DOI: 10.3303/CET24111080 Paper Received: 16 January 2024; Revised: 1 June 2024; Accepted: 10 August 2024 Please cite this article as: Palma D., Ubaldi S., Quinterno M., Gentile P., Mazzaro M., Russo P., 2024, Experimental & Theoretical Study on the Solid, Liquid and Gaseous Residues After Fire Extinction of Lithium-ion Batteries, Chemical Engineering Transactions, 111, 475-480 DOI:10.3303/CET24111080 CHEMICAL ENGINEERING TRANSACTIONS VOL. 111, 2024 A publication of The Italian Association of Chemical Engineering Online at www.cetjournal.it Guest Editors: Valerio Cozzani, Bruno Fabiano, Genserik Reniers Copyright © 2024, AIDIC Servizi S.r.l. ISBN 979-12-81206-11-3; ISSN 2283-9216 Experimental & Theoretical Study on the Solid, Liquid and Gaseous Residues after Fire Extinction of Lithium-Ion Batteries Davide Palmaa, Sofia Ubaldia, Mario Quinternob, Pierpaolo Gentileb, Michele Mazzarob, Paola Russoa* a Dipartimento Ingegneria Chimica Materiali Ambiente, Sapienza Università di Roma, Via Eudossiana 18, Roma, Italy bDirezione Centrale per la Prevenzione e la Sicurezza Tecnica Antincendio ed Energetica. Corpo Nazionale dei Vigili del Fuoco, Largo S. Barbara 2, Roma, Italy. paola.russo@uniroma1.it The electric vehicles are a good option for aiming the Energy transition, in a such important sector such as the mobility. The use of a huge quantity of high energy and high power Lithium-Ion Batteries (LIBs) opens the discussion of many topics such as competitiveness compared to other technologies, recycling of the materials and the safety. This paper shows an experimental study based on fire extinguishing tests on NMC Lithium-Ion pouch cells, each with approximately 100 Wh energy stored. Different extinguishing agents were tested: Foam, Aqueous Vermiculite Dispersion (AVD). After tests, solid samples were taken from the burnt cell and the residue extinguishing liquid was collected. Gases produced from the combustion of the cell were also sampled during the test and collected in gas bags. Chemical analyses were performed on solid, liquid and gaseous residues in order to evaluate presence of hazardous compounds for health and environment. Gas Chromatography Mass Spectrometry and a Fourier-transform Infrared Spectroscopy were used to handle this type of analysis. The aim of this experimental work is to identify the hazardous compounds present in the residue of LIB after the fire is extinguished (i.e. electric vehicles) to evaluate their health and environmental impact and then to propose a methodology for their treatment or disposal. In particular, a comparison between different extinguishing agents is reported. 1. Introduction To meet the emission limits imposed by the European Community the electric mobility will play an important role for the next 30 years. For supporting the required increasing electric vehicles (EVs) range a huge penetration of Lithium-Ion Batteries (LIBs) will take place. Incidents related to this technology should be investigated for a deep comprehension of the environmental impact that they may entail. In particular, the development of a fire from the LIBs stored in a EV might generate toxic and flammable gases. These gases are the product of several reactions occurring during the so called thermal runaway (TR), in which the chemical compounds within a cell react. Therefore, as consequence of