Environ. Sci. Technol. 2003, 37, 68-74
Kinetic Speciation of Co(II), Ni(II), Cu(II), and Zn(II) in Model Solutions and Freshwaters: Lability and the d Electron Configuration AMINA L. R. SEKALY, JOHN MURIMBOH, NOURI M. HASSAN, RUPASRI MANDAL, MUFIDA E. BEN YOUNES, CHUNI L. CHAKRABARTI,* AND MARGARET H. BACK Ottawa-Carleton Chemistry Institute, Department of Chemistry, Carleton University, 1125 Colonel By Drive, Ottawa, Ontario, K1S 5B6 Canada D . C O N R A D G R EÄ G O I R E Geological Survey of Canada, 601 Booth Street, Ottawa, Ontario, K1S 0E8 Canada
The kinetic speciation of Co(II), Ni(II), Cu(II), and Zn(II) in model solutions of a well-characterized fulvic acid (Laurentian fulvic acid), freshwater samples from the Rideau River (Ottawa, Ontario), and freshwater samples from the Sudbury (Ontario) area were investigated by the competing ligand exchange method using Chelex 100 as the competing ligand and by inductively coupled plasma-mass spectrometry to measure the dissociation kinetics. The metal species were quantitatively characterized by the rate coefficient for the first-order dissociation of metal complex to free metal ion. This technique can be applied to almost all elements and represents an important advance in our ability to investigate the kinetic availability of metal species in the freshwater environment. The order of the lability of the metal complexes, Co(II) > Ni(II) > Cu(II) < Zn(II), follows the reverse order of the ligand field stabilization energy with the exception of Cu(II); the behavior of Cu(II) is also due to the Jahn-Teller effect, which shortens the equatorial bonds and lengthens the axial bonds of a tetragonally distorted Cu(II)-L6 complex. This study has demonstrated a relationship between the lability of metal-DOM complexes of the 3d transition metals in freshwaters and their d electron configuration. This is the first time that the importance of the d electron configuration on the lability of metal complexes in the freshwater environment has been demonstrated. The slow complexation kinetics of both Ni(II) and Cu(II) suggest that the usual equilibrium assumption for freshwaters may be invalid.
Introduction The biological effects and biogeochemical cycling of trace metals are profoundly influenced by their chemical speciation (1, 2). The interactions of metal ions with humic substances influences the transport and bioavailability of metals; acidbase balance; and solubility in waters, sediments, and soils. As a result, much research has focused on studying the metal * Corresponding author telephone: (613)520-2600, ext 3839; fax: (613)520-3749; e-mail:
[email protected]. 68
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 1, 2003
and proton binding by dissolved organic matter (DOM), such as humic substances, which is aimed at both elucidating mechanisms and providing quantitative information. DOM is a particularly important because of its high concentration; strong affinity for metals; and ubiquity in freshwaters, soils, and sediments. In competition studies, it is important to consider competition between trace metals (3, 4), proton competition (5, 6), and competition with major cations (7). The dominant tool for studying metal speciation has been the local equilibrium (or pseudoequilibrium) approximation (8). The complexation reactions are assumed to be fast relative to other processes, and an equilibrium distribution among dissolved organic and inorganic species is assumed despite the fact that the total metal concentration may vary as a result of some controlling biogeochemical process. However, natural systems are often not equilibrium systems. DOM, which is a heterogeneous, polyelectrolytic ligand system, can depart considerably in kinetic behavior from simple ligands and may result in systems that are dynamic and far removed from equilibrium (8-12); hence, equilibrium speciation methods may not apply (13). In some instances, the rates of biogeochemical reactions may be influenced or even controlled by the rates of metal complexation reactions (12). Furthermore, not much is known about the kinetics of metal complexation of trace metals by DOM and the relative reactivity of metals in freshwaters. Hence, new kinetics-based chemical speciation methodologies are required to reflect more correctly the realities of freshwater systems, in which metal release is often kinetically controlled. The objective of this study was to investigate the kinetic speciation of cobalt, nickel, copper, and zinc within the context of natural waters as dynamic systems.
