当前位置:首页 >> 交通运输 >>

Ambient temperature light-off for automobile emission control


Applied Catalysis B: Environmental 18 (1998) 123±135

Ambient temperature light-off for automobile emission control
David S. Lafyatis1,a,*, Graham P. Ansella, Steven C. Bennetta, Jonathan C. Frosta, Paul J. Millingtona, Raj R. Rajarama, Andrew P. Walkera, Todd H. Ballingerb
a

Johnson Matthey Technology Centre, Blount's Court, Sonning Common, Reading RG4 9NH, UK b Johnson Matthey CSD, 436 Devon Park Drive, Wayne PA 19087, USA

Received 21 November 1997; received in revised form 10 March 1998; accepted 13 March 1998

Abstract The time taken for an exhaust emission-control catalyst to reach its operating temperature for hydrocarbon oxidation is a major barrier to achieving ultra-low emissions from vehicles. A new approach for achieving rapid catalyst light-off from a vehicle cold-start has been devised and demonstrated on an automobile. An important component in the system is a Pd±Ptbased catalyst which, under net lean conditions in an exhaust stream containing high levels of CO, is active at ambient temperature for the highly exothermic CO oxidation reaction. The Pd±Pt catalyst has positive-order kinetics with respect to CO for the CO oxidation reaction; hence, increasing the level of CO in the feed leads to increasing reaction rates and a faster temperature rise for the catalyst. In practice, this means that enriching the air to fuel mixture supplied to the engine at coldstart (with a secondary air source to provide at least the required amount of oxygen for complete CO conversion in the exhaust) enables the catalyst to reach operating temperature within seconds of starting the engine. In the present work, this ? ) and hydrocarbon trap (zeolites H-ZSM-5 ambient temperature catalyst is combined with an upstream water trap (zeolite 5 A or H-Beta). The traps delay the exposure of the catalyst to these potentially inhibiting species until it has reached a temperature at which it can effectively combust hydrocarbons. When tested fresh, this system demonstrated high levels of hydrocarbon conversion throughout the start-up phase of a Federal Test Procedure cycle.# 1998 Elsevier Science B.V. All rights reserved. Keywords: Automotive emissions; Three-way catalysts; CO oxidation; Ambient temperature light-off; H2O trap; Hydrocarbon trap

1. Introduction Over the last 20 years, the regulated emissions of CO, hydrocarbons (HC) and NOx from mobile sources have been consistently reduced [1]. Legislation in California, such as the ultra-low emission vehicle
*Corresponding author. Tel.: 610-341-8208; fax: 610-341-3495; e-mail: Lafyad@matthey.com 1 Present address: 436 Devon Park Dr, Wayne, PA 19087, USA 0926-3373/98/$19.00 # 1998 Elsevier Science B.V. All rights reserved. PII S0926-3373(98)00032-0

(ULEV) standard, is particularly demanding with respect to lowering hydrocarbon emissions. It has long been recognized that the key to reducing hydrocarbon emissions is rapidly increasing the temperature of the catalyst when the vehicle is ?rst started (a cold-start) [2]. Typically, 80±90% of the hydrocarbon emissions from automobiles equipped with modern three-way catalysts (TWC) occur during the cold-start. Thus, a typical vehicle may fail stringent emission standards such as ULEV within the ?rst 30 s of the 30 min

124

D.S. Lafyatis et al. / Applied Catalysis B: Environmental 18 (1998) 123±135

Federal Test Procedure (FTP) cycle if the hydrocarbon emissions from cold-start are not controlled. Several potential solutions to the cold-start problem have been proposed. These can be divided into two categories. First, there is a series of ideas which are based upon methods of quickly bringing the catalyst to working temperature. Close-coupled or manifoldmounted catalysts are placed in positions very near the engine, thus reducing the time necessary for the heat of the engine exhaust to increase the catalyst temperature [2,3]. In the electrically heated catalyst (EHC), electrical power is provided to heat the catalyst at start-up [4±7]. Exhaust-gas ignition (EGI) works by deliberately running the engine under very rich conditions (air-to-fuel (A/F) ratio%9) so that large quantities of hydrogen are produced in the exhaust, which is then ignited by a glow-plug upstream of the catalyst [8,9]. Another proposal is the combustion heated catalyst (CHC) in which hydrogen and oxygen are fed to the catalyst prior to starting the engine, and the exothermic combustion reaction heats the catalyst [10]. A variety of heat storage devices have also been suggested, all of which work on the principle of retaining heat from the time the car was last shutdown until the following cold-start. The second category of cold-start solutions involves trapping hydrocarbons during cold-start for release after the catalyst has reached operating temperature. Various solutions involving hydrocarbon adsorbents such as activated carbons or zeolites have been proposed [11±17]. The simple solution of placing a monolith coated with a hydrocarbon adsorbent upstream from the catalyst is problematic, because the thermal mass of the adsorbent monolith delays the heat from the engine exhaust reaching the catalyst. As the adsorbent temperature rises, the stored hydrocarbons desorb and pass over the catalyst which is generally still below its light-off temperature and, therefore, unable to convert the hydrocarbons. Ideas to get around this heat management problem have often involved the use of bypass valves in the exhaust [12,13]. In one recent example, a ?uidic valve is used to direct the exhaust stream through a hydrocarbon trap at cold-start; after the exhaust stream has become hot the gas is re-directed to the catalyst while the hydrocarbons slowly desorb from the trap [14]. Catalyzed hydrocarbon traps, in which the HC trap and the catalyst are coated onto the same monolith, have

