 10ex burns in the exhaust passage 10ex before reaching the reduction catalyst.

Further, the present inventors consider the reason why SO ₂ can be desorbed from the reduction catalyst in a low temperature environment when the fuel is partially oxidized and reformed as compared with the case where the fuel is not reformed. Hereinafter, referring to FIG. 12, a mechanism in which SO ₂ adsorbed on the reduction catalyst (silver catalyst) on the carrier 15b (alumina oxide) reacts with the reformed fuel in a lean environment and desorbs, that is, the reduction catalyst. Will explain the mechanism of recovery from S poisoning.

As shown in (a) of FIG. 12, first, the aldehyde added to the exhaust passage 10ex is adsorbed by the silver catalyst. The silver catalyst at this time is in a state where sulfur is adsorbed in addition to the aldehyde (a state in which S is poisoned). Since this S poisoning recovery process is performed during lean combustion, oxygen molecules are contained in the exhaust gas that contacts the silver catalyst. By this oxygen molecule, the aldehyde adsorbed on the silver catalyst is oxidized and converted into acetate as shown in (b). Thus, the acetate produced | generated on the silver catalyst is a substance which is easy to couple | bond with a sulfur component. Therefore, the acetate in the state adsorbed on the silver catalyst is combined with the sulfur component adsorbed on the silver catalyst and changed to R-SCO as shown in (c). Thereafter, R-SCO adsorbed on the silver catalyst is oxidized by oxygen molecules in the exhaust as shown in (d), and as a result, it is decomposed into S, SO ₂ , H ₂ O, CO ₂ and desorbed from the silver catalyst. Release.

  In short, when a partial oxide is allowed to flow into the reduction catalyst in a lean environment, the partial oxide chemically reacts with sulfur adsorbed on the reduction catalyst and oxygen in the exhaust, thereby desorbing sulfur from the reduction catalyst. The present inventors have obtained the knowledge that the phenomenon of reduction occurs. In view of this knowledge, in the present embodiment, the partially oxidized reformed fuel is added to the exhaust passage 10ex on the condition that the internal combustion engine 10 is lean burning at the time of the S poison recovery request.

  Therefore, it is possible to perform the S poison recovery by causing the phenomenon of the above knowledge while eliminating the need for the control to switch from lean combustion to rich combustion (rich switching control) as in the prior art. Therefore, fuel consumption deterioration due to the conventional rich switching control can be avoided. Also, contrary to the present embodiment, when sulfur is desorbed from the reduction catalyst in a rich environment, malodorous hydrogen sulfide is generated, but in this embodiment, sulfur is desorbed in a lean environment. Generation of hydrogen can be suppressed.

  In order to desorb sulfur from the reduction catalyst, it is necessary to set the exhaust gas of the internal combustion engine 10 to a temperature higher than that during normal control so that the reduction catalyst has a higher temperature. However, according to this embodiment, the conventional rich switching control is performed. Compared with this, sulfur can be desorbed at a lower temperature. Although these effects are confirmed by experiments, the present inventors consider that they are caused by the reaction of the partial oxide as shown in FIG.

  Furthermore, according to the present embodiment, the recovery control means is applied to a combustion system provided with the DPF 14 and the combustion removal control means 81c, on the condition that the combustion removal (DPF regeneration) is being performed by the combustion removal control means 81c. Addition of reducing agent according to 81b is carried out. Therefore, it is not necessary to raise the exhaust temperature for the purpose of recovery of S poisoning separately from the DPF regeneration, and fuel consumption can be improved. Note that, due to the above-described effect that sulfur can be desorbed at a low temperature, it is possible to achieve S poison recovery at the exhaust temperature during DPF regeneration.

  Now, there is an optimum temperature for the DPF temperature for the DPF regeneration process, and there is also an optimum temperature for the reduction catalyst temperature for the S poison recovery process. According to the present embodiment in view of this point, the exhaust passage 10ex is set such that the temperature of the reduction catalyst rises as the temperature rises due to the DPF regeneration process, and the temperature of the reduction catalyst falls within the temperature range suitable for S poison recovery. The separation distance L between the DPF 14 and the NOx purification device 15 is set. Therefore, it is possible to simultaneously achieve both the DPF temperature and the reduction catalyst temperature at optimum temperatures. Therefore, in recovering S poisoning at the exhaust temperature during DPF regeneration, both DPF regeneration and S poison recovery can be performed with high efficiency.