Kinetic Model Competing Ligand Exchange Method (4, 14). Adapting the kinetic model proposed by Olson and Shuman (15), the following kinetic model has been developed (16, 17) to study the dissociation kinetics of metal complexes in an aqueous mixture of n components in which each component (MLi) exists in equilibrium with its dissociation products: the free metal ion (i.e., metal aqua complex) (M) and a naturally occurring, heterogeneous complexant (Li) such as humic substances that are ubiquitous in the aquatic environment (charges have been omitted for simplicity). The subscripted Li represents different binding sites for the metal, M. The derivation of the model has been presented in our earlier publications (4, 14): kd,i
MLi {\ } M + Li (slow) k f,i
(1)
where the formation and dissociation rate coefficients (kf,i and kd,i) are coupled by the stability constant (K ) kf/kd) through the principle of microreversibility (18). When a large excess of a competing ligand, such as Chelex 100 chelating resin, is added to the sample, M-Chelex is formed, driving reaction 1 to the right:
M + Chelex a M-Chelex
(fast)
(2)
If each complex (MLi) dissociates simultaneously and independently (at a rate that depends on the nature of the functional group, its position on the macromolecule, and the residual charge), the total concentration of all complexes (cML) at any time (t) is given by a summation of exponentials (eq 3): 10.1021/es025805g CCC: $25.00
2003 American Chemical Society Published on Web 12/04/2002
n
cML(t) )
∑c i)1
o MLi exp(-kd,it)
(3)
o where cML,i is the initial concentration of MLi complex and cML,i(t) is the concentration of MLi at any time (t). The measured rate coefficients for each component represent the dissociation rate coefficient of the metal complex bound to the corresponding complexing site. Disjunctive and Adjunctive Pathways. Two reaction pathways for the overall ligand-exchange reactions may be considered (18): (i) slow dissociation of ML to give M and L, followed by a fast reaction with the competing ligand, Chelex 100, (eqs 1 and 2), which is known as the disjunctive pathway, and (ii) direct attack by the competing ligand followed by loss of the original ligand (Li), which is known as the adjunctive pathway, shown in eqs 4 and 5. The adjunctive and disjunctive pathways are stoichiometric mechanisms inferred from the rate laws and do not imply any knowledge of the transition state of the substitution reaction:
MLi + Chelex f Chelex - Mi - Li
(4)
Chelex - M - Li f M - Chelex + Li
(5)
Dissociation of ML is a fundamental process of the natural sample and is the type of reaction for which the theory of ligand substitution processes can be reliably extrapolated over concentrations and pH (12). The first-order rate coefficient of the disjunctive pathway is simply the rate coefficient for the slow step (eq 1), dissociation of the metal complex (ML). The observed rate coefficients for the adjunctive pathway are strongly influenced by the nature of the arbitrarily chosen probe ligand (i.e., steric and electrostatic factors and by protonation of the incoming ligand) since either formation or dissociation of the intermediate ternary complex can be rate-limiting. Consequently, little relevant information can be inferred about the processes in natural systems from the adjunctive pathway (8). Although both the disjunctive and adjunctive pathways generally contribute to the observed, overall rate coefficient (19, 20), one pathway may predominate.
Experimental Section Chemicals and Reagents. Stock solutions (1000 mg/L) of copper and nickel were prepared separately by dissolving high-purity metal powder (99.999% pure, SPEX) in ultrapure nitric acid (Ultrex II, JT Baker) with heating and diluting the solution to the appropriate volume with ultrapure water (18.2 MΩ‚cm) acidified to 1% (v/v) with ultrapure nitric acid. Stock solutions (1000 mg/L) of cobalt and zinc were purchased from BDH (atomic absorption grade). Working standard solutions were prepared daily by dilution of the stock solutions with ultrapure water acidified to 1% (v/v) with ultrapure nitric acid. Analytical grade (minimum 99% pure) Chelex 100 resin (100-200 mesh, sodium form, Bio-Rad) was pretreated by equilibrating it with an acetic acid-sodium acetate buffer solution (pH 5.0 ( 0.1) for at least 24 h and storing it in the buffer solution until use. The wet capacity of Chelex 100 resin is 0.61 mequiv/g (21). A stock solution of Laurentian fulvic acid (1.0000 g/L) was prepared by dissolving 1.0000 g of freeze-dried Laurentian fulvic acid (Fredriks Research Products, The Netherlands) in ultrapure water. The concentration of carboxylic and phenolic groups in the Laurentian fulvic acid is 11.6 mmol/g (22, 23). All standards and test solutions were prepared with ultrapure water, which was obtained direct from a Milli-Q UF Plus water purification system (Millipore), fitted with an organic purification column to remove organic matter.
