Deterioration of concrete due to ASR: Experiments and multiscale modeling

Abstract

The process of ASR (Alkali-Silica Reaction) induced expansion and damage in pavement concrete specimens is investigated using laboratory experiments and computational modeling. In the experimental program, the concrete specimens are subject to CS-CPT (climate simulation concrete prism test) to obtain ASR induced expansion with and without external supply of alkali. The dissolution rates of the granodiorite used in the concrete mix and the gel formation rates are determined under concrete-like conditions (pH 13.8, with/without Ca(OH)₂ and NaCl) at different temperatures. A micromechanics based computational model with aggregate-scale diffusion and reaction kinetics coupled to an Eigenstrain based micromechanics damage model is developed for the simulation of ASR induced expansion and damage. Data from the experimental program are used to calibrate and validate the computational model. Model predictions show that for the given concrete mixture, ASR induced expansion in the specimen exposed to water is predominantly governed by microcracking in the aggregate, while the expansion in the specimen subjected to external alkali supply is governed by microcracking in both the aggregates and the cement paste.

Introduction

Concrete structures containing reactive aggregates are susceptible to deterioration caused by the Alkali-Silica Reaction (ASR) occurring in the presence of sufficient moisture and temperature. During ASR, silica in the reactive aggregates react with alkali, calcium and hydroxide ions in the pore-fluid. The product of this reaction is an expansive ASR gel. In the presence of moisture, the ASR gel swells and generates a pressure leading to distributed microcracking in concrete. At the structural scale, these processes at the microscale are manifested as simultaneous expansion and degradation of the material. Thus, the processes involved in ASR induced expansion and deterioration of concrete consist of: i) the transport of hydroxide, alkali and calcium ions into the reactive aggregates, ii) the dissolution of silica and the formation of a swelling gel, iii) swelling of this gel in pre-existing defects, voids and pores of the aggregates and iv) finally microcracking due to expansion of the swollen gel, which leads to an overall expansion and damage of the concrete (see Fig. 14).

In Germany, the number of concrete highway pavements that were damaged by ASR over the last decades is unusually high [1]. This was the motivation for a series of research projects aimed at understanding the causes and influencing factors responsible for ASR in concrete pavements. The external supply of NaCl de-icing salts and pre-existing microcracks were found to have a significant impact on ASR induced deterioration of highway pavement concrete containing alkali-reactive aggregates [[2], [3], [4], [5]]. With the introduction of the German guideline ARS 04/2013 [6], new regulations and testing methods are now available to reduce damage due to ASR in concrete pavements.

Given the wide range of multiphysical processes and scales involved in ASR induced expansion and damage in concrete, several modeling approaches have been proposed in the past. An overview of these models can be found in [7,8]. In general, these models can be classified according to the scale (micro-, meso-, macro/structural scale) at which the physical mechanisms involved in ASR induced expansion and damage in concrete are resolved. ASR induced expansion and damage at the macroscale, i.e., the scale of concrete structure, is usually modeled using phenomenological approaches such as thermo-chemo-plastic models [9,10], thermo-chemo-damage models [[11], [12], [13]] and thermo-hygro-chemo-damage models [[14], [15], [16], [17], [18]]. In these models, the mechanisms at the concrete microscale are not explicitly considered, but are represented at the structural scale by parameters that are obtained using experimental data. These models have a good predictive capability in the context of durability analyses on the structural scale when correctly calibrated, but cannot provide additional insight into the physical mechanisms involved in ASR induced deterioration in concrete.

In contrast to the aforementioned models, the processes involved in ASR as well as microcraking mechanism at lower spatial scales can be described using microstructure characterizing models such as semi-analytical micromechanics models [[19], [20], [21], [22], [23], [24], [25]] and computational mesoscale models [[26], [27], [28]]. These microstructure characterizing models adopt different hypotheses regarding the mechanisms of gel production and expansion and can roughly be classified according to the manner in which the local expansive processes in concrete are represented. Expansion initiated in the cement paste matrix is assumed in [[29], [30], [31]], while in [[32], [33], [34], [35], [36], [37]] gel expansion occurs in the aggregate. In [[38], [39], [40]] it is assumed, that gel penetrates the Interfacial Transition Zone (ITZ), where it induces the gel pressure. As features at the level of aggregates are considered in these models, they are also capable of modeling ASR induced time-dependent swelling processes at this scale.

Due to the numerous physical mechanisms determining ASR kinetics, these modeling strategies differ in how the spatial distribution and the transport of the reactants and reaction products are characterized. One of the modeling approaches to describe the ASR kinetics neglects the rate of the Alkali-Silica Reaction and focuses on the transport processes in and around a reactive aggregate. Hence, in this class of models [32,41], the ingress of the reaction front into reactive aggregates is modeled as a diffusive process of moisture transport. A more detailed description of ASR is achieved by taking into account the transport of individual reactants and reaction products (e.g., alkali diffusion and gel permeation in [42,43].

In contrast to the aforementioned strategies focusing on transport processes in the concrete and the aggregates, a description of the chemical reactions involved in ASR neglecting the transport processes can still provide insights into the dependencies between the individual reaction components and the reaction products [44]. A more detailed characterization of ASR kinetics can be obtained by considering both aspects of ASR, i.e., transport and chemical reaction by means of diffusion-reaction equations [[45], [46], [47], [48]]. However, the availability of the required data for the choice of the kinetic constants and other relevant model parameters is scarce.

When modeling gel formation at the aggregate scale, the growth of existing distributed microcracks caused by the ASR gel pressure can be described in the framework of continuum micromechanics. Besides initial microcracks and weak zones along the ITZ resulting from autogeneous and drying shrinkage [49,50], such distributed microcracks are observed during fatigue loading [51] or in fiber reinforced cementitious composites [52,53]. Continuum micromechanics models have also been successfully used for the description of these processes characterized by distributed microcracking (e.g., [54,55]).

The major goal of the present study is to link specific experimental results with a numerical model for a multiscale analysis of pavement concrete subjected to ASR. The experimental program is described in Section 2 with the results presented in Section 3. Details of the developed multiscale modeling approach and the analysis of the model parameters are given in 4 Modeling of ASR and ASR induced concrete deterioration, 5 Analysis of model parameters, respectively. Comparison between the experimental observations and numerical predictions are presented in Section 6. Conclusions from the results of the presented study are drawn in Section 7.