Selective Electrochemical Detection of Ionic and Neutral Species

Indiana University Northwest, Gary, Indiana 46408. An electroanalytical method has been developed to investigate the uptake of redox-active species by...
0 downloads 0 Views 338KB Size
Environ. Sci. Technol. 2004, 38, 2161-2166

Selective Electrochemical Detection of Ionic and Neutral Species Using Films of Suwannee River Humic Acid K . V I N O D G O P A L , * ,† V A I D Y A N A T H A N S U B R A M A N I A N ‡, A N D PRASHANT V. KAMAT‡ Radiation Laboratory and Department of Chemical & Biomolecular Engineering, University of Notre Dame, Notre Dame, Indiana 46556, and Department of Chemistry, Indiana University Northwest, Gary, Indiana 46408

An electroanalytical method has been developed to investigate the uptake of redox-active species by the humic acid substances. The Suwannee River humic acid (SHA) films were first cast on a glassy carbon electrode using an electrophoretic approach. The binding of a series of redox-active species to these SHA films was then probed using cyclic voltammetry at a rotating disk electrode. Neutral molecules such as hydroquinone and cationic species such as methyl viologen are able to bind with the humic membrane and exhibit high redox activity within the film. On the other hand, anionic species such as ferrocyanide are unable to attach themselves to the SHA films and thus exhibit negligible electrochemical activity. Cyclic voltammetric study of SHA films has also facilitated the determination of the partitioning constants and identification of the effect of coadsorbed ions (Ca2+) on the binding of redox species. The electroanalytical method described in this study opens up new avenues to examine the interactions and transport of charged species in a humic acid environment.

Introduction Humic substances (HSs) are naturally occurring complex polymeric oxidation products of decaying plant and animal wastes. The study of these materials is important from an environmental perspective because they are abundantly found around the world in soil and fresh waters. The interaction of HSs with various environmentally significant chemical moieties is dependent upon the pH, ionic strength, temperature, and other coexisting species. These parameters influence the interaction of the various functional groups of the humic substances with the chemicals in the environment. Evaluation and proper understanding of such interactions between HSs and other chemicals is essential for the proper application and utilization of humic materials for environmental remediation and pollution abatement.1,2 The complexity of humic materials makes this task even more difficult. Advances in characterizing techniques such as luminescence measurements and pulsed NMR methods have led to a greater understanding of the HS interactions with different materials in solutions.3,4 The focus of these studies has been * Corresponding author phone: (219) 980-6688; fax: (219) 9806673; e-mail: [email protected]. † Indiana University Northwest. ‡ University of Notre Dame. 10.1021/es034988j CCC: $27.50 Published on Web 03/04/2004

 2004 American Chemical Society

the elucidation of structural information on the humic acid. In an earlier paper, we presented details of a simple and novel electrophoretic method to deposit humic films on conducting glass electrodes.5 By controlling the applied voltage and the concentration of the solution, we succeeded in obtaining films from Suwannee River humic acid (SHA) of varying thickness and porosity. Such SHA films are appropriate for studying mass-transfer behavior as well as the intercalation of dissolved chemical species. Electrophoretically deposited SHA films are therefore useful for evaluating the interaction of various ionic species found in the environment and possible use of humic films as natural ion exchange membranes. A variety of groups have modified electrode surfaces with polymer films to produce materials for numerous applications including electrocatalysis and corrosion prevention. On the basis of experiments performed in the early 1980s by Anson et al.,6,7 Bard et al,8-10 and Murray et al.,11-14 we can generally classify these polymer-coated electrodes into two types: (A) those containing redox-active groups on the backbone of the polymer14 and (B) those in which redox-active ions are bound electrostatically to polar groups on the polymer. Oyama and Anson6,7 used redox polymers (type A) derived from poly(4-vinylpyridine) to coat electrodes and showed that such an arrangement resulted in almost ideal Nernstian responses. The type B electrodes however show deviations from an ideal Nernstian response arising from a host of factors including the thickness of the film and limitations in the rate of the redox process due to the film. Ghantous et al.15 have used a membrane humic acid modified carbon electrode to study the incorporation of metal ions and the organic compound methyl viologen in humic acids. In this arrangement, the humic acid is sandwiched between the electrode surface and a dialysis membrane. The redox species under examination pass through the membrane and become incorporated into the material. Diffusion of the species to the electrode surface is a function of the interactions with the humic material in the entrapped region. Several groups have examined the permeability, ion transport, and equilibrium in layered polyelectrolyte films.16-19 In most such studies, the experimental technique of choice to obtain a direct measure of the flux of the probe ion has been cyclic voltammetry at rotating disk electrodes coated with the polyelectrolyte. Polyelectrolyte membranes have been shown to have selectivity for ions of low charge, and such membranes have been tailored for specific applications. Our expectation in this case is that the SHA films should likewise show differences in permeability between different species. SHA has a large negative surface potential20 and therefore can be deposited as a film on a conducting surface by application of an external dc voltage.5 We show in this paper that these SHA films can be deposited on a variety of electrode surfaces including glassy carbon (GC) electrodes, thereby giving us the option to use electrochemical methods such as rotating disk electrode (RDE) cyclic voltammetry to explore the intercalation and transport of redox-active species across such humic films. We have exploited the polyelectrolyte nature of SHA1 and the relative ease of making SHA films by electrophoretic deposition to investigate the interaction of a series of redox-active species (both ionic and neutral) with such humic films. The interaction of three representative species, viz., methyl viologen (MV2+), hydroquinone (HQ), and ferrocyanide ion (Fe(CN)64-), with humic films has been the focus of this study. These three species have been chosen because (a) they are electroactive species with wellVOL. 38, NO. 7, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

