AN UPDATE ON FLUOROCARBONS

Update on the atmospheric degradation of HFOs and HCFOs

Three recent academic papers have further enhanced our understanding of the atmospheric degradation of HFOs and HCFOs.

Reaction of Ozone with HFOs/HCFOs. The minor reaction of HFOs and HCFOs with ozone in the atmosphere was already known, but a recent paper by McGillen et. al. [1] has identified the formation of extremely small yields of HFC-23 for some HFOs/HCFOs. The yields vary depending on the specific HFO/HCFO and according to the paper are related to the molecular structure. No HFC-23 was formed from HFO-1234yf, the calculated overall yield of HFC-23 from data reported in the paper for HFO-1336mzz(Z) is 0.0005%, and from HFO-1234ze(E), 0.092%.  The overall yield is calculated from the percentage of HFO/HCFO that reacts with ozone compared to reaction with hydroxyl radical (the main reaction pathway) and the yield of HFC-23 following reaction with ozone. This paper, discussed in more detail in the next newsletter item, was reported in Science with this comment:

“Still, the actual implications of the work—published in the Proceedings of the National Academy of Sciences—are unclear, says Stephen Montzka, an atmospheric scientist at the U.S. National Oceanic and Atmospheric Administration who was not involved in the research. The amount of fluoroform [HFC-23] produced by HFOs is minute, he notes, so they still have a lower global warming potential than chemicals used in the past.”

Potential emissions of HFC-23 from reaction with ozone (ozonolysis) in the context of reported global HFC-23 emissions are discussed in the third newsletter item. 

Reaction of HO2 radical with trifluoroacetaldehyde. A second paper by Long et. al. [2] discussed in the fourth newsletter item, describes a theoretical investigation of the reaction of trifluoroacetaldehyde (CF3CHO), an intermediate degradation product from some HFOs/HCFOs (not HFO-1234yf), with HO2 radical in the atmosphere, and concludes that that the reaction of CF3CHO with HO2 radical is the dominant reaction pathway compared to the reaction with OH radical. This may influence the yields of important degradation products from the degradation of CF3CHO.

Photolytic Lifetime of Trifluoroacetaldehyde. The third paper by Sulbaek Andersen et. al. [3], also reported in the November 2023 EFCTC newsletter, provided a revised atmospheric lifetime for the photolysis reaction of CF3CHO, which is discussed, together with the HO2 radical reaction pathway.

Reaction of Ozone with HFOs and HCFOs

The presence of an olefinic bond (C=C bond) in HFOs and HCFOs results in their relatively rapid reaction with hydroxyl (OH) radicals present in the atmosphere, leading to short atmospheric lifetimes measured in days or weeks and small direct global warming potentials.

However, the olefinic bond also allows minor reaction with ozone (ozonolysis), and whilst these reactions are slow, McGillen et. al. [1] report that HFC-23 may be produced in very small yields from some HFOs/HCFOs. According to McGillen et. al. no HFC-23 is formed from HFO-1234yf.  The paper accounts for this process in atmospheric chemical and transport modelling simulations and reports the indirect global warming potentials due to the formation of HFC-23. The results reported in the paper are summarised in the table below. The calculated indirect GWPs due to HFC-23 formation are also shown in the table. It is worthwhile noting that some hydrocarbon refrigerants have indirect GWPs [2] of similar magnitude (propane, 9.50; n-butane, 6.5). All of these refrigerants (HFOs, HCFOs and HCs) are classified as ultralow GWP refrigerants (0-30 GWP) [3].  

The ozonolysis experiments for HFO-1234yf also showed no formation of CF4 which might, potentially, have been formed by a similar mechanism to HFC-23 (CF3H) formation for some HFOs.

