Publications

A critical evaluation of the effects of mixture proportioning (binder paste content and coarse aggregate: fine aggregate ratio) on the workability, compressive strength, and tensile strength of concrete made with alkali-activated fly ash (AAF) binder was undertaken. Tensile cracking in the fly-ash-based geopolymer concrete (FGC), as influenced by its fracture properties, was studied to understand the failure of the material. Microstructural analyses were conducted using scanning electron microscopy and energy-dispersive X-ray spectroscopy to ascertain product formation. The FGC mixtures demonstrated significantly higher slump and compressive strength than concrete made with Portland cement. The slump or flow of the FGC mixtures was determined by the binder paste and alkaline solution content, and the values increased with the fine aggregate content. An increasing relationship between the coarse aggregate content in the concrete mixture and its compressive strength was identified. A similar trend was observed for the splitting tensile strength, which reached 7% of the FGC compressive strength. The high strength of the FGC was attributed to the formation of sodium aluminosilicate gel and a denser interfacial transition zone (ITZ) near the aggregate interface. The dense ITZ around the aggregate contributed to an enhanced bond between the aggregate and AAF paste, resulting in aggregate fracture. The decrease in strength with binder content was attributed to the maximum paste thickness per unit volume, which contributed to a larger number of voids per unit volume of concrete.

Partial replacement of portland cement with supplementary cementitious materials (SCMs), such as fly ash, is an effective strategy for improving durability and reducing the CO2 footprint of concrete. However, using high-volume fly ash (HVFA) binders in precast and prestressed concrete is currently limited; largely due to reduced early-age strength development that impedes rapid production and prestressing of precast concrete. To investigate and address this challenge, HVFA mortars with a minimum of 40% fly ash by mass of cementitious materials were developed and tested in this study. Two fresh fly ashes (an ASTM C618 Class F and a Class C) and a landfilled fly ash (Class F) were included. Various strategies for improving the early strength were evaluated, including gypsum optimization, chemical accelerators, steam curing, use of CSA cements, and adding other reactive SCMs like silica fume, calcined clay, and slag cement. Steam curing and the use of CSA cement at high dosages (40% of total binder) were found to be the most successful strategies across all three fly ashes. Additionally, significant improvements were observed with gypsum optimization (for Class C fly ash) and the use of accelerators (for Class F fly ashes), and these strategies are likely to be more feasible considering later-age strength and economic viability. Interestingly, HVFA mixtures made with the landfilled fly ash used in this study were able to achieve high early strengths with water-to-cementitious materials ratio adjustment alone. These HVFA mixtures were also found to be less responsive to accelerators when compared to the fresh Class F fly ash, highlighting an important distinction between the materials despite the similarity in chemical composition.

Formulating alkali-activated binders requires a clear understanding of the role of activator concentration and source material composition on product formation and strength gain. Alkaline-activation of a low-calcium fly ash binder with slag replacement at 30 and 50% by mass is evaluated. The reactivity of the binder and the compressive strength gain are evaluated using activating solutions of different NaOH molarities. Increasing the NaOH molarity produces a higher compressive strength in the activated binder. The primary reaction product formed in the activated binder is a calcium silicate hydrate with aluminium substitution (C-(A)-S-H). There is an increase in the C-(A)-S-H content, and a reduction in the porosity, with an increase in NaOH molarity. Slag contributes to early reactivity in the binder and the initial strength development while fly ash contributes to later age strength by silica enrichment of C-(A)-S-H. While increasing the NaOH molarity produces an increase in the early reactivity, the Na does not directly contribute to the reaction product formation. The Na from alkaline activator is not chemically bound to the C-(A)-S-H and can be removed by leaching in water. The increasing basicity with NaOH molarity enhances the contribution of low-calcium fly ash leading to higher silica enrichment of C-(A)-S-H.

The stability of different phases in a binder exposed to seawater is essential when assessing the performance of structures in a marine environment. The understanding of stability of alkali-activated slag upon interaction with seawater is insufficient. In this study, phase changes in alkali-activated slag systems upon interaction with seawater were investigated by employing thermodynamic calculations. The results show that C-N-A-S-H and Mg-Al LDH, the main reaction product phases in alkali-activated slag systems, are stable at lower seawater quantities and destabilize to ettringite and M-S-H with an increase in the seawater content. M-S-H, Al(OH)3 and calcite are observed to be the stable phases within the interaction mass of seawater considered here. The Ca/Si ratio of C-N-A-S-H reached a value of ∼1 at a higher seawater content regardless of the activator used. A significant increase in the volume of the system is observed during the destabilization of C-N-A-S-H. The current study improves our understanding of the spurious volume changes of different phases in alkali-activated slag systems upon interaction with seawater.

