Introduction

In the rice production area of the Yangtze River Delta, soil heavy metal contamination is a serious issue, especially paddy soils (MEP & MLP 2014; Chen et al. 2015). Of the total area of contaminated lands, 64.8% is polluted by heavy metal (of which 46.7% is slightly polluted), with lands contaminated by Cd taking up the largest area. Compared to other toxic heavy metals, Cd is readily mobile metal and is toxic even at lower concentration (He et al. 2015; Rizwan et al. 2016; Ran et al. 2016). It is recognized that Cd could be the key factor behind unavailability of safe food on a sustainable basis (Ran et al. 2016). This causes a concern for rice cultivation as it is known that rice readily accumulates Cd. Therefore, major effort should be on the proper remediation of slightly polluted agricultural soils, especially paddy soils, aiming to break the bottle neck of remediation cost and technology so as to have the agricultural production capacity recover as soon as possible.

Heavy metal mining is one of the major reasons for soil heavy metal pollution, and acidic mine wastewater is a worldwide environmental problem causing a broad concern (Natarajan et al. 2006). Due to disorderly exploitation and unregulated mine wastewater emission, nearby water bodies used for irrigation have been contaminated. Consequently, a large area of downstream agricultural lands shows very high concentrations of heavy metals, causing a concern on agricultural product security and severely constraining local agricultural production. China is a major tungsten production and consumption country. High concentrations of Cd, Cu, Zn, and Pb were found in the agricultural soils close to the Dayu Tungsten Ore in Jiangxi, which is the largest reserve and the longest exploiting history in China; the Cd had the largest accumulation factor of up to 4.0 (Shao et al. 2013). Thus, there is an urgent need to employ remediation techniques to reduce Cd availability and uptake by crops in such area.

Various remediation techniques have been proposed for reduction in Cd contents and availability in polluted agricultural soils (Shaheen and Rinklebe 2015; Antoniadis et al. 2017). To employ low Cd-accumulating rice varieties in combination with amendments for Cd decontamination, as an efficient remediation technique, is mainly used for slightly contaminated agricultural soil especial dealing with large area and sustainable agricultural outputs (Yu et al. 2006; Qayyum et al. 2017). Biochar has extremely large surface area, rich carboxyl groups and minerals, and high pH and cation exchange capacity (CEC). Therefore, it can not only improve soil fertility and structure, but also effectively immobilize toxic elements, contributing to low plant accumulation and significant efficacy in Cd-contaminated paddy soil remediation (Beesley et al. 2011; Zhang et al. 2013; Bian et al. 2014; Yan et al. 2014; Li et al. 2016). Previous field studies by others showed that application of biochar (at 20 and 40 t ha−1) sharply reduced the bioavailability of heavy metals and the Cd concentration in rice grains to safe levels (Bian et al. 2013, 2014). Up till now, most studies on heavy metal immobilization were carried out via pot experiments, and pollutants were artificially added. Few field studies have been conducted to compare the effects of biochar on Cd uptake by different rice varieties, and the remediation potential of different types of biochars under different soil Cd contamination level. Therefore, in this study, slightly and moderately Cd-contaminated agricultural fields irrigated with tungsten mine wastewater were chosen and cultivated with super high yield hybrid rice and conventional late japonica rice. Biochars derived from two different raw materials were applied at different dosages to investigate their effects on soil Cd immobilization and Cd accumulation in rice grains.

Materials and methods

Site and soil

The field trail was conducted in the site (N29°56′, E119°13′) locating approximately 8 km downstream a tungsten ore in the northwest of Zhejiang Province, China. The ore was discovered in 1960s and used to be the largest tungsten ore in Zhejiang Province. The area has a subtropical monsoon climate and an altitude of 400 m a.s.l. Disorderly mine exploitation and unregulated emission of mine wastewater have led to heavy metal pollution of nearby water bodies and downstream agricultural soils. As a result, the safety of agricultural products has been a concern and local agricultural production has been restricted. In this study, two fields (Fields A and B) in a same irrigation system using river water were about 300 m apart from each other. Traditionally in rice–wheat (oilseed rape) rotation, the two fields were of the same landform and had close soil properties (Table 1) except for heavy metal concentrations (Table 2). Soils were Ferralic Acrisols.

Table 1 Soil properties of the two study fields
Full size table
Table 2 Soil heavy metal concentrations (mg/kg) of the two study fields
Full size table

Biochar for soil amendment

The two types of biochars, MC and KC, used in this study were derived from maize straw and hickory nut shell, respectively. For the production of MC, maize straws were pyrolyzed at 550 °C, for KC, hickory nut shells were pyrolyzed at 500 °C. The specific surface area of MC and KC was 57.68 and 93.46 m2 g−1, respectively. The SEM images in Fig. 1 show the microstructures of MC and KC.

Fig. 1
figure 1

SEM images of the maize straw-derived biochar (MC, left) and the hickory nut shell-derived biochar (KC, right)

Full size image

Experimental design

In this study, two rice varieties, high yield single cropping hybrid rice (Zhongzheyou#1) and conventional late japonica rice (Xiushui#9), were cultivated. In Field A, hybrid rice was planted and two biochar treatments were set up: 0(CK) and 15 t ha−1 (MC1). In Field B, five biochar treatments were set up: 0, 15 t ha−1 (MC1 and KC1), and 30 t ha−1 (MC2 and KC2), and each biochar treatment was further divided into two for cultivation of hybrid rice and japonica rice, respectively. The treatments were performed in triplicate, and the individual treatment plots were arranged in a complete randomized block design. Each plot of 2 m × 20 m in area was separated with surrounding protection rows 0.5 m in width and with separated irrigation inlets and drainage outlets. In Field A, there were three biochar treatments and total nine individual plots. In Field B, there were five biochar treatments and total 15 individual plot. The biochars were applied and mixed with the soils when the paddy soil were turn over in April before the rice transplanting in June 21. The plow depth is 20 cm. When the rice was harvested in October 24, 30 clusters of rice plants were randomly selected from each plot for productive ear counting and then the straws and grains of 3 clusters were oven-dried and weighed. Meanwhile, soil was sampled from each plot by collecting 3 randomly selected cores (0–10 cm deep), which were mixed to yield one composite sample per plot. Soil was air-dried and ground to pass through a 2-mm sieve for the detection of pH and through 150 µm for DTPA-Cd contents.

Chemical analysis

Soil pH was measured with a pH meter (FE28-Standard, Mettler Toledo). For Cd determination, soil samples were digested with HNO3-HF-H2O2, while brown rice and straw were digested with HNO3-H2O2, Soil DTPA-Cd was extracted using DTPA extraction method, and then Cd concentration of was measured with an ICP-MS (ICPMA8300, Perkin-Elmer).

Calculation of Cd bio-concentration factor (BCF)

A plant’s BCF of heavy mental is an important parameter for plant’s potential in remediating soil heavy metal contamination. The Cd BCF in rice straw or grain = Cd concentration in rice straw or grain/soil available Cd (i.e., DTPA-Cd) (Sun et al. 2009; Arao et al. 2010; Tang et al. 2015).

Statistics

Statistical analysis was performed using SAS 8.0 (SAS Institute, Cary, NC, USA); means were separated by LSD test at P < 0.05; and graphs were plotted using SigmaPlot 10 (Systat software Inc., San Jose, USA).

Results and analysis

The effects of biochar type and dosage on soil DTPA-Cd

Two-way ANOVA results showed that DTPA-Cd in the MC1 treatment of Field A was decreased by an average of 20.0% (from 0.44 ± 0.17 to 0.35 ± 0.07 mg kg−1) compared to CK (Fig. 2). In Field B, DTPA-Cd was reduced by 34.4–50.4% in the MC treatments while by 42.4–53.6% in the KC treatments (Fig. 3). Particularly, DTPA-Cd in the MC1 treatment of Field B was significantly reduced by 34.4% on average from 1.26 ± 0.12 to 0.82 ± 0.14 mg kg−1, a much higher percentage than that in MC1 of Field A. The larger dosage (30 t ha−1) reduced DTPA-Cd more than the less dosage. However, for the same dosage, there were no significant differences in soil DTPA-Cd between treatments of different biochar types. That is, there was no significant difference in DTPA-Cd between MC1 and KC1 or between MC2 and KC2. Biochar slightly increased the soil pH value, which ranged from 4.7 to 5.6.

Fig. 2
figure 2

Effects of biochar application on soil DTPA-Cd in Field A. MC1: maize straw-derived biochar (MC) applied at 15 t ha−1. Different letters indicate significant difference between treatments at P ≤ 0.05

Full size image
Fig. 3
figure 3

Effects of biochar type and dosage on soil DTPA-Cd in Field B. MC1: maize straw-derived biochar (MC) applied at 15 t ha−1; MC2: MC applied at 30 t ha−1; KC1: hickory nut shell-derived biochar (KC) applied at 15 t ha−1; KC2: MKC applied at 30 t ha−1. Different letters indicate significant difference between treatments at P ≤ 0.05

Full size image

Effects of biochar application on Cd accumulation of the two rice varieties

The results showed that the severer the soil contamination, the higher the Cd accumulation in the rice. The Cd concentration in the straws and grains of the hybrid rice was significantly higher than that of the late japonica (Table 3). For the hybrid rice, its grain Cd concentration was 3.57 ± 0.51 mg kg−1 and up to 8.06 ± 0.18 mg kg−1 in CK of Field A and B, respectively; and its straw Cd concentration in Field B was 2.2 times that in Field A. Biochar application significantly reduced grain and straw Cd concentration by 21.3 and 28.3%, respectively, for the hybrid rice in Field A, while 18.1 and 26.5%, respectively, in Field B. In Field B (Fig. 4), grain Cd concentration of the late japonica rice under CK was 3.23 ± 0.25 mg kg−1, 38.0% that of the hybrid rice, while straw Cd concentration of the former was 65.6% that of the latter. Through all treatments in Field B, the ratio of grain to straw in Cd concentration of the late japonica rice was 0.13–0.17, while that of the hybrid rice was 0.26–0.40. Results showed biochar can decreased the Cd concentrations both in the hybrid rice and late japonica rice. As for the late japonica rice, grain and straw Cd concentrations under treatments of MC2 and KC2 were lower than those of MC1 and KC1. As for the hybrid rice, grain and straw Cd concentration under treatments of KC2 was lower than that KC1, and they had no significant difference between treatments of MC1 and MC2. Above results showed that after biochar application, straw Cd concentration was decreased by a larger percentage than grain Cd concentration, which led to increase in the grain/straw ratio in Cd concentration or increase in the transfer coefficient (TC) of Cd from straw to grain.

Table 3 The effects of biochar application on Cd concentration in the brown rice grains and straws of the hybrid rice
Full size table
Fig. 4
figure 4

The effects of biochar type and dosage on Cd concentration of the brown rice grains and straws in Field B. MC1: maize straw-derived biochar (MC) applied at 15 t ha−1; MC2: MC applied at 30 t ha−1; KC1: hickory nut shell-derived biochar (KC) applied at 15 t ha−1; KC2: MKC applied at 30 t ha−1. Different letters indicate significant difference between treatments at P ≤ 0.05

Full size image

The effects of biochar application on rice growth and Cd BCF

The Cd BCF values of brown grain and straw of the hybrid rice were 7.75–9.25 and 25.50–38.55, respectively (Table 4). Values in Field A were significantly higher than those of Field B. Biochar application lowered Cd BCF values in Field A, but not significantly, compared to no biochar application (CK). In Field B, Cd BCF of the hybrid rice was higher than that of the late japonica rice, 140–270 and 20–90% higher for brown grains and straws, respectively (Fig. 5). ANOVA results (Table 5) showed that biochar application significantly raised Cd BCF of rice grain and straw; for Cd BCF increase, grain > straw and the hybrid rice > the japonica rice. The Cd BCF value was higher in KC1 than in KC2, while the opposite was true for MC.

Table 4 Effects of biochar application and Cd pollution level on the Cd BCF of grain and straw of the hybrid rice
Full size table
Fig. 5
figure 5

Cd bio-concentration factor (BCF) of brown rice grain and straw in Field B. MC1: maize straw-derived biochar (MC) applied at 15 t ha−1; MC2: MC applied at 30 t ha−1; KC1: hickory nut shell-derived biochar (KC) applied at 15 t ha−1; KC2: KC applied at 30 t ha−1. Different letters indicate significant difference between treatments at P ≤ 0.05

Full size image
Table 5 Two-way ANOVA on the effects of biochar type and dosage on Cd bio-concentration factor (BCF) of rice grain and straw
Full size table

The influence of biochar on rice yield index, such as productive ear count and rice per spike, is not significant both in field A and field B. Yield composition analysis showed that there were no significant differences in productive ear count of the hybrid rice between the treatments with it being 15.9 ± 1.2 and 15.9 ± 0.8 in Field A and B, respectively. In Field A, the unhusked rice biomass the hybrid rice took up 1.49 ± 0.53 and 1.88 ± 0.64 of the straw biomass (aboveground) in CK and MC1, respectively. In Field B, the unhusked rice biomass of the hybrid rice took up 0.38–0.87 of the total straw biomass under various treatments. Generally, the hybrid rice grew obviously better in Field A than in Field B, but biochar application did not make significant differences in the ratio of unhusked rice biomass to aboveground biomass. In Field B (Table 6), the late japonica rice weighed 2.2–3.2 g per spike, higher than the hybrid rice. But with a count of 12.8–15.9 per cluster, the former had fewer productive ears per cluster than the latter. Assuming a rice plant density of 240,000 clusters per ha, theoretical rice yield in the different treatments ranged 5258 to 10,245 kg ha−1, close to the yield level in the local region. Biochar application did not exert a significant effect on rice yield, due to large soil heterogeneity in this field experiment.

Table 6 Rice yield composition in Field B
Full size table

Discussion

Immobilization of soil Cd by biochars in paddy fields

In this study, 7 months after biochar application, soil DTPA-Cd decreased by over 20%, and plant Cd concentration also decreased significantly. The severer the soil pollution was, by a larger percentage soil DTPA-Cd and rice grain Cd concentration were decreased, indicating that after a crop growing season, heavy metal adsorption by the biochars was not yet to reach equilibrium and heavy metal immobilization would continue. It is pointed out by previous studies that the heavy metal immobilization ability of biochar is mainly attributed to its surface functional groups, which interact with heavy metal via chelation, precipitation, redox, electrostatic attraction, etc.(Beesley et al. 2011). Short-term adsorption experiments demonstrated that Cd2+ adsorption by an MC can be well described by a Langmuir equation, and the adsorption capacity was 23.52–23.59 mg/g at 25–35 °C (Kołodyńska et al. 2012; Xu et al. 2014). Findings of 1–3-year field experiments on agricultural soil remediation showed that biochar has a long-term immobilizing effect on Cd, Pb, and Cu and improves the uptake of Si, P, and K by rice and wheat (Bian et al. 2014; Li et al. 2016; Abbas et al. 2017). The alternate wetting and drying water management and fertilization in paddy fields lead to continuous increase in the surface organic and inorganic layers of biochar, which could be one of the major reasons for biochar’s long-term adsorption effect. Another reason might be that the reaction of the iron oxides on the surface and in the nano-pores of heavy metal-loaded biochar would increase the surface functional group density of biochar. Consequently, heavy metal nano-particles are continuously adsorbed and react with the surface inorganic and organic compounds.

Though biochar application slightly raised soil pH, soil pH was not significantly related to soil DTPA-Cd due to high heterogeneity in soil heavy metal concentration in this study. This implies that soil pH increase was not the major reason for heavy metal immobilization by biochar. It was in line with the result from Rinklebe et al.(2016) which found that biochar addition for 1 year into a flooding soil has little effects on soil EH/pH and while decreased concentrations of dissolved heavy metals under dynamic redox conditions. Biochar have abundant binding sites and a large surface area and thus can reduce cadmium mobilization, while El-Naggar et al. (2018) found short-term addition of biochar into aerobic soil, exhibited a wider range of EH and a lower pH than the non-treated soil, biochar addition had increased the dissolved concentrations of Cd, which might be due to the lower pH values under oxic conditions. In addition, MC and KC biochars differed in the function of soil Cd immobilization. Soil DTPA-Cd was reduced by a larger percentage by KC than by MC. It is attributed to the more abundant pore structure and binding sites in KC than MC and thus had higher heavy metal immobilization capacity. Li et al. (2016) found that corn straw-derived biochar (CB) had more O-bearing groups than hardwood-derived biochar (HB) and in turn more adsorbing sites for Cd and Cu. Lignin content is an important factor influencing the oxidation degree of biochars and in turn the sustainability of heavy metal immobilization. In this study, it was found that soil DTPA-Cd was lower with KC application than with MC application, which might be due to the high porosity and high lignin content in KC.

Grain Cd accumulation characteristics and responses to biochar application of the two rice varieties

The results of this study showed that grain Cd concentration of the hybrid rice in Field B was twice that in Field A. Meanwhile, for the same biochar treatment, grain Cd concentration was reduced by a larger percentage in Field B than in Field A, which was directly related to the larger decrease in soil DTPA-Cd in Field B. It is believed that biochar reduces Cd accumulation in rice grains by reducing Cd mobility. Bian et al. (2013) found grain Cd concentration of both super rice and conventional rice decreased significantly to 0.4 mg kg−1 with 40 t ha−1 biochar in Cd-contaminated soil and the brown grain Cd concentration was positively and significantly related to soil exchangeable Cd. Biochar addition improve the rice tiller number in this study. It improved rice growth in the slightly polluted field, while in the moderately polluted field, its effect was not so prominent. In rhizosphere, significant soil–rice plant interaction occurs, explaining plant behavior in remediated contaminated soils. In biochar remediation soil, Cd photoavailability and accumulation in rice plants were altered by changed microflora and root exudates in rhizosphere with biochar amendments, and strategies were employed for Cd uptake and translocation of different rice varieties (Antoniadis et al. 2017). Long-term biochar addition can improve the microliving environment, such as increasing soil pH, moisture, active nutrients and the C/N ratio, and indirectly strong influence on microbe compositions, especially fungals in paddy soil, as well as increased root growth and raised plant biomass (Dai et al. 2016; Zheng et al. 2016).

Liu et al. (2016) found that the BCF of Cd, Cu, Pb, and Zn of the underground part and aboveground part of lettuce (Lactuca sativa L. var. capitata L.) tended to decrease with increasing biochar application dosage. However, the transfer coefficient of Cd, Cu, and Zn in the aboveground part tended to increase, demonstrating that biochar can significantly lower heavy metal accumulation in plant tissues but increase heavy metal upward transfer. It was also found in this work that the higher the biochar application dosage, the higher the grain Cd/straw Cd ratio was, and the more Cd was transferred from straw to grain. Chen et al. (2016) found that biochar works better in lowering Cd uptake by low Cd-accumulating rice varieties than by high Cd-accumulating rice varieties. The results from this work shows that biochar can significantly raise the Cd BCF of the straws and grains for the hybrid rice, but not significantly for the conventional late japonica rice. Additionally, Cd BCF responded differently to the two types of biochars. Compared to in KC1, soil DTPA-Cd was lower, less Cd was accumulated by rice plants, and Cd BCF was lower in KC2. Similar to KC, soil DTPA-Cd was lower in MC2 than in MC1. However, Cd accumulation by rice plants was not significantly reduced with higher MC application dosage. This indicates that KC has a higher potential than MC in lowering Cd bioavailability.

Grain Cd concentration of different rice varieties can differ by several times due to their different strategies employed for Cd uptake and translocation (Zhou et al. 2013; Zhang et al. 2015; Cai et al. 2016). Cd accumulation in rice plants mainly differ in its accumulation and transfer in stems and leaves (Liu et al. 2007; Uraguchi et al. 2009; Yang et al. 2015; Cai et al. 2016), as they are one of the paths for Cd to reach rice grains and also an important regulating factor. Many protein families involving in metal transfer and playing important roles in Cd transfer, such as gene OsHMA2 was found to involve in the transfer of metals from the underground parts to the aboveground parts, facilitating the loading of Cd and Zn in xylem (Takahashi et al. 2012; Yamaji et al. 2013). It is also suggested the accumulation of Cd in rice roots should be attributed to the sequestration of intense Fe plaque in rhizosphere of paddy soil (Yu et al. 2016; Li et al. 2017). The results from this work show that whether for the hybrid rice or the conventional rice, Cd BCF of straw was higher than that of grain and Cd BCF in Field A was higher than in Field B. In addition, Cd BCF of rice straw was more sensitive to rice variety and pollution level than that of grain, confirming that straw Cd accumulation is an important factor regulating grain Cd accumulation. Grain Cd accumulation in hybrid rice was more sensitive to soil Cd contamination than that in late japonica rice, and Cd BCF in the straws and grains of the hybrid rice was higher than that of the japonica rice. In terms of the ratio of grain Cd concentration to straw Cd concentration, the hybrid rice was also twice the japonica rice, indicating a stronger Cd-transferring capacity of the former. This may be a major reason for high Cd accumulation in the high yield hybrid rice. Therefore, it is suggested that caution should be taken when choosing high yield single cropping hybrid rice for Cd-contaminated fields.

Conclusions

Results of the field experiment showed biochar had significantly decreased the soil Cd mobilization and reduced Cd uptake by rice and accumulation in rice grains. More biochar inputs, less soil DAPT-Cd concentration. With application of larger dosage, hickory nut shell-derived biochar led to larger reduction both of plant Cd accumulation and soil DAPT-Cd concentration than maize straw-derived biochar. The hybrid rice variety commonly grown by local people accumulated much more Cd in its grains than the conventional late japonica rice, so it is not suggested to be grown in area with Cd-contaminated soil or high environmental original value of cadmium.