Explanatory notes for the table:

The yields of HFC-23 following reaction with ozone (ozonolysis) are experimental measurements. The percentage of degradation by reaction with ozone route is calculated from atmospheric modelling simulations, with the main degradation route being the reaction with hydroxyl (OH) radical. The indirect GWP due to HFC-23 formation in this table is taken from Fig.4 of the paper. The supporting information for the paper explains that the GWP was calculated for each HFO under consideration and also for CHF3. The calculated direct GWPs for the HFOs in Fig.4 of the paper are slightly different from the 100 year AR6 GWP values, which are included in the table for completeness. HCFO-1233xf and HFO-1243zf are not used commercially as refrigerants but are used in the study due to their molecular structure to improve understanding of the mechanism and structural effects.

References

[1] Ozonolysis can produce long-lived greenhouse gases from commercial refrigerants, Max R. McGillen, Zachary T. P. Fried, M. Anwar H. Khan, Keith T. Kuwata, Connor M. Martin, Simon O’Doherty , Francesco Pecere, Dudley E. Shallcross, Kieran M. Stanley , and Kexin Zhang, PNAS 2023 Vol. 120 No. 51 e2312714120, https://doi.org/10.1073/pnas.2312714120

[2] Lifetimes, direct and indirect radiative forcing, and global warming potentials of ethane (C2H6), propane (C3H8), and butane (C4H10), Atmos. Sci. Lett. 2018;19:e804. wileyonlinelibrary.com/journal/asl2 https://doi.org/10.1002/asl.804  Øivind Hodnebrog, Stig B. Dalsøren, Gunnar Myhre, Center for International Climate and Environmental Research-Oslo (CICERO), Oslo, Norway

[3] Montreal Protocol on Substances that Deplete the Ozone Layer, UNEP 2022 Report of the Refrigeration, Air-conditioning and Heat Pumps Technical Options Committee 2022 Assessment. Table 3-1: Classification of 100-year GWP levels, available at Technology and Economic Assessment Panel (TEAP) | Ozone Secretariat (unep.org)

Potential emissions of HFC-23 due to reaction of ozone with HFOs/HCFO

The potential emissions of HFC-23 due to reaction of ozone (ozonolysis) with HFOs/HCFOs reported by McGillen et. al. [1] are extremely small and have not materially contributed to the increase in atmospheric HFC-23 emissions observed in recent years. The SAP 2022 Assessment provides an estimate of HFO emissions and assumes that 50% of the HFO emissions are HFO-1234yf. For 2020, it estimates HFO 1234yf emissions at 30,000 tonnes [2]. Therefore, the other HFO emissions are also estimated to be 30,000 tonnes, and this may have resulted in less than 30 tonnes of HFC-23 emissions in 2020, based on the reported yields of HFC-23 from HFOs by reaction with ozone (ozonolysis).  This is less than 0.2% of the reported emissions of HFC-23 in 2020. According to McGillen et. al., no HFC-23 is formed from HFO-1234yf by reaction with ozone.  

The SAP 2022 Assessment reports a rise in HFC-23 emissions derived from atmospheric monitoring to 17.2 ± 0.8 Gg yr–1 (17,200 tonnes) in 2019, and a similar value of 16.5 ± 0.8 Gg yr–1 (16,500 tonnes) in 2020. Up until 2013, global bottom-up emissions, derived from reported data, track (within ± 2 Gg yr–1, 2,000 tonnes) the global emissions derived from atmospheric measurements. Similarly, up until 2013, the ratio of HFC-23 emissions to HCFC-22 production (E23/P22) derived from atmospheric data closely matched that derived from bottom-up estimates. Between 2015 and 2019, as reported abatement increased dramatically in China and India, bottom-up emissions (derived from reported data) and E23/P22 declined substantially. However, emissions and E23/P22, derived from atmospheric data, increased. By 2019, the difference between top-down and bottom-up emissions and E23/P22 was the largest since atmospheric records began [3]

The SAP 2022 Assessment provides a forecast of the reduction in emissions as CO2e from the replacement of HFCs by low-GWP HFOs and non-halogenated alternatives, assuming full compliance with the provisions of the Montreal Protocol, including the Kigali Amendment. Figure 7-5 from the SAP 2022 Assessment (see Figure abover), reproduced here, shows historical and projected emissions of ODSs, HFCs and alternatives. The explanation notes that “The contributions of the low-GWP HFOs in panels b and c, are smaller than the thickness of the green curves.” Any potential emissions of HFC-23 due to ozonolysis of some HFOs would not change this observation.

