Distribution Characteristics of Cadmium in Soil Aggregates and Their Regulating Effects on Cd Bioavailability

Abstract

Soil aggregates play critical roles in regulating the behavior of heavy metal in soils. To understand the distribution of cadmium (Cd) in aggregates of different soil types, as well as their roles in regulating the Cd bioavailability of bulk soils, four major arable soils, including acidic, neutral, and calcareous purple soils and calcareous yellow soil (APS, NPS, CPS, and CYS), were sampled from Chongqing, China, for aggregate separation and determination of the total Cd(T-Cd) distribution, fractionation, and extractability in various-sized aggregates. A pot experiment with ryegrass (Lolium perenne L.) was conducted to evaluate the Cd bioavailability in bulk soils as influenced by aggregates. The results show that the composition of soil aggregates varies a lot among soils: lower soil pH tends to increase the proportion of macroaggregates while decreasing that of smaller aggregates. The Cd distribution, HCl-extractability, and active fraction (AF, T-Cd/HCl-Cd) in aggregates are all soil type-dependent, with pH and particle size being the main determining factors; the distribution pattern of Cd concentrated in smaller aggregates is only found for CPS and CYS (pH > 7.5) upon exogenous Cd addition, though the finest aggregates (silt–clay, <0.053 mm) consistently exhibited the highest Cd enrichment for all tested soils. The Cd extractability and AF values in all aggregates show a sequence of APS > NPS > CPS > CYS, indicating the fundamental influence of soil pH on Cd availability. Higher AF values over bulk soils, either in silt–clay aggregates or in microaggregates (0.053–0.25 mm), whereas lower AF in macroaggregates (1–2 mm) are found for APS and NPS, which correspond to the relative portions of Ex-Cd and Fe/Mn oxide-bound Cd (Fe/Mn-Cd) in these aggregates. In contrast, less variation of AF values among aggregates is observed for CPS and CYS and for APS/NPS upon Cd addition. Pot experiments demonstrated strong positive correlations between ryegrass Cd uptake and HCl-Cd in silt–clay aggregates and T-Cd in microaggregates, while a negative correlation was observed with T-Cd in macroaggregates. These findings supply new insight into the mechanisms of aggregates in controlling Cd bioavailability in bulk soils and shed light on the development of new strategies for remediating Cd-polluted soils.

1. Introduction

Heavy metal contamination of soils has been an environmental concern worldwide due to its widespread sources, high toxicity, and non-biodegradability. According to the National Soil Pollution Survey Report of China, cadmium (Cd) is the most widespread soil pollutant in China, with 7.0% of soil sampling sites exceeding the risk screening values set by the “Risk control standard for soil contamination of agricultural land in China (GB15618-2018)”. Understanding the Cd distribution and transformation in soils is crucial for controlling farmland Cd pollution, reducing its accumulation in the food chain, and ensuring sustainable agriculture [1,2,3]. Bioavailability can be defined as the amounts of any given metal that can be taken up by plants or microorganisms such that a physiological response is observable [4]. Generally, it is believed that the bioavailability of metals in soils depends not only on the total concentration but also on their specific chemical forms [5]. To date, considerable research has been conducted for understanding heavy metals’ migration and transformation processes in soils and their impacts on the growth of plants [6,7,8]. However, the roles of soil aggregates in regulating the bioavailability of heavy metals have been overlooked and remained unclear.

Soil aggregates, formed by soil particles conjoined with various cementing matters such as organic matter, are the basic functional units in soil [9,10], which regulate the physicochemical and biological properties of soils and thus deeply affect the behaviors of nutrients and pollutants in soils [11,12]. Hence, understanding the distribution, forms, and migration characteristics of heavy metals in soil aggregates is crucial. Commonly, fine-sized soil aggregates have a higher capacity for Cd adsorption due to their larger specific surface area and higher active metal (hydro)oxide content [13,14,15]. Duong, Schlenk, Chang, and Kim [16] reported that the majority of elements involved in soil heavy metal contamination, including Cu, Zn, and Pb, are preferentially concentrated in tiny particle-sized portions, with the tiniest particles typically showing the highest contents. Acosta, Faz, Kalbitz, Jansen, and Martínez-Martínez [17] also found that the contents of heavy metals like Cd, Cu, and Zn in soil aggregates increase dramatically as their particle size steadily decreases, and their adsorption capability also increases. However, Lombi, Sletten, and Wenzel [18] observed that the distribution of As varies noticeably across aggregate particles of different sizes in As-contaminated soil, with As in larger aggregate particles having higher biological toxicity. Under certain circumstances, small-sized soil aggregate particles can aggregate into large-sized particles, and the presence of heavy metals in large-sized particles will also be significantly enriched. The distribution of heavy metals among varying sizes of aggregates differs greatly among soil types. Furthermore, there may be variations in the chemical forms and bioavailability of heavy metals in various aggregates. Li, Wu, Luo, and Christie [19] found that hyperaccumulators could uptake more Cd and Zn from coarse particle fractions than from fine particles.

To date, few studies have addressed the differences of the distribution characteristics of Cd in aggregates among various soil types, and how they regulate the Cd availability in bulk soil remains unclear. In this paper, we compared the distribution patterns of both native and exogenous Cd in various-sized aggregates in four types of typical farmland soils with apparent differences in basic properties, and the existing states of native Cd and the transformation of exogenous Cd in aggregates of different soils were also investigated by sequential fractionation. In combination with a pot experiment, where ryegrass (Lolium perenne L.) was used as an indicator, the bioavailability of Cd in soils with varied Cd distribution patterns and the existing states in soil aggregates were evaluated, and thus the roles of soil aggregates in regulating Cd availability in bulk soils were elucidated via correlation analyses. The results supplied bases for the risk assessment of soil Cd pollution and the precise remediation of Cd-contaminated soils.

