Overview and thermodynamic modelling of deep eutectic solvents as co-solvents to enhance drug solubilities in water

Abstract

The poor water solubility of active pharmaceutical ingredients (APIs) is a major challenge in the pharmaceutical industry. Co-solvents are sometimes added to enhance drug dissolution. A novel group of co-solvents, the Deep Eutectic Solvents (DES), have gained interest in the pharmaceutical field due to their good solvent power, biodegradability, sustainability, non-toxicity, and low cost. In this study, we first provide an overview of all the literature solubility studies involving a drug or API + water + DES, which can be a valuable list to some researchers. Then, we analyze these systems with focus on each individual drug/API and provide statistical information on each. A similar analysis is carried out with focus on the individual DESs. An investigation of the numeric values of the water-solubility enhancement by the different DESs for various drugs indicates that DESs are indeed effective co-solvents, with varying degrees of solubility enhancement, even up to 15-fold. This is strongly encouraging, indicating the need for further studies to find the most promising DESs for solubility enhancement. However, time-consuming and costly trial and error should be prevented by first screening, using theoretical-based or thermodynamic-based models. Based on this conclusion, the second part of the study is concerned with investigating and suggesting accurate thermodynamic approaches to tackle the phase equilibrium modeling of such systems. For this purpose, a large data bank was collected, consisting of 2009 solubility data of 25 different drugs/APIs mixed with water and 31 different DESs as co-solvents at various DES concentrations, over wide ranges of temperatures at atmospheric pressure. This data bank includes 107 DES + water + drug/API systems in total. The solubility data were then modeled according to the solid–liquid equilibrium framework, using the local composition activity coefficient models of NRTL, and UNIQUAC. The results showed acceptable behavior with respect to the experimental values and trends for all of the investigated systems, with AARD% values of 9.65 % and 14.08 % for the NRTL and UNIQUAC models, respectively. In general, the lower errors of NRTL, as well as its simpler calculation process and the requirement of fewer component parameters, suggest the priority of NRTL over UNIQUAC for use in this field.