TR fire and/or explosion may occur. Different agents were proposed in the literature for LIBs fires. Novec 1230 as extinguishing agent for LIBs fire was studied by Liu et al. (2018). Rao et al. (2015) showed that ABC powder extinguishing agent has low efficiency against LIBs fires. The efficiency of different water-based extinguishing agents, such as pure water, water + F500 (1%), water + Firesorb (1.8%) was studied by Egelhaalf et al. (2013). In our previous studies (Ubaldi et al., 2022; Palma et al., 2023), we demonstrated that water based agents (i.e., water mist, F500-water spray and F500-water mist) exhibited a higher extinction efficiency than CO2, foam and AVD. After the extinction the solid and liquid residues may also cause and environmental contamination. Hence, in this paper an experimental work about the chemical analysis of gas, liquid and solid residues after fire extinction of LIBs is reported. In the literature, Golubkov et al. (2014) studied the production of gases of burning cells from three different cathode compositions such as lithium iron phosphate (LFP), lithium cobalt oxide (LCO), lithium nickel cobalt oxide (NMC). Ribière et al. (2012) analysed the combustion gases of pouch LIBs at different State of Charge (SOCs). Larsson et al. (2017) investigated HF 475 mailto:paola.russo@uniroma1.it https://www.bing.com/ck/a?!&&p=14f50dbc9dec8779JmltdHM9MTcwMDAwNjQwMCZpZ3VpZD0wMzdmMzE2NC1lYThiLTY0MzgtMDI0ZS0zZmMxZWI2YTY1YzgmaW5zaWQ9NTIyNA&ptn=3&ver=2&hsh=3&fclid=037f3164-ea8b-6438-024e-3fc1eb6a65c8&psq=ftir+spectroscopy&u=a1aHR0cHM6Ly9lbi53aWtpcGVkaWEub3JnL3dpa2kvRm91cmllci10cmFuc2Zvcm1faW5mcmFyZWRfc3BlY3Ryb3Njb3B5&ntb=1 production and composition from fire of LIBs with different cathode chemistry such as LCO, LFP and lithium nickel cobalt aluminium oxide (NCA). They found that the amount of the HF produced increases with the higher SOCs. They also impute a higher HF concentration from pouch cells than cylindrical cells due to the lower venting pressure. Ubaldi et al. (2023) studied the key events, the gas and particulate emissions due to thermal abuse of NCA cells, evidencing the formation of toxic concentrations of HF and particle sizes of the order of PM2.5. Hynynen et al. (2023) performed a large scale fire test for measuring the fire behaviour of internal combustion engine vehicle (ICEV) and EV, highlighting the differential HF production between the two technologies. Jia et al. (2023) proposed a comparison between the gas composition of LIBs under overcharging and overheating conditions. Funk et al (2023) realized full scale EV fire tests, measuring the HF production. Andersson et al (2013) studied how the combustion gases composition (in terms of HF and POF3) is influenced by the electrolyte composition, testing different solutions of electrolytes and lithium hexafluorophosphate (LiPF6). But, far as the authors know, analytical studies on solid and liquid residues are not reported in the literature. The aim of these experimental tests is to compare how the extinguishing actions of the AVD and Foam (fire class 27A 233B 40F), certified for LIBs fires, influence the gas, liquid and solid residues composition. A theoretical calculation model has been developed to carry out the gas quantitative analysis. 