Hazardous Reaction Mixtures

The potential for chemicals to interact in a violent and uncontrolled manner should be foremost in the mind of everyone concerned with the planning and execution of chemical operations. Not only can syntheses and purifications go disastrously wrong if the elementary principles of chemistry are overlooked, but the inadequate storage of incompatible chemicals has led to many a gutted and blackened factory, warehouse and laboratory. 

Luckily for the laboratory chemist, many of these mishaps of yesteryear have been collated, most notably (and authoritatively) by Leslie Bretherick. Bretherick s Handbook of Reactive Chemical Hazards, which by 2006 had reached its 7th edition, details the predictable and the unexpected from the literature of reactive chemical hazards. In a review, published in Hazards in the Chemical Laboratory, 5th edn, ed. S.G. Luxon, Royal Society of Chemistry, Cambridge, 1992, Bretherick has also summarised some frequently encountered incompatible chemicals that present either a reactive hazard or a toxic hazard if combined. These two lists are reprinted here as Tables 11.4 and 11.5 by kind permission of the Royal Society of Chemistry. In addition, potentially explosive combinations of some commonly-encountered laboratory reagents are shown in Table 11.6 (reproduced with permission from Chemical Safety Matters, lUPAC-IPCS, Cambridge University Press, Cambridge, 1992). 

---

Solid-solid reactions have also been successfully studied by these techniques, as have decompositions of high explosives. The applicability of DTA/DSC to very small samples is of obvious value here. Similarly, preliminary screening of potentially hazardous reaction mixtures is conveniently carried out in this way as a guide to the likelihood of exothermic reaction, the corresponding temperature range and magnitude, and possible effects of pressure and atmospheric conditions (see also Section 26.2.4.5). 

To maintain this temperature at the rate of addition of the nitrite indicated will probably require the occasional addition of cracked ice to the reaction mixture. Only occasional checking of the temperature is necessary. If the temperature is kept below 0°, slightly better yields are obtained. The addition of Dry Ice directly to the ether layer accomplishes this easily. Temperatures above 10° cause a loss in yield, but apparently create no hazard. 

A small amount of hydrazine hydrate was present in the reaction mixture at this point, but a safety evaluation indicated the final reaction mixture had a very low thermal potential (AH = 15.3 J/g). This poses a minimum thermal hazard for vacuum distillation. 

As mentioned before, the vast majority of accidents in batch processing arise when the control of the temperature of the reaction mixture is lost. This situation often leads to a temperature thermal) runaway, i.e. a temperature overshoot that can result in undesired reactions (decompositions), evaporation, or gas formation. As a consequence, pressure is built up inside a reactor and this can cau.se an explosion. The explosion is usually accompanied by damage to the equipment and release of hazardous (toxic, explosive, or flammable) species to the. surroundings. 

A process is inherently safe in a rigorous sense, when no fluctuation or disturbance can cause an accident. To search for synthetic routes that avoid hazardous reactants, intermediates, and reaction mixtures, is an impetus to be seriously considered by chemists and process designers. Nevertheless, there will always be a need to cope with potentially hazardous materials and reaction mixtures in future process design work, the more so because process streams are expected to become potentially more dangerous in the future. The process streams will be more concentrated to increase energy efficiency, to ease purification, and to decrease the load of wastewater and spent acids. More concentrated process streams have a higher specific content of latent energy and are hence less stable. 

Safety. The MR is much safer than the MASR. (1) The reaction zone contains a much smaller amount of the reaction mixture (hazardous material), which always enhances process safety. (2) In case of pump failure, the reaction automatically stops since the liquid falls down from the reaction zone. (3) There is no need to filter the monolithic catalyst after the reaction has been completed. Filtration of the fine catalysts particles used in slurry reactors is a troublesome and time-consuming operation. Moreover, metallic catalysts used in fine chemicals manufacture are pyrophoric, which makes this operation risky. In a slurry reactor there is a risk of thermal runaways. (4) If the cooling capacity is insufficient (e.g. by a mechanical failure) a temperature increase can lead to an increase in reaction, and thus heat generation rate. 

According to an O.S. amendment sheet, the procedure as described [1] is dangerous because the reaction mixture (dicyanodiamide and ammonium nitrate) is similar in composition to commercial blasting explosives. This probably also applies to similar earlier preparations [2]. An earlier procedure which involved heating ammonium thiocyanate, lead nitrate and ammonia demolished a 50 bar autoclave [3], TGA and DTA studies show that air is not involved in the thermal decomposition [4], Explosive properties of the nitrate are detailed [5], An improved process involves catalytic conversion at 90-200°C of a molten mixture of urea and ammonium nitrate to give 92% conversion (on urea) of guanidinium nitrate, recovered by crystallisation. Hazards of alternative processes are listed [6],... 

