Annex 2 EXPERIMENTAL METHODS

The determination of the system type should be done experimentally. The experimental method will depend on the type of.calorimetric equipment used. Some information is given in Annex 2. v. . . -... 

Data about the rate of the runaway reaction, (characterised by the adiabatic rate of temperature rise, dT/dt, and the rate of permanent gas generation, QG, as appropriate), is best obtained experimentally. Methods are given in Annex 2. The relief system sizing method will depend on whether the system is tempered or untempered (see 4.2 and Annex 2). 

A9.5.1.3 Data on bioconcentration properties of a substance may be available from standardized tests or may be estimated from the structure of the molecule. The interpretation of such bioconcentration data for classification purposes often requires detailed evaluation of test data. In order to facilitate this evaluation two additional appendixes are enclosed. These appendixes describe available methods (Appendix III of Annex 9) and factors influencing the bioconcentration potential (Appendix IV of Annex 9). Finally, a list of standardized experimental methods for determination of bioconcentration and Kow are attached (Appendix V of Annex 9) together with a list of references (Appendix VI of Annex 9). 

The EC study methods are given in Annex V of the Directive. The Base Set tests are published as Commission Directive 92/69/EEC [23], which is the update to the methods of Commission Directive 84/449/EEC [24] used for Sixth Amendment notifications. Some of the methods fw Level 1 studies are contained in Commission Directive 87/302/EEC [25]. The EC methods closely parallel the OECD Guidelines [26]. The Annex V methods must be followed unless deviation is scientifically justified. Omission of a Base Set test for technical reasons must also be justified. Tests must be performed in compliance with GLP. In view of Council Directive 86/609/EEC [27] on the protection of experimental animals, tests on animals are to be pexformed as humanely as possiUe, and limit tests are accqrtable for substances only toxic under extreme conditions. 

Leung s method is given in 6.3.2 below. The method.is an approximate solution to the differential mass and energy balances for the reactor during relief and takes account of both emptying via the relief system and the tempering effect of vapour production due to relief. The method makes use of adiabatic experimental data for the rate of heat release from the runaway reaction (see Annex 2). Nomenclature is given in Annex 10. 

No simplifying assumptions are involved to this point in the derivation. Simplifying assumptions are involved to obtain the mass, m, at the point when the temperature begins to fall. The required relief flow rate can then be calculated on the basis that, at the maximum temperature, the heat released by the reaction is balanced by the heat removed by pressure relief. The method makes use of adiabatic experimental data, which can be obtained as described in Annex 2. 

Fauske s method14,51 (given below) is based on emptying the reactor (or achieving vapour/ liquid disengagement) before the pressure has risen from the relief pressure to the maximum accumulated pressure in a vented system. The method incorporates the simplified equilibrium rate model, ERM, for saturated liquid inlet (see 9.4.2) together with the Clausius-Clapeyron relationship (discussed in 6.3.3). The method makes use of adiabatic experimental rate data for the runaway, whose measurement is described in Annex 2. 

Material for which there is no appropriate place in the body text (e.g. material that is subsidiary to or separate from the body text, is included in support of statements in the main text, or describes methods of calculation, experimental procedures or limits and conditions) may be presented in appendices or annexes. 

Credible evaluation uses robust statistical techniques to construct the counterfactual. These techniques include randomization or experimental design quasi-experiments such as matching, regression discontinuity design, or double differences and nonexperimental or instrumental variable methods (annex 6.3). All these techniques try to guarantee the comparability of the treatment and control groups by avoiding what statisticians refer to as bias (box 6.7). 

The methods used to select the counterfactual or control group, known as evaluation designs, can be broadly classified into three categories experimental, quasi-experimental, and nonexperimental. These three evaluation designs vary in terms of feasibility, cost, degree of clarity, validity of results, and extent of selection bias. This annex, adapted from Ravallion (2008), summarizes these designs. 

---