The Texas Carbon Management Program focuses on carbon capture from coal and natural gas power plant flue gas to help mitigate anthropogenic greenhouse gas emissions. We investigate post-combustion amine scrubbing, which is a flexible, tail-end technology that can be retrofitted onto existing power plants or be included as part of a greenfield installation. In the amine scrubbing process, CO2 is absorbed into an aqueous amine solvent at low temperature in an absorber, and then the dissolved CO2 and amine solution is sent to a stripper where it is heated to high temperatures causing the CO2 to desorb cleanly. The CO2 is then compressed for geological sequestration or used for enhanced oil recovery.
Our goal is to understand and improve all aspects of amine scrubbing. Thirteen graduate students are collecting thermodynamic and rate measurements, testing amine degradation, mitigating nitrosamines, quantifying aerosol formation, creating process models, improving process design and efficiency, and understanding pilot plant results. These efforts have established aqueous piperazine (PZ), used with the advanced flash stripper, as the most efficient, open-literature amine scrubbing system.
The lab has a wide variety of analytical equipment used to study alternative solvents. Any solvent must be thermally and oxidatively stable, possess a high absorption rate and capacity, and be minimally volatile.
Amine solvents are heated to high temperatures and sampled over a period of weeks to check for thermal stability. Degradation products are analyzed using in-house cation and anion chromatography. PZ is more stable than monoethanolamine (MEA), the previous industry standard.
The oxidative stability is checked by cycling the solvent from low to high temperature while sparging with air. This simulates the absorber and stripper conditions.
If a solvent is found to be stable, its rate and capacity are quantified using a wetted wall column. This experiment provides both the rate of CO2 absorption and the CO2 solubility from 20 to 100 °C. These data allow for the rapid screening of solvents.
Other studies include: using NMR to determine the amount of free amine and products in loaded solvents, using FTIR to determine amine volatility, and using the wetted wall column with NO2 to study nitrosamine formation kinetics.
The experimental data collected is regressed into a rigorous Aspen Plus® process model that can predict plant performance, design new process configurations, and provide understanding of limiting mechanisms.
Rigorous models have been created for MEA, PZ, PZ/monodiethanolamine (MDEA), and PZ/2-amino-2-methyl-1-propanol (AMP). These models have shown that a higher heat of absorption results in a lower cost of capture, an idea that is counterintuitive to minimizing energy performance. Absorber intercooling has been advanced by understanding the temperature bulge that results from the exothermic absorption of CO2. The advanced flash stripper was simulated and showed better performance than other stripper configurations before it was tested in a pilot plant campaign. Economic analysis shows that using PZ with the advanced flash stripper would require 25% less capital than the existing Econamine FG PlusSM process.
We have also developed a dynamic model of the amine scrubbing process. Dynamic models are necessary to understand the effects of disturbances on process operation and for controller design. We are currently evaluating different process control strategies that will enable stable and optimal operation while achieving process objectives such as maintaining a removal set point in the absorber.
Pilot Plant Studies
Located on the Pickle Research Campus (PRC), our 0.1 MWe pilot plant provides data to validate our models and explore other phenomena, such as aerosol emissions and corrosion. The pilot plant started with a simple stripper and absorber running MEA, and has investigated a number of solvent compositions. The equipment has been modified with each test to study novel absorber and stripper configurations. The latest test used 5 m PZ with an intercooled absorber and the advanced flash stripper. This test was the first to incorporate aerosol generation using H2SO4 or SO2 injection and demonstrated the use of a phase doppler interferometer to measure drop size distribution
Also located at PRC is another 16.8-inch diameter column used to characterize kl, kg, and effective area of random, structured, and hybrid packing. Previously, packing characterization has used dilute aqueous solutions. Studies are under way to measure these properties as a function of viscosity in aqueous glycerol.