TABLE 1. Instrumental Operating Conditions and Data Acquisition Protocol for Inductively Coupled Plasma-Mass Spectrometry inductively coupled plasma rf power (kW) coolant argon flow rate (L min-1) carrier argon flow rate (L min-1) auxiliary argon flow rate (L min-1) data acquisition parameters dwell time (ms) scan mode signal measurement points/spectral peak resolution
1 15 0.6 2.4 100-1000 peak hop counts s-1 1 normal
Cleaning Procedures for Containers. All containers used were made of Teflon, and they were cleaned as follows. After being cleaned with ultrapure water, they were completely filled with 10% nitric acid (AR grade) and allowed to stand at room temperature for 1 week. Then they were rinsed 5 times with ultrapure water (18.2 MΩ‚cm), filled with ultrapure water, and allowed to stand until they were used; the filling water was renewed periodically to ensure continued contact with clean water. Model Solutions and Freshwater Samples. Model solutions were prepared to contain Laurentian fulvic acid (FA) and equimolar concentrations of Ni(II), Cu(II), and Zn(II). The concentration of each metal was 7.0 × 10-7 mol/L. The Rideau River water sample was collected from a location close to the Steacie Building at Carleton University in Ottawa (ON, Canada) using an acid-precleaned Teflon container. Immediately after the collection, the samples were filtered through 0.45-µm filters of mixed cellulose ester membranes (Pall) using a peristaltic pump. Since it was known from previous experiments that the concentration of the metals in the Rideau River water sample was too low for kinetic analysis, the filtered Rideau River water sample was spiked with an appropriate amount of a standard solution containing one of the following metals: Ni(II), Cu(II), or Zn(II). The concentration of each metal was 3.2 × 10-7 mol/L. The pH of the fulvic acid model solutions and the spiked Rideau River water sample was adjusted to 5.0 ( 0.2 with 2 mol/L ultrapure nitric acid and/or 2 mol/L purified sodium hydroxide, and the samples were left to equilibrate overnight (12 h). Freshwater samples were collected from two sites in the Sudbury (ON, Canada) area in pretreated 2.2-L Teflon bottles, and the pH and the conductivity of the original samples were measured at the sampling sites using an Horiba D-24 pH/conductivity meter. On the following day, the samples were brought to our laboratory, where they were immediately filtered through 0.45-µm filters of mixed cellulose ester membranes (Pall) using a peristaltic pump. The filtrates were used as test solutions at their initial pH values. Kinetic Experiments. The dissociation kinetics of Co(II), Ni(II), Cu(II), and Zn(II) complexes was studied using Chelex 100 as the competing ligand and an ELAN 5000 inductively coupled plasma-mass spectrometry (ICP-MS) (Perkin-Elmer SCIEX). The instrument operating conditions and the data acquisition protocol are presented in Table 1. Three grams (1%, w/v) of Chelex 100 was added to the sample solution in a cylindrical Teflon Reactor (500 mL volume), and the sample solution was stirred continuously with a Tefloncoated stirring bar. The data acquisition was initiated 15 s before the Chelex 100 resin was added to the Teflon reactor so that any contamination of the resin from its pretreatment would be indicated by a sudden increase in the signal. In the case of any contamination, the experiment was terminated, and a fresh test solution was studied. The sample solution was filtered with an online 0.45-µm GN-6 membrane filter VOL. 37, NO. 1, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
69
TABLE 2. Natural Water Sample Characteristics
Rideau River Onaping River Moose Lake
pH
conductivity (S/m)
DOC (mg/L)
5.0 6.8 6.9
0.0308 0.0093 0.152
7.7 4.9 1.5
FIGURE 2. Dissociation kinetics of metal complexes in the Rideau River water sample, collected on June 13, 1996, from Ottawa (ON, Canada) measured by ICP-MS. pH 5.0 ( 0.1, DOC ) 7.7 mg/L, ionic strength 4.9 × 10-3 mol/L. (0) Ni(II)-DOM complexes; (4) Cu(II)DOM complexes; ()) Zn(II)-DOM complexes. FIGURE 1. Dissociation kinetics of metal-FA complexes in the model solution of Laurentian FA, measured by ICP-MS. pH 5.0 ( 0.2; ionic strength 4.2 × 10-5 mol/L. (0) cNi/cFA ) 0.8; (4) cCu/cFA ) 0.8; ()) cZn/cFA ) 0.8. (Pall) to filter the Chelex 100 resin before introducing the filtrate into the plasma torch through the solution nebulizer using a peristaltic pump at a flow rate of 1 mL/min. The time interval between the data points and the total time of the data acquisition was set for each experiment on the basis of the dissociation kinetics of the metal complexessshort time intervals and short total data acquisition times for rapid kinetics and longer time intervals and longer total data acquisition times for slow kinetics. Quality control for measurements of cobalt, nickel, copper, and zinc included periodic analysis of a Certified Reference Material, NIST 1643d. A prior set of samples was re-run if the analyzed value differed from the certified value by >20%. The relative standard deviation of replicate determinations was typically e5%.
