Proceedings of the 17th International Conference on Alkali-Aggregate Reaction in Concrete

The alkali metal concentration and pH values within the pore solution of a concrete strongly influence the occurrence and the extent of alkali-silica reaction (ASR). At (Na + K) concentrations <300 mmol/L, no significant ASR expansion is observed. The alkali metal and hydroxide concentration can be lowered by using either a low alkali cement or by blending Portland cement with silica-rich supplementary cementitious materials (SCMs) such as silica fume, fly ash, calcined clays or blast-furnace slag.The reaction of silica-rich SCMs lowers the Ca/Si-ratio in C-A-S-H, which increases the uptake of Na and K by C-A-S-H and lowers both alkali concentrations and pH values in the pore solution, efficiently preventing ASR.Thermodynamic modelling and the empirical Taylor model were used to predict the changes in the pore solution composition as a function of the amount of the respective SCM added to the concrete. Comparison with literature data confirms that the models can be used to predict the trends in alkali concentrations.The literature data and thermodynamic modelling indicate that the replacement of Portland cement with ≥ 20 wt.% silica fume or metakaolin, ≥ 50 wt.% silica-rich fly ash or ≥ 65 wt.% blast-furnace slag can suppress ASR expansion in concrete. The study was conducted within the framework of working group 2 of RILEM TC 301 ASR.

ASR expansion is caused by the confined precipitation of an alkali gel, formed from the reaction of Alkalies in the cement paste pore solution and reactive silica in the aggregates. The role of alkali concentration on the development of the reaction has led to the formulation of guidelines imposing alkali thresholds when using potentially reactive aggregates. However, it is difficult, experimentally to verify the effect of very low alkali concentrations, due to alkalies being systematically present in cement paste. Even low-alkali cements can provide sufficient alkalies for the reaction of highly reactive aggregates. Further, some aggregates are suspected of releasing their own alkalies, and testing methods to evaluate this potential are still in development. Here, a clinker has been specially produced at a high temperature (above 1700 ℃) to almost eliminate all alkalies through volatilisation. Ground and sulphated, we obtain an alkali-free cement from this clinker. We test the expansion of samples prepared with this cement and known reactive and unreactive aggregates. We use the results from these experiments to evaluate the effect of very low alkali concentrations and aggregate alkali leaching in mortars.

When alkali reactive or marginally reactive aggregates are used in concrete, the concrete mixtures are mitigated by using SCMs or by limiting alkali loading. The level of mitigation depends on the reactivity of the aggregate. These mixtures may still undergo low levels of reaction with the alkalis in the binder system without causing deleterious expansion. During this time, the alkalinity of the system normally decreases as the alkalis are bound in the reaction products. However, if an aggregate contributes, or releases, alkalis into the binder system over time, it can drive additional ASR reaction, leading to long-term damage. RILEM has published the AAR-8 method to measure the potential for alkali release of an aggregate but there is no correlation between the results from that test and the impact on the pore solution in the concrete yet.In this study, four either reactive or innocuous aggregates were tested according to AAR-8 and used to make mortar specimens. The impact of these aggregates on the pore solution of mortar was studied by extracting the pore solutions at various ages. The mortar specimens were cured at 60 ℃ to accelerate the ASR reaction and to see if alkali release can impact the pore solution as the alkalis are consumed in the reaction.Results showed that both reactive and innocuous aggregates can potentially release alkalis. However, no impact was observed on the pore solution of mortar made with the aggregates. The two methods used to obtain the pore solution of the mortar showed similar results.

Although the chemical basics of the ASR are largely revealed and widely accepted, the nature of the ASR expansion mechanism in concrete is still not yet sufficiently well understood. Recent observations showed that ASR products could be considered as colloids. In order to clarify, if and to what extent ASR products exhibit colloidal properties, 10 ASR products of different composition and water content were synthesized and stored for 1.5 years at 40 and 60 ℃. Among others, the ASR products were investigated every 24 weeks by means of NTA, 29Si NMR, XRD and by an osmotic cell test (OCT). The key results show that ASR products contain particles with a diameter of about 50–600 nm, what confirms their colloidal nature. The particle concentration increased with the Si concentration and the degree of polymerization in the ASR products respectively. The particles are unable to pass pores with a size smaller than their own, what represents a mechanism of semi-permeability leading to the Donnan effect and osmosis. By adding Ca, the particles start to link irreversibly to a gel that increasingly loses its osmotic potential. Over time, different crystalline products are formed, depending on composition and temperature.

In the context of concrete structures ageing, the study of cementitious material durability is of critical importance, particularly the degradations caused by the Alkali-Silica Reaction (ASR). This reaction is driven by the reactive silica dissolution of the concrete aggregate and can unfold in different manner according to the aggregate. The goal of the study is to improve the evaluation of the aggregate reactivity, in terms of quantity of reactive silica and dissolution rate, linked to the form and state (crystalline, amorphous, micro-crystalline) of the aggregate structure. Four aggregates were selected for this study: Potsdam sandstone, Springhill greywacke, Spratt limestone, and a non-reactive Quartzite (used as a reference). For dissolution properties assessment, the aggregates were immersed in a basic solution to promote the degradation of their reactive structure. To reproduce an idealized highly alkaline solution close to a concrete pore solution, dissolution tests were performed on a 1M NaOH solution. Two particle size fractions for each aggregate: 0.5 mm–1 mm and 1–2 mm (with a solid to solution mass ratio of 1:4) and three temperatures: 25, 38, and 60 ℃ were considered. Silica concentration in the solution was measured by complexometric titration. The study displays the dissolution rate constants and analyses the cause of the differences in dissolution.

Alkali-silica reaction (ASR) is a deterioration mechanism in concrete through which reactive silica from the aggregates react with alkalis in the pore solution to form an expansive, hygroscopic gel, producing internal pressures within the concrete pores. The expansion caused by ASR may continue and reach a plateau upon full consumption of the reactants; alkalis or silica. Concrete expansion and damage caused by ASR may be promoted when exposed to an external alkali source. Sodium chloride (NaCl), commonly applied to concrete as a deicing salt, may act as an external alkali source and promote ASR-induced damage. This paper investigates the effects of NaCl on aged concrete, which has previously reached an expansion plateau caused by ASR. The expansions of aged concrete prisms, containing different aggregate and cementitious materials, exposed to NaCl solution were examined. Considerable expansions after NaCl exposure were observed in aged concrete made with non-boosted alkalis (0.99% $${{\text{Na}}}_{2}{{\text{O}}}_{{\text{e}}}$$ Na 2 O e per mass of Portland cement) in comparison to concrete with boosted alkalis (1.25% $${{\text{Na}}}_{2}{{\text{O}}}_{{\text{e}}})$$ Na 2 O e ) , likely due to the presence of locations of unreacted silica. Moreover, aggregates of high reactivity provided greater late ASR expansions. The use of supplementary cementitious material was effective in limiting further ASR-induced expansion within the first year of testing.

We investigated the effect of partial replacement of Portland cement with SCMs such as limestone, fly ash, or volcanic pozzolan on the alkali metal distribution in cement paste. This is important to understand the underlying mechanisms of how SCMs mitigate the alkali-silica reaction. Cement pastes were cured at 20, 38, and 60 ℃, and the pore solution was analyzed using ICP-OES/MS. Solid phases were analyzed using TGA, XRD and SEM-EDS. Concrete prisms with the same cement were monitored for expansion according to the Norwegian 38 ℃ CPT and the RILEM AAR 11 60 ℃ CPT. We discuss the implications of our findings for accelerated ASR testing.

Although it is well established that the expansion of concrete due to the alkali–silica reaction (ASR) is very sensitive to temperature, and real structures are inevitably subject to temperature variations, studies on ASR expansion at low temperatures are scarce. This paper presents the results of laboratory tests on concrete specimens subjected to different temperature conditions, including cyclic temperature variations. The experimental results indicate that high temperatures significantly accelerate early-stage expansion, but have minimal impact on late-stage expansion. A simplified simulation demonstrates that laboratory-derived data on expansion behavior at different temperatures cannot be directly applied to real-world scenarios with temperature variations. Further, averaging real-world environmental temperatures significantly reduces the accuracy of the simulation.

In order to control alkali-silica reaction (ASR), various standards set an upper limit for the total alkali content (Na2Oeq) of concrete, and performance tests of specific aggregates have been conducted to determine the alkali threshold (AT) to be tolerated in terms of ASR risk. It well established that ASR expansion does not always show a positive correlation with concentration or quantity of Na2Oeq. In other words, AT is not simply fixed at a constant value. In JASS 5N, the construction specification for nuclear power plants by Japan Architectural Society Standard, the case where the expansion would be greater at smaller Na2Oeq is also considered. However, it is unclear under what conditions the expansion becomes greater when the Na2Oeq is smaller. Therefore, in this study, alkali-wrapped concrete prism test (AW-CPT) was conducted for a long period of time on Norwegian cataclasite and four types of Japanese reactive aggregates with different Na2Oeq and temperatures to identify ATs. The test results strongly suggested that AT cannot be simply determined. The difficulty of determining AT by CPT under a single condition was clearly shown in this paper.

