JEQ Grow Your Career With ASA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 5 January 2006
Published in J Environ Qual 35:334-341 (2006)
DOI: 10.2134/jeq2005.0181
© 2006 American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via ISI Web of Science (2)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Xu, S.
Right arrow Articles by Jaffé, P. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Xu, S.
Right arrow Articles by Jaffé, P. R.
Agricola
Right arrow Articles by Xu, S.
Right arrow Articles by Jaffé, P. R.
Related Collections
Right arrow Wetland Soils
Right arrow Evapotranspiration
Right arrow Remediation
Right arrow Heavy Metals
Right arrow Plant and Soil Interactions

TECHNICAL REPORTS

Wetlands and Aquatic Processes

Effects of Plants on the Removal of Hexavalent Chromium in Wetland Sediments

Shangping Xu* and Peter R. Jaffé

Department of Civil and Environmental Engineering, Princeton University, Princeton, NJ 08544

* Corresponding author (shangping.xu{at}yale.edu)

Received for publication May 10, 2005.

   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
The effect of two wetland plants, Typha latifolia L. (cattail) and Phragmites australis (Cav.) Trin. ex Steud (common reed), on the fate of Cr(VI) in wetland sediments was investigated using greenhouse bench-scale microcosm experiments. The removal of Cr(VI) was monitored based on the vertical profiles of aqueous Cr(VI) in the sediments. The Cr(VI) removal rates were estimated taking into account plant transpiration, which was found to significantly concentrate dissolved species in the sediments. After correcting for evapotranspiration, the actual Cr(VI) removal rates were significantly higher than would be inferred from uncorrected profiles. On average, the Cr(VI) removal rates were 0.005 to 0.017 mg L–1 d–1, 0.0003 to 0.08 mg L–1 d–1, and 0.004 to 0.13 mg L–1 d–1 for the control, T. latifolia, and P. australis microcosms, respectively. The fate of the removed Cr(VI) was examined by determining the quantity and chemical speciation of the Cr in the sediment and plant materials. Chromium(III) was the dominant form of Cr in both the sediment and plants, and precipitation of Cr(III) in the sediment was the major pathway responsible for the disappearance of aqueous Cr(VI) from the pore water. Incubation results showed that abiotic reduction was the primary mechanism underlying Cr(VI) removal in the microcosm sediments. Organic compounds produced by plants, including root exudates and mineralization products of dead roots, are thought to be the factor that is either directly or indirectly responsible for the gap between Cr(VI) removal efficiencies in the sediments of the vegetated and unvegetated microcosms.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
CHROMIUM has many applications in modern industry. Its production and industrial applications have resulted in large quantities of this element being discharged into the environment (Nriagu and Nieboer, 1988), resulting in significant contamination. For instance, as many as 110 sites in the USA have been identified as contaminated by waste mud from the processing of chromite ores, and the total amount of Cr in these sites was estimated to be >72 million kg (Palmer and Wittbrodt, 1991). In the natural environment, Cr exists mainly in two stable oxidation states: trivalent Cr and hexavalent Cr. In terms of toxicity, mobility and bioavailability, Cr(III) and Cr(VI) differ remarkably from each other (Bartlett and James, 1988; Calder 1988; Holdway 1988; Nieboer and Shaw, 1988; Wong and Trevors, 1988; Yassi and Nieboer, 1988). Chromium(VI), generally in the forms of chromate and dichromate, is toxic and very soluble in water. Since chromate and dichromate ions are negatively charged across a wide pH range (pH > 3; Dragun, 1998) and their adsorption to minerals and organic materials is thus limited, mobility of Cr(VI) in the environment is considered to be very high. In contrast, Cr(III) is much less soluble and thus less mobile, and it is a necessary micronutrient for animals (Blowes, 2002). For plants, however, both Cr(III) and Cr(VI) are nonessential and often toxic (Shanker et al., 2005) and in wetland sediments, mobility and bioavailability of Cr(III) could potentially be enhanced by the presence of some organic acids (Srivastava et al., 1999). Overall, the reduction of Cr(VI) to Cr(III) is an important process for the mitigation of environmental problems associated with Cr contamination (Blowes, 2002).

