Heavy metal displacement in chelate-irrigated soil during phytoremediation

Heavy metal displacement in chelate-irrigated soil during phytoremediation

Journal of Hydrology 272 (2003) 107–119 www.elsevier.com/locate/jhydrol Heavy metal displacement in chelate-irrigated soil during phytoremediation F...

297KB Sizes 1 Downloads 6 Views

Journal of Hydrology 272 (2003) 107–119 www.elsevier.com/locate/jhydrol

Heavy metal displacement in chelate-irrigated soil during phytoremediation F. Madrida,1, M.S. Liphadzib,2, M.B. Kirkhamb,* a

Instituto de Recursos Naturales y Agrobiologı´a de Sevilla, Consejo Superior de Investigaciones Cientı´ficas (CSIC), Apartado de Correos 1052, 41080 Sevilla, Spain b Department of Agronomy, Kansas State University, 2004 Throckmorton Hall, Manhattan, KS 66506-5501, USA

Abstract Heavy metals in wastewater sewage sludge (biosolids), applied to land, contaminate soils. Phytoremediation, the use of plants to clean up toxic heavy metals, might remove them. Chelating agents are added to soil to solubilize the metals for enhanced phytoextraction. Yet no studies follow the displacement and leaching of heavy metals in soil with and without roots following solubilization with chelates. The objective of this work was to determine the mobility of heavy metals in biosolids applied to the surface of soil columns (76 cm long; 17 cm diam.) with or without plants (barley; Hordeum vulgare L.). Three weeks after barley was planted, all columns were irrigated with the disodium salt of the chelating agent, EDTA (ethylenediamine tetraacetic acid) (0.5 g/kg soil). Drainage water, soil, and plants were analyzed for heavy metals (Cd, Cu, Fe, Mn, Ni, Pb, Zn). Total concentrations of the heavy metals in all columns at the end of the experiment generally were lower in the top 30 cm of soil with EDTA than without EDTA. The chelate increased concentrations of heavy metals in shoots. With or without plants, the EDTA mobilized Cd, Fe, Mn, Ni, Pb, and Zn, which leached to drainage water. Drainage water from columns without EDTA had concentrations of these heavy metals below detection limits. Only Cu did not leach in the presence of EDTA. Even though roots retarded the movement of Cd, Fe, Mn, Ni, Pb, and Zn through the EDTA-treated soil from 1 d (Cd) to 5 d (Fe), the drainage water from columns with EDTA had concentrations of Cd, Fe, Mn, and Pb that exceeded drinking water standards by 1.3, 500, 620, and 8.6 times, respectively. Because the chelate rendered Cd, Fe, Mn, Ni, Pb, and Zn mobile, it is suggested that the theory for leaching of soluble salts, put forward by Nielsen and associates in 1965, could be applied to control movement of the heavy metals for maximum uptake during chelate-assisted phytoremediation. q 2002 Elsevier Science B.V. All rights reserved. Keywords: Heavy metals; Leaching; Drainage water; Chelate; Phytoremediation; Biosolids

1. Introduction Heavy metals in sewage sludge, applied to land, contaminate soils. While recycling of wastewater * Corresponding author. Tel.: þ1-785-532-5731; fax: þ 1-7855326094/5391850. E-mail addresses: [email protected] (M.B. Kirkham), [email protected] irnase.csic.es (F. Madrid), [email protected] (M.S. Liphadzi). 1 Fax: þ34-954624002. 2 Fax: þ1-785-532-6094.

residuals back to land is important to conserve water and nutrients, the metals can runoff and pollute surface waters, move to groundwater through cracks, or be ingested by children playing on sludge-fertilized soil. For environmental safety, the heavy metals should be removed. Phytoremediation is the use of green plants to remove pollutants from soil. Addition of chelates, in conjunction with phytoremediation, is advocated for enhancing the clean up of soil contaminated by heavy metals (Thayalakumaran et al., 2000).

0022-1694/03/$ - see front matter q 2002 Elsevier Science B.V. All rights reserved. PII: S 0 0 2 2 - 1 6 9 4 ( 0 2 ) 0 0 2 5 8 - 5