2. Experimental 2.1 Materials Cells used were pouch cells Kokam SLPB 25 Ah (Kokam), with a nominal energy of 92 Wh. Cells have square shape of 22 cm x 22 cm x 1 cm. The cathode chemistry is NMC and the anode is graphite. The layers of anode and cathode are immersed in an electrolyte made of a mixture of ethylene carbonate (EC), ethyl methyl carbonate (EMC) and LiPF6. 2.2 Fire extinguishing tests & sampling Experimental tests were performed to investigate the extinction efficiency of each extinguishing agent for a fire of a single cell. Each cell was placed horizontally on a metal grate and heated by two electric plates. 6 cm distance between the electric plates was set for ensuring enough space for the swelling of the cell (Figure 1). A portion of the combustion gases was sampled, filtered and collected in a gas bag (Figure 2). A gas pump set on PDISCHARGE = 1.15 bar and FDISCHARGE = 2.5 L/min was chosen and a sampling time of 110 s was set for a 5 L gas bag. A multy-layer gas bag (15 μm nylon, 7.62 μm aluminum and 50 μm Polyethylene) was used. For each test the liquid residues, the solid burnt cells, the Activated Carbon (A.C.) filters and the combustion gases were sampled for the subsequential chemical analysis. The liquids were collected within falcon test tubes (mL 50). The burnt cells were stored into high resistant plastic bag and preserved before solid sampling. In the following a data elaboration of the chemical analysis carried out on the residues of test of Free burning cell without extinguishing agent (Test 16), of test with foam extinguishing agent (Test 19) and with AVD (Test 20) is reported. Foam composition is: EWAB15 (2 % SC 6 + 10 % Inilam AX + 88 % Water). 2.3 Analytical methods A quantitative analysis of the combustion gases was performed through a Perkin Elmer Spectrum 3TM FT-IR Spectrometer with the fallowing acquisition parameters: resolution 4 cm-1, spectral range between 4500 and 650 cm-1, 8 scans per spectrum and MCT as detector. Spectra were acquired continuously. A previous calibration obtained by standard gas was used for identifying and quantifying the species. A qualitative analysis of volatile organic compounds (VOCs) in the liquid, solid and A.C. filters was performed by Solid Phase micro Extraction Gas Chromatography-Mass Spectroscopy (SPME-GC-MS). Fiber Type: PK3, FFA SPME 100 µm PDMS, 23 Ga – RED. Column type: Agilent: HP – 5MS UI – 60 m x 0.250 mm x 0.25 µm. Figure 1: Scheme of the gas sampling system Figure 2: Gas bag 1 m Ci COMB (t) Ci StB (t)Ci HOOD (t) GAS BAG TO SAFETY RELEASE CELL L = 5m ID = 8 mm L = 1m ID = 4 mm GAS BAG TO SAFETY RELEASE L = 5m ID = 8 mm 476 Regarding the quantitative analyses, the gas stored in the gas bag is sent to the FT-IR with constant flow rate (FG B). Before it reaches the analysis cell, a dilution with N2 is required for avoiding signal saturation. Once diluted, only a portion of the total flow, (FG B + FN2)*Φ = FG F+FN2 F, is sent to the analysis. 3. Results 3.1 Gas Analysis results The gas stored in the gas bag was analyzed by FT-IR. The quantifiable compounds were CH4, CO, CO2, DMC, DEC, EC, HCl and HF. H2 is not detectable by this analysis. The results of free burning test, no extinction (Test 16), of test with foam as extinguishing agent (Test 19) and that with AVD agent (Test 20) are compared in order to understand how the different extinguishing agent influences the combustion gas composition. In table 1 are reported the concentration (Ci B) and the mass (mi B) of the i-components in the gas collected in each test. Table 1: Gas concentration and gas mass in the gas bag for Test 16, Test 19 and Test 20 Test EXT UOM CH4 CO2 EC HCl HF ∑ 16 - Ci B ppm(v) 10.9 141.1 0.0 20.9 4.0 176.8 19 Foam Ci B ppm(v) 1.5 303.6 60.1 7.5 9.1 381.8 20 AVD Ci B ppm(v) 32.8 2122.3 157 1.8 9.3 2323.2 16 - mi B g 2.1*10-5 7.6*10-4 0 9.4*10-5 9.8*10-6 8.9*10-4 19 Foam mi B g 3.0*10-6 1.6*10-3 6.5*10-4 3.4*10-5 2.2*10-5 2.4*10-3 20 AVD mi B g 6.5*10-5 1.1*10-2 1.7*10-3 7.9*10-6 2.3*10-5 1.3*10-2 3.2 Gas results interpretation Since FTIR analysis does not detect all the gas compounds produced during the fire, it is useful to write some definitions. Let’s call mC B the mass of all the compounds sampled (i.e. not including the atmospheric air). Moreover, let’s call mC B° = ∑ mi B the total mass of the compounds detected and mC BØ = ∑mj B the total mass of the not detected compounds such that mC B = mC B° + mC BØ. The subscript “i” indicates a detected compound, while the subscript “j” indicates a not detected compound. The apex “°“ refers to the total detected compounds (hence not including air), while apex “Ø” to the total not detected compounds (no air). Figure 3a,b,c show the percentages of the i-th volumetric gas composition in the bag referred only to the detected compounds (Ci B°). 𝐶𝑖 𝐵⋄ (% 𝑉 𝑉 ) = 𝑉𝑖 𝐵 𝑉𝐶 𝐵° ∗ 100 = 𝑉𝑖 𝐵 ∑ 𝑉𝑖 𝐵 ∗ 100 = 𝑉𝑖 𝐵 𝑉𝐺 𝐵⁄ ∑(𝑉𝑖 𝐵 𝑉𝐺 𝐵)⁄ ∗ 100 = 𝐶𝑖 𝐵 ∑ 𝐶𝑖 𝐵 ∗ 100 (1) (a) (b) (c) Figure 3: Gas composition (% V/V) of the total compounds detected in Test 16 (a), Test 19 (b) and Test 20 (c) The concentrations of the sampled and detected gases referred to the total detected compounds (Ci B°) are comparable between the 3 tests. Figure 3a,b,c show that CO2 is the main compound produced during the fire and the extinction phases. However, it can be highlighted how in the free burning test (Test 16) HCl is in a percentage of 11.8 %V/V with respect to the all the detected gases, while is 2.0 %V/V and 0.1 %V/V respectively for Test 19 and Test 20. On the contrary, the electrolyte EC is completely burnt in Test 16, while it is present in Test 19 as 15.7 %V/V and in Test 20 as 6.8 %V/V. It is likely that this evidence is related to the cooling effect provided by the extinguishing agents, which restrained the burning process. CH4 6.2% C02 79.8% EC 0.0% HCl 11.8% HF 2.3% Test 16 - (%vol) CH4 C02 EC HCl HF CH4 0.4% C02 79.5% EC 15.7% HCl 2.0% HF 2.4% Test 19 - (%vol) CH4 C02 EC HCl HF CH4 1.4% C02 91.4% EC 6.8% HCl 0.1% HF 0.4% Test 20 - (%vol) CH4 C02 EC HCl HF 477 3.3 A comparison between the Gas composition in the tests This paragraph presents a physical model of the hood and the gas sampling system, which is intended to facilitate an understanding of the roles of all the parameters. It is assumed that, due to the hood aspiration velocity, the vertical diffusion characteristic time (ԏz) is much smaller than the horizontal diffusion characteristic time (ԏx,y), for all the compounds. Thus, it is possible to assume that Ci HOOD° only depends on z and time (see Figure 4). On the other hand, the total gas flowing through the hood (mG HOOD) is the sum of the total combustion compounds mass (mC COMB) and the atmospheric air mass (mA ATM), during aspiration. These considerations bring to Eq(2a) and Eq(2b): 𝛿2𝐶𝑖 𝐻𝑂𝑂𝐷(𝑥,𝑦,𝑧,𝑡) 