Erosion Resistance of Iron Ore Tailings as Aggregate for Manufacturing of Cement-Based Materials

Abstract

Cement-based materials used in China’s coastal and salt lake areas in the northwest are exposed to long-term chloride corrosion, which deteriorates the materials and substantially reduces the durability of the structures. This study investigates the chlorine ion erosion resistance in salt spray environments of cement-based materials made with iron ore tailings (IOTs) as an aggregate (namely, IOTCs). The compressive strength, mass loss, and relative dynamic elastic modulus (RDEM) macroscopic performance of IOTC undergoing different chloride diffusion times (0–180 d) were explored in detail. Chloride ion profiles at 0–180 d were analyzed via chemical titration, while X-ray computed tomography (CT) and scanning electron microscopy (SEM) were employed to characterize microstructural evolution. The results demonstrate that IOTC exhibited superior chloride resistance compared to conventional concrete (GC). While both materials showed early strength gain (<60 d) due to hydration and pore-filling effects, IOTC experienced only a 23.9% strength loss after long-term exposure (180 d) significantly less than the 37.2% reduction in GC. Chloride profiling revealed that IOTC had 43.5% lower free chloride ions (C_f) and 32% lower total chloride ions (C_t) at 1 mm depth after 180 d, alongside reduced chloride diffusion coefficients (D_a). The CT analysis revealed that IOTC exhibited a significantly denser and more uniformly distributed pore structure than GC, with a porosity of only 0.67% under chloride-free conditions. SEM confirmed IOTC’s more intact matrix and fewer microcracks.

1. Introduction

With the global emphasis on green, sustainable development and resource recycling, the reuse of iron ore tailings (IOTs), an industrial by-product [1,2,3], in construction materials is gaining increasing attention. Global IOT reserves exceed 230 billion tons, yet the utilization rate remains below 10% [4]. The long-term storage of large volumes of IOT not only occupies valuable land but also poses environmental risks, including heavy metal leaching [5,6,7]. Meanwhile, concrete, the most widely used construction material globally, faces a major decarbonization challenge due to the high carbon emissions associated with cement production, which accounts for approximately 8% of global CO₂ emissions. The adoption of solid waste upcycling technologies is essential for the sustainable transition of concrete. Therefore, investigating the large-scale application of IOT as an eco-friendly aggregate and powder in cementitious materials is crucial for promoting the development of green construction materials [8,9].

It is worth noting that extensive research has consistently shown that using IOT as an fine aggregate facilitates the clean production of cementitious composite materials [10,11,12]. It was reported by Lv et al. [13] that replacing 15% of cement with diatomite and fully substituting river sand with IOT improved the grading and compactness of aggregates at a 25% IOT replacement level, resulting in a 12–18% increase in concrete strength. However, at higher replacement levels (more than 75%), the dilution of cementitious components and disruption of the packing effect begin to dominate, leading to a reduction in mechanical strength. Similarly, Zhao et al. [14] and Zhang et al. [15] reported that incorporating tailings into concrete in appropriate proportions optimizes its density and improves both compressive strength and durability. Fine IOT particles fill the voids in concrete, effectively reducing its permeability. However, when the IOT content exceeds 40% the optimal limit, it increases porosity and promotes microcrack formation, ultimately compromising concrete performance. Shi et al. [16] evaluated the use of IOT as a supplementary cementitious material (SCM) and as a replacement for fine aggregate in ultra-high-performance concrete (UHPC). Their results showed that when fine aggregate was replaced by IOT, the mechanical properties decreased. The most significant reduction occurred at a 100% replacement level. Shettima et al. [17] found that some strength indices of concrete with IOT of intermediate quality belonging to the 882 standards (weights of 25%, 50%, 75%, and 100%) were better than those of the control group at each level of replacement, but the workability declined with higher IOT proportions. Mendes Protasio et al. [18] investigated the use of IOT from the Germano Dam in Brazil as a substitute for cement and sand. The study found that replacing 30% of the natural river sand with IOT reduced pore size, thereby improving compressive strength. Research has established that the innovative use of IOT as a substitute for natural fine aggregates in traditional concrete and mortar mixtures contributes to notable enhancements in material performance.

