Journal of Environmental Quality 31:745-751 (2002)
© 2002 American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America
TECHNICAL REPORTS
Ecological Risk Assessment
Sorption of Iron–Cyanide Complexes on Goethite in the Presence of Sulfate and Desorption with Phosphate and Chloride
Thilo Rennert and
Tim Mansfeldt*
Arbeitsgruppe Bodenkunde und Bodenökologie, Fakultät für Geowissenschaften, Ruhr-Universität Bochum, D-44780 Bochum, Germany
* Corresponding author (tim.mansfeldt{at}ruhr-uni-bochum.de)
Received for publication October 27, 2000.
 |
ABSTRACT
|
|---|
Soils are contaminated with potentially toxic iron–cyanide complexes by some industrial activities. The influence of sulfate on the sorption of the iron–cyanide complexes ferricyanide, [Fe(CN)6]3-, and ferrocyanide, [Fe(CN)6]4-, on goethite was investigated in batch experiments. The experiments were conducted as influenced by pH and varying sulfate/iron–cyanide complex concentration ratios. Furthermore, the desorption of iron–cyanide complexes sorbed on goethite was studied using phosphate and chloride solutions as influenced by pH and anion concentration. Over the whole pH range (pH 3.5 to 8), ferricyanide and sulfate showed similar affinities for the goethite surface. The extent of ferricyanide sorption strongly depended on sulfate concentrations and vice versa. In contrast, ferrocyanide sorption was only decreased (approximately 12%) by sulfate additions at pH 3.5. Ferricyanide was completely desorbed by 1 M chloride, ferrocyanide not at all. Unbuffered phosphate solutions (pH 8.3) desorbed both iron–cyanide complexes completely. Even in 70-fold excess, pH-adjusted phosphate solutions could not desorb ferrocyanide completely at pH 3.5. For ferricyanide we propose a sorption mechanism that is similar to the sulfate sorption mechanism, including outer-sphere and weak inner-sphere surface complexes on goethite. Ferrocyanide appears to form inner-sphere surface complexes. Additionally, we assume that ferrocyanide precipitates probably as a Berlin Blue–like phase at pH 3.5. Hence, ferrocyanide should be less mobile in the soil environment than ferricyanide or sulfate.
|
INTRODUCTION
|
|---|
INDUSTRIAL ACTIVITIES of mankind are the main sources of cyanides in soils, because cyanides from natural sources are consumed by soil microorganisms and do not persist (Fuller, 1985). Soils on sites of former manufactured gas plants (MGP) and coke ovens are contaminated with cyanides (Shifrin et al., 1996). This contamination is caused by the deposition of purifier wastes, which originate from stripping of raw gas onto iron oxide particles by fixation as the pigment Berlin Blue, Fe4[Fe(CN)6]3. In these soils the ferric ferrocyanide Berlin Blue exists in an impure and/or less well crystallized form (Mansfeldt et al., 1998). Although it is slightly soluble (solubility product constant Ksp = 10-84.5; Meeussen et al., 1992b), dissolved iron–cyanide complexes are detected in the ground water (Meeussen et al., 1994) according to a pH-dependent dissolution of Berlin Blue:
 | [1] |
The speciation of iron–cyanide complexes depends on redox potential:
 | [2] |
The anions in Eq. [2] are ferricyanide, [Fe(CN)6]3-, and ferrocyanide, [Fe(CN)6]4-. The purifier wastes also contain large amounts of amorphous FeS and elemental S. The total S content of purifier wastes can be up to 60% (Environmental Resources Limited, 1987). Sulfuric acid is produced by oxidation of reduced sulfur species. Thus, the soils acidify and sesquioxide surfaces are protonated. Anions bearing large charges such as ferri- and ferrocyanide can be adsorbed on these surfaces. Furthermore, large amounts of sulfate are released by oxidation. Hence, both sulfate and dissolved iron–cyanide complexes are present in the soil solution of these sites and compete for adsorption sites.
Additionally, iron–cyanide complexes occur in the environment due to the use of road salt (Ohno, 1990) and in soils on which blast furnace sludge has been deposited (Mansfeldt and Dohrmann, 2001).
Two major adsorption mechanisms have been recognized. They are inner-sphere and outer-sphere surface complexation. Outer-sphere surface complexation is caused by electrostatic attraction and is relatively weak. Chloride is sorbed on goethite, 
-FeOOH, by this mechanism (Hingston et al., 1972). Inner-sphere surface complexation is more stable. The adsorbent becomes part of the surface and a covalent bonding can be formed via ligand exchange. Phosphate forms inner-sphere surface complexes on goethite (Tejedor-Tejedor and Anderson, 1990).