108

F. Madrid et al. / Journal of Hydrology 272 (2003) 107–119

Ethylenediamine tetraacetic acid (EDTA) is the most commonly used chelating agent. Chelates increase solubility of heavy metals for plant uptake during phytoremediation (Brooks, 1998; Salt et al., 1998). The enhanced phytoextraction results in high metal concentrations in plants (Deram et al., 2000). The contaminated plants then are ashed and placed in a confined disposal area. If the metals are valuable, they can be extracted from the ash and recycled (Anderson et al., 1998). Chelates are used not only in phytoremediation but also in production agriculture to increase uptake of essential heavy metals (Cu, Fe, Mn, Zn). The synthetic chelates are crucial micronutrient sources in many of the leading brands of hightechnology, soluble fertilizers used in fertigation (Leymonie, 2001). Except for the pot experiment by Kirkham (2000), who found that EDTA enhanced Pb uptake by sunflower (Helianthus annuus L.), no information is available concerning the efficacy of chelates applied to biosolids-treated soils for metal removal, and no work reports if chelate-solubilized heavy metals in biosolids applied to soil can pollute drainage water. One would expect that the metals might not leach as readily in chelate-amended soil as free metals because heavy metals are bound to the organic matter in the biosolids. Work with EDTA-treated agricultural soil contaminated by salts of heavy metals shows that the possibility for drainage water pollution is high. In laboratory columns without plants, Kedziorek et al. (1998) observed leaching of Cd and Pb from a smelterpolluted soil during the percolation of EDTA. Vogeler et al. (2001) added to the surface of a large lysimeter (1.15 m deep) a solution of copper nitrate to simulate Cu contamination that occurs at timber treatment sites. A poplar tree grew in the lysimeter. Two weeks after the Cu was added, a pulse of EDTA was applied to the soil. Copper was mobilized by the chelate, and 7 d after addition, Cu started to breakthrough into the leachate. Concentrations peaked 16 d after EDTA addition. Measurement of Cu in the soil at the end of the experiment showed that solubilization as a result of chelation had moved the Cu down to the 700 cm depth. In follow-up studies, Clothier et al. (2002) found that large quantities of Cu in chelate-treated soil columns growing pasture plants escaped to drainage water (20% of the total Cu in the columns). The results confirmed that chelate-treated root zones are leaky

(Grcˇman et al., 2001). Clothier et al. (2002) highlighted the need for greater understanding of pollutant transport in soil with roots. Half of the population in United States obtains its drinking water from groundwater (Josephson, 1976), a percentage that has remained fairly constant throughout recent decades. The concentration of heavy metals, including Cd, Fe, Mn, Pb, and Zn, is limited in drinking water. Cadmium is highly toxic, especially in water compared to food because of antagonistic effects of components in the food. Cadmium concentrates in the liver and kidney, and minute amounts (5 £ 1026 M) of Cd are responsible for adverse effects on the kidneys (renal arterial hypertension) (Public Health Service, 1962). Even though Fe is essential in the human diet, it is a highly objectionable constituent in water supplies for either domestic or industrial use. Domestic consumers complain of the brownish color that Fe imparts to laundered goods. The taste of Fe in beverages can be readily detected at 1.8 mg/ml. The amount of Fe permitted in water is set to prevent objectionable taste, laundry staining, and toxicological significance in the diet (Public Health Service, 1962). Even though Mn, like Fe, is essential for humans, there are two reasons for limiting the concentration of Mn in drinking water: (1) to prevent esthetic and economic damage, and (2) to avoid possible physiological effects from excessive intake. The domestic consumer finds that Mn produces a brownish color in laundered goods and impairs the taste of beverages including coffee and tea, and complaints arise when the level of Mn exceeds 0.15 mg/ml. Hepatic cirrhosis has been produced in rats when treated orally with large doses of Mn (Public Health Service, 1962). Lead taken into the body can be seriously injurious to health, even lethal. It is a cumulative poison from three sources: food, air, and water. The amount of Pb ingested by a healthy individual per day is 0.32 mg, and, if intakes of Pb from its different sources exceed this amount, health could be endangered. Zinc is an essential and beneficial element in human metabolism and concentrates in the retina and the prostrate. Water can contain 50 mg/ml Zn with no noticeable harm. High levels of Zn (675 – 2280 mg/ml are gastrointestinal irritants (Public Health Service, 1962). Transport of contaminants depends upon root-zone processes (Clothier and Green, 1997), yet no study has

F. Madrid et al. / Journal of Hydrology 272 (2003) 107–119

compared movement of heavy metals in EDTAtreated soil with and without roots. Our objective was to determine the effect of EDTA on movement of heavy metals in biosolids applied as a surface layer to columns of soil with and without roots. We chose to grow barley, because it is the most salt tolerant of economically important plants (Richards, 1954), and we wanted a crop that might survive high-metal content resulting from EDTA treatment. The heavy metals studied were four essential ones for plant growth (Cu, Fe, Mn, Zn) and three toxic ones (Cd, Ni, Pb). The three toxic metals were chosen because concentrations of them, along with Cu, in biosolids frequently exceed average background soil concentrations by 10 to more than 100 fold (McBride, 1998). They are potentially dangerous contaminants in biosolids applied to land.

2. Materials and methods The experiment was carried out between 8 September and 21 November 2000 in a greenhouse at Kansas State University in Manhattan, KS (398080 N, 968370 W, 314 m ASL). Twelve clear plastic columns, each 17 cm in inner diameter and 76 cm long, were filled with 65 cm of a Haynie very fine sandy loam soil (coarse-silty, mixed, superactive, calcareous mesic Mollic Udifluvents) at a constant bulk density of 1.3 Mg/m3. It has a pH of 7.6 (Madrid et al., 2002), and its field capacity is about 0.33 m3/m3 (Song et al., 1999), and its saturated volumetric water content is about 0.40 m3/m3 (Tarara and Ham, 1997; Song et al., 1998). Columns were packed in layers, and each layer was weighed before it was put in a column and then watered and packed. The average weight of dry soil in each column for the 12 columns was 19,265 g (range: 19,052 – 19,429 g). Approximately 7 l of water were used to pack each column. The columns then stood for 5 d and were watered a couple of times with 100 ml each time. During this time, there was drainage. Each column stood in a 22 cm diameter pie pan, which collected the drainage water. Drainage water was removed daily by pipetting it out of the pie pan. The columns were wrapped in aluminum foil to prevent algae from growing along the sides of the columns. Over a 2-day period (18 – 19 September), liquid, digested biosolids (sewage

