Alkali-silica reaction in concrete: Mechanisms, mitigation and test methods

Construction and Building Materials 222 (2019) 903–931

Contents lists available at ScienceDirect

Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat

Review

Alkali-silica reaction in concrete: Mechanisms, mitigation and test methods R.B. Figueira a,⇑, R. Sousa a, L. Coelho b, M. Azenha c, J.M. de Almeida b,d, P.A.S. Jorge b, C.J.R. Silva a a

Centro de Química, Universidade do Minho, Campus de Gualtar, 4710-057 Braga, Portugal CAP/INESC TEC—Technology and Science and FCUP—Faculty of Sciences, University of Porto, 4169-007 Porto, Portugal c ISISE, Escola de Engenharia, Universidade do Minho, Campus de Azurém, 4800-058 Guimarães, Portugal d Department of Physics, School of Sciences and Technology, University of Trás-os-Montes e Alto Douro, 5001-801 Vila Real, Portugal b

h i g h l i g h t s

g r a p h i c a l a b s t r a c t


Review of the most recent and

relevant achievements concerning ASR process.
Conditioning factors, diagnostic and preventive measures of ASR were debated.
Proper models to monitor ASR is a major concern for concrete service life prediction.
Several efforts have been devoted in understanding the fundamentals of ASR.

a r t i c l e

i n f o

Article history: Received 15 March 2019 Received in revised form 15 July 2019 Accepted 19 July 2019

Keywords: ASR Concrete Expansion

a b s t r a c t In the last few decades, the alkali–silica reaction (ASR) has been reported as one of the major concrete concerns regarding durability, leading to high maintenance and reconstruction costs. The occurrence of ASR in numerous concrete infrastructures all over the world points to the need for research regarding measures for its detection in an initial stage (and further mitigation) either in new or existing structures. Furthermore, the chemical and physical mechanisms for ASR remain poorly understood. This lack of knowledge leads to incapacity to assess risk, cost-effectively predict service life, and efficiently mitigate the deterioration process due to ASR in concrete structures. This manuscript aims to review the most recent and relevant achievements and the existing knowledge concerning the reaction mechanisms of ASR. Additionally, this manuscript is focused on the conditioning factors, diagnostic and prognostic methodologies, preventive measures and test methods (including their limitations) of ASR conducted at an academic level. The perspectives for future research challenges are also identified and debated. Ó 2019 Elsevier Ltd. All rights reserved.

Abbreviations: AAR, Alkali–Aggregate Reaction; AET, Acoustic Emission Test; ASR, Alkali–Silica Reaction; AMBT, Accelerated Mortar Bar Test; BFS, Blast Furnace Slag; C-AS-H, Calcium Aluminium Silicate Hydrate; CIM, Chemical Index Model; C-S-H, Calcium Silicate Hydrate; CPT, Concrete Prism Test; DIC, Digital Image Correlation; FA, Fly Ash; HRM, High Reactivity Metakaolin; MK, Metakaolin; NZ, Natural Zeolite; LSP, Limestone Powder; PC, Portland Cement; RCS, Reinforced Concrete Structures; RH, Relative Humidity; SCM, Supplementary Cementitious Materials; STD, Stiffness Damage Test; SF, Silica Fume; w/c, Water to Cement ratio. ⇑ Corresponding author. https://doi.org/10.1016/j.conbuildmat.2019.07.230 0950-0618/Ó 2019 Elsevier Ltd. All rights reserved.

904

R.B. Figueira et al. / Construction and Building Materials 222 (2019) 903–931

Contents 1. 2.

3.

4. 5.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Concrete degradation due to alkali-silica reaction (ASR) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. ASR mechanisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. Conditioning factors of ASR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.1. Reactive aggregate content . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.2. Alkalis and soluble calcium source . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.3. Moisture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.4. Impact of alkali leaching . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.5. Effect of irradiation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3. Diagnostic and prognostic methodologies of ASR – State of the art . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4. Preventive measures of ASR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4.1. Reduction of the alkalinity of the interstitial solution of the concrete . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4.2. Use of non-reactive aggregates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4.3. Concrete moisture control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4.4. Modification of expandable properties of ASR gel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4.5. Mineral additives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Test methods – State of the art and their limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Diagnosis methodology and sampling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. Aggregate assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3. Limitations of standard tests for aggregates reactivity assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4. Tests for ASR assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ASR modelling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions and recommendations for further research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1. Introduction It is undeniable that concrete is the most used construction material worldwide. Consequently, this material is actively present in the economy of most countries, leveraging large investments, thus leading to its technological development. Furthermore, studies that contribute to increase the service life of reinforced concrete structures (RCS) are effective contributions to the conservation of non-renewable natural resources, thus increasing the importance of the environmental performance of this material. The increase of the service life of concrete and RCS will have direct impact in the environment and in the population health by reducing depletion of natural resources, including metals and fuels used to manufacture them, pollution due to escaping products from corroded equipment or due to a corrosion product itself. The choice of concrete as building material is primarily justified by its durability, high fire resistance, low-cost, benefit ratio and low maintenance. Nevertheless, aggressive environments may lead to premature degradation of concrete and loss of structural integrity of concrete or RCS. The degradation of concrete properties is a complex process and results from the combination of several factors namely the physico-chemical properties of the concrete and by the environment to which the concrete is exposed. The main degradation processes of RCS include carbonation [1], alkali–silica reaction (ASR) [2], ingress of chloride ions [3] and/or depletion of oxygen [4], steel corrosion [5], spalling and debonding of concrete [4]. The ASR, identified in 1940 [6], is the chemical reaction between the reactive constituents of the cement aggregates and the alkali (K+ and Na+) and hydroxyl (OH–) ions present in the interstitial concrete pore solution [2,7,8]. ASR involving rocks and minerals containing reactive silica forms are currently considered as the cause of early deterioration of an increasing number of concrete structures. Moreover, ASR, which causes concrete expansion and fissures, significantly favours other deterioration processes, namely, the corrosion of the reinforcement in RCS. ASR induced damage has been widely studied in the last forty years. Several papers have been devoted to the comprehension of the chemicalphysical mechanisms involved [2,7,9–20] in ASR, some were

904 905 905 905 905 906 909 909 910 910 911 912 912 913 913 913 918 918 918 921 921 922 923 923 924

focused on reviewing the state-of-the-art [21–25] while others were devoted to modelling concrete expansion induced by ASR [15,26–37]. Preventive measures, including limiting the alkali content in concrete, the use of supplementary cementitious materials (SCMs) such as fly ashes (FA), granulated blast-furnace slag (BFS) and silica fume (SF) [18,19,38–49], and lithium compounds [16,50–54] have been widely studied. SCMs controls the ASR by reducing the amount of alkalis available, e.g., by reducing the concentration of Na+, K+ and OH– ions in the interstitial concrete pore solution and therefore limiting their availability to react with the aggregate [22,23,25]. Moreover, it has been reported that the ability of SCMs to bind alkalis seems to be strongly related with the CaO/SiO2 ratio of the SCMs [22]. SCMs that are high in silica and low in alkali and calcium, tend to be more effective in reducing the concrete pore solution alkalinity and consequently mitigating the ASR [23]. One of the advantages of these SCMs is that they are generally used at low replacement rates [22]. On the other hand, SCMs with increased amounts of alkali and calcium are used at higher replacement rates. Hong and Glasser [55], in 2002, showed that alumina play an important role in determining the alkali-binding capacity of SCMs. However, the precise role of alumina is not obvious from the data available [22,55]. Despite all these developments, the occurrence of ASR in numerous concrete infrastructures of different types; including bridges, dams, spillways and buildings all over the world [18,56], is a reality and points to the need for research work for its detection in initial stages and promoting mitigation, either in new or existent structures. This manuscript aims to review the most recent and relevant achievements and the existing knowledge concerning the reaction mechanisms of ASR. The relevant manuscripts published in the last few decades, until 2018, concerning the fundamentals of ASR mechanism are reviewed and debated. Regarding the use of SCMs the relevant manuscripts published about the use of lithium compounds, ground BFS, FA, SF and metakaolin (MK) for ASR are mitigation are also included. Additionally, the conditioning factors, diagnostic and prognostic methodologies, preventive measures and test methods (including their limitations) of ASR conducted

R.B. Figueira et al. / Construction and Building Materials 222 (2019) 903–931

at an academic level are discussed. The perspectives for further research and future research challenges are also identified and discussed. 2. Concrete degradation due to alkali-silica reaction (ASR) 2.1. ASR mechanisms The fundamentals of ASR mechanisms have been widely discussed in several text books and manuscripts [2,7,11–14,57]. The profusion of literature and information concerning ASR can be mainly credited to the multiplicity of ASR impacts on concrete that depend on several factors namely the stoichiometry and nature of alkalis [58–60], reactive silica [26,44,61], humidity [62–64], temperature [36,59,63,65,66], concrete stiffness [67–70] or aggregate size [26,71,72]. In 2015, F. Rajabipour et al. [17] published, perhaps one of the most complete reviews about the ASR mechanisms. The chemical reactions together with specific thermodynamic information were discussed. Generically, the authors stated that the ASR mechanism resulted from several sequential reactions (Fig. 1), namely: (a) dissolution of silica, (b) formation of sol, (c) formation of gel, and (d) swelling of the gel. According to Multon [73], the expansion of the gel includes two phases namely alkali-silica gel hydration and increased volume followed by the gel diffusion in the cement paste. During the hydration of alkali-silica gel the chemical potential difference between the solution within the gel and the concrete pore solution induces higher water adsorption and expansion of the gel [57]. The gel swelling behaviour was explained by Diamond based on osmotic pressure [57]. The most relevant manuscripts published in the last few decades concerning the fundamentals of ASR mechanisms as well as the main conclusions are summarised in Table 1. In the last few decades, significant progresses have been achieved in comprehending the micromechanics of ASR [2,7,10,11,13,14,17,51,79]. This has led to the development of predictive models suitable for laboratory samples [27,28,36,78,80]. A number of macroscopic models [30,34,81] have also been developed for real structures management. Though several expansion mechanisms have been proposed and advanced the actual stateof-art (Table 1), the capacity to comprehend, predict and describe the development of ASR remains limited and incomplete [26,30,32,36,44,51]. The development of suitable models to monitor ASR deleterious effect is a major concern to understand the mechanism of ASR expansion and predict the structural service life degradation [36,82–84].

905

2.2. Conditioning factors of ASR The ASR is essentially the attack to certain forms of reactive silica unstable in an environment with high pH due to the presence of sodium (Na+), potassium (K+) and hydroxide (OH–) ions in the concrete pore solution producing a hydrous alkali silicate gel [85]. Generically, the ASR may be described by the neutralisation reaction of the groups -Si-OH by the OH–, Na+ and K+ ions (eq. (1.1)) followed by the rupture of the siloxane bridges ASiAOASiA (Eq. (1.2)):

BSiAOH + OH— + Xþ ! BSiAOAX + H2 O

ð1:1Þ

where X is the K+ or the Na+ ions

BSiAOASiB + 2OH— ! BSiAO +  OASiB + H2 O

ð1:2Þ

The kinetic of the silica attack will depend on the structure of the silica and on concentration of the alkali hydroxides in the concrete pore solution. Well crystallised or dense forms of silica (e.g. quartz) are generally non-reactive in alkaline solutions [86,87] since the alkali hydroxide attack occurs essentially on its surface (eqs. (1.1) and (1.2)). The penetration rate of the alkali hydroxides into the well crystallized quartz structure is low due to its reduced internal surface area. In quartz with a cryptocristaline structure it is higher due to the increase of the superficial area [86]. The same behaviour is observed in large grains of quartz that are deformed or have defects suggesting that no siliceous aggregate can be considered initially inert with regard to ASR [86,88,89]. The information available allows to state that ASR only takes place in certain conditions. The four main requirements that must be simultaneously satisfied for the development of ASR in concrete, are: (i) the existence of reactive aggregates with concentrations within a critical range i.e., the source of reactive silica, (ii) high alkalis concentration, or more precisely high OH ions concentration in interstitial concrete pore solution (for silica attack), (iii) a source of soluble calcium, such as portlandite, to react with dissolved silica and form the deleterious gel, and finally (iv) high humidity conditions as the access to moisture allows the gel expansion [17,90,91]. If any of the mentioned factors does not exist then the ASR does not occur [23,65,92]. Therefore, all the specifications recommended and mitigation methods to prevent depreciation by ASR seek to exclude at least one of these. 2.2.1. Reactive aggregate content The aggregates for concrete may come from igneous, metamorphic or sedimentary rocks [93]. The presence of reactive aggregates to alkalis is a condition for the occurrence of ASR. According to Kurtis et al. [51], in theory, any aggregate having silica in its

Fig. 1. Sequential steps of ASR (adapted from [17]).

906

R.B. Figueira et al. / Construction and Building Materials 222 (2019) 903–931

Table 1 The most relevant manuscripts published in the last few decades concerning the fundamentals of ASR mechanism. Year

Ref.

Main results and conclusions

1987

[10]

1991

[13]

2005

[8]

2006

[14]

2007

[15]

Pozzolanic and ASR phenomena were briefly reviewed. A mechanism was proposed. It was concluded that the presence of Ca(OH)2 aggravates the deleterious effect of ASR. The fundamentals of ASR were related with the test of a suspect aggregate. A chemical method was proposed to test the alkali-silica reactivity of the aggregate itself. A mechanism for the damage induced by the ASR was proposed. The mechanism was decomposed in three periods: (i) latency, (ii) swelling and (iii) cicatrisation. A three-step model was proposed claiming that the ASR did not caused the deterioration of concrete, if the ASR is completed before the formation of the reaction rim. A model to predict the chemical evolution of concrete affected by ASR was proposed to describe the influence of the reactive aggregate grading on ASR. Review focused on the mechanisms by which SCM controls the ASR. Composition effect and test methods for detection of SCM amount to minimize the risk of ASR expansion were debated. Review of ASR in concrete including background, chemistry, affecting factors, and symptoms. None of the methods available for ASR assessment was considered ideal to assess the alkali-silica reactivity of an aggregate, instead several methods should be combined. Review on the properties and ASR-influencing mechanisms of polymer additions was performed together with a critical review of the existing literature. Review focused on the mechanisms and symptoms of ASR. It was concluded that further research should be conducted involving the choice of materials to prevent ASR appearance. Development and model validation to describe ASR. Two mechanisms, one induced by the chemical reaction and another induced by the external loads, were considered and described. Micro chemo–mechanical model was developed to predict the expansion of concrete structures in service. The model was reliable and effective in predicting the ASR expansion. The current state of art, the existing knowledge gaps concerning to ASR mechanisms, the role of aggregate properties, pore solution composition and exposure conditions were discussed. Efforts were made to develop a mathematical framework in order to gain deeper knowledge on the kinetics of the chemical reaction governing ASR. Changes in the concentrations of alkali ions were explained by a 1st order reaction. The rate of the ASR in systems containing highly reactive aggregate was also discussed. A calculation method to predict the expansion induced by ASR was proposed. The calculation and experimental results showed that the proposed model can predict the ASR expansion. Recent findings on the effects of coal FA as a SCM on ASR of reactive aggregates were debated. ASR mitigation mechanisms of FA were also analysed.

[26]

2011

[22]

2013

[74]

[75]

[76]

[30]

2014

[28]

2015

[17]

[32]

[77]

2016

[78]

2018

[18]

constitution has the potential to participate in ASR. The alkali reactivity of aggregates depends on their geological origin, mineralogical composition and texture. Reactive aggregates contain reactive forms of silica that react quickly with sodium and potassium hydroxides (e.g. opal, tridymite, cristobalite, acid volcanic glass) and those which react slowly (e.g. chalcedony, cryptocrystalline quartz and strained quartz) [53,94,95]. Furthermore, aggregates from the same rock, may show different potential reactivity [96]. Ponce and Batic [61] showed that the mineralogy and fabric of the rocks involved are responsible for different demonstrations of ASR in the aggregate and in the interfacial transition zone. The methodologies that are generally used to detect the reactivity of the aggregates to the alkalis are based, essentially, on the performance of petrographic tests together with mortar prism expansion tests [97].

In the last two decades, numerous manuscripts have been focused on the pessimum effect [19,98–100] which is generally described as the balance between the alkali hydroxide concentration of a pore solution and the aggregates reactivity. It has been showed, for the first time, by Stanton [6] that a definite proportion of certain reactive siliceous aggregate induced the highest concrete expansion. However, that expansion decreased when the content of the reactive aggregate in the concrete was increased or decreased from that proportion – pessimum effect. It was also reported that, for a fixed proportion of reactive aggregate, the expansion was maximum at a certain grain size. Furthermore, that expansion decreased when the grain size increased or decreased from the pessimum size. Several papers have discussed the effect of the particle size of reactive aggregates [72,101,102] as well as the pessimum effect for numerous types of aggregates in a wide interval of particle sizes [26,71,72,98,103]. Some reported irrelevant expansion when the sizes of the reactive particles were under 50–160 lm [71,99,104]. Multon et al. [71], concluded that small particles (lower than 160 lm) did not cause expansion while coarse particles (0.63–1.25 mm) showed the highest expansion. These authors also showed that when the mortars contained two sizes of aggregates (0–80 lm and 1250–3150 lm) the ASRexpansion decreased with the quantity of small particles [71]. Several mechanisms have been proposed to explain the pessimum effect regarding the size and the proportion of reactive aggregate [19,71,98,100,105]. In 2009, Ichikawa reported a study focused on the pessimum proportion and size effects based on an ASR model. The author proved that using very reactive siliceous minerals under antipessimum conditions was very effective for ASR inhibition and to improve the strength and durability of concrete structures [98]. One year later, Multon et al. [71] proposed two models to study the correlations between the expansions measured and the alkali and reactive silica contents and the size of the aggregates. The authors assessed the pessimum effect of the reactive aggregate size. The first model proposed predicted the expansions of the mortars containing two sizes of reactive aggregates. In the second model, the authors introduced a time dependence of the expansion showing that, for a certain test period and a given alkali content, the expansion was maximised in a specific range of aggregate size – pessimum effect. Three years later, Gao et al. [100] investigated and quantified the combined effects of aggregate and specimen sizes on ASR expansion together with the influence of the reactive silica content of aggregate. The authors concluded that the pessimum size effect of stabilised expansion was not an intrinsic phenomenon of ASR expansion. However, it depended on the size of the specimen used to perform the expansion test as for the largest specimens, the pessimum effect was not detected. Moreover, the data obtained allowed the improvement a microscopic model. This model took into account the scale effect through an empirical relationship. The calculations were in agreement with the experimental expansion tests performed for ratios, between specimen and aggregate sizes, higher than 10. The information already published suggests that the mechanism of the pessimum behaviour, considering the proportion of reactive silica, seems to be well-explained [17]. Furthermore, the influence of the proportion together with the size of reactive aggregate on ASR deleterious effect already achieved a high knowledge level. The proof of that is the number of studies [26,71,102,106,107] already available in assessing the synergetic effects between different factors (e.g. aggregate size, nature and reactivity). 2.2.2. Alkalis and soluble calcium source The hydration of Portland cement (PC) produces an interstitial concrete pore solution composed mainly of calcium hydroxide (Ca(OH)2), potassium hydroxide (KOH) and sodium hydroxide (NaOH). The concentration of OH–, Na+ and K+ ions depends on

R.B. Figueira et al. / Construction and Building Materials 222 (2019) 903–931

the alkali content in the anhydrous cement. The alkaline compounds in the clinker are constituted basically by alkali silicates, alkali aluminates and alkali sulphates [90]. According to Fournier and Bérubé [21,89], the alkali sulphates are solubilised in the mixing water, while the aluminates and the silicates are slowly released during the hydration process. The reactive aggregates are attacked at a significant rate, over time, if the concentration of OH– ions in the interstitial concrete pore solution reach a level between 0.2 and 0.25 M [108,109]. Although alkalis in concrete are mainly from cement, it should be kept in mind that alkalis from other sources, such as de-icing salts, additives (e.g. pozzolanic materials [98,110,111], SF [107,112–115], FA [18,38,43,46,48–50], etc.) and mixing water, may contribute to the alkalinity of the concrete interstitial solution. Moreover, the alkali-rich aggregates [8,25,116–121], such as micas and feldspars, may contribute to an increase in the alkali content of concrete [91,122]. The risk assessment of ASR linked to alkali content, considering the contribution of sodium and potassium, in PC can be expressed in terms of equivalents of sodium oxide (Na2Oeq) using the following equation [90]:

Na2 Oeq = Na2 O + 0.658 K2 O

ð1:3Þ

where the 0.658 is the ratio between the molecular weight of Na2O e K2O. According to Hobbs [90], concrete with a high alkali content will have a pH ranging between 13.5 and 13.9, while the concrete with low alkali content will have a pH between 12.7 and 13.1. The high concentration of alkalis optimises the appearance of ASR due to the increase of OH– concentration and therefore the increase of pH [123,124]. Two types of reactions may contribute to the increase of pH, i.e. the hydrolysis of the anions of weak acids such as silicates (Eq. (1.4)) or the formation of insoluble calcium salts like sulphates instead of Ca(OH)2 (Eq. (1.5)).

Xn + H2O ! HXðn1Þ + OH—

ð1:4Þ

2/n Xn + Ca(OH)2 ! CaX2=n + 2OH—

ð1:5Þ

The first phase involves the attack of the aggregates by the hydroxyl and alkali ions. The aggregates containing silica poorly crystallised are the most vulnerable. The attack on wellcrystallised silica forms occurs mainly on the surface [86] and is a reaction with slow kinetics producing silicate ions that pass to the fluid phase. On the other hand, the poorly crystallised silica allows the penetration of the OH–, Na+ and K+ into the interior, causing its dissolution. This process follows, basically, the reactions described in Eqs. (1.1) and (1.2). The second phase involves the chemical reaction of the OH–, K+ and Na+ with silica (attack on the silanol groups and siloxane bridges) forming the gel (Eqs. (1.1) and (1.2)). To maintain the electroneutrality, the Na+ and K+ ions diffuse into the ionized aggregate being adsorbed, and since they tend to attract water, their presence increases the tendency to form a hydrated gel. However, not every alkali existent in the concrete participates in the ASR. Only the ones that are free (e.g. not fixed in the aggregates or in the calcium silicate crystalline network), known as active alkalis, can participate [125]. Bérubé et al. [91,126] confirmed that significant amounts of alkalis can be released, through time, to the interstitial concrete pore solution by aggregates, particularly by feldspar-rich ones. The authors considered that the main cause for the development of ASR was the use of cement with high alkalis content, higher than 0.60% Na2Oeq. However, some studies showed alkali contents between 0.45% and 0.60% of Na2Oeq may react while others showed that contents of 0.40% or less rarely did [127]. According to the Engineering Technical Letter 06-2 [127] the 0.60% Na2Oeq represents a compromise between economic production and technical considerations stating

907

that the use of low-alkali cements alone is not sufficient to prevent ASR. Other authors showed that the best way to prevent ASR would be the imposition of a threshold value for the active alkali content of concrete [128]. In 2014, Kim and Olek published a two-part series study [129,130] focused on the kinetics and sequence of the chemical steps of the ASR process. The authors, discussed the experimental data obtained by the model reactant method. Several chemical and physical parameters were obtained in the early stages of ASR and information on the chemical kinetics parameters such as activation energies and dissolution rate constants were provided. The authors showed that the ASR was initiated by the silica dissolution due to OH– ions (step 1 of Fig. 2) and that the rate constants silica dissolution in the presence of Ca(OH)2 of NaOH solution were higher than those for KOH solution concluding that the Na+ ions accelerate the reaction more than K+ ions. Furthermore, the results indicated the dissolution of silica minerals are the rate limiting step of ASR. The results obtained about the chemical kinetics of the ASR process allowed the identification of the associated distinct chemical steps. The main steps proposed are schematised in Fig. 2. The mechanism proposed for ASR gel formation proposed by Kim and Olek is in agreement with a previous mechanism suggested by Hou et al. [85]. It is widely accepted that Ca(OH)2 has an important role in ASR as source of calcium ions (Ca2+) that react with dissolved silica forming the deleterious calcium-rich ASR gel rim around aggregate particles (Fig. 3). Ca2+, whose main source is portlandite formed by the reactions of cement hydration, leads to the formation of calcium, sodium and potassium silicate gels of variable composition around the aggregates. These gels absorb water molecules and expand. The swelling caused by this absorption will generate expansive forces and tensile stress causing microcracking near the reaction site (Fig. 3) and subsequent expansion and cracking of the concrete. The intensity of the expansive force changes with the composition of the gel. Recently A. Gholizadeh-Vayghan and F. Rajabipour [131] studied the impact of the ASR gels composition according to the ratios of Na/Si, K/Si and Ca/Si, at a RH of 95%, on the gels swelling behaviour. The authors concluded that when the ratios of Na/Si and K/Si (e.g. alkali content) in the ASR gels increase the free swelling of gels and water absorption also increases. For the Ca/Si ratio the effect was

Fig. 2. Schematic representation of the reaction steps proposed by the Kim and Olek [129].

908

R.B. Figueira et al. / Construction and Building Materials 222 (2019) 903–931

Fig. 3. Schematic representation of the mechanism of the ASR in concrete, adapted from [102].

not straightforward. For lower values (e.g. from 0.05 to 0.18) the water adsorption of the gels decreased. For Ca/Si ratios between 0.18 and 0.40 the gel swelling and water absorption increased. For higher values (i.e. above 0.40) the swelling ability of the gels gradually reduced with no swelling reported for Ca/Si ratio equal to 0.55. In the last decades, much research has been undertaken to explain the role of Ca(OH)2 on ASR [13,77,132-138] and the precise Ca2+ role in ASR is not unanimous. A review conducted by Feiteira et al. in 2013 [75] discussed the most relevant information concerning the models found in the literature to explain the role of Ca(OH)2 on ASR. The authors showed that those models are not in agreement concerning the calcium-rich ASR gel expansion causes in concrete. Nevertheless, there is agreement that this cation is essential for the development/rate of the ASR gel expansion [77]. Some authors such as Powers and Steinour [139] assumed that the Ca2+ concentration in concrete controlled the formation of expansive alkali-silica complex or non-expansive lime-alkalisilica complex. Others such as Chatterji [12] did not accept the non-expansive nature of the lime-alkali-silica complex and proposed that Ca2+ concentration would control the relative rates of diffusion of silica ions in and out of the reactive grains. Later, Wang [13] proposed a model in which the Ca2+ exchange for alkali ions on silica gel, forming a non-swelling lime-alkalisilica complex. It was also proposed that in this exchange, a partial releasing of alkali ions in a mechanism known as ‘‘alkali recycling” occurred regenerating these ions for further production of swelling alkali-silica complex [13,140,141]. In 1992, A. Poole proposed a model (Fig. 4) where it was suggested the breaking of the siloxane bonds (equation (1.2)) adjacent to the silanol groups (SiAOH) by distorting the silica structure and reducing the local stability of the lattice. The siloxane bonds break forming new bonds as water and Na+ and K+ ions penetrate the structure.