2. Materials and Methods

2.1. Soil Preparation

The tested soils represented the main farmland soil types in southwest China: acidic purple soil (APS), neutral purple soil (NPS), calcareous purple soil (CPS), and calcareous yellow soil (CYS). According to the World Reference Base for Soil Resources (WRB, 2022), APS and NPS are classified as Cambisol, while CPS and CYS are classified as Calcisol. Soil samples were collected from Jiangjin, Hechuan, Beibei, and Tongnan districts (Chongqing, China) in May 2024, each corresponding to the respective soil types. From each site, the topsoil layer (0–20 cm) was sampled, air-dried at room temperature (~25 °C), and homogenized. Visible impurities (e.g., plant roots, gravel) were manually removed before sieving through a 2 mm mesh to ensure uniformity for subsequent analyses. The basic physicochemical properties of the tested soils are shown in

Table 1. Basic physicochemical properties of soil.

Soil Type	pH	Organic Matter /g·kg−1	Available N /mg·kg−1	Available P /mg·kg−1	Available K /mg·kg−1	CEC /cmol·kg−1	Total Cd /mg·kg−1	Soil Texture %
								Sand	Silt	Clay
APS	6.04	29.42	205.59	23.40	88.28	22.00	0.39	32.00	36.00	32.00
NPS	7.30	31.63	105.91	24.26	233.22	24.71	0.40	44.00	34.00	22.00
CPS	7.81	22.55	80.99	25.69	161.04	23.00	0.42	8.00	30.00	62.00
CYS	7.67	19.01	67.28	24.93	193.48	22.14	0.45	20.00	44.00	36.00

. The total cadmium (T-Cd) content of all tested soils did not exceed the risk screening value stipulated in the Chinese standard “Soil Environmental Quality Agricultural Land Soil Pollution Risk Control Standard (Trial)” (GB 15618-2018).

2.2. Preparation of Cd-Contaminated Soil

To investigate the distribution of exogenous Cd in the aggregates of different soils, four levels of exogenous Cd were introduced into the soils using a Cd(NO₃)₂·4H₂O (≥99.0%, Macklin Biochemical Co., Ltd., Shanghai, China). solution. The soils were incubated in pots and aged for 90 days in a greenhouse (25 ± 2 °C, at Southwest University) to represent the control (CK, 0 mg·kg⁻¹), low (Cd1, 1 mg·kg⁻¹), medium (Cd2, 2 mg·kg⁻¹), and high (Cd4, 4 mg·kg⁻¹) pollution levels, in reference to GB 15618-2018. The soil moisture was maintained at 60% of the field water-holding capacity (WHC) through regular replenishment with deionized water during the aging period. All treatments were conducted in triplicate.

2.3. Soil Aggregates Separation

Soil aggregates were fractionated into five size classes via ultrasonic dispersion-coupled wet sieving (Cambardella and Elliot 1993) [20] using a soil aggregate analyzer (Model XY-100, Beijing Xiangyu Weiye Instrument Equipment Co., Ltd., Beijing, China): macroaggregate (1–2 mm), medium aggregate (0.5–1 mm), small aggregate (0.25–0.5 mm), microaggregate (0.053–0.25 mm), and silt–clay aggregate (<0.053 mm). The procedure for soil aggregate separation was as follows: 300 g of air-dried soil was evenly distributed onto the top sieve of a nested nylon sieve set, with aperture sizes decreasing sequentially from top to bottom. The sieve stack was immersed in deionized water for 15 min to pre-wet soil samples, followed by up–down oscillation with an oscillator (30 mm amplitude) for 5 min under ultrasonic dispersion (25 kHz, 0.2 W·cm⁻³, KQ-300DE, Kunshan Ultrasonic Instruments Co., Ltd., Kunshan, China). Aggregates retained on each sieve were rinsed with deionized water into pre-labeled containers, oven-dried at 50 °C for 48 h, and then weighed. Dried fractions were stored in airtight containers for subsequent analyses.

2.4. Soil Determination and Analysis

Soil pH was measured in a 2.5:1 water-to-soil suspension using a pH meter (PB-10, Sartorius AG, Göttingen, Germany). Soil organic matter (SOM) content was determined using the potassium dichromate volumetric method. Available nitrogen (AN) was measured using the alkaline hydrolysis diffusion method. Available potassium (AK) was extracted with 1.0 M ammonium acetate and measured by flame photometry. Available phosphorus (AP) was extracted with 0.5 M NaHCO₃ and measured using the molybdenum blue colorimetric method. All aggregate fractions (from the four soil types) and bulk soils were analyzed for total Cd (T-Cd), available Cd, and Cd fractionation. T-Cd was determined following acid digestion with HNO₃-HClO₄-HF, according to the method described by [21]. Available Cd content was measured by HCl extraction (soil-to-solution ratio of 1:10, with shaking for 2 h), and Cd speciation was determined via Tessier sequential extraction (five-step procedure). The Cd contents in all solutions were determined by graphite furnace atomic absorption spectrometry (GFAAS; PinAAcle 900T, PerkinElmer Inc., Waltham, MA, USA). The national standard materials of GBW07404 (GSS-4) and GBW07428 (GSS-14) were used as soil internal standard references for quality control. All analyses were conducted in triplicate.

