Stepwise Recycling of Red Mud Through Electrochemical Activation for Enhanced Cementitious Performance and Magnetically Separable Iron Recovery

Abstract

Red mud, a major solid waste from the alumina industry, suffers from an extremely low utilization rate due to its high alkalinity, complex chemistry, and particularly low cementitious activity, which drives the need for novel activation strategies. This study presents a new method for red mud activation through electrochemical treatment, which simultaneously enables iron recovery as a valuable by-product. The electrochemical activation was systematically investigated by performing experiments in alkaline, neutral, and acidic electrolytes. The alkaline system showed a pronounced enhancing effect on the electrochemical process. Under alkaline conditions, the average Faradaic efficiency exceeded 80%. The electrochemical treatment modified the microstructure of red mud particles and transformed iron oxides into magnetic species, which could be effectively separated via magnetic separation. More importantly, this activation process significantly enhanced the cementitious activity of the treated red mud by removing iron oxide that encapsulates reactive aluminosilicate phases and increasing surface reactivity. When used as a supplementary cementitious material with ordinary Portland cement and gypsum, the electrochemically activated red mud demonstrated remarkably improved mechanical properties, with 28-day compressive strength reaching up to 69 MPa. Characterization analysis revealed that the electrochemical activation promoted the formation of key hydration products, including C-S-H gel (formed through both OPC hydration and pozzolanic reactions between activated red mud and portlandite), ettringite, and portlandite. This work provides a green and low-carbon pathway for the stepwise utilization of red mud through activation and resource recovery.

1. Introduction

Red mud is a highly alkaline solid waste generated during the extraction of alumina from bauxite via the Bayer process in the alumina industry. As of 2025, the global annual production of red mud has reached approximately 200 million tons, with a cumulative historical stockpile of about 5 billion tons. This solid waste occupies extensive land resources. Its high alkalinity and heavy metal content are readily mobilized through rainwater leaching, causing persistent contamination of soil and water bodies and thereby imposing severe pressure on the global ecological environment. However, despite such massive generation volumes, the comprehensive utilization rate of red mud remains extremely low, with the current global average below 5% [1]. The fundamental challenge in its valorization lies in its complex physicochemical properties. Due to the low hydraulic activity of red mud, it contributes only limited cementitious strength when used directly as a cement supplementary material [2]. On the other hand, the high iron content (typically 30–50%) results in high energy consumption and poor economics of conventional treatment processes, constituting a critical bottleneck for large-scale application.

Current research on red mud valorization primarily focuses on two directions: building material applications and the recovery of valuable metals [3,4,5]. In building material applications [6,7], red mud as a supplementary cementitious material can achieve compressive strengths exceeding 60 MPa under alkali-activated conditions [8] and improve the mechanical properties of concrete [9,10]. However, due to its inherently low hydraulic activity, high dosages of alkali activators are required, leading to increased costs and limiting industrial-scale implementation. Red mud contains abundant valuable metals, particularly iron (primarily as hematite), making metal recovery an economically attractive approach. In terms of valuable metal recovery [10,11,12,13,14,15,16], conventional processes such as carbothermal reduction, acid leaching, and magnetic separation have established mature technology systems [17,18,19] but universally suffer from high energy consumption and high carbon emissions. More critically, existing studies predominantly focus on valuable metal extraction while neglecting the valorization potential of the post-extraction residue. Metal extraction processes like thermal reduction and acid leaching alter the mineral structure and chemical composition of the residue, weakening its cementitious activity. Consequently, residues rich in calcium, silicon, and aluminum components become difficult to utilize further. Therefore, developing technological pathways capable of achieving synergistic transformation of all components in red mud and promoting stepwise and efficient utilization of each component has become critical for achieving large-scale valorization of red mud.

In recent years, electrochemical reduction has emerged as a green and efficient separation method, demonstrating significant potential in the field of solid waste valorization [20,21,22]. This technology applies a specific reduction potential to induce electron transfer reactions of target metal oxides at the cathode, enabling their selective reduction and separation. Compared to conventional high-temperature pyrometallurgical processes, electrochemical reduction can operate under mild conditions, effectively avoiding the high energy consumption and carbon emissions associated with calcination [23]. Currently, a relatively comprehensive theoretical framework for electrochemical metallurgy in oxide reduction has been established. Molten salt electrolysis systems, utilizing novel anode materials, have achieved efficient oxide electrolysis with current efficiency and product purity meeting industrial application standards [24]. Meanwhile, aqueous systems have established correlations among pH, potential, and phase, allowing for precise control of the reaction process [25]. These advancements fully highlight the notable advantages of electrochemical methods, including high selectivity, mild operating conditions, and process controllability.

However, existing research has predominantly focused on systems with single-component iron oxides [25], leaving the electrochemical behavior of complex multi-phase systems like red mud poorly understood. The high alkalinity of Bayer-process red mud can alter the double-layer structure at the electrode/electrolyte interface, thereby affecting electron transfer kinetics [26,27,28]. Simultaneously, the interlocked microstructure, in which iron oxides are encapsulated within aluminosilicates [13], increases charge transfer resistance. The coexistence of multiple reducible components also readily triggers competitive reactions, reducing the selectivity for the target product. Furthermore, the impact of electrochemical treatment on the mineral phase transformation of red mud and the activation mechanism for its cementitious activity [7] remains unclear, which limits the application potential of this technology for the full-component utilization of red mud [29]. Therefore, systematically investigating the electrochemical reduction mechanisms in the red mud system, elucidating the interaction mechanisms among its multiple components, and deeply exploring the influence of electrochemical treatment on the cementitious properties of red mud hold significant scientific importance and practical value.

