should

SECTION 2: Hazards identification

Further, physico-chemical data are essential for the correct planning of (eco)toxicological studies and for the optimisation of the test conditions. R.7.1.1.5 References for introduction of Physicochemical properties Recommendations on the transport of dangerous goods, Manual of Test and Criteria, United Nations. http://www.unece.org/trans/danger/publi/manual/manual_e.html Guidance on the Application of the CLP Criteria, Version 4.0 - 2013, ECHA. http://echa.europa.eu/web/ guest/support/quidance-on-reach-and-clp-implementation OECD? (2004) Principles for the Validation of (Q)SARs htto://ecb.jrc.it/ QSAR/home.php 7CONTENU= / QSAR/background/background oecd principles. php OECD? (2004) series on testing and assessment Number 49 The report from the expert group on (quantitative) structure activity relationships [(Q)SARs] on the principles for the validation of (Q)SARs. 2nd Meeting of the ad hoc Expert Group on QSARs http://www.oecd.org/officialdocuments/displaydocumentpdft/ ?cote= env/jm/mono(2004)248&do clanguage=en Version 2.4 — February 2014 31 R.7.1.2 Melting point/ freezing point R.7.1.2.1 Type of property The melting point contributes to the indentification of a substance and to the designation of its physical state (liquid or solid!2) of a substance. A number of

Physical hazards

classes are distinguished based on the physical state. Therefore the melting point of a substance and the consequent designation as liquid or solid has also consequences for the assignment of the correct hazard class. Furthermore, the melting/freezing point together with vapour pressure serves as an indicator for the physical state (liquid or solid) of a substance under specific conditions (e.g environmental conditions, manufacturing process conditions). As a result, with regard to environmental relevance the melting point can give an indication of the distribution of the substance within and between the environmental media (water, soil and air). R.7.1.2.2 Definition The melting temperature is defined as the temperature at which the phase transition from the solid to the liquid state occurs at atmospheric pressure and this temperature ideally corresponds to the freezing temperature. As the phase transition of many substances takes place over a temperature range, it is often described as the melting range. For some substances, the determination of the freezing or solidification point is more appropriate. Where, due to the particular properties of the substance, none of the above parameters can be conveniently measured, a pour point may be appropriate. R.7.1.2.3 Test method(s) Method A.1 of Regulation (EC) 440/2008 or OECD Test Guideline 102 should be generally used for testing. Any procedure given in A.1 may be used within the scope and applicability specifications. However, it is advisable to use the Differential Scanning Calorimetry (DSC) or Differential Thermo-Analysis (DTA) method since they give additional information about the thermal stability of the substance like decomposition onset and energy. If decomposition occurs during the melting point study, determination of the boiling point need not be carried out. In this case, if DSC has been used, conducting the experiment under inert gas should be considered. R.7.1.2.4 Adaptation of the standard testing regime Adaptation possibilities according to column 2 of Annex VII to REACH Column 2 of REACH Annex VII provides the following specific rules for adaptation of the standard information requirement for melting/freezing point: ‘The study does not need to be conducted below a lower limit of - 20 °C.’ Therefore, Annex VII to REACH does not require determination of the melting point below a lower limit of -20 ° C. The lower limit should be confirmed through testing, except where a (Q)SAR indicates a melting point of -50 ° Cor lower. ‘2 Definitions of physical states can be found in Section 1.0. of Annex | to the CLP Regulation. 32 Version 2.4 — February 2014 Adaptation possibilities according to Annex XI to REACH USE OF EXISTING DATA: DATA ON PHYSI CAL-CHEMICAL PROPERTIES FROM EXPERI MENTS NOT CARRIED OUT ACCORDING TO GLP OR THE TEST METHODS REFERRED TO IN ARTICLE 13 (3) OF REACH If experimental data are available (study reports or literature data) meeting the criteria of Annex XI, section 1.1.1, these could be used to meet the endpoint data requirements. If an estimation method is used as a source of information according to Column 2 of Annex VII, the QSAR model must meet the criteria set out in Annex XI, section 1.3. WEIGHT OF EVIDENCE Where no single source of existing data (study reports, QSAR, literature data) is considered sufficiently reliable, thus not fully meeting the criteria of Annex Xl, section 1.1.1 or where several sources of similar reliability with deviating results exist, a weight-of-evidence approach may be used. The criteria of Annex XI, section 1.2 must then be met }3. (Q)SAR For the determination of the melting point, (Q)SAR approaches are discouraged, because the accuracy is not sufficient (+ 25 °C or more) for the purposes of

2.1 Classification of the substance or mixture

