Synergistic Effects of Silicon and Ferrous Sulfate on Reducing Arsenic and Cadmium Accumulation in Rice from Co-Contaminated Soil

Abstract

The co-contamination of arsenic (As) and cadmium (Cd) in paddy soils threatens rice safety, yet synergistic mitigation strategies using silicon (Si) and ferrous sulfate (FeSO₄) remain underexplored. This study integrated hydroponic and soil pot experiments to evaluate Si-FeSO₄ interactions on As/Cd accumulation and rice growth. Hydroponic trials employed 21-day-old rice seedlings exposed to 0.5 mg As(III)/Cd(II) L⁻¹ with/without 70 mg Si L⁻¹ and 30–70 mg Fe L⁻¹, followed by sequential harvesting at 14 and 21 days. Soil experiments utilized co-contaminated paddy soil (50 mg As kg⁻¹ and 1.2 mg Cd kg⁻¹) amended with Si (80 or 400 mg kg⁻¹) and Fe (100 or 1000 mg kg⁻¹), with pore water dynamics monitored over 120 days. Hydroponic results demonstrated that 70 mg Si L⁻¹ combined with 30 or 70 mg Fe L⁻¹ enhanced shoot biomass by 12–79% under As stress, while simultaneously reducing shoot As concentrations by 76–87% and Cd concentrations by 14–33%. Iron plaque induced by FeSO₄ exhibited contrasting adsorption behaviors: hydroponic roots immobilized both As and Cd (p < 0.01), whereas roots in soil primarily retained Cd (p < 0.05). In soil experiments, the optimal treatment of 100 mg Fe kg⁻¹ and 400 mg Si kg⁻¹ (Fe₁ + Si₂) increased grain biomass by 54%, while reducing As and Cd concentrations by 37% and 42%, respectively. However, a higher Fe dosage (Fe₂: 1000 mg kg⁻¹ Fe) paradoxically increased grain Cd concentrations. Mechanistically, Si amendment elevated soil pH (Δ + 0.72), facilitating Cd immobilization, while FeSO₄ lowered pH (Δ−0.07–0.53), increasing Cd mobility. A strong correlation between soluble Cd and plant uptake was observed (p < 0.01), while changes in As accumulation were unrelated to aqueous behavior. The optimized Si/Fe molar ratio of 7.95:1 effectively mitigated As and Cd co-accumulation, offering a dual-functional strategy for safe rice cultivation in contaminated soils.

1. Introduction

Accelerated industrialization has resulted in the substantial discharge of pollutants into agricultural soils, with excessive pesticide and fertilizer application, atmospheric deposition, and wastewater irrigation collectively intensifying soil contamination [1]. This contamination is a pervasive global challenge, with potentially toxic elements (PTEs) emerging as critical environmental pollutants. Globally, it is estimated that 14–17% of agricultural land soils are contaminated with at least one metal that exceeds safe limits, affecting approximately 24.2 million ha and impacting around 900 million to 1.4 billion people living in regions with high ecological and health risks [2]. In many countries, including China, the prevalence of contaminants such as As and Cd in agricultural soils is alarmingly high. Recent surveys indicate that approximately 19.4% of monitored arable lands in China exceed permissible PTEs thresholds, with As and Cd being the most common contaminants, showing exceedance rates of 7.0% and 2.7%, respectively [3]. Rice (Oryza sativa L), as a staple crop for over half of the global population, is particularly susceptible to bioaccumulating these PTEs [4]. This accumulation poses significant health risks, including renal dysfunction and various carcinogenic outcomes [5,6]. Consequently, PTEs remediation in agricultural soils has emerged as a strategic imperative for the Chinese government.

China’s limited arable land (0.09 ha/capita [7]) intensifies challenges in reconciling food security with soil contamination. National policies, including the Soil Pollution Prevention and Control Action Plan, mandate >95% safe utilization of contaminated farmland by 2030 [8,9], demanding urgent remediation solutions to enable continued cultivation. Despite these policy targets, achieving safe utilization is hindered by the unique biogeochemical challenges of As-Cd co-contaminated paddies. Unlike industrial sites, agricultural remediation prioritizes cost-effectiveness and soil fertility preservation. Electrokinetic or chemical leaching methods, though effective for rapid PTE removal elsewhere, risk nutrient depletion in farmlands [10]. Consequently, agricultural remediation prioritizes in-situ stabilization via soil amendments to immobilize PTEs and curb crop uptake [11]. For As-Cd co-contaminated paddies, redox-driven antagonism complicates remediation: under anaerobic flooding, reduced Eh mobilizes As (reduced to mobile As(III)) but immobilizes Cd (forming CdS precipitates) [12,13,14]. Conversely, drainage increases Eh, mobilizing Cd while promoting As adsorption (as As(V)) onto Fe/Al oxides [12,13]. This contrasting biogeochemical behavior under shifting redox conditions poses critical technical barriers to simultaneous As-Cd mitigation, necessitating field-validated strategies that balance efficacy and scalability.

Silicon demonstrates potential for mitigating As-Cd toxicity through pH elevation and competitive root uptake mechanisms [15,16], yet field studies show limited As reduction in heavily contaminated soils [17,18]. FeSO₄ enhances iron plaque formation on rice roots, immobilizing both elements [19,20], but synergistic effects with Si remain unexplored. Furthermore, Cd adsorption depends strongly on pH, while As(III) mobility shows pH insensitivity—a dichotomy suggesting combined Si-FeSO₄ application could strategically target both elements. To validate this hypothesis, we designed a two-phase study to bridge the knowledge gap between single-amendment mechanisms and co-contamination field scenarios.

Based on these considerations, this study aims to (1) investigate the effects of exogenous Si and FeSO₄ amendments on As or Cd uptake in rice seedlings under hydroponic systems with single-element exposure (preventing As-Cd precipitation) to reveal amendment-specific responses to individual element stress; and (2) evaluate the synergistic potential of combined Si-FeSO₄ application in reducing As and Cd accumulation in rice grown in co-contaminated paddy soils. Additionally, we addressed the potential environmental risks associated with chemical amendments, ensuring a comprehensive understanding of their implications for soil health and agricultural productivity. Our findings contribute to the development of more effective remediation strategies for contaminated agricultural soils.

2. Materials and Methods

2.1. Hydroponic Experiment

2.1.1. Rice Seedling Preparation

Rice cv. Yongyou 1540 was cultivated in a controlled hydroponic system. Seeds were surface-sterilized with H₂O₂ (Gr, 30%) for 15 min, rinsed thoroughly with deionized water, and germinated on moist quartz sand (pre-treated with 5% HCl for 24 h and washed to pH neutrality) at 25 °C for 15 days. Uniform seedlings were transferred to plastic containers (25 cm × 30 cm × 20 cm, 48 seedlings per container, total of 5 containers, 240 seedlings) and sequentially cultured in 1/5- and 1/4-strength Hoagland nutrient solutions for 7 days each. Subsequently, two representative seedlings were transplanted into 1 L black polyethylene pots (110 mm diameter × 118 mm height, total of 36 pots, 72 seedlings) filled with 1/3-strength Hoagland solution. The full-strength Hoagland solution composition is as follows (mmol L⁻¹): KH₂PO₄ 1.0, KNO₃ 5.0, Ca(NO₃)₂·4H₂O 5.0, MgSO₄·7H₂O 2.0, CuSO₄·5H₂O 3.2 × 10⁻⁴, ZnSO₄·7H₂O 7.7 × 10⁻⁴, MnCl₂·4H₂O 9.2 × 10⁻³, H₃BO₃ 4.6 × 10⁻², H₂MoO₄·4H₂O 3.85 × 10⁻⁴, and EDTA-Fe(II) 5 × 10⁻². The solution pH was adjusted to 5.5 using 0.1 mol L⁻¹ HCl/NaOH and refreshed every 3 days. Experiments were conducted in a controlled greenhouse at 25–35 °C in the Xia’an area of Fujian Agriculture and Forestry University with a 12–14 h daily photoperiod.

2.1.2. Experimental Treatments

At 21 days post-transplanting, iron plaque induction was initiated through sequential pretreatment as follows: (1) Root systems of plaque groups (Fe30, Fe70) were immersed in deionized water for 24 h, followed by 3-day exposure to FeSO₄·7H₂O solutions (30 or 70 mg Fe²⁺ L⁻¹) in Fe/P-depleted Hoagland solution; (2) Non-plaque controls received equivalent washing but without Fe removal.

The experimental design followed a complete factorial structure comprising six treatment combinations per abiotic stress factor: Control, Si, Fe30, Fe70, Si + Fe30, and Si + Fe70 (three biological replicates per treatment; total n = 36 experimental units containing two seedlings each, equivalent to 72 individual plants).

Post-induction, all seedlings were transferred to 1/3-strength Hoagland solution for 48 h prior to PTEs exposure. To avoid the reduced availability of As and Cd caused by precipitation when added simultaneously in the hydroponic system [21], the subsequent hydroponic experiments employed a single exposure to either As or Cd. Two parallel PTEs stress systems were established: (A) 0.5 mg As(III) L⁻¹ (NaAsO₂) and (B) 0.5 mg Cd(II) L⁻¹ (Cd(NO₃)₂·4H₂O), both with/without 70 mg Si L⁻¹ (Na₂SiO₃·9H₂O). The hydroponic pot experiment was initiated in April 2021, spanning 75 days from rice seed germination to plant harvest, with plants harvested on 7 July 2021.

2.1.3. Sample Collection and Analysis

Rice seedlings were harvested in batches at 14 and 21 days after treatment with As or Cd, aiming to investigate whether prolonged exposure to these elements influenced seedling growth. Harvested seedlings underwent thorough rinsing with deionized water to eliminate surface contaminants. Roots and shoots were separated using sterilized steel scissors at the root base, with root samples transferred to 100 mL beakers for subsequent iron plaque extraction using a modified dithionite-citrate-bicarbonate (DCB) method according to Taylor and Crowder [22].

Specifically, roots were immersed in 40 mL of a mixed solution containing 0.03 mol L⁻¹ Na₃C₆H₅O₇·2H₂O and 0.125 mol L⁻¹ NaHCO₃. To initiate the reductive dissolution of iron oxides, 1 g of Na₂S₂O₄ was added under dark conditions at 25 °C for 60 min. The resulting extract was transferred to a 100 mL volumetric flask, with roots rinsed three times using deionized water; all washings were incorporated into the flask. Final extracts were filtered through 0.45 μm aqueous-phase membranes prior to analysis.

