Many different reports have studied the crystallization behaviour of lactose, e.g., by exposing samples of amorphous lactose to different relative humidity at constant temperatures. However, only few reports are available investigating the formation of α-lactose monohydrate and β-lactose during re-crystallization.
Applying the static ampoule method in the microcalorimeter, the enthalpies of amorphous lactose were reported to be constantly 32 and 48 J/g, respectively, considering the mutarotation of lactose at 25 °C and 58% RH, 75% RH and 100% RH. In this study, an alternative microcalorimetric technique, the relative humidity-perfusion cell (RH-perfusion cell) was chosen. The RH-perfusion cell is able to deliver a constant and controlled flow of humidified air to the sample. Investigated compounds were purely amorphous lactose and different powder mixtures of lactose. They consisted of α-lactose monohydrate (Pharmatose 325M®), β-lactose (Pharmatose DCL21®) or a combination (1:1) thereof as carriers, and different concentrations of amorphous lactose. The determination of the enthalpy of desorption of the just re-crystallization lactose by the RH-perfusion cell was used to discriminate whether the monohydrate or the anhydrous form of lactose was produced. Differences in the re-crystallization behaviour of lactose at 25 °C and 58–100% RH were found. At 60–80% RH purely amorphous lactose showed a high heat of desorption which can be attributed to a very high content of formed β-lactose. Powder mixtures containing high contents of amorphous lactose (8% and 15%, respectively) blended with α-lactose monohydrate as a carrier resulted in similar results at the same RH ranges. The high amount of β-lactose can be due to the equilibrium anomeric composition. Whereas powder mixtures containing β-lactose as a carrier and amorphous lactose in a concentration of 1%, 8% and 15%, respectively, formed less β-lactose than the mixtures containing α-lactose monohydrate as a carrier. At a relative humidity of 90% none of the powder mixtures showed desorption as to the fact that in all cases only α-lactose monohydrate was formed at the surface of the re-crystallized lactose. Furthermore, mixtures of α-lactose monohydrate and β-lactose (1:1) and 8% amorphous lactose were investigated. An increase in formed α-lactose monohydrate by increasing RH was found. To consolidate the results, the same mixtures were re-crystallized at different RH in desiccators and subsequently investigated in the solution calorimeter. The results of the pre-mix were confirmed by the solution calorimeter. In summary, purely amorphous lactose and mixtures containing α-lactose monohydrate as a carrier show different re-crystallization behaviour compared to mixtures containing β-lactose as a carrier.