risk assessment. GROUPING OF SUBSTANCES AND READ-ACROSS APPROACH For the determination of the melting point read-across is usually not possible. However interpolation may still be possible within homologous series. TESTING 1S TECHNICALLY NOT POSSI BLE Some substances will decompose or sublime before the melting point is reached. Further adaptation possibilities Not foreseen. R.7.1.2.5 Impurities; uncertainties Impurities can have a significant influence on the melting point, as they will generally lower the melting point noticeably. Therefore utmost care should be taken in the selection of the key study(s), or weight-of-evidence approaches, that the data selected is representative of the substance being registered by the respective companies. R.7.1.2.6 Endpoint specific information in the registration dossier/ |UCLID Materials and methods e type of method or reference to the standard or the test method applied. Results and discussion '3 National Institute of Standards and Technology (NIST) have a useful statistical approach which has been used for the evaluation of literature melting point data (ref.: http://webbook.nist.gov/chemistry/site-cal.html# AVG). Version 2.4 — February 2014 33 e melting point value (°C) as measured; e rate of temperature increase if available; e decomposition or sublimation temperature (if applicable) ; e measurement uncertainty if available; e if testing is waived, the reasons for waiving must be documented in the dossier. Any deviation from the guideline method used or any other special consideration should be reported. In cases where there is more than one source of data, the endpoint summary under results and discussion should provide a justification for the selection of the key study chapter. Reference to other ECHA Guidance Documents Further detailed guidance on melting point/freezing point can be found in the following chapters: IUCLID Section REACH Annex Endpoint title IUCLID 5 End User ECHA Practical Manual Chapter Guide 3 4.2 VII 7.2 Melting E.4.3 3.2 point/freezing point R.7.1.3 Boiling point R.7.1.3.1 Type of property The boiling point is a property: e which contributes to the characterisation of a substance and to the designation of its physical state (gas or liquid); e which is the basis for the assignment of the correct hazard class because a number of physical hazard classes are distinguished based on the physical state; e which is needed for the classification of flammable liquids into categories; e which gives an indication of the distribution of the substance within and between the environmental compartments (air, soil and water); e which have correlations with vapour pressure and therefore gives indications whether a substance may be available for inhalation as a vapour or may form flammable/explosive vapour-air mixtures, too; e which is important for physical hazard assessment. R.7.1.3.2 Definition The normal boiling point is the temperature at which the vapour pressure of a liquid equals 101.3 kPa. Note: If the vapour pressure equals 101.3 kPa or more at a given temperature this means the substance is completely gaseous at that temperature. If this is the case at temperatures < 20 °C the substance is a gas also according to the CLP Regulation. 34 Version 2.4 — February 2014 R.7.1.3.3 Test method(s) Method A.2 of Regulation (EC) 440/2008 or OECD Test Guideline 103 should be used for testing. Any determination method may be used within the scope and applicability specifications. DSC allows the determination of the melting and boiling point in a single test. Likewise, for some substances a single test can be used to determine both ‘boiling point’ and ‘vapour pressure’, as when the dynamic method is applied. For high-boiling liquids or liquids which may decompose, auto-oxidize etc. before the boiling point at 101.3 kPa or more is reached, it is recommended to determine the boiling point either under inert gas or at reduced pressures, in order to derive the boiling point at reduced pressures from the vapour pressure curve. If explosive substances, pyrophoric substances or self-reactive substances are to be characterized, determination of the boiling point is in general not practicable. For pyrophoric substances testing under inert gas or reduced pressures should be considered. Where standards are applicable, the use of the most recent updates is advised; they are accessible via numerous websites, see above in section R.7.1.1.3. R.7.1.3.4 Adaptation of the standard testing regime Adaptation possibilities according to column 2 of Annex VII to REACH Column 2 of REACH Annex VII provides the following specific rules for adaptation of the standard information requirement for boiling point: ‘The study does not need to be conducted: e for gases; or e for solids which either melt above 300 °C or decompose before boiling. In such cases the boiling point under reduced pressure may be estimated or measured; or e for substances which decompose before boiling (e.g. auto-oxidation, rearrangement, degradation, decomposition, etc.).’ Therefore the Annex VII to REACH does not require determination of the boiling point if: e the substance is a gas; However, for some gases the boiling point may be relevant. In the CLP Regulation, the boiling point is the main criterion to distinguish gases from liquids (see Annex |, section 1.0: Gas means a substance which (i) at 50 °C has a vapour pressure greater than 300 kPa (absolute); or (ii) is completely gaseous at 20 °C at a standard pressure of 101.3 kPa). Therefore it is important to report the boiling point in borderline cases where the transition from liquid to gas occurs close to 20 °C. e the melting point of the substance is above 300 °C or when any chemical change occurs during the melting point study; e the substance decomposes before boiling at ambient pressure. In such cases the boiling point under reduced pressure (down to 0.2 kPa) should be determined if possible without decomposition. Version 2.4 — February 2014 35 Adaptation possibilities according to Annex XI to REACH USE OF EXISTING DATA: DATA ON PHYSI CAL-CHEMI CAL PROPERTIES FROM EXPERI MENTS NOT CARRIED OUT ACCORDING TO GLP OR THE TEST METHODS REFERRED TO IN ARTICLE 13 (3) OF REACH If experimental data are available (study reports or literature data) meeting the criteria of Annex XI, section 1.1.1, these could be used to meet the endpoint data requirements. If an estimation method is used as a source of information according to Column 2 of Annex VII, the QSAR model must meet the criteria set out in Annex XI, section 1.3. WEIGHT OF EVIDENCE Where no single source of existing data (study reports, QSAR, literature data) is considered sufficiently reliable, thus not fully meeting the criteria of Annex Xl, section 1.1.1 or where several sources of similar reliability with deviating results exist, a weight-of-evidence approach may be used. The criteria of Annex XI, section 1.2 must then be met /4. (Q)SAR For the determination of the boiling point, (Q)SAR approaches are discouraged for the purpose of classification / risk assessment, except when the mean absolute error of the method is lower than 2 K. GROUPING OF SUBSTANCES AND READ- ACROSS APPROACH For the determination of the boiling point read-across is usually not possible. However interpolation may still be possible within homologous series. TESTING IS TECHNICALLY NOT POSSI BLE Testing is not possible if: the substance is an explosive; the substance is self-reactive; e any chemical change occurs during the melting point study; the liquid decomposes before the boiling point is reached even at reduced pressures below 0.2 kPa. In such cases the decomposition temperature in relation to the (reduced) pressure should be reported, in order to allow determination of whether it is the substance itself or its decomposition products that should be considered under environmental conditions for the purpose of risk assessment. The details of the determination method should also be reported. Further adaptation possibilities Data generated with the same tests and principles as specified in the CLP Regulation on boiling point generated in conjunction with

Transport classification

can be deemed to satisfy the REACH requirements on a case-by-case basis. As stated in Annex |X of the REACH Regulation, when for certain endpoints, it is proposed to not provide information for '4 The NIST have a useful statistical approach which has been used for the evaluation of literature boiling point data (ref.: http://webbook.nist.gov/chemistry/site-cal.html# AVG). 36 Version 2.4 — February 2014 other reasons than those mentioned in column 2 of that Annex or in Annex XI of REACH, this fact and the reasons must also be clearly stated. Such an approach may then be used. R.7.1.3.5 Impurities; uncertainties Impurities can have a significant influence on the boiling point. Therefore utmost care should be taken in the selection of the key study(s), or weight-of-evidence approaches, that the data selected is representative of the substance being registered by the respective companies. R.7.1.3.6 Endpoint specific information in the registration dossier / in IUCLID Materials and methods e type of method or reference to the standard or the test method applied. Results and discussion e boiling point value ( °C) as measured; e pressure value and unit; e rate of temperature increase if available; e decomposition (if applicable) ; e measurement uncertainty if available; e boiling point value in °C (corrected to standard pressure, except where the boiling point has been determined at specified reduced pressures) (as above, but in a separate block of fields); e if testing is waived, the reasons for waiving must be documented in the dossier. A Note: In cases where the boiling point is determined at reduced pressure a determination at ambient pressure is obviously not possible. A boiling point at standard pressure could then only be derived by extrapolation of the vapour pressure curve in cases where a vapour pressure curve is known. Even in such cases this corrected/extrapolated boiling point could only be nominal one and would be potentially misleading because it is not possible to determine it at ambient pressure. Any deviation from the guideline method used or any other special consideration should be reported. In cases where there is more than one source of data, the endpoint summary under results and discussion should provide a justification for the selection of the key study chapter. Reference to other ECHA Guidance Documents Further detailed guidance on boiling point can be found in the following chapters: IUCLID Section REACH Annex Endpoint title IUCLID 5 End User ECHA Practical Manual Chapter Guide 3 4.3 VII 7.3 Boiling point E.4.4 3.3 Version 2.4 — February 2014 37 R.7.1.4 Relative density R.7.1.4.1 Type of property For gaseous materials, relative density is of value in determining the tendency to settle or to disperse when discharged at high concentrations into the atmosphere. The relative density of gaseous substances can be calculated from molecular weight using the Ideal Gas Law. For insoluble liquids and solids, (absolute) density will be a determining factor in the settling of the substance. R.7.1.4.2 Definition Density (p) of a substance is the quotient of the mass m and its volume V: p= m/V SI units (kg/m?) The relative density is related to a standard, the density of which is set to 1. It has no dimension. For gases air is used as standard so that gases with a relative density of less than 1 are lighter than air (and and those with a value above 1 heavier). The relative density, D,?