JP2012527560A - Method for purifying exhaust gas of an internal combustion engine having a catalytic converter - Google Patents

Abstract

  The present invention relates to a method for purifying exhaust gas of an internal combustion engine having a catalytic converter containing an oxygen storage component. In particular, the present invention restores the optimal filling degree of oxygen storage components for controlled stoichiometric operation after the engine has been operated under lean conditions for a relatively short time or for a relatively long time. About.

Description

The present invention relates to a method for purifying exhaust gas of an internal combustion engine having a catalytic converter containing an oxygen storage component. In particular, the present invention relates to the recovery of the optimum filling degree of oxygen storage components for controlled (lambda-adjusted) stoichiometric operation of the engine after it has been operated under lean conditions.

  In order to purify the exhaust gas of such engines, so-called three-way catalytic converters are used that simultaneously remove carbon monoxide (CO), hydrocarbons (HC) and nitrogen oxides (NOx) from the exhaust gas.

  The air ratio lambda (λ) is often used to describe the composition of the air / fuel mixture supplied to the engine. The air ratio is an air / fuel ratio normalized with respect to stoichiometric conditions. The air / fuel ratio describes how many kilograms of air per kilogram of fuel is supplied to the internal combustion engine. The air / fuel ratio for combustion in stoichiometry is 14.7 for normal engine fuel. At this point, the air ratio lambda is 1. An air / fuel ratio of less than 14.7, or an air ratio of less than 1 is referred to as rich, and an air / fuel ratio of greater than 14.7 or an air ratio of greater than 1 is referred to as lean.

  If there is no storage effect for a particular component of the exhaust gas in the internal combustion engine, the air ratio of the exhaust gas corresponds to the ratio of the air / fuel mixture supplied to the engine. In order to obtain a high conversion of the three pollutants, the air ratio lambda must be set in a very narrow range around λ = 1 (stoichiometric conditions). The interval around λ = 1 where all three contaminants are converted by at least 80% is often referred to as the lambda window.

To compensate for fluctuations in oxygen content in the exhaust gas, the three-way catalytic converter stores oxygen under lean exhaust gas conditions (λ> 1) and oxygen under rich exhaust gas conditions (λ <1). An oxygen storage component (OSC), which sets the exhaust gas stoichiometry to λ = 1. Compounds that allow a change in oxidation state are suitable as oxygen storage components in catalytic converters. Preferably, cerium oxide is used, which can exist both as Ce ₂ O _{3 and} as CeO ₂ . In order to stabilize the cerium oxide, it is used, for example, as a mixed oxide with zirconium oxide.

  In the following, the storage capacity of the oxygen storage component should be understood to mean the mass of oxygen that can be absorbed by the oxygen storage component per gram. Thus, the degree of filling relates to the ratio of the mass of oxygen actually stored to the storage capacity. Storage capacity can be determined experimentally using various methods known to those skilled in the art.

  The aim of controlling the air ratio is to completely fill the oxygen store or prevent substantial emptying. In the case of complete filling of the oxygen reservoir, leakage of lean exhaust gas occurs and therefore nitrogen oxides are released. In the case of substantial emptying, a rich leak occurs, ie carbon monoxide and hydrocarbons are released.

  The signal of the oxygen probe (lambda probe) placed upstream of the catalytic converter (pre-cat probe) in the exhaust gas flow direction is used to control the air ratio. By means of the probe, the air / fuel mixture supplied to the engine is controlled so that the exhaust gas has a stoichiometric composition before entering the catalytic converter. Within the context of this invention, the control is denoted as lambda control. The oxygen probe is usually incorporated downstream of the catalytic converter drive system in addition to the pre-catalyst probe. The target stoichiometry of lambda control can be readjusted by the post-catalyst probe. This is shown as post-cat regulation. Post-catalyst control works in particular to monitor and adjust the degree of filling of the oxygen storage of the catalytic converter.

  The probe generates a voltage according to the oxygen content of the exhaust gas. For this, a two-point lambda probe is usually used, which is also indicated as a step change lambda probe. Under lean exhaust gas conditions, the lambda probe has a voltage of about 0.2V, which is from 0.2V to above 0.7V in a very narrow lambda section during the transition to rich exhaust gas. Jump up. Here, the post-catalyst control is configured to produce a probe voltage of about 0.65V. This point is on the steepest branch of the characteristic curve of the probe and corresponds to an optimum filling degree of about 50% oxygen storage. In this method, it is possible to easily detect and correct the deviation of the exhaust gas from the stoichiometry to the higher side or the lower side.