  Furthermore, in this embodiment, a substance containing at least silver is used as the reduction catalyst. According to this, it was confirmed that S poisoning can be recovered at a lower temperature and in a shorter time. This is because the phenomenon that “the partial oxide chemically reacts with sulfur adsorbed on the reduction catalyst and oxygen in the exhaust gas to desorb and reduce sulfur from the reduction catalyst” uses a silver catalyst. The present inventors consider that this occurs actively.

Explaining this consideration in more detail, when platinum is used as the catalyst, platinum has a stronger oxidizing power than silver, so the partial oxide flowing into the catalyst is in the state of acetate shown in FIG. It is completely oxidized without going through the process and decomposed into H ₂ O and CO ₂ . On the other hand, when silver is used for the catalyst, since silver has a lower oxidizing power than platinum, the partial oxide that has flowed into the catalyst is once in the state of acetate before being completely oxidized (FIG. 12 ( b)). Therefore, a reaction in which the reducing agent (acetate) is combined with sulfur can be caused, and the reducing agent is completely oxidized and decomposed together with sulfur in a state of being combined with sulfur. Thus, the present inventors consider that the above phenomenon occurs actively by using a silver catalyst having an appropriate oxidizing power.

  A fuel in a high temperature environment oxidizes with oxygen contained in the surrounding air and ignites and burns even at atmospheric pressure. Such an oxidation reaction by self-ignition combustion is also called a hot flame reaction in which carbon dioxide and water are generated while generating heat. However, when the ratio of fuel to air (equivalent ratio) and the ambient temperature are within a predetermined range, the period of staying in the cold flame reaction described below becomes longer, and then a hot flame reaction occurs. That is, an oxidation reaction occurs in two stages, a cold flame reaction and a hot flame reaction (see FIG. 4).

  This cold flame reaction is a reaction that easily occurs when the ambient temperature is low and the equivalence ratio is small, and is a reaction in which the fuel is partially oxidized by oxygen contained in the surrounding air. Then, when the ambient temperature rises due to heat generated by the cold flame reaction and then a certain period of time elapses, the partially oxidized fuel (for example, aldehyde) is oxidized and the above-described hot flame reaction occurs. When a partially oxidized fuel such as an aldehyde generated by a cold flame reaction is used as a reducing agent for NOx purification, the NOx purification rate is improved as compared with the case where a fuel that is not partially oxidized is used.

  In this embodiment in view of this point, the hydrocarbon compound (reducing agent) heated to a predetermined temperature or higher by the heater 30 is partially oxidized by oxygen contained in the air to be reformed. Therefore, the hydrocarbon compound is reformed to aldehyde by a cold flame reaction, and the reformed fuel is used as a reducing agent for NOx purification, so that the NOx purification rate can be improved. And since the partial oxide (aldehyde etc.) used as a reducing agent for NOx purification in this way is used for S poison recovery, the hardware constitution only for S poison recovery can be made unnecessary.

  In addition, the present inventors have obtained knowledge that a cold flame reaction occurs before a hot flame reaction if the atmospheric temperature and the equivalent ratio are adjusted within a predetermined range. In the present embodiment made based on this knowledge, a reaction vessel 25 is provided which internally forms a vaporization chamber 25a for oxidizing the fuel with oxygen in the air. Then, the temperature and equivalent ratio of the vaporizing chamber 25a are adjusted so that a cold flame reaction occurs, and fuel partially oxidized by the cold flame reaction is used as a reducing agent for NOx purification. Therefore, the NOx purification rate can be improved compared to the case where fuel that is not partially oxidized is used as it is as a reducing agent.

  Now, regardless of the case where the reducing agent is added in any of the above-described strong partial oxidation mode, weak partial oxidation mode, and oxidation stop mode, the NOx purification rate changes according to the NOx catalyst temperature. In any mode, there is a catalyst temperature (peak temperature) when the purification rate is maximized. However, the peak temperature differs depending on the mode. Accordingly, in the region on the higher temperature side than the predetermined temperature T1 (first boundary temperature), a part of the partially oxidized reformed fuel is completely oxidized and converted into carbon dioxide and water (oxidation alteration) before reducing NOx. ) And lose NOx reduction ability. For this reason, the purification rate is higher when the modification by partial oxidation is not performed. Similarly, even in a region on the higher temperature side than the predetermined temperature T2 (second boundary temperature), a part of the partially oxidized reformed fuel is oxidized and deteriorated and loses the NOx reduction ability, and the higher the catalyst temperature is, the higher the catalyst temperature is. The degree of oxidative alteration increases. Therefore, the purification rate is higher when ozone is not supplied. However, in the region on the lower temperature side than the second boundary temperature, even if ozone is supplied and partial oxidation is promoted, the oxidation deterioration hardly occurs. Therefore, the purification rate is higher when ozone is supplied.

  Based on the above consideration, according to the present embodiment, when the reduction catalyst temperature is higher than the predetermined temperature T1, the reforming is stopped by switching to the oxidation stop mode. Therefore, for example, when the catalyst temperature rises and shifts from the low temperature side to the high temperature side of the first boundary temperature, the mode is switched from the weak partial oxidation mode to the oxidation stop mode. Therefore, the chance of falling into a situation where the partially oxidized fuel is oxidized and deteriorated to lose the NOx reduction ability is reduced, and the purification rate is improved. For example, when the catalyst temperature rises and shifts from the low temperature side to the high temperature side of the second boundary temperature, the high partial oxidation mode is switched to the weak partial oxidation mode. Therefore, the chance of falling into a situation where the partially oxidized fuel is oxidized and deteriorated to lose the NOx reduction ability is reduced, and the purification rate is improved.

  Furthermore, in this embodiment, the discharge reactor 20 is provided, and ozone generated by the discharge reactor 20 is supplied when a cold flame reaction is caused. Therefore, it is possible to accelerate the start time of the cold flame reaction and shorten the cold flame reaction time. Therefore, even if the reaction vessel 25 is downsized so that the residence time of the fuel in the reaction vessel 25 is shortened, the cold flame reaction can be completed within the residence time. Therefore, the reaction vessel 25 can be downsized. And since the discharge reactor 20 used for such reforming is used for S poison recovery, a hardware configuration dedicated to S poison recovery can be eliminated.

Further, in the present embodiment, when the reduction catalyst is lower than the activation temperature T3, the fuel generated by the injection valve 40 is stopped and the ozone generated by the discharge reactor 20 is supplied to the air passage 23b, so that the exhaust gas is exhausted. Ozone is added to the passage 10ex. According to this, although the reduction catalyst of the NOx purification device 15 is not activated, it can be prevented that the reformed fuel as the reducing agent is added. Then, by adding ozone, NO in exhaust gas is oxidized to NO ₂ and adsorbed to the NOx purification catalyst, so that the amount of NOx adsorption to the NOx purification device 15 can be increased.

  Furthermore, in the present embodiment, cracking is performed in which the fuel is decomposed into a hydrocarbon compound having a small number of carbon atoms by the heater 30. Since the hydrocarbon having a reduced number of carbon atoms due to cracking has a low boiling point, the vaporized fuel is suppressed from returning to a liquid.

(Other embodiments)
The preferred embodiments of the present invention have been described above, but the present invention is not limited to the above-described embodiments, and various modifications can be made as illustrated below. Not only combinations of parts that clearly show that combinations are possible in each embodiment, but also combinations of the embodiments even if they are not explicitly stated unless there is a problem with the combination. Is also possible.

  In the above embodiment, a catalyst containing silver is used as the reduction catalyst. However, the present invention is not limited to such a silver catalyst. For example, a catalyst containing copper or iron may be used as the reduction catalyst. Good. Since these copper catalysts or iron catalysts also have lower oxidizing power than platinum catalysts in the same manner as silver catalysts, it is assumed that the phenomenon of sulfur desorption shown in FIG. 12 occurs.

  In the embodiment described above, the heater 30, the air pump 20p, the injection valve 40, and the discharge reactor 20 are operated with preset control amounts when the S poisoning recovery process shown in FIG. 8 is performed. On the other hand, the sulfur component amount (S poisoning amount) adsorbed on the reduction catalyst may be estimated, and the control amount may be variably set according to the estimated S poisoning amount.

  In the embodiment shown in FIG. 1, by providing the discharge reactor 20, ozone is supplied to the air passage 23b. On the other hand, the discharge reactor 20 may be abolished, the strong partial oxidation mode may be abolished, control may be performed so as to switch between the partial oxidation mode and the oxidation stop mode, and ozone supply during the S poison recovery process may be abolished. . According to this, although the reaction speed of the fuel by the supply of ozone cannot be improved and the time required for the S poison recovery process becomes long, the discharge reactor 20 can be eliminated and the apparatus can be downsized. .

  In the embodiment shown in FIG. 1, the NOx purification device 15 is applied to a combustion system including the DPF 14 and the S poisoning recovery process is performed during the DPF regeneration process. On the other hand, when the DPF regeneration process is not performed, the S poison recovery process may be performed separately from the DPF regeneration process. In this case, the NOx purification device 15 may be applied to a combustion system that does not include the DPF 14.

  In the embodiment shown in FIG. 1, air is supplied to the discharge reactor 20 by the air pump 20p. On the other hand, a part of the intake air of the internal combustion engine 10 may be branched and allowed to flow into the discharge reactor 20. Specifically, the high-temperature intake air before being cooled by the cooler 12 may be branched to the discharge reactor 20, or the low-temperature intake air after being cooled by the cooler 12 may be branched to the discharge reactor 20.

  In the above embodiment shown in FIG. 1, a reduction catalyst that physically captures (that is, adsorbs) NOx is employed. However, the combustion system employs a reduction catalyst that traps (that is, stores) NOx by chemical bonding. A reducing agent addition device may be applied.

  When the internal combustion engine 10 is burning in a state leaner than the stoichiometric air-fuel ratio, the reducing agent addition device is applied to a combustion system in which the NOx purification device 15 adsorbs NOx and reduces NOx in other than lean combustion. Also good. In this case, ozone may be generated during lean combustion, and reformed fuel may be generated at times other than lean combustion. As a specific example of the catalyst that captures NOx during lean combustion in this way, an occlusion reduction catalyst using platinum and barium supported on a carrier can be cited.

  In the embodiment shown in FIG. 1, a reducing agent addition device is applied to a combustion system mounted on a vehicle. On the other hand, a reducing agent addition device may be applied to a stationary combustion system. In the embodiment shown in FIG. 1, a reducing agent addition device is applied to a compression self-ignition diesel engine, and light oil used as a fuel for combustion is used as a reducing agent. On the other hand, gasoline used as a fuel for combustion may be used as a reducing agent by applying a reducing agent addition device to an ignition ignition type gasoline engine.

  DESCRIPTION OF SYMBOLS 10 ... Internal combustion engine, 10ex ... Exhaust passage, 15 ... NOx purification apparatus, 20 ... Discharge reactor (reforming means), 30 ... Heater (reforming means), 81a ... Purification control means, 81b ... Recovery control means.

Claims (6)

A NOx purification device (15) for purifying NOx contained in the exhaust gas of the internal combustion engine (10) on the reduction catalyst is provided in a combustion system provided in the exhaust passage (10ex), and upstream of the reduction catalyst in the exhaust passage. In a reducing agent addition apparatus for adding a reducing agent that is a hydrocarbon compound to the side,
Modifying means (20, 30) for partially oxidizing and modifying the reducing agent;
A purification control means (81a) for adding a reducing agent reformed by the reforming means to the exhaust passage at the time of a purification request that requires NOx purification on the reduction catalyst;
Recovery for adding the reducing agent reformed by the reforming means to the exhaust passage on the condition that the internal combustion engine is lean burning at the time of the recovery request for recovery of the sulfur-poisoned reduction catalyst Control means (81b);
An apparatus for adding a reducing agent, comprising: The combustion system includes:
A particulate collection device (14) disposed upstream of the NOx purification device in the exhaust passage and collecting particulates contained in the exhaust; and
Combustion removal control means (81c) for burning and removing fine particles deposited on the fine particle collection device by raising the temperature of the fine particle collection device;
With
2. The reducing agent addition apparatus according to claim 1, wherein the reducing agent is added by the recovery control means on condition that the combustion removal by the combustion removal control means is in progress.   The particulate collection device and the NOx purification device in the exhaust passage so that the temperature of the reduction catalyst rises as the temperature rises by the combustion removal control means, and the temperature of the reduction catalyst falls within a temperature range suitable for the recovery. The reducing agent addition apparatus according to claim 2, wherein a separation distance (L) is set.   The said reducing catalyst is a substance containing at least silver, The reducing agent addition apparatus as described in any one of Claims 1-3 characterized by the above-mentioned. The reducing agent is a hydrocarbon compound;
The reforming means includes a heating means (30) for heating the reducing agent, and the reducing agent heated to a predetermined temperature or higher by the heating means is partially oxidized by oxygen contained in air. The reducing agent addition apparatus according to claim 1, wherein the reducing agent addition apparatus is reformed.   The said reforming means has an ozone production | generation apparatus (20) which produces | generates ozone, The ozone produced | generated by the said ozone production | generation apparatus is included in the air used for the oxidation of the said reducing agent. The reducing agent addition apparatus described in 1.

Images