𝛿𝑥𝛿𝑦 ≠ 0 ; 𝛿2𝐶𝑖 𝐻𝑂𝑂𝐷°(𝑧,𝑡) 𝛿𝑥𝛿𝑦 = 0 (2a) ; (2b) Ci HOOD° is the i-th gas compound concentration in the hood referred only to the detectable compounds. From Eq(2a), the concentration of the compounds in the gas through the hood (Ci HOOD) is a function of the tridimensional space (x,y,z), depending on the random distribution of the combustion compounds in air across the horizontal section (see Figure 4), during the aspiration. Hence, the mass of the total compounds sampled in the gas bag (mC B) might be affected by a random error due to the sampling conditions. Consequentially, mi B might be affected too, such that Eq(3) can be written, where EA is the extinguishing agent action: 𝐶𝑖 𝐵 𝑓𝑢𝑛𝑐𝑡𝑖𝑜𝑛 (𝐸𝐴, 𝐶𝑖 𝐻𝑂𝑂𝐷) (3) On the other hand from Eq(2b), at the sampling point height (HSP) Eq(4) can be assumed: 𝐶𝑖 𝐻𝑂𝑂𝐷°(𝑡, 𝑧 = 𝐻𝑆𝑃) ≡ 𝐶𝑖 𝑆𝑡𝐵°(𝑡) ∀ x,y ∈ hood section (4) Because the distance between the gas source height (HGS) and HSP is small (1 m), Eq(5) can be assumed: 𝐶𝑖 𝐶𝑂𝑀𝐵°(𝑡) ≃ 𝐶𝑖 𝐻𝑂𝑂𝐷°(𝑡, 𝑧 = 𝐻𝑆𝑃) ∀ t ∈ ΔtSAMPLING (5) Thus, 𝐶𝑖 𝐶𝑂𝑀𝐵°(𝑡) ≃ 𝐶𝑖 𝑆𝑡𝐵°(𝑡) → 𝐶𝑖 𝐶𝑂𝑀𝐵° = ∫ 𝐶𝑖 𝐶𝑂𝑀𝐵°(𝑡)𝑑𝑡 = ∫ 𝐶𝑖 𝑆𝑡𝐵°(𝑡)𝑑𝑡 = 𝐶𝑖 𝐵° (6) Figure 4: Hood & Aspiration Point physical model For making the comparison between the gas composition of the two tests the ratio of the total produced masses of the i-th compound (Ω°) has to be considered. From Eq(6): 𝛺° = 𝑚𝑖 𝐶𝑂𝑀𝐵(Test 19) 𝑚𝑖 𝐶𝑂𝑀𝐵(Test 20) = 𝐶𝑖 𝐵°(19) 𝐶𝑖 𝐵°(20) ∗ 𝑚𝐶 𝐶𝑂𝑀𝐵°(19) 𝑚𝐶 𝐶𝑂𝑀𝐵°(20) 𝑓𝑢𝑛𝑐𝑡𝑖𝑜𝑛 (𝐶𝑖 𝐵°(𝐸𝐴), 𝑚𝐶 𝐶𝑂𝑀𝐵°(𝐸𝐴)) (7a) Due to the fact that mC COMB° is unknown from the tests, the calculation is expressed as ω° in Eq(9b): 𝜔° = 𝑚𝑖 𝐶𝑂𝑀𝐵(19) 𝑚𝐶 𝐶𝑂𝑀𝐵°(19) ⁄ 𝑚𝑖 𝐶𝑂𝑀𝐵(20) 𝑚𝐶 𝐶𝑂𝑀𝐵° (20) ⁄ = 𝑚𝑖 𝐵(19) 𝑚𝐶 𝐵°⁄ (19) 𝑚𝑖 𝐵(20) 𝑚𝐶 𝐵°⁄ (20) = 𝐶𝑖 𝐵°(19) 𝐶𝑖 𝐵°(20) 𝑓𝑢𝑛𝑐𝑡𝑖𝑜𝑛 (𝐶𝑖 𝐵°(𝐸𝐴) ) (7b) With regard to Eq(3), Ci B is a function of EA and the three-dimensional space (Ci HOOD(x,y,z)), using Ci B° in Eq(7a), instead of Ci B (i.e. relating the calculation to the detectable compounds), allows Ω° to be decoupled from the random composition of the total gas flowing through the aspiration hood. This allows a comparison to be made z HSP y x 1 m mC COMB + AIRATM Ci COMB (t) HGS Ci StB (t) Ci HOOD(t,x,y,z) HOOD Ci HOOD°(t,z) 478 between different tests regardless of the mixing conditions between the compounds and the air across the hood section. Furthermore, since mC COMB° is unknown the calculation model is expressed as a normalized ratio in Eq(7b). This ratio (ω°) is calculated from known quantities (mi B and mC B°). For clarity, since mC COMB° is a function of EA, the influence of EA on the formation and evolution of the detected combustion compounds mass (mC COMB°) is lost when going from Ω° of Eq(7a) to ω° of Eq(7b). Table 2: Ratio between the compound’s emissions in the two extinction tests with foam (19) and AVD (20) ω° CH4 CO2 EC HCl HF 0.3 0.8 2.2 24.0 5.5 From Table 2, when ω ° > 1 → the AVD agent (Test 20) shot down more gas emission than the Foam agent (Test 19). Where ω° < 1 → the Foam agent shot down more gas emission than the AVD. This calculation shows that the Foam extinguishing agent is more efficient in inhibiting the production (and the evolution) of compounds as CH4 and CO2. On the other hand, AVD was more efficient in reducing the emissions of EC, HF and HCl. 3.4 Liquid, Solid, Filter Analysis results In Table 3 and 4 the VOCs identified (yes) and searched, but not identified, (no) by