Cementitious Calorimetry
An investigation into the thermal properties of Corex slag and the relationship between thermal profiles and setting times
Kyle Cullen van der Westhuizen - 4th year Dissertaion - CIV4044S
CONCLUSIONS
Ultimately the results indicate that slag pastes have lower rates of heat development and therefore may be beneficial in the reduction of thermal cracking in large concrete pours. Although Corex slag does not offer the same heat reduction benefit as blastfurnce slag, it still produced a 32% reduction in the peak temperature of 100% PC paste over a 24 hour period. This finding is contrary to results presented by Alexander et al. (2003), which indicated no appreciable reductions in the heat produced through Corex slag addition. However it is possible that the Corex mixture has a slower decrease in heat rate after the initial peak that occurs within 24 hours. This characteristic could lead to a total heat production that shows less severe differences between the Corex slag mixture and 100% PC mixture after a 7 day period. In order to validate this claim heat profiles would have to be developed over a longer time period with a fully adiabatic calorimeter to avoid heat loss.
THERMAL PROPERTIES OF GGBS AND GGCS
The setting times of the experimental mixtures corresponded to very lower proportions of the main peak response. This could be attributed to the high level of insulation provided by the semi-adiabatic calorimetric setup and the low w/b ratios used for the mixtures. The level of insulation has been found to amplify the peak of the thermal profile (ASTM draft 2012), which in turn would result in smaller fractions of the MPR corresponding to setting time. Taking this into account isothermal calorimetry may be a better method for determining comparative characteristics because the level of insulation would not influence the profile developed as the heat generated is allowed to flow away from the sample providing consistent and reproducible profiles. However a semi-adiabatic calorimeter is far cheaper and easier to construct and a standard that explicitly specifies the level of insulation, sample volume, environment and temperature sensors required may eliminate the large degree of variability associated with is method of calorimetry.
RELATIONSHIP BETWEEN THERMAL PROFILES AND SETTING TIMES
The semi-adiabatic calorimetric setup constructed is a convenient and cost effective system as clear thermal profiles were developed and the thermocouples embedded in the samples could easily be retrieved for reuse. Semi-adiabatic calorimetry is still a relatively new means by which to monitor and characterise cements and concrete within industry, however it is growing and simply lacks standardisation and empirical evidence to support its use. The exothermic nature of the hydration reaction makes the measurement of the thermal profiles an insightful graphical tool to classify and characterise cements, blended cements and concrete mix designs. However semi-adiabatic calorimetry does not give clear indications of the actual heat liberated during the curing period and therefore cannot be used to develop input data for heat prediction model to monitor thermal cracking. Its use should be focused on comparative data but even in this case caution must be taken because there are many variables that may affect the results such as the level of insulation, sample volume and test environment. For this reason the accuracy of a calorimeter is an important parameter to be aware of even if comparative analysis is the key concern as the relative difference between two systems may result in analysis methods becoming invalid within a certain accuracy level. One way to combat this problem is to establish a classification method for semi-adiabatic systems relative to a fully adiabatic system. A semi-adiabatic system could be classified by comparing the results obtained for a standard mixture to those established by a fully adiabatic system. Comparative test methods may then be tailored to semi-adiabatic calorimeters that fall within a certain classification. Ultimately a fully adiabatic calorimetry is important for developing quantifiable heat rate values for temperature prediction models for structural analysis. However more in-complex and cost effective semi-adiabatic systems are more useful for industry to adopt due to the robust nature and mobility that can be offered.
SEMI-ADIABATIC CALORIMETRY
There was a clear indication that slag addition increased both the initial and final setting times when compared to 100% PC. No clear difference was found between the initial setting times of Corex slag and blastfurnace slag; however the Corex slag did reach its final set 30 min faster, which verifies the increased reactivity. The use of CEM I 52.5 N also showed a decrease in final setting with both slags when compared to CEM I 42.5 N, which is expected due to the greater reactivity associated with a higher strength class of cement.
SETTING TIME PROPERTIES OF GGCS AND GGBS