2.1 Soil characterization
A greenhouse study was performed in the city of Lavras, Minas Gerais, Brazil. The surface layer (0 – 0.2m) of an Haplic Gleisol was collected and used for rice cultivation in pots (7 dm³). Samples were air dried and physically and chemically characterized according to Teixeira et al. (2017) and PTE’s were analyzed following the methodology preconized by U.S. Environmental Protection Agency (USEPA) 3051A (USEPA 2007) (Table 1).
2.2 Steel mill wastes characterization
The steel mill wastes chosen to be used in this study were: Metallurgical press residue (FPM), Filter press mud (FPM) and Phosphate mud (PM), collected in the city of Juiz de Fora, Minas Gerais, Brazil, where a metallurgical industry is installed. The available contents of macro and micronutrients in these wastes were analyzed following Mehlich-1 extraction and they were also analyzed regarding PTE’s following USEPA 3051A (USEPA 2007) (Table 2). The elements analyzed were: P, S, Mg, K, Ca, Cr, Mn, Fe, Ni, Cu, Zn, Cd and Pb, all of them determined by spectrophotometry of atomic absorption with air-acetylene flame, except for phosphorus that was determined by colorimetry and K that was analyzed by flame photometry (Malavolta et al. 1989).
2.3 Experiment conduction
The experiment was set in a completely randomized design with a factorial scheme (3×7) in triplicates, consisting of three steel mill wastes combined at seven wastes doses (0; 0.5; 1; 2; 4; 8; 16 t ha-1) which corresponds to the following doses per pot: 0; 1; 2; 4; 8; 16 and 32 g pot-1.
Soil samples and metallurgical wastes were weighed, mixed homogeneously in plastic bags and incubated for 60 days. Then, soil samples (7 dm³) were fertilized with N, K, P, Mg, S, B, Cu, Zn and Mo (450; 450; 200; 30; 50; 0.5; 1.5; 5 and 0.2 mg kg-1 respectively) following recommendations of Malavolta (1981) and using analytical grade reagents. Iron and Mn were not added due to the natural content of them in soil and their low demand by plants. Micronutrients and S were added at once during cultivation while N and K were split out in topdressing applications. The first nutritional supplementation occurred 32 days after installing the crop. From then on, N and K were supplied weekly. Cultivation was performed in pots with 7 dm3 capacity.
Rice crop (Oryza sativa L.) cultivar Curinga was chosen due to the interest in studying PTE’s behavior under flooding and because rice is an important staple food all over the world (Laborte et al. 2012). Twelve seeds were cultivated per pot but only five seedlings were carried on until the end of five months. Most of rice crops are conducted in floodplain soils, and in order to approach such condition, a 5 cm water blade was maintained until the end of the experiment.
2.4 PTE’s in rice parts
Plant material was cleansed with distilled water in order to remove any soil particle and avoiding contamination, especially roots. After drying the plant material in an oven with forced air circulation at 65 ºC up to constant mass, the shoots, husk, grain and roots dry matter of rice was determined. Then, the material was ground in a Wiley mill and digested with nitroperchloric method in a digester block. About 1 g of plant material was added to each tube in the digester, together with 6 ml of HClO4 + HNO3 solution in a 1:2 ratio. Samples were allowed to pre-digestion for approximately 4 hours and then, they were gradually heated until temperature reached 190ºC. When samples became colorless and a 2 mL aliquot was left, digestion was completed. After cooling down, the extracts were diluted with distilled water to 16 mL and filtered through Whatman 40 filter papers (Scott 1978). PTE’s concentration (Mn, Ni, Cu, Zn, Cd and Pb) was then determined in an atomic absorption spectrophotometer (AAS).
2.5 Bioaccumulation Factor (BAF) and Translocation Factor (TF)
PTE’s concentrations (Ni, Cd, Pb, Cu, Zn and Mn) in the plant and in the soil were calculated based on dry weight. The bioaccumulation factor (BAF) is an index that shows the plant's ability to accumulate an element according to the concentration of that element in the soil. Thus, BAF was calculated according to the following equations (Galal and Shehatab 2015; Usman et al. 2013; Usman and Mohamed 2009):
BAF (grain) = Mgrain / Msoil (1)
BAF (husk) = Mhusk / Msoil (2)
BAF (shoot) = Mshoot / Msoil (3)
BAF (roots) = Mroot / MCsoil (4)
Where Mgrain, Mhusk, Mshoot and Mroot are the concentrations of Ni, Cd, Pb, Cu, Zn and Mn in the grains, in the husk, in shoot and root, respectively, and Msoil is the concentration in the soil.
