Internally-Cured Low-Cracking High-Performance Concrete (IC-LC-HPC) Bridge Decks: Durability and Cracking Performance

Abstract

The laboratory portion of this study investigates the effects of internal curing (IC) water in pre-wetted lightweight aggregates (LWA) between 8.2 and 9.0% and between 12.0 and 13.1% by weight of binder and total internal (TI) water in all aggregates between 3.4 and 12.5% by weight of binder on freeze-thaw durability, scaling resistance, shrinkage, and ion transport properties of concrete mixtures with different binder compositions (100% portland cement or a ternary composition with 30% slag cement and 3% silica fume as partial replacements for portland cement [by total weight of cementitious materials]), paste as a percentage of concrete volume (23.7, 24.6, 26.7, or 33.7%), and water-to-cementitious material ratios (w/cm, 0.45 or 0.41). Normalweight aggregates consisted of three types of coarse aggregates and river sand.

The results show that for paste contents between 23.7 and 33.7% of concrete volume, the freeze-thaw durability of internally-cured concrete mixtures is a function of the percentage of IC water by the weight of binder, rather than total IC water per unit volume of concrete; all IC mixtures assessed for freeze-thaw durability in accordance with ASTM C666-Procedure A exhibited durability factors below 90% and failed the freeze-thaw test and would not be considered acceptable under MnDOT specifications, while some mixtures at w/c ratio of 0.45 and all mixtures at a w/c ratio of 0.41satisfied the requirements of ASTM C666-Procedure B and KTMR-22 and would be considered acceptable under KDOT specifications. The results also demonstrate that the freeze-thaw resistance of the mixtures decreased markedly when the TI water exceeded 12.0% by the weight of binder. Scaling test results show that as the paste content increases from 23.7 to 33.7%, the scaling resistance of the specimens decreases. At a w/cm ratio of 0.45 and a paste content of 23.7%, mixtures with an IC water content of 8.8% passed the scaling test; at a w/cm ratio of 0.41 and a paste content of 23.7%, mixtures with IC water contents less than or equal to 13% passed the scaling test. None of the mixtures with a paste content of 33.7% passed the scaling test at either w/cm ratio. Moreover, for a given binder composition and type of coarse aggregate, increased TI water resulted in higher scaling resistance. The type of coarse aggregate also had effects on scaling resistance. The ternary mixtures with granite as the coarse aggregate, had lower mass losses than the ternary mixtures with low-absorption limestone and similar quantities of TI water. As the TI water content increased, shrinkage decreased for concretes with both binder compositions. Mixtures with IC water exhibited more expansion at the end of the curing period than mixtures with no IC water. Increases in the TI water content in mixtures did not affect the rapid chloride permeability or surface resistivity measurements, while the binder composition did, with the ternary mixtures, on average, showing higher and lower SRM and RCP values, respectively, than mixtures containing 100% portland cement.

The second portion of the study involved the construction, crack surveys, and evaluation of 12 bridge decks (nine in Minnesota and three in Kansas) containing IC water and supplementary cementitious materials (SCMs) that were constructed between 2016 and 2021 following IC-LCHPC specifications (of Minnesota or Kansas) and two associated Control decks without IC. The decks were monolithic with the exception of three of the Minnesota decks, which had overlays. The results show that the use of overlays on bridge decks results in high crack densities and should be avoided. Low-cracking bridge decks require concrete with a paste content of 27.2% or less based on concrete volume. Paste contents above 27.2% correlate with increased cracking, and for decks with paste contents greater than 27.2%, the addition of IC and SCMs does not overcome the negative effects of high paste content. The results also indicate that the combination of low paste, internal curing, and good construction procedures offer the potential to reduce cracking. Under circumstances, good construction practices are needed for low-cracking decks. If poor construction practices are employed, even decks with low paste content and IC can exhibit high cracking and scaling damage.