1 Introduction

Food crops often suffer abiotic stress, such as pesticide and heavy metal residues, from contaminants in soil (Bigott et al. 2021; Feng et al. 2021; Li et al. 2023a; Rose et al. 2022). Herbicides residues in soil may cause yield loss and contamination of agricultural products. As a representative of long residual herbicides, atrazine (6-chloro-N2-ethyl-N4-isopropyl-1,3,5-triazine-2,4-diamine, ATZ), a triazine herbicide that kills weeds by interrupting photosystem II (PSII) by blocking the electron transport chain in the chloroplasts (Qu et al. 2021), played an important role in maize weeding in every continent except the European Union, particularly in the United States, Brazil, and China (Chang et al. 2022; Hu et al. 2023). However, repeated application of ATZ may lead to accumulation and long-term persistence of ATZ in the soil, as evidenced by the recent detection of ATZ in German soil 21 years after being banned (Vonberg et al. 2014). ATZ residues in soil have posed a serious threat to the survival of nontarget organisms (Chowdhury et al. 2021), and the most direct impact is on sensitive rotation crops (Ramanathan et al. 2022). The agricultural system of corn-soybean rotation has been widely implemented in China (Liu et al. 2019). Residual ATZ in soil may injure the succeeding soybeans and other ATZ-sensitive crops (Huang et al. 2022), which is an urgent problem to be solved in this system. Therefore, it is necessary to develop strategies for minimizing crop uptake of herbicides from soils to ensure grain security and food safety.

It is important to determine the critical value of atrazine injury to soybean seedlings in soil pore water to prevent soybean injury. Several studies have reported that the highest tolerance concentration of ATZ in soil for soybean seedlings ranged from 0.791 to 1 mg kg−1 (Li et al. 2014; Wang and Ji 2002); however, recent research has shown that contaminants dissolved in soil pore water are the available part for crop uptake (Li et al. 2019). This means that, rather than the total content in the soil, that determines whether plant seedlings are injured or not. Therefore, reducing the concentration of ATZ in soil pore water is an optional strategy to reduce uptake by crops. Several technologies have been proposed to maintain healthy soil, such as microbial degradation, plant absorption and metabolism, photolysis, advanced oxidation, etc. (Rostami et al. 2021). However, these technologies often have the disadvantages of not adapting to the actual environment, not adapting to the actual agricultural production or high costs. Carbonaceous sorbents can adsorb numerous inorganic and organic contaminants from water, thus reducing their bio-accessibility (Li et al. 2020). Among them, biochar, a carbon-rich material pyrolyzed from biomass under oxygen-limited conditions, has been extensively studied as a low-cost soil amendment for immobilizing organic contaminants via surface adsorption. Biochar amendment in soils can help crops effectively resist growth stress caused by herbicides (Wu et al. 2019), polycyclic aromatic hydrocarbons (Zhang et al. 2021), pharmaceuticals (Li et al. 2020), and emerging organic contaminants (Hurtado et al. 2017). Previous studies have reported that biochar can mitigate the inhibitory effect of herbicides on crops, such as ATZ, oxyfluorfen, sulfentrazone, fomesafen, and isoproturon (Gao et al. 2022; Meng et al. 2019; Sopeña et al. 2012; Wang et al. 2022b; Wu et al. 2019). However, it is more important to focus on relationship between herbicide concentrations in soil pore water and crop bioavailability in biochar amended soil so as to use biochar in farmland more scientifically and economically. Biochar from many raw materials has been reported to have good adsorption properties to atrazine, such as corn stalk, peanut shell, rice stalk, etc. (Mandal et al. 2017; Wang et al. 2020). A study showed that compared with other raw materials, biochar prepared from rice straw has more environmental practical significance for atrazine adsorption (Mandal et al. 2017), so it was selected in this study.

This study aimed to evaluate the effect of biochar amendment on the distribution of ATZ in soil and bioavailability of soybean seedlings. Accordingly, we investigated (1) the no-effect concentration (NOEC) of ATZ in artificial culture solution (simulate soil pore water) to the seedling stage of conventional soybean, (2) the effect of biochar amendment on the distribution of ATZ in soil pore water, and (3) the change of ATZ concentration in soil pore water following the addition of biochar and used the results to discuss methods of improving atrazine-induced stress injury to soybean in seedling stage via biochar. In addition, as a guiding method for soil amendment via biochar, a model was proposed to predict the change in ATZ concentration in the pore water of various soils. This study provides useful information that can help better evaluate the potential effects of biochar amendment on the distribution and bioavailability of ATZ in soil pore water.

2 Experiment and methods

2.1 Chemicals and reagents

ATZ reference material; analytical-grade acetonitrile, methanol, NaCl and anhydrous MgSO4; HPLC-grade acetonitrile, methanol and formic acid; ultrapure water and Octadecylsilane (C18) cleaner were used in this study. The details are described in Text S1 and Text S2 of the Supporting Information.

2.2 Soils and biochar

To examine the influence of soil properties on herbicide distribution, five representative soil types in China were collected. They were fluvo-aquic (Cambisols), black (Phaeozems), cinnamon (Luvisols), red (Acrisols), and paddy soils (Hydragric anthrosols) (FAO (Food and Agriculture Organization of the United Nations) 2015; Zhang et al. 2022). The details are described in Text S3 and Table S1. The biochar was produced from rice straw according to a previously published method (Wang et al. 2020). The detailed process is described in Text S4.

2.3 Phytotoxicity of ATZ toward soybean seedlings in a culture solution

The seed pre-culture method and conditions are detailed in Text S5. The nutrient solution used was a quarter-strength standard Hoagland solution (composition is listed in Table S2) as described by Palermo et al. (2012).

The selected seedlings had a similar size (root length, 10 ± 1 cm; shoot height, 5 ± 1 cm) when they reached the first real leaf. The roots were washed with sterile water. Subsequently, they were transferred to 100-mL conical flasks containing 130 mL exposure solutions in six concentrations (0, 0.01, 0.02, 0.05, 0.1, and 0.2 mg L−1) of ATZ. We distributed one plant per flask and performed 10 replicates per herbicide level. The roots were immersed immediately below the surface of the solution. The containers without ATZ were used as controls to monitor the possibility of cross-contamination, whereas the containers without a seedling were used as unplanted controls to monitor ATZ loss in the absence of seedlings. The outer walls of the glass conical flasks and the caps were covered with aluminum foil paper to prevent possible algal blooms and ATZ photolysis. Fresh sterile nutrient solution without ATZ was replenished to a total volume of 120 mL. After 10 days of culture, chlorophyll fluorescence analysis was conducted to assess phytotoxicity.

2.4 Chlorophyll fluorescence analysis

Chlorophyll fluorescence analysis of the leaves was performed using a Dual-PAM-100 chlorophyll fluorometer (Heinz Walz GmbH, Effeltrich, Germany) that was computer-controlled by the Windows software Dual-PAM. PSII activities were quantified using chlorophyll fluorescence and P700 absorbance changes. The final results were expressed as the maximum quantum yield of PSII (Fv/Fm), the effective photochemical quantum yield of PSII [Y(II)], the quantum yield of light-induced non-photochemical fluorescence quenching [Y(NPQ)] and the quantum yield of non-light-induced non-photochemical fluorescence quenching [Y(NO)] (Wang et al. 2013). The detailed process is described in Text S6.

2.5 Biochar adsorption experiment in soil

Adsorption experiments were conducted using batch equilibration method (Man et al. 2021). About 30 ± 0.1 g of each type of soil was weighed in a 40-mL amber glass bottle. ATZ solution (1000 mg L−1) was added to each bottle and mixed with soil to prepare the samples with initial ATZ content of 0.05, 0.5, and 5 mg kg−1 (according to the actual residual concentration range in the field). After solvent volatilization, the bottles were with the lids, and shaken to mix the soil. Subsequently, 0.006, 0.03, 0.15, and 0.6 g of test biochar were weighed into the bottles and mixed with soil. The biochar in the samples was 0.02%, 0.1%, 0.5%, and 2% (w/w), respectively (according to the pre-experiment, the data are not shown here). Five replicates were set for each treatment. In addition, soil samples containing three concentrations of ATZ without biochar and samples mixed with biochar but without ATZ were prepared separately as control treatments. To activate the adsorption process in the samples, ultrapure water with 100 mg L−1 NaN3 (to suppress microorganisms) was added to the samples up to 30%. The bottles were stoppered with lids and stored in a PRX-1500D incubator (GREEN, Shanghai, China) at 25 °C in the dark. The samples were collected on days 0 (6 h), 1, 2, 4, 6, 9, 12, and 15. The pore waters were extracted immediately and the remaining soil samples were freeze-dried and stored at − 18 °C until extraction.

2.6 Soil remediation application

Pot experiments were conducted to verify biochar mitigation on the phytotoxicity of soil residual ATZ following previous literature methods (Wu et al. 2019). The test soil (black soil) was weighed for 400 ± 2 g in a 17 × 24 cm plastic ziplock bag. Then the ATZ solution was added to make the initial content of ATZ in the soil 2.0 mg kg−1. After solvent volatilization, 2.0 ± 0.1 g biochar was added in soil samples and made the proportion of biocahr in the soil 0.5%. Soil, ATZ and biochar were thoroughly mixed for soybean seedling culture experiment. All seeds were washed with 10% sodium hypochlorite, then planted in the treated soil and maintained in a controlled growth incubator for 15 days (14 h light/10 h dark, with 75% humidity and 22 °C, 40% soil moisture). Three treatments were set up in blank, ATZ and ATZ/ biochar, and each treatment was conducted in quintuplicate. At the end of this experiment, chlorophyll fluorescence analysis was conducted.

2.7 Sample extraction and analysis

Soil pore water was extracted using a high-speed centrifugation method which were described in a previous study (Li et al. 2022), with slight variations in the specification of the experimental equipment. Each soil sample (at least 0.5 g) was placed in a 5-mL disposable plastic syringe. Glass wool was placed at the bottom of the syringe to act as a filter sieve for separating soil and water. Each syringe was inserted into a 10-mL centrifuge tube and centrifuged at 2077×g for 10 min. The soil pore water was collected at the bottom of the centrifuge tube. An aliquot of the soil pore water was diluted to 100 times. The resulting liquid aliquot was passed through a 0.22-μm polytetrafluoroethylene membrane to an autosampler vial before analysis. Dry soil samples were extracted using a modified QuEChERS method. The detailed processes are described in Text S7.

The soil and pore water samples after pre-treatment were analyzed using an Agilent 1290 high-performance liquid chromatography system coupled to a G6470A triple-quadrupole mass spectrometer equipped with an electrospray ionization source (Agilent Technologies, Santa Clara, CA, USA). Analysis conditions are listed in Tables S3 and S4. The processes and results of the analytical methods validation are described in Text S8 and Table S5.

2.8 Statistics analysis

Student’s t-test and one-way analysis of variance (ANOVA) were performed to assess the significant differences between treatments and controls. The post-hoc least significant difference test was used to establish significant differences among the treatments. Statistical significance was accepted when P value was less than 0.05. Statistical analysis was conducted using SPSS 22.0 software for Windows (IBM Corp., Armonk, NY).

3 Results

3.1 Phytotoxicity of atrazine toward soybean seedlings in a culture solution

It has been mentioned in the introduction that the absorption of pollutants in soil by plants is directly related to their concentration in soil available water. This is because although the soil provides the physical basis and material conditions for the growth of plants (Fan et al. 2021), the material transmission is completed through free water as the transmission medium (Bazilevskaya et al. 2018). Soil pore water is the main form of available water absorbed by plants (Zangiabadi et al. 2020). In order to avoid the adhesion of soil particles to ATZ and to control the concentration in the water environment more precisely, soil pore water was simulated with artificial culture solution. ATZ induced significant changes in leaf photosynthetic potential quantum yield and photosynthetic activity following exposure to various ATZ concentrations for 10 days, as revealed by the results obtained for Fv/Fm (F [5, 12] = 14.75; P < 0.011), Y(II) (F [5, 12] = 3.81; P < 0.05), Y(NPQ) (F [5, 12] = 12.17; P < 0.01), and Y(NO) (F [5, 12] = 16.85; P < 0.01]. The growth of soybean seedlings is shown in Fig. 1.

Fig. 1
figure 1

Growth of soybean seedlings in culture solution of atrazine at various concentrations. Significance markers are statistical for plant height, n = 5. P ≤ 0.05 among different significance markers

Full size image

The Fv/Fm values decreased with increasing ATZ concentration (Fig. 2a). When ATZ concentration in the culture solution was 0.01, 0.02, and 0.05 mg L−1, the Fv/Fm values decreased by 0.5%, 4.0%, and 0.8%, respectively, compared with the blank control 0.806 (P > 0.05). The statistical results with no significant difference showed that a low concentration of ATZ caused fewer negative effects on soybean seedlings. However, the Fv/Fm values rapidly decreased when ATZ concentration reached ≥ 0.1 mg L−1. Under 0.1 and 0.2 mg L−1 ATZ treatments, the Fv/Fm values decreased by 7.4% and 53%, respectively (P < 0.05). The trend of Y(II) values also decreased with increasing ATZ concentration, although a small increase was observed at 0.05 and 0.1 mg L−1 ATZ (Fig. 2b). Compared with the blank treatment (0.218), the Y(II) value decreased by 26.9%, 42.2%, 21.3%, and 13.9% in the culture solution with ATZ concentrations of 0.01, 0.02, 0.05, and 0.1 mg L−1, respectively. However, a steep drop of 79.8% (P < 0.01) was observed at a concentration of 0.2 mg L−1 ATZ. Similar to the Fv/Fm values, Y(NPQ) did not significantly decrease (P > 0.05) when ATZ concentration was ≤ 0.05 mg L−1, being 6.5%, 8.6%, and 4.9% for ATZ concentrations of 0.01, 0.02, and 0.05 mg L−1, respectively (Fig. 2c). However, when the ATZ concentration reached 0.1 mg L−1, Y(NPQ) decreased by 41.0% (P < 0.01), and at an ATZ concentration of 0.2 mg L−1, Y(NPQ) decreased sharply by 93.2% (P < 0.01) compared with the blank control 0.370. Y(NO) increased with increasing ATZ concentrations compared with the blank treatment 0.412 (Fig. 2d). Similarly, when the concentration of ATZ was between 0.01 mg L−1 and 0.1 mg L−1, although it was significantly increased (P < 0.05), the amplitude was small; however, a significant increasing range was observed when the ATZ concentration reached 0.2 mg L−1 (P < 0.01). The same phenomenon was observed for fresh weight and leaf area as biomass assessment indicators (Fig. 2e, f), that is, when the ATZ concentration was 0.1 mg L−1, the reduction was only ~ 10% (P > 0.05) compared with the blank treatment (5.67 g for fresh weight and 67.6 cm2 for leaf area); however, when the ATZ concentration was 0.2 mg L−1, the observed data sharply reduced by 31.7% and 39.9%, respectively (P < 0.01). According to one-way ANOVA, only the 0.2-mg L−1 ATZ treatment made a significant difference in the chlorophyll fluorescence parameters of soybean seedlings compared with the blank treatment group; hence, 0.1 mg L−1 ATZ in culture solution was the NOEC and 0.2 mg L−1 was the lowest observed effect concentration of the soybean seedlings.

Fig. 2
figure 2

Fv/Fm (a), Y(II) (b), Y(NPQ) (c), Y(NO) (d), fresh weight (e), and leaf area (f) in soybean seedlings exposed to increased concentrations (0, 0.01, 0.02, 0.05, 0.1, and 0.2 mg L−1) of atrazine. P ≤ 0.05 among different significance markers, n = 5

Full size image

3.2 Distribution characteristics of atrazine in soil pore water system and regulation by biochar

The absorption of organic pollutants by plant roots is related to the effective concentration of organic pollutants in soil pore water (Li et al. 2022). To predict and manage the phytotoxicity of residual ATZ in the soil to soybeans, the distribution of ATZ between soils and pore water was studied. According to the data of ATZ concentration in each treatment, the adsorption process reached equilibrium around the 4th day. Therefore, in order to correct for error, the data of 4–15 days were selected, and the mean value after correcting natural dissipation was the adsorption equilibrium data. The results are shown in Table 1. When 50 μg kg−1 ATZ was added to dry soil and the absolute water content was 30%, the amount of distribution of ATZ in soil pore water was 0.0101, 0.036, 0.053, 0.031, and 0.016 μg for red, cinnamon, fluvo-aquic, paddy, and black soils, respectively. For 500 μg kg−1 ATZ samples, the values of 0.919, 0.598, 0.538, 0.354, and 0.085 μg in pore water were measured for the five soils, respectively. For 5000 μg kg−1 ATZ samples, 8.572, 6.687, 5.462, 3.246, and 1.743 μg were observed in soil pore water. In addition, ATZ was not detected in the control treatments.

Table 1 Atrazine distribution in soil pore water without biochar (μg/10 g soil)
Full size table

Biochars exhibit a strong adsorption property toward ATZ (Hernandes et al. 2022; Wang et al. 2022a). This study simulated the application of biochar in soil environments containing three ATZ concentrations. As shown in Fig. 3, the concentration of ATZ in pore water decreased significantly (P < 0.01) with the biochar treatment in all five soils. Among them, 0.02%, 0.1%, 0.5%, and 2% biochar treatment reduced the concentration of ATZ in the pore water of the five soils by 2.7 to 44.2%, 8.2 to 92.7%, 62.3 to 99.3%, and 85.7% to 99.8%, respectively. Because AZT is stable in the environment and has the characteristics of a long-term residue, the initial total amount of ATZ in all treatments did not change significantly during the period of this study.

Fig. 3
figure 3

Effect of increasing the proportion of biochar (0, 0.02%, 0.1%, 0.5%, and 2% w/w) on the concentration of atrazine in soil pore water. a The total atrazine content in soil was 50 μg kg−1. b The total atrazine content in soil was 500 μg kg−1. c The total atrazine content in soil was 5000 μg kg−1. All the same soil and atrazine levels showed significant differences (P ≤ 0.01) between each biochar content treatments, except for No-BC and 0.02% BC treatments in (c) (P > 0.05). n = 5

Full size image

3.3 Effect of biochar on soil adsorption capacity

The adsorption isotherm describes the equilibrium performance of an adsorbent at a constant temperature (Al-Ghouti and Da'ana 2020). Herein, Henry’s isotherm, the most fundamental adsorption isotherm (Chen et al. 2022), was selected to investigate the adsorption performance of biochar. Henry’s partition coefficient (Kd, unit is meaningless) and the determination coefficients (R2) are shown in Table 2.

Table 2 Kd (R2 in parenthesis) of atrazine in soils containing different proportions of biochar
Full size table

The values of R2 were at least 0.990, indicating that the adsorption performance was suitably described by Henry’s isotherm. In the soils without biochar, Kd was 0.90, 1.81, 1.95, 3.98, and 8.33 for red, cinnamon, fluvo-aquic, paddy, and black soils, respectively. The data show that ATZ is most easily combined with the black soil with the highest Kd in the natural state. Conversely, ATZ is can more easily become free of the moisture held in the red soil, causing soil water pollution. Soil organic matter content (OMC) was considered the key factor affecting the adsorption property. Their values in the five soils demonstrated a significant correlation (P < 0.05) with Kd values consistent with the findings of previous studies involving other organic pollutants (Alvarez-Esmorís et al. 2021; Zhang et al. 2020). In addition, pH and cation exchange capacity are not correlated with AZT adsorption (P > 0.05, data are not shown, and soil parameters are listed in Table S1).

The Kd of soil–biochar systems increased with increasing soil–biochar content. Compared with non-biochar treatments, the Kd of the five soils increased 0.2–40.6%, 15.8–101.7%, 3.0–31.5 times and 32.8–632 times when the biochar contents were 0.02%, 0.1%, 0.5% and 2%, all with significant differences (P < 0.05). Although there is a positive correlation between biochar content and Kd, the nonlinear trend of Kd produced skewness (owing to the common influence of various factors in the soil), thus rendering the data inconsistent with normal distribution. Herein, the reasonable log-transformation of Kd was employed to assess the normality assumption, and the results revealed that biochar content and Kd were significantly correlated (P < 0.05) except for in red soil, but there was still a correlation trend in red soil.

Generally, Kd is considered an inherent property of an adsorbent and is a constant. Therefore, Kd and the content of adsorbent in soil should be in an approximate linear proportional relationship. However, as shown in Table 2, Kd increased nonlinearly concerning biochar content in all the soil systems. This phenomenon was speculated to be associated with the competition of other substances in the soil for adsorption sites on biochar. This theory has also been supported by previous studies (Cao et al. 2009; Zhao et al. 2019).

.

3.4 Mathematical model for predicting the effect of biochar on atrazine concentration in soil pore water

To guide the management of ATZ residue in environmental soils using biochar, a mathematical model for predicting ATZ concentration in pore water under (or non) biochar amendment conditions was developed termed as Adsorption Regulation Model. Based on this model, ATZ bioavailability to soybean seedlings in soil pore water and biochar regulation strategies can be determined alongside the critical concentration discussed in Sect. 4.1.

According to Henry’s isotherm, under natural conditions, ATZ distribution in soil particles and pore water exhibits a linear relationship, and mathematical deduction reveals that the total ATZ content in water-bearing soil also demonstrates a linear relationship with pore water distribution. The model is as follows and the mathematical derivation can be found in the Text S9.

$${C}_{w}=\frac{{C}_{t}(r+1)}{r+{K}_{d}}$$
(1)

Cw is the concentration of ATZ in soil water (pore water), μg L−1; Ct is the total content of ATZ in water-bearing soil, μg kg−1; r is the absolute moisture content of the soil, indication as decimals; Kd is the partition coefficient, and it is the same as in Henry’s isotherm.

The theory that Kd is positively correlated with soil OMC and proportion of biochar in soil has been discussed in Sect. 3.3. Therefore, the conversion of Kd from OMC and biochar content in soils should be simultaneously considered. A zero-intercept linear function (Y = 0.1146X, R2 = 0.94) was fitted, where the OMCs of the five soils are X and the corresponding Kd is Y. Similarly, the association between biochar content and Kd can be fitted as an equation Y = 125X2 with the R2 ranging from 0.996 to 0.999 and a relative standard deviation of 7.7%. Finally, the approximate functions of Y = 0.1146X and Y = 125X2 were selected as the conversion coefficient between soil organic matter and Kd. Equation 1 can be rewritten as

$${C}_{w}=\frac{{C}_{t}(r+1)}{r+0.1146*OMC+125*{B}^{2}}$$
(2)

OMC is organic matter content of soils, indicated as percentages; B is biochar content in soil, indicated as percentages.

Equation 2 is suitable for obtaining relatively accurate ATZ concentration in soil pore water from the OMC of soils and biochar proportion in the absence of Kd. The accuracy of the model was verified and analyzed through 80 observation data (Shown in Table S6) of ATZ concentrations (shown in Fig. 4). The mean absolute percentage error (MAPE) and linear fitting of predicted and measured data were used as evaluation indices of the model accuracy. According to the MAPE, the statistical results of 14.1% indicated that the model has a small error. The slope of linear fit was 1.0, indicating that the model has a high prediction accuracy. The R2 represents the goodness of the fitting curve. The statistical result of R2 in this study was 0.91, indicating that the model has a strong precision in predicting ATZ distribution in soil pore water (The data are listed in Table S7). This study indicates that ATZ concentration regulation in soil pore water using biochar can be predicted accurately via the model Eq. 2, which is suitable for guiding the control of ATZ pollution in farmland. We believe that this model is suitable for non-sandy soils where ATZ does not leach quickly.

Fig. 4
figure 4

Predicted and observed atrazine concentrations in soil pore water by the Adsorption Regulation Model. The fitting line is a linear regression result

Full size image

3.5 Verification of soil remediation application

The chlorophyll fluorescence analysis data are presented in Table 3. The growth (data of Fv/Fm, Y(II), Y(NPQ), and Y(NO)) of soybean seedlings of the blank treatment was basically the same as that in the herbicide-free treatment group in Sect. 3.1. When the content of ATZ in dry soil was 2 mg kg−1, the theoretical concentration in pore water under 40% soil moisture was 0.47 mg L−1. The data showed that the soybean seedlings in this condition had the similar degree of injury as those in 0.2 mg L−1 ATZ culture solution (Sect. 3.1). Compared with ATZ treatment alone, the four data of soybean seedlings in ATZ-0.5% biochar treatment group were 95.6%, 76.0%, 79.4%, and 77.7% of healthy plants in blank treatment, respectively (Fig. 5), showing significant differences (P < 0.01). The results of this study confirm the management effect of biochar on the phytotoxicity of residual ATZ in soil.

Table 3 Results of soil remediation application experiment
Full size table
Fig. 5
figure 5

Comparison of blank, atrazine and biochar remediation treatments

Full size image

4 Discussion

4.1 Maximum no-effect concentration of atrazine residues in soil for soybean seedlings

Herein, we examined the toxic effects of ATZ toward soybean seedlings because of the widespread occurrences of ATZ residues in farmland soil (Dou et al. 2020) and crop injury. As the most prominent PSII inhibitor herbicide, ATZ obstructs normal photosynthesis in plants by blocking photosynthetic electron transport (Zharmukhamedov and Allakhverdiev 2021). Therefore, herein, the quantum yields of PSII and energy dissipation were analyzed to detect the responses of PSII activities to ATZ treatment and to detect the critical concentration of soybean seedling toxicity under ATZ stress. As described in Sect. 3.1, when ATZ concentration in environmental water reached 0.01 mg L−1, Y(II) of soybean seedlings growing in it began significantly decreasing, indicating that soybean seedlings exhibited a strong sensitivity toward ATZ. However, Fv/Fm, Y(NPQ), and Y(NO) were relatively stable when ATZ was < 0.1 mg L−1. These results indicate that although stress was induced in soybean seedlings under 0.01, 0.02, and 0.05 mg L−1 ATZ, they did not lose the self-photoprotection ability under excess light conditions (Osmond et al. 2022). Fv/Fm did not significantly change (< 10%), indicating that the maximum quantum yield of PSII and potential maximum photosynthetic efficiency of soybean seedlings grown in environmental water exhibited limited changes. Therefore, 0.1 mg L−1 is the NOEC of ATZ in soil pore water that does not exert phytotoxicity toward soybean seedlings. If the ATZ in soil pore water is lower than or equal to this concentration, soybean seedlings may self-adapt and recover under ATZ stress. Otherwise, soybean seedlings will suffer severe and irreversible herbicide injury.

Several NOEC data of ATZ or other herbicides for numerous crops have been reported. Among them, two NOEC data were noted, 0.791 and 1 mg kg−1 ATZ residue in the soil of soybean seedlings (Li et al. 2014; Wang and Ji 2002). However, recent research has shown that compared with the highly variable factors derived from soils, relatively constant factors based on pollutant concentration in soil pore water were superior for describing organic pollutant accumulation in plants grown in soils (Li et al. 2022). This is related to adsorptivity controlled by physicochemical properties in different soils. Therefore, herein, the NOEC of ATZ in soil pore water toward soybean seedlings was proposed as a unified index in various soils, which signifies that the contaminant treatment targeting NOEC and method for purifying soil pore water will be further promoted.

4.2 Biochar reduced the concentration of ATZ in soil pore water through adsorption and cuts off transfer to soybean

Many studies confirmed the adsorption of ATZ by various biochar through adsorption experiments (Cheng et al. 2022; do Nascimento et al. 2022; Wang et al. 2022a). In this study, the total amount of AZT was constant before and after the experiment while the concentration in pore water decreased significantly with biochar increased, by up to 99.8%, which confirmed the existence of adsorption during the experiment. The soybean seedling culture tests discussed in Sect. 3.1 confirmed that a positive correlation was observed between the injury degree of plants and the ATZ concentration in pore water. Therefore, the low concentration ATZ in pore water caused by biochar adsorption is the direct reason for protecting plants from herbicide stress. Except for adsorption, previous studies also suggested that biochar may promote natural degradation (Kumari et al. 2021) and biological metabolism (Yang et al. 2017) of AZT, and the growth of plant probiotics (Huang et al. 2022). Many studies also reported that other abiotic stress of the plants that coexist with ATZ in soil, such as organic pollutants (Lin et al. 2022) and heavy metals (Yang et al. 2023), may also be avoided by the use of biochar.

Although adsorption is the main mechanism of pollutant isolation, the protection strategy of biochar to plants under natural conditions is the result of a synergistic effect. Biochar not only intercepts the transmission of pollutants to plants but also accelerates the dissipation of pollutants and promotes the healthy growth of protected objects. Therefore, biochar has a good contribution to the regulation of farmland soil and ecology (Bao et al. 2022).

4.3 Practical implications of biochar regulation models in agricultural production

Generally, soils exhibit a strong water-holding capacity, and the separation and analysis of the pore water contained within it is difficult under low water content conditions (Di Bonito et al. 2008). Therefore, the level of abiotic stress in plants is commonly predicted using the total amount of pollutants in the soil. Based on research results showing that ATZ concentration in pore water is more suitable for predicting phytotoxicity than the total content in soil, we developed a mathematical model (Eq. 2) to predict ATZ concentration in soil pore water based on the soil content. Although only ATZ was studied for the model, common parameters that impact the concentration of organic contaminants in pore water of various soils were found: water content and OMC. Changes in soil adsorption properties owing to biochar application were also considered in the model. Therefore, this model may not be limited to ATZ and may provide a foundation for studying pore water distribution and biochar regulation for more pollutants in soil. Several reported prediction models can also accurately predict the distribution of organic contaminants in soil (Albergaria et al. 2010; Arands et al. 1997). These models have the advantage of considering more factors and can be applied in more places. Compared to previous models, the model constructed herein excluded numerous complex parameters that can be ignored and introduced soil amendments/pollution remediation agents (biochar) as new parameters. These modifications were intended to make the model more suitable for predicting ATZ distribution in soils amended by biochar because the powerful adsorption capacity of biochar can greatly alter the adsorption–desorption equilibrium relationship of an organic contaminant in the soil. In addition, many soils were selected for the experiments and validation studies conducted to establish a prediction model, and they were collected from 15 regions in 11 provinces of China (shown in Table S1 and S6 for details). It can be considered that this model is suitable for more types of soil. Based on the prediction of this model, we conducted an applied study of ATZ pollution remediation in a black soil collected from Jilin Province, China. The OMC in the soil was 46.1 g kg−1. ATZ treatment level was 2 mg kg−1 Dw. According to the model obtained in this study, when the humidity is 40%, the ATZ concentration in pore water is 0.49 mg L−1 without remediation, which is greater than the injury critical value of soybean seedlings (0.1 mg L−1 in pore water), thus causing a serious injury. The model predicts that when the content of biochar in soil is 0.5%, the concentration of ATZ in pore water will drop to 0.076 mg L−1, which is lower than the critical concentration. The results of no symptoms confirmed the usability of the model.

The regulation of soil pollutant distribution via biochar is developing from theory to application. Previous studies have reported that ≥ 2% biochar is required to control the phytotoxicity of pollutants (Ji et al. 2022; Li et al. 2023b; Szmigielski et al. 2018), which may add up to an unimaginable cost. Herein, it was concluded that under the general ATZ residue in soil (0.5 mg kg−1 for maximum accumulation), a smaller proportion of biochar (0.1%) could control the phytotoxicity toward soybean seedlings (< 0.1 mg L−1 in pore water). The model proposed the optimal proportion of biochar to achieve the control target (predetermined pore water concentration) using known AZT content in soil. Therefore, the findings of this study can contribute to guidance regarding the economic use of biochar in agricultural production.

5 Conclusion

Based on two previous academic views: (1) concentration of organic pollutants in soil pore water is the determining factor of plant uptake and (2) biochar has an adsorption capacity for organic pollution in water, the potential application value of biochar in protecting seedling soybean from injury by removing ATZ from soil pore water was confirmed in this study. The critical concentration (NOEC) of ATZ-induced stress injury in soybean seedlings was 0.1 mg L−1 in soil pore water. This critical value provided the target and basis for the precise management of ATZ-induced stress in crops. Based on this, a study was conducted to purify ATZ in soil pore water using biochar. Compared with the adsorption experiment conducted in pure water, biochar in soil still exerts a strong adsorption capacity for residual ATZ in soil, although it is disturbed to a certain extent. The ATZ concentration in soil pore water was effectively reduced by a lower proportion of biochar–soil mixed system. The study regarding ATZ distribution in the soil pore water system revealed that ATZ distribution in pore water differs in the five soils with the same ATZ content; however, it is directly associated with the soil OMC. Combined with the effect of biochar content on ATZ distribution in soil, a Biochar Regulation of Organic Contaminant Soil Distribution Model was successfully developed and verified to be suitable for predicting ATZ concentration in soil pore water with or without biochar treatment. This model can contribute to environmental remediation applications to estimate biochar usage.

Although there is room for further optimization, the results of this study are sufficient for service agriculture production. It should be recognized that the situation in the actual farmland is complex. For example, soil moisture, soil temperature, soil fertility, air temperature and humidity may affect crop herbicide injury critical point and biochar adsorption. In regard to the model, more accurate and more universal is a topic worth studying in the future. In order to simplify the mathematical model to the maximum extent within a certain precision range (MAPE < 20%), we only selected a few pivotal parameters to study. However, it is undeniable that selecting more parameters and calculating more complex models are of great value in further improving the accuracy. This study, under the most common conditions, provides universal basic data regarding ATZ-induced stress damage in soybean seedlings alongside the application of biochar adsorption research. It is particularly useful to model the fate and distribution of pesticides in agricultural soils, which provides a solution for controlling herbicide residues. In addition, this study exhibits significant application potential for managing more nonpoint source pollutants in natural and engineered environments.