also been proposed to lessen the heat management problem involved with HC traps [15]. Another suggestion for heat management in a hydrocarbon adsorbent system is to coat a catalyst onto a heat-exchanger with two ?ow paths. In this system, the exhaust gas ?rst passes through the catalyst/heat exchanger, then through the HC trap, and, ?nally, again through the same catalyst/heat exchanger [16,17]. A concept which does not involve the use of an adsorbent is the storage of the entire exhaust stream into a collection bag underneath the car during cold-start, the contents of which are released to the catalyst after it has reached operating temperature [18]. In general, all of these cold-start solutions bene?t further through the application of improved catalyst technology (so-called `low-light-off catalysts') which begin to operate at lower temperatures [2,19], and (with the exception of EGI) leaner starting engines which provide an exhaust with less unburnt hydrocarbon and more oxygen at start-up. However, all of the solutions discussed above have inherent disadvantages. Close-coupled catalysts are located in the valuable space near the engine compartment (which is inconvenient for engine design and can result in a loss of power) and also must be robust to very high temperature exposure. EHCs are bulky and require large amounts of power at start-up, often requiring the use of a second battery. EGI systems require the complexity of implementing a glow-plug into the exhaust stream. Feeding hydrogen to the catalyst prior to starting the engine is inconvenient, and requires a ready source of hydrogen on-board. Heat storage devices are bulky, involve the use of expensive materials, and may be dif?cult to fabricate. HC trap strategies involving valves in the exhaust are unpopular due to their inherent complexity, and there are concerns about the durability of a valve in the corrosive exhaust environment. Obtaining proper overlap between the hydrocarbon desorption temperature and the catalyst light-off temperature is dif?cult in catalyzed HC trap systems. The catalytic heat exchanger requires a catalyst with multi-directional ?ow channels, and inevitably leads to a 3608 turn in the exhaust pipe. The exhaust collection bag is bulky, particularly for small vehicles. This paper discusses a novel method for controlling cold-start hydrocarbon emissions. Many of the oxidation reactions which occur over an autocatalyst are

D.S. Lafyatis et al. / Applied Catalysis B: Environmental 18 (1998) 123±135

125

highly exothermic. For example, the oxidation of 1% CO in a nitrogen stream leads to an adiabatic temperature rise of 978C. At start-up, a typical vehicle runs at an A/F ratio of between 12 and 13, providing an exhaust stream containing between 3 and 6% CO and, additionally, 0.5±2% hydrogen. The catalytic oxidation of CO and hydrogen at temperatures well-below ambient is known in the literature [20]. Indeed, the use of hydrogen prior to start-up to heat a converter from ambient temperature was demonstrated in the CHC [10]. The current paper demonstrates the use of a catalyst with ambient temperature activity for CO oxidation in an automotive exhaust. Use of such a catalyst in an exhaust stream containing high levels of CO (with suf?cient oxygen provided to satisfy the stoichiometric requirements of the combustion reaction) provides a method for very rapid catalyst heating from cold-start. Combining this with an HC trap system leads to excellent cold-start performance, providing the basis for a system capable of meeting very demanding emission standards. 2. Experimental 2.1. Catalyst and trap materials The catalyst used in this study contained Pt and Pd and CeO2. The catalyst was prepared by co-precipitation of palladium nitrate, chloroplatinic acid and cerium nitrate (further details of the preparation of such catalysts may be found in Refs. [21±23]). The catalyst was calcined in air at 5008C prior to use. This catalyst is referred to as the Pd±Pt ambient temperature light-off (ATL) catalyst for the remainder of this paper. The zeolites 5A, H-ZSM-5 and H-Beta (silica : alumina ratio of 50 : 1) were investigated as water and hydrocarbon traps. 2.2. Microreactor studies Catalyst and trap materials were studied in a stainless steel microreactor with an internal volume of %0.95 cm3 (1 cm long, 1.1 cm diameter). The catalyst was examined as a 250/355 mm particle size powder. In these powder experiments, the active catalyst phase was diluted with cordierite to simulate the thermal