Results and Discussion Table 2 presents some characteristics of the natural water samples. Kinetic data for the dissociation of Ni(II), Cu(II), and Zn(II) complexes in model solutions of a well-characterized fulvic acid (Laurentian fulvic acid) and in RRSW samples (collected on June 13, 1996, from a site at Carleton University) are shown in Figures 1 and 2, respectively. Because the metal concentrations in the Rideau River water samples were below their limits of detection, the samples were spiked with 3.2 × 10-7 mol/L of each metal, and the spiked samples were equilibrated overnight (12 h). Freshwater samples collected from the Onaping River and Moose Lake, which contained measurable concentrations of the metals, were studied without spiking the samples. The results of kinetic analysis of Co(II), Ni(II), Cu(II), and Zn(II) complexes from the Onaping River and Moose Lake water samples are shown in Figures 3 and 4, respectively. The data were fitted to the kinetic model (eq 3) by nonlinear regression analysis based on the Marquardt-Levenberg algorithm (24).The results of regression analysis are presented in Tables 3-6. The uncertainties represent the 95% confidence limits of the nonlinear regression analysis. In some cases, a quickly dissociating kinetically distinguishable component (kd ∼ 10-2 70
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 1, 2003
FIGURE 3. Dissociation kinetics of metal complexes in the Onaping River water sample, collected on October 27, 1998, from the Sudbury (ON, Canada) area, measured by ICP-MS. pH 6.8 ( 0.1, DOC ) 4.9 mg/L, ionic strength 1.5 × 10-3 mol/L. (O) Co(II)-DOM complexes, cCo ) 3.6 × 10-8 mol/L; (0) Ni(II)-DOM complexes, cNi ) 1.5 × 10-6 mol/L; (4) Cu(II)-DOM complexes, cCu ) 1.3 × 10-7 mol/L; ()) Zn(II)-DOM complexes, cZn ) 7.2 × 10-8 mol/L. s-1) was not observed from the nonlinear regression analysis. Two kinetically distinguishable components were observed in the fulvic acid model solutions and the Moose Lake sample, whereas three kinetically distinguishable components were observed in the spiked Rideau River surface water sample and the Onaping River sample. Since it is not possible to define the individual binding sites of heterogeneous complexants, such as natural organic matter, the problem in applying eq 3 lies in attempting to fit the kinetic data to a set of rate coefficient (ki) and cono centration (cML,i ) parameters without any a priori knowledge about the total number of binding sites. Since eq 3 is nonlinear, this is a difficult task. Furthermore, the actual binding sites may be a nearly continuous distribution (25). The data for each experiment was fitted to eq 3 using kinetic models with one, two, three, and four kinetically distinguishable components. The model chosen was the one that provided the “best fit”. Hence, the measured rate coefficients do not necessarily imply the presence of only two or three discrete metal complexes. They only represent the minimum
FIGURE 4. Dissociation kinetics of metal complexes in the Moose Lake water sample, collected on October 27, 1998, from the Sudbury (ON, Canada) area, measured by ICP-MS. pH 6.9 ( 0.1, DOC ) 1.5 mg/L, ionic strength 2.4 × 10-2 mol/L. (O) Co(II)-DOM complexes, cCo ) 3.2 × 10-8 mol/L; (0) Ni(II)-DOM complexes, cNi ) 9.0 × 10-7 mol/L; (4) Cu(II)-DOM complexes, cCu ) 1.1 × 10-7 mol/L; ()) Zn(II)-DOM complexes, cZn ) 1.1 × 10-7 mol/L. number of components required to model the metalcomplex systems. Sekaly et al. (26, 27) have shown that the binding sites of humic substances probably exist as distributions of sites, but they can be separated into recognizable groups that may be approximated by a small collection of rate coefficients. The steep section at the beginning of each curve (kd ∼ 10-2 s-1) in Figures 1-4 can be attributed to one or several of the following species: metal aqua complex, quickly dissociating inorganic complexes, and quickly dissociating organic complexes of the metal. Rate coefficients for the uptake of Ni2+, Cu2+, and Zn2+ aqua complexes by Chelex 100 were reported to be (4 ( 2) × 10-2, (2.2 ( 0.2) × 10-2, and (2.2 ( 0.6) × 10-2 s-1 (28), respectively. No artifacts were observed in the uptake of the metal aqua complex by Chelex 100. The results suggest that complexes with dissociation rate coefficients >10-2 s-1 cannot be experimentally distinguished from the metal aqua complex using this method. In the context of metal speciation, the lifetime (first-order dissociation) of the metal complex (ML) is inversely related to its dissociation rate coefficient (τML ) 1/kd). On a time scale much larger than the lifetimes of M and ML (i.e., kft and kdt . 1, where t is the time scale of measurement), the metal frequently interconverts (i.e., flip flops) between M and ML, and the system is in dynamic equilibrium. The other limiting case is when the lifetimes are much larger than the analytical time scale of measurement (i.e., kft and kdt , 1). In this case, the system is static, and the complexes are defined as inert because both M and ML are essentially frozen in time (29). The slowly falling sections of each curve are probably due to the slow dissociation of strong metal complexes. The fraction of metal complexes remaining at the end of each experiment represents the fraction of complexes that did not dissociate within the duration of the experiment. The dissociation rate coefficients for these complexes are estimated to be Ni(II) d8 > Cu(II) d9 < Zn(II) d10. This trend was observed in the model solutions of Laurentian fulvic acid, the spiked river water samples from the Rideau River, and the freshwater sample from the Onaping River. An opposite trend was reported for the stability constants of complexes of 3d transition metal ions with simple ligands
