Title: Impact of materials science and stereology on the design of experiments in concrete technology

Authors: Piet Stroeven

Addresses: Faculty of Civil Engineering and Geosciences, Delft University of Technology, Stevinweg 1, 2628 CN Delft, The Netherlands

Abstract: Experimentation in concrete technology relies heavily on correct fulfilment of the various sampling principles from stereology and materials science. Random sampling is uneconomical and technically difficult, hence in this paper it is replaced by orthogonal tripod (orthrip) sampling. It is indicated how the efforts can be minimised. The cases of dispersed fibres and cracks are elaborated, making use of projections and sections. Size sampling is a major issue in the paper, requiring introduction of the materials science principles of stochastic heterogeneity and structural sensitivity. Samples are generally of subrepresentative size in concrete technology, so that the estimated characteristics of material structure or properties that are to some degree dependent on configuration (instead of solely on composition) will be biased. Yet comparison of the effects of technological parameters can be achieved in an unbiased way when sample size and sensitivity of observations are fixed proportions of the respective representative elements. This is illustrated with two examples.

Keywords: concrete technology; stochastic heterogeneity; representative volume-area element; sampling strategy; stereology; structure sensitivity; particulate composite; mesostructure; microstructure; experimental design; element size; dispersed fibres; cracks.

DOI: 10.1504/IJMMP.2009.028637

International Journal of Microstructure and Materials Properties, 2009 Vol.4 No.2, pp.250 - 260

Published online: 21 Sep 2009 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article