Mixing hazard evaluation of organic peroxides with other chemicals

Organic peroxides (POs) have been widely used in chemical industries as initiators of polymerization, hardening or bridge formation agents, and they are known for its self-reactive and also mixing hazard characteristics when mixed with other chemicals such as acids and alkalis. It is the purpose of this investigation to propose a simple but useful evaluation flow of mixing hazards of POs with other chemicals using conventional experimental techniques such as a glass test tube test, Dewar vessel test and DSC is proposed. 7 kinds of POs (DEPD, THP, TBEH, TBTC, MEKPO, DTBP, THHP) were mixed with sulfuric acids with various concentration, sodium hydroxide solutions, -iron(III) oxide and acrylonitrile (AN).

Based on the proposed evaluation flow the testing results were classified into 4 ranks due to the hazard criteria. Futhermore DEPD/acrylonitrile mixtures were investigated in more detail and the influences of the mixing ratio and the stirring speed were discussed.     

Risk reduction concept to provide design criteria for Emergency Systems for onshore LNG plants

The functional safety requirement is widely applied in the process plant industry in accordance with the international standards, such as IEC and ISA. The requirement is defined as safety integrity level (SIL) based on the risk reduction concept for protection layers, from original process risk to tolerable risk level. Although the standards specify both, the Prevention System and the Emergency System, as level of protection layers, the standards specify in detail only the use of the Prevention System (i.e., Safety Instrumented System (SIS)). The safety integrity level is not commonly allocated to the Emergency System (e.g., Fire and Gas System, Emergency Shutdown System and Emergency Depressuring System). This is because the required risk reduction can be normally achieved by only the Prevention System (i.e., SIS and Pressure Safety Valve (PSV)). Further, the risk reduction level for the Emergency System is very difficult to be quantified by the actual SIL application (i.e., evaluated based on the single accident scenario, such as an accident from process control deviation), since the escalation scenarios after Loss of Containment (LOC) greatly vary depending on the plant design and equipment. Consequently, there are no clear criteria for evaluating the Emergency System design. This paper aims to provide the functional safety requirement (i.e., required risk reduction level based on IEC 61508 and 61511) as design criteria for the Emergency System.In order to provide clear criteria for the Emergency System evaluation, a risk reduction concept integrated with public’s perception of acceptable risk criteria is proposed and is applied to identify the required safety integrity level for the Emergency System design. Further, to verify the safety integrity levels for the Emergency Systems, the probabilistic model of the Emergency Systems was established considering each Emergency System (e.g., Fire and Gas System, Emergency Shutdown System and Emergency Depressuring System) relation as the Overall Emergency System. This is because the Overall Emergency System can achieve its goal by the combined action of each individual system, including inherent safe design, such as separation distance.The proposed approach applicability was verified by conducting a case study using actual onshore Liquefied Natural Gas Plant data. Further, the design criteria for Emergency Systems for LNG plants are also evaluated by sensitivity analysis.     

Influence of physical properties of ammonium nitrate on the detonation behaviour of ANFO

In order to obtain a better understanding of the non-ideal detonation behaviour of ammonium nitrate based explosives, detonation velocities of ANFO (ammonium nitrate and fuel oil) prepared with different kinds of ammonium nitrate (AN) were measured in steel tubes. In this series of test six kinds of AN were used and the influence of the pore diameter, the pore volume and the particle diameter of the AN particle on the detonation velocity of ANFO was investigated.

It was found that the pore diameter and the pore volume had a strong influence on the detonation velocities of ANFO. In the case of ANFO samples which were prepared with AN that had the same pore diameter and the pore volume, when tested the highest detonation velocity (3.85 km/s) was observed when the smallest particle diameter (<0.85 mm) was used. This value corresponded to 75% of the ideal detonation velocity, which was theoretically predicted by the CHEETAH code with the JCZ3-EOS.

The 12 months aging showed the change of the detonation velocities of ANFO and the reaction of ANFO was influenced both by the physical and the chemical properties of AN particles and oil during the storage period.     

Behavior of flames propagating through lycopodium dust clouds in a vertical duct

The structure of flame propagating through lycopodium dust clouds has been investigated experimentally. Upward propagating laminar flames in a vertical duct of 1800 mm height and 150×150 mm square cross-section are observed, and the leading flame front is also visualized using by a high-speed video camera. Although the dust concentration decreases slightly along the height of duct, the leading flame edge propagates upwards at a constant velocity. The maximum upward propagating velocity is 0.50 m/s at a dust concentration of 170 g/m³. Behind the upward propagating flame, some downward propagating flames are also observed. Despite the employment of nearly equal sized particles and its good dispersability and flowability, the reaction zone in lycopodium particles cloud shows the double flame structure in which isolated individual burning particles (0.5–1.0 mm in diameter) and the ball-shaped flames (2–4 mm in diameter; the combustion time of 4–6 ms) surrounding several particles are included. The ball-shaped flame appears as a faint flame in which several luminous spots are distributed, and then it turns into a luminous flame before disappearance. In order to distinguish these ball-shaped flames from others with some exceptions for merged flames, they are defined as independent flames in this study. The flame thickness in a lycopodium dust flame is observed to be 20 mm, about several orders of magnitude higher than that of a premixed gaseous flame. From the microscopic visualization, it was found that the flame front propagating through lycopodium particles is discontinuous and not smooth.     

