Improving consistency at testing cementitious materials in the Dynamic Surface Leaching Test on the basis of the European technical specification CEN/TS 16637-2 - Results of a round robin test.

The environmental impact assessment of materials is usually based on laboratory tests, mostly in combination with models describing the longterm fate of the substances of interest in the targeted environmental compartment. Thus, laboratory tests are the fundamental link to achieve appropriate assessment conclusions which makes it essential to generate consistent results. This just as applies to the leaching of cementitious materials. In Europe, the leaching behavior of monolithic building materials is tested in the Dynamic Surface Leaching Test following the specification CEN/TS 16637-2. An interlaboratory comparison on European level regarding this technical specification showed relatively high intra- and interlaboratory variations for the tested materials (monolithic copper slag and cement stabilized coal fly ash). Therefore the German Committee for Structural Concrete (DAfStb) framed a guideline to specify additional testing conditions for cementitious materials. To assess the possible improvement by this guidelines measures, a round robin test with 11 participants from Germany and the Netherlands was conducted. This work aims to provide insight into the factors to be considered in the testing of alkaline materials, including sample preparation, and highlights crucial procedures and their manifestation in the results. All evaluated parameters showed improved results compared to the earlier round robin test. The relative standard deviations for repeatability (RSD) and reproducibility (RSD) of the elements calcium, barium, antimony, chromium, molybdenum and vanadium, which are the parameters evaluated in both round robin tests, were RSD = 4%, 4%, 2%, 5%, 5%, and 5% respectively (4% in average) for this work, in comparison to the European round robin test with an average RSD of 29% (17%, 17%, 20%, 40%, 36%, and 42%). The RSD improved from 41% (30%, 36%, 29%, 57%, 40%, and 56%) to 14% (12%, 8%, 6%, 28%, 15%, and 12%). CO ingress during testing and the inaccuracy of eluate analytics for concentrations close to the determination limits were identified as the main sources of error.