  Spark ignition engines are primarily operated with stoichiometric air / fuel mixtures. However, if the engine no longer outputs power, the fuel supply is usually shut off. In this so-called overrun fuel cutoff, only air is supplied to the engine, so that the exhaust gas composition corresponds to the ambient air.

  During inertial running fuel shut-off, the oxygen storage component of the catalytic converter is fully saturated or filled with oxygen. Post-catalyst control is not possible during inertial fuel cutoff. Apart from inertial running fuel shut-off, complete filling of the oxygen store can also occur in other driving situations, for example due to control errors in lambda control.

  After the end of inertial fuel cutoff, controlled stoichiometry operation should be resumed as quickly as possible. For this purpose, however, it is first necessary that the filling of the oxygen store is restored to an optimum value of about 50%. For this reason, the engine is usually run for a while using a rich air / fuel mixture after the inertial running fuel shut off. The short run with a rich air / fuel mixture is also indicated as a rich pulse. Regular post-catalyst control is resumed only when the oxygen storage charge is returned to about 50%. Alternatively, it is also known that post-catalyst control is directly reactivated after the end of inertial running fuel shut-off. Both methods have the disadvantage that it takes a relatively long time to set the optimum conditions for lambda control. Undesirable release may occur during these times.

  DE 10 2004 038 482 B3 relates to the setting of the degree of filling of the oxygen store after a transient operating state of the engine, for example after inertial fuel cut-off. In the case of an inertial running fuel shut-off, the oxygen store should be quickly evacuated to an optimum value of about 50% of its filling. For this purpose, a rich air / fuel ratio λ <1 is set, and then adjusted to return to 1 at an optimum speed.

  DE 102004019831 A1 provides for the undesired oxygen loading of an exhaust gas catalytic converter during the coasting fuel cutoff phase by supplying the catalytic converter with a defined mass flow rate having a predetermined lambda value. To prevent.

  DE 102006044458 A1 likewise relates to fuel injection after inertial running fuel shut-off. Here, during the first fuel injection after the end of inertial running fuel shut-off, the fuel pulse width is set so that the fuel supply volume increases significantly with respect to the air volume at the inlet, and the ignition time is One delay ignition time is set. During the second and subsequent fuel injections, a fuel pulse width having a smaller increase in fuel is set, and the ignition time is set to a second delayed ignition time, the second delayed ignition time being the first Delayed to a degree less than the delayed ignition time.

  In the known method, the inventors have observed that a rich pulse after the inertial running fuel shut-off triggers a temporary release of carbon monoxide and hydrogen. Said release lasts for about 100 seconds and has a carbon monoxide concentration of up to 10-500 ppm, so that after-catalyst control is interrupted and delayed after the inertial fuel cutoff.

  Accordingly, an object of the present invention is to define a method that can facilitate the transition from inertial running fuel shut-off to controlled stoichiometric operation.

  This problem is solved by the method defined in the main claim. Preferred embodiments are claimed in the dependent claims.

  The method comprises an internal combustion engine having a catalytic converter comprising an oxygen store composed of oxygen storage components, equipped with an electronic engine control mechanism and controlled over most of its operating period. The present invention relates to purification of exhaust gas of the internal combustion engine, which is operated using a fuel mixture, and a temporary lean operation phase is generated depending on an operation state.

  The method allows the engine to be rich after a temporary lean operating phase of the engine with a lean air / fuel mixture, associated with substantial filling of oxygen storage, and before resuming controlled engine operation. The amount of oxidizing component supplied to the catalytic converter by the lean pulse is restored to the optimum level for stoichiometric operation, thanks to being supplied using the pulse and then the lean pulse. Is characterized by being lower than may be required to fully compensate for the amount of rich exhaust gas component supplied by the rich pulse.

