Cold weather admixture systems for cement-based materials applied to masonry mortar binder

Masonry construction in cold weather is a challenge due to the slow or non-existent hydration reaction of the cementitious material at low and subfreezing temperatures. The construction industry’s need for an alternative to providing thermal protection has motivated the exploration of using antifreeze admixtures. The main objectives of this project were to develop and evaluate an effective antifreeze admixture for masonry mortars from available products, and to identify the active components responsible for promoting the strength gain and the mechanisms by which they act. In the first stage of the experimental program, an incomplete response surface design approach was used to develop an antifreeze admixture. The approach consisted of combining a total of six off-the-shelf concrete admixtures, up to five at a time at three dosage levels each. The target optimization function of the system was the minimization of the freezing point of the mortar, which was measured using an embedded thermocouple in the center of the mortar cylinders. Several combinations of admixtures were effective at lowering the freezing point of the mortar mix; however, the compressive strength was found not to be systematically correlated to the freezing temperature. The compressive strengths of mortar samples prepared with the best candidates, when cured at 10°C and 15°C, reached acceptable levels. However, a pre-curing (heat protection) period of between 6 and 12 hours was necessary for the mortar to reach these strength levels. The best performing candidate from the previous stage was selected to undergo further investigation to identify the active compound and to study its effect on the hydration process. Elemental and mineral characterization of the admixture, using mainly X-ray fluorescence (XRF) and X-ray diffraction (XRD), revealed a high concentration of sodium nitrite with some mullite, in addition to an unidentified amorphous phase. The characterization of the hydration products did not reveal any uncommon phases, suggesting the presence of a certain amount of unfrozen water in the pore structure that allowed the hydration reaction to proceed and the C S H phase to develop. The suspected active ingredient (sodium nitrite) was tested as a stand-alone admixture to confirm its action as an antifreeze agent, and produced masonry mortar with an acceptable 28-day compressive strength when cured at 10°C. No pre-curing period was required in this phase of testing. The dosage of sodium nitrite was also optimized and found to be approximately 5% by cement weight to maximize the strength gain. Given that no hard evidence of any unusual ongoing chemical reactions was found using the characterization techniques, the physical action of the antifreeze was investigated. The working hypothesis was that a certain amount of liquid water was present at subfreezing temperatures, which allowed the hydration reaction to proceed. As an indirect way of confirming the hypothesis, the non-destructive time domain reflectometry (TDR) technique was used to measure the bulk dielectric constant of the plain and treated cement pastes during the curing process up to an age of three weeks. A mixing model was formulated to quantitatively track the individual constituents of the cement paste, with a particular interest in the available liquid water at temperatures below the normal freezing temperature. The results showed clear evidence of the existence of liquid water in the antifreeze treated samples, as well as evidence of the consumption of water and unreacted cement at subfreezing temperatures