Simultaneously, low-carbon cement-based materials face significant durability challenges in chloride-laden environments, particularly in civil engineering applications [19,20,21,22]. Chloride ions penetrate cement-based materials pores and disrupt the passive film on steel reinforcement, initiating corrosion [23]. This corrosion subsequently leads to visible damage, such as cracking, delamination, and spalling [24,25]. As a result, the structural integrity and service life of cement-based materials infrastructures are severely compromised. Globally, such corrosion-related degradation is estimated to cause economic losses exceeding USD 2.5 trillion each year [26]. In recent years, increasing attention has been devoted to enhancing the chloride resistance of low-carbon cement-based materials, aiming to ensure structural durability under aggressive chloride exposure. A reduction in chloride permeability was observed when manufactured sand was replaced with IOT, attributed to pore structure refinement through the filling effect [15]. In contrast, an increase in electric flux, along with acceptable chloride resistance levels, was reported by Shettima et al. [17]. Similarly, an approximately 21.85% rise in electric flux was documented by Shi et al. [16], suggesting a reduction in chloride resistance; however, acceptable performance was maintained due to a decrease in cumulative pore volume. It can be observed that the effects of IOT as fine aggregate on the chloride resistance of cement-based materials vary, and further research is needed.

Although the mechanical properties of concrete have been extensively studied, the effects of incorporating IOT into low-carbon cementitious systems on chloride ion penetration, as well as the underlying mechanisms, remain critical scientific issues and underexplored areas in current research. Therefore, studying the chloride ion penetration behavior in IOT as aggregate for manufacturing cement-based material is crucial for advancing their use in eco-friendly construction materials. This study comprehensively examines the influence of IOT as an aggregate on the chloride resistance of cement-based materials by assessing compressive strength, mass loss, relative dynamic modulus (REDM), chloride ion diffusion, pore structure, and microstructure, providing theoretical and technical support for its application.

2. Materials and Test Methods

2.1. Raw Materials and Specimen Preparation

Ordinary Portland cement (OPC) PO 42.5R (Shaanxi Tianzhu Cement Manufacturing Co., Ltd., Xi’an, China), first-grade fly ash (FA) (Huaneng Power International, Inc. Hebei Branch, Beijing, China), metakaolin (MK) (Shaanxi Yulin Pubai Kaolin Co., Ltd., Hanzhong, China), silica sand (S) (Xi’an Jieli Quartz Materials Co., Ltd., Xi’an, China), and iron ore tailings (IOTs) (Qingke Huaxing New Materials Technology Co., Ltd., Dongguan, China) fine aggregate as the main solid raw materials manufactured by the IOTC were used.

Table 1. Primary chemical compositions in OPC, FA, MK, S, and IOT (wt.%).

	OPC	FA	MK	S	IOT
SiO2	20.5	52.9	22.13	98.21	70.26
Al2O3	4.3	29.9	63.81	1.02	6.42
CaO	64.1	3.6	5.1	0.32	4.21
MgO	/	1.5	0.94	0.14	2.85
Fe2O3	2.8	7.9	5.51	0.11	13.75
TiO2	0.2	/	/	/	/
SO3	/	0.6	2.15	0.02	0.24
Others	0.3	3.2	0.62	0.16	1.22

details the chemical compositions of the solid raw materials.

In line with the Chinese national standard (GB/T14684-2011 [27]) and ISO 13320, both IOT and S were tested for their key physical properties, and their particle size distributions were analyzed using a laser particle size analyzer (Bettersize 2600, Bettersize Instruments Ltd., Dandong, China). The samples were dispersed in deionized water using ultrasonic treatment for 2 min prior to measurement to ensure adequate dispersion. The measurement range was 0.01–3500 μm, and the refractive index of the dispersant and materials was set accordingly. Each test was repeated three times to ensure repeatability, as seen in Figure 1. The data reveal that IOT has a smaller particle size relative to S. However, the water absorption of IOT is high. This difference in water absorption performance may be due to the rough surface and internal porosity of IOT particles, which provide more pathways for water ingress and thereby increase the overall water absorption capacity of the material.