The use of HFOs and HCFOs with a good balance of safety, technical and environmental properties (low or no flammability, ultra-low GWP) has greatly contributed to the reduction in use of HFCs to meet the requirements of the F-gas Regulation and the Kigali Amendment. The SAP 2022 Assessment [4] states that “In short, the replacement of old equipment containing HFCs with high Global Warming Potentials (GWPs) by new installations and low-GWP alternatives, as well as not-in-kind solutions, has the potential for multiple positive effects on climate change. For example, the emissions of low-GWP alternatives will directly reduce projected radiative forcing of climate. Also, and thought to have greater potential climate benefit, the transition to new refrigerants is an opportunity to implement design changes for achieving higher energy efficiency and therefore lowering greenhouse gas emissions from energy use.”

References

[1] Ozonolysis can produce long-lived greenhouse gases from commercial refrigerants, Max R. McGillen, Zachary T. P. Fried, M. Anwar H. Khan, Keith T. Kuwata, Connor M. Martin, Simon O’Doherty , Francesco Pecere, Dudley E. Shallcross, Kieran M. Stanley , and Kexin Zhang, PNAS 2023 Vol. 120 No. 51 e2312714120, https://doi.org/10.1073/pnas.2312714120

[2] World Meteorological Organization (WMO). Scientific Assessment of Ozone Depletion: 2022, GAW Report No. 278, 509 pp.; WMO: Geneva, 2022. Section 7.2.5.1

[3] TEAP Report, September 2023: Volume 6: Response to Decision XXXIV/7: Strengthening institutional processes with respect to information on HFC-23 by-product emissions: 3.3.1 HFC-23 emissions derived from atmospheric monitoring

[4] Scientific Assessment of Ozone Depletion: 2022 Section 7.2.2.3 Energy Efficiency

Reaction of HO2 radical with trifluoroacetaldehyde (CF3CHO) in the atmosphere

Trifluoroacetaldehyde (CF3CHO) is an intermediate degradation product for some HFOs and HCFOs (not HFO-1234yf) produced by their reaction with hydroxyl radical (OH). Long et. al. [1] report investigations that are mainly focused on obtaining quantitative rate constants for the reactions of aldehydes with HO2 radical, using computational chemistry. Such rate constants are fundamental parameters for atmospheric modelling. The paper provides new results for the full range of atmospheric temperatures and pressures for the HO2 reaction with CF3CHO and concludes that the present findings also have probable implications in understanding the atmospheric lifetimes of perfluorinated aldehydes. The results show that the theoretical methods and computational strategies used in the paper have accuracies comparable to available experiments, where experimental data is available, i.e. for HCHO and CH3CHO, but no experimental data was reported in the Long, et. al. paper for CF3CHO reaction with HO2 radical, and according to Long et. al., there are no reported data elsewhere for HO2 + CF3CHO kinetics.

From the calculated rate constants for CF3CHO with HO2 radical, the paper concludes that the reaction of CF3CHO with HO2 radical is the dominant reaction pathway compared to reaction with OH radical.  The paper comments that in the literature, the major reaction for these aldehydes is considered to be reaction with OH. However, the reported data shows that the reaction of HO2 with CF3CHO dominates over OH + CF3CHO at 0−50 km altitude [the troposphere is up to about 14 km].

According to the paper, the intermediate from the reversible reaction of CF3CHO with HO2 radical can further react with NO. Although the paper does not discuss the atmospheric fate of the NO adduct, it may result in efficient removal of the aldehyde, according to a paper [2] referenced by Long et. al.