Calcium complexes or calcium silica hydrates (CASAH) are a stable and non-soluble variety of ASR gel; and in the absence of Ca2+, silica ions dissolve and remain in the solution. Leemann et al. [133] showed that the calcium reacts with dissolved silica, forming CASAH. The removal of the dissolved silica from the alkaline solution progresses until saturation of CASAH forms. However, since the calcium controls the silica dissolution, this dissolution endures while the calcium is present in the solution. Liaudat et al. [143] reported a model that took in account the role of calcium in the development of ASR and the relationship between the ASR product composition and its swelling capacity. Furthermore, three diffusion processes were taken into account. It was proposed that the alkalis (Na+, K+) and Ca2+ ions would diffuse from the pore solution to the aggregate interfaces or cracks. The OH– ions would attack certain forms of silica in the aggregates, dissolving it in the form of silicate ions that would diffuse in the opposite way. The proposed model was implemented and compared to experimental results, reproducing aspects such as the effect of calcium availability on silica dissolution, expansion development and ASR products on the diffusivity coefficient. Later, the same group [144] improved this model by including three main diffusion equations for Na+, K+, Ca2+ and silicate ions concentrations. The model accurately reproduced the propagation of ASR products. The main limitation found was the meso-scale used. Moreover, the reproducibility of the model on large-scale structures was not described. Bazant and Rahimi-Aghdam proposed a diffusion-based model [145] to estimate the diffusion of ASR gel into the transition zones between aggregate and cement paste and into the cracks. The model was afterwards calibrated and validated by using finite element fitting [146]. The authors obtained a good agreement between the data obtained from the laboratory tests and the proposed model. It was concluded that this model has the potential to realistically predict the multi-decade evolution of the ASR in large structures.

Fig. 4. Schematic of the model proposed by Poole with the formation of alkali-silica gel due to attack of sodium (Na+) or potassium (K+) ions. Only the Na+ ions are represented. Adapted from [142].

R.B. Figueira et al. / Construction and Building Materials 222 (2019) 903–931

Others, such as Maraghechi and collaborators [137] studied the kinetics and products of silica dissolution in terms of temperature, pH and calcium availability. At the same pH and temperature levels, it was shown that a portlandite saturated solution can slow down the dissolution of silica. Kleib et al. [147] proposed a detailed ASR mechanism based in 3 steps, adapted from Bulteel et al. [148], in which the OH ions present in the pore solution attack the silanol and siloxane groups. The continuous attack results in the formation of dissolved silica ions. These ions then precipitate by action of the cations forming CASAH and/or CA(Na,K)ASAH formation. The three steps of the proposed mechanism are detailed below: Step 1: The OH– ions in the pore solution attack the silanol and siloxane groups and the formation of silica-sodium/potassium stabilised forms occurs.

2SiO2 + OH— ! SiO5=2  + SiO5=2 H (silanol formation)

ð1:6Þ

SiO5=2 H + OH— = SiO5=2  + H2 O (silanol neutralisation equilibria) ð1:7Þ SiO5=2  + Naþ ;Kþ ! SiO5=2 Na; K (charge stabilisation)

ð1:8Þ

Step 2: Attack of SiO-5/2 ions by OH– which contributes to reduce the existent solid (SiO2).

SiO5=2  + OH— + H2 O ! H2 SiO4 2

ð1:9Þ

The presence of the ions H2SiO2 4 formed is determined by the acid-base equilibrium that is described as:

HSiO4 2 + H2 O $ H3 SiO4  + OH—

ð1:10Þ

Step 3: The cations in the pore solution of concrete react with the H2SiO2 4 forming CASAH and/or CA(N,K)ASAH precipitates

H2 SiO4 2 + Ca2þ + H2 O ! CASAH

ð1:11Þ

2H2 SiO4 2 + Ca2þ + 2Naþ + H2 O ! CANASAH

ð1:12Þ

2H2 SiO4 2 + Ca2þ + 2 Kþ + H2 O ! CAKASAH

ð1:13Þ

The information detailed above shows that in spite of different explanations provided to describe the Ca2+ ions role it seems that these ions are essential for ASR and are linked to a higher ASR development. Furthermore, all the information published indicates that the ASR mechanism relies on the control of the alkali and Ca2+ ions in the interstitial concrete pore solution. 2.2.3. Moisture The water effect on ASR expansions has been widely discussed in the literature [62,63,131,149–151]. According to Larive et al. [149], the water appears to have a double effect on ASR; as reactive agent since it influences the rate of expansion at the time of formation of the reaction product and as transport agent of different reactive species. Reducing the exposure to water may constrain or halt the reaction and expansion. However, the expansion will re-develop at a rapid pace when the concrete has retained the necessary RH (relative humidity). In 1987, Olafsson [152] showed that the limit of 80% of RH falls with the increase of the temperature. The author observed significant expansions for values of 70% RH at 38 °C. Tomosawa et al. [153], showed that there is a correlation between the alkali content and the RH level necessary for the development of ASR. The authors found, for the same concrete composition, that the increase in alkali content requires a lower RH to trigger ASR [153]. Fournier and Bérubé reported [89] that for ASR development concrete must be exposed to high humidity, over 80–85% RH. The same authors concluded that the ASR should

909

not occur when the concrete is exposed to low humid conditions, even with highly reactive aggregates [89]. Furthermore, it was reported that for concrete depths exceeding 10 cm regardless of external weather conditions, excepting desert climates, the RH was higher than 80% [89]. Multon [73] advocates that the expansion by ASR is not necessarily dependent on water ingress. Instead the expansion due to ASR may occur as a result of the moisture inside the concrete. This idea is supported by the appearance of cracking in dry areas compared to permanently immersed areas [73]. However, once the reaction has occurred and enough gel has formed, any increase in the humidity may lead to quick expansion and then to cracking [154], that is why wetting-drying cycles can endanger a structure [90]. Multon et al. [62] also studied the ASR expansion due to changes of the moisture conditions. The authors showed that, if the maximum expansion potential has not been reached (due to lack of available water), late water supply leads to new expansions in concrete. Furthermore, it was concluded that the introduced strains implied longitudinal compressive stresses reducing the longitudinal strains and causing expansion in directions free of restraint. Generally speaking, all concrete structures are susceptible to ASR excepting the ones where the concrete, by itself, can control the internal humidity [155–159] to values lower than 80% and that are permanently protected from the atmospheric conditions and other sources of moisture. Permeability, which is related to water to cement ratio (w/c), is one of the most important factors that control the concrete durability [160,161]. It is widely known that excessive water content, high w/c ratios, leads to strength reduction of cement mortar. However, deficient water contents affect the workability. Haach et al. [162] showed that increasing w/c lead to reduced values of mechanical properties and increased the workability. The publications found concerning the effect w/c on the development of ASR are contradictory [131,149,150,163]. Gillot [163] showed that lower w/c conducted to lower expansions while Bérubé and Fournier [89] showed that concrete expansion increases with the w/c decrease. Lindgård et al. [164] showed that lower w/c (i.e. 0.30) compared to higher ratios (i.e. 0.45 and 0.60) show lower internal RH. This was justified by the higher extent of self-desiccation and due to the higher concentration of ions in concrete pore solution. Moreover, the lower relative diffusion coefficient was justified based on the finer pore structure that would lead to slower water uptake. 2.2.4. Impact of alkali leaching Alkali leaching of aggregates has been reported for the first time in 1946 by Blanks and Meissner [165] and may be explained as a two-steps process. In the 1st step the transport of the alkalis to the prism surface takes place followed by the external reception of the alkalis (2nd step) [164]. In the last few years, some studies were focused on this problem and on the inclusion of alkali leaching phenomena in the most used test methods for ASR. In the case of CPT, alkali leaching may lead, for certain aggregates, to inaccurate conclusions. If the test methods used do not take into account the alkali ions leached out of the aggregates, false negatives may be obtained. Therefore, this phenomenon is of major concern in the application of ASR test methods. Alkali leaching may lead to reduced expansions during laboratory expansion and the specimens found as reactive in one test may be found as non-reactive in other tests or when used in the field. In 2006, Thomas et al. [166] showed that the ASTM C 1293 test, due to high alkali leaching by the aggregates (around 35% of the alkalis initially in the concrete), was not suitable to define the minimum level of alkali that could cause expansion neither the amount of mineral addition required. Later, Lindgard et al. [164,167] focused on CPT and studied the impact of different parameters (e.g. influence of specimen pre-treatment, exposure

910

R.B. Figueira et al. / Construction and Building Materials 222 (2019) 903–931

conditions and prism size) on alkali leaching and expansion. The main objective was to assess whether or not the CPTs were suitable for ASR general performance assessment. The authors showed that the CPT procedure could influence the final results. Furthermore, it was evidenced that generally higher amounts of alkalis were leached out and a correlation between the leakage and the prism expansion was established. The authors concluded that the major shortcoming of the CPT was the loss of alkalis during exposure. The authors also recommended that the development of performance tests for ASR potential assessment should consider test procedures that limit, compensate for or eliminate alkali leaching during testing. Drolet et al. [121] in 2017 studied the alkali leaching using 3 types of aggregates i.e. one alkali reactive, one alkali-free reactive and one alkali-free non-reactive. The study showed that the first type released up to 3% wt. of Na2O into the pore solution after one year, while the other two types did not. Furthermore, it was found that the amount of Na2O bounded to ASR gel was proportional to the ASR expansion, however there was no evidence that Na2O influenced the expansion of the gel. The two studies mentioned previously also considered the influence of temperature in alkali leaching and expansion. On one hand, Drolet et al. [121] reported, due to lower humidity conditions, higher expansion at 38 °C than at 60 °C. On the other hand, Lindgard et al. [164,167] showed that alkali leaching was similar for both 38 °C and 60 °C in the first 4 weeks of exposure. Nevertheless, beyond this time, the experiments at 60 °C showed that higher amounts of alkalis were leached out. In the same year, Costa et al. [168] studied the CPT at 38 °C. The authors showed that the higher the mass vapour the higher was the leaching of the alkalis. A correlation was found between the ratio test container volume / concrete volume and alkali leaching. The smaller the volume of vapour present in the container, the smaller was the alkali leaching. In 2018, Huang et al. studied different methods for mortars curing methods. The authors showed that the common curing method for PC concrete, due to the high humidity (>95%), was not proper for alkali-activated materials leading to alkalis leaching. It was also found that the soaking curing was not suitable for the alkaliactivated materials as the water penetrated into the sample increasing the leaching of alkalis. In the end, the authors concluded that the seal curing method (sealing the specimens with plastic film before curing) effectively prevented the leaching of alkalis being the most suitable curing method for alkali-activated materials. Studies focused on modelling the alkali leaching contribution on ASR were also found. For instance, Martin et al. [169] presented a chemo-mechanical model to assess structures affected by ASR. This model took into account several coupling effects including the role of alkali leaching depending on the diffusion within the samples studied. The authors considered that the alkali leaching was the cause of deviation in the estimation of the kinetics of expansion. Other authors such as Multon and Sellier [105] studied the impact of alkali leaching on expansion tests through a multiscale analysis. Considering the multi-scale modelling proposed, the authors concluded that it was possible to assess the causes of ASR scale effect due to alkali leaching. The authors showed that the alkali leaching must not be neglected when modelling ASR processes. Grymin et al. [170] proposed a mathematical model for the ASR development and alkali diffusion which was validated by experimental data. The proposed model took into account hygrothermal behaviour of cement-based materials and external alkali sources in the ASR development. Berra et al. [171] proposed a simple model to predict the effect of alkali leaching on ASR expansion. The proposed model was estimated based on different parameters including the initial alkali

content used, the threshold alkali level of the aggregate and long-term alkali contribution. The authors reported that the model was consistent with the field data and with ASR prevention criteria recommended by RILEM specifications. 2.2.5. Effect of irradiation Nuclear power stations and related industries rely on concrete structures as a protective material from nuclear radiation. The knowledge of nuclear radiation effect on concrete, namely c-rays and neutrons, is crucial for assuring the stability and safety of nuclear buildings. The properties of concrete are not affected up to a dose of 1010 Gy of c-rays, however, expansion of aggregates and the contraction of cement occurs when concrete is subjected to doses of fastneutrons higher than 1019 cm2, leading to deterioration of concrete. These threshold doses, attained under the enhanced conditions, correspond to >50 years of irradiation of concrete placed near the core of a nuclear power plant. However, lower doses of irradiation activate selected chemical reactions producing deterioration of concrete. This means that irradiation makes the aggregate more sensitive to ASR. The reactivity of quartz to alkali was found to increase 700 times for that the irradiation of a-quartz at doses of 1012Gy for c and b rays and 1020 cm2 for fast neutrons and one order of magnitude lower for amorphous quartz [15,172]. Critical doses for irradiation induced ASR of aggregates containing plagioclase are much lower: 108Gy for c and b rays and 1016 cm2 for fast neutrons [173]. The most pertinent reviews concerning the effects of nuclear radiation on concrete were conducted by Rosseel et al. [174] and Field et al. [175]. The first author reviewed the state of the art on the effects of nuclear radiation on concrete and its components considering the role of several factors such as temperature, creep, neutron energy spectra on radiation-induced volumetric expansion of aggregate-forming minerals were considered. Rosseel et al. [174] evidenced that fundamental and detailed comprehension of both separated and combined effects of irradiation, temperature and internal moisture on aggregate and hardened cement paste is still needed. Furthermore, the authors concluded that the stability of alkali-silica hydrates under c irradiation is unknown and requires further research. Field et al. [175] reviewed the current state of the art of neutron-irradiated concrete in light water reactor power plants and concluded that the radiation tolerance was dependent on the aggregate type. The authors confirmed that the siliceous aggregates showed the highest risk for deleterious effects of both irradiation and elevated temperatures on concrete. 2.3. Diagnostic and prognostic methodologies of ASR – State of the art Diagnostic and prognostic methodologies of ASR aim to establish whether or not the degradation was initiated by the ASR, determine the extension and impact on the structure and estimate the evolution of the service life of the structure [176]. It follows that the information about the concrete in a structure and an accurate diagnostic will contribute to establish suitable maintenance and repairing strategies. These procedures allow the adoption of preventive measures to mitigate the risk of ASR development [177,178]. The appearance of cracking and expansion together with other visual symptoms, such as localised crushing of concrete, extrusion of joint material, surface pop-outs, surface discolouration and gel exudations may give warning of ASR [176]. The main visual indication of ASR is the abnormal expansion and cracking on the surface of concrete in service, which is more or less pronounced according to the content of moisture [90,179]. Concrete without reinforcement shows random cracking (Fig. 5) while RCS

R.B. Figueira et al. / Construction and Building Materials 222 (2019) 903–931

911

Fig. 5. ASR induced map-cracking observed in two different dams in Portugal.

show cracking in two directions (Fig. 6). This pattern difference is mainly due to the fact that the cracking and expansion are restrained by the rebars. Cracking in two directions occurs mainly in RCS with reduced concrete cover and the cracking will match the distribution of the rebar net (Fig. 6). If the visual inspection of concrete raises suspicions of ASR, the methodology to confirm this preliminary diagnosis will involve several phases, namely: gathering information on the constituents of concrete, extraction of several samples in different areas of the structure and laboratory tests. All the information concerning the construction of the structure should be obtained, including the date of construction, diagrams with the location and type of reinforcement, origin of the materials, composition of the concrete, reports of previous inspections and/or interventions, and the dates of the appearance of the first signs of degradation [180]. One of the most important phases of the diagnosis is sampling by extraction of specimens from specific areas of the structures for later observation and laboratory testing. The number of samples required is dependent on the type and complexity of the structure. Generally, three areas should be selected in each structure (e.g. column, slab, foundation, etc.). These must be representative of both moderate and highly degraded concrete. Usually, a more representative diagnosis is obtained by performing few tests on a large number of samples. The analysis of the samples should include the record of the length and diameter of each sample, the type of aggregates and visual observation of specific elements such as filled pores, cracks, borders in aggregates, etc. Whenever possible the visual observation of the samples should be complemented with the binocular loupe observation. This observation will aid in the selection of specimens for mineralogical characterisation tests by X-ray diffractometry and microstructural characterisation by

scanning electron microscope combined with energy dispersive spectroscopy (SEM/EDS) [180]. A simple way to confirm the occurrence of ASR in concrete is by using luminescent methods [181–183]. These methods make use of the cationic exchange affinities of the products generated by the ASR which are essentially constituted by hydrated silicate of alkali and calcium [21,181,184,185]. However, luminescence is not sufficient to confirm ASR and random areas should be checked using SEM/EDS and X-ray diffraction mineralogical analysis [180]. The SEM/EDS analysis of concrete specimens is considered by many the safest methodology to confirm the presence of ASR and its level of progress [23,88,186–191]. Furthermore, SEM/EDS analysis allows the detection of typical morphologies of ASR products and assesses the presence of products from other concrete degradation processes [192–194]. In case of confirmation of ASR, the selection of analyses should serve as a complement to the diagnostic tests. Complementary tests to ASR diagnostic include physical tests of concrete specimens including water absorption and permeability, tests of compressive strength and modulus of elasticity, determination of soluble alkali content, residual reactivity tests on alkalis of concrete samples and confirmation of the reactivity of aggregates [195,196]. Once the ASR has been diagnosed in a certain structure, its future management should include the installation of an adequate observation system on selected sections. The selected sections should be inspected periodically to monitor cracking and expansion changes and internal RH. Generally, the structures should be inspected twice a year, spring and autumn, in order to avoid the impact of extreme temperatures [195]. The results obtained from monitoring and inspection plans allow to decide which repair procedures and implementation of mitigation strategies (e.g. sealers, coatings, or lithium) should be adopted. The need of having a correct diagnosis of the causes of a concrete structure degradation is decisive to choose the suitable repair method. Some cases of failure, when choosing a certain repair method, were justified by incorrect or incomplete diagnosis. For example, internal sulphatic reaction shows, in macroscopic terms, similarity with ASR which may lead to incorrect diagnosis and therefore unsuitable repairing methodologies. 2.4. Preventive measures of ASR

Fig. 6. Schematic representation of ASR cracking in two directions in RCS.

As mentioned in Section 2.2 the preventive measures of ASR seek to exclude at least one of the four requisites (e.g. source of reactive silica, high alkalis concentration, a source of soluble calcium and high humidity). The following sections will discuss in detail the main achievements on how to control each requirement. The use of chemical additives, to modify the expansive properties of the ASR gel, will also be discussed.

912

R.B. Figueira et al. / Construction and Building Materials 222 (2019) 903–931

2.4.1. Reduction of the alkalinity of the interstitial solution of the concrete The main source of alkalis in concrete is the cement. Therefore, the alkali content of the concrete is obtained from the alkali content of the cement (expressed as Na2Oeq.) multiplied by the cement dosage and addition of a factor which translates the content from other constituents. The control of the alkali content assumes that the manufacturers and/or suppliers of cement certify the average alkali content and its variability. This measure is already used in several countries as part of the concrete control and standardisation process. In addition to cement, other sources of alkalis for concrete such as pozzolan additions, aggregates, kneading water and additives should be considered. However, different recommendations can be found in the literature. For instance the RILEM TC 191 working group [180] recommends that the alkali content for values of above 40% of slag and 25% of ash should not be included in the alkalis concrete level. Another specification is the one from the Portuguese National Laboratory for Civil Engineering – LNEC Specification E 461 [97] which recommends that the alkali content of the concrete should be equal to the sum of the alkali content of its constituents, excluding FA and granulated BFS if present in a percentage higher than 30% of the total binder. The reduction of the alkali content of the interstitial concrete pore solution can be performed by (i) controlling the alkali content in concrete by limiting the content of concrete soluble alkalis or using suitable binders; (ii) using a low alkali cement (e.g. a total alkali content <0.6% Na2Oeq); and (iii) mineral addition to the concrete [23]. Some authors state that ASR is unlikely to occur when the alkali content of the cement is below 0.60% Na2Oeq [6] while others showed that ASR took place either in lab or in situ conditions when low-alkali cements are used [197,198]. It has been reported that the ASR expansion increases with the increase of alkalis content [199–201] however others showed that the ASR expansion declines with the increase of alkalis content [25,200,202,203]. Despite all these controversial studies, the threshold of 0.60% Na2Oeq is generally accepted. According to the ASTM C150 [204] this value is accepted as the maximum limit for cement to be used with reactive aggregates and is an optional limit when concrete contains deleteriously reactive aggregate. However, in 2008 Leemann and Lothenbach [205] showed that concrete mixtures produced with cements with similar Na2Oeq but different K/Na ration may expand differently. Therefore, the Na2Oeq parameter should be considered carefully when used to assess the potential reactivity of concrete mixtures. 2.4.2. Use of non-reactive aggregates Not all aggregates are susceptible to ASR and this process can be mitigated by using non-reactive aggregates. The aggregates are classified in three reactivity classes, namely: class I – aggregates are not alkali-reactive or very unlikely to be; class II – aggregates are potentially reactive to alkalis or the alkali-reactivity is uncertain; class III – aggregates are very likely to be alkali reactive. Aggregates that are essentially siliceous, or carbonates with a potentially reactive silica content, are known as class II-S or class III-S. Aggregates that are either essentially carbonate, or mixtures that include possibly reactive types of carbonate, are known as class II-C or III-C. Aggregates of mixed composition are known as class II-SC or III-SC [186]. Nevertheless, the aggregates classified as II or III may be used when, for example, mixed with other aggregates and the mixture is not alkali-reactive. The reactive silica content depends on several factors such as the temperature, type and concentration of solution in contact with silica, chemical composition, level of crystallinity, among others. A certain type of silica in concrete may be reactive in a certain environment and inert in another. Therefore, reliable methodologies are necessary to assess the alkali-reactivity of concrete

aggregates. An analysis based only on lithological aspects does not provide enough certainties whether a given aggregate may or may not be reactive to alkalis. The reactivity concept of an aggregate or mixture of aggregates is generally expressed by the results obtained from the aggregate using different tests such as petrographic characterisation [191,206,207], measurement of silica dissolved and expansion tests. Gao et al. [208] tested and compared three methods (basic attack (NaOH) at 100 °C and 60 °C) and two acid attacks (HCl–KOH and HF) to quantify the amount of reactive silica, potentially available for ASR, in different aggregates. The reliability of the methods was assessed by comparing the reactive silica contents with expansion measurements. The authors showed that the acid attack with HF provided enhanced correlation with the mortar tests. The acidic attacks showed improved results considering the high correlation with the expansion tests, its simplicity and quickness (24 h). Generally, most of the recommendations mention that for reactivity assessment the petrographic analysis should be the first test. However, some limitations remain. For instance, there is no established criterion with basis on the threshold value for the quantity of reactive species identified. Countries such as France, Germany and Denmark indicate for aggregate reactivity a threshold value of 2% where the presence of reactive species above 2% is not allowed. Other countries do not indicate any threshold value, mentioning only the need for additional tests when in the presence of reactive aggregates. Furthermore, petrographic analysis is a technique highly dependent on the knowledge and experience of the technician [89]. Moreover, the wide diversity of names, sometimes contradictory, given to the minerals or rocks from different countries can lead to ambiguities [209]. In order to avoid these limitations some standards and recommendations have been proposed such as ASTM C 295 [210], RILEM AAR-1 [211], and RILEM TC 191-ARP [180]. The measurement of silica dissolved based on the ASTM C289 was a test widely used for the determination of aggregate reactivity [212] mainly due to its ease and speed. This test was based on the chemical attack of a ground fraction of the aggregate (150– 300 lm) with a solution of sodium hydroxide (NaOH) and residue filtration. The alkalinity reduction and the silica dissolved in the NaOH solution was then assessed [212]. However, this standard was withdrawn in 2016 due to contradictory and incongruent results [89,90,213]. The incongruences found all over the world, may be explained due to the presence of different carbonates (e.g. MgCO3, FeCO3, CaCO3), magnesium silicate hydrated, plaster, zeolites, clay minerals, organic matter, and iron oxides. These species may lead to an underestimation of the dissolved silica content due to their interference or precipitation during its determination. Furthermore, the reactions with Na+ and K+ ions may lead to higher alkalinity values. On the other hand, the fine grinding of the aggregates leads to new surfaces for the attack of the alkaline solution. The surfaces of the ground aggregate will have more broken siloxane bridges and silanol groups when compared to particles that are not ground. This will lead to a higher solubilization of the silica. Besides, the grinding and sieving can also remove reactive phases hindering the detection of aggregate reactivity [89,213]. ASTM C227 [214], which was withdrawn recently (October 2018), was a method that allowed to assess the reactivity of the aggregates. The specimens fabricated with cement rich in alkalis were maintained during 6 months in a chamber at 38° C with a HR higher than 95%. Expansion measures of the specimens were conducted at defined ages. The aggregate was considered reactive if the expansion value at 6 months was higher than 0.10%. One of the main restrictions of this test was its duration, which is usually incompatible with the construction requirements. Furthermore, it has been shown that the results obtained by this method are dependent on the alkali content of the cement, the w/c and the conditioning conditions of the test specimens [89].

R.B. Figueira et al. / Construction and Building Materials 222 (2019) 903–931

ASTM C1260 [215] is perhaps the most used test for aggregate reactivity assessment. According to this standard, the aggregate is considered reactive if at the 14th day of trial the expansion is higher than 0.20%. If the result at the 14th day is between 0.10% and 0.20%, it is advisable to extend the test until to the 28th day. In this case the aggregate reactivity should be classified as doubtful if the expansion does not exceed the value of 0.20%. RILEM also adopted a test, known as AAR-2, to qualify the reactivity of aggregates for concrete [216]. According to the ASTM C1260 or RILEM AAR-2 any aggregate that shows a result above 0.10% at 14th day, in the absence of local information about its use, should be considered as potentially reactive to alkalis. This situation has raised criticism since a number of authors [89,217–221] consider that this method is very severe, leading to false positives – i.e. the rejection of many aggregates that would not contribute to the development of ASR. For some authors [89,217] the ASTM C 1260 should not be used to reject aggregates except in situations where the reactivity is confirmed using ASTM C 1293 [222] or the petrographic analysis confirm the result of the ASTM C 1260 or RILEM AAR-2 tests [221,223]. Whenever data are available from both ASTM C 1260 and ASTM C 1293, the ASTM C 1293 results should prevail [224,225]. Despite these limitations, the methods described in ASTM C 1260 or AAR-2 have proved to be very useful with the advantage of allowing the assessment of aggregates in only 2 weeks. Therefore, this method is considered to be an important tool in decision making and is generally included in the main characterisation methodologies of aggregate reactivity [219,220,223,224]. 2.4.3. Concrete moisture control As mentioned previously a certain content of RH (above 80%) is necessary to initiate ASR in concrete [89]. Several studies have been devoted to study the effects of moisture conditions [62,63,167,226–228]. Multon and Toutlemonde [62] have shown that if the water is supplied, regardless of the age of an ASR damaged structure, it swells if the ASR gel is already produced. It was also proposed that the ASR reaction could have been stopped by lack of water in certain areas of the structure [62]. Therefore, any methodology that restrains the water access to the concrete with reactive aggregates mitigates the risk of ASR development. These include, for instance, equipment or project solutions that avoid the accumulation of water. Project solutions that allow the water to drain away from the concrete structure should be considered. The application of coatings or sealants is also, generally, beneficial. However, it should be kept in mind that the implementation of these measures should be considered during the project development. Furthermore, the employment of such solutions are far from being a sustainable treatment since these will only postpone the progression of the ASR [229]. The use of surface treatments for ASR mitigation is mainly influenced by its efficiency in controlling the moisture exchange between the concrete and the external atmosphere. Ideally, surface treatments for ASR mitigation should allow the escape of water vapour, for progressive drying of the concrete. Several manuscripts have reported the effectiveness of the use of silanes in delaying the appearance of ASR and therefore increasing the service life of concrete structures [230–233]. Bérubé et al. [230] tested the behaviour of different sealant materials in ASR mitigation namely silane, oligosiloxane, polysiloxane, linseed oil, or epoxy resin on specimens prepared with a w/c = 0.50 exposed to different conditions. The authors showed that all the specimens, sealed early with silane, oligosilixane, or polysiloxane did not expanded considerably nor showed map-cracking pattern. The epoxy resin was not effective and samples sealed with linseed oil showed poorer results compared to silane-based sealers. In the end, the authors concluded that a good sealer may provide

913

improved aesthetic appearance and stop the ASR expansion of the concrete specimens. Tosun et al. [231] studied the use of iso-butyl-triethoxy-silane for ASR mitigation. The authors showed that the effectiveness of iso-butyl-triethoxy-silane depended on extended prewaiting periods. The formation of strong oxygen bridging bonds between silane molecules and mortar provided a surface with hydrophobic character. Recently, Deschenes Jr. et al. [63,233] studied the influence of RH on the development of ASR. The authors showed that enough drying may prevent ASR deterioration in non-air-entrained concrete. The authors also showed that silane could be used as a mitigation measure for concrete exposed to a combination of ASR, freezing and thawing, and wetting and drying cycles. 2.4.4. Modification of expandable properties of ASR gel The use of lithium salts to prevent ASR was reported for the first time by McCoy and Caldwell in 1951 [234]. The authors tested >100 different compounds, including organic products, metallic salts, acids and adjuvants and reported that the lithium compounds were the most effective [234]. Later, some studies concerning the effectiveness of lithium compounds to control ASR were conducted [21,235–239]. In 1993, it was reported by Stark et al. [240] that insufficient dosages of lithium compounds may increase the ASR expansion, due to some aggregates, rather than reducing it, known as the pessimum effect. Furthermore, the mechanisms proposed to explain the effectiveness of lithium compounds against ASR are not fully understood and several mechanisms have been proposed [16,236,241]. The effect of lithium has been described by different ways. Some proposed that lithium may reduce silica dissolution [53,242,243] others advocate that the lithium may decrease the repolymerisation of silica and silicates [51,244]. It was also proposed that lithium may reduce repulsive forces between colloidal ASR gel particles [243,245] or is responsible for changing the ASR product composition, leading to less expansion or the formation of non-expansive products [241,246,247]. It was also reported that the lithium ASR mitigation mechanism could be explained by the ability of lithium to replace calcium in ASR gel rather than K+ and Na+ [244,247]. Another proposal included the formation of a protective barrier surrounding the ASR gel leading to a reduction of the affinity and absorption of water which is responsible for the gel swelling [244]. Table 2 shows the most representative studies published in the last few years (since 2007) regarding the use of lithium compounds to mitigate ASR. From Table 2 can be observed that the most studied compound was the LiNO3 followed by LiOH. According to RILEM the LiNO3 dosage is dependent of the aggregate’s reactivity. For instance class III aggregates should use 5.95 kg of LiNO3 per kg of Na2Oeq in concrete [258]. Despite the research efforts and the number of publications concerning the lithium ASR mitigation mechanism (Table 2), contradictory results were found in the literature [16,244]. Therefore, further studies to fully understand the lithium mitigation mechanism must be conducted. 2.4.5. Mineral additives In the last few decades, mineral additives have shown an important role in the construction industry and are responsible for the technological development of concrete and mortars. The use of natural additives dates back to ancient Greece such as the use of pozzolanic materials from Santorini island. The actual state of conservation of many ancient buildings, still in use today, is the paramount evidence of the excellent behaviour of pozzolanic additives. The use of mineral additives such as ground BFS [259–261], FA [262–264], SF [115,265,266] and MK [42,43,267] among other materials [40,268,269] with pozzolanic properties have been advocated as effective in inhibiting ASR. This section aims to show and

914

R.B. Figueira et al. / Construction and Building Materials 222 (2019) 903–931

Table 2 The most relevant manuscripts published in since 2007 concerning the use of lithium compounds to mitigate ASR. Year

Ref.

Li compounds

Main results and conclusions

2007

[248]

LiOH

2008

[249]

Li2CO3

2010 2012

[250] [16,251] [252]

LiOH LiNO3 LiNO3

2013

[253]

2014

[244] [50]

Li2SO4, LiNO3, Li2CO3, LiBr LiNO3 LiOH and LiNO3

2015

[52]

LiNO3, Aluminium

2016

[254]

LiNO3

[255]

LiNO3

[53]

LiNO3

2017

[54]

LiOH

2018

[256]

Li2CO3, LiOH and LiNO3 Li2SO4, LiNO3, Li2CO3, LiBr, LiF

The higher the proportion of Li the less ASR expansion took place. The diffusion of SiO2 from opal grains was limited in the presence of LiOH. Li2CO3 was more and less effective in the presence of highly and less reactive aggregates, respectively. Its efficacy was function of the mineral properties and reactivity of the aggregates. LiOH acted as an ASR inhibitor. LiNO3 suppresses the ASR and a mechanism for those effects was proposed. LiNO3 was applied via topical sprays and vacuum impregnation on sections of affected areas. The lithium salts applied were unable to penetrate and mitigate ASR expansion. Li compounds applied at the ratio 0.5–3.0% proved to be effective in reducing length changes. The lowest changes were obtained with 3% of Li2CO3. Li decreased the ASR expansion and bound preferentially leading to a faster depletion in the pore solution. Changes of water-soluble alkali and Li content were studied to understand the effect on ASR by adding FA and lithium compounds simultaneously. LiNO3 addition exhibited improved performance in increasing water-soluble Li/(Na + K) ratio compared to LiOH addition. The effect of aluminium (from metakaolin (MK) and calcium aluminate clinker) and LiNO3 on the mitigation of ASR was investigated. MK lead to a slower SiO2 dissolution and formation of reaction products. LiNO3 suppressed ASR. Li prevented the dissolution of reactive silica. The ASR control was explained by the formation of physical barrier in certain areas of the reactive aggregate exposed surface. ASR was dependent on the dosage of LiNO3. LiNO3 was more and less effective in the presence of highly and less reactive aggregates, respectively. A good agreement was found between the required optimum experimental and analytical Li dosages to inhibit the excessive mortar expansion. The addition of a molar ratio of Li/(Na + K) = 0.74 decreased the expansion of mortars made with reactive gravel aggregate until reach the safe and non-destructive level. Studies on the influence of Li+ migration on different levels of ASR development. Na+ and K+ removal, Li+ migration (combined with Na+ and K+ removal) and Li+ diffusion on ASR expansion were assessed. Li+ migration led to the lowest posttreatment expansion levels. Type and concentration of Li compounds in the anolyte to be used on electrochemical migration repair technique were studied. The concentration of the solution, rather than the type of Li salt, affected migration. The optimum amount of Li additives changed according to the compound (e.g. 3% Li2SO4, 1.5–3% LiNO3, 0.5–3% Li2CO3, 0.5–3% LiF). ASR was affected by the type and additive ratio.

[257]

discuss the latest knowledge and developments on the use of mineral additives in mitigating/preventing ASR. The mechanisms proposed for each type of addition on ASR mitigation will also be debated. In the last few years, a number of review manuscripts have been published concerning the use of mineral additives [18,22]. Thomas in 2011 [22] reviewed the mechanisms by which supplementary cementing materials (SCM) controls the ASR, the SCM composition effect and the test methods used to determine the amount of SCM required to mitigate the risk of ASR expansion. Saha et al. [18], in 2018, published a review focused on ASR mechanism of reactive aggregates in concrete and its mitigation by FA. Both authors, based on the available research, stated that the use of SCM has been found to be the most efficient way of ASR mitigation technique. However, much has been debated about the action mechanism of mineral additions in ASR as well as the minimum content of each type of addition to be employed to mitigate/inhibit ASR. The most relevant studies reported since 2001 concerning the incorporation of different SCM are summarised. The SCM/mineral additives considered in the following sections were BFS (ground or granulated), FA, SF and MK. The main results and impact of SCM as well as the weight percentage used in each case for mitigating/delaying the ASR reported was also considered. 2.4.5.1. Blast furnace slag (BFS). Blast furnace slag (BFS) is a nonmetallic by-product of pig iron production obtained in igneous conditions in a blast furnace. BFS is a granular non-crystalline glassy form composed essentially by silicates and calcium alumino-silicates with latent hydraulic properties [270]. It has been reported that the use of blast furnace slag (ground or granulated) in concrete has numerous benefits and its use as a SCM, to inhibit ASR, dates back to the 1950s [229]. When BFS is used as SCM several properties of concrete are improved such as abrasion resistance, workability and rheological properties, longterm compressive strength, shrinkage, permeability and water diffusion, chloride ions and gas diffusion reduction, resistance to attack by sulphate, and ASR expansion [271].

The mechanisms proposed for the mitigation of ASR due to BFS include generically three steps [140,272], namely (i) permeability reduction; (ii) alkali retention and (iii) Ca(OH)2 consumption. Experiments, where the slag contents were above 50% together with reactive aggregates and high alkali cements, showed that, besides diluting agent of alkalis, the main effect of slag was the reduction of the OH– ions mobility [90]. The BFS are less effective when the cement have low alkali content [90]. Other studies [273], showed that a replacement level of 50% of PC with BFS significantly reduced the expansion of the concrete in expansion tests. However, it was also indicated that the alkali level of the slag was not a contributory factor concerning to the level of replacement used. Özbay et al. [271], published a review focused on understanding the use of BFS considering potential environmental impacts and technical benefits for sustainable construction. The authors concluded that the use of BFS had a tremendous effect on the environmental protection in terms of the CO2 emission and natural resources. Studies performed on the use of BFS, using contents higher than 50%, with highly reactive aggregates and high alkali cements showed that the main effect of BFS was as alkali diluent [274] reducing the mobility of OH– ions. Others concluded that BFS are less beneficial when the alkali content of the cement is low [90,275]. While others have shown that in addition to these effects, BFS also acted to reduce cementitious paste alkalinity regardless the alkali content of BFS, suggesting that other factors influenced the composition of the CASAH gel [276]. Table 3 displays the most representative studies published since 2001 regarding the use of BFS in concrete to mitigate ASR. Table 3 shows that different percentages of BFS have been studied. The data published are generally in agreement when considering that the incorporation of BFS in concrete effectively mitigates the ASR. Generally, most of the publications show that the degree of expansion decreases as the replacement level of BFS increases and for improved concrete behaviour the content of BFS should be above 50% [271]. Nevertheless, no threshold value for the use of BFS to effectively mitigate the ASR was established.

915

R.B. Figueira et al. / Construction and Building Materials 222 (2019) 903–931 Table 3 The most relevant manuscripts published since 2001 concerning the use of BFS to mitigate ASR. Year

Ref.

Main results and conclusions

BFS/Cement (wt %)

2001 2002

[40] [277]

50% ground BFS decreased the ASR expansion to lower values than those obtained with low-alkali cement. The rate of ASR in alkali-activated ground BFS mortars was less than that of ordinary PC mortars due to the competition between the slag and the aggregate for the alkalis. ASR mitigation techniques were assessed. Mitigation procedures were developed by using low-alkali cement and the replacement of part of it by FA Class F, or ground BFS or the combination of both. The alkali level of ground BFS on potential alkali–silica reactivity was studied. The partial replacement of PC with slag (50%) significantly reduced the ASR expansion. The influence of SF, slag and FA on ASR under 70 °C was studied. ASR was not inhibited (only delayed) for contents between 30% and 70%. Tests to determine the durability aspects of concrete containing 50% or 65% of slag replacement in different curing environments. 65% suffered less expansion and microcracking compared to 50%. The influence of blended cements and sulphate resistance of concrete on ASR were studied. FA and BFS reduced ASR and sulphate damages compared to PC. BFS showed improved results. Study on the combination of brass-coated steel microfiber and BFS on ASR mitigation. ASR expansion was reduced and BFS was effective at preventing the mechanical property loss. Study of alkali dosages impact on ASR in activated slag mortars. Alkali activated slag mortars showed improved performance compared to PC under accelerating condition. Alkali-activated Portland hybrid blended cement and a Portland blended cement were compared to 100% ordinary PC. The material based on PC had highest susceptibility to ASR, followed by the alkali-activated hybrid material and the blended cement with 80% ground BFS.

35 and 50 NR

[278] 2005

[273]

2008

[279]

2009

[280]

2011

[281]

2014

[282]

2017

[201]

2018

[260]

2.4.5.2. Fly ash (FA). FA is a fine powder, in the form of small spherical particles with pozzolanic and binding properties. FA due to its reactive capacity is suitable for multiple applications, particularly as a building material. The type of reactivity depends on its properties and according to the calcium content FA can be classified as ‘‘poor” or ‘‘rich” in calcium, generally known as Ca-rich or Ca-poor FA. The chemical and mineralogical composition of Ca-rich and Capoor FA together with the glass content and degree of granulometry, influences their properties, especially the pozzolanic effect [134,283]. According to ASTM C 618-19 [284], FA in concretes, are divided in two classes according to the lime content. Class F has a lime content <10%, and Class C has a content higher than 10%. FA is used in concrete for two main purposes i.e. as correction of the granulometry of the fine aggregate and/or as partial replacement of the cement [285]. The use of FA as a SCM aids the pumping, delays the setting time, increases the workability and reduces the hydration heat in the concrete [286]. Additionally, FA revealed to be important in improving the durability of concrete structures particularly as SCM in concretes for marine [287] and sulphate [288] environments. The combined effects of reducing the penetration of aggressive agents and the calcium hydroxide reduction content of the hydrated cement paste are the main reasons for FA success in these environments. In 2017, Kawabata and Yamada [19] discussed the inhibitory role of FA in ASR, particularly the effect of FA at the pessimum proportion. The authors proved that the inhibitory effect of FA on ASR expansion was drastically reduced at the pessimum proportion. It was concluded that a larger amount of cement should be replaced with FA when a highly reactive aggregate, exhibiting a strong pessimum effect, was mixed with non-reactive aggregates. Several mechanisms have been proposed in an attempt to explain the inhibiting process of FA in the ASR [18,19,289]. A generic mechanism was proposed by Thomas in 1996 [289] and included basically three steps: (i) Reduction of the ionic mobility paste permeability due to the formation of CASAH; (ii) Absorption of the alkaline ions by the CASAH formed previously leading to alkalinity decreasing of the concrete pore solution; (iii) Ca(OH)2 consumption. The concrete permeability decreasing is directly linked to the quantity of hydrated cementitious material, formed at a certain age, diminishing as the cure period is increased. In 1994, Malhotra

40 and 50 5, 20, 30, 50 and 60 30, 50 and 70 50 and 65 10, 20, 30, 40 and 45 20 and 40 NR 80

and Ramezenianpour found that concrete with FA has lower permeability than concrete without any FA [290]. Nevertheless, the same authors failed to explain the influence of the chemical composition of FA, namely the alkalis and calcium contents, on the ASR. The information found in the literature regarding the higher capacity of CASAH formed from the pozzolanic reaction with FA, compared to ordinary PC, in retaining the alkalis present in the cement paste, is almost unanimous. Most of the authors consider that the addition of FA promotes the reduction of the ratio CaO/ SiO2 of CASAH gel which will have higher capacity to retain alkalis [289]. This high capacity in retaining the alkalis is highly dependent on the chemical composition of the FA. FA poor in calcium content has shown a higher capacity in retaining the alkalis [291]. The retention alkali capacity by the CASAH gel was explained by Hong and Glasser [292]. The authors showed that the alkalis would react with SiAOH groups present in the CASAH gel. This process may be described as an acid-base reaction where the alkalis behave like bases (electron accepter) and the SiAOH groups behave like acids (electron donor). The number of SiAOH groups present and their ionic strength, which is high in the CASAH gels of lower ratio CaO/SiO2, can be the explanation to the higher or lower capacity that the CASAH gel have in retaining the alkalis. This capacity is even higher when in the CASAH gel are introduced other ions with acidic character, namely aluminium (Al3+) [55]. It has been widely reported that the presence of this ion can mitigate ASR. This may explain the improved results obtained with FA with high contents of silica and alumina compared to other SCMs [140,293]. Nevertheless, not all the alkalis are retained in the structure of the CASAH or CAAASAH gels since part are adsorbed on the gel surface being available for ASR. Other studies showed that for the same content of FA the reduction in the expansion was regardless the content of the alkalis present in the system [262,289,291]. These results suggest that other factors interfere in the FA mechanism regarding the suppression of ASR. In 2011, Thomas [22] reported that the presence of alumina contents contribute to ASR mitigation. The Al3+ ions can incorporate the CASAH system, forming calcium aluminium silicate hydrate (CAAASAH). The CAAASAH system has different properties compared to CASAH system and promotes the binding of alkali ions from pore solutions, stopping the continuous recycling of these ions and formation of the swelling complex. Aluminium adsorption allows the formation of inactive aluminosilicate complexes, providing an inhibitory effect [294]. Chappex and Scrivener [295] reported a study that was focused on understanding if the presence of additional aluminium provided by the SCM reduces

916

R.B. Figueira et al. / Construction and Building Materials 222 (2019) 903–931

the formation of gel in reactive aggregates. The authors concluded that systems with SCMs containing alumina expand less due to ASR than those containing only silica additions. The authors also showed that aluminium can reduce the dissolution of disordered silica at high pH and this reduction was most effective in the presence of a continuous reserve of aluminium. Leeman et al. [52] showed that the addition of Al3+ slowed down the dissolution of SiO2 without stopping the reaction only affecting the kinetics. Szeles et al. [296], in 2017, performed a comprehensive study on the mitigation of ASR by Al(OH)3 and proposed five possible mechanisms behind this process, namely: (i) pH and alkalis content reduction in concrete pore solution; (ii) Portlandite consumption and reduction and dissolved calcium reduction in the concrete pore solution; (iii) Silica dissolution reduction and aggregates damage at high pH; (iv) ASR gel composition changes and producing innocuous gels; (v) Porosity and pore size reduction (reducing water and ion transport). The potential ASR suppression mechanisms by Al(OH)3 were based on combatting the well-known pre-requisites needed for onset of ASR. Mechanism (i) was assessed by determining the pH and ionic concentrations of the concrete pore solution; mechanism (ii) was assessed by conducting TGA to determine the solid portlandite and Al(OH)3 contents of two paste mixtures over time; mechanisms (iii) and (iv) were assessed by determining the aggregate damage and ASR gel formation. The fifth mechanism was assessed by determining the porosity and the pore structures of the paste samples. The authors proved that alumina mitigated the ASR mainly through (i), (ii) and (iii) mechanisms. The consumption of Ca(OH)2 due to the pozzolanic reaction with FA is another way to explain the inhibitive mechanism of ASR. The reduction of the Ca(OH)2 availability by the FA, contributes to reduce the potential development of ASR. Furthermore, the efficiency of FA addition is linked to the oxide’s contents (Al2O3, SiO2 and Fe2O3) and their capacity to react with the Ca(OH)2 to decrease the pH of the cement paste [297]. Table 4 shows the most relevant information concerning the use of FA for mitigation of ASR expansion published in the last few years (since 2001) and it is clear that extensive research has been conducted to explore the effects of FA on ASR mitigation with more than a yearly publication. Though there is an intensive scientific debate among researchers, most of the work developed clearly shows the positive contribution of FA on ASR mitigation. The information found in the literature (Table 4) shows that FA effectively mitigates the ASR expansion although some contradictory references [298] show that FA with poor calcium contents are more efficient at inhibiting the ASR than the FA with high calcium contents [291]. The divergences found in the results can be explained by the use of different materials and the conditions employed to study FA impact on ASR. On the other hand, most of the existent data focused on the efficacy of FA, was obtained from accelerated tests which do not reproduce the real conditions of a structure. Moreover, Table 4 shows that there are, so far, few publications mentioning cases of structures affected by ASR when FA is incorporated in the concrete. Furthermore, the cases reported in the literature on the replacement of cement by FA for ASR mitigation are connected to the use of lower FA dosages than those recommended, or to the combination of other concrete degradation processes [299]. Another specific concern of the scientific community regarding FA is its inhibitory mechanism in ASR mitigation which is not fully understood [17–19,269,285].

2.4.5.3. Silica fume (SF). Silica fume (SF) or micro-silica is a very fine non-crystalline silica and was first tested in concrete in Norway in the early 1950s [315]. The use of SF concrete has several advantages. The incorporation of small amounts of SF increases the durability properties of concrete and higher strength concretes were obtained. Moreover, the performance of SF concretes in regarding ASR expansion and exposure to sulphate environments is enhanced when compared to PC concretes. Concerning the ASR mitigation the recommended replacement dosages range from 10 to 15% of SF. Replacements below 5% did not show a suppressive effect, only delaying ASR evolution [316,317]. The mechanism by which SF acts in ASR is not fully understood and is still a topic of discussion [266,318]. According to Hobbs [90], the SF in the fresh state works as a reactive aggregate due to the large amount of ultrafine amorphous silica particles. If the replacement content is controlled, the alkalis released by the cement and by the SF, are consumed in the pozzolanic reaction with the silica even before the concrete setts. Furthermore, this mechanism allows to understand the failure, in certain cases, of SF regarding ASR as reported by different authors [107,319]. Table 5 shows the most relevant information concerning the use of SF for mitigation of ASR expansion published in the last few years (since 2001). The authors all agree that for SF replacements of cement below 10% the ASR suppression is not effective (Table 5). To minimise/ avoid some of the limitations related to the use of SF, studies in which SF is employed in blends with other additions such as FA have been reported [300,323,324]. The data published (Table 5) also shows that further studies should be performed in order to determine some of the limitations related to the use of SF in ASR mitigation, notably the mechanism responsible for the role of SF. Other concerns are essentially technological and are related to the agglomeration of SF particles in the concrete [107,322,325]. 2.4.5.4. Metakaolin. The use of calcined clays as additives in concrete was reported for the first time by Andriolo and Sgaraboza [326]. It was found that for the incorporation of 15% of calcined clay the ASR expansion was reduced from approximately 0.18– 0.02%. Five years later, Jones et al. [327], showed that ASR in concrete was inhibited by the incorporation of 10% of MK, even when exposed to chloride ions. It was also reported that a 15% of MK replacement would remove all the free calcium hydroxide. However, the authors stated that the minimum MK replacement levels were not well defined, although the experiments suggested that a content close to 10% was enough. Later, Wild and Khatib [328,329] studied and tested cement mortars and pastes containing 0, 5, 10 and 15% replacement of cement with MK. The authors concluded that the incorporation of MK in cement pastes led to the refinement of the pore structure [328] and the threshold value for paste decreased as the MK content in the paste increased. Furthermore, the authors showed that the proportion of pores with radii smaller than 20 nm was increased as the replacement level of cement by MK increased. These findings suggest that the presence of MK may influence the reduction of the transport of the alkali ions to the sites with reactive silica in the concrete. One year later, the same authors, [329] found that the Ca(OH)2 in mortars was less than that found in pastes. Considering that the consumption of Ca(OH)2 can prevent ASR, the MK have been considered a SCM with high potential in ASR mitigation [329]. In addition to the effects, which may be directly related to the mechanism of suppression of ASR by MK, other effects on the durability of hardened concrete have been reported in the literature [42,267,330,331]. The incorporation of MK as replacement of cement increases the resistance to penetration of chloride ions [332,333], reduction of capillary absorption [334], and high resistance to external chemical attacks [335,336].

917

R.B. Figueira et al. / Construction and Building Materials 222 (2019) 903–931 Table 4 The most relevant manuscripts published since 2001 concerning the use of FA for mitigation of ASR. Year

Ref.

Main results and conclusions

FA/ Cement (wt %)

2001

[40]

20, 40

2002

[300]

2003

[301]

2004

[135]

2005

[302]

2006

[303]

FA-A (low calcium and alkali contents), FA-B (moderate calcium and low alkali contents), FA-C (high calcium and alkali contents) were studied. FA-A and FA-B decreased the expansions to lower values than those obtained with low-alkali cement. FA-C was ineffective even at a 40% cement replacement by mass. Blends of SF and FA were studied using the CPT. The levels of SF and FA introduced into high alkali cement systems were effective in reducing ASR expansion to levels < 0.04%. Different contents of FA and Natural Zeolite (NZ) have been tested. The ASR expansion was suppressed by replacing cement with 20% of a mixture of NZ and FA. The alkali reactivity of chert and the effect of moderate-calcium FA on the ASR was studied. The alkali-reactive chert used showed pessimum proportion behaviour in the range of 5–15%. High proportions of chert in the aggregate, i.e. > 50%, did not resulted in deleterious expansion. As the FA replacement increased, the expansion decreased. The effect of modified zeolite (MZ) on ASR expansion was studied and compared to other mineral admixtures such as FA and ground BFS. The MZ controlled the ASR expansion. The 14-day expansion of the specimens with 5% of MZ was < 0.1%, while the percentage of the FA and slag with the same efficiency, was 25% and 40%, respectively. Investigation whether the Ca/Si molar ratio of the C-S-H in PC is reduced when a part of cement is replaced by FA and whether this reduction, results in increased alkali retention. Alkali binding was sensitive to the Ca/Si ratio of the C-S-H, decreasing as this ratio increases. A link between C-S-H composition and sorption of alkali was established. The releasable alkali from granite, gneiss and feldspar used in Three-Gorges dam was assessed. Low calcium FA was used and ASR was not found. Samples with PC or FA were compared. The systems based on FA were less susceptible to ASR than traditional PC systems. Influence of SCMs on ASR (70 °C) were studied. 10% of FA did not influenced the ASR. 20%-30% of FA only delayed the expansion. FA > 50% inhibited effectively ASR. Comparisons between concretes modified with glass powder or FA, at the same replacement level, were performed. FA showed improved behaviour. Class C FA was assessed for its ability to resist damage by ASR. Binary blends of MK or Class C FA reduced the ASR expansion compared to control samples. Ternary blends of MK and Class C FA resulted in higher expansion than binary blends incorporating the same amount of MK. The effect of fineness of FA on mechanical properties and ASR resistance was studied. Grinding process improved the mechanical properties of all samples. Incorporation of FA with different fineness values and ratios decreased the ASR expansions. The fineness of FA seems to not affect the ASR expansion compared to coarser FA replacements. Concrete blocks with reactive aggregates, different levels of high-alkali cement and two sources of FA were assessed outdoor. All blocks without FA showed excessive expansion and cracking within 5–10 years of production. FA replacement levels of 25% or 40% were effective in reducing ASR expansion. Study on the influence of blended cements with different types of pozzolans on ASR. NZ, FA, and ground BFS were used. NZ, FA, and ground BFS reduced the ASR expansion. Study on the effect of FA from biomass combustion in ASR mitigation. The biomass FA incorporation mitigates the ASR. Expansion decreased with the increasing content of FA. FA fineness influenced the ASR within the range of the average particle size (APS) of FA (10–30 lm). ASR had an exponential relation with SiO2, CaO, SiO2 + Al2O3 + Fe2O3, CaO + MgO + SO3, SiO2equi and CaOequi for FA with APS<10 lm. Expansion was a linear function of CaO, CaO + MgO + SO3, and CaOequi and a logarithmic function of SiO2, SiO2 + Al2O3 + Fe2O3 and SiO2equi for FA with APS < 5 lm. Study on the contribution of different mechanisms for ASR mitigation by FA using ASTM C1567 test. Experimental and computer simulation were combined. FA ability to control ASR generated by recycled glass sand. Capacity of CIM [312] to predict the FA dosage for ASR was assessed. It was concluded that the model parameters must be revised. Ternary blends of high-calcium FA and slag for ASR mitigation were studied. Ternary blends did not offered advantage over binary blends and of individual material for the same SCM. Capacity to retain alkalis increased with the blend ability to consume Ca(OH)2. The alkali leaching test was proposed as a tool to compare the efficiency of different blends. FA’s effect, was studied by ASTM C 227, ASTM C 1260 and autoclave methods. The 12-month results agree with the AMBT. Samples whose mixing water was pure were in a good correlation on the basis of 12-month results. The ability of high CaO and/or high-alkali (Na2Oeq) FA to mitigate ASR in mortar was assessed. Low and moderate CaO content FA were more effective compared to higher CaO and Na2Oeq FA. The methodology worked for low CaO and alkali FA. It was concluded that to predict replacement levels of moderate to high-alkali FA adjustments were necessary. Expansion, pore solution, compressive strength and alkali leaching of biomass and coal FA were investigated. Class F mixes showed improved behaviour. Class C reduced expansion with restricted efficiency. Biomass FA showed equivalent expansion reduction to that of Class C. Extended CIM was developed to predict the FA dosage for ASR mitigation. Extended CIM offered acceptable prediction accuracies for both Class C and F FA. Alkali-activated FA concrete showed promising performance compared to ordinary PC concrete in ASR mitigation. Study on the role of FA in ASR mitigation and at pessimum proportion. ASR was reduced at pessimum proportion. FA increased the latency time. Study on the effect of trass and FA under short- and long-term experiments. FA was more effective than trass in strength development. 20% FA was the optimum content. Study of binary and ternary systems (FA and LSP) on ASR. ASR decreased with FA increase. Ternary blend of 20% FA/LSP was the most effective. Study on effect of FA and MK on ASR in water–glass-activated slag mortars. Both FA and MK mitigated the ASR of the water– glass-activated slag mortars.