Figure 2 presents the results of X-ray diffraction (XRD) and SEM analyses, which were employed to investigate the primary components and microstructure of IOT and S. The primary crystal composition of IOT closely resembles that of S, consisting predominantly of quartz with stable chemical properties. SEM observations further revealed that the surface morphology of IOT is more angular than that of S.

This study aimed to investigate the chloride ion erosion resistance of IOTC under a salt spray environment. The mixtures were divided into two groups: a conventional cement-based mixture (control group) with 0% IOT replacement and an IOTC mixture with 25% IOT replacement [28]. The two types of IOT were designated as GC and IOTC, respectively.

Table 2. Mix proportions (kg/m³).

No.	Fine Aggregate		Cementitious Materials			W	SP	Flow Diameter	28 d fcu (MPa)
	S	IOT	OPC	FA	MK			(mm)
GC	565.6	0	692	377.4	188.7	377	5.04	215	37.4
IOTC	424.2	141.4	692	377.4	188.7	377	5.69	185	45.2

shows the mix proportions for both groups of specimens. Cubic specimens measuring 100 mm × 100 mm × 100 mm and prismatic specimens measuring 100 mm × 100 mm × 400 mm (three parallel specimens for each group) were prepared according to the Chinese standard GB/T 50082-2019 [29] and the experimental design. These specimens were used to evaluate compressive strength, mass loss, RDEM, chloride ion distribution, and industrial CT non-destructive analysis after exposure to different durations of salt spray corrosion.

2.2. Test Methods

2.2.1. Test for Chloride Resistance of Specimens Under Salt Spray Environment

The test was performed in an automatic salt spray corrosion chamber, in accordance with ASTM B117-18 [30], GB/T 10125-2012 [31], and GB/T 50082-2019 [29] standards. The temperature was maintained at 35 ± 2 °C, with a salt spray deposition rate of 1–2 mL/80 cm²·h. The salt solution concentration was 5% NaCl, and the pH was between 6.5 and 7.2. The spray method used was intermittent alternating spraying. Each test cycle lasted for 30 d, with a total maximum corrosion duration of 180 d (6 cycles). After the specified corrosion time, the 100 mm × 100 mm × 100 mm and 100 mm × 100 mm × 400 mm samples were removed, cleaned to remove surface salts, and air-dried. Subsequently, the compressive strength was measured at a loading rate of 0.5 MPa/s with a 1000 kN MTS microcomputer-controlled electro-hydraulic servo universal testing machine. The mass loss was measured with an electronic analytical balance with an accuracy of 0.01 g. Prior to weighing, the surface of each specimen was wiped clean and air-dried until no visible moisture remained. The RDEM was determined using an ultrasonic pulse velocity tester (Proceq Pundit Lab+, Screening Eagle Technologies, Zurich, Switzerland). Each test group consisted of three samples, and the average values were calculated to determine the final data.

2.2.2. Test for Chloride Ion Content in Specimens Under Salt Spray Environment

In accordance with JGJ/T 322-2013 [32] and ASTM C1152/C1218 [33,34] standards, the chloride ion content of the samples, exposed to the specified corrosion time, was determined using the silver nitrate titration method (Mohr method). This measurement includes both free chloride ions (C_f) and total chloride ions (C_t). Powder samples were collected using a handheld electric drill equipped with a depth control device. The specimens were sliced in successive layers from the exposed surface at depths of 1 mm intervals for the range of 1–10 mm, and at 2 mm intervals from 10–20 mm. This sampling scheme was designed to capture the high-resolution chloride ion distribution in the near-surface region, where the gradient is steep, and lower resolution at deeper depths where ion migration slows down. Each powdered sample was passed through a 0.15 mm sieve before chemical analysis.

2.2.3. Test for Pore Structure of Specimens Under Salt Spray Environment

Industrial X-ray computed tomography (CT) (MS-Voxe1450, MTS Industrial Testing Co., Ltd., Shenzhen, China) scanning was conducted on cube specimens pre- and post-chloride ion corrosion exposure. Scanning parameters included a 380 kV operating voltage, 1.3 mA current, and 0.5 s exposure duration. Cross-sectional scans at a 0.5 mm depth were acquired at 20 mm vertical intervals, yielding 200 two-dimensional tomographic images (1200 × 1200 px resolution). AVIZO 3D image processing software (Thermo Fisher Scientific, Waltham, MA, USA) reconstructed the CT image data into a 3D pore distribution model.