SPME-GM-MS analysis in the solid, liquid and A.C. filter are reported. Table 3: SPME-GC-MS analysis for the solid residues and the A.C. filter Test Solid residue A.C. filter DMC EMC DEC EC DMC EMC DEC EC Test 16 yes yes yes no no yes no no Test 19 yes yes yes no no yes no no Test 20 yes yes yes no no yes no no Table 4: SPME-GC-MS analysis of the liquid residues Liquid residue 1- optanol 3-nonanol 4-nonanol 2-decanol 2-propil, eptanol 2-butil, optanol 1-decanol Test 16 LNA LNA LNA LNA LNA LNA LNA Test 19 yes yes yes yes yes yes yes Test 20 LNA LNA LNA LNA LNA LNA LNA As Test 16 is a free burning test (no extinction), the liquid residue is not available: “LNA”. Due to the extinction mode of the AVD in Test 20, no liquid was found after the fire. AVD agent is composed of 80 – 95 % water and 20 – 5 % exfoliated vermiculite flakes (other ingredients < 1%). When AVD is applied to the burning cell, the water evaporates and the Vermiculite forms a shell around the cell. After the extinction phase, only a dry solid film of Vermiculite remains. In the liquid residue of Test 19 with foam C8 – C10 alcohol molecules are present due to the Foam composition. 4. Conclusions The aim of this work is to examine the impact of foam and AVD agents on the composition of gas, solid and liquid extinction residues. A theoretical calculation model was developed in order to describe the quantitative chemical analysis. This calculation model is based on the ratio of the masses of the i-th gas compound produced and developed by the cells during the fire and the extinction phases in the two tests normalized by the detectable compounds total combustion mass produced (ω°). The use of this model negates the sampling experimental error due to the mixing condition through the hood, but the EA influence on the total mass is lost. Consequentially, only a comparative result is obtained for each compound. The calculation model demonstrated that AVD agent was more effective than foam for reducing EC, HCl, HF emissions, but less effective for CH4 and CO2. In this calculation the non-detection of certain compounds (i.e. H2) by the FTIR analysis does not introduce any error. 479 Nomenclature FG B – gas flowrate from bag , L/min FN2 – Nitrogen flowrate to analysis , L/min FG F – gas flowrate to FTIR , L/min FN2 F – Nitrogen flowrate to FTIR , L/min Φ - flux analysis fraction Ci B – i-th conc. in bag, ppmv mi B – detect. i-th compound mass in gas bag, g mj B – not detect. j-th compound mass in gas bag, g mC B – compounds mass in gas bag, g mC B° – detect. compounds mass in gas bag, g mC BØ – not detect. compounds mass in bag, g Ci B° – i-th conc. in bag referred to detect., %V/V Vi B – detect. i-th compound volume in gas bag, L VG B – sampled gas volume in gas bag, L VC B° – detect. compounds volume in bag, L Ci HOOD – i-th compound concentration in hood, g/g Ci HOOD° – i-th conc. in hood referred to detect., g/g EA – Extinguishing agent action Ci StB – i-th compound conc. sampled to bag, g HSP – Sampling point height, m HGS – Gas source height, m ΔtSAMPLING – Sampling time, sec Ci COMB – i-th conc. in combustion gas, g/g Ci COMB° – i-th conc. in combustion gas referred to detected, g/g mi COMB – detect. i-th compound mass in combustion gas, g mC COMB – mass of compounds in combustion, g mC COMB° – mass of detect. compounds in combustion, g