Root and shoot samples were dehydrated in kraft paper envelopes at 60 °C for 72 h and ground for digestion using HNO₃-H₂O₂ (4:1 v/v) according to Yu et al. [23]. In detail, 0.2 g of the pulverized sample was mixed with 4 mL of ultrapure HNO₃ (Gr, 65.0~68.0%) and allowed to soak for 10 h. Subsequently, 1 mL of H₂O₂ (Gr, 30%) was added to initiate oxidative digestion. The temperature programming involved ramping up to 110 °C over 15 min, maintaining that temperature for 30 min, and then cooling back to ambient conditions over 20 min using a microwave digestion system (MARS6; CEM, Matthews, NC, USA). Digested solutions were diluted to 100 mL with deionized water and filtered through 0.45 μm syringe filters into pre-cleaned centrifuge tubes. Reagent blanks and citrus leaf certified reference material (the citrus leaves: GBW10020) were used for quality control of the digestion procedure, with recovery rates ranging from 90% to 105%.

2.2. Soil Pot Experiment

2.2.1. Soil Collection and Treatment

The experimental design comprised 9 treatments in triplicate: Control, Fe₁, Fe₂, Si₁, Si₂, Fe₁ + Si₁, Fe₁ + Si₂, Fe₂ + Si₁, and Fe₂ + Si₂.

Surface soil (0–20 cm depth) from uncontaminated agricultural fields in Minhou County, Fuzhou City, China, was air-dried, homogenized, and sieved (<1 cm). Comprehensive physicochemical characterization of the prepared soil, including particle size distribution, total As/Cd content, and cation exchange capacity, is systematically provided in Table S1.

Specific additives were introduced into the soil: Fe (as FeSO₄) at two concentrations: 100 mg kg⁻¹ (Fe₁: 1.79 mmol) and 1000 mg kg⁻¹ (Fe₂: 17.9 mmol); Si (as Na₂SiO₃·9H₂O) at 80 mg kg⁻¹ (Si₁: 2.85 mmol) and 400 mg kg⁻¹ (Si₂: 14.24 mmol).

Stock solutions were separately prepared by dissolving 8 g As (as NaAsO₂) and 200 mg Cd (as Cd(NO₃)₂·4H₂O) in 100 mL deionized water each, achieving final As and Cd concentrations of 50 mg kg⁻¹ and 1.2 mg kg⁻¹, respectively, in subsequent experimental treatments. A 3 cm floodwater layer was maintained over the soil surface. After two months of flooding for stabilization, the soil was air-dried, re-crushed, and thoroughly mixed with a base fertilizer (N: 0.2 g kg⁻¹, P₂O₅: 0.15 g kg⁻¹, K₂O: 0.2 g kg⁻¹). The amended soil was then repacked into pots (cylindrical PVC containers: 28 cm height × 15 cm diameter; 5 kg dry weight equivalent) and re-flooded.

Next, 15-day-old rice seedlings pre-cultured in quartz sand were transplanted following hydroponic acclimatization (1/5 → 1/4 strength Hoagland solution, 7 days per stage). One month after re-flooding the soil, these acclimatized 29-day-old rice seedlings were transplanted into the soil to initiate the growth phase.

Rhizon MOM pore water samplers (10 cm length, Wageningen, The Netherlands) were installed at a 45° inclination for periodic sampling.

2.2.2. Sample Collection and Analysis

At maturity, plants were partitioned into roots, stems, leaves, and grains. Root iron plaques were DCB-extracted as described in Section 2.1.3. Vegetative tissues were heat-treated (105 °C, 1 h), oven-dried (70 °C) to a constant mass, and then weighed. Grains were sun-dried, hulled, and pulverized for microwave-assisted digestion. Pore water was collected at 30, 60, 90, and 120 days post-transplantation. A 10 mL aliquot was immediately analyzed for pH (Mettler Toledo FE28, Zurich, Switzerland), while the remaining 15 mL was filtered through 0.45 μm cellulose acetate membranes, acidified with 2% ultrapure HNO₃, and stored at 4 °C prior to elemental analysis. Soil pH determination followed a 1:2.5 (w/v) soil-to-water ratio: 8.00 g air-dried soil (<2 mm) was mixed with CO₂-free deionized water, shaken horizontally (200 rpm, 30 min), and equilibrated for 1 h before measurement. Elemental concentrations (As, Cd and Fe) in both plant digests and pore water were quantified by inductively coupled plasma mass spectrometry (ICP-MS, PerkinElmer NexION 300×) with Cd-111, As-75, and Fe-57 as the isotopes. The operating parameters of the ICP-MS instrument are shown in Table S2. The pot-based cultivation experiment was carried out from 15 April to 28 September 2022, covering the complete growth cycle of rice from seed germination to grain harvest at physiological maturity (≥90% filled grains).

2.2.3. Data Processing

Data processing utilized Excel 2021, SPSS 23.0, and SigmaPlot 14.0. Results are presented as mean ± standard deviation (n = 3). Significant differences (p < 0.05) were determined by Fisher’s LSD test following one-way ANOVA. Multivariate ANOVA in SPSS evaluated main effects and interaction terms (Si × Fe) on plant biomass, soil parameters, and As/Cd distribution patterns.

3. Results

3.1. Dry Weight of Rice Plant Exposed to As and Cd

Initial harvest data revealed distinct phytotoxic responses in rice seedlings to As and Cd exposure. Under As stress, Fe70, Si + Fe30, and Si + Fe70 treatments significantly enhanced shoot biomass by 16% (p < 0.05), 35% (p < 0.001), and 36% (p < 0.001) respectively compared to untreated controls (1.03 ± 0.03 g pot⁻¹) (Figure 1A). Neither Si alone nor Fe30 application showed significant effects on shoot biomass under As exposure. Notably, root biomass remained unaffected across all treatment groups during initial As exposure, with a control root biomass of 0.25 ± 0.02 g pot⁻¹ (Figure 1B). In contrast to As effects, Cd exposure elicited no statistically significant alterations in either shoot or root dry weights among treatment groups (Figure 1C,D).

