Engine bench test

Figure 51. Conversion of CO, HC and NO., over a three way catalyst as a function of the exhaust gas lambda value (monolith catalyst with 62cellscm", three-way formulation with Pt 1.42gl , Rh 0.28gl ), fresh engine bench test A/F scan at 623 K exhaust gas temperature space velocity 60000NIC h dynamic conditions frequency 1 Hz amplitude 0.5 A/F).

Figure 60. Influence of space velocity on the conversion of CO, HC and NO.v, reached over a three-way catalyst in the fresh state and after engine aging, at fixed exhaust gas temperature and exhaust gas composition (monolith catalyst with 62 cells cm , three-way formulation with Pt 1.42gl->, Rh 0.28gr engine bench test at 723 K exhaust gas temperature exhaust gas composition lambda 0.995 dynamic frequency 1 Hz amplitude 1 A/F engine bench aging during 200 h). Reprinted with permission from ref. [76], ...

Figure 66. Influence of the washcoat loading of a ceramic monolith on the conversion of NO t (monolith catalyst with 62cells cm" three-way formulation with Reprinted with permission from ref. [34], 1991 Society of Automotive Engineers, Inc. Pt 1.42gl" , Rh 0.28gl" after aging on an engine bench 20 h engine bench test space velocity 60000N1E h exhaust gas temperature 723 K exhaust gas composition lambda 0.999 dynamic frequency 1 Hz amplitude 1 A/F). Reprinted with permission from ref [34], 1991 Society of Automotive Engineers, Inc.

Bartolomeo, E.D. and GriUi, M.L., YSZ-based electroehemical sensors From materials preparation to testing in the exhausts of an engine bench test, J. Eur. Ceram. Soc. 25 (2005) 2959-2964. 

The objective of this study is to extend this methodology to the case of a commercial three-way catalyst deposited on a ceramic monolith and to follow the evolution of the ceria surface area at different stages of its life-time, including long time engine-bench testing. In this case, it is important to examine the impact of poisons on TPR curves. 

The analysis and distribution of the poisons were done by SEM coupled with an EDX analysis. A macroporous layer of pollutants was evidenced on the surface. The analysis was done on the elements of the support (Al, Ba and Ce) as well as the poisons usually found after such a treatment, i.e. P, Ca and Zn [11]. Sulfur was searched for but was not detected. In the front side of the converter, the poisons were the only elements detected, with almost 50% Zn, 40% P and 10% Ca. It means that the poison layer is thicker than that analysed by EDX, i.e. about Ifim. The Zn concentration decreased quickly in a few miUimeters axially and then was not detected (<1%), whilst P and Ca were always found in high proportion. However their concentration decreased also continuously along the axis of the converter, from 20 to 7% for Ca, and from 40 to 30% for phosphorus. It results that, after an engine bench test, the washcoat is covered by a thick layer of poisons which becomes progressively thinner when arriving to the outlet. 

Even if the engine bench test was considered as not very severe, with a catalyst temperature of about 1130 K, the BET areas and the reduction extents taken as a criterion of the aging, seem to indicate that the EB catalyst was submitted locally to temperatures at least higher than 1273 K. Indeed, the BET surface area was found lower than for the catalyst aged at 1273 K, 66-60 instead of 77-72 m g-i. However the calculated ceria surface remained rather high, 22-23 m gi for the upstream and the downstream face. Since we observed that stabilization under reactants leads to lower ceria surface areas, this less severe... 

Engine bench test was found equivalent to an hydrothermal aging at 1273 K, with however a better resistance of ceria to sintering. The surface of these engine bench aged samples is covered by a more or less compact layer of pollutants. Its thickness decreases from the inlet to the outlet of the converter. However, the presence of these poisons does not modify the accessible ceria surface area measured by the TPR method. 

From these results a laboratory test has been defined enabling the simulation of engine bench tests. 

Nevertheless, the light-off temperatines determined on laboratory tests are inferior by about 100°C to those detennined on engine bench tests or on vehicle under similar conditions [6]. This difference could only be the consequence of major phenomena that have not yet been identified. It was thought tliat, under real conditions, there could be mixture effects due to interactions between the catalyst active sites and hydrocarbons belonging to different families, in the presence of CO, NO, O2, CO2, H2O and SO2. Thus, a study was undertaken to determine the influence of the nature of various HC on the oxidation reactions of CO and HC by O2 and NO. Special emphasis was directed toward HC able to strongly coordinate to tlie catalyst siufaces. 

On an engine bench test perfonned with a catalyst of the same composition, under the same space velocity, the light-off temperatures for CO, NO and HC are identical and very high 305°C (Fig. 7-b). This confinns that neither propane nor propene is adequate to model the properties of the real hydrocarbon mixture. 

Prototype converters can be evaluated based on the moderate cost engine bench tests. A rating criterion, which is commonly used, is the light-off temperature, i.e., the temperature in which a certain conversion is achieved, typically 50 percent. These kind of engine bench tests have been done for four converters. The light-off temperatures of the selected converters are shown in Table 3. 

Simplihed inlet stream dynamics has also been studied by simulation. Firstly, stepwise change from ambient temperature (298 K) to 650 K has been assumed while the other inlet stream variables are kept constant. Secondly, stepwise change to 6(X) K temperature, 60 s after stepwise change from 298 K to 440 K in the inlet gas temperature, is done. The above stream temperature dynamics mimics the main features of the temperature dynamics of the NEDC vehicle test. Finally, the inlet stream dynamics of the NECD test is modified in such a way that the concentrations are kept constant and approximately same as in the engine bench tests. The inlet stream temperature and flow rate are varying similarly as in the NEDC vehicle test. In table 4 the rating of the converters is shown based on these simulations. 

---