Abstract Hydrogen‐blended natural gas (HBNG) is an effective transitional solution for large‐scale hydrogen energy transportation, whose safety requires quantitative assessment from a practical perspective. This study constructed a 60 L cylindrical deflagration experimental setup, conducting deflagration tests on 30% HBNG at an equivalence ratio of 1 under initial pressures of 1, 1.2, 1.5, and 2 atm, with a control test of 80% HBNG at 1 atm. High‐speed schlieren imaging and dynamic pressure monitoring were adopted to reveal how different conditions affect the HBNG deflagration characteristics. Results show that raising the initial pressure from 1 to 2 atm increased the maximum deflagration overpressure from 0.947 to 2.091 MPa and the maximum pressure rise rate from 5.66 to 17.7 MPa/s, along with faster flame propagation. Initial pressure has a more significant effect on HBNG maximum explosion pressure, while hydrogen blending ratio has a more pronounced impact on flame development. The peak pressure of 30% HBNG at 2 atm is 44.2% higher than that of 80% HBNG at 1 atm, with a 90.32% longer time to reach peak pressure. These findings provide critical data for pressure specifications in the safety design of transportation pipelines and differential risk assessment for hydrogen‐blending scenarios.
With the surging demand for ethane as a petrochemical feedstock, high-pressure liquid ethane pipeline networks are expanding rapidly. However, under complex conditions involving rapid flash vaporization, the interplay among transportation pressure, leak aperture, and leak direction on accident impact ranges remains inadequately quantified. This study employs FLACS software to assess the risks of accidents following leakage and dispersion. The findings indicate that, when considering flash vaporization and turbulent entrainment, a pressure increase expands the 1.5 kW/m2 thermal radiation contour by less than 3%, highlighting a significant sublinear trend. In contrast, increasing the aperture from 5 to 25 mm raises the thermal radiation impact ranges by 280%, underscoring leak geometry as the primary factor in risk amplification. Vertical releases reduce the thermal radiation impact range by approximately 21% compared to horizontal leaks, while a 2.6 m/s (6 MPH) headwind has a negligible mitigating effect. These parameters collectively define a multidimensional hazard boundary that single-factor correlation equations cannot capture. By quantifying these nonlinear coupling effects, this study provides a framework for designing safety distances and preventing leaks in ethane pipelines, thereby establishing a foundation for risk assessment in long-distance ethane transmission pipelines.
Abstract This paper describes a new set of 45 laboratory experiments where about 180 g of ammonia was released from pressurized liquefied storage in a still 51.3 m 3 chamber (i.e., a small room) and the variations of the ammonia gas concentrations and liquid deposition over a 60 min time period were investigated. Chamber temperatures ranged from −18 to 40°C, and relative humidities from 11% to 90%. A liquid collection (catch) tray was placed so that the 45° downwards‐directed two‐phase jet would strike its center, and a scale under the tray measured the liquid mass. In three trials, the tray was inverted; in six trials, a water spray system was operated; and, in four trials, CO 2 gas was introduced. The initial ammonia deposition to the tray ranged from about 10%–50% of the released mass, with the larger fractions occurring during colder conditions. The strong ammonia jet generated eddies lasting for many minutes. The addition of CO 2 gas produced only minor differences in the concentrations. The addition of water spray resulted in about a 50% reduction in ammonia vapor. These results should be helpful in planning emergency response to ammonia accidents in confined locations.
Chemical incompatibility management in process safety depends on accurate chemical reactive group identification. When chemicals are not covered by existing databases, classification often relies on expert judgment and may be inconsistent. This study was conducted based on 68 commonly used petrochemical products and the reactive group list embedded in the Chemical Reactivity Worksheet (CRW) database. Using this reference framework, the accuracy of six large language models (LLM) in identifying chemical reactive groups was systematically evaluated. The tests used repeated trials and two evaluation criteria: correct number of reactive groups and correct category assignment. The conclusion is that performance is only moderate; the models frequently underestimate or misclassify reactive groups, and outputs are not stable enough for safety-critical use.
TBP/kerosene mixtures are vital extraction solvents in nuclear fuel reprocessing, yet their explosion characteristics under coupled high temperature and sub-atmospheric pressure are poorly understood. This study experimentally investigated the flammability limits and explosion intensities of these vapor mixtures at 60 degrees C and 120 degrees C (75-100 kPa, 0.75 bar-1.0 bar). Results show that reducing initial pressure significantly suppresses explosion intensity and narrows the flammable range. At 120 degrees C, the explosive range (1.0%-11.0% at 100 kPa) converges to a singular critical point (3.5% at 75 kPa); when at 60 degrees C, the critical pressure threshold rises to 95 kPa. The maximum explosion pressure consistently occurred at a stoichiometric concentration of 7.0%. Sensitivity analysis indicates the upper flammability limit is more pressure-sensitive than the lower flammability limit at 120 degrees C, driven by reduced molecular collision frequency and oxygen limitation in fuel-rich zones. These findings provide essential thermodynamic data for safety design and risk assessment in nuclear chemical facilities under extreme conditions.