109

sludge) from the Manhattan, KS, wastewater treatment plant was applied to the surface of all columns at a rate of 120 kg N/ha. This was equivalent to 1 kg of liquid sludge per column (44 kg/m2 on a wet-weight basis or 0.32 kg/m2 on a dry-weight basis) with 500 g per column added on each day. One kg of biosolids was approximately equal to 1000 ml, and the sludge layer was 1 –2 mm thick. On 21 September 2000, 20 barley (Hordeum vulgare L. ‘Weskan’) seeds were planted in half of the columns. Seeds germinated on 25th September, 4 days after planting (4 DAP). The plants were thinned to 10 plants per column on 5th October (14 DAP). Three weeks after germination, the disodium salt of EDTA [EDTANa2(H2O)2], dissolved in water, was added to half of the columns at a rate of 0.5 g/kg soil, the rate recommended for chelate-assisted phytoremediation (Robinson et al., 1999). The amount of the EDTA salt added (0.14 M) was based on the average weight of the soil in each of the 12 columns (19,240 g). Thus there were four treatments: no plants, no EDTA; no plants, with EDTA; plants, no EDTA; plants, with EDTA. Each treatment was replicated three times (three columns). Throughout the experiment, height of shoots was measured by choosing at random three plants per column on each measurement day. Depth of root penetration was measured by removing the aluminum foil, so roots could be observed through the plastic column. Four days after the addition of EDTA on 28 DAP, three plant samples were taken, when detrimental effects due to the EDTA had already been observed for 2 days. One sample was a composite sample from the three columns with plants and no EDTA. Another sample was a composite from two columns with plants and with EDTA, and the third sample was from one column with plants and with EDTA that showed the most severe damage. The three samples were digested using a nitric-perchloric acid digest (Kirkham, 2000), and they were analyzed for Cd, Cu, Fe, Mn, Ni, Pb, and Zn using ICP-ES (inductively coupled plasma-atomic emission spectroscopy). Detection limits in mg/kg for the ICP-ES were Cd, 0.05; Cu, 0.20; Fe, 1.00; Mn, 0.60; Ni, 0.10; Pb, 0.10, and Zn, 0.10. Before EDTA addition, columns were kept well watered and the drainage water was removed, but volumes were not recorded. Columns were watered

110

F. Madrid et al. / Journal of Hydrology 272 (2003) 107–119

about 10 days during this period with 225 ml (10 mm) each time. After EDTA was added, 225 ml water was added daily to each column between 25 and 41 DAP, and 450 ml water was added daily to each column between 42 and 58 DAP for a total of 11,700 ml water added to each column. This was equivalent to two pore volumes, using the equation, one pore volume ¼ volume £ us, where us ¼ saturated volumetric water content. The volume of each column was 14,746 cm3 and us ¼ 0.40 m3/m3, so one pore volume ¼ 5898 cm3. The experiment was terminated after the two pore volumes had been added. Drainage water was collected every day from the first day of EDTA addition (25 DAP) until the end of the experiment (60 DAP). At harvest (21 November; 61 DAP; 37 days after EDTA addition), shoots were cut and fresh and dry weights were determined. Plants were dried at 70 8C for 48 h. Roots were extracted from 0 to 5 cm depth by washing in distilled water. Roots were below this depth, sometimes even more than 60 cm deep, but they were too fine to extract. Shoots were analyzed for Cu, Fe, Mn, and Zn, and roots were analyzed for Cd, Cu, Fe, Mn, Ni, Pb, and Zn, again using the nitricperchloric acid digest and the ICP-ES. The soil was divided into 10 cm layers, and analyzed for total concentrations of Cd, Cu, Fe, Mn, Ni, Pb, and Zn using a method similar to that of Sposito et al. (1982). In their method, total concentration of the heavy metals in the soil is determined on filtered extracts obtained from 2 g samples, which are digested overnight with 12.5 ml 4 M HNO3 at 80 8C. We used 2 g samples, but added 20 ml 4 M HNO3 and heated the mixture for 18 h at 85 8C in a water bath. The extract was analyzed using ICP-ES. The digested biosolids were analyzed for the same heavy metals using the modified method of Sposito et al. (1982). Table 1 shows the heavy metals in the biosolids. They are classified as Class A, as defined by the US Environmental Protection Agency, and heavy-metal limits for Class A biosolids are given in Table 1. Analysis of the biosolids showed that they had 0.72% dry solids, a pH of 6.3, an electrical conductivity of 2.6 mS/cm, 37,200 mg/kg N, and 23,560 mg/kg P. The table also gives the total concentrations of the heavy metals in the Haynie very fine sandy loam, determined in another experiment (Madrid et al., 2002). The concentration for each metal fell within normal ranges found in soils (Kirkham, 1979). Drinking water standards also are shown in Table 1,

Table 1 Heavy metal concentration in biosolids from Manhattan, Kansas, and the Haynie very fine sandy loam soil. Concentration limits set by the US Environmental Protection Agency (EPA) for heavy metals in biosolids applied to agricultural land also are given, along with limits of the heavy metals in drinking water set by the US Public Health Service Heavy metal

