OLI software provides multiple modes to address chemical reactions between impurities beyond traditional thermodynamic equilibria, particularly in complex systems such as dense-phase CO2 with impurities. Some of the reactions between impurities in CO2 systems involve redox, characterized by changes in the oxidation states of elements such as sulfur or nitrogen. In contrast, other reactions among impurities do not involve any change in oxidation states and are instead governed by acid-base equilibria, complexation, or physical dissolution mechanisms. While thermodynamic equilibria (e.g., non-redox) and some of the redox reactions are automatically handled through oxidation state balancing and electrochemical equilibrium in the software, the users also have the flexibility to implement specific reaction kinetics as desired (typically for redox reaction), especially when system is not at equilibrium. This allows greater control over reaction pathways and enables better representation of impurity behavior under certain dynamic conditions. When a reaction is kinetically controlled, the extent of reaction needs to be included in the material balance equations, and the speciation can be modeled dynamically over time. Depending on the system and modeling objectives, users can select the most suitable approach (thermodynamic non-redox, redox, or kinetic) to capture the phase and chemistry of their systems. Table 1 compares key features of the two modeling approaches available in OLI software for impurity reactions: manual kinetic implementation, where users define specific rate laws and reaction mechanisms, versus redox equilibria, which relies on thermodynamic solvers to predict species distribution and mass balance based on oxidation states and equilibrium assumptions.
This blog is intended to guide users in choosing the appropriate mode in OLI software, using a few use cases. It briefly highlights how OLI’s solutions can support the implementation and interpretation of impurity reactions in different CO2 streams, depending on the user’s specific goals and simulation needs.
Table 1. Comparison of redox and kinetic reaction modes in OLI software (Kinetic vs. Redox)
The composition of impurities in a CO2 stream varies based on the type of capture technology employed and the specific industry sector, whether related to capture, transport, injection, utilization, storage. Table 2 provides some possible examples of selected reactions, including both redox ones and established non-redox equilibria among common impurities that contribute to the formation of liquid or solid acid dropout as well as elemental sulfur or other products. Many more reactions may exist that are not listed in this table. Phase equilibrium reactions such as VLE, SLE, and LLE are auto-generated separately by the software solver to support phase quantification calculations for different chemistries and are not listed in the table.
Table 2. Some of the possible reactions in CCTUS with impurities (Redox vs. Non-Redox)
Redox equilibria mode: An example of the redox-equilibrium mode within a model that incorporates nitrogen and sulfur species in the oxidation states of N(+2), N(+4), N(+5), S(−2), S(0), S(+4), and S(+6), with specific monitoring of S(+6) species (e.g., sulfuric acid), is shown in Figure 1. Figure 1 illustrates predictions from the OLI’s Mixed-Solvent Electrolyte (MSE) model showing how temperature and the amount of NO2 affect the composition of the liquid dropout. The water dropout, shown in green, and the total concentration of S(+6)-bearing species, shown in orange. When NO2 levels are below 10 ppm, the amount of liquid acid remains steady regardless of the temperature changes. However, as NO2 increases from 1 to 5 ppm, the liquid acid concentration increases from 32 to 36 ppm and reaches 41 ppm at 10 ppm NO2. Beyond this point, higher NO2 concentrations cause precipitation of corrosive S(+6) compounds such as liquid acid or sulfuric acid monohydrate (not shown). Figure 1 also shows that water condensation occurs alongside the formation of these sulfur species at all NO2 levels, which can increase the risk of acid corrosion in CO2 transport and storage systems. At NO2 concentrations above 25 ppm, the model predicts little change in the amount of water and sulfur-containing dropouts at different temperatures. It is important to note that these results are based purely on thermodynamic equilibrium calculations (redox + non-redox). They do not consider possible kinetic effects from rate-controlling reaction steps, which can influence the speed and extent to which equilibrium is reached. Under such conditions, the user should switch to the kinetic mode, an alternative modeling approach available in OLI software, disable the redox mode, and implement their own reactions and corresponding rate expressions as needed. The kinetic mode will be discussed in the following case.
Figure 1. Redox-equilibria mode: Predicted NO2 limits for dropout of water (dark green) and acid or S(6+) species (orange). Condition: 25 °C, 100 bar. H2O: 90 ppm, O2: 70 ppm, H2S: 36 ppm, SO2: 20 ppm.
Kinetic mode: speciation analysis over time: There is strong evidence from experiments, laboratory studies, and field data, particularly from IFE in Norway, confirming oxidation pathways through consecutive oxidations of lower oxidation state sulfur species, such as H2S to elemental sulfur, then to S(+4), and finally to S(+6), through reactions of SO2 with oxidizing agents such as NO2 and O2 in coalesced aqueous phases dropped out of the dense-phase CO2 liquid. Such rate-controlling reactions can significantly govern the phase precipitation incidents and their chemistry. Under such conditions, the user can construct and implement custom rate equations tailored to specific reaction mechanisms based on their experimental observations. Three key reactions may be considered to explain or support the observations: the oxidation of H2S to elemental sulfur via its reaction with O2, as well as two oxidation pathways that convert S(4+) species such as SO2 to S(+6) through reactions with either O2 or NO2 to produce SO3. Upon SO3 hydration, it can form aqueous sulfuric acid, which has been commonly observed in practice across real-world CO2 assets. Sensitivity analyses were performed to identify the appropriate sensitivity domain for the kinetic variables of these three reactions and to better align the predictions with field data, literature, and experimental investigations (interpretation not shown). Each reaction was subsequently assigned a user-defined rate equation with the following form within the OLI Flowsheet software:
where kf and kr represent forward and reverse rate constants, and the products ar and ap denote activities of reactants and products raised to their respective reaction orders (ri and pi, respectively). This approach enabled dynamic prediction of species distribution across all phases, including liquid and vapor CO2, precipitated solids, and any condensed aqueous phases. Figure 2 shows the evolution of impurity concentrations and dominant species in both the liquid CO2 and the aqueous acidic phase that drops out of the liquid CO2 over time under pipeline transport conditions, as predicted using the MSE framework.
Figure 2. Kinetic mode: Predicted concentration of impurities in (a) acid dropout and (b) liquid CO2 over time. Condition: 25 °C, 100 bar. H2O: 10 ppm, O2: 140 ppm, H2S: 65 ppm, SO2: 35 ppm, NO2: 48 ppm.
Depending on their systems, users can select from different modes within the OLI software to model reactive interactions between species, with the ability to optimize kinetic variables for improved model alignment with laboratory or field data. To learn more and share ideas, visit OLI’s contact page.