Different mechanisms for sulfate adsorption on goethite are presented in the literature. Inner-sphere surface complexation forming a binuclear complex was proposed by Parfitt and Russell (1977) using infrared spectroscopy. Hansmann and Anderson (1985) found that sulfate is weakly bound through electrostatic attraction on goethite using electrophoresis. Zhang and Sparks (1990) explained the mechanism by outer-sphere surface complexation and simultaneous protonation of the mineral surface using pressure-jump relaxation. Peak et al. (1999) determined both outer-sphere and inner-sphere surface complexes on goethite at pH < 6 and only outer-sphere surface complexes at pH > 6 using in situ attenuated total reflectance Fourier transform infrared (ATR–FTIR) spectroscopy. Similarly, Wijnja and Schulthess (2000) found monodentate inner-sphere sulfate surface complexes on goethite at pH < 6 and predominantly outer-sphere surface complexes at pH > 6 using Raman spectroscopy. Alewell and Matzner (1986) found that in forest soils total inorganic sulfate was extracted by phosphate while chloride desorbed only exchangeable sulfate, which was assumed to be bound as outer-sphere surface complexes. A distinction of adsorbates into inner-sphere and outer-sphere based on macroscopic adsorption data is probably not possible (Eggleston et al., 1998). A spectrum of intermediate behaviors is likely, as shown on spectroscopic information by Peak et al. (1999) and Wijnja and Schulthess (2000) for sorption of sulfate on goethite and by Eggleston et al. (1998) for sorption of sulfate on hematite.
The sorption mechanisms of iron–cyanide complexes on goethite have not been investigated as intensively as the sulfate bonding mechanism. Both iron–cyanide complexes formed outer-sphere surface complexes on 
-Al2O3 as inferred from surface complexation modeling using the triple-layer model (Cheng and Huang, 1996; Cheng et al., 1999). In kinetic batch experiments ferricyanide sorption on goethite was completely reversible (Theis et al., 1988). Therefore, Theis et al. (1988) concluded outer-sphere surface complexation of ferricyanide. Fuller (1985) found ferricyanide to be very mobile in soils mainly controlled by pH as judged from column experiments. Research by Ohno (1990) showed that pH was the most important parameter controlling ferrocyanide sorption by soils. Ferricyanide was not adsorbed by negatively charged kaolinite (Stein and Fitch, 1996). Rennert and Mansfeldt (2001) assumed outer-sphere and weak inner-sphere surface complexation of ferricyanide, whereas ferrocyanide was probably sorbed as an inner-sphere surface complex and by precipitation of a Berlin Blue–like phase on the goethite surface. The presence of sulfate even at a molar ratio (sulfate to ferricyanide) of 100 had no effect on the sorption of ferricyanide on activated carbon (Saito, 1984). The competitive effect of sulfate on iron–cyanide complex sorption on soils or minerals has not yet been investigated.
Goethite is a common Fe oxide in soils, has a well-defined crystal structure and can be readily synthesized in the laboratory (Cornell and Schwertmann, 1996). Hence, it has been used as a model substance for numerous adsorption experiments (summarized by Cornell and Schwertmann, 1996). One objective of this study was to investigate the sorption of iron–cyanide complexes on goethite as influenced by the competitive presence of sulfate in batch experiments. The extent of sorption was investigated including the effects of pH and various sulfate/iron–cyanide complex concentration ratios. The second objective was to study the desorption of sorbed iron–cyanide complexes. We did this by addition of chloride or phosphate solutions, the same method Alewell and Matzner (1996) used to desorb different sulfate surface complexes. All these aspects were used to test the proposed sorption mechanisms of iron–cyanide complexes on goethite by Rennert and Mansfeldt (2001). In the following, we use the term sorption for any retention mechanism including two-dimensional adsorption and three-dimensional processes such as diffusion into crystals and precipitation.
|
MATERIALS AND METHODS
|
|---|
Goethite: Preparation and Properties
Goethite was prepared following Schwertmann and Cornell (1991) by precipitating ferrihydrite from alkaline solution [reaction of 1 M Fe(NO3)3 x 9 H2O with 5 M KOH]. The product was aged for 60 h and converted to goethite at 343 K. It was cleaned by pressure filtration until the electrical conductivity of the filtrate was <3 µS cm-1. Mößbauer spectra showed a hematite content <0.5%. The specific surface area (Brunauer–Emmett–Teller [BET]) measured on freeze-dried samples was 30 m2 g-1, and the point of zero charge (pzc) was 8.3 (Rennert and Mansfeldt, 2001).
