A hybrid activator formulation combining equal parts sodium carbonate and sodium hydroxide successfully mitigates early reaction retardation in slag-based cemented paste backfill while achieving long-term mechanical strength comparable to traditional Portland cement systems. In a detailed experimental investigation published in Discover Concrete and Cement, researchers evaluated how blended alkaline activators transform the early-age hydration kinetics, microstructural development, and mechanical properties of underground mine backfill slurries. Operating under realistic high water-to-binder conditions, the team discovered that ground granulated blast-furnace slag activated with a balanced fifty-fifty mass ratio of sodium carbonate to sodium hydroxide at a ten percent total dosage achieved a 210-day unconfined compressive strength of approximately 1.6 megapascals. This structural capacity closely approached the 1.8 megapascals measured in control mixtures formulated with ordinary Portland cement. The findings establish that integrating soluble hydroxide ions directly into carbonate-activated slag binders accelerates the dissolution of latent hydraulic components, offering a viable route to overcome sluggish early curing without surrendering the cost, handling, and environmental benefits associated with carbonate-based chemistry in underground mining operations.

Underground mining operations rely extensively on cemented paste backfill to stabilize excavated subterranean voids, enhance structural mine safety, improve secondary ore recovery, and limit the substantial environmental footprint of surface tailings storage impoundments. The engineering technique involves blending mine tailings with water and a modest binder fraction to create a pumpable slurry that is routed through distribution pipelines to cure into a self-supporting solid in situ. Historically, ordinary Portland cement has served as the dominant industrial binder because of its dependable strength gain and predictable hydration reactions. Nevertheless, the manufacture of ordinary Portland cement is responsible for approximately seven to eight percent of global anthropogenic carbon dioxide emissions, alongside consuming significant thermal energy. In deep underground mines that place large volumes of paste backfill, binder procurement represents one of the largest ongoing operational expenditures. Furthermore, Portland cement backfills remain susceptible to severe chemical durability problems in aggressive underground subterranean environments, particularly where aggressive groundwater or sulfide minerals induce acid mine drainage and external sulfate attack over extended operating periods.

To overcome these environmental and economic drawbacks, researchers have increasingly investigated ground granulated blast-furnace slag as a low-carbon, latent hydraulic alternative to conventional clinker. Among various prospective activating agents, sodium carbonate has emerged as an appealing candidate because it is non-corrosive, presents fewer industrial handling hazards, costs less, and carries a much lower manufacturing carbon footprint than sodium hydroxide or sodium silicate. In spite of these operational merits, pure sodium carbonate activation exhibits severe early kinetic retardation. When calcium dissolves from the slag grains, it rapidly reacts with carbonate anions in the pore fluid to precipitate calcium carbonate phases. This early carbonation buffers the pore-solution alkalinity, preventing the rapid pH rise required to sustain aggressive slag breakdown. Consequently, the precipitation of primary calcium aluminosilicate hydrate gels is delayed, leaving early strength development inadequate for rapid mine cycling.

To resolve this fundamental chemical trade-off, the research team designed an experimental program evaluating hybrid alkaline activation combining sodium carbonate and sodium hydroxide across different activator proportions and dosages under representative field backfill conditions. Silica tailings sourced to reflect the grain-size distribution of average tailings from nine operational mines across Eastern Canada served as the mineral aggregate for all backfill mixtures. X-ray diffraction confirmed that these tailings were composed of 99.8 percent crystalline quartz by weight, with only 0.2 percent minor unidentified crystalline phases, preventing confounding mineralogical side reactions and isolating binder hydration chemistry. The ground granulated blast-furnace slag binder contained approximately 41.1 percent calcium oxide, 34.2 percent silica, 9.5 percent alumina, and 11.0 percent magnesium oxide by weight, featuring a specific surface area of 2.2 square meters per gram. Type I Portland cement was evaluated alongside the slag as a reference binder. Both anhydrous sodium carbonate and sodium hydroxide were batched as solid powders into tap water at a nominal water-to-binder ratio of 7.4.

Investigating the internal structural evolution under such dilute water contents required a multi-parameter sensor framework capable of tracking ionic activity and pore fluid reorganization without destroying the hardening specimens. The researchers cast fresh backfill slurries into large cylindrical plastic molds measuring 100 by 200 millimeters, embedding Decagon 5TE and MPS-6 sensors at their geometric centers to avoid mold boundary effects. The sensors continuously recorded pore fluid electrical conductivity, volumetric water content, and matric suction at five-minute intervals over the initial 28 days of sealed laboratory curing at 20 degrees Celsius. In parallel, the investigators prepared 120 smaller cylinders measuring 50 by 100 millimeters to track unconfined compressive strength, specific gravity via water pycnometer testing, dry density, void ratio, water content, and total porosity across curing intervals spanning 1, 3, 7, 14, 28, 90, 150, and 210 days. Tailings-free paste specimens prepared at a water-to-binder ratio of 1.0 were also cured for 150 days to enable qualitative phase identification using copper K-alpha X-ray diffraction.