Based on the current state of research, this study proposes a synergistic approach that couples electrochemical activation of red mud with iron extraction, thereby enabling efficient utilization of the activated residue and recovered iron and exploring new pathways for full-component valorization of red mud. Electrochemical reduction experiments on red mud were conducted in alkaline, neutral, and acidic electrolyte systems to systematically investigate the effects of pH, ionic concentration, and coordination environment on the reduction efficiency of iron oxides and on Faradaic efficiency. The kinetic characteristics of reductions in different systems were clarified through these investigations. The electrochemically treated red mud was further utilized in cement-based materials to evaluate its cementitious performance. The focus was on analyzing the regulatory effects of the electrochemical process on the mineral composition and microstructure of red mud, particularly the evolution of aluminosilicate phase reactivity after iron oxide removal. Through comprehensive characterization, the underlying mechanisms linking electrochemical reduction to the activation of red mud cementitious properties were revealed.

2. Materials and Methods

2.1. Materials

The Bayer red mud used in this study was obtained from Shandong Aluminum Co., Ltd. (Zibo, China). P·O 52.5 ordinary Portland cement (OPC) and gypsum were commercially available products. The chemical compositions of raw materials are presented in

Table 1. Oxide composition of raw materials (wt.%).

Materials	Mass Fraction/%
	Fe2O3	Al2O3	SiO2	Na2O	TiO2	CaO	MgO	SO3	Other	L.O.I. *
Red mud	32.42	21.24	17.37	11.76	5.96	1.02	0.13	0.31	0.83	8.96
Cement	4.73	3.41	13.23	N.D. #	0.32	71.30	1.09	2.84	0.65	2.43
Gypsum	0.23	0.68	2.31	0.08	0.04	32.85	3.65	35.87	0.4	23.89

#: Not detected. *: L.O.I. loss at 1100 °C.

. The mineralogical composition and particle size distribution of raw materials are shown in Figure 1.

2.2. Electrochemical Reduction Process

Electrochemical experiments were performed using a three-electrode Pyrex glass cell (500 mL) equipped with a CS electrochemical workstation (Wuhan Corrtest Instruments Co., Ltd., Wuhan, China). The working electrode was a 316 stainless steel mesh (16 cm²), the counter electrode was a platinum–titanium mesh (30 × 30 mm), and the reference electrode was a saturated Hg/HgO electrode (E = +0.098 V vs. NHE at 25 °C). For each experiment, 40 g of red mud was dispersed in 400 mL of electrolyte solution and subjected to electrochemical treatment under constant magnetic stirring (300 rpm) at a controlled temperature.

To investigate the effect of electrolyte type on electrochemical reduction of red mud, comparative experiments were conducted using alkaline (NaOH), neutral (NaCl), and acidic (H₂SO₄) electrolyte systems. The experimental groups were classified based on electrolyte pH: the neutral group (EW-0) used deionized water (pH ≈ 7), the alkaline groups (EN-1 to EN-4) employed NaOH solutions at concentrations of 1, 2, 3, and 4 mol/L, respectively, and the acidic groups (EH-1 to EH-4) used 0.1 mol/L H₂SO₄ solution at volumes of 10, 30, 50, and 70 mL, respectively, added to 400 mL deionized water (resulting in progressively lower pH values). The specific electrolyte compositions are listed in

Table 2. Formulation design of electrolytic solutions for red mud treatment.

Group	Number	Deionized Water	Red Mud	NaOH	H2SO4 Solution (0.1 mol/L)
Neutral	EW-0	400 mL	40 g	/	/
Alkaline	EN-1			1 mol/L	/
	EN-2			2 mol/L
	EN-3			3 mol/L
	EN-4			4 mol/L
Acidic	EH-1			/	10 mL
	EH-2				30 mL
	EH-3				50 mL
	EH-4				70 mL

. All reagents were of analytical grade and purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China).

2.3. Cementitious Material Preparation from Treated Red Mud

Following electrochemical reduction, the systematic separation and collection of products was performed. Magnetic separation was employed to recover the magnetic iron-bearing components from the electrolyzed slurry. The magnetic fraction was thoroughly washed with deionized water, dried at 40 °C for 12 h, and reserved for subsequent characterization. The remaining slurry after magnetic separation was then centrifuged at 3500 rpm for 15 min. The obtained solid was dried to constant weight at 40 °C to yield the electrochemically treated red mud.

The mass recovery of the magnetic product was determined gravimetrically. The phase composition and morphology of both the magnetic product and the electrochemically treated red mud were characterized using X-ray diffraction (XRD) and scanning electron microscopy (SEM). Finally, the electrochemically treated red mud was utilized as the primary constituent to prepare red mud-based cementitious materials (as shown in Figure 2) according to the mix proportions presented in

Table 3. Mix proportions of red mud-based hardened pastes.

Sample	Treated Red Mud	OPC	Gypsum	W/C
Red mud-based pastes	8%	89%	3%	0.35

[2,7]. Specimen preparation and curing conditions: The dry powders (treated red mud, OPC, and gypsum) were dry-mixed for 3 min, then water (W/C = 0.35) was added, and the specimen was mechanically mixed for 5 min at low speed (140 ± 5 rpm), followed by 2 min at high speed (285 ± 10 rpm). The fresh paste was cast into 20 mm × 20 mm × 20 mm cubic molds and vibrated for 2 min to remove air bubbles. After 24 h at room temperature (20 ± 2 °C), specimens were demolded and transferred to a standard curing chamber (20 ± 2 °C, relative humidity ≥ 95%) in accordance with GB/T 17671-2021. Specimens were cured for 3, 7, and 28 days before testing. Three parallel specimens were tested at each age, and the average value was reported. The mechanical properties and hydration mechanisms of these materials were then investigated.

2.4. Test Procedure and Analysis

2.4.1. Electrochemical Treatment Process of Red Mud

The electrochemical reduction of iron oxides in red mud requires precise control of electrode potential to achieve selective reduction. The solution resistance of the red mud suspension was first measured dynamically, and the IR compensation mechanism was activated to eliminate potential drift caused by solution resistance. This ensured that the applied potential acted accurately at the electrode interface, with potential fluctuations controlled within ±0.03 V after compensation.

Subsequently, cyclic voltammetry (CV) was employed to identify the characteristic reduction peak potential of iron oxides. The CV scan was performed from −2.0 to +2.0 V vs. Hg/HgO at a scan rate of 10 mV/s, providing parametric guidance for subsequent potentiostatic electrolysis.

Based on the reduction peak potential determined by CV, potentiostatic electrolysis was conducted for 180 min using chronoamperometry. Continuous magnetic stirring at 300 rpm was maintained to ensure uniform dispersion of red mud particles. The progress of the reaction was monitored via the current–time response. Electrolysis was terminated when the current decayed to 95% of the steady-state value, achieving an optimal balance between reduction efficiency and energy consumption.

2.4.2. Calculation of Faradaic Yield

Faradaic efficiency (FE) is a critical parameter for evaluating charge utilization in electrochemical reactions and is defined as the ratio of charge consumed by the target reaction to the total charge passed. The electrochemical reduction of iron oxides in red mud can be described by a stepwise reaction mechanism. Under cathodic polarization, Fe₂O₃ undergoes partial reduction to Fe₃O₄:

The electrochemical reduction of iron oxides in red mud can be described using a stepwise reaction mechanism. Under cathodic polarization, Fe₂O₃ undergoes partial reduction to Fe₃O₄:

3Fe₂O₃ + 2H⁺ + 2e⁻ → 2Fe₃O₄ + H₂O, E° = −0.047 V vs. SHE

(1)

which is subsequently reduced further to metallic Fe:

Fe₃O₄ + 8H⁺ + 8e⁻ → 3Fe + 4H₂O, E° = −0.085 V vs. SHE

(2)

Meanwhile, the hydrogen evolution reaction (HER) is a parasitic side reaction:

2H₂O + 2e⁻ → H₂ + 2OH⁻, E° = −0.828 V vs. SHE

(3)

This process will compete with iron oxide reduction at the cathode surface for electrons, thereby affecting the Faradaic efficiency.

The overall charge balance can be expressed as:

$$\mathrm{Qtotal} = {\int }_0^{\mathrm{t}}\mathrm{I}\left(\mathrm{t}\right)\mathrm{dt}$$

(4)

where I(t) is the current as a function of time (A), t is the electrolysis time (s), and Q_total is the total charge (C).

The reaction selectivity primarily depends on the electrode potential, current density, and solution pH. Under highly reductive conditions (E < −1.0 V vs. SHE), a complete reduction in iron oxides becomes the dominant reaction.

Since the magnetic products are oxidized to Fe₂O₃ upon exposure to air, the composition of the original reduced products must be deduced. By assuming two limiting cases (entirely Fe₃O₄ or entirely Fe), a reasonable range for the Faradaic efficiency can be established, as shown in the following equations [26].

Assuming that the product is entirely Fe₃O₄ (conservative estimation), the Faradaic efficiency (lower bound) is:

$$\left[\mathrm{m}\right({\mathrm{Fe}}_2{\mathrm{O}}_3,\mathrm{measured}) \times 1.247]/\mathrm{Qtotal} \times 100\%$$

(5)

Assuming that the product is entirely Fe (optimistic estimation), the Faradaic efficiency (upper bound) is:

$$\left[\mathrm{m}\right({\mathrm{Fe}}_2{\mathrm{O}}_3,\mathrm{measured}) \times 3.632]/\mathrm{Qtotal} \times 100\%$$

(6)

The actual Faradaic efficiency lies between the lower and upper bounds.

2.4.3. Test Methods

Compressive strength tests were conducted on specimens cured for 3, 7, and 28 days under standard conditions using a TYE-3000 compression testing machine (Jinan Testing Equipment Co., Ltd., Jinan, China) at a loading rate of 1 kN/s.

X-ray diffraction (XRD, D8 Advance, Bruker AXS GmbH, Karlsruhe, Germany) analysis was performed on the electrochemically treated products (magnetic iron powder and electrolyzed residue) and red mud-based cementitious materials at different curing ages. The scans were conducted from 5° to 70° (2θ) with a step size of 0.02° and a scan rate of 5°/min.

The hydration heat evolution of the cementitious materials was monitored using a TAM Air isothermal calorimeter (TA Instruments, New Castle, DE, USA). Paste samples (15 ± 0.2 g) were placed in 20 mL ampoules with 7 g of deionized water as a reference. Heat flow and cumulative heat release were continuously recorded at 25 °C for 10 days.

Thermal analysis of the hardened pastes was performed using a simultaneous thermal analyzer (Netzsch STA 449 F3, NETZSCH-Gerätebau GmbH, Selb, Germany). Samples were heated from 30 °C to 1000 °C at a heating rate of 10 °C/min under a nitrogen atmosphere.