Sorption and Desorption Experiments
For the sorption experiments, K3[Fe(CN)6], K4[Fe(CN)6], and Na2SO4 (reagent grade; Riedel-de Haën, Seelze, Germany) were used. Stock salt solutions containing 2000 mg CN- L-1 or 1200 mg SO2-4 L-1 were prepared. Sorption experiments were carried out by a batch technique at a temperature of 283 K in 0.01 M (NaNO3) in 50-mL polyethylene bottles with 25-mL subsamples of a goethite suspension (10 g L-1). Prior to the sorption experiments, six goethite suspensions were adjusted to different pH values ranging from 3.5 to 8.0 using 0.025 M HNO3 or 0.025 M NaOH. It took up to 10 d until the suspension pH was constant. Subsequently, various amounts ranging from 0 to 1 mL (iron–cyanide complexes) and 0 to 2 mL (sulfate) of the stock solutions were added to gain initial concentrations (ci) ranging from 0 to 0.9 mmol L-1 for sulfate and from 0 to 0.5 mmol L-1 for iron–cyanide complexes. In the case of the competitive sorption experiments, aliquots of two stock solutions were added, while for the single sorption experiments, aliquots of only one stock solution were used. The samples were shaken horizontally for 24 h at 150 rpm. This time is sufficient to reach an equilibrium for the sorption of both iron–cyanide complexes on goethite (Rennert and Mansfeldt, 2001). As mentioned above, soils on sites of former coke ovens are very acid. Hence, the competitive sorption experiments with varying concentrations of sulfate or iron–cyanide complexes were conducted at pH 3.5.
For the desorption experiments goethite was equilibrated at a suspension pH of 3.5, 5.0, and 7.0 with ferrocyanide or ferricyanide (ci = 0.25 mmol L-1) for 24 h at 283 K. Then a NaCl solution (seven concentrations ranging from 0.04 to 1 M) or a NaH2PO4 solution (five concentrations ranging from 0.2 to 1.5 mM) was added. As the NaH2PO4 solution is alkaline (pH 8.3), further experiments were conducted using pH-adjusted solutions. The solutions were adjusted to pH 3.5 or 5.0 using 0.01 M HNO3, added to the goethite suspensions (six concentrations ranging from 0.24 to 15 mM NaH2PO4), and then shaken for an additional 24 h.
After separating the phases by membrane filtration (cellulose nitrate, 0.45-µm filter), sulfate, chloride, and phosphate were determined by ion chromatography with a Dionex (Idstein, Germany) DX 500 with a conductivity detector. A micro-distillation technique was used to determine cyanide (Mansfeldt and Biernath, 2000).
Data Interpretation
Selectivity coefficients were calculated as described by Hingston et al. (1971). The concentrations of two competing sorbates as surface phases and in solution are expressed by:
 | [3] |
where Sligand is the sorbed amount of a ligand (mmol kg-1), Kligand is the Langmuir constant of a sorbed ligand and c(ligand) the equilibrium concentration of a ligand (mmol L-1). Selectivity coefficients result from combining the Langmuir constants:
 | [4] |
Hence, the selectivity coefficient concerning the competition between an iron–cyanide complex and sulfate is:
 | [5] |
The influence of phosphate on the dissolution of Berlin Blue was investigated using the speciation program ECOSAT 4.7 (Keizer and van Riemsdijk, 1999). In the system we investigated, the following compounds were initially present: NaH2PO4 (0.5 to 1.5 mM), goethite (10 g L-1), NaNO3 (0.01 M), Berlin Blue (5 x 10-3 mM), and K4[Fe(CN)6] (0.15 to 0.21 mM). The pH was varied in the range of 3.5 to 10.