[304] 2007 2008

[305] [279] [306]

2010

[307]

[308]

2011

[262]

[281] 2012

[309]

2013

[310]

[311] 2014

[44] [132]

2014

[45]

2015

[313]

[46]

2016

[263]

2017

[314] [19] [38] [49]

2018

[43]

Table 6 shows the most relevant information concerning the use of MK for mitigation of ASR expansion published in the last few years (since 2001). Concerning MK’s mechanism in ASR mitigation, the available publications differ in conclusions. Some authors reported that the

10,15, 20,30,40,45,60 25, 30 10, 20, 30, 40

30 and 40

30 and 45

2025 100 10, 20, 30, 50, 70 5, 10 and 20 25

20, 40, 60

25, 40

10, 20, 30, 40, 45 20, 30 25

15, 20, 25, 30, 35 10, 15, 20, 25, 30, 35 20, 30, 40, 50

10, 20, 30, 40, 50 25, 35, 45

15, 25, 35

N.A. 45 10, 20, 30 10, 20, 30 10, 15, 20, 30 30

MK is very effective in reducing Ca(OH)2 and reducing ionic mobility (Table 6). Others have shown that the hydration products formed in the pozzolanic reaction of MK had the ability to absorb alkalis, which is an inhibition factor of ASR by MK [320,337]. Table 6 also shows that the number of publications regarding the

918

R.B. Figueira et al. / Construction and Building Materials 222 (2019) 903–931

Table 5 The most relevant manuscripts published since 2001 concerning the use of SF for mitigation of ASR. Year

Ref.

Main results and conclusions

SF/Cement (wt. %)

2001

[320]

The influence of high reactivity metakaolin (HRM) and SF on ASR was studied. SF and HRM showed analogous results. The Ca content of ASR products increased with time and as ASR reaction proceeded. Ca/Si ratio of the reaction products increased in a linear trend indicating that the Ca content may be linked to expansion. Two types of SF for ASR mitigation were studied. 5% SF was not enough to control ASR expansion. Use of 10% SF restricted the expansion to the level given by the low-alkali control. SF with high contents of silica and less alkali was more efficient in reducing the ASR expansion. Mortar mixtures with SF with low contents of SiO2 (68%) failed to control ASR of Spratt aggregate up to 12%. SF with normal contents of SiO2 (88%) was effective at 12%. SiO2 contents affect the ability of SF to control ASR. Microstructure and morphology of agglomerated and sintered SF particles on ASR were studied. Large SF particles decreased and increased the expansion. Agglomerated SF decreased expansion. Replacements with sintered SF aggregates increased the expansion. Expansion of mortar bars containing various amounts of SF, expanded perlite, and natural perlite were studied. Both expanded and natural perlite showed potential to suppress the ASR. Expansion decreased with increased SF content. 8% and 12% SF had a similar expansion rate. Samples with 16% SF met ASTM C1260 limit. Three types of SF using reactive and non-reactive aggregates were tested. Large SF agglomerates were not linked to ASR cracking. When SF was alkali silica reactive, there was a pessimum effect with expansion related to the percentage of SF used. Lower amounts resulted in higher expansions. Influence of SF, BFS and FA on ASR at 70 °C was studied. SF, BFS and FA inhibited the ASR only under appropriate content. SF contents < 10% did not influenced the ASR, between 15%  20% delayed the expansion.

10

[40]

2003

[113]

2004

[107]

2005

[321]

2007

[322]

2008

[279]

5 and 10

4, 8 and 12 5

4, 8, 12 and 16

2, 4, 6 and 10

5, 10, 15 and 20

Table 6 The most relevant manuscripts published since 2001 concerning the use of MK for mitigation of ASR. Year

Ref.

Main results and conclusions

MK/cement (wt. %)

2003

[337]

8, 15, 20

2010

[307]

2012

[309]

2014

[338]

2015

[339]

2018

[43]

Study on the expansion behaviour of heat-cured mortars containing pozzolans and slag. Its addition reduced the long-term expansion, the expansion rate and delayed the onset of expansion. Efficacy of the additions depended on the Al2O3 content. MK was the most effective at relatively low replacement level. Two MK (different particle size distributions) and FA (binary and ternary blends) for ASR mitigation were studied. MK was more effective than FA due to smaller particle size, higher degree of reactivity and chemical composition. MK with higher surface area was less effective. Ternary blends of MK with FA seemed to provide no benefit on ASR expansion over MK used alone in concrete. The use of MK was effective in limiting the AMBT expansion when used with biomass FA (20% FA + 10% MK). The effectiveness of MK was justified by the finer particle size and the chemical composition. Studies on the effects of cement replacement by MK on mechanical properties, ASR, resistance to sulphate, absorption capacity and permeability. 10–20% of MK increased the ASR resistance. Enhanced results were obtained for mortars containing 15–20%. ASR of geopolymer mortars containing only MK in the presence of six different sands were studied. Sands in MK based geopolymer mortars activated by sodium silicate did not lead to ASR characteristic of PC mortars. Mitigation effects of FA and MK on ASR in water-glass-activated slag mortars were investigated. Both mitigated the ASR. Optimum dosage of FA was 30%. Expansion decreased with MK increase and was suppressed when the slag was replaced by 70% of MK.

use of MK are much less than the ones found for FA. This may be explained by the fact that earlier than 2000 it was found that MK was efficient in ASR mitigation [327]. 3. Test methods – State of the art and their limitations In this section a comparative consideration of the different test methods used to assess ASR was performed. The advantages and disadvantages as well as the present shortcomings are also discussed. 3.1. Diagnosis methodology and sampling The diagnosis methodology is an effective way of detection whether or not the concrete structure is being degraded by ASR. Moreover, the type and number of products detected allows to assess the degree of ASR development. Generally, a simple and an accurate methodology for ASR diagnosis should be established and followed. Fig. 7 shows a schematic example of the steps that may be involved in the ASR diagnosis. The methodology shown in Fig. 7 (or other methodologies already studied/implemented) allow to diagnose the presence of ASR and, depending on the type and quantity of products detected, to assess its degree of development. After ASR diagnosis, the following steps involve performing residual alkali reactivity evaluation tests that, when necessary, should be complemented with physical tests. These must be performed before and after the reac-

3, 5, 8, 15

10 5, 10, 15, 20 NR 10, 30, 50, 70

tivity test assessment. A system to monitor the cracking evolution must be considered and installed. The residual reactivity to the alkalis should be conducted on concrete specimens (tiles) collected from the structure. Sampling should always include areas that show different stages of degradation. However, during the extraction of the concrete specimens, the interception of cracks or reinforcements must be avoided. Furthermore, the concrete samples to be extracted should have a diameter between 75 and 100 mm and a length twice its diameter. After extraction, the concrete specimens must be protected immediately, in plastic bags, in order to avoid moisture loss (drying) and carbonation. Drying may influence the extent of carbonation and this will affect the successive expansiveness of the gel of the ASR [341]. 3.2. Aggregate assessment The performance of a given aggregate is one of the most reliable methods of knowing the alkali reactivity of a concrete aggregate. Nevertheless, the information on that performance may not be conclusive, since depend on different factors e.g. exposure conditions of the structure, alkali content of the concrete used, mixture of aggregates employed and type of structure [21,213]. The need to obtain information on the reactivity of the aggregate, in a short time frame, has led to the implementation of reliable and simple laboratory methods under controlled conditions [117,342]. According to RILEM recommendations [258], specifically RILEM test method AAR-1 [343], the assessment of aggregate combination

R.B. Figueira et al. / Construction and Building Materials 222 (2019) 903–931

919

Fig. 7. Schematic example of the steps that may be adopted in the ASR diagnose. The SEM/EDS image was adapted from [340].

for ASR potential should begin with petrographical examinations of the aggregates. These should be included as a preselection method. Petrographic tests allow to determine the individual and combined composition, identify the types and concentrations of possible reactive constituents and to classify the aggregates according to three categories mentioned previously (Section 2.4.2) [258]. If the assessment of aggregates by petrography categorises the aggregates in Class II or III, further tests are necessary. The main disadvantages and limitations of the petrographic method lay in the high specificity and on the knowledge that this technique requires. Nevertheless, it is certain that when reactive species to alkalis are identified by petrographic analysis as Class II aggregates it is necessary to further confirm the relevance of its presence in the development of ASR using expansion tests. Fortunately, if the aggregates are classified either as Class I or III, this will lead directly to conclusive outcome. As mentioned previously (Section 2.4.2) different test methods for alkali aggregate reactivity assessment have been proposed. According to RILEM TC 06 [344] and TC 219 [345] the most used methods for alkali aggregate reactivity were ASTM C295 [210], ASTM C289 [212] which was withdrawn in 2016, ASTM C227 [214] which was withdrawn in 2018, ASTM C1260 [215], and NF P18-594 [346]. In 2014 two new Australian Standard methods to assess ASR of aggregates were published, namely AS 1141.60.1 AMBT [347] and AS 1141.60.2 CP [348] adopting test procedures consistent with ASTM C1260 and ASTM C1293, respectively. Nonetheless, improved performance limits, leading to a new class of slowly reactive aggregates, were introduced. Two years later, the accuracy of the proposed methods was assessed by V. Sirivivatnanon et al. [349]. The authors concluded that AS 1141.60.1 was an accurate test in classifying aggregates as ‘‘slowly reactive” and ‘‘reactive”. However, a poorer assessment was obtained for nonreactive aggregates when compared to the results provided by ASTM C1260. AS 1141.60.2 and ASTM C1293 were found to be more reliable than AS 1141.60.1 since both classified most of the aggregates studied, against known field performance, correctly [349]. Table 7 shows the most common tests that are still being used to assess aggregate reactivity and preventive measures, the main specifications, as well as the advantages and disadvantages of each test used. Among the wide variety of test methods available for ASR expansion assessment, the CPT performed at 38 °C ± 2 °C and RH above 95%, is one of the most frequently tests used worldwide

[168]. Furthermore, this test is often used as a reference test to calibrate other accelerated CPT methods [168]. The CPT [350,351] allows to assess the behaviour of fine or coarse aggregates, or their mixture, in concrete prisms with dimensions of 7.5 cm  7.5 cm  25 ± 5 cm. The samples are produced with a cement with high alkali content, generally higher than 0.9% Na2Oeq. Afterwards the alkali content is adjusted with the addition of NaOH to the kneading water, in order to reach a content of 1.25% Na2Oeq by mass of cement. The specimens, after demoulding, are preserved for 12 months in a chamber at 38 °C and with a RH higher than 95%. The expansion measures are performed at defined ages and the aggregate is considered reactive if the expansion value at 12 months is higher than 0.04% considering the ASTM C1293 [222] or higher than 0.05% considering the RILEM recommendation [352]. The AAR-3 [216] test can be used in two ways e.g. for assessing the alkali-reactivity of the combination of aggregates or as a test to establish the alkali threshold of a certain aggregate combination. AAR-3 allows to test coarse and fine aggregates together in a standard mix combination in cases where it is suspected and/or unknown a pessimum behaviour [258]. AAR-3 was considered, for some time, as the reference test for the assessment of the alkali-reactivity of aggregates mainly due to the accumulated experience of its use in different forms. Still, its duration, which is of one year, is one of the main limitation turning it, generally, incompatible with the requirements of construction industry [258]. Therefore, AMBT AAR-2 and concrete bar test AAR-5 were developed in order to get results earlier. A CPT for ASR has been developed as AAR-4.1. This test is an accelerated version of the AAR-3 test to assess the reactivity of an aggregate combination. However, there is still no agreement on the criteria established for the interpretation of the results of the AAR-4.1. This test is similar to the AAR-3 excepting the temperature, which changes from 38 °C to 60 °C, and the duration, which is reduced from 1 year to 4 months [258]. It is also worth mentioning measures of the residual expansion potential of the concrete due to alkali-aggregate reactions (AAR) [196,353]. Residual expansion tests (RET) compared to on-site measurements allow a fast and reliable assessment of concrete expansion potential. Additionally, the data obtained can also be used in numerical tools [196,353]. Generally, RET are carried out on cores extracted from structures damaged by the deleterious effect of AAR. The samples collected are then stored and the expansion is monitored at controlled temperature (generally 38 °C) and

920

Table 7 The most common tests used to assess aggregate reactivity and preventive measures, the main specifications, the advantages and disadvantages of each test. Purpose

Sample form and dimensions

Advantages

Disadvantages

Maximum expansion limits

ASTM C295 [210]

Determination of potential silica aggregate reactivity by petrographic analysis

Fragments, thin sections or polished surfaces.

Fast method. Useful tool in linking the aggregate from a given source to field structures; Allows to identify most of the potentially reactive minerals; Track changes in source materials; Relate aggregate sources/mineralogy to field structures.

N.A.

Completes RILEM AAR-2 in assessing reactivity of carbonate aggregates.

Complementary tests are generally necessary; The sureness of the analysis depends on the operator experience; Hard to identify some reactive minerals; Lack of a recommended value as a classification criterion. Need for an experienced user; Adjustment for each country. Does not detect the reactivity of aggregates of slow reactivity; Considered a severe test that tends to classify as reactive aggregates with good behaviour in field. Severe test; It is not representative of the field conditions.

Good predictor of aggregate reactivity. Suitable for assessing SCMs and Li admixtures. No need to crush coarse aggregates; Used to validate other test methods; All materials used in the concrete mixture are used in their full amounts; Identifies both rapidly reactive aggregates as well as slower reacting ones.

Alkalies leach from prisms during course of test therefore it is impossible to assess alkali threshold for a given aggregate; Too slow (one year for aggregates, two years for preventive measures); Depend on the test conditions (e.g. w/c ratio, cement alkali content, etc.)

This test is not suitable for evaluating the potential for ASR of hydraulic cement combinations and aggregate (e.g. in the absence of pozzolans or ground BFS); Likely to produce acceptable expansions when tested in concrete. This test has no correlation to the performance of actual aggregates; Pyrex can contain large and variable amounts of alkalies that can be released during the test, adding variability; Does not assess the suitability of pozzolans or slag for use in concrete. Severe test and it is not representative of the field conditions; Depend on the test conditions (e.g. w/c ratio, cement alkali content, etc.)

0.05% after 1 year. Inconclusive if > 0.03% after 1 year and 0.04% after 2 years 0.03% after 1 year, and/or 0.04% after 2 years 0.10% at 14 days

RILEM AAR-1 [342] ASTM C1260 [215]; RILEM AAR-2.1 [216]

Production of specimens, that are then stored in a solution of NaOH 1 N at 80 °C ± 2 °C for 14 days. Expansion measures at defined ages.

RILEM AAR-2.2 [216]; AAR-5 [355] ASTM C1293 [222]

The alkalis (NaOH) are added to the mixing water to obtain a total alkali content of 1.25% Na2Oeq (by mass of cement). Samples are stored in water at 38 °C ± 2 °C.

Aggregate samples. Three mortar prisms with: 25  25  285 mm Three mortar prisms with: 40  40  160 mm Three mortar prisms with: 75  75  250 mm

RILEM AAR-3.1; AAR3.2 [351]

Quick test (14 days); It allows assessing mineral additives.

ASTM C1567 [356]

Specimens are stored in NaOH 1 N at 80 °C ± 2 °C for 14 days. Expansion measures at defined ages.

Three mortar prisms with: 25  25  285 mm

Quick test (16 days) to assess the ASR potential of combinations of cementitious materials and aggregate in mortar bars.

ASTM C441 [357]

Storing developed bars over water at 38 °C ± 2 °C. Expansion measures at defined ages.

Three mortar prisms with: 25  25  285 mm

AAR – 4.1 [358]

Specimens are kept at 60 °C ± 2 °C, RH = 100%. Expansion measures at defined ages.

Three mortar prisms with: 75  75  285 mm

Screening test to evaluate the relative effectiveness of different materials that are considered to prevent ASR; Used to assess materials for a particular job to prevent excessive ASR; These materials should comply with ASTM C618 or ASTM C1240. Fast and reliable method to distinguish between reactive and non-reactive aggregates; Allows to access the effectiveness of mineral additions.

AASHTO T 303 [359]

Specimens stored in NaOH 1 N at 80 °C ± 2 °C for 14 days. Expansion measures at defined ages.

Three mortar prisms with: 25  25  285 mm

Allows to detect deleterious expansion potential of mortar bars due to the ASR within 14 days.

Is a severe test. High temperature and infinite supply of alkalies can cause some aggregates to expand that will not expand using ASTM C1293 or in field.

0.10% at 14 days

0.08% and AAR5 < AAR-2

0.020% at 14 days

0.02% at 15 weeks, and/or 0.03% at 20 weeks or longer 0.10% at 14 days

R.B. Figueira et al. / Construction and Building Materials 222 (2019) 903–931

Test Method

R.B. Figueira et al. / Construction and Building Materials 222 (2019) 903–931

saturated air [353]. Others performed the RET in water or NaOH solutions [354]. It is generally agreed that these tests should be conducted in accelerated conditions increasing both temperature and RH [353]. Nevertheless, no consensus was found on different parameters namely how much time during conditioning should be considered, subsequent testing that should be used as the initial measure and which phase of expansion is proper to measure the residual expansion potential. Furthermore, the prominent expansion in the beginning of the tests that may be explained by moisture uptake and is dependent on the current swelling, freshly formed reaction products and on the degree of water saturation. The end of reaction phase is, generally, considered after the asymptotic decrease of the expansion. This decrease may be explained by the leaching of alkalis and takes, usually, several months to be reached. However, in certain cases the expansion of the cores collected does not cease, remaining steady at a certain expansion level for years [353,355]. Therefore, these tests are sometimes criticised and some authors advocate that RET induce divergences namely due to water-uptake expansion (during drying prior to the RET), stress released after coring or alkali leaching [196]. Furthermore, since the absolute expansion values can be high and are dependent on the test duration several researchers and engineers have doubts if late stage expansions should be considered or not. Considering the shortcomings regarding the use of RET, additional studies must be conducted. Additionally, the influence of parameters such as core size and temperature, different expansion phases, timeframe involved and alkali leaching on expansion measures determined by RET need to be totally clarified. The authors Merz and Leeman [353] reported a study focused in providing detailed information on the parameters above mentioned. The authors showed that cores with larger diameters were less prone to alkali leaching, the expansion rates were faster at 38 °C compared to 20 °C. In the end, the authors concluded that the residual expansion potential was a precious tool to assess AAR in a structure. Nevertheless, supplementary information and further research was necessary [353]. 3.3. Limitations of standard tests for aggregates reactivity assessment The precision of the methods AAR-2, AAR-3 and AAR-4 was assessed and conducted according to ISO 5725-2:1994 by the RILEM partners [258]. AAR-2 method showed a good precision being able to differentiate reactive from non-reactive materials. The results concerning AAR-3 indicated that the repeatability was good however the reproducibility was poor even so the precision was enough to differentiate between reactive and nonreactive materials. The results for AAR-3 showed that this test was not appropriate to identify reactive aggregates with slower kinetics, unless the duration of the test was extended. AAR-4 test method showed improved precision than AAR-3. Both AAR-3 and AAR-4 methods are able to distinguish between non-reactive and reactive materials, but not between minor differences of reactivity [258]. Therefore, the assessment of the reactivity of an aggregate should always be performed by using more than one test method. The aggregate reactivity assessment must be performed in an integrated way and a given aggregate should not be rejected based only on the result of a single test. Considering the advantages and limitations of the accelerated tests presented in Table 7, ASTM C1260 is the most comprehensive. This method does not have major limitations in terms of operative mode, and allows rigorous assessment of most aggregates to potential reactivity to alkalis. The expansion value of 0.10% obtained on 14th day, should be used as the cut-off value to distinguish between innocuous and potentially reactive aggregates. Expansion values between 0.10% and 0.20% on the 14th day should

921

be considered as potentially reactive however of slow alkali reactivity type. Expansion values above 0.20% should be considered with high alkali reactivity. However, it should be kept in mind that the rejection of a certain aggregate should not be conducted based only on results obtained by the ASTM C1260 test. Moreover, the assessment of the reactivity to alkalis of aggregates should be performed using an integrated approach. The final decision, on whether using or not an aggregate, should be complemented with results obtained, for example, by AAR-3 which when compared to AAR-4 test is conducted under less aggressive settings and closer to field conditions. Additionally, it should be kept in mind that the accuracy of the standards results is conditioned by the type of reactive aggregates. In 2016 Ramos et al. [360] concluded that the ASTM C1260 should not be used exclusively when assessing granitic aggregates. The authors showed that AAR-4.1 was the expansion test most accurate for the detection of potentially reactive granitic aggregates. Additionally, the authors showed that the duration of the AAR-3.1 test must be extended (above one year) for an accurate assessment. In the end it was concluded that AAR-4.1 was the most conservative test in the assessment of potentially reactive granitic aggregates and should be preferred mainly due to its short duration [360]. Generally, the standards fail in providing, simultaneously, the information concerning the reactivity of the aggregates in a reasonably short time and establishing an acceptance limit for the long-term expansion of a reactive combination aggregatecement. This contradiction has led to the development of more expeditious tests [8,67–69]. 3.4. Tests for ASR assessment The shortcomings found when using the available tests/standards have led researchers to improve the existing ones and to search for new ones. In 2005, Chatterji examined different methods in the light of fundamental understanding of the ASR mechanism [8]. Emphasis was given on the reactivity of the aggregates and not on the acceptance limit. The author concluded that the reactivity of the aggregates was best assessed in a solution of moderate OH concentration such as Ca(OH)2 and high ionic strength. It also showed that this was a simple and fast method to be employed on a daily basis to assess the quality of a chosen aggregate. In 2013, Donnell et al. [361], by means of microwave signal attenuation, proposed and demonstrated a new potential approach for ASR characterisation. The authors showed the potential of microwave diagnosis techniques for distinguishing between mortar samples containing reactive and non-reactive aggregates and concluded that significant amount of information could be extracted and used in combination with other measures (such as expansion length and microstructural data). ASR reduces concrete stiffness significantly [67]. Therefore the Stiffness Damage Test (SDT) was another technique for assessing concrete structures affected by ASR [67–70]. SDT is a method based on cyclic loading. Generally, five compression cycles (continuous loading and unloading) of concrete specimens are carried out using a fixed load at a certain loading rate. SDT suffered some changes through the time [67,68,362,363]. Recently, Islam and Ghafoori [27] tested several input parameters through SDT in order to verify its influence on the SDT responses. The authors concluded that the use of the SDT output values (e.g. hysteresis area and plastic deformation) for concrete damage characterisation due to ASR can be misleading since SDT is extremely sensitive to the concrete design strength and properties such as the aggregate’s type and contents. Furthermore, it was recommended that, for practical purposes, SDT should be conducted at 40% or even less damaged concrete at the time of the assessment in order to become a quantitative assessment tool. Finally, the authors concluded that the SDT was

922

R.B. Figueira et al. / Construction and Building Materials 222 (2019) 903–931

a powerful tool for concrete ASR damage assessing in spite of further work needed to be implemented in practical engineering applications. Acoustic emission technique (AET) shows several advantages since it is a non-invasive technique and excludes or limits the need of collecting samples for petrographic examination. Additionally, the method is practically immune to changes in environmental conditions (e.g. RH and temperature) while length measurements should be corrected for such variations to avoid errors and misinterpretation issues. The availability of self-powered and wireless AE equipment provides additional advantage particularly for ASR long-term assessment and monitoring. M. Abdelrahman et al. [364] reported a study using AET monitoring for ASR detection. Accelerated tests were implemented at the University of South Carolina Structures and Materials Laboratory by continuously monitoring the samples with AET. The authors performed length change measurements and petrographic examination, periodically, to serve as benchmarks. The micro-cracking related with ASR damage was detected by AE and the rate of AE activity was correlated to the rate of ASR damage. The authors concluded that the AET could detect ASR damage and the rate of AET could be related to the rate of ASR degradation. Furthermore, an acoustic emission intensity analysis chart for ASR damage classification was proposed. The chart was obtained by correlating the AET results with the petrographic examination and could be used for health monitoring to enable proper identification of the extent of ASR damage. Nevertheless, it was concluded that more data was needed to validate the proposed limits and extend the chart to include heavy ASR damage [364]. Rashidi et al. [365] proposed a multi-physics approach for the assessment of ASR. The authors obtained results from four test methods including expansion, nonlinear acoustics, microwave measurements, and quantitative petrographic analysis using the damage rating index. A connection between gel formation, damage, and expansion using the mentioned methods was established. [365]. In 2018, Teramoto et al. envisioned the development of micro-cracks due to ASR in two dimensions, using a digital image correlation (DIC) technique [366]. The authors concluded that the ASR-induced micro-cracking in concrete was effectively visualized using the DIC and the DIC-based estimations were in good agreement with the expansion values measured using a micrometer and contact gauge.

4. ASR modelling Modelling ASR is of tremendous importance to understand the mechanisms behind this process. Modelling allows the assessment of the current state of existing concrete structures, allows to improve the stability and monitor the need for repair. Therefore, the service life of concrete structures can be significantly improved with the aid of accurate predictive models. Nevertheless, there are multiple factors involved in the ASR process which leads, naturally, to the development of several modelling approaches with different purposes. Commonly, ASR models may be numerical or analytical. Rajabipour et al. [17] reported that analytical models are fast, but imprecise. On the other hand, numerical and finite element models are expensive, and provide additional information about the studied processes. Modelling can also be focused on different scales, e.g. from aggregate scale to large-scale concrete specimens. Furthermore, some of the most recent, and reliable models are based on a multi-scale pattern, extrapolating the results from the smaller scales to larger scales [367,368]. One of the most studied and reliable models for concrete-related aspects is the Lattice Discrete Particle Model (LDPM), proposed by Cusatis et al. in 2011 [369]. This

model has been applied specifically to modelling ASR expansion and cracking. The authors concluded that LDPM simulated concrete at the length scale of coarse aggregate pieces (meso-scale). The proposed LDPM simulated, accurately, the behaviour of concrete under unconfined compression which was mainly governed by tensile fracturing and cohesive/frictional shearing at the mesoscale. Liaudat et al. [370] developed a numerical chemo-mechanical model for ASR expansion based on the finite element model. The authors showed that the proposed model was capable to reproduce experimental observations and was able to capture the effect of external alkali diffusion in expansion development. However, some aspects such as pore filling effect of ASR gel were not included. Islam et al. [29] proposed an ASR decay model (ADM) to predict mortar expansion. The authors reported that the ADM model was able to predict the expansion of mortars, showing a good correlation between the analytical and the experimental results. Furthermore, this model, compared to others, have the advantage of predicting the mortar expansion in solutions with different alkali concentration. Other authors such as Qian et al. [78] developed and validated a model to predict the ASR expansion. Factors such as aggregate size, mechanical properties and alkali concentration were included. The chemical reactions were determined based on solid state reaction theory and the amount of ASR gel was assumed to be according to alkali consumption. Also, elastic plastic mechanics were used to simulate the change of stress within the concrete. The experimental results were similar to the model results proving its reliability. Ishkakhov et al. [367] proposed a multi-scale micromechanics model to describe the expansion of ASR based on linear elastic fracture mechanics. Two different ASR mechanisms linked to different types of reactive aggregates were modelled. The experimental results were in agreement with theoretical results, both qualitative and quantitative. Furthermore, the influence of the effect of external alkali supply was also included and studied. This model also allows to distinguish between different types of reactive aggregate. Alnaggar et al. [371] formulated an aggregate-scale model ASR – Lattice Discrete Particle Model (ASR-LDPM) that can precisely model several parameters such as expansion, cracking distribution, alkali concentration and temperature effect as well as the anisotropy of the expansion. The last parameter is one of the topics which is still open for further understanding/development. The same group [35] used LDPM to couple creep, shrinkage and ASR, controlling parameters such as temperature, humidity and cement hydration. LDPM has also been recently employed in a multi-scale approach [368]. The precise effect of SCMs was only considered in ASR modelling a few years ago. Chemical Index Model (CIM) was first proposed by Malvar and Lenke in 2006 [312], while studying the efficiency of FA on mitigating ASR based on the chemical composition. The authors proposed a model to estimate the expansion on test method ASTM C 1567 of FA-cement-aggregate mortars. CIM was later studied extensively by different authors [44,263,313,372]. In 2014, CIM was validated to reactive glass aggregates by Wright [44] and extended to other SCMs such as slag and natural pozzolans by Mahyar et al. in 2018 [372]. Lithium impregnation is also one of the topics related to ASR that was modelled in the last few years by several authors [77,255,373]. Recently, Mao et al. [373] developed a model to study the efficiency of the impregnation of lithium and the removal of chloride ions through an established electrochemical treatment (ELM-ECR: Electrochemical Lithium Migration and Electrochemical Chloride Removal). The proposed numerical model describes the interaction between aggregates and lithium ions in the mitigation of ASR, considering the ionic interaction between species, binding and the electrochemical reaction at electrodes. Several parameters such

R.B. Figueira et al. / Construction and Building Materials 222 (2019) 903–931

as time, temperature, current density and lithium concentration were taken into account. ASR mechanism is still open for debate in the scientific community since it is not completely understood. Further research is also needed to model the anisotropy of expansion. ASR modelling must also be approached from multiple points of view. Multi-scale approaches seem to be the best way to correctly predict all the parameters related to ASR, with applications that go from aggregate-level models to large concrete structures. A number of studies are keen to understand a single aspect, such as alkali leaching or the effect of SCMs. However, as the ASR is so complex, the effect of a single parameter may change considerably when studied in conjunction with others. Validation results can also be very dissimilar from one model to another, making it difficult to evaluate and compare the different models between each other. The compilation of several described models and the application in a more reliable and complete model, in which several aspects and parameters of ASR are considered, can be of interest in the near future.

5. Conclusions and recommendations for further research ASR in concrete is a major concern in civil engineering, affecting concrete structures all over the world. ASR is a complex chemical reaction that depends and is influenced by multiple factors. In this manuscript, a review of the most recent and relevant achievements, as well as the existing knowledge, concerning the reaction mechanisms of ASR was performed. The conditioning factors, diagnostic and prognostic methodologies, preventive measures and test methods of ASR were also debated. Recent publications have shown that multi-physics approaches generate new foundational comprehension of the nature of the ASR, that eventually may be used for the development of techniques and tools that allow the assessment and monitoring the evolution of ASR in new and existing concrete structures. The information gathered shows that several efforts have been devoted in understanding the fundamentals of ASR, which is reflected by the huge amount of data published in the last few decades. In spite of the significant progresses achieved, namely the development of predictive models and advancement of the actual state-of-art, the capacity to fully understand and describe the development of ASR remains incomplete. Concerning the preventive measures, lithium-containing additives are employed to control ASR. The most studied compound, at different levels of contents, was LiNO3. However, the mechanism by which lithium acts is not yet fully understood. To date, the research published on lithium additives has not provided a complete and comprehensive understanding of the mechanism or mechanisms by which this type of additive is effective in reducing the expansion due to the ASR. Therefore, it is challenging to predict the timeframe in which the lithium additives will act to suppress the ASR expansion. Data found in the references are unanimous in considering that, for the same level of suppression, it is necessary, compared to the FA, a higher content of BFS. Furthermore, the recommendations on the use of BFS for ASR are more conservative and less flexible than that for the use of FA. Most publications have indicated that the mechanism by which BFS inhibits the formation of ASR in concrete is not completely understood. The same was found in the case of FA. This may be due to the combination of more than one factor, when BFS or FA are used, that contributes for ASR mitigation. Most of the work developed and published using FA clearly shows its positive contribution. The few contradictory cases reported in the literature, on the replacement of cement by FA, are linked to the use of lower FA dosages or to the combination of other concrete degradation processes.

923

Concerning the use of SF further studies should be performed in order to clarify the mechanism responsible for its role. The available publications do not agree on how the MK works. The main flaws are directly related with models and mechanism involved in the ASR and the mechanistic role of additives in inhibiting/mitigating this reaction. When it comes to the role of aluminium, most of the studies found were performed using SCMs such as MK or FA. Therefore, the effects described may be only due to the aluminium ions or to the combination of several factors. The complexity of ASR is directly related to the numerous parameters that must be taken into account. Therefore, the search for systems that allow to assess and monitor, simultaneously, different parameters is of utmost importance. To the best of the authors knowledge, no sensor systems have been reported that focus on monitoring multiple parameters simultaneously, that directly or indirectly would provide information on ASR initiation process and further evolution. In spite of the amount of information published it is undeniable that several aspects require further investigation both in terms of preventive measures and palliative measures. For that, crossfertilisation is necessary and this implies involving several scientific areas such as chemistry, materials science, civil engineering and geology in order to overcome the limitations of each individual area. The paramount objective of this cross-fertilisation is the key to ensure that concrete fulfils its functions during the projected lifetime without requiring unplanned repairs. The information available allows to conclude that several research aspects should be pursued, namely:
A process of surveying and assessing the use of other types of materials available, namely reactive fine aggregates or other types of FA, as alternative to the SCMs currently in use.
Development of new methodologies (tests and sensors) that are both fast and reliable, to detect and monitor the ASR expansion.
Development and improvement of models to accurately predict the chemico-mechanical behaviour of concrete structures affected by ASR. Furthermore, the pursuit on improving the existing models and on the development of new, complete and multi-scaled approaches will, undoubtedly, dominate this research area. It is widely accepted that early detection of ASR implies saving costs in repair and rehabilitation and therefore increase the service life of existing structures. Moreover, ASR mitigation will also contribute to a reduction in greenhouse gas emissions, thus promoting sustainability. Despite there being a number of test methods available, none are completely safe or reliable. Nowadays, only the combination of several tests can reliably assess the reactivity of an aggregate. Acknowledgments The authors would like to thank the financial support provided by the project ‘‘SolSensors — Development of Advanced Fiber Optic Sensors for Monitoring the Durability of Concrete Structures”, with reference POCI-01-0145-FEDER-031220 supported by the Program Budget COMPETE — Operational Program Competitiveness and Internationalization — COMPETE 2020 and the Lisbon Regional Operational Program in its FEDER component and by the budget of FCT Foundation for Science and Technology, I.P and to Hugo Gomes for assisting in the execution of Figs. 1, 3 and 6. The authors acknowledge the support of Centro de Química, CQUM, and the Institute of Sustainability and Innovation in Structural Engineering, ISISE, which are financed by national funds through the FCT Foundation for Science and Technology, I.P. under the projects UID/

924

R.B. Figueira et al. / Construction and Building Materials 222 (2019) 903–931

QUI/00686/2016, UID/QUI/00686/2019 and UID/ECI/04029/2019 respectively. The authors acknowledge the important output provided by the reviewers. References [1] M. Raupach, P. Schiessl, Monitoring system for the penetration of chlorides, carbonation and the corrosion risk for the reinforcement, Constr. Build. Mater. 11 (1997) 207–214, https://doi.org/10.1016/S0950-0618(97)00039-1. [2] S. Diamond, A review of alkali-silica reaction and expansion mechanisms 1. Alkalies in cements and in concrete pore solutions, Cem. Concr. Res. 5 (1975) 329–345. [3] R.B. Figueira, A. Sadovski, A.P. Melo, E.V. Pereira, Chloride threshold value to initiate reinforcement corrosion in simulated concrete pore solutions: the influence of surface finishing and pH, Constr. Build. Mater. 141 (2017) 183– 200, https://doi.org/10.1016/j.conbuildmat.2017.03.004. [4] H. Böhni, Corrosion in Reinforced Concrete Structures, Elsevier, 2005. [5] L. Bertolini, Steel corrosion and service life of reinforced concrete structures, Struct. Infrastruct. Eng. 4 (2008) 123–137, https://doi.org/10.1080/ 15732470601155490. [6] T.E. Stanton, Expansion of Concrete through Reaction between Cement and Aggregate, 2008. https://trid.trb.org/view/868520 (accessed July 19, 2018). [7] S. Diamond, A review of alkali-silica reaction and expansion mechanisms 2. Reactive aggregates, Cem. Concr. Res. 6 (1976) 549–560. [8] S. Chatterji, Chemistry of alkali–silica reaction and testing of aggregates, Cem. Concr. Compos. 27 (2005) 788–795, https://doi.org/10.1016/j. cemconcomp.2005.03.005. [9] S. Chatterji, The role of Ca (OH) 2 in the breakdown of Portland cement concrete due to alkali-silica reaction, Cem. Concr. Res. 9 (1979) 185–188. [10] S. Urhan, Alkali silica and pozzolanic reactions in concrete. Part 1: Interpretation of published results and an hypothesis concerning the mechanism, Cem. Concr. Res. 17 (1987) 141–152. [11] S. Chatterji, N. Thaulow, A.D. Jensen, Studies of alkali-silica reaction, part 6. Practical implications of a proposed reaction mechanism, Cem. Concr. Res. 18 (1988) 363–366. [12] S. Chatterji, N. Thaulow, A.D. Jensen, Studies of alkali-silica reaction. Part 5. Verification of a newly proposed reaction mechanism, Cem. Concr. Res. 19 (1989) 177–183. [13] H. Wang, J.E. Gillott, Mechanism of alkali-silica reaction and the significance of calcium hydroxide, Cem. Concr. Res. 21 (1991) 647–654. [14] E. Garcia-Diaz, J. Riche, D. Bulteel, C. Vernet, Mechanism of damage for the alkali–silica reaction, Cem. Concr. Res. 36 (2006) 395–400, https://doi.org/ 10.1016/j.cemconres.2005.06.003. [15] T. Ichikawa, M. Miura, Modified model of alkali-silica reaction, Cem. Concr. Res. 37 (2007) 1291–1297, https://doi.org/10.1016/j.cemconres.2007.06.008. [16] X. Feng, M.D.A. Thomas, T.W. Bremner, K.J. Folliard, B. Fournier, New observations on the mechanism of lithium nitrate against alkali silica reaction (ASR), Cem. Concr. Res. 40 (2010) 94–101, https://doi.org/10.1016/ j.cemconres.2009.07.017. [17] F. Rajabipour, E. Giannini, C. Dunant, J.H. Ideker, M.D.A. Thomas, Alkali–silica reaction: current understanding of the reaction mechanisms and the knowledge gaps, Cem. Concr. Res. 76 (2015) 130–146, https://doi.org/ 10.1016/j.cemconres.2015.05.024. [18] A.K. Saha, M.N.N. Khan, P.K. Sarker, F.A. Shaikh, A. Pramanik, The ASR mechanism of reactive aggregates in concrete and its mitigation by fly ash: a critical review, Constr. Build. Mater. 171 (2018) 743–758, https://doi.org/ 10.1016/j.conbuildmat.2018.03.183. [19] Y. Kawabata, K. Yamada, The mechanism of limited inhibition by fly ash on expansion due to alkali–silica reaction at the pessimum proportion, Cem. Concr. Res. 92 (2017) 1–15, https://doi.org/10.1016/j.cemconres.2016. 11.002. [20] C. Hu, B.P. Gautam, D. Shang, F. Wang, D.K. Panesar, Atomic force microscopy characterisation of alkali-silica reaction products to reveal their nanostructure and formation mechanism, Ceram. Int. 44 (2018) 7310–7314, https://doi.org/10.1016/j.ceramint.2018.01.069. [21] B. Fournier, M.-A. Bérubé, Alkali-aggregate reaction in concrete: a review of basic concepts and engineering implications, Can. J. Civ. Eng. 27 (2000) 167– 191, https://doi.org/10.1139/l99-072. [22] M. Thomas, The effect of supplementary cementing materials on alkali-silica reaction: a review, Cem. Concr. Res. 41 (2011) 1224–1231, https://doi.org/ 10.1016/j.cemconres.2010.11.003. [23] J. Lindgård, Ö. Andiç-Çakır, I. Fernandes, T.F. Rønning, M.D.A. Thomas, Alkali– silica reactions (ASR): literature review on parameters influencing laboratory performance testing, Cem. Concr. Res. 42 (2012) 223–243, https://doi.org/ 10.1016/j.cemconres.2011.10.004. [24] F. Moradi-Marani, P. Rivard, Nondestructive assessment of alkali-silica reaction in concrete: a review, in: O. Büyüköztürk, M.A. Tasßdemir, O. Günesß, Y. Akkaya (Eds.), Nondestruct. Test. Mater. Struct., Springer Netherlands, 2013, pp. 317–322. [25] C. Shi, Z. Shi, X. Hu, R. Zhao, L. Chong, A review on alkali-aggregate reactions in alkali-activated mortars/concretes made with alkali-reactive aggregates, Mater. Struct. 48 (2015) 621–628, https://doi.org/10.1617/s11527-014-0505-2.

[26] S. Poyet, A. Sellier, B. Capra, G. Foray, J.-M. Torrenti, H. Cognon, E. Bourdarot, Chemical modelling of alkali silica reaction: influence of the reactive aggregate size distribution, Mater. Struct. 40 (2007) 229. [27] M.S. Islam, N. Ghafoori, Evaluation of alkali–silica reactivity using ASR kinetic model, Constr. Build. Mater. 45 (2013) 270–274, https://doi.org/10.1016/ j.conbuildmat.2013.03.048. [28] L.F.M. Sanchez, S. Multon, A. Sellier, M. Cyr, B. Fournier, M. Jolin, Comparative study of a chemo–mechanical modeling for alkali silica reaction (ASR) with experimental evidences, Constr. Build. Mater. 72 (2014) 301–315, https://doi. org/10.1016/j.conbuildmat.2014.09.007. [29] M.S. Islam, Prediction of ultimate expansion of ASTM C 1260 for various alkali solutions using the proposed decay model, Constr. Build. Mater. 77 (2015) 317–326, https://doi.org/10.1016/j.conbuildmat.2014.12.087. [30] R. Pignatelli, C. Comi, P.J.M. Monteiro, A coupled mechanical and chemical damage model for concrete affected by alkali–silica reaction, Cem. Concr. Res. 53 (2013) 196–210, https://doi.org/10.1016/j.cemconres.2013.06.011. [31] S. Chatterji, P. Christensen, Studies of alkali-silica reaction. Part 7. Modelling of expansion, Cem. Concr. Res. 20 (1990) 285–290. [32] V.E. Saouma, R.A. Martin, M.A. Hariri-Ardebili, T. Katayama, A mathematical model for the kinetics of the alkali–silica chemical reaction, Cem. Concr. Res. 68 (2015) 184–195, https://doi.org/10.1016/j.cemconres.2014.10.021. [33] M.M. Karthik, J.B. Mander, S. Hurlebaus, ASR/DEF related expansion in structural concrete: model development and validation, Constr. Build. Mater. 128 (2016) 238–247, https://doi.org/10.1016/j.conbuildmat.2016.10.084. [34] Y. Kawabata, J.-F. Seignol, R.-P. Martin, F. Toutlemonde, Macroscopic chemomechanical modeling of alkali-silica reaction of concrete under stresses, Constr. Build. Mater. 137 (2017) 234–245, https://doi.org/10.1016/ j.conbuildmat.2017.01.090. [35] M. Alnaggar, G. Di Luzio, G. Cusatis, Modeling time-dependent behavior of concrete affected by alkali silica reaction in variable environmental conditions, Materials 10 (2017), https://doi.org/10.3390/ma10050471. [36] B. Li, L. Baingam, K. Kurumisawa, T. Nawa, L. XiaoZhou, Micro-mechanical modelling for the prediction of alkali-silica reaction (ASR) expansion: Influence of curing temperature conditions, Constr. Build. Mater. 164 (2018) 554–569, https://doi.org/10.1016/j.conbuildmat.2018.01.007. [37] W. Wallau, S. Pirskawetz, K. Voland, B. Meng, Continuous expansion measurement in accelerated concrete prism testing for verifying ASRexpansion models, Mater. Struct. 51 (2018), https://doi.org/10.1617/ s11527-018-1205-0. [38] A. Joshaghani, The effect of trass and fly ash in minimizing alkali-carbonate reaction in concrete, Constr. Build. Mater. 150 (2017) 583–590, https://doi. org/10.1016/j.conbuildmat.2017.06.034. [39] J. Feiteira, M.S. Ribeiro, Polymer action on alkali–silica reaction in cement mortar, Cem. Concr. Res. 44 (2013) 97–105, https://doi.org/10.1016/j. cemconres.2012.09.008. [40] J. Duchesne, M.-A. Bérubé, Long-term effectiveness of supplementary cementing materials against alkali–silica reaction, Cem. Concr. Res. 31 (2001) 1057–1063, https://doi.org/10.1016/S0008-8846(01)00538-5. [41] A. Santos Silva, D. Soares, L. Divet, A.C. Ribeiro, Prevenção da reação sulfática interna no betão. Resultados a longo prazo do efeito de adições minerais, 2016. [42] P. Shen, L. Lu, W. Chen, F. Wang, S. Hu, Efficiency of metakaolin in steam cured high strength concrete, Constr. Build. Mater. 152 (2017) 357–366, https://doi. org/10.1016/j.conbuildmat.2017.07.006. [43] Z. Shi, C. Shi, J. Zhang, S. Wan, Z. Zhang, Z. Ou, Alkali-silica reaction in waterglass-activated slag mortars incorporating fly ash and metakaolin, Cem. Concr. Res. 108 (2018) 10–19, https://doi.org/10.1016/j. cemconres.2018.03.002. [44] J.R. Wright, S. Shafaatian, F. Rajabipour, Reliability of chemical index model in determining fly ash effectiveness against alkali-silica reaction induced by highly reactive glass aggregates, Constr. Build. Mater. 64 (2014) 166–171, https://doi.org/10.1016/j.conbuildmat.2014.04.042. [45] K. Yıldırım, M. Sümer, Comparative analysis of fly ash effect with three different method in mortars that are exposed to alkali silica reaction, Compos. Part B Eng. 61 (2014) 110–115, https://doi.org/10.1016/ j.compositesb.2014.01.004. [46] S. Wang, Cofired biomass fly ashes in mortar: reduction of Alkali Silica Reaction (ASR) expansion, pore solution chemistry and the effects on compressive strength, Constr. Build. Mater. 82 (2015) 123–132, https://doi. org/10.1016/j.conbuildmat.2015.02.021. [47] A.P. More, S.T. Mhaske, Anticorrosive coating of polyesteramide resin by functionalized ZnO-Al2O3-Fly ash composite and functionalized multiwalled carbon nanotubes, Prog. Org. Coat. 99 (2016) 240–250, https://doi.org/ 10.1016/j.porgcoat.2016.06.005. [48] V. Mazarei, D. Trejo, J.H. Ideker, O.B. Isgor, Synergistic effects of ASR and fly ash on the corrosion characteristics of RC systems, Constr. Build. Mater. 153 (2017) 647–655, https://doi.org/10.1016/j.conbuildmat.2017.07.097. [49] K. Turk, C. Kina, M. Bagdiken, Use of binary and ternary cementitious blends of F-Class fly-ash and limestone powder to mitigate alkali-silica reaction risk, Constr. Build. Mater. 151 (2017) 422–427, https://doi.org/10.1016/ j.conbuildmat.2017.06.075. [50] W.C. Wang, Effects of fly ash and lithium compounds on the water-soluble alkali and lithium content of cement specimens, Constr. Build. Mater. 50 (2014) 727–735, https://doi.org/10.1016/j.conbuildmat.2013.10.024.

R.B. Figueira et al. / Construction and Building Materials 222 (2019) 903–931 [51] K.E. Kurtis, P.J.M. Monteiro, Chemical additives to control expansion of alkalisilica reaction gel: proposed mechanisms of control, J. Mater. Sci. 38 (2003) 2027–2036. [52] A. Leemann, L. Bernard, S. Alahrache, F. Winnefeld, ASR prevention—effect of aluminum and lithium ions on the reaction products, Cem. Concr. Res. 76 (2015) 192–201, https://doi.org/10.1016/j.cemconres.2015.06.002. [53] J. Zapała-Sławeta, Z. Owsiak, The role of lithium compounds in mitigating alkali-gravel aggregate reaction, Constr. Build. Mater. 115 (2016) 299–303, https://doi.org/10.1016/j.conbuildmat.2016.04.058. [54] L.M.S. Souza, R.B. Polder, O. Çopurog˘lu, Lithium migration in a two-chamber set-up as treatment against expansion due to alkali–silica reaction, Constr. Build. Mater. 134 (2017) 324–335, https://doi.org/10.1016/ j.conbuildmat.2016.12.052. [55] S.-Y. Hong, F.P. Glasser, Alkali sorption by C-S-H and C-A-S-H gels: Part II. Role of alumina, Cem. Concr. Res. 32 (2002) 1101–1111, https://doi.org/10.1016/ S0008-8846(02)00753-6. [56] A. Carse, P.F. Dux, Measurement of concrete expansive strains due to alkalisilica reaction in Australian concrete structures, Cem. Concr. Res. 20 (1990) 376–384. [57] S. Diamond, ASR-Another Look at Mechanisms, 8-th International Conference on Alkali-Aggregate Reaction, 1989, pp. 83–94. [58] M.T. Blanco-Varela, S. Martínez-Ramírez, T. Vázquez, S. Sánchez-Moral, Role of alkalis of aggregate origin in the deterioration of CAC concrete, Cem. Concr. Res. 35 (2005) 1698–1704, https://doi.org/10.1016/j.cemconres.2004.08.015. [59] M.A.T.M. Broekmans, Deleterious reactions of aggregate with alkalis in concrete, Rev. Mineral. Geochem. 74 (2012) 279–364, https://doi.org/ 10.2138/rmg.2012.74.7. [60] R. Tänzer, Y. Jin, D. Stephan, Effect of the inherent alkalis of alkali activated slag on the risk of alkali silica reaction, Cem. Concr. Res. 98 (2017) 82–90, https://doi.org/10.1016/j.cemconres.2017.04.009. [61] J.M. Ponce, O.R. Batic, Different manifestations of the alkali-silica reaction in concrete according to the reaction kinetics of the reactive aggregate, Cem. Concr. Res. 36 (2006) 1148–1156, https://doi.org/10.1016/j. cemconres.2005.12.022. [62] S. Multon, F. Toutlemonde, Effect of moisture conditions and transfers on alkali silica reaction damaged structures, Cem. Concr. Res. 40 (2010) 924– 934, https://doi.org/10.1016/j.cemconres.2010.01.011. [63] R.A.D. Jr, E.R. Giannini, T. Drimalas, B. Fournier, W.M. Hale, Effects of moisture, temperature, and freezing and thawing on alkali-silica reaction, Mater. J. 115 (2018) 575–584, https://doi.org/10.14359/5170219. [64] I. Dehri, M. Erbil, The effect of relative humidity on the atmospheric corrosion of defective organic coating materials: an EIS study with a new approach, Corros. Sci. 42 (2000) 969–978, https://doi.org/10.1016/S0010-938X(99) 00126-2. [65] B.P. Gautam, D.K. Panesar, The effect of elevated conditioning temperature on the ASR expansion, cracking and properties of reactive Spratt aggregate concrete, Constr. Build. Mater. 140 (2017) 310–320, https://doi.org/10.1016/ j.conbuildmat.2017.02.104. [66] T. Ueda, J. Kushida, M. Tsukagoshi, A. Nanasawa, Influence of temperature on electrochemical remedial measures and complex deterioration due to chloride attack and ASR, Constr. Build. Mater. 67 (2014) 81–87, https://doi. org/10.1016/j.conbuildmat.2013.10.020. [67] L.F.M. Sanchez, B. Fournier, M. Jolin, J. Bastien, Evaluation of the Stiffness Damage Test (SDT) as a tool for assessing damage in concrete due to alkalisilica reaction (ASR): input parameters and variability of the test responses, Constr. Build. Mater. 77 (2015) 20–32, https://doi.org/10.1016/ j.conbuildmat.2014.11.071. [68] L.F.M. Sanchez, B. Fournier, M. Jolin, J. Bastien, D. Mitchell, Practical use of the Stiffness Damage Test (SDT) for assessing damage in concrete infrastructure affected by alkali-silica reaction, Constr. Build. Mater. 125 (2016) 1178–1188, https://doi.org/10.1016/j.conbuildmat.2016.08.101. [69] E.R. Giannini, L.F.M. Sanchez, A. Tuinukuafe, K.J. Folliard, Characterization of concrete affected by delayed ettringite formation using the stiffness damage test, Constr. Build. Mater. 162 (2018) 253–264, https://doi.org/10.1016/ j.conbuildmat.2017.12.012. [70] M.S. Islam, N. Ghafoori, A new approach to evaluate alkali-silica reactivity using loss in concrete stiffness, Constr. Build. Mater. 167 (2018) 578–586, https://doi.org/10.1016/j.conbuildmat.2018.02.047. [71] S. Multon, M. Cyr, A. Sellier, P. Diederich, L. Petit, Effects of aggregate size and alkali content on ASR expansion, Cem. Concr. Res. 40 (2010) 508–516, https:// doi.org/10.1016/j.cemconres.2009.08.002. [72] C.F. Dunant, K.L. Scrivener, Effects of aggregate size on alkali–silica-reaction induced expansion, Cem. Concr. Res. 42 (2012) 745–751, https://doi.org/ 10.1016/j.cemconres.2012.02.012. [73] S. Multon, Évaluation expérimentale et théorique des effets mécaniques de l’alcali-réaction sur des structures modèles, thesis, Marne-la-Vallée, 2003. http://www.theses.fr/2003MARN0181 (accessed September 28, 2018). [74] M.S. Islam, S. Akhtar, A critical assessment to the performance of Alkali-Silica Reaction (ASR) in concrete, Can. Chem. Trans. 253–266 (2013), https://doi. org/10.13179/canchemtrans.2013.01.04.0026. [75] J. Feiteira, J. Custódio, M.S.S. Ribeiro, Review and discussion of polymer action on alkali–silica reaction, Mater. Struct. 46 (2013) 1415–1427, https://doi.org/ 10.1617/s11527-012-9983-2. [76] I. Fernandes, M.A.T.M. Broekmans, Alkali–silica reactions: an overview. Part I, Metallogr. Microstruct. Anal. 2 (4) (2013) 257–267, https://doi.org/10.1007/ s13632-013-0085-5.