  In the present invention, when the short rich pulse is followed by the short lean pulse after the coasting fuel cutoff, the optimal filling degree of the oxygen storage for stoichiometric control of the air / fuel ratio is Based on the observation that it can be recovered very quickly. Here, the rich and lean pulses are generated by a corresponding adjustment of the air / fuel ratio supplied to the engine. This preferably occurs thanks to a pre-catalytic lambda probe that predefines a corresponding time series lambda profile. Pre-catalyst control after the lambda profile is finished and an optimal charge of the oxygen store has been reached (it can be identified by a post-catalyst signal voltage of about 0.6 to 0.7 volts, preferably 0.65 volts) And regular lambda control using post-catalyst control is resumed.

  The inventors have found that the oxidation (filling) or reduction (evacuation) of the oxygen store in the exhaust gas constitutes an equilibrium process. The paper entitled “Is Oxygen Storage in Three Way Catalysts an Equilibrium Controlled Process?” By the present inventors has been accepted by the journal “Applied Catalysis B: Environmental”.

  In steady state operation, the equilibrium state of the oxygen store is always set using the reduction and oxidizing components of the exhaust gas, i.e., the oxygen store by carbon monoxide, hydrogen or hydrocarbons in the equilibrium state. The reduction of is accurately compensated by the corresponding oxidation together with carbon dioxide and water.

  An important consequence of the equilibrium behavior is that the maximum attainable evacuation of the deposit depends on the stoichiometry of the exhaust gas. For example, at lambda = 0.95, the deposit is more fully reduced (emptied) than lambda = 0.99.

  A further consequence of the equilibrium behavior is that the fully evacuated oxygen store is also partially re-oxidized by the moderately rich exhaust gas until a new equilibrium state is established using the moderately rich exhaust gas. Is Rukoto. Here, the oxygen reservoir forms carbon monoxide and hydrogen components by reaction with water or carbon dioxide. The situation occurs according to the prior art when the oxygen store is emptied again with rich pulses only after the inertial running fuel shut-off. The single rich pulse significantly reduces (thoroughly evacuates) the oxygen store. Said exhausted oxygen store reacts with stoichiometry or slightly rich exhaust gas after a rich pulse, which produces carbon monoxide and hydrogen for a period of 10 to several hundred seconds To do. Typical concentrations of the carbon monoxide and hydrogen emissions are about 10 ppm to 500 ppm. The pollutant emissions may be reduced slightly if the rich pulse does not end abruptly but rather slowly returns to the stoichiometric value. However, this prolongs the time between the end of coasting fuel cutoff and the resumption of controlled operation, with the risk of further pollutant emissions.

  In a more accurate analysis of the method, it is also necessary to take into account the distribution of oxygen storage oxidation and reduction along the catalytic converter. The rich pulse first strikes the inlet end surface of the catalytic converter. Even if the rich pulse is sized so that the entire oxygen storage of the catalytic converter can only be partially emptied, a thorough evacuation of the oxygen storage is still present at the front of the catalytic converter. And therefore delayed release of carbon monoxide and hydrogen occurs as a result of the pulse. In the method, the rear part of the catalytic converter is only partially emptied. In the most preferred case, carbon monoxide and hydrogen released from the front of the catalytic converter can empty the rear of the catalytic converter to a desired degree. However, in this case, due to the slow release of carbon monoxide and hydrogen, the oxygen store is completely evacuated throughout the length of the catalytic converter, and the stoichiometry downstream of the exhaust gas of the catalytic converter is steady state. It takes 10 to 100 seconds to correspond to the value of.

  The aforementioned carbon monoxide and hydrogen emissions in vehicles operated by the prior art adversely affect the stability of restarting post-catalyst control. A probe located downstream of the catalytic converter is used for post-catalytic control. As a result of the carbon monoxide and hydrogen emissions that occur in the catalytic converter, post-catalytic control is directed in the wrong direction by the overall rich exhaust gas. Post-catalyst control attempts to make the lean offset more lean by compensating for the rich offset by setting the air / fuel mixture supplied to the engine. As a result of the leaning, the oxygen store is again filled with oxygen, contrary to the actual purpose of control. In the filled state, nitrogen oxide leakage occurs at the slightest lean excursion of the exhaust gas. The results of the above phenomenon prove that it is often difficult to switch to correct operation of post-catalyst control after inertial fuel cutoff. One solution to the problem is to stop the post-catalyst control for a specific time after the inertial fuel cutoff. However, the solution described above is not optimal because the catalytic converter is operated uncontrolled for a relatively long time.