2.5. Pot Experiment

A pot experiment was conducted in a greenhouse from October to December 2024 to examine the Cd bioavailability in bulk soils consisting of various aggregates, in which four Cd treatments (CK: 0, Cd1: 1, Cd2: 2, Cd4: 4 mg·kg⁻¹) were applied to each soil type (n = 4), with triplicate pots per treatment (total 48 pots = 4 soils × 4 treatments × 3 replicates). The aged contaminated soils (1.5 kg of soil per pot) were packed into PVC pots (diameter × height: 15 cm × 12 cm), and the soils were amended with urea 0.483 g·kg⁻¹, sodium dihydrogen phosphate 0.330 g·kg⁻¹, and potassium sulfate 0.416 g·kg⁻¹ as base fertilizers. Ryegrass (Lolium perenne L.) seeds were sown (10 seeds/pot) as a Cd bioaccumulation indicator. Soil moisture was maintained at 60% WHC by weekly replenishment with deionized water (adjusted by weighing each pot). At the two-true-leaf stage, seedlings were thinned to retain eight uniform plants per pot. After 60 days of cultivation, the plant height was recorded, and the aboveground and underground parts were separately harvested. Samples were rinsed three times with deionized water, blotted dry, and fresh weights recorded. After 30 min of enzyme deactivation at 105 °C, the plant samples were dried to a constant weight at 70 °C. The dry weight of each treatment was weighed and then crushed for analyses. Dried tissues were ground and microwave-digested (HNO₃:HClO₄, 3:1 v/v) for GFAAS Cd analyses (PinAAcle 900T, PerkinElmer Inc., Waltham, MA, USA). Standard addition recoveries and certified plant materials (GBW07603) were adopted in each batch of analyses for the control of analytical quality. The relative standard deviation (RSD) was maintained below 5%.

2.6. Data Analysis

The experimental data were primarily treated in Microsoft Excel 2007 (Microsoft Corporation, Redmond, WA, USA) and then processed in IBM SPSS Statistics 18.0 (IBM Corp., Armonk, NY, USA)for statistical analyses. OriginPro 8.5 (Origin Lab Corporation, Northampton, MA, USA) was used to plot the graphs. Statistical differences were analyzed using one-way ANOVA followed by Duncan’s multiple range test, with significance set at p < 0.05.

3. Results

3.1. Characteristics of Aggregate Mass Distribution

The mass recovery rates of the aggregate separation method ranged from 89% to 95%, indicating the method’s reliability. As shown in Figure 1, the composition of soil aggregates varies with soil types. In purple soils, soil pH was the key determinant of aggregate distribution patterns. As pH decreased, the mass fraction of silt–clay aggregate decreased from 49.6% in CPS to 23.8% in APS, while the fractions of other aggregates all increased; for example, the fraction of macroaggregate increased from 3.7% in CPS to 9.8% in APS. For calcareous yellow soil, mass fractions of aggregates vary less, except for the macroaggregate, which accounts for the least portion (4.8%). Consequently, the dominant aggregate fractions varied significantly among soil types, with small aggregate (24.2%) prevailing in APS, silt–clay aggregate (34.5% and 49.6%) in NPS and CPS, and medium aggregate (31.8%) in CYS. This may be attributed to the differences in soil properties and aggregating processes.

3.2. The Enrichment of T-Cd in Various Soil Aggregates

The T-Cd content in aggregates of different sizes reflects the degree of cadmium enrichment associated with each fraction. The recovery rates of T-Cd in native soils and soils spiked with 1, 2, and 4 mg·kg⁻¹ exogenous Cd were 87.8–104.8%, 89.1–100.1%, 86.7–99.8%, and 97.1–103.3%, respectively, indicating the reliability of the analytical method. The T-Cd contents in various-sized soil aggregates across the four soils types are presented in Figure 2.

In native soil (CK), the commonly reported trend of T-Cd enrichment in finer aggregate was not consistently observed in our study. Instead, the T-Cd enrichments in aggregates of various soils were particle size- and pH-dependent, with a particle size of 0.053–0.25 mm (the size of microaggregate) playing a critical role: for APS and NPS (pH < 7.5), the Cd contents in microaggregate were lowest, and lower- or higher-sized aggregates resulted in higher Cd enrichment, while for CPS and CYS (pH > 7.5), the opposite tendency was observed, with the Cd contents in microaggregate being highest. When exogenous Cd was added, its enrichment patterns in aggregates were in general similar to those of CK: microaggregate showed lower Cd enrichment in APS and NPS and higher enrichment in CPS and CYS. However, the smallest silt–clay aggregate showed the highest exogenous Cd enrichment for all soils, and a clear tendency of Cd concentrated in smaller aggregate was only found for CPS and CYS. Contrary to the conventional understanding that the smaller the particle size is, the higher its metal retention, our results emphasize that the enrichment of Cd in soil aggregates is not only particle size-determined but also differs with soil types, and soil pH shows a profound influence.

3.3. The Distribution of T-Cd Loads in Soil Aggregates

The T-Cd distribution in aggregates is affected by both the mass proportion of each aggregate and its corresponding Cd content. Therefore, the concept of heavy metal mass loading factors [22] is introduced to evaluate the mass contribution of Cd in each particle-sized aggregate in soils. The calculation formula is as follows:

$$GS{F}_{loading}=\left[{\displaystyle \frac{X_i\times G{S}_i}{{\sum }_{i=1}^5{X}_i\times G{S}_i}}\right]\times 100$$

where X_i represents the Cd content (mg·kg⁻¹) in aggregate size fraction I, and GS_i is the mass fraction of size i in the whole soil.