°, of solids or liquids is the ratio between the mass of a volume of substance to be examined, determined at 20 °C, and the mass of the same volume of water, determined at 4 °C (at which temperature, water has its maximum density, i.e. 999.975 kg/m®). R.7.1.4.3 Test method(s) Test methods for determining (absolute) density are applicable to solids and liquids. Table R.7.1-2 lists the respective test methods. Table R.7.1-3 Test methods for determining density Method Applicability Maximum Dynamic Viscosity (Liquids only)/ Pa.S Hydrometer Liquids 5 Hydrostatic balance Solids and Liquids 5 Immersion ball Liquids 20 Pycnometer Solids and Liquids 500 Air comparison pycnometer | Solids - Oscillating densitim eter Liquids 5 EU Test guideline A.3 for relative density Regulation (EC) No 440/2008 includes a list of standards with technical information about the different methods and actual measuring of different types of substances. 38 Version 2.4 — February 2014 R.7.1.4.4 Adaptation of the standard testing regime Adaptation possibilities according to column 2 of Annex VII to REACH Column 2 of REACH Annex VII provides the following specific rules for adaptation of the standard information requirement for relative density: ‘The study does not need to be conducted if: e the substance is only stable in solution in a particular solvent and the solution density is similar to that of the solvent. In such cases, an Indication of whether the solution density is higher or lower than the solvent density is sufficient; or e the substance is gaseous at room temperature. In this case, an estimation based on calculation can be made from its molecular weight and the Ideal Gas Laws.’ For liquids, it is useful to have some indication of the dynamic viscosity as this can affect the choice of method. The physical state of test substances should always be homogeneous, this is particularly relevant for highly viscous substances where internal bubbles can be formed; in these cases, the test substance should be allowed to rest until all internal bubbles have disappeared. The summary should include the numerical value for density and temperature at which it was measured, test material identity, purity of the sample used, physical state, method and guideline used and reference substance (if any). Adaptation possibilities according to Annex XI to REACH USE OF EXISTING DATA: DATA ON PHYSI CAL-CHEMI CAL PROPERTIES FROM EXPERI MENTS NOT CARRIED OUT ACCORDING TO GLP OR THE TEST METHODS REFERRED TO IN ARTICLE 13 (3) OF REACH lf experimental data are available (study reports or literature data) meeting the criteria in section 1.1.1 of Annex XI to REACH, these could be used to meet the endpoint data requirements. If an estimation method is used as a source of information according to Column 2 of Annex VII, the QSAR model must meet the criteria set out in section 1.3 of Annex XI to REACH. WEIGHT OF EVIDENCE Where no single source of existing data (study reports, QSAR, literature data) is considered sufficiently reliable, thus not fully meeting the criteria in section 1.1.1 of Annex XI to REACH, or where several sources of similar reliability with deviating results exist, a weight of evidence approach may be used. The criteria in section 1.2 of Annex XI to REACH must then be met !. (Q)SAR is generally not applicable for determination of relative density. For this endpoint there are often experimental measurements and therefore QSPR models for this property have not received special attention in the environmental literature. Several '? The NIST have a useful statistical approach which has been used for the evaluation of literature data (ref.: http://webbook.nist.gov/chemistry/site-cal.html# AVG. Version 2.4 — February 2014 39 software programs can be used to calculate the density of a given substance but the documentation and validation of the methods is limited. GROUPING OF SUBSTANCES AND READ-ACROSS APPROACH For the determination of the relative density read-across is usually not possible. However interpolation may still be possible within homologous series. TESTING IS TECHNICALLY NOT POSSI BLE Testing should always be considered, if none of the waiving possibilities applies. Waiving relative density testing on the basis of not being technically possible is not applicable. Further adaptation possibilities Not foreseen. R.7.1.4.5 Impurities; uncertainties Impurities can have a significant influence on the density. This influence depends on the amount and density of the impurity; thus, the higher the amount of impurity and the higher the difference between the densities of the main component and the impurity, the higher the influence. Therefore utmost care should be taken in the selection of the key study(s), or weight-of-evidence approaches, that the data selected is representative of the substance being registered by the respective companies. Density is temperature dependant. Whenever possible, determinations should be performed at 20 °C. R.7.1.4.6 Endpoint specific information in the registration dossier / in IUCLID Materials