This work package develops new experimental and computational methods for determining the breakdown products of HFC replacements, HFOs and HCFOs, and their fates in the atmosphere.

WP 4.1: Laboratory studies of HFO and HCFO oxidation (Bristol)

Oxidation of HFOs or HCFOs that contain a -CF3 group, including HFOs 1234yf, 1234ze(E), 1336mzz(Z), 1336mzz(E), 1225ye(Z), and 1225ye(E), and HCFOs 1233zd(E) and 1233zd(Z) leads to the production of trifluoracetic acid (TFA), a highly soluble and stable organic acid, and a persistent, mobile, and ubiquitous environmental pollutant. Previous suggestions that TFA has natural sources have recently been challenged (Joudan et al., 2021), highlighting the need to account more fully for its anthropogenic sources. Recent atmospheric modelling by our team suggests a 33-fold increase in the annual global burden of TFA if all emissions of HFC-134a were replaced by HFO-1234yf (Holland et al., 2021).

HFO or HCFO oxidation is also a potential source of potent GHGs such as HFC-23, CF4 or CF3Cl through multiple pathways. If confirmed, these pathways could increase the GWPs of HFOs and HCFOs. Our preliminary studies have shown that ozonolysis of some HFOs is one such pathway. Another is via the production of CF3CHO (Javadi et al., 2008). CF3CHO has recently been argued to produce HFC-23 at solar actinic wavelengths, but this finding remains uncertain (Campbell et al., 2021, Sulbaek Andersen and Nielsen, 2022). Reactions of Criegee intermediates with CF3CHO and other halo-aldehydes from HFO or HCFO oxidation have large rate coefficients (Taatjes et al. 2012), but these reactions have not yet been factored into estimates of secondary HFC-23 production.

Our experimental studies of HFO and HCFO degradation pathways will use specialist laboratory capabilities at the University of Bristol. The atmospheric EXTreme RAnge (EXTRA) chamber (Leather et al., 2010) will be used to study reactions of HFOs and HCFOs with ozone under different conditions of temperature (190 – 480 K), pressure (300 – 3750 Torr) and humidity (0 – 30% RH), to determine the effects of these parameters on the yields of CF3CHO, HFC-23, CF4, CF3Cl, and TFA. Studies over broader pressure and temperature ranges than encountered in the troposphere are important to extract quantitative mechanistic information. Chamber experiments will be performed with simultaneous high-precision monitoring of HFO or HCFO consumption by ozone, and CF3CHO, HFC-23, CF4 and CF3Cl formation using the Bristol Medusa GC-MS. Using this chamber and detection method with project partner Dr Max McGillen (CNRS, Orléans), we showed that ozonolysis of certain HFOs produces HFC-23 directly via a rearrangement of Criegee intermediates to dioxiranes and then to vibrationally hot TFA which eliminates CO2. These steps compete with collisional thermalization of the reactive intermediates; hence, their efficiencies are likely to depend strongly on pressure, and perhaps also temperature. We will also explore the open question of the outcomes of a possible competing epoxidation pathway from ozone reaction with electron-deficient HFOs and HCFOs.

The OH-radical induced degradation of HFOs via CF3CHO or CF3CFO cannot be studied directly in the Bristol EXTRA chamber. Instead, a collaboration with Dr Max McGillen, using instrumentation in his CNRS laboratory, will provide the required data for the yields of HFC-23 from CF3CHO photochemistry.

Data from EXTRA and the CNRS laboratory measurements will be shared with Dr Keith Kuwata (Macalester College, USA) who will use quantum computational chemistry methods to predict the thermochemistry and activation energy barriers for HFO degradation pathways. He will use the outputs in master-equation kinetic modelling to calculate pressure and temperature dependent rate coefficients. The integration with our laboratory data will provide in-depth understanding of HFO oxidation mechanisms and quantitative predictions of TFA yields. Similar methods can be applied to HCFO degradation.

A second existing apparatus in Bristol will be used to study the impacts of Criegee intermediate chemistry on HFO and HCFO degradation pathways, motivated by a recent report of higher levels of these reactive intermediates in the troposphere than predicted by current models (Caravan et al., 2022). Our prior laboratory work revealed a possible atmospheric removal pathway for TFA by reaction with Criegee intermediates (Chhantyal-Pun et al., 2017), but the consequences of Criegee intermediate chemistry for HFO and HCFO oxidation products need to be better quantified. Recently, the Bristol apparatus was used to measure p and T-dependent rate coefficients for the reaction of the simplest Criegee intermediate CH2OO with CF3CFO (Chhantyal-Pun et al., 2022), and measurements are now needed of the rates of reaction of this and other Criegee intermediates (e.g. (CH3)2COO and those from isoprene ozonolysis) with the halo-aldehydes from HFO and HCFO oxidation. In collaboration with Dr D. Papanastasiou (Honeywell, USA), we will also explore the formation of secondary ozonide products of these halo-aldehyde reactions, characterize their structures by IR spectroscopy and seek to detect and identify their final decomposition products under tropospheric conditions (i.e., in the presence of OH radicals, O2 and NOx). This work will use capabilities in the Honeywell research labs and will provide a fuller understanding of how such chemistry impacts TFA yields in the environment.

WP 4.2: Modelling the consequences of HFO and HCFO oxidation in the atmosphere (Bristol, Cambridge)

To assess the impact of HFO and HCFO oxidation pathways in the Earth’s atmosphere, modelling will be undertaken on global and regional scales using STOCHEM-CRI (a Lagrangian chemical transport model in which chemistry and transport of molecules are uncoupled) and WRF-Chem-CRI (an Eulerian chemical transport model in which chemistry and transport are fully coupled), respectively. Both models incorporate the Common Representative Intermediate (CRI, v2-R5) chemical mechanism (Jenkin et al., 2008; Watson et al., 2008; Utembe et al., 2009), a reduced form of the Master Chemical Mechanism (MCM).

STOCHEM-CRI global modelling will identify regions where HFO or HCFO degradation chemistry is most significant. WRF-Chem-CRI will predict distributions of TFA production from HFOs and HCFOs for important regions, providing data that can be incorporated into models of the environmental and ecosystem impacts of this additional TFA burden. Box-model simulations of surface water systems will be used to evaluate TFA lifetimes and quantify aqueous uptake as a function of pH, metal ion loadings and other parameters. UKCA-CRIStrat modelling will explore the effects of changing climate on the oxidation of HFOs to TFA under low warming (SSP1-26) and high warming (SSP3-70) scenarios.

Using the MCM and CRI protocols, we will devise degradation schemes for the HFOs and HCFOs so that they can be incorporated into the models used in WP 5. The oxidation reactions of HFOs and HCFOs, for which p- and T-dependent rate-coefficients will come from WP 4.1 and the existing literature, and the newly explored roles of CF3CHO, CF3CFO, HFC-23, CF4 and CF3Cl will be incorporated into the CRI mechanisms common to STOCHEM-CRI, UKCA-CRIStrat and WRF-

Chem-CRI models. Modelled GHG production from HFO and HCFO degradation pathways will feed into calculation of REs in WP 5.

References