The results presented in Figure 3 indicate that the mass distribution of Cd in various aggregates in purple soils showed similar patterns with and without exogenous Cd addition: with increasing soil pH, the mass loading factors of Cd in the finest aggregate—silt–clay aggregate—significantly increased from 22% in APS to 47% in CPS for native soil, 26% in APS to 53% in CPS for soils with 1 mg/kg exogenous Cd added, 29% in APS to 51% in CPS for soils with 2 mg/kg exogenous Cd added, and 30% in APS to 47% in CPS for soils with 4 mg/kg exogenous Cd added. Meanwhile, those in larger aggregates apparently decreased; for instance, the loading factors of Cd in macroaggregate decreased from 12% in APS to 4% in CPS for native soil, 10% in APS to 3% in CPS for soils with Cd1 and Cd2, and 10% in APS to 4% in CPS for soils with Cd4. The dominating Cd loads were found in silt–clay aggregate for NPS and CPS and in small aggregate for APS, regardless of the Cd addition. This suggests that higher soil pH promotes Cd sequestration in the finest fraction for purple soil. However, the allocation of Cd in various-sized aggregates in CYS varied to less extent, except for macroaggregate, which had a load of only 3–5% of T-Cd. Commonly, heavy metals in acidic soils are more soluble; thus, they have higher availability to plants and higher environmental risks to water bodies due to their higher mobility. Our observations point out that enriched Cd in smaller aggregates in neutral and calcareous purple soils may also endow them with greater environmental risks via mass transportation driven by runoff, implying that the environmental risks of soil Cd are soil type- and pH-dependent.

3.4. The Extractability of Cd in Soil Aggregates by HCl

The extractable soil Cd contents by various chemical agents are often used to evaluate the soil Cd availability to plants; our previous study [23] demonstrated that Cd extracted using HCl (HCl-Cd) showed the strongest correlation with crop uptake in southwest China. To evaluate the contribution of soil aggregates to the bulk soil regarding Cd availability, the Cd extractability by HCl in various-sized aggregates of the tested soils was measured, as shown in Figure 4. It shows that the Cd extractability in aggregates varied significantly with the particle size, pH, soil type, and Cd addition.

Soil type shows a fundamental influence on the Cd extractability in soil aggregates with or without exogenous Cd added, following a general tendency of APS > NPS > CPS > CYS, which is in accordance with the conventional knowledge that Cd in acidic soils is more labile than that in neutral or calcareous soils. Within each soil type, the extractability of native soil Cd among aggregates varies significantly, whereas the variations diminished upon exogenous Cd addition. In native soils, the HCl-Cd content was consistently highest in silt–clay aggregate across all soil types. Conversely, macroaggregate exhibited the lowest HCl-Cd levels, except for NPS, where the differences among aggregates were less pronounced, with small aggregate showing the lowest extractability. With exogenous Cd addiction, the HCl-Cd contents in microaggregate were apparently lower than those in other aggregates in APS and NPS, which were in accordance with their Cd enrichment ability (see Figure 1). For the calcareous soils (CPS and CYS), the HCl-Cd contents in aggregates tend to decrease with their increased sizes, suggesting that the smaller aggregates in these soils have higher HCl-extractable Cd due to their higher Cd loads.

The significance of Cd in aggregates to its availability in the bulk soil depends not only on its extractability but also on the total Cd loads in each aggregate. To better evaluate the availability of Cd in aggregates, we introduce the concept of active fraction (AF), defined as the ratio of HCl-extractable Cd to the total Cd in each aggregate (AF = [HCl − Cd]/[T − Cd] × 100). The results are shown in

Table 2. AFs of Cd in bulk soil and aggregates of tested soils.

Soil Type	Aggregates in Native Soil					Bulk Soils
	<0.053 mm	0.053–0.25 mm	0.25–0.5 mm	0.5–1 mm	1–2 mm	0~2 mm
APS	65.45 ± 0.47 a 1	65.05 ± 4.21 a	47.56 ± 7.26 b	43.81 ± 10.57 b	22.15 ± 4.25 c	48.77 ± 0.24
NPS	34.31 ± 1.05 c	49.29 ± 2.62 a	42.55 ± 0.00 b	36.79 ± 2.21 c	41.96 ± 1.70 b	44.10 ± 0.16
CPS	37.08 ± 0.48 a	24.32 ± 4.24 bc	30.15 ± 1.18 b	17.85 ± 0.44 c	9.79 ± 0.26 d	26.02 ± 0.25
CYS	23.35 ± 4.90 a	12.50 ± 0.26 b	11.23 ± 1.90 b	5.79 ± 1.41 c	2.9 ± 1.03 d	20.94 ± 0.05
Soil Type	Aggregates in Soils with Cd1 Addition					Bulk Soils
	<0.053 mm	0.053–0.25 mm	0.25–0.5 mm	0.5–1 mm	1–2 mm	0~2 mm
APS	43.30 ± 0.00 d	52.06 ± 0.00 a	46.03 ± 0.00 b	45.94 ± 0.29 b	45.06 ± 0.45 c	51.25 ± 0.17
NPS	38.64 ± 0.21 d	43.83 ± 0.26 b	42.90 ± 0.36 c	38.97 ± 0.00 d	46.23 ± 0.47 a	39.93 ± 0.28
CPS	12.54 ± 0.30 a	10.30 ± 0.79 ab	12.33 ± 1.09 a	9.97 ± 0.69 b	12.10 ± 1.07 ab	12.22 ± 0.41
CYS	8.81 ± 1.43 a	8.86 ± 1.06 a	6.75 ± 0.00 b	5.91 ± 1.09 b	3.73 ± 0.29 c	8.68 ± 0.04
Soil Type	Aggregates in Soils with Cd2 Addition					Bulk Soils
	<0.053 mm	0.053–0.25 mm	0.25–0.5 mm	0.5–1 mm	1–2 mm	0~2 mm
APS	31.28 ± 0.45 d	40.28 ± 0.07 a	39.78 ± 0.00 a	38.69 ± 0.00 b	36.35 ± 0.74 c	39.57 ± 0.21
NPS	33.22 ± 0.28 c	48.59 ± 0.24 a	39.27 ± 1.03 b	34.29 ± 0.41 c	37.71 ± 0.45 b	43.37 ± 0.09
CPS	10.36 ± 1.26 b	9.50 ± 0.51 b	12.09 ± 0.47 b	13.35 ± 0.34 a	12.06 ± 0.29 b	11.79 ± 0.16
CYS	7.91 ± 0.32 a	6.57 ± 0.08 b	6.42 ± 0.01 b	6.21 ± 0.14 b	5.02 ± 0.03 c	7.08 ± 0.07
Soil Type	Aggregates in Soils with Cd4 Addition					Bulk Soils
	<0.053 mm	0.053–0.25 mm	0.25–0.5 mm	0.5–1 mm	1–2 mm	0~2 mm
APS	30.71 ± 0.31 c	39.12 ± 0.00 a	38.86 ± 0.17 a	36.66 ± 0.06 b	35.34 ± 0.51 b	38.43 ± 0.03
NPS	32.86 ± 0.64 c	49.47 ± 0.00 a	39.61 ± 0.00 b	34.05 ± 0.58 c	37.44 ± 1.09 b	42.22 ± 0.25
CPS	8.91 ± 0.00 b	10.05 ± 0.01 a	9.28 ± 0.76 ab	8.91 ± 0.22 b	10.32 ± 0.15 a	10.39 ± 0.06
CYS	5.47 ± 0.13 a	5.36 ± 0.05 a	5.20 ± 0.02 a	5.17 ± 0.00 a	4.86 ± 0.03 b	6.02 ± 0.25