2.2.4. Test for Microstructure of Specimens Under Salt Spray Environment

After compressive strength testing, fragments were selected from the core region of the fractured concrete cubes to avoid edge effects. These fragments were manually cut into smaller pieces and subsequently polished and ground using a series of progressively finer silicon carbide abrasive papers until the particle size ranged between 3 and 5 mm. To remove residual surface moisture, the prepared samples were placed in a vacuum drying oven at 50 °C for 24 h prior to further characterization. The microstructure was thoroughly analyzed using an S-4800 (Hitachi High-Tech Corporation, Tokyo, Japan) cold field emission scanning electron microscope (SEM).

3. Results and Discussion

3.1. Chloride Resistance

3.1.1. Compressive Strength

Figure 3 illustrates the impact of chloride corrosion duration on compressive strength. The compressive strength of GC and IOTC samples exhibits a two-stage behavior: an initial enhancement followed by a subsequent deterioration. After 60 d of corrosion, the compressive strength of GC and IOTC samples increases by 8% and 4.2%, respectively, compared to the non-corroded state, reaching a peak for both. The primary reason lies in the continued hydration of cementitious materials under chloride-rich conditions, as well as the reaction between chloride ions and aluminates present in the cement matrix. This interaction leads to the formation of secondary products, such as Friedel’s, which can fill capillary pores and refine the pore structure, thereby enhancing the material’s density and compressive strength. Li et al.’s [35] report also demonstrates this phenomenon. In addition, the strengthening effect is more pronounced in GC, likely due to its higher aluminate content, which promotes greater formation of chloride-binding products and results in a denser microstructure during the early stage of corrosion. As the corrosion time exceeds 90 d, the compressive strength of both GC and IOTC samples decreases with prolonged corrosion. Specifically, after 180 d of corrosion, the compressive strength of GC and IOTC samples decreased by 37.2% and 23.9%, respectively, compared to their non-corroded counterparts. This decline is primarily due to the accumulation of chloride ions within the material as corrosion advances, leading to the reaction and decomposition of some C-S-H gels. Concurrently, the volume expansion of the corrosion byproducts further enlarges the pores and microcracks within the material, ultimately reducing its density and compressive strength. Moreover, it is evident that the initial increase in compressive strength of GC samples during early-stage corrosion may surpass that of IOTC samples, while the subsequent strength reduction during later-stage corrosion is more pronounced. This is primarily because the GC specimens exhibit a higher initial porosity compared to IOTC, allowing corrosion products to accumulate and fill the voids more rapidly and extensively, thereby resulting in a faster strength gain at the early stage. However, as the deposition of corrosion products continues, it leads to a progressive build-up of internal stresses and significant propagation of microcracks. This is particularly evident in GC, where the interfacial transition zone (ITZ) is more fragile. In addition, prolonged chemical attack causes the decomposition of C–S–H gel, further weakening the microstructure. In contrast, the IOTC specimens possess a denser and more homogeneous pore structure, which effectively limits chloride ion penetration and the rapid accumulation of corrosion products. This delays the degradation process and results in a more gradual decline in mechanical strength at later stages.

3.1.2. Mass Loss

Figure 4 demonstrates the mass loss for specimens exposed to varying diffusion times, highlighting a two-stage mass loss pattern for both GC and IOTC samples. Initially, there is a slight increase in mass at the early stage (60 days), measuring 0.26% for GC and 0.15% for IOTC. This phenomenon aligns with findings by Liu et al. [36], suggesting that the quality of cement-based materials improves due to ettringite crystallization in the initial phase of salt spray corrosion. Beyond 60 d, mass loss gradually escalates for both samples, with a significant increase observed after 120 d of corrosion. This acceleration is primarily attributed to the dissolution of corrosion products and the propagation of micro-cracks [37]. Notably, throughout the testing period, the mass loss of IOTC samples consistently remains lower than that of GC samples, with final mass losses of 2.29% and 1.77% at 180 d, respectively. This disparity underscores the superior salt spray corrosion resistance of IOTC samples.