Alkali-silica gels formed during alkali-silica reaction (ASR) can vary in composition and structure. This variation depends on the concrete composition, environmental conditions, gel maturation and the location of gel formation within the concrete matrix. An effective and well-known method to mitigate ASR is the use of aluminum-rich supplementary cementitious materials (SCMs). To investigate the effect of aluminum on ASR gel properties, concrete prisms with quartz sand, borosilicate glass, 90 wt.% Portland cement and 10 wt.% metakaolin were mixed. As an aluminum-free non-reactive “SCM”, limestone was used with the same replacement level. On these concretes, the expansion up to 140 days at 40 ℃ above water, the ASR gel composition and microstructure, as well as the spatial distribution of the gels, were investigated. Due to the small gel amounts formed, the structural investigations (29Si, 27Al NMR and 1H NMR spectroscopy) were performed on synthetic ASR gels with similar compositions as in the concretes. The concrete expansion for the mix with limestone was approximately three times higher than with metakaolin (0.5 vs. 1.5 mm/m). However, the concrete with metakaolin contained larger amounts of an Al-ASR gel than the concrete with limestone. Analytical investigations on the synthetic ASR gels indicate increased connectivity for the Al-ASR gel compared to the Al-free ASR gel. This connectivity change could alter the swelling properties of the gels, thus limiting the expansion pressure in concrete.

During alkali-carbonate reaction (ACR), dolomite present in the aggregate particles reacts with the hydroxide ions present in the concrete pore solution. The mechanism of expansion of ACR has generated controversial discussion in the scientific community. A detailed microstructural investigation was performed to characterize the changes in a concrete containing ACR-susceptible Kingston carbonate rock as aggregates. Four types of reaction products are present in the aggregates resulting from the dissolution of dolomite, illite and quartz. However, no alkali-silica reaction (ASR) products were observed. The cement paste adjacent to reactive aggregates is altered with the formation of a patchy layer of calcite embedded in de-calcified C-S-H with a high alkali content. A decrease of pH together with this alteration was indicated by the absence of S-containing phases within the reaction rim.

Alkali-carbonate reaction (ACR) is chemical reaction in concrete which may induce significant damage in concrete structures. Researchers have been focusing on studies of the symptoms of alkali carbonate reaction (ACR), namely expansion and loss of mechanical properties. Not enough attention has been paid to the processes of dedolomitization in dolomitic rocks. As shown in previous studies, dedolomitization in dolostones leads to the formation of a reaction rim which is similar to a nanofiltration structure, which, when thick enough, can prevent the diffusion of alkali ions and inhibit the dedolomitization process. This paper investigates the diffusion of alkali ions in dolomitic limestones and argillaceous dolomitic limestones and the process of dedolomitization of dolomite in the aggregates. Chemical, physical, and microstructural analysis of dolomitic rocks and the degree of dedolomitization were studied based on X-ray diffraction (XRD) and field emission scanning electron microscopy (SEM). It was found that the dedolomitization reaction occurred in dolomitic limestones is similar to that of dolostones where the product layer formed by the dolomitization reaction consists of an area of Mg(OH)2 crystals mixed with CaCO3 crystals. For argillaceous dolomitic limestones, some of the dolomite crystals were found to be covered by 0.5-2μm thick layer of a Mg-Al-Si phase, and the dolomite crystals wrapped by this Mg-Al-Si phase did not undergo the dedolomitization reaction, while the dolomite without such a layer shows obvious dedolomitization reaction. This suggests that the Mg-Al-Si phase prevents the dedolomitization reaction.

The composition of the pore solution is a main parameter determining the risk for alkali-silica reaction (ASR) in concrete. The relation between the pore solution composition and ASR expansion was established based on literature data, where concrete or mortar expansion as well as pore solution composition were reported.The general trend was that expansion only occurred at (Na + K) concentration > 300 to 400 mmol/L in the pore solution, or at a hydroxide concentration > 250 mmol/L. The pH of the pore solution alone is not suitable to assess the expansion risk as the pH values strongly depend on the temperature, are demanding to measure and are determined using diverse methods, which are often poorly described.Our literature search showed also that the Al concentration is not a selective criterion for ASR expansion as most pore solutions had Al concentrations below 1 mmol/L and as both the presence and absence of expansion was observed at low Al concentrations. All expanding systems showed also a strong undersaturation of < 10–4 with respect to quartz, which complies with a fast dissolution of silica and thus a high potential for ASR.This study showed that the sum of (Na + K) concentration is a reliable and relatively easily accessible parameter to indicate the potential for ASR in concrete.The study was conducted within the framework of working group 2 of RILEM TC 301 ASR.

The most common standards ruling the assessment of the potential reactivity of aggregates consider that the petrographic analysis should be followed, in the first step, by accelerated mortar-bar tests and, in case of a positive result, by concrete-prism tests.However, experience has shown that these different approaches often provide contradictory results in the classification of an aggregate as innocuous or potentially reactive. Discussion about the inaccuracy of the accelerated mortar-bar test for some slow reactive aggregates has been puzzling the scientific community and some explanations have been suggested to explain this fact.In the present work, deformed rocks used as aggregates, previously submitted to accelerated mortar-bar tests and to concrete-prism tests, are analyzed regarding different grain sizes. The research aims to define the role of crushing mechanisms in the destruction of crystals originated from sub-graining, due to tectonic deformation, during the production of the smaller particles requested by the mortar-bar test. With this purpose, the petrographic analysis of the rocks selected, and their correspondent sand size aggregates has been done using the image-based open software JMicrovision.

Though an important subject that has been investigated for several decades, the full understanding of the alkali release from aggregates, its contributing factors and role in the alkali-silica reaction still remains a challenge for the scientific community. Recently, the RILEM AAR-8 test method was developed and several improvements to this methodology have already been proposed. In this paper, automated scanning mineralogy (QEMSCAN®) is introduced as a complementary tool for petrographic characterization in order to better understand the factors contributing to the alkali release from aggregates with a similar mineralogy. It was used to examine distinctive textures, grain sizes and alteration of a mylonite, a cataclasite, a gneiss, a granite and an altered granite. The QEMSCAN® analysis enabled a more thorough understanding of the aggregates whilst complimenting the optical petrography.

In this study, the characteristics of the alkali-silica reaction (ASR) of glass cullet and the effect of fly ash on suppressing ASR were investigated by microscopic observation for the purpose of recycling glass cullet obtained by crushing the glass of photovoltaic panels as fine aggregate. As a result, it was shown that ASR can be effectively suppressed by replacing high-quality fly ash with 15% or more of cement for glass cullet, which exhibits high alkali-silica reactivity. Furthermore, chemical compositions were compared by SEM-EDS composition analysis of the reaction layer of the glass interface, the pozzolanic reaction layer of the fly ash, and the ASR gel. As a result, it was found that the chemical composition of the reaction layer of the glass interface and the pozzolanic reaction layer of the fly ash were similar and the calcium oxide contents was high. In comparison, the chemical composition of ASR gel was found to be high in silica and sodium oxide contents.

To investigate the effect of alkali dissolution from aggregates on ASR expansion, alkali metal-rich nepheline syenite powder (NS) was substituted as a fine aggregate replacement in reactive chart aggregate concrete and its expansion behavior was evaluated by concrete prism tests with alkaline wrapping. The results showed that the addition of 5wt% NS had no effect in the short ages but promoted expansion in the long ages. The effect in terms of total alkali content was an increase of 1.75 kg/m3, while the RILEM AAR-8 evaluation resulted in an alkali metal increase of 0.18 kg/m3, which was an underestimate. This may be due to too little Ca(OH)2 added in the test procedure of RILEM AAR-8. As the gradual dissolution of alkali metals from the NS, and the alkali silica gel produced slowly, contributes efficiently to the expansion, suggesting that expansion may continue over the long term in low total alkali content and low temperature environments, when alkali dissolving minerals are present.

Many structures damaged by Delayed ettringite formation (DEF) have been reported in France while Japan has not yet reported structures clearly affected by DEF even if they were exposed to high temperature over 70 ℃. In this paper, the difference of cement characteristic and environment between Japan and France were focused on. Cubic concrete specimens with French and Japanese cements, French aggregate and high temperature curing at early age were exposed at four places in Japan. The expansion strain was continuously measured with observation of expansion cracking. As a result, the effect of exposed environment on DEF expansion rate in Japan was varied according to cement types while the ultimate expansion was likely to be not affected by the environment but by cement type. The laboratory DEF expansion tests using the same concrete mix proportion and heating process as those in field experiment also suggest that the temperature dependency of expansion rate is different in each cement type.

In regions characterized by cold and snowy climates, the combined deterioration mechanisms resulting from the salt-induced steel corrosion and the expansion alkali-silica reaction (ASR) has emerged notably due to the influence of anti-freezing agents spread out on the road. Consequently, it becomes imperative to precisely diagnose and prognosis the occurrence of these phenomena and incorporate them into the comprehensive maintenance and management plan for the entire bridge structure. This investigation was undertaken to assess the extent of chloride penetration into concrete and the expansivity of concrete due to ASR in RC deck slabs along the Hokuriku Expressway to elucidate the impact of this compounded deterioration on the extracted RC deck slab. The findings revealed that in a significant proportion of RC slab decks where the erosion had transpired, steel corrosion manifested after the infiltration of chloride ions from the anti-freezing agent into the concrete. Furthermore, the formation of cracks, as a consequence of the combined deterioration mechanisms of steel corrosion and ASR, was observed to diminish the physical integrity of the concrete in RC deck slab.

DEF is a deleterious reaction which may cause cracking in concrete structures. In Australia, reported observations of DEF have been associated with the presence of ASR. Although DEF is an issue of concern for the construction industry, with each state authority having their own specifications for temperature limits in the manufacture of heat-cured precast concrete elements, little work has been done on the susceptibility of structural concrete to DEF or DEF in the presence of reactive aggregates. This paper presents the outcomes of an investigation into the concrete systems at risk of deleterious DEF in the presence of ASR reactive aggregates.

Similarly to Alkali Aggregate Reaction, Delayed Ettringite Formation (DEF) is highly affected by moisture. Understanding quantitatively the coupling between moisture and expansion is mandatory to be able to predict the field behavior of concrete exposed to variable water supply by using results of laboratory tests performed in controlled conditions. This paper presents an experimental investigation that consists in exposing concrete specimens to wetting/drying cycles. To be more representative of field exposure, the wetting duration was set to relatively short periods. For the concrete mixes investigated, the tests showed that not only the kinetics of expansions but also their final magnitude is affected by the water-supply history. Some preliminary assumptions are proposed to explain this phenomenon.