CF3CHO + HO2 CF3CH(OH)OO

CF3CH(OH)OO + NO à CF3CH(OH)O + NO2

EFCTC comment

CF3CH(OH)O radical could subsequently break down on the C-C bond which leads to CF3 radical and HCOOH (formic acid, one of the most abundant organic molecules in Earth’s atmosphere). The CF3 radical would react further leading to formation of CO2 and HF via initial reaction with O2. Further work would be required to determine the different alkoxy radical pathways. Based on the kinetic data presented in the paper, and the average atmospheric concentrations of HO2 and OH radicals [3], a simple calculation, as very rough estimate, is that 70% of CF3CHO degradation may potentially occur via the HO2 reaction pathway, with the remainder via photolysis or reaction with OH radical. However, a more detailed transport model that accounts for the different physicochemical processes would provide a more accurate estimate. Potentially, there are three important pathways for CF3CHO degradation in the troposphere:

• Photolysis which leads to complete degradation in the troposphere. According to Sulbaek-Andersen, et. al. [4] any CF3CHO produced within the lower troposphere will be a negligible source of CF3
• Reaction with OH radical lifetime which leads to minor formation of TFA.
• Reaction with HO2 radical which may potentially lead to complete degradation.

The photolytic lifetime of trifluoroacetaldehyde (CF3CHO): A revised atmospheric lifetime for the photolysis reaction of CF3CHO has been reported by Sulbaek-Andersen, et. al. [4].  The revised lifetime of about 13 days, is longer than the previously determined value of ~ 2 days [5]. The significance of this revision is to (i) reduce the relative contribution of photolysis to the atmospheric removal of CF3CHO compared to earlier evaluations, and (ii) increase the relative contribution of other removal processes (i.e., reaction of CF3CHO with HO or HO2 radicals) compared to earlier evaluations. This Sulbaek-Andersen, et. al. paper was also reported in the November 2023 EFCTC newsletter.

References

[1] Quantitative Kinetics of HO2 Reactions with Aldehydes in the Atmosphere: High-Order Dynamic Correlation, Anharmonicity, and Falloff Effects Are All Important, Bo Long, Yu Xia, and Donald G. Truhlar, Journal of the American Chemical Society 2022 144 (43), 19910-19920, https://doi.org/10.1021/jacs.2c07994

[2] Kinetics of α-Hydroxy-alkylperoxyl Radicals in Oxidation Processes. HO2•-Initiated Oxidation of Ketones/Aldehydes near the Tropopause, Ive Hermans, Jean-François Müller, Thanh Lam Nguyen, Pierre A. Jacobs, and Jozef Peeters, J. Phys. Chem. A 2005, 109, 19, 4303–4311, https://doi.org/10.1021/jp044080v

[3] Concentrations of OH and HO2 radicals during NAMBLEX: measurements and steady state analysis, S. C. Smith, J. D. Lee, W. J. Bloss, G. P. Johnson, T. Ingham, and D. E. Heard, Atmos. Chem. Phys., 6, 1435–1453, 2006, www.atmos-chem-phys.net/6/1435/2006/

[4] Photolysis of CF3CHO at 254 nm and potential contribution to the atmospheric abundance of HFC-23, Mads Peter Sulbaek Andersen, Sasha Madronich, Joanna May Ohide, Morten Frausig, and Ole John Nielsen, Atmospheric Environment Volume 314, 1 December 2023, 120087, https://doi.org/10.1016/j.atmosenv.2023.120087

[5] Atmospheric Chemistry of Perfluoroaldehydes (CxF2x+1CHO) and Fluorotelomer Aldehydes (CxF2x+1CH2CHO): Quantification of the Important Role of Photolysis, Malisa S. Chiappero, Fabio E. Malanca, Gustavo A. Arguello, Steven T. Wooldridge, Michael D. Hurley, James C. Ball, Timothy J. Wallington, Robert L. Waterland, and Robert C. Buck, J. Phys. Chem. A 2006, 110, 11944.  https://pubs.acs.org/doi/10.1021/jp064262k