Intrinsic Continuous Process Safeguarding
Cornelis Van 't Land is a PhD student in the department Sustainable Process Technology. (Co)Promotors are prof.dr.ing. M.B. Franke; prof.dr. S.R.A. Kersten and prof.dr.ir. A. Nijmeijer from the faculty of Science & Technology (TNW), University of Twente.
Inherent to the large-scale manufacture of certain chemicals is the danger of explosions, fires, and releases of toxic gases. Process safety is the science of preventing accidents in the chemical industry or mitigating their effects. This thesis deals with an approach to process safety and is structured in seven chapters.
Chapter 1. ICPS is continuously present and minimizes the probability of the occurrence of „runaway” reactions resulting from equipment failure or human error. It is compared to Extrinsic Process Safeguarding (EPS) that starts working upon a signal. EPS can therefore not be accepted for primary process protection. A different methodology is Inherently Safer Design (ISD). ICPS comes under the umbrella of ISD. ICPS excludes EPS for primary process protection, ISD does not.
Chapter 2 – Literature Overview - Three serious accidents are treated in Section 2.2. The section starts with a description of the explosion of a gaseous mixture of mainly cyclohexane and air and fires at Flixborough in the U.K. in a caprolactam plant in 1974. Caprolactam is an intermediate for the manufacture of Nylon 6 and Nylon 66. The next incident is the release of toxic dioxin at Seveso in Italy in 1976. Dioxin was released at the manufacture of a material for the production of herbicides and antiseptics. The section is concluded by a description of the release of toxic methyl isocyanate (MIC) at Bhopal in India in 1984. MIC was produced as an intermediate for the manufacture of an insecticide. Section 2.3 describes explosions and a fire at a refinery at Texas City in the U.S.A. in 2005. That incident is highlighted because, due to an unsatisfactory design, operators were seduced to disobey instructions concerning the bottom liquid level in a distillation column with catastrophic consequences.
Section 2.4 treats the fact that improvements concerning the safe handling and use of materials had already been achieved before the introduction of ICPS and ISD. For instance, the captive use of chlorine as an alternative for transport to other locations had been recommended.
An incident regarding a nuclear reactor at Harrisburg (Three Mile Island, TMI) in the U.S.A. in 1979 is discussed in Section 2.5. The heart of the matter is that the heat development of a nuclear reactor cannot immediately be fully stopped when the reactor is turned down. At Harrisburg, the remaining heat development could only be dealt with by systems that needed activation. Those systems failed and a serious accident could just be avoided. The incident is described because it illustrates that EPS is unsuitable for primary process protection.
Section 2.6 concerns methods. ISD is discussed in Section 2.6.1. Possibly, a quick way to characterise ISD is by one-liners like: think twice, go the extra mile, and think about what can happen. Hazop and Hazan are discussed in Section 2.6.2. Hazop stands for Hazard and Operability Study and Hazan for Hazard Analysis. Generally, Hazop and Hazan are preceded by ISD and ICPS. Hazop is a systematic technique for a team to discuss possible hazards and operating problems. Hazan follows and it addresses the following questions: how often does a hazard occur, how serious are the consequences, and is it necessary to take actions. A quantitative method to estimate how often a protective system fails is treated. Section 2.6.2 is concluded by a discussion of the concept of the Fatal Accident Rate (FAR).
Chapter 3 contains the original article regarding ICPS, published in 1985. The basis was experience obtained from the manufacture of organic peroxides. Process safety receives attention in this chapter. Reaction calorimetry to measure the isothermal and adiabatic heat developments of chemical reactions is mentioned.