  The carbon monoxide and hydrogen emissions described above following substantial reduction of the oxygen store have adverse effects, not only after the inertial running fuel shut off. Even in normal operation, short control errors can occur, especially in the dynamic operating phase, which leads to a complete filling of the oxygen reservoir. If the reservoir is substantially filled and at the same time the exhaust gas stoichiometry deviates to the lean region for some time, a lean leak occurs and is recorded by the post-catalyst probe. As described in the introduction, the signal of the post-catalyst probe is used to readjust the lambda controlled target stoichiometry. In this case, the post-catalyst control results in an air / fuel mixture being supplied to the engine and enriched again. The enrichment has the same effect as the rich pulse after the inertial running fuel shut-off: The oxygen store is first very thoroughly emptied. Said thorough emptying results in the emission of carbon monoxide and hydrogen as described above. Post-catalyst control is the reaction of the leaning of the air / fuel mixture supplied to the engine, which can react to the new lean leakage as the post-catalyst probe voltage decreases. Decreasing the post-catalyst probe voltage initiates the above-described processes of carbon monoxide and hydrogen emissions, leaning, and lean leakage from the beginning. Thus, periodic oscillations in the exhaust gas stoichiometry occur with periodic lean leakage and corresponding nitrogen oxide emissions. The vibration behavior is well known to control engineers. In order to prevent vibrations, the control response time must be increased by adjusting the control parameters. The solution described above is naturally not optimal, because the lambda displacement that inevitably occurs in the driving operation as a result of the reduced control speed can only be compensated with an unnecessarily prolonged response time.

  According to the present invention, after the end of the lean operation phase, using at least one rich pulse and one lean pulse, the filling of the oxygen store is returned to the optimum level for the subsequent post-catalyst control. The aforementioned problems of conventional methods are reduced or even completely eliminated.

  Here, the amount of rich exhaust gas component is higher than required to set the optimum filling degree for stoichiometric operation, but the complete emptying of oxygen storage capacity Preferably less than the amount of rich exhaust gas component that may be required for

  Thus, according to the present invention, a rich pulse is first used that can empty the catalytic converter over its entire length. Here, the front of the storage is emptied thoroughly. Said exhaustive emptying at the front is terminated by a relatively small lean pulse. Lean pulses inevitably fill the small zone at the inlet of the catalytic converter again, beyond the optimum degree of filling, in order to also fill the back of the zone that has been thoroughly exhausted beforehand. This can be compensated by an additional rich pulse, such that the amount of rich component supplied thereby is less than that required to fully compensate the preceding lean pulse. Selected.

  The amount of reducing agent in the first rich pulse must be greater than the equivalent amount of oxygen to be withdrawn from the catalytic converter during the transition from the fully oxidized state to the stoichiometric operating state. Therefore, first, the catalytic converter is thoroughly emptied. However, the amount of reducing agent in the first rich pulse is preferably selected to be less than the oxygen equivalent that can be withdrawn from the catalytic converter by the steady state rich operating mode.

  The pulse sequence is preferably also set during the controlled operation of the catalytic converter at the operating point after the end of the pulse sequence, depending on the operating condition of the engine and the aging condition of the catalytic converter. Configured to correspond to the distribution obtained. The optimum pulse sequence can be identified in that the post-catalyst probe voltage stably exhibits the post-catalyst control set value after the end of the pulse sequence. The amplitude and / or duration of rich and lean pulses can be used as a dominant variable for the optimization. The amplitude and / or duration can be optimized depending on exhaust gas temperature and space velocity and / or aging conditions of the catalytic converter.

  If the sequence of rich and lean pulses is not sufficient to fully return the fill level to the optimum value, the engine is supplied with further rich and lean pulses after the first rich and lean pulses And the amount of rich component supplied with each rich pulse is greater than the amount that can be compensated with the oxidizing component of the next lean pulse. The optimum number of subsequent lean / rich pulses can be determined according to the operating conditions after the inertial running fuel shut-off in the preliminary test.

  The method is preferably used for exhaust gas purification of stoichiometrically operated internal combustion engines where inertial running fuel shut-off occurs when engine power is no longer needed. In this case, inertial running fuel shut-off forms a temporary lean operating phase. However, the temporary lean operating phase can also be caused by unwanted control fluctuations in stoichiometric operation.