¹ A distinct letter denotes a significant difference (p < 0.05) between treatments in the same soil, whereas the same lowercase letter in the same row indicates no significant difference between treatments.

.

In native soil, AF values in silt–clay aggregate were the highest for all soils, and the lowest AF was found in macroaggregate for all soils, except for NPS, where AF variations among aggregates were minimal, indicating that the Cd in the silt–clay aggregate of native soil showed the highest availability. Upon exogenous Cd addition, the AF values in aggregates became primarily soil type-dependent, with reduced variations among aggregates. Notably, in purple soils, AF decreased sharply with increasing soil pH, following the trend APS > NPS > CPS, confirming higher Cd availability in acidic conditions. For instance, the AF value for silt–clay aggregate in APS was nearly 3.5 times higher than that in CPS. Despite both being alkaline, CPS aggregates displayed significantly higher AF values than CYS across all size fractions, even if CPS has a higher pH (7.81) than CYS (7.67), implying the higher Cd availability in CPS.

The AF values of bulk soils (Table 2) may reflect their overall Cd availability to plants, which is closely associated with that in various-sized aggregates for a specific soil type. For the native soils, higher AF values were found for microaggregate in NPS (49.25% versus 44.10%), both microaggregate and silt–clay aggregate in APS (65.05% and 65.05%, respectively, versus 48.77%), and silt–clay aggregate in CPS (37.08% versus 26.02%) and CYS (23.35% versus 20.94%), as compared to the corresponding bulk soils. Upon exogenous Cd addition, only microaggregate in APS and NPS maintained higher AF values (39.12–50.06%, 43.83–49.47%, respectively) than bulk soil (38.43–51.25%, 39.93–43.37%, respectively), which was consistent with the pattern observed in native soil. Higher AF values in aggregates relative to bulk soils suggest that these aggregates contribute more significantly to bulk soil Cd availability. In calcareous soils (CPS and CYS), AF values in all aggregates were lower than those for APS and NPS and are comparable to the bulk soils, suggesting that the contributions of aggregates to the Cd availability in CPS and CYS might mainly rely on their total Cd loads in each aggregate.

3.5. The Fractionation of Cd in Soil Aggregates

The extractability of heavy metals in soil or its aggregates are closely related to their occurring states [9]. Figure 5 illustrates the fractionation of Cd across different soil aggregate sizes (detailed data provided in Appendix A,

Table A1. The contents and recovery of each morphometric of Cd in soil aggregates with varying particle sizes.

Soil Type (CK)	Particle Size (mm)	Ex-Cd (mg·kg−1)	CA-Cd (mg·kg−1)	Fe/Mn-Cd (mg·kg−1)	OM-Cd (mg·kg−1)	Res-Cd (mg·kg−1)	Rate of Recovery (%)
APS	<0.053	0.0213	0.0006	0.0227	0.0052	0.0187	94
	0.053–0.25	0.0173	0.0006	0.0145	0.0031	0.0285	88
	0.25–0.5	0.0204	0.0019	0.0201	0.0065	0.0449	107
	0.5–1	0.0203	0.0004	0.0285	0.0077	0.0346	93
	1–2	0.0193	0.0002	0.0221	0.0058	0.0202	82
NPS	<0.053	0.0212	0.0000	0.0255	0.0036	0.0309	88
	0.053–0.25	0.0189	0.0001	0.0145	0.0019	0.0494	113
	0.25–0.5	0.0191	0.0001	0.0169	0.0043	0.0302	103
	0.5–1	0.0202	0.0002	0.0152	0.0014	0.0338	87
	1–2	0.0201	0.0000	0.0192	0.0041	0.0488	118
CYS	<0.053	0.0147	0.0000	0.0362	0.0080	0.0419	116
	0.053–0.25	0.0080	0.0000	0.0561	0.0112	0.0214	101
	0.25–0.5	0.0133	0.0000	0.0433	0.0077	0.0308	112
	0.5–1	0.0123	0.0000	0.0214	0.0081	0.0360	105
	1–2	0.0083	0.0000	0.0420	0.0091	0.0196	106
CPS	<0.053	0.0266	0.0002	0.0263	0.0102	0.0208	109
	0.053–0.25	0.0204	0.0002	0.0177	0.0085	0.0511	105
	0.25–0.5	0.0354	0.0002	0.0211	0.0079	0.0244	111
	0.5–1	0.0290	0.0002	0.0243	0.0050	0.0217	100
	1–2	0.0237	0.0002	0.0129	0.0010	0.0295	96
Soil Type (Cd2)	Particle Size (mm)	Ex-Cd (mg·kg−1)	CA-Cd (mg·kg−1)	Fe/Mn-Cd (mg·kg−1)	OM-Cd (mg·kg−1)	Res-Cd (mg·kg−1)	Rate of Recovery (%)
APS	<0.053	0.0890	0.0008	0.1348	0.0103	0.0587	98
	0.053–0.25	0.0839	0.0008	0.1121	0.0095	0.0729	112
	0.25–0.5	0.0868	0.0008	0.1528	0.0180	0.0334	102
	0.5–1	0.0827	0.0008	0.1062	0.0150	0.0401	88
	1–2	0.0759	0.0009	0.0904	0.02119	0.0246	83
NPS	<0.053	0.0670	0.0006	0.1466	0.0506	0.0394	101
	0.053–0.25	0.0636	0.0006	0.1605	0.0148	0.0244	102
	0.25–0.5	0.0646	0.0006	0.1471	0.0205	0.0294	94
	0.5–1	0.0419	0.0006	0.1043	0.0426	0.0272	83
	1–2	0.0583	0.0005	0.1014	0.0282	0.0238	88
CYS	<0.053	0.0435	0.0003	0.1804	0.0271	0.0414	97
	0.053–0.25	0.0412	0.0004	0.1500	0.0261	0.0589	92
	0.25–0.5	0.0401	0.0004	0.1490	0.0257	0.0829	98
	0.5–1	0.0419	0.0004	0.1324	0.0203	0.0798	98
	1–2	0.0436	0.0004	0.1102	0.0129	0.0895	97
CPS	<0.053	0.0582	0.0008	0.1420	0.0199	0.0560	97
	0.053–0.25	0.0597	0.0008	0.1481	0.0061	0.0745	104
	0.25–0.5	0.0670	0.0008	0.1138	0.0067	0.0737	99
	0.5–1	0.0552	0.0008	0.1598	0.0063	0.0579	14
	1–2	0.0553	0.0008	0.0666	0.0040	0.0594	86