3.1.3. RDEM

The RDEM serves as a crucial indicator for assessing the evolution of microstructural damage in cement-based composites subjected to salt spray corrosion [38]. The variation in RDEM can effectively elucidate the degradation mechanism of material durability. Figure 5 illustrates the evolution characteristics of RDEM for GC and IOTC at different durations of chloride ion erosion. The data in Figure 5 reveal that RDEM exhibits pronounced fluctuations during the erosion process due to its high sensitivity to changes in the micro-pore structure of the material, with a test sensitivity as low as a 0.1% strain level. Moreover, the significant fluctuations in RDEM values for GC and IOTC can be attributed to temperature and humidity variations during the testing, highlighting a noticeable contrast with compressive strength and mass loss trends. The degradation patterns of GC and IOTC RDEM exhibit a nonlinear trend, with degradation indicators initially increasing and then decreasing. Moreover, the degradation rates differ significantly between the two materials. Specifically, the RDEM of GC exhibits a prominent 9.2% increase at 90 d of erosion, while IOTC’s RDEM peaks with an 8.5% increase at 60 d of erosion. This suggests that IOTC demonstrates faster response in the early stages of erosion, facilitating quicker filling of internal pores through the generation of Friedel’s and C-S-H gels, leading to the significant short-term enhancement of RDEM at a macroscopic level [39]. With prolonged erosion, the ongoing intrusion of chloride ions gradually intensifies material degradation. The degradation rate of RDEM for IOTC is comparatively gradual, whereas RDEM of GC experiences a substantial decrease after 120 d of erosion. Following 180 d of erosion, GC and IOTC RDEM decreased by 13% and 6.9%, respectively, underscoring IOTC’s superior resistance to chloride ion erosion and its ability to sustain greater integrity in long-term erosive conditions.

3.2. Chloride Ion Diffusion Behavior

3.2.1. Free Chloride Ions and Total Chloride Ions

Figure 6 illustrates the distribution patterns of free chloride ions (Figure 6a) and total chloride ions (Figure 6b) along the diffusion depth for GC and IOTC specimens at various erosion ages (30, 60, 90, 120, 150, and 180 d). The contents of free chloride ions (C_f) and total chloride ions (C_t) exhibited a steep decline from the surface to a depth of approximately 12 mm, beyond which the concentrations tended to stabilize, forming a plateau. This indicates that chloride ions primarily accumulate near the surface in the early stages, while their inward migration is notably restricted. This diffusion behavior is primarily governed by the capillary pore transport mechanism. Chloride ions penetrate the material through interconnected pore networks driven by concentration gradients and capillary suction. Near the surface, the relatively open and porous microstructure offers little resistance, facilitating quick ion migration. In contrast, at greater depths, the pore structure becomes denser and more refined, increasing diffusion resistance and thereby slowing further chloride ingress. A comparison between GC and IOTC specimens reveals that, at the same age and diffusion depth, the chloride ion content in IOTC specimens is significantly lower than in GC specimens. This suggests that GC materials tend to adsorb and retain more chloride ions, likely due to their higher porosity or more complex microstructure. For instance, at 180 d of erosion, the C_f and C_t contents in IOTC specimens at a 1 mm depth were 32.2% and 43.89% lower, respectively, than those in GC specimens. Additionally, at the same diffusion depth, both GC and IOTC specimens exhibited rapid chloride accumulation during the early stage (30–90 d) and slower diffusion during the later stage (120–180 d), with a noticeable plateau at 180 d. In IOTC samples, chloride contents at a 1 mm erosion depth were 32% and 43.5% lower, respectively, than those in GC samples. For example, in IOTC specimens at a 1 mm depth, the growth rates of C_f and C_t between 30–90 d were 94.39% and 22.32%, respectively, while between 120–180 d they were 36% and 19.76%, respectively. These results indicate that chloride ion diffusion in the material gradually stabilizes as internal adsorption and binding sites become saturated, leading to a reduced penetration capacity over time.

