Quasi-brittle porous material: Simulated effect of stochastic air void structure on compressive strength

https://doi.org/10.1016/j.cemconres.2020.106255Get rights and content

Highlights

  • A simulation procedure to study the effect of porosity on the compressive strength of a quasi-brittle material is introduced

  • Based on the simulated results, the relationship between compressive strength and air void fraction (porosity) is presented.

  • A linear relationship was achieved between the cubic root of porosity and the simulated compressive strength.

Abstract

The effect of porosity comprised of spherical air voids on the compressive strength of quasi-brittle material was studied via simulations. The simulated porous structures were based on pore size distributions of two mortar samples measured by X-ray microtomography. While the simulation method set practical limits on the size of sample, the base of the statistics was established by simulating 128 small structures generated by sampling from pore structures of two mortars. By studying the application of the classical strength-porosity formulas to the simulated data, a new simple model was formed. A linear relationship was achieved between the cubic root of air void fraction (porosity) and the simulated compressive strength. The reasons for scattering of simulated strength around fitted trend remained unresolved in this study; no clear dependence on pore number or other distribution properties was observed. With the presented simulation approach, the dependence of compressive strength on porosity is achieved independently of disturbances that occur in experimental studies creating understanding of compressional behavior of low porosity materials.

Introduction

The strength and behavior of a material depends on its internal structure. In particular, in the case of a porous material, it is known that porosity affects strongly e.g. elastic modulus, yield behavior and tensile and compressive strength. Concrete and other cement-based materials are widely used porous materials where porosity has a crucial role in the suitability of the materials for their numerous uses. For this reason, and in addition to the need to make these materials more environmentally friendly by utilizing additives, the effect of the porosity on the behavior of the material has been extensively studied for several decades. The importance of porosity, pore distribution and potential additives for the compressive strength of cement-based materials has been clarified in several studies, see for example, [[1], [2], [3], [4], [5], [6], [7], [8], [9], [10]]. These studies have shown that porosity has a strong effect on compressive strength and, in addition, in the case of additives, changes in compressive strength come at least in part due to a change in porosity. The phenomenological, experiment based, equations introduced by Balshin [11], Ryshkewitch [12], Schiller [13] and Hasselman [14] have been utilized to describe the dependence of the compressive strength of cement-based materials on porosity for decades. Analytical approaches have also been developed to describe this dependence [15,16].

The pores of cement-based materials are often classified into different categories based on their size and type: e.g. gel (<10 nm), capillary (0.005–10 μm), and air (macropores 0.005–5 mm) pores. Also, the cracks due to shrinkage at aggregate-cement interface are common. The ranges of pore categories are not strictly defined and different sources cite slightly different ranges, see, e.g., [[17], [18], [19], [20], [21], [22], [23]]. Capillary pores and other larger pores are found to be responsible for reduction in strength of cement paste, while the effect of gel pores are noticed to be negligible [1,4]. Several measurement methods have been applied to the porous structure characterization of cement-based materials, such as mercury intrusion porosimetry (MIP), vapor sorption, scanning electron microscopy, and X-ray tomography. Frequently used mercury intrusion as well as vapor sorption techniques are indirect methods, which mean that the interpretation of the results usually requires some assumptions and theoretical simplifications [[24], [25], [26], [27]]. Mercury intrusion technique allows pore sizes over a broad range to be measured, whereas the vapor sorption techniques are more sensitive to gel scale pores. In MIP, the closed pores remain nonintruded, while for those large pores, which are only accessible by very narrow throats, the size is misinterpreted; this mechanism is commonly referred to as the “ink bottle” effect [22]. To observe a direct physical structure of the microstructure different imaging techniques have been utilized. Scanning electron microscopy produce images of microstructure features from 2D cross sections with resolution of submicron ranging from capillar porosity scale [23,28,29]. For the acquisition of 3D information of the pores, some studies using X-ray computed tomography have been presented [[30], [31], [32]]. Micro-tomography provides resolution in the order of few microns.

In recent decades, the finite element method has been extensively utilized in the study of the behavior of cement-based heterogeneous materials. These studies are focused on, e.g., the dependence of the elastic properties on the aggregate content and on the shape and distribution of the aggregate, see for example, [33,34]. In addition to the elastic properties, the strength and fracture behavior of the material and its dependence on aggregate properties have been considered, e.g., in references [[35], [36], [37], [38], [39]]. The most common approach in these studies is to divide the cement-based material into the cement matrix, the aggregate, and the interfacial transaction zone, each of which has its own material properties. In addition to the effect of aggregate, the effect of porosity on the mechanical behavior of cement-based material has also been studied by the finite element method [40,41] and by the discrete element method [42].

This paper aims to investigate the effect of variations in porous structure on the compressive strength of quasi-brittle material through a numerical approach. X-ray microtomography was used to determine 3D pore structure of two mortar samples, without and with 10% green liquor dregs (GLD). GLDs, which are side streams of pulping mills containing environmentally hazardous metals, are a recycling challenge. Its treatment and recycling are under investigation by several parties. However, in this study, the role of GLD as an additive of mortar is limited to the generation of different structures for porosity studies. To simulate the compressive strength, 128 cubes with the sides of 0.5 mm and different pore volume ratios, distributions and positionings were constituted based on the measured pore size distributions. Due to the cube size of the simulations and the measurement resolution the pore sizes were limited to between 20 μm and 500 μm. The statistically relevant number of pore structures was used in the simulations to reveal the dependence of compressive strength solely on porosity, without interference with other factors due to composition and manufacturing process.

Section snippets

Preparation of mortar samples

Mortar samples with dimensions 40 mm × 40 mm × 160 mm were prepared according to the standard EN 196-1. Two different mortars were produced: Sample 1 consisted of CEN standard sand, water and cement and in Sample 2 10% of cement weight was replaced by dried and milled green liquor dregs. The six parallel samples with size approximately 5 mm × 5 mm × 5 mm were cut from the produced samples.

X-ray tomography imaging and image analysis

Internal 3D-structure of mortar samples was measured using SkyScan 1172 X-ray microtomographic scanner;

Constitutive model

The constitutive model used is the damaged plasticity model available in Abaqus [46]. The original model is presented in the reference [47] and its modifications in the reference [48]. The stress-strain relation is governed byσ=1dD0elεεplwhere σ, d, D0el, ε and εpl are the Cauchy stress, scalar stiffness degradation variable, undamaged elastic stiffness of the material, total strain and plastic strain, respectively. The stress-strain behavior corresponding to the undamaged (d = 0) material

Simulation

In the simulations, the side of the subcube was 0.5 mm. An example of structure is shown in Fig. 4. The x3-directional displacement was set to zero at the bottom of the sample and 2 × 10−3 mm displacement was applied into negative x3-direction at the top of the sample, see Fig. 4. Displacements in the x1- and x2-directions were not restricted at any point in the subcube to ensure equal freedom in the x1- and x2-directions for each node. Examples of the simulated stress-strain behavior of eight

Results

The simulation procedure was performed on 128 subcube specimens. The measured pore distributions of Sample 1 and Sample 2 were the basis for two sets of structures, S1 and S2; 64 subcube specimens were taken from each. The simulated maximum stresses as a function of porosity and number of pores in the subcube specimens are presented in Fig. 12 (see also Table A.5, Table A.6 (Appendix)). The number of pores divides the subcubes into two separate groups; the number of pores in S1 varies between 0

Discussions

The simulations were performed on CSC's (the Finnish IT center for science) new supercomputer Puhti. For one standard sample simulation, a wall clock time of 36 h was set aside, during which time all samples exceeded the maximum stress point. The computation time required to simulate the entire compressible distance was considerably longer, varying from several days to two weeks. The wall clock time needed for the simulations was decreased by paralleling the computation, see Fig. 16. In the

Conclusions

A large number of simulations were performed to reveal the compressive strength-porosity relation for quasi-brittle material at low porosity region from zero to 3.5%. 128 stochastic structures that included only variability due to porosity, pore distribution and pore placement were generated for simulations based on measured distributions. Practical factors related to material, measurement method and simulation approach limited the porosity, the side length and the number of pores of a single

CRediT authorship contribution statement

Anna-Leena Erkkilä: Conceptualization, Methodology, Software, Formal analysis, Writing - Original Draft, Writing - review & editing, Visualization

Teemu Leppänen: Conceptualization, Methodology, Software, Validation, Formal analysis, Writing - Original Draft, Writing - Review & Editing, Visualization

Jussi Virkajärvi: Formal analysis, Data Curation, Writing - Original Draft and Review

Joni Parkkonen: Investigation, Data Curation, Writing - Original Draft

Leena Turunen: Resources, Project

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This research was supported by Regional Council of Central Finland and European Regional Development Fund (ERDF) - Sustainable Bioresidual Concrete. Simulations were performed by CSC's (the Finnish IT center for science) Puhti supercomputer and commercial software Abaqus which was licensed to CSC. Samples were prepared in the concrete testing laboratory of JAMK University of Applied Science.

References (59)

  • X. Du et al.

    Macroscopic effective mechanical properties of porous dry concrete

    Cem. Concr. Res.

    (2013)
  • P. Choi et al.

    Air-void structure, strength, and permeability of wet-mix shotcrete before and after shotcreting operation: the influences of silica fume and air-entraining agent

    Cem. Concr. Compos.

    (2016)
  • R. Kumar et al.

    Porosity, pore size distribution and in situ strength of concrete

    Cem. Concr. Res.

    (2003)
  • S. Diamond

    Mercury porosimetry: an inappropriate method for the measurement of pore size distributions in cement-based materials

    Cem. Concr. Res.

    (2000)
  • A.B. Abell et al.

    Mercury intrusion porosimetry and image analysis of cement-based materials

    J. Colloid Interface Sci.

    (1999)
  • M.A. Laskar et al.

    Some aspects of evaluation of concrete through mercury intrusion porosimetry

    Cem. Concr. Res.

    (1997)
  • N. Alderete et al.

    Pore structure description of mortars containing ground granulated blast-furnace slag by mercury intrusion porosimetry and dynamic vapour sorption

    Constr. Build. Mater.

    (2017)
  • R.S. Edwin et al.

    Quantitative analysis on porosity of reactive powder concrete based on automated analysis of back-scattered-electron images

    Cem. Concr. Compos.

    (2019)
  • S. Häfner et al.

    Mesoscale modeling of concrete: geometry and numerics

    Comput. Struct.

    (2006)
  • P. Wriggers et al.

    Mesoscale models for concrete: homogenisation and damage behaviour

    Finite Elem. Anal. Des.

    (2006)
  • J.P.B. Leite et al.

    Computational model of mesoscopic structure of concrete for simulation of fracture processes

    Comput. Struct.

    (2007)
  • X.Q. Zhou et al.

    Modelling of compressive behaviour of concrete-like materials at high strain rate

    Int. J. Solids Struct.

    (2008)
  • I. Comby-Peyrot et al.

    Development and validation of a 3d computational tool to describe concrete behaviour at mesoscale. Application to the alkali-silica reaction

    Comput. Mater. Sci.

    (2009)
  • N. Benkemoun et al.

    Anisotropic constitutive model of plasticity capable of accounting for details of meso-structure of two-phase composite material

    Comput. Struct.

    (2012)
  • F. Montero-Chacón et al.

    Mesomechanical characterization of porosity in cementitious composites by means of a voxel-based finite element model

    Comput. Mater. Sci.

    (2014)
  • T.T. Nguyen et al.

    Experimental and numerical investigation of influence of air-voids on the compressive behaviour of foamed concrete

    Mater. Des.

    (2017)
  • J. Lubliner et al.

    A plastic-damage model for concrete

    Int. J. Solids Struct.

    (1989)
  • A.S. Genikomsou et al.

    Finite element analysis of punching shear of concrete slabs using damaged plasticity model in ABAQUS

    Eng. Struct.

    (2015)
  • M. Kõrgesaar et al.

    Influence of mesh size, stress triaxiality and damage induced softening on ductile fracture of large-scale shell structures

    Mar. Struct.

    (2014)
  • Cited by (6)

    • X-ray microtomography applied to mortars: Review of microstructural visualization and parameterization

      2023, Micron
      Citation Excerpt :

      Alves et al. (2020) studied the effect of temperature rise on the microstructure of cementitious materials and used µ-CT, thermogravimetry and derived thermogravimetry to verify changes in the porosity and mass losses of the material, concluding that the increase in temperature leads to an increase in porosity. Erkkilä et al. (2021) used X-ray microtomography to study pore distribution, directing the study focus to the air pores and thus establishing the use of spherical pores in numerical simulations on compressive strength. Regarding the solid structure, X-ray microtomography is capable of providing analysis instruments for situations that involve structure damage as well as self-healing techniques.

    • Electrochemical modulation of pore structure inside structural supercapacitor solid electrolyte

      2022, Microporous and Mesoporous Materials
      Citation Excerpt :

      According to the desorption isotherms, the pore size distributions of pore structure inside modulated PAA cement electrolytes are analyzed by Barret-Johner-Halenda (BJH). From Fig. 2c, the pore size distribution inside modulated PAA cement electrolyte gradually widens and amount of macroporous (50 nm–300 nm) increases with the increase of CV modulation operation voltage, which can influence mechanical performance of PAA cement electrolytes [32,33]. Based on the desorption isotherms in the relative pressure (P/P0) range of 0.05–0.30, the specific surface areas of pore structure inside PAA cement electrolytes are calculated.

    • On the role of compaction in disputes over the quality of the supplied concrete

      2023, Materials and Structures/Materiaux et Constructions
    View full text