| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Remediation and Containment Branch, National Risk Management Research Lab., USEPA, 5995 Center Hill Ave., Cincinnati, OH 45224-1702
b Dep. of Agronomy, Throckmorton Plant Sciences Center, Kansas State Univ., Manhattan, KS 66506-5501
* Corresponding author (hettiarachchi.ganga{at}epa.gov)
Received for publication May 4, 2000.
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: AR, active repository • HP, hydroxypyromorphite • PA, phosphoric acid • PBET, physiologically based extraction test • PR, phosphate rock • TCR, time critical repository • TSP, triple superphosphate • XRD, X-ray diffraction
|
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Before this technique can be adopted, several issues need to be resolved including the long-term stability of the newly formed Pb phosphates, the appropriate amounts of P to add, P source, and the effects of P on other metals commonly associated with Pb in contaminated soils (e.g., Zn and Cd). Because this in situ technique does not change the total Pb concentration in soil, information on the amount of time required for reducing soil Pb bioavailability after P application as well as the long-term effects of P on soil Pb bioavailability is very important.
Issues related to how soil Pb bioavailability is estimated also are unresolved. In particular, assessments of human health risk from Pb frequently employ the integrated exposure uptake biokinetic model (IEUBK; USEPA, 1994) using Pb bioavailabilities in soil and dust as model inputs. Typically, the USEPA relies on feeding studies using immature swine as the model animal, because their digestive systems are similar to those of humans. The animals are given various doses of Pb-contaminated soil, and Pb uptake is assessed to determine the relative bioavailability compared with a very soluble form of Pb such as Pb acetate. However, these feeding studies are expensive and considerable research has been done with the objective of finding a less expensive alternative. An in vitro bioaccessibility test (also known as the physiologically based extraction test, PBET) that mimics the human digestive system has been developed (Ruby et al., 1996). The PBET has been validated for use in determining soil Pb bioavailability with animal feeding studies done with weanling rats (Ruby et al., 1996) and young swine (Medlin, 1997).
In vivo and in vitro assays have indicated that Pb availability in mammalian gastrointestinal systems depends on the form and relative dissolution rates of Pb solids (Ruby et al., 1994). In vitro bioaccessibility of Pb from different Pb-containing minerals also has been tested, and only 1 to 5% of Pb from pyromorphite was bioaccessible (Medlin, 1997). However, bioaccessible Pb in soils high in Pb phosphates was approximately 50%. The reason for this difference is not known, but it might be due to differences in crystallinity. Other studies have used electron microprobe techniques to identify Pb associated with P (Cotter-Howells, 1996; Cotter-Howells and Caporn, 1996; Laperche et al., 1997), but the results of Medlin (1997) suggested that the mere association of Pb with P may not be a good indicator of bioavailability.
Lead shows greater solubility at low pH. Zhang and Ryan (1999a)(b) studied the formation of pyromorphite from combinations of different Pb solids and apatite suspensions under varying pH conditions. They found that the dissolution of apatite or dissolution of both apatite and Pb bearing solids was the rate-limiting step of the reaction involving formation of pyromorphites. This suggests that acidification prior to P addition might enhance the formation of pyromorphite in soils.
The objectives of this study were to evaluate the influence of P amendment on soil Pb bioavailability in five metal-contaminated soils and mine spoils, and to determine the effects of time, level of P addition, P source, and preacidification on soil lead bioavailability. In addition, mineralogical changes produced by the soil amendments were determined using X-ray diffraction.
|
MATERIALS AND METHODS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Methods
Soil pH was determined in a 1:1 soil–deionized water mixture with a ROSS combination pH electrode (Thermo Orion, Beverly, MA). Organic carbon (OC) content was determined by the Walkley–Black procedure (Jackson, 1958). Sand, silt, and clay contents were determined by the pipette method as described by Gee and Bauder (1986) and Bray-1 extractable P by the method of Bray and Kurtz (1945). For total metal concentrations, 2 g of material (
2-mm or 250-µm fractions) were digested with 20 mL 4 M HNO3 (trace metal grade) acid at 80 to 85°C for 4 h. Filtered digestions then were analyzed for Pb, Cd, and Zn by inductively coupled plasma–atomic emission spectrometry (ICP–AES).