Chromium-bearing industrial wastewater is typically treated by chemical means, which generally involves the reduction of Cr(VI) to Cr(III) by reductants like Fe(II) and the subsequent adjustment of the solution pH to near-neutral conditions to precipitate the Cr(III) ions produced (Blowes et al., 1997; Chen and Hao, 1998; Eary and Rai, 1988). Microbial reduction of Cr(VI) to Cr(III) is also possible and is becoming a promising alternative to the remediation of Cr(VI)-contaminated sites via chemical reduction (Chen and Hao, 1998; Lovley and Coates, 1997).

Both natural and constructed wetlands are effective in transforming and removing both organic and inorganic contaminants as well as pathogens via either biological or abiotic pathways (Begg et al., 2001; Chendorain et al., 1998; Debusk et al., 1996; Hammer 1989; Hansen et al., 1998; Lin and Terry, 2003; Machemer et al., 1993; Martin et al., 2003; Mitsch and Wise, 1998; Pilon-Smits et al., 1999; Scholes et al., 1998; Scholes et al., 1999; Shutes et al., 2001; Stearman et al., 2003; Vidales et al., 2003). An advantage of the multiple mechanisms by which wetlands transform pollutants is that several contaminants can be degraded or immobilized simultaneously (Begg et al., 2001; Chendorain et al., 1998; Doyle and Otte, 1997; Hansen et al., 1998). Given that wetlands are often the transition zones between uplands and streams or lakes, or in some cases a significant point for groundwater recharge, they play an important role on the water quality of these systems. Therefore, a thorough understanding is needed of the mechanisms by which wetlands remove or immobilize specific pollutants.

The objective of this project was to investigate the immobilization of Cr(VI) in wetland sediments. Special attention was given to the effect of plants, which are ubiquitous in wetland environments and are usually the main source of organic matter that serves as the driving force for Cr(VI) reduction.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 REFERENCES
 
Four microcosms were constructed from plastic buckets (diameter, 26 cm; depth, 30 cm) by adding a drainage outlet at the center of the bottom. When the experiment was started, ASTM 20/30 silicon sand (US Silica, Ottawa, IL) was added to a depth of 20 cm, resulting in a porosity of 0.38, and shoots of T. latifolia (one microcosm) and P. australis (two microcosms) were transplanted to individual microcosms. Typha latifolia (cattail) and P. australis (common reed) were selected for this study because they are common in both natural wetlands and constructed wetland systems (e.g., Begg et al., 2001; Hansen et al., 1998; Mitsch and Wise, 1998). Furthermore, the ecological effects of P. australis are of interest due to its invasive nature (Chambers et al., 1999). A control microcosm without any plants was also maintained under similar experimental conditions. The microcosms were operated in a greenhouse under ambient temperature, light intensity, and humidity.

A modified Hoagland nutrient solution, containing 0.5 mg L–1 Cr(VI) (in the form of sodium chromate), 136 mg L–1 KH2PO4, 202 mg L–1 KNO3, 246 mg L–1 Ca(NO3)2, 241 mg L–1 MgSO4, 66 mg L–1 (NH4)2SO4, 2.42 mg L–1 Fe(NO3)3, 1.37 mg L–1 MnSO4, 2.86 mg L–1 H3BO3, 0.12 mg L–1 ZnSO4, 0.016 mg L–1 NaMoO4, and 0.06 mg L–1 CuSO4, was freshly prepared and pumped into the microcosms. The pH of freshly prepared nutrient solution is ~6.8. The nutrient solution prepared for the control and one of the two P. australis microcosms also contained 164 mg L–1 Na acetate as an extra C source. This was done so that reducing conditions could be established in the control (since unvegetated sediments in natural systems do have C inputs) and one P. australis microcosm was operated without the acetate addition so that its effect on the planted microcosms could be assessed.