925

[77] T. Kim, J. Olek, H. Jeong, Alkali–silica reaction: kinetics of chemistry of pore solution and calcium hydroxide content in cementitious system, Cem. Concr. Res. 71 (2015) 36–45, https://doi.org/10.1016/j.cemconres.2015.01.017. [78] C. Qian, Y. Zhuang, H. Huang, Numerical calculation of expansion induced by alkali silica reaction, Constr. Build. Mater. 103 (2016) 117–122, https://doi. org/10.1016/j.conbuildmat.2015.11.041. [79] P. Krivenko, R. Drochytka, A. Gelevera, E. Kavalerova, Mechanism of preventing the alkali–aggregate reaction in alkali activated cement concretes, Cem. Concr. Compos. 45 (2014) 157–165, https://doi.org/ 10.1016/j.cemconcomp.2013.10.003. [80] R. Esposito, M.A.N. Hendriks, A multiscale micromechanical approach to model the deteriorating impact of alkali-silica reaction on concrete, Cem. Concr. Compos. 70 (2016) 139–152, https://doi.org/10.1016/j. cemconcomp.2016.03.017. [81] Q. Huang, P. Gardoni, D. Trejo, A. Pagnotta, Probabilistic model for steel– concrete bond behavior in bridge columns affected by alkali silica reactions, Eng. Struct. 71 (2014) 1–11, https://doi.org/10.1016/j. engstruct.2014.03.041. [82] X.X. Gao, S. Multon, M. Cyr, A. Sellier, Optimising an expansion test for the assessment of alkali-silica reaction in concrete structures, Mater. Struct. 44 (2011) 1641–1653, https://doi.org/10.1617/s11527-011-9724-y. [83] H.W. Reinhardt, O. Mielich, A fracture mechanics approach to the crack formation in alkali-sensitive grains, Cem. Concr. Res. 41 (2011) 255–262, https://doi.org/10.1016/j.cemconres.2010.11.008. [84] L. Charpin, A. Ehrlacher, A computational linear elastic fracture mechanicsbased model for alkali–silica reaction, Cem. Concr. Res. 42 (2012) 613–625, https://doi.org/10.1016/j.cemconres.2012.01.004. [85] X. Hou, L.J. Struble, R.J. Kirkpatrick, Formation of ASR gel and the roles of C-SH and portlandite, Cem. Concr. Res. 34 (2004) 1683–1696, https://doi.org/ 10.1016/j.cemconres.2004.03.026. [86] M.A.T.M. Broekmans, Structural properties of quartz and their potential role for ASR, Mater. Charact. 53 (2004) 129–140, https://doi.org/10.1016/ j.matchar.2004.08.010. [87] Chapter 2 Alkali-Silica Reaction – Guidelines for The Use of Lithium to Mitigate Or Prevent Alkali-Silica Reaction (Asr), July 2003 – FHWA-RD-03047, (n.d.). https://www.fhwa.dot.gov/publications/research/infrastructure/ pavements/pccp/03047/02.cfm (accessed March 4, 2019). [88] Š. Šachlová, A. Kucharˇová, Z. Pertold, R. Prˇikryl, M. Fridrichová, Quantitative assessment of alkali silica reaction potential of quartz-rich aggregates: comparison of chemical test and accelerated mortar bar test improved by SEM-PIA, Bull. Eng. Geol. Environ. 76 (2017) 133–144, https://doi.org/ 10.1007/s10064-015-0812-z. [89] M.-A. Bérubé, B. Fournier, Canadian experience with testing for alkaliaggregate reactivity in concrete, Cem. Concr. Compos. 15 (1993) 27–47, https://doi.org/10.1016/0958-9465(93)90037-A. [90] D.W. Hobbs, Alkali-silica reaction in concrete, Telford, 1988. [91] M.-A. Bérubé, J. Duchesne, J.F. Dorion, M. Rivest, Laboratory assessment of alkali contribution by aggregates to concrete and application to concrete structures affected by alkali–silica reactivity, Cem. Concr. Res. 32 (2002) 1215–1227, https://doi.org/10.1016/S0008-8846(02)00766-4. [92] B. Fournier, J.H. Ideker, K.J. Folliard, M.D.A. Thomas, P.-C. Nkinamubanzi, R. Chevrier, Effect of environmental conditions on expansion in concrete due to alkali–silica reaction (ASR), Mater. Charact. 60 (2009) 669–679, https://doi. org/10.1016/j.matchar.2008.12.018. [93] ASTM C294-12(2017), Standard Descriptive Nomenclature for Constituents of Concrete Aggregates, 2017. https://www.astm.org/Standards/C294.htm. [94] H.F.W. Taylor (Ed.), Cement Chemistry, Thomas Telford, 1997. [95] W. Kurdowski, Cement and Concrete Chemistry, Springer Netherlands, 2014. https://www.springer.com/us/book/9789400779440 (accessed September 7, 2018). [96] S. Multon, Evaluation experimentale et theorique des effets mecaniques de l’alcali-reaction sur des structures modeles, ETUDES Rech. Lab. PONTS CHAUSSEES, 2004. https://trid.trb.org/view/967507 (accessed May 23, 2019) [97] LNEC E 461-2007 – Betões. Metodologias para prevenir reações expansivas internas, Laboratório Nacional de Engenharia Civil, 2007. [98] T. Ichikawa, Alkali–silica reaction, pessimum effects and pozzolanic effect, Cem. Concr. Res. 39 (2009) 716–726, https://doi.org/10.1016/j. cemconres.2009.06.004. [99] E. Garcia-Diaz, D. Bulteel, Y. Monnin, P. Degrugilliers, P. Fasseu, ASR pessimum behaviour of siliceous limestone aggregates, Cem. Concr. Res. 40 (2010) 546–549, https://doi.org/10.1016/j.cemconres.2009.08.011. [100] X.X. Gao, S. Multon, M. Cyr, A. Sellier, Alkali–silica reaction (ASR) expansion: pessimum effect versus scale effect, Cem. Concr. Res. 44 (2013) 25–33, https://doi.org/10.1016/j.cemconres.2012.10.015. [101] K. Ramyar, A. Topal, Ö. Andiç, Effects of aggregate size and angularity on alkali–silica reaction, Cem. Concr. Res. 35 (2005) 2165–2169, https://doi.org/ 10.1016/j.cemconres.2005.03.010. [102] H. Du, K.H. Tan, Effect of particle size on alkali–silica reaction in recycled glass mortars, Constr. Build. Mater. 66 (2014) 275–285, https://doi.org/ 10.1016/j.conbuildmat.2014.05.092. [103] S. Diamond, S. Sahu, Densified silica fume: particle sizes and dispersion in concrete, Mater. Struct. 39 (2006) 849–859, https://doi.org/10.1617/s11527006-9087-y. [104] S. Multon, M. Cyr, A. Sellier, N. Leklou, L. Petit, Coupled effects of aggregate size and alkali content on ASR expansion, Cem. Concr. Res. 38 (2008) 350– 359, https://doi.org/10.1016/j.cemconres.2007.09.013.

926

R.B. Figueira et al. / Construction and Building Materials 222 (2019) 903–931

[105] S. Multon, A. Sellier, Multi-scale analysis of alkali–silica reaction (ASR): impact of alkali leaching on scale effects affecting expansion tests, Cem. Concr. Res. 81 (2016) 122–133, https://doi.org/10.1016/j. cemconres.2015.12.007. [106] Š. Šachlová, Microstructure parameters affecting alkali–silica reactivity of aggregates, Constr. Build. Mater. 49 (2013) 604–610, https://doi.org/10.1016/ j.conbuildmat.2013.08.087. [107] M.C.G. Juenger, C.P. Ostertag, Alkali–silica reactivity of large silica fumederived particles, Cem. Concr. Res. 34 (2004) 1389–1402, https://doi.org/ 10.1016/j.cemconres.2004.01.001. [108] J.J. Kollek, S.P. Varma, C. Zaris, Measurement of OH- concentrations of pore fluids and expansion due to alkali-silica reaction in composite cement mortars, in: Proc 8th Int Congr. Chem. Cem. vol. 3, 1986, pp. 183–189. [109] S. Diamond, Effects of mcrosilica (silica fume) on pore-solution chemistry of cement pastes C-82-C–84, J. Am. Ceram. Soc. 66 (1983), https://doi.org/ 10.1111/j.1151-2916.1983.tb10060.x. [110] K. Zheng, Pozzolanic reaction of glass powder and its role in controlling alkali–silica reaction, Cem. Concr. Compos. 67 (2016) 30–38, https://doi.org/ 10.1016/j.cemconcomp.2015.12.008. [111] S.M.S. Kazmi, M.J. Munir, I. Patnaikuni, Y.-F. Wu, Pozzolanic reaction of sugarcane bagasse ash and its role in controlling alkali silica reaction, Constr. Build. Mater. 148 (2017) 231–240, https://doi.org/10.1016/ j.conbuildmat.2017.05.025. [112] K. Pettersson, Effects of silica fume on alkali-silica expansion in mortar specimens, Cem. Concr. Res. 22 (1992) 15–22, https://doi.org/10.1016/00088846(92)90131-E. [113] A.M. Boddy, R.D. Hooton, M.D.A. Thomas, The effect of the silica content of silica fume on its ability to control alkali–silica reaction, Cem. Concr. Res. 33 (2003) 1263–1268, https://doi.org/10.1016/S0008-8846(03)00058-9. [114] M. Manera, Ø. Vennesland, L. Bertolini, Chloride threshold for rebar corrosion in concrete with addition of silica fume, Corros. Sci. 50 (2008) 554–560, https://doi.org/10.1016/j.corsci.2007.07.007. [115] A.A. Ramezanianpour, M.A. Moeini, Mechanical and durability properties of alkali activated slag coating mortars containing nanosilica and silica fume, Constr. Build. Mater. 163 (2018) 611–621, https://doi.org/10.1016/ j.conbuildmat.2017.12.062. [116] S. Chatterji, An accelerated method for the detection of alkali-aggregate reactivities of aggregates, Cem. Concr. Res. 8 (1978) 647–649. [117] M. Berra, Alkali-silica reactivity criteria for concrete aggregates, Mater. Struct. 38 (2005) 373–380, https://doi.org/10.1617/14144. [118] Y. Boukari, D. Bulteel, P. Rivard, N.-E. Abriak, Combining nonlinear acoustics and physico-chemical analysis of aggregates to improve alkali–silica reaction monitoring, Cem. Concr. Res. 67 (2015) 44–51, https://doi.org/10.1016/j. cemconres.2014.08.005. [119] H. Yang, P. Li, M. Rao, Long term investigation and inhibition on alkaliaggregates reaction of Three Gorges Dam concrete, Constr. Build. Mater. 151 (2017) 673–681, https://doi.org/10.1016/j.conbuildmat.2017.06.060. [120] C. Rößler, B. Möser, C. Giebson, H.-M. Ludwig, Application of Electron Backscatter Diffraction to evaluate the ASR risk of concrete aggregates, Cem. Concr. Res. 95 (2017) 47–55, https://doi.org/10.1016/j. cemconres.2017.02.015. [121] C. Drolet, J. Duchesne, B. Fournier, Effect of alkali release by aggregates on alkali-silica reaction, Constr. Build. Mater. 157 (2017) 263–276, https://doi. org/10.1016/j.conbuildmat.2017.09.085. [122] J.H.P. van Aardt, S. Visser, Formation of hydrogarnets: calcium hydroxide attack on clays and feldspars, Cem. Concr. Res. 7 (1977) 39–44, https://doi. org/10.1016/0008-8846(77)90006-0. [123] A.C.I. Committee 221, A.C. Institute, State-of-the-art report on alkaliaggregate reactivity, Farmington Hills, Mich. : American Concrete Institute, 1998. https://trove.nla.gov.au/version/43881016. [124] M. Santos, J. de Brito, O panorama nacional das reacções álcalis-sílica em betões, Rev. Eng. Civ. UM. 32 (2008) 57–71. [125] R.E. Obersholster, Alkali-aggregate reaction in South Africa: Some recent developments in research, in: Kyoto, Japan, 1986: pp. 77–82. [126] M.A. Bérubé, C. Tremblay, B. Fournier, M.D. Thomas, D.B. Stokes, Influence of lithium-based products proposed for counteracting ASR on the chemistry of pore solution and cement hydrates, Cem. Concr. Res. 34 (2004) 1645–1660, https://doi.org/10.1016/j.cemconres.2004.03.025. [127] H. Afcesa, T.A. Fl, ETL 06-2 Alkali-Aggregate Reaction in Portland-Cement Concrete (PCC) Airfield Pavements, (n.d.) 30. [128] B.J. Wigum, Examination of microstructural features of Norwegian cataclastic rocks and their use for predicting alkali-reactivity in concrete, Eng. Geol. 40 (1995) 195–214, https://doi.org/10.1016/0013-7952(95)00044-5. [129] T. Kim, J. Olek, Chemical sequence and Kinetics of Alkali-Silica Reaction Part I. Experiments, J. Am. Ceram. Soc. 97 (2014) 2195–2203, https://doi.org/ 10.1111/jace.12992. [130] T. Kim, J. Olek, Chemical Sequence and Kinetics of Alkali-Silica Reaction Part II. A thermodynamic model, J. Am. Ceram. Soc. 97 (2014) 2204–2212, https:// doi.org/10.1111/jace.12830. [131] A. Gholizadeh-Vayghan, F. Rajabipour, The influence of alkali–silica reaction (ASR) gel composition on its hydrophilic properties and free swelling in contact with water vapor, Cem. Concr. Res. 94 (2017) 49–58, https://doi.org/ 10.1016/j.cemconres.2017.01.006. [132] S. Kandasamy, M.H. Shehata, The capacity of ternary blends containing slag and high-calcium fly ash to mitigate alkali silica reaction, Cem. Concr.

[133]

[134]

[135]

[136]

[137]

[138]

[139]

[140]

[141]

[142]

[143] [144]

[145]

[146]

[147]

[148]

[149] [150]

[151]

[152]

[153]

[154]

[155]

Compos. 49 (2014) 92–99, https://doi.org/10.1016/j. cemconcomp.2013.12.008. A. Leemann, G.L. Saout, F. Winnefeld, D. Rentsch, B. Lothenbach, Alkali-silica reaction: the influence of calcium on silica dissolution and the formation of reaction products, J. Am. Ceram. Soc. 94 (2011) 1243–1249, https://doi.org/ 10.1111/j.1551-2916.2010.04202.x. H. Myadraboina, S. Setunge, I. Patnaikuni, Pozzolanic Index and lime requirement of low calcium fly ashes in high volume fly ash mortar, Constr. Build. Mater. 131 (2017) 690–695, https://doi.org/10.1016/ j.conbuildmat.2016.11.038. F. Bektas, L. Turanli, T. Topal, M.C. Goncuoglu, Alkali reactivity of mortars containing chert and incorporating moderate-calcium fly ash, Cem. Concr. Res. 34 (2004) 2209–2214, https://doi.org/10.1016/j.cemconres.2004.02.007. S. Chatterji, A discussion of the paper ‘‘Mechanism of alkali-silica reaction and the significance of calcium hydroxide” by H. Wang and JE Gillot, Cem. Concr. Res. 22 (1992) 190–192. H. Maraghechi, F. Rajabipour, C.G. Pantano, W.D. Burgos, Effect of calcium on dissolution and precipitation reactions of amorphous silica at high alkalinity, Cem. Concr. Res. 87 (2016) 1–13, https://doi.org/10.1016/j. cemconres.2016.05.004. S. Guo, Q. Dai, R. Si, Effect of calcium and lithium on alkali-silica reaction kinetics and phase development, Cem. Concr. Res. 115 (2019) 220–229, https://doi.org/10.1016/j.cemconres.2018.10.007. T.C. Powers, H.H. Steinour, an interpretation of some published researches on the alkali-aggregate reaction Part I – the chemical reactions and mechanism of expansion Part II – a hypothesis concerning safe and unsafe reactions with reactive, Am Concr. Inst J. Proc. (1955). https://trid.trb.org/view/95712 (accessed May 27, 2019). J. Duchesne, M.A. Bérubé, The effectiveness of supplementary cementing materials in suppressing expansion due to ASR: another look at the reaction mechanisms part 1: Concrete expansion and portlandite depletion, Cem. Concr. Res. 24 (1994) 73–82, https://doi.org/10.1016/0008-8846(94)900841. A. Sellier, E. Bourdarot, S. Multon, M. Cyr, E. Grimal, Combination of structural monitoring and laboratory tests for assessment of alkali-aggregate reaction swelling: application to gate structure dam, Mater. J. 106 (2009) 281–290, https://doi.org/10.14359/56553. A.B. Poole, Alkali silica reactivity mechanisms of gel formation and expansion, in: Proc. 9th Int. Conf. Alkali-Silica React. Concr. ICAAR, London, England, 1992: pp. 782–789. J. Liaudat, C.M. López, I.B. Carol, Diffusion-reaction model for alkali-silica reaction in concrete, 2013. G. Meschke, B. Pichler, J.G. Rots, Computational Modelling of Concrete Structures: Proceedings of the Conference on Computational Modelling of Concrete and Concrete Structures (EURO-C 2018), February 26 – March 1, 2018, CRC Press, Bad Hofgastein, Austria, 2018. Zdeneˇk P. Bazˇant, Saeed Rahimi-Aghdam, Diffusion-controlled and creepmitigated ASR damage via microplane model. I: Mass concrete, J. Eng. Mech. 143 (2) (2017) 04016108, https://doi.org/10.1061/(ASCE)EM.19437889.0001186. Saeed Rahimi-Aghdam, Zdeneˇk P. Bazˇant, Ferhun C. Caner, Diffusioncontrolled and creep-mitigated ASR damage via microplane model. II: Material degradation, drying, and verification, J. Eng. Mech. 143 (2) (2017) 04016109, https://doi.org/10.1061/(ASCE)EM.1943-7889.0001185. J. Kleib, G. Aouad, G. Louis, M. Zakhour, M. Boulos, A. Rousselet, D. Bulteel, The use of calcium sulfoaluminate cement to mitigate the alkali silica reaction in mortars, Constr. Build. Mater. 184 (2018) 295–303, https://doi.org/10.1016/ j.conbuildmat.2018.06.215. D. Bulteel, N. Rafaï, P. Degrugilliers, E. Garcia-Diaz, Petrography study on altered flint aggregate by alkali–silica reaction, Mater. Charact. 53 (2004) 141–154, https://doi.org/10.1016/j.matchar.2004.08.004. C. Larive, A. Laplaud, O. Coussy, The role of water in alkali-silica reaction, alkali-aggregate reaction in concrete, Can. Underwrit. (2000) 61–69. S. Poyet, A. Sellier, B. Capra, G. Thè-venin-Foray, J.-M. Torrenti, H. TournierCognon, E. Bourdarot, Influence of water on alkali-silica reaction: experimental study and numerical simulations, J. Mater. Civ. Eng. 18 (2006) 588–596, https://doi.org/10.1061/(ASCE)0899-1561(2006) 18:4(588). E. Grimal, A. Sellier, Y.L. Pape, E. Bourdarot, Creep, shrinkage, and anisotropic damage in alkali-aggregate reaction swelling mechanism-Part II: identification of model parameters and application, Mater. J. 105 (2008) 236–242, https://doi.org/10.14359/19819. H. Olafsson, The effect of relative humidity and temperature on alkali expansion of mortar bars, in: Proc. 7th Int. Conf. Alkali-Aggreg. React. Concr., Ottawa, Canada, 1987: pp. 461–465. F. Tomosawa, K. Tamura, M. Abe, Influence of water content of concrete on alkali-aggregate reaction, in: Proc. 8th Int. Conf. Alkali-Aggreg. React. Concr., Kyoto, Japan, 1989: pp. 881–885. C. Larive, Apports combinés de l’expérimentation et de la modélisation à la compréhension del’alcali-réaction et de ses effets mécaniques., Ecole Nationale des Pontset Chaussée, 1997. chrome-extension:// oemmndcbldboiebfnladdacbdfmadadm/https://pastel.archives-ouvertes. fr/file/index/docid/520676/filename/1997TH_LARIVE_C_NS20683.pdf. F.N. Okoye, J. Durgaprasad, N.B. Singh, Effect of silica fume on the mechanical properties of fly ash based-geopolymer concrete, Ceram. Int. 42 (2016) 3000– 3006, https://doi.org/10.1016/j.ceramint.2015.10.084.

R.B. Figueira et al. / Construction and Building Materials 222 (2019) 903–931 [156] M. Wyrzykowski, P. Lura, Controlling the coefficient of thermal expansion of cementitious materials – a new application for superabsorbent polymers, Cem. Concr. Compos. 35 (2013) 49–58, https://doi.org/10.1016/j. cemconcomp.2012.08.010. [157] F.C.R. Almeida, A.J. Klemm, Efficiency of internal curing by superabsorbent polymers (SAP) in PC-GGBS mortars, Cem. Concr. Compos. 88 (2018) 41–51, https://doi.org/10.1016/j.cemconcomp.2018.01.002. [158] A. Pourjavadi, S.M. Fakoorpoor, A. Khaloo, P. Hosseini, Improving the performance of cement-based composites containing superabsorbent polymers by utilization of nano-SiO2 particles, Mater. Des. 42 (2012) 94– 101, https://doi.org/10.1016/j.matdes.2012.05.030. [159] Y. Wei, W. Guo, X. Zheng, Integrated shrinkage, relative humidity, strength development, and cracking potential of internally cured concrete exposed to different drying conditions, Dry. Technol. 34 (2016) 741–752, https://doi.org/ 10.1080/07373937.2015.1072549. [160] R.J. Thomas, E. Ariyachandra, D. Lezama, S. Peethamparan, Comparison of chloride permeability methods for Alkali-Activated concrete, Constr. Build. Mater. 165 (2018) 104–111, https://doi.org/10.1016/ j.conbuildmat.2018.01.016. [161] Cui Xinzhuang, Zhang Jiong, Huang Dan, Liu Zequn, Hou Fei, Cui Sheqiang, Zhang Lei, Wang Zhongxiao, Experimental study on the relationship between permeability and strength of pervious concrete, J. Mater. Civ. Eng. 29 (2017) 04017217, https://doi.org/10.1061/(ASCE)MT.1943-5533.0002058. [162] V.G. Haach, G. Vasconcelos, P.B. Lourenço, Influence of aggregates grading and water/cement ratio in workability and hardened properties of mortars, Constr. Build. Mater. 25 (2011) 2980–2987, https://doi.org/10.1016/ j.conbuildmat.2010.11.011. [163] J.E. Gillott, Alkali-aggregate reactions in concrete, Eng. Geol. 9 (1975) 303– 326, https://doi.org/10.1016/0013-7952(75)90013-7. [164] J. Lindgård, M.D.A. Thomas, E.J. Sellevold, B. Pedersen, Ö. Andiç-Çakır, H. Justnes, T.F. Rønning, Alkali–silica reaction (ASR)—performance testing: Influence of specimen pre-treatment, exposure conditions and prism size on alkali leaching and prism expansion, Cem. Concr. Res. 53 (2013) 68–90, https://doi.org/10.1016/j.cemconres.2013.05.017. [165] R.F. Blanks, H.S. Meissner, The expansion test as a measure of alkali-aggregate reaction, J. Proc. 42 (1946) 517–540, https://doi.org/10.14359/8720. [166] M. Thomas, B. Fournier, K. Folliard, J. Ideker, M. Shehata, Test methods for evaluating preventive measures for controlling expansion due to alkali–silica reaction in concrete, Cem. Concr. Res. 36 (2006) 1842–1856, https://doi.org/ 10.1016/j.cemconres.2006.01.014. [167] J. Lindgård, E.J. Sellevold, M.D.A. Thomas, B. Pedersen, H. Justnes, T.F. Rønning, Alkali–silica reaction (ASR)—performance testing: influence of specimen pretreatment, exposure conditions and prism size on concrete porosity, moisture state and transport properties, Cem. Concr. Res. 53 (2013) 145– 167, https://doi.org/10.1016/j.cemconres.2013.05.020. [168] U. Costa, T. Mangialardi, A.E. Paolini, Minimizing alkali leaching in the concrete prism expansion test at 38°C, Constr. Build. Mater. 146 (2017) 547– 554, https://doi.org/10.1016/j.conbuildmat.2017.04.116. [169] R.-P. Martin, O. Omikrine Metalssi, F. Toutlemonde, Importance of considering the coupling between transfer properties, alkali leaching and expansion in the modelling of concrete beams affected by internal swelling reactions, Constr. Build. Mater. 49 (2013) 23–30, https://doi.org/10.1016/ j.conbuildmat.2013.08.008. [170] W. Grymin, M. Koniorczyk, F. Pesavento, D. Gawin, Numerical model of the alkali-silica reaction development with external source of alkalis, Proc. Eng. 193 (2017) 509–516, https://doi.org/10.1016/j.proeng.2017.06.244. [171] M. Berra, T. Mangialardi, A.E. Paolini, Alkali release from aggregates in longservice concrete structures: laboratory test evaluation and ASR prediction, Materials 11 (2018) 1393, https://doi.org/10.3390/ma11081393. [172] T. Ichikawa, H. Koizumi, Possibility of radiation-induced degradation of concrete by alkali-silica reaction of aggregates, J. Nucl. Sci. Technol. 39 (2002) 880–884, https://doi.org/10.1080/18811248.2002.9715272. [173] T. Ichikawa, T. Kimura, Effect of nuclear radiation on alkali-silica reaction of concrete, J. Nucl. Sci. Technol. 44 (2007) 1281–1284, https://doi.org/10.1080/ 18811248.2007.9711372. [174] T.M. Rosseel, I. Maruyama, Y.L. Pape, O. Kontani, A.B. Giorla, I. Remec, J.J. Wall, M. Sircar, C. Andrade, M. Ordonez, Review of the current state of knowledge on the effects of radiation on concrete, J. Adv. Concr. Technol. 14 (2016) 368– 383, https://doi.org/10.3151/jact.14.368. [175] K.G. Field, I. Remec, Y.L. Pape, Radiation effects in concrete for nuclear power plants – Part I: Quantification of radiation exposure and radiation effects, Nucl. Eng. Des. 282 (2015) 126–143, https://doi.org/10.1016/j. nucengdes.2014.10.003. [176] Alkali-Aggregate Reactivity (AAR) Facts Book, (n.d.) 224. [177] 4 Structural effects and implications and repair, in: Alkali-Silica React. Concr., Thomas Telford Publishing, 1988: pp. 73–87. https://doi.org/10.1680/aric. 13179.0004. [178] W.-F. Chen, L. Duan, Bridge Engineering Handbook Second Edition: Construction and Maintenance, CRC Press, 2014. [179] S. Urhan, Alkali silica and pozzolanic reactions in concrete. Part 2: Observations on expanded perlite aggregate concretes, Cem. Concr. Res. 17 (1987) 465–477. [180] T.C. Rilem, 191-ARP, RILEM TC 191-ARP ‘‘Alkali-reactivity and prevention – Assessment, specification and diagnosis of alkali-reactivity” AAR- 5: Rapid preliminary screening test for carbonate aggregates, Mater. Struct. 38 (2005) 787–792, https://doi.org/10.1617/14382.

