Time-dependent Deformation of Ultra-high Performance Concrete
| dc.contributor.advisor | Bennett, Terry | |
| dc.contributor.advisor | Visintin, Phillip | |
| dc.contributor.author | Sun, Ming | |
| dc.contributor.school | School of Civil, Environmental and Mining Engineering | en |
| dc.date.issued | 2022 | |
| dc.description.abstract | When compared to conventional concrete, ultra-high performance concrete (UHPC) is characterised by its superior compressive strength and high durability, and this makes UHPC potentially revolutionary in the construction industry. The characteristics of UHPC are achieved by high binder contents and low water to binder (w/b) ratios, and these mix design characteristics also impact time-dependent deformations such as shrinkage and creep. Time-dependent deformations can lead to development of cracks and loss of prestress, which can contribute to deflection of members and reduction in strength. Existing design models of shrinkage and creep are formulated and calibrated for lower strength concretes and therefore may not be applicable to UHPC, due to the great difference in behaviour and mechanisms of shrinkage and creep, and this may limit the ability to apply UHPC in practice. Through the literature review, the identified lack of studies and understanding of shrinkage and creep of UHPC results in difficulties in extending or developing models. This thesis seeks to address this limitation by further experimental investigation of UHPC shrinkage and creep. Past studies on shrinkage and creep were reviewed and assessed and several issues were identified. When considering shrinkage, a large number of past studies do not report details of UHPC mix design, chemical and physical properties of mix constituents, or test methodology. In addition, it was found that the effects of w/b ratios and silica fume dosages on shrinkage of UHPC are inconsistently reported in literature, due to competing effects. For instance, the increase in w/b can result in increased pore size and decreased capillary pressures, leading to decrease in autogenous shrinkage; however, increased w/b can promote hydration, causing increase in skeleton stiffness, restraining shrinkage development. An increase in silica fume dosage can decrease pore size and increase capillary pressures, causing increased autogenous shrinkage, but when silica fume dosage exceeds the threshold value, agglomerate can occur and conversely increase pore size. In addition to autogenous shrinkage, their effects on total shrinkage should be analysed based on both autogenous and drying shrinkage components, which requires further studies. Therefore, an experimental investigation on very early-age shrinkage of UHPC from 3 hours after water addition to the age of 7 days was performed to understand the effects of w/b and silica fume dosage on shrinkage of UHPC, development pattern of early-age shrinkage and underlying mechanisms. The test results demonstrated that the development of very early-age shrinkage of UHPC is very intense in the first 24 hours; therefore, the time-zero of autogenous shrinkage of UHPC should be chosen carefully, as a significant difference in magnitude can be introduced in later ages based on the selected time-zero. The effects of w/b and silica fume dosage on shrinkage are a combination of competing effects and the competing effects not only occur in autogenous shrinkage but also in between autogenous and drying shrinkage components. Shrinkage size effect is a phenomenon that shrinkage strain of concrete samples with the same mix design has dependency on sample size and shape and it is important in shrinkage modelling. For conventional concrete, the shrinkage size effect of total shrinkage is due to drying shrinkage component and its uneven moisture distribution of moisture. However, whether this is true for UHPC with low w/b is questionable. When considering shrinkage modelling of UHPC, the existing codified design models only consider total shrinkage and do not consider drying and autogenous components, except AS3600 and RILEM code B4. In addition, these two models attribute the observed shrinkage size effect to the drying shrinkage component, which may not be true for UHPC because drying shrinkage of UHPC is relatively negligible compared to the magnitude of autogenous shrinkage. Therefore, the effect of sample sizes on shrinkage of UHPC was also investigated and it can be concluded from test results that unlike normal and high strength concrete, UHPC demonstrated size effect on autogenous shrinkage, due to uneven distribution of temperature and chemical reaction degree within the specimens. The effect of drying on shrinkage of UHPC showed limited size dependency, resulting from minor drying shrinkage component of total shrinkage. In addition, the existing models are not able to capture shrinkage and shrinkage size effect of UHPC and a major modification is needed in future studies. When considering creep of UHPC, the number of studies on creep is limited, with only six papers available for compressive creep of UHPC. In addition, as creep can be regarded as additional shrinkage under stress, it is closely correlated to shrinkage; therefore, similar to shrinkage of UHPC, the behaviour of basic and drying creep of UHPC may also differ from conventional concrete. However, there is no study in the literature measuring basic and drying compressive creep of UHPC simultaneously; therefore, in order to investigate basic and drying creep of UHPC and potential creep size effect, experiments were performed on cylinder samples with different sizes. In addition, the prediction results of creep models have the unit of either creep compliance (creep strain per stress) or creep coefficient (creep strain per elastic strain), indicating creep strain is proportional to load levels; therefore, different load levels were utilised in the test. The performance of model fib MC2010 on creep modelling of UHPC was evaluated. The test results demonstrated that similar to shrinkage of UHPC, creep of UHPC was also governed by the basic creep component and the effect of drying is not evident. Model fib MC2010 can be recalibrated to fit test data, but with limited data available in literature, the calibration and extension of this model is difficult and more compressive creep data of UHPC is needed in the future to further calibrate this model. | en |
| dc.description.dissertation | Thesis (Ph.D.) -- University of Adelaide, School of Civil, Environmental & Mining Engineering, 2022 | en |
| dc.identifier.uri | https://hdl.handle.net/2440/136409 | |
| dc.language.iso | en | en |
| dc.provenance | This electronic version is made publicly available by the University of Adelaide in accordance with its open access policy for student theses. Copyright in this thesis remains with the author. This thesis may incorporate third party material which has been used by the author pursuant to Fair Dealing exceptions. If you are the owner of any included third party copyright material you wish to be removed from this electronic version, please complete the take down form located at: http://www.adelaide.edu.au/legals | en |
| dc.subject | UHPC | en |
| dc.subject | Creep | en |
| dc.subject | Shrinkage | en |
| dc.subject | Model | en |
| dc.subject | Autogenous shrinkage | en |
| dc.title | Time-dependent Deformation of Ultra-high Performance Concrete | en |
| dc.type | Thesis | en |
Files
Original bundle
1 - 1 of 1
No Thumbnail Available
- Name:
- Sun2022_PhD.pdf
- Size:
- 11.18 MB
- Format:
- Adobe Portable Document Format
- Description: