Expansion behavior of reinforced concrete elements due to alkali-silica reaction
Abstract
The influence of reinforcement on the development and distribution of multiaxial expansions in reinforced concrete elements due to alkali-silica reaction (ASR) requires further research. Understanding how passive restraint provided by reinforcement in a given direction may simultaneously affect expansions in that direction as well as in other reinforced or unreinforced directions is important for assessment of ASR-affected structures. Few experimental studies to date have endeavored to monitor expansions in more than one direction for field-scale reinforced concrete specimens. A parametric study at The University of Texas at Austin was conducted in which thirty-three 19 in. reinforced concrete cubes were fabricated and monitored for ongoing ASR expansions. The cubes had variable uniaxial, biaxial, and triaxial reinforcement layouts and ratios. Three concrete mixtures of varying reactivity, controlled by altering the types of coarse and fine aggregate, were used. The cubes were conditioned in an outdoor, climate-monitored environmental facility and were regularly measured in three orthogonal directions to track expansions. The expansions monitored physically corresponded to micro- and macro-cracks that form due to ASR. The specimens exhibited typical surface cracking due to ASR. Random “map cracking” occurred on the surfaces of the triaxially restrained and unreinforced specimens. Large, discrete surface cracks formed between layers of reinforcement for the uniaxially and biaxially restrained specimens as the specimens expanded in the unreinforced directions. Cylinder tests were performed over the course of the program to gauge material property degradation due to ASR-induced expansions; however, due to the small size and fluctuating conditioning of the cylinders the results from these tests were not considered representative of the material property degradation associated with the cube specimens. Prism expansion monitoring was used to predict the free expansion potential of the cube specimens under idealized ASR conditioning. However, much like with mechanical property degradation, the small size and differential conditioning of the prisms limited effective correlation of prism expansions to cube specimen expansions. This study primarily focused on the development of ASR-induced reinforced concrete expansions over time. Cubes fabricated from the three mixes exhibited variable expansions due to the different reactivities of the coarse and fine aggregates; however, the use of different mixes did not change the overall axial expansion distribution trends documented across all specimens. Expansions in unreinforced directions exceeded those of the reinforced directions in all specimens. The rate of expansion of the reinforced directions of the uniaxially and biaxially restrained specimens was reduced once the expansion of the restrained axes reached and exceeded the steel yield strain. The unreinforced and triaxially reinforced specimens exhibited similar proportional axial expansions in all directions; however, the reinforcement did reduce expansions. Analysis of the axial distribution of volumetric expansions, i.e. the summation of the axial expansions, made it possible to remove the influence of concrete mixture reactivity and variable moisture and temperature conditioning during the timeline of specimen production and placement within the storage facility. The axial expansion distribution trends depicted the axial expansions influenced by the reinforcement and not the conditioning of the specimens. This permitted the development of expansion trends that were solely impacted by axial reinforcement conditions, and were independent of other extraneous factors.