2. Experimental and computational details

Calculation method

In this paper, the Gaussian12 software package was used to study the thermal decomposition and fire-extinguishing mechanism of CF3I by means of Ab initio quantum chemistry and DFT. The geometrical configuration optimization and vibration analysis of all stationary points involved in the reactions were obtained at the B3LYP/LanL2DZ level, and at the same level group, the relationship between reactants, products, intermediates and transition states (TS) was analyzed by using the theory of intrinsic reaction coordinate (IRC), and the correctness of each reaction path was verified. For the analysis for transition states, TS, QST2 and QST3 methods are employed when necessary. In order to obtain more accurate energy value, a more accurate method was used to calculate the single point energy of the stationary point optimized in the reaction at the level of CCSD/LanL2DZ basis set, and the accurate energy barrier value was obtained. All bond dissociation energies (BDEs) and energy barriers are corrected by zero point energy (ZPE).
In addition, on the basis of calculating the energy barrier of reaction path, the classical variational transition state theory (CVT) method, which is the most widely used in VTST, is used to calculate the reaction rate constants of each reaction path under the condition of considering Eckart tunneling effect, to further verify the possibility of each reaction path. The above calculation was completed by the Kisthelp13 package.

2.2 Experimental method

2.2.1 Thermal decomposition analysis

The thermal decomposition of CF3I (Yuji Tech, 99%) was conducted in a quartz tube reactor under argon (Tianjin sizhi gas Co. Ltd, 99.9%) flow (Fig.S1 ). In the experiment, CF3I was first premixed with the argon carrier gas in a mixing chamber, and their entering flow rates were adjusted separately by standard mass flowmeters. At the outlet of the chamber, samples were extracted for GC-MS (Thermo Fisher Scientific, Trace 1310) instrument equipped with a DB-VRX column (Agilent, 30 m × 0.25 mm i.d., 1.4um film thickness). Then, CF3I was further carried into the tubular furnace, and its residence time was regulated by the gas flowing rate. The residence time was chosen to be 10 s and the volume fraction of CF3I was set as 20%. The pyrolysis temperature ranged from 200 to 800℃. After the thermal paralysis in the quartz tube, the decomposition products were analyzed by GC-MS. At different pyrolysis temperatures, the sampling and GC-MS detection were repeated three times at a time interval of 30 min, to ensure the reliable and reproducible of the results. The temperature was increased at a ramping rate of 5℃/min to the next targeting temperature and maintained stable for 30 min.

2.2.2 Fire extinguishing concentration measurement

The FEC of CF3I for extinguishing methane-air flame and propane-air flame were measured by the cup burner (Fig.S2 ) method according to ISO14520-1 and NFPA 2001 standards14, 15. Volumetric flow rates of synthetic air (Tianjin sizhi gas Co. Ltd, O2 20.9%, N2 79.1%) and gas fuel were fixed at 40 L/min, 356 mL/min (methane), and 118 mL/min (propane), respectively, to achieve a visible flame length of 80 mm. After the flame was pre-burned for 60 s, the extinguishing agent of CF3I was delivered into the flame burner, until flame blow-off occurred. The mean FEC was determined based on five consecutive test trial results. All the flow meters deployed in this paper were calibrated by the soap-film method or the drainage method before the experiments, and the total uncertainty of FEC obtained by the experiment was estimated as 5%. The flame extinguisher was recorded by using a high-speed camera (Phantom Miro LAB110) which operated at 500 frames per second, with an exposure time of 40μs.