The effect of specimen size on autogenous and total shrinkage of ultra-high performance concrete (UHPC)

Unlike normal strength concretes, in which drying is the dominant form of shrinkage, in concretes with very low water to cement ratios autogenous and chemical shrinkage mechanisms can dominate. While the impact of specimen size and shape on drying shrinkage is well understood, the same is not true for autogenous and chemical shrinkage, and this lack of understanding may limit model precision and accuracy. To address this issue, this paper presents the results of a series of experiments conducted to measure the dependency of shrinkage of UPHC on specimen size. Results, recorded from 2 days after water addition, demonstrate a strong specimen size dependency when tested under both sealed and unsealed conditions, thereby indicating that the underlying mechanism is fundamentally different from normal strength concrete, with autogenous shrinkage exhibiting a large influence. Existing shrinkage models (AS3600, B4, CEB-FIP, GL2000 and ACI209) are evaluated for their potential calibration and/or extension to low water to binder ratio concretes and it is shown that commonly used parameters to account for size dependency in normal strength concrete (volume to surface area ratio and hypothetical thickness) do not capture size dependency in UHPC.