Recent advances in phytoremediation of arsenic-contaminated soils

Arsenic contamination in soils occurs widely in a range of ecosystems resulting from geological origins and anthropogenic activities. On average, arsenic concentration ranges from 5 to 10 mg kg−1 in uncontaminated soils and above 10 mg kg−1 in contaminated soils (Hossain, 2006). Increased buildup of arsenic in irrigated soils has been widely recognized in South and South-east Asia (Brammer and Ravenscroft, 2009), posing significant threats to agriculture sustainability. In Bangladesh, long-term irrigation with arsenic-rich groundwater from shallow aquifers in dry season adds >1000 tons of arsenic to the agricultural soils (Ali et al., 2003). In addition, arsenic contamination in soils results from various anthropogenic activities, such as mining and smelting (Williams et al., 2009), and using arsenic-containing wood preservatives (Chirenje et al., 2003), pigment, pesticides, herbicide (Sarkar et al., 2005) and feed additives (Arai et al., 2003). As a cost-effective and ecology-friendly technology, phytoremediation of arsenic-contaminated soils has been widely studied. Among phytoremediation technologies, phytoextraction and phytostabilization are two predominant approaches in remediation of soils contaminated with heavy metals. Phytoextraction takes advantage of plants to remove contaminants from soils by concentrating the targeted contaminant to the harvestable tissues (Salt et al., 1998). To achieve effective arsenic removal from soils, the plant should be highly tolerant to arsenic and efficient in accumulating arsenic into sufficient aboveground biomass. Therefore, phytoextraction efficiency depends on both aboveground biomass yield and plant arsenic concentration. Bioconcentration factor (BF), which is defined as the ratio of element concentration in plant shoots to that in soil, has been used to measure a plant’s efficiency in phytoextraction. Based on mass balance calculation, phytoextraction is feasible only by using plants with BF much greater than 1, regardless of how large the harvestable biomass (McGrath and Zhao, 2003). Furthermore, to achieve efficient removal of contaminant in a reasonable time frame with high plant survival and biomass yield, the initial and target soil contaminant concentrations should be taken into account to predict the applicability of phytoextraction, which is in most cases appropriate for soils with low contamination (Zhao and McGrath, 2009). For heavily contaminated sites (e.g., industrial and mining degraded sites), indigenous tolerant species with extensive root system and low translocation factor (TF, the ratio of contaminant concentration in shoots to that in roots) provide valuable plant resources to immobilize the pollutant in the rhizosphere, and simultaneously stabilize the degraded sites by establishing vegetation cover. Soil amendments, in some cases, are essential to assist the success of the survival of pioneering species by mitigating contaminant toxicity and improving substrate conditions (Vangronsveld et al., 2009). In this way, ecological restoration of contaminated sites can be gradually achieved through revegetation, which is termed as phytostabilization. Beside these two major phytoremediation techniques, other methods include phytoexclusion and rhizofiltration. To remediate large-scale agricultural soils contaminated by arsenic, phytoexclusion is more practical to reduce arsenic transfer from soil to crops. Based on the well-established knowledge with regard to arsenic biogeochemistry and arsenic transport mechanisms in rice, a range of strategies including water management, Si fertilization, and rhizosphere manipulation