Strengthening Comparison of Carbonated Stabilized Soils Under Different Organic Matters

Abstract

Organic matter critically restricts the effectiveness of conventional cement stabilization of soft soils and undermines the performance of reinforced soil layers. However, how different organic matter components affect reactive MgO carbonation reinforcement remains poorly understood. This study adopted Portland cement (PC) stabilization, MgO stabilization and MgO carbonation to treat silty clay with 0~8% fulvic acid (FA) and humic acid (HA). The engineering performance and microstructural features of stabilized soils were investigated via dry density, unconfined compressive strength (UCS) and pH tests, combined with X-ray diffraction (XRD), scanning electron microscopy (SEM) and pore structure analysis. The results show that MgO carbonation achieved the optimal densification effect, delivering distinctly higher dry density and UCS than the other two methods. Increased organic matter content reduced soil alkalinity and hindered strength development, with FA exerting a stronger inhibitory impact than HA due to lower UCS and higher dosage sensitivity. Carbonation effectively mitigated the adverse effects of FA and HA and yielded superior soil strength. XRD identified nesquehonite, dypingite and hydromagnesite as dominant carbonation products; organic matter reduced the formation of these strength-enhancing phases and retained uncarbonated brucite. SEM further verified that organic matter modified carbonate morphology and distribution, forming loose microstructures despite reduced pore volume. Overall, reactive MgO carbonation presents stronger anti-interference capability against organic matter than conventional PC stabilization. It improves the performance of organic-rich soft soils and realizes CO₂ sequestration. This study provides experimental support and practical insights for the engineering application of low-carbon MgO-CO₂ soil stabilization technology.

1. Introduction

Soft soils are extensively distributed in coastal zones, estuarine deposits, inland plains and reclamation areas worldwide [1]. They are inherently characterized by high natural water content, high compressibility, low shear strength and insufficient bearing capacity [2]. These unfavorable engineering properties frequently trigger excessive settlement, poor foundation stability and long-term residual deformation in infrastructure construction. Accordingly, chemical stabilization has evolved as one of the most widely adopted ground improvement techniques for soft soil treatment, attributable to its practical applicability, high in situ soil utilization rate and well-established construction technology [3,4]. Ordinary Portland cement (PC) is the most prevalent binder for soft soil reinforcement [3]. It enhances the strength and stiffness of stabilized soils by binding loose soil particles, filling internal pores and forming a stable cemented skeleton, which relies on the generation of hydration products including calcium silicate hydrate (C-S-H), calcium aluminate hydrate (C-A-H) and portlandite [5,6]. Nevertheless, the stabilization efficiency of cement is significantly compromised when soft soils contain abundant soil organic matter, particularly humic substances [4,7]. Existing studies have demonstrated that organic matter can delay or even inhibit cement hydration reactions, hinder the formation of cementitious products, and thereby reduce the unconfined compressive strength (UCS) of cement-stabilized soils [8,9,10].

The detrimental effects of organic matter on soil stabilization are governed by a combination of physical and chemical mechanisms [6]. From a physical perspective, fine organic particles with superior water retention and adsorption capacities can adhere to the surfaces of PC and soil particles, forming a barrier that blocks the effective contact among PC particles, water and reactive soil minerals [11,12]. From a chemical perspective, humic acid (HA) and fulvic acid (FA), the primary active fractions of humic substances, are rich in oxygen-containing functional groups (e.g., carboxyl, phenolic hydroxyl and carbonyl groups). These functional groups can complex with Ca, Mg and Al ions in pore solutions and consume alkaline cement hydration products [5,10,13]. Notably, FA exhibits a lower molecular weight, higher solubility and stronger mobility than HA, which strengthens its interaction with hydration products and further exacerbates the deterioration of the cemented soil microstructure [14]. In this context, distinguishing the individual effects of HA and FA is critical to fully elucidate the inhibitory mechanism of organic matter in chemically stabilized soils [4,15].

In recent years, low-carbon alternatives to conventional cement stabilization have attracted growing research attention, driven by the massive energy consumption and carbon dioxide (CO₂) emissions associated with cement production [16,17]. Among these sustainable alternatives, reactive MgO-based carbonation stabilization has emerged as a promising ground improvement technology, owing to its dual functions of soil strength enhancement and CO₂ sequestration [18,19]. In this technique, reactive MgO firstly undergoes hydration to form brucite (Mg(OH)₂), which subsequently reacts with CO₂ to generate various hydrated magnesium carbonates (HMCs), including nesquehonite, dypingite, hydromagnesite [20,21,22]. These carbonation products fill interparticle pores and bridge discrete soil particles, constructing a denser soil microstructure and thus achieving rapid strength growth and improved engineering performance of treated soils [23,24,25,26,27].

