1 31 st Cement and Concrete Science Conference Paper Number XX Novel Developments and Innovation in Cementitious Materials 12-13 September 2011 Imperial College London, United Kingdom Measuring thermal properties of hydrating cement pastes B. Milovanovic, I. Banjad Pecur; I.Gabrijel Department of materials, Faculty of Civil Engineering, University of Zagreb ABSTRACT Critical inputs for heat transfer computations in concrete are the thermo-physical properties of the material as a function of time, including its density, heat capacity, and thermal conductivity. It is widely accepted that the conductivity and the heat transfer coefficient have an influence on temperature gradient along the concrete structure. Meanwhile, the exothermic phenomenon during the hardening process is mainly affected by the specific heat and density of concrete. While thermal and mechanical properties of the hardened concrete have been investigated in detail, only few studies of thermal properties fresh and young hardening concrete have been performed. Measurements of thermal properties of cement materials as heterogeneous, wet and porous materials by conventional steady-state methods can produce large errors. In order to avoid water migration during the long run-time of the steady thermal tests, transient measurement methods are preferable. In this paper, a transient plane source (TPS) method is applied to measuring heat capacities and thermal conductivities of hydrating cement pastes cured under both saturated and sealed conditions at room temperature. 1. INTRODUCTION One of the challenges in the design of mass concrete structures is to avoid the initiation of cracks regardless of the concrete element size, concreting procedure, weather conditions and material properties. Temperature and stress analysis due to the hydration of concrete is highly non-linear problem because a vide variety of time- dependent boundary conditions and strongly time and temperature dependent thermal and mechanical properties of early-age concrete. Modeling the generation and transfer of heat in early-age concrete is essential to understanding the behavior of concrete block as a whole. Thermal gradients can cause internal stresses which can lead to cracking on a microscopic or macroscopic scale (Mehta, 2006). Hence, the temperature of concrete during hardening is a major design consideration, especially since the time- temperature history affects both strength and durability (De Schutter, 2001). Various models to predict the distributions of temperature (Bošnjak, 2000; Mikulić, 2007) and stress (Milovanović, 2009) in young hardening concrete have been developed in recent years. Critical inputs for these computations are the thermo physical properties of the concrete as a function of time, including its density, heat capacity, and thermal conductivity. As important input variables in the heat conduction model, the thermal conductivity and heat capacity need to be determined depending on time or the maturity of concrete. It is widely accepted that conductivity and heat capacity have an influence on temperature gradient along the concrete structure and also that it could be taken into the model with the value of hardened concrete or even the values of aggregates used to make concrete mixture. Thermal conductivity values of concrete range from 1.98 to 2.94 W/(mK) in accordance with different aggregates, as reported by the ACI Committee 207 or even wider range reported for normal concretes, from 1.20 up to 3.00 W/(mK) (Bošnjak, 2000). However, thermal properties of concrete are depended by the concrete’s mixture proportions, the thermo physical properties of the aggregates that it contains, and those of its hydrating cement (binder) paste component, (Bentz, 2007; Bentz, 2011). While the necessary densities are generally well known and easily measurable, less information is available on the heat capacity and thermal conductivity, particularly those of the hydrating cement paste. Previous measurements of thermal conductivity for cement-based materials were conveniently summarized by (De Schutter, 1995). Since that time, several additional studies have been published (Morabito, 2001; Demirboga, 2003; Kim, 2003). Obtained values from these studies still exhibit considerable scatter of thermal conductivity and heat capacity (Bentz, 2007). Contradictory results on thermal conductivity of concrete at early ages can be summarized as shown in the figure 1. An analytical evaluation of the effect showed a slight increase of the