and methods e type of method or reference to the standard or the test method applied. Results and discussion e temperature (°C); e relative (for gases)/ absolute (for liquids and solids) density value (dimensionless) ; e measurement uncertainty if available; if testing is waived, the reasons for waiving must be documented in the dossier. Any deviation from the guideline method used or any other special consideration should be reported. In cases where there is more than one source of data, the endpoint summary under results and discussion should provide a justification for the selection of the key study chapter. Reference to other ECHA Guidance Documents Further detailed guidance on relative density can be found in the following chapters: IUCLID Section REACH Annex Endpoint title I!UCLID 5 End User ECHA Practical 40 Version 2.4 — February 2014 Manual Chapter 4.4 VII 7.4 Relative density E.4.5 3.4 Version 2.4 — February 2014 41 R.7.1.5 Vapour pressure R.7.1.5.1 Type of property Vapour pressure is a property: e for substance characterisation; e which serves as a key parameter for assessing some toxicological and environmental e which gives indications whether a substance may be available for inhalation as a vapour or may form flammable/explosive vapour-air mixtures; e which allows determination of the volatility of a substance from an aqueous medium or soil, in terms of the Henry’s Law constant (Appendix R.7.1-1) and partition coefficient air/soil, respectively; e which allows determination of the right container/vessel to ensure safety during storage, transport and use; e which is importiant for physical hazard assessment. R.7.1.5.2 Definition The vapour pressure of a substance is defined as the saturation pressure above a solid ora liquid substance at constant temperature. At the thermodynamic equilibrium, the vapour pressure of a pure substance is a function of temperature only. R.7.1.5.3 Test method(s) Method A.4 of Regulation (EC) 440/2008 or OECD Test Guideline 104 (Vapour pressure) should be used for testing. It is useful to have preliminary information on the structure, the melting point and the boiling point of the substance to perform this test. There is no single measurement procedure applicable to the entire range of vapour pressure values. Therefore, several methods are recommended to be used for the measurement of vapour pressure from < 10°'° to 10° Pa. For the selection of the test method the scope and applicability specifications have to be taken into account. The results should be checked for consistency with other physical data like boiling point, flash point etc. It is recommended to determine the vapour pressure at least for two temperatures, for volatile substances (boiling point up to 150 °C) preferably at 20 °C and at 50 °C. Where standards are applicable, the use of the most recent updates is advised, please check section R.7.1.1.3 for further information. R.7.1.5.4 Adaptation of the standard testing regime Adaptation possibilities according to column 2 of Annex VII to REACH Column 2 of REACH Annex VII provides the following specific rules for adaptation of the standard information requirement for vapour pressure: ‘The study does not need to be conducted if the melting point is above 300 °C. 42 Version 2.4 — February 2014 If the melting point is between 200°C and 300°, a limit value based on measurement or a recognised calculation method is sufficient.’ Vapour pressure testing is also not required for substances with a standard boiling point of < 30 °C, as these substances will have vapour pressures above the limit of measurement (i.e. 10° Pa). Adaptation possibilities according to Annex XI to REACH USE OF EXISTING DATA: DATA ON PHYSI CAL-CHEMI CAL PROPERTIES FROM EXPERI MENTS NOT CARRIED OUT ACCORDING TO GLP OR THE TEST METHODS REFERRED TO IN ARTICLE 13 (3) OF REACH lf experimental data are available (study reports or literature data) meeting the criteria in section 1.1.1 of Annex XI to REACH, these could be used to meet the endpoint data requirements. If an estimation method is used as a source of information according to Column 2 of Annex VII, the QSAR model must meet the criteria set out in section 1.3 of Annex XI to REACH. WEIGHT OF EVIDENCE Where no single source of existing data (study reports, QSAR, literature data) is considered sufficiently reliable, thus not fully meeting the criteria in section 1.1.1 of Annex XI to REACH, or where several sources of similar reliability with deviating results exist, a weight of evidence approach may be used. The criteria in section 1.2 of Annex XI to REACH must then be met. (Q)SAR For the determination of the vapour pressure, (Q)SAR approaches may be used if determination by experiment is not possible. The vapour pressure depends on the temperature. This dependence was modelled by Grain (Grain, 1982), based on thermodynamic principles. The estimation methods differ for vapour pressure that can be applied for compounds that are liquid or gaseous at the temperature of interest, and for solid and liquid compounds. The former can be estimated by the Antoine equation, while the latter could be predicted by the Watson correlation, which accounts also for the heat of vaporisation. Another method, described by Mackay et a/. (1982), is applicable only for hydrocarbons and halogenated hydrocarbons. Further, the Grain model was modified to be applicable for all solids, liquids, and gases. These methods are still in