Compared with traditional PC stabilization, MgO-CO₂ carbonation possesses multiple superiorities, such as rapid early strength development, lower pore solution alkalinity after carbonation, reduced carbon footprint and refined pore structure [28,29]. Both laboratory tests and model-scale experiments have verified that pressurized CO₂ carbonation can significantly improve the mechanical properties of MgO-treated soils, and the cementation and pore-filling effects of nesquehonite, dypingite and hydromagnesite are the dominant contributors to strength enhancement [30,31]. Furthermore, recent advances in bio-carbonation, internal carbonation and MgO-based cementitious systems have further proven that carbonation efficiency, moisture condition, CO concentration, MgO reactivity and carbonation duration jointly determine the formation of carbonate phases and the final mechanical performance of stabilized soils [23,32,33]. Despite the above research progress, existing studies on reactive MgO carbonation stabilization mainly focus on sand, silt, clay, contaminated soil and construction waste, while organic-rich soils have received limited research attention [20,34]. More importantly, the distinct influences of specific organic matter fractions (especially HA and FA) on MgO hydration, CO₂ diffusion, carbonate precipitation and microstructural evolution remain poorly understood [13,35]. Given the significant differences between HA and FA in particle morphology, solubility, molecular size, functional group composition and metal-ion complexation capacity, the two substances may interfere with the MgO carbonation process through disparate pathways [36,37]. For instance, organic acids can reduce the alkalinity of soil pore solutions, adsorb Mg released from MgO hydration, alter carbonate nucleation behavior, and modify the morphology and yield of magnesium carbonate products [22,38,39].

Despite the recognized potential of reactive MgO-based carbonation as a sustainable alternative to Portland cement, its performance in organic-rich soils remains poorly understood. Specifically, the disparate inhibitory pathways of humic acid (HA) and fulvic acid (FA) on MgO hydration and carbonation require systematic differentiation. This study comparatively investigates the macroscopic and microstructural evolution of silty clay stabilized by PC, MgO, and MgO-CO₂ carbonation across varying HA and FA dosages (0–8%). By integrating mechanical and multi-scale microscopic analyses (UCS, pH, XRD, SEM, MIP), we aim to quantify the anti-interference robustness of carbonation stabilization, identify the mechanisms by which organic fractions inhibit carbonate nucleation versus pore-filling, and establish the critical organic matter thresholds for low-carbon ground improvement in heterogeneous soils. These findings provide a theoretical framework for deploying carbonation technology in challenging organic-rich foundations.

2. Experimental Plan and Methods

2.1. Materials

The raw materials used in this study consisted of silty clay, reactive MgO, high-purity carbon dioxide (CO₂), and organic matter. The silty clay was collected from an in situ construction site in Nanjing, China. Prior to testing, the retrieved soil sample was air-dried to constant mass, mechanically crushed using a geotechnical hammer, and sieved through a 1 mm mesh sieve to remove coarse impurities and ensure uniform particle size. The reactive MgO powder, supplied from Hebei Meishen Biotechnology Co., Ltd. (Xingtai, China), was a white powdery material with a measured reactivity index of 100, as determined via the iodine adsorption method. PC was purchased from a hardware store, appearing as a gray powder, with the brand Conch and a 32.5 strength grade. Commercial CO₂ gas with a purity of 99% was purchased from Yingju Gas Co., Ltd., in Nanjing, China. A series of fundamental physicochemical properties of the test soil, including natural water content, liquid and plastic limits, specific gravity, pH, electrical conductivity, particle size distribution, and chemical composition, were measured in accordance with the standard specifications of JTG 3430-2020 [40] and GB/T 50123-2019 [41]. The primary physicochemical parameters and chemical composition of the soil are summarized in

Table 1. Physical–chemical properties and main chemical composition of test soils.

Physicochemical Properties		Chemical Composition
Parameter	Value	Constituent	Proportion/%
w/%	22.3	MgO	2.61
wL/%	41.4	Al2O3	17.40
wP/%	20.7	CaO	1.25
IP	20.7	SiO2	64.30
GS	2.67	Fe2O3	8.00
pH	8.39	K2O	3.10
EC/(mS/cm)	0.38	TiO2	1.18

, and its particle size distribution curve is illustrated in Figure 1.

Table 2. Main chemical composition of improved materials.

Material	MgO	Al2O3	CaO	SiO2	Fe2O3	K2O	TiO2
MgO	93.50	0.21	4.91	1.02	0.10	0.16	-
PC	5.90	5.53	62.00	18.30	2.69	0.69	0.18

presents the main chemical composition of the stabilizing materials. Two typical analytical-grade organic matters, namely fulvic acid (FA) and humic acid (HA), were adopted in this experiment. The morphological features and particle size distributions of FA and HA are displayed in Figure 2. Considering the water solubility of organic matter, absolute ethanol was selected as the dispersing medium for testing. It can be clearly observed that FA contains a higher content of coarse particles compared with HA.

