Article pubs.acs.org/est
Sparingly-Soluble Phosphate Rock Induced Significant Plant Growth and Arsenic Uptake by Pteris vittata from Three Contaminated Soils Jason T. Lessl† and Lena Q. Ma‡,†,* †
Soil and Water Science Department, University of Florida, Gainesville, Florida 32611, United States State Key Lab of Pollution Control and Resource Reuse, School of the Environment, Nanjing University, Jiangsu 210046, China;
‡
S Supporting Information *
ABSTRACT: We evaluated the ability of As-hyperaccumulator Pteris vittata (PV) to remove As from As-contaminated soils over five harvests in 2.5 years in raised beds (162 kg soil/bed). We tested the hypothesis that a P-limiting environment would enhance PV growth and As uptake owing its unique ability to uptake P under As-rich environment. In Dec. 2009, PV was transplanted to three As-contaminated soils (pH of 5.5−7.2) containing 25−129 mg kg−1 As, which was amended with sparingly-soluble phosphate rock (PR-soil) or soluble P fertilizer (P-soil). During the 2.5-year, PV obtained sufficient P (1882 vs 2225 mg kg−1) from PR-soils, with increased root biomass (33%) and root exudation (53%) compared to P-soils. In addition, its frond biomass increased by 20% consecutively with each harvest (six month interval) from 18 to 36 g plant−1. Its frond biomass in PR-soils (52.2 g plant−1 year−1 or ∼12 mt ha−1 year−1) averaged 39% more than that in P-soils. To our knowledge, this represented the largest PV frond biomass reported, demonstrating the unique ability of PV in using insoluble P from PR in alkaline soils. In addition to biomass increase, PV from PR-soils had ∼1.5 times more As in fronds (2540, 780, and 920 mg kg−1) than those from P-soils (1740, 570, and 400 mg kg−1), with soils containing 129, 25, and 30 mg kg−1 As, respectively. The low available P in PR-soils induced substantial plant growth and As uptake by PV. This translated into significantly more As removal from soil, averaging 48% reduction in PR-soils and 36% in P-soils in 2.5 years. With multiple harvests and PR amendments, our results showed As removal by PV from contaminated soils was ∼7 times faster than published studies.
■
INTRODUCTION
protocols exist for residential and public spaces, suggesting the presence of a potential risk to public health.6 Many engineering technologies have been developed to remediate As−contaminated soils, but they are costly and invasive. Remediation methods either disturb the environment (excavation) or do not remove the As (stabilization), allowing potential for future exposure. Alternatively, the use of phytoremediation preserves the topsoil while removing hazardous contaminants. This technique requires no special equipment or high operating costs and can be aesthetically pleasing, garnering more public acceptance. Arsenic hyperaccumulator Pteris vittata L. (PV; Chinese brake fern) can accumulate up to 22 630 mg As kg−1 in the above ground biomass (fronds), indicating its capacity for high tolerance and accumulation of As.7 Numerous studies have demonstrated its ability for substantial As uptake, but they are mostly based on short growing periods (∼12 weeks) under conditions unrepresentative of soil environments (i.e., hydroponic or small pots in a
Due to its toxicity and carcinogenicity, arsenic is ranked by the Agency for Toxic Substances & Disease Registry as the No. 1 contaminant in the environment.1 From 1930 to 1980, application of arsenical pesticides added ∼1300 mt yr−1 As to soils.2 The U.S. Environmental Protection Agency (USEPA) regional screening level for soil As is 0.39 mg kg−1 for residential use and 1.6 mg kg−1 for industrial use while the Florida Soil Cleanup Target Level (SCTL) is 2.1 mg kg−1 under residential use and 12.0 mg kg−1 under industrial use.3 Despite being a known issue, As accumulation in soils is a continual problem. Natural leaching and inappropriate disposal of As-treated wood pose a threat to public health and the environment. By 2002, > 90% of outdoor wood structures in the U.S. were treated with copper chromate arsenate (CCA) pesticide.4 With high concentrations of As (∼1200 mg kg−1), CCA-treated wood has a long life-span (20−50 years) and acts as a long-term source of As contamination in the vicinity.5 Even though CCA wood was banned for residential use in 2004, ∼6.1 × 103 mt of As is used annually for wood treatments in the U.S.2 Normally, soil As concentrations exceeding the limit results in regulatory actions at waste sites, but no such © 2013 American Chemical Society
Received: Revised: Accepted: Published: 5311
January 2, 2013 April 12, 2013 April 22, 2013 April 22, 2013 dx.doi.org/10.1021/es400892a | Environ. Sci. Technol. 2013, 47, 5311−5318
Environmental Science & Technology
■
greenhouse).8−12 To remediate slightly to moderately contaminated soils, phytoremediation can take years to decades to achieve. Thus, to evaluate the potential of PV for phytoremediation, a long-term study to elucidate conditions conducive to maximizing As uptake and biomass production is needed. Arsenate, the most prevalent form of As in soils, is taken up by higher plants including PV via phosphate transporters.13 Due to their shared homology, phosphate and arsenate act as competitive inhibitors, which can exacerbate P-deficiency when roots assimilate arsenate.14 During P starvation, many plants respond by increasing root length and density, and by exuding organic compounds to mobilize insoluble P in soils.15 P. vittata is native to soils, which are characterized by low available P and/or high concentrations of As.7,16 Not only does PV tolerate low P availability, the presence of As has been shown to improve its biomass.17 These properties make PV unique in its ability to scavenge for P, especially in the presence of As. Application of low-available P to soils can improve plant As uptake, but it may adversely impact plant growth. Phosphate rock (PR), which is the raw material used to manufacture phosphate fertilizers, could provide a source of P with limited plant availability on a long-term basis. However, PR is typically unsuitable as a direct P fertilizer because its rate of dissolution is inadequate to meet plants demands. In a study comparing PR and (NH4)2HPO4 amendments in 16 Brassica species, plant biomass in PR treatments was reduced >2.5 times and plant P concentration dropped ∼1.5 times in all cultivars.18 In a different study using white clover (Trifolium repens) and ryegrass (Lolium perenne), dry matter yields from PR treatment at pH 5.3 were equivalent to CaHPO4 treatment, but decreased by 24−28% at pH 5.6−6.4.19 These results suggest that soil acidity is a key factor dictating PR dissolution by plants.20 However, certain plants have the capacity to enhance P solubilization. In a study with white lupin (Lupinus albus), the limited P availability from PR stimulated root growth and exudation allowing for increased root-induced dissolution of PR, even at a neutral soil pH, although plant biomass was adversely impacted.21 In PV, low soluble P conditions have been shown to increase As uptake as well as promote root growth in greenhouse experiments.22 Furthermore, the presence of As is known to increase PV biomass in a hydroponic experiment.23 These unique responses could be used to improve long-term phytoremediation using PV. In addition to maintaining low available P, appropriate clipping techniques are needed to maximize plant regrowth. In a 16−month pot-study by Gonzaga et al.,9 clipping of PV fronds at the rhizome base hindered its regrowth, leading to biomass declines of 74 and 40% in the second and third harvests. We hypothesized that PV can effectively remove As on a long-term basis, over multiple harvests by maintaining low− available P to increase its plant growth and As uptake. The field experiment was conducted over 2.5 years and 5 harvests using three As-contaminated soils with PR as a low−available P source. The objectives were to (1) evaluate the ability of PV in efficient P uptake under PR amendment, (2) determine the effect of PR amendment on enhancing plant growth and As uptake by PV, and (3) demonstrate the effectiveness of proper husbandry practices to improve PV regrowth over multiple harvests.
Article
MATERIALS AND METHODS
Soil Collection and Characterization. Three soils were collected from As−contaminated sites in central Florida. Two soils (Arenic Albaqualfs) were from abandoned cattle dipping vats (DVA and DVB), which were contaminated with an arsenical tickicide. A third soil (Grossarenic Paleudult) was collected from an abandoned wood treatment facility, contaminated with copper chromate arsenate (CCA) where PV was originally discovered to be an As hyperaccumulator.7 Soils were air-dried, sieved through a 2 mm mesh screen and analyzed for pH (1:2 soil to water), organic matter content (Walkley-Black method), cation exchange capacity (ammonium acetate method) and particle size (pipet method).24 Soil samples were subjected to HNO3/H2O2 digestion (USEPA Method 3051) on a hot block (Environmental Express, Ventura, CA). The supernatants were filtered (0.45 μm) and analyzed for total As concentration using graphite furnace atomic absorption spectroscopy (GFAAS, Perkin-Elmer SIMMA 6000, Perkin-Elmer Corp., Norwalk, CT). Total P was measured spectrophotometrically (UVI1800U, Shimadzu Corp., Columbia, MD) at 880 nm using the modified molybdenum-blue reaction. Due to arsenate (AsV) interference with the molybdenum reaction, samples were first incubated with cysteine at 80 °C for 5 min to reduce arsenate to arsenite (AsIII).25 In addition to total As and P, we have also determined plant available As and P using 0.05 M (NH4)2SO4 solution (1:25 soil to solution ratio for 4 h) following the protocols of Wenzel et al.26 Plant available Fe, Al, Ca, and Mg in soils were extracted using 0.2 M ammonium oxalate (1:100 soil to solution ratio for 1 h) and Mehlich-3 (1:10 soil to solution ratio for 5 min) and determined by inductively coupled plasma mass spectrometry (ICP-MS Perkin-Elmer Corp., Norwalk, CT). Besides phytoavailability, we have determined the impact of continued plant As removal on As bioavailability to humans in the soils using simplified