practical use today. The OECD guideline 104 reports that the Watson correlation is applicable over the pressure range from 10° Pa to 10° Pa. It should in any case be pointed out that estimated values for vapour pressure can be subjected to great uncertainty if the computed pressure is lower than 1 Pa, especially when the boiling point has not been experimentally determined (OECD monograph 67). The uncertainty is even greater if the estimated value is used together with water solubility in order to estimate the Henry’s Law constant. The environment monograph 67 of the OECD describes all of the above mentioned methods and the OECD guideline 104 supports the use of the Watson correlation for the calculation of vapour pressure, but does not specifically reject other calculation methods. The handbook for estimating the physico-chemical properties of organic compounds (Reinhard and Drefahl, 1999) reports another method based on thermodynamic properties and elaborated by Mishra and Yalkowsky that discussed the application of the method of Mackay (Mackay et a/., 1982). Version 2.4 — February 2014 43 The equation by Mishra and Yalkowsky gave significantly better estimates than the method of Mackay on the same data set (Mishra and Yalkowsky, 1991). Another methodology that proved to be effective in estimating vapour pressure relies on group contribution approaches. Several models using this strategy have been proposed (Reinhard and Drefahl, 1999; see Table R.7.1-4). Table R.7.1-4 Group contribution approach and vapour pressure Compounds Authors Methodology Statistics Alkyl aromatic compounds | Amidon and Anik | Group contribution | Standard error approach . 1.1 kJ on the estimation for the free energy of vaporisation Mono-, di-, tri- and tetra} Hoshino et al. Group contribution | Average error 3.7 % substituted approach Max. Error 30.9 % Perfluorinated saturated | Kelly et al. Group contribution | Arithmetic mean deviation < 0.5 hydrocarbons approach % Numerous other models are available for the estimation of vapour pressure, and Schwarzenbach et al. (1993), Delle Site (1996), Sage and Sage (2000) and Dearden (2003) have reviewed many of these. The descriptors used in vapour pressure QSPRs include physico- chemical, structural and topological descriptors, and group contributions. Katritzky ef al. (1998) used 4 CODESSA descriptors to model the vapour pressure (in atmospheres at 25 °C) of 411 diverse organic chemicals, with r° = 0.949 standard error = 0.331 log unit. A number of studies (Andreev ef a/. 1994, KUhne et al. 1997, Yaffe & Cohen 2001) allow of the estimation of vapour pressures over a range of temperatures. GROUPING OF SUBSTANCES AND READ-ACROSS APPROACH For the determination of vapour pressure read-across is usually not possible. However interpolation may still be possible within homologous series. TESTING IS TECHNICALLY NOT POSSI BLE Vapour pressure testing is not required for substances with a standard boiling point of < 30 °C, as these substances will have a vapour pressure value above the limit of measurement (i.e. 10° Pa). For substances which decompose during measurement or which are unstable or explosive, determination of the vapour pressure may not be technically possible. This also applies to self- reactive substances and organic peroxides. Pyrophoric substances may be difficult to handle experimentally. If fully inert conditions cannot be maintained during sample preparation and measurement, use of an appropriate calculation method is recommended. A calculation method should also be applied in the case of some corrosive substances which would destroy essential metallic parts of the measurement apparatus. 44 Version 2.4 — February 2014 Further adaptation possibilities Not foreseen. R.7.1.5.5 Impurities; uncertainties Impurities can have a large influence on vapour pressure. The influence depends on the amount of the impurity and the vapour pressure of that impurity. Small amounts of volatile impurities may increase the vapour pressure by several orders of magnitude. This has to be kept in mind when performing the measurements and for the interpretation of results. Therefore utmost care should be taken in the selection of the key study(s), or weight-of- evidence approaches, that the data selected is representative of the substance being registered by the respective companies. Where there are volatile impurities in the sample which could affect the result, the substance may be purified. Test method A.4 states that it may also be appropriate to quote the vapour pressure for the technical material. However, in consideration of the large effect that impurities may have (see above), doing so is strongly discouraged. R.7.1.5.6 Endpoint specific information in the registration dossier / in IUCLID Materials and methods e type of method or description of the apparatus or reference to the standard or the test method applied. Results and discussion e if testing is waived, the reasons for waiving must be documented in the dossier; e measured value of the vapour pressure for at least two temperatures; e estimate of the vapour pressure at 20 or 25 °C (if not measured at these temperatures); e if a transition (change of state, decomposition) is observed, the following should be noted: e nature of change; e temperature at which change occurs. Any deviation from the guideline method used or any other special consideration should be reported. In cases where there is more than one source of data, the endpoint summary under results and discussion should provide a justification for the selection of the key study chapter. Reference to other ECHA Guidance Documents Further detailed guidance on vapour pressure can be found in the following chapters: IUCLID Section REACH Annex Endpoint title I!UCLID 5 End User ECHA Practical Manual Chapter Guide 3 4.6 VII 7.5 Vapour pressure E.4.7 3.6 R.7.1.5.7 References on vapour pressure Version 2.4 — February 2014 45 OECD Guidelines for the Testing of Chemicals / Section 1: Physical-Chemical properties, Test No. 104 Vapour Pressure, OECD Code 979910401E1, July 2006 Andreev N.N, Kuznetsov S.E, Storozhenko S.Y. (1994) Prediction of vapour pressure and boiling points of aliphatic compounds. Mendeleev Commun. 