Fourier-transform infrared spectroscopy (FTIR, PerkinElmer Ltd., Beaconsfield, UK) was employed to analyze changes in functional groups in the samples before and after hydration. Spectra were collected in the 400–4000 cm⁻¹ range with 32 scans.

The microstructure of red mud before and after electrochemical treatment, magnetic products, and hardened pastes was examined using scanning electron microscopy (SEM, QUANTA FEG 450, FEI Company, Hillsboro, OR, USA). Samples were coated with platinum prior to imaging to enhance conductivity. Imaging was performed at an accelerating voltage of 15 kV with a working distance of 10 mm. Energy-dispersive X-ray spectroscopy (EDS) was coupled with SEM for elemental analysis.

3. Results and Discussion

3.1. Characterization of Raw Materials

Prior to electrochemical treatment, the raw materials were characterized to establish baseline properties. XRD analysis (Figure 1a) reveals that the primary phases in red mud are cancrinite, hematite, kaolinite, muscovite, and quartz. The presence of hematite (α-Fe₂O₃) as the main iron-bearing phase is consistent with typical Bayer process red mud [13]. The particle size distributions of raw materials are shown in Figure 1b. The measured specific surface area of red mud was 4.68 m²/g, with a surface area mean diameter of 1.282 μm and a volume mean diameter of 19.359 μm. For P·O 52.5 OPC, the specific surface area was 1.55 m²/g, with a surface area mean diameter of 3.862 μm and a volume mean diameter of 16.079 μm. Gypsum exhibited a specific surface area of 0.754 m²/g, with a surface area mean diameter of 7.955 μm and a volume mean diameter of 43.127 μm. These particle characteristics are important for understanding the subsequent electrochemical behavior and cementitious performance, as smaller particle sizes and higher specific surface areas generally enhance reactivity.

3.2. Mechanism of Electrochemical Reduction in Red Mud

Figure 3 presents the cyclic voltammograms (CV) of red mud electrochemical reduction under different pH conditions. Under acidic conditions (Figure 3a), the reduction potential shifted negatively to −1.32~−1.75 V with increasing H₂SO₄ concentration. In contrast, alkaline conditions (Figure 3b) exhibited a significant promotional effect, with the reduction current density increasing substantially compared to neutral and acidic conditions. The distinct CV characteristics between alkaline (Figure 3b) and acidic (Figure 3a) systems reflect fundamentally different electrochemical mechanisms. In an acidic electrolyte, iron exists predominantly as Fe³⁺ cations, and the reduction process is accompanied by significant hydrogen evolution, resulting in lower current efficiency and more negative reduction potentials. Conversely, the alkaline system features anionic iron complexes ([Fe(OH)₄]⁻ and FeO₂⁻) with more favorable reduction kinetics.

The superior performance of the alkaline system arises from multiple synergistic mechanisms. First, OH⁻ ions facilitate the dissolution of Fe₂O₃ to form [Fe(OH)₄]⁻ complex species, thereby enhancing the electrochemical activity of reactants [27]. Second, the alkaline environment significantly reduces the overpotential for the stepwise reduction pathway of Fe³⁺→Fe²⁺→Fe. Furthermore, elevated ionic strength improves both electrolyte conductivity and mass-transfer kinetics. In contrast, under acidic conditions, competitive hydrogen evolution and surface passivation substantially limit the reduction efficiency of iron oxides [28]. Consequently, alkaline electrolyte conditions were selected as the optimal system for iron recovery due to high Faradaic efficiency exceeding 80%, as demonstrated in subsequent potentiostatic electrolysis.

The current–time curves obtained via chronoamperometry are shown in Figure 4. Although all nine experimental groups are presented, the alkaline group curves (EN-1 to EN-4) appear to overlap due to their relatively small differences in current density compared with variations among electrolyte types. These groups exhibit similar steady-state current densities and current decay behavior, suggesting comparable mass transfer characteristics under alkaline conditions. However, this apparent overlap does not imply identical electrochemical performance. As shown in

Table 4. Faradaic efficiency of different experimental groups.

Faradaic Efficiency (%)	EW-0	EH-1	EH-2	EH-3	EH-4	EN-1	EN-2	EN-3	EN-4
Min	0.05	0.34	0.06	0.15	0.03	7.57	14.09	45.88	78.74
Max	0.14	0.98	0.16	0.43	0.09	22.04	41.04	100	100

, the Faradaic efficiency increases markedly from EN-1 (7.57–22.04%) to EN-4 (78.74–100%), indicating that higher NaOH concentrations enhance the selectivity toward iron oxide reduction while suppressing parasitic hydrogen evolution, even though the differences in total current are relatively subtle and visually indistinguishable in Figure 4. All experimental groups exhibited rapid current decay during the initial stage of electrolysis, attributed to concentration polarization caused by the rapid depletion of Fe³⁺ at the electrode surface. As the surface iron ion concentration decreased, the reaction gradually transitioned to diffusion control following Fick’s law, and the current stabilized. The steady-state current reflects the mass transfer characteristics of different electrolyte systems. Acidic conditions (EH groups) maintained higher steady-state currents by preserving high Fe³⁺ solubility and electrochemical activity at low pH [30]. In contrast, alkaline conditions (EN groups) exhibited relatively lower steady-state currents, primarily because Fe³⁺ forms Fe(OH)₃ precipitates at high pH, reducing the free iron ion concentration [27].