In addition to the alkali-silica reaction (ASR), delayed ettringite formation (DEF) has recently received more attention among deterioration due to the internal swelling reaction of concrete. In this study, concrete specimens in which both ASR and DEF occur simultaneously were prepared using ASR-reactive aggregates, covering specimens with nonwoven cloths containing an alkaline solution to promote ASR, and giving a high-temperature history at an early age to promote DEF to investigate the mechanism of combined ASR and DEF deterioration. At any given age, we measured the expansion, performed tissue observations, and analyzed the elements using an electron probe micro analyzer. When concrete was prepared with NaOH added primarily to promote ASR and with K2SO4 added primarily to promote DEF, ASR occurred in both cases, but DEF occurred only with K2SO4 addition. When both ASR and DEF occurred, the expansion was relatively large, and the estimated pH values calculated from the alkali concentration gradually decreased with age.

Concrete infrastructure in cold climates is susceptible to rapid deterioration due to exposure to a combination of degradation mechanisms that can lead to a reduction in service life. While damage caused by reinforcement corrosion is one of the primary concerns of concrete structures subjected to harsh environmental conditions, its effects are intensified with the presence of frost damage. This is due to the fact that microcracks associated with freezing and thawing cycles (FTC) may accelerate ion transport and corrosion conditions of embedded reinforcement. Moreover, the expected rise in the frequency of FTC in cold climates resulting from climate change will magnify the deleterious effects on concrete infrastructure. Concrete deck slabs are among the critical and vulnerable elements of a bridge that are exposed to the ingress of chloride ions resulting from the use of winter de-icing salts, coupled with FTC. At present, few researchers have explored the synergetic impacts of reinforcement corrosion and FTC on the structural performance of affected prestressed concrete field members. This paper presents the results of the condition assessment of a corroded post-tensioned concrete deck slab element using visual assessment and non-destructive testing (NDT) techniques. The slabs were obtained from a major Canadian bridge after nearly 60 years of service. The findings of this study can ultimately aid in developing more effective maintenance and repair strategies for existing and affected field structures in order to enhance public safety and reduce the economic costs associated with premature structural deterioration.

Delayed ettringite formation (DEF) is a deleterious reaction which can result in expansion and cracking of concrete. The reaction occurs in the cement paste and is associated with the recrystallisation of the ettringite at a later age after hardening of the concrete. As ettringite is an expansive phase and as recrystallisation occurs in the hardened state, its formation may lead to cracking of the concrete. Laboratory tests of DEF induced expansion use specific conditions to initiate expansion which include elevating the alkali and sulphate contents (1% Na2Oe and 4% SO3) of the cement as well as heat curing the specimens at elevated temperatures (90 ℃) for extended periods (12 h). These conditions coupled with the appropriate calcium aluminate content have yielded significant expansion within a year of heat curing the specimens. The expansion process is characteristic of an underlying reaction which once initiated accelerates before depleting the available reactants. In this paper, simple solid state kinetic models have been applied to modelling the expansion of concrete prisms. The Prout-Tompkins’ auto-acceleration model was found to best fit the expansion data.

Delayed Ettringite Formation (DEF) is among the Internal Swelling Reactions (ISR) that has a detrimental effect on concrete thus compromising the material strength, stiffness, and integrity. Previous research shows that the multi-level assessment applied to concrete affected by ISR is a reliable approach to evaluate cause and extent of damage. However, the Stiffness Damage Test (SDT) has been more readily accepted for engineering applications while further understanding on the application of the Damage Rating Index (DRI) is necessary. This work therefore presents a condition assessment of a DEF-affected unreinforced concrete beam subjected to a moisture gradient to induce varying degrees of damage across the beam height by adapting the multi-level assessment. Correlations between the DEF damage features as depicted through the DRI and outputs from the SDT were found, highlighting the impact of the frequency of feature observation at a scale representative of the measured damage to the material.

It is important to distinguish between DEF (delayed ettringite formation) from ASR (alkali-silica reaction), which show similar macroscopic deterioration phenomena, using microscopic techniques.In addition to the petrographic observation, a residual expansion test of a concrete core has been proposed to distinguish DEF from ASR. In this method, concrete cores are immersed in a solution saturated with Ca(OH)2 (or water) and a 1 mol/L NaOH solution. The DEF potential is evaluated under the former condition and the ASR potential under the latter. Alkali leaching promotes DEF and alkali supply promotes ASR. However, the use of this type of the test method has been limited, and its validity has not been adequately verified.In this study, relatively large concrete specimens with DEF and ASR expansion potential were prepared under 80 ℃ pre-curing and using ASR reactive aggregate, and cores were taken for residual expansion testing. The results of the tests conducted by four institutes confirmed that the specimen with only DEF potential expanded in water and gaps filled with ettringite were observed around the aggregates. Specimens with both DEF and ASR potential expanded in both water and NaOH immersion, and the characteristics of DEF and/or ASR were identified by microscopic observation. In addition to the petrographic observation, the water and NaOH immersion tests were useful in distinguishing DEF from ASR.

Concrete construction is currently facing a significant challenge to reduce its environmental impact to minimal levels. Portland cement (OPC), the major contributor towards concrete’s carbon footprint, is increasingly being replaced or blended with supplementary cementitious materials (SCMs), alternative materials, and/or by-products; among these, the use of limestone fillers interground with PC resulting in Portland limestone cement (PLC) is probably the most accepted trend by the market. The full implications of PLC in alkali-aggregate reaction (AAR) contexts, however, remain to be understood, demanding the evaluation of current testing methods (i.e., AMBT and CPT) for assessing the potential reactivity of aggregates and mixtures when incorporating PLC. By a comparative analysis of expansion, dynamic modulus of elasticity, and damage rating index over time (i.e., 1 and 6 moths) for both OPC and PLC mortar and concrete, similar kinetics between the two cement types were observed. Yet, PLC has shown different performances depending on aggregate reactivity level.

In previous research, volcanic rocks from the Azores Islands (Portugal), Brazil, Canada, Canary Islands (Spain), Hawaiian Islands (USA), Iceland, Japan, Mozambique, New Zealand, Norway, and Turkey were studied to assess their potential reactivity to alkalis. The analyzed samples correspond to basalt, andesite and rhyolite. Several tests were performed, namely, petrographic characterization by optical microscopy and scanning electron microscopy, bulk rock chemical analysis, accelerated mortar-bar test and concrete prism test.In volcanic rocks, the possible reactive forms of silica are not only of minute size, and hard to identify, but are also usually present in very low content, and often below the detection limits used in chemical bulk rock analysis.This study reports the results obtained in additional tests to characterize the aforementioned aggregates: 1) the phosphoric acid method, by dissolution of silicates of the rock reduced to powder, preserving the free silica forms of the sample; and 2) the gel-pat test, applied on lengthwise slabs of concrete prisms prepared with the selected aggregates, to obtain the possible presence of reactive components, in particular volcanic glass and free silica polymorphs.The results from the two tests are presented and the possible correlations with the results from the petrographic characterization and from the expansion tests are discussed.

Although the alkali content of cement required to initiate and sustain alkali-silica reaction (ASR) was thought to be >0.60% Na2Oeq, the data from exposure sites show that concrete blocks made with low-alkali portland cements can still show deleterious expansion. The more important parameter is the total available alkali content in kg/m3 which will lead to ASR. The alkali threshold of an aggregate can be defined as the alkali loading of a mixture below which ASR will not initiate and/or cause deleterious expansion. Obtaining information about the alkali threshold of different aggregates has been largely ignored but now merits significant attention as a key to developing effective long-term ASR preventive measures. Most ASR performance tests use a fixed cement content and alkali level; therefore they cannot be used to determine alkali threshold. In this paper, several features of the miniature concrete prism test (MCPT) and the accelerated mortar bar test (AMBT) were modified to determine the alkali threshold of several different known ASR aggregates. Low- and high-alkali portland cements, and a combination of both, were used to obtain a spectrum of alkali loadings to determine the alkali level at which ASR expansion would occur for different aggregates. The alkalinity of the soak solution was matched to that of the pore solution for each mixture. The results showed that MCPT can differentiate between mixtures with different alkali loadings. The AMBT results show that there is potential for this method, but further interpretation of results is needed since this is a mortar bar method and the cement content (and thus alkali loading) needs to be scaled appropriately to concrete.

The amount of available alkalis to the pore solution is an important property for alkali-silica reaction (ASR) prevention when supplementary cementitious materials (SCMs) are used. Supplementary cementing materials (SCM) are known to produce cement hydrates that increase alkali binding compared to portland cement only systems. The alkali-binding capacity of a specific SCM coupled with the alkali threshold of an aggregate will determine the required SCM dosage to prevent deleterious ASR for that specific reactive aggregate., Pore solution extraction (PSE) and analysis has been the most direct and reliable way to determine available alkalis of cementitious pastes. PSE requires advanced equipment and can be time intensive. In addition, the representativity of the pore solution extracted is a matter of debate. This paper introduces a new method called the alkali leaching test (ALT) to determine the available alkalis of binders containing portland cement and SCMs. Cement pastes at different alkali levels and containing various types and dosages of SCMs were tested. The results of the test showed different alkali-binding capacity for different fly ashes; also as the replacement level of SCMs increased, the available alkalis of cement pastes decreased.