The water level was kept at 3 cm above the sediments by pumping the nutrient solution into the top of the microcosms and allowing the excess to drain from an outlet located 3 cm above the sediments. The downward flow rate was maintained at about 350 mL d–1 using a peristaltic pump to drain the solution from the bottom of the containers. After 9 mo of operation, the drainage rate for the control, T. latifolia, and P. australis (with acetate input) microcosms was increased to 600 mL d–1 to test the impact of the drainage rate on the immobilization of Cr(VI). After an additional 2 mo of operation, these three microcosms were dismantled.

The operation of the P. australis microcosm without acetate addition was discontinued 6 mo after the plants were transplanted. The difference between the P. australis microcosms with and without acetate input during this time period was minor in terms of Cr(VI) profile and plant transpiration, showing that the acetate addition had a minimum impact, if any, on the planted microcosms.

Pore water samples were collected during daytime using stainless steel needles inserted in 2- or 3-cm increments from the water–sediment interface up to a depth of 20 cm. Initial tests showed that no detectable Cr was introduced into the water samples from new or reused stainless steel needles. At each depth, approximately 10 mL of pore water were collected from five different locations. Water samples collected at each depth were passed through 0.2-µm nylon filters (Fisher Scientific, Hampton, NH). The pH of filtered pore water samples was determined with a Fisher Scientific AccurapH electrode. Since the size of the microcosms used in the experiment was relatively large and many plants were growing in a single microcosm, this sampling strategy was implemented to help account for possible variations in the horizontal dimension that might be a concern for microcosm experiments performed with a single plant. This same sampling procedure was shown to be reliable for other chemical species sampled on the same microcosms (Xu et al., 2004).

A Dionex Ion Chromatograph (IC) (Model LC20, Dionex Corp., Sunyvale, CA) with a conductivity detector (Dionex CD25) was used to quantify concentrations of SO42– and NO3–. The injection loop was 25 µL, and the columns were Dionex IonPak AS-14 and AG14, both with a diameter of 4 mm. The eluant contained 3.5 mM Na2CO3 and 1 mM NaHCO3. The flow rate was set at 1 mL min–1. Total dissolved organic carbon (TOC) contents in the filtered pore water samples were determined using a TOC analyzer (Shimadzu TOC-5000A, Shimadzu Scientific Instruments, Kyoto, Japan). For the IC and TOC analyses, standards were prepared and analyzed for every three or four samples for quality control purposes.

Chromium(VI) was determined colorimetrically within 24 h using the 1,5-diphenylcarbazide method (American Public Health Association et al., 1989). The concentration of Cr(VI) in a few samples was checked after a few days and no difference was observed. For total Cr and Cr(VI) analyses, new calibration curves were prepared before a batch of samples was analyzed. When all the calibration curves [Cr(VI) and total Cr] were compared, no significant (<2.5%) variations were observed. Variations in ionic strength did not show any significant impact on Cr(VI) determination. It was previously reported that a variety of wetland plants can take up Cr(IV) and reduce it to Cr(III) (Howe et al., 1999; Lytle et al., 1998). To quantify the Cr accumulation in the shoot and roots of T. latifolia and P. australis, plant samples were collected after the microcosm operation was discontinued. Duplicate plant samples were rinsed, dried, and then thoroughly digested with H2SO4 and HNO3. Chromium(VI) contents were measured colorimetrically as mentioned above and total Cr determined by inductively coupled plasma–atomic emission spectrometry (ICP-AES). The ICP-AES was calibrated with both Cr(III) and Cr(VI) and no difference was observed between the two calibration relationships. Good agreement was found between the ICP-AES and colorimetry methods by measuring standard solution as blind samples. Chromium(III) was calculated as the difference between total Cr and Cr(VI).

