The original version of the 4sight computer program was written to facilitate the performance assessment of buried concrete structures based deterministic calculations of service life, using data supplied by the user. In an effort to assure reliability in the results, a means of validation was required. Since the results from the calculations depend strongly on the diffusive and advective transport of ionic species, validating the transport portion of 4sight was paramount.The validation of the transport portion of 4sight used both reference and laboratory data. The reference data include electrolyte densities, activity coefficients, and binary salt diffusion coefficients. The laboratory experiments included diffusion experiments performed on relatively inert ceramic frits, using a range of ionic species and concentrations. The laboratory experiments were also used to confirm the fundamental assumption that multi-species transport in non-reacting systems can be characterized by two numbers, the porosity and the formation factor, regardless of the number of diffusing species.A default analysis for estimating the spacing and thickness of flexural and drying shrinkage cracks was added to 4sight. The flexural cracks are calculated from a rudimentary structural analysis that assumes the roof of the vault acts as a one-way slab. The drying shrinkage cracks are estimated, in large part, from the total water content of the concrete mixture, the exposure environment, and the geometry of the slab. The subsequent predicted crack widths, crack spacings, and location of the neutral axis, are used to predict the bulk permeability of the concrete structure as a function of depth; the portion above the neutral axis being uncracked, the portion below the neutral axis being cracked.Another addition to 4sight was the means to perform a Monte Carlo calcuation, based on parameter uncertainty, of the duration of the useful service life. The service life calculation is repeated multiple times using a random number generator to calculate random deviates for the physical parameters. The results can then be used to make a probabilistic statement about the performance of the concrete structure.The resulting computer program is a state-of-the-art attempt to characterize the response of a concrete to its environment. However, there can be no assurances that the model is exact for every scenario possible, regardless of parameter uncertainty. Therefore, in cases that concern public safety, the only way to confirm the response of the concrete to its environment is through continuous monitoring and periodic testing. The computer program, in these circumstances, can also serve as a useful tool in establishing a testing and monitoring protocol. The spatial and temporal distribution of the monitoring and testing protocol can be established using the computer program as a guide.
NIST Interagency/Internal Report (NISTIR) - 6747
building technology, concrete, cracking, degradation, Monte Carlo, service life, transport