In the oil and gas industry, quantitative risk assessment (QRA) often relies on generic failure-rate databases, even though facilities operate with different levels of asset integrity and process safety maturity. These differences create operational "fingerprints," so safeguards that perform well at one site may fail at another-leaving the question unanswered: "Why do accidents still happen?" This paper proposes a risk amplification factor (RAF) that bridges organizational reality and hard QRA frequencies using a "No size fits all" adjustment. RAF integrates systematically four auditable organizational factors-maintenance and inspection, safety culture, emergency response, and HSE management system robustness-mapped to both preventive and mitigative bowtie barriers. Factor influence is derived from 20 major accident hazard cases through frequency-severity weighting and normalized into exponents, yielding a transparent, multiplicative structure that is stable for decision-making. The resulting RAF is calibrated to remain bounded (approximately 0.6-3.0), enabling consistent comparison across sites without destabilizing baseline calculations. A concise checklist converts documented practices (backlog control, reporting and training, drills and equipment readiness, and action closure) into 1-5 scores. By capturing escalation sensitivity-where "the first five minutes are the most critical"-RAF helps practitioners prioritize improvements and align risk controls with organizational capacity.
Traditional risk assessment methodologies for major industrial hazards predominantly rely on the binary combination of probability and severity. This paper proposes a comprehensive multi-criteria risk assessment methodology integrating six key indicators: occurrence probability, hazardous phenomenon intensity, phenomenon kinetics, target exposure, target vulnerability, and non-detectability. Each indicator is evaluated on a standardized 1-10 scale, enabling the integration of both qualitative and quantitative data. The methodology incorporates the ALARP principle through three risk tolerance zones (Acceptable, ALARP, Unacceptable). The methodology is applied to a BLEVE (Boiling Liquid Expanding Vapor Explosion) scenario of an LPG storage sphere at the Hassi R'Mel industrial complex in Algeria. The results demonstrate that the proposed approach provides more nuanced risk characterization than traditional methods, with the radar chart visualization offering intuitive comparison of risk profiles. A 16.7% reduction in risk was achieved through inventory limitation, moving the scenario from Unacceptable to the ALARP zone, thereby validating the methodology's practical utility for industrial risk management.
A commonly applied "rule of thumb" for combustible dust is that a light bulb would be nearly obscured by a 6-foot-thick cloud of dust if the dust density was near that required to be explosible (i.e., the minimum explosible concentration [MEC]). This heuristic is based on Eckhoff's normative text on this topic: "a cloud of 40 g/m3 of coal dust in air is so dense that a glowing 25-W lightbulb can hardly be seen through a dust cloud 2 m thick." The coal dust cloud described by Eckhoff would be close to the MEC. This heuristic is well known, and the author has used it repeatedly to illustrate how dense a dust cloud must be to be explosible. This paper demonstrates the heuristic is correct by evaluating light transmission fractions for combustible dust clouds over a range of densities and thicknesses. Large-scale open-air tests are presented to illustrate the appearance of an explosible dust cloud. It is also shown that horizontal surface dust deposition rates at cloud densities well below the MEC would create an intolerable housekeeping burden, and that the highest density tolerable from a worker respiratory protection standpoint is over 3 orders of magnitude less than the MEC.
The long-term storage of foam extinguishing agent may lead to metal corrosion, which poses substantial risks to the safety and operational reliability of fire suppression systems. This study systematically investigated the performance evolution of foam extinguishing agent interacting with aluminum alloys through accelerated aging tests under elevated temperatures. The results showed that the aluminum alloy surface underwent significant corrosion. Meanwhile, the expansion ratio of the foam decreased by 37.46% and 12.84%, accompanied by precipitation and gelation. Molecular dynamics simulation revealed the mechanism by which aluminum ions coordinate and bridge with xanthan gum through electrostatic interactions, leading to foam concentrate changes. Fire suppression tests demonstrated that prolonged contact between the foam extinguishing agent and metal materials leads to performance deterioration. Moreover, the aged extinguishing agent exhibited poor miscibility with water during discharge, thereby hindering the effective progression of the fire suppression process. The findings highlight the critical importance of compatibility between foam agent and metallic material in process safety management. These results provide a scientific basis for the full life-cycle management of foam extinguishing agent and the maintenance of system reliability.