Sediment samples were also collected in layers of 2 cm after the microcosms were dismantled. The remaining water content of these sediments was determined by measuring the weight loss after drying at 80°C for 48 h, to express the concentrations in terms of mass per dry mass of sediment and to correct for the dissolved concentrations. Chromium [either Cr(VI) or Cr(III)] and Fe were extracted from the gently rinsed sediments with 6 M HCl for 72 h. Complete recovery of spiked Cr(VI) to one of the sediment samples was achieved, suggesting insignificant conversion of Cr(VI) to Cr(III). Chromium(VI) and total Cr were determined after filtering the samples (0.2-µm nylon, Fisher Scientific) and adjusting the pH to ~1.0 using concentrated NaOH solution. The Cr(III) content was obtained by subtracting the Cr(VI) content from the total Cr content. Iron(II) was determined by the ferrozine method (Stookey 1970). Total Fe was obtained after Fe(III) was reduced with hydroxylamine. The effectiveness of reduction of Fe(III) by hydroxylamine was checked during the experiment and was found to be satisfactory. Ferric iron was expressed as the difference between total Fe and Fe(II).

Both biotic and abiotic reduction of Cr(VI) have been reported (Chen and Hao, 1998; Deng and Stone, 1996a; Lovley and Coates, 1997; Wittbrodt and Palmer, 1995, 1996, 1997). Incubation tests were performed to investigate the contributions from biological and abiotic pathways to the overall reduction of Cr(VI) in the microcosms. About 50 g of sediments obtained from a depth between 10 and 12 cm of the T. latifolia and P. australis (with acetate) microcosms were immediately frozen on the termination of the microcosm experiment. Six serum bottles (Fisher Scientific) with 100 mL of deoxygenated (purged with 80% N2 and 20% CO2 for 45 min) nutrient solution amended with 5 mM Na acetate and 2.0 mg L–1 Cr(VI) were prepared. The bottles were then divided into three groups of two. The first group of bottles did not contain any sediment, while 4 g of sediment from either the T. latifolia or the P. australis microcosms were added to the bottles of Group 2 and 3, respectively. One serum bottle out of each group was then autoclaved. All serum bottles were sealed with butyl rubber stoppers and stored in a dark incubator at 30°C and shaken periodically. Duplicate water samples were collected (every 24 h for the first 96 h and then with gradually increasing time intervals) from each bottle under constant flow of a 80% N2 and 20% CO2 mixture and filtered through a 0.2-µm nylon filter (Fisher Scientific) for the Cr(VI) analysis.


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
REFERENCES
 
Hexavalent Chromium Removal Rates
Vertical profiles of Cr(VI) measured after 6 and 11 mo of operation clearly show that Cr(VI) was removed from the aqueous phase in all microcosms (Fig. 1 ), with greater removal in the vegetated microcosms than in the unvegetated control. The vertical Cr(VI) profiles in the two P. australis microcosms were very similar, indicating that acetate addition did not have a significant effect on these microcosms. Although Cr(VI) concentration in freshly prepared nutrient solution was always 0.5 mg L–1, its concentration in the water samples collected at the water–sediment interface was sometimes ~0.4 mg L–1, which was slightly lower than the 0.5 mg L–1 of Cr(VI) in the inflow to the microcosms, suggesting that some Cr(VI) removal occurred in the overlying water column.



View larger version (20K):
[in this window]
[in a new window]
  		Fig. 1. Pore water Cr(VI) concentrations: (A) after 6 mo of operation; and (B) after 11 mo of operation. The Phragmites australis (without acetate) microcosm was only sampled once after 6 mo of operation. Lines represent nonlinear regression curves of the concentrations against depth. Chromium(VI) removal rates were calculated based on the regression curves to minimize the effects of experimental variations.

 
In the control microcosm with a constant drainage rate and under steady state conditions, the rate of Cr(VI) removal at each depth in the control equals the concentration difference multiplied by the flow rate. For the vegetated systems, however, the Cr(VI) removal rates in the sediments must be evaluated accounting for plant transpiration. If the transpiration rate is significant compared with the drainage rate, this process could substantially concentrate the dissolved compounds in the rhizosphere, as illustrated by Fig. 2 . A consequence of the increased downward water flow induced by plant transpiration is a higher loading of dissolved constituents such as nutrients and pollutants [i.e., Cr(VI) in this case] to the sediments. At the water–sediment interface of the vegetated microcosms, the downward flow rate is the summation of the drainage rate from the bottom of the microcosms plus the transpiration rate. The actual factor by which dissolved species are concentrated is determined by the ratio between the plant transpiration rate and the drainage rate.



View larger version (27K):
[in this window]
[in a new window]
  		Fig. 2. Conceptual graph showing the mechanism by which plant transpiration concentrates conservative dissolved species. For a thin layer of sediment with transpiration rate of Qtranspiration and inflow rate Qinflow with concentration of Cinflow, the outflow rate Qoutflow = Qinflow – Qtranspiration, and its concentration in the outflow Coutflow = (QinflowCinflow)/(Qoutflow) > Cinflow. With the known drainage rate and the vertical concentration profile of a conservative species, the transpiration and inflow rates for a specific layer can be estimated: QN – 1 = CNQN/CN – 1; QN – 1t = QN – 1 – QN; Qi – 1 = CiQi/Ci – 1; Qi – 1t = Qi – 1 – Qi. Here N represents total number of layers, while i is the sequential number of one specific layer; i = 0 for the sediment right beneath the water–sediment interface; QN – 1 is the inflow rate of layer N – 1, QN – 1 is the outflow rate of layer N – 1, and QN – 1t represents the transpiration rate in layer N – 1.

 
Using a conservative tracer, one can correct for the effect of plant transpiration and determine the water lost from a specific sediment layer due to transpiration (Fig. 2). The steady state Cr(VI) removal rate (reduction plus plant uptake) in a specific sediment layer can then be estimated through mass balance, in which the water inflow and outflow rates are now known. Bromide and Cl–, two widely used tracers in hydrological tests, were not suitable because of significant plant uptake of both ions (Xu et al., 2004). A tracer that can potentially be considered to be conservative in these microcosm experiments is SO42–. Three potential pathways could lead to the loss of SO42– in the microcosm sediments: dissimilatory SO42– reduction, precipitation or adsorption, and plant uptake. Since NO3– was present at millimolar levels in the pore water throughout the domain, and denitrification is energetically favorable over SO42– reduction (Park and Jaffe, 1996), SO42– reduction did not occur to any measurable degree. The fact that SO42– concentrations remained constant in the control microcosm (Fig. 3 ) suggests that no significant amount of SO42– was precipitated or adsorbed to the solid matrix. Sulfur is an essential nutrient and its content in dry plant materials usually ranges from 0.1 to 0.5% (Marschner 1995). Given that the loading of SO42– applied to the microcosms was >42 mg d–1, loss of SO42– due to plant uptake was also negligible.



View larger version (18K):
[in this window]
[in a new window]
  		Fig. 3. Pore water SO42– profiles: (A) after 6 mo of operation; and (B) after 11 mo of operation. The Phragmites australis (without acetate) microcosm was only sampled once after 6 mo of operation. Lines represent nonlinear regression curves of the concentrations against depth.

 
Sulfate profiles, which illustrate the relative strength of the concentration process in the rhizosphere due to plant transpiration, are shown in Fig. 3. Based on the vertical SO42– profiles, the spatial distribution of transpiration rates could be determined as follows:

[1]

[2]

[3]

[4]
Chromium(VI) removal rates in the vegetated microcosms could be then calculated:

[5]
where i represents the sequential number of a layer starting from the water–sediment interface, µi is the Cr(VI) removal rate, Qi – 1 and Qi are inflow and outflow rates for layer i, Qit is the transpiration rate in layer i, Ci – 1 and Ci are SO42– (for Eq. [1]GoGo–[4]) or Cr(VI) (Eq. [5]) concentrations in the inflow and outflow of layer i, Vi is the total volume of layer i, layer N is the bottom layer and it is obvious that QN is the drainage rate, which is known.