2.2. Sample Preparation and Carbonation Curing

This study was designed to investigate the effects of organic matter characteristics on the performance of modified soil under different curing regimes, with organic matter dosage and type as the core variables. The experimental variables included six organic matter dosage levels (0%, 0.5%, 1%, 2%, 4%, and 8%) and two typical organic matter types, namely fulvic acid (FA) and humic acid (HA). Meanwhile, three distinct curing methods were adopted: ordinary Portland cement (PC) curing, reactive magnesium oxide (MgO) curing, and MgO carbonation curing, to systematically explore the coupled influence of organic matter conditions and curing approaches on the properties of modified soil materials.

All specimens in this test were prepared strictly in accordance with the Chinese Standard for Geotechnical Testing Methods (GB/T 50123-2019). Firstly, the field-collected test soil was naturally air-dried for 48 h and then oven-dried at a constant temperature of 105 °C for 24 h, after which the dried soil blocks were crushed and sieved through a 1 mm standard sieve to obtain qualified test soil samples. Subsequently, according to the preset target density of specimens (approximately 1.89 g/cm³) and the internal geometric volume of the test mold, the required masses of dry soil, reactive MgO, deionized water, and organic matter were accurately calculated and weighed, where the organic matter content was defined as the mass ratio of organic matter to dry soil. In the material mixing process, dry components including sieved dry soil and organic matter were poured into a mixing container and stirred uniformly for 3 min; water was then added in three separate batches, with 30 s of stirring after each water addition, and the total mixture was continuously stirred for 5 min after the final water supplementation to ensure uniform mixing. Standard cylindrical molds with an inner diameter and height of 50 mm (50 mm × 50 mm) were adopted for specimen fabrication. The well-mixed mixture was filled into the mold in three layers, and each layer was compacted with a metal rod; the surface of each compacted layer was scored appropriately to enhance interlayer bonding, and the top surface was smoothed after the completion of filling. The mass error of all parallel specimens was controlled within ±2 g to guarantee test consistency.

After fabrication, all specimens together with molds were transferred to an indoor curing chamber with a constant temperature of 20 ± 5 °C for standard curing. The specimens were kept in this chamber for a duration of 28 days to allow for pre-curing. Upon the completion of pre-curing, the specimens were demolded, and their basic dimensions and mass parameters were recorded. Then, a barrel-type carbonation test device was used for carbonation treatment (Figure 3). To ensure the reproducibility of the experiment, the carbonation process was strictly controlled under a constant temperature of 20 °C, a relative humidity of 70%, and a gas supply pressure of 200 kPa. Two parallel MgO-stabilized specimens from each organic matter dosage group were placed vertically in the carbonation barrel in a vertical stack (one above the other), which helped minimize the adverse effects of overburden pressure and bottom moisture accumulation on the carbonation uniformity of the lower specimens. After 6 h of continuous carbonation, the specimens were taken out and placed in a natural environment for 24 h of standing stabilization. Afterwards, the dimensional parameters, mass, and unconfined compressive strength of the specimens were retested, and some crushed soil samples were collected to measure the water content and pH value of the modified soil.

2.3. Test Methods

2.3.1. Physical Tests

The dimensional and mass parameters of specimens were measured to calculate the basic physical properties before and after carbonation curing. Specifically, the diameter and length of cylindrical specimens were measured via a vernier caliper, and the specimen mass was weighed using an electronic balance. The volume of specimens before carbonation was approximated as the inner volume of the used mold, while the post-carbonation volume was calculated based on the measured dimensional parameters of demoulded specimens. The specimen mass change rate before and after treatment was calculated according to Equation (1):

$${\delta }_{\mathrm{m}}={\displaystyle \frac{m_2-m_1}{m_1}}\times 100\%$$

(1)

where δ_m is the mass change rate of the specimen, m₁ is the specimen mass before treatment (g), and m₂ is the specimen mass after treatment (g). This index was adopted to characterize the CO₂ absorption capacity of MgO-carbonated specimens.

The water content of failed specimens after carbonation was determined by the drying method and calculated via Equation (2):

$$w={\displaystyle \frac{m_{\mathrm{W}}-m_{\mathrm{D}}}{m_{\mathrm{D}}-m_{\mathrm{C}}}}\times 100\%$$

(2)

where w represents the specimen water content (%), m_W and m_D are the masses of the aluminum container filled with wet and dried crushed soil specimens (g), respectively, and m_C is the mass of the empty aluminum container (g), as above.