effect of the catalyst being coated onto an inactive ceramic support. The catalyst was also examined following coating onto a 400 cells per square inch (cpsi) ceramic monolith core at a loading of 190 g/l. Hydrocarbon and water adsorbers were also examined in the microreactor. The zeolite adsorbents were coated onto a 400 cpsi monolith core at a loading of 95 g/l. All microreactor experiments utilizing monolith cores were performed at GHSV?30 000 h?1 (i.e. comparable to exhaust stream space velocities). A single thermocouple located immediately in front of the catalyst bed was used to measure the microreactor inlet temperature. The reactor was designed to allow sharp steps in inlet concentrations to simulate the abrupt introduction of reactant to the catalyst at automobile key-on. Prior to an experiment, argon and oxygen were allowed to ?ow over the catalyst at ambient temperature. The experiment began by the introduction of a selected series of additional components to the inlet stream. The simulated exhaust streams were monitored by the following analyzers: infrared analyzers to monitor CO and CO2 concentrations, a chemiluminescence analyzer to monitor NOx concentration, an electrochemical cell for oxygen analysis and a quadrupole mass spectrometer (QMS) to monitor propene, toluene and water. The reactor could be operated either with a constant inlet temperature or by ramping the inlet temperature. 2.3. Engine testing Ambient temperature light-off was also investigated on a 1995 model year, federally calibrated (i.e. approved for use in 49 U.S. states, excluding California) vehicle, with a 2.0 l engine. The original aftertreatment system for this vehicle consisted of two close-coupled catalysts. This system was replaced by an exhaust pipe with room for two 42-in3 monoliths in an under?oor position (%1 m from the engine manifold). Catalyst and trap monoliths were coated to give loadings of 190 and 95 g/l, respectively. Thermocouples were inserted in the monoliths, 1 inch from the front of each of the 3.25-inch long blocks. Testing consisted of repeated Federal Test Procedure (FTP) runs on a dynamometer using fuel containing no sulfur or lead, with modal analysis of CO, hydrocarbons and NOx. Hydrocarbon concentrations were measured using a FID analyzer, CO was measured by infrared,

126

D.S. Lafyatis et al. / Applied Catalysis B: Environmental 18 (1998) 123±135

and NOx was measured by a chemiluminescence analyzer. The catalysts were not aged prior to FTP testing, although at least one pre-conditioning FTP run was performed on each catalyst con?guration. The systems were kept dry in-between FTP runs by a nitrogen ?ow. When the catalysts were not dry prior to FTP testing, a substantial drop in low-temperature performance was apparent. All data shown from engine trials are the averaged results from two consecutive FTP runs. To simulate the effect of increasing the engine richness at start-up and/or providing secondary air for a lean condition over the ambient temperature light-off catalyst, CO and/or air were added to the exhaust stream via metering valves upstream from the catalyst during the ?rst 120 s of some FTP runs. 3. Microreactor results and discussion 3.1. Ambient temperature catalyst behavior There are many examples in the literature of catalysts which will convert CO to CO2 at ambient temperature. For application on a vehicle to achieve a lowering of hydrocarbon emissions, three additional requirements were critical for the success of the lowtemperature catalyst: 1. the catalyst must operate effectively under feeds with high levels of CO (so that the exotherm from this reaction provides suf?cient heat to rapidly increase the catalyst temperature to the point where it will perform hydrocarbon conversion); 2. the catalyst must perform effectively in the presence of other exhaust gas species, such as CO2, H2O, hydrocarbons and NOx; and 3. the catalyst must be durable under engine conditions (both thermally and resistant to poisons). The rate of CO oxidation for many catalysts is negative or zeroth-order in CO and positive-order in oxygen in the range of concentrations normally found in the start-up of an automobile [24]. However, the Pd±Pt ATL catalyst utilized in this work demonstrates much different behavior. The reaction orders for this catalyst are 0.84 for CO (in the 0±0.5% CO range), and 0.00 for oxygen (in the 0±1.0% O2 range). This change in catalyst kinetics means that increasing the level of CO increases the rate of reaction for the highly

