ICME Multi-scale Modelling of Ultra High Performance Concrete (UHPC)

From EVOCD
Jump to: navigation, search

Objective

Concrete is the most commonly used manmade construction material in the world, and is the second most consumed substance behind water. Approximately 10 billion tons of concrete are produced annually. Because of the prevalent use of concrete, it is important to understand the fracture behavior. Concrete’s fracture behavior has been an area of much study and research for the last several decades and poses great challenges due to the complex heterogeneous microstructure spanning over a wide range of length scales from nanometers to millimeters. The objective of this proposed research is to capture concrete’s fracture behavior using multiscale modelling and FEA simulations in an effort to bridge the information provided from each length scale. Three representative volume element (RVE) cubes were generated, each of a different size and including various inclusion types, using Python script created by William Lawrimore. These RVE cubes will be used in mesoscale finite element simulations using Abaqus software, boundary conditions consisting of displacement/strain information provided from the macroscale simulations and applied to lower and lower length scales. The information obtained from these simulations determines the structure property relationships within the concrete at lower length scales and provides the desired information on mechanical properties, concrete failure, fracture behavior, and damage evolution.

Megan Burcham has laid a foundation for this work by quantifying the structure-property relationships of Ultra-High Performance Concrete (UHPC) using imaging techniques to characterize the multiscale hierarchical heterogeneities and the mechanical properties. Different sized inclusion types such as steel fibers, sand grains, unhydrated cement grains, and voids where used to construct UHPC’s composite like structure. Through image analysis Burcham was able to analyze the average inclusion size, percent area, nearest neighbor distance, and relative number density of each inclusion type. The information Burcham acquired was used to create the aforementioned RVE cubes that will be used in this multiscale modelling research. Mingzhong Zang and Andrey P. Jivkov have also developed a site-bond model with elastic brittle spring bundles for analysis of the mechanical behavior of cement paste. The model length and elasticity was calculated using the volume fraction and size distribution of anhydrous cement grains. They also used porosity and pore size distributions to allocate local failure energies. They discovered that the macroscopic damage emerges from micro-crack population represented by bond removals. They also confirmed that this technology could predict mechanical and fracture behavior of cementitious materials based exclusively on microstructural information.[1] [2] [3]


Multiscale Bridging Diagram; Bottom Left Image taken from Pellene, R., Lequeux, N., and van Damme, H., “Engineering the bonding scheme in C-S-H: The Iono-Covalent framework,” Cement and Concrete Research, No. 38, pp. 159–174, 2008.


Multiscale Approach

Multiscale concrete modelling takes into account the needed information provided across various length scales ranging from the electronic scale up to the macroscopic scale. Understanding the fracture behavior requires information about crack coalescence provided through FEA or DEM analysis. From crack coalescence information about fiber-bridging is needed and can be obtain through FEA or DEM analysis. Next information about fiber/paste bonding is required from the micromechanics scale. Then information about concrete’s bonding strengths and the effect of water is needed. Lastly information about inter-particle potentials, contact laws, interfacial energy, elasticity, and atomic potentials is required from the atomistic and electronics scale. Once this information has been collected quantities such as the elastic moduli, high rate mechanisms, creep and shrinkage, crack nucleation, and crack growth can be bridged to the macroscale continuum model.


References

  1. "Concrete Facts." Concrete Helper- A Concrete Industry Resource. N.p., 28 Oct. 2011. Web. 07 Feb. 2017.
  2. Zhang, Mingzhong, and Andrey P. Jivkov. "Microstructure-informed modelling of damage evolution in cement paste." Construction and Building Materials 66 (2014): 731-742.
  3. Burcham, Megan. “Multiscale structure-property relationships of ultra-high performance concrete.”
Personal tools
Namespaces

Variants
Actions
home
Materials
Material Models
Design
Resources
Projects
Education
Toolbox