2.3.2. pH Test

The pore fluid pH value of carbonated soil specimens was tested in accordance with ASTM D4972-19 (2019) standard [42]. The failed carbonated specimens were crushed and sieved to obtain soil particles with a size less than 2 mm. A total of 10 g of the sieved soil sample was placed in a 20 mL plastic container, mixed with 10 mL of distilled water, and stirred thoroughly. After 1 h of static standing, the suspension pH value was measured using a PHS-3C digital pH meter (Shanghai Yidian Scientific Instrument Co., Ltd., Shanghai, China). To ensure test accuracy, the reading was recorded after numerical stabilization, and three parallel measurements were conducted for each specimen to obtain the average value.

2.3.3. Unconfined Compressive Strength Test

The unconfined compressive strength test of carbonated specimens was performed following the T0148-1993 [43]. The test was carried out on a microcomputer-controlled electronic multifunctional permafrost testing machine (WDT-100B (Jinan Ruipu Instrument Co., Ltd., Jinan, China)) under zero lateral confining stress. A constant loading rate of 1 mm/min was adopted, corresponding to an axial strain rate of 1~3% per minute, ensuring the test was completed within 8 min. The test was terminated when the specimen was crushed and damaged, and the failure morphology of each specimen was recorded synchronously. The UCS of each specimen was calculated based on the maximum failure load, and the final strength value was determined by averaging the test results of parallel samples. After the mechanical test, typical fragment samples were collected from the failed specimens, dried in a 60 °C oven, and used for subsequent moisture content and microscopic tests.

2.3.4. Microscopic Tests

To explore the influence mechanism of organic matter on soil carbonation and clarify the soil improvement mechanism, systematic microscopic tests including X-ray diffraction (XRD), scanning electron microscopy (SEM), and mercury intrusion porosimetry (MIP) were conducted on typical specimens with different organic matter contents. All specimen pretreatment procedures were unified to preserve the original microstructure and chemical properties of carbonated soil. Representative small fragments (approximately 1 cm³) of failed specimens were rapidly frozen with liquid nitrogen (−195 °C) and then vacuum freeze-dried at −80 °C for about 48 h using an FD5-series freeze dryer (Beijing Boyikang Experimental Instrument Co., Ltd., Beijing, China). The frozen water in the specimens was removed by sublimation to avoid microstructure damage caused by conventional high-temperature drying. The freeze-dried samples were divided into three parts for XRD, SEM, and MIP tests respectively.

For XRD mineral composition analysis, the freeze-dried samples were ground into fine powder with an agate mortar and sieved through a 75 μm geo-sieve. A Rigaku Ultima IV multifunctional horizontal X-ray diffractometer (Rigaku Corporation, Tokyo, Japan) was used for scanning under Cu-Kα radiation (wavelength λ = 1.54059 Å, working parameters: 40 kV, 30 mA). The scanning range of 2θ was 5~65°, with a step length of 0.02° and a scanning speed of 10°/min, to identify the mineral phase changes generated by carbonation reaction. For SEM microscopic morphology observation, the freeze-dried soil fragments were gold-sprayed and tested via an FEI Quanta 200 environmental scanning electron microscope (FEI Company, Hillsboro, OR, USA) with a high voltage of 20 kV. The microscopic morphology of specimens was observed under magnifications of 1500 times and 5000 times to analyze the structural changes before and after carbonation. The pore structure characteristics of specimens were tested by an AutoPore IV 9510 automatic mercury intrusion porosimeter (Micromeritics Instrument Corporation, Norcross, GA, USA). The test pore diameter range was 0.004~340 μm, and the maximum intrusion pressure reached 345 MPa, which was used to quantitatively analyze the pore size distribution and porosity changes of carbonated soil.

3. Results and Analysis

3.1. Visual Appearance and Physicochemical Properties

Figure 4 illustrates the changes in the appearance of specimens with different dosages of FA. Due to the different degrees of reaction of FA and HA with water, as well as the differences in the types of stabilizers, different phenomena occurred during the specimen preparation process. The following observations were made: (1) With increasing organic matter content, the PC-mixed soil appeared relatively moist, whereas the MgO-mixed soil appeared relatively dry. (2) FA powder was relatively fine and became viscous upon contact with water; when mixed with water, it readily formed small soil lumps of high viscosity, which required manual crushing during mixing. (3) As the FA content increased, the viscosity of the mixed soil increased, and demolding became more difficult. At an FA content of 8.0%, the mixture already exhibited very high viscosity during mixing, and the specimens prepared by static compaction showed high viscosity, high porosity, low strength, difficulty in forming and demolding, and poor shape retention after demolding.

