DisclaimerThe content of this site has been prepared by the members of the thematic network "S2S - A Gateway for Plant and Process safety". Since the conditions of use are beyond our control we disclaim any liability, including patent infringement, incurred in connection with the use of these products, data or suggestions.
The thematic network S2S is a European Community Project carried out in the "Competitive and Sustainable Growth" programme and funded in part by contract number G1RT-CT-2002-05092.
Training - Evaluation Chemical Reaction Hazards
The purpose of this section is to familiarise the reader with some of the techniques which may be applied to the prediction of potential reactive chemicals hazards.
Any expert in the field of Reactive Chemicals will attest to the fact that most reactive chemicals incidents can be readily explained by chemistry which is already known from public literature. For those willing to take the minimal time, (and if they know where to look) a literature search can be extremely beneficial in the Reactive Chemicals hazard assessment process. In the following paragraphs, several works of interest will be discussed briefly.
The knowledge of binary chemical compatibility is an important part of the safe handling, transport, and use of chemicals. A significant amount of Reactive Chemicals incidents in industry (and elsewhere) results from incompatibility issues, especially with regard to waste streams, waste storage and transportation. Therefore it is important to consider the "reactive" compatibility of materials at an early stage in process scale-up. Generally compatibility information can be documented and presented in chart form for easy reference. These charts may be compiled from easy to obtain information from the chemical literature or from in-house "process knowledge". It is very important, when discussing the issue of chemical compatibility, to define the scenario under which the mixing (or potential mixing) will take place.
The following movie shows the incompatibility between a chlorinated compound and the 2,4,5,6 tetrahydro-1,2-dimethil pymiridine.
Representing the compatibility of materials in an easy-to-use chart can be an efficient way to convey compatibility information. This can be especially useful in general education in the chemical plant or lab, or for areas and operations where commonly performed tasks might lead to a chemical mix such as might occur during co-shipment in compartmentalized containers, storage in common diked areas, or common waste containers in the lab.
Documentation for these charts is another important consideration. If testing was performed to make a decision about a particular binary combination in a chart, then reference this test in the chart. As mentioned earlier, exceptions to the chart should also be noted. Details of how to prepare a compatibility chart may be found in ASTM Standard E2012-99 as well as by Frurip et al. (1997).
There are at least two publicly available compatibility charts in general use in the United States. One is the Coast Guard Chart (1994) and the other is the ASTM "P-168" chart (1986). The Coast Guard chart was compiled to address the scenario of inadvertent mixing during sea transport.
The general objective of thermal screening of chemicals and mixtures is to identify whether the sample can undergo an exotherm process and the temperature range of its occurrence, which provides a preliminary indication of potential chemical reaction hazards. The focus lies here on secondary reactions. Depending on the calorimeter used additional information on the amount and rate of released heat, kinetics (normal or autocatalytic) and pressure build up can be obtained. Thermal screening is the first experimental stage in a hazard assessment.
Calorimeters for thermal screening are available from various producers in different specifications and in different sensibilities. The following picture shows as the same exothermic effect can be recognized at different temperatures depending on the sensibility of the used calorimeter.
DTA measures the difference of temperature of a sample compared with the temperature of a reference (an empty or filled with inert material pan). Sensors measure the temperature of the sample or sample container relative to a reference.
In most of the available apparatuses this temperature difference can be calibrated in units of rate of heat generation. Samples which absorb a known amount of heat at a specific temperature, for instance the melting of a known quantity of indium, are used for calibration.
DSC directly measures the heat flow and the resulting thermal effect (endo or exothermic) can be integrated to obtain the enthalpy of the transition. DSC measure can be based both on ?power compensation? or ?heat flow? principle.
Power compensation DSC
Heat Flow DSC
In a typical screening apparatus a small amount of sample is placed in a pressure resistant metal or glass container and is heated at a constant rate (0.1 to 10K/min) in the temperature range of 20 to 500 ?C.
Besides this scanning mode, isothermal experiments are also used for certain applications.
There is either no mixing or only very limited mixing in the test cells.
All screening calorimeters are characterised by the small sample size they require (mg- to g-scale) and by the speed at which measurements can be performed. They are particularly useful for:
In screening calorimeters results can strongly depend on some important variables:
Besides the scanning mode, isothermal experiments are also used for certain applications, i.e. when the sample must be thermally stressed for a certain time (drying, distillation, etc). Beside this cases, it must be underlined that the temperature of decomposition is not a thermodynamic parameter: a decomposition can occur at all temperatures but with different times. A dynamic scanning on a sample can reveal a decomposition temperature but the sample can decompose also if heated at a lower temperature for a long period. The following picture shows the typical DSC results in isothermal mode.