(30) and with humic substances (31). The reaction kinetics of Ni(II) is known to be slow relative to that of other first-row divalent transition metals because of its electron configuration (32, 33). Slow Ni(II) ligand-exchange kinetics between DOM and dimethylglyoxime (half-life: 5-95 h) was observed by Xue et al. (34) in an oligotrophic lake and in a small river affected by inputs from sewage effluents and agriculture in Switzerland. Our results on the slow dissociation kinetics of Cu(II) complexes agree with those of Town and Filella (35), who found that Cu(II)-fulvic acid complexes are nonlabile under anodic stripping voltammetry; with that of a previous study by Sekaly et al. (36) using Chelex 100 as a competing ligand (kd ∼ 10-3-10-6 s-1); and with that of Achterberg et al. (37), who reported that equilibrium between organically complexed Cu and suspended particulate matter was only reached after 4-15 h. Figures 1-3 and Tables 3-5 show that the dissociation kinetics of Ni(II) and Cu(II) complexes are kinetically slow relative to Zn(II) and Co(II) complexes. However, Figure 4 and Table 6 show that the dissociation rate coefficients of the Ni(II) complexes became larger than those of the Co(II) and Zn(II) complexes at sufficiently low concentrations of dissolved organic carbon (DOC), i.e., high nickel to DOC ratios (Moose Lake: Ni/DOC ) 0.0006 mol/g; Onaping River: Ni/DOC ) 0.0003). The same trend was reported by Sekaly et al. (14) for the kinetic speciation of Ni(II), Pb(II), Cu(II), and Cd(II) in model solutions of Armadale fulvic acid and by Mandal et al. (38) for the kinetic speciation and toxicity of Ni(II) in freshwaters from the Sudbury (ON) area. The importance of the metal to DOC ratio on metal speciation reflects the essential role of DOM in buffering free metal ion concentrations (38, 39) in the aquatic environment and suggests that the strong binding sites became saturated as the metal loading of the system increased and the system became progressively dominated by the weak sites (which are labile) at high nickel to DOC ratios. A general increase in the percentage of quickly dissociating species (over the sample from the Onaping River) was observed for all the metal complexes except copper in the sample from Moose Lake, as shown in Tables 5 and 6. The results probably reflect competition for the binding sites by large concentrations major cations, such as Ca2+, which results in the high conductivity of 0.152 S/m for the sample. Although the concentrations of calcium and magnesium were not measured in this sample, they were measured in another sample collected from the same site in June 1999, which was similar to the sample collected in October 1998, had a similarly high conductivity of 0.167 S/m, and had a high calcium concentration of 7.5 mmol/L (the concentration of magnesium was below the detection limit). The DOC concentrations of the samples were 1.5 and 3.4 mg/L for the October 1998 and the June 1999 samples, respectively. Although Ca2+ (a major cation) forms weak complexes with humic substances, it is present in large excess relative to the concentrations of trace metals in freshwaters. The overwhelming concentration of Ca2+ in the sample from Moose Lake probably resulted in the presence of a large amount of Ca2+ in the electrical double layer of the DOM polyanions, screening the charge of the DOM polyanions and thereby significantly decreasing the total binding energy of the metal-DOM complexes (18). The influence of major cations such as Ca2+ and Mg2+ on metal speciation has been demonstrated by Mandal et al. (7), who have shown that overwhelming amounts of major cations (Ca2+ and Mg2+) promote the release large fractions of DOMbound Ni(II). No significant effect was observed on the kinetic speciation of Cu(II) probably because the Cu(II) was strongly bound as Cu(II)-DOM complexes. In this case, the concentration of Ca2+ in the sample may not have been sufficient to compete with Cu(II) for the binding sites of the DOM in Moose Lake. VOL. 37, NO. 1, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
71
TABLE 3. Dissociation Kinetics of Metal-FA Complexes in Model Solution of Laurentian FAa metal
CM,total (mol/L)
%C1
k1 × 102 (s-1)
%C2
k2 × 103 (s-1)
%C3
k3 × 104 (s-1)
Ni(II) Cu(II) Zn(II)
7.0 × 10-7 7.0 × 10-7 7.0 × 10-7
65 ( 3 76 ( 1
1.3 ( 0.3 3.0 ( 0.2
32 ( 1 -
0.24 ( 0.03 -
41 ( 3 68 ( 1 24 ( 1
3.0 ( 0.1 1.3 ( 0.1 4.5 ( 0.3
a Measured by ICP-MS, c /c -5 M. -, not observed from the nonlinear regression analysis. k and M FA ) 0.8, pH 5.0 ( 0.1, ionic strength 4.2 × 10 1 k2, dissociation rate coefficients of the kinetically distinguishable components. %C1 and %C2, percentage concentrations of the first and second components, respectively. The uncertainties represent the 95% confidence limits of the nonlinear regression analysis.