The technical specifications for the construction of airport or road pavements usually include the requirement of the appropriate selection of non-reactive aggregate for concrete. This ensures an appropriate safety factor against the occurrence of a harmful alkali-silica reaction (ASR). However, it is difficult to predict the effects of external sources of alkalis, including alkaline deicers used for winter maintenance of pavements in wet-freeze climate zone. This study is aimed to evaluate the ASR performance of concrete using RILEM AAR-12 procedure with external alkali supply comparing it to the standard evaluation in Austria (1 molar NaOH, 38 ℃) according to ÖNORM B 3100. Tested materials included reactive and moderately reactive siliceous gravel and crushed rock aggregate. Concrete specimens were subjected to cyclic exposure to high humidity storage at 60 ℃, drying, and soaking in de-icing salt solution (sodium chloride and sodium formate). Measurements of the expansion of concrete specimens, mass changes, and changes in the dynamic elastic modulus were systematically taken. The microstructural characterization of concrete was performed to identify alkali-silica reaction products using scanning electron microscopy SEM/EDS. The ASR related damage in concrete exposed to the performance testing with external alkali supply is moderately increased when NaCl solution is replaced with HCOONa solution at the same concentration.

The alkali-silica reaction (ASR) still causes considerable damage to concrete structures and furthermore the most beneficial method to avoid ASR is to test the aggregate for reactivity. There are internationally different but similar test methods for a reliable assessment of the damage potential of an aggregate, but these methods require specially manufactured specimens and test durations up to several months. Furthermore, a method for the short-term characterization of the aggregate is still missing. Therefore, the Institute of Materials, Physics and Chemistry of Buildings at Hamburg University of Technology (TUHH) is working on the development of a new fast test method.The idea of this new test method is an accelerated dissolution experiment. However, in this approach the time dependent change of chemical parameters due to the dissolution process in hot basic solution with an excess of alkalis is used to estimate the reactivity. The pH, the electrical conductivity, and the redox potential are recorded as a function of time. The assumption is that the different alkali reactivity of the aggregates is significantly reflected in the chemical parameters, which should make a categorization regarding the alkali reactivity possible.

The Ministry of Transportation (MTO) evaluates the reactivity of concrete aggregates using the MTO laboratory methods, Accelerated Mortar Bar Test (LS-620) and Concrete Prism Test (LS-635), demanding the use of Portland Cement (GU). The introduction of Portland-limestone (GUL) cement in the industry calls for an investigation of its effect on the two methods.This paper reports on LS-620 test data on two MTO reference aggregates (a quarried dolomitic limestone (ACR CA1) from Kingston area and a quarried siliceous limestone (ASR CA6) from Eastern Ontario) using both GU and GUL cements. The data were produced through 4 years of MTO Aggregate and Soil Proficiency Programs.2019 and 2022 programs involved testing of ACR CA1 using GU (2019) and GUL (2022). The expansion of mortar bars at 14 days produced by GU and GUL from the same supplier are very similar. ANOVA analysis of the expansion data showed no significant difference. However, the precision estimates calculated for ACR CA1 suggested that the variations in expansion with GUL cement are much greater than the variations for GU cement from the same supplier.The findings from the 2021 program showed that with ASR CA6, the use of GUL cement has no significant impact on the test method. The expansion range produced by using GU cement should only be exceeded one time in twenty if GUL cement of the same supplier is used.The paper also includes analysis of the data from 2023 program involving testing of ASR CA6 using GU and GUL cements from different suppliers. The analysis further verifies the effect of GU vs. GUL on LS-620 testing of aggregate reactivity and supports establishment of expansion ranges for the AMBT of ASR CA6.

The Ministry of Transportation of Ontario (MTO) has been actively involved in developing and establishing AAR reference materials for over 35 years. Currently, the primary function of these materials is for test method control and calibration for the critical purpose of ensuring the highest quality testing in support of durable Ontario infrastructure. A secondary and equally important function of these materials is to promote research into AAR. The availability and use of materials with established or “known” parameters is often essential to the success of research projects. As a result, many applied, practical and experimental researchers engaged in test method development and innovative technologies development request MTO’s AAR reference materials in support of their research programs. MTO has historically both directly and indirectly greatly benefitted from these types of external research as a result of the practice of providing reference materials. This paper provides an overview of MTO’s reference material program including its development and evolution. MTO’s new reference material code system is presented, as well as the procedure for requesting materials and MTO’s policies for making materials available to requestors. Focus is placed on the ASR reference materials past and present including each material’s physical and engineering property test ranges, geochemistry and petrography, as well as descriptions of each material’s field performance and geological environment of origin, where available.

This paper presents the results of the Ontario Ministry of Transportation’s (MTO’s) work to validate a new alkali-silica reactive (ASR) reference material (RM) as a replacement for the internationally recognized Spratt No. 3. Availability of Spratt No. 3 has been restricted in recent years due to the limited quantities remaining, and the need for MTO to conserve remaining resources for control and calibration of AAR test methods until a comparable replacement material could be established and validated. Prospecting for a new MTO ASR RM source in the same geological formation as, and with comparable characteristics to Spratt No. 3 was undertaken by MTO between 2018 and 2020. Screening of prospective sources included sampling and testing, i.e., accelerated mortar bar testing (AMBT) to assess the level of reactivity, and insoluble residue (IR) testing to assess the general composition. Once exploration identified a suitable candidate source, further investigations including geological examination, sampling and testing were conducted to pinpoint the most highly ASR areas and strata within the quarry in preparation for aggregate production. MTO RM ASR CA6 was produced (drilled, blasted, crushed, transported) and placed in stockpile in MTO’s Aggregate Storage Building in September 2020. MTO RM ASR CA6 has been included in several Soil and Aggregate Proficiency Sample Testing Programs (SASTP) conducted by the Soils and Aggregates Section of the Engineering Materials Office since 2021. Data generated by the SASTP has been utilized to obtain material properties and to validate and establish MTO RM ASR CA6.Preliminary SASTP results indicate that MTO RM ASR CA6 gives similar or higher expansions in both the 14-day AMBT and 1-year CPT as compared with Spratt No. 3. Additional geochemical and petrographic data further confirm the comparable nature of MTO RM ASR CA6 and the Spratt aggregate.

The release of alkalis from aggregates can be a source of alkalis in concrete that has been exposed to a moist environment for a long time. Several types of aggregates can lixiviate different amounts of alkalis; this depends not only on an aggregate´s mineralogy, but also its shape, that is, if it is a crushed or a naturally abraded aggregate. In concrete production, it is required to estimate the potential amount of alkalis releasable in order to determine the concrete mix proportions. Several solutions and test conditions are used to evaluate aggregate alkali release. The potential alkali release of different Spanish aggregates was assessed in this study utilizing different solutions and test conditions; the effect of the ions’ type, concentration, and temperature in the solution was studied. The main findings of this work were that (1) the ultrapure water extraction method provides an easy and fast way to estimate the potential leachable alkali content from aggregates, and (2) in the method of extraction on concentrated alkaline solutions, reactors at 150 ºC are used; (3) the test times are reduced, allowing a greater extraction of alkalis in less time. To estimate the potential contribution of alkalis from the aggregates in a concrete structure, the more accurate extraction methods are those that partially simulate the concentration of the porosity phase of concrete.

The newly developed ‘DTU concrete prism test method’ (DTU CPT) ‘Accelerated test method for alkali silica reaction in concrete’ is based on the NORDTEST method NT BUILD 295 ‘Sand-alkali-silica reactivity accelerated test’. The changes from NT BUILD 295 in DTU CPT primary concerns test specimen size, aggregate size and expansion limit. The DTU CPT can be used for ASR-reactivity assessment of a single coarse aggregate or a combination in the range (4–16 mm) hence the method can reproduce representative results for concrete. The method applies concrete prisms of size 75 × 75 × 285 mm in saturated NaCl solution at 50 ℃ up to 52 weeks.The paper shows measurement results by use of DTU CPT for seven coarse aggregates, both fast reactive and slow reactive aggregates, in relation with two innocuous reference aggregates. The results from DTU CPT tests are compared with results from ASTM C 1293 tests. The results from the test show that DTU CPT seems to be useful for assessing both fast and slow coarse reactive aggregate types, seemingly also for aggregates with pessimum behaviour.

The prevailing practice entails a uniform, relatively modest level of alkali loading (typically 2.1–2.4 kg/m3) across all concrete mixes, disregarding the specific application. Essentially, this one-size-fits-all approach overlooks the crucial aspect of determining Aggregate Threshold Alkalinity (THA) for Alkali-Silica Reactivity (ASR). There is a pressing demand for innovative, swift, and dependable methods to gauge aggregate reactivity and THA.To tackle these challenges, the authors have devised expeditious ASR testing techniques, including an Aggregate Chemical Test (AASHTO T364) and an Accelerated Concrete Cylinder Test (ACCT, AASHTO TP142). They have also introduced a novel performance-based method for estimating aggregate THA. Arrhenius rate theory and insights from materials science on reaction and expansion mechanisms were utilized to interpret the data and determine the Composite Activation Parameter (CAP) as an indicator of aggregate reactivity. A lower CAP signifies higher reactivity. This CAP-based approach exhibited a better comparison with ASTM C1293-based reactivity measurements and facilitated a consistent identification of aggregates belonging to false positive and negative attributes.A procedure based on AASHTO T364 has been developed to ascertain Aggregate THA by analyzing the CAP-alkalinity relationship. The relationship between Pore Solution Alkalinity (PSA) and Aggregate THA (PSA ≤ THA) serves as the foundation for estimating Supplementary Cementitious Material (SCM) dosage to mitigate ASR in concrete. This approach facilitates rapid assessment of aggregate THA, determines appropriate concrete alkali loading, and estimates SCM dosage for effective ASR mitigation.