The total charge was obtained by integrating the current–time curves. The actual moles of reduced iron were calculated according to Faraday’s law, enabling quantitative evaluation of the reduction efficiency of each system [31]. The results are summarized in Table 4. The calculated Faradaic efficiency (FE) demonstrates that electrochemical treatment conditions significantly affect the reduction efficiency of iron in red mud. The neutral electrolyte group (EW-0) exhibited relatively low FE values (0.05–0.14). Among the acidic groups (EH), only EH-1 showed high FE (0.34–0.98), while the others remained low. The FE of alkaline groups (EN) increased markedly with increasing NaOH concentration, with the EN-4 group achieving an average FE exceeding 80%. These results indicate that strongly alkaline environments effectively suppress side reactions such as hydrogen evolution, thereby significantly improving charge utilization efficiency in the electrochemical reduction of iron oxides and favoring efficient iron recovery from red mud.

Based on the analysis described above, the electrochemical reduction mechanism of red mud was elucidated (Figure 5). In alkaline electrolyte, iron oxides in red mud (primarily Fe₂O₃) partially dissolve to form soluble iron species such as FeO₂⁻ and [Fe(OH)₄]⁻. Upon application of a reductive potential, these dissolved iron species and solid-phase Fe₂O₃ undergo stepwise electrochemical reduction at the working electrode (cathode) via the Fe₃O₄ intermediate phase, ultimately yielding metallic Fe. Simultaneously, the oxygen evolution reaction (OER) occurs at the counter electrode (anode), where OH⁻ ions are oxidized to O₂ and released into the solution. Under alkaline conditions, the elevated OH⁻ concentration facilitates both the dissolution of iron oxides and electron transfer, while effectively suppressing cathodic hydrogen evolution. This enhances Faradaic efficiency, consistent with experimental observations that EN groups exhibited higher FE at elevated NaOH concentrations.

3.3. Characterization of Treated RM

After 3 h of electrolytic reduction, the products obtained via magnetic separation were dried, weighed, and characterized by XRD (Figure 6). The XRD pattern reveals that the primary phase of the magnetic product is hematite (α-Fe₂O₃), which is likely attributed to re-oxidation of the reduced products during the drying process. Although the final product remains in the form of iron oxide, its magnetic properties confirm that the electrolytic reduction process indeed altered the iron speciation in red mud. This provides a viable pathway for subsequent iron recovery through magnetic separation.

After 3 h of electrolytic reduction, magnetic materials were separated from the red mud slurry using a magnetic rod. The remaining slurry was centrifuged and dried to obtain the treated red mud. Figure 7 shows the XRD patterns of red mud after electrochemical iron removal. As shown in Figure 7a, with increasing H₂SO₄ concentration in the electrolyte, the characteristic diffraction peaks of muscovite gradually weakened, whereas the diffraction peak intensities of magnetite and quartz exhibited an overall increasing trend. This indicates that H₂SO₄ reacts with muscovite under acidic conditions, disrupting its crystal structure and forming soluble salts. Simultaneously, electrochemical reduction progressively converts Fe₂O₃ to Fe₃O₄. As shown in Figure 7b, the characteristic diffraction peak intensities of both muscovite and kaolinite gradually decreased with increasing NaOH concentration. This is likely attributed to the strong alkaline environment, which facilitates the dissolution of muscovite and kaolinite. Their layered aluminosilicate structures are disrupted, with aluminum and silicon components converting to soluble aluminates or silicates in the solution [32].

The significant reduction in crystalline phase peak intensities (muscovite, kaolinite, and cancrinite) after electrochemical treatment indicates a decrease in crystalline content and a corresponding increase in amorphous material. The electrochemical process disrupts the long-range crystalline order of these aluminosilicate minerals, converting them into disordered, amorphous phases with enhanced reactivity. This amorphization is particularly pronounced in the EW-0, EH-4 and EN-4 groups, where the peak intensities of muscovite and kaolinite show greater reduction compared to other groups. The increased amorphous content provides more reactive Si and Al species for subsequent pozzolanic reactions, which contribute to the improved cementitious performance observed in these groups.

Figure 8 presents the microstructure of red mud and electrolytic products after electrochemical treatment. EDS analysis shows that the magnetic powder is composed of iron oxides with high Fe and O content. Successful magnetic separation confirms the formation of magnetic iron species through electrochemical reduction, as the original hematite (α-Fe₂O₃) in red mud is non-magnetic or weakly magnetic and cannot be effectively separated magnetically. This indicates that electrochemical reduction successfully converted hematite to strongly magnetic species such as magnetite (Fe₃O₄) or metallic Fe, demonstrating the effectiveness of the electrochemical process for iron mineral transformation and recovery.

Comparison of the microstructure between raw red mud and electrochemically treated groups (EW-0, EH-4, EN-4) reveals that the surface of raw red mud is populated with numerous fine particles (bright regions in SEM images). According to EDS spot analysis, these bright regions exhibit significantly higher Fe content than the surrounding matrix, corresponding to iron oxide phases. After electrochemical treatment, the visible abundance of these iron oxide particles decreased in all three groups (EW-0, EH-4, and EN-4), and the particle surfaces became noticeably rougher. Notably, some iron mineral particles that were not removed by magnetic separation remained in the EN-4 group. This indicates that the EN-4 group produced the largest amount of magnetic iron oxides through electrochemical reduction, demonstrating the most pronounced reduction effectiveness.