927

[181] G. Igarashi, K. Yamada, Y. Xu, H. Wong, S. Hirono, S. Ogawa, IMAGE analysis of alkali-aggregate gel in concrete prism test with alkali-wrapping, (n.d.) 11. [182] C.E. Efstathiou, A.B. Mitsana-Papazoglou, T.P. Hadjiioannou, A sensitive spot test for the detection of sodium based on the fluorescence of sodium-zincuranyl acetate, Microchem. J. 28 (1983) 341–343, https://doi.org/10.1016/ 0026-265X(83)90069-3. [183] K. Natesaiyer, Some applications of the gel fluorescence test for alkaliaggregate reaction in concrete, Cem. Concr. Compos. 15 (1993) 3–6, https:// doi.org/10.1016/0958-9465(93)90033-6. [184] G.D. Guthrie, J.W. Carey, A simple environmentally friendly, and chemically specific method for the identification and evaluation of the alkali-silica reaction, Cem. Concr. Res. 27 (1997) 1407–1417, https://doi.org/10.1016/ S0008-8846(97)00123-3. [185] K.C. Natesaiyer, K.C. Hover, Further study of an in-situ identification method for alkali-silica reaction products in concrete, Cem. Concr. Res. 19 (1989) 770–778, https://doi.org/10.1016/0008-8846(89)90047-1. [186] On behalf of the membership of RILEM TC 219-ACS, P.J. Nixon, I. Sims, RILEM Recommended Test Method: AAR-0—Outline Guide to the Use of RILEM Methods in the Assessment of the Alkali-Reactivity Potential of Aggregates, in: P.J. Nixon, I. Sims (Eds.), RILEM Recomm. Prev. Damage Alkali-Aggreg. React. New Concr. Struct., Springer Netherlands, Dordrecht, 2016: pp. 5–34. https://doi.org/10.1007/978-94-017-7252-5_2. [187] T. Knudsen, N. Thaulow, Quantitative microanalyses of alkali-silica gel in concrete, Cem. Concr. Res. 5 (1975) 443–454, https://doi.org/10.1016/00088846(75)90019-8. [188] N. Thaulow, U.H. Jakobsen, B. Clark, Composition of alkali silica gel and ettringite in concrete railroad ties: SEM-EDX and X-ray diffraction analyses, Cem. Concr. Res. 26 (1996) 309–318, https://doi.org/10.1016/0008-8846(95) 00219-7. [189] M. Brouxel, The alkali-aggregate reaction rim: Na2O, SiO2, K2O and CaO chemical distribution, Cem. Concr. Res. 23 (1993) 309–320, https://doi.org/ 10.1016/0008-8846(93)90096-R. [190] M. Miler, B. Mirticˇ, Accuracy and precision of EDS analysis for identification of metal-bearing minerals in polished and rough particle samples, Geologija 56 (2013) 5–17, https://doi.org/10.5474/geologija.2013.001. [191] N. Castro, B.E. Sorensen, M.A.T.M. Broekmans, Quantitative assessment of alkali-reactive aggregate mineral content through XRD using polished sections as a supplementary tool to RILEM AAR-1 (petrographic method), Cem. Concr. Res. 42 (2012) 1428–1437, https://doi.org/10.1016/j. cemconres.2012.08.004. [192] V. Bulatovic´, M. Melešev, M. Radeka, V. Radonjanin, I. Lukic´, Evaluation of sulfate resistance of concrete with recycled and natural aggregates, Constr. Build. Mater. 152 (2017) 614–631, https://doi.org/10.1016/ j.conbuildmat.2017.06.161. [193] A.M. Bazán, J.C. Gálvez, E. Reyes, D. Galé-Lamuela, Study of the rust penetration and circumferential stresses in reinforced concrete at early stages of an accelerated corrosion test by means of combined SEM, EDS and strain gauges, Constr. Build. Mater. 184 (2018) 655–667, https://doi.org/ 10.1016/j.conbuildmat.2018.06.195. [194] H. Morillas, I. Marcaida, C. García-Florentino, M. Maguregui, G. Arana, J.M. Madariaga, Micro-Raman and SEM-EDS analyses to evaluate the nature of salt clusters present in secondary marine aerosol, Sci. Total Environ. 615 (2018) 691–697, https://doi.org/10.1016/j.scitotenv.2017.09.299. [195] M.D.A. Thomas, Folliard, K. J., Fournier, B., Rivard, P., Drimalas, T, Methods for Evaluating and Treating ASR-Affected Structures: Results of Field Application and Demonstration Projects Volume I: Summary of Findings and Recommendations Final Report, 2013. [196] R.-P. Martin, L. Sanchez, B. Fournier, F. Toutlemonde, Evaluation of different techniques for the diagnosis & prognosis of Internal Swelling Reaction (ISR) mechanisms in concrete, Constr. Build. Mater. 156 (2017) 956–964, https:// doi.org/10.1016/j.conbuildmat.2017.09.047. [197] D.O. Woolf, Reaction of aggregate with low-alkali cement, Public Roads (1952). [198] D. Stark, Alkali-silica reactivity: some reconsiderations, Cem. Concr. Aggreg. 2 (1980) 92–94, https://doi.org/10.1520/CCA10189J. [199] S. Al-Otaibi, Durability of concrete incorporating GGBS activated by waterglass, Constr. Build. Mater. 22 (2008) 2059–2067, https://doi.org/10.1016/ j.conbuildmat.2007.07.023. [200] C. Yang, X. Pu, F. Wu, Research on alkali aggregate reaction expansion of alkali-slag mortar (in Chinese), J. Chin. Ceram. Soc. 27 (1999) 651–657. [201] Z. Shi, C. Shi, S. Wan, Z. Ou, Effect of alkali dosage on alkali-silica reaction in sodium hydroxide activated slag mortars, Constr. Build. Mater. 143 (2017) 16–23, https://doi.org/10.1016/j.conbuildmat.2017.03.125. [202] M. Chi, Effects of dosage of alkali-activated solution and curing conditions on the properties and durability of alkali-activated slag concrete, Constr. Build. Mater. 35 (2012) 240–245, https://doi.org/10.1016/ j.conbuildmat.2012.04.005. [203] A.M. Rashad, A comprehensive overview about the influence of different additives on the properties of alkali-activated slag – a guide for Civil Engineer, Constr. Build. Mater. 47 (2013) 29–55, https://doi.org/10.1016/ j.conbuildmat.2013.04.011. [204] C01 Committee, ASTM C150/C150M-18, Standard Specification for Portland Cement, (n.d.). https://doi.org/10.1520/C0150_C0150M-18. [205] A. Leemann, B. Lothenbach, THE NA2O-EQUIVALENT OF CEMENT: A UNIVERSAL PARAMETER TO ASSESS THE POTENTIAL ALKALI-AGGREGATE REACTIVITY OF CONCRETE?, (n.d.) 12.

928

R.B. Figueira et al. / Construction and Building Materials 222 (2019) 903–931

[206] C09 Committee, Guide for Petrographic Examination of Aggregates for Concrete, ASTM International, n.d. doi: 10.1520/C0295_C0295M-18A. [207] Water-Cement Ratio – Petrographic Methods of Examining Hardened Concrete: A Petrographic Manual, NOVEMBER 1997 – FHWA-RD-97-146, (n.d.). https://www.fhwa.dot.gov/publications/research/infrastructure/ pavements/pccp/97146/chaptr09.cfm (accessed November 12, 2018). [208] X.X. Gao, M. Cyr, S. Multon, A. Sellier, A comparison of methods for chemical assessment of reactive silica in concrete aggregates by selective dissolution, Cem. Concr. Compos. 37 (2013) 82–94, https://doi.org/10.1016/j. cemconcomp.2012.12.002. [209] T.C. Rilem, A-TC 106–2-Detection of potential alkali-reactivity of aggregatesthe ultra-accelerated mortar-bar test B-TC 106–3-detection of potential alkali-reactivity of aggregates-method for aggregate combinations using concrete prisms, Mater. Struct. 33 (2000) 283–289, https://doi.org/10.1007/ BF02479697. [210] C09 Committee, ASTM C295/C295M-18a, Standard Guide for Petrographic Examination of Aggregates for Concrete, 2018. doi: 10.1520/C0295_C0295M18A. [211] P.I. Sims, P. Nixon, RILEM Recommended Test Method AAR-I: Detection of potential alkali-reactivity of aggregates-Petrographic method, (n.d.) 17. [212] C09 Committee, ASTM C289-07, Standard Test Method for Potential AlkaliSilica Reactivity of Aggregates (Chemical Method) (Withdrawn 2016), (n.d.). doi: 10.1520/C0289-07. [213] B. Fournier, M.A. Bérubé, Recent applications of a modified gel pat test to determine the potential alkali-silica reactivity of carbonate aggregates, Cem. Concr. Compos. 15 (1993) 49–73, https://doi.org/10.1016/0958-9465(93) 90038-B. [214] C09 Committee, ASTM C227-10, Standard Test Method for Potential Alkali Reactivity of Cement-Aggregate Combinations (Mortar-Bar Method) (Withdrawn 2018), (n.d.). doi: 10.1520/C0227-10. [215] C09 Committee, ASTM C1260 – 14, Standard Test Method for Potential Alkali Reactivity of Aggregates (Mortar-Bar Method), (n.d.). https://doi.org/10.1520/ C1260-14. [216] P.J. Nixon, I. Sims, RILEM Recommended Test Method: AAR-2—Detection of Potential Alkali-Reactivity—Accelerated Mortar-Bar Test Method for Aggregates, in: P.J. Nixon, I. Sims (Eds.), RILEM Recomm. Prev. Damage Alkali-Aggreg. React. New Concr. Struct. State—Art Rep. RILEM Tech. Comm. 219-ACS, Springer Netherlands, Dordrecht, 2016, pp. 61–77. doi: 10.1007/ 978-94-017-7252-5_4. [217] B. Fournier, M.A. Bérubé, Application of the NBRI accelerated mortar bar test to siliceous carbonate aggregates produced in the St. Lawrence Lowlands (Quebec, Canada) part 2: Proposed limits, rates of expansion, and microstructure of reaction products, Cem. Concr. Res. 21 (6) (1991) 1069– 1082, https://doi.org/10.1016/0008-8846(91)90067-R. [218] P.E. Grattan-Bellew, A critical review of ultra-accelerated tests for alkali-silica reactivity, Cem. Concr. Compos. 19 (1997) 403–414, https://doi.org/10.1016/ S0958-9465(97)00049-8. [219] A. Shayan, H. Morris, A comparison of RTA T363 and ASTM C1260 accelerated mortar bar test methods for detecting reactive aggregates, Cem. Concr. Res. 31 (2001) 655–663, https://doi.org/10.1016/S0008-8846(00)00491-9. [220] E.R. Latifee, S. Akther, K. Hasnat, A Critical Review of the Test Methods for Evaluating the ASR Potential of Aggregates, 2015, 15. [221] M.J. Munir, S. Abbas, A.U. Qazi, M.L. Nehdi, S.M.S. Kazmi, Role of test method in detection of alkali–silica reactivity of concrete aggregates, Proc. Inst. Civ. Eng. – Constr. Mater. 171 (2018) 203–221, https://doi.org/10.1680/ jcoma.16.00058. [222] C09 Committee, ASTM C1293-18, Standard Test Method for Determination of Length Change of Concrete Due to Alkali-Silica Reaction, (n.d.). https://doi. org/10.1520/C1293-18. [223] A. Santos Silva, D. Soares, L. Matos, I. Fernandes, M.M. Salta, Alkaliaggregate reactions in concrete: methodologies applied in the evaluation of alkali reactivity of aggregates for concrete, Mater. Sci. Forum. (2012) 730409–732414, https://doi.org/10.4028/www.scientific.net/MSF.730732.409. [224] Cruz Carlos, Jr., Mauricio Mancio, Kome Shomglin, John Harvey, Paulo Monteiro, and Abdikarim Ali, Accelerated Laboratory Testing for AlkaliSilica Reaction Using ASTM 1293 and Comparison with ASTM 1260, Pavement Research Center Institute of Transportation Studies University of California, Berkeley University of California, Davis, 2004. [225] L.J. Malvar, L.R. Lenke, Alkali Silica Reaction Criteria for Accelerated Mortar Bar Tests Based on Field Performance Data, (n.d.) 17. [226] T. Kurihara, K. Katawaki, Effects of Moisture Control and Inhibition on Alkali Silica Reaction, in: Proc 8th ICAAR Kyoto Jpn., 1989, pp. 629–634. [227] J.F. Seignol, T.T. Ngo, F. Toutlemonde, Modeling of the coupling between moisture and alkali-silica reaction in concrete, in: 2006: pp. 639–646. https:// www.scopus.com/inward/record.uri?eid=2-s2.0-77950628513&partnerID= 40&md5=cc0b9f8017e6d965ce0bdf32bf82bd00. [228] W. Zhang, H. Min, X. Gu, Temperature response and moisture transport in damaged concrete under an atmospheric environment, Constr. Build. Mater. 123 (2016) 290–299, https://doi.org/10.1016/j.conbuildmat.2016.07.004. [229] I. Sims, A.B. Poole, Alkali-Aggregate Reaction in Concrete: A World Review, CRC Press, 2017. [230] M.-A. Bérubé, D. Chouinard, M. Pigeon, J. Frenette, L. Boisvert, M. Rivest, Effectiveness of sealers in counteracting alkali-silica reaction in plain and airentrained laboratory concretes exposed to wetting and drying, freezing and

[231]

[232]

[233]

[234] [235] [236]

[237]

[238]

[239]

[240] [241]

[242]

[243]

[244]

[245] [246]

[247]

[248]

[249]

[250]

[251]

[252]

[253]

[254]

[255]

[256]

thawing, and salt water, Can. J. Civ. Eng. 29 (2002) 289–300, https://doi.org/ 10.1139/l02-011. K. Tosun, B. Felekoglu, B. Baradan, Effectiveness of alkyl alkoxy silane treatment in mitigating alkali-silica reaction, Mater. J. 105 (2008) 20–27, https://doi.org/10.14359/19203. R.A.D. Jr, C.D. Murray, W.M. Hale, Mitigation of alkali-silica reaction and freezing and thawing through surface treatment, Mater. J. 114 (2017) 307– 314, https://doi.org/10.14359/51689493. R.A.D. Jr, E.R. Giannini, T. Drimalas, B. Fournier, W.M. Hale, Mitigating alkalisilica reaction and freezing and thawing in concrete pavement by silane treatment, Mater. J. 115 (2018) 685–694, https://doi.org/10.14359/ 51702345. W.J. McCoy, A.G. Caldwell, New approach to inhibiting alkali-aggregate expansion, J. Am. Concr. Inst. 22 (1951) 693–706. J.S. Lumley, ASR suppression by lithium compounds, Cem. Concr. Res. 27 (1997) 235–244, https://doi.org/10.1016/S0008-8846(97)00003-3. X. Mo, C. Yu, Z. Xu, Long-term effectiveness and mechanism of LiOH in inhibiting alkali–silica reaction, Cem. Concr. Res. 33 (2003) 115–119, https:// doi.org/10.1016/S0008-8846(02)00934-1. K. Ramyar, O. Çopurog˘lu, Ö. Andiç, A.L.A. Fraaij, Comparison of alkali–silica reaction products of fly-ash- or lithium-salt-bearing mortar under long-term accelerated curing, Cem. Concr. Res. 34 (2004) 1179–1183, https://doi.org/ 10.1016/j.cemconres.2003.12.007. C.L. Collins, J.H. Ideker, G.S. Willis, K.E. Kurtis, Examination of the effects of LiOH, LiCl, and LiNO3 on alkali–silica reaction, Cem. Concr. Res. 34 (2004) 1403–1415, https://doi.org/10.1016/j.cemconres.2004.01.011. X. Mo, T. Jin, G. Li, K. Wang, Z. Xu, M. Tang, Alkali-aggregate reaction suppressed by chemical admixture at 80 °C, Constr. Build. Mater. 19 (2005) 473–479, https://doi.org/10.1016/j.conbuildmat.2004.07.012. D. Stark, P. Okamoto, S. Diamond, Eliminating or minimizing alkali-silica reaktivity, Washington, DC, 1993. X. Mo, Laboratory study of LiOH in inhibiting alkali–silica reaction at 20 °C: a contribution, Cem. Concr. Res. 35 (2005) 499–504, https://doi.org/10.1016/j. cemconres.2004.06.003. X. Feng, M.D.A. Thomas, T.W. Bremner, B.J. Balcom, K.J. Folliard, Studies on lithium salts to mitigate ASR-induced expansion in new concrete: a critical review, Cem. Concr. Res. 35 (2005) 1789–1796, https://doi.org/10.1016/j. cemconres.2004.10.013. I. Mohd, S. Yasutaka, H. Hidenori, D. Yamamoto, An experimental study on mitigating alkali silica reaction by using lithium hydroxide monohydrate, AIP Conf. Proc. 1903 (2017), https://doi.org/10.1063/1.5011512 030005. A. Leemann, L. Lörtscher, L. Bernard, G. Le Saout, B. Lothenbach, R.M. Espinosa-Marzal, Mitigation of ASR by the use of LiNO3—characterization of the reaction products, Cem. Concr. Res. 59 (2014) 73–86, https://doi.org/ 10.1016/j.cemconres.2014.02.003. T. Kim, J. Olek, Influence of lithium ions on the chemistry of pore solutions in pastes and mortars with inert aggregates, (n.d.) 11. I.X.Y. Mo, Z.Z. Xu, K.R. Wu, M.S. Tang, Effectiveness of LiOH in inhibiting alkali-aggregate reaction and its mechanism, Mater. Struct. 38 (2005) 57–61, https://doi.org/10.1007/BF02480575. C. Tremblay, M.A. Bérubé, B. Fournier, M.D. Thomas, K.J. Folliard, Experimental investigation of the mechanisms by which LiNO3 is effective against ASR, Cem. Concr. Res. 40 (2010) 583–597, https://doi.org/10.1016/j. cemconres.2009.09.022. Y. Qi, Z. Wen, Effects of lithium hydroxide on alkali silica reaction gels created with opal, Constr. Build. Mater. 21 (2007) 1656–1660, https://doi.org/ 10.1016/j.conbuildmat.2006.05.022. X. Mo, Y. Zhang, J. Yao, G. Li, Y. Feng, Influence of various parameters on Li2CO3 against alkali–silica reaction, Constr. Build. Mater. 22 (2008) 1668– 1674, https://doi.org/10.1016/j.conbuildmat.2007.06.005. D. Bulteel, E. Garcia-Diaz, P. Dégrugilliers, Influence of lithium hydroxide on alkali–silica reaction, Cem. Concr. Res. 40 (2010) 526–530, https://doi.org/ 10.1016/j.cemconres.2009.08.019. X. Feng, M.D.A. Thomas, T.W. Bremner, K.J. Folliard, B. Fournier, Summary of research on the effect of LiNO3 on alkali–silica reaction in new concrete, Cem. Concr. Res. 40 (2010) 636–642, https://doi.org/10.1016/j. cemconres.2009.08.021. A.F. Bentivegna, T. Drimalas, J.H. Ideker, S. Hayman, Mitigation of ASR affected concrete in Boston, MA, USA: A case study, in: Concr. Repair Rehabil. Retrofit. III – Proc. 3rd Int. Conf. Concr. Repair Rehabil. Retrofit. ICCRRR 2012, 2012: pp. 944–949. https://utexas.influuent.utsystem.edu/en/publications/ mitigation-of-asr-affected-concrete-in-boston-ma-usa-a-case-study (accessed December 11, 2018). _ Demir, M. Arslan, The mechanical and microstructural properties of Li2SO4, I. LiNO3, Li2CO3 and LiBr added mortars exposed to alkali-silica reaction, Constr. Build. Mater. 42 (2013) 64–77, https://doi.org/10.1016/ j.conbuildmat.2012.12.059. T. Kim, J. Olek, The effects of lithium ions on chemical sequence of alkalisilica reaction, Cem. Concr. Res. 79 (2016) 159–168, https://doi.org/10.1016/j. cemconres.2015.09.013. M.S. Islam, N. Ghafoori, Experimental study and empirical modeling of lithium nitrate for alkali-silica reactivity, Constr. Build. Mater. 121 (2016) 717–726, https://doi.org/10.1016/j.conbuildmat.2016.06.026. L.M.S. Souza, R.B. Polder, O. Çopurog˘lu, The influence of the anolyte solution type and concentration on lithium migration in mortar specimens, Constr.

R.B. Figueira et al. / Construction and Building Materials 222 (2019) 903–931

[257]

[258]

[259]

[260]

[261]

[262]

[263]

[264]

[265]

[266]

[267]

[268]

[269]

[270]

[271]

[272]

[273]

[274]

[275]

[276]

[277]

[278]

[279]

[280]

[281]

Build. Mater. 186 (2018) 123–130, https://doi.org/10.1016/ j.conbuildmat.2018.07.073. _I. Demir, Ö. Sevim, I. _ Kalkan, Microstructural properties of lithium-added cement mortars subjected to alkali–silica reactions, Sa¯dhana¯ 43 (2018) 112, https://doi.org/10.1007/s12046-018-0901-3. P.J. Nixon, I. Sims, RILEM Recommendations for the Prevention of Damage by Alkali-Aggregate Reactions in New Concrete Structures: State-of-the-Art Report of the RILEM Technical Committee 219-ACS, Springer, 2015. Z. Shi, C. Shi, R. Zhao, S. Wan, Comparison of alkali–silica reactions in alkaliactivated slag and Portland cement mortars, Mater. Struct. 48 (2015) 743– 751, https://doi.org/10.1617/s11527-015-0535-4. D.E. Angulo-Ramírez, R. Mejía de Gutiérrez, M. Medeiros, Alkali-activated Portland blast furnace slag cement mortars: performance to alkali-aggregate reaction, Constr. Build. Mater. 179 (2018) 49–56, https://doi.org/10.1016/ j.conbuildmat.2018.05.183. D. Wang, Q. Wang, S. Zhuang, J. Yang, Evaluation of alkali-activated blast furnace ferronickel slag as a cementitious material: reaction mechanism, engineering properties and leaching behaviors, Constr. Build. Mater. 188 (2018) 860–873, https://doi.org/10.1016/j.conbuildmat.2018.08.182. M. Thomas, A. Dunster, P. Nixon, B. Blackwell, Effect of fly ash on the expansion of concrete due to alkali-silica reaction – exposure site studies, Cem. Concr. Compos. 33 (2011) 359–367, https://doi.org/10.1016/j. cemconcomp.2010.11.006. A. Gholizadeh Vayghan, J.R. Wright, F. Rajabipour, An extended chemical index model to predict the fly ash dosage necessary for mitigating alkali– silica reaction in concrete, Cem. Concr. Res. 82 (2016) 1–10, https://doi.org/ 10.1016/j.cemconres.2015.12.014. M.L. Schafer, K.A. Clavier, T.G. Townsend, C.C. Ferraro, J.M. Paris, B.E. Watts, Use of coal fly ash or glass pozzolan addition as a mitigation tool for alkalisilica reactivity in cement mortars amended with recycled municipal solid waste incinerator bottom ash, Waste Biomass Valorization (2018), https:// doi.org/10.1007/s12649-018-0296-8. M.-A. Bérubé, J. Duchesne, Does silica fume merely postpone expansion due to alkali-aggregate reactivity?, Constr Build. Mater. 7 (1993) 137–143, https://doi.org/10.1016/0950-0618(93)90050-M. M. Rostami, K. Behfarnia, The effect of silica fume on durability of alkali activated slag concrete, Constr. Build. Mater. 134 (2017) 262–268, https:// doi.org/10.1016/j.conbuildmat.2016.12.072. A.M. Rashad, Alkali-activated metakaolin: a short guide for civil engineer – an overview, Constr. Build. Mater. 41 (2013) 751–765, https://doi.org/10.1016/ j.conbuildmat.2012.12.030. B. Lothenbach, K. Scrivener, R.D. Hooton, Supplementary cementitious materials, Cem. Concr. Res. 41 (2011) 1244–1256, https://doi.org/10.1016/j. cemconres.2010.12.001. M.C.G. Juenger, R. Siddique, Recent advances in understanding the role of supplementary cementitious materials in concrete, Cem. Concr. Res. 78 (2015) 71–80, https://doi.org/10.1016/j.cemconres.2015.03.018. S.C. Pal, A. Mukherjee, S.R. Pathak, Investigation of hydraulic activity of ground granulated blast furnace slag in concrete, Cem. Concr. Res. 33 (2003) 1481–1486, https://doi.org/10.1016/S0008-8846(03)00062-0. _ Durmusß, Utilization and efficiency of ground E. Özbay, M. Erdemir, H.I. granulated blast furnace slag on concrete properties – a review, Constr. Build. Mater. 105 (2016) 423–434, https://doi.org/10.1016/j.conbuildmat. 2015.12.153. S. Chatterji, N.F. Clausson-Kaas, Prevention of alkali-silica expansion by using slag-Portland cement, Cem. Concr. Res. 14 (1984) 816–818, https://doi.org/ 10.1016/0008-8846(84)90006-1. D. Hester, C. McNally, M. Richardson, A study of the influence of slag alkali level on the alkali–silica reactivity of slag concrete, Constr. Build. Mater. 19 (2005) 661–665, https://doi.org/10.1016/j.conbuildmat.2005.02.016. N. Arano, M. Kawamura, Comparative consideration on the mechanisms of ASR suppression due to different mineral admixtures, 11th ICAAR, 2000, pp. 553–562. P.M. Gifford, J.E. Gillott, Alkali-silica reaction (ASR) and alkali-carbonate reaction (ACR) in activated blast furnace slag cement (ABFSC) concrete, Cem. Concr. Res. 26 (1996) 21–26, https://doi.org/10.1016/0008-8846(95)001824. I. Canham, C.L. Page, P.J. Nixon, Aspects of the pore solution chemistry of blended cements related to the control of alkali silica reaction, Cem. Concr. Res. 17 (1987) 839–844, https://doi.org/10.1016/0008-8846(87)90046-9. A. Fernández-Jiménez, F. Puertas, The alkali–silica reaction in alkali-activated granulated slag mortars with reactive aggregate, Cem. Concr. Res. 32 (2002) 1019–1024, https://doi.org/10.1016/S0008-8846(01)00745-1. L.J. Malvar, G.D. Cline, D.F. Burke, R. Rollings, T.W. Sherman, J.L. Greene, Alkali-silica reaction mitigation: state of the art and recommendations, Mater. J. 99 (2002) 480–489, https://doi.org/10.14359/12327. C. Zhang, A. Wang, Effect of mineral admixtures on alkali-silica reaction, J. Wuhan Univ. Technol.-Mater Sci Ed. 23 (2008) 16, https://doi.org/10.1007/ s11595-006-1016-y. A. Bouikni, R.N. Swamy, A. Bali, Durability properties of concrete containing 50% and 65% slag, Constr. Build. Mater. 23 (2009) 2836–2845, https://doi.org/ 10.1016/j.conbuildmat.2009.02.040. _ C. Karakurt, I.B. Topçu, Effect of blended cements produced with natural zeolite and industrial by-products on alkali-silica reaction and sulfate resistance of concrete, Constr. Build. Mater. 25 (2011) 1789–1795, https:// doi.org/10.1016/j.conbuildmat.2010.11.087.

