Document Type
Dissertation
Date of Award
Fall 10-31-1995
Degree Name
Doctor of Philosophy in Chemical Engineering - (Ph.D.)
Department
Chemical Engineering, Chemistry and Environmental Science
First Advisor
Henry Shaw
Second Advisor
Robert J. Farrauto
Third Advisor
Dana E. Knox
Fourth Advisor
Robert Benedict Barat
Fifth Advisor
S. Mitra
Abstract
The effects of HCl, H2S, tributyl phosphate (TBP), P2O5, and zinc dialkyldithophosphate (ZDP) as catalyst poisons for methane oxidation over palladium catalysts supported on γ-alumina were investigated in this study. The overall reaction rate of methane oxidation over palladium catalysts can be simplified into a pseudo first order reaction under excess oxidation conditions.
The deactivation of the catalyst poisoned by HCl is attributed to the loss of active sites due to volatilization of PdCl2. H2 treatment at 200 °C can reactivate the catalyst poisoned by HCl at temperatures of up to 300 °C, since no volatilization of PdCl2 occurs at this relatively low temperature. It was found that 1.5 % water vapor in air can retard HCl induced catalyst poisoning and prolong catalyst life.
The effects of H2S induced catalyst poisoning are attributed to a pore diffusion deactivation mechanism resulting from the reaction of the support with H2S. Catalyst activity decreased with increasing H2S concentration and temperature. Surface sulfite and sulfate groups were detected by FT-IR on poisoned catalysts and are believed to be the cause of the decrease of 25 % BET surface area. In the poisoning test at 400 °C, the decreases of activation energy suggest that the formation of Al2(SO4)3 which transfers the reaction from surface reaction control to pore diffusion control. H2 treatment at 600 °C removes the sulfite and sulfate from the surface of poisoned catalysts and regenerates most of the fresh catalyst activity. The deactivation of methane oxidation can be described by first order deactivation rate law in presence of 80 ppm H2S in the feed stream.
In the case of TBP induced poisoning, it is believed that the deposition of product P2O5 and other unburned phosphorous compounds cause a decrease of 64% BET surface area and catalyst activity. The catalyst activity decreases with increasing dose of TBP and temperature. The formation of AJPO4 was detected by FT-IR on the aged catalyst. It is believed that a strong impervious film of phosphate blocks access to the pores and is the major deactivation mechanism.
In the case of P2O5 induced poisoning, a glass-like P2O5 film is produced at elevated temperatures that masks the catalyst surface as indicated for the TBP oxidation and causes a decrease of 92 % BET surface area and catalyst activity. The catalyst activity decreases with increasing dose of P2O5 and temperature. The formation of AlPO4 is also detected by FT-IR indicating a similarity between the poisoning mechanism due to TBP oxidation and P2O5 addition. Thermal gravimetric analysis (TGA) shows that P2O5 can be partially removed by heat treatment in the range of 500 to 820 °C, but AlPO4 is formed at these temperatures.
Studies of ZDP induced poisoning, treatment at 500 °C and 820 °C results in the deposition of a glaze containing a ZDP combustion product, (β-Zn2P2O7, which was identified by x-ray diffraction crystallography. The glaze masks the catalyst surface causing a decrease of BET surface area as well as catalyst deactivation. The catalyst activity decreases with increasing dose of ZDP and temperature. Heat treatment in air at 650 °C and H2 treatment at 600 °C can slightly improve catalyst activity. TGA and FT-IR measurements show that (β-Zn2P2O7 is difficult to remove from the aged catalyst. However, H2 treatment at 800 °C removes the Zn2P2O7 from the catalyst surface, but the regenerated surface has poor catalytic activity. This may be due to palladium sintering at the severe reducing atmosphere used for regeneration.
Recommended Citation
Yu, Tai-Chiang, "The effect of chlorine, sulfur and phosphorus compounds on deactivation of palladium catalysts" (1995). Dissertations. 1128.
https://digitalcommons.njit.edu/dissertations/1128