3.4. Hydration Mechanism of Treated RM-Based Cementitious Materials

3.4.1. Mechanical Properties

Figure 9 presents the compressive strength of cementitious materials prepared from red mud treated with different electrolyte systems at various curing ages. The untreated control group (REF) exhibited 3-day and 28-day strengths of 59.8 MPa and 63.9 MPa, respectively, with a strength gain of only 6.9%, indicating low hydraulic activity of raw red mud. In contrast, neutral (EW-0) and acidic (EH-4) electrolyte treatments achieved optimal mechanical performance, with 28-day compressive strengths reaching 68.9 MPa and 69.1 MPa, respectively, representing significant improvements over the control group.

Electrochemical treatment markedly improved the mechanical properties of red mud-based cementitious materials. The neutral electrolyte treatment group (EW-0) achieved a 3-day strength of 56.2 MPa and a 28-day strength of 68.9 MPa, demonstrating optimal strength development. Previous studies have shown [33] that layered aluminosilicates undergo structural decomposition under acidic conditions, releasing reactive SiO₂ and Al₂O₃ components. These components participate in subsequent hydration reactions to form C-S-H gel, thereby enhancing mechanical performance. The acidic electrolyte treatment group (EH-4) reached a 28-day strength of 69.1 MPa, comparable to the EW-0 group. Both groups exhibited high rates of strength gain, indicating that electrochemical treatment improved the hydration process in the later stages. These compressive strengths (68.9–69.1 MPa) are comparable to or exceed those of alkali-activated red mud systems (typically 40–60 MPa) [8], demonstrating that electrochemical activation effectively enhances cementitious performance.

The alkaline electrolyte treatment group (EN-4) showed 3-day and 28-day strengths of 47.4 MPa and 59.0 MPa, respectively, with relatively lower absolute strength values but a strength gain of 24.5%. This suggests that excessively high alkalinity may inhibit hydration product formation, thereby affecting the development of strength. However, combined with the Faradaic efficiency analysis presented earlier, the EN-4 group demonstrated significant advantages in iron recovery. Therefore, practical applications require a balanced consideration between iron recovery efficiency and the mechanical performance of the cementitious material. Overall, red mud treated with neutral and acidic electrolyte systems achieved high 28-day strength while maintaining adequate early-age strength, showing substantial performance improvements compared to the control group. Although the alkaline system exhibited slightly inferior mechanical properties, it offers higher iron recovery efficiency. The appropriate electrolytic conditions can be selected based on specific application requirements.

3.4.2. pH Value

Table 5. pH values of red mud in different groups.

Group	REF	EW-0	EH-4	EN-4
pH Value	10.45	10.35	10.07	12.65

presents the pH values of red mud after electrochemical treatment. The EN-4 group exhibited the highest pH values, significantly higher than other groups. The EW-0 and REF groups showed comparable pH values, whereas the EH-4 group exhibited the lowest pH. The elevated pH of the EN-4 group originates from residual alkaline species following treatment with 4 mol/L NaOH. As shown in Figure 7b, high NaOH concentration induced the dissolution of minerals such as cancrinite and muscovite. Table 1 shows that the original red mud already contains 11.76% Na₂O. After EN-4 treatment, the pH reached 12.65, indicating a substantial increase in system alkalinity. Studies have demonstrated that excessive alkali content in cement-based materials adversely affects long-term strength [34].

The EH-4 group exhibited the lowest pH due to the consumption of alkaline components in red mud by H₂SO₄ treatment. Figure 7a demonstrates that acid treatment reduced the diffraction peak intensities of alkaline minerals such as muscovite. The pH value of the EW-0 group was comparable to that of the REF group, indicating that neutral electrolyte treatment had minimal effect on red mud alkalinity. Red mud with different pH values affects the overall alkalinity of the cement-based system when subsequently used as a supplementary material. The 28-day strength results in Figure 9 demonstrate that the EW-0 and EH-4 groups (pH 10.07–10.35) achieved the highest strengths (68.9 and 69.1 MPa), while the EN-4 group (pH 12.65) exhibited the lowest strength (59.0 MPa). This is primarily attributed to excessive alkalinity, which, although accelerating early-age hydration kinetics and degree of hydration, causes a more rapid decline in internal relative humidity [35]. This leads to faster saturation of larger pores, thereby restricting further hydrate growth and consequently reducing strength. These results demonstrate that treatment of red mud with electrolytes at appropriate pH values can achieve superior cementitious performance.

3.4.3. Hydration Process Analysis

Figure 10 shows the hydration heat evolution curves of cementitious materials prepared from red mud pretreated with different electrolyte systems. The first exothermic peak corresponds to the hydration of cement clinker minerals, while the second exothermic peak corresponds to the pozzolanic reaction of reactive aluminosilicate components in red mud.

As shown in the heat flow curves (Figure 10a), compared to the REF group, all electrochemically treated samples exhibited earlier occurrence and higher intensity of the second exothermic peak, indicating that electrochemical treatment promoted early-age hydration reactions. Notably, the EW-0 group displayed the most pronounced second hydration peak, suggesting that electrochemical treatment under neutral electrolyte conditions is more favorable for activating the hydration of reactive components in red mud. Figure 8 demonstrates that particle surfaces became rougher after iron oxide removal, increasing the contact area with the hydration solution and enabling the exposed reactive phases to participate fully in early-age hydration.