exothermic CO oxidation reaction. This heat of reaction may then be used to rapidly raise the catalyst temperature. For a complete combustion of CO, there is the obvious requirement to provide enough oxygen to keep the catalyst at least under stoichiometric conditions. However, the zeroth-order behavior with oxygen concentration indicates that there is little advantage to further increases in the oxygen concentration in the feed. A typical automobile exhaust may contain many components including hydrocarbons, CO2, H2O, NOx and SO2, which could inhibit ambient-temperature CO light-off. This work was carried out in sulfur-free fuel, eliminating the possibility of SO2 inhibition or poisoning. At start-up, very little NOx is produced by the cold engine. Microreactor studies of inhibition effects on the Pd±Pt ATL catalyst showed that CO2 was not a strongly inhibiting species, but that hydrocarbons (e.g. propylene) and water could severely inhibit the low temperature light-off behavior of this catalyst. This is likely due to competition of CO with the hydrocarbons and/or water on the catalyst surface, thus slowing down the rate of CO oxidation. A series of water and hydrocarbon traps were placed upstream of the catalyst to prevent these inhibitors from reaching the catalyst during light-off. These traps are discussed in Sections 3.2 and 3.3. The investigation on the durability of the catalyst used in this study was limited to exposure to repeated runs of the legislated FTP cycle, which approximates real-world driving conditions. The catalyst was shown to be stable under these conditions. 3.2. H2O trap behavior Automotive exhaust contains approximately 10% H2O, and the inhibition effects discussed above show the importance of preventing this water from reaching the catalyst prior to ambient temperature CO light-off. It is well known that zeolite materials are hydrophilic; thus, the use of a zeolite adsorbent to trap moisture from an exhaust stream at light-off was investigated. Experiments were conducted in which 10% H2O was introduced at the inlet of the microreactor. Fig. 1 shows the breakthrough of water at the outlet of the microreactor during a ramped temperature experiment for a monolith core coated with the hydrophilic zeolite 5A (curve B) and a blank (uncoated) monolith core

D.S. Lafyatis et al. / Applied Catalysis B: Environmental 18 (1998) 123±135

127

Fig. 1. Effect of adsorbent on H2O breakthrough, 10% steam, GHSV?30 000 h?1: (a) blank monolith; and (b) zeolite 5A-coated monolith.

(curve A). It is apparent from these data that a zeolitic coating can effectively be used to postpone the elution of H2O (note that due to dif?culties in getting high levels of steam into the system at low temperature, the blank experiment also shows some apparent delay in water breakthrough). This implies that a zeolite 5A trap may be used in an automobile exhaust to reduce the inhibition by water vapor of ambient temperature light-off. 3.3. Hydrocarbon trap behavior The use of zeolite adsorbents to trap hydrocarbons from a vehicle exhaust at engine start-up has been well documented [11±17]. As discussed in Section 3.1, hydrocarbons such as propene inhibit the ambient temperature CO oxidation activity of the Pd±Pt ATL catalyst. Preventing these inhibitors from reaching the catalyst surface at engine start-up will promote the low-temperature light-off behavior of the catalyst. Fig. 2(A) and (B) illustrate the function of the hydrocarbon trap in microreactor experiments. In these temperature-ramped breakthrough experiments, the reactor feed containing 600 ppm propene, 90 ppm toluene and 4% O2 was introduced as a sharp step at the beginning of the experiment. Experiments were performed in the presence, and in the absence, of 10% H2O. The hydrocarbon breakthroughs at the reactor

outlet for a blank core and for a core coated with zeolite Beta were compared. Fig. 2(A) plots the result for propene elution. For the blank reactor (curve A), propene breakthrough occurs immediately and remains at a stable value throughout most of the experiment. At high temperatures (e.g. after 200 s in the experiment), hydrocarbon is burned on the hot metal pipework leading to the reactor and, thus, the propene signal falls off. The same breakthrough experiment was then conducted on a core coated with zeolite Beta (curve B). In the presence of 10% steam, propene elution is delayed for a short time (roughly 10 s) before elution. Thus, there is some adsorption shown by the hydrocarbon trap, but propene breakthrough is still rapid. For the ?nal experiment on zeolite Beta, the coated core was studied in the absence of steam (curve C). In this case, propene was strongly held by the dry zeolite, with elution occurring much more slowly. Fig. 2(B) plots the breakthrough of toluene from the same set of experiments. It is evident that, even in the presence of 10% steam, toluene is effectively trapped by the hydrocarbon adsorbent. In the experiment conducted in the absence of steam, toluene adsorption was also improved slightly. It is clear from these experiments that a hydrocarbon trap may be used to delay hydrocarbon elution from an exhaust stream. From the results in Fig. 2(A)

128

D.S. Lafyatis et al. / Applied Catalysis B: Environmental 18 (1998) 123±135

Fig. 2. (A) Effect of hydrocarbon adsorbent on propene breakthrough, 600 ppm propene, 90 ppm toluene, 4% O2, balance Ar, GHSV?30 000 h?1: (a) blank monolith; (b) zeolite Beta coated monolith, in the presence of 10% steam; and (c) zeolite Beta coated monolith, dry. (B) Effect of hydrocarbon adsorbent on toluene breakthrough, 600 ppm propene, 90 ppm toluene, 4% O2, balance Ar, GHSV?30 000 h?1: (a) blank monolith; (b) zeolite Beta coated monolith, in the presence of 10% steam; and (c) zeolite Beta coated monolith, dry.

and (B), it can also be inferred that the presence of a water trap upstream from the hydrocarbon trap and the Pd±Pt ATL catalyst will not only have the bene?t of removing the inhibiting effect of moisture on catalyst light-off, but will also improve the performance of the hydrocarbon trap for certain hydrocarbon species such as propene. Improvement in hydrocarbon trapping due

to the presence of an upstream water trap has already been demonstrated on a vehicle [25], and is most likely due to competitive adsorption effects within the zeolite framework. Heavier species such as toluene compete effectively with water for adsorption sites, and are much less affected by the presence of the water trap.

