A COMPARISON STUDY OF DIFFERENT EMPIRICAL MODELLING METHODS IN PREDICTING CO2 SOLUBILITY IN DIETHANOLAMINE, N-METHYLDIETHANOLAMINE AND THEIR MIXTURES

• Main Author: MOHD FADZLI, MOHD FAREEZ AKMAL
• Format: Final Year Project
• Language: English
• Published: Universiti Teknologi PETRONAS 2013
• Subjects: TP Chemical technology
• Online Access: http://utpedia.utp.edu.my/8415/1/Dissertation%20Mohd%20Fareez%20Akmal%20Bin%20Mohd%20Fadzli%2012677.pdf; http://utpedia.utp.edu.my/8415/

Acid gas removal from natural gas, synthesis gas and refinery gas stream is very important in plant industry to prevent corrosion in the subsequent piping and as per requirement by various organizations and companies. Because of the corrosiveness of H2S and CO2 the sales gas is required to be sweetened to contain no more than a quarter grain H2S per 100 standard cubic feet ( 4 parts per million) and to have a heating value of no less than 920 to 980 Btu/SCF, depending on the contract. The most widely used process to remove acid gas from natural gas is by using alkanlomines, and of the alkanolamines the most common are nmethyldiethanolamine (MDEA) and diethanolamine (DEA). In this study, data from Khalid Osman et al (2012), A. Benamor et al (2005) and Zhang et al (2002) will be used to simulate the solubility of CO2 in DEA and MDEA mixtures using Multiple Linear Regression (MLR) and Artificial Neural Network (ANN) and the performance will be compared to show which model is better for CO2 absorption. Furthermore, data from Jou et al (1982) and Lee et al (1972) will be used to study the solubility of CO2 in pure DEA and MDEA aqueous solution and simulation of the models will be compared between the models and the reference research works mentioned earlier. MLR has proved it cannot be used to predict CO2 for pure DEA, MDEA and their mixtures. The results clearly shown that the model is pressure dependent as it has large coefficient compared to other parameters which is very small and becomes dominant in the equation thus neglecting them in predicting the CO2 loading data. ANN proved the model can be used to predict CO2 solubility in the alkanolamines and their mixtures. Developed model for DEA and MDEA mixture has an absolute relative deviation δAAD 10.47 % while for data from Khalid Osman et al (2012), A. Benamor et al (2005) and Zhang et al (2002) are 17.06%, 12.09% and 9.82% respectively. In pure alkanolamines prediction, ANN model of CO2 solubility predicted in pure DEA has δAAD 4.02% while from the experimental data of A. Benamor et al (2005) has absolute relative deviation of 4.72%. As for prediction of CO2 in pure MDEA, the model resulted δAAD of 9.77% compared to the reference paper from A. Benamor et al (2005) with 10.76%.