Differential scanning calorimetry (DSC) measurements of chemical thermal stability are highly sensitive to test conditions, which can compromise data accuracy. In accordance with ASTM E537-24, this study systematically evaluated the effects of crucible types and headspace gases (nitrogen vs. air) on DSC results, and elucidated the underlying mechanisms responsible for these effects. The results show that non-sealed and pin-holed crucibles lead to sample loss and distorted heat-flow data, while air atmospheres introduce oxidative interference. High-pressure gold-plated crucibles meet the core requirements of ASTM E537-24; however, their relatively large headspace and specific surface area of sample promote evaporation. To ensure data reliability, the ratio of total internal volume to sample volume must be kept at 5 or less, which, for certain energetic compounds, increases the risk of container rupture during the test. To overcome these limitations, a flame-sealed glass capillary crucible (5 mu L volume) was introduced, enabling precise and safe testing with minimal sample quantities. Validation using 2,4,6-TNT confirmed that this approach achieves results in excellent agreement with theoretical values, and the method enhances both accuracy and safety. The study provides experimental guidance for optimizing DSC test conditions and improving the fidelity of thermal-stability evaluations under ASTM E537-24.
Within the complex environment of university laboratories that integrate teaching and research, the direct cause of accidents in university laboratories is often manifested in human error, but their root causes often reside in organizational defects. To develop targeted prevention strategies, this study applies the 24Model, an accident causation model, to examine 35 Chinese laboratory safety accidents, aiming to systematically analyze the causal relationship between accidents and various influencing factors. By categorizing the influencing factors into immediate, indirect, radical, and root causes, this study statistically identifies the specific manifestations and frequencies of unsafe factors revealing that cultivating a good safety culture and building a robust management system are key to ensuring laboratory safety at the root. Then, a series of systematic prevention strategies is proposed, providing practical guidance for laboratory safety managers to implement root cause governance and enhance overall safety at the organizational level.
The safe operation of liquefied petroleum gas (LPG) storage systems is critical due to the potential hazards. This paper develops a comprehensive risk assessment framework to improve the safety evaluation of LPG storage systems in Algeria. First, hazard and operability study (HAZOP) identifies possible hazards and consequences, which are then assigned as top events in the fuzzy fault tree analysis (FFTA) in order to address the inherent uncertainties. Subsequently, event tree analysis (ETA) evaluates different accident scenarios and consequence pathways. Finally, the results are transformed into a failure mode, effects, and criticality analysis (FMECA) framework, where basic events are assessed to prioritize risks and recommend mitigation measures. By combining HAZOP's systematic hazard identification, FFTA's uncertainty handling, ETA's consequence mapping, and FMECA's risk prioritization, the proposed framework offers a holistic approach that is both robust and adaptable. The proposed methodology is applied to an Algerian LPG storage facility, demonstrating its effectiveness in identifying high-risk and providing actionable safety improvements. A comparison with a standard FTA/ETA methodology is applied to validate the approach.
Management of change (MOC) is a critical process safety control used to govern engineering and organizational modifications in safety-critical systems; however, despite widespread procedural adoption, it often underperforms in practice. This study identifies the organizational and managerial mechanisms that erode the practical authority of MOC and provides insight into strengthening change governance. A qualitative explanatory study was conducted in a safety-critical gas distribution organization. Twenty-two semi-structured interviews with managers, supervisors, and HSE and technical experts were analyzed using inductive content analysis, followed by realist synthesis based on context-mechanism-outcome configurations. Findings show that MOC degradation is primarily driven by governance erosion rather than isolated procedural gaps. Organizational conditions triggered mechanisms such as managerial disengagement, discretionary enforcement, formalistic compliance, and risk-signal dilution, leading to outcomes including informal change implementation, bypassed HSE review, superficial documentation, and incomplete closure. Training and documentation gaps were identified as downstream consequences of weakened decision authority. Overall, MOC underperformance is best understood as a gradual loss of practical authority rather than a simple compliance issue. Improving effectiveness requires embedding MOC as a mandatory approval gate within managerial decision-making and reinforcing accountability and consistent enforcement.
As the energy industry is exploring new opportunities in the blue ammonia productions, there is the need to increase awareness of the opportunity to incorporate inherently safer design (ISD) considerations particularly in the front end of the project development where there is the greatest opportunity to implement ISD measures. The hazards associated with each of these process modules may not always be well recognized. As a result, the pre-FEED (front-end engineering design) phase may miss the greatest opportunity to introduce corresponding ISD measures that will greatly reduce the required engineering barriers in the later project phases. The aim of this paper is to illustrate a risk-based approach to this process, starting with hazard identification, followed by ISD goal identification and implementation. This paper will also highlight key lessons learned, including important ISD measures that may be missed by typical engineering contractors, and the key focus areas that process safety engineers should address in pre-FEED to mitigate process safety risks in a typical integrated blue ammonia production facility.