  A further field of use of the invention is the purification of exhaust gases of lean-operated internal combustion engines that are partly stoichiometric and partly lean. In urban traffic, the demand for power is low, so the engine runs leaner and saves fuel. If higher levels of power are needed, the engine must be switched to stoichiometric operation. Therefore, in this case, the oxygen storage in the catalytic converter is completely filled in the lean operation mode, exactly as in the case of coasting fuel cutoff. Switching to stoichiometric operation has the same problems as those encountered after shutting off inertial fuel.

  Undesirable temporary lean operating phases as a result of control errors are preferably detected because the post-catalyst probe indicates lean exhaust gas. For this, a step change probe can be used. If their signal voltage drops below a predefined threshold, there is a temporary lean operating phase according to the present invention. The threshold can be selected according to the exhaust gas temperature and space velocity, according to the exhaust gas stoichiometry, and according to the aging state of the catalytic converter. The threshold is preferably recorded in a table of the engine control device.

  The oxygen storage component of the exhaust gas purification catalytic converter continuously loses its storage capacity as a result of thermal aging. The method makes it possible to measure the remaining storage capacity. The output signal of the oxygen probe located downstream of the catalytic converter in the exhaust zone can be used for this purpose. If the signal voltage is below the expected voltage after the transient lean operation phase to the controlled stoichiometry operation, the catalytic converter's residual oxygen storage capacity is lower than expected. Therefore, the residual oxygen storage capacity can be measured from the signal voltage in the stoichiometric operation mode after the inertia running fuel is cut off. If the residual oxygen storage capacity falls below a predefined value, a corresponding warning signal can be activated.

  By measuring the residual oxygen storage capacity, the amount of rich and lean components supplied to the catalytic converter by rich and lean pulses is adapted to the residual oxygen storage capacity and thereby controlled from stoichiometric fuel cutoff. It is possible to optimize the transition to metric operation. This is preferably done thanks to the rich and lean pulse amplitudes being reduced by a factor corresponding to the residual oxygen storage capacity. This factor can be recorded in a table of the engine control unit depending on the residual oxygen storage capacity.

  It is advantageous to record an average value for the oxygen storage capacity in the engine controller, from which the oxygen storage capacity for the various operating points of the engine can be calculated using correction factors.

  The present invention will be described in more detail with reference to the following drawings.

FIG. 1 shows the release of carbon monoxide / hydrogen in stoichiometric mode of operation following a rich pulse. FIG. 2 shows the resulting profile of the voltage downstream of the catalytic converter's lambda probe for a conventional lambda profile after inertial running fuel shutoff and for two different rich pulses after inertial running fuel shutoff. FIG. 3 shows the resulting profile of the lambda profile according to the present invention after inertial running fuel shut off and the voltage downstream of the lambda probe of the catalytic converter.

  FIG. 1 shows coastal fuel cutoff and carbon monoxide and hydrogen release after returning to stoichiometric mode of operation using a single rich pulse. For the measurement, a conventional three-way catalytic converter was tested in a model gas system.

  The upper figure shows the profile of the air ratio lambda as a function of time (lambda profile). In the first 10 seconds, a coasting fuel cutoff with a lambda value of 1.1 was simulated. After completion of the inertial running fuel shut-off, the oxygen storage of the tested three-way catalytic converter was evacuated to a degree of filling for stoichiometric operation with lambda = 1 using a single rich pulse. The two lower figures show the measured profiles of hydrogen and carbon monoxide concentrations downstream of the catalytic converter, respectively. After a delay time after the rich pulse, hydrogen and carbon monoxide are released by the catalytic converter. The discharge of the two pollutants continues for a time longer than 40 seconds.

  FIG. 2 shows the results of a simulation calculation for the case of a conventional lambda profile after the inertial running fuel shut-off, fully filled with oxygen storage. The calculation was performed for two rich pulses of different lengths with a lambda value of 0.9. The lambda profile upstream of the catalytic converter is shown in the figure above. The lower figure shows the calculated signal voltage of the post-catalyst probe.

  The signal voltage of the post-catalyst probe starts at about 0.1 V and shows a very lean exhaust gas (lean working phase) with a high oxygen content. Oxygen storage has virtually 100% fill. In order to empty the oxygen storage, the exhaust gas is concentrated for some time upstream of the catalytic converter.