2161

FIGURE 1. (a) AFM image of SHA particles assembled as thin films on a conducting glass electrode in an aqueous solution at pH 2.0. The films were prepared from an SHA solution of concentration 190 mg/L, with 50 V dc and a 1 min deposition time. The AFM image in the fluid media was recorded in the tapping mode using a silicon nitride probe. The parameters used for the AFM measurements in the Nanoscope IIIa were drive amplitude 1400 mV, amplitude setpoint 1.2 V, and drive frequency 8.9 kHz. (b) Sectional view indicating the contours of the surface of a cross-section of the SHA film on a conducting glass electrode. (c) 3-D AFM image of the SHA film shown in (a). characterized redox behavior and (b) they represent in turn cationic, neutral, and anionic chemical species whose interaction with humic membranes can shed light on the complex polyelectrolyte nature of the latter.

Methods Reference SHA was obtained from the International Humic Substances Society (IHSS). Solutions of SHA in a mixture of 4% ethanol and 96% acetonitrile at a concentration of 190300 mg/L were used for the electrophoretic deposition. Films of SHA were made by electrophoretic deposition on glassy carbon and a conducting glass electrode obtained from Pilkington, Libbey Owens Ford, Ohio. Details of the deposition process can be found elsewhere.5,21,22 The extent of SHA deposition can be controlled by varying the SHA solution concentration, applied dc voltage, and time of deposition. Typically the films were deposited from SHA solutions (190 mg/L) of a 4% ethanol/96% acetonitrile mixture at an applied voltage of 200 V for 20 min. The choice of solvent was a function of the dissolution of the SHA and the need for a solvent of low dielectric constant, thereby ensuring that the applied voltage can be kept to a minimum during the deposition process. Cyclic voltammetry studies were performed on SHA electrophoretically deposited on glassy carbon electrodes using a Bio-analytical System 100 W electrochemical analyzer. The SHA film deposited on the glassy carbon was used as a working electrode. A platinum wire and saturated calomel electrode were used as auxiliary and reference electrodes, respectively. Rotating disk experiments were performed with a Pine Instrument Co. RDE assembly. The cell used was a standard three-arm electrochemical cell for the electrodes. 2162

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 7, 2004

Unless otherwise specified the electrolyte used was an aqueous solution of the target redox species: hydroquinone, methyl viologen (MV2+), or ferrocyanide ion maintained at pH 2.0 using HClO4. Metal effects were studied using aqueous solutions of calcium chloride (10 mM) adjusted to pH 2.0. The surface morphology of SHA films deposited on a conducting glass electrode (1 cm × 1 cm size) was analyzed using a digital Nanoscope IIIa atomic force microscope. Characterization of the surface was carried out in the tapping mode in a Digital Instruments liquid cell using triangular probes with silicon nitride tips (Nanoprobe NP-S-20 contact mode tips). The liquid cell can hold ∼0.3 mL of an aqueous solution (adjusted to pH 2.0 with HClO4) inside an O-ring.

Results and Discussion Electrophoretic Deposition of SHA Films on a GC Electrode. Electrochemistry of adsorbed species is conveniently characterized on rotating disk electrodes, which are usually GC electrodes enclosed in an insulating sleeve. Deposition of the SHA on such a GC surface was carried out by immersing the carbon electrode in the SHA solution while a conducting glass electrode kept parallel to the GC electrode surface acted as the negative terminal. The SHA is deposited exclusively on the positively charged GC electrode. Deposition was made from an SHA solution of 190 mg/L for 20 min at 100 V. The resulting SHA film that was obtained on the GC electrode was stable at low pH (pH < 3), and its robustness can be tested at rotations up to 10000 rpm in the rotating disk assembly. Such an SHA-coated RDE was used as the working electrode in the subsequent voltammetric analysis and dipped in the electrolyte of interest, viz., an aqueous solution at pH 2.0 of Fe(CN)64-, HQ, or MV2+.

CHART 1. SHA-Coated Rotating Disk Electrode in (a) Water, (b) Water Containing Hydroquinone at 0 rpm, and (c) Water Containing Hydroquinone at 6000 rpma

a The partitioning between aqueous and humic phases can be monitored by decreasing the diffusive layer.

FIGURE 2. (A, top) Linear scan voltammogram of 0.13 mM hydroquinone in 100 mL of deionized water (adjusted to pH 2 with 0.1 M HClO4 solution at a stationary 6 mm diameter glassy carbon electrode: (a) uncoated and (b) coated with an SHA film. Scan rate: 50 mV/s. (B, bottom) Linear scan voltammogram of 0.13 mM hydroquinone in 0.1 M HClO4 solution at a rotating disk electrode: (a, b) GC electrode coated with an SHA film at 0 and 6000 rpm, respectively; (c) bare electrode. The electrode was 6 mm diameter glassy carbon. Scan rate: 50 mV/s. Surface Morphology of SHA Films in Liquid. We have examined the morphology of the SHA film on a conducting glass electrode while immersed in an aqueous solution at pH 2.0 using tapping mode atomic force microscopy (AFM). Figure 1a shows a representative tapping mode AFM image. The film is an assembly of small SHA particulates with the SHA particles ranging in height from 50 to 125 nm. The reproducibility of the image was confirmed by monitoring the topography profiles at more than three different locations and with different samples. Tip-induced sample deformations are minimal in the tapping mode. Blank imaging of an OTE in an aqueous solution maintained at pH 2.0 does not show any characteristic particles similar to those observed with the humic film. Figure 1b shows the sectional view of the humic membrane. The 3-D image shown in Figure 1c shows the close network of humic acid clusters that provide a large surface area for interaction with the redox species present in solution. We estimate the thickness of the film to be in the range of 0.5-1.0 µm. Such a nanostructured morphology of the SHA film is likely to play an important role in controlling the mobility of redox species within the humic film. Electrochemistry of HQ at SHA/RDE. The cyclic voltammograms of hydroquinone at a bare RDE and at an SHAcoated RDE at 0 rpm are shown in Figure 2A. The anodic and cathodic peaks represent oxidation of HQ to quinone (Q) in the forward scan (0 f 1.2 V) and reduction of quinone to HQ in the reverse scan (1.2 f 0 V) (eq 1). The mobility of the

C6H4O2 + 2H+ + 2e- T C6H4(OH)2

(1)

hydroquinone or quinone toward a bare GC electrode surface

occurs with a mass-transfer limitation across the diffusion layer, the thickness of which is controlled by the rotation rate of the GC electrode. On the other hand, with an SHAcoated electrode, HQ has to permeate through the SHA film to undergo a redox process at the electrode. The difference between the redox peaks of HQ at an SHA-modified RDE is smaller than that recorded on a bare RDE (∆Ep ) 50 mV). This suggests that most of the HQ is bound to the SHA film and the observed current is a function of the HQ bound to the HA. For an ideal Nernstian reaction under Langmuir isotherm conditions, Epa (anodic peak potential) for a surfacebound species should be equal to Epc (the cathodic peak potential) or ∆Ep ) 0 V. Deviations from this ideal behavior suggest the presence of a film resistance and/or diffusion of species within the SHA film. Confirmation that the hydroquinone is bound to the SHA film was further ascertained from cyclic voltammetry of the HQ by spinning the electrode (Figure 2B). For the SHA-coated RDE, we observe only a surface wave. This is in contrast to the diffusion-limited waves (curve c in Figure 2B) observed for HQ with an increased scan rate at the bare RDE. The peak current arising from the SHA-bound HQ at 6K (curve b in Figure 2B) remains more or less as the one observed at 0K (curve a in Figure 2B), and its magnitude remains relatively small at about 7µA. The failure to observe a diffusion-limiting current in the SHA-coated electrode arising from the oxidation of HQ in solution confirms that the HQ is bound strongly to the SHA film and HQ from solution does not permeate through the SHA film. Chart 1 gives an illustration of the binding of the HQ to the SHA film at an RDE. The limiting current arising from diffusion across a diffusion layer observed between the solution and a bare RDE is governed by the Levich equation:23