929

[282] A. Beglarigale, H. Yazici, Mitigation of detrimental effects of alkali-silica reaction in cement-based composites by combination of steel microfibers and ground-granulated blast-furnace slag, J. Mater. Civ. Eng. 26 (2014) 04014091, https://doi.org/10.1061/(ASCE)MT.1943-5533.0001005. [283] A. Bicer, Effect of fly ash particle size on thermal and mechanical properties of fly ash-cement composites, Therm. Sci. Eng. Prog. 8 (2018) 78–82, https://doi. org/10.1016/j.tsep.2018.07.014. [284] C09 Committee, Specification for Coal Fly Ash and Raw or Calcined Natural Pozzolan for Use in Concrete, ASTM International, n.d. https://doi.org/10. 1520/C0618-19. [285] G. Xu, X. Shi, Characteristics and applications of fly ash as a sustainable construction material: a state-of-the-art review, Resour. Conserv. Recycl. 136 (2018) 95–109, https://doi.org/10.1016/j.resconrec.2018.04.010. [286] K. Wesche, Fly Ash in Concrete: Properties and performance, CRC Press, 2004. [287] E.G. Moffatt, M.D.A. Thomas, A. Fahim, Performance of high-volume fly ash concrete in marine environment, Cem. Concr. Res. 102 (2017) 127–135, https://doi.org/10.1016/j.cemconres.2017.09.008. [288] Q. Nie, C. Zhou, H. Li, X. Shu, H. Gong, B. Huang, Numerical simulation of fly ash concrete under sulfate attack, Constr. Build. Mater. 84 (2015) 261–268, https://doi.org/10.1016/j.conbuildmat.2015.02.088. [289] M.D.A. Thomas, Review of the effect of fly ash and slag on alkali-aggregate reaction in concrete, Building Res. Establishment (1996). [290] V.M. Malhotra, A.A. Ramezanianpour, Fly Ash in Concrete, in: Second Edition, Natural Resources Canada, Ottawa, Ontario, 1994, CANMET – Canadian Centre for Mineral and Energy Technology, 1994. [291] M.H. Shehata, M.D.A. Thomas, The effect of fly ash composition on the expansion of concrete due to alkali–silica reaction, Cem. Concr. Res. 30 (2000) 1063–1072, https://doi.org/10.1016/S0008-8846(00)00283-0. [292] S.-Y. Hong, F.P. Glasser, Alkali binding in cement pastes: Part I. The C-S-H phase, Cem. Concr. Res. 29 (1999) 1893–1903, https://doi.org/10.1016/ S0008-8846(99)00187-8. [293] T. Chappex, K. Scrivener, Alkali fixation of C–S–H in blended cement pastes and its relation to alkali silica reaction, Cem. Concr. Res. 42 (2012) 1049– 1054, https://doi.org/10.1016/j.cemconres.2012.03.010. [294] A. Brykov, A. Anisimova, Efficacy of aluminum hydroxides as inhibitors of alkali-silica reactions, Mater. Sci. Appl. 04 (2013) 1, https://doi.org/10.4236/ msa.2013.412A001. [295] T. Chappex, K.L. Scrivener, The influence of aluminium on the dissolution of amorphous silica and its relation to alkali silica reaction, Cem. Concr. Res. 42 (2012) 1645–1649, https://doi.org/10.1016/j. cemconres.2012.09.009. [296] T. Szeles, J. Wright, F. Rajabipour, S. Stoffels, Mitigation of alkali-silica reaction by hydrated alumina, Transp. Res. Rec. (2017), https://doi.org/ 10.3141/2629-04. [297] A. Gharzouni, L. Ouamara, I. Sobrados, S. Rossignol, Alkali-activated materials from different aluminosilicate sources: effect of aluminum and calcium availability, J. Non-Cryst. Solids. 484 (2018) 14–25, https://doi.org/10.1016/j. jnoncrysol.2018.01.014. [298] A. Shayan, R. Diggins, I. Ivanusec, Effectiveness of fly ash in preventing deleterious expansion due to alkali-aggregate reaction in normal and steamcured concrete, Cem. Concr. Res. 26 (1996) 153–164, https://doi.org/10.1016/ 0008-8846(95)00191-3. [299] A.A. Ramezanianpour, Cement Replacement Materials: Properties, Durability, Sustainability, Springer Science & Business Media, 2013. [300] M.H. Shehata, M.D.A. Thomas, Use of ternary blends containing silica fume and fly ash to suppress expansion due to alkali–silica reaction in concrete, Cem. Concr. Res. 32 (2002) 341–349, https://doi.org/10.1016/S0008-8846 (01)00680-9. [301] F. Xiao-xin, F. Nai-qian, H. Dong, Effect of the composite of natural zeolite and fly ash on alkali-silica reaction, J. Wuhan Univ. Technol.-Mater Sci Ed. 18 (2003) 93, https://doi.org/10.1007/BF02838402. [302] N. Quanlin, F. Naiqian, Effect of modified zeolite on the expansion of alkaline silica reaction, Cem. Concr. Res. 35 (2005) 1784–1788, https://doi.org/ 10.1016/j.cemconres.2004.10.030. [303] W. Fengyan, L. Xianghui, L. Yinong, X. Zhongzi, Effect of pozzolanic reaction products on alkali-silica reaction, J. Wuhan Univ. Technol.-Mater Sci Ed. 21 (2006) 168–171, https://doi.org/10.1007/BF02840910. [304] D. Lu, X. Zhou, Z. Xu, X. Lan, M. Tang, B. Fournier, Evaluation of laboratory test method for determining the potential alkali contribution from aggregate and the ASR safety of the Three-Gorges dam concrete, Cem. Concr. Res. 36 (2006) 1157–1165, https://doi.org/10.1016/j.cemconres.2006.01.004. [305] I. García-Lodeiro, A. Palomo, A. Fernández-Jiménez, Alkali–aggregate reaction in activated fly ash systems, Cem. Concr. Res. 37 (2007) 175–183, https://doi. org/10.1016/j.cemconres.2006.11.002. [306] N. Schwarz, H. Cam, N. Neithalath, Influence of a fine glass powder on the durability characteristics of concrete and its comparison to fly ash, Cem. Concr. Compos. 30 (2008) 486–496, https://doi.org/10.1016/j. cemconcomp.2008.02.001. [307] R.D. Moser, A.R. Jayapalan, V.Y. Garas, K.E. Kurtis, Assessment of binary and ternary blends of metakaolin and Class C fly ash for alkali-silica reaction mitigation in concrete, Cem. Concr. Res. 40 (2010) 1664–1672, https://doi. org/10.1016/j.cemconres.2010.08.006. [308] S. Aydın, Ç. Karatay, B. Baradan, The effect of grinding process on mechanical properties and alkali–silica reaction resistance of fly ash incorporated cement mortars, Powder Technol. 197 (2010) 68–72, https://doi.org/10.1016/j. powtec.2009.08.020.

930

R.B. Figueira et al. / Construction and Building Materials 222 (2019) 903–931

[309] T.C. Esteves, R. Rajamma, D. Soares, A.S. Silva, V.M. Ferreira, J.A. Labrincha, Use of biomass fly ash for mitigation of alkali-silica reaction of cement mortars, Constr. Build. Mater. 26 (2012) 687–693, https://doi.org/10.1016/ j.conbuildmat.2011.06.075. [310] H. Kizhakkumodom Venkatanarayanan, P.R. Rangaraju, Decoupling the effects of chemical composition and fineness of fly ash in mitigating alkalisilica reaction, Cem. Concr. Compos. 43 (2013) 54–68, https://doi.org/ 10.1016/j.cemconcomp.2013.06.009. [311] S.M.H. Shafaatian, A. Akhavan, H. Maraghechi, F. Rajabipour, How does fly ash mitigate alkali–silica reaction (ASR) in accelerated mortar bar test (ASTM C1567)?, Cem Concr. Compos. 37 (2013) 143–153, https://doi.org/10.1016/j. cemconcomp.2012.11.004. [312] L.J. Malvar, L.R. Lenke, Efficiency of fly ash in mitigating alkali-silica reaction based on chemical composition, ACI Mater. J. 103 (2006), https://doi.org/ 10.14359/18153. [313] K.A. Schumacher, J.H. Ideker, New considerations in predicting mitigation of alkali-silica reaction based on fly ash chemistry, J. Mater. Civ. Eng. 27 (2015) 04014144, https://doi.org/10.1061/(ASCE)MT.1943-5533.0001021. [314] T. Williamson, M.C.G. Juenger, The role of activating solution concentration on alkali–silica reaction in alkali-activated fly ash concrete, Cem. Concr. Res. 83 (2016) 124–130, https://doi.org/10.1016/j. cemconres.2016.02.008. [315] A.A. Ramezanianpour, Silica Fume, in: A.A. Ramezanianpour (Ed.), Cem. Replace. Mater. Prop. Durab. Sustain., Springer Berlin Heidelberg, Berlin, Heidelberg, 2014, pp. 193–223, https://doi.org/10.1007/978-3-642-367212_4. [316] G. Davies, R.E. Oberholster, Alkali-silica reaction products and their development, Cem. Concr. Res. 18 (1988) 621–635, https://doi.org/10.1016/ 0008-8846(88)90055-5. [317] A.M. Boddy, R.D. Hooton, M.D.A. Thomas, The effect of product form of silica fume on its ability to control alkali–silica reaction, Cem. Concr. Res. 30 (2000) 1139–1150, https://doi.org/10.1016/S0008-8846(00)00297-0. [318] D.D.L. Chung, Review: Improving cement-based materials by using silica fume, J. Mater. Sci. 37 (2002) 673–682, https://doi.org/10.1023/ A:1013889725971. [319] S.L. Marusin, L.B. Shotwell, Alkali-silica reaction in concrete caused by densified silica fume lumps: a case Study, Cem. Concr. Aggreg. 22 (2000) 90– 94, https://doi.org/10.1520/CCA10468J. [320] W. Aquino, D.A. Lange, J. Olek, The influence of metakaolin and silica fume on the chemistry of alkali–silica reaction products, Cem. Concr. Compos. 23 (2001) 485–493, https://doi.org/10.1016/S0958-9465(00)00096-2. [321] F. Bektas, L. Turanli, P.J.M. Monteiro, Use of perlite powder to suppress the alkali–silica reaction, Cem. Concr. Res. 35 (2005) 2014–2017, https://doi.org/ 10.1016/j.cemconres.2004.10.029. [322] A.J. Maas, J.H. Ideker, M.C.G. Juenger, Alkali silica reactivity of agglomerated silica fume, Cem. Concr. Res. 37 (2007) 166–174, https://doi.org/10.1016/j. cemconres.2006.10.011. [323] T. Nochaiya, W. Wongkeo, A. Chaipanich, Utilization of fly ash with silica fume and properties of Portland cement–fly ash–silica fume concrete, Fuel 89 (2010) 768–774, https://doi.org/10.1016/j.fuel.2009.10.003. [324] N.B. Singh, M. Kalra, M. Kumar, S. Rai, Hydration of ternary cementitious system: portland cement, fly ash and silica fume, J. Therm. Anal. Calorim. 119 (2015) 381–389, https://doi.org/10.1007/s10973-014-4182-8. [325] A. Brykov, M. Voronkov, M. Mokeev, Ultrafine silica additives behavior during alkali-silica reaction long-term expansion test, Mater. Sci. Appl. 2014 (2014), https://doi.org/10.4236/msa.2014.52010. [326] F.R. Andriolo, B.C. Sgaraboza, The use of pozzolan from calcined clays in preventing excessive expansion due to alkali-aggregate reaction in some brazilian dams, in: P.E. Grattan-Bellew (Ed.), Concrete Alkali-Aggregate Reactions, Noyes Publication, New Jersey, 1987. [327] T.R. Jones, G.V. Walters, J.A. Kostuch, Role of metakaolin in suppressing ASR in concrete containing reactive aggregates and exposed to saturated NaCl solution, Proc. 9th Int. Conf. Alkali-Aggreg. React. Concr. Lond. UK., 1992, pp. 485–496. [328] J.M. Khatib, S. Wild, Pore size distribution of metakaolin paste, Cem. Concr. Res. 26 (1996) 1545–1553, https://doi.org/10.1016/0008-8846(96)00147-0. [329] S. Wild, J.M. Khatib, Portlandite consumption in metakaolin cement pastes and mortars, Cem. Concr. Res. 27 (1997) 137–146, https://doi.org/10.1016/ S0008-8846(96)00187-1. [330] J.M. Khatib, Metakaolin concrete at a low water to binder ratio, Constr. Build. Mater. 22 (2008) 1691–1700, https://doi.org/10.1016/j.conbuildmat. 2007.06.003. [331] N.R. Rakhimova, R.Z. Rakhimov, Reaction products, structure and properties of alkali-activated metakaolin cements incorporated with supplementary materials – a review, J. Mater. Res. Technol. (2018), https://doi.org/10.1016/j. jmrt.2018.07.006. [332] A.H. Asbridge, G.A. Chadbourn, C.L. Page, Effects of metakaolin and the interfacial transition zone on the diffusion of chloride ions through cement mortars, Cem. Concr. Res. 31 (2001) 1567–1572, https://doi.org/10.1016/ S0008-8846(01)00598-1. [333] H.S. Al-alaily, A.A.A. Hassan, Time-dependence of chloride diffusion for concrete containing metakaolin, J. Build. Eng. 7 (2016) 159–169, https://doi. org/10.1016/j.jobe.2016.06.003. [334] J.M. Khatib, R.M. Clay, Absorption characteristics of metakaolin concrete, Cem. Concr. Res. 34 (2004) 19–29, https://doi.org/10.1016/S0008-8846(03) 00188-1.

[335] J.M. Khatib, S. Wild, Sulphate resistance of metakaolin mortar, Cem. Concr. Res. 28 (1998) 83–92, https://doi.org/10.1016/S0008-8846(97)00210-X. [336] R. Abbas, S.A. Abo-El-Enein, E.-S. Ezzat, Properties and durability of metakaolin blended cements: mortar and concrete, Mater. Constr. 60 (2010) 33–49, https://doi.org/10.3989/mc.2010.50609. [337] T. Ramlochan, P. Zacarias, M.D.A. Thomas, R.D. Hooton, The effect of pozzolans and slag on the expansion of mortars cured at elevated temperature: Part I: Expansive behaviour, Cem. Concr. Res. 33 (2003) 807– 814, https://doi.org/10.1016/S0008-8846(02)01066-9. [338] Sß. Yazıcı, H.S ß. Arel, D. Anuk, Influences of metakaolin on the durability and mechanical properties of mortars, Arab. J. Sci. Eng. 39 (2014) 8585–8592, https://doi.org/10.1007/s13369-014-1413-z. [339] R. Pouhet, M. Cyr, Alkali–silica reaction in metakaolin-based geopolymer mortar, Mater. Struct. 48 (2015) 571–583, https://doi.org/10.1617/s11527014-0445-x. [340] T. Ramos, A.M. Matos, B. Schmidt, J. Rio, J. Sousa-Coutinho, Granitic quarry sludge waste in mortar: effect on strength and durability, Constr. Build. Mater. 47 (2013) 1001–1009, https://doi.org/10.1016/ j.conbuildmat.2013.05.098. [341] J.A. Farny, S.H. Kosmatka, Diagnosis and Control of Alkali-Aggregate Reactions in Concrete, Portland Cement Association, 2007. [342] R. Johnson, M.H. Shehata, The efficacy of accelerated test methods to evaluate alkali silica reactivity of recycled concrete aggregates, Constr. Build. Mater. 112 (2016) 518–528, https://doi.org/10.1016/j.conbuildmat.2016.02.155. [343] P.J. Nixon, I. Sims, RILEM recommended test method: AAR-1.1—detection of potential alkali-reactivity—Part 1: petrographic examination method, in: P.J. Nixon, I. Sims (Eds.), RILEM Recomm. Prev. Damage Alkali-Aggreg. React. New Concr. Struct. State—Art Rep. RILEM Tech. Comm. 219-ACS, Springer Netherlands, Dordrecht, 2016, pp. 35–60. https://doi.org/10.1007/978-94017-7252-5_3. [344] P. Nixon, I. Sims, Testing aggregates for alkali-reactivity, Mater. Struct. 29 (1996) 323–334, https://doi.org/10.1007/BF02486340. [345] RILEM TC 219-ACS-P: Literature survey on performance testing SINTEF Bokhandel, (n.d.), https://www.sintefbok.no/book/index/1007/rilem_tc_219acs-p_literature_survey_on_performance_testing (accessed January 29, 2019). [346] NF P18-594 - Juillet 2015, (n.d.), https://www.boutique.afnor.org/norme/nfp18-594/granulats-methodes-d-essai-de-reactivite-aux-alcalis/article/ 807484/fa180908 (accessed January 29, 2019). [347] AS 1141.60.1:2014 | Methods for sampling and testing a... | SAI Global, (n.d.), https://infostore.saiglobal.com/en-gb/Standards/AS-1141-60-1-2014111591_SAIG_AS_AS_233418/ (accessed June 11, 2019). [348] AS 1141.60.2:2014/Amdt 1:2016 | Methods for sampling a... | SAI Global, (n. d.), https://infostore.saiglobal.com/en-gb/Standards/AS-1141-60-2-2014Amdt-1-2016-111590_SAIG_AS_AS_281822/ (accessed June 11, 2019). [349] V. Sirivivatnanon, J. Mohammadi, W. South, Reliability of new Australian test methods in predicting alkali silica reaction of field concrete, Constr. Build. Mater. 126 (2016) 868–874, https://doi.org/10.1016/ j.conbuildmat.2016.09.055. [350] J.H. Ideker, B.L. East, K.J. Folliard, M.D.A. Thomas, B. Fournier, The current state of the accelerated concrete prism test, Cem. Concr. Res. 40 (2010) 550– 555, https://doi.org/10.1016/j.cemconres.2009.08.030. [351] S. Beauchemin, B. Fournier, J. Duchesne, Evaluation of the concrete prisms test method for assessing the potential alkali-aggregate reactivity of recycled concrete aggregates, Cem. Concr. Res. 104 (2018) 25–36, https://doi.org/ 10.1016/j.cemconres.2017.10.008. [352] Recommendations of RILEM TC 106-AAR: Alkali aggregate reaction A. TC 1062 – Detection of potential alkali-reactivity of aggregates – The ultraaccelerated mortar-bar test B. TC 106-3 – Detection of potential alkalireactivity of aggregates – Method for aggregate combinations using concrete prisms, (n.d.). [353] C. Merz, A. Leemann, Assessment of the residual expansion potential of concrete from structures damaged by AAR, Cem. Concr. Res. 52 (2013) 182– 189, https://doi.org/10.1016/j.cemconres.2013.07.001. [354] S. Multon, F.-X. Barin, B. Godart, F. Toutlemonde, Estimation of the residual expansion of concrete affected by alkali silica reaction, J. Mater. Civ. Eng. 20 (2008) (accessed July 1, 2019) https://trid.trb.org/view/849812. [355] A. Carles-Gibergues, M. Cyr, Interpretation of expansion curves of concrete subjected to accelerated alkali–aggregate reaction (AAR) tests, Cem. Concr. Res. 32 (2002) 691–700, https://doi.org/10.1016/S0008-8846(01)00747-5. [356] ASTM C1567-13, Standard Test Method for Determining the Potential AlkaliSilica Reactivity of Combinations of Cementitious Materials and Aggregate (Accelerated Mortar-Bar Method), ASTM, International, West, Conshohocken, PA, 2013. www.astm.org. [357] ASTM C441 / C441M-17, Standard Test Method for Effectiveness of Pozzolans or Ground Blast-Furnace Slag in Preventing Excessive Expansion of Concrete Due to the Alkali-Silica Reaction, ASTM, International, West, Conshohocken, PA, 2017. www.astm.org. [358] P.J. Nixon, I. Sims, RILEM Recommended Test Method: AAR-4.1—Detection of Potential Alkali-Reactivity—60 °C Test Method for Aggregate Combinations Using Concrete Prisms, in: P.J. Nixon, I. Sims (Eds.), RILEM Recomm. Prev. Damage Alkali-Aggreg. React. New Concr. Struct. State—Art Rep. RILEM Tech. Comm. 219-ACS, Springer Netherlands, Dordrecht, 2016: pp. 99–116. doi:10. 1007/978-94-017-7252-5_6. [359] AASHTO T 303 - Standard Method of Test for Accelerated Detection of Potentially Deleterious Expansion of Mortar Bars Due to Alkali-Silica Reaction

R.B. Figueira et al. / Construction and Building Materials 222 (2019) 903–931

[360]

[361]

[362]

[363]

[364]

[365]

[366]

| Engineering360. https://standards.globalspec.com/std/10182267/AASHTO% 20T%20303 (accessed January 30, 2019). V. Ramos, I. Fernandes, A. Santos Silva, D. Soares, B. Fournier, S. Leal, F. Noronha, Assessment of the potential reactivity of granitic rocks — Petrography and expansion tests, Cem. Concr. Res. 86 (2016) 63–77, https://doi.org/10.1016/j.cemconres.2016.05.001. K.M. Donnell, R. Zoughi, K.E. Kurtis, Demonstration of microwave method for detection of alkali–silica reaction (ASR) gel in cement-based materials, Cem. Concr. Res. 44 (2013) 1–7, https://doi.org/10.1016/j.cemconres.2012.10.005. N. Smaoui, M.-A. Bérubé, B. Fournier, B. Bissonnette, B. Durand, Evaluation of the expansion attained to date by concrete affected by alkali–silica reaction. Part I: experimental study, Can. J. Civ. Eng. 31 (2004) 826–845, https://doi. org/10.1139/l04-051. L.F.M. Sanchez, B. Fournier, M. Jolin, J. Bastien, Evaluation of the stiffness damage test (SDT) as a tool for assessing damage in concrete due to ASR: test loading and output responses for concretes incorporating fine or coarse reactive aggregates, Cem. Concr. Res. 56 (2014) 213–229, https://doi.org/ 10.1016/j.cemconres.2013.11.003. M. Abdelrahman, M.K. ElBatanouny, P. Ziehl, J. Fasl, C.J. Larosche, J. Fraczek, Classification of alkali–silica reaction damage using acoustic emission: a proof-of-concept study, Constr. Build. Mater. 95 (2015) 406–413, https://doi. org/10.1016/j.conbuildmat.2015.07.093. M. Rashidi, M.C.L. Knapp, A. Hashemi, J.-Y. Kim, K.M. Donnell, R. Zoughi, L.J. Jacobs, K.E. Kurtis, Detecting alkali-silica reaction: a multi-physics approach, Cem. Concr. Compos. 73 (2016) 123–135, https://doi.org/10.1016/j. cemconcomp.2016.07.001. A. Teramoto, M. Watanabe, R. Murakami, T. Ohkubo, Visualization of internal crack growth due to alkali–silica reaction using digital image correlation,

[367]

[368]

[369]

[370]

[371]

[372]

[373]

931

Constr. Build. Mater. 190 (2018) 851–860, https://doi.org/10.1016/ j.conbuildmat.2018.09.168. T. Iskhakov, J.J. Timothy, G. Meschke, Expansion and deterioration of concrete due to ASR: micromechanical modeling and analysis, Cem. Concr. Res. 115 (2019) 507–518, https://doi.org/10.1016/j.cemconres.2018.08.001. R. Rezakhani, M. Alnaggar, G. Cusatis, Multiscale homogenization analysis of alkali-silica reaction (ASR) effect in concrete, Engineering (2019), https://doi. org/10.1016/j.eng.2019.02.007. G. Cusatis, D. Pelessone, A. Mencarelli, Lattice Discrete Particle Model (LDPM) for failure behavior of concrete. I: theory, Cem. Concr. Compos. 33 (2011) 881–890, https://doi.org/10.1016/j.cemconcomp.2011.02.011. Liaudat J., Martínez A., López C. M., Carol I., Numerical Modelling of ASR Expansions in Concrete, CONCREEP 10 (n.d.) pp. 445–454. https://doi.org/10. 1061/9780784479346.054. M. Alnaggar, G. Cusatis, G.D. Luzio, Lattice Discrete Particle Modeling (LDPM) of Alkali Silica Reaction (ASR) deterioration of concrete structures, Cem. Concr. Compos. 41 (2013) 45–59, https://doi.org/10.1016/j. cemconcomp.2013.04.015. ˘ M. Mahyar, S.T. Erdogan, M. Tokyay, Extension of the chemical index model for estimating Alkali-Silica reaction mitigation efficiency to slags and natural pozzolans, Constr. Build. Mater. 179 (2018) 587–597, https://doi.org/ 10.1016/j.conbuildmat.2018.05.217. L. Mao, Z. Hu, J. Xia, G. Feng, I. Azim, J. Yang, Q. Liu, Multi-phase modelling of electrochemical rehabilitation for ASR and chloride affected concrete composites, Compos. Struct. 207 (2019) 176–189, https://doi.org/10.1016/ j.compstruct.2018.09.063.