Doctoral Degrees (Civil Engineering)

Permanent URI for this collection

https://scholar.sun.ac.za/handle/10019.1/764

Browse

Browsing Doctoral Degrees (Civil Engineering) by Author "Alexandre, Vital Jorge Fisch"

Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    The time-dependent behaviour of cracked textile reinforced concrete (TRC)
    (Stellenbosch : Stellenbosch University, 2023-03) Alexandre, Vital Jorge Fisch; Combrinck, Riaan; Boshoff, William Peter; Stellenbosch University. Faculty of Engineering. Dept. of Civil Engineering.
    ENGLISH ABSTRACT: The use of concrete faces drawbacks in the form of anthropogenic carbon dioxide emissions and a lower tensile strength threshold. The use of textile-reinforced concrete (TRC) is sought to tackle both these issues. Firstly, TRC elements provide an option for thinner structural elements and secondly, it improves the concrete’s post-crack behaviour when considering the short-term loading. However, there is limited research on the behaviour of TRC when subjected to long-term uni-axial loading, mainly due to the time-consuming nature of such related investigations. The purpose of this study is to investigate the behaviour of TRC composites when subjected to various sustained uniaxial loading levels for sustained time periods. This is accomplished by conducting both short- and long-term tests. The short-term investigations focus on the interaction between the textile and matrix and are assessed by delving into the pull-out of yarns from the matrix and uniaxial tensile strength tests. The long-term behaviour is investigated by performing tensile creep tests on predamaged and non-damaged specimens. The sustained loads applied ranged between 10 % and 75 % of the ultimate tensile static load test results. The single-yarn pull-out tests showed that the pull-out behaviour depends on the embedment length. Shorter lengths (25 and 35 mm) exhibited strain-softening behaviour with pull-out being the dominant failure mechanism. If the embedment length increases (30, 40, and 60 mm), then strain-hardening dictates the pull-out behaviour. Pull-out was found to be the failure mechanism for the 30 and 40 mm lengths and rupturing for the 60 mm embedment length. Additionally, the dynamic stage of the pull-out response was also identified to be associated with a bottleneck mechanism forming. This mechanism results from the imprint the warp yarn leaves in the matrix, constricting and dilating the extraction pathway. Particle fragments also cause congestion of the pathways, increasing the pull-out resistance. The uniaxial static pull-out tests were conducted with specimens containing two to six layers of textiles. It was discovered that the number of weft yarns in the observed section dictated the maximum number of cracks formed when considering the crack saturation. All samples also showed a ductile failure attributed to the telescopic failure mechanism. Moreover, the stiffness degradation showed that the samples failed when the secant modulus lowered to 2 GPa. The latter occurred regardless of the number of cracks and number of textiles, indicating that stiffness was related to the failure as opposed to the reinforcing area. The stiffness degradation is argued to be tied to the telescopic mechanism taking place. The sustained uniaxial load tests showed that the time-dependent strain increased with time and increasing sustained load level. The samples with a stress-level below 60 % did not fracture during the period over which the loads were sustained. However, the samples loaded at 75 % stress levels fractured within 10 minutes of loading. The residual strength tests also highlighted that the straightening of fibres during the sustained loading period, known as the training effect, enhanced the strength of the specimens compared to samples that were not subjected to sustained loads but with the same specimen age. Pre-damaged specimens also exhibited lower time-dependent strains compared to those with no predamage. This observation is also attributed to the training effect. The implications of the training effect was also noticed when considering the loading history of samples by increasing the stress level for a select few specimens.