Manhattan KS, biosolids (mg/kg)

EPA limitb (mg/kg)

Haynie soilc (mg/kg)

Drinking watera (mg/ml)

Cd Cu Fe Mn Ni Pb Zn

12.5 ^ 0.14 533.9 ^ 50.6 29,238 ^ 1999 1469 ^ 92 40.8 ^ 1.6 110.5 ^ 1.8 591.0 ^ 29.7

39 1500 None None 420 300 2800

0.78 ^ 0.12 7.29 ^ 0.57 7691 ^ 158 130 ^ 3 8.04 ^ 0.05 17.6 ^ 1.7 14.6 ^ 0.4

0.01 1.0 0.3 0.05 None 0.05 5

a b c

From Public Health Service (1962). From Bastian (1997); for high quality biosolids. From Madrid et al. (2002).

for later reference to concentrations of the heavy metals in the drainage water. Extractable amounts of Cu, Fe, Mn, and Zn were determined in the soil at harvest. The soil was extracted with DTPA (diethylenetrinitrilo pentaacetic acid), and concentrations were determined using the ICP-ES. It was recognized that DTPA is also a chelating agent, as is EDTA. However, the standard procedure of Lindsay and Norvell (1978) for extracting these four essential heavy metals was followed. Drainage water was analyzed for Cd, Cu, Fe, Mn, Ni, Pb, and Zn, starting 17 days after EDTA addition (41 DAP). This was the first day that a metal (Ni) became detectable in the drainage water. Drainage water then was analyzed daily until the end of the experiment (60 DAP) using the ICP-ES. Detection limits for water were the same as those for plants, except the detection limit for Cd was set at 0.005 mg/kg. Means and standard deviations are shown in the figures and tables.

3. Results 3.1. Leachate Without EDTA, levels of heavy metals in the leachate were below the set detection limits. With EDTA, all the heavy metals, except Cu, leached (Fig. 1).

F. Madrid et al. / Journal of Hydrology 272 (2003) 107–119

111

Fig. 1. Concentrations of six heavy metals (Cd, Fe, Mn, Ni, Pb, and Zn) in leachate from 76 cm long columns of Haynie very fine sandy loam with a 1–2 mm layer of sludge at the surface. Columns were treated with 0.5 g EDTANa2(H2O)2 per kg soil 25 days after barley was planted in half of them. Concentrations of the heavy metals were below detection limits in control columns (no EDTA added). Vertical bars show ^standard deviation (n ¼ 3).

112

F. Madrid et al. / Journal of Hydrology 272 (2003) 107–119

It is not known why Cu was below detection limits, even though its concentration in the biosolids was significant (one-third of the EPA limits). Other studies have shown that Cu leaches readily through EDTAtreated soil to drainage water (Vogeler et al., 2001; Clothier et al., 2002). The organic matter in the sludge might have bound the Cu making it relatively unavailable for leaching, even in the presence of EDTA. Cadmium leached into the drainage water, and the concentration reached levels above the one permitted in drinking water (0.01 mg/ml; Table 1). Peak concentration of Cd in columns without plants was 0.007 mg/ml and occurred 1 d before the peak concentration appeared in columns with plants. This peak was 0.013 mg/ml and was sustained for 3 d (50 – 52 DAP). It was 1.3 times the drinking water standard (0.013 4 0.010 mg/ml). If the chelate had been added to biosolids-treated soil in the field, the Cd could have leached to groundwater and contaminated it. Columns with plants leached more Cd than columns without plants. One can see by the relative areas under the curves in Fig. 1 the greater yield of mobilized Cd with roots. If one multiplies the amount of leachate (daily data not shown; between 41 and 60 DAP when 450 ml was added daily, average amount of daily drainage water and standard deviation from columns with EDTA was 347 ^ 25 ml, when no plants were present, and 302 ^ 26 ml, when plants were present) by the concentration of Cd in the leachate (Fig. 1, top left), one calculates that 25.87 mg leached without plants and 40.80 mg leached with plants. Therefore, about 1.5 times more Cd leached with plants than without. The amount leached was small compared to the total amount of Cd in the columns (15.03 mg Cd in soil and 0.99 mg Cd in biosolids). Iron leached in high amounts, and the leachate from columns with EDTA was yellow, probably due to the Fe. Leachate from columns without EDTA was clear. The concentrations of Fe in the drainage water were as much as 500 times the amount allowed in drinking water (Table 1) (150 4 0.3 mg/ml). Peak concentration of Fe in leachate from columns with plants appeared 5 d after the peak for columns without plants (48 versus 53 DAP, Fig. 1, upper right). The peak concentration of Fe in leachate without plants (150 mg/ml) was higher than the peak concentration of Fe in leachate from columns with plants