The RILEM AAR-2 method and the ASTM C 1260 method are nearly identical ultra-accelerated mortar bar test methods. The RILEM AAR-2 method has two versions AAR-2.1 for specimen size 25 × 25 × 285 mm and AAR-2.2 for specimen size 40 × 40 × 160 mm. The NORDTEST NT BUILD 295 method ‘Sand-alkali-silica reactivity accelerated test’ is equal to the Danish TI-B 51 method, which applies specimen size 40 × 40 × 160 mm in saturated NaCl solution at 50 ℃ up to 52 weeks. The TI-B 51 method has been used for 37 years for testing Danish sand (0–4 mm) with (fast reactive) porous opaline and porous calcedonian flint types. The argument for using saturated NaCl solution is that structures often are exposed to alkalis from sea water or de-icing salt (NaCl). In Denmark, many buildings from before 1987 contain a critical amount of fast reactive opaline flints, thus resulting in the structures being sensitive to exposure to NaCl.The paper shows measurement results 1) by use of RILEM AAR-2.1 and AAR-2.2 test methods which by comparison of results gives a ratio for conversion in expansion recordings between the two test methods, and 2) a comparison of results between ASTM C 1260, RILEM AAR-2 and TI-B 51 test methods on both fast and slowly reactive rock types.

In Denmark, TI-B 51 is used to test the reactivity of sand containing porous flint, but it is also the reference test to document the ASR performance of binders that do not fulfil the requirements on the total alkali content, orto decide on the declared alkali content of a binder.In TI-B 51, mortar bars are first cured in water for four weeks and then immersed in a saturated NaCl solution at 50 ℃. However, the type of binder affects different properties, such as the chemistry of the pore solution and the ingress resistance to external compounds. The present study investigated these two aspects and intended to address the following question: what is the most influential factor on the TI-B 51 outcome between the free alkali content and the resistance to NaCl ingress?Three binders were used: 1) a low-alkali OPC, 2) the same OPC with 35 wt.% replacement by fly ash and 3) a composite cement with 35 wt.% clinker replacement by equal amounts of calcined clay and limestone (same clinker as the OPC). Investigations were carried out on mortar bars cast with a highly reactive sand containing porous opaline flint. The experimental program included: TI-B 51 expansions, penetration depths of chloride over time, thin sections and µXRF analyses. In parallel, corresponding paste samples were cast to determine the free alkali metal concentration in the pore solution by Cold Water Extraction.

During an investigation of the potential of quartz rich aggregate to cause alkali-silica reaction, a number of concrete mixtures were made with different stone contents. One set was made at the proportion of 70% coarse aggregate defined in the ASTM C1293 version of the test. A second set was made with 52% coarse aggregate which was the concrete intended to be used on the job site. Other than the stone content the mixture proportions and subsequent storage regimes were identical. The concrete made with 70% coarse aggregate did not expand significantly and showed no sign of deleterious alkali-silica reaction. The concrete made with 52% coarse aggregate gave an average expansion of 0.083% at one year and 0.13% at two years, exceeding the usual limit of a maximum of 0.040% at one year. The 70% coarse aggregate mixture gave an average expansion of 0.038% expansion at two years. Petrographic examination showed alkali-silica gel and associated cracking in the deleteriously expansive prisms made with 52% coarse aggregate. Gel and cracks were not found in the prisms made with 70% coarse aggregate.This observation appears to be associated with a pessimum effect. This indicates that concrete prism testing is best done at stone contents likely to be those used in actual practice and that testing at unusual aggregate proportions may be misleading.

The key question for “ASR performance testing” when subjecting aggregates containing alkalis - and during service life possibly releasing alkalis (“ARA aggregates”) - is whether such properties will be accounted for during the performance testing. Further, how should test results from a testing method assessing the potential alkali contribution from an aggregate (e.g. RILEM AAR-8) be interpreted in context of performance testing and in lab/field relation of such tests?To investigate this, an attempt is made to detect the alkali release as well as the change of expansion when subjecting ARA aggregates to RILEM AAR-10 testing (and AAR-8). The “ARA-proof-of-concept” in concrete expansion testing is demonstrated (preliminary results) in the present study by including “non-ASR-reactive” alkali releasing fine aggregate to a medium reactive concrete mix design. Further, because the alkali release expectedly is more pronounced in the fine fraction including filler, the attempt is made to explicitly address the effect of filler content and properties on the release.The testing methods used in the study include RILEM AAR-10, RILEM AAR-8 (on “bulk” 0/4 mm and – modified – on individual fractions with and without filler) and high-pressure pore water extraction. The background context, program outline and preliminary results are presented.The preliminary results support the hypothesis that a significant part of “ARA” is accounted for during the concrete prism testing, but this should be validated also for other aggregates. The work is still ongoing, and studies on granulometric effects are also in progress.

When densified silica fume (DSF) is used in the ASTM C1567 test method and mixed as prescribed in the standard, the silica fume agglomerates (SFAs) may not break up. The lack of dispersed silica fume can result in significantly increased expansions of C1567 mortar bars relative to the same mix containing well-dispersed silica fume.Testing was performed to determine the difference in expansions between C1567 mortar bars prepared with as-received DSF and those prepared using different methods to break up and disperse the silica fume. The potential of the SFAs to act as reactive silica particles that can contribute to expansion of the mortar bars was also investigated. Petrographic examinations (ASTM C856) were performed on the bars following the testing. Point count analyses (ASTM C457) were performed to determine the volume of the SFAs in the tested mortar bars. Recommendations are made regarding modifications to the C1567 test method when using silica fume.

Supplementary cementitious materials (SCMs) are often used in concrete to reduce the risk of ASR, primarily through alkali dilution and binding. However, some SCMs also contain a high level of alkalis, and these may be soluble in the pore solution of concrete, leading to increased OH- concentration and ASR risk. This may be especially problematic for many non-conventional and rapidly growing SCMs, such as volcanic ashes, marginal coal ashes, and ground glass, with Na2Oeq > 3.0%wt. This paper presents a new soluble alkali test to quantify the soluble fraction of alkalis in an SCM. The test was applied to 14 SCMs, including volcanic ashes, calcined clays, coal ashes, and ground glass, and their alkali release was monitored for six months. The results are compared with ASTM C311’s Available Alkali Test method. The pore solution compositions of cement pastes containing the SCMs were also analyzed over one year. The findings indicate that a considerable fraction of the total alkalis in SCMs is soluble in concrete's pore solution. However, the pozzolanic reaction can bind most of the dissolved alkalis, resulting in a net alkali sink for a majority of the tested SCMs. Additionally, regression analysis revealed that the reduction in [OH-] of the pore solution of concrete by using SCM is significantly dependent on the SCM’s soluble alkalis, pozzolanic reactivity, and Ca/(Si + Al) oxide ratio.

As sustainability measures in the construction industry aim to reduce cement use, the reactive potential of low-alkali mixtures must be further evaluated. While the concrete prism test (CPT) is the most reliable laboratory procedure for determining aggregate’s reactivity, leaching that occurs during the test can compromise the accuracy of results. To overcome this limitation, alkalis are boosted in CPT specimens by adding NaOH to the mixing water, but further studies are required to understand if this approach is suitable for evaluating the reactivity of aggregates in low-alkali mixtures. This study compares three test setups (CPT, soaked, wrapped, and encapsulated) to evaluate ASR-induced expansion and damage development of boosted and non-boosted mixtures with low amounts of alkalis. The results show that regardless of the testing protocol, mixtures reach similar ultimate expansion when the total system alkali content is as low as to 3.61 kg/m3, yet the ASR-kinetics and crack features of CPT mixtures are accelerated and exacerbate due to alkali boosting when compared to non-boosted mixtures.

The use of non-reactive aggregate is the most reliable measure for avoiding alkali aggregate reaction (AAR) in concrete. Rapid and reliable identification of the distribution of alkali-reactive aggregate in an area is important for the construction and long-term service of concrete infrastructures. In present study, about 100 river sand samples distributed over 2100-km-long of the Yarlung Tsangpo River in Qinghai-Tibet Plateau were collected, and their alkali-silica reactivity as well as the type and source of reactive constituent of expansive sample were examined by petrographic analysis and Chinese autoclave method. Results show that 13% of these sands are reactive, and the main reactive constituent are microcrystalline quartz and acid volcanics. Based on expansion measurement, petrography quantification and geological mapping, a map showing the distribution of the alkali-silica reactivity of sands in drainage basin of Yarlung Tsangpo River is established.

This study investigates the effect of pore solution on alkali-silica reaction (ASR) expansion and mitigation using the simulated pore solution method. The simulated pore solution method is an accelerated ASR test developed at EPFL, Switzerland which uses the pore solution alkali concentration of the binder of interest at 28 days as basis of the storage solution. This study presents the 4-year expansion results of concrete prisms with reactive Australian aggregate (dacite), with and without 25% fly ash stored at 38 ℃. Results show no expansion in concrete prisms with 25% fly ash even if the cement alkali content was boosted up to 1%Na2Oeq. Extensive cracking was notable in the concrete without fly ash while the concretes with fly ash show only minimal presence of fine cracks.

Accelerated test methods have long and successfully been used to assess and categorise the reactivity of aggregates to the alkali-silica reaction (ASR). Test methods (e.g. RILEM AAR-3.2) have also been developed to assess the potential for alkali content threshold determination for aggregates that have been categorised as reactive and to investigate the efficacy of supplementary cementitious materials (SCMs) in mitigating ASR (e.g. RILEM AAR-10). Typically, these tests are based on the concrete prism test (CPT). The CPT is, however, susceptible to leaching of the alkali from the prisms. As testing alkali threshold content or SCM efficacy is dependent on the alkali content of the concrete prism, alkali leaching has the potential for providing misleading results. This study reports the outcomes of threshold testing using CPT as well as the use of a novel pore solution immersion test method for their assessment of threshold determination and SCM efficacy in mitigation of ASR for a course reactive aggregate.

