Uniaxial Tensile Creep Result
After subjecting the specimens to the respective creep loads for 8 months, Fig. 8 shows the results of the average time-dependent crack opening at the various stress levels investigated. It should be noted that the effect of shrinkage has already been deducted from the overall creep measured as previously mentioned.
It is acknowledged that there is variability in the test results of specimens within each stress level due to the distribution and number of fibres, however, the goal here is to consider the average creep response based on the stress levels tested. Evidently, from the result presented in Fig. 8, the creep response is stress dependent. The higher the stress level, the higher the creep measured as crack opening. At higher stress levels (60 and 70 %), the creep was significantly pronounced leading to creep fracture after less than a day for one of the two specimens subjected to 70 % stress level. Similarly, one of the specimens tested at 60 % stress level was also unloaded from its creep frame after 15 days when the crack opening had reached the maximum measuring capacity of the LVDTs (10 mm). So, results for 60 and 70 % stress levels are not averages but the results of the single specimen left. At 40 % stress level, the result represented the time-dependent behaviour up to 6 months because the test was started much later from the others. However, the crack opening still lie between that of 30 and 50 % stress levels and expected would have followed the same trend up to 8 months.
Fig. 8 Average tensile creep results at different stress levels 
It is noteworthy to observe that, the results presented in Fig. 8 indicate that even at the lowest stress level of 30 % (i.e. 0.23 MPa), a total crack opening of 1.2 mm is recorded, which is quite significant. This means that even at such low levels of loading, the crack width increase is significant during uni-axial sustained loading.