غیر فعال سازی کاتالیزور های سه راهه با تشکیل CePO4 Deactivation of real three way catalysts by CePO4 formation

۱٫ Introduction Three way catalysts (TWC) have become the widely adopted solution of the car industry to fulfil legislation concerning the emission of more toxic components of the gas exhaust pipe of engine car. The installation of TWC in a catalyst cartridge at the exit of the spark ignition engine considerably reduces the concentration of NOx , unburned hydrocarbon and CO [1–۳]. Although TWC deteriorates under the tough conditions to which they are subjected (high temperatures and deposition of poisons) they have demonstrated their robustness to keep their efficiency after long time under running conditions (high mileage). However, the deactivation mechanisms is becoming a vivid topic of research [4–۱۰] due to the fact that the legislation are becoming more stringent [11]. Consequently, the durability of the catalyst must be expanded if the increasingly low level of emissions must be fulfilled during the lifetime of the catalysts (more than 100 000 km). It is obvious that the understanding of the deactivation mechanisms is one of the key aspects in the achievement of more durable catalyst cartridges. The deactivation by thermal effects are rather well known: sintering of active components results in a loss of specific surface what it is clearly detrimental for the activity of the catalysts [7,8,12–۱۴]. However, the role of chemical poisoning is less understood. It is well known that P, Ca, S, Pb and Zn are the most concentrated contaminants (P, Ca and Zn arising from lubricants and Pb and S from fuel) [5,6,9,15–۲۰]. These studies must be updated and the information revisited because the nature and composition of the catalysts and the fuel are rapidly changing [11]. Moreover the phases that the contamination and the components of the catalyst built and the effect that these new phases have in the multiple functions the catalyst must operate are not well studied. This work is a contribution to the understanding of the role that P contamination has in the poisoning of the TWC. Real used catalyst extracted from the exhaust pipe of a gasoline car kindly supplied by Ford Spain with ca. 30 000 km was studied and characterised by several techniques especially fitted to this purpose. A fresh catalyst supplied also by Ford Spain with 0 km was used as a blank when required. ۲٫ Experimental ۲٫۱٫ Sample preparation The catalytic container of a gasoline car usually carries two monoliths. The specimens were taken from the very beginning (first millimetres) of the upstream monolith of a Ford Focus 2.0 (1999 model) catalytic converter with 29 900 km. The samples studied in this work were small pieces separated from the monolith by breaking under gently pressure the honeycomb structure with a tweezers. The further treatment of the sample is critical for the characterisation of the sample: an optimal physical shape (pieces of monolith wall or powder obtained by grinding those pieces) must be selected according to the characterisation technique used and with the intention of improving the detection and stressing the chemical features that contamination brings about in the catalysts. Therefore, the physical shape will be clearly stated when describing the equipment for characterisation. ۲٫۲٫ Characterisation techniques TXRF analysis was performed in a Seifert EXTRAII spectrometer (Rich Seifert & Co., Ahrensburg, Germany), equipped with two X-ray fine focus lines, Mo and W anodes, and a Si(Li) detector with an active area of 80 mm2 and a resolution of 157 eV at 5.9 keV (Mn K). To carry out the TXRF analysis X-ray tungsten source was used for P determination. The radiation was filtered with a Cu film of 10
m thickness, in order to optimise the energy range (0–۱۰ keV) used in the analysis. The X-ray molybdenum source was used for the analysis of the rest of elements, previously filtered with a Mo film of 50
m. The exciting conditions used were a potential difference of 50 kV and a variable intensity—between 5 and 25 mA—to yield a count rate of about 5000 cps in the spectra acquired with the tungsten source and a potential difference of 50 kV and a variable intensity—between 5 and 30 mA—to yield a count rate of about 5000 cps in the spectra acquired with the molybdenum source. The analysed samples were subjected to the preparation process suggested [6,21] for the analysis of solid samples by TXRF [21–۲۳]. First, 10 mg of a sample were ground to a powder of particle size less than 30
m in an agate mortar. The powder was then further grounded for 20 min in a vibrating micro-pulveriser having a ball and a base of agate (Fritsch GmbH, Oberstein, Germany). Subsequently, 1 ml of high-purity water was added to the powder. Next, the mixture was poured into a test tube in which up to 2 ml high-purity water was added. The sample was homogenised for 10 min by ultrasonic desegregation in order to disperse possible agglomeration of particles. The 2
l of the suspension was taken and placed on a flat carrier of plastic where the water was evaporated by vacuum. The errors (S.D.) was achieved with five replicates of the fresh and used samples.