3.2.2. Bound Chloride Ions

In unsaturated concrete, chloride ions exist in two forms: free and bound chloride ions. Components in the hydration products of concrete, such as C-S-H gel, AFm phase, or mineral admixtures like fly ash and slag, can immobilize C_t through physical adsorption or chemical bonding, converting them into bound forms. Martin et al. [40] modified Fick’s model to account for chloride ion adsorption and bonding effects in unsaturated concrete, as shown in Equations (1)–(3):

$${\displaystyle \frac{\partial C_{\mathrm{t}}}{\partial t}}={\displaystyle \frac{\partial }{\partial x}}\left(D_0{\omega }_e{\displaystyle \frac{\partial C_{\mathrm{f}}}{\partial x}}\right)$$

(1)

$$C_{\mathrm{t}}={\omega }_e·C_{\mathrm{f}}+C_{\mathrm{b}}$$

(2)

$$C_{\mathrm{b}}=\alpha C_{\mathrm{f}}$$

(3)

where C_t is the total chloride ion content in unsaturated concrete (%); C_f is the free chloride ion content in unsaturated concrete (%); C_b is the bound chloride ion content in unsaturated concrete (%); ω_e is the volume ratio of evaporable water in the pores of unsaturated concrete to the total volume of concrete, assumed to be 1; D₀ is the diffusion coefficient of free chloride ions in the concrete pore solution (m²/s); and α is the chloride binding coefficient of concrete.

As shown in Figure 7, the chloride binding capacity of the specimens changes significantly over time. The chloride binding capacity initially increases, followed by a gradual decrease. This trend reflects the dynamic interaction between free chloride ions and available binding sites in the cementitious matrix during exposure. The increase indicates that chloride ions initially react with the material’s surface and binding sites. However, over time, as the chloride ions accumulate, the binding sites begin to saturate, leading to a slower rate of chloride uptake. During the initial erosion stage (up to 30 d), the chloride binding capacity increases markedly. This is primarily due to the abundance of binding sites in hydration products, such as calcium silicate hydrate (C-S-H) gel and AFm phases. These compounds interact with chloride ions through physical adsorption and chemical substitution. Additionally, the pore solution remains highly alkaline, which promotes the retention of chloride in bound form. After 30 d of exposure, the chloride binding capacity begins to decline. This reduction is likely due to the saturation of available binding sites, as well as the gradual breakdown of the binding phases under prolonged chloride attack. The material’s capacity to fix additional chloride ions diminishes, as the interactions between chloride ions and the binding sites become less effective due to the depletion of available sites and the structural degradation of binding phases. Continued chloride ingress may cause the decalcification of C-S-H and the decomposition of AFm phases, the material’s capacity to fix additional chloride ions diminishes. These findings indicate that although concrete initially provides an effective barrier to chloride penetration, its resistance weakens over time due to a reduced binding capacity.

3.2.3. Chloride Diffusion Coefficient

Figure 8 illustrates the evolution of chloride diffusion coefficients (D_a) in GC and IOTC specimens across erosion ages from 30 to 180 d. Both materials exhibit a continuous decrease in D_a with increasing diffusion time, indicating a progressive weakening of chloride ion mobility. This decline is primarily attributed to ongoing hydration reactions and the secondary pozzolanic activity of reactive mineral components, particularly pronounced in IOTC. These reactions consume calcium hydroxide and produce dense C–S–H gels, which refine the pore structure and reduce the connectivity of capillary pores, thereby effectively limiting further chloride penetration. The apparent D_a decreases markedly during the early exposure period (30–90 d). This indicates that chloride ions predominantly migrate into the outer layer of the specimens, where a relatively porous and interconnected capillary pore structure still exists. At this stage, the diffusion resistance is low, facilitating rapid ion transport. In contrast, during the later stages (≥90 d), the rate of decline in D_a becomes more gradual. This shift reflects the progressive densification of the cementitious matrix, reduced pore connectivity, and the increasing difficulty of ion movement through the refined microstructure. The continued hydration and pozzolanic reactions contribute to this refinement, thereby impeding further chloride ingress.

