The use of alkali-activated slag (AAS) has been recognised for a long time as an alternative solution to Ordinary Portland cement-based binders. However, the durability of AAS-based binders is still unknown1, which limits their application in engineering practice. Carbonation is of the great interest for AAS since it induces both chemical and physical changes. The change of the pore structure after carbonation and CO2 uptake for AAS are rarely reported. This paper investigates the carbonation of AAS pastes in natural (laboratory and outdoors (unsheltered)) and accelerated conditions. The effect of the different exposure conditions on AAS pastes was studied in terms of the gel pore structure, chemical and mechanical properties.@en

The alkali-activated slag (AAS) material addresses the energy and environmental concerns associated with Portland cement production without compromise on the properties and performance1. Pore structure, as an essential component of AAS material, its volume (the total porosity), shape, size and tortuosity determine water and ionic transport. And thus, they govern the durability of AAS materials. Mercury intrusion porosimetry (MIP) has been widely used to study the pore structure in cement-based materials. However, the standard MIP test method cannot reveal the real pore structure because MIP test results show underestimation of large pores and overestimation of small pores due to ink-bottle pore effect. In order to overcome the ink-bottle pore effect, a recently developed pressurisation-depressurisation cycling mercury intrusion porosimetry (PDC-MIP) testing method was used2. In this study, the pore structures of AAS specimens activated with sodium hydroxide in terms of different dosages of Na2O and cured up to 1 year were tested using PDCMIP and compared with standard MIP. Through the PDC cycles the throat pore size distribution and ink-bottle pore size distribution were determined. The influence of Na2O dosage as well as curing age on pore structure were discussed.@en

The carbonation-induced degradation in alkali-activated materials (AAMs) is unknown. This is due to the current level of understanding of the carbonation mechanism in accelerated conditions, whereas the results from the long-term performance of AAMs in service are not available.

In this paper, the natural laboratory carbonation of alkali-activated fly ash (FA) and blast furnace slag (BFS)-based pastes was studied. The aim of experiments was to investigate the influence of the binder composition on the carbonation mechanism. The microscope techniques, i.e. polarization, fluorescent and Environmental Scanning Electron Microscope (ESEM) enabled the identification of the carbonation front and local defects due to carbonation, such as microcracks, and pore structure changes. The pH of the simulated pore solution in carbonated and noncarbonated samples was analyzed. The pore structures after carbonation were characterized in terms of capillary and gel pores by Mercury Intrusion Porosimetry (MIP) and nitrogen sorption. The effect of the change in the porosity on the mechanical properties was examined by Nanoindentation tests.

Results show that the natural carbonation did not reduce the alkalinity of the pore solution to below the pH 9. The pH was kept above 10.5 in all the mixtures. The samples with 50 wt% and less BFS content, were uniformly carbonated. The increase of the total porosity in those samples has been attributed to the simultaneous effect of the carbonation and drying shrinkage of the pastes. The change in the porosity for a consequence has a reduction in the modulus of elasticity. The samples containing more than 50 wt% of BFS were fully resistant to carbonation or carbonation was induced along the cracks as observed from the microscopic analysis.@en

Carbonation of the pore solution in alkali-activated materials (AAMs) produces alkali and/or alkali-earth carbonates. When the carbonate solubility in the water is very high (case of the most alkali carbonates), it is very hard to determine the carbonation depth in AAMs with the phenolphthalein indicator frequently used in Ordinary Portland Cement (OPC)-based materials. Carbonation gradually decreases the alkalinity of the pore solution, while the color after spraying phenolphthalein changes from colorless to pink when pH< 13 and changes back to colorless when pH< 8.2. The color change with phenolphthalein indicator may still exist in the less alkaline areas where carbonation may have already occurred. Therefore, using the indicator test is likely to underestimate the depth to which carbonation reaction has occurred in AAMs and more complete assessment is required. This study investigates the carbonation front in alkali-activated fly ash (FA) and blast furnace slag (BFS) pastes in natural laboratory conditions. Monitoring carbonation front in the samples after one year of exposure has been carried out under polarized light microscope (PLM), and environmental scanning electron microscope (ESEM). The carbonation products were sharply distinguished from the other constituents of the paste, by their crystallographic and optical characteristics under PLM, and characterized by X-Ray diffraction (XRD). @en