TABLE 4. Dissociation Kinetics of Metal Complexes in Rideau River Water Sample, Collected on June 13, 1996, from Ottawa (ON, Canada)a metal
CM,total (mol/L)
%C1
k1 × 102 (s-1)
%C2
k2 × 103 (s-1)
%C3
k3 × 104 (s-1)
Ni(II) Cu(II) Zn(II)
3.2 × 10-7 3.2 × 10-7 3.2 × 10-7
41 ( 0.5 55 ( 1
1.7 ( 0.4 2.9 ( 0.4
36 ( 1 47 ( 4 37 ( 3
2.0 ( 0.3 0.9 ( 0.2 3.9 ( 0.1
23 ( 1 51 ( 4 8(1
1.5 ( 0.1 1.0 ( 0.2 2.7 ( 0.1
a Measured by ICP-MS, pH 5.0 ( 0.1, DOC ) 7.7 mg/L, ionic strength 4.9 × 10-3 mol/L. -, not observed from the nonlinear regression analysis. k1-k3, dissociation rate coefficients of the kinetically distinguishable components. %C1-%C3, percentage concentrations of the first-third components, respectively. The uncertainties represent the 95% confidence limits of the nonlinear regression analysis.
TABLE 5. Dissociation Kinetics of Metal Complexes in Onaping River Water Sample, Collected on October 27, 1998, from the Sudbury (ON, Canada) Areaa metal
CM,total (mol/L)
%C1
k1 × 102 (s-1)
%C2
k2 × 103 (s-1)
%C3
k3 × 105 (s-1)
Co(II) Ni(II) Cu(II) Zn(II)
3.6 × 10-8 1.5 × 10-6 1.3 × 10-7 7.2 × 10-8
40 ( 1 33 ( 1 32 ( 1
4.5 ( 0.2 1.7 ( 0.1 3.0 ( 0.3
33 ( 1 30 ( 1 11 ( 1 32 ( 1
1.7 ( 0.1 1.6 ( 0.1 1.1 ( 0.1 9.5 ( 0.3
28 ( 1 38 ( 1 89 ( 1 36 ( 1
8.4 ( 0.1 4.7 ( 0.1 4.9 ( 0.1 9.2 ( 0.1
a Measured by ICP-MS, pH 6.8 ( 0.1, DOC ) 4.9 mg/L, ionic strength 1.5 × 10-3 mol/L. -, not observed from the nonlinear regression analysis. k1-k3, dissociation rate coefficients of the kinetically distinguishable components. %C1-%C3, percentage concentrations of the first-third components, respectively. The uncertainties represent the 95% confidence limits of the nonlinear regression analysis.