The second harvest under prolonged As stress demonstrated synergistic interactions between Si and Fe. While Si application alone increased shoot biomass by 54% (p < 0.001), combined Si + Fe30 and Si + Fe70 treatments produced greater enhancements of 68% (p < 0.001) and 79% (p < 0.001), respectively, compared to a control shoot biomass of 2.73 ± 0.09 g pot⁻¹ (Figure 1A). Improvements in the root system were observed with the Fe70 treatment (36%, p < 0.005), as well as with the Si + Fe30 (33%, p < 0.01) and Si + Fe70 (43%, p < 0.005) treatments, in comparison to a control root biomass of 0.86 ± 0.12 g pot⁻¹ (Figure 1B). Conversely, Cd-exposed plants exhibited contrasting responses, with Si and Si + Fe70 treatments causing 26% root biomass reduction (p < 0.05) compared to controls (1.40 ± 0.23 g pot⁻¹) (Figure 1C). Other Cd exposure treatments maintained non-significant differences in plant biomass parameters (Figure 1C,D).

The soil culture experiments revealed distinct enhancement effects of Si and Fe supplementation on rice biomass accumulation. Notably, grain dry weight exhibited the most pronounced response to Si₁ and Fe₁ + Si₂ treatments, with respective increases of 66% (p < 0.005) and 54% (p < 0.05) compared to the control (

Table 1. Dry weight of rice plant exposed to As-Cd co-contaminated soil.

Treatments	Root (g pot−1)	Straw (g pot−1)	Grain (g pot−1)
Control	1.48 ± 1.03 cd	44.8 ± 22.8 cd	4.17 ± 1.45 bc
Fe1	1.36 ± 0.60 cd	43.7 ± 13.0 cd	2.75 ± 0.53 c
Fe2	1.43 ± 0.46 cd	45.2 ± 9.74 cd	3.02 ± 0.35 c
Si1	2.23 ± 0.11 abc	63.3 ± 6.18 abc	6.94 ± 1.59 a
Si2	2.63 ± 0.14 ab	66.6 ± 8.43 ab	5.56 ± 1.27 ab
Fe1 + Si1	1.41 ± 0.45 cd	41.6 ± 7.74 d	3.24 ± 0.27 c
Fe1 + Si2	2.79 ± 0.98 a	73.1 ± 15.5 a	6.41 ± 1.78 a
Fe2 + Si1	1.69 ± 0.26 bcd	52.5 ± 8.22 d	2.79 ± 0.15 c
Fe2 + Si2	1.19 ± 0.18 d	36.5 ± 4.25 bcd	2.48 ± 0.15 c
F value	3.24	3.44	7.97
P	<0.05	<0.05	<0.0005

Fe₁: 100 mg kg⁻¹, Fe₂: 1000 mg kg⁻¹, Si₁: 80 mg kg⁻¹, Si₂: 400 mg kg⁻¹. Degrees of freedom = 8. Within a column, different lowercase letters indicate significant differences among treatments (p < 0.05) based on Fisher’s LSD post-hoc test, with three replicates per treatment (data presented as mean ± standard deviation).

). Similar growth-promoting patterns were observed in vegetative tissues, where Si₂ and Fe₁ + Si₂ treatments significantly elevated straw biomass by 49% (p < 0.05) and 63% (p < 0.01), respectively (Table 1). Root system development showed comparable enhancement, with corresponding treatments increasing root dry weight by 77% (p < 0.05) and 88% (p < 0.05) (Table 1). Other experimental groups did not induce statistically significant biomass alterations in any plant compartment.

3.2. Dynamic Accumulation of As and Cd in Rice Tissues

3.2.1. As Sequestration Patterns

In the hydroponic experiments, As accumulation patterns in rice tissues demonstrated similar response patterns across both sampling periods, with most amendments inducing significant As uptake reduction compared to untreated controls. Notably, the Fe30 treatment showed divergent behavior, failing to decrease As levels in either plant compartment (Figure 2). During the initial harvest, compared to a control shoot As level of 92.6 ± 11.3 mg kg⁻¹, Si and Fe-based treatments (Si, Fe70, Si + Fe30, Si + Fe70) induced marked decreases in shoot As concentrations by 76%, 54%, 87%, and 86%, respectively (all p < 0.001) (Figure 2A). These treatments also led to corresponding reductions in root As levels of 64% (p < 0.005), 53% (p < 0.05), 42% (p < 0.05), and 64% (p < 0.005), compared to a control root As concentration of 304 ± 38.2 mg kg⁻¹ (Figure 2B). Subsequent harvest analysis revealed sustained mitigation efficacy, with shoot As concentrations declining by 78% (Si), 34% (Fe70), 81% (Si + Fe30), and 79% (Si + Fe70) (all p < 0.001), compared to a control shoot As level of 127 ± 11.8 mg kg⁻¹ (Figure 2C). Root systems exhibited comparable As exclusion patterns, showing 69% (Si), 65% (Si + Fe30), and 51% (Si + Fe70) reductions (all p < 0.001), compared to a control root As concentration of 354 ± 25.9 mg kg⁻¹ (Figure 2D). A temporal comparison indicated elevated As accumulation in second-harvest plants across all treatment groups relative to initial harvest specimens.

In the soil experiments, all treatments significantly reduced the As concentration in rice grains compared to the control, except for the Si₁ treatment (

Table 2. As concentration of rice plants exposed to As-Cd co-contaminated soil.

Treatments	Root (mg kg−1)	Straw (mg kg−1)	Grain (mg kg−1)
Control	211 ± 48.1 a	19.5 ± 4.38 a	0.92 ± 0.05 a
Fe1	147 ± 57.3 ab	16.2 ± 2.30 ab	0.70 ± 0.15 bc
Fe2	138 ± 56.8 ab	11.8 ± 1.12 cd	0.49 ± 0.03 d
Si1	109 ± 69.0 b	14.5 ± 1.90 bc	0.89 ± 0.12 a
Si2	179 ± 47.9 ab	12.8 ± 1.12 bcd	0.70 ± 0.07 bc
Fe1 + Si1	130 ± 28.4 b	19.4 ± 2.22 a	0.72 ± 0.04 b
Fe1 + Si2	148 ± 33.3 ab	12.2 ± 0.49 bcd	0.58 ± 0.04 cd
Fe2 + Si1	118 ± 29.5 b	9.94 ± 1.35 d	0.57 ± 0.10 cd
Fe2 + Si2	163 ± 28.2 ab	15.0 ± 3.48 bc	0.49 ± 0.02 d
F value	1.38	6.00	11.3
P	0.27	<0.001	<0.0001

Fe₁: 100 mg kg⁻¹, Fe₂: 1000 mg kg⁻¹, Si₁: 80 mg kg⁻¹, Si₂: 400 mg kg⁻¹. Degrees of freedom = 8. Within a column, different lowercase letters indicate significant differences among treatments (p < 0.05) based on Fisher’s LSD post-hoc test, with three replicates per treatment (data presented as mean ± standard deviation).

). Specifically, the Fe₁, Fe₂, Si₂, Fe₁ + Si₁, Fe₁ + Si₂, Fe₂ + Si₁, and Fe₂ + Si₂ treatments decreased the As concentration in rice grains by 24% (p < 0.005), 46% (p < 0.001), 24% (p < 0.005), 21% (p < 0.01), 37% (p < 0.001), 38% (p < 0.001), and 47% (p < 0.001), respectively (Table 2). In the rice straw, treatments with Fe₂, Si₁, Si₂, Fe₁ + Si₁, Fe₁ + Si₂, Fe₂ + Si₁, and Fe₂ + Si₂ significantly reduced the As concentration by 39% (p < 0.001), 26% (p < 0.05), 35% (p < 0.005), 37% (p < 0.005), 49% (p < 0.001), and 23% (p < 0.05), respectively (Table 2). For the rice roots, the As concentration was significantly reduced by 48% (p < 0.05), 38% (p < 0.05), and 44% (p < 0.05) following the Si₁, Fe₁ + Si₁, and Fe₂ + Si₁ treatments, respectively. In contrast, other treatments, particularly Fe₁ and Fe₂, also led to a reduction in As concentration in the rice roots; however, these changes were not statistically significant (Table 2).

3.2.2. Cd Accumulation Heterogeneity

In the hydroponic experiments, the Cd concentration in the shoots of both rice samples exhibited consistent trends, with the Si, Si + Fe30, and Si + Fe70 treatments significantly reducing Cd levels (Figure 3A,C). However, none of the treatments had a significant effect on the Cd concentration in the roots (Figure 3B,D). Specifically, the Si, Si + Fe30, and Si + Fe70 treatments reduced the Cd concentrations in the shoots by 19% (p < 0.001), 14% (p < 0.005), and 17% (p < 0.005), respectively, compared to a control shoot Cd level of 41.3 ± 4.93 mg kg⁻¹ in the first harvest (Figure 3A). In the second harvest, these treatments achieved reductions of 20% (p < 0.001), 33% (p < 0.001), and 29% (p < 0.001), respectively, compared to a control shoot Cd level of 76.9 ± 0.68 mg kg⁻¹ (Figure 3C). Notably, the Cd concentrations in the rice plants from the second harvest were higher than those from the first harvest.

In the soil experiments, the results demonstrated that, compared to the control, the Cd concentration in the rice grains was significantly reduced by 36% and 42% with the Si₂ and Fe₁ + Si₂ treatments, respectively (

Table 3. Cd concentration of rice plants exposed to As-Cd co-contaminated soil.

Treatments	Root (mg kg−1)	Straw (mg kg−1)	Grain (mg kg−1)
Control	1.88 ± 0.03 abc	0.30 ± 0.02 b	0.29 ± 0.01 d
Fe1	1.49 ± 0.24 bc	0.19 ± 0.04 b	0.31 ± 0.04 d
Fe2	2.59 ± 0.35 a	0.91 ± 0.05 a	1.21 ± 0.09 a
Si1	1.33 ± 0.35 bc	0.26 ± 0.11 b	0.30 ± 0.03 d
Si2	1.98 ± 0.43 ab	0.28 ± 0.07 b	0.18 ± 0.02 e
Fe1 + Si1	1.02 ± 0.06 c	0.19 ± 0.05 b	0.33 ± 0.06 d
Fe1 + Si2	1.33 ± 0.35 bc	0.27 ± 0.08 b	0.17 ± 0.05 e
Fe2 + Si1	2.63 ± 0.57 a	1.19 ± 0.60 a	0.77 ± 0.10 b
Fe2 + Si2	2.69 ± 1.18 a	0.81 ± 0.44 a	0.65 ± 0.05 c
F value	4.67	6.65	118
P	<0.005	<0.0005	<0.0001

Fe₁: 100 mg kg⁻¹, Fe₂: 1000 mg kg⁻¹, Si₁: 80 mg kg⁻¹, Si₂: 400 mg kg⁻¹. Degrees of freedom = 8. Within a column, different lowercase letters indicate significant differences among treatments (p < 0.05) based on Fisher’s LSD post-hoc test, with three replicates per treatment (data presented as mean ± standard deviation).

). In contrast, the Fe₂, Fe₂ + Si₁, and Fe₂ + Si₂ treatments significantly increased the Cd concentration by 317%, 166%, and 125%, respectively (Table 3). Similarly, the Cd concentration in the straw was significantly elevated by the Fe₂, Fe₂ + Si₁, and Fe₂ + Si₂ treatments, resulting in increases of 202%, 296%, and 171%, respectively (Table 3). Although the Cd concentration in the rice roots exhibited a trend similar to that observed in the grains and straw, none of the treatments significantly affected the Cd concentration in the roots compared to the control (Table 3).

3.2.3. Iron Plaque-Mediated Metal(loid) Binding

The hydroponic cultivation system demonstrated robust elemental interactions, with Fe concentrations showing statistically significant positive correlations (p < 0.01) with both As and Cd levels in the root surface iron plaque across consecutive harvest periods (Figure 4). Notably, these synergistic relationships remained consistent in both experimental cycles. In comparison, soil-grown conditions revealed differential metal(loid) binding patterns: while a significant positive correlation (p < 0.05) was observed between Fe and Cd concentrations in the rhizospheric iron plaque, As accumulation exhibited no statistical dependence on Fe content under identical soil cultivation protocols (Figure 5).

3.3. Changes in Soil pH and As and Cd in Pore Water

3.3.1. Dose-Dependent pH Regulation

The soil pH exhibited differential responsiveness to Fe and Si amendments under varying application regimes (Figure 6). Individual applications of Fe₁ and Si₁ induced marginal pH reductions (Δ pH = −0.07 to −0.08), though these alterations lacked statistical significance (p > 0.05). In contrast, elevated doses of Fe₂ triggered substantial acidification (Δ pH = −0.53, p < 0.0001), while high-dose Si₂ unexpectedly elicited a marked alkalization effect (Δ pH = +0.72, p < 0.0001), revealing divergent dose-dependent mechanisms between the two elements.

Combined treatments revealed non-linear interactions influencing pH dynamics. Co-application of low-dose Fe₁ and Si₁ showed negligible pH variation (Δ pH = −0.15), suggestive of the additive neutralization of their individual effects. Notably, the pairing of low-dose Fe₁ with high-dose Si₂ reversed Fe’s inherent acidifying tendency, elevating pH by 0.47 units (p < 0.0001). Conversely, high-dose Fe₂ dominated pH regulation in composite treatments: when combined with low-dose Si₁, severe acidification ensued (ΔpH = −0.70, p < 0.0001), whereas synergistic interaction with high-dose Si₂ attenuated proton release, yielding a moderated pH decline (Δ pH = −0.18, p < 0.05).

3.3.2. Contrasting As/Cd Solubility Dynamics

Longitudinal monitoring of pore water metal(loid) concentrations revealed distinct temporal patterns in aqueous-phase behavior between As and Cd (Figure 7). The terminal sampling at the 120th day demonstrated the lowest dissolved As concentrations across all treatments, showing 26–90% reduction compared to initial measurements (Figure 7A). All amendment regimens achieved significant As concentration suppression relative to the unamended control (p < 0.05), with the maximum suppression magnitude reaching 76% in Si₂-treated systems. Contrasting mobilization trajectories emerged for Cd, where pore water concentrations peaked at the 120-day interval, achieving 1.1–3.6 fold increases from initial baselines (Figure 7B). Notably, Fe₂ and Fe₂ + Si₁ treatments exhibited Cd concentration amplification effects, elevating aqueous concentrations to 0.071 ± 0.037 mg L⁻¹ and 0.056 ± 0.038 mg L⁻¹, respectively, which is—statistically distinct from control levels (0.032 ± 0.024 mg L⁻¹, p < 0.05).

Multi-temporal monitoring of pore water Cd concentrations demonstrated a robust positive correlation (p < 0.01) with Cd accumulation in rice grains, in striking contrast to the non-significant relationship observed between As aqueous-phase levels and grain As content (Figure 8).

4. Discussion

4.1. Interactive Effects of Si and Fe on Rice Growth Under As and Cd Stress

4.1.1. Phytotoxic Impacts of As/Cd Contamination

In our experiments, hydroponic culture involved individual treatments of As and Cd, while soil culture adopted a combined treatment of As and Cd. The results consistently demonstrated the phytotoxicity of As, Cd, and their co-contamination. Prolonged exposure to both elements (14 days in hydroponics; full growth cycle in soil) resulted in significant biomass reductions, with root dry weight, shoot biomass, and grain yield decreasing by 67%, 56%, and 78%, respectively (Figure S1 and Figure S2). These findings align with established mechanisms of PTEs toxicity in rice agroecosystems, where exposure duration and concentration synergistically determine phytotoxicity severity [24,25,26].

The temporal dynamics of element toxicity exhibited concentration-dependent thresholds. For As(III), short-term exposure (5 days) to 0.8 mg kg⁻¹ unexpectedly enhanced biomass by 8–66%, potentially through hormetic stimulation [27]. However, this effect reversed under prolonged exposure, with 10-day treatments at 1.6–4.0 mg kg⁻¹ As(III) causing progressive biomass losses of 7–26% [26]. Lower As(III) concentrations (0.1–0.5 mg kg⁻¹) showed no significant phytotoxicity within 14 days, whereas 2.5 mg kg⁻¹ triggered ≥35% biomass reduction [28]. Similarly, Cd exposure displayed cumulative toxicity patterns: hydroponic systems exhibited 14–21% shoot biomass reduction under 0.56–2.25 mg kg⁻¹ Cd within 7 days [29], while soil experiments recorded 38% biomass loss at 1 mg kg⁻¹ Cd over equivalent durations [30]. Notably, even sub-lethal Cd concentrations (0.056–0.225 mg kg⁻¹) suppressed seedling growth after 13-day exposure [31], emphasizing the importance of exposure duration in toxicity assessments.

The binary contamination regime revealed synergistic interactions between As and Cd, with combined exposures inducing supra-additive phytotoxic effects. The most pronounced toxicity occurred under 3 mg kg⁻¹ Cd and 30 mg kg⁻¹ As co-exposure, causing systemic biomass reductions of 35–80% across roots, shoots, and grains [23].

4.1.2. Si/Fe-Mediated Stress Alleviation Mechanisms

Our experimental results revealed differential effects of Si and Fe in mitigating As/Cd co-contamination toxicity across cultivation systems. Hydroponic studies demonstrated that Si, Fe70, and their combination enhanced rice growth under As stress through significant biomass promotion, whereas no comparable stimulatory effects emerged under Cd exposure (Figure 1). This system-specific response was further corroborated in soil trials, where both sole Si application and the Fe₁ + Si₂ combination achieved significant biomass enhancement. These findings collectively suggest that optimized Si-Fe co-application can partially counteract As/Cd co-contamination impacts, with Fe application rate being a critical determinant—lower application rates may be ineffective for biomass enhancement, while excessive doses could induce phytotoxicity. Previous studies have mainly focused on the individual effects of Si and Fe on plant growth under single-element stress. Our research, however, highlights the importance of the interaction between Si and Fe in the context of As-Cd co-contamination. Fe excess, a common abiotic stress in paddy fields, severely impacts rice yield and grain quality, with yield reductions ranging from 12% to 100% depending on Fe stress intensity and varietal tolerance [32].

Silicon has been documented to improve photosynthetic efficiency and yield in rice [33]. As reported by Li et al. [34], Si fertilizer application increased rice yield by 17.15–25.45% in calcareous paddy soils. However, Fe application requires careful dosage control [34]. Excessive Fe supply (1.4 mg kg⁻¹) reduced photosynthetic efficiency and grain yield compared to 0.28 mg Fe kg⁻¹ treatment [35], though Si supplementation could mitigate these adverse effects. Field and greenhouse studies showed that Fe toxicity (1000 mg Fe kg⁻¹) decreased grain yield by 34.2% and 28.3%, respectively, compared to 300 mg Fe kg⁻¹ treatment, with most of the 19 tested varieties exhibiting significant reductions in both grain yield and root-shoot biomass at higher Fe levels [36]. The interaction between Fe and Cd stress appears complex. Liu et al. [37] observed that 1000 mg kg⁻¹ Fe application showed no significant effect on rice biomass under Cd-free or low Cd (2 mg kg⁻¹) conditions but significantly enhanced biomass under high Cd stress (10 mg kg⁻¹), with 2 g kg⁻¹ Fe maintaining non-phytotoxic effects. Hydroponic studies further clarified Fe dosage effects: Xu et al. [38] reported that 30 mg Fe L⁻¹ (3-day Fe-plaque induction) caused no biomass reduction, whereas 100 mg Fe kg⁻¹ significantly inhibited growth. Conversely, 50 mg Fe L⁻¹ (3-day Fe-plaque induction) could positively influence rice biomass under hydroponic conditions [24].

4.2. Si-Fe Interactions in Regulating As/Cd Accumulation

4.2.1. Si’s Tripartite Detoxification Pathway

Our integrated analysis of hydroponic and soil cultivation systems demonstrates that Si amendment consistently reduces As and Cd accumulation in rice plants through three synergistic mechanisms: (1) enhanced Si assimilation in plant tissues, (2) activation of antioxidant defense systems, and (3) transcriptional regulation of element transporter genes. These findings corroborate established models of Si-mediated PTEs mitigation [16,34], particularly through rhizospheric uptake inhibition and xylem transport restriction of toxic elements to aerial organs. Compared to previous research, our study provides more in-depth evidence for these mechanisms in the context of As-Cd co-contamination. Notably, rice genotypes with elevated shoot Si concentrations exhibited an inverse correlation with grain As levels, suggesting this trait could inform breeding strategies for low-As cultivars [39]. Molecular analyses revealed that Si application downregulates key transporter genes (OsNramp5 and OsHMA2), effectively limiting Cd uptake at root interfaces and subsequent vascular transport [40]. This dual-phase inhibition (root-level uptake and vascular translocation) provides a mechanistic foundation for Si’s protective effects.

4.2.2. FeSO₄’s Dual-Phase Modulation of Element Mobility

Our findings indicate that the application of FeSO₄ effectively alleviates As accumulation in rice plants, yet its effectiveness in reducing Cd is limited. Notably, soil experiments revealed a dosage-dependent Cd accumulation pattern, with 1000 mg Fe kg⁻¹ (as FeSO₄) increasing grain Cd by 317% compared to controls (Table 3). This paradoxical phenomenon can be attributed to two competing mechanisms: the formation of iron plaque and pH-mediated changes in bioavailability.

First, exogenous FeSO₄ application promotes the formation of iron plaque on the root surfaces. Under anaerobic flooded conditions, Fe²⁺ from FeSO₄ facilitates the enhanced development of iron plaque, which acts as an amphoteric colloid capable of adsorbing both cations and anions [19]. This surface adsorption effectively immobilizes As and Cd at the root interface, thereby reducing their translocation to the aerial parts of the plant. Numerous studies have confirmed that the adsorption by iron plaque significantly inhibits the migration of As and Cd in plants [23,29,41].

However, in addition to the iron plaque formation, FeSO₄-induced alterations in soil pH also play a crucial role in modifying As and Cd bioavailability. While an increase in pH typically enhances the mobilization of As into the soil solution [42], our experimental results showed that application of FeSO₄ decreased soil pH (Figure 6), consequently reducing the soluble As concentrations (Figure 7) and contributing to lower As accumulation in rice tissues (Table 2). In contrast, a reduction in pH plays a key role in Cd activation. Although low doses of FeSO₄ had minimal effects on pH, high application rates significantly acidified the soils (Figure 6 and Figure S3). Previous research has shown that decreasing the soil pH significantly increased the exchangeable Cd fractions [43]. This finding is consistent with our observations of elevated pore water Cd concentrations and subsequent plant uptake following FeSO₄ treatment (Table 3 and Figure 7B).

This differential pH sensitivity underscores critical environmental implications: while pH serves as a secondary regulator for As mobility [44], it acts as a primary determinant for Cd activation in soils [12]. These findings highlight the necessity for comprehensive pH monitoring when applying soil amendments to co-contaminated systems, as pH modifications may inadvertently enhance the mobility of both toxic elements.

Compared to previous studies, our research provides a more detailed understanding of the dual-phase modulation of As and Cd mobility by FeSO₄. The dosage-dependent Cd accumulation pattern we observed is a novel finding that can help in optimizing the use of FeSO₄ in soil remediation. By emphasizing the importance of pH monitoring, we also offer practical guidance for farmers and environmental managers dealing with As- and Cd-co-contaminated soils.

4.2.3. Synergistic Si-Fe Effects and Nutritional Homeostasis

Our experimental findings revealed that both the Si₂ treatment and the combined Si₂ + Fe₁ treatment effectively reduced As and Cd accumulation in rice grains, with the Si₂ + Fe₁ combination achieving the most pronounced reduction. Hydroponic experiments under identical pH conditions demonstrated that the standalone Si treatment likely suppressed As(III) uptake by competing for binding sites on ion channel proteins. This aligns with the established mechanism wherein the rice Si transporter Lsi1 serves as the primary pathway for As(III) root uptake [26]. In the Si₂ + Fe₁ treatment, the addition of FeSO₄ further diminished As accumulation through enhanced iron plaque formation on root surfaces, which exhibited strong adsorption affinity for As. Previous studies have not fully explored the synergistic effect of Si and Fe on the reduction in As and Cd accumulation in grains. Notably, previous studies have shown that Si supplementation promotes the development of a denser crystalline structure in iron plaque, increasing its thickness and improving Cd sequestration efficiency [45].

Intriguingly, while As-Cd co-contamination is known to suppress nutrient uptake in rice [46,47], our data suggest that controlled Si-Fe co-application may counteract this effect by restoring micronutrient homeostasis. This observation is corroborated by earlier reports of improved micronutrient bioavailability following exogenous Si or Fe supplementation [48]. Collectively, these results underscore the dual benefits of ratio-optimized Si-Fe combinations: the (1) simultaneous mitigation of toxic element accumulation, and (2) preservation of rice nutritional quality. The observed synergy between Si and FeSO₄ not only elucidates mechanistic interactions at the root−soil interface but also provides a scientific foundation for developing more precise remediation strategies in PTE -contaminated paddy systems.

5. Conclusions

The optimized co-application of Si and Fe at a molar ratio of 7.95:1 effectively mitigates As and Cd co-contamination in rice through dual mechanisms. Specifically, Si raises soil pH, facilitating the immobilization of Cd, while iron plaques formed from FeSO₄ application selectively adsorb both As and Cd. Our findings indicate that the synergistic treatment of 400 mg Si kg⁻¹ and 100 mg Fe kg⁻¹ achieved maximum reductions in As (37%) and Cd (42%) in rice grains, along with a significant 54% increase in yield.

However, it is crucial to recognize that excessive application of Fe (1 g kg⁻¹) led to soil acidification, resulting in a concerning 317% increase in Cd accumulation in grains. This underscores the importance of careful management of FeSO₄ applications to mitigate adverse effects on food safety and environmental health.

In summary, our study demonstrates that pH-plaque engineering can effectively balance food safety and agricultural productivity in contaminated paddies. This approach offers a theoretical framework for the safe utilization of As-Cd co-contaminated soils. Future research should prioritize refining FeSO₄ application rates to optimize pH levels while also considering the potential environmental risks associated with chemical amendments. Additionally, further investigation into the long-term effects of Si and Fe co-application on soil health and crop safety is essential. Addressing the limitations of our study, including the specific environmental conditions tested, will be essential for validating these findings across a range of agricultural systems.