|
RESULTS
|
|---|
Sorption of Iron–Cyanide Complexes on Goethite in the Presence and Absence of Sulfate
The sorption of iron–cyanide complexes on goethite in the presence and in the absence of equimolar sulfate concentrations is shown in Fig. 1
. Furthermore, the amounts of sulfate sorbed in the presence of iron–cyanide complexes are also presented. With the exception of sulfate sorption in the presence of ferrocyanide (Fig. 1b), the extent of sorption of each anion decreased with increasing pH due to increasing negative charge of the goethite surface. Sulfate decreased ferrocyanide sorption only at pH 3.5 by 12% (Fig. 1b). At other pH values, sorption of ferrocyanide was not influenced by equimolar additions of sulfate. The amount of sulfate sorbed in the presence of ferrocyanide was nearly constant at approximately 2 mmol kg-1 over the entire pH range. At all pH values, only 4 to 7% of the sulfate initially added was sorbed. Hence, in an equimolar solution of both anions, ferrocyanide was preferentially sorbed. At pH 3.5 and 4, ferricyanide sorption was reduced up to 54% by sulfate additions (Fig. 1a). Over the whole pH range, the average ratio of sorbed ferricyanide to sorbed sulfate, S([Fe(CN)6]3-) to S(SO2-4), was 1.7. This is slightly more than the charge ratio, 1.5. At all pH values used here, ferricyanide is completely deprotonated, since all pKa for ferricyanic acid values are <1 (Jordan and Ewing, 1962). Therefore, the charge of ferricyanide was fixed at -3 when calculating the charge ratio. Thus, in equimolar solutions and in the pH range of 3.5 to 8.0 ferricyanide had a slightly greater affinity for the goethite surface than sulfate. Selectivity coefficients calculated according to Eq. [5] are summarized in Table 1
. The selectivity coefficients for ferricyanide and sulfate reflect their similar sorption behavior as mentioned above. In the case of ferrocyanide, the selectivity coefficients decreased with increasing pH, but at any pH ferrocyanide sorption was favored. Hingston et al. (1971) explained variations of selectivity coefficients with pKa values of the conjugate acids of the anions used. However, for ferrocyanic acid pK4 is 4.2 and pK3 is 2.2 (Jordan and Ewing, 1962). Hence, the speciation of ferrocyanide cannot explain selectivity coefficient variations at pH > 6.0. Possibly there are other factors influencing the selectivity coefficients, for example, various sorption mechanisms depending on pH.
View larger version (9K):
[in this window]
[in a new window]
|
Fig. 1. Sorption of iron–cyanide complexes on goethite in the presence and absence of equimolar amounts of sulfate and sorption of sulfate in the presence of iron–cyanide complexes as influenced by pH (a, ferricyanide; b, ferrocyanide). Initial concentration of each anion was 0.3 mmol L-1.
|
|
The sorption of iron–cyanide complexes and sulfate on goethite at pH 3.5 with increasing concentrations of iron–cyanide complexes and constant sulfate concentrations is shown in Fig. 2 . The variable r is the initial molar concentration ratio of the anions. The lines represent exponential functions. In the case of ferricyanide (Fig. 2a), the point of intersection was at 1.35. This is nearly the charge ratio of the anions and reflects the similar affinity of ferricyanide and sulfate at pH 3.5 over the range of concentration ratios. Ferrocyanide was sorbed to a greater extent than sulfate for r > 0.48 (Fig. 2b). These experiments show that ferrocyanide is sorbed preferentially on goethite compared with sulfate.
The sorption of iron–cyanide complexes and sulfate on goethite at pH 3.5 with increasing concentrations of sulfate and constant iron–cyanide complex concentrations is shown in Fig. 3
. Again, ferricyanide and sulfate showed similar affinities for goethite (Fig. 3a), but the point of intersection was at 0.88, which is higher than the charge ratio, 0.67. In contrast, ferrocyanide sorption was not decreased by sulfate, even when it was present in nearly fourfold excess (Fig. 3b).
Desorption of Iron–Cyanide Complexes by Chloride and Phosphate
The effect of chloride additions on the desorption of iron–cyanide complexes is shown in Fig. 4
. Only ferricyanide was desorbed by chloride. With increasing chloride concentration more ferricyanide was desorbed. Complete desorption was achieved with 0.5 M chloride. As mentioned before, the final chloride concentrations in the filtrate were measured. They were not lower than the initial ones. Therefore, it can be concluded that chloride was not sorbed on goethite. Thus, ferricyanide desorption was caused by an increase in ionic strength and not by an ion exchange reaction. Rennert and Mansfeldt (2001) showed that ferricyanide sorption on goethite depends on ionic strength adjusted with NaNO3. In contrast to ferricyanide, ferrocyanide was not desorbed by chloride significantly even at concentrations up to 1 M chloride. This is consistent with the observation that ferrocyanide sorption on goethite does not depend on ionic strength (Rennert and Mansfeldt, 2001). Again, final chloride concentrations were not lower than the initial ones.
