Arsenic Behavior in Paddy Soils: Sorption Capacity and the Role of Algal Addition

Abstract

Rice is one of the world’s most consumed foods, and the cereal that most efficiently uptakes and accumulates As, contributing to human health risk. Flooded rice fields alter Eh-pH conditions and, consequently, the proportion of As(III)/As(V), favoring their accumulation in the crop. The use of algae in paddy soils can improve fertility and C-stock and affect chemical conditions and As availability. This study aimed to evaluate the effect of algae application on: As adsorption capacity in paddy soils from Sado, Portugal, changes in pH-Eh conditions in the soil–water environment, and consequent As speciation. Batch-based As adsorption assays were performed with different solid–solution ratios and Chlorella minutissima algae application, and fitted to the Freundlich and Langmuir linear models. In semi-continuous column assays, simulating rice field conditions, the effect of algae on the pH-Eh of soil pore water was evaluated. The soil quality assessment showed pseudo-total contents of As and other elements higher than Portuguese agriculture limits (11 mg As kg⁻¹), but their availability was low, posing no environmental risk. The studied soils had great As adsorption, which increased with algae application (1.07 mg g⁻¹). Algae application favored oxygenation, increasing Eh values, and maintaining As(V) species. This indicated a potential approach to reducing As(III) mobility.

1. Introduction

Rice (Oryza sativa L.) is one of the most widely consumed crops in the world, forming part of the staple diet of most of the world’s population [1]. It is one of the most efficient cereals in the uptake and accumulation of As in its shoots and grain, which can often lead to enrichments that may pose risks to human health, as it is a highly toxic and carcinogenic element [2,3,4]. Human toxicity resulting from the consumption of rice enriched with As depends on several factors (e.g., quantity consumed and number of days of consumption, age, gender, and body weight), but, in general, its continuous consumption causes lethal diseases such as cancer, high blood pressure, and cardiovascular or neurological disorders [4].

The source of As in the paddy fields can be natural, being associated with the geochemical background of the soils, or by anthropic activities such as the use of As-based agrochemicals and irrigation with As-enriched water [5].

The concentration of As in the soil solution directly depends on various biogeochemical processes. These include variations in pH-Eh conditions that regulate the As(III)/As(V) ratio, as well as the element interaction with different soil solid phases that lead to sorption–desorption processes or plant uptake [2,6,7]. In rice cultivation, most plantations around the world continue to be flooded during the growth phase. This creates favorable conditions for the conversion of arsenate (As(V)) into more toxic species, such as arsenite (As(III)) [8,9,10,11]. Arsenite is more readily available than As(V) because it is less retained by soil constituents, being more readily taken up by plants. Therefore, it is important to improve knowledge of cost-effective techniques that can be applied to rice fields to decrease availability and uptake of As in rice crop soil–plant systems, and that allow compliance with national and EU regulations for the maximum As contents in rice, currently 0.15 mg/kg [12]. The speciation of As and its availability to plants in rice field soils is controlled by several factors, including pH, redox potential, concentration of organic matter, phosphates, sulfates, and Fe or Mn oxides, among others. Therefore, mechanisms to avoid this availability or decrease its mobility focus on altering these soil characteristics [4,7]. The main control pathways are based on the application of additives or fertilizers, such as phosphates, to compete for the same plant absorption channels, the use of Fe or Mn oxides for surface complexes formation with As, or a microbial consortium that can exert control over As uptake and decrease its availability [7,13].

The use of biological processes to modulate oxidation–reduction conditions is a fascinating and sustainable approach to rice cultivation. This method involves the application of algae or other organisms capable of oxygenating the water column during the flooding phase of the rice cultivation process. The implementation of this technique is straightforward and offers a promising solution to the challenges posed by traditional rice cultivation practices, as it can take advantage of natural communities or organisms within the fields themselves (e.g., unicellular algae as genus Chlorella) to perform this function. In fact, the environmental conditions of paddy fields can be a sustainable support for algae growth [4]. The presence of algae can alter the oxidation–reduction conditions in the system, directly or indirectly affecting the availability of elements and their speciation and performing additional biogeochemical and edaphic functions, such as C fixation and changes in nutrient availability [4,11,14,15]. The oxygenation of the system, as a consequence of algae photosynthesis, can facilitate Fe oxides formation, which has a strong As immobilization capacity, and promote the oxidation of As(III) to As(V), decreasing the availability of As(III) or total As for rice uptake ([4] and references therein). Moreover, As(V) is more retained by different soil components such as clay minerals and oxides/hydroxides of Fe/Mn [16,17].

However, the new conditions resulting from the application of these biological techniques must also be considered to avoid changes in the microbiota environment and reduced crop yield resulting from nutrient consumption (e.g., the availability of phosphorus) or cause synergistic or antagonistic effects in processes typical of paddy soils. These effects can influence soil sorption capacity, which already naturally controls or influences the availability of As in the environment [18,19,20].

While most studies on algae have been conducted in aquatic systems, the geochemical behavior of water–soil systems, particularly in rice paddy soils, is limited. The focus of these studies has been primarily on direct mechanisms of algae affecting soil fertility or the availability of metals [4]. Therefore, it is crucial to ascertain the combined effect that algae application can exert on the soil’s sorption mechanisms. Furthermore, it is also fundamental to understand how paddy soils respond to an increase in As in water and whether this poses an environmental risk. Since information on these interactions and combined effects is relatively scare, this study was conducted with the objectives of determining the effects of algae application on: (i) As adsorption capacity in paddy soils from Sado (South of Portugal) under a wide range of As concentration in water, and (ii) on changes in the pH-Eh conditions in the soil–water environment, and consequent As speciation.

2. Materials and Methods

2.1. Soil Sampling and Characterization

The collected soil samples correspond to the surface horizon, until 25 cm of depth, from areas used for rice cultivation on the riverbanks of the Sado River (Alcácer do Sal, South of Portugal). These soils were developed on alluvial plains and are classified as Irragric Anthrosols [21]. Rice has been cultivated in these soils since the 18th century [22]. Nowadays, the rice fields in Alcácer do Sal (38°22′06.7″ N 8°32′14.2″ W) cover a total area of 78.2 km² [22].

Sampling was conducted prior to the planting season. Composite soil samples (total ≈ 3 kg) were collected from three different rice fields using plastic tools. Representative composite soil samples were obtained from each rice field by the collection of multiple samples from different points across the area (n = 5). The samples were then stored in labeled polyethylene bags, homogenized, and divided into two parts: one part was stored at 4 °C, and the other was air-dried and sieved to 2 mm.

The dry samples (fraction < 2 mm) was used to the determination of the following parameters: pH, electrical conductivity (EC), and redox potential (Eh) in water at a ratio of 1:2.5 (m/V) [23]; organic C content by wet oxidation [24] expressed as organic matter by Van Bemmelen conversion factor (OM = OC × 1.72); total N by Kjeldahl method [25]. The macro- and micro-nutrients and potentially hazardous elements in the pseudo-total fraction, after acid digestion [26], and in the available fraction, extracted with an aqueous solution at 1:20 m/V [27], were determined by ICP-OES and ICP-MS in an international certified laboratory (Actlabs, Ancaster, Canada). In these aqueous solutions, which simulate soil pore water, pH, EC, and Eh were also determined. Arsenic concentrations associated with soil organic matter, total and non-crystalline Fe oxides, and Mn oxides were obtained, in the parallel method, with selective chemical extractions using sodium pyrophosphate [28], citrate–dithionite [29], acid ammonium oxalate [30], and hydroxyl amine chloridrate [31], respectively. These extractable soil solutions were analyzed by ICP-OES (Thermo Scientific iCAP 7000 series, Waltham, MA, USA).

Quality assurance and quality control of the analyses was made by analytical replicate samples, blanks, use of certified standard solutions and reference materials (OREAS 45d, 922, 907, 263, 130, 521, 602, 620, 623, 610; CLV-1; CDV-1; IV-STOCK-1643) and laboratory standards (Actlabs and Laboratorio de Pedologia of Instituto Superior de Agronomia–Universidade de Lisboa). The results obtained for certified materials had a recovery range from 86 to 110% while blanks were usually below the detection limit.

2.2. Arsenic Adsorption Assay

The adsorption assays were carried out as batch experiments in a growth chamber under controlled conditions (25 ± 1 °C; in darkness) with the same interaction time and procedure similar to that described by Arán et al. [32,33]. The initial concentration of As used in the assays varied between 0 and 150 mg L^–1, of As(V) a wide range of contents, and the upper limit raised to test the limits of the sorption process. The As solutions were prepared using a sodium arsenate (Na₂HAsO₄·7H₂O; (CAS. 10048-95-0; ≥98.0% in purity), supplied by Sigma–Aldrich (Burlington, MA, USA). Suspensions of the soil sample (dry fraction < 2 mm) were prepared at different solid: solution ratios of 12.5, 25, and 50 g L⁻¹. The pH of the suspensions was adjusted to pH 6 by the addition of 0.1 or 1.0 mol L⁻¹ HNO₃ (CAS. 7697-37-2, 70% in purity, Sigma–Aldrich) or NaOH (CAS. 1310-73-2, ≥99% in purity, Sigma–Aldrich), to achieve the pH of natural field conditions.

The suspensions were shaken at 170 rpm for 24 h, which is sufficient time to reach the adsorption equilibrium. At the end of the agitation, the soil suspensions were filtered through 0.45 μm filters, and the total concentration of As in solution (As_s) was measured by ICP-OES. The As sorption fraction was determined by subtracting the amount of As in solution (As_ad) from the total amount of added As.

The effect of algae on the sorption processes was evaluated at a soil: solution ratio of 50 g L⁻¹ and with the constant application of 1 mL of Chlorella minutissima suspension (0.2247 g of dry mass L⁻¹) in batch-type experiments as previously described. The effect of the algae alone was also determined with the As contents used in the adsorption assays.

2.3. Empirical Adsorption Models

Data obtained in the As adsorption assays were described using two of the most common empirical models in the descriptive analysis of adsorption processes, which allow checking how the system behaves as a homogeneous or heterogeneous surface, being the linear models for Freundlich (1) and Langmuir (2), as well as the equilibrium factor (R_L) (3) was determined [34,35,36,37]. The models correspond to the following equations:

$$Ln\left(qe\right)=\ln \left(K_F\right)+{\displaystyle \frac{1}{n}}\times \mathrm{l}\mathrm{n}(C_e)$$

(1)

Ce/qe = 1/(Q_max·K_L) + Ce/Q_max

(2)

R_L = 1/(1 + K_L·Ce)

(3)

In both models, qe is the concentration of As adsorbed per unit of mass of soil (mg g^–1), and Ce is the concentration of As in solution (mg L^–1) at equilibrium. In the Freundlich model, the parameter K_F is a constant related to the adsorption capacity (mg g⁻¹) (L mg^–1)^1/n, while the parameter n is a constant related to the intensity of adsorption and the heterogeneity of the binding sites. For the Langmuir model, the parameter Q_max represents the maximum adsorption capacity of the soil (mg g^–1), while the parameter K_L is the adsorption constant related to the affinity between the solute and adsorbent. In both linear models, the correlation factor (R²) and the sum of squares of the residuals (SSR) were calculated to evaluate the degree of fit of the model to the obtained data.

On the other hand, the parameter R_L is the separation factor or equilibrium factor, which is a dimensionless constant used to predict whether the adsorption process is unfavorable (R_L > 1), linear (R_L = 1), favorable (0 < R_L < 1), or irreversible (R_L = 0). All calculations for model fitting and graphical representation were performed using the Excel software.

2.4. Effect of Algae Application on Soil pH-Eh Conditions and As Speciation

To evaluate the effect of Chlorella minutissima algae on the pH-Eh conditions of the soil–water environment, a column assay under semi-continuous conditions was conducted to simulate field conditions during the rice emergence phase. During this phase, the soil was saturated and covered by a ~20 cm water column. This condition is used in the field to protect against pest outbreaks. However, it is also the phase in which As is more mobile and available [9,19].

Three glass columns, each 40 cm high and 4.5 cm in diameter, were set up. The following materials were placed inside them: glass wool at the base to prevent obstruction of the column (0.5 cm thick); 8 g of washed quartz sand (0.5 cm thick); 175 g of fresh rice field soil (total fraction) kept at 4 °C (10 cm thick). Distilled water was then added to saturate the layers, reaching a water column height above the soil of 20 cm and obtaining a total volume of 415 mL. During the first 24 h, the system was kept at rest, and subsequently, 20 mL of water (<5% of the total system volume) was renewed daily. This volume was collected by percolation at the bottom, and the same amount of distilled water was added to the surface to maintain a constant height of the water column above the soil. The pH, Eh, and EC were determined in these percolated water samples. The soil–water system was evaluated for 20 consecutive days, prior to the single application of 35 mL of the algal inoculum (7.865 mg of C. minutissima) in each column (equivalent to 49.46 kg ha⁻¹). After the application of the algae, the system was evaluated for an additional 60 days using the same procedure. The assay was placed in room conditions (~25 °C; 12 h light/12 h darkness).

The variation in As speciation in the mobile/available soil fraction along the assay period was evaluated by thermodynamic modeling using the program PHREEQC Interactive (version 2.14) [38]. According to the pH and redox conditions obtained from this column assay and the As contents obtained initially in the pore water fraction of the studied paddy soils, which represent the real amount of As in the system, the speciation of the As forms was determined. In the modeling, the pH value and available As content were kept constant, while the ‘pe’ values (electron activity calculated by the Nernst equation) were varied according to the Eh results obtained in the column over time. To simplify the behavior, the average redox value was considered every three days.

A descriptive analysis was performed on the results obtained from the columns. Moreover, to evaluate the effect of the algae application on pH-Eh changes in the columns, a statistical analysis was performed in two different ways. First, the results obtained for both variables (Eh, pH) over time were used directly with the Mann–Whitney U test at a significance level of p ≤ 0.05. Second, the slopes between the values of the entire time series and for each variable were calculated, subsequently performing the same test and significance level. This test does not assume a normal distribution and is appropriate for ordinal or continuous data that do not meet parametric assumptions. All statistical analyses were performed using IBM SPSS Statistics for Windows, version 26 (IBM Corp., Armonk, NY, USA).

3. Results and Discussion

3.1. Physicochemical Quality of Paddy Soils

The soils presented a slightly acidic reaction condition (pH = 6.16 ± 0.18) in a non-saline environment, with low average electrical conductivity (313.75 ± 243.45 µS cm⁻¹). These conditions are considered suitable for growing rice [39,40]. The organic matter content was medium (22.69 ± 4.57 g kg⁻¹; [39]), while the N content varied from medium to low (1.97 ± 0.17 g kg⁻¹). These concentrations contribute to C/N ratios of ~11.5, which are consistent with normal conditions and indicate a slight tendency towards mineralization of organic matter [40].

The contents of macro- and micro-nutrients in pseudo-total and available fractions are presented in Figure 1. In general, the available concentrations of Ca, Mg, K, and P in the soils were less than 1.5% of the pseudo-total content, while the available fractions of Na (354 mg kg⁻¹) and S (170 mg kg⁻¹) were significant, representing 46 and 34% of their total amounts, respectively (Figure 1). The available P concentration is considered low, with mean values from 2.1 mg kg⁻¹, a level that is deemed unsuitable for growing rice [39,40].

The Al and Fe contents in the pseudo-total fraction were high (g kg⁻¹-Al: 32.9 ± 3.9, Fe: 40.4 ± 0.4) while their availability represented less than 0.01% (mg kg⁻¹-Al: < 10, Fe: 2.9 ± 1.1). Given the importance of some Fe and Mn solid phases in the sorption mechanisms with anionic species, the concentration of total and non-crystalline Fe oxides and Mn oxides was determined, as well as the concentration of As associated with these solid phases and organic matter. The concentration of Mn associated with Mn oxides varied between 284.7 and 403.1 mg kg⁻¹, corresponding to 23 and 47% of the pseudo-total Mn concentration. The greatest contribution of Fe was associated with reactive solid phases with low crystallinity and non-crystalline, accounting for 53.9% of the pseudo-total Fe concentration. Meanwhile, Fe associated with crystalline forms and OM reached 16.2% and 29.91% of the pseudo-total fraction, respectively. These data highlight the importance of Fe oxides in the soil in controlling the mobility and availability of various metals and metalloids, such as As [10]. It also suggests that Fe oxides may affect the availability of P, which exhibits similar interaction mechanisms to As [41,42,43].

The pseudo-total contents of some metals and metalloids in the soils are presented in

Table 1. Concentration of metals and metalloids in the pseudo-total and available fractions of the studied rice paddy soils from Sado (mean ± standard deviation; n = 3). Maximum allowed values (MAV) for soils with agriculture use according to Portuguese legislation [44]. Nd: not determined.

Elements	Total	Available	MAV
	mg·kg−1
As	16.4 ± 6.72	<2.0	11
Sb	0.36 ± 0.21	<0.2	1
Cd	0.37 ± 0.34	<0.2	1
Cu	72.5 ± 11.5	0.46 ± 0.25	62
Zn	152 ± 96.8	4.85 ± 5.64	290
Ni	36.2 ± 4.88	<0.6	37
Pb	31.3 ± 1.63	<1.0	45
Co	17.95 ± 5.16	<0.10	19
Cr	53.50 ± 3.54	<2.0	67
Hg	0.17 ± 0.04	Nd	0.16

Highlighted indicates greater than MAV.

. The concentrations of As and Cu in the pseudo-total fraction exceeded the maximum allowed values (MAV) of the Portuguese legislation for agriculture use [44], while Hg is close to MAV. Nonetheless, the availability of the same elements was low regardless of the pseudo-total concentrations. The available fraction of As represented 12% of the pseudo-total content, corresponding to ~0.1 mg L⁻¹ in soil pore water (Table 1). Most of the As in the soils was associated with Fe oxides (~65% of the pseudo-total Fe concentration), followed by organic matter (~25%), and Mn oxides, without any relationship. The available concentration of Cu varied between 0.28 and 0.64 mg kg⁻¹, with this element being mainly associated with complexes with organic matter (~56% of the pseudo-total Cu concentration).

The presence of As in different rice paddy systems in Portugal was also previously reported by other authors, such as Signes-Pastor et al. [45], as well as recent studies in the mid-northern region of the country, both for As, Cu, Pb, and U in both rice cultivation soils and abandoned fields [46]. The pseudo-total concentrations of As in cultivated rice soils from Vouga (15.56–24.13 mg kg⁻¹; [46]) as well as from south of Portugal and other localizations from Spain (~3–20 mg kg⁻¹; [45]) were in the same range as those obtained in the present study. However, the abandoned rice soils from Vouga had higher variability and maximum values (27.36–82.92 mg kg⁻¹; [46]).

3.2. The Soil Adsorption Capacity of As and the Effect of Algae Application

The adsorption capacity of As in the studied soils, as well as the effect of algae application on adsorption mechanisms, was determined using adsorption curves in linear form (Figure 2; Figure S1). The results were fitted to the Langmuir and Freundlich models, and the obtained constants and parameters from these fits are presented in

Table 2. Parameters of linear Langmuir and Freundlich models for As adsorption isotherms in the studied rice paddy soils from Sado, without algae at different solid–solution ratios and with combined application of algae.

Sorption Model	Adsorption Parameter	Relation Soil–Solution (g·L−1)
		Soil			Soil + Algae
		12.5 (n = 13)	25.0 (n = 16)	50.0 (n = 26)	50.0 (n = 33)
Langmuir model	Qmax (mg·g−1)	0.1770	0.3270	0.4538	1.0717
	KL (L·mg−1)	4.8486	2.6757	0.9250	0.4792
	RL	0.002–0.026	0.003–0.046	0.010–0.219	0.014–0.3610
	R2	0.9041	0.9257	0.9575	0.9408
	SSR	0.4050	0.7737	12.543	30.645
Freundlich model	KF (L·g−1)	0.2385	0.2628	0.1798	0.2000
	n	2.9958	3.7651	2.3987	3.2072
	R2	0.9619	0.9665	0.9525	0.9502
	SSR	0.0551	0.0713	0.8480	1.0529

.

The Freundlich model was a better fit than the Langmuir model in all the evaluated conditions, obtaining higher R² values (>0.95) and lower values for the sum of the squares of the residuals (SSR), demonstrating greater model fit and response (Table 2).

It was evident that the reaction in the adsorption processes was favorable, with R_L values between zero and one, although with a decrease in the adsorption constant (K_L) with the increase in the solid/solution ratio. This suggests that the adsorption affinity between As and the soil decreased by 0.0997 L mg⁻¹ with an increase in the solid–solution ratio (Table 2). Although the adsorption affinity decreased, the number of interaction sites or adsorption positions increased, evidencing the highest adsorption capacities (Q_max) at the soil ratios used of 50 g L⁻¹, with retention values of 0.45 mg As g⁻¹ (Table 2).

The effect of Chlorella minutissima application in batch assays showed an improvement in the Q_max, although with a strong decrease (~50%) in the As–soil affinity constant (Table 2). The fit of the adsorption curves to the Freundlich model also presented a behavior consistent with the Langmuir model, showing that, although the processes are favorable, these are of low affinity, with the K_F constants presenting low values in all the evaluated ratios and not showing a clear relationship between changes depending on the ratio m/V used (Table 2). The fit of the adsorption curves in the presence of the algae showed a slight increase in the affinity, the opposite condition to that observed by the Langmuir model fit, which occurs in the presence of a more heterogeneous surface (Table 2). The maximum As retention capacities obtained from the fit to the Langmuir model, as well as the other parameters of both models, were similar to those observed in other rice fields and soils under saturated conditions [36,37,47].

The combined application of algae with soil did not show a clear improvement in the immobilization or removal of As from the aqueous medium. Although the Qmax, maximum adsorption capacity obtained with the Langmuir model adjustment, was higher, the affinity constants in this model and in the Freundlich model were low. This low affinity can significantly influence the efficiency of this process or its reversibility in the medium. In summary, the presence of the algae affects the affinity between As and the soil solid surface and may also be associated with the surface morphology itself or changes in surface charge.

3.3. Effect of Algae Application on pH-Eh Conditions and As Speciation

The evaluation of the pH-Eh conditions in the pore water soil system from percolation columns, on which the dynamics of soil saturation were evaluated, as well as the subsequent application of the algae, is shown in Figure 3.

The biogeochemical behavior of As in the water–soil–plant system depends highly on its oxidation state. The soil saturation conditions quickly led to obtaining low or negative redox potential values, reaching a minimum value of −78.85 ± 13.15 mV after percolation of 510 mL, while the average pH values for that period were 7.36. These conditions were also obtained by field measurements carried out in a flooded rice field in Alcácer do Sal, reaching values in the range of −40 to −170 mV, with pH values of 7.01 ± 0.29, electrical conductivity of 341 ± 105 µS/cm, and water temperature of 24.4 ± 1.58 °C. In this initial stage, where greater reduction processes were obtained, the dominant As species obtained by thermodynamic modeling was H₃AsO₃, favoring the change in As(V) to As(III) (Figure 4). The As(III) form is less retained by soil constituents, resulting in greater amounts of As (III) species present in pore water. This makes As more readily available for uptake by rice plants.

The application of the algae contributed to an increase in the oxidation conditions, with an effect observed from the third day, and Eh values increasing in the range of 150–200 mV. These Eh values were maintained throughout the test period (Figure 3) even with the periodic entry of fresh water into the system. Both statistical analyses showed significant differences between data obtained before and after algae application (p < 0.01). The pH maintained a very slight and constant upward trend over time, showing significant differences before and after the algae application (p < 0.01). While these differences were observed across the entire data set in the time series, evaluating the slope changes at each point in the time series showed no differences with the effect of the algae application (p = 0.408).

Under these new oxidation conditions, because of the algae application, the As forms observed through the modeling performed in PHREEQC were As(V) species such as HAsO₄⁻² or H₂AsO₄⁻, each representing ~50% of the As species in the medium (Figure 4). This conversion of As(III) to As(V) is conditioned by the oxygenation exerted by the algae photosynthesis process in the water column above the soil, favoring its conversion into forms less available for plant uptake [11,48]. Moreover, As(V) can be more adsorbed by soil components from soil [16,17], leading to a decrease in the available As level in the water–soil–plant system. Therefore, this change resulting from algae addition can play an important role in the biogeochemical behavior of As in flooded paddy soils.

4. Conclusions

Rice paddy soils from Sado had great As adsorption capacity, being able to retain up to ~0.45 mg As g⁻¹. This effect increased with the application of Chlorella minutissima algae. However, despite the increase in maximum retention capacity to 1.07 mg As g⁻¹, adsorption affinity was significantly reduced by almost half. This synergistic effect of algae should be approached with caution.

The application of algae to the water column of flooded paddy soil and their photosynthetic activity oxygenated the system, leading to significant changes in Eh-pH conditions, particularly the transition from suboxic to oxic conditions. Algae development along the assay maintained this oxic condition even when fresh water entered the system. Consequently, the As(V)/As(III) ratio in the percolated pore water varied, raising As(V) and reducing mobility and potential accumulation risk of the more toxic As species (As(III)).

These results support the potential use of Chlorella minutissima as a strategic approach to rice cultivation with lower As risk. However, further research is needed to better understand its interaction with other elements in the soil solution, the conditions that favor As desorption processes, and the optimal application rates of algae to improve this biogeochemical behavior at the lowest possible cost.