Carbon dioxide solubility was studied in 3.5 m (23.5 wt %) morpholine (MOR), 17.7 m (65 wt %) 2-aminoethoxyethanol (diglycolamine or DGA), and 3.6 m MOR + 14.7 m DGA (11 wt % MOR + 53 wt % DGA). CO 2 solubility was determined by dynamic measurements with a wetted wall contactor. Carbamate and bicarbonate concentrations were determined by 13 C NMR in solutions loaded with 13 CO 2 . The data are represented by the electrolyte NRTL model. At a given CO 2 loading (mol/mol amine), the CO 2 vapor pressure over 3.5 m MOR is 10 to 1000 times greater than 17.7 m DGA. In 3.6 m MOR + 14.7 m DGA, the CO 2 vapor pressure is 5 to 7 times greater than in 17.7 m DGA at high CO 2 loading, but the same below 0.2 loading. MOR carbamate is less stable than DGA carbamate by a factor of 7 to 10 from (300 to 333) K. The model predicts that MOR vapor pressure is 100 times greater than DGA over 3.6 m MOR + 14.7 m DGA from (313 to 333) K. The heat of CO 2 absorption in the blend is equivalent to 17.7 m DGA up to 0.35 loading but is 40 % lower at 0.5 loading. The working capacity of the blend is 17 % less than 17.7 m DGA.

The kinetics of ethylene epoxidation network were studied experimentally over a cesium-doped silver catalyst supported on α-Al2O3 pellets in a differential reactor. A variety of rate functions were considered and among these, expressions based on a dual-site Langmuir−Hinselwood mechanism:  r i = k i P E /(1 +   P E)2, fitted the data best with an average error of 13.0 and 10.7% for the epoxidation and the complete combustion reactions, respectively. Under the experimental conditions, both reactants influence both reaction rates; however, product influence is negligible because of their small partial pressures present in the reactor. High oxygen/ethylene ratios and lower temperature favored selectivity to the epoxidation reaction. The kinetic parameters obtained from the differential reactor experiments were refined to match the fixed-bed reactor experiments. With this, the average error in predicting both ethylene conversion and selectivity to ethylene oxide was 4.3%. The apparent activation energies for the epoxidation and combustion reactions were 60.7 and 73.2 kJ/mol, respectively.

No abstract

Absorption of CO 2 into aqueous DGA (Diglycolamine) was performed at 25-60 °C in a wetted-wall column. The absorption data were analyzed using a rigorous model based on eddy diffusivity theory and approximations assuming pseudo-first-order (PFO) and interface pseudo-first-order (IPFO) reactions. The PFO is a good approximation for CO 2 absorption into DGA at a CO 2 loading of less than 0.2 mol/mol DGA. At CO 2 loadings greater than 0.4, instantaneous reactions are approached. The IPFO model matches the rigorous model very well. The second-order rate constant in 65 wt % DGA at 25 °C for the reaction with CO 2 is four times larger than previously published values, but 25 wt % DGA yields a rate constant which is in good agreement with literature values. This finding suggests that the second-order rate constant is probably a function of DGA concentration. The second-order rate constant in 65 wt % DGA increases by a factor of 5 from 0 to 0.4 mol CO 2 /mol DGA. Experiments with 65 wt % DGA + glycolic acid and 65 wt % DGA + potassium formate at 25 °C and 40 °C showed similar trends. The rate constant increases a factor of 2 to three in these solutions, suggesting that the rate constant is a function of ionic strength.

No abstract