The translocation factor (TF) evaluates the relative displacement of Ni, Cd, Pb, Cu, Zn and Mn from plants’ roots towards other plants’ parts such as shoots, grains and husk. The TF is calculated using the following equations (Azzia et al. 2017):
TF (grain) = Mgrain / Mroot (5)
TF (husk) = Mhusk / Mroot (6)
TF (shoot) = Mshoot / Mroot (7)
2.6 Health Risk Assessment
2.6.1. Health Risk Index (HRI)
Health risk index is calculated to verify exposure risks of PTE’s through food intake. It reveals the risk of consuming contaminated foodstuffs. HRI is the ratio between the exposure and the reference oral dose (RfD) (USEPA, US Environmental Protection Agency 2002). It is generally used in EPA non-cancer health assessments. The RfD values used in this study were 0.02, 0.04, 0.004, 0.001, 0.03, 0.30 mg kg-1 for Ni, Cu, Pb, Cd, Mn and Zn, respectively. (USEPA, 2010; WHO, 1993; Food and Nutrition Board, 2004). Therefore, the health risk index was calculated as in the following equation:
HRI = (DI) × (Cmetal) / RfD × B (8)
Where DI is the daily rice intake rate (kg per day), Cmetal is the metal concentration in the grain (mg kg-1), RfD is the oral reference dose for the metal (mg kg -1 per weight per day) and B is the human body mass (kg). The DI used was 34 kg year-1 for adults and for children (0 to 6 years) the consumption was established as 1/3 of the rate for adults (IBGE 2004, 2006; WHO 1993). The average B value was 70 kg for adults and 19.5 kg for children (IBGE 2004, 2006; Guerra et al. 2012; WHO 1993).
HRI greater than 1 for any metal in rice grains is considered unsafe for human health and serious health risks are possible. However, HRI < 1 means that oral exposure to PTE’s is considered safe. (Storelli 2008).
2.6.2 Hazard Index (HI)
The potential health risk from exposure of several metals in rice grains (HI) was calculated according to the EPA guides for health risk assessment (Ogunkunle et al. 2016) using the following equation:
HI = ΣHRI= HRICd + HRINi + HRIPb + HRICu + HRIMn + HRIZn (9)
For the Health Risk Index (HRI) and Harzard Index (HI) calculations, the three metallurgical wastes (Metallurgical press residue; FPM: Filter press mud; PM: Phosphate mud) were considered at 8 t ha-1 dose.
2.7 Quality Control and Quality Assurance
The qualitative detection limit for each analytical method (MDL) was calculated after determining Ni, Cd, Zn, Pb, Mn, Cu in seven blank samples and using the equation below (APHA 2012): MDL = (x + t × s)×d, whereas, x is the average content of the substance in seven blank samples, t is the Student value at 0.01 probability and n-1 degrees of freedom (for n+7, α = 0.01 and t = 3,14), s is the standard deviation of these seven samples, d is the diluton eventually applied. The MDL for Zn, Cu, Cd, Mn, Ni and Pb were 2.9, 2.35, 7.7, 11.6, 1.36 e 0.72 µg kg-1, respectively. All glassware used was rinsed several times with 10% HNO3 in order to avoid contamination. An internal sample was used to ensure quality control, and determination was only run if recovery was higher than 90%.
2.8 Statistical Analysis
Statistical analysis was performed by conducting a variance analysis (F test, P <0.05). The effect of steel-mill wastes application and their doses on the soil regarding the contents of Zn, Cu, Cd, Mn, Ni and Pb in husks, grains, shoots and roots were evaluated with the Tukey test (P <0.05).
For bioaccumulation factor (BAF) and translocation factor (TF), only the effect of steel-mill wastes was evaluated through the mean of all doses (n = 18) by applying the Tukey test (P <0.05).