PERFORMANCE EVALUATION OF RHEOLOGICAL MODELS FOR CARBON NANOTUBE-ENHANCED CEMENTITIOUS GROUTS
Ground modification enhances soil properties and mechanical behavior to support structural integrity through various mechanical, hydraulic, physical, chemical, and inclusion-confinement methods. Grouting, a technique that combines physical and chemical processes, involves injecting pressurized fluidized materials to strengthen, densify, or reduce soil permeability. Portland cement is the most commonly used binder for grouting applications, as it is in other areas of civil engineering. To improve cement-based grouts' mechanical, chemical, and engineering properties, admixtures, fibers, and nanomaterials are employed. Nanomaterials, such as carbon nanofilaments (SWCNTs, MWCNTs, CNFs), offer extraordinary mechanical properties (e.g., TPa-level Young’s modulus, GPa-level tensile strength), making them promising reinforcements. The workability of cementitious grouts, defined by rheological behavior, viscosity, and yield stress, is crucial for practical application. Using the right rheological model is essential for accurately assessing grout workability by characterizing flow behavior, including yield stress and viscosity, under different conditions. Selecting an appropriate model ensures precise flowability, stability, and performance predictions, which are critical for optimizing grout applications in various field conditions. To assess the performance of rheological models in carbon nanotube-enhanced cementitious grouts, 20 mix designs were prepared with varying water-to-binder (w/b) ratios (0.50, 0.75, 1.00, 1.25, and 1.50) to suit different grout applications (e.g., sealing, compaction, or injection). The binder proportions (Portland cement and CNT matrix) were varied by adding 0.05%, 0.10%, and 0.30% SWCNTs by weight of the solution, with each ratio including a control grout containing only Portland cement for comparison. Each mixture was tested in a rotational rheometer immediately after stirring. The performance of rheological models (i.e., Bingham, Modified Bingham, Herschel-Bulkley, Cross, and Carreau-Yasuda) was compared for each mixture’s rheometer results using statistical metrics and Akaike Information Criterion (AIC) to account for the trade-off between the goodness of fit of the model and its simplicity. For high w/b ratios and low CNT additions, the Herschel-Bulkley model excelled in prediction, fit, simplicity, and compatibility with high shear rates, which were beyond the scope of this study. Similarly, the Cross model demonstrated the best performance in these aspects for low w/b ratios and high CNT additions.