TABLE 6. Dissociation Kinetics of Metal Complexes in Moose Lake Water Sample, Collected on October 27, 1998, from the Sudbury (ON, Canada) Areaa metal
CM,total (mol/L)
%C1
k1 × 103 (s-1)
Co(II) Ni(II) Cu(II) Zn(II)
3.2 × 10-8 9.0 × 10-7 1.1 × 10-7 1.1 × 10-7
80 ( 1 93 ( 1 8.4 ( 0.1 64 ( 1
3.1 ( 0.1 2.7 ( 0.1 2.0 ( 0.1 2.0 ( 0.1
%C2
k2 × 105 (s-1)
19 ( 1 Ni(II) d8 > Cu(II) d9 < Zn(II) d10. Since the observed rate coefficients for the adjunctive pathway are strongly influenced by the nature of the arbitrarily chosen probe ligand, no general trend can be predicted from the adjunctive pathway. Hence, the results suggest that the disjunctive pathway probably predominates and that eq 3 is probably valid for the fulvic acid model solutions and the three freshwater systems studied. The significance is that the observed rate coefficients in the test solution can be reliably related to the dissociation rate coefficients of the metal complexes in the natural water samples because the relative rates tend to remain unchanged (25). By contrast, results for the adjunctive pathway cannot be attributed to natural water samples because of the influence of the competing ligand on the experimental results. The disjunctive pathway was also found to predominate in kinetic studies of the dissociation of Ni(II) complexes in model solutions of fulvic acid at sufficiently high concentrations of DOM (20). The slow complexation kinetics of Ni(II) and Cu(II) suggests that the distribution of Ni(II) and Cu(II) in the aquatic environment may be determined by the slow exchange rates between Ni(II) or Cu(II) complexes and the competing metals and ligands, involving simultaneous ligand- and metal-exchange reactions of four reacting species, i.e., double-exchange reactions (12). The systems may reach equilibrium only very slowly. The ability of the kinetic model to correctly identify the relative rates of metal complex dissociation in both model solutions and freshwaters supports the conclusion that the speciation model has chemical significance. The practical importance of this finding is that it is possible to extrapolate results from model solutions to freshwater systems because the relative rates are correctly diagnosed. The results from this work have a direct bearing on predicting free metal ion availability of the above metals and, hence, their bioavailability (43, 44) in the freshwater environment.
Acknowledgments Financial support was received from Nickel Producers Environmental Research Association (USA), Inco Ltd., and Falconbridge Ltd. Research grants from the Natural Sciences and Engineering Research Council of Canada (NSERC), the NSERC Metals in the Environment-Research Network,
(1) Fillela, M.; Town, R.; Buffle, J. In Chemical Speciation in the Environment; Ure, A. M., Davidson, C. M., Eds.; Blackie Academic & Professional: London, UK, 1995; pp 160-200. (2) Hughes, K.; Meek, M. E.; Newhook, R.; Chan, P. K. L. Regul. Toxicol. Pharmacol. 1995, 22, 213. (3) Xue, H.-B.; Kistler, D.; Sigg, L. Limnol. Oceanogr. 1995, 40, 1142. (4) Mandal, R.; Sekaly, A. L. R.; Murimboh, J.; Hassan, N. M.; Chakrabarti, C. L.; Back, M. H.; Gre´goire, D. C.; Schroeder, W. H. Anal. Chim. Acta 1999, 395, 323. (5) Kinniburgh, D. G.; Tipping, E. Environ. Sci. Technol. 2001, 35, 2049. (6) Christl, I.; Kretzschmar, R. Environ. Sci. Technol. 2001, 35, 2505. (7) Mandal, R.; Salam, M. S. A.; Murimboh, J.; Hassan, N. M.; Chakrabarti, C. L.; Back, M. H.; Gre´goire, D. C. Environ. Sci. Technol. 2000, 34, 2201. (8) Langford, C. H.; Cook, R. L. Analyst 1995, 120, 591. (9) Anderson, M. A.; Morel, F. M. M. Limnol. Oceanogr. 1982, 27, 789. (10) Sunda, W. G.; Guillard, R. R. J. Mar. Res. 1976, 34, 511. (11) Anderson, D. M.; Morel, F. M. M. Limnol. Oceanogr. 1978, 23, 283. (12) Hering, J. G.; Morel, F. M. M. In Aquatic Chemical Kinetics, Reaction Rates of Processes in Natural Waters; Stumm, W., Ed.; John Wiley: New York, 1990; pp 145-171. (13) Cabaniss, S. E. Environ. Sci. Technol. 1990, 24, 583. (14) Sekaly, A. L. R.; Mandal, R.; Hassan, N. M.; Chakrabarti, C. L.; Back, M. H.; Gre´goire, D. C.; Schroeder, W. H. Anal. Chim. Acta 1999, 402, 211. (15) Olson, D. L.; Shuman, M. S. Geochim. Cosmochim. Acta 1985, 49, 1371. (16) Lu, Y.; Chakrabarti, C. L.; Back, M. H.; Gre´goire, D. C.; Schroeder, W. H.; Szabo, A. G.; Bramall, L. Anal. Chim. Acta 1994, 288, 131. (17) Lu, Y. Ph.D. Thesis, Carleton University, Ottawa, Ontario, Canada, 1995. (18) Morel, F. M. M.; Hering, J. G. Principles and Applications of Aquatic Chemistry; John Wiley: New York, 1993; pp 398-403. (19) Hering, J. G.; Morel, F. M. M. Environ. Sci. Technol. 1990, 24, 242. (20) Celo, V.; Murimboh, J.; Salam, M. S. A.; Chakrabarti, C. L. Environ. Sci. Technol. 2001, 35, 1084. (21) Chelex 100 and Chelex 20 Chelating Ion Exchange Resin Instruction Manual; Bio-Rad: Hercules, CA; p 13. (22) Tramonti, V. M.Sc. Thesis, Concordia University, Montreal, Canada, 1988. (23) Wang, Z. D.; Pant, B. C.; Langford, C. H. Anal. Chim. Acta 1990, 232, 43. (24) Marquardt, D. W. J. Soc. Ind. Appl. Math. 1963, 11, 431. (25) Langford, C. H.; Gutzman, D. W. Anal. Chim. Acta 1992, 256, 183. (26) Sekaly, A. L. R.; Back, M. H.; Chakrabarti, C. L.; Gre´goire, D. C.; Lu, J. Y.; Schroeder, W. H. Spectrochim. Acta 1998, 53B, 837. (27) Sekaly, A. L. R.; Back, M. H.; Chakrabarti, C. L.; Gre´goire, D. C.; Lu, J. Y.; Schroeder, W. H. Spectrochim. Acta 1998, 53B, 847. (28) Sekaly, A. L. R. Ph.D. Thesis, Carleton University, Ottawa, Canada, 2000. (29) van Leeuwen, H. P. Electroanalysis 2001, 13, 826, (30) Buffle, J. Complexation Reactions in Aquatic Systems: An Analytical Approach; Ellis Horwood: Chichester, UK, 1988; p 320. (31) Gamble, D. S.; Schnitzer, M.; Kerndorff, H.; Langford, C. H. Geochim. Cosmochim. Acta 1983, 47, 1311. (32) Margerum, D. W.; Cayley, G. R.; Weatherburn, D. C.; Pagenkopf, G. K. In Coordination Chemistry, Vol. 2; Martell, A. E., Ed.; ACS Monograph; American Chemical Society: Washington, DC, 1978; pp 1-220. (33) Wilkins, R. G. Kinetics and Mechanism of Reactions of Transition Metal Complexes; VCH: New York, 1991. (34) Xue, H.-B.; Jansen, S.; Prasch, A.; Sigg, L. Environ. Sci. Technol. 2001, 35, 539. (35) Town, R. M.; Filella, M. J. Electroanal. Chem. 2000, 488, 1. (36) Sekaly, A. L. R.; Mandal, R.; Hassan, N. M.; Murimboh, J.; Chakrabarti, C. L.; Back, M. H.; Gre´goire, D. C.; Schroeder, W. H. Anal. Chim. Acta 1999, 402, 211. VOL. 37, NO. 1, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
73
(37) Achterberg, E. P.; van Eltern, J. T.; Kolar, Z. I. Environ. Sci. Technol. 2002, 36, 914. (38) Mandal, R.; Hassan, N. M.; Murimboh, J.; Chakrabarti, C. L.; Back, M. H.; Rahayu, U.; Lean, D. R. S. Environ. Sci. Technol. 2002, 36, 1477. (39) Buffle, J.; Altmann, R. S.; Filella, M.; Tessier, A. Geochim. Cosmochim. Acta 1990, 54, 1535. (40) Douglas, B.; McDaniel, D.; Alexander, J. Concepts and Models of Inorganic Chemistry; John Wiley: New York, 1994; pp 407408, 487-489. (41) Butler, I. S.; Harrod, J. F. Inorganic Chemistry Principles and Applications; Benjamin/Cummings: Redwood City, CA, 1989; p 422.
74
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 1, 2003
(42) Langford, C. H. In Ionic Interactions, Vol. 2; Petrucci, S., Ed.; Academic Press: New York, 1971; p 16. (43) Morel, F. M. M. Principles of Aquatic Chemistry; John Wiley: New York, 1983; pp 300-304. (44) Campbell, P. G. C. In Metal Speciation and Bioavailability in Aquatic Systems; Tessier, A., Turner, D. R., Eds.; IUPAC Series on Analytical and Physical Chemistry of Environmental Systems; John Wiley: Chichester, UK, 1995; pp 45-102.
Received for review May 19, 2002. Revised manuscript received September 30, 2002. Accepted October 16, 2002. ES025805G
