Modelling of heat capacity–temperature data for sucrose–water systems

Abstract

The significance of the point of intersection of the ‘ice-melting curve’ and glass-transition curve (measured by coordinates T′g and C′g on a ‘state’ diagram) as a characteristic feature of the behaviour of aqueous solutions of small sugar (and other) molecules on freezing, has been much emphasised in recent years. In consequence, accurate measurement of T′g and C′g for a solute–water system is important, and at present much reliance is placed upon differential scanning calorimetry as a means of achieving this. Recently, however, controversy has arisen regarding the validity of this approach, particularly where studies are restricted to the freezing of dilute solutions. In an attempt to clarify this matter, the present paper focuses on the sucrose–water system, and describes simulation of experimental heat capacity–temperature scan data for a range of sucrose concentrations. The simulations suggest that a true value for C′g can be obtained from DSC scans performed on dilute solutions only if a correction is made for the temperature dependence of the latent heat of melting of ice and, more importantly, the correct heat capacity ‘background’ is known. The simulations also suggest that, in the ideally reversible situation in which the state diagram is followed precisely, the heat capacity step occurring at T′g will contain a substantial extra contribution from initiation of ice dissolution. In real situations, however, as a result of imperfect annealing of samples during preparation, and irreversibility arising during DSC experiments, a separation and broadening of these events occurs, with the practical consequence that the measured temperatures corresponding to them bracket the true T′g value.