(105 mg/ml). Columns with plants leached less Fe than columns without plants. The difference was almost two fold; 410,611 mg of Fe leached out of the columns without plants and 236,763 mg leached with plants. The amount leached was small compared to the total amount of Fe in the columns (148 g Fe in the soil plus 0.210 g Fe in the biosolids). It is not known why in both columns with and without plants, Mn peaked two times in leachate (Fig. 1, middle left). The oxidation state of Mn, in the presence of EDTA, depends upon the pH and the amount of excess EDTA (Klewicki and Morgan, 1998). The oxidation state of Mn also depends on aeration status of the soil. A constant amount of water was added daily to the columns and a constant amount drained out, so it is assumed that the redox potential of the soil did not vary as a result of watering regime. As for Cd and Fe, the concentration of Mn in the leachate exceeded the drinking water limit, which is 0.05 mg/ml, and the peak excess was 620 times the limit (31 4 0.05 mg/ml). Also, as for Cd and Fe, the plants retarded the movement of Mn through columns, both when the first and second peaks appeared, and the difference was 2 d for the first peak (44 versus 46 DAP) and 1 d for the second peak (53 versus 54 DAP). Almost two times more Mn leached out of columns without plants (61,133 mg) than with plants (35,468 mg), but amount leached was small compared to the total amount of Mn in the columns (2.50 g Mn in soil plus 0.011 g Mn in the biosolids). Plants retarded the movement of Ni through the columns, and peak concentration of Ni in leachate from columns without plants occurred 48 DAP versus 52 DAP for columns with plants. Peak concentrations in columns with and without plants were similar (about 5 mg/ml). Significant amounts of Ni leached out of the soil, and the amount was about the same with plants (14,424 mg) as without plants (14,155 mg). Total amount of Ni in each column was 155.2 mg (154.9 mg from the soil and 0.294 mg from the biosolids). Peak concentration of Pb in leachate from columns without plants occurred 48 DAP and with plants, 51 DAP. As for Cd, more Pb leached with plants (1017 mg) than without (310 mg). It is not known why the plants enhanced the movement of Pb. Maximum amount of Pb in the leachate (0.43 mg/ml) exceeded the drinking water standard (0.05 mg/ml),

F. Madrid et al. / Journal of Hydrology 272 (2003) 107–119

and the excess was 8.6 times the limit. If a human drank the average amount of water consumed each day (2 l) (Public Health Service, 1962, p. 45) and it contained the maximum amount of Pb found in the leachate in our experiment (0.43 mg/ml), then that person would ingest 0.86 mg/d, 2.7 times the normal amount. An intake of Pb in excess of 0.6 mg/d results in the accumulation of a dangerous quantity of Pb in the body during a lifetime (Public Health Service, 1962). Total amount of Pb in each column was 339.8 mg (339 mg in the soil and 0.796 mg in the biosolids). Peak concentration of Zn in columns without plants occurred 48 DAP and with plants, 52 DAP. About the same amount of Zn leached with plants (2620 mg) as without plants (2042 mg). Concentrations of Zn in the leachate did not exceed the drinking water limit (5 mg/ml). The total concentration of Zn in each column was 285.5 mg (281.3 mg in the soil and 4.25 mg in the biosolids). The average total amounts of leachate collected for the four treatments during the 37 days after the EDTA was added to the columns were as follows: columns with no plants and no EDTA: 9446 ml; columns with plants and no EDTA, 6136 ml; columns with no plants and with EDTA, 9519 ml; columns with plants and with EDTA, 7781 ml. Therefore, in columns without plants, EDTA had little effect on the amount of leachate (73 ml difference). This suggested that EDTA did not change the structure and flow in the soil. However, when plants were present, columns with EDTA leached 1645 ml more than columns without EDTA. The greater leaching in columns with plants and EDTA showed that the EDTA was damaging the roots, which was confirmed by the root length measurements (see below). 3.2. Soil Total concentrations of the toxic heavy metals (Cd, Ni, and Pb) in the columns at the end of the experiment generally were lower in the top 30 cm in soil with EDTA than without EDTA, both in columns without plants (Fig. 2, left) and with plants (Fig. 2, right). Plants had no effect on concentration of heavy metals in the soil. Below 30 cm, concentrations were similar, even though there was a tendency for Cd and Pb in the columns with EDTA to have a higher concentration at depth than the columns without EDTA, suggesting that the EDTA pushed these heavy

113

metals from the top half of the columns to the bottom half. Total concentrations of the essential heavy metals (Fig. 3), in general, followed a similar pattern, except for Fe. At each depth, total concentration of Fe in columns without EDTA was not significantly different from that in columns with EDTA, either with plants or without plants. Extractable concentrations of the essential heavy metals (Fig. 4) paralleled concentrations for total amounts (Fig. 3), except extractable Fe in columns with plants and with EDTA was lower at the surface than that in columns with plants, but without EDTA. 3.3. Plants Damage to the plants (wilting) was observed within 48 h after the EDTA was added. Root elongation (Fig. 5, left) and shoot growth (Fig. 5, right) stopped immediately. At harvest, the fresh weights of shoots in each column with and without EDTA (^ standard deviation) were 15.9 ^ 11.0 and 30.7 ^ 8.5 g, respectively, and the dry weights were 3.6 ^ 2.4 and 7.3 ^ 1.6 g, respectively. The dry weights of the roots in each column with and without EDTA were 0.737 ^ 0.363 and 1.093 ^ 0.146 g, respectively. Thus, the EDTA had reduced the shoot fresh and dry weights by half and the root dry weights by a third. These results showed that the plants stopped growing when EDTA was added, and EDTA greatly reduced fresh and dry weights. With EDTA, the concentrations of the heavy metals in the shoots were increased, but those in the roots were not, except for Fe (Table 2). Even though plant shoots took up more metals with EDTA than without, because the EDTA stopped growth, the total amount of metals in shoots did not differ due to EDTA treatment. Combining roots and shoots, the plants in each column without EDTA had 55.0 mg Cu; 1285 mg Fe; 393 mg Mn, and 101 mg Zn. With EDTA, the plants in each column had 53.2 mg Cu; 1550 mg Fe; 350 mg Mn; and 101 mg Zn. Plants grown with EDTA accumulated miniscule amounts of Fe, Mn, and Zn compared to the amounts leached. As noted previously, columns with plants and EDTA leached 236,763 mg Fe; 35,468 mg Mn; and 2620 mg Zn.