The phase changes in alkali-activated slag samples when exposed to supercritical carbonation were evaluated. Ground granulated blast furnace slag was activated with five different activators. The NaOH, Na2SiO3, CaO, Na2SO4, and MgO were used as activators. C-S-H is identified as the main reaction product in all samples along with other minor reaction products. The X-ray diffractograms showed the complete decalcification of C-S-H and the formation of CaCO3 polymorphs such as calcite, aragonite, and vaterite. The thermal decomposition of carbonated samples indicates a broader range of CO2 decomposition. Formation of highly cross-linked aluminosilicate gel and a reduction in unreacted slag content upon carbonation is observed through 29Si and 27Al NMR spectroscopy. The observations indicate complete decalcification of C-S-H with formation of highly cross-linked aluminosilicates upon sCO2 carbonation. A 20–30% CO2 consumption per reacted slag under supercritical conditions is observed. 

The assessment of the extent of carbonation and related phase changes is important for the evaluation of the durability aspects of concrete. The phase assemblage of Portland cements with different clinker compositions is evaluated using thermodynamic calculations. Four different compositions of cements, as specified by ASTM cements types I to IV, are considered in this study. Calcite, zeolites, and gypsum were identified as carbonation products. CO2 content required for full carbonation had a direct relationship with the initial volume of phases. The CO2 required for portlandite determined the initiation of carbonation of C-S-H. A continual decrease in the pH of pore solution and a decrease in Ca/Si is observed with the carbonation of C-S-H. Type II cement exhibited rapid carbonation at relatively less CO2 for full carbonation, while type III required more CO2 to carbonate to the same level as other types of cement. The modeling of carbonation of different Portland cements provided insights into the quantity of CO2 required to destabilize different hydrated products into respective carbonated phases. 

The rheology of fly ash-slag geopolymer binders is evaluated for different forms and dosages of dissolved silica in the alkaline activating solution. The addition of silica decreases the elastic resistance and the yield stress of the binder. The binder rheology is influenced by the polymeric form of silica in the activating solution determined by the silica modulus (MS = SiO2/Na2O molar ratio). There is a larger decrease in the elastic component and a larger increase in the viscosity at a lower MS. Silica addition at low MS transforms the yield behavior of the binder to viscous Maxwell flow response. The changes in the fresh state influenced by thixotropy and setting are related to the early kinetics and reactivity of the slag in the binder. The silica content and the MS in the activating solution influence reaction kinetics in the binder. While the addition of silica delays the hydration of slag, there is a higher level of early reactivity associated with gel formation from the dissolved silica at a lower MS. The early chemical reactivity produced by the silica in the activating solution influences the buildup of storage modulus (G/) and produces set. However, the influence of early reactivity in the systems on the measured increase in G/ and penetration resistance varies with silica content. While high early reactivity produces a faster setting, G/ does not scale with the kinetics. Higher silica content produces a more uniform distribution of reaction products, not producing a proportional increase in stiffness. 

An evaluation of strength-gain in binders with equal mass proportions of low-calcium fly ash and slag activated with alkaline solutions containing dissolved silica is conducted. The roles of Si and Na in the activating solution on compressive strength, reaction kinetics, and content and phase composition of reaction products are evaluated. The early reactions within the activated binder are due to the dissolution and hydration of slag. In the presence of silica, there is early precipitation of calcium silicate hydrate (CSH) before the hydration of slag. Silica in the activating solution delays the primary hydration of slag. The reactivity of silica depends on the SiO2/Na2O ratio in the activating solution. A less polymerized form of silica with higher reactivity is produced at a higher Na content in the activating solution. More reactive silica form produces a more rapid early precipitation of CSH following the initial dissolution of slag. Fly ash in the binder provides additional Si and Al to the calcium aluminosilicate hydrate (CASH) formed from the slag reaction. The Na identified in the reaction products is in a water-soluble form and not chemically bound to the CASH. The Na does not contribute directly to the compressive strength. In the long-term, the content and the silica incorporation in the CASH scales with the silica content in the activating solution and it does not depend on the form of silica. Increasing the activating solution silica content results in its larger incorporation in CASH and produces compressive strength enhancement. The form of silica has a negligible effect on the dissolution of fly ash in the binder and ultimate strength achieved.

The X-ray diffraction (XRD) signature of the glassy phase in blast furnace slag undergoing alkaline dissolution is evaluated. The intensity signature of the glassy phase present in slag can consistently be decomposed into three underlying pseudo-Voigt (PV) peaks. It is shown that the fundamental underlying characteristics of the XRD signature of the undissolved glassy phase of slag in terms of the underlying PV peaks do not change after dissolution in an alkaline solution. The stability of the calcium ions depends on the [OH−] concentration in the solution. An intensity-based procedure is developed for quantifying the unreacted glassy phase content in alkali-activated slag. The XRD profile information of the glassy phase in raw slag is suitable for fitting the intensity profile of the dissolved glassy phase. The mass percentage of the unreacted glassy phase of slag within alkali-activated slag is validated with selective acid dissolution. A procedure for determining the degree of reaction in alkali-activated slag is established. The procedure developed here could be used to determine the activity of slag in an alkaline environment. 