ilim ) 0.620nFADaq2/3ω1/2υ-1/6Caq

(2)

where A is the area of the electrode rotating at an angular velocity ω, n is the number of electrons transferred, F is Faraday’s constant, Daq is the solution diffusion constant of the redox species, υ is the kinematic viscosity of the electrolyte, and Caq is the solution concentration. As can be seen from eq 2, the limiting current arising from diffusion of species in solution is proportional to the square root of the rotation rate. In fact as shown in Figure 3, curve a, HQ at a bare RDE shows the expected linear relationship between the limiting current and the square root of the rotation rate for the bare RDE. However, the current at the SHA-coated electrode remains constant at different rotation rates (Figure 3, curves b-e). As the concentration of HQ in solution is increased, the peak current arising from the SHA-bound HQ increases, but at any given concentration, this peak current VOL. 38, NO. 7, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

2163

Another interesting feature is the shape of the voltammogram, which resembles more a diffusion-limited behavior, possibly arising from HQ diffusion within the film. Please note that this diffusion of HQ within the film is different from the one observed from solution at the bare electrode. The observed voltammograms on the SHA-coated electrode can be analyzed as a function of the HQ concentration in solution to determine its relationship to the surface-bound HQ. Since the peak current (ip at 0.4V) arises entirely from HQ in the film, it is proportional to the concentration of surface-bound HQ:

ip ∝ [HQ]bound FIGURE 3. Levich plots of current vs rotation rate for HQ at a bare GC electrode (curve a) and at an SHA-coated GC electrode as a function of decreasing HQ concentration (curves b-e).

The ratio of the peak current to the HQ concentration in solution is a measure therefore of the partition coefficient of HQ between film and solution. Since the solution concentration of HQ is proportional to the peak current at a bare GC electrode, we can equate

partition coefficient ) ip(HQ@SHA)/ip(HQ@bare)

FIGURE 4. (A, top) Linear scan voltammogram of hydroquinone as a function of concentration at a rotating disk electrode coated with an SHA film. All HQ solutions were in 100 mL of water acidified to pH 2 with 0.1 M HClO4. The electrode was 6 mm diameter glassy carbon rotated at 6000 rpm with a scan rate of 50 mV/s. (B, bottom) A plot of the peak current (ip) vs concentration of HQ in solution for the oxidation of HQ at a bare GC electrode (0) and at an SHAcoated GC electrode (b) at 0 rpm. The inset shows the expanded plot for the concentration region below 0.3 mM. remains quite steady and constant as a function of the rotation rate. This provides confirmation of the surface origin of this redox peak. As the concentration of HQ is increased, there is a significant increase in the currents but no change in the position of the surface-bound peak. Figure 4A shows the voltammograms of three different concentrations of HQ at an SHA-coated RDE at 6000 rpm. The overall shape of the cyclic voltammogram is dependent on the concentration of HQ in solution and not on the rotation rate (Figure 4A), thus confirming the surface origin of the HQ. We can conclude that the primary result of increasing the HQ concentration in solution is that more of it is bound to the SHA film. 2164

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 7, 2004

(3)

(4)