To minimize the impacts of experimental variability on these calculations, pore water SO42–, as well as Cr(VI) concentrations, were regressed against the depth (z). These regression curves (solid lines in Fig. 1 and 3) were then used to obtain the inflow, outflow, and Cr(VI) removal rates in each layer of the vegetated microcosms, using the equations given in Fig. 2.

Chromium(VI) removal rates as a function of the aqueous Cr(VI) concentration are presented in Fig. 4 . These results show that, after accounting for the concentration factor due to plant transpiration, the enhancement of Cr(VI) removal in the vegetated microcosms is substantial. A linear relationship between the pore water Cr(VI) concentration and the Cr(VI) removal rates is observed.



View larger version (18K):
[in this window]
[in a new window]
  		Fig. 4. Relationship between aqueous Cr(VI) concentrations and Cr(VI) removal rates: (A) after 6 mo of operation; and (B) after 11 mo of operation.

 
Chromium(VI) removal rates were higher in all planted microcosms than the microcosm without plants, suggesting that the presence of plants contributes to the Cr(VI) removal in the sediments. Such a removal enhancement was caused, at least in part, by the higher loading of Cr(VI) into the sediment created by plant transpiration. At the end of the 11-mo period, NO3– was measured and found to be uniform in the control, T. latifolia, and P. australis microcosms. When plant transpiration is considered, a uniform NO3– profile means that there was active NO3– removal. Again, as is the case for SO42–, this amount of NO3– removal cannot be due to plant uptake given the loading of NO3– to the microcosms, and indicates that active denitrification is occurring in these microcosms. The overall denitrification activity, based on the transpiration-corrected NO3– profiles, was higher in the plant microcosms than in the control.

Chromium Accumulation in the Plants and the Sediments
Major pathways that may have contributed to the removal of Cr(VI) in this study include uptake by plants, reduction of Cr(VI) to Cr(III), which then precipitates in the sediments, and adsorption of Cr(VI) onto the mineral surfaces.

Table 1 lists the Cr contents in plant materials, as well as the size of the Cr reservoir. Chromium in plant shoots was low and the roots of both plant species had the highest degree of Cr accumulation. Although the speciation of Cr in plant materials could not be reliably determined due to the limitations of the digestion method used in this experiment, available experimental evidence clearly suggests the reduction of Cr(IV) in various wetland plants (Howe et al., 2003; Lytle et al., 1998).


View this table:
[in this window]
[in a new window]
  		Table 1. Total Cr in Typha latifolia and Phragmites australis.

 
The other important sink in the removal of Cr(VI) via reduction is the precipitation and accumulation of Cr(III) in the sediments. Contents of Cr(III) and Cr(VI) in the sediments were thus determined by extraction. Results revealed that only a negligible (<2%) fraction of the total Cr was in the form of Cr(IV). The content of Cr(III) as a function of the depth into the sediments, expressed in terms of dry weight of sediment, is shown in Fig. 5 . The highest Cr(III) levels were observed at the water–sediment interface, where Cr(VI) concentrations were also highest.



View larger version (19K):
[in this window]
[in a new window]
  		Fig. 5. Concentrations of Cr(III) precipitated in the sediment.

 
The bulk density of the silica sand used in this experiment was 1.6 x 103 kg m–3. Therefore, the total amounts of Cr(III) in the sediments of the control, T. latifolia, and P. australis microcosms were 24.6, 56.6, and 115.0 mg, respectively. Comparing these amounts of Cr(III) in the sediments to the mass of Cr in plant tissue (Table 1) shows that there is a significant difference in the fraction of total Cr immobilized in T. latifolia (19.6%) and P australis (4.9%). It was also observed that the pH in the rhizosphere of the T. latifolia microcosm was slightly acidic (pH = 5.2), while that of the P. australis and control microcosms was near neutral (pH = 6.7 and 6.8, respectively). The relatively lower pH in the T. latifolia microcosm might have affected the degree of precipitation of Cr(III). In that case, a higher dissolved Cr(III) concentration in the T. latifolia microcosm sediments would result in a higher Cr(III) uptake, explaining, at least in part, why the concentration of Cr in the roots of T. latifolia was much higher than that in the roots of P. australis.

