One method that has shown much promise due to its simplicity and effectiveness in homogeneous catalyst recovery is the use of thermomorphic solvent systems (TMS). In this contribution, a novel method for TMS solvent selection based on quantum chemical predictions of catalyst solubility and phase equilibrium is presented. This allows for solvent effects on the catalyst to be incorporated directly into the solvent screening process. A framework for TMS design is developed and implemented using the hydroformylation of 1-dodecene and the rhodium-Biphephos catalyst as an example reaction system. In this way, several promising TMS systems were identified. Experiments were then performed to validate the model based on catalyst partitioning and phase equilibrium. This was followed by conducting a series of reactions to investigate feasibility of the new TMS systems in the actual hydroformylation. In the end it was shown that although some problems arise from inconsistencies in phase equilibrium predictions, the method does provide a functioning a priori basis for TMS development. (C) 2015 The Authors. Published by Elsevier B.V.