D.S. Lafyatis et al. / Applied Catalysis B: Environmental 18 (1998) 123±135

129

Fig. 3. (A) Light-off test under lean conditions with 0.5% CO in feed, Pd±Pt ATL catalyst, GHSV?30 000 h?1: (a) CO conversion; (b) propene conversion; and (c) toluene conversion. (B) Light-off test under lean conditions with 4% CO in feed, Pd±Pt ATL catalyst, GHSV?30 000 h?1: (a) CO conversion; and (b) propene and toluene conversions.

3.4. Simulated catalyst light-off experiments Temperature-ramped microreactor light-off experiments were also performed on cores coated with the Pd±Pt ATL catalyst (190 g/l). No water or hydrocarbon traps were utilized in these experiments, so the CO light-off behavior is expected to be degraded. Fig. 3(A) and (B) show the results from two light-

off experiments (the temperature ramp is shown on the ?gures). The reactor concentrations for these lean light-off experiments are: NO ? 500 ppm, propene ? 400 ppm, toluene ? 100 ppm, CO2 ? 15%, H2O ? 10%, O2 ? 3%, CO ? 0.5% (Fig. 3(A)) or 4.0% (Fig. 3(B)), Ar ± balance. Fig. 3(A) shows the CO (curve A) and hydrocarbon (curves B and C) conversion during a light-off experi-

130

D.S. Lafyatis et al. / Applied Catalysis B: Environmental 18 (1998) 123±135

ment conducted with the low CO concentration in the feed. The CO conversion in this experiment is initially high, then falls to roughly 40%, and ?nally increases to 100%. This behavior can be explained as follows. The Pd±Pt ATL catalyst is able to convert CO at low temperatures, leading to the high initial conversion. However, as inhibitors such as H2O and hydrocarbons accumulate on the catalyst surface at low temperatures, conversion begins to decline. Eventually, the inlet temperature to the catalyst increases to the point where the inhibitors (particularly water) begin to desorb, allowing the CO conversion to reach 100%. The conversion of propene (curve B) and toluene (curve C) begins much later, and does not occur to any signi?cant extent until 75 s into the experiment. In this system, the ambient temperature activity of the catalyst for CO oxidation has not been fully utilized for hydrocarbon removal. Fig. 3(B) shows the same experiment with the high (4%) CO concentration in the feed. The high level of CO is completely converted instantaneously in this experiment, leading to a large exotherm in the catalyst bed. This exotherm causes the catalyst temperature to rise swiftly, preventing hydrocarbon and water inhibition and leading to high levels of hydrocarbon conversion from the very beginning of the experiment (the propene and toluene conversion lines overlay each other on this plot). The laboratory light-off behavior of this catalyst was unchanged following calcination at 6008C for 2 h. 4. Engine results and discussion 4.1. Engine conditions Following these successful laboratory tests showing the possibilities of ambient temperature light-off, FTP testing was performed on a standard production vehicle. Fig. 4(A)±(C) show the essential characteristics
Table 1 Catalyst configurations tested on vehicle Configuration title ATL catalyst/blank Blank/ATL catalyst Traps/ATL catalyst Front position

during start-up for the vehicle utilized during these tests. Fig. 4(A) shows the air/fuel ratio (AFR) trace during the cold-start portion of the FTP test. Also shown for reference is the vehicle speed trace for the test. As can be seen, the car starts at a rich AFR condition, which quickly moves toward stoichiometric AFR as the vehicle warms up. After %60 s, the engine control becomes reasonably tight around the stoichiometric AFR of 14.6. Fig. 4(B) shows the CO concentration in the vehicle exhaust during the cold-start portion of the FTP test. As would be expected from the AFR trace of Fig. 4(A), during the rich start-up there is a fairly high level of CO in the exhaust, but this level quickly reduces as the engine warms and begins to move toward stoichiometric conditions. As was shown in Fig. 3(B), there are potential light-off bene?ts of using increased CO levels at start-up. Thus, also shown for comparison in Fig. 4(B) is the CO level in the exhaust during runs in which additional CO and air were injected into the exhaust stream (curve B), simulating the effect of a richer start from the engine combined with a secondary air source to give net lean conditions over the catalyst. Fig. 4(C) shows the inlet temperature (curve A) to the under?oor catalyst system measured during cold-start of the FTP test, and also measurements made 1 inch into two uncatalyzed monoliths in the under?oor position (curves B and C). The heat of the exhaust gas from the engine causes the temperature to rise rapidly at the inlet to the under?oor system; however, due to their large thermal mass the temperatures of the under?oor monoliths increase much more slowly. For example, the temperature of the rear uncatalyzed brick remains below 508C for the ?rst 25 s of the test after start-up. 4.2. Ambient temperature light-off on an engine The three system con?gurations examined are shown in Table 1. All tests for a given con?guration