173-174. Grain C.F., (1982) Handbook of chemical property estimation methods. New York, Mc Graw- Hill Delle Site A. (1996) The vapour pressure of environmentally significant organic chemicals: a review of methods and data at ambient temperature. J. Phys. Chem. Ref. Data 26:157-93. Dearden JC. (2003) Quantitative structure-property relationships for prediction of boiling point, vapour pressure, and melting point. Environ Toxicol Chem 22(8):1696-709. Katritzky AR, Y. W, Sild S, Tamm T, Karelson M. (1998) QSPR studies on vapour pressure, aqueous solubility, and the prediction of water-air partition coefficients. J. Chem. Inf. Comput. Sci. 38:720-5. Kuhne R, Ebert RU, Schitrmann G. (1997) Estimation of vapour pressures for hydrocarbons and halogenated hydrocarbons from chemical structure by a neural network. Chemosphere 34:671-86. Mackay D, Bobra A, Chan W, Shiu WY. (1982) Vapour pressure correlation for Low- Volatility Environmental Chemicals. Environ. Sci. Technol. 16:645-9. Mishra DS, Yalkowsky SH. Estimation of vapour pressure of some organic compounds. Ind. Eng. Chem. Res. 1991;30:1609-12. OECD Guidelines for Testing of Chemicals, Method 104 “Vapour Pressure Curve” Reinhard M, Drefahl (1999).A. Handbook for Estimating Physico-Chemical Properties of Organic Compounds. New York: Wiley. Sage M.L, Sage G.W.(2000) Handbook of Property Estimation Methods for Chemicals. Boca Raton, FL: Lewis. Schwartzenbach, R.P., Gswend, P.M., Imboden, D.M. (1993). Environmental Organic Chemistry. John Wiley and Sons. Yaffe D, Cohen, Y (2001) Neural network based temperature-dependent quantitative structure property relationships (QSPRs) for predicting vapour pressure of hydrocarbons. J. Chem. Inf. Comput. Sci. 41:463-477. 46 Version 2.4 — February 2014 R.7.1.6 Surface tension R.7.1.6.1 Type of property Surface tension measurements of aqueous solutions are significant since decreasing the surface tension of water may impact on the properties of the solution and other physicochemical measurements. R.7.1.6.2 Definition e Surface tension: ‘The free surface enthalpy per unit of surface area is referred to as surface tension’ (Council Regulation (EC) No 440/2008). The surface tension is given as: N/m (SI unit) or mN/m (SI sub-unit). 1 N/m = 103 dyne/cm or 1mN/m = 1 dyne/cm in the obsolete cgs system. The surface tension of an aqueous solution of a substance can be used to determine whether the substance is surface active. e Surface active substance (surfactant): “Surfactant’ means any organic substance and/or preparation [mixture] used in detergents, which has surface-active properties and which consists of one or more hydrophilic and one or more hydrophobic groups of such a nature and size that it is capable of reducing the surface tension of water, and of forming spreading or adsorption monolayers at the water-air interface, and of forming emulsions and/or microemulsions and/or micelles, and of adsorption at water- solid interfaces’ (see Article 2(6) of Council Regulation (EC) No 648/2004). R.7.1.6.3 Test method(s) Testing should be done in accordance with one of the methods specified under section A.5 of Regulation (EC) No 440/2008. These methods are applicable to most chemical substances. It is useful to have preliminary information on the water solubility, the structure, the hydrolysis properties and the critical concentration for micelles formation of the substance before performing the test. Surface tension measurements require a test material that is stable against hydrolysis during the test period and soluble in water at concentrations of > 1 mg/l. Measurements should be performed on a solution at either 90 % of the solubility limit or 1 g/l (where viscosity permits), whichever is smaller. R.7.1.6.4 Adaptation of the standard testing regime Adaptation possibilities according to column 2 of Annex VII to REACH Column 2 of REACH Annex VII provides the following specific rules for adaptation of the standard information requirement for surface tension: ‘The study need only be conducted if: e based on structure, surface activity is expected or can be predicted; or e surface activity is a desired property of the material. Version 2.4 — February 2014 47 If the water solubility is below 1mg/I at 20 °C the test does not need to be conducted. ’ Adaptation possibilities according to Annex XI to REACH USE OF EXISTING DATA: DATA ON PHYSI CAL-CHEMI CAL PROPERTIES FROM EXPERI MENTS NOT CARRIED OUT ACCORDING TO GLP OR THE TEST METHODS REFERRED TO IN ARTICLE 13 (3) OF REACH lf experimental data are available (study reports or literature data) meeting the criteria in section 1.1.1 of Annex XI to REACH, these could be used to meet the endpoint data