In the example the sample decomposes at 200 ?C after 15 minutes, at 180 ?C after 40 minutes and it seems it doesn?t decompose after 170 minutes in the furnace at 160 ?C. To verify if the sample doesn?t really decompose after 170 min at 160 ?C it is advisable to perform a new dynamic test on the aged sample and to compare the result with a dynamic screen on the fresh sample. If the 2 tests give the same result the sample is really stable for 170 min at 160 ?C; if the test on the aged sample is different from the test on the fresh sample (it is the aged sample is stable in dynamic test and the fresh sample is reactive in a dynamic test) it means that the sample is not stable in the isothermal conditions but the exothermic effect is hidden in the noise of the baseline.
The interpretation and evaluation of screening experiments are often difficult because the peak limits can not be clearly recognised and only rarely does an exothermic peak appear independent from other peaks. However, using a different heating rate can sometimes separate previously unseparated peaks.
The most obvious result of a screening test is the direction of the resulting peak indicating an exothermic or endothermic process. The other information is the onset temperature1, Tonset, the peak temperature and the peak shape, where the onset temperature characterises the thermal stability, and the peak shape (size and sharpness) gives an indication on the hazard potential of the sample
The preliminary assessment of the thermal stability of chemicals or mixtures at production scale is often based on a rule-of-thumb using Tonset of the secondary reaction. Such a rule-of-thumb states that if the maximum operating temperature of a process is X Kelvin lower than Tonset, the operation will not experience this secondary reaction, and it is not necessary to obtain more detailed information by other means. For instance, a safety margin of 100K is often used for DSC measurements. This kind of rule-of-thumb can easily be misused, since the value of Tonset depends on the sensitivity of the apparatus, the sample container, the heating rate and is sometimes difficult to determine - especially if the signal is noisy and/or curvy. The DSC safety margin of 100K is based on the assumption that the decomposition follows zeroth order kinetics and has a defined temperature dependence. It has been shown that reactions with activation energies below 80kJ/mol will violate the 100K rule (Hofelich, 1989). Hence, the 100K rule was extended to ensure that low activation energy decompositions are excluded. Beside the 100K safety margin, a shift of the peak temperature of less then 40K is demanded when changing the heating rate by a factor of ten (Steinbach, 1995).
The baseline represents the theoretical profile, which would have been measured if the reaction/decomposition would have taken place without any heat release. The difference between the measured heat flow and the baseline is the heat production rate of the reaction. For a baseline correction the peak limits have to be fixed, which can be quite difficult - especially with a noisy and/or curvy heat flow profile. In the reaction interval the unknown baseline must be interpolated, which is often an error prone task. Commercial evaluation software often offers a large variety of interpolation methods, e.g. linear, integral, spline, horizontal etc. The baseline type determines the obtained results and should be chosen carefully. If there is any doubt about the peak limits or the baseline profile, a second experiment on the sample may help with the determination of the peak limits and baseline profile.
The integration of the baseline-corrected heat flow over time gives the heat of reaction, which is a measure of the thermal hazard. Based on the amount of released heat decisions are made how to proceed with further investigations.
The thermal hazard is usually related to the adiabatic temperature rise, which can be calculated by dividing the heat of reaction with the heat capacity of the sample. An adiabatic temperature rise of 50K, which equals a heat of reaction of about 100J/g for a typical organic liquid, is often used as lower hazard limit. This means, if an exothermic process generates less than 100J/g than this process is regarded as not hazardous providing gaseous products or vapour are not generated in significant amounts. If the sample undergoes several exothermic processes, the generated heat of each peak relative to the temperature differences of their onset temperatures has to be considered.
Some calorimeters enable the measurement of the pressure on-line indicating the amount and rate of gas generation, which can afford valuable information about hazardous gas production. The plot of the logarithm of the pressure vs. the reciprocal absolute temperature gives an indication whether the investigated sample generates gaseous products. An approximately straight line in such an Antoine plot indicates that the pressure, in the sample container, rises due to vaporisation rather than the generation of non-condensable products.
Considerable deviations of the slope from 10.5 times the absolute boiling temperature of the liquid indicate possible generation of non-condensable products.