bioaccessibility extraction test (SBET).26 Experimental Setup. Twenty-four raised beds were constructed (0.36 m2 to a 35 cm depth) and filled with CCA, DVA, and DVB soil with PR (PR-soil; 15 g kg−1) and without (P-soil) (four beds soil−1 treatment−1). The PR used was a coarsely ground (0.5−2.0 mm) material containing 8% P and 24% Ca [Ca10(PO4)6F2 (CaCO3)x; PCS Phosphate, White Springs, Florida]. The volume of soil in each bed was 108 L weighing ∼162 kg (dw). Containers (12 L volume) were also filled with each soil (18 kg container−1) for growth of PV without P amendments (four replicates soil−1). Plantless soil containers were also setup with PR-soils being setup to monitor non-PV mediated As loss. Hydrated lime was applied (40 g bed−1, 4.4 g container−1) to the surface of DVA and DVB soils to raise the pH to ∼6. The beds and containers were watered to field capacity, which was maintained for two weeks and had a final soil depth of 30 cm. In December 2009, three month old PV (3−4 fronds ∼15 cm in length) from Milestone Agriculture Inc. (Apopka, Florida) were washed clean of potting mix and transplanted 15 cm apart (9 per bed; 1 per container; 228 total) in hand-dug holes ∼5 cm deep. At transplant, P-free granulated fertilizer (N:P:K ratio of 10:0:10; 42 g bed−1, 4.7 g container−1; Rite Green; Sunniland Corporation, Sanford, Florida) was surface applied to all PRsoils while a granulated fertilizer with P (N:P:K ratio of 6:4:6; 70 g bed−1) was used in P-soils, both of which was repeated bimonthly. Overhead and drip irrigation were employed to 5312
dx.doi.org/10.1021/es400892a | Environ. Sci. Technol. 2013, 47, 5311−5318
Environmental Science & Technology
Article
maintain soil moisture (60−80% field capacity), which was measured with a Kelway HB-2 acidity and moisture tester (Kel Instruments, Wyckoff, New Jersey). Transparent corrugated roof panels with >90% light transmittance were used as a canopy for the plots. Plant Harvest. Five harvests were conducted in six month intervals (July 2010 to July 2012). Frond biomass was collected by cutting mature fronds ∼15 cm above the rhizome, leaving a few leaflets and young fiddleheads intact. Samples were ovendried at 60 °C for 96 h, weighed, and ground through a 2 mm mesh screen in a Wiley Mill (Thomas Scientific, Swedesboro, NJ). Frond samples (0.1 g) were subjected to HNO3/H2O2 digestion and analyzed for As and P as previously described. Soil and Root Sampling. At planting and each harvest, 30 cm soil cores were taken with an auger (3 cm diameter) approximately 7.5 cm away from the base of the ferns. Samples were taken before subsequent fertilizer applications to minimize influence of soluble P. Two cores were taken from each bed, separated by depth (0−15 cm and 15−30 cm) and composited. Samples were sieved through a 2 mm screen to separate root tissue and weighed. Soil samples were dried at 60 °C for 48 h and analyzed for elemental analysis as previously described. Pore Water Analysis. Rhizon soil moisture samplers were obtained from Soil Moisture Equipment Corp., Santa Barbara, CA. The samplers consisted of a length of porous plastic, capped with nylon, and attached to polyethylene tubing. Samplers were inserted into the soil at the base of PV plants, and equilibrated for two months before drawing off pore water under vacuum. Dissolved organic carbon (DOC) in pore water was analyzed using TOC-5050A total organic carbon analyzer (Shimadzu). Organic acids from root exudates in pore water were identified using a Waters 2690 Separations Module, a SUPELCOGEL C610H HPLC column (30 cm × 7.8 mm ID) and a Waters 2478 Dual λ absorbance detector set at 210 nm with a flow rate of 30 mL h−1. Total As and P in pore water samples were also assessed using previously described methods. Statistical Analysis. Data are presented as the mean of all replicates with standard error. Using one-way analysis of variance (ANOVA), significant differences were determined with treatment means compared by Duncan’s multiple range tests, at p < 0.05 using JMP PRO, Version 10 (SAS Institute Inc., Cary, NC, 1989−2010).
■
Table 1. Properties of As-Contaminated Soils Collected from a Former Wood Treatment Facility (CCA) and Two Abandoned Dipping Vat (DVA and DVB) Sites soil characteristic
CCA
DVB
DVA
pH total As (mg kg−1) bioaccessible As (mg kg−1)a available As (mg kg−1)b available P (mg kg−1)b Mehlich III P (mg kg−1) Mehlich III Ca (mg kg−1) Mehlich III Mg (mg kg−1) Mehlich III K (mg kg−1) Amorphous Al (mg kg−1)c Amorphous Fe (mg kg−1)c organic matter (%) CEC (cmol+ kg−1) sand (%) silt (%) clay (%) textural class
7.22 129 40.6 3.57 0.38 75.5 1541 115 26.8 854 2004 1.12 7.81 86.3 9.91 3.85 loam sand
5.06 29.9 12.8 1.53 0.11 41.5 256 54.0 22.9 481 29.0 0.38 12.4 80.7 6.58 12.7 sand loam
5.14 25.5 13.2 1.75 0.09 24.3 132 18.2 12.2 543 82.8 2.23 3.32 95.5 2.66 1.81 sand
a Glycine, 0.05 M. bAmmonium sulfate, 0.05 mM. cOxalic acid + ammonium oxalate, 0.2 M.
Prior to amendments, Mehlich-3 soil P concentrations were 76, 24, and 42 mg kg−1 in CCA, DVA and DVB soils, respectively, and water-soluble P concentrations were