Treatments
Seven treatments in triplicate were evaluated as follows: no P (control); 2500 mg P/kg soil as TSP fertilizer (TSP2500), PR (PR2500), or phosphoric acid (PA2500); preacidification to pH 5.0 with acetic acid for 24 h followed by addition of 2500 mg P/kg soil as TSP fertilizer (acetic); and 5000 mg P/kg soil as TSP fertilizer (TSP5000) or PR (PR5000). Triple superphosphate is a common agricultural fertilizer made by reacting PR with orthophosphoric acid. The principal ingredient is monocalcium orthophosphate [Ca(H2PO4)2 · H2O]. The PR was from Occidental Corp. (White Springs, FL) and previously was shown to be effective for immobilizing Pb in soil (Ma et al., 1995). The amount of P required for each treatment was calculated on the basis of the total P content of the P source. Treatments were evenly applied to the soil and thoroughly mixed. Deionized water was added to bring the samples to a gravimetric water content (
m) of 20%; soils were thoroughly mixed for a second time. Twenty-four hours after P additions, predetermined amounts of CaO were added to samples as needed to increase the soil pH to 7.0 to 7.5. Samples were kept in plastic containers covered with Parafilm "M" (American National Can, Menasha, WI) to allow air exchange while minimizing the moisture loss, and incubated at 20% m at 25°C. Once a month, deionized water was added to maintain the samples at 20% m. Subsamples of approximately 100 g were taken from the containers at 3, 28, 84, 252, and 365 d. They were air-dried and analyzed for pH (1:1 soil to deionized water), and Bray-1 P (Bray and Kurtz, 1945). Materials were sieved through a stainless steel sieve with an opening of 250 µm before determining bioavailable Pb and performing the dry particle size separation procedure for X-ray diffraction (XRD) analyses.
Physiologically Based Extraction Test Procedure
Bioavailable Pb was determined by a modified PBET (Ruby et al., 1996). The gastric solution was prepared by adding 5 g of pepsin (activity of 800–2500 units/mg), 2 g of anhydrous citric acid, 2 g of DL-malic acid, 1.68 mL of DL-lactic acid (Sigma, St. Louis, MO), and 2 mL of glacial acetic acid (Fisher Scientific, Pittsburgh, PA) to 4 L of deionized water. Variable amounts of trace-metal grade, concentrated HCl were added to the gastric solution to ensure a pH of 2.0. One hundred milliliters of gastric solution prewarmed to 37°C was combined with 1 g of test material (250-µm fraction) in a 125-mL wide-mouth HDPE bottle that was covered with a cap containing a rubber septum. The head space was replaced with argon gas, and the bottles were shaken for 1 h in a chamber maintained at 37°C. The pH was maintained at 2.00 ± 0.2 throughout this period. After 1 h, a 10-mL aliquot of gastric solution was removed for analysis using a 10-mL disposable syringe attached with a cellulose acetate membrane filter. This aliquot was replaced by 10 mL of fresh gastric solution at 37°C. The contents of the bottle were brought to pH 
6.5 by adding a 15-cm length of dialysis tubing (100000 MWCO, Spectra/Por cellulose ester membrane tubing; Spectrum Laboratories, Rancho Dominguez, CA) containing 2.5 g of NaHCO3. The tubing was removed after 30 min, and 0.175 g of bile extract (Porcine, 8008-63-7; Sigma) and 0.05 g of pancreatin (Porcine pancreas-4X 8049-47-6; Sigma) were added. The cap was replaced and the head space was replaced again with argon. The sample was shaken for another 1 h at 37°C. Another 10 mL of aliquot was removed for analysis as before. A drop of concentrated HNO3 (trace-metal grade) was added to each aliquot to prevent precipitation of metals. Extracts were stored at 4°C until analysis using ICP–AES. Quality assurance–quality control consisted of 15% of samples run as duplicates, check samples, or blanks for each extraction. Extractions were done in batches of eight with seven samples plus a QA–QC sample to reduce random error. Bioavailable Pb is expressed as a percentage of bioavailable Pb in the control sample.
Mineralogical Analyses
Air-dried samples (250-µm fraction) from 28, 84, and 365 d were sieved using an ATM sonic sifter (ATM Corporation, Milwaukee, WI) to an effective diameter of 10 µm before XRD analyses. The analyses were conducted with a Philips X-ray diffractometer (Philips Electronic Instruments, Mahwah, NJ) using Cu K
radiation at a wavelength of 1.54 Å. The potential was 35 kV, and the amperage 20 mA. Measurements were made using a continuous scanning technique from 2 to 60° 2
and a scan-speed of 2° 2 per minute.
Data Analyses
The experimental design was a split plot design with three replications. The main plot factor was treatment and was arranged in a randomized complete block. Sampling time was the subplot factor. Statistical analyses were performed using SAS for Windows Version 6.12 (SAS Institute, 1985). For mean separations, least significant difference (LSD) values were used.
|
RESULTS AND DISCUSSION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
250-µm fraction of the three soils ranged from 1521 to 3022 mg/kg, and was lower than those in the 2-mm fraction, except for total Pb in AR. In contrast, the mine waste materials always had higher concentrations of Pb in the 250-µm fraction compared with the 2-mm fraction (2310 and 12414 mg/kg for Chat and Dearing materials, respectively).
Effects of Treatments on Soil pH and Available Soil Phosphorus
The acetic, PA2500, and TSP5000 treatments reduced soil pH to 5.2 ± 0.3 in all materials except Chat. The addition of TSP2500 reduced soil pH to 5.7 ± 0.2 in all materials, whereas PR had no effect on soil pH even at the highest level. Calcium oxide additions increased soil pH to near 7.0 in most samples. The samples that received TSP5000 had the lowest pH of all materials tested (6.8), indicating that the CaO added was not sufficient to neutralize the acidity created by the dissolution of TSP.
Plant-available P in soils can be estimated by one of several methods depending on soil characteristics (Kuo, 1996). Bray-1 extractable P generally is used for acidic to near-neutral soils. Soils treated with TSP (TSP2500, TSP5000, and acetic) and PA2500 had levels of Bray-1 P well above those required for normal plant growth (Table 2). However, PR2500 and PR5000 did not change Bray 1-P levels significantly, indicating the relative insolubility of PR compared with the other P sources. Although excessive levels of plant-available P are not harmful to plants, they have been correlated with higher P losses in runoff. Thus, PR may present less of an environmental risk from enhanced eutrophication compared with the other P sources.
|
In general, the reproducibility of bioavailable Pb values was higher in the stomach phase than in the intestinal phase. Ruby et al. (1996) also reported poor reproducibility for the intestinal data. They preferred the stomach phase as a measure of Pb bioaccessibility and reported correlations between their in vivo (Sprague–Dawley rat) and in vitro test results of r2 = 0.93 for the stomach phase (n = 7) and r2 = 0.76 for the intestinal phase (n = 7). Medlin (1997) used a revised version of the PBET procedure, similar to that used in this study, and found that data from the stomach phase of the test correlated well (r2 = 0.85) with in vivo data using young swine for the stomach phase (n = 15). Moreover, Medlin (1997) calculated the relative percent difference (RPD) of duplicate sample pairs by dividing the difference between duplicate measurements by the mean. From the 98 duplicate sample pairs examined, the average RPD for the stomach phase was 12.2%. Average coefficients of variance in the current study were 6 ± 3% for the stomach phase and 14 ± 6% for the intestinal phase. Oliver et al. (1999) also reported acceptable reproducibility for the intestinal phase using the PBET procedure as described by Ruby et al. (1996).
Trends for all three soil material were similar, and, therefore, only data for the Joplin material are presented. Time was included as a factor in the statistical analysis of the in vitro results. The interactions between time and treatments for each soil were not significant. Therefore, data are reported as averaged over sampling times. Without exception, all soil amendments reduced bioavailable Pb in the stomach phase compared with the control (Fig. 1–3). Similar results were found with the intestinal phase.
|
|
|
The PR5000 treatment was the most effective for the AR, Chat, and Joplin materials, whereas the PR5000 and TSP5000 treatments were equally and most effective for the TCR material. The addition of P did produce larger reductions in bioavailable Pb for Chat compared with any other material, even though the relative treatment differences were similar to those observed for Joplin, TCR, and AR (Fig. 3). In all cases, increasing the amount of P added from 2500 to 5000 mg/kg resulted in a significantly greater reduction in bioavailable P (Fig. 1–3). Brown et al. (1999) conducted a field study on the same vacant lot in Joplin, MO that served as the source of our Joplin sample. They found that 10000 mg of P/kg soil as TSP was not effective in reducing either in vitro Pb or Pb uptake by tall fescue (Festuca arundinacea Schreb.), whereas TSP at 32000 mg P/kg soil was the most effective treatment. The authors suggested that adding TSP at 10000 mg/kg did not lower soil pH enough to cause significant reductions in either bioavailable Pb or plant Pb uptake.
For all materials except Dearing, PR was equally or more effective than TSP or H3PO4 in reducing bioavailable Pb. The dissolution of PR and subsequent formation of pyromorphite can be expressed as follows:
 | [1] |
 | [2] |
The effects of time on treatment efficacy were of interest in order to test the long-term viability of this remediation approach. For example, a more complete crystallization of newly formed pyromorphite could result in further reductions in Pb bioavailability. Conversely, changes in soil pH or other factors could act to increase Pb bioavailability over time. In this study, Pb bioavailability was influenced little by time, and results were similar for all materials. The lack of a time effect suggests that either the reactions between soil Pb and P occurred within the first 3 d of the incubation and changed little thereafter, or that the reactions between soil Pb and P that occurred in the stomach phase solution were not influenced by the contact time for soil Pb and P.
Mineralogical Analyses
Direct identification of mineralogical changes resulting from P addition is possible with physical methods. Unfortunately, only crystalline materials present at concentrations 
10 to 20 g/Kg can be detected (Ma et al., 1994a), and direct identification of solid forms of many elements is not always possible with XRD. Particle size separation could increase the possibility of direct identification of Pb solids using XRD, if newly formed Pb phosphates were present at higher concentrations in the smaller size fractions compared with whole soil. However, the use of traditional procedures for particle size separation involving sedimentation in water may allow reactions between Pb and P to occur that would not have occurred in the soil. We attempted to overcome this problem by using a dry separation procedure to separate a smaller sized, and presumably Pb-rich, fraction. Our attempt was successful for Dearing material, which contained the highest amount of total Pb.
X-ray diffraction patterns for the Dearing material are shown in Fig. 4. Of the three most prominent peaks of hydroxypyromorphite [Pb5(PO4)3OH, HP], two can be seen without interference from quartz (2.85 and 2.97 Å). Both of these peaks also were present in the control, indicating that the materials contained pyromorphite prior to P amendment. This was not surprising, because Nriagu (1974), Cotter-Howells and Thornton (1991), and Cotter-Howells et al. (1994) have observed pyromorphite as a common weathering product of Pb in mine waste materials and roadside soils. The addition of P from the soluble P sources increased the intensity of the 2.85 Å peak, indicating that more pyromorphite formed after P addition. For the Dearing sample, the phosphoric acid–treated samples (PA2500) had the most intense pyromorphite peak and showed the greatest reduction in bioavailable Pb by PBET (Fig. 1). Similarly, the significantly lower Bray-1 extractable P concentrations for the Dearing sample compared with similar amounts of P from TSP (TSP2500 and acetic, Table 2) provided indirect evidence of the most extensive reaction between soil Pb and P.
|
The reduced intensity of the FA peaks in PR-treated Dearing material with time might be an indication of further dissolution of PR (Fig. 5). The presence of metals can induce further dissolution of PR by removal of soluble phosphates from soil solution through precipitation reactions. Moreover, reduced FA peaks also could indicate increased sorption of metals onto PR. Increased sorption of metals onto mineral surfaces can be responsible for reduced intensity of XRD peaks because of the increased mass absorption coefficient of X-rays upon adsorption of metals (McKenzie, 1980). Strong adsorption of Pb onto apatite surfaces or simultaneous adsorption of Pb and surface precipitation of Pb phosphate on the apatite surfaces also might be possible in PR-treated soils (Ma et al., 1994a). Regardless, no indications of further reductions in Pb bioavailability with time were observed. This might have been due to the dissolution of any unreacted PR in the stomach phase solution. In addition, the peak at 2.79 Å developed two distinctive shoulders at 2.76 and 2.78 Å. This could be an indication of coprecipitation of apatite with other metals, which can be expected in the presence of metals like Pb, Cd, and Zn (Nriagu, 1984; Bigi et al., 1991; Ma et al., 1994a).
|
|
CONCLUSIONS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Both PBET data and XRD data suggested that phosphoric acid was the most effective P source for the Dearing material. Similarly, preacidification prior to P addition was significantly more effective in reducing bioavailable Pb compared with the same level of P addition without preacidification only for the Dearing material, indicating that acidification prior to or with P addition may reduce Pb bioavailability further in some materials. Moreover, the reductions in Pb bioavailability occurred between 0 and 3 d after treatment addition, and Pb bioavailability did not appear to change with time.
The greatest reductions in Pb bioavailability observed in the stomach phase ranged from 25 to 38% compared with the corresponding control soil, which may not be sufficient for some contaminated materials. Therefore, additional improvements to this method may be necessary before this technology is adopted for remediation of Pb-contaminated soils.
|
NOTES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
REFERENCES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. O. Mbila and M. L. Thompson Plant-Available Zinc and Lead in Mine Spoils and Soils at the Mines of Spain, Iowa J. Environ. Qual., March 1, 2004; 33(2): 553 - 558. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. M. Lanfranco, A. M. Lanfranco, P. F. Schofield, P. J. Murphy, M. E. Hodson, J. F. W. Mosselmans, and E. Valsami-Jones Characterization and identification of mixed-metal phosphates in soils: the application of Raman spectroscopy Mineralogical Magazine, December 1, 2003; 67(6): 1299 - 1316. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. F. Jaynes, R. E. Zartman, R. E. Sosebee, and D. B. Wester Biosolids Decomposition after Surface Applications in West Texas J. Environ. Qual., September 1, 2003; 32(5): 1773 - 1781. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. M. Hettiarachchi, G. M. Pierzynski, F. W. Oehme, O. Sonmez, and J. A. Ryan Treatment of Contaminated Soil with Phosphorus and Manganese Oxide Reduces Lead Absorption by Sprague-Dawley Rats J. Environ. Qual., July 1, 2003; 32(4): 1335 - 1345. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. M. Hettiarachchi and G. M. Pierzynski In Situ Stabilization of Soil Lead Using Phosphorus and Manganese Oxide: Influence of Plant Growth J. Environ. Qual., March 1, 2002; 31(2): 564 - 572. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
