Environ. Sci. Technol. 2000, 34, 3180-3185
Interaction Kinetics of I2(aq) with Substituted Phenols and Humic Substances J E F F R E Y A . W A R N E R , * ,† W I L L I A M H . C A S E Y , †,‡ A N D RANDY A. DAHLGREN† Department of Land, Air and Water Resources and Department of Geology, University of California, Davis, California 95616
We evaluate the hypothesis that the reactivity trend for iodination of natural humic substances (HS) resembles that for the iodination of some substituted phenols. The hypothesis was tested by comparing the rates of reaction of I2(aq) with HS and a series of eight substituted phenolic compounds. Rates of iodination for all of the phenolic compounds, except salicylate, are described with an empirical rate law R ) kobs[phenolic compound]1[I3-]1[[H+]-1[I-]-2] with the values of kobs related to the structure of the substituted phenol. The values of kobs, corresponding to iodination of the simple substituted phenols, range from 5.6 × 10-8 to 4.7 × 10-5 M s-1 at 25 °C. These rate coefficients can be predicted over at least three orders-of-magnitude from a modified Hammett relation. The rates of iodination of HS fall within the range measured for substituted phenols, suggesting that iodination of the natural HS proceed by similar pathways. The humic substances differ markedly in their reactivity toward I2(aq) in several important ways. First, unlike the substituted phenols, the HS typically reacted in two stages. An initial stage involves rapid uptake of I2(aq) and is followed by a much slower reaction. Surprisingly, each stage of the reaction follows a similar rate law with respect to the reactants. Second, rates of iodination of the HS are characterized by noninteger rate orders with respect to the concentration of protons and the concentration of dissolved iodide [I-(aq)].
Introduction The interaction between I2(aq) and humic substances (HS) is important due primarily to the radioactive isotopes of iodine, 129I (t1/2 ) 1.6 × 107 years) and 131I (t1/2 ) 8 days). They are released to the environment through nuclear low-level waste disposal, fuel reprocessing, and nuclear weapons testing (1, 2). The transport and fate of radioiodine, after release to the environment, is determined by its tendency to be adsorbed onto minerals in backfill material surrounding the waste or by minerals in the rock, its solubility in groundwater, and its volatility. Iodine-129, in the form of I-(aq), tends to migrate readily with water through soils because it forms no insoluble salts in natural waters and * Corresponding author phone: (650)723-4152; fax: (650)725-2199; e-mail:
[email protected]. Present address: Department of Geological and Environmental Sciences, Stanford University, Stanford, CA 94305-2115. † Department of Land, Air and Water Resources. ‡ Department of Geology. 3180
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 34, NO. 15, 2000
does not significantly sorb to any common soil minerals, as suggested by its very low partition coefficient (Kd) of 0.1-1.0 mL/g onto clays and many soil minerals (3). As a result, iodine has a high estimated solubility under various conditions of redox potential and pH in the environment (4). Complicating predictions of reactivity for iodine in the environment is the likelihood that iodide (I-(aq)) in soils is transformed to the I2(aq) species through oxidation by Mn(IV) and Fe(III)-containing mineral phases (5). Iodide oxidation is potentially important because it leads to the formation of the more volatile and reactive electrophilic species I2(aq). There is evidence that the subsequent fate of I2(aq) is dominated by reactions with soil organic matter (6-8). Here, we test the hypothesis that uptake of I2(aq) by HS exhibits a reactivity trend similar to the iodination of substituted phenolic compounds. We reason that an important reactive component of HS are activated aromatics, such as the phenolic constituents that make up about 20-30 wt % of humic and fulvic acids (9). The phenolic constituents would vary considerably in reactivity toward I2(aq) depending upon the nature and location of substituents on the aromatic ring. Therefore we determined the rates for iodination of a series of substituted phenols (phenol, 3-methylphenol (m-cresol), 4-methylphenol (p-cresol), 2-methoxyphenol (guaiacol), 3-methoxyphenol, 3-hydroxyphenol (resorcinol), 2-hydroxybenzoate (salicylate), and 4-hydroxybenzoate) and compared these to the rates of iodination of four HS. The four humic materials include the following: (i) an unpurified, soil-derived humic acid (HA), (ii) a soil fulvic acid (FA) from the International Humic Substances Society (IHSS), (iii) an IHSS soil HA, and (iv) an IHSS Suwannee River FA.
Experimental Section Reagents including p-cresol, m-cresol, 3-methoxyphenol, guaiacol, resorcinol, 4-hydroxybenzoic acid, sodium iodide dihydrate, anhydrous sodium thiosulfate, sodium perchlorate, potassium iodate, potassium hydrogen phthalate, and methylene chloride were obtained from Fisher Scientific and used without further purification. Salicylic acid (Aldrich, 99+%) was used as purchased. Molecular iodine, I2 (Baker), was doubly sublimed prior to use (10). All solutions were prepared with distilled, deionized (DD) water (18 MΩ). Iodine solutions