In their recent paper in Building Simulation (Trabelsi et al. 2012), “Assessment of temperature gradient effects on moisture transfer through thermogradient coefficient”, Trabelsi et al. introduce their assertions on ‘the occurrence of significant thermal diffusion’. In this comment, the premises and outcomes of their analysis are challenged, based on my own recent paper on that same topic (Janssen 2011).
Firstly Trabelsi et al.’s physical model and measurement methodology are critiqued, partially based on their own measured results. Reinterpretation of their data indicates that no consistent nor significant thermal diffusion can be found, confirming vapour pressure as sole significant transport potential for diffusion. This suggests that their physical model, built upon vapour density and temperature, is physically confusing and needlessly complicated. The model additionally requires a thermogradient coefficient, the measurement of which is complex and unreliable. A physical model for vapour diffusion based on vapour pressure, making use of the easy to measure vapour permeability, is thus pragmatically and fundamentally preferable. The comment finally ends by disputing Trabelsi et al.’s findings on the impact of thermal diffusion on moisture buffering: it is argued that they are practically and fundamentally faulty.