| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Dep. of Biological and Environmental Engineering, Graduate School of Agricultural and Life Sciences, The University of Tokyo, 1-1-1 Yayoi, Bunkyo, Tokyo, 113-8657, Japan
b Inst. of Terrestrial Ecology, ETHZ, CH-8952 Schlieren, Switzerland
c Dep. of Plant and Soil Sciences, University of Delaware, Newark, DE 19717-1303
Corresponding author (noriko{at}soil.en.a.u-tokyo.ac.jp)
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
-Ni hydroxide on gibbsite. This difference can be explained by the differing Al solubilities of the two minerals. Pyrophyllite is relatively soluble, causing the rapid formation of amorphous Al hydroxide, which, in turn, is a necessary precursor for the formation of Ni–Al LDH. In spite of the complexation of Al by organic ligands, sufficient amorphous Al hydroxide was available to promote the formation of Ni–Al LDH. Gibbsite, on the other hand, is much less soluble, and the smaller amount of initially released Al may be fully complexed by citrate and salicylate. The subsequent lack of amorphous Al hydroxide prevented the formation of Ni–Al LDH, and, instead, -Ni hydroxide formed. Only after a longer period of 30 d and at a low citrate concentration did enough Al become available to transform -Ni hydroxide into the thermodynamically more stable Ni–Al LDH.
Abbreviations: DRS, diffuse reflectance spectroscopy • HS-gibbsite, high surface area gibbsite • LDH, layered double hydroxide • LS-gibbsite, low surface area gibbsite • PZSE, point of zero salt effect • XAS, x-ray absorption spectroscopy
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
X-ray absorption spectroscopy (XAS) studies revealed that Ni(II), Co(II), and Zn(II) formed LDH precipitates on Al-bearing minerals and in soil at pH 
7 (d'Espinose de la Caillerie et al., 1995; Towle et al., 1997; Scheidegger et al., 1997; Scheidegger et al., 1998; Roberts et al., 1999; Ford and Sparks, 2000). These LDH phases consist of brucite-type mixed-metal hydroxide sheets, which are separated from each other by water and charge-balancing anions. Their formula is 
x+ 
A-n mH2O, where M2+ represents a range of transition metals. The net positive layer charge is balanced by anions such as NO-3, Cl-, CO2-3, and ClO-4 (A-n) (Hashi et al., 1983; Génin et al., 1991). In the presence of Al-free minerals, structurally very similar but thermodynamically less stable -type metal hydroxides with the formula M(OH)2-x (x/n)A-nmH2O have been identified by DRS and XAS (Scheinost et al., 1999; Scheinost and Sparks, 2000). Both types of precipitates create a sink for Ni and are more stable than Ni bound as outer-sphere or inner-sphere sorption complexes (Bradbury and Baeyens, 1997). However, Ni–Al LDH is more resistant to dissolution than -Ni hydroxide. (Scheckel et al., 2000). Therefore, to accurately predict the fate of Ni in soils and sediments, it is important to understand the controls for the formation of specific precipitates.
Scheidegger et al. (1998) suggested that the rate-limiting step for the formation of Ni–Al LDH is Al dissolution from the mineral surface. This is in consistent with the observation that Ni–Al LDH formed after only 5 min in the presence of the relatively soluble pyrophyllite, but only after 24 h in the presence of the more stable gibbsite (Scheinost et al., 1999). In both cases, dissolution of the mineral surfaces may be enhanced by Ni-promoted dissolution (d'Espinose de la Caillerie et al., 1995). Evidence for pyrophyllite dissolution was suggested by increasing Si concentrations in solution. However, the Al concentrations in gibbsite and pyrophyllite systems remained below 1 µmol L-1, most likely due to precipitation of amorphous Al hydroxide (Thompson et al., 1999). Together with an initial Ni hydroxide phase, the Al hydroxide is a necessary precursor for the formation of Ni–Al LDH (Boclair and Braterman, 1999; Taylor, 1984). The progression of the dissolution explains the constant growth of Ni–Al LDH, which has been observed as long as sufficient Ni was in solution (Scheinost et al., 1999).
Organic ligands form complexes with both surface-bound and aqueous cations. Depending on pH, and type and concentration of organic ligands, mineral dissolution may or may not be enhanced (Drever and Stillings, 1997; Kraemer et al., 1998), and metal adsorption by mineral surfaces may be suppressed or enhanced by the presence of ligands (Brooks and Herman, 1998; Bryce et al., 1994; Boily and Fein, 1996). Consequently, organic ligands may affect the formation process of Ni–Al LDH and -Ni hydroxide.
In this context, we used citrate and salicylate as representatives of tricarboxylic and monocarboxylic ligands in soils (Tan, 1986) and investigated their influence on the formation of surface-induced Ni hydroxides. Our working hypothesis was that they may affect the precipitate formation by two main processes:
-Ni hydroxide.
Pyrophyllite and gibbsite were chosen as sorbents because of their differing stability towards Al release. To monitor precipitate formation we used DRS, which is capable of detecting and discriminating Ni–Al LDH and -Ni hydroxide (Scheinost et al., 1999).
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
-Ni hydroxide reference compound was synthesized by adding 550 mL of 30% ammonia to 500 mL of 1 M Ni(NO3)2 (Génin et al., 1991). After vigorously stirring for 2 h the precipitate was washed and dried as described above. The synthesized precipitates were identified as Ni–Al LDH and -Ni hydroxide by x-ray diffraction and Fourier transform infrared spectroscopy (Scheinost et al., 1999; Scheinost and Sparks, 2000).
Nickel Sorption Studies
Sorbents were suspended in 0.1 M NaNO3 background electrolyte and equilibrated for 24 h, then adjusted to pH 7.5 by addition of 0.1 M NaOH or 0.1 M HNO3. Next, 0.1 M Ni(NO3)2 solution was mixed with sodium citrate or sodium salicylate. After 2 h, the Ni-ligand solution was slowly added to the sorbent suspension. The final solid concentration was 20 g L-1. A pH of 7.50 ± 0.01 was maintained by continuous addition of 0.1 M NaOH using a pH-stat system (Radiometer, Copenhagen, Denmark). Initial metal and ligand concentrations were 1.5 mM Ni(II), up to 3 mM citrate, and up to 1.5 mM salicylate. The suspensions were vigorously stirred and purged with N2 gas to exclude CO2. Thirty milliliters of suspension were collected periodically and centrifuged at 27000 g for 3 min to separate wet pastes from solution. The pastes were washed once with background electrolyte and stored in a refrigerator for a maximum of 3 d before collecting DRS spectra. However, to prevent aging effects, the spectra of short-term samples (<24 h) were collected immediately. Supernatants were filtered through a 0.2-µm membrane filter and analyzed for Ni, Al, and Si by inductively coupled plasma atomic emission spectroscopy. Citrate and salicylate were determined with a dissolved organic C analyzer (URA-106, Shimadzu Scientific, Kyoto, Japan). Amounts of sorbed Ni, citrate, and salicylate were calculated from the difference between initial and final concentrations in solution.
Diffuse Reflectance Spectroscopy Studies
The DRS experiments were conducted with a Perkin-Elmer double-beam Lambda 9 spectrophotometer equipped with a Spectralon-coated integrating sphere 5 cm in diameter (Perkin-Elmer, Norwalk, CT). Spectra were collected from 1000 to 500 nm (1-nm steps, 60 nm min-1 scan speed, 2-s response time). The wet pastes were filled in an aluminum holder coated with parafilm, 10 mm in diameter and 1 mm in depth, and the surface of the paste was covered with a microscope cover slide. The reflectance was calibrated against a Spectralon standard (Labsphere, North Sutton, NH) covered with another slide to compensate for the absorbance of the glass. The spectra were ratioed against those of blanks; that is, the minerals were prepared the same way as the sorption samples but without Ni or citrate addition. The resulting reflectance spectra were converted into absorbance using the Kubelka–Munk equation, then the 
2 band positions of the spectra were determined by deconvolution with Gaussian line shapes using GRAMS/32 ver.4.2 (Galactic Industries Corp., Nashua, NH). Detailed procedures are described in Scheinost et al. (1999).
Preparation of Model Mixtures
To calibrate the 2 band positions for the quantification of surface precipitates consisting of mixtures of Ni–Al LDH and -Ni hydroxide, we prepared physical mixtures of freeze-dried Ni–Al LDH and -Ni hydroxide samples. The DRS analysis of these samples was conducted on dry powders to prevent chemical reactions.
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
25% of aqueous Ni (calculated using the data by El-Ezaby and El-Khalafawy, 1981). As expected, the suppression of Ni sorption by salicylate was weaker than in the presence of citrate (Fig. 1c). Furthermore, with 3% reduction on HS-gibbsite, 8% on pyrophyllite, and 55% on LS-gibbsite, the order of the reduction was different from the Ni–citrate system. The smaller influence of citrate on Ni sorption by the HS-gibbsite than by the LS-gibbsite and the pyrophyllite is in line with the presence of a substantial amount of chemisorbed Ni on HS-gibbsite as has been detected by x-ray absorption fine structure spectroscopy (Yamaguchi and Scheinost, 2000, unpublished data).
|
2 band positions of the sorption samples are plotted together with gray bands representing the range of Ni–Al LDH and -Ni hydroxide reference compounds (Scheinost et al., 1999). In the absence of citrate, the 2 positions of the LS-gibbsite system increased with time from 15230 to 15410 cm-1, but remained in the range indicative of Ni–Al LDH (Fig. 2a). A similar blue shift with aging has been previously observed and explained by crystallite growth (Scheinost et al., 1999). At molar ratios of citrate to Ni of 0.7 and 1 (labeled cit/Ni = 0.7 and 1 in Fig. 2, with Ni = 1.5 mM, citrate = 1.0 mM and Ni = 1.5 mM, citrate = 1.0 mM), the 2 band was close to the -Ni hydroxide region, with a slight blue shift with time. For cit/Ni = 0.3, 2 was initially in between Ni–Al LDH and -Ni hydroxide, then dropped down to -Ni hydroxide, and was up in the Ni–Al LDH range after 30 d. For cit/Ni = 2, the band intensity was too low to reliably determine the band position.
|
-Ni hydroxide was observed in the presence of both HS- and LS-gibbsite, while the precipitate formation was almost completely suppressed for pyrophyllite (the 2 band was too weak to be fitted). However, a pyrophyllite sample with a lower citrate/Ni ratio of 0.3 confirms the general trend that increasing citrate causes a red shift of 2 positions towards that of -Ni hydroxide (Fig. 2b).
In the presence of salicylate, the red shift was also observed (Fig. 2c). The 2 band positions are predominantly intermediate to those of Ni–Al LDH and -Ni hydroxide. We assumed that these intermediate positions reflect mixtures of Ni–Al LDH and -Ni hydroxide. In fact, physical mixtures of both Ni–Al LDH and -Ni hydroxide showed only one slightly broadened 2 band (Fig. 3a)
. The differing positions of the 2 band of Ni–Al LDH and -Ni hydroxide do not resolve, because they are only 500 cm-1 apart, but have a large width at one-half height of 4500 cm-1. The fitted positions of this broadened band are intermediate to those of the two single components consistent with the observed band positions (Fig. 3b). Relatively small fractions of -Ni hydroxide shift the band drastically toward red.
|
|
Aluminum and Silicon Dissolution Kinetics
In the presence of citrate, the Al concentration in solution increased to more than 100 µmol L-1 (Fig. 5a)
. This high Al concentration is caused by citrate-promoted pyrophyllite dissolution, and the subsequent formation of aqueous Al–citrate complexes, which lower the activity of free Al3+ and consequently its tendency to form Al hydroxide. The addition of Ni lowered the Al concentration in solution. This may be explained by Al and Ni both competing for citrate complexes, which increases the relative amount of uncomplexed Al and reduces the amount of free citrate available for dissolution of pyrophyllite. Al concentration decreased in the sequence: pyrophyllite >> LS-gibbsite > HS-gibbsite (Fig. 5b), corresponding with the sequence in which citrate suppressed the sorption of Ni. This confirms that the suppression mechanism of Ni sorption by citrate is related to Al dissolution. In spite of the fact that little citrate was sorbed on the pyrophyllite surface, Al dissolution was strongly promoted by the presence of citrate. While the addition of Ni suppressed the Al dissolution from the pyrophyllite surface, Si dissolution was not affected (Fig. 5c). Scheidegger et al. (1997) found that Si was dissolved from the pyrophyllite surface by 0.1 M NaNO3 at pH 7.5, whereas Al was not detectable. This is due to the low solubility of Al at pH 7.5, causing the Al detached from the pyrophyllite structure to reprecipitate as Al hydroxide (Thompson et al., 1999). While citrate is able to keep more Al in solution by forming an Al–citrate complex, the competitive formation of Ni–citrate complexes reduces the Al solubility in the presence of Ni. Al was not dissolved from gibbsite or pyrophyllite in the presence of salicylate at pH 7.5, neither with nor without Ni. This is in line with results by Kraemer et al. (1998) showing that the salicylate-promoted dissolution of 
-Al2O3 had a minimum at pH 7.50. Likewise, the presence of salicylate had no effect on Si dissolution from pyrophyllite.
|
-Ni hydroxide was achieved at pH 7.50 and [Ni] > 3 mM. Therefore, our system was also undersaturated with respect to -Ni hydroxide. Nevertheless, DRS confirmed that this phase formed in the presence of gibbsite. Our results are in agreement with several other studies, showing that the formation of precipitates at the mineral–water interface occurs under conditions that are undersaturated with respect to the homogeneous solution (Fendorf et al., 1992; Xia et al., 1997). This precipitation may be explained by the combination of several processes. First, the electric field of the mineral surface attracts Ni ions through adsorption, leading to a local supersaturation at the mineral–water interface. Second, the solid phase may act as a nucleation center for polyhydroxy species and catalyze the precipitation process (McBride, 1994, p. 154). Third, the physical properties of water molecules adsorbed at the mineral surface are different from those of free water (Sposito, 1984), potentially causing a lower solubility of metal hydroxides at the mineral–water interface. Layered double hydroxides are commonly synthesized by the addition of base to a mixture of M(II) and M(III). For the formation of LDH, the M(III) hydroxide that precipitates first must be sufficiently soluble, and the M(II) hydroxide that precipitates second must be sufficiently insoluble (Boclair and Braterman, 1999). Consequently, Al hydroxide is a necessary precursor for the formation of Ni–Al LDH. Our solution data as well as those of Scheidegger et al. (1996)(1997, 1998) and Scheinost et al. (1999) show a substantial Si release from pyrophyllite, indicating the dissolution of this mineral. Although one would expect that the pyrophyllite dissolution leads to a congruent release of Al and Si, Al concentration was below the detection limit in all experiments cited. Since the detection limit for Al in the references cited was above the saturation of amorphous Al hydroxide, it is very likely that the missing Al from pyrophyllite dissolution was precipitated as such an amorphous Al hydroxide.
This explanation is in line with the observed difference between pyrophyllite and gibbsite. While the formation of Ni–Al LDH is rapid on pyrophyllite (5 min), it is slow on gibbsite (1 d) (Scheidegger et al., 1996, 1997, 1998; Scheinost et al., 1999). Our sorption experiments with citrate, which keeps Al in solution, showed that more Al was dissolved from pyrophyllite than from gibbsite (Fig. 5b), making evident that pyrophyllite is more soluble than gibbsite. Therefore, the dissolution of pyrophyllite produces a substantial amount of amorphous, easily soluble Al hydroxide precipitate, which is responsible for the fast formation of Ni–Al LDH. In contrast, the lower solubility of gibbsite prevents both the immediate reaction of this crystalline Al hydroxide with Ni to form Ni–Al LDH, as well as the formation of a sufficient amount of secondary Al hydroxide. Therefore, the formation of Ni–Al LDH on pyrophyllite and the formation of -Ni hydroxide on gibbsite give clear evidence that the kinetics of Ni–Al LDH formation is controlled by the solubility of Al from sorbent phases. Due to the higher thermodynamic stability of Ni–Al LDH, however, -Ni hydroxide precipitates transform into Ni–Al LDH as soon as sufficient Al is available.
The reaction scheme outlined above explains the influence of the organic ligands on both the amount and the composition of the surface-induced precipitates. The formation of stable Ni–organic complexes reduces the formation of Ni hydroxide, which is either the end product of the surface precipitation (-Ni hydroxide) or the precursor for the subsequent formation of Ni–Al LDH in the presence of an amorphous Al hydroxide. Therefore, the organic ligands generally reduce the amount of Ni surface precipitates. Furthermore, the organic ligands altered the type of precipitates. In spite of the fact that Al was released into solution from both LS- and HS-gibbsite, the formation of Ni–Al LDH was suppressed by citrate. This, and the fact that citrate kept Al effectively in solution, shows that citrate dissolved the secondary amorphous Al hydroxide and subsequently reduced or prevented the formation of Ni–Al LDH. At the lowest citrate/Ni ratio of 0.3, -Ni hydroxide formed first on LS-gibbsite, but after 30 d Ni–Al LDH predominated (Fig. 2a). Simultaneous to this phase transformation, Al concentration in solution dropped below the detection limit (Fig. 5a). In contrast to the gibbsite system, however, citrate did not prevent the formation of Ni–Al LDH in the presence of pyrophyllite, although it drastically suppressed its amount. Again, this can be explained with the reaction scheme outlined above. Citrate promotes the dissolution of pyrophyllite, but also reduces formation of the secondary, amorphous Al hydroxide phase. Consequently, formation of Ni–Al LDH is reduced. The precipitation of amorphous Al hydroxide is responsible for the rapid formation of Ni–Al LDH (Taylor, 1984). At low citrate concentration ([Ni] = 1.5 mM, [cit] = 0.5 mM), Al hydroxide was still available to induce the rapid formation of Ni–Al LDH on pyrophyllite though a part was dissolved by citrate. At citrate/Ni = 1 ([Ni] = [cit] = 1.5 mM), Al concentration in solution was very high; hence, amorphous Al hydroxide may not have been present. At this reaction condition, no Ni precipitate was formed on pyrophyllite.
Salicylate may suppress the precipitation of Ni by two mechanisms, the complexation of Ni in the aqueous phase and the formation of an adsorption complex with the mineral surface. At the circum-neutral pH of our study, salicylate most likely forms a monodentate sorption complex with the aluminol surface (Kubicki et al., 2000). This adsorbed salicylate may block the surface of Al hydroxide and subsequently suppress the coprecipitation of Al hydroxide with Ni hydroxide. Therefore, the precipitates were dominated by -Ni hydroxide (Fig. 2c). At the pyrophyllite surface, which is probably coated by secondary Al hydroxide (Thompson et al., 1999), salicylate should preferentially bind to the sorption sites on this secondary Al hydroxide, therefore reducing the nucleation of Ni hydroxide polymers. The weak affinity of salicylate for Ni could also suppress the approach of Ni to the salicylate-adsorbed Al hydroxide surface. However, because of the relatively high solubility of Al at the pyrophyllite surface, the formation of Ni–Al LDH still dominated.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
-Ni hydroxide. The observed effects could be explained by the ligand-enhanced dissolution of the sorbent phases, the formation of aqueous organo-metal complexes, and the interaction of these complexes with the surfaces. The kinetics of precipitate formation was largely controlled by either the less soluble primary Al hydroxide (gibbsite) or more soluble secondary Al hydroxide (pyrophyllite). Both the suppression of precipitates in general and the trend from more stable Ni–Al LDH towards the less stable -Ni hydroxide may significantly enhance the availability of Ni in soils compared with systems free of organic ligands. |
ACKNOWLEDGMENTS |
|---|
Received for publication February 3, 2000.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
-Al2O3. Geochim. Cosmochim. Acta 60: 2929–2938.
-alumina with Ni(II) or Co(II) ions at neutral pH: Hydrotalcite-type coprecipitate formation and characterization. J. Am. Chem. Soc. 117:11471–11481.[Web of Science]-type nickel hydroxide obtained by chemical precipitation: Study of the anionic species. Eur. J. Solid State Inorg. Chem. 28:505–518.This article has been cited by other articles:
|
|
 |
|
  R. Dahn, M. Jullien, A. M. Scheidegger, C. Poinssot, B. Baeyens, and M. H. Bradbury IDENTIFICATION OF NEOFORMED Ni-PHYLLOSILICATES UPON Ni UPTAKE IN MONTMORILLONITE: A TRANSMISSION ELECTRON MICROSCOPY AND EXTENDED X-RAY ABSORPTION FINE STRUCTURE STUDY Clays and Clay Minerals, April 1, 2006; 54(2): 209 - 219. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. U. Yamaguchi, N. U. Yamaguchi, A. C. Scheinost, and D. L. Sparks INFLUENCE OF GIBBSITE SURFACE AREA AND CITRATE ON Ni SORPTION MECHANISMS AT pH 7.5 Clays and Clay Minerals, December 1, 2002; 50(6): 784 - 790. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Journal of Natural Resources and Life Sciences Education |
Vadose Zone Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||