Total Cr added to each microcosm was estimated based on the following equation:

[6]
where TOTCr is the total amount of Cr input, CCr is the concentration of Cr in the nutrient solution (0.5 mg L–1), FET is the evapotranspiration factor, which is the ratio of SO42– concentrations in the drainage and nutrient solution, QDrainage is the drainage rate, and T is the duration of the experiment. The total amounts of Cr that the control, T. latifolia, and P. australis microcosms received were 65.3, 102.9, and 158.5 mg, respectively. The enhanced Cr(VI) input to the vegetated microcosms was due to plant transpiration. The amounts of Cr retained in the control, T. latifolia, and P. australis microcosms were 24.6, 70.2, and 120.9 mg, respectively.

The results shown in Fig. 6 demonstrate the linear relationship between the weighted average of the Cr(VI) removal rate and the mass of Cr(III) precipitated in the sediments. The results were weighted in terms of the relative duration of the different flow rates used in the microcosms (0.8 for the low flow rate and 0.2 for the high flow rate).



View larger version (15K):
[in this window]
[in a new window]
  		Fig. 6. Relationship between Cr(III) precipitation and Cr(VI) removal rate.

 
Incubation of Sediment Samples
Chromium(VI) can be reduced and immobilized by either abiotic (Deng and Stone, 1996a, 1996b; Gu and Chen, 2003) or biological processes (Middleton et al., 2003; QuiIntana et al., 2001; Sani et al., 2002; Schmieman et al., 2000; Viamajala et al., 2002a, 2002b), both of which are driven by organic compounds as the electron donor. To examine the contributions from these two processes to the overall Cr(VI) removal, an incubation experiment was conducted. In the incubation test, sediment samples collected and frozen upon the dismantlement of the microcosms were suspended in glass vials containing similar growth nutrient solution amended with 2 mg L–1 of Cr(VI) and 5 mM Na acetate. The results are shown in Fig. 7 .



View larger version (17K):
[in this window]
[in a new window]
  		Fig. 7. Loss of Cr(VI) in the incubation experiment with 100 mL of degassed Hoagland solution with 5 mM Na acetate and 2.0 mg L–1 Cr(VI): A = solution only; B = autoclaved solution only; C = solution with 4 g of sediment samples (10–12 cm) from the Typha latifolia microcosm; D = same as C but autoclaved; E = solution with 4 g of sediment samples (10–12 cm) from the Phragmites australis microcosm that received 2 mM acetate from the nutrient solution; and F = same as E but autoclaved. Error bars are smaller than symbols.

 
For the autoclaved sample, there was a significant drop in Cr(VI) concentration during or immediately after autoclaving. Organic materials might get hydrolyzed during autoclaving and produce small-molecule organic acids that could reduce Cr(VI) rapidly and result in the initial drop in Cr(VI) concentration in the autoclaved vial. After this rapid initial drop in Cr(VI) concentrations, rates of Cr(VI) reduction in both the autoclaved and the unautoclaved vials with sediment samples were comparable. This suggests that reduction of Cr(VI) was most likely a process that did not require bacterial growth. The abiotic reduction of Cr(VI) catalyzed on mineral surfaces such as TiO2, goethite, and Al oxide can be achieved by various organic compounds including lactic acid, mandelic acid, tartaric acid, oxalic acid, or salicylic acid (Deng and Stone, 1996a). Many of these compounds are found in root exudates of various plant species (Aulakh et al., 2001; Dakora and Phillips, 2002; Ryan et al., 2001) or may be produced during root or bacterial biomass turnover. Both humic and fulvic acids can also reduce and immobilize Cr(VI) (Wittbrodt and Palmer, 1995, 1996, 1997). In the control microcosm, such organic compounds may have originated from the turnover of bacteria, since acetate does not lead to the abiotic reduction of Cr(VI) (Deng and Stone, 1996a, 1996b). Average DOC contents in the filtered pore water samples were 3.6 (±0.6), 4.6 (±2.2), and 4.2 (±1.3) mg L–1 for the control, T. latifolia, and P. australis (with acetate) microcosms, respectively. In the vegetated microcosms, both plant exudates and microbial turnover may have contributed to the removal of Cr(VI); however, their contributions could not be quantitatively separated based on available experimental data.

