The mechanisms behind the expansion of samples immersed in solutions containing sulfate ions have been controversial. Expansion is believed to originate in the formation of ettringite, but it is not possible to find any direct correlation between these two processes or to relate expansion purely to the increase in solid volume. Crystallization pressure is considered to be the most likely mechanism. A key aspect of this mechanism is the existence of supersaturation in the solution with respect to the growing crystal and it is not possible to directly measure the composition of the pore solution at different points within the specimen. In this paper we present a refinement of the crystallization pressure hypothesis to explain sulfate related expansion. We provide evidence that expansion is related to the transformation of monosulfate crystals embedded in the C-S-H to ettringite. Initially incoming sulfate reacts with aluminate containing hydrates in large pores to give ettringite without expansion. During this process the composition of the solution is buffered by the precipitation of solids so the rate of ingress does not depend strongly on the solution composition. After all this "freely available" aluminate has reacted, the concentration of sulfate ions in the pore solution increases, as indicated by the rise of sulfate absorbed on C-S-H. When the solution concentration reaches a critical level it provides the driving force for the precipitation of ettringite crystals in small pores within the C-S-H. Due to the confinement and the supersaturation of the solution this formation of ettringite leads to expansion of the paste. (C) 2012 Elsevier Ltd. All rights reserved.