Optimizing Purity And Recovery Of Hydrogen From Syngas By Equalized PSA Using Palm Kernel Shell Activated Carbon Adsorbent

• Main Author: Gobi, Shanthini
• Format: Monograph
• Language: English
• Published: Universiti Sains Malaysia 2018
• Subjects: T Technology; TP Chemical Technology
• Online Access: http://eprints.usm.my/53938/1/Optimizing%20Purity%20And%20Recovery%20Of%20Hydrogen%20From%20Syngas%20By%20Equalized%20PSA%20Using%20Palm%20Kernel%20Shell%20Activated%20Carbon%20Adsorbent_Shanthini%20Gobi_K4_2018.pdf; http://eprints.usm.my/53938/

This thesis is concerned with the adsorbent preparation, design and operation of Pressure Swing Adsorption (PSA) unit at optimum condition for the effective syngas separation processes. The aim of the study is directed towards optimizing the purity and recovery of hydrogen gas from the syngas. The first detailed study presented in this work deals with the preparation of activated palm kernel shell (ACPKS) and its characterization study. In adsorbent preparation, optimum activation temperature of about 800ºC has been used. Whereas, the characterization study reveals that the specific surface area of the prepared adsorbent is around 697.67m²/g with pore volume and pore size of 0.35m3/g and 2.01nm respectively. The average particle size of the adsorbent sample is recorded as 0.11µm. Scanning Electron Microscope (SEM) and X-Ray Diffraction (XRD) analysis uncovers the spongy structure of the highly amorphous ACPKS. Breakthrough studies at varying adsorption pressure were executed and 3 bar adsorption pressure is chosen for the subsequent optimization study. Based on the software analysis, the optimum operating condition of the PSA unit is predicted at 5 minutes adsorption and blowdown time. A systematic and rigorous methodology are employed towards attaining the similar data as predicted by the software. The PSA study which operated at the optimum operating condition yielded hydrogen purity of up to 99.978% with recovery of 80.014%. The results obtained shows a close agreement with the predicted data given by the software.