requirements. If an estimation method is used as a source of information according to Column 2 of Annex VII, the QSAR model must meet the criteria set out in section 1.3 of Annex XI to REACH. WEIGHT OF EVIDENCE Where no single source of existing data (study reports, QSAR, literature data) is considered sufficiently reliable, thus not fully meeting the criteria in section 1.1.1 of Annex XI to REACH, or where several sources of similar reliability with deviating results exist, a weight of evidence approach may be used. The criteria in section 1.2 of Annex XI to REACH must then be met. (Q)SAR At the time of writing, no reliable (Q)SAR methods exist for sufficiently accurate predictions of surface tension. GROUPING OF SUBSTANCES AND READ-ACROSS APPROACH For the determination of the surface tension read-across is usually not possible. However interpolation may still be possible within homologous series. TESTING IS TECHNICALLY NOT POSSI BLE Testing should always be considered, if none of the waiving possibilities applies. Testing may not be possible for reactive substances which react with water or air (hydrolyse, are pyrophoric, evolve gas, etc). Further adaptation possibilities Not foreseen. R.7.1.6.5 Impurities; uncertainties For the measurement of surface tension the ring or plate tensiometer methods are preferred. The error on the measurement is in the order of 0.1-0.3 mN/m. Use of the standard protocols and GLP procedures are recommended. Surface active impurities in substances may in some cases lead to false-positive surface tension measurements. R.7.1.6.6 Endpoint specific information in the registration dossier / in IUCLID Materials and methods * description of the apparatus and dimensions or reference to the standard or the test method applied; 48 Version 2.4 — February 2014 * test material identity: apart from general issues, if surface tension of active impurities affects results, it should be noted. Results and discussion e surface tension value and unit (preferably mN/m or N/m but other units are also acceptable); concentration of the solution* !6: age of solution’ ; type of water or solution used* ; results from repeated measurements with varied equilibrium time (of the solution); several measurement results should be provided to assess the possible time- dependency of the measurement. Equilibration times may vary from minutes to hours. Measurements should be sufficient to prove that a constant surface tension was reached; e if testing is waived, the reasons for waiving must be documented in the dossier. Any deviation from the guideline method used (and reasons for it) or any other special consideration should be reported. In cases where there is more than one source of data, the endpoint summary under results and discussion should provide a justification for the selection of the key study chapter. Reference to other ECHA Guidance Documents Further detailed guidance on surface tension can be found in the following chapters: IUCLID Section REACH Annex Endpoint title I!UCLID 5 End User ECHA Practical Manual Chapter Guide 3 4.10 VII 7.6 Surface tension E.4.11 3.9 16 * As indicated in test A.5. Surface tension described in Council Regulation (EC) No 440/2008). Version 2.4 — February 2014 49 R.7.1.7 Water solubility Advice to registrants with regard to nanomaterials characterisation of water solubility can be found in Appendix R7-1 Recommendations for nanomaterials applicable to: Chapter R7a Endpoint specific guidance, section 2.2.1 Water solubility. R.7.1.7.1 Type of property Water solubility is a significant parameter for a number of reasons: e the mobility of a test substance is largely determined by its solubility in water. In general, highly soluble substances are more likely to be distributed by the hydrological cycle; e water soluble substances gain access to humans and other living organisms; e knowledge of the water solubility is a prerequisite for setting up test conditions for a range of fate (e.g. biodegradation, bioaccumulation) and effects studies; e it is also used to derive other environmental parameters, such aS Koy, Kop and Henry’s Law Constant (Appendix R.7.1-1). It is also used as input for some QSAR models; e water solubility is used as a regulatory trigger for waiving certain physicochemical and ecotoxicological endpoints. R.7.1.7.2 Definition ‘The solubility of a substance in water is specified by the saturation mass concentration of the substance in water at a given temperature. The solubility in water is specified in units of mass per volume of solution. The SI unit is kg/m? (grams per litre may also be used)’ (see Regulation (EC) No 440/2008, A.6, section 1.2). Mixtures of organic compounds, e.g. petroleum substances, behave differently from their single constituent compounds when brought into contact with water. Petroleum substances are typically hydrophobic and exhibit low solubility in water. However, reflecting the range of structures, constituent hydrocarbons will exhibit a wide range of water solubility. Therefore, water solubility measurements for these substances are loading rate dependent due to their complex composition. This water solubility behaviour impacts on both the conduct and interpretation of aquatic toxicity tests for these complex substances. The complex composition, and generally low water solubility, impact also on the choice and conduct of biodegradation studies. Consequently, the above definition for solubility of a single