View larger version (26K):
[in this window]
[in a new window]
|
Fig. 4. Desorption of iron–cyanide complexes by chloride solutions (0.04 to 1 M NaCl) at three initial pH values.
|
|
Additions of unbuffered phosphate solutions resulted in a complete desorption of both iron–cyanide complexes sorbed on goethite (Fig. 5)
. At any initial suspension pH, ferricyanide was desorbed completely with 0.8 mM phosphate. To desorb ferrocyanide completely, 1.2 mM phosphate at pH 3.5 and pH 5.0 and 0.8 mM phosphate at pH 7.0 had to be added. As the phosphate solution is alkaline (pH 8.3), partial desorption of the complexes by unbuffered phosphate was due to raising the pH. Actually, after the desorption experiments, the pH of samples with initial pH 3.5 rose to values between 4.8 and 5.2; the group of samples with initial pH 5 rose to between 6.3 and 6.7; and the group of samples with initial pH 7 rose to between 7.5 and 7.8. Therefore, a distinction between desorption caused by the increase of pH and desorption caused by the exchange of sorbed iron–cyanide complexes with phosphate was not possible. In contrast to the chloride experiments, in any desorption experiment using phosphate, the final phosphate concentrations were smaller than the initial ones. Hence, phosphate must have been sorbed, and the desorption of iron–cyanide complexes was the result of an exchange reaction of the surface complexes. Furthermore, additions of phosphate solutions to the goethite suspension increased ionic strength, which certainly influenced ferricyanide desorption as well. To exclude pH-induced desorption caused by the use of unbuffered phosphate solutions, pH-adjusted phosphate solutions (pH 3.5 and 5.0) were added to goethite that was equilibrated with ferrocyanide (Fig. 6)
. At pH 5.0, sorbed ferrocyanide was completely desorbed with 1.7 mM phosphate. However, independent of phosphate concentrations, at pH 3.5 a fraction of ferrocyanide was not desorbed (13% of initially sorbed ferrocyanide), although the phosphate concentration was 70 times greater than the initial ferrocyanide concentration. This excess phosphate should have desorbed all the adsorbed ferrocyanide. The amounts of phosphate sorbed were evaluated with the Langmuir equation (not shown here). At both pH 3.5 and pH 5.0, the sorption maximum was in the range of 93 to 95 mmol kg-1 or, with respect to the specific surface area, 3.1 µmol m-2. Theoretically, it is possible that ferrocyanide, which was not desorbed, is bound in the interior of goethite crystals and cannot be desorbed in the time range we used. However, ferrocyanide diffusion into the pores of goethite crystals is unlikely due to the observed lack of time-dependent sorption within 120 h under similar reaction conditions (Rennert and Mansfeldt, 2001). These authors proposed that ferrocyanide, which is sorbed on goethite at pH 3.5, may form a precipitated Berlin Blue–like phase on goethite. Modeling the pH-dependent dissolution of Berlin Blue with phosphate concentrations that were used in the desorption experiments showed that Berlin Blue was dissolved at pH > 7 only. For lower pH values, the saturation index was 1. The dissolution was not increased by the presence of phosphate in the concentration range of 0.5 to 1.5 mM phosphate. Therefore, if the fraction of sorbed ferrocyanide not being desorbed at pH 3.5 is a Berlin Blue–like phase, it should not be desorbed by phosphate solutions adjusted to pH 3.5.
View larger version (23K):
[in this window]
[in a new window]
|
Fig. 5. Desorption of iron–cyanide complexes by phosphate solutions (0.2 to 1.5 mM Na2HPO4) at three initial pH values.
|
|
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 6. Desorption of ferrocyanide by pH-adjusted phosphate solutions (0.24 to 15 mM Na2HPO4) at two initial pH values.
|
|
|
DISCUSSION
|
|---|
We showed that the affinities of sulfate and ferricyanide for goethite are very similar, with ferricyanide having a slightly higher affinity. The affinity of ferrocyanide for goethite is higher than that of sulfate. Soil solutions of manufactured gas plant (MGP) site soils contain more sulfate than iron–cyanide complexes (up to 125 mmol SO2-4 kg-1 soil; Byers et al., 1994). Therefore, in MGP site soils, sulfate should be sorbed to a greater extent than ferricyanide. Consequently, ferricyanide should be mobile. However, it is not very probable that ferricyanide exists to a great extent in soil solutions of MGP sites. Even under oxic conditions, ferrocyanide should be the dominant species in these soils due to its kinetic stability (Holleman and Wiberg, 1985). Theis et al. (1994) found ferrocyanide to be the dominant cyanide species in batch experiments with MGP site soils. Hence, high sulfate concentrations should influence cyanide mobility in these soils only to a small extent. Ferrocyanide is potentially more toxic than ferricyanide. It is converted to extremely toxic free cyanide, CN-, when transported to surface waters and exposed to daylight (Meeussen et al., 1992a). Hence, the presence of ferrocyanide in seepage, ground, and surface water is potentially hazardous because of its photodegradation to free cyanide.
