Experimental Study of De-icing Salt-frost Scaling in Concrete with Low-calcium Fly Ash or Slag : Influence of Drying and Carbonation, and Air Content

In order to reduce the net climate impact of concrete production, it is necessary for the industry to undertake several actions. Among these is to include fly ash and slag as supplementary cementitious materials. Accordingly, this project has studied concrete containing low-calcium fly ash or ground-granulated blast-furnace slag. In incorporating fly ash or slag as supplementary material, the cement reaction is affected and the products of reaction are not the same. Consequently, the material properties and durability also change.This PhD-project has studied a superficial damage on concrete called de-icing saltfrost scaling (DISFS), which occurs when a solution with a low concentration of deicing agent freezes in contact with concrete. DISFS has been studied ever since the beginning of the 20th century in order to understand the damage mechanism, and how concrete recipes should be designed to minimize the damage. Early studies have shown that a salt solution with a low concentration (approx. 3%) results in more DISFS damage than when the concentration is 0 or 10%. Moreover, studies have shown that an entrained air void content, created by using an air entraining agent, reduces the DISFS. Furthermore, studies have also shown that preconditioning of the concrete surface, especially drying and carbonation, also have an effect on the DISFS. The reason for this is that the microstructure changes depending on the preconditioning process. If a sample is cured for a longer period, more products are created from the binder reaction and the microstructure becomes denser. Two factors that affect this are carbonation and drying. Carbonation is the reaction between the concrete surface and CO2 (carbon dioxide), which changes the microstructure of the surface. Drying does not by itself affect the microstructure, but it can cause micro cracks to form and it lowers the initial moisture content. Prior studies have resulted in some theories that try to explain the mechanism causing DISFS. However, there is a lack of studies that have tested what influence the factors mentioned above (drying and carbonation, air content, and varying salt solution concentrations) have on the DISFS on concrete containing fly ash or slag. Concrete containing fly ash or slag has different microstructures compared to ordinary concrete containing 100% cement. Because curing and carbonation change the microstructures of the material, there is a need for a broad study analysing how these different factors (drying and carbonation, air content, and varying salt solution concentrations) affect the DISFS in concrete containing fly ash or slag. As the microstructures change with time due to curing and carbonation, there can be preconditioning processes that are favourable to some materials and unfavourable to others.This project has focused on how drying, carbonation, air content, and salt concentration affect the DISFS in concrete containing fly ash or slag. To determine 9 the influence of air content on the DISFS, concrete batches with various air contents are required. A method was developed to find one combination of admixtures (one superplasticiser and one air-entraining agent) for each binder. Linear traverse analysis was used to quantify the varied air-void systems created by the airentraining agent. Furthermore, other measurements have been performed to characterise the materials. Low-temperature calorimetry has been conducted to assess the amount of ice formed and acquire information on the microstructure of pores with sizes below 50 nm. Because a coarse porosity results in a high absorption rate, and a fine porosity results in a low absorption rate, capillary absorption measurements have been conducted to obtain some information on capillary porosity (pores over 50 nm in size); moreover, total porosities are also determined. Measurements of pH have been conducted to determine whether the samples used for low-temperature calorimetry and capillary absorption measurements are completely carbonated. To analyse the effect of carbonation on the DISFS, thinslice analysis has been performed to measure the carbonation depth on the samples used for the DISFS test. It is possible to correlate the carbonation depth with the DISFS mass by using the approximation that 1 kg/m² equals a 1-mm scaling depth. A DISFS test is developed and designed to contribute to a small variation in the DISFS results.A study has been made where different characteristics of the porosity have been correlated to the DISFS for concrete that have hydrated for over 300 d. The binder combinations used for these tests were CEM I mixed with 20, 35% LCFA, or 20, 35, 70% slag, or 25+10% slag+limestone filler. These tests focused on the effect on the DISFS from the increased fraction of SCM, the effect from drying and carbonation, and the effect from various air contents for each binder combination. Moreover, a broader DISFS study was performed on concrete with 100% CEM I, 35% LCFA, or 35% slag, where the concrete hydrated for 8 to 63 d. These tests were conducted for several reasons. Partly to acquire more knowledge regarding how different preconditioning processes, air contents, and salt solution concentrations affect the DISFS in concrete containing fly ash or slag exposed to a short period of hydration. The results also bridge the gap between the effect drying and carbonation has of the DISFS on concrete exposed to a short period of hydration in relation to well hydrated concrete (results from concrete that have hydrated for over 300 days). Moreover, they bridge the gap between the effect air content has on the DISFS of concrete exposed to a short period of hydration in relation to well hydrated concrete (results from concrete that have hydrated for over 300 days). The broad approach on this DIFS study contributed to some interesting and unexpected results.The study that aimed to correlate different characteristics of the porosity with DISFS showed that carbonation contribute to a lower DISFS for concrete containing 100% CEM I, 20, and 35% slag, unchanged DISFS for concrete containing 20% LCFA, 10 or 25+10% slag+limestone filler, and increased DISFS for concrete containing 35% LCFA, or 70% slag. The study also found a correlation between the carbonation depth and the DISFS with the approximation that 1 kg/m² DISFS equals to 1 mm DISFS depth. Thereby the following conclusions could be made. A fine structure of pores below 50 nm (that reduces the fraction of ice that can form at a given temperature), combined with low permeability according to the capillary suction measurements (which infer a fine porosity for the structure of pores larger than 50 nm), generally results in a low DISFS. (The only exception was the noncarbonated concrete with 25% slag and 10% limestone filler). A coarse structure of pores below 50 nm (which enable more water to freeze at a given temperature), with a high permeability according to the capillary suction measurements (which infer a coarse porosity for the structure of pores larger than 50 nm) generally results in a high DISFS. The DISFS results on the concrete that have hydrated for 8 to 63 d clearly show that the preconditioning process affect the DISFS. The different materials can be affected by the same preconditioning process in various ways. The results also show that the air content has different effects on the DISFS depending on whether or not the test surface is dried and carbonated. The tests with different NaCl concentrations resulted in some unexpected results where a high NaCl concentration resulted in a high DISFS in two cases. The cause for these two unique cases for 35% LCFA is believed to be that the critical degree of saturation was exceeded. Otherwise the results seemed to agree with the so called ‘glue spall theory’ where a high salt solution concentration results in a low DISFS.This project has shown that these three factors (drying and carbonation, air content, and NaCl concentration) affect the DISFS in various ways depending on the material. Accordingly, this highlights the complexity of attempting to test the DISFS in the laboratory. Some researchers have criticised the standard test methods for being too conservative and so that some concrete recipes containing fly ash or slag which seem DISFS resistant in the field do not pass. Because, there is a critical need to lower the CO2 emissions due to global warming, there is a reason to fine-tune the standard preconditioning process making it less conservative, so that concrete containing fly ash or slag that reduce the net CO2 emissions, which works in the field tests, does not fail the test. This thesis contributes with important information if a fine-tuning of the standard preconditioning process would be carried through of the European standard DISFS test method