We have plotted the peak current at 0 rpm as a function of increasing HQ concentration (Figure 4B) for both the bare and SHA-coated RDEs. The dependence of ip versus HQ concentration is linear for the bare GC electrode, confirming the diffusion-limited electron transfer at the electrode. However, the SHA-film-coated GC electrode shows two distinctive slopes. At low concentrations, the slope of the plot is identical to the one observed at a bare GC electrode (see inset in Figure 4B). This indicates that, at low concentrations (0.3 mM). The exclusive nature of surface-bound voltammetric behavior of HQ further shows that its binding to SHA is quite strong and the film is remarkably resistant to the diffusion of HQ from the solution though the film. On the basis of the interaction of metal ions with humics, a number of groups have concluded that there are two classes of binding sites in humic substances.3,4 Both hydrophobic and electrostatic interactions are likely to play a major role in providing different binding strengths within the film. The two different partition coefficients observed in the present RDE voltammetry experiments support the presence of differing binding sites. Electrochemistry of MV2+ at SHA/RDE. MV2+ undergoes one-electron reversible reduction at -0.68 V vs SCE. In our studies, we have monitored the reduction of MV2+ over the scan range of 0 to -0.8 V.

MV2+ +e- T MV+

(5)

The MV2+ solution was purged with nitrogen for ∼30 min to ensure deaerated conditions prior to scanning. The cyclic voltammogram (Figure 5) of the MV2+ was performed by using a bare RDE and SHA-coated RDE at 0K and 6K rotation rates. As in the HQ system, the classic sigmoidal limiting current is observed when the rotation rate is increased from 0 to 6000 rpm for the bare RDE in a solution of MV2+ (curves c and d). This is attributed to the increased flux of the MV2+ to the RDE as the diffusion layer is decreased at higher rotation

FIGURE 5. Linear scan voltammogram of 0.10 mM MV2+ in an aqueous solution acidified to pH 2.0 with 0.1 M HClO4 at a rotating disk electrode at a scan rate of 50 mV/s: (a, b) RDE coated with an SHA film. The electrode was 6 mm diameter glassy carbon. In (a) the electrode was not rotated, 0 rpm, while in (b) the electrode was rotated at 6000 rpm. (c, d) Bare RDE at 0 and 6000 rpm, respectively.

CHART 2. SHA-Coated Rotating Disk Electrode in Contact with Two Differently Charged Species Representing (a) Binding of Positively Charged MV2+ and (b) the Impenetrability of Fe(CN)64-

FIGURE 6. (A, top) Effect of Ca2+ and H+ ions on the peak current arising from HQ bound to an SHA film on an RDE. Decay of the peak current arising from SHA-bound HQ at an RDE as a function of time after immersion for 10 min in (a) 10 mM CaCl2 solution at pH 2.0 (O) and (b) pH 2.0 water (b). (B, bottom) Effect of Ca2+ and H+ ions on the peak current arising from MV2+ bound to an SHA film on an RDE. Decay of the peak current arising from SHA-bound MV2+ at an RDE as a function of time after immersion for 10 min in (a) 10 mM CaCl2 solution at pH 2.0 (O) and (b) pH 2.0 water (b).

rates. At the SHA-modified RDE, we observe an increase in the reduction current (curves a and b) compared to that at the bare GC. The presence of redox peaks in the voltammogram confirms the binding of MV2+ to the humic membrane. Little change in the oxidation peak current (ip) is observed at higher rotation speeds (6000 rpm, Figure 5). This observation further confirms that the redox process is a surface wave effect observed due to the incorporation of methyl viologen in the film (Chart 2a). The peak current recorded at 0K is higher for the SHA-coated RDE. The increase in the reduction current observed at the SHA-coated electrode shows that the MV2+ preferentially partitions into the SHA film and attains a higher concentration in the film than in solution. The methyl viologen is preferentially adsorbed in the SHA film in much the same way as other charged dyes have been incorporated into Nafion polymer films.8-10,24 Ghantous et al. have observed a similar enrichment of methyl viologen in their membrane humic acid carbon electrode.15 Effect of Calcium Ions on Binding of HQ and MV2+ to SHA. The apparent strength of the binding of the two species HQ and MV2+ with SHA does raise the question of whether these species can be displaced from the SHA film under appropriate conditions. Positively charged metal ions such as Ca2+ are known to complex with SHA easily and therefore could displace species from the SHA film.25 This displacement in the case of HQ and MV2+ would manifest itself in decreasing oxidation currents. Our experimental procedure in this case is to use the humic film incorporated with either the hydroquinone or the methyl viologen. The cyclic voltam-

mogram of these SHA films with the bound HQ or MV2+ was monitored at regular intervals of time after immersion in two different electrolytes: (a) a degassed dilute solution of perchloric acid at pH 2.0 and (b) a degassed 10 mM calcium ion solution at pH 2.0. With continued exposure to Ca2+ ions, we expect HQ or MV2+ from the film to diffuse into the solution and be replaced by calcium ions till equilibrium is established between the species in the film and solution. Under these conditions, the decrease in the peak current reflects a decrease in the bound species with increasing time. Figure 6 shows the peak current for HQ observed at 0.33 V and for MV2+ observed at -0.77 V as a function of time in both the electrolyte solutions given above. For both HQ and MV2+ the peak current decreases over time, with the decrease being faster in the case of calcium ions as compared to H+. This is not unexpected given the ability of Ca2+ to complex with humic acid. More interesting however are the differences in the oxidative current decay rate between the neutral HQ and the cationic MV2+ in both electrolytes. While a 40% decrease in peak current is observed with neutral HQ species in 5 min as the Ca2+ displaces the HQ, a similar 40% decrease in the case of MV2+ takes nearly 20 min. The results from these experiments confirm both the reversible nature of the binding between HQ and MV2+ to the SHA film and the electrostatic nature of the interaction. The reversible nature of the interaction between SHA binding sites and the adsorbed species was confirmed from the recovery of the oxidation peak following the immersion of the Ca2+-bound SHA film back into the HQ or MV2+ solution. These observations confirm that the binding of these VOL. 38, NO. 7, 2004 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

2165

The binding of redox species and their interaction with humic films have been probed using rotating disk voltammetry. We have shown that electrostatic and hydrophobic interactions play a major role in binding the redox species to the humic film. Neutral HQ molecules bind strongly to the SHA film depending upon their concentration in the surrounding solution. Cations such as MV2+ are preferentially incorporated into the SHA membranes. On the other hand, ferrocyanide ions fail to penetrate or bind to the SHA membrane. Such a screening property demonstrated by the SHA films can have potential applications in incorporating selectivity in environmental waste treatment using humic membranes. FIGURE 7. Linear scan voltammogram of 0.13 mM ferrocyanide solution in an aqueous solution acidified to pH 2.0 with 0.1 M HClO4 at a rotating disk electrode at 6000 rpm: (a) uncoated, (b) coated with an SHA film, (c) coated with an SHA film as in (b) and then dipped in 10 mM CaCl2 solution at pH 2.0 for 3 h. The electrode was 6 mm diameter glassy carbon. species is dependent on the coexisting ionic species. The diffusion of the species from the SHA film in the presence of Ca2+ and H+ ions can be attributed to the varying strength of their electrostatic and or hydrophobic interactions with the SHA membranes. Greater affinity of the bivalent Ca2+ with the negatively charged SHA membrane than the H+ is expected to accelerate the displacement of HQ or the MV2+ at a higher rate from the SHA film. Electrochemistry of Ferrocyanide at SHA/RDE. The experimental observations of the SHA clusters depositing on the positive electrode during electrophoresis and the experimental data from positively charged methyl viologen would suggest that the humic acid film is expected to have negatively charged binding sites. If indeed such an argument is valid, a negatively charged species such as ferrocyanide ion would not show similar binding behavior. Fe(CN)64- undergoes a one-electron reversible redox process as shown in eq 6. The voltammograms at 6000 rpm for the ferrocyanide ion at a bare RDE electrode and an SHA-

Fe(CN)64- T Fe(CN)63- + e-

(6)

coated RDE are shown in Figure 7. At 6000 rpm, the currentvoltage scan for ferrocyanide at the bare RDE displays the classical sigmoidal shape, reflecting the greater mass flux of unoxidized ferrocyanide ions to the electrode surface under the well-defined hydrodynamic conditions of a rotating disk. Higher oxidation currents for ferrocyanide are observed with increased scan rate at the bare RDE. However, in the case of the RDE coated with the SHA at 6000 rpm, the limiting current for ferrocyanide oxidation is completely suppressed as can be seen from Figure 7. In the presence of SHA, the voltammogram is featureless and no redox peaks corresponding to the ferrocyanide ion are observed. Furthermore, the complete absence of a ferrocyanide redox wave for the SHA-coated electrode at both 0 and 6000 rpm suggests that there is no surface-bound ferrocyanide and these ions are unable to permeate through the SHA film (Chart 2b). This screening effect of the SHA film is likely to arise from the negative charge of the SHA film that deters ferrocyanide anions from attaching themselves to the film, or its bulkiness prevents permeation through the humic acid film. If the former is true, we should facilitate diffusion of ferrocyanide by neutralizing the charges with Ca2+ ions. When the SHA-coated electrode was immersed in a 10 mM solution of Ca2+ at pH 2 for 1 h, the voltammogram of Fe(CN)64- at this Ca2+-pretreated-SHA-coated RDE (Figure 7) did not exhibit any redox peaks. This would imply that bulky negative ions such as ferrocyanide neither attach themselves to SHA nor permeate through the film. 2166