The alkalinity of concrete has long been recognized as an important factor in alkali-silica reaction (ASR). The introduction of electrical resistivity for evaluating ASR potential adds a new dimension of importance to the role of alkalis that are the primary conductive ions in concrete. A sand with a high alkali affinity and a non-reactive sand were used in this study. High-pressure extraction was done to obtain pore solution for analyses by XRF. The electrical resistivity of the hardened mortars and extracted solutions was measured. The effect of aggregate absorption capacities on the pore solution concentrations was accounted for using a pore partitioning model (PPM) to help isolate the influence of aggregates on alkalinity. The capability of aggregate to adsorb alkali in hardened mortars was supported by the results .

This paper is based on the hypothesis that expansion due to alkali-aggregate reaction (AAR) is caused by alkali-silica gel (ASG) constrained inside the aggregate, which is produced by the alkali-silica reaction (ASR) caused by high pH concrete porewater and reactive aggregates. An attempt was then made to describe the four points of studies required to bridge the gap between materials science fundamentals and engineering applications. 1) Petrographic diagnosis is essential for the analysis of the AAR, but its implementation has to be based on an assessment of what is not visible by polarized light microscopy, i.e. the presence of cryptocrystalline quartz and highly viscous ASG. 2) It is known that the trend of AAR expansion does not match that of accelerated tests and exposure conditions, and it is explained that this is due to the fact that the AAR expansion is affected by the change in the constraining conditions of the ASG due to the accelerated ASR. This can be a reason why the produced amount of ASG does not show a proportional correlation with the expansion. 3) It was then explained that this working mechanism can explain various complex pessimum phenomena (aggregate composition, aggregate grain size, total alkali content, temperature, and age of the evaluation). 4) The rate of ASR depends on the pH of the porewater, but the working mechanism by which the pH is determined and the process by which it influences AAR expansion was presented, explaining the necessity for long-term evaluation. One important suggestion of this study is that simply, slow ASR will result in more AAR expansion by the same amount of ASG.

In 2015 the Pennsylvania Department of Transportation (PennDOT) and the Pennsylvania Aggregates and Concrete Association (PACA) implemented a unique concrete aggregate testing and evaluation protocol to mitigate the risk of deleterious alkali-silica reaction (ASR). The novel/cutting-edge approach, at that time, utilized widespread ASTM C1293 Concrete Prism tests to evaluate aggregate reactivity. The reactivity testing was required to be repeated once every five years. Based on the aggregate reactivity levels, prescriptive amounts of supplementary cementitious materials (SCMs) are utilized to mitigate the reaction, where increased reactivity levels require increased amounts of SCMs. The implementation of this specification brought about beneficial results, along with some known and a few unanticipated challenges. With the implementation of the new specifications and the increase in SCMs, the northeast portion of the state saw a dramatic increase in scaling of flatwork. This presentation will review the reasons for adopting these specifications, the benefits realized, challenges faced during implementation, and current challenges.

One of the most destructive problems affecting concrete infrastructure worldwide is the Alkali-silica reaction (ASR). There are well-established tools and protocols for detecting and addressing ASR in the field. Appropriately using supplementary cementing materials (SCMs) is widely accepted to prevent ASR-induced expansion and deterioration. However, traditional standards and protocols require ongoing improvement to ensure optimal material use without sacrificing performance. This paper provides an extensive review of the use of SCMs to prevent and/or mitigate the effects of ASR in concrete. The discussion emphasizes the chemical composition of the blended binder mixtures used in various studies and their effectiveness against ASR. A critical analysis is conducted, and a performance-based framework is proposed that centers on the main ternary oxides of binder materials for selecting raw materials to mitigate ASR in concrete. The proposed ternary oxides approach yields promising results, providing valuable data to assist in selecting the best options for applying different SCMs and quantities in concrete structures exposed to ASR development. The analysis sheds light on the importance of each main oxide of binder material (CaO, SiO2, and Al2O3) while offering valuable insights to support decision-making.

It is widely accepted that ASR-induced expansion and deterioration may be prevented by appropriately using supplementary cementing materials (SCMs). Moreover, traditional standards and protocols for concrete production often require improvements and adaptations to optimize material use while maintaining performance. To address this, a new performance-based framework is proposed to optimize the performance-based selection of raw materials to mitigate ASR in concrete. To achieve this goal, concrete cylinders (100 × 200 mm) were produced using two types of reactive aggregates (fine and coarse) and five distinct binders: Portland cement, slag, fly ash, silica fume, metakaolin, and calcium hydroxide. These specimens were then subjected to the concrete prism test (ASTM C1293), and their efficiency in mitigating ASR was assessed through expansion measurements. The data gathered reveal promising results from using the proposed ternary oxides approach. By comparing the effect of different portions of Al2O3, SiO2, and CaO, it was demonstrated that higher content of either Al2O3, SiO2, or both resulted in lower ASR-induced expansion development. When keeping the amount of CaO constant, mixtures with more SiO2 than Al2O3 were more effective in mitigating ASR. These results provide valuable information for making informed decisions when selecting the best options (i.e., combining different SCMs and their quantities) to apply in concrete structures where ASR development can be expected.

In recent years, long-term exposure sites have shown a disconnect between laboratory test methods and field exposure blocks. To improve standard guidance documents such as ASTM C1778 and AASHTO R80, three universities cast over 450 exposure blocks placed at 8 different exposure sites. These exposure blocks were cast at lower alkali loading levels compared to traditional exposure blocks and included traditional and non-traditional supplementary cementitious materials (SCMs) to prevent ASR. Non-traditional SCMs included natural pozzolans, reclaimed fly ash, and blended fly ashes. New test methods were evaluated as well, so that six test methods will be bench-marked against these new exposure blocks as well as historical (15–20 year) exposure blocks. This paper provides the results of a study aimed at improving the guidance currently contained in ASTM C1778 for mitigating alkali-silica reaction (ASR) in concrete. The improvement to these guidelines were drawn upon an extensive, historic database of laboratory and exposure site data. The results from this study provide recommended changes to both the prescriptive and performance-based approaches for preventing ASR.

Nano-Fibrillated Cellulose (NFC) is a new high-performance biomaterial produced from wood pulp. Recent work has demonstrated that low levels of NFC can dramatically reduce swelling due to the Alkali Silica Reaction (ASR) in concrete. A 92–97% reduction in swelling was observed with just 0.1% addition of NFC (by weight of binder).In previous work, it was demonstrated that very low loadings of NFC in concrete have exceptional abilities to reduce early age shrinkage cracking (through internal curing), reduce corrosion (through pore restructuring), and improve adhesion in repair mortars [2, 12].Using ASTM C 1260-21, samples containing 0.1 wt % NFC exhibited ~ 97% less expansion than their non-NFC-containing counterparts.It is hypothesized that this pronounced effect is due to the sequestration of alkali ions at the surface of the NFC fibrils, reduced permeability of the concrete matrix leading to reduced alkali transport, as well as through physical restraint of swelling by the nanofibrils. The very high surface area of NFC (~120–150 m2/g), along with the readily available hydroxyl and carboxyl groups on the NFC surface, is expected to contribute to the alkali sequestration effect.These results suggest that, in addition to NFC’s potent crack-reduction, corrosion prevention, and adhesion promotion capabilities, there is also an excellent case for using NFC as a low-cost highly effective additive for mitigation of ASR-induced swelling in cementitious composites.

Alkali-silica reaction continues to be a challenging durability issue for portland cement-based concrete. In general, the research community has supported industry with practical solutions based on empirically derived relationships, mostly from accelerated test methods and to a lesser extent realistic exposure/field structures. The latest research shows that both the accelerated mortar bar test and the concrete prism test, with their current limits, do not match well to high alkali loading field exposed concrete blocks. The research questions driving the development of a new approach include: 1) Are these high alkali loading blocks (e.g., 3.78 to 5.25 kg/m3 Na2Oeq) too aggressive?; 2) Would low and moderate alkali loading blocks that are representative of a broader range of concretes (e.g., pavements, bridge decks, dams, foundations) show a better correlation?; and 3) Can we develop a better approach to ASR prevention that relies more on science and our current understanding of ASR rather than pure empiricism? The research team represented in this paper is investigating a new methodology that combines the alkali availability needed to initiate ASR (aggregate specific) with the available alkali from the total cementitious blend. The alkali sensitivity of aggregates is explored using a modification to the miniature concrete prism test (MCPT), the accelerated mortar bar test (AMBT), the University of New Brunswick concrete cylinder test (UNBCCT) and the T-FAST alkali threshold test (ATT). These methods are compared, and the most efficacious method (s) will be highlighted. A newly developed alkali leaching test (ALT) will be demonstrated to assess the available alkali from the entire cementitious blend (SCM + cement or portland limestone cement). The relationship between reactivity of a supplementary cementitious material and ASR expansion is also explored. Chemical admixtures capable of reducing alkali silica reaction will also be included in this new approach.

Ternary Portland cements composed of coarse silica fume (SF), limestone (LS), and Portland cement will help to reduce the clinker factor target from 0.78 to 0.60 by 2050 with the aim to be climate neutral. Silica fume (SF) possesses pozzolanic properties that enhance mechanical strength and durability. The physical-chemical properties of the raw materials used to develop the ternary cements are studied in this work. In addition, a physical and chemical characterization of the designed mixtures is carried out considering durability criteria. Empirical information is obtained on the determination of density, setting time, hydration heat, and the potential reactivity of mixtures in relation to alkali-silica reaction. This study shows that, compared to Portland type cement (CEM I), cement pastes with higher silica fume and limestone contents have densities that are around 7% lower. On the other hand, setting time decreases in mixtures containing between 5% and 7% of silica fume, and mixtures containing less than 3% of silica fume have setting times similar to the reference cement. Likewise, the expansion caused by the alkali-silica reaction decreases significantly as we increase the content of silica fume and limestone in all cases. The use of these industrial byproducts, such as cement additions, is crucial because they improve circular economy efficiency and resource efficiency, reduce the amount of carbon dioxide produced by the cement industry, give the sector competitiveness, and improve the performance of the raw materials used in cement.