After specimen preparation, the changes in mass and dimensions before and after carbonation (or curing) were measured. Figure 5 shows the mass change rates for different organic matter types and dosages before and after curing or carbonation. For the FA-containing specimens, the carbonation-treated group consistently exhibited the highest mass change rate, showing an initial increase followed by a gradual decrease with curing time. This trend indicates that CO₂ uptake and carbonate precipitation contributed significantly to mass gain during the early stage, whereas the carbonation reaction gradually approached equilibrium at later ages. In contrast, the MgO-cured specimens displayed a continuous increase in mass change rate before reaching a relatively stable level, suggesting the sustained formation of hydration and carbonation products. The PC-cured specimens showed a progressive decline in mass change rate with increasing curing age, implying that the contribution of hydration-induced mass gain was gradually offset by moisture loss and microstructural evolution.

A similar trend was observed in the HA-containing specimens, where the carbonation-treated group maintained the highest and most stable mass change rate throughout the curing period, indicating the persistent effectiveness of carbonation in promoting mass accumulation. The MgO-cured specimens exhibited a noticeable increase in mass change rate at later ages, reflecting the enhanced development of secondary reaction products. By comparison, the PC-cured specimens showed relatively large fluctuations and lower mass change rates, suggesting that HA may inhibit PC hydration to some extent and consequently weaken the mass gain effect. Overall, carbonation treatment resulted in the most significant mass increase for both organic matter systems, while MgO curing demonstrated a pronounced mass-gain effect during the later curing stage. In contrast, conventional PC curing was less effective, particularly in the presence of HA, where the mass change rate remained relatively low and unstable.

Figure 6 illustrates the variation in dry density of specimens containing different contents of FA and HA under three curing regimes: PC curing, MgO curing, and MgO carbonation curing. All curing methods increased the dry density of the stabilized soils, but the extent of densification varied among them. Under PC curing, the dry density increased moderately, with FA-modified specimens reaching the highest values at low dosages (0–0.5%), while higher FA contents slightly reduced the density. HA-modified specimens showed a similar trend, although the decrease with increasing dosage was less pronounced. MgO hydration curing further enhanced the dry density compared with PC curing, particularly for specimens with moderate FA or HA contents, while higher additive levels limited the improvement. The most significant densification occurred under MgO carbonation curing, with dry density values generally 0.15–0.20 g/cm³ higher than those under PC curing and 0.05–0.10 g/cm³ higher than under MgO hydration curing. Across all curing regimes, FA-modified specimens consistently exhibited slightly higher densities than HA-modified specimens, especially at low to moderate dosages, indicating that the type of organic additive influences the degree of densification. Overall, the results demonstrate a clear hierarchy in densification efficiency: MgO carbonation curing > MgO curing > PC curing, and show that both the type and dosage of organic additives, together with the curing method, strongly affect the final dry density of the stabilized soils.

3.2. Unconfined Compressive Strength

After curing or carbonation, the surface moisture of the specimens was wiped off, and their dimensions and mass were measured. Subsequently, unconfined compressive strength tests were conducted. The failure patterns of selected specimens are shown in Figure 7. The PC-cured specimens exhibited relatively low strength, and no obvious through-going cracks developed upon failure. For the carbonated specimen with an FA content of 8%, through-going cracks appeared, but the specimen did not split.

Figure 7 illustrates the effect of organic matter content on the UCS of specimens containing FA and HA under different curing conditions. For FA-containing specimens, carbonation-treated samples consistently exhibited the highest UCS. Interestingly, UCS initially decreased with FA content but showed a recovery at higher concentrations. This trend reflects a competing mechanism between chemical inhibition and physical reinforcement. At low FA dosages, the abundant carboxyl and phenolic hydroxyl groups in FA preferentially complex with Mg²⁺ ions, significantly suppressing the nucleation of strength-contributing carbonation products (e.g., nesquehonite). However, as FA dosage exceeds a critical threshold (~2.0%), the chemical inhibition effect tends toward saturation, while the physical contribution of FA becomes dominant. Its viscous nature promotes the formation of a cohesive organic-mineral interfacial layer that enhances inter-particle bonding, thereby compensating for the loss of rigid crystalline strength. In contrast, MgO-cured specimens showed limited response, and PC-cured specimens exhibited a sharp strength reduction, confirming that FA significantly inhibits PC hydration by hindering the formation of a stable cemented skeleton.

For HA-containing specimens, a similar superiority of carbonation treatment was observed. The UCS increased progressively with HA content, reaching its peak at the maximum investigated concentration. This recovery at high HA dosages suggests that the physical cohesive effect of HA, combined with its ability to form a denser carbonated microstructure, effectively overcomes the initial chemical retardation observed at low dosages. While PC-cured specimens showed moderate strength with an initial decrease followed by a slight recovery, MgO-cured specimens displayed the lowest UCS and a consistent downward trend, indicating that HA exerts a more persistent negative influence on MgO hydration than FA.

Overall, increasing organic matter content generally exerts a negative effect on strength development, particularly under conventional PC and MgO curing conditions. However, carbonation treatment effectively mitigates the adverse influence of both FA and HA, resulting in substantially higher UCS values. The experimental data reveals that the influence of OM on strength is not merely inhibitory but follows a dual-role pattern: acting as a crystallization inhibitor at low concentrations and a cohesive physical binder at higher concentrations. This competing mechanism explains the strength recovery trend observed, highlighting that carbonation treatment possesses a unique capacity to leverage the cohesive properties of organic matter to enhance the mechanical performance of organic-rich soils.

3.3. pH Variation

Figure 8 presents the variation in pH value with increasing organic matter content for specimens containing FA and HA under different curing conditions. For the FA-containing specimens, the pH value of the PC-cured group was consistently higher than those of the MgO-cured and carbonation-treated groups throughout the investigated range. However, with increasing FA content, the pH value exhibited a pronounced downward trend, particularly at higher concentrations, indicating that FA effectively consumed alkaline components generated during PC hydration and weakened the alkaline environment of the system. In contrast, the MgO-cured specimens showed relatively stable pH values with only minor fluctuations, suggesting that the buffering capacity provided by MgO hydration products partially offset the acidification effect of FA. The carbonation-treated specimens maintained the lowest pH values and exhibited only slight variations with increasing FA content. This behavior can be attributed to the consumption of hydroxyl ions during carbonation and the formation of stable carbonate phases, which reduced the overall alkalinity of the system.

A similar trend was observed for the HA-containing specimens, although the magnitude of the variation differed from that observed in the FA system. The PC-cured specimens maintained the highest pH values among all treatment groups, indicating that PC hydration continuously supplied alkaline substances despite the presence of HA. Nevertheless, a gradual decline in pH value was still evident with increasing HA content, reflecting the neutralization effect of humic functional groups on the alkaline environment. The MgO-cured specimens exhibited moderate pH values and a slight decreasing trend as HA content increased, while the carbonation-treated specimens showed the lowest pH values and a continuous decline throughout the investigated range. Compared with FA, HA exerted a relatively weaker influence on the pH value of the PC-stabilized specimens but produced a more noticeable reduction in the carbonation-treated system.

Overall, increasing organic matter content reduced the alkalinity of all treated specimens to varying degrees. The reduction was most pronounced in the PC-cured groups, owing to the strong interaction between organic acids and hydration-generated alkaline products. MgO curing exhibited a comparatively stable pH response, demonstrating a certain buffering capacity against organic acid interference. Carbonation treatment resulted in the lowest pH values because the carbonation reaction consumed alkaline constituents and transformed them into carbonate minerals. These findings indicate that organic matter can substantially modify the chemical environment of stabilized soils, thereby influencing the hydration, carbonation, and strength development processes of different curing systems.

3.4. Microscopic Mechanism Analysis

3.4.1. XRD

After the XRD measurement, the obtained patterns were analyzed using JADE 6.5 software to identify the phases corresponding to the diffraction peaks. Subsequently, the XRD patterns of the samples were plotted using Origin 2021 software.

Table 3. Summary of XRD phase.

Name of Mineral	Symbol	Name of Mineral	Symbol
Quartz	Q	Dypingite	D
Kaolinite	K	Hydromagnesite	H
Brucite	B	Artinite	A
Portlandite	P	Hydrotalcite	Ht
Magnesite	M	Gypsum	G
Calcite	C	Calcium silicate hydrate	CSH
Nesquehonite	N	Dypingite	D

below lists the main products and their corresponding symbols.

Figure 9 shows XRD patterns of carbonated soil with different organic matter (OM) contents under various binder mixing ratios. As shown in the figure, significant nesquehonite (N) peaks can be detected at 2θ = 13.60°, 23.04°, 29.44°, 32.16°, 34.18°, 47.18°, and 53.30°, indicating that nesquehonite (N) is an important component of the carbonated soil. This result is consistent with the findings of Cai et al. Compared with the specimen without OM, the specimen containing 6% OM exhibits weaker nesquehonite (N) peak intensities, and brucite (B) peaks with higher intensities are detected at 2θ = 18.59°, 38.09°, and 58.72°, suggesting that the presence of OM hinders the carbonation reaction of brucite. According to Liu Songyu’s work [44], nesquehonite (N), dypingite (D), and hydromagnesite (H) are important components contributing to the strength of carbonated soil. The peak intensities of N, D, and H in the carbonated soil with 6% OM are significantly lower than those in the specimen with 0% OM, which is a key reason for the lower strength of the carbonated soil containing 6% OM.

