(561g) Investigating Effectiveness of Diethylhydroxylamine (DEHA) As Inhibitor for Oxygen-Involved Polymer Fouling Using Automatically Generated Multi-Phase Kinetics Model

Polymer fouling is a ubiquitous problem in steam cracking plants, where undesired polymeric materials accumulate in a processing unit. At a unit scale, polymer foulant can clog pipelines, elevating the pressure drop and increasing the energy requirement. It can also deposit on the surface of a heat exchanger or the trays of a distillation column, reducing the thermal conductivity of the unit and limiting the thermal efficiency. On the plant level, polymer fouling decreases overall throughput, increases downtime required for maintenance, and may sometimes cause an unscheduled shutdown. Certain types of polymer foulant can cause a fire hazard and endanger both property and human life. An in-depth understanding of the polymer fouling mechanism is necessary for developing effective mitigation strategies.

This work investigates the effectiveness of a commercially available polymerization inhibitor, diethylhydroxylamine (DEHA), under different concentrations of dissolved trace oxygen in a typical debutanizer using an automatically generated multi-phase kinetics model. A workflow developed in our previous work for automated generation and fast simulation of multi-phase kinetic models for polymer fouling is applied to generate the model. The workflow is built upon two software packages developed by MIT Green research group, Reaction Mechanism Generator (RMG) and ReactionMechanismSimulator.jl (RMS). The generated model contains two submodels for three phases: the vapor-liquid submodel and the polymer film submodel. The vapor-liquid submodel contains 896 vapor-phase species, 896 liquid-phase species, and 25,421 liquid-phase reactions, which capture the detailed mechanism of oxygen-involved oligomerization between olefins and aromatics in the tray liquid, as well as DEHA’s activation and regeneration cycles. Moreover, the vapor-liquid submodel incorporates the inter-tray vapor/liquid flow and the intra-tray vapor-liquid mass transfer and phase equilibria. The film submodel keeps track of the polymer film growth caused by reactions between the liquid phase components and the film phase reacting sites. This submodel is constructed using fragment-based modeling, where reactive functional groups on the film are considered reduced-representation species since standalone species is not well-defined for the film phase. It contains 896 liquid phase species, 7 film phase fragments, and 23,166 liquid-film reactions. Rate of production analyses and flux diagram analyses were applied to each submodel to study the main chain initiation, termination, and propagation pathways, and the influences from the inter-tray vapor and liquid flows and the intra-tray vapor-liquid mass transfer. The effectiveness of DEHA as a polymer fouling inhibitor is evaluated using the change in polymer growth time constant. Moreover, a grid evaluation for a wide range of concentrations for trace dissolved oxygen and DEHA is used to study the oxygen-sensitivity of DEHA chemistry and investigate the optimal amount of DEHA to reach desired performance. This work showcases the applicability of our previously proposed workflow and paves a preliminary path for fully automated studies for polymer fouling mitigation strategies.