Abstract:

At present, the reaction safety risk assessment technology in China's fine chemical industry mainly uses the Stoessel criticality classification method and ranks the reaction process hazard levels from 1 — 5 class. This method is based on the Gygax cooling failure model, which evaluates the reactivity hazard level of reaction systems under the assumption of a cooling failure condition. However, statistics on chemical reactivity safety incidents, both domestically and internationally, indicate that the cooling failure condition only accounts for a very small portion of the causes of such incidents (about 3%). Taking nitration processes as an example, the literature reports using the Stoessel criticality classification method indicate that most nitration reaction processes are classified as low risk levels. This contradicts with the fact that based on actual production incidents, nitration process is the dangerous process with the highest fatalities in China. The primary reason for this is that the Stoessel method only assesses the cooling failure scenarios in batch reaction vessels while it overlooks many other complex incident causes. This indicates that this classification method cannot accurately assess the reactivity hazards in actual chemical production, because it fails to cover most of the causes of chemical reactivity safety incidents and to propose effective accident control measures. On the other hand, the full-process reactivity hazard analysis (RHA) is based on assessing the accident chain of a chemical process from identifying the root causes of abnormal conditions to simulating the occurrence of reaction runaways, proposing corresponding protective layers, and thereby determining specific accident prevention measures. This approach is more suitable for enhancing actual process safety compared to the Stoessel criticality classification method. Therefore, it is recommended to conduct a comprehensive RHA analysis for the reactivity risk assessment of actual chemical plants, especially for hazardous processes rated at levels 4 and 5 high risk.

Key words: Stoessel criticality classification, cooling failure scenario, full-process reactivity hazard analysis, safety, reaction, chemical processes

摘要：

目前我国精细化工的反应安全风险评估技术主要是使用Stoessel反应危害分级方法而得出1~5级的工艺危险度等级，本文提出该方法是基于Gygax冷却失效模型建立的，即评估冷却失效工况下反应体系的工艺危险度等级。而国内外的化工反应安全事故统计结果表明，冷却失效工况仅是反应安全事故起因的很小部分（约3%）。以硝化工艺为例，文献报道的基于冷却失效的反应危害分级方法得出大部分硝化反应工艺处于低风险，这与实际生产中硝化反应事故数与死亡数占比最多存在矛盾。主要原因是由于该方法只评估间歇反应釜的冷却失效工况，而忽略其他众多复杂的事故起因。这表明该分级方法无法准确评估出实际工艺生产中的反应危害，无法较全面地涵盖化工反应安全事故起因并提出有效的事故管控措施，具有一定局限性。而化工过程全流程的反应危害分析（RHA）是根据从异常工况起因至最后反应事故发生的事故链以提出相应保护层，进而确定具体事故防范措施。相比Stoessel反应危害分级方法更适合实际生产工艺安全的提升。因此，建议对化工过程的反应安全风险评估作全流程RHA分析，特别是危险工艺和4~5级的高风险工艺。

关键词: Stoessel反应危害分级, 冷却失效工况, 全流程反应危害分析, 安全, 反应, 化学过程