The EN-4 group exhibited only one exothermic peak. The cumulative heat release curve shows that the 72 h cumulative heat of the EN-4 group was the lowest, only slightly higher than that of the REF group. Figure 7b reveals that the diffraction peak intensities of cancrinite, muscovite, and kaolinite in the EN-4 group decreased to the lowest levels, indicating the most severe mineral phase decomposition. Studies have shown [31] that aluminosilicate minerals undergo rapid dissolution in high-concentration NaOH solutions. This pre-dissolution reaction is essentially a premature hydration process that consumes some reactive components. Table 1 shows that the raw red mud already contains 11.76% Na₂O. Combined with the newly introduced 4 mol/L NaOH (Table 2), the Na⁺ concentration in the system is extremely high. Excessive alkali concentration inhibits the development of later-stage strength [36]. The earlier but broader exothermic peak and limited cumulative heat release of the EN-4 group explain the lowest compressive strength observed in Figure 9.

3.4.4. Phase Composition Analysis

Figure 11 presents the XRD patterns of cementitious materials prepared from red mud pretreated with different electrolyte systems after 28 days of curing. The EN-4 group exhibited the highest diffraction peak intensities of Ca(OH)₂ (portlandite), tricalcium silicate (alite), and dicalcium silicate (belite), while the EH-4 group showed the lowest intensities. The peak intensity of portlandite in the EW-0 group was higher than that of the REF group. These differences are closely related to the chemical characteristics of each electrolyte system. It should be noted that while XRD identifies crystalline phases, the amorphous C-S-H gel—which is the primary binding phase—does not produce sharp diffraction peaks. The EW-0 and EH-4 groups, with their superior compressive strengths, likely contain higher amorphous gel content that contributes to diffuse scattering rather than discrete crystalline peaks.

The EN group underwent electrochemical treatment in an alkaline electrolyte. According to the XRD results in the figure, the high-alkalinity environment promotes the dissolution of calcium-bearing minerals such as cancrinite. The released Ca²⁺ subsequently forms Ca(OH)₂ during cement hydration. Additionally, Figure 9 shows that the EN-4 group exhibited lower 28-day strength. This indicates that although portlandite content is high, the formation of cementitious phases that contribute more significantly to strength may be insufficient. This is consistent with the aforementioned finding that excessive alkalinity inhibits the formation of later-stage hydration products.

The EH group was treated in an acidic electrolyte. Figure 7a demonstrates that acid treatment markedly reduced the peak intensities of muscovite and kaolinite. Decomposition of these layered aluminosilicates releases reactive SiO₂ and Al₂O₃, which react with Ca(OH)₂ during hydration to form C-S-H gel, thereby consuming portlandite. The EH-4 group exhibited lower residual peak intensities of C₃S and C₂S. Combined with the higher cumulative heat release in Figure 10 and the high 28-day strength in Figure 9, this indicates more complete hydration of these cement minerals. The EW-0 group used deionized water as an electrolyte, maintaining a near-neutral system. Electrochemical treatment had minimal effect on red mud alkalinity, resulting in moderate portlandite peak intensity.

3.4.5. Characterization of Hydration Degree

Figure 12 presents the TG-DTG curves of cementitious materials prepared from red mud pretreated with different electrolyte systems after 28 days of curing. All samples exhibited four distinct mass-loss regions located before 105 °C, at 125~200 °C, 400~470 °C, and 550~720 °C.

The mass loss before 105 °C primarily corresponds to the removal of weakly bound water and adsorbed water from C-S-H gel. The mass loss at 125~200 °C corresponds to the dehydration of structural water from C-S-H gel, ettringite, and gypsum dihydrate [34,35]. The mass loss at 400~470 °C mainly corresponds to the dehydroxylation of Ca(OH)₂ and the removal of some C-S-H structural water. The mass loss peak at 550~720 °C primarily corresponds to the decomposition of carbonates [37]. These results indicate that the hydration products of red mud-based cementitious materials mainly comprise C-S-H gel, ettringite, and Ca(OH)₂. Further comparison of mass loss among groups reveals that the neutral electrolyte treatment group (EW-0) and acidic electrolyte treatment group (EH-4) exhibited significantly higher mass loss in the 125~200 °C region compared to other groups, indicating greater C-S-H gel formation. This result is consistent with the superior mechanical performance of the EW-0 and EH-4 groups observed in Figure 9, confirming the critical role of C-S-H gel as the primary cementitious phase in strength development. Therefore, electrochemical pretreatment under neutral and acidic electrolyte conditions effectively enhances the degree of hydration of red mud-based cementitious materials, thereby improving their mechanical performance.

3.4.6. Functional Group Evolution

Figure 13 presents the FTIR spectra of cementitious materials prepared from red mud pretreated with different electrolyte systems after 28 days of curing. The intensities of the characteristic absorption peaks can be placed in the following order (from highest to lowest): neutral treatment group, acidic treatment group, alkaline treatment group, and untreated group. Red mud-based cementitious materials treated under neutral electrolyte conditions exhibited the strongest infrared absorption peaks, indicating the most significant formation of hydration products.

The absorption peaks observed near 1647 cm⁻¹ and 3445 cm⁻¹ correspond to the H-O-H bending vibration and O-H stretching vibration of water molecules, respectively. The peak at 3644 cm⁻¹ is attributed to the O-H stretching mode of Ca(OH)₂ [38]. The absorption peaks at 713 cm⁻¹, 875 cm⁻¹, and 1424 cm⁻¹ originate from the in-plane bending, out-of-plane bending, and asymmetric stretching vibrations of CO₃²⁻ in CaCO₃, respectively. These are related to calcium carbonate present in the raw materials and may also result from the reaction between cement and atmospheric CO₂ during curing [39]. The absorption peak at 969 cm⁻¹ corresponds to Si-O stretching vibration, while the absorption near 460 cm⁻¹ is attributed to O–Si–O bending vibration. The peak near 518 cm⁻¹ is associated with Fe-O stretching vibration [40]. Additionally, the absorption peak at 1109 cm⁻¹ reflects the asymmetric stretching vibration of Si-O-Si or Si-O-Al bonds in SiO₄ or AlO₄ tetrahedra [41].

