|
|
||||||||
a QSAR Risk Assessment Service Inc., 360 St-Jacques W., Suite 800, Montréal, QC, Canada H1Y 2P5
b Dep. of Soil, Crop, and Atmospheric Sci,, Bradfield Hall, Cornell Univ., Ithaca, NY 14853 USA
c Dep. of Natural Resources Sci., McGill Univ., Macdonald Campus, Ste-Anne-de-Bellevue, QC, Canada H9X 3V9
sebastien.sauve{at}inrs-sante.uquebec.ca
| ABSTRACT |
|---|
|
|
|---|
Abbreviations: DPASV, differential pulse anodic stripping voltammetry EC, electrolytic conductivity FTIR, Fourier transform infrared spectra XRD, x-ray diffraction
| INTRODUCTION |
|---|
|
|
|---|
Iron oxides occur in soils in varying concentrations, depending on the pedogenetic processes. The high affinity of Fe oxides for certain trace elements, Pb in particular, makes them a likely long-term sink for cationic polyvalent metals. Since a large part of trace elements, measured as soil total by acid digestion of the solid phase, is not bioavailable, it has been argued that elevated concentrations of sparingly soluble metals such as Pb do not pose a threat to the environment (Cook and Hendershot, 1996). From a bioavailability perspective, it is critical to improve our understanding of natural systems to be able to predict under which circumstances metals will be dissolved and possibly bioavailable, relative to situations where the metals are strongly bound to the solid phase and represent only a limited potential for toxicity.
Most studies on the metal adsorption properties of Fe oxides use laboratory-synthesized, well characterized materials. The actual adsorption properties of pedogenic oxides may differ from those of pure materials. Soil oxides are heterogeneous, albeit predominantly made of Fe. They are of mixed composition and often amorphous, as the presence of silica and organic matter hinders the crystallization process (Schwertmann and Taylor, 1989). The adsorption properties of naturally occurring pedogenic soil oxides must be compared with laboratory-synthesized Fe oxides before assuming that the results of adsorption studies using pure minerals can be used to predict the environmental fate of metals.
Soil organic matter has a pronounced impact on the solubility and soil adsorption of metals. Prior research suggests that soil organic matter can either enhance or inhibit adsorption, depending on other soil properties (pH, cation-exchange capacity, competing ions, etc.) (Gerritse and van Driel, 1984; Basta et al., 1993; Harter and Naidu, 1995; Römkens and Dolfing, 1998). Soil humic acids can increase adsorption, reducing both metal concentration and free metal activity. On the other hand, soil organic matter can increase dissolved organic matter and fulvic acid concentrations, which increase the total dissolved metals via complexation reactions in the soil solution, resulting in higher metal mobility (Bruemmer et al., 1986; McBride et al., 1997a; Sauvé et al., 1998b). Therefore, it is important to quantify the metal-adsorption capacity of soil organic matter relative to Fe oxides; it is also necessary to assess the extent that soil organic matter affects the adsorption properties of pedogenic Fe oxides.
Increasing pH reduces the solubility of most Pb-bearing minerals, while increasing the adsorption affinity of iron oxides, organic matter, and other adsorptive surfaces. However, increasing pH also increases Pb hydrolysis, inorganic ion-pair formation, and organic matter solubility, promoting higher dissolved concentrations of Pb (Bruemmer et al., 1986; Sauvé et al., 1998b, 1998c). It is important to identify which soil components are responsible for controlling Pb concentrations in the aqueous phase and the conditions under which solubility and chemical speciation are optimum to minimize Pb bioavailability. Furthermore, although the total dissolved concentrations of Pb are important for certain aspects of environmental soil modeling (Sauvé et al., 1998c; Bruemmer et al., 1986), bioavailability and chemical reactivity may be more directly related to the free metal pool (Sauvé et al., 1998a; Knight et al., 1998; Knight and McGrath, 1995).
The objective of this study was to determine the pH-dependent Pb-adsorption properties of two pedogenic oxides compared with synthesized ferrihydrite and a leaf compost, thus exploring the relative contributions of soil oxides in relation to leaf compost in controlling free Pb2+ activity in contaminated soils.
| Materials and methods |
|---|
|
|
|---|
4°C for 6 mo before use (as slurries of 7.9 and 19% solids). A sugar maple (Acer saccharum L.) leaf compost was collected from a rural location near Ithaca, NY, air-dried, ground in a blender, and sieved to obtain the size fraction <1 mm (used in prior research, Sauvé et al., 1998c). We preferred leaf compost to reagent-purified standardized humic materials because we were interested in characterizing natural materials and contrasting their properties to laboratory-synthesized products. Synthetic ferrihydrite was prepared by slow (0.03 mL min-1) precipitation to a pH 6 of a 4.95 mM Fe(NO3)3 solution with 1 mM KOH (further details in Martínez and McBride, 1998).
Characterization of Solid Phases
Basic chemical properties of the materials used can be found in Table 1
. Elemental composition of the samples was determined by digestion in concentrated HNO3 (Mench et al., 1994) and analysis by inductively coupled plasma emission spectrometry. The impurities in the synthesized ferrihydrite come at least partially from the reagents used. The organic C and N content was measured using a combustion method. Specific surface was determined by three-point N2 adsorption measurements after a 1-h outgas procedure with dry N2 at 100°C. Fourier transform infrared spectra (FTIR) were obtained using a pellet containing 2 mg of sample mixed with 170 mg of KBr. Measurements were made using a Fourier-transform Perkin-Elmer 1720-X spectrometer (Perkin-Elmer, Buckinghamshire, UK) at 2 cm-1 resolution and 100 co-added scans. X-ray diffraction (XRD) analyses were made using a theta-theta diffractometer (Scintag, Cuttertino, CA) with a solid-state intrinsic germanium detector. Specific surface, FTIR, and XRD measurements showed that the adsorbents vary in degree of crystallinity (Martínez et al., 1999). The synthetic ferrihydrite had a specific surface that was typical of a non-crystalline oxide material (170 m2 g-1); SMS-1 had an intermediate specific surface (97.9 m2 g-1); and SMS-2 had a specific surface that was consistent with that of a crystalline material (26.4 m2 g-1). These results agree with the XRD patterns for each adsorbent. The x-ray diffraction patterns showed that a two-line ferrihydrite was synthesized in the laboratory. The XRD pattern of SMS-1 also shows two broad lines, consistent with the presence of ferrihydrite, and there was no evidence of the presence of well-crystallized iron oxides. For SMS-2, the XRD pattern showed some quartz and feldspar; however, there was no indication of the presence of non-crystalline (ferrihydrite) or crystalline (hematite, goethite, or lepidocrocite) Fe oxides. Infrared analysis indicated that a combination of various Fe oxides was present in the samples. Hematite-like (in ferrihydrite) and lepidocrocite-like (in SMS-1 and SMS-2) microcrystalline structures were present, but they were not ordered well enough to produce a sharp XRD pattern. The FTIR spectra of the leaf compost was dominated by two strong absorption bands (typical of organic matter): (i) the OH stretching band, and (ii) a very intense band at
1640 nm, indicating a high COOH content and thus a high metal-binding capacity (Stevenson, 1994). The XRD and FTIR spectra are presented and discussed in more detail in Martínez et al. (1999).
|
The spiked materials were then shaken every other day for 15 d. The bottles were kept loosely covered (to prevent dust deposition) but not closed, to allow gas exchange and keep the solutions aerobic and in equilibrium with ambient CO2. Then, the tubes were centrifuged at 15000 x g to separate the supernatants. The pH and electrolytic conductivity were determined in the supernatants before filtration. The pH was measured using a Fisher 805MP meter (Fisher Scientific, Pittsburgh, PA) and a combination glass electrode (Orion model 91-55, Orion Research, Boston, MA). Electrolytic conductivity (EC) was measured using a YSI model 31 meter (YSI, Yellow Springs, OH). Electrolytic conductivity was also used to estimate the ionic strength (IS), assuming
(where IS in unitless and EC is in mS cm-1) (Griffin and Jurinak, 1973). The solutions were then passed through 0.22-µm cellulosic membranes and analyzed for labile Pb using differential pulse anodic stripping voltammetry (DPASV) (Florence, 1986; Opydo, 1989; Sauvé et al., 1998b, 1998c). For the DPASV analysis, a hanging mercury drop electrode was used. The samples were prepurged for 8 min with N2 (gas) to remove dissolved oxygen. We used a -0.8 V reducing potential deposition step of 1 min (with stirring) followed by a 30-s homogenization period (without stirring). We then carried out metal stripping using current ranges (sensitivity) between 0.1 µA and 0.5 mA, as necessary. The DPASV determinations allow the discrimination of organically bound Pb from the inorganic labile Pb (Florence, 1986). A chemical equilibrium model (MINEQL+, Schecher and McAvoy, 1994), combined with the constants compiled in Sauvé et al. (1998b), can then be used to partition the measured DPASV-labile Pb into those inorganic ion-pairs calculated to be most prevalent in the solution (namely PbOH+, Pb
02, Pb
+3, PbHCO-3, PbCO03, Pb
2-2, PbNO+3, PbCl-, and PbSO04. An estimate of free Pb2+ is obtained by taking the difference between DPASV and the sum of inorganic chemical species (Sauvé et al., 1998b). Actual free Pb2+ may be lower than this estimate if significant levels of ASV-labile organo-Pb complexes exist in solution (Florence, 1986). Thus in organic-rich systems, this method might overestimate free Pb2+ activity, although a comparison with other free Pb2+ speciation methods showed that ASV gave the lowest, most restrictive free metal activity results (Sauvé et al., 1997a; Sauvé, 1999).
| Results and discussion |
|---|
|
|
|---|
106, the difference is negligible. Most of the variability in the solution free Pb2+ activity is explained by pH and total Pb content. Thus, both adsorption and free Pb2+ activity are related directly to pH and metal loading. The data were fit to a semi-mechanistic model of the following form:
![]() | (1) |
|
and the leaf compost
. Ferrihydrite represents an inorganic material as opposed to the leaf compost, which is mostly organic. The coefficients for the two pedogenic oxides are intermediate between ferrihydrite and leaf compost. They apparently behave as mixtures of the inorganic and organic materials (Table 2), showing a more curvilinear relationship (Fig. 1) that reduces the degree of fit to the linear regression model (Eq. [1], Table 2). Adding a quadratic pH component (+c pH2) improves the R2 of the regressions for SMS-1 and SMS-2 to >0.85. However, adding more and more parameters also makes the model a simple curve-fitting exercise, whereas Eq. [1] is simple and best explains the variability.
|
The adsorption properties of the various materials can be compared more directly using the data in Table 2 to calculate the expected Pb2+ activities over the same range of pH and fixed total Pb loads (Fig. 2) . Figure 2 is simply an illustration of the numerical consequence of regressions given in Table 2 (experimental data shown in Fig. 1). Figure 2 suggests that Pb-binding by ferrihydrite, as a surrogate for laboratory-synthesized amorphous Fe oxides, overestimates Pb binding and consequently underestimates equilibrium Pb2+ activity by one to six orders of magnitude relative to the sorption properties of soils (see Fig. 2). Even when comparing ferrihydrite with the pedogenic oxides, it is clear that these naturally occurring materials have a lower affinity for Pb. The discrepancy between adsorption by synthetic and pedogenic oxides is lowest at low pH and increases with pH. Similarly, the discrepancy is most marked at low total Pb loadings and decreases with higher loadings (remembering that the lowest Pb levels are more realistic). At the highest loading illustrated, the amorphous pedogenic oxide (SMS-1) is similar to ferrihydrite, whereas the more crystalline pedogenic Fe oxide (SMS-2) shows adsorption properties similar to soils (Fig. 2). The leaf compost supports a free Pb2+ activity in solution similar to that in soils, and much higher than that resulting from equilibrium with the pedogenic and synthetic oxides.
|
Data in Fig. 2 suggest that increasing organic matter content and the heterogeneity of the pedogenic oxide tend to reduce the Pb adsorption affinity of the material studied (in increasing order of adsorption affinity: ferrihydrite > pedogenic oxides > soils > organic matter). This is somewhat surprising considering that organic matter has a high affinity for metals, including Pb (Stevenson, 1976; Logan et al., 1997). We do not believe this is because Pb has a low affinity for organic matter, but rather because Pb has a much higher affinity for certain mineral surfaces (Fe oxides and phosphates in particular) relative to organic matter. Therefore, organic matter may block the reactive sites on Fe-oxides and indirectly increase Pb solubility. Furthermore, the formation of dissolved organo-Pb complexes will increase the concentration of Pb in solution and prevent its adsorption.
| Conclusions |
|---|
|
|
|---|
| ACKNOWLEDGMENTS |
|---|
Received for publication November 9, 1998.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
J. M. Kaste, B. C. Bostick, A. J. Friedland, A. W. Schroth, and T. G. Siccama Fate and Speciation of Gasoline-Derived Lead in Organic Horizons of the Northeastern USA Soil Sci. Soc. Am. J., August 22, 2006; 70(5): 1688 - 1698. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Levonmaki, H. Hartikainen, and T. Kairesalo Effect of Organic Amendment and Plant Roots on the Solubility and Mobilization of Lead in Soils at a Shooting Range J. Environ. Qual., May 31, 2006; 35(4): 1026 - 1031. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Appel, L. Q. Ma, R. D. Rhue, and W. Reve Selectivities of Potassium-Calcium and Potassium-Lead Exchange in Two Tropical Soils Soil Sci. Soc. Am. J., November 1, 2003; 67(6): 1707 - 1714. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Brown, R. L. Chaney, J. G. Hallfrisch, and Q. Xue Effect of Biosolids Processing on Lead Bioavailability in an Urban Soil J. Environ. Qual., January 1, 2003; 32(1): 100 - 108. [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 |
Journal of Environmental Quality |
||||