Evaluation of Modified Kent-Eisenberg Model to Predict Carbon Dioxide Solubility in Methyldiethanolamine (MDEA) Under Extended Operating Conditions

The aim of this study is to evaluate the reliability of Modified Kent-Eisenberg model
for predicting the carbon dioxide (CO2) solubility in methyldiethanolamine (MDEA)
under wider range of carbon dioxide partial pressure. Model from Haji-Sulaiman et.
al. (1998) is used as the reference in this work. The model is re-evaluated by rederiving
the equations used in the paper. The re-derived equation is verified by using
the equation taken from Pahlavanzadeh et. al. (2010). Data from Haji-Sulaiman et. al
(1998) is fitted into the model to validate the model. The modified model is then
applied in MATLAB to obtain the solubility data of carbon dioxide in 3.04-4.28
kmol/m3 aqueous MDEA solutions at temperature ranging from 313K to 373K and
CO2 partial pressures ranging from 0.876 to 1013 kPa, with reference to Xu et. al.
(1998). The maximum error obtained from the model is 16.585% which is at 4.28M
MDEA. The parameters in non-ideality equation are re-regressed and the new values
for the parameters g and k generated are -0.0458 and 0.6772 respectively. The new
model which use the new parameters are fitted once again to the experimental data
and the results shows decrease in errors with maximum error of 14.739% at 3.46M
MDEA. Data from Jou et. al. (1982) which range from 2-4M aqueous MDEA
solution, temperature from 298K to 393K, and CO2 partial pressure from 0 to 6630
kPa are fitted to both models to further test their performance. The results once again
show that the new model performs better in predicting the CO2 solubility in aqueous
MDEA solution.