Analysis of absorption peak intensity differences reveals that the EW-0 group exhibited the highest peak intensities at 969 cm⁻¹ (Si-O) and 1109 cm⁻¹ (Si-O-Si/Si-O-Al). Studies have shown [41] that the absorption peak at 1109 cm⁻¹ is associated with Si-O-Si bridging bonds in C-S-H gel. Increased peak intensity indicates enhanced connectivity of silicate tetrahedra. Electrochemical treatment under neutral electrolyte conditions most effectively promoted the transformation of SiO₄ or AlO₄ tetrahedra from isolated to polymerized states, favoring the formation of aluminosilicate network structures. The EH-4 group showed moderate absorption peak intensities, whereas the EN-4 group exhibited lower intensities, indicating that excessive alkalinity inhibited the effective dissolution and polycondensation of silicate and aluminate species, resulting in insufficient gel-phase development. The REF group displayed the lowest peak intensities, confirming the limited degree of hydration in untreated red mud. FTIR is particularly sensitive to amorphous phases through the detection of short-range bonding environments. The enhanced Si-O-Si/Si-O-Al band intensity at 1109 cm⁻¹ in the EW-0 and EH-4 groups indicates higher amorphous gel content [42]. Combined with the reduced crystalline aluminosilicate peaks observed in Figure 7, this confirms that electrochemical treatment converts low-reactivity crystalline minerals into reactive amorphous phases, which is the key mechanism for enhanced cementitious performance. These results demonstrate that electrochemical treatment significantly enhances the degree of hydration of red mud-based cementitious materials, with the most pronounced effect observed under neutral electrolyte conditions.

3.4.7. Microstructure and Morphology Analysis

Figure 14 shows the microstructure of cementitious materials prepared from red mud pretreated with different electrolyte systems after 28 days of curing. The results demonstrate that cementitious materials prepared from electrochemically treated red mud exhibit a denser microstructure than those in the control group. The types of hydration products are generally similar across all groups, primarily consisting of needle-like ettringite, plate-like Ca(OH)₂, and network-like C-S-H gel. The needle-like ettringite crystals interlace to form a spatial framework structure, while plate-like Ca(OH)₂ fills the pore spaces. C-S-H gel, in its network-like morphology, encapsulates and binds hydration products and unreacted particles. Together, these three phases constitute a dense microstructure that provides the primary cementitious strength.

Based on the comprehensive characterization results, electrochemical treatment with different electrolyte systems significantly influences the hydration behavior and mechanical performance of red mud when subsequently used as a supplementary cementitious material. Although the EN-4 group treated with alkaline electrolyte (4 mol/L NaOH) exhibited a high heat evolution rate in the early stage, the hydration of cement minerals at later stages was insufficient. This resulted in low hydration product content, loose microstructure, and a 28-day strength of only 59.0 MPa, the lowest among all electrochemically treated groups. In contrast, the EW-0 and EH-4 groups showed more complete hydration of cement minerals, higher content of hydration products, and a denser microstructure. Their 28-day strengths reached 68.9 MPa and 69.1 MPa, respectively, representing significant improvements over the untreated group.

4. Conclusions

This study investigated Bayer red mud and developed an efficient and green electrochemical reduction method for iron removal. The effects of different electrolyte systems (neutral, acidic, and alkaline) on the electrochemical reduction mechanism and Faradaic efficiency of iron oxides were systematically explored. The activation mechanisms by which electrochemical treatment modifies the microstructure of red mud and enhances its hydraulic activity and mechanical performance were comprehensively investigated. The main conclusions are as follows:

(1)

Electrochemical reduction demonstrated excellent iron extraction performance under alkaline conditions, with average Faradaic efficiency exceeding 80%. The alkaline environment facilitates the formation of soluble iron complexes and suppresses hydrogen evolution, enabling the efficient conversion of non-magnetic hematite to magnetically separable species.

(2)

Electrochemical treatment significantly enhanced the cementitious activity of red mud by disrupting crystalline aluminosilicate structures and increasing amorphous phase content. The process removes iron oxide that encapsulates reactive minerals and converts crystalline phases (muscovite, kaolinite, cancrinite) into disordered, reactive amorphous species. This mineralogical transformation—from high-crystallinity, low-reactivity minerals to amorphous, high-reactivity phases—is the fundamental activation mechanism. Neutral (EW-0) and acidic (EH-4) treatments achieved 28-day compressive strengths of 68.9–69.1 MPa by enhancing C-S-H gel formation and improving pozzolanic reactivity.

(3)

Electrolyte pH determines performance trade-offs between iron recovery and cementitious properties. Alkaline conditions maximize iron extraction efficiency but compromise strength development due to excessive residual alkalinity. Neutral and acidic conditions optimize cementitious performance while achieving moderate iron recovery. This flexibility enables tailored valorization strategies based on application priorities. Future work should address scalability, long-term durability, and techno-economic feasibility for industrial implementation.

This study established a stepwise valorization approach for red mud that integrates electrochemical activation for enhanced cementitious performance with simultaneous iron recovery. The activation mechanisms and the influences of different electrolyte systems on both cementitious performance enhancement and iron recovery efficiency were systematically elucidated, providing new insights for the full-component valorization of red mud.