The generative power of metaphor: long‐term action research on disaster recovery in a small Japanese village

Disaster recovery is a dynamic process of creating, maintaining, and changing the meaningful context of survivors. It is completed when they redevelop their self‐reliance and resume managing their social relations with a sense of community. This study employed action research to examine how researchers and survivors collaborated to change disaster recovery through the generative power of metaphor in a small village in Japan that experienced the Niigata–Chuetsu earthquake on 23 October 2004. It outlines long‐term collaborative practices as survivors undertook new activities owing to the power of the metaphor of ‘school’. Once ‘school’ was adopted as the metaphor for where survivors learnt new skills and passed on traditional knowledge, they created new metaphors and performed new activities independently, which is critical for recovery as it demonstrates self‐reliance. The paper assesses the reasons why generative metaphors worked effectively in this case and highlights some academic and practical implications for disaster recovery.     

Investigation of an accidental explosion due to unintended mixing involving reactive chemicals at a waste storage tank

An accidental explosion occurred in a waste storage tank at an incineration plant in Kawasaki, Japan, on May 11, 1997. The accident was caused primarily by unintended mixing involving reactive chemicals, such as organic peroxides (POs) and acrylonitrile (AN). The PO initiated polymerization of AN and the heat released during the polymerization led to a runaway reaction and explosion. POs are widely used in the chemical industry and can be self-reactive and hazardous when mixed with other chemicals such as acids and alkalis. The goal of this study was to obtain a better understanding of the mixing hazard of chemicals through an evaluation of POs with other chemicals using conventional experimental techniques such as glass test-tube tests, Dewar vessel tests, and differential scanning calorimetry (DSC). Seven types of POs were mixed with AN. Test results were classified into four ranks based on the hazard criteria. In addition, di(2-ethylhexyl)-peroxydicarbonate/AN mixtures were investigated in detail and the influences of the mixing ratio and the stirring rate were examined.     

Approach enhancing inherent safety application in onshore LNG plant design

This study aims to provide the approach for inherent safety design of onshore LNG plants to be applied at the very early stages (concept definition phase) of the project development. Onshore LNG plant development project starts from the “Concept Definition” phase, where financial feasibility is estimated and major conditions, such as site location and plant foot print, are set.The inherent safety design basic criteria and design measures should be identified and selected when setting the basic conditions during the Concept Definition phase of the project development, such as the site location (relative location from populated areas), site condition (prevailing wind direction) and plant production capacity (number of process train, number of product tanks). The safety measures, which are usually not fully developed at the project early stages in the current design execution practices, are the emergency systems, which mitigate an accident escalation, the modularized plant and layout, and the tank selection.The inherent safety design measures discusses in this paper were identified based on the categories of plot plan, emergency system, and module plant application.The proposed approach will contribute to improve inherent safety design of onshore LNG plants and it will also yield schedule and cost benefits.     

Decomposing deflagration properties of acetylene under low temperatures

In this study, the decomposing deflagration properties of acetylene under temperatures down to −60 °C and pressures up to 0.2 MPa in a 1-L cylindrical closed vessel were experimentally investigated. The gases were ignited by an electric spark at the center of the vessel. The lower-limit pressures of decomposing deflagration by electric spark ignition were determined. The lower-limit pressure at 10 °C was 0.15 MPa, and it gradually increased with decreasing temperature. The lower-limit pressure at −60^oC was 0.18 MPa. The flame propagation properties, such as the pressure, were measured with pressure transducers mounted along the vessel. The maximum decomposing deflagration pressures and pressure rising rates also increased with decreasing temperature.     

Prediction of the relief of gas explosion in a tubular vessel by venting using a mathematical model

The relief of a gas explosion in a tubular vessel by venting can be predicted by using a mathematical model. In this model, the flame acceleration is represented by an increase in the burning velocity. The movement of a vent cover can be included. The model assumes that the vent is blocked by the vent cover prior to the explosion. the venting ratio was the most influential parameter in terms of relieving the pressure. In the case of a large venting ratio, the flame acceleration made a highly significant contribution, whereas for small venting ratios, the weight of the vent cover contributed to the relief more than the flame acceleration. When the pressure is required to be reduced significantly, the venting ratio, the vent open pressure and the weight of the vent cover must all be reduced.