114

F. Madrid et al. / Journal of Hydrology 272 (2003) 107–119

Fig. 2. Total concentrations of three toxic heavy metals, Cd, Ni, and Pb, in columns of Haynie very fine sandy loam with a 1 –2 mm layer of sludge at the surface. Columns were treated with 0.5 g EDTANa2(H2O)2 per kg soil 25 days after barley was planted in half of the columns. Control columns received no EDTA. Left: columns with no plants; right: columns with plants. W and %, columns with no EDTA; A and k , columns with EDTA. Horizontal bars show ^standard deviation (n ¼ 3).

F. Madrid et al. / Journal of Hydrology 272 (2003) 107–119

115

Fig. 3. Total concentrations of four essential heavy metals, Cu, Fe, Mn, and Zn, in columns of Haynie very fine sandy loam with a 1 –2 mm layer of sludge at the surface. Columns were treated with 0.5 g EDTANa2(H2O)2 per kg soil 25 days after barley was planted in half of the columns. Control columns received no EDTA. Left: columns with no plants; right: columns with plants. W and %, columns with no EDTA; A and k , columns with EDTA. Horizontal bars show ^standard deviation (n ¼ 3).

116

F. Madrid et al. / Journal of Hydrology 272 (2003) 107–119

Fig. 4. Extractable concentrations of four essential heavy metals, Cu, Fe, Mn, and Zn, in columns of Haynie very fine sandy loam with a 1–2 mm layer of sludge at the surface. Columns were treated with 0.5 g EDTANa2(H2O)2 per kg soil 25 days after barley was planted in half of the columns. Control columns received no EDTA. Left: columns with no plants; right: columns with plants. W and %, columns with no EDTA; A and k , columns with EDTA. Horizontal bars show ^standard deviation (n ¼ 3).

F. Madrid et al. / Journal of Hydrology 272 (2003) 107–119

117

Fig. 5. Root length (left) and plant height (right) of barley growing in columns of Haynie very fine sandy loam with a 1–2 mm layer of sludge at the surface. Columns were treated with 0.5 g EDTANa2(H2O)2 per kg soil 25 days after barley was planted. Arrow indicates day of EDTA addition. Control columns received no EDTA. W, no EDTA; A, with EDTA. Vertical bars show ^standard deviation (n ¼ 3).

Table 2 Concentration of heavy metals in plants

4. Discussion

Heavy Shoots metal No EDTA (mg/kg)

Clothier et al. (2002) found that EDTA was effective at mobilizing Cu and Fe, and transporting them as rapidly as an inert solute, bromide, a conservative tracer. Data (this experiment; Clothier et al., 2002) show that in chelate-facilitated phytoremediation large quantities of heavy metals are leached out of the root zone, despite the increased uptake of metals into the plants. The amount that the plants take up is minor compared to that leached. Clothier et al. (2002) used a lower concentration of EDTA (0.01 M) than we did (0.14 M). Yet even with the lower concentration, 20% of the Cu in their columns leached out, while only 0.2% was taken up by pasture grass. Because chelates render immobile heavy metals mobile like conservative tracers, the principles described by Nielsen et al. (1965) and used by Miller et al. (1965) might be applied. Miller et al. (1965) showed that chloride movement in soil depends upon the method of water application. They found that intermittently ponding a field soil with 51 mm increments of water was more efficient in leaching applied chloride from the soil surface than continuous

4 days Cd Cu Fe Mn Ni Pb Zn

Roots With EDTA (mg/kg)

after EDTA addition (28 days after planting) 2.5a 3.8 ^ 0.3b – c 11.9 19.7 ^ 2.5 – 175.0 336.3 ^ 77.1 – 59.2 120.7 ^ 46.1 – 5.5 9.2 ^ 1.0 – 11.0 19.7 ^ 2.3 – 16.5 23.5 ^ 2.6 –

At harvest (61 days after planting) Cd – – Cu 6.0 ^ 0.0 11.7 ^ 3.0 Fe 50.7 ^ 3.8 210 ^ 162 Mn 44.3 ^ 13.6 86.3 ^ 176 Ni – – Pb – – Zn 11.3 ^ 1.5 25.0 ^ 4.6 a

No EDTA (mg/kg)

With EDTA (mg/kg)

– – – – – – –

NDd ND 10.2 ^ 1.4 15.0 ^ 4.0 837.0 ^ 40.3 1077.5 ^ 178.6 63.4 ^ 15.7 53.1 ^ 4.2 1.8 ^ 0.7 1.1 ^ 0.3 ND ND 16.6 ^ 4.7 15.3 ^ 1.7