Desorption with chloride and phosphate seems to be an adequate tool to distinguish between ferrocyanide and ferricyanide sorbed on iron oxides, since chloride desorbs ferricyanide only. In subsoils of former coke oven sites a fraction of iron–cyanide complexes exists, which has been transported there as Berlin Blue colloids (Mansfeldt et al., 1998). A distinction between colloidal solid phases and adsorbed iron–cyanide complexes in soils seems to be possible using phosphate solutions: alkaline phosphate solutions desorb all ferrocyanide; phosphate solutions adjusted to pH 3.5 desorb adsorbed ferrocyanide and do not dissolve Berlin Blue colloids. However, Berlin Blue colloids, which may be present in the filtrate, are digested by the distillation technique we used. Therefore, the colloids have to be separated from the bulk filtrate by ultracentrifugation.
Ferrocyanide sorption is only slightly decreased by sulfate additions at pH 3.5. Sulfate forms inner-sphere surface complexes on goethite at pH 3.5 (Peak et al., 1999) and can thus compete for surface sorption sites with ferrocyanide. At higher pH values, the outer-sphere component of sulfate sorption increases, hence its ability to lower ferrocyanide sorption decreases. Ferrocyanide cannot be desorbed by chloride. This suggests inner-sphere surface complexation of ferrocyanide. Desorption with phosphate adjusted to pH 3.5 is not complete. When using phosphate as a desorbing agent, the phosphate maximum sorption is 3.1 µmol m-2 as shown above. This is higher than the phosphate maximum sorption on goethite of 2.51 µmol m-2 calculated for bidentate linkage (Schwertmann, 1988). However, the specific surface area of goethite in suspensions is probably higher than that of dried goethite on which the Brunauer–Emmett–Teller (BET) surface area was measured. Nevertheless, we can conclude from this result that phosphate has occupied all available adsorption sites. This is also shown by the complete desorption of sorbed ferrocyanide at pH 5.0 when using pH-adjusted phosphate solutions. If ferrocyanide is not desorbed completely at pH 3.5, although all adsorption sites are occupied by phosphate, it is not irreversibly adsorbed on goethite in terms of hysteresis. Hence, ferrocyanide sorbed at pH 3.5 does not only exist as an adsorbed fraction, but may also exist as a precipitated fraction, as a Berlin Blue–like phase. This has been proposed by Rennert and Mansfeldt (2001). Additions of phosphate solutions adjusted to pH 3.5 do not dissolve a Berlin Blue–like phase, this was demonstrated by the modeling results. However, this could be expected, because the solubility product constant (Ksp) of iron(III)phosphate is 9.9 x 10-29 (Lide, 1995), which is much greater than the Ksp of Berlin Blue. The dissolution of Berlin Blue strongly depends on pH with greatest dissolution occurring under alkaline conditions (Mansfeldt et al., 1998). This process is controlled by dissolution kinetics (Meeussen et al., 1992b). The proposed precipitation of a Berlin Blue–like phase might also explain the decrease of the selectivity coefficients concerning ferrocyanide and sulfate with increasing pH. Due to the pH-dependent dissolution of Berlin Blue, it is very probable that the precipitate only exists in the acidic range. The selectivity coefficients at pH 7.0 and 8.0 may reflect the competitive adsorption of ferrocyanide and sulfate without formation of a Berlin Blue–like phase. This, in turn, means that variations of the selectivity coefficients are, in addition to dependence on the pKa values of the conjugate acids (as stated by Hingston et al., 1971), caused by differences in sorption mechanisms, which may be pH dependent.
Ferricyanide and sulfate compete for the same exchange sites. The extent of sorption can be explained with the value of their charges. Rennert and Mansfeldt (2001) proposed a sorption mechanism for ferricyanide that is similar to the sulfate sorption mechanism. Ferricyanide is desorbed by chloride just as sulfate sorbed as an outer-sphere surface complex would be (Alewell and Matzner, 1996); its sorption on goethite is highly influenced by ionic strength and pH; it is quickly and completely desorbed as the pH is raised (Rennert and Mansfeldt, 2001); and it is not sorbed on goethite when pH > point of zero charge (pzc), like sulfate (Geelhoed et al., 1997). Hence, we propose that the strength of ferricyanide and sulfate surface complexes on goethite is very similar.
|
CONCLUSIONS
|
|---|
According to our results, high sulfate concentrations should not diminish the immobilization of potentially hazardous ferrocyanide. Ferrocyanide and sulfate differ in their affinity for goethite. Due to its strong sorption on goethite, ferrocyanide cannot be regarded as a highly mobile anion in soils. Ferricyanide and sulfate sorb similarly on goethite.