Rear position blank Pd±Pt ATL catalyst Pd±Pt ATL catalyst

Pd±Pt/ATL catalyst blank 1/2 brick zeolite 5A H2O adsorber 1/2 brick zeolite ZSM-5 HC adsorber

D.S. Lafyatis et al. / Applied Catalysis B: Environmental 18 (1998) 123±135

131

Fig. 4. (A) AFR and speed trace for test vehicle at start of FTP test. (B) CO concentration in the vehicle exhaust during an engine cold-start: (a) standard vehicle start-up; and (b) vehicle start-up with enhanced CO level to simulate a richer start-up. (C) Temperatures in exhaust during cold-start measured: (a) at the inlet to the underfloor catalysts (1 inch in front of front monolith); (b) 1 inch into the front uncatalyzed monolith; and (c) 1 inch into the rear uncatalyzed monolith.

132

D.S. Lafyatis et al. / Applied Catalysis B: Environmental 18 (1998) 123±135

Fig. 5. (A) CO and hydrocarbon conversions at start of FTP in ATL catalyst/blank configuration, using normal vehicle start-up conditions: (a) CO conversion; and (b) hydrocarbon conversion. (B) CO and hydrocarbon conversions at start of FTP in ATL catalyst/blank configuration, using increased CO level at vehicle start-up: (a) CO conversion; and (b) hydrocarbon conversion.

were run on the same set of traps and catalysts. However, when a new con?guration was tested, the catalyst was changed to a fresh sample. Fig. 5(A) and (B) show the results during start-up of the ATL catalyst/blank experiment. Fig. 5(A) shows the CO and hydrocarbon conversions for the case where additional air was added during the ?rst 120 s after start-up in order to keep the under?oor catalyst lean, but no additional CO has been added. As can be seen from Fig. 5(A), the performance of the

catalyst is excellent for CO oxidation (curve A) under these conditions, with 50% CO conversion after roughly 20 s and reaching 100% CO conversion at 35 s. The hydrocarbon performance (curve B) is also good under these conditions, but it is apparent that there is virtually no hydrocarbon conversion during the ?rst 30 s of the test cycle. It should be noted that the apparent negative hydrocarbon conversion during the ?rst 20 s is due to the desorption of hydrocarbons that adsorb on the catalyst, exhaust walls, and sample

D.S. Lafyatis et al. / Applied Catalysis B: Environmental 18 (1998) 123±135

133

lines between tests. As discussed in Section 1, this car could fail strict hydrocarbon emission standards in the ?rst 30 s based on this poor hydrocarbon performance. Fig. 5(B) shows the conversions from the same con?guration, but with additional air and CO added (as shown in Fig. 4(B)) to simulate an enriched startup strategy with secondary air. Comparison of Fig. 5(A) with Fig. 5(B) shows that the light-off of CO is much more rapid in the case where additional CO has been added to the feed. This illustrates the effect of the positive order kinetics that this catalyst possesses for CO oxidation. The addition of higher levels of CO leads to an increased reaction rate and, thus, more heat generation from the exothermic reaction. The temperature 1 inch into the catalyst block is 308C higher after 20 s and 608C higher after 30 s when the run was performed with the increased level of CO. Comparing the hydrocarbon conversions of Fig. 5(A) with Fig. 5(B) shows how this more rapid temperature rise translates into improved hydrocarbon lightoff performance for the catalyst. Substantial (50%) hydrocarbon conversion now occurs within 20 s of start-up, compared to the 40 s necessary for this level of conversion to be obtained with the original calibration. This ?rst set of tests indicated the advantages of increased levels of CO under net lean conditions for this catalyst which performs CO oxidation at low temperatures. Fig. 6(A) shows the results from experiments on the blank/ATL catalyst con?guration under the increased CO condition. This con?guration clearly demonstrates the effects of the ambient temperature activity of the catalyst. The CO performance (curve A) is remarkable, starting at >60% instantaneously at light-off, and quickly increasing to 100% within 15 s. The hydrocarbon performance (curve B) is also outstanding, with conversion rapidly increasing to >70%, 20 s after engine start-up. These experiments clearly show the ambient temperature light-off activity of the catalyst because, as illustrated in Fig. 4(C), the rear under?oor position receives virtually no heat from the engine during the early stages of the test. The temperature rise measured on the front and rear monoliths during light-off is shown in Fig. 6(B). The temperature of the catalyzed rear monolith (curve B) rises quickly to >2508C, 20 s into the run, while the temperature of the front uncatalyzed monolith (curve A) is still below 508C at this point.