Ω° - Ratio of produced gas referred to detect., g/g ω° - Normalized ratio of produced gas referred to detected., g/g ppmv – volumetric concentration (mL/m3) Acknowledgments This paper is the result of a collaborative work of people with different skills and experience. All the extinguishing experimental tests were made in collaboration with the NIA (Nucleo Investigativo Antincendio) of Rome, which provided very professional operators, instrumentation and facilities. References Andersson P, Blomqvist P., Lorén A. and Larsson F., 2013, Investigation of fire emissions from Li-ion batteries, SP Technical Research Institute of Sweden, SP Report 2013:15. Egelhaaf M., Kress D., Wolpert D., Lange T., Justen R., Wilstermann H., 2013, Fire Fighting of Li-Ion Traction Batteries, SAE International Journal of Alternative Powertrains, Vol. 2, pp. 37-48. Funk E., Flecknoe-Brown K.-W., Wijesekere T., Husted B. P., Andres B., 2023, Fire extinguishment tests of electric vehicles in an open sided enclosure, Fire Safety Journal 141 103920. Golubkov A.W., Fuchs D., Wagner J., Wiltsche H., Stangl C., Fauler G., Voitic G., Thaler A. and Hacker V., 2014, Thermal-runaway experiments on consumer Li-ion batteries with metal-oxide and olivin-type cathodes, The Royal Society of Chemistry, 4, 3633–3642. Jia Z., Wang S., Qin P., Li C., Song L., Cheng Z.., Jin K., Sun J., Wang Q., 2023, Comparative investigation of the thermal runaway and gas venting behaviors of large-format LiFePO4 batteries caused by overcharging and overheating, Journal of Energy Storage 61 106791. Larsson F., Andersson P., Blomqvist P. & Mellander B-E., 2017, Toxic fluoride gas emissions from lithium-ion battery fires, Scientific Reports | 7: 10018. Liu Yujun, Duan Qiangling, Xu Jiajia, Chen Haodong, Lu Wei, Wang Qingsong, 2018, Experimental study on the efficiency of dodecafluoro-2-methylpentan-3-one on suppressing lithium-ion battery fires, Royal Society of Chemistry, 8, 42223–42232. Hynynen J., Willstrand O., Blomqvist P., Quant M., 2023, Investigation of extinguishing water and combustion gases from vehicle fires, RISE Research Institutes of Sweden, Report 2023:22. Palma D., Gentile P., Mazzaro M., Russo P., 2023, Experimental Study on Different Extinguishing Agents for Fire of Lithium Ion Batteries for Electric Mobility, Chemical Engineering Transactions, 104, 43-48 DOI:10.3303/CET23104008 Rao H., Huang Z., Zhang H., Xiao S., 2015, Study of fire tests and fire safety measures on lithium-ion battery used in ships, The 3rd International Conference on Transportation Information and Safety, June 25 – 28. Ribière P., Grugeon S., Morcrette M., Boyanov S., Laruelleand S. Marlair G., 2012, Investigation on the fire- induced hazards of Li-ion battery cells by fire calorimetry, The Royal Society of Chemistry, 5, 5271–5280. Ubaldi S., Di Bari C., De Rosa A., Michele M., Russo P., 2022, Investigation on Effective Fighting Technology for Lib Fire, Chemical Engineering Transactions, 91, 505-510 DOI:10.3303/CET2291085 Ubaldi S., Conti M., Marra F., and Russo P., 2023, Identification of Key Events and Emissions during Thermal Abuse Testing on NCA 18650 Cell, Energies, 16, 3250, doi.org/10.3390/en16073250. 480 https://doi.org/10.3390/en16073250 68palma.pdf Experimental & Theoretical Study on the Solid, Liquid and Gaseous Residues after Fire Extinction of Lithium-Ion Batteries