You are here:
Combustion of C1 and C2 PFAS: Kinetic Modeling and Experiments
Citation:
Krug, Jonathan D., P. Lemieux, C. Lee, J. Ryan, P. Kariher, E. Shields, L. Wickersham, M. Denison, K. Davis, D. Swensen, R. Burnette, J. Wendt, AND Bill Linak. Combustion of C1 and C2 PFAS: Kinetic Modeling and Experiments. In Proceedings, 38th International Conference on Incineration and Thermal Treatment Technologies, NA, January 27 - 28, 2021. AWMA, Pittsburgh, PA, NA, (2021).
Impact/Purpose:
The paper will be delivered at the International Conference on Incineration and Thermal Treatment Technologies, Jan 27-28, 2021. The conference was originally going to be held in West Palm Beach, FL, but it has been turned into a virtual conference due to COVID. This is a long-standing international conference sponsored by the Air & Waste Management Association that usually has a few hundred researchers and practitioners of the science of combustion and thermal treatment of waste and associated pollution control technologies. There is a mix of industry, states, consultants, academia, and varying numbers of EPA attendees depending on the relevance of current regulatory activities to the audience. Thermal treatment of PFAS wastes is the major focus of this upcoming conference. Based on the level of interest in the PFAS incineration keynote delivered by OLEM’s Mike Galbraith at last year’s conference, it is expected a high level of interest in seeing this EPA work presented. The paper reflects a preliminary block of work performed on the Rainbow Furnace in the EPA’s RTP facilities, where model PFAS compounds (CF4, C2F6, and CHF3) were fed into the furnace through the flame as well as through several ports downstream of the flame at various temperatures and residence times. Real-time measurements, taken with a Fourier Transform Infrared (FTIR) instrument are reported, and are compared to results from a computer incinerator simulation (the Configured Fireside Simulator, CFS) originally developed by Reaction Engineering International for the HSRP, with PFAS-specific additions funded from both SHC and the PITT. The paper compares predictions from the model with experimental measurements taken as part of the SHC program in EPA’s RTP Combustion Laboratory under the RTP in-house Jacobs contract. The paper’s authors include researchers from CESER and CEMM. At this point experimental results are preliminary, but very exciting. This represents a unique application of what can only be called “the holy grail” of combustion science, where experiments were performed using model compounds in a real combustion system, real-time measurements were performed of the injected compound and trace products of incomplete combustion at operationally relevant concentrations, and the results were successfully compared to model predictions of those same trace gas-phase constituents. This work represents a significant potential enhancement in available tools to support effective management of PFAS-containing waste. The paper was internally reviewed by 2 ORD researchers then was sent to the conference for external peer review by the conference session chair(s). This version of the paper addresses all ORD reviewer comments; comments from the conference are still pending. The paper additionally has been sent to Mike Galbraith of OLEM and Charlene Spells of OAQPS for a review to assess policy implications. As the research progresses and the model is updated, there will be ongoing status updates to ORD and our Program Office partners. The incinerator model was developed under HSRP Project HS19-02.06-1056, and the pilot-scale experimental work was done under Project SHC5.4.4. Since this is a PFAS related project, Susan Burden has been included in these advance notifications. Although the CFS software has been previously presented at various venues as an HSRP product, this is the first time that research using the model on PFAS compounds and comparing model predictions with experimental PFAS measurements on a pilot-scale combustor will be released outside of EPA.
Description:
An existing computational fluid dynamic (CFD) combustion model, originally developed for the Department of Defense (DoD) to model the destruction of chemical warfare agents in demilitarization incinerators was modified to include C1-C3 fluorinated organic chemical reactions and kinetics compiled by the National Institute of Standards and Technology (NIST) and available from the literature. As part of an initial study, a simplified plug flow reactor version of this model was used to predict the destruction efficiency (DE) and products of incomplete combustion (PICs) for three C1 and C2 per- and poly-fluorinated alkyl substances (PFAS, CF4, CHF3, and C2F6) and compare predicted values to limited FTIR-based measurements made from a small pilot-scale EPA research combustor (40-45 kW, natural gas, 20% excess air). PFAS were introduced through the flame, and at selected post flame locations along a time-temperature profile. Results indicate that CF4 is particularly difficult to destroy with maximum DE of ~60% when introduced through the flame at these conditions. Compared to CF4, CHF3 and C2F6 were easier to destroy, exhibiting DEs >99% even when introduced downstream of the flame. This is likely due to the presence of lower energy C-H and C-C bonds to initiate molecular dissociation reactions. However, these lower bond energies may also lead to the formation of CF2 and CF3 radicals at thermal conditions unable to fully de-fluorinate these species, and lead to the formation of fluorinated PICs. DEs determined by the model agreed well with the measurements for CHF3 and C2F6, but overpredicted DEs at high temperatures and underpredicted DEs at low temperatures for CF4. However, high DEs do not necessarily mean absence of PICs, with both model predictions and limited FTIR measurements indicating the presence of similar fluorinated PICs in the combustion emissions. FTIR-measured hydrofluoric acid (HF) concentrations did not agree with modeled HF concentrations, suggesting significant adsorption, reaction, or loss to refractory wall surfaces. Results from this initial study are encouraging, in that additional model development may improve prediction of PFAS destruction.