It may be noted when comparing Fig. 6(A) with Fig. 5(B) that the light-off behavior seems better in the blank/ATL catalyst con?guration than in the ATL catalyst/blank con?guration. This may be explained by the condition of the ATL catalyst/blank sample. The catalyst in the blank/ATL catalyst con?guration was probably in a fresher condition (due to fewer coldstart experiments, and being more remote from the engine) than the ATL catalyst/blank con?guration catalyst. It is also possible that the blank brick adsorbs a small amount of water at start-up, thus leading to improved light-off. The data from the ATL catalyst/blank and blank/ ATL catalyst con?gurations show excellent light-off capability. However, data discussed previously indicated that the presence of water and hydrocarbons in the exhaust at engine start-up may inhibit the light-off behavior. It is also clear from Fig. 5(B) and Fig. 6(A) that the hydrocarbon performance for these systems immediately at start-up, before the catalyst has been heated by the exothermic CO oxidation reaction, was still rather poor. Thus, the blank brick in the blank/ ATL catalyst con?guration was replaced with a water and hydrocarbon trapping system. The results from the traps/ATL catalyst con?guration in which additional CO was added to the exhaust stream are shown in Fig. 7. The CO performance is again outstanding, with initial CO conversions at engine start-up being >70% (curve A). However, more signi?cant is the hydrocarbon conversion (curve B) displayed by the system. Comparison with Fig. 6(A) shows the advantage obtained by combining the ambient temperature activity of the Pd±Pt ATL catalyst with an upstream trapping system. The trap system successfully holds %80% of the hydrocarbon emissions present at engine start-up, allowing the catalyst to be heated by the exothermic combustion reactions. When the trap releases these hydrocarbons the catalyst is hot enough for their conversion. Thus, a system has been constructed in which hydrocarbon conversion never falls below 50% throughout the start-up phase of the FTP test. These results demonstrate that a novel Pd±Pt catalyst with positive-order kinetics for the CO oxidation reaction combined with a water and hydrocarbon trap system can be applied effectively to an automotive exhaust treatment system. For example, 68.5% of the cold-start hydrocarbon emissions (initial 120 s of the

134

D.S. Lafyatis et al. / Applied Catalysis B: Environmental 18 (1998) 123±135

Fig. 6. (A) CO and hydrocarbon conversions at start of FTP in blank/ATL catalyst configuration, using increased CO level at vehicle start-up: (a) CO conversion; and (b) hydrocarbon conversion. (B) Monolith temperatures at start of FTP in blank/ATL catalyst configuration, using increased CO level at vehicle start-up, measured (a) 1 inch into front (uncatalyzed) monolith. and (b) 1 inch into rear (catalyzed) monolith.

FTP) were removed in the ATL catalyst/blank con?guration using the normal engine start-up conditions (Fig. 5(A)), while 86.7% of the cold-start hydrocarbon emissions were removed in the traps/ATL catalyst con?guration with the increased CO at start-up (Fig. 7). Such an improvement in hydrocarbon emissions in the initial portion of a test cycle will signi?cantly impact the ability of an automobile to meet stringent hydrocarbon emission regulations.

5. Conclusions A new method for achieving rapid catalyst light-off from a cold-start has been devised and demonstrated under fresh conditions on a vehicle. The technique relies upon a catalyst which will combust CO at ambient temperatures in a vehicle exhaust. The CO oxidation reaction is highly exothermic and, thus, by increasing the levels of CO in the exhaust at start-up

D.S. Lafyatis et al. / Applied Catalysis B: Environmental 18 (1998) 123±135

135

Fig. 7. CO and hydrocarbon conversions at start of FTP in traps/ATL catalyst configuration, using increased CO level at vehicle start-up: (a) CO conversion; and (b) hydrocarbon conversion.

(i.e. by tuning the engine stoichiometry rich) a very rapid temperature increase can be achieved over the catalyst. The overall performance of the system is enhanced by a series of zeolite adsorbents upstream from the catalyst, which delay inhibiting species such as hydrocarbons and water from reaching the catalyst during the cold-start portion of an engine test cycle. This work demonstrates that combining a catalyst containing such low temperature activity sites with other components in a standard automotive three-way catalyst provides an emission system with the potential to reach very stringent emission standards. References
[1] J.C. Summers, J.F. Skowron, M.J. Miller, SAE Paper 930386, 1993. [2] N.R. Collins, G.R. Chandler, R.J. Brisley, P.J. Andersen, P.J. Shady, S.A. Roth, SAE Paper 960799, 1996. [3] B. Pfalzgraf, M. Rieger, G. Ottowitz, SAE Paper 960261, 1996. [4] J.E. Kubsh, SAE Paper 941996, 1994. [5] K.P. Reddy, S.T. Gulati, D.W. Lambert, P.S. Schmidt, D.S. Weiss, SAE Paper 940782, 1994. [6] H. Mizuno, F. Abe, S. Hashimoto, T. Kondo, SAE Paper 940466, 1994. [7] P.F. Kuper, W. Maus, H. Swars, R. Bruck, F.-W. Kaiser, SAE Paper 940465, 1994. [8] T. Ma, N. Collings, T. Hands, SAE Paper 920400, 1992. [9] T. Tsoi-Hei Ma, UK patent application GB 2280128 A, 1995.

