Aging of zeolite based automotive hydrocarbon traps with applications to diagnostics
Presented is an experimental and theoretical study of the aging of zeolite-based automotive hydrocarbon traps with applications to diagnostics. The central hypothesis is that the aging of zeolite-based hydrocarbon traps changes the transient response to engine cold-starts. Furthermore, these changes are predictable and can be used as the basis of a diagnostic algorithm. A survey of the work on hydrocarbon traps and also adsorption theory suggests that water may play a central role. An experimental cold-start study also suggests water might be an important factor. The cold-start experiments demonstrate a correlation between aging and the transient thermal behavior. Thus the combination of the literature survey and cold-start experiments leads to a causal chain leading from aging to changes in the zeolite to changes in adsorption behavior to changes in the transient response. ^ To prove the hypothesized chain, additional measurements and analysis are performed. Water adsorption measurements are used to understand the impact of aging on water uptake. As the zeolite ages, the water adsorption decreases. Furthermore, as the zeolite ages the cold-start HC performance declines. These experimental results provide a missing piece in the research literature between fundamental zeolite studies and applied engine research. ^ While experimental studies provide the link between aging and adsorption, the link between adsorption and transient behavior is provided primarily by mathematical modelling. Statistical analysis of the cold-start data indicates aging and thermal response are strongly related. A mathematical model of the system is developed, and the model provides a detailed understanding of the system dynamics. The system response is interpreted in terms of propagating transient fronts. A simulated sensitivity study leads to selection of several potential diagnostic metrics. Noise factors are discussed along with methods to mitigate their impact. ^
Matthew A. Franchek, Purdue University, Peter Meckl, Purdue University.
Off-Campus Purdue Users:
To access this dissertation, please log in to our