Experimental and numerical characterization of scalable cellulose nano-fiber composite.

摘要

Fiber-reinforced polymer composites have been used in recent years as an alternative to the conventional materials because of their low weight, high mechanical properties and low processing temperatures. Most polymer composites are traditionally made using reinforcing fibers such as carbon or glass fibers. However, there has been recent interest in making these reinforcing fibers from natural resources. The plant-derived cellulose nano-fibers (CNF) are a material with remarkable mechanical properties at the nano-scale that are much superior to the mechanical properties of the traditional natural fibers (such as jute, hemp, kenaf, etc) used in the natural-fiber based polymer composites. Because CNF is bio-based and biodegradable, it is an attractive 'green' alternative for use in automotive, aerospace, and other engineering applications. However, efforts to produce CNF based nano-composites, with successful scaling-up of the remarkable nanoscale properties of CNF, have not met with much success and form an active area of research.;The main goals of this research are to characterize the scalable CNF based nano composites using experimental methods and to develop effective models for flow of polymeric resin in the CNF-based porous media used during the proposed manufacture of CNF nano-composites. In the CNF composite characterization section, scalable isotropic and anisotropic CNF composites were made from a porous CNF preforms created using a freeze drying process. Formation of the fibers during freeze-drying process can change the micro skeleton of the final preform structure as non-aligned or isotropic and aligned or anisotropic CNF. Liquid Composite Molding (LCM) processes form a set of liquid molding technologies that are used quite commonly for making the conventional polymer composites. An improvised vacuum-driven LCM process was used to make the CNF-based nanocomposites from CNF preforms using a 'green' epoxy resin with high bio-content. Under the topic of isotropic CNF, formation of the freeze-dried CNF preforms' porous network strongly affects the mechanical, microstructural and tribological properties of the composite, therefore experimental testing was performed to characterize the effects of pore structure on global properties of isotropic CNF composites. Level of curing was investigated by experimental methods such as DSC in order to analyze its effects on the mechanical properties. The causes of failure in the composites were discussed by analyzing the SEM micrographs of fractured surfaces. The investigations revealed that the silane treated samples show superior mechanical behavior and higher storage modulus compared to the untreated (no silane) samples. The DMA and DSC results indicated a reduction in the glass transition temperature for the CNF composites compared to the pure resin samples. The tensile results showed higher elastic moduli in composites made from silane treated CNF preforms compared to those made from non silane-treated preforms. The tribological behavior of the silylated CNFs composites showed lower coefficient of friction and wear volumes than the neat bio-epoxy due to the formation of a transfer film on the mating surfaces, which led to a decrease in the 'direct contact' of the composite with the asperities of the hard metallic counterface. Under the topic of anisotropic CNF nanocomposites, a recently-discovered new type of CNF preform with more-aligned pore structure was used in our improvised LCM process to make the CNF-based anisotropic nanocomposite. The effect of such aligned pore structure on the mechanical and microstructural properties of CNF-based nanocomposites was investigated. As before, we used the tensile test, DMA and SEM to characterize this new material. Our investigation revealed that anisotropic CNF preform improved the overall mechanical properties of CNF composites due to better interfacing between the CNF and resin inside aligned pore structure of anisotropic CNF. Also, DMA results showed an improvement in the glass transition temperature of the anisotropic samples compared to the isotropic ones. For flow modeling in the CNF-based porous medium, the closure formulation, developed as a part of the derivation of Darcy's law developed by Whitaker [1], was used to develop novel numerical and experimental methods for estimating the permeability and absorption characteristics of a porous medium with a given pore-level microstructure. The permeability of such a porous medium was estimated numerically while the absorption characteristics were analyzed through experiments. In order to use real micrograph in permeability simulations, 2D SEM pictures of the CNF-based porous media were considered. The falling head permeameter was used for measuring the experimental permeability in order to test the accuracy of the permeability tensor obtained by the proposed numerical simulation. The permeability values were also compared with the theoretical models of Kozeny-Carman. A good agreement between the numerical, experimental and analytical methods demonstrated the accuracy of the closure formulation and the resulting simulation. These results also present the closure formulation based method as a viable method to estimate the permeability of porous media using 2D SEM micrographs; such a method harnesses the micro-macro coupling and is marked with absence of any constitutive-relation based assumption for such upscaling. Such a method is also faster, less expensive and less problematic than the corresponding 3D micro-CT scan based method because of much smaller degrees-of-freedom, memory and storage requirements. Under the absorption characteristics study, absorption characteristics of paper-like CNF porous medium was modeled using theoretical derivation of governing equation for single-phase flow and swelling behavior and absorption coefficient were investigated through experiments. In derivation part, unique form of mass conservation was developed using volume averaging theorem in the swelling, liquid-absorbing CNF-based preform. The case of the absorption coefficient, b being unity, which corresponds to the liquid absorption rate into fibers being equal to the fiber expansion rate, results in the classical form of the continuity equation that is originally derived for a rigid, non-deforming porous medium. The value of b was determined using a novel dipping experiment conducted with the help of a microbalance and was found to be unity for flow models in swelling porous media made of the CNF.