Analysis of one sample only. Analysis of two samples. c Not analyzed. d Not detected. Detection limit for Cd was 0.05 mg/kg and for Pb, 0.10 mg/kg. b

118

F. Madrid et al. / Journal of Hydrology 272 (2003) 107–119

ponding or leaching intermittently with 152 mm increments. For example, after 305 mm of water had been applied in 152 or 51 mm increments, the chloride concentration at the 152 mm depth was 75 and 20 mmol/l, respectively. Irrigating chelate-treated soil with large, infrequent irrigations might keep the solubilized heavy metals near the soil surface for plant uptake. However, the fact that plants exacerbated leaching for some metals (e.g. Cd, Fig. 1, top left) and reduced others (e.g. Fe, Fig. 1, top right) shows that the metals are not moving through the soil as simple soluble salts in the presence of EDTA. If the metals behaved as soluble salts whose leaching could be predicted using the theory of Nielsen et al. (1965), it seems that all metals rendered soluble by EDTA would show consistent trends in the presence or absence of plants. It is not known why in some cases the concentration of a metal leached was greater from columns with plants than from columns without plants. It could be related to essentiality of the element. In the cases of Cd and Pb, both non-essential elements, columns with plants had a higher concentration of the metal in the leachate than did columns without plants (Fig. 1). In the cases of Fe and Mn, both essential elements, columns without plants had a higher concentration of Fe and Mn in the leachate than columns with plants. The plants may have been taking up the Fe and Mn, reducing their concentration in the leachate. Plant roots may have evolved special biochemical pathways to absorb essential elements, which are lacking for non-essential elements. The effect of plant roots on uptake and leaching of essential and toxic heavy metals solubilized by EDTA needs further study. The fact that EDTA leaches heavy metals out of the soil might be used to clean up soils even without plants. Decontamination efforts during weapons production has involved the generation of chelated radionuclides (Jardine and Taylor, 1995). Heavy metals might be leached out of soil with EDTA, like radionuclides. The heavy metals would have to be confined in drainage water, which would have to be collected and then evaporated. The heavy metals then could be placed in a safe disposal area. Chelate-facilitated phytoremediation still might be used, even if the leachates contain high concentrations of heavy metals. The metal-enriched drainage water could be collected using a dual-pipe

subirrigation– drainage system (Kirkham and Horton, 1993) and recycled back to crops for further phytoremediation. In the system described by Kirkham and Horton (1993), water intercepted by the drain tubes is returned by pumping it back into the irrigation supply. The method conserves nutrients and reduces the potential for groundwater pollution. A barrier is needed at depth to prevent solutes from moving to groundwater. If no natural barrier exists, it might be cheaper to place a barrier at depth and use chelateassisted phytoremediation, rather than excavating contaminated soil. Crops will have to be identified that can grow in the presence of high concentrations of chelate-solubilized heavy metals. Genetic variability in willows and poplars for Cd tolerance exists (Robinson et al., 2000). With tolerant plants and new irrigation strategies, it should be possible to improve tactics for chelate-assisted phytoremediation that minimize groundwater pollution.

Acknowledgements We thank Dr Brett H. Robinson for his helpful comments in the review of this manuscript. This is contribution no. 02-252-J from the Kansas Agricultural Experiment Station, Manhattan, Kansas 665065501, USA.

References Anderson, C.W.N., Brooks, R.R., Stewart, R.B., Simcock, R., 1998. Harvesting a crop of gold in plants. Nature 395, 553 –554. Bastian, R.K., 1997. Biosolids management in the United States. Water Environment and Technology 9 (5), 45–50. Brooks, R.R., 1998. General introduction. In: Brooks, R.R., (Ed.), Plants that Hyperaccumulate Heavy Metals, CAB International, Wallingford, pp. 1–14. Clothier, B.E., Green, S.R., 1997. Roots: the big movers of water and chemical in soil. Soil Science 162, 534 –543. Clothier, B.E., Green, S.R., Robinson, B.H., Thayalakumaran, T., Scotter, D.R., Vogeler, I., Mills, T.M., Deurer, M., van der Velde, M., Granel, Th., 2002. Contaminants in the rootzone: bioavailability, uptake and transport, and their implications for remediation. In: Naidu, R., (Ed.), Chemical Bioavailability in the Terrestrial Environment, CSIRO Publishing, Collingwood, In review. Deram, A., Petit, D., Robinson, B.H., Brooks, R.R., Gregg, P.E.H., van Halluwyn, C., 2000. Natural and induced heavy-metal accumulation by Arrhenatherum elatius: implications for

