Soil Science Society of America Journal 66:1198-1206 (2002)
© 2002 Soil Science Society of America
DIVISION S-2—SOIL CHEMISTRY
Desorption Kinetics of Yttrium, Lanthanum, and Cerium from Soils
Bei Wen,
Xiao-quan Shan*,
Jin-ming Lin,
Gui-gang Tang,
Nai-bin Bai and
Dong-an Yuan
Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences, P.O. Box 2871, Beijing 100085, China
* Corresponding author (xiaoquan{at}mail.rcees.ac.cn)
 |
ABSTRACT
|
|---|
Study on desorption kinetics of Y, La, and Ce from soils is of importance because it relates to the bioavailability and potential toxicity of rare-earth elements. In the present study, a column-flow method and three models (first-order, two site first-order, and log-normal distribution first-order kinetics models) were used to describe the desorption kinetics of Y, La, and Ce from four Chinese soils with different physicochemical properties. A high desorption percentage of Y (87.1–96.6%), La (89.9–98.5%), and Ce (57.6–96.4%) from Yingtan soil was attributed to the low soil pH 5.43 and low organic matter of 1.53%. In contrast, a low percentage of Y (27.5–45.7%), La (27.6–53.6%), and Ce (1.09–50.8%) sorbed by Beijing, Tongjiang, and Haerbin soils desorbed probably because of the higher soil pH values of 8.24, 7.16, 7.23, and increased organic matter (36.4%) in Haerbin soil. The results also suggest that the first-order kinetics model did not offer an acceptable description of the data (R2 < 0.90). However, excellent agreements were achieved between the experimental data and fits to the latter two kinetic models (R2 > 0.99). The parameters derived from the kinetic equations indicated that increasing the initial sorption period from 1 to 20 wk could lead to a strong binding of rare-earth elements, resulting in slower desorption.
Abbreviations: CEC, cation-exchange capacity
|
INTRODUCTION
|
|---|
WITH THE WIDESPREAD APPLICATION of rare-earth elements in industry, agriculture, forest, and animal husbandry, more and more rare-earth elements enter into the surface terrestrial and food chain. The forecast in the next few decades is for increasing agricultural usage of rare-earth elements. However, an improper use of rare-earth elements might cause environmental contamination (Volokh et al., 1990; Mermut et al., 1996). In China, millions of tons of fertilizers containing rare-earth elements were used in agriculture to increase crop yields. However, exogenous rare-earth elements and their possible long-term hazardous environmental effects are of concern although no acute toxicity of rare-earth elements is observed (Brown et al., 1990). This work is trying to reduce uncertainty to some extent.
Along with the increased interest in the physiological effect of rare-earth elements, research of rare-earth element chemistry in soils has also accelerated. Jones (1997) reported that adsorption of La, Y, Pr, and Gd depended on soil pH and cation-exchange capacity. The adsorption appeared to conform well to the single Langmuir equation. Equilibrium release experiments (Cao et al., 2001) demonstrated that the release of La, Ce, Gd, and Y increased with decreasing pH or EH. The release of rare-earth elements were correlated with the release of Fe and Mn, that the release of rare-earth elements might originate from dissolution of Fe-Mn oxyhydroxides under reduced and low soil pH conditions (Li et al., 1998). The adsorption and desorption of rare-earth elements by individual components of soil were affected by soil pH, demonstrating that adsorption increased with increasing soil pH (Ran and Liu, 1992, 1993). Li et al. (2001) studied the kinetics of adsorption and desorption of Ce(III) on soil using a batch method and isotope 141Ce. It was indicated that the Elovich equation proved to fit the data on desorption of Ce (III) from fluvoaquic and black soils well, while the parabolic-diffusion equation were the best models for red earth and loess soils.
A first-order kinetic model was successfully applied to deal with anion and cation desorption from kaolinite and soils (Selim et al., 1976; Bar-Yosef and Kafkafi, 1978), but failed to describe the dissociation of heavy metal-humic acid complexes (Rate and McLaren, 1992). However, Rate and McLaren (1992) used a log-normal distribution first-order kinetic model to describe the dissociation data successfully. It is also recognized that sorption-desorption processes are found to be characterized by a rapid reversible stage followed by a much slower, nonreversible stage (Sparks, 1995). It seems that both the two-site first-order and log-normal distribution first-order models described the desorption kinetics of Cd and Co from Fe and Mn oxides and clay well (Backes et al., 1995; McLaren et al., 1998). To our knowledge there are no reports on the desorption of rare-earth elements using two-site first-order and log-normal distribution first-order kinetics models in the literature.
A knowledge of desorption behavior of rare-earth elements from soil allows one to predict transformation, bioavailability, and mobilization of rare-earth elements in soil, thus it helps to predict the long-term effect on the environment. Therefore, the aim of this study was to investigate the desorption kinetics of rare-earth elements from four Chinese soils with different physicochemical properties using a continuous column-flow technique and three kinetics models, first-order, two-site first-order, and log-normal distribution first-order kinetics. In addition, the effect of the sorption period of rare-earth elements in soils on the subsequent desorption was also investigated to improve our knowledge of the fate of rare-earth elements in soils.
|
MATERIALS AND METHODS
|
|---|
Soils
Four Chinese cultivated soils were collected from Jiangxi Province, southern China (Yingtan soil); Beijing, northern China (Beijing soil); and Heilongjiang Province, northeastern China (Tongjiang and Haerbin soils). All the soil samples were taken from the surface layer (0–20 cm) of cultivated soils. The soils were air dried, ground, and screened through a 1-mm nylon fiber sieve to remove stone, plant root, and other large particles.
Soils were classified according to the American Classification System. Soil pH was measured in deionized water using a 1:1 (w:v) soil/solution ratio. Organic matter was determined by the Walkley-Black procedure (Nelson and Sommers, 1982). Cation-exchange capacity (CEC) was determined by the method described by Rhoades (1982). Amorphous Fe and Al oxides were determined by ammonium-oxalate extraction (Blakemore et al., 1987). Crystalline Fe and Al oxides were determined by the oxalate-ascorbic acid extraction method of Shuman (1982). Manganese oxide content was determined by extraction with 0.1 mol L-1 hydroxyl-ammonium hydrochloride (Shuman 1982). The selected properties of the soils are given in Table 1.
Sorption-Desorption Procedure
A batch technique was used in the sorption procedure. Samples of soil (50.0 mg) were weighed into 25-mL polypropylene vials, and 10 mL of Y, La, or Ce solutions were added as nitrate solution to give a final concentrations of 1.0 mmol L-1 in a background electrolyte solution of 10 mmol L-1 Ca(NO3)2. The Ca(NO3)2 was used to inhibit outer-sphere adsorption so that adsorption of rare-earth cations was specific in nature (Sauve et al., 2000; Romkens and Dolfing, 1998; Backes et al., 1995). The pH values of the equilibrium solutions were adjusted to 6.0 by adding a small amount of Ca(OH)2 solution. The samples were shaken at 20°C for 24 h, and then incubated for either 1 or 20 wk. After incubation, the samples were filtered through a weighed filter-holder with a 0.45-µm Millipore cellulose ester membrane (Millipore Corp., Milford, MA). Filtrate pH values were measured. To determine the weight (volume) of solution remaining entrained with the soil on the membrane, the filter holder plus the wet soil was weighed again after filtration. The concentrations of Y, La, and Ce in the filtrates were determined as described below, so that both the amounts of Y, La, and Ce sorbed and the amounts entrained could be determined.
Desorption was conducted by continuous peristaltic pumping of 10 mmol L-1 Ca(NO3)2 through the soil layer on the filter at a flow rate of 2.0 mL min-1. The pH of Ca(NO3)2 solution was adjusted to the value of the final pH of the equilibrium solution, and the whole desorption procedure was performed at 20°C for about 5 h. Elute fractions were collected every 180 s using a fraction collector. The amounts of entrained rare-earth elements were subsequently subtracted from the amounts present in the first-desorption fraction. The concentrations of Y, La, and Ce were measured in all the fractions from each desorption run. Each experimental treatment was run in duplicate.
Yttrium, Lanthanum, and Cerium Determination
Yttrium, La, and Ce concentrations were determined by inductively coupled plasma mass spectrometry (VG Elemental, Winsford, UK). 115Indium was added as an internal standard to monitor matrix effects and signal drifts (Wen et al., 1999).
Speciation Analysis of Rare-Earth Elements in Soil
A portion of 50.0 g of Yingtan, Beijing, and Haerbin soils were weighed into a 100-mL beaker, and 50 mL of La and Ce at the levels of 10 and 20 µg mL-1, respectively, were added as their nitrate solution. The mixture was homogenized occasionally during the incubation period of 20 wk. After incubation the soils were air dried and ground to pass through a 1-mm sieve for speciation analysis.
Rare-earth elements exogenuously added to soil were fractionated as water soluble, exchangeable and carbonate bound (B1), Fe-Mn oxide bound (B2), organic matter, and sulfide bound (B3) by a three-stage procedure (Quevauviller, 1993). The detailed procedure is listed in Table 2.
Desorption Kinetics Models
Three models, first-order, two-site first order, and log-normal distribution first-order kinetics models were used to describe the desorption of Y, La, and Ce from soils. The mathematical expressions are as follows:
First-order kinetic model (Sivasubramaniam and Talibudeen, 1972).
 | [1] |
where CMS is the concentration of Y, La, or Ce sorbed by soil at time t; C0 is the initial concentrations (t = 0) of Y, La, and Ce bound to soil; t is time in seconds; k'd is apparent desorption time-rate coefficient in units of per second. Fitting the experimental data was achieved by optimizing the values of k'd.
Two-site first-order kinetic model (Willis et al., 1970).
Desorption of rare-earth element ions takes place from soils, which have two discrete, independent sites for rare-earth ion binding. Simultaneous first-order desorption for this system may be expressed mathematically as:
 | [2] |
where CMS is the concentration of Y, La, or Ce sorbed to the soil at time t; C1 and C2 are the initial concentrations (t = 0) of Y, La, and Ce bound to sites with first-order desorption rate constants k1 and k2, respectively. It was assumed that all Y, La, and Ce would eventually desorb, C1 + C2 equals Y, La and Ce sorbed (Tables 3–5). Fitting to experimental data was achieved by optimizing values of C1, k1, and k2.
View this table:
[in this window]
[in a new window]
|
Table 5. Lanthanum sorbed by soils prior to desorption and proportions desorbed after 1.8 x 104 s (5 h) (La added = 200 mmol kg-1).
|
|
View this table:
[in this window]
[in a new window]
|
Table 4. Yttrium sorbed by soils prior to desorption and proportions desorbed after 1.8 x 104 s (5 h) (Y added = 200 mmol kg-1).
|
|
Log-normal distribution first-order kinetic model (Rate et al., 1992).
It was assumed that binding sites of rare-earth element ions form a continuous distribution with respect to the first-order desorption constant. If this distribution is log-normal distribution, the appropriate rate equation is as:
 | [3] |
where CMS is the concentration of Y, La, or Ce sorbed to soil at time t, Cdes is the amount of Y, La, or Ce that desorbs, µ is the mean of the normal distribution in lnk, 
is the corresponding standard deviation, and K(=lnk) is the variable of integration. It was assumed that all Y, La, or Ce would eventually desorb. Equation [3] may be fitted to experimental data by optimizing values of and µ.
Parameter sets were optimized for Eq. [1], [2], and [3] by nonlinear regression, using the multidimensional simplex method of Nelder and Mead (1965) to minimize a residual sum-of-squares term. For the log-normal distribution model, integrals were solved numerically using the Romberg method (Press et al., 1986). Optimization programs were written in C++ using adaptations of code presented by Liu et al. (1995).
|
RESULTS AND DISCUSSION
|
|---|
Accuracy Check of Determination of Rare-Earth Elements
Accuracy of the elemental analysis was confirmed by the determination of rare-earth elements in the standard reference materials of soils. The results are listed in Table 3, and a good agreement was achieved between the data obtained by the present method and the certified values.
Sorption
For the sorption studies, a pH 6 for the Y(NO3)3, La(NO3)3, and Ce(NO3)3 solutions was chosen. However, because of the different pH values of the soils investigated, the final pH of the equilibrium solutions for sorption experiments varied from soil to soil (Tables 4–6). Nevertheless, the final pH was quite close in the replicate experiments of the same soil. The contents of Y, La, and Ce sorption and desorption after incubation of 1 or 20 wk are also shown in Tables 4–6. The highest amount of Y (83.1–84.9 mmol kg-1), La (77.6–80.8 mmol kg-1), and Ce (82.6–83.2 mmol kg-1) were adsorbed by Haerbin soil. The reason for this was ascribed to larger CEC and organic matter present in Haerbin soil (Table 1). It is generally recognized that more colloidal-size mineral and organic matter can provide a large number of sites capable of binding metals (Stevenson, 1982), and may dominate sorption in many soils. It should be pointed out that the least amount of Y (18.5–20.9 mmol kg-1), La (12.2–15.3 mmol kg-1), and Ce (15.8–15.9 mmol kg-1) was adsorbed by Yingtan soil. Many factors can influence the apparent adsorption of metals by soils. Although Yingtan and Beijing soils have a quite similar amount of organic matter and the content of clay in Yingtan soil was higher, less sorption of rare-earth elements in Yingtan soil was observed. In this case lower pH and low CEC of Yingtan soil may play a dominant role in controlling adsorption of rare-earth elements (Table 1). This finding was consistent with previous reports that rare-earth elements sorption by soil and synthetic oxides increased with increasing pH (Ran and Liu, 1992) and presumably because of an increase in the negatively charged sorbed site (Jones, 1997).
View this table:
[in this window]
[in a new window]
|
Table 6. Cerium sorbed by soils prior to desorption and proportions desorbed after 1.8 x 104 s (5 h) (Ce added = 200 mmol kg-1).
|
|
Generally, the amounts of Y, La, and Ce sorption by each soil are quite alike owing to the similar electric configuration and coordination chemistry of these elements (Brown et al., 1990). However, small differences exist that arise from the "lanthanide contract" (Bailar et al., 1973). The radius of La3+, Ce3+, and Y3+ are 1.06, 1.03 and 0.9 Å, respectively, with cationic radius decreasing the coulombic attraction for organic acids in soil increases. Therefore, the amounts of Y sorbed by Yingtan, Beijing and Haerbin soils were the highest while those of La were the least (Tables 4–6) although this trend was not noticeable for Tongjiang soil.
Comparison of Desorption Kinetics of Rare-Earth Elements
The first-order kinetic model did not provide an acceptable description of Y, La, or Ce desorption from the soils studied, with R2 values between 0.554 and 0.904.
Equation [2] assumed that desorption of Y, La, and Ce from soils was controlled by two simultaneously discrete and independent sites. One is fast desorption reaction with t0.5 (k1) of <1 h, the time required for 50% of Y, La, and Ce associated with a particular reaction to desorb from soil. The other is a much slower reaction, with t0.5 (k2) between 3 and 200 h, where k1 and k2 represent the rate constants for the relatively faster and much slower desorption reactions, respectively. Excellent fits were obtained between the experimental data and model calculation for all four soils and for all three elements, with R2 > 0.99 (Tables 7–9). Lanthanum sorbed versus desorption time by Haerbin soil is chosen as an example and shown in Fig. 1
. However, it should be pointed out that it seems rather unlikely that there are only two types of reaction sites (Barrow, 1999). As noted by Sparks (1995) and Backes et al. (1995), apparent desorption rate constants may be comprised of numerous diffusional and chemical reaction rates. The optimized desorption-rate constants k1 and k2 could only be thought of as representing apparent averages for either relatively fast or relatively slow reactions. Certainly, soil heterogeneity is ascribed to different particle sizes, properties, and types of retention sites. Nevertheless, decrease in k2 values as a result of increasing sorption time could represent a movement of reactions of rare-earth elements in soil from the soil external sites with fast reaction rates to interlayer sites or micropores with slow reaction rates.
View this table:
[in this window]
[in a new window]
|
Table 7. Fitted parameters derived from application of Eq. [2] and [3] to Y desorption kinetics data (C1 and C2 are initial concentrations; k, k1 and k2 are first-order rate constants for desorption; and Czero is the total amount of metal sorbed at time = 0).
|
|
View this table:
[in this window]
[in a new window]
|
Table 9. Fitted parameters derived from application of Eq. [2] and [3] to Ce desorption kinetics data (C1 and C2 are initial concentrations; k, k1 and k2 are first-order rate constants for desorption; and Czero is the total amount of metal sorbed at time = 0).
|
|
View this table:
[in this window]
[in a new window]
|
Table 8. Fitted parameters derived from application of Eq. [2] and [3] to La desorption kinetics data (C1 and C2 are initial concentrations; k, k1 and k2 are first-order rate constants for desorption; and Czero is the total amount of metal sorbed at time = 0).
|
|
Soil physicochemical properties play an important role in controlling the rate of many soil diffusional and chemical reactions. Therefore, the model selected is based on the hypothesis that rare-earth element binding sites form a continuous log-normal distribution with respect to the first-order rate constant (Eq. [3]). Mean log10k (µ) value obtained from Eq. [3] was comparable with the k'd obtained from Eq. [1]. Figure 2
shows good agreement between data and Eq. [3] predicted amounts of sorbed La (Cms) during the desorption experiment. With few exceptions, Eq. [3] also reproduced the amounts of Y and Ce in the solid phase during the desorption experiments (R2 > 0.99). Because the La, Y, and Ce results were similar only the La data and model are presented.
The optimized parameters of desorption experiments of Y, La, and Ce were summarized in Tables 7 through 9. There were small differences between k1 value calculated for Y, La, or Ce desorbed from different soils. For Beijing, Tongjiang, and Haerbin soils, the C2/Czero values were found to be 60.49 to 81.95, 56.12 to 78.13, and 54.67 to 99.85% for Y, La, and Ce, respectively. However, for Yingtan soil, much lower C2/Czero values of 8.95 to 15.66, 6.21 to 26.95, and 46.88 to 53.14% were found for Y, La, and Ce, indicating a more rapid desorption rate. This finding was consistent with the desorption values listed in Tables 4 through 6. Of course, the lower pH of Yingtan soil may be responsible for the increased desorption of Y, La, and Ce relative to other soils with greater pH values. The effect of soils pH on desorption of rare-earth elements may be because of the differences in the soil surface charge (Barrow 1999, Jones, 1997).
To further explain why the desorption rate differed from different soils, a three-stage speciation procedure was performed. Rare-earth elements in these soils were fractionated into water soluble, exchangeable, and carbonate bound (B1), Fe-Mn oxide bound (B2), and organic matter and sulfide bound (B3). The results of fractions of rare-earth elements exogenously added are given in Table 10. As it can be seen for the Beijing soil, <4% of La and Ce were in B1, while the B2 and B3 accounted for 51.6 to 56.4 and 41.9 to 45.4%, respectively. For Haerbin soil, La and Ce in fractions B1, B2, and B3 were 1.40 to 1.98, 45.3 to 56.4, and 41.9 to 53.1%, respectively. However, for Yingtan soil, more than 10% of added La and Ce were present in B1, and B2 was the predominate form as much as 65 to 70% of the La and Ce. This finding is consistent with the higher content of Fe and Mn present in Yingtan soil (Table 1). Generally, fractions B1 and B2 are recognized as easily or relatively easily desorbed forms and B3 as more difficult to desorb from soil. This was additional evidence that rare-earth elements easily desorbed from Yingtan soil.
For Beijing and Tongjiang soils, k1 values decreased slightly when the initial sorption period was increased from 1 to 20 wk (Tables 7–9). In addition, increasing the initial incubation period also increased the proportion of the sorbed Y, La, and Ce associated with the slower reaction (higher C2/Czero, see Tables 7–9). A similar, but less obvious effect was obtained for Y, La, and Ce desorption from Haerbin soil. The effects of initial sorption time on desorption of Y, La, and Ce is shown in Fig. 3 through 5
. The common feature is that increasing the initial sorption period also increased the proportion of rare-earth elements associated with the slower reaction sites (higher C2, see Tables 7–9) because longer sorption period allowed metal ions to penetrate small soil pores (Barrow, 1999). It is much evident that there is an initial rapid adsorption reaction on the surface of the soil particles and reaches equilibrium within hours. This is then followed by a much slower reactions, and hence the term sorption to encompass both rapid and slow reactions (Barrow, 1999). Results obtained by Barrow and Whelan (1989) showed that the sorption rates were correlated with a measure of the affinity of the ions for the surface. Liu et al. (1999) observed the transformation of added rare-earth elements, and showed that as the sorption time increased, the amount of exchangeable rare-earth elements decreased rapidly, while the content of rare-earth elements bounded to organic matter increased. This indicated the continuing reaction of sorbed rare-earth elements. For Ce, another interpretation may exist. Cerium readily undergoes a valence change from Ce3+ to Ce4+ within the range of pH prevalent in soil. Quadrivalent Ce could be fixed in soil by precipitation or coprecipitation as a highly insoluble compound or by being held tenaciously in ion-exchanging materials of the soil. So the fixed Ce would not easily be available nor would it be extracted by the usual ion-exchanging solution (Robinson, 1958).
For Yingtan soil, some exceptional results were obtained especially for Y, the value of C2/Czero decreased (Table 7) with increasing sorption time. It may be because of a higher pH unit of 0.2 of the 1-wk sorption than that of the 20-wk sorption. In Tables 7 through 9, the µ values of Yingtan soil are greater than that of other soils, and there is a clear shift to lower µ value with the increasing sorption period for Y, La, or Ce. This is consistent with the ideas discussed above of continuing reaction result in an overall movement of rare-earth elements to more slowly desorbing sites, or slower interparticle diffusion rates associated with the movement of rare-earth elements to more inaccessible site (McLaren et al., 1998). Alternatively, it is possible that the lower µ values result from chemical reconfiguration at the original binding sites. Although the log-normal model offered excellent fits to the experimental data, the model gives no specific information regarding chemical reaction mechanism (Rate and McLaren, 1992).
It should be mentioned again that soils were heterogeneous complex of organic matter; Fe-, Al-, and Mn-oxides; clay minerals; and miscellaneous other minerals. In addition, soils have micropores and macropores that have different impact on solutes entering the soil. Since numerous chemical reactions and diffusion processes coexist, the relative contributions of chemical and physical processes are difficult to isolate, and thus, only apparent rate parameters were determined. Therefore, the rate parameters are apparent ones. However, Eq. [3] is based on the hypothesis that rare-earth element binding sites form a continuous log-normal distribution with respect to the first-order rate constant rather than just two discrete reaction sites. The change in the desorption with sorption period suggests an overall movement of rare-earth elements to slow desoprtion reaction sites (Backes et al., 1995; McLaren et al., 1998).
Error Analysis
In this experiment, the filter flow methods were applied to investigate the desorption behavior of rare-earth elements from soils. These methods have some advantages over batch techniques for kinetic studies. However, as McLaren et al. (1998) discussed, there are still some potential problems, including the possibility of preferential flow through the material on the filter, and the occurrence of film and interparticle diffusion, which can introduce artifact, particularly when comparing the kinetics of different materials. In this study, no check was made to determine whether desorption kinetics were altered by preferential flow of Ca(NO3)2 through the filter.
The second potential source of error associated with the unstable flow rate. To compensate for clogging of the membrane, the peristaltic pump was adjusted during the experiment such that the flow rate was nearly constant. In this study, the flow rate was checked by measurement of filtrate volume. The error introduced by such unstable flow rate was estimated to be <5%.
The final error source came from the entrained Y, La, and Ce in the filtered soil. Although it was corrected at least to certain extent by weighing, but because the exactly specific gravity of entrained solution is unknown, the amount of rare-earth elements entrained could not be precisely determined. However, the total weight of entrained solution is <0.05 g, showing that the proportion of the entrained solution in the first fractionation is 2 to 10%. It was therefore considered that the error arising from entrained rare-earth elements from the first data point was negligible.
|
CONCLUSION
|
|---|
Among three kinetic models compared, the first-order kinetic model did not provide acceptable descriptions of rare-earth elements from soil. However, both the two-site first-order and log-normal distribution first-order kinetic models gave excellent fits to the experimental kinetic data. The apparent parameters obtained from these two models provided useful information regarding possible changes in rare-earth elements-soil binding with increasing sorption periods. Based on the above study, one may conclude that the desorption behavior of Y, La, and Ce is influenced not only by different soil pH, but also by content of Fe and Mn and organic matter in soil. The distribution of rare-earth elements among different speciation fractions were also responsible for different desorption. Increasing the sorption period resulted in decreased rates of metal desorption, implying that reactions had not reached equilibrium even though no further rare-earth elements were adsorbed. In addition, it should be recognized that none of these models is a panacea for kinetic analyses. Many efforts must be made to develop both the mathematical models and experimental approaches, which can obtain somewhat detailed information to explain the mechanisms of the complicated reactions taking place in the heterogeneous surface of soils.
|
ACKNOWLEDGMENTS
|
|---|
We thank Prof. Jeffrey Charrois of the University of Alberta for his English polishing, comments and suggestions on the manuscript. We express appreciation to Prof. R.G. McLaren for his help. This work was supported by the National Natural Science Foundation of China_29890280-1(2)_, Chinese Academy of Sciences (KZ952-J1-048) and the Research Center for Eco-Environmental Sciences (KIP-9901).
Received for publication April 9, 2001.
|
REFERENCES
|
|---|
- Backes, C.A., R.G. McLaren, A.W. Rate, and R.S. Shift. 1995. Kinetic of cadmium and cobalt desorption from iron and manganese oxides. Soil Sci. Soc. of Am. J. 59:778–785.[Abstract/Free Full Text]
- Bailar, J.C., Jr., H.J. Emeleus, R. Nyholm, and A.F. Trotman-Dickenson. 1973. Comprehensive inorganic chemistry. Pergamon Press, New York.
- Barrow, N.J. 1999. The four laws of soil chemistry: The Leeper lecture 1998. Aust. J. Soil Res. 37:787–829.
- Barrow, N.J., and B.W. Whelan. 1989. Testing a mechanistic model. VII. The effects of time and temperature of incubation on the sorption and subsequent desorption of selenite and selenate by a soil. J. Soil Sci. 40:29–37.
- Bar-Yosef, B., and U. Kafkafi. 1978. Phosphate-calcium exchange equibria in soil: The location of non-specific (Gapon) and specific exchange sites. J. Soil Sci. 18:244–262.
- Blackemore, L.C., P.L. Searle, and B.K. Daly. 1987. Methods for chemical analysis of soils. New Zealand Bureau Scientific Report No. 80. NZ Soil Bureau. Lower Hutt, New Zealand.
- Brown, H.P., A.H. Rathjen, and D.E. Tribe. 1990. Rare earth elements in biological system. p. 423–451. In K.A. Gschncidner Jr. et al. (ed.) Hand book on the physics and chemistry of rare earths. Elsevier Science Publisher B.V.
- Cao, X., Y. Chen, X. Wang, and X. Deng. 2001. Effects of redox potential and pH value on the release of rare earth elements from soil. Chemosphere 44:655–661.[Medline]
- Jones, D.L. 1997. Trivalent metal (Cr, Y, Rh, La, Pr, Gd) sorption in two acid soils and its consequences for bioremediation. Eur. J. Soil Sci. 48:697–702.
- Li, D., S. Huang, W. Wang, and A. Peng. 2001. Study on the kinetics of cerium(III) adsorption-desorption on different soils of China. Chemosphere 44:663–669.[Medline]
- Li, F.L., X.Q. Shan, T.H. Zhang, and S.Z. Zhang. 1998. Evaluation of plant availability of rare earth elements in soils by chemical fractionation and multiple regression analysis. Environ. Pollut. 102:269–277.
- Liu, Y., J.S. Kang, and Y.P. Chen. 1995. Nonnumeric parallel algorithm-Generic algorithm. Science Press, Beijing.
- Liu, Y.L., J. Lin, Z.J. Wang, and A. Peng. 1999. Speciation transformation of added rare earth element in soil. Environ. Chem. 18:393–397.
- McLaren, R.G., C.A. Backes, A.W. Rate, and R.S. Swift. 1998. Cadmium and cobalt desorption kinetics from soil clays: Effect of sorption period. Soil Sci. Soc. Am. J. 62:332–337.[Abstract/Free Full Text]
- Mermut, A.R., J.C. Jain, S. Li, R. Kerrich, L. Kozak, and S. Jana. 1996. Trace element concentrations of selected soils and fertilizers in Saskatchewan, Canada. J. Environ. Qual. 25:845–853.[Abstract/Free Full Text]
- Nelder, J.A., and R. Mead. 1965. A simplex method for function minimization. Comput. J. 7:308–313.
- Nelson, D.W., and L.E. Sommers. 1982. Total carbon, organic carbon, and organic matter. p. 539–579. In A.L. Page et al. (ed.) Methods of soil analysis: Part 2. 2nd ed. ASA and SSSA, Madison, WI.
- Press, W.H., B.P. Flannery, S.A. Teukolsky, and W.T. Vetterling. 1986. Numerical recipes—The art of scientific computing. Cambridge Univ. Press, Cambridge, England.
- Quevauviller, Ph., G. Rauret, and B. Griepink. 1993. Single and sequential extraction in sediments and soils. Intern. J. Environ. Anal. Chem. 51:231–235.
- Ran, Y., and Z. Liu. 1993. Adsorption and desorption of rare earth elements on soils and synthetic oxides. Acta Scientiae Circumstantiae 13:288–293.
- Ran, Y., and Z. Liu. 1992. Investigation of the mechanism of rare earth elements adsorbed by soils and synthetic oxides. Kexue Tongbao 18:1705–1709.
- Rate, A.W., and R.G. McLaren. 1992. Evaluation of a log-normal distribution first-order kinetic model for copper (II)-humic acid complex dissociation. Environ. Sci. Technol. 26:2477–2483.
- Rhoades, J.D. 1982. Cation-exchange capacity. p. 149–1158. In A.L. Page et al. (ed.) Method of soil analysis. Part 2. 2nd ed. ASA and SSSA, Madison, WI.
- Robinson, W.O., H. Bastron, and K.J. Murata. 1958. Biogeochemistry of the rare-earth element, with particular reference to hickory rare earth elements. Geochim. Cosmochim. Acta 14:55–67.
- Romkens, P.E.A.M., and J. Dolfing. 1998. Effect of Ca on the solubility and molecular size distribution of DOC and Cu binding in soil solution samples. Environ. Sci. Technol. 32:363–369.
- Sauve, S., W.A. Norvell, M. McBride, and M. Hendershot. 2000. Speciation and complexation of cadmium in extracted soil solutions. Environ. Sci. Technol. 34:291–296.
- Selim, H.M., R.S. Mansell, and L.W. Zelazny. 1976. Modeling reactions and transport of potassium in soils. Soil Sci. 122:77–84.
- Shuman, L.M. 1982. Separating soil iron and manganies oxide fractions for microelement analysis. Soil Sci. Soc. Am. J. 46:1099–1102.[Abstract/Free Full Text]
- Sivasubramaniam, S., and O. Talibudeen. 1972. Potassium-aluminum exchange in acid soil. I. Kinetic. J. Soil Sci. 23:163–176.
- Sparks, D.L. 1995. Kinetics of soil chemical processes. Academic Press, New York.
- Stevenson, F.J. 1982. Humus chemistry, genesis, composition, reactions. John Wiley and Sons, New York.
- Volokh, A.A., A.V. Gorbunov, S.F. Gundorina, B.A. Revich, M.V. Frontasyeva, and S.P. Chen. 1990. Phosphorus fertilizer production as a source of rare-earth elements pollution of the environment. Sci. Total Environ. 95:141–148.[Medline]
- Wen, B., X.Q. Shan, and S.G. Xu. 1999. Preconcentration of ultratrace rare earth elements in seawater with 8-hydroxyquinoline immobilized polyacrylonitrile hollow fiber membrane for determination by inductively coupled plasma mass spectrometry. Analyst. 124:621–626.
- Willis, B.G., W.H. Woodruff, J.R. Frysinger, D.W. Margerum, and H.L. Pardue. 1970. Simultaneous kinetic determination of mixtures by on-line regression analysis. Anal. Chem. 42:1350–1355.
TOP
ABSTRACT
INTRODUCTION
.
= experimental observations, lines = fits to 
) Yingtan soil, 20 wk; (
) Tongjiang soil, 1 wk; (
) Tongjiang soil, 20 wk; (
) Beijing soil, 1 wk; (
) Beijing soil, 20 wk; (
) Haerbin soil, 1 wk; (•) Haerbin soil, 20 wk.