Figure 10 also shows the XRD patterns of carbonated soils containing different types of organic matter (OM) under various binder mixing ratios. The product compositions of the carbonated soils containing different types of OM are consistent with those shown in the figure. For the FA-containing specimens, only dypingite (D) and hydromagnesite (H) were detected at 2θ values of 8.86°, 19.87°, 22.09°, 25.54°, and 41.87°, whereas nesquehonite (N) was scarcely detected. The HA- and HA-Na-containing specimens exhibited similar carbonation product peaks. However, at 2θ values of 29.65° and 34.38°, the N peak intensity of the HA-Na-containing specimen was slightly higher than that of the HA-containing specimen, which may account for the slightly higher strength of the carbonated HA-Na-containing specimen compared with its HA-containing counterpart.

3.4.2. SEM

Figure 11 presents SEM images of specimens with organic matter (OM) contents of 0% and 6% (FA and HA). As shown in the figure, in the absence of OM, abundant rod-like nesquehonite (N) in the carbonated specimen interconnected the soil particles, forming a particle-cemented structure and filling the interparticle pores, thereby rendering the internal soil structure dense. When the OM content was increased to 6%, product N was scarcely observed in the microstructure. Instead, a large number of spherical carbonate products with smooth surfaces were distributed throughout the interior of the specimen. Based on a literature review, these products were identified as hydromagnesite (H).

Furthermore, in the present study, the morphology of product H differed from that reported in previous studies. This discrepancy may be related to the type of reactive MgO used and the degree of carbonation. Moreover, although product H was abundantly distributed in the carbonated specimen containing 6% FA, it did not effectively cement the soil particles, resulting in a relatively loose internal structure. When HA was used as the OM type, nesquehonite (N) and brucite (B), the hydration product of MgO, were clearly observed within the specimen. Compared with the OM-free specimen, the internal structure of the HA-containing specimen was looser, which explains why, under the same binder dosage, the strength of the carbonated HA-containing specimen was slightly lower than that of the OM-free specimen.

3.4.3. Pore Characteristics of Stabilized Soil

Existing studies [45] have indicated that the influence of organic matter content on the strength of PC-stabilized soil can be interpreted from two perspectives. From the viewpoint of chemical interactions, the presence of organic matter reduces the pH of the pore solution, adsorbs calcium ions required for the formation of hydration products, and simultaneously decomposes the hydration products. These effects lead to a reduction in the final amount of hydration products, thereby decreasing the strength of the stabilized soil. As shown in Figure 12. From the structural perspective, when PC is added to silt, the hydration process generates products that bind soil particles together, forming a network skeleton structure that enhances the structural strength of the soil. When organic matter is present in the silt, it can hinder and delay the PC hydration reaction. The higher the organic matter content, the greater the hindrance and delay, and consequently the lower the strength of the skeleton.

The study by Fan Zhaoping [46] showed that there exists a critical content of organic matter that limits its effect on the solidification efficiency of silt. When the organic matter content is below this critical value, the strength of the stabilized soil decreases with increasing organic matter content. When the organic matter content exceeds this critical value, further increases in organic matter content have little effect on the strength gain of the stabilized soil.

4. Discussion

The scientific advancement of this work is that it distinguishes the effects of two representative organic fractions, fulvic acid (FA) and humic acid (HA), on PC curing, MgO curing, and MgO carbonation stabilization under comparable conditions. Unlike previous studies that mainly focused on the overall influence of organic matter content, this study links the chemical characteristics of FA and HA with strength development, pH variation, pore structure evolution, and microstructural observations. Therefore, the main novelty lies in clarifying the different interference mechanisms of FA and HA during MgO carbonation and demonstrating the stronger resistance of carbonation stabilization to organic matter compared with PC and uncarbonated MgO stabilization.

The higher mass gain, dry density, and UCS of the MgO-carbonated specimens can be attributed to MgO hydration and subsequent carbonation. During this process, MgO hydrates to form Mg(OH)₂, which then reacts with dissolved CO₂ to generate hydrated magnesium carbonates such as nesquehonite, dypingite, and hydromagnesite. These products fill pores, bridge soil particles, and form a denser carbonate-bonded skeleton. The decrease in pH after carbonation also indicates the consumption of alkaline species during carbonate formation. However, the mass-change index used here only provides an indirect indication of CO₂ uptake rather than a direct quantitative carbonation degree. In addition, the XRD and SEM results are mainly qualitative because the current XRD data were not collected with an internal standard and the SEM images were not obtained for statistically rigorous image analysis. Therefore, the interpretation of carbonation efficiency and microstructural evolution is discussed cautiously. Future studies should include direct CO₂ uptake measurement, TGA, calibrated Rietveld refinement, and quantitative image analysis.