In this paper, carbonation resistance of alkali-activated slag (AAS) pastes exposed to natural and accelerated conditions up to 1 year was investigated. Two aspects of carbonation mechanism were evaluated. The first was the potential carbonation of the main binding phases in finely powdered AAS pastes. The second was the reactivity and diffusivity of CO2 within the bulk AAS paste. From Fourier transform infrared spectroscopy and thermogravimetric analysis coupled with mass spectroscopy time-series measurements, it was found that powdered AAS was largely carbonated within 28 days with a CO2 uptake of 14 wt%. The main carbonation products were calcium carbonates. Nevertheless, the bulk paste samples were highly resistant to carbonation, regardless of the exposure conditions. The findings showed that the pH value (initial pH[12) and strength of the samples did not decrease under accelerated carbonation compared to those of the samples exposed under natural conditions. The mineralogy of the samples in these two carbonation exposures did not alter either, except for outdoor conditions. The gel pores were dominant in the pastes (pore size in range of 2–15 nm). The dense microstructure was the main barrier for CO2 to diffuse and further react with binding phases.@en

Alkali-activation technology as an environmental friendly approach to produce construction materials can lead to a great CO2 emission reduction and provide high potential for waste reutilization. Alkali activated materials (AAMs) generally exhibit better durability and comparable or superior mechanical performance in comparison to traditional cementitious materials. However, like cementitious binders, AAMs are inherently brittle (quasi brittle). When it comes to engineering practices, the size effect also contributes to a more severe brittleness, which limits the applications in large scale structures. Fiber reinforcement is one of the solutions, which can be used to control the brittleness of AAMs. This paper presents a comparative study on the deflection-hardening behavior of ductile alkali-activated slag/fly ash (DAASF) composites reinforced with three types of synthetic fibers. All the developed DAASFs exhibited a deflection hardening with multiple cracking behavior under four-point bending tests. The influence of different fiber types, different slag/fly ash ratio as well as liquid/solid (L/S) ratio were investigated. The obtained results are a first step contribution to understanding the feasibility of using different fiber types in AAMs and development of mixture design of DAASF composites.@en

In this paper, PARC software was used to determine the spatial distribution of the dominant reaction products, their chemical composition and area/volume percentage in alkali-activated fly ash (FA) and ground granulated blast furnace slag (GBFS) pastes. The pastes reacted at different rates and dominant compositional domains were Mg rich gel around GBFS particles, i.e. Ca-(Na-)Al-Si+Mg, while the domains further form the GBFS and FA particles consisted mainly of Ca-(Na-)Al-Si. The reaction degree of GBFS and FA was calculated. Alkali-activated GBFS reacted faster in the studied alkaline conditions, reaching 66 % reaction at 28 days as GBFS contains 99 vol.-% of highly-reactive calcium and magnesium-rich aluminosilicates. The reaction degree in alkali-activated FA and GBFS paste was lower, as FA contains 31 vol.-% of nonreactive phases (quartz, mullite, hematite). The results obtained from PARC were compared with independent bulk analytical techniques X-ray fluorescence (XRF) and X-ray diffraction (XRD).@en

Pore solution of hardened alkali-activated fly ash paste was extracted by the steel-die method. The aqueous phase composition of pore solution was analyzed using ICP-OES analysis technique. The results show that the concentrations of Si, Al, Ca, K and OH- decrease with curing time regardless of the curing temperatures (40°C/60°C) and alkaline activators (sodium hydroxide with/without sodium silicate). On the contrary, the concentration of S increases with curing time. A higher temperature curing decreases the solubility of Si, Al, Ca and K in the alkali-activated fly ash, while it doesn’t show much influence on the solubility of S. The plot of the concentration of Al versus the concentration of Si displays a quasi-linear logarithmic relationship. This relationship implies congruent removal of Si and Al from the frameworks of fly ash.@en