? nig, G. Eigenberger, in: N. [10] T. Kirchner, A. Donnerstag, A. Ko Cruse, A. Frennet, J.M. Bastin (Eds.), 4th International Congress on Catalysis and Automotive Pollution Control, 1997, p. 39, preprints. [11] B.H. Engler, D. Lindner, E.S. Lox, K. Ostgathe, A. SchaferSindlinger, W. Muller, SAE Paper 930738, 1993. [12] J.L. Williams, M.D. Patil, W. Hertl, SAE Paper 960343, 1996. [13] M.J. Heimrich, L.R. Smith, J. Kitowski, SAE Paper 920847, 1992. [14] M.D. Patil, W. Hertl, J.L. Williams, J.N. Nagel, SAE Paper 960348, 1996. [15] R.J. Brisley, N.R. Collins, D. Law, European Patent Application 95308884, 1995. [16] J.K. Hochmuth, P.L. Burk, C. Tolentino, J.J. Mignano, SAE Paper 930739, 1993. [17] P.L. Burk, J.K. Hochmuth, D.R. Anderson, S. Sung, S.J. Tauster, C.O. Tolentino, J. Rogalo, G. Miles, M. Niejako, A. Punke, U. Dahle, SAE Paper 950410, 1995. [18] S. Birch, Automotive Eng. 104 (1996) 50. [19] S.E. Golunski, H.A. Hatcher, R.R. Rajaram, C.J. Bennett, T.J. Truex, SAE Paper 950408, 1995. [20] M. Haruta, H. Kageyama, N. Kamijo, T. Kobayashi, F. Delannay, in: T. Inui (Ed.), Successful Design of Catalysts, Elsevier, Amsterdam, 1988, p. 33. [21] R.R. Rajaram, G.P. Ansell, H.A. Hatcher, European Patent Application 5,480,854, 1993. [22] J.C. Frost, S.C. Bennett, D.S. Lafyatis, A.P. Walker, World Patent Application WO 96/39576, 1996. [23] J.C. Frost, S.C. Bennett, D.S. Lafyatis, A.P. Walker, European Patent Application EP 0747581A1, 1996. [24] Y.F. Yu Yao, J. Catal. 87 (1984) 152. [25] T.H. Ballinger, W.A. Manning, D.S. Lafyatis, SAE Paper 970741, 1997.


相关文章:
电器工程与自动化 专业英语词汇
ambient temperature 环境温度 ambient temperature range 环境温度范围 ambient ...control (AVC) 自动音量控制 automatic welding 自动焊接 automobile lighting 汽车...
汽车专业英语(A篇)
Temperature Air Cleaner Temperature Sensor Override ...Ambient Absolute Pressure Aspirator Air System ...Cutoff Air Fuel Ratio Control Valve Automobile ...
故障对照表(f文件)
control_cabinet_temperature off 控制柜冷却温度故障 control_cabinet_temperature .... ambient temperature error_ambient_temperature_global off 环境温度故障 ambient...
汽车专业英语 缩写大全
Temperature 空气滤清器温度传感器超压 Sensor Override...Ambient Absolute Pressure Aspirator Air System Auto...Cutoff Air Fuel Ratio Control Valve Automobile ...
基本汽车知识英语说明
automobile control system 汽车综合控制系统, camshaft...( h light dependent resistor 光敏电阻 1 g) p)... `% AAT ambient air temperature 环境温度 ABS ...
新视野试卷
________ of the light is sufficiently great....temperature 10. It is well-known that the ...The room is dark; Mr. Smith ________ to ...
发动机术语
Temperature Sensor Override Valve 空气滤清器温度...Ambient Absolute Pressure 外界绝对压力 AAS Aspirator...Emission Control 汽车排气控制 AECD Auxiliary ...
电气专业词汇
电气专业词汇集散控制系统——Distributed Control ...ambient temperature 环境温度 ambient temperature ...automobile lighting 汽车照明 auxiliary discharge 辅助...
汽车维修技术缩略语词典
Temperature Sensor Override Valve A/CL-VCD A/CL...Cutoff Air Fuel Ratio Control Valve Automobile ...Ambient American Motors Corporation Air Management ...
汽车相关常用英语缩写词典
Ambient Absolute Pressure Aspirator Air System Auto...Control Valve Automobile Fuel Economy Air-Fuel ...Temperature Control After Top Center Automatic ...
更多相关标签: