2.1 Physicochemical properties of sepiolite, goat manure and soil
Soil specimen with natural Cd was collected from the 0 to 20 cm layer in Chenzhou, Hunan, China (112º43'48"N, 25º43'48"E). The basic characteristics of soil are showed in Table 1. Natural sepiolite was purchased from Hebei Huakai Environmental Protection Material Co. Ltd. The goat manure (GM) was purchased from a local fertilizer market in Tianjin, China. Selected characteristics of goat manure were listed in Table 2.
2.2 Experimental setup and water management
Pot experiment in the first year.
Pot experiments were performed in a greenhouse in the Agro-Environmental Protection Institute, Tianjin, China. The soil samples were air-dried, homogenized and passed through a 4mm sieve. Soil (6 kg) was placed in each pot (23.3 cm in diameter, 34 cm in height). Basal fertilizer, containing 3.87 g urea and 6.24 g monopotassium phosphate, was added to soil and mixed thoroughly (Xiao et al., 2015). SP and GM were mixed in to each pot as shown in Table 3. Then, soils were equilibrated for one month, maintaining approximately 75% of field water-holding capacity with tap water (without Cd).
Seedlings of rice (Oryza sativa L.) were germinated on perlite and transplanted on June 11, 2019. The rice plants were grown under uniform flooded conditions for 21 d before the water treatment. There were three water management conditions, involving continuous flooded (CF, maintained 2–3 cm water layer in pot throughout the growth period), intermittent water condition (IW, repeated every 7-d with 3-d flooding followed by 4-d drainage), and aerobic condition (AO, soil maintained moisture during the growth period). Drainage from the pots was prevented during the experiments, and the pots were arranged in random blocks and rotated intermittently to ensure the same conditions for each pot.
Pot experiment in the second year.
In order to test and verify the long-term stability and effectiveness of sepiolite, the pot experiment with the same design was conducted in the second year. The pot have applied goat manure in the first year were added with the same dosage of GM to replenish soil fertility.
2.3 Sample preparation and analysis
After 135 d of growth, the plants were harvested and separated into root, straw, rachis, husk, and rice grains. The rice grains were oven-dried at 75 °C and weighed immediately after removing the husk and unfilled grains. Each part of the plant were ground and passed a 1.85-mm sieve for further analysis. Cd contents in plant samples were digested using a block digester and soil solutions was determined by inductively coupled plasma mass spectrometry (iCAP Q; Thermo Scientific, Waltham, MA, USA). (Yin et al., 2017).
Rhizosphere soil was collected from a small amount of soil shaken directly from the plant roots. Soil samples were air-dried at room temperature and sieved to <1 mm, and stored in a plastic container for further analysis.
Soil pH was measured with a pH meter (PB-10, Sartorius, Germany) at the solid-to-liquid ratio (m:v) of 1:2.5(Zhou et al. 2018a).
DOM content was measured using a total organic carbon analyzer (Vario TOC, Elementar Germany element) at a solid-to-water ratio of 1:5(Gao et al. 2018).
The bioavailability of Cd in soil was determined by diethylenetriaminepentaacetic acid (DTPA) at a solid-to-water ratio of 1:5 (Liang et al. 2016; Wang et al. 2016).
Changes in the fractions of Cd in paddy soil were analyzed by sequential extraction (Tessier et al., 1979), which separates heavy metals into five different fractions: extractable with 1.0 M MgCl2 (pH=7) (exchangeable, EXC-Cd), extractable with 1.0 M NaOAc (pH=5) (carbonate-bound, CB-Cd), extractable with 0.04 M NH2OH·HCl in 25% (vol.%) HOAc solution (associated with Fe–Mn oxides, OX-Cd), extractable with 0.02 M HNO3 in 30% (wt.%) H2O2 (complexed with organic matter, OM-Cd), and HNO3–HF–HClO4 digested (residual, Res-Cd).
2.4 Quality control
To monitor the accuracy and quality of chemical analyses, quality control measures were adopted using soil (GBW(E)-070009) and plant reference materials (GBW-10045 (GSB-23)) obtained from the Institute of Geophysical and Geochemical Exploration (IGGE, China). Three reagent blanks and three standard reference materials were included in each batch of extraction, preparation, and analysis of every 40 samples.. The recovery rate were 95–103% and 92–100% for Cd of plants and soil, respectively.
2.5 Statistical analyses
All treatments were replicated three times. The means and standard deviations (SD) were calculated using Microsoft Office Excel 2003. Two way analysis of variance was carried out with SAS 9.2 (SAS Inc., NC, USA). The mean values of the experimental parameters were compared using Student’s t-test at p≤0.05 (significant).