Delayed ignition of a hydrogen jet release from a vent or accidental release can lead to a jet explosion with blast and thermal hazards. Such hazards should be accounted for in vent stack design and separation distances and when performing an overall facility risk assessment. The authors, with others, have previously presented a model for predicting the jet explosion blast overpressure for high-pressure choked hydrogen releases at warm temperatures. The jet explosion phenomenon was subsequently observed during lower pressure non-choked releases, with an initial fireball that extended downwards below the release elevation. To investigate the non-choked jet explosion and fireball phenomenon, Air Products and BakerRisk conducted more than 20 delayed ignition hydrogen jet release tests over a wide range of release velocities and temperatures. Moreover, new engineering models were developed to predict blast overpressures and initial fireball dimensions due to a hydrogen jet explosion. This paper presents the new test data, a comparison with computational fluid dynamics (CFD) predictions, and the new models for blast overpressure and fireball predictions.
On the night of December 2-3, 1984, water contamination of a tank of methyl isocyanate in Bhopal, India initiated a series of events that led to a catastrophic toxic release, killing more than 20,000 residents and injuring more than 200,000. Immediately after, leaders from the chemical industry asked AIChE to lead a collaborative effort to eliminate catastrophic process incidents by advancing the state-of-the-art in technology and management practices, serving as the premier resource for information on process safety, supporting process safety in engineering, and promoting process safety as a key industry value. In the spring of 1985, CCPS was founded. Since that time, CCPS has incorporated experience-based learning from major and minor accidents and near-misses into a world-class collection of guideline documents, resource books, and online content. This paper will reflect on 40 years of CCPS's accomplishments.
Fuming nitric acid and fuming sulfuric acid, as critical industrial chemicals, can exhibit high reactivity with metals, posing significant thermal safety risks. This study systematically evaluated the thermal stability of fuming nitric acid, fuming sulfuric acid, and their mixed acid, alongside their chemical compatibility with common structural metals. Utilizing differential scanning calorimetry with custom glass crucibles effectively prevented side reactions with the container, allowing for precise characterization of thermal behavior. Findings revealed that while none of the three acids exhibited exothermic decomposition, their thermal stability varied. Metal compatibility results were distinct: gold was inert across all systems; titanium demonstrated good compatibility only with fuming nitric acid, reacting exothermically with the others; both aluminum and copper triggered vigorous exothermic reactions in all acids, indicating a high thermal runaway risk-the copper/mixed acid system further displayed autocatalytic behavior; the stainless steels showed complex, composition-dependent reactivity.
Trichloroisocyanuric acid (TCCA) is a widely used novel and highly efficient disinfectant, bactericide, and bleaching agent. However, numerous accidents have occurred during the production, storage, and usage of TCCA, and its thermal stability has not been systematically studied. This study investigates the thermal stability of TCCA in a closed environment using dynamic differential scanning calorimetry (DSC), adiabatic accelerating rate calorimetry (ARC), and a combined testing method; the Friedman method was employed to conduct two decomposition kinetics studies based on DSC test data, as well as a combination of DSC and ARC test data. The results show that TCCA is thermally sensitive, with an onset decomposition temperature of 237.81 degrees C and a decomposition heat release of 940 J/g. The decomposition process is strongly exothermic, and loss of control may lead to severe consequences. The kinetic parameters obtained from the two decomposition kinetic studies are highly consistent. The starting temperature corresponding to a time to maximum rate (TMR) of 24 h for the thermal decomposition of TCCA, T-D24, is 216.1 degrees C, which provides important technical and data support for determining safety limits in the production, storage, and use of TCCA.
Outsourced maintenance in South Korea's chemical industry exposes subcontractor workers to complex risks such as fires, explosions, and chemical exposure, as existing safety standards often fail as practical control systems. This study and pilot application addresses these gaps through a case study of a manufacturing company, implementing an improved operational framework centered on a collaborative, job safety analysis (JSA)-based risk assessment process. This framework institutionalizes a joint review process between the primary contractor and subcontractors and is supported by a digital platform that integrates JSA results with work permits and toolbox meetings in real-time. Following its implementation, the new system yielded substantial improvements in safety performance: the number of lost-time accidents decreased from 19 to 5, and the Lost Time Injury Rate (LTIR) fell dramatically from 0.289 to 0.04 within a year. The findings demonstrate that integrating a collaborative assessment framework with a digital system creates a practical, effective safety management model. This study provides an empirically validated framework for enhancing safety in outsourced operations within the chemical sector and other high-hazard industries.