Notably, the IOTC specimens exhibited a more pronounced decline in chloride diffusion coefficient over time, highlighting their superior resistance to chloride ion transport. This behavior can be attributed to the refined pore structure and enhanced binding phases formed by the incorporation of industrial by-products. These characteristics effectively limit long-term chloride migration and reinforce the durability performance of IOTC under chloride exposure. Overall, IOTC specimens exhibit significantly lower Da than GC specimens at all exposure durations, indicating superior resistance to chloride ion ingress. For instance, as exposure time increased from 30 to 180 d, the Da of IOTC were reduced by 20.02%, 46.25%, 33.96%, 31.40%, 29.37%, and 33.70% compared to those of GC. This enhanced performance is primarily due to IOTC’s denser pore structure and lower capillary porosity, which jointly inhibit chloride ion transport. Additionally, the Da of IOTC samples decreased more sharply and stabilized after 120 d, suggesting the system approached a quasi-steady state of chloride ion transport. These findings highlight the excellent long-term resistance of IOTC to chloride-induced erosion.

3.3. Pore Characteristics

3.3.1. Porosity

Figure 9 illustrates the distribution of planar pore area along the height (z-axis) for GC and IOTC specimens subjected to various chloride ion diffusion times. The planar pore area percentage (PAP) is defined as the ratio of pore area within each cross-sectional slice to the total area of that slice. It can be found that except for the 60 d diffusion period, both specimens exhibit pronounced non-uniform pore distribution curves. These fluctuations are mainly caused by particle settling during material preparation and microstructural heterogeneity introduced by vibratory densification, resulting in the spatial dispersion of pores along the specimen’s height. Despite local fluctuations, both types of samples consistently exhibit a typical “surface-loose to internal-dense” pore gradient, which becomes more pronounced as chloride ion exposure time increases. For instance, the GC and IOTC samples exhibited higher porosity in the 0–10 mm region near the exposed surface at diffusion 30 d, indicating the presence of numerous interconnected pores that facilitated rapid chloride ion transport. In contrast, in the deeper region (after 12–15 mm), the porosity decreased and stabilized, suggesting that this phenomenon can be attributed to the physical and chemical interactions between chloride ions and hydration products, including the formation and deposition of secondary phases such as Friedel’s within capillary pores. These deposits occupy pore spaces and block continuous pathways, thereby reducing the measured porosity. Beyond a depth of 12–15 mm, the porosity stabilized at a lower level, indicating minimal structural disturbance and a more intact microstructure in the inner matrix. This stabilization suggests that chloride ingress beyond this zone is greatly hindered by the densified surface layer. The porosity of GC and IOTC samples was determined by averaging the pore area percentage, representing pore volume as a fraction of the total sample volume. Over time, the porosity of both GC and IOTC specimens followed a non-linear evolution—initially decreasing and then increasing slightly at later stages. The initial decline reflects the pore-blocking effects of corrosion product deposition, while the subsequent increase may result from microcracking or localized dissolution of binding phases due to prolonged chloride attack. Importantly, IOTC consistently exhibited a lower porosity than GC across all depths and exposure durations. This lower porosity reflects the denser microstructure developed through the pozzolanic activity of industrial by-products, which generate additional C-S-H gels and reduce capillary porosity. As a result, fewer pathways are available for chloride transport, effectively enhancing the material’s resistance to ion ingress. This process illustrates a time-dependent degradation mechanism: initial product filling enhances densification, while prolonged exposure leads to structural damage accumulation. Additionally, IOTC consistently exhibited a lower porosity than GC at all diffusion times. For example, within the 0–180 d range, IOTC specimens showed porosity reductions of 27.13%, 35.95%, 39.47%, 8.33%, 8.15%, 13.29%, and 1.18% compared to GC. This advantage is mainly attributed to the “filling effect” of IOT in IOTC, which reduces the initial capillary porosity and highlights the critical role of microstructural densification in resisting ionic intrusion. The reduced porosity of IOTC plays a critical role in its superior long-term durability under chloride-rich conditions. A lower porosity not only limits the physical space for ion movement but also reduces the connectivity of transport pathways. Combined with improved chloride binding capacity, this contributes to a more robust barrier against chloride-induced degradation.