A mixture of flake potassium hydroxide and sodium hydroxide was added to a reaction mixture without the agitator running. When this was started the batch erupted, owing to the sudden solution exotherm. Although this is a physical hazard rather than a chemical hazard, similar incidents have occurred frequently. 

When following the original route [1] to 3-hydroxythietane-1,1-dioxide, it is essential to dilute and evaporate the hydrogen peroxide-acetic acid reaction mixture slowly from a large dish, to prevent explosions arising from concentration of peroxyacetic acid [2], Alternative routes to avoid this hazard are available [3,4], See Peroxyacetic acid... 

Several reports have been made of the application of this procedure for the preparation of vinylphosphonates129-131 as well as tertiary vinylphosphine oxides132 in good yield. When formaldehyde is used, this approach provides a convenient method for the preparation of chloromethylphosphonic dichloride.133 Extreme caution needs to be exercised in the performance of this reaction because the extremely hazardous bis-chloromethyl ether is generated as an intermediate and may remain in the product or escape the reaction mixture during performance of the reaction. 

The main subject of this chapter is the identification of hazardous chemicals, materials, mixtures, and reaction masses. The chapter deals with undesired decompositions and hazardous reactions. A basic knowledge of the chemistry involved, and, in particular, with the thermodynamics and kinetics, is required. Furthermore, it is important to have a test strategy to recognize and assess the hazards associated with the energetic materials, mixtures, and reaction masses. 

Example 1 Typical Outputs of Thermal Stability Test Methods As discussed in detail later in Section 2.3, various techniques with different working principles are available to identify the thermal reactivity hazards of individual substances and reaction mixtures. Some examples are presented here. 

In DSC instruments, heat production (q) can be determined directly as a function of temperature. The shape of the heat production curve is also important for hazard identification. A sharp rise in energy release rate (i.e., a steep slope of the exotherm), whether due to a rapid increase of the rate constant with temperature or to a large enthalpy of reaction, indicates that the substance or reaction mixture may be hazardous. Figure 2.14 illustrates an example of a DSC curve with a gradual exothermic reaction, while Figure 2.15 is an example of a steep exothermic rise. 

Responses to the CSB industry survey50 indicate that most companies consult a variety of information sources as a first step in compiling data on reactive hazards. However, respondents prefer literature sources and expert opinion over computerized tools such as CHETAH, The Chemical Reactivity Worksheet, or Bretherick s Database of Reactive Chemical Hazards. Such programs can be used to predict the thermal stability of compounds, reaction mixtures, or potential chemical incompatibilities. In some cases, they provide an efficient means of identifying reactive hazards without having to conduct chemical testing. Survey responses showed that five of nine companies consider computer-based tools not valuable. Only two of the surveyed companies use The Chemical Reactivity Worksheet.51... 

In both the OSHA PSM Standard and the EPA RMP regulation, the PHA element does not currently specify the factors that must be considered to effectively manage reactive hazards. Present requirements should be augmented to explicitly require an evaluation of such factors as rate and quantity of heat generated maximum operating temperature to avoid decomposition thermostability of reactants, reaction mixtures, byproduct waste streams, and products effect of charging rates, catalyst addition, and possible contaminants and understanding the consequences of runaway reactions or toxic gas evolution. 

Methoxy-2-nitrophenylacetyl chloride was never isolated from the reaction mixture. It is known5 6 that 2-nitrophenylacetyl chloride is a potentially explosive material, but it is not documented that the 5-methoxy derivative is also hazardous. 5-Methoxy-2-nitrophenylacetyl chloride was also prepared as described by Rosenmund and his co-workers,7 and was used in the crude form. 

It is most important that the temperature of the reaction mixture be not allowed to rise prematurely, since if it reaches 50° during the early stages the reaction velocity increases so rapidly that the contents are apt to boil out of the flask. For the same reason it appears necessary that a certain ratio between reacting mass and cooled surface be not exceeded. Runs of twice the above size have been carried out in a 5-I. flask without mishap, but temperature control was difficult and these conditions are extremely hazardous. A run of 4 kg. of ethylene chlorohydrin in a 12-I. flask was attempted, but rapidly went out of control. 

Safety Handbook", Part III. Hazardous Compounds, Mixtures, and Reactions, Expl Res Sect, Expls Propints Lab, Pica tinny Arsenal, Dover, NJ (Dec 1959), pp 33-46 (List of compds, mists, and reactions which are, or may be, dangerous) 5) "Explosive Accident/lncident Abstracts" (A compilation listing description, causes preventive measures of 219 expl incidents reported to the Armed Services Explosives Safety Board by companies, governmental agencies other groups from Sept 1961 thru June 1967. The Abstracts are available as ASTIA Document 660020 from die National Technical Information Service, US Dept of Commerce, PO Box 1553, Ravensworth, Va 22151)... 

---