Slag is activated with alkali-silicate solutions made with nano-silica and waterglass. The reaction kinetics and the reaction products formed are evaluated for different silica and sodium contents. Isothermal calorimetric measurements with in-situ mixing reveal increased early reactivity with added silica, while the main hydration reaction is delayed. The presence of Na with dissolved silica in the activating solution does not influence the initial dissolution and early reactivity of slag. At the same time, it has an accelerating effect on the main hydration peak of slag. Dissolution studies of slag in alkaline solution with added silica indicate the immediate formation of calcium silicate hydrates after dissolution contributes to the early reactivity. The enhanced early reaction product formation in the presence of dissolved silica leads to the very rapid development of the internal structure. Rheological measurements indicate an early loss of workability due to increased early reactivity. The increase in time for the main hydration reaction of slag in the presence of added silica, however, delays the setting behavior. The primary reaction product in alkali-activated slag is calcium silicate hydrate with aluminum uptake, C-(A)-S-H. The addition of silica in the activating solution leads to its larger incorporation in the C-(A)-S-H and higher compressive strength. The sodium content in the activating solution does not contribute to the ultimate strength. The sodium in the alkali-activated slag is present in the form of a water-soluble amorphous product. 

The reaction of blast furnace slag in sodium hydroxide (NaOH) solutions of different molarities is evaluated. The compressive strength of the activated slag does not scale with the molarity of NaOH. The primary reaction product in the activated slag is identified with calcium aluminosilicate hydrate [C(A)SH]. While the early reactivity of slag is enhanced at higher alkalinity, and the dissolution of slag increases with the molarity of NaOH, the quantity of C(A)SH in the hydrating system does not scale with the molarity of NaOH in the activated slag. From X-ray diffraction (XRD) analysis, an additional water-soluble, sodium-based amorphous product is identified in the reaction products. The water-soluble product, which does not contribute to strength, increases proportionately with the initial Na content in the solution. At higher molarity, there is a larger proportion of the water-soluble product relative to C(A)SH in the reaction products. The Ca/Si ratio and Al/Si ratios in the C(A)SH gel are relatively invariant of the NaOH molarity in the activating solution. The compressive strength gain in the alkali-activated slag is determined by the quantities of C(A)SH and the intrinsic sodium-filled water-soluble product. 

An X-ray diffraction (XRD)-based evaluation of the crystalline and amorphous phases in slag hydrating in an alkaline environment is presented. A method is developed for the quantification of the amorphous phases present in hydrating slag in a sodium hydroxide solution. In hydrating slag, the amorphous reaction product is identified as calcium aluminosilicate hydrate. A water-soluble sodium-based amorphous reaction product is also produced. The XRD-based quantification method relies on the direct decomposition of the XRD intensity pattern of the total amorphous phase present in partially hydrated slag into the intensity patterns of the amorphous unreacted slag, the hydrate and the sodium-based product. The unreacted slag content in partially hydrated slag is also determined from the decomposition of the intensity signature of the total amorphous phase. An independent verification of the amorphous unreacted slag content in hydrating slag is obtained from measurements of blends of unhydrated and partially hydrated slag. The XRD-based phase-quantification procedure developed here provides a basis for evaluating the extent of reaction in hydrating slag. 

An experimental investigation on the fracture test response of notched concrete beams with two types of discrete macrosynthetic fibres is presented. The influence of high-modulus polypropylene macro fibres on crack propagation and opening was evaluated using the digital image correlation technique. The surface displacements measured close to the tip of the notch were analysed to evaluate the crack opening profile in concrete in relation to the observed load response. A method for precisely estimating the displacement discontinuity across the crack from measured surface displacements was developed. Post-peak softening in the flexural load response was found to be associated with crack advance in the cementitious matrix. The measured crack opening profiles in the post-cracking response indicated a hinge-type behaviour in the beam. The physical hinge in the post-cracking flexural response of the beam was directly determined from the surface measurements. It was found that the addition of macrosynthetic fibres up to 8 kg/m3 (0·9% by volume) did not significantly influence crack propagation but provided resistance to opening of the hinge, which resulted in an increase in toughness and significant load recovery in the post-peak flexural load response. 

An analytical formulation for flexure behaviour of concrete considering a multi-linear stress-crack separation (σ-w) relationship is developed using the cracked hinge model. An inversion procedure for obtaining the multi-linear cohesive stress response from the flexural load response of a beam is presented. The procedure is applied to obtain the σ-w relationship for macro-synthetic fiber reinforced concrete. An experimental investigation of the crack propagation in flexural response of macro-synthetic fiber reinforced concrete is presented using the digital image correlation technique. The post-cracking response of macro-synthetic fiber reinforced concrete during the initial softening and the subsequent load recovery is experimentally shown to be associated with a hinge-type behaviour and is produced by crack closing stresses contributed by fibers. From the optical measurements the hinge length is identified with a zone of length equal to twice the aggregate size. Using the measured hinge length, the multi-linear σ-w relationship for macro-synthetic fiber reinforced concrete obtained by matching the experimental and the analytical load responses exhibits a stress recovery following initial softening. The cohesive stress subsequently decreases following the recovery at large crack separation. The crack closing stresses contributed by the pullout of fibers produce stress recovery in the σ-w relationship and are primarily active after the formation of the hinge resulting in significant contribution to fracture energy at large crack openings. There is a good correlation in the fracture energy obtained from load response and the σ-w relationship at different values of crack opening displacements.