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 38, NO. 7, 2004

Acknowledgments The work described herein was supported by the Office of Basic Energy Sciences of the U.S. Department of Energy. K.V. acknowledges the support of Indiana University Northwest through a Grant-In-Aid. This is Contribution No. NDRL 4471 from the Notre Dame Radiation Laboratory.

Literature Cited (1) Understanding humic substances. Advanced methods, properties and applications; Ghabbour, E. A.; Davies, G., Eds.; Royal Society of Chemistry: Cambridge, U.K., 1999. (2) Perdue, E. M. In Aquatic humic substances. Ecology and biochemistry; Hassen, D. O., Tranvik, L. J., Eds.; Springer: Berlin, 1998. (3) Yoon, T. H.; Moon, H.; Park, Y. J.; Park, K. K. Environ. Sci. Technol. 1994, 28, 2139. (4) Shin, H. S.; Rhee, S. W.; Lee, B. H.; Moon, C. H. Org. Geochem. 1996, 4, 523. (5) Vinodgopal, K.; Subramanian, V.; Carrasquillo, S.; Kamat, P. V. Environ. Sci. Technol. 2003, 37, 761. (6) Oyama, N.; Anson, F. C. Anal. Chem. 1980, 52, 1192. (7) Shigehara, K.; Oyama, N.; Anson, F. C. J. Am. Chem. Soc. 1981, 103, 2552. (8) Rubinstein, I.; Bard, A. J. J. Am. Chem. Soc. 1981, 103, 5007. (9) Martin, C. R.; Rubinstein, I.; Bard, A. J. J. Am. Chem. Soc. 1982, 104, 4817. (10) Rubinstein, I.; Bard, A. J. J. Am. Chem. Soc. 1980, 102, 6641. (11) Calvert, J. M.; Schmehl, R. H.; Sullivan, B. P.; Facci, J. S.; Meyer, T. J.; Murray, R. W. Inorg. Chem. 1983, 22, 2151. (12) Nakahama, S.; Murray, R. W. J. Electroanal. Chem. 1983, 158, 303. (13) Surridge, N. A.; Sosnoff, C. S.; Schmehl, R.; Facci, J. S.; Murray, R. W. J. Phys. Chem. 1994, 98, 917. (14) Dalton, E. F.; Murray, R. W. J. Phys. Chem. 1991, 95, 6383. (15) Ghantous, L.; Lojou, E.; Bianco, P. Electroanalysis 1998, 10, 1249. (16) Farhat, T.; Schlenoff, J. B. Langmuir 2001, 17, 1184. (17) Schlenoff, J. B.; Dubas, S. T.; Farhat, T. Langmuir 2000, 16, 9968. (18) Dubas, S. T.; Schlenoff, J. B. Langmuir 2001, 17, 7725. (19) Harris, J. J.; Bruening, M. L. Langmuir 2000, 16, 2006. (20) Green, S.; Morel, F.; Blough, N. Environ. Sci. Technol. 1992, 26, 294. (21) Chandrasekharan, N.; Kamat, P. V. Nano Lett. 2001, 1, 67. (22) Kamat, P. V.; Barazzouk, S.; Hotchandani, S. Adv. Mater. 2001, 13, 1614. (23) Bard, A. J.; Faulkner, L. R. Electrochemical Methods. Fundamentals and Applications; John Wiley and Sons: New York, 1980. (24) Gopidas, K. R.; Kamat, P. V. Proc. Ind. Acad. Sci. (Chem. Sci.) 1993, 105, 505. (25) Stevenson, F. J. Humus Chemistry: Genesis, Composition, Reactions, 2nd ed.; John Wiley and Sons: New York, 1994.

Received for review September 8, 2003. Revised manuscript received December 22, 2003. Accepted January 23, 2004. ES034988J