The continuous sensor records revealed that the hydration of hybrid-activated slag cemented paste backfill progresses through three clearly distinct physicochemical stages. Stage one represents an initial ionization and retardation period during which electrical conductivity remains elevated while volumetric water content holds nearly constant at a plateau. In this opening regime, activator solids dissolve into the pore solution and initial ionic species release from slag particle surfaces, but water consumption is minimal because carbonate buffering temporarily restrains bulk slag breakdown. The onset of stage two marks a rapid deionization and acceleration phase characterized by a steep, synchronized drop in electrical conductivity and volumetric water content. In this stage, dissolved calcium, silicon, and aluminum ions precipitate rapidly into solid reaction gels, chemically binding free pore water into structural hydrates. Finally, stage three corresponds to matrix densification, where conductivity and water content level off into stable plateaus while matric suction climbs steadily, reflecting capillary stress development as interparticle pores refine.

The physical properties monitored over 150 days reinforced the sensor data, confirming that hybrid activation accelerates internal consolidation and tightens the backfill skeleton. While ordinary Portland cement specimens retained the lowest dry density and highest porosity throughout testing, slag specimens activated with balanced carbonate-to-hydroxide ratios densified substantially. Specifically, backfill mixtures formulated with equal proportions of sodium carbonate and sodium hydroxide at ten and fifteen percent total activator dosages consistently recorded the lowest void ratios and porosities alongside the highest dry densities. The fifty-fifty activator blend at ten percent dosage achieved the lowest stabilized volumetric water content of roughly 0.22 cubic meters per cubic meter, indicating superior chemical water binding. Furthermore, this balanced mixture generated a 28-day matric suction of approximately 50 kilopascals, the highest suction value recorded among all alkali-activated slag formulations. This capillary suction surge provided direct physical evidence that precipitation of hydration products progressively narrowed pore throats, interrupted hydraulic connectivity, and increased internal compaction.

Qualitative X-ray diffraction patterns gathered from 150-day paste samples revealed that the hybrid activation mechanism completely reconfigures the underlying phase assemblage compared to conventional Portland cement systems. The Portland cement reference paste exhibited standard reflections corresponding to portlandite, residual clinker phases, ettringite, and a diffuse background hump indicative of poorly crystalline calcium silicate hydrate gel. In stark contrast, crystalline portlandite was entirely absent from all hybrid-activated slag diffractograms. Instead, the alkali-activated pastes displayed a broad amorphous hump spanning 25 to 35 degrees two-theta, corresponding to an extensive amorphous calcium aluminosilicate hydrate gel network.

These microstructural and physical adaptations directly dictated the mechanical performance of the backfill across 210 days of compressive strength testing. At early curing ages between one and seven days, all alkali-activated slag mixtures exhibited negligible strength development due to the initial carbonate-induced kinetic retardation, whereas Portland cement controls developed early load-bearing capacity from rapid clinker hydration. However, between 7 and 90 days, hybrid-activated slag specimens experienced an accelerated surge in compressive strength, driven by extensive gel precipitation as hydroxide ions stimulated slag dissolution. Continued gel precipitation and ongoing pore refinement drove this balanced mixture to approximately 1.6 megapascals at 210 days. The researchers noted that increasing total activator dosages above ten percent failed to yield proportional mechanical gains, proving that chemical balance rather than excessive activator mass governs structural efficiency.

The authors emphasized that their findings establish an integrated mechanistic model explaining how blended carbonate and hydroxide chemistry governs slag backfill performance, while acknowledging essential experimental boundaries. Because the laboratory testing relied on qualitative X-ray diffraction, the relative proportions of amorphous binding gels could not be quantified through Rietveld refinement. Furthermore, the multi-parameter sensor arrays were monitored over 28 days, leaving very late-stage kinetic transitions unrecorded by real-time instrumentation. The empirical program also evaluated a targeted selection of activator ratios and focused exclusively on a single quartz-dominated silica tailings stream, meaning the identified fifty-fifty blend represents the optimal mixture within the tested parameter space rather than an absolute universal standard for every mine site. Nonetheless, demonstrating that hybrid alkaline activation overcomes early kinetic stagnation to produce 1.6 megapascals of compressive strength provides mining engineers with a scientifically validated framework to formulate low-carbon, slag-based backfills that simultaneously enhance geotechnical stability, reduce cement expenditures, and safely repurpose industrial mineral wastes.

Subject of Research: Hydration kinetics, microstructure, and compressive strength of hybrid-activated slag cemented paste backfill

Article Title: Hydration mechanisms and strength development of hybrid sodium carbonate–sodium hydroxide activated slag-cemented tailings backfill

Article References: Bathily, S., Fall, M., & Mohammed, A. S. (2026). Hydration mechanisms and strength development of hybrid sodium carbonate–sodium hydroxide activated slag-cemented tailings backfill. Discover Concrete and Cement, 2(1), Article 62. https://doi.org/10.1007/s44416-026-00104-1

Image Credits: AI Generated

DOI: 10.1007/s44416-026-00104-1

Keywords: cemented paste backfill, ground granulated blast-furnace slag, alkali-activated materials, sodium carbonate, sodium hydroxide, hydration kinetics, electrical conductivity, volumetric water content, matric suction, uniaxial compressive strength, calcite precipitation, hydrotalcite-like phases, C-(A)-S-H gel, mine tailings management, pore structure refinement, underground mining