Chapter 4 is a reprint of Chapter 8 in the book Safety in Design. Five cases are discussed in this chapter. The first case concerns an explosion and a subsequent fire at a batch reaction in a chemical plant. The batch reaction turned, because of equipment failure, into a „runaway” reaction when the reactor cooling failed. There was neither an alternative nor a back-up for the cooling that failed. The second case regards a batch reaction that started, due to a human mistake, at a too high temperature. At that temperature, the reaction could no longer be kept under control with the existing equipment. An explosion resulted. The reaction was modified into a semi-batch reaction. The third case concerns a „runaway” reaction causing two explosions during the activation step of a catalyst in packed beds in a petrochemical plant. The type of the catalyst had been replaced by a different type. The new type of catalyst reacted violently with a chemical used at the activation step. The company had wrongly considered the new catalyst to be a „drop-in” catalyst. The activation step was modified. The fourth event regarded an explosion in a furnace of a continuous chemical plant during a start-up. Instructions to remove a mixture of natural gas and nitrogen from a cleaned filter were not followed because that was by the operators considered to be a roundabout and time-consuming way. The purge of gaseous mixtures from the cleaned filter was modified. The fifth incident concerned an explosion in a double-coned contact dryer. A batch had been left in the dryer at a temperature of 120-130 oC and self-heating occurred. The practice of storing a warm product in process equipment was stopped.
Chapter 5 is a reprint of Chapter 14 of the book Drying in the Process Industry (2012). Section 5.2 treats safeguarding of the drying operation. Temperature elevation, often in combination with the presence of air, may cause an explosion, a fire, or a release. Tests concerning the possibility of a fire are discussed. The phenomenon that a large sample has a lower self-ignition temperature in air than a small sample is treated. Dust explosions is the next subject. The phenomenon that the rate of pressure rise caused by a dust explosion in large equipment is smaller than in small equipment is discussed. Prevention (by, for instance, using nitrogen as a drying gas) and curing (by, for instance, relief venting or suppression) are dealt with. The options regarding the prevention of fires and dust explosions for the four important convective industrial dryer types (fluid bed, rotary, flash, and spray dryer) are treated. The section is concluded by mentioning contact dryers. The heat needed for evaporation is supplied by indirect heating in those dryers which makes safeguarding simpler than at convective drying.
Chapter 6 is a reprint of an article in the journal ACS Chemical Health & Safety (2024). The batch laboratory reaction at which the concept of ICPS was developed is discussed. The descriptions of two serious accidents with chemical reactors that occurred between 2000 and 2020 follow. The first accident concerns an explosion and fire caused by an inadvertent release of subsequently flashing vinyl chloride monomer (VCM) in open floor drains at the batchwise manufacture of polyvinyl chloride (PVC). Red lights on reactors indicating pressure in the reactors and avoiding open floor drains could have prevented the accident or mitigated its consequences. The second accident regards an explosion and fire caused by a too high jacket temperature of a batch reactor in the pharmaceutical industry. The manufacturer had studied product safety and had not checked process safety. Such a check is a cornerstone of ICPS. ICPS would recommend to apply Accelerating Rate Calorimetry (ARC) to actual reaction mixtures to determine a safe reactor jacket temperature. ARC was already recommended by the investigating body.
Chapter 7 contains seven Conclusions. In Conclusion No. 1, the statement is made that the application of ICPS could have prevented or mitigated the consequences of three serious accidents in the chemical industry between 2000 and 2020. It is mentioned in Conclusion No. 2 that the principles of ICPS can be applied to various fields of society. It is stated in Conclusion No. 3 that, generally, the quality of the reports of the Chemical Safety and Hazard Investigation Board (CSB, U.S.A.) concerning accidents with chemical reactors between 2000 and 2020 is good. However, the quality of a few reports could be better. Conclusion No. 4 mentions that the chemical and chemical engineering input in the Board of CSB is at present insufficient. Conclusions Nos. 5 and 6 treat the facts that thestructure of CSB is relatively strong and the structure of eMARS (European Major AccidentReporting System) is less pronounced. Conclusion No. 7 mentions the past existence of the Japanese Failure Knowledge Database. Conclusion No. 8 states that it was possible for the author of this dissertation to apply the principles of Intrinsic Continuous Process Safeguarding in several cases within Akzo Nobel in the period 1985-2000.
The Conclusions are followed by an Outlook. Much attention has been paid to the possibility of improving the process safety of chemical plants in the fourth quarter of the previous century. The text of this thesis proves that it is still possible to further improve the process safety of chemical reactors by applying the principles of ISD and ICPS.