The technology of alkaline activated materials (AAMs) has become a sustainable alternative to cement production with respect to CO2 emission and energy consumption. This technology is based on by-products usage from industrial energy and manufacturing processes, such as fly ash or blast furnace slag. The long-term performance of AAMs is decisive for their structural application. Generally, physical properties, such as porosity, pore size distribution, internal pore surface and connectivity determine material long-term behavior. Characterization of these properties in activated fly ash, slag and their blends was performed by Mercury intrusion porosimetry (MIP), Nitrogen adsorption and ESEM. Since the microstructure is greatly influenced by dissolution rate of fly ash and slag in selected activator, the pore solution was characterized by inductively coupled optical emission spectrometry (ICP-OES). Mixtures with the following fly ash and slag content (100:0, 70:30, 50:50, 30:70, 0:100 wt.%) have been prepared using sodium hydroxide/sodium silicate solutions. The increase of slag amount produces highly dense and refined microstructure, where only gel pores were identified and measured by Nitrogen adsorption. ICP-OES analysis has shown high concentrations of sodium and sulfur in solutions of slag rich mixtures, while Al, Si, Ca were almost completely absent from solution suggesting their total incorporation into reaction products of pastes. @en

A newly developed thermodynamic approach was introduced from the literature to investigate the influence of NaOH content on the hydrate assemblage and chemistry of aqueous solution of activated slag paste (Na2O/slag=4%, 6% and 8%). The ICP-OES test was performed to obtain the aqueous species profiles such as Si, Ca, Al and Na etc. The elemental concentrations in aqueous solution show a similar trend between thermodynamic calculations and experimental results. The modeling results showed that the influence of NaOH content on the uptake of Na into the primary reaction products and chemical shrinkage depends on slag reaction extent and NaOH content. The NaOH content does not show an obvious influence on the porosity development. The volume fraction of primary products in total products is between 40% and 60%, and particularly increases slightly after a certain degree of reaction for activated slag with Na2O/slag=4% and 6%. The volume ratio between total products and reacted slag is around 1.5~1.7, smaller than the value from 2.06 to 2.2 reported for Portland cement.@en

Alkali-activated materials (AAM) have received increasing interest during the past few decades due to their environmental and technological advantages. However the durability of these binders is still considered as an unproven issue that needs to be addressed before their commercial adoption and large-scale production. The present work is focussed on the investigation of durability performances of concretes manufactured through the alkaline activation of fly ash and slag blends using a low-concentrated activating solution (formulated by blending commercial sodium silicate and sodium hydroxide). The aim of this study is to provide new insights on durability properties of AAM and to assess their resistance under severe conditions. For that purpose, concrete specimens have been exposed to accelerated carbonation (varying the curing and the exposure times from 7 to 28 days). After 28 days curing, their chloride resistance has been assessed using the non-steady-state chloride migration experiments following the NordTest method (NT Built 492). Prior to durability test, the compressive strength of investigated mixtures was determined at early ages (1, 7 days) and after 28, 56, 90 days curing. The slag-rich mixtures have shown high compressive strength values (~45MPa after 1 day and ~80MPa after 28 days curing). Excellent mechanical properties have been also found for the fly ash-rich mixtures at early and late curing ages (15MPa after only 1 day and ~50MPa after 28 days curing). From carbonation test, it has been found that the carbonation depth increases as increasing fly ash/slag ratio or the exposure time. However when curing time is increased, a decrease on carbonation depth was observed. Slag-rich concretes were practically not carbonated after 7, 14 days exposure and showed 3-7mm carbonation depth after 28 days exposure. A higher carbonation depth (~18mm) was observed in fly ash-rich mixture. When the concrete specimens were exposed to chloride ingress, a high chloride permeability associated with higher concrete porosity was found for fly ash-rich mixtures with a chloride penetration depth of 18mm and chloride migration coefficient near 15-17×10-12m2s-1. These parameters decreased considerably as increasing slag content reaching values near 5mm and 2×10-12m2s-1, respectively. @en

