Influence of the Biot number on error in measuring enthalpy using the T-history method

Abstract

The T-history method measures the enthalpy and the phase transition temperatures in bulk scale samples. Despite its widespread use in characterizing phase change materials (PCMs), a systematic investigation of errors in T-history measurements is still lacking. While a lumped capacitance is often assumed in the data analysis, actual experimental conditions may not justify such use. To help identify when such analysis can be significantly erroneous, this paper investigates the quantitative relationship between the Biot number (Bi) and error using both experiments and computational modeling. We use two different sample tubes in combination with convection conditions to vary Bi. At Bi<0.1, T-history shows excellent agreement with Differential Scanning Calorimetry (DSC) measurements. As Bi approaches 1, the error increases progressively due to the development of a vertical non-uniformity in temperature. At low Bi, the thermal time constant of the sample can still introduce significant uncertainty in data analysis that is often overlooked. Finally, the analysis is also sensitive to the specific heat of the reference material. By identifying experimental conditions that properly justify the use of the T-history data analysis, this work helps improve the overall accuracy and reliability of the method most commonly used in characterizing PCMs.