3.3.2. Size of Pores

GC and IOTC specimens’ pores were categorized into four groups based on size: gelatinized pores (<10 nm), transition pores (10–100 nm), capillary pores (100–1000 nm), and macropores (>1000 nm). Figure 10 presents the changes in the percentage of pore size subjected to chloride exposure over different diffusion times. It can be observed from Figure 10a,b that pores in the ranges of (<10 nm) and (10–100 nm) dominate in GC and IOTC specimens subjected to chloride salt erosion. When the diffusion time was less than 90 d, there was a notable increase in the proportion of pores (<10 nm), accompanied by a gradual decline in pores (>100 nm). This shift indicates that the formation and deposition of chloride-induced corrosion products rapidly filled existing gel pores and refined both capillary and macropores. Consequently, the overall porosity decreased, and the microstructure became more compact and less permeable. This refinement reduced the connectivity of capillary pores and limited diffusion pathways, effectively inhibiting chloride ion migration. However, after 90 d of diffusion, the proportion of pores larger than 1000 nm began to increase, while the share of (<10 nm) pores declined. This indicates that prolonged chloride exposure led to the degradation of the internal pore structure. At 180 d, the pores of (>1000 nm) accounted for 10.6% in the GC sample and 7.7% in the IOTC sample. Compared to earlier measurements, this represents a 27.35% reduction in the IOTC sample, indicating better pore structure stability.

3.4. Microstructure

The microstructural observations of GC and IOTC specimens subjected to different diffusion times are illustrated in Figure 11. Evident morphological evolution over diffusion time suggests a strongly time-governed mechanism underlying the internal degradation of the material. After 60 d of chloride exposure, the matrix mainly featured isolated pores, indicating that the pore network had become less interconnected. Compared to the unexposed specimens, a slight improvement in overall compactness was observed. This densification effect is attributed to the partial filling of microcracks and interconnected pores by corrosion products, which likely acted as physical barriers and contributed to refining the pore structure. However, with prolonged exposure up to 180 d, the internal stress induced by the formation of expansive secondary phases—such as Friedel’s salt—exceeded the local tensile strength of the matrix. This led to the initiation and propagation of microcracks, particularly along the interfacial transition zones (ITZs) and weakened regions. These cracks could potentially reopen previously blocked pathways, allowing deeper chloride ingress and undermining structural integrity. Notably, the IOTC specimens maintained a more coherent and compact microstructure throughout the entire exposure period, in contrast to the GC specimens. The reduced microcrack density and more stable pore network in IOTC are indicative of its superior resistance to chloride-induced degradation. This enhanced durability is supported by both microstructural observations and lower chloride diffusion coefficients, underscoring the effectiveness of industrial by-products in improving long-term performance.

4. Conclusions

This study systematically investigated the chloride resistance of cementitious composites (IOTCs) prepared by partially replacing fine aggregates with iron tailings (IOTs) in a salt spray environment. The main conclusions were drawn by comparing the evolution of compressive strength, mass loss, relative dynamic elastic modulus (RDEM), chloride ion distribution, pore structure, and micro-morphology between ordinary cementitious materials (GC) and IOTC over different corrosion periods, as follows:

(1) In the early stage of corrosion (within 60 d), the compressive strengths of both IOTC and GC increased, mainly due to ongoing hydration and the pore-filling effect of corrosion products. As corrosion progressed, the material properties gradually deteriorated. After 180 d, the compressive strength of IOTC decreased by 23.9%, significantly less than the 37.2% reduction observed in GC. In addition, IOTC exhibited significantly lower mass loss and reduced RDEM degradation compared to GC, which reflects its enhanced durability and improved structural integrity under chloride exposure;

(2) Chloride ions mainly accumulate in the surface layer and diffuse slowly into the interior. The diffusion depth stabilizes over time. After 180 d of chloride exposure, IOTC exhibited 43.5% and 32% lower C_f and C_t contents, respectively, at a depth of 1 mm compared to GC. Moreover, the Da of IOTC decreased more sharply and stabilized earlier, indicating a more effective barrier to chloride ion ingress;

(3) CT analysis revealed that IOTC consistently exhibited lower total porosity and a finer pore structure throughout the exposure period. Although the proportion of large pores (>100 nm) increased over time due to microstructural deterioration, this change was significantly less in IOTC, which maintained a more uniform and compact pore network;

(4) SEM results further demonstrated that in the later stages of chloride exposure, IOTC retained a more coherent matrix structure with fewer and less severe microcracks, compared to GC. This indicates excellent crack resistance under the expansion stress of corrosion products.