). The Cd recovery rates for sequential fractionation varied between 82% and 116%, matching the analytical quality requirement.

In the test soils, the Cd in aggregates mainly exists as exchangeable (Ex-Cd), Fe/Mn oxide-bound (Fe/Mn-Cd), and residual (Res-Cd) forms. The carbonate-associated (CA-Cd) and organic matter-bound (OA-Cd) fractions contributed only 0–2% and 6–12%, respectively, in native soils. However, the proportion of OA-Cd increased substantially to 5–23% following exogenous Cd addition. The allocation of Cd in aggregates mainly varies among Ex-Cd, Fe/Mn-Cd, and Res-Cd, and CA-Cd and OA-Cd are less variable, regardless of Cd addition. In native soils, the Ex-Cd fraction exhibited a significant pH-dependent decline across all aggregate sizes (p < 0.05), with APS displaying the highest proportion—up to 32% in silt–clay aggregate. Notably, in calcareous soils (CPS and CYS), a clear size-dependent decline of Ex-Cd was observed: macroaggregate contained 40–50% less Ex-Cd than silt–clay aggregate. The content distribution trend of Ex-Cd among various-sized aggregates was in accordance with the Cd activity percentage (AF) in aggregates, as mentioned above. Following exogenous Cd introduction, two contrasting patterns emerged: (1) microaggregate in two kinds of purple soils (APS and NPS; pH < 7.5) exhibited the lowest Ex-Cd concentrations, whereas (2) the Ex-Cd contents followed the same size-dependent pattern in the aggregates for the two calcareous soils (CPS and CYS; pH > 7.5), where the Ex-Cd content progressively decreased with increasing aggregate size. For the Fe/Mn-Cd fraction, the relative distribution patterns across different aggregate sizes remained consistent with those in the native soils after exogenous Cd introduction. However, a significant increase in its proportion was observed compared to the native soils, exhibiting the following characteristics: it ranged from 9 to 31% in acidic purple soils, 17% to 44% in neutral purple soils, and 5% to 26% in calcareous purple soils. Notably, Fe/Mn-Cd dominated across all aggregate fractions in CYS, representing the dominant Cd species. This distribution pattern can be attributed to the abundant Fe/Mn oxides in calcareous systems, which provide substantial sorption sites through their high density of negative surface charges and large specific surface areas. These physicochemical properties facilitate the formation of Fe/Mn oxide-bound complexes, as demonstrated by spectroscopic evidence [24,25]. Overall, the observed fractionation patterns were strongly correlated with both the T-Cd and HCl-Cd pools, reinforcing the importance of aggregate-mediated controls in Cd behavior in soils.

3.6. Cd Contamination on Ryegrass Growth

Figure 6 presents the plant height and aboveground biomass of ryegrass grown under varying Cd concentrations. The response of ryegrass growth to Cd stress varied markedly among different soil types. In NPS, Cd exposure significantly stimulates the growth of ryegrass: the aboveground biomass and the plant height are increased by 36.49–44.94% and 9.13–14.62%, respectively, as compared to the control. In APS and CPS, ryegrass growth showed minor variations in biomass and plant height with Cd addition; no statistically significant differences (p > 0.05) exist relative to the control. Conversely, in CYS, Cd exposure caused significant growth inhibition; especially at 2 mg·kg⁻¹ Cd, plant height decreased by 36.85%, and the aboveground biomass dropped to 63.14% of control levels.

3.7. Accumulation of Cd in Ryegrass

The uptake and accumulation of heavy metals in plants are the final indicators reflecting their bioavailability in soils. Among these, the Cd content in plant tissues is of particular concern due to its implications for food safety. Figure 7 displays the Cd contents in aboveground and belowground parts (roots). It shows that the Cd contents in both parts increased nearly linearly with the soil Cd levels, with the Cd content in the aboveground part being systematically higher than that in the roots. At a given soil Cd level, the Cd contents in both parts show a clear soil type dependence, following a sequence of APS > NPS > CPS in purple soils. Notably, for the two calcareous soils (CPS and CYS), the Cd accumulated in the ryegrass plant in CPS is consistently higher than that in CYS. The pH-dependent Cd accumulation pattern, as well as the difference of the Cd accumulation in ryegrass plants between calcareous purple soil and yellow soil, corresponded well to the availability of Cd as evaluated by HCl extraction, as described in the previous sections. The results prove that (1) HCl extraction could be well adapted for the evaluation of soil Cd availability, and (2) soil pH is crucial to Cd bioavailability but not the sole factor controlling Cd bioavailability, as a similar pH in different soil types may result in different plant accumulations.

3.8. Ryegrass Cd Accumulation in Relation to Cd States in Soil Aggregates

To further reveal the roles of soil aggregates in regulating the Cd bioavailability to plants, we determined the critical elements influencing the Cd distribution; it is important to comprehend the relationship between the cadmium uptake in ryegrass (Plant-Cd) and the T-Cd and HCl-Cd content across aggregate size fractions. As shown in

Table 3. Correlation coefficients between Cd concentrations in ryegrass and T-Cd in soil aggregates.

T-Cd	<0.053 mm	0.053–0.25 mm	0.25–0.5 mm	0.5–1 mm	1–2 mm
Plant-Cd (CK)	0.329	0.703 * 1	0.423	0.571	−0.990 **
Plant-Cd (Cd1)	0.018	0.806 **	0.053	0.361	−0.729 **
Plant-Cd (Cd2)	0.214	0.811 **	0.416	0.516	−0.807 **
Plant-Cd (Cd4)	0.475	0.793 **	0.327	0.104	−0.795 **

¹ * Indicates a significant correlation at the 0.05 level, and ** indicates a significant correlation at the 0.01 level.

, microaggregate in native soils exhibited a strong positive correlation between Plant-Cd and T-Cd (r = 0.703, p < 0.05), whereas macroaggregate showed an exceptionally strong negative correlation (r = −0.990, p < 0.01). Following exogenous Cd addition, the T-Cd content in microaggregate has a very significant positive correlation with Plant-Cd, in macroaggregate has a strong negative correlation, and the correlation of other particle size aggregates remains non-significant. As shown in

Table 4. Correlation coefficients between Cd concentrations in ryegrass and HCl-Cd in soil aggregates.

HCl-Cd	<0.053 mm	0.053~0.25 mm	0.25~0.5 mm	0.5~1 mm	1~2 mm
Plant-Cd (CK)	0.715 **	0.615 * 1	0.480	0.513	0.185
Plant-Cd (Cd1)	0.834 **	0.669 *	0.484	0.522	0.540
Plant-Cd (Cd2)	0.817 **	0.652 *	0.311	0.489	0.372
Plant-Cd (Cd4)	0.779 **	0.695 *	0.405	0.547	0.468

¹ * Indicates a significant correlation at the 0.05 level, and ** indicates a significant correlation at the 0.01 level.

, the HCl-Cd content exhibited highly significant correlations (p < 0.01) with silt–clay aggregate and significant correlations (p < 0.05) with microaggregate in both native and exogenous Cd-contaminated soils. These results identify silt–clay, microaggregate, and macroaggregate as the key fractions controlling Cd bioavailability.

The T-Cd amounts in microaggregate in APS and NPS and in macroaggregate in CPS and CYS all are the smallest content, but they have the best correlation with plant absorption in both native and exogenous Cd-contaminated soils. This compelling evidence establishes aggregate size—rather than mass—as the primary determinant of Cd bioavailability. Notably, the HCl-Cd content in silt–clay aggregate consistently exhibited the highest extraction efficiency among all four soil types and demonstrated a significant positive correlation with plant Cd uptake (p < 0.01). This finding further confirms that more phytoavailable Cd tends to accumulate in finer particle fractions, which aligns with previously reported experimental patterns. The fine-sized aggregates (silt–clay and microaggregate) typically exhibit a higher specific surface area and greater abundance of variable charge sites, which facilitate continuous Cd release through ion exchange processes. This phenomenon can be further enhanced by low-molecular-weight organic acids that promote Cd desorption, consequently increasing Cd bioavailability. In contrast, macroaggregate exhibits well-developed pore structures that physically encapsulate heavy metals, creating a “shielding effect” that impedes both root contact and activation by root exudates. Furthermore, macroaggregate generally contains higher concentrations of Fe/Mn oxides, organic matter, and mineral components. These constituents can immobilize Cd through specific adsorption or form stable complexes/mineral-bound fractions, ultimately reducing heavy metal uptake by plants. In summary, these findings confirm that aggregate particle size is a critical factor governing Cd mobility and bioavailability in soil–plant systems.

4. Discussion

4.1. pH-Dependent Regulation of Cd Distribution and Bioavailability by Soil Aggregates

The bioavailability of Cd in soil is affected by a series of individual physical, chemical, and biological interactions that determine the exposure of plants and animals to chemicals [26]. Our results demonstrate that the distribution of Cd and its chemical species within soil aggregates, as well as Cd bioavailability, exhibit significant pH-dependent characteristics. For the purple soils, the physical mass proportion of silt–clay aggregate increased significantly from 23.8% to 49.6% with rising pH, while the proportions of all other aggregate size fractions showed decreasing trends (Figure 1). This phenomenon results from H⁺ accumulation in acidic soils, which promotes particle sloughing and reduces permeability, thereby facilitating the transformation of small aggregates into larger ones [27]. Figure 3 clearly shows that in purple soils (pH 6.07–7.87), fine-sized aggregate exhibited significantly higher Cd enrichment, consistent with the results described by Mo et al. [26]. The speciation of Cd(II) in aqueous solutions is dominantly in the form of the hexahydrate cation under acidic-to-weakly alkaline pH conditions, which exhibits its higher solubility and mobility. The lower Cd enrichment in calcareous yellow soils may be attributed to their higher pH levels and lower organic matter content, which collectively inhibit Cd adsorption by soil aggregates—a finding consistent with the experimental results of Peng et al. [28]. The Ex-Cd fraction exhibited a significant pH-dependent decrease across all aggregate sizes (p < 0.05), with acidic purple soil having the highest proportion (32% in silt–clay aggregate). Ex-Cd is widely acknowledged as the most bioavailable of morphologies. Increasing soil pH and converting Cd into a stable form can reduce Cd availability [29].

4.2. Particle Size-Dependent Control of Cd Distribution and Bioavailability in Soil Aggregates

Soil behavior and heavy metal bioavailability are substantially regulated by the dispersion of aggregates [30]. Our results demonstrate that silt–clay aggregate (<0.053 mm) exhibited the highest Cd enrichment across all the tested soils, which is consistent with the findings of Shen et al. [31]. Cd preferentially attaches to fine aggregates. The combination of increased surface area, organic matter content, and unique microenvironmental conditions (e.g., redox potential and microbial activity) in fine-sized aggregate contributes to its enhanced capacity for heavy metal adsorption [32]. Moreover, fine-sized aggregate typically exhibits narrower pore diameters and higher clay content, which collectively reduce hydraulic conductivity and limit the physical mobility of Cd [33]. Consequently, such aggregate enhances Cd retention, explaining the consistently higher Cd enrichment observed in silt–clay aggregate. Beyond physicochemical adsorption, aggregate size also mediates Cd bioavailability through microbial processes. In metal-contaminated soil, the aggregate particle size affects the biomass and community structure of soil microorganisms [12]. The large specific surface area of fine aggregate provides a large area for microorganisms to attach, thus increasing its biomass. Meanwhile, the study by Yang et al. [34] demonstrated that microbial respiration is stronger in macroaggregate. The intensive respiration leads to oxygen consumption, consequently forming an anaerobic environment. Over time, Cd increasingly exists in fewer bioavailable fractions (e.g., organic-bound and sulfide-bound fractions) while decreasing as Ex-Cd, which is consistent with our experimental results (Figure 5).

4.3. Mechanisms and Perspectives for Soil Aggregate in Regulating Cd Bioavailability

As the basic structural unit, soil aggregate plays a critical role in regulating the behavior and bioavailability of heavy metals such as Cd through the integrated influence of its physical architecture, chemical composition, and biological activity, as well as the dynamic interactions among these factors. This study systematically examined the Cd distribution and speciation within soil aggregates across various soil types and evaluated their influence on Cd bioavailability through a pot experiment using ryegrass. Findings identified soil pH and aggregate size as key determinants of Cd fractionation and plant uptake, and proposed a mechanistic framework for understanding Cd bioavailability at the soil aggregate scale. Nevertheless, the regulation of Cd by soil aggregates involves a broader range of mechanisms that warrant deeper exploration. From a physical perspective, the current findings suggest that aggregate size influences Cd release, likely due to differences in internal pore structure and surface accessibility. Recent studies utilizing μ-XRF and XANES have revealed that the pore connectivity and mineral organic spatial distribution within aggregates control Cd diffusion pathways; yet, further in situ investigations are needed to clarify how structural heterogeneity regulates Cd mobility under dynamic environmental conditions [35,36]. From a chemical perspective, adsorption is the initial step by which heavy metals enter the soil, and it is inextricably linked to aggregation. Soil aggregates adsorb heavy metals to varying degrees, with microaggregate often showing stronger retention due to a greater surface area and higher organic matter content [37,38]. Cd tends to accumulate at the organic–mineral micro-interfaces within soil aggregates, where its retention is strongly affected by the composition and complexity of plant-derived organic matter and its association with Fe/Al oxides, particularly in finer fractions such as silt and clay. From a biological perspective, studies have shown that external disturbances such as long-term cultivation can reduce the complexity and metabolic activity of microbial networks within macroaggregate, thereby weakening microbially mediated carbon stabilization. Future research should further explore whether the structural degradation of macroaggregate under long-term cultivation also impairs its capacity to retain heavy metals [39], and how biological processes such as microbial respiration, root exudation, and faunal activity modulate Cd dynamics within the aggregate microenvironment.

5. Conclusions

Our systematic investigation identifies soil pH and aggregate particle size as pivotal factors governing Cd distribution dynamics and bioavailability in soil systems. A distinct pH-dependent pattern was identified, whereby microaggregate (0.053–0.25 mm) exhibited minimal Cd retention in acidic soils (APS/NPS; pH < 7.5) but became a primary Cd reservoir in alkaline environments (CPS/CYS; pH > 7.5). Notably, silt–clay aggregate in acidic soils displays a superior affinity for exogenous Cd across the measured parameters (T-Cd, HCl-Cd, and Ex-Cd), and microaggregate paradoxically achieves peak Cd activity fractions (AFs). Pot experiments further validated these observations, revealing significant positive correlations between ryegrass Cd uptake and both HCl-Cd in silt–clay aggregate and T-Cd in microaggregate, whereas macroaggregate (1–2 mm) was negatively correlated with Cd uptake. These trends collectively indicate that aggregate size, rather than total Cd content alone, plays a dominant role in regulating Cd bioavailability. Our results strengthen the fundamental effects of soil pH and highlight the critical roles of fine-sized aggregates (silt–clay and microaggregate) and macroaggregate (1–2 mm) in controlling cadmium bioavailability in bulk soils, and thus provide a mechanistic framework for developing novel remediation strategies for Cd-polluted soils. Despite the importance of soil aggregates in regulating Cd behaviors, the detailed mechanisms underlying them are far from fully understood due to their complexities in governing soil physical, chemical, and biological properties, as well as the interactions among these properties, which necessitates the need for more advanced studies in this regard.