  For a rich pulse duration of only 1.0 second (dashed curve), it takes about 17 seconds to raise the signal voltage of the post-catalyst probe to 0.65V. For a 1.4 second rich pulse duration, a signal voltage of 0.65 V has already been achieved after only about 3.5 seconds. However, in both cases, the post-catalyst probe records further shifts in exhaust gas stoichiometry to rich values. After 40 seconds, the probe voltage remains at about 0.75V. This significant rich shift is caused by the aforementioned emissions of carbon monoxide and hydrogen.

  FIG. 3 shows the result of a simulation calculation for the case of a lambda profile according to the present invention. In this example, the exhaust gas upstream of the catalytic converter has two pairs of rich and lean pulses over a duration of approximately 20 seconds to empty the oxygen reservoir. The figure with the signal voltage of the post-catalyst probe reaches the desired 0.65V and already stays at this voltage level after about 4 seconds. Thus, the oxygen store has reached an optimal filling level, averaged over its entire length already after the short time using only one rich / lean pulse pair. Nevertheless, because of the above-mentioned axial distribution of filling levels, further rich / lean pulse pairs are required to optimally set the filling degree over the length of the catalytic converter. The post-catalyst control remains stopped until the end of the last rich / lean pulse pair at about 20 seconds after the end of the preceding lean operating phase at time zero. Only then is post-catalyst control resumed.

Claims (13)

A stoichiometric air / fuel mixture comprising an oxygen storage comprising an oxygen storage component and having a catalytic converter, equipped with an electronic engine controller and controlled over most of its operating period In the method for purifying exhaust gas of the internal combustion engine, which is operated using the engine, and a temporary lean operation phase is generated depending on the operation state,
After the engine's temporary lean operation phase, which is associated with substantial filling of the oxygen storage, and before the restart of the controlled engine operation, the oxygen storage charge is determined by the engine following a rich pulse. The amount of oxidizing component supplied to the catalytic converter by the lean pulse is returned to the optimum level for stoichiometric operation, and the rich exhaust gas supplied by the rich pulse is used. A method characterized in that it is less than the amount that may be required to fully compensate the amount of the gas component.   The amount of rich exhaust gas component supplied by the rich pulse is higher than required to set the optimal filling for stoichiometric operation, but the oxygen storage capacity is full 2. A method according to claim 1, characterized in that it is less than the amount of rich exhaust gas components that may be required for complete emptying.   After the first rich pulse and lean pulse, the engine is supplied with further rich pulses and lean pulses, and the amount of rich component delivered by each rich pulse is compensated by the oxidizing component of the next lean pulse. Method according to claim 2, characterized in that there are more than obtained.   The rich and lean pulses have an amplitude and duration, and the amplitude and / or duration is adapted according to the temperature and space velocity of the exhaust gas and / or the aging condition of the catalytic converter A method according to claim 1 or 2, characterized.   5. The method according to claim 4, characterized in that the rich and lean pulse amplitudes are reduced by a factor corresponding to the aging state of the catalytic converter.   The method of claim 1, wherein the temporary lean operating phase is inertial running fuel shut-off.   The method according to claim 1, characterized in that the temporary lean operating phase is a lean operating phase of an internal combustion engine that is operated both stoichiometric and lean depending on the operating conditions.   The method according to claim 1, characterized in that the temporary lean working phase is caused by control fluctuations in stoichiometric operation.   A temporary lean operating phase is detected by an oxygen probe located downstream of the catalytic converter indicating lean exhaust gas when the signal voltage of the oxygen probe drops below a threshold value The method according to claim 8.   The method according to claim 9, characterized in that the threshold value is selected according to the exhaust gas temperature and space velocity, according to the exhaust gas stoichiometry and according to the aging state of the catalytic converter.   An oxygen probe is located downstream of the exhaust zone of the catalytic converter, and the signal voltage that the oxygen probe actually reaches after the transient lean operation phase to controlled stoichiometric operation is the oxygen storage from them. The method according to claim 1, wherein the method is used for measuring the residual oxygen storage capacity of an object.   12. A method according to claim 11, characterized in that a signal is activated when the residual oxygen storage capacity falls below a predefined value.   13. A method according to claim 12, characterized in that the amount of rich and lean components supplied to the catalytic converter by rich pulses and lean pulses is adapted to the residual oxygen storage capacity.

Images