Earticle

Fires caused by the concentration of solar radiation through transparent objects have been discussed mainly from a theoretical perspective, while experimental verification based on actual residential fire cases remains limited. This study analyzes the ignition mechanism of a residential fire suspected to have resulted from solar radiation concentration through a spherical transparent lighting device. In January 2026, a fire occurred in a detached house in Yongin, Republic of Korea. Fire scene investigation identified localized burn patterns near a window, and no clear evidence of electrical or mechanical failure was found. A spherical transparent lighting device was located adjacent to combustible materials at the presumed point of origin, indicating the possibility of a non-electrical ignition source. To verify this hypothesis, reproduction experiments were conducted under actual sunlight conditions using a spherical transparent lighting device identical to that found at the fire scene, with black Hanji paper used as an indicator combustible material to clearly observe localized thermal response due to its high absorptivity and thermal sensitivity. The results showed that solar radiation transmitted through the spherical transparent lighting device was refracted and concentrated into a localized focal region, resulting in smoke generation and localized surface charring. These findings suggest that spherical transparent lighting devices can act as non-electrical ignition sources under specific solar exposure conditions and highlight the necessity of considering solar radiation concentration as a potential non-electrical ignition factor in fire origin investigations.

This study investigates the thermal runaway behavior of portable lithium-ion batteries and proposes design criteria for fire-resistant containment packs. Since Sony commercialized lithium-ion batteries, energy density has increased, but so has the risk of thermal runaway. Portable batteries, with multiple cells in confined structures, are prone to heat accumulation, pressure rise, and flame ejection. Forced-heating experiments on pouch-type lithium cobalt oxide (LCO) batteries (10,000–40,000 mAh) showed that temperatures reached 200 °C within 4–9 minutes, with swelling and off-gas release. Maximum temperatures reached 814 °C, and thermal runaway lasted up to 38 seconds. Sealed vessel tests indicated rapid pressure rise (>0.8 MPa) and explosion risk upon oxygen ingress. Fire-resistant pack tests using double-layer aramid fiber demonstrated that opposite inlet orientation prevented external flame ejection while allowing off-gas release. FDS-based PyroSim simulations confirmed delayed flame ejection and moderated pressure. Effective packs should withstand ≥814 °C for ≥40 seconds, allow off-gas ventilation, maintain ≥600 N tensile strength, and employ multi-layer structures. This study provides practical safety design guidelines and highlights recycled firefighter protective clothing as a sustainable material.

The rapid advancement of industrial processes has led to an increase in the use of complex hazardous materials with dual property-flammable and oxidizing-, yet comprehensive hazard assessment criteria remain underdeveloped. Unlike widely recognized contact hazards such as hypergolic materials that react readily upon mixing, the dual-property mixtures exhibit latent hazards, remaining stable under ambient or standard operating thermal conditions. However, this “reliability on stability” often becomes a primary cause of critical judgment errors in the process of fire scene analysis. When these mixtures, which are stable at room temperature, are exposed to threshold energy due to external fire sources or equipment system failures, the thermal decomposition of oxidizer initiates. At this point, the fire’s characteristics deviate from conventional combustible fires, entering a stage of dominant hazards where the oxidizer completely dictates the rate and intensity of combustion. While the current “Act on the Safety Control of Dangerous Substances” regulates materials with multiple properties by defining a hierarchy of dominant hazards, clear classification criteria or hazard determination guidelines remain absent for mixtures of flammable liquids and oxidizing substances. Consequently, existing hazard data and information collected from fire scenes are insufficient to explain the unique synergistic interactions of complex hazardous materials which constitutes a decisive obstacle to accurate fire cause analysis and investigation. Specifically, through step-by-step flammability and oxidizing property tests according to the UN GHS international standards and domestic “Act on the Safety Control of Dangerous Substances” regulations, this study aims to quantitatively verify the transition patterns of hazards depending on mixing conditions. Furthermore, this research will extract precise analytical data on s

This study aimed to identify fire safety measures for electric vehicle (EV) fire prevention in underground parking lots and prioritize them through an Importance-Performance Analysis (IPA). A third-round Delphi analysis with 18 experts identified 19 key factors. An IPA analysis with 52 practitioners identified “installation of battery thermal runaway (off-gas) detectors” and “establishment of a dedicated smoke exhaust mode” as the most urgent improvement tasks (Quadrant 2). The results of this study can serve as a foundation for establishing EV fire safety policies and implementing them in practice.

This study explored the possibility of supplementing the existing qualitative analysis method by applying artificial intelligence (AI) technology to the fire investigation field. By applying and analyzing ChatGPT and Circle-to-Search functions to the actual fire case of Nonsan Fire Station, it showed high accuracy and efficiency in tasks such as fire cause prediction, fire pattern classification, and field photo analysis. In particular, automatic analysis of visual evidence through image-based neural network model and automatic summary and classification of reports based on natural language processing suggested the possibility of greatly improving the speed and objectivity of fire investigation. In addition, by constructing the Vision AI prototype model, meaningful results were obtained to identify oil patterns on the floor, and a systematic fire investigation system was prepared by establishing an NSFI shared system. Lastly, the deployment of data collection experts to the fire investigation team was proposed.

This study empirically validates the limitations of Vision AI(GPT-5.2) and the efficacy of Human-in-the-Loop collaboration using 280 real-world fire cases(N=280). While independent AI analysis showed potential discriminative capability (AC1=0.645), it exhibited significantly low reliability (κ=0.296) and distributional heterogeneity (Stuart-Maxwell test, p<.001), driven by “Multimodal Hallucination” and “Confirmation Bias” based on visual patterns. Conversely, incorporating expert contextual information effectively rectified misjudgments and reconstructed the AI’s causal reasoning logic. Consequently, this study demonstrates that AI cannot replace human investigators due to limitations in “Sensory Extension” and “Legal Agency,” suggesting that AI must be established as an “Intelligent Collaborative Tool” operating under the control of human investigators.

The widespread use of capacitors in compact, high-performance electronics has heightened concerns regarding fire risks under thermal stress. This study experimentally investigated the degradation of various capacitor types at elevated temperatures to evaluate their ignition potential. Results showed that stepwise heating induced significant capacitance fluctuations, structural deformation, and dielectric leakage. Microscopic analysis via SEM-EDS confirmed the presence of carbon-rich residues in the leaked materials, which facilitate the formation of conductive paths and promote electrical tracking. Furthermore, the combination of leaked substances and narrow lead spacing compromised insulation reliability, thereby increasing the risk of arc discharges. These findings demonstrate that thermal degradation and dielectric leakage are critical precursors to tracking and ignition, providing substantiated forensic evidence for capacitor-related fire investigations.