A public database for amine degradation data was established during the LAUNCH project. This is a unique platform for sharing and searching for data related to the stability and degradation of amine solvents for CO₂ capture and facilitates knowledge sharing and application of real data in models and simulations of the CO₂ capture process. Data from a range of lab and pilot scale degradation tests are made available and searchable in this database, which is currently administered by SINTEF Industry. The database currently contains 45 different datasets, which are all connected to published literature. In the database, a collection of data, metadata and additional information about setups and experiments can be found and shared. This memo introduces how the database works and how it can be used.
WP1 - Predicting Degradation
The main objective of WP1 is to develop methodologies and models to predict solvent degradation behaviour. This will enable an estimation of the extent of solvent degradation as well as give an input to the design and development of countermeasures tailored to specific solvent/flue gas combinations.
Results will also provide a tool for screening potential solvents. This publicly available resource will be of great value for solvent developers when designing new formulations.
This fundamental work will be based on a multi-layer modelling approach, from data mining to complex chemistry connections.
WP1 is divided into 3 key tasks:
Task 1.1 Solvent degradation database
Task 1.2 Predicting degradation based on process data: Big Data modelling
Task 1.3 Optimised design of CO2 capture plants: Process modelling
Coordinator profile
Coordinator Name: Andreas Grimstvedt
Coordinator Job Role: Research Scientist, SINTEF
Andreas Grimstvedt (PhD) is a Research Scientist working in the Chemical and Environmental Process Engineering group, Department of Process Technology at SINTEF Industry. He has been working in the field of CCS since 2008, in particular in the area of identification and quantification of degradation compounds in solvent-based CO2 capture processes. His main fields of interest include statistical methods for quality management and improvements, chemometrics and multivariate analysis of experimental data, and he has been actively working with solvent characterisation, solvent development, chemical analysis and thermodynamics. Holding project manager or work package leader roles in several projects, Andreas is also an author and co-author of numerous journal papers and conference presentations.
Results
D1.2.1 Big Data tools for identifying key degradation predictors
In this deliverable, we have explored the use of Big Data tools for improving the knowledge on solvent degradation, and particularly for identifying key degradation predictors. Initially, process and solvent analysis data from the ALIGN-CCUS RWE campaign with MEA were used. However, and despite working with data from the longest ever open MEA campaign, not enough degradation data was available. The attempts to analyse the process data did not lead to any new insights. There is a clear need to generate larger data sets on solvent degradation. However, doing that by solvent sampling and laboratory analysis would become prohibitively expensive.
Download document (1.48 MB) D1.3.1 and D4.2.1 Assessing the representativeness of accelerated degradation tests using the LAUNCH rigs and the DNM
Different accelerated degradation techniques were tested, and the results were compared between TERC pilot plant (1000 kg CO₂/day) and the LAUNCH rig (25 kg CO₂/day). Four accelerating degradation techniques were tested: increased oxygen levels in the flue gas, increased solvent concentration, increased stripping temperature, and addition of NOx. Increasing the MEA content in the solvent, the stripping temperature and the addition of NOx was studied in combination with increased O₂ content in the gas.
Download document (1.35 MB) D1.3.2 Generalizing and Validating the DNM-LAUNCH rigs and pilot runs & D1.3.3 Optimised design and operation of capture units to reduce degradation
This report pertains to two sub tasks taken over by the University of Sheffield on the request of Doosan.
• Subtask 1.3.2 verifying the representativeness of accelerated degradation tests
• Subtask 1.3.3 Utilise the degradation network model to optimise design and operation of
capture units to reduce degradation
The oxidative degradation model developed by TNO as part of WP1 was used in conjunction with a thermal degradation model published by Braakhuis et al, 2022. A tool was developed that simulates the long-term operation of a CO₂ capture plant. Both oxidative and thermal degradation contributions are reported for each plant component as well as the total predicted solvent consumption rate resulting from degradation and the resulting variation in predicted impurity levels with time. The output from this tool was compared to results from the LAUNCH rigs to estimate a ratio of MEA molar consumption to non- volatile degradation compounds as 1:0.32. This ratio then forms an input in the continued use of the tool.
Download document (2.19 MB) D1.3.4 Application: Degradation Network Model: a tool for predicting solvent degradation in industrial CO₂ capture applications
In this deliverable, the Degradation Network Model (DNM) is presented. The model can be applied to any solvent, as it relies solely on oxygen consumption data. The oxygen consumption data used in the current version of the DNM is generated with fresh solvent, in the absence of metals and degradation products. In order to obtain representative values, the stoichiometric factor is treated as a fitting parameter, and used to anchor the DNM results to those observed in the RWE pilot. The DNM methodology is thoroughly discussed in the current deliverable, and the model can be used to predict solvent losses for MEA and CESAR1 at different operational conditions. To improve the model prediction, more data is needed, as extensively discussed throughout this report.
Download document (1.23 MB) More About the Project
Objectives
The primary objective of the LAUNCH project is to accelerate the implementation of CO2 capture across the energy and industry sectors by developing novel solvents and establishing a fast-track, cost-effective de-risking mechanism to predict and control degradation of capture solvents.
Outcomes
The LAUNCH project will deliver the necessary knowledge and tools to allow CO2 capture plants to operate in a more controlled and cost-efficient way. The project will also provide solvent developers with the tools to assist in designing and validating novel solvents. By avoiding uncontrolled build-up of degradation products, LAUNCH will improve the performance and economics of CO2 capture.