The performances of alkali activated materials (AAM) are strongly related to their microstructures. The microstructure development is mainly affected by chemistry of the primary raw materials. In the present study the influence of different proportions of fly ash and slag (100:0, 70:30, 50:50, 30:70, 0:100 wt.%, named S0, S30, S50, S70, S100, respectively) on the microstructure and mineralogy of the alkali activated fly ash and slag based pastes has been investigated through XRD, TGA/DTGA, ESEM/EDX techniques. XRD has been used for the qualitative and quantitative analysis of different reaction products. The Rietveld refinement method and TGA/DTGA analysis have demonstrated that the increase of slag content increased the amount of amorphous phase associated with gel formation. From ESEM results, a more refined and denser matrix has been deduced in slag rich mixtures. These results are in agreement with higher strength values observed for S70 and S100 mixtures (~100 MPa at 28d and ~120 MPa after 90d curing). Depending on the fly ash to slag ratio, two types of gel have been distinguished by EDX analysis, a N-A-S-H gel in the fly ash rich mixtures and a C-A-S-H gel in the slag rich pastes.@en

This study extracted pore solution from the hardened blast furnace slag pastes activated with sodium silicate and/or sodium hydroxide through the steel-die method and analyzed the aqueous phase composition via ICP-OES technique. The results show that the concentrations of Si, Al, Na and OH" decrease with progress of slag reaction and increase with increase of Na20, while the concentration of Ca shows no substantial change. The concentrations of Na, Al and Ca are positively correlated to the concentration of OH" in a quasi-linear relationship, while the concentration of Si has a great dependence on the initial concentration of 0H'. Two regions were identified in the plots between the concentrations of Si and Ca, by which two kinds of C-S-H gels are expected. The Al/Si is linearly correlated to Na/Si for both sodium hydroxide and sodium silicate activated systems. @en

The technology of alkaline activation of calcium aluminosilicate materials, as an alternative to Portland cement to produce a clinker-free material, can address the energy and environmental concerns associated with Portland cement production. As this technology attracts more and more attention in recent years, there are increasing researches carried out on the thermodynamics of alkali-activated slag cement. However, none of the thermodynamic modeling studies were carried out in combination with the reaction kinetics of alkali-activated slag. As a result, the modeling results were expressed in term of reaction extent instead of time, which makes it difficult and inconvenient to compare with the experimental results that are usually indicated as a function of time. In order to deal with this issue in this study, the reaction kinetics of sodium hydroxide activated slag (SHAS) was studied through the measurement of heat evolution rate and quantified as a function of time using the modified Jander equation. The quantification of reaction kinetics enables the correlation between the reaction extent and time, by which the hydration of SHAS can be thermodynamically simulated in a time scale. The simulated elemental concentrations in the aqueous solution show a good agreement with the experimental results in the altering trend, and match the experimentally measured solubility data to be within ±1 order of magnitude. The phase assemblages of SHAS simulated as a function of time using thermodynamic modeling, shows no obvious dependence on the Na2O content in the range of 4% to 8%. Based on the modeled phase assemblages, some properties can be calculated as a function of time, such as chemical shrinkage, capillary porosity, etc.@en

Alkali-activated materials (AAMs) have high potential as alternative binder to ordinary portland cement (OPC), because of their high performance beside lower CO₂ emissions. While there is a general consensus about their strength advantages over OPC, there is a widespread debate regarding their durability. Some groups believe that the availability of wide scientific/technical background, together with the already-known OPC durability problems, are sufficient for their commercialization; however, others consider the durability of AAMs to be an unproven issue. This controversy represents one of the limitations facing their bulk applications. The present work provides an overview of the latest developments on durability of fly ash/slag-based AAMs with the aim to update recent findings regarding their behavior under aggressive conditions (sulfates, freeze-thaw, chloride, carbonation, acid, efflorescence). This review will provide a better understanding of the durability issues of AAMs, which will stimulate further research to develop the appropriate testing methods and help to promote their commercialization.

@en