Both iron–cyanide complexes adsorbed on goethite were desorbed by phosphate solutions adjusted to pH 
5. Since acid to neutral phosphate solutions do not dissolve Berlin Blue–like precipitates, they could be used to separate adsorbed from precipitated fractions of ferrocyanide. This could be used as a tool to investigate the different forms of iron–cyanide complexes in contaminated soils.
The results of competitive sorption of ferrocyanide and sulfate and of desorption using chloride and phosphate solutions suggest strong sorption, which may be explained by inner-sphere surface complexation of ferrocyanide and partial precipitation as a Berlin Blue–like phase at low pH. The results for both ferricyanide sorption and desorption indicate that its sorption is very similar to sulfate sorption. This may be attributed to similar sorption mechanisms. However, since these conclusions are inferred from macroscopic observations, spectroscopic investigations on the nature of iron–cyanide surface complexes formed on goethite are necessary.
|
ACKNOWLEDGMENTS
|
|---|
This paper represents Publication no. 160 of the Priority Program 546 "Geochemical processes with long-term effects in anthropogenically affected seepage- and groundwater." Financial support was provided by Deutsche Forschungsgemeinschaft. Assistance in the laboratory was given by H. Biernath and G. Wilde, both of Ruhr-Universität Bochum, Germany.
|
REFERENCES
|
|---|
- Alewell, C., and E. Matzner. 1996. Water, NaHCO3–, NaH2PO4– and NaCl-extractable SO2-4 in acid forest soils. Z. Pflanzenernähr. Bodenkd. 159:235–240.
- Byers, W., M.B. Meyers, and D.E. Mooney. 1994. Analysis of soil from a disused gasworks. Water Air Soil Pollut. 73:1–9.
- Cheng, W.P., and C. Huang. 1996. Adsorption characteristics of iron–cyanide complex on -Al2O3. J. Colloid Interface Sci. 181:627–637.
- Cheng, W.P., C. Huang, and J.R. Pan. 1999. Adsorption behavior of iron-cyanide on -Al2O3 interface: A coagulation approach. J. Colloid Interface Sci. 213:204–207.[ISI][Medline]
- Cornell, R.M., and U. Schwertmann. 1996. The iron oxides. VCH, Weinheim, Germany.
- Eggleston, C.M., S. Hug, W. Stumm, B. Sulzberger, and M. dos Santos Afonso. 1998. Surface complexation of sulfate by hematite surfaces: FTIR and STM observations. Geochim. Cosmochim. Acta 62:585–593.
- Environmental Resources Limited. 1987. Problems arising from the redevelopment of gas works and similar sites. 2nd ed. Dep. of the Environment, London.
- Fuller, W.H. 1985. Cyanides in the environment with particular attention to the soil. p. 19–44. In D. van Zyl (ed.) Cyanide and the environment. Colorado State Univ., Fort Collins.
- Geelhoed, J.S., T. Hiemstra, and W.H. van Riemsdijk. 1997. Phosphate and sulfate adsorption on goethite: Single anion and competitive adsorption. Geochim. Cosmochim. Acta 61:2389–2396.
- Hansmann, D.D., and M.A. Anderson. 1985. Using electrophoresis in modeling sulfate, selenite, and phosphate adsorption onto goethite. Environ. Sci. Technol. 19:544–551.
- Hingston, F.J., A.M. Posner, and J.P. Quirk. 1971. Competitive adsorption of negatively charged ligands on oxide surfaces. Discuss. Faraday Soc. 52:334–342.
- Hingston, F.J., A.M. Posner, and J.P. Quirk. 1972. Anion adsorption on goethite and gibbsite. J. Soil Sci. 23:177–193.
- Holleman, A.F., and E. Wiberg. 1985. Lehrbuch der anorganischen Chemie. 91st to 100th ed. Walter de Gruyter, Berlin.
- Jordan, J., and G.J. Ewing. 1962. The protonation of hexacyanoferrates. Inorg. Chem. 1:587–591.
- Keizer, M.G., and W.H. van Riemsdijk. 1999. ECOSAT 4.7. A computer program for the calculation of speciation and transport in soil–water systems. Dep. of Soil Sci. and Plant Nutrition, Wageningen Agric. Univ., Wageningen, the Netherlands.
- Lide, P.R. (ed.) 1995. CRC handbook of chemistry and physics. 76th ed. CRC Press, Boca Raton, FL.
- Mansfeldt, T., and H. Biernath. 2000. Determination of total cyanide in soils by micro-distillation. Anal. Chim. Acta 406:283–288.
- Mansfeldt, T., and R. Dohrmann. 2001. Identification of a crystalline cyanide-containing compound in deposited blast furnace sludge. J. Environ. Qual. 30:1927–1932.[Abstract/Free Full Text]
- Mansfeldt, T., S.B. Gehrt, and J. Friedl. 1998. Cyanides in a soil of a former coking plant site. Z. Pflanzenernähr. Bodenkd. 161:229–234.