The non-monotonic UCS variation with increasing FA or HA content reflects the competition between chemical inhibition and physical contribution. At low organic matter contents, functional groups such as carboxyl, phenolic hydroxyl, and carbonyl groups can complex Mg²⁺, reduce alkalinity, and inhibit the nucleation and growth of hydrated magnesium carbonate crystals, leading to strength reduction. At higher organic matter contents, this chemical inhibition may approach saturation, while the physical effects of organic matter become more evident. FA can form viscous organic-mineral coatings, whereas HA particles may enhance local interlocking and frictional resistance. These effects partly compensate for the loss of carbonate cementation and explain the partial UCS recovery observed at higher FA or HA contents.

FA and HA affect MgO carbonation through different pathways. FA has higher solubility and mobility, allowing it to interact more strongly with Mg-bearing species and reduce the availability of free Mg²⁺ for carbonate precipitation. This explains the more pronounced strength reduction and higher dosage sensitivity in FA-containing specimens. In contrast, HA is less soluble and more particulate, so its influence is more localized and related to adsorption, pore blocking, water retention, and particle-scale interaction. As a result, HA shows a relatively weaker but more persistent influence on MgO hydration and carbonation.

The SEM and MIP results further support this interpretation. In specimens without added organic matter, carbonate products form relatively continuous particle bonds. After FA or HA addition, the distribution of carbonation products becomes less uniform, and the structure changes from a carbonate-dominated cemented skeleton to a mixed organic-mineral-carbonate system. Although MIP shows that pore volume may decrease in some organic-matter-bearing specimens, lower porosity does not necessarily produce higher strength because strength also depends on the rigidity, continuity, and bonding capacity of the solid skeleton. Organic coatings may refine pores but weaken carbonate crystal interlocking, which explains the inconsistent relationship between porosity and UCS.

The reliability of the macroscopic trends is supported by the consistency among UCS, pH, dry density, XRD, SEM, and MIP results. Nevertheless, organic-rich stabilized soils are heterogeneous, especially at high FA or HA contents. Therefore, the revised figures should present mean values with error bars or standard deviations where available, and the discussion should focus on consistent trends rather than isolated data points.

From an engineering perspective, MgO carbonation stabilization shows potential for organic-rich soft soils because it provides higher early strength, denser structure, lower alkalinity after carbonation, and possible CO₂ sequestration benefits. However, field application still requires effective CO₂ delivery, moisture control, adequate gas–soil contact, and suitable pre-curing to promote MgO hydration before carbonation. This technique may be more feasible for shallow mixing layers, road subgrades, or prefabricated treated-soil elements, whereas deep in situ stabilization may be limited by CO₂ diffusion and carbonation uniformity. Long-term durability, field-scale carbonation efficiency, and economic feasibility should be further evaluated before practical implementation.

5. Conclusions

Based on the above investigations and analyses, the following conclusions and implications can be drawn:

(1)

This study clarifies the scientific advancement of applying reactive MgO carbonation to organic-rich soils by separately evaluating the effects of FA and HA and by comparing PC stabilization, MgO stabilization, and MgO carbonation under the same organic-matter dosage range. The results show that MgO carbonation has stronger resistance to organic-matter interference than conventional PC and uncarbonated MgO stabilization.

(2)

MgO carbonation produced the most favorable macroscopic performance among the three treatment methods. The higher mass change, dry density, and UCS of the carbonated specimens are attributed to CO₂ absorption and the formation of hydrated magnesium carbonates, which fill pores and cement soil particles. However, mass change should be regarded as an indirect indicator of carbonation, and direct measurements such as CO₂ uptake or TGA are recommended in future work to quantify the carbonation degree.

(3)

The influence of organic matter on strength development is governed by competing chemical and physical mechanisms. At low FA or HA contents, organic functional groups complex Mg²⁺ ions, reduce alkalinity, and inhibit carbonate nucleation, resulting in strength reduction. At higher contents, the viscous or particulate nature of organic matter can increase cohesion, particle interlocking, or organic-mineral bonding, which partly explains the UCS recovery observed after the initial decrease. Therefore, the effect of organic matter should not be interpreted as purely inhibitory.

(4)

FA and HA affect MgO carbonation differently. FA shows a stronger inhibitory effect because of its higher solubility, mobility, and interaction with Mg-bearing species, whereas HA mainly acts through localized adsorption, pore blocking, and frictional effects. This difference explains the lower minimum UCS and greater dosage sensitivity of FA-containing specimens compared with HA-containing specimens.

(5)

XRD and SEM observations indicate that nesquehonite, dypingite, and hydromagnesite are the main carbonation products, while organic matter changes their morphology and distribution. Together with MIP results, the microstructural evidence suggests that strength is controlled not only by total porosity but also by the continuity and rigidity of carbonate bonding. Because the present XRD and SEM analyses are qualitative, future studies should include Rietveld refinement, quantitative image analysis, and direct carbonation measurements to strengthen the microstructural interpretation.