F. Madrid et al. / Journal of Hydrology 272 (2003) 107–119 phytoremediation. Communications in Soil Science and Plant Analysis 31, 413– 421. Grcˇman, H., Velikonja-Bolta, Sˇ, Vodnik, D., Kos, B., Lesˇtan, D., 2001. EDTA enhanced heavy metal phytoextraction: metal accumulation, leaching and toxicity. Plant and Soil 235, 105–114. Jardine, P.M., Taylor, D.L., 1995. Fate and transport of ethylenediamine tetraacetate chelated contaminants in subsurface environments. Geoderma 67, 125 –140. Josephson, J., 1976. Quality assurance for groundwater. Environmental Science and Technology 10, 226–227. Kedziorek, M.A.M., Dupuy, A., Bourg, A.C.M., Compere, B., 1998. Leaching of Cd and Pb from a polluted soil during the percolation of EDTA: laboratory column experiments modeled with a non-equilibrium solubilization step. Environmental Science and Technology 32, 1609–1614. Kirkham, M.B., 1979. Trace elements. In: Fairbridge, R.W., Finkl, C.W. Jr. (Eds.), Encyclopedia of Soil Science. Part 1. Physics, Chemistry, Biology, Fertility, and Technology, Dowden, Hutchinson and Ross, Stroudsburg, pp. 571 –575. Kirkham, M.B., 2000. EDTA-facilitated phytoremediation of soil with heavy metals from sewage sludge. International Journal of Phytoremediation 2, 159–172. Kirkham, D., Horton, R., 1993. Modeling water flow from subirrigation with drainage. Soil Science Society of America Journal 57, 1451–1457. Klewicki, J.K., Morgan, J.J., 1998. Kinetic behavior of Mn(III) complexes of pyrophosphate, EDTA, and citrate. Environmental Science and Technology 32, 2916–2922. Leymonie, J.-P. (Ed.), 2001. Lost in the Jungle of Micronutrient Products?, New Ag International. September Issue, NAI Ltd, Teddington, pp. 40–48. Lindsay, W.L., Norvell, W.A., 1978. Development of a DTPA soil test for zinc, iron, manganese, and copper. Soil Science Society of America Journal 42, 421–428. Madrid, F., Liphadzi, M.S., Kirkham, M.B., 2002. EDTA-assisted phytostabilization by barley roots contaminated with heavy metals. In: Naidu, R., (Ed.), Chemical Bioavailability in the Terrestrial Environment, CSIRO Publishing, Collingwood, In review. McBride, M.B., 1998. Growing food crops on sludge-amended soils: problems with the US environmental protection agency method of estimating toxic metal transfer. Environmental Toxicology and Chemistry 17, 2274–2281. Miller, R.J., Biggar, J.W., Nielsen, D.R., 1965. Chloride displacement in Panoche clay loam in relation to water movement and distribution. Water Resources Research 1, 63– 73.

119

Nielsen, D.R., Biggar, J.W., Luthin, J.N., 1965. Desalinization of soils under controlled unsaturated flow conditions. International Commission on Irrigation and Drainage. Sixth Congress, R. 2, Question 19, 19.15–19.24. Public Health Service, 1962. Drinking water standards. Publication No. 956. United States Department of Health, Education, and Welfare, Washington, DC, 61 pp. Richards, L.A. (Ed.), 1954. Diagnosis and Improvement of Saline and Alkaline Soils, Agricultural Handbook, vol. 60. United States Department of Agriculture, Washington, DC, 160 pp. Robinson, B., Mills, T., Clothier, B., 1999. Phytoremediation of heavy-metal contaminated soils. WISPAS, A Newsletter about Water in the Soil-Plant-Atmosphere System No. 72. Environment Group, HortResearch, Palmerston North, New Zealand, pp. 1–2. Robinson, B.H., Mills, T.M., Petit, D., Fung, L.E., Green, S.R., Clothier, B.E., 2000. Natural and induced cadmium-accumulation in poplar and willow: implications for phytoremediation. Plant and Soil 227, 301–306. Salt, D.E., Smith, R.D., Raskin, I., 1998. Phytoremediation. Annual Review of Plant Physiology and Plant Molecular Biology 49, 643 –668. Song, Y., Ham, J.M., Kirkham, M.B., Kluitenberg, G.J., 1998. Measuring soil water content under turfgrass using the dualprobe heat-pulse technique. Journal of the American Society for Horticultural Science 123, 937 –941. Song, Y., Kirkham, M.B., Ham, J.M., Kluitenberg, G.J., 1999. Dual probe heat pulse technique for measuring soil water content and sunflower water uptake. Soil and Tillage Research 50, 345–348. Sposito, G., Lund, L.J., Chang, A.C., 1982. Trace metal chemistry in arid-zone field soils amended with sewage sludge. I. Fractionation of Ni, Cu, Zn, Cd, and Pb in solid phases. Soil Science Society of America Journal 46, 260– 264. Tarara, J.M., Ham, J.M., 1997. Measuring soil water content in the laboratory and field with dual-probe heat-capacity sensors. Agronomy Journal 89, 535–542. Thayalakumaran, T., Vogeler, I., Scotter, D.R., Robinson, B.H., Clothier, B.E., Green, S.R., 2000. Chelation therapy for dirty soils. 2000 Annual Meetings Abstracts. American Society of Agronomy, Crop Science Society of America, Soil Science Society of America, Madison, Wisconsin, pp 202. Vogeler, I., Green, S.R., Clothier, B.E., Kirkham, M.B., Robinson, B.R., 2001. Contaminant transport in the root zone. In: Iskandar, I., Kirkham, M.B. (Eds.), Trace Elements in Soils: Bioavailability, Fluxes and Transfer, Lewis, Boca Raton, FL, pp. 175–197.