- Meeussen, J.C.L., M.G. Keizer, and F.A.M. de Haan. 1992a. Chemical stability and decomposition rate of iron cyanide complexes in soil solutions. Environ. Sci. Technol. 26:511–516.
- Meeussen, J.C.L., M.G. Keizer, W.H. van Riemsdijk, and F.A.M. de Haan. 1992b. Dissolution behavior of iron cyanide (Prussian Blue) in contaminated soils. Environ. Sci. Technol. 26:1832–1838.
- Meeussen, J.C.L., M.G. Keizer, W.H. van Riemsdijk, and F.A.M. de Haan. 1994. Solubility of cyanide in contaminated soils. J. Environ. Qual. 23:785–792.[Abstract/Free Full Text]
- Ohno, T. 1990. Levels of total cyanide and NaCl in surface waters adjacent to road salt storage facilities. Environ. Pollut. 67:123–132.[Medline]
- Parfitt, R.L., and J.D. Russell. 1977. Adsorption on hydrous oxides. IV. Mechanisms of adsorption of various ions on goethite. J. Soil Sci. 28:297–305.
- Peak, D., R.G. Ford, and D.L. Sparks. 1999. An in situ ATR–FTIR investigation of sulfate bonding mechanisms on goethite. J. Colloid Interface Sci. 218:289–299.[ISI][Medline]
- Rennert, T., and T. Mansfeldt. 2001. Sorption of iron–cyanide complexes on goethite. Eur. J. Soil Sci. 52:121–128.
- Saito, I. 1984. The removal of hexacyanoferrate(II) and (III) ions in dilute aqueous solution by activated carbon. Water Res. 18:319–323.
- Schwertmann, U. 1988. Some properties of soil and synthetic iron oxides. p. 203–250. In J.W. Stucki et al. (ed.) Iron in soils and clay minerals. D. Reidel, Dordrecht, the Netherlands.
- Schwertmann, U., and R.M. Cornell. 1991. Iron oxides in the laboratory. VCH, Weinheim, Germany.
- Shifrin, N.S., B.D. Beck, T.D. Gauthier, S.D. Chapnick, and G. Goodman. 1996. Chemistry, toxicology, and human health risk of cyanide compounds in soils at former manufactured gas plant sites. Regul. Toxicol. Pharmacol. 23:106–116.[ISI][Medline]
- Stein, J.A., and A. Fitch. 1996. Effect of clay type on the diffusional properties of a clay-modified electrode. Clays Clay Miner. 44:381–392.[Abstract]
- Tejedor-Tejedor, M.I., and M.A. Anderson. 1990. Protonation of phosphate on the surface of goethite as studied by CTR–FTIR and electrophoretic mobility. Langmuir 6:602–611.[ISI]
- Theis, T.L., R. Iyer, and L.W. Kaul. 1988. Kinetic studies of cadmium and ferricyanide adsorption on goethite. Environ. Sci. Technol. 22:1013–1017.
- Theis, T.L., T.C. Young, M. Huang, and K.C. Knutsen. 1994. Leachate characteristics and composition of cyanide bearing wastes from manufactured gas plants. Environ. Sci. Technol. 28:99–106.
- Wijnja, H., and C. Schulthess. 2000. Vibrational spectroscopy study of selenate and sulfate adsorption mechanisms on Fe and Al (hydr)oxide surfaces. J. Colloid Interface Sci. 229:286–297.[ISI][Medline]
- Zhang, P.C., and D.L. Sparks. 1990. Kinetics and mechanisms of sulfate adsorption/desorption on goethite using pressure-jump relaxation. Soil Sci. Soc. Am. J. 54:1266–1273.[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
T. Rennert, T. Mansfeldt, K. U. Totsche, and K. Greef
Sorption and Transport of Iron-Cyanide Complexes in Goethite-coated Sand
Soil Sci. Soc. Am. J.,
May 1, 2003;
67(3):
756 - 764.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
= 0.23 mmol L-1; I = 0.01 M. Regression equations for sorption data: ferricyanide, S 
= 20.5 - 19.8 exp
; ferrocyanide, S 
= 22.2 - 24.1 exp
; sulfate (presence of ferricyanide), S 
; sulfate (presence of ferrocyanide), S 
.
= 0.24 mmol L-1; I = 0.01 M. Regression equations for sorption data: ferricyanide, S 
; ferrocyanide, S 
; sulfate (presence of ferricyanide), S 
; sulfate (presence of ferrocyanide), S 
.