Iron(III) in the Sediments
Chromium(VI) usually exists in the environment in the form of chromate and dichromate, both of which are negatively charged, and ferric (hydr)oxides provide good adsorption sites at mineral surfaces. The abiotic reduction of Cr(VI) by various organic reductants is a function of the strength of adsorption of Cr(VI) onto mineral surfaces (Deng and Stone, 1996a). As discussed above, abiotic Cr(VI) reduction appears to be responsible for Cr(VI) reduction in this study. During the extraction experiment, Fe(III) was also quantified to assess if the observed differences in Cr(VI) removal rates were due to differences in the density of ferric (hydr)oxide adsorption sites in the solid matrix. Distribution of Fe(III) in three microcosms at the end of the experiment showed no significant difference in the Fe(III) content along the vertical dimension of any microcosm or among different microcosms, all of which had an Fe(III) content of 0.35 ± 0.03 g kg–1. Therefore, the difference between the vegetated and unvegetated microcosms in terms of Cr(VI) removal efficiency was not due to differences in the adsorption of Cr(VI) onto ferric (hydr)oxides. The ferric (hydr)oxides, however, could have catalyzed the abiotic reduction of Cr(VI), driven by a wide variety of organic compounds (Deng and Stone, 1996a). And input of such organic compounds from plants through various pathways could be the primary reason for the observed difference in Cr(VI) removal efficiencies achieved with and without the presence of plants.

Environmental Significance
This study illustrates the significant effect that wetland plants can have on the reduction of Cr(VI) in wetland sediments and its immobilization as Cr(III). Plant root exudates, plant litter, and root turnover provide the electron donor required for the reduction of some organic (i.e., chlorinated solvents) and inorganic [i.e., Cr(VI), U(VI), and NO3–] contaminants. In addition to affecting the sediment biogeochemistry, many plants absorb, transform, and store specific pollutants (Tu and Ma, 2002; Windham et al., 2001, 2003), which for the case of Cr, varies significantly between T. latifolia and P. australis.

As illustrated here, wetland plants can significantly increase the loading of dissolved pollutants into the sediments through active transpiration. This process leads to a higher pollutant mass that enters the sediments, as is the case for Cr(VI) in this study, resulting in a larger mass of pollutant degradation or immobilization in these sediments. This is illustrated by the significantly larger mass of Cr(III) in the sediments of the P. australis microcosm, which had the highest evapotranspiration rate. Furthermore, evapotranspiration concentrates individual dissolved constituents in the rhizosphere, which, in the absence of toxic effects, may result in increased reaction rates, plant uptake rates, or both (Xu et al., 2004). This concentration effect requires careful consideration to link spatial or temporal changes of pollutant concentrations in wetland sediments to the kinetics of their transformations.


   ACKNOWLEDGMENTS
 
This research was funded by USEPA–Science To Achieve Results (STAR) Program, Grant no. R827288.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 





This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via ISI Web of Science (2)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Xu, S.
Right arrow Articles by Jaffé, P. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Xu, S.
Right arrow Articles by Jaffé, P. R.
Agricola
Right arrow Articles by Xu, S.
Right arrow Articles by Jaffé, P. R.
Related Collections
Right arrow Wetland Soils
Right arrow Evapotranspiration
Right arrow Remediation
Right arrow Heavy Metals
Right arrow Plant and Soil Interactions

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Crop Science
Vadose Zone Journal Journal of Plant Registrations
Journal of Natural Resources
and Life Sciences Education		 Soil Science Society of America Journal