The current performance-based approach to determine the optimal supplementary cementitious materials (SCM) dosage for alkali-silica reaction (ASR) prevention in concrete requires testing at multiple replacement levels, which is time-consuming and inefficient. Moreover, the effects of soluble alkalis from certain supplementary cementitious materials (SCMs) on ASR remain undetected because of the alkali-boosted test conditions of the current tests. Thus, a chemical screening tool (CST) was developed to predict optimum SCM dosage for ASR prevention by establishing concrete pore solution alkalinity (PSA) and aggregate threshold alkalinity (THA) based criteria. The TTI Model-1, a pore solution model, was developed within CST to estimate concrete PSA, considering the combined soluble alkali effects from cement and SCMs. Soluble alkali contributions from SCMs were assessed using quantitative X-ray diffraction and water-soluble alkali methods, followed by PSA estimation through the TTI Model-1. AASHTO TP142 and ASTM C1567 tests were used to validate CST’s dosage predictions. The current study shows that CST reliably predicted dosages for both conventional (Class C & F ashes) and unconventional (blended/reclaimed ashes and natural pozzolans) SCMs within 1–2 days. The CST effectively highlighted the shortcomings of the rapid ASTM C1567 test for certain SCMs and identified SCMs that require further validation by reliable but long-term concrete ASR testing like TP 142 (90 days) and C1293 (2 years).

In 2020, the Danish concrete industry established a roadmap to reduce its CO2 emissions by 50% compared to 1990, by 2030. The plan highlighted the necessity to introduce new supplementary cementitious materials (SCMs) to face the expected shortage of coal fly ash. This paper reports the main outcomes of an industrial PhD project, which aimed to suggest an update of the Danish ASR regulations to allow the use of more SCMs without putting concrete durability at risk.The Danish regulations for binders are based on the total alkali content, with a few exemptions for well-documented materials. Since current regulations hinder the use of most alternative SCMs due to their high alkali content, this project focused on other ways to compare binders on an ASR-relevant basis. For this purpose, Cold Water Extraction (CWE) was carried out to determine the free alkali content in blended cement pastes. CWE was selected due to its relative simplicity, which makes it a promising method to be implemented by industrial stakeholders. As the study intended to be generic, various SCMs were tested to cover a wide range of material composition: coal fly ash, two calcined clays, two biomass ashes, sewage sludge ash, crushed brick and glass beads. The results were compared with ASR expansion tests (ASTM C1567, TI-B 51, AAR-10) and field exposure cubes to assess the ASR performance of the same binders.The project emphasised the need for new guidelines, better suited for screening and assessing all types of SCMs.

Several approaches have been developed to assess aggregates’ potential reactivity and the efficacy of preventive measures to mitigate alkali-aggregate reactions (AAR). Despite this, nanoparticles (NPs) of titanium dioxide (TiO2), highly effective mineral additives with hydrophilic/hydrophobic surface capabilities, have not been investigated in AAR mitigation. In this paper, an experimental program is presented to investigate the use of commercial TiO2 in both powder and colloidal suspension forms alongside two highly reactive aggregates. The powder will be incorporated as a filler (5%) into the mix, while the suspension will be applied as a protective coating to the concrete surface. To measure the effectiveness of TiO2 against AAR, part of the specimens will be subjected to artificial radiation to emulate photocatalysis conditions, while others were maintained under standardized AAR test environments. Finally, specimens will be assessed through compressive strength (CS), stiffness damage test (SDT), damage rating index (DRI), scanning electron microscopy (SEM), and porosity at different damage levels.

Deterioration mechanisms pose a significant challenge to the durability, serviceability, and safety of concrete structures. Hence, proper protection is essential to combat potential deterioration issues effectively. However, it is important to note that concrete structures are often exposed to multiple deterioration mechanisms simultaneously, which enhances the importance in searching efficient methods to protect the structures against all of these mechanisms. Since water is a common factor in many deterioration mechanisms, minimizing moisture in concrete and implementing proper waterproofing measures can be a reliable way to protect concrete. In this paper, we summarize several studies investigating the effects of a hydrophilic crystalline waterproofing admixture (CA) on concrete performance concerning corrosion, freeze-thaw cycles (FT), and alkali-silica reaction (ASR). The admixture was added during the concrete production, and the concrete specimens were examined using various test methods to evaluate their performance under different conditions. The research studies results showed that the investigated waterproofing admixture provided significant protection against corrosion by reducing its rate and preventing the initiation and propagation of corrosion-induced cracks, ultimately increasing the service life of the structures. The admixture also improved the FT resistance of concrete by reducing the damage caused by repeated freezing and thawing cycles. Additionally, the admixture could minimize ASR development, a major cause of concrete deterioration in many regions.

Alkali-silica reaction (ASR) is one of the main causes of damage in aging concrete structures, resulting in microcracking, loss of material integrity, and reduced functionality. Despite its widespread occurrence, there is still no consensus on the most effective ways to mitigate or rehabilitate structures or members affected by ASR. While various waterproofing materials have shown promise, their effectiveness is limited by the continued progression of ASR and the formation of new cracks. Self-healing waterproofing materials have been introduced recently, but their performance under physicochemical distress mechanisms is poorly understood. This study aims to evaluate the ability of three types of surface treatments, silane/siloxane high-performance penetrating sealers, crystalline waterproofing coating mixtures with self-healing properties and with or without fibers, to mitigate concrete deterioration caused by ASR in its early and moderate stages. The effectiveness of these treatments in mitigating ASR is assessed using microscopic analysis. The results indicate that surface treatments alter the kinetics of ASR and that the induced expansion results in the loss of mechanical properties and microscopic features.

Alkali-Silica Reaction (ASR) is one of the most harmful distress mechanisms affecting the durability and serviceability of concrete infrastructure worldwide. Its development leads to microcracking, loss of material integrity and functionality of the affected structure. It has been verified that some products, such as crystalline admixtures, could enhance the healing properties of concrete, thus presenting an interesting “physical” solution for durability-related distress due to ASR. However, their behaviour under the critical development of physicochemical distress mechanisms is relatively unknown. This paper aims to evaluate the influence of crystalline admixtures on ASR development mechanisms and kinetics. Therefore, this study is divided into two phases. In the first phase, concrete cylinders containing non-reactive aggregates and a commercially available crystalline admixture developed based on Krystol technology were fabricated. After 28 days of curing at 20℃ and 100% RH, pre-loaded using up to 90% of their corresponding compressive strength, then microscopic analysis was performed. Afterwards, the samples were restored at the same curing conditions for 90 days and reevaluated microscopically. In the second phase, concrete mixtures containing highly reactive coarse aggregates, the commercially available and modified versions of the CA, were fabricated, exposed to ASR development, and evaluated over time. At specific time points, microscopic tests were conducted to appraise ASR-induced development. Results show that CAs decreased ASR-induced expansion and changed its mechanism. The data suggest that the concrete's enhanced self-healing ability impacted ASR development.

Previous field investigation of concrete blocks with different ASR reactive aggregate types with and without surface hydrophobic silane impregnation in the period 2018–2023 has shown promising results. The aggregate in some of the first series of test is fast reactive as they contain a high amount (5–6 vol %) of porous calcareous opaline flint (silicified limestone). Over a period of more than 5 years, moisture content (RH) has been continuously measured inside the concrete cubes and registered whether the concrete has developed harmful expansions and/or visible cracks. These measurements have now been continued for more than 5.5 years on the exposure field on the Technical University of Denmark (DTU) near Copenhagen, Denmark.

In this study, the causes of degradation of high-strength concrete are determined using a petrological approach that incorporates polarizing microscopy, electron microscopy, and EDS analysis, and the factors that lead to the formation of a DEF-like degradation textures are clarified, taking into account the aggregate structure, constituent minerals, and compressive strength of the concrete. Precast RC box culverts for bridge and pretensioned PC hollow girders were used as observation samples. For the bridge box culverts, ASR was the cause of deterioration in the box culverts where meta-gabbro was used as the coarse aggregate, and drying shrinkage cracking was the cause of deterioration in the box culverts where serpentinite was used as the coarse aggregate. Observations showed that the PC hollow girder was primarily ASR in granite cataclasite of the coarse aggregate and chert in the fine aggregate, which contains late expansive reactive minerals. In both cases, the cracks propagated through the relatively weak aggregate interface and into the cement paste. Inside the cracks, ettringite was formed due to the high cement content, forming a microstructure similar to DEF.

A promising prognosis approach developed at the Laboratoire des Matériaux et de la Durabilité de Constructions (LMDC) based on the reactive silica consumption kinetics was applied to an hydro-electric dam located in province of Quebec, Canada. Accelerated expansion tests in immersion conditions (1N NaOH solution) were carried out on mortar specimens of three different sizes made with unreacted rock material (metagrauwacke) crushed into four different particle sizes. The optimal combination of aggregate size (600–1180 µm) and mortar bar size (252 × 25 × 285 mm) for this aggregate was determined based on expansion results measured after 546 days. Accelerated tests in similar conditions with unreacted rock material and two gradations (5–20 and 20–40 mm) of aggregates recovered from the dam’s concrete and crushed to optimal aggregate gradation showed results well aligned with the reactive silica consumption model implied in LMDC’s method. LMDC’s testing method appeared effective in consuming reactive silica. However, mechanism of consumption of the reactive silica in site conditions differs from those of the accelerated expansion tests in alkali rich immersion conditions.

The influence of boundary effects on the experimental results of concrete specimens is a phenomenon known for several decades. They are believed to be the main cause of size effects affecting physico-chemical concrete parameters assessed in the laboratory. For the specific context of hydraulic structures affected by ASR, they have a considerable impact on the predictions of numerical multi-physical simulations. This work highlights and quantifies the chemical and mechanical boundary effects that have arisen during a large experimental program to characterize the mechanical properties and ASR kinetics of a reactive mass concrete mixture used in the construction of an existing hydro-electric facility (maximum aggregate size of 76 mm). The focus is made on the fracture energy and the ASR free expansion curve, two important parameters for constitutive macro chemo-mechanical concrete models. Influence of specimen free boundary effects is highlighted on both parameters using a chemical-mechanical analogy. Also, the laser-induced breakdown spectroscopy (LIBS) with fast high-resolution scan technology is explored for the context of concrete samples affected by ASR. LIBS was found in this work in-progress study, as an interesting quantitative tool to characterize the chemical size effects.

This paper presents a new methodology to analyze data provided with the Damage Rating Index (DRI) method on field structures, which is a popular engineering tool for the assessment of concrete damage in North America. In an effort to optimize the analytical time with respect to a selected margin of error, the inherent variability of the DRI method between cores extracted in a “restricted” area (± 1 m2) was investigated. Four to six cores were assessed in seven ASR-affected component locations and thorough statistical analysis was performed. This analysis was applied down to the maximum depth available.A notable variability between cores extracted in a “restricted” area was assessed, but it significantly varied between the seven investigated locations. For instance, the average 90% confidence prediction range between cores was equal to 263 when considering their whole area. A conservative correction factor is suggested based on statistical analysis from the obtained data. This correction factor implies that assessing squares from more than one core approximately doubles the precision of the obtained DRI number. Further implications on the uncertainty of the DRI number are discussed. An optimized approach to assess the precision level of the obtained DRI number is provided.

A unique opportunity was given to groups of researchers to study the condition of a large bridge structure that has been subjected to severe service conditions for more than 50 years. A number of concrete blocks were extracted from various structural elements of the bridge (top of a pier, slabs, exposed and protected sections of a crosshead beam, sections of exterior and interior longitudinal beams). Non-destructive testing, mechanical testing (compressive strength, Stiffness Damage Test), semi-quantitative petrographic examination (Damage Rating Index) and residual expansion testing were conducted on cores extracted from those elements. This paper provides a summary of the results obtained so far from the laboratory investigations on the cored specimens.

An accelerated expansion test was carried out using concrete cores taken from outdoor-exposed experimental prestressed concrete girders which were prepared by using aggregate from the Shogawa and Joganjigawa Rivers in the Hokuriku region in Japan. The purpose of the study was to investigate the reactivity of the aggregates and the deterioration found after the outdoor exposure and then compare the findings with the results of the accelerated expansion test. The test results showed that expansion depended significantly on the presence or absence of alkali-silica reaction (ASR) at the time of core sampling.

For millennia, the medical field has utilized the sense of smell for qualitative assessment of health, but recent research shows we can tap into volatile organic compounds (VOCs), which create the odors that we perceive, for quantitative detection and analysis. In this paper, volatile organic compounds produced by the microbes in aged concrete, actively undergoing deterioration due to alkali-silica reaction were analyzed. Volatile organic compounds metabolites, an oft unused resource of chemical information that are produced by concrete-associated microbial communities, were used to detect, and characterize concrete deterioration. In this talk, preliminary results of volatile detection on long-term samples (e.g., ~ 7 years) will be provided. Scanning electron microscopy with energy disperse x-ray analysis was used to confirm deterioration mechanisms identified using volatilomics. Volatiles were analyzed using direct thermal extraction (DTE) and comprehensive two-dimensional gas chromatography - time-of-flight mass spectrometry (GC × GC–TOFMS).Identifying VOC biomarkers of concrete health will lay the groundwork for the development of sensors that can provide early deterioration warning, enabling more effective remediation/repair strategies. Microbes may also be sensitive to environmental stress (e.g., climate change), and a sensor based on microbial volatile organic compounds could be used to monitor infrastructure health and provide data to detect environmental stressors relevant to other fields.

When alkali-silica reaction (ASR) occurs in concrete structures, the residual expansion is evaluated by conducting accelerated expansion tests on sampled cores. However, there is still an issue in terms of quantitative prediction of ASR potential: when and to how much the expansion will occur. In this study, test methods to properly detect residual expansion and future prediction methods were investigated. Several cores were sampled in the early age of concrete blocks under conditions causing ASR, and expansion tests were conducted on these cores under several accelerated conditions. The observed expansion behaviors were regressed numerically on the curves. The core expansion curves were then converted to the case at outdoor temperatures using temperature correction factors obtained by concrete prism test (CPT) at different accelerated temperatures with similar concretes. The predicted results were validated by comparison with the expansion data of concrete blocks of the same mix exposed in field conditions for 6 years. Furthermore, the results were compared with previous cases of residual expansion experiments using cores taken during the accelerated stage of expansion.

Internal swelling reactions (ISR), such as the alkali-aggregate reaction, are an ongoing threat to concrete infrastructure. ISR can cause significant reductions in the mechanical properties of the concrete such as reductions in compressive/tensile strength and stiffness at relatively low expansion. Visual assessment, including the use of the cracking index (CI), is useful in assessing structures affected by ISR. However, determining the diagnosis (cause and extent of deterioration) of a concrete affected by ISR relies heavily on assessment of cores extracted from the structure using microscopic (i.e. the damage rating index - DRI) and mechanical (i.e. the stiffness damage test - SDT) assessments. In this work visual, non-destructive, microscopic and mechanical assessments are utilized on two elements from a reinforced concrete structure constructed in the early 1960’s. The two elements represent a variety of exposure conditions: a pier cap which was exposed to chlorides with the east side in close proximity to a water body, and the dry and semi-submerged sections of a pier shaft which is located in the waterbody. Data collected from visual, non-destructive testing and assessment of cores suggests high damage with the presence of mixed mechanisms, the alkali-aggregate reaction, freeze-thaw, and reinforcement corrosion.

Much interest surrounds the ability to quantitatively characterize alkali silica reaction (ASR) damage in concrete, particularly as it relates to the progression of ASR and its resulting damage pattern. Characterization methods which can provide both spatial and compositional information, then, can provide insights into the extent and progression of ASR damage in affected concrete. Micro x-ray fluorescence (µXRF) allows elemental mapping and quantification based on x-ray excitation, with spatial resolution approaching 20 µm, but to date only limited work has been published using this technique for ASR investigations. The Damage Rating Index (DRI) has been used as a semi-qualitative method using weighted scoring of concrete damage in 1 cm2 grid areas using a stereoscope to visually quantify the observed damage. Both techniques operate on similar scales (e.g., can be performed on concretes or mortars), and it is believed they can complement each other in the types of provided information. This paper is a first step in evaluating the potential benefit in combining both µXRF and DRI to investigate ASR-affected concrete with the goal of comprehensively quantifying damage in ASR-affected concrete by coupling damage features from DRI to µXRF elemental mapping. DRI and µXRF were both performed on cores retrieved from ASR-affected concrete highway (“Jersey”) barriers. Using the same referenced areas, information provided from DRI and µXRF scans are presented to showcase potential for a more comprehensive characterization of damage.

Understanding the potential for further deterioration due to internal swelling reaction (ISR) is paramount for managing ISR-affected concrete structures. Despite several studies conducted on this topic, it is still an open question when combined ISR mechanisms occur. Therefore, cores extracted from ASR+ISA-distressed concrete sleeper were subjected to conditions known to enhance the progress of ASR (1M NaOH at 38 ℃), ISA (lime water at 38 ℃), and ASR+ISA (100% RH at 38 ℃). The results reveal that the initial damage degree and leading deterioration mechanism significantly influence further damage kinetics, integrity reduction, and mechanical properties losses. A comparative evaluation of damage progression under individual and combined conditions highlights the importance of specific environmental factors for each mechanism. In conclusion, an accurate prognosis of ISR-affected concrete structures depends heavily on a comprehensive diagnosis, as the cause, the extent of deterioration, and the leading mechanism play pivotal roles in the outcomes of prognostic tests.

Physical and chemical interactions between different internal swelling reactions (ISR) play an important role in predicting future deterioration, structural implications, and management strategies of ISR-affected structures. Estimating Alkali-Silica Reaction (ASR) and Internal Sulfate Attack (ISA) contributions to the overall damage in combined attacks may support decisions on prognosis investigations (i.e., experimental and modeling). However, determining the contribution of each mechanism has been challenging. In this context, concrete infrastructure members distressed by ASR+ISA were assessed through a multi-level protocol to identify the cause and extent of induced deterioration. A novel approach was proposed to quantify the damage features generated by ASR and ISA separately. Additionally, SEM-EDS was performed to support the multi-level evaluation. The results show that the novel approach provides interesting results for estimating the influence of each mechanism in the overall deterioration process. Furthermore, it was identified that the leading mechanism may change depending on the overall degree of damage.

Automating the detection and segmentation of objects in images has been the basis of many self-driven protocols. Point-count stereomicroscopy such as the data collection procedure to calculate the damage rating index (DRI) follows such protocol in which objects (features) are detected, segmented, and counted. This work therefore presents the beginnings of a novel automated DRI while evaluating the variability and subjectivity of the method. The Mask R-CNN model trained on 110 annotated images taken from a severely damage laboratory made concrete specimen affected by alkali-silica reaction (ASR) and produced a model loss of 0.8, targeting 0.1 in this study. Further evaluation to graphically represent variability has shown that a cumulative DRI number converges towards the expected value as the sample size, in terms of number of analyzed squares, increases. Moreover, the observed occurrences and frequencies of the counted features were plotted as distributions which helps to reduce subjectivity in the result interpretation.