Fictive temperatures of pharmaceutical glasses—A comparison of two methods for determining the enthalpy and entropy integrals

Fictive temperature of a glass, TfH is determined by matching the Cp dT   integrals in a single DSC heating scan (method I), a method also useful for determining the fictive temperature Tfσ by matching the Cp dln T   integrals. But for quench-formed glasses, the heating scan contains a broad minimum that prevents one from determining TfH and Tfσ by method I. Therefore, an additional heating scan is used for the purpose (method II). For some quench-formed glasses, method II has yielded Tfσ considerably different from TfH, causing concerns on the use of the method and on the conclusions on the entropy of glass. To resolve it, we use method II to determine TfH and Tfσ of normally formed and annealed   glasses of acetaminophen, carbamazepine and griseofulvin, and find that both the TfH and Tfσ values are identical to the values obtained by using method I. This confirms that the entropy production on the liquid–glass path is too small to affect the residual entropy of glass. Contribution from dynamically heterogeneous, β   or JG relaxation is not observed for these low-energy glasses. The difference between TfH and Tfσ previously observed for quench-formed glasses is due to data errors and/or extrapolation.

► Fictive temperatures of pharmaceutical glasses determined by two methods.
► Enthalpy and entropy integrals give the same fictive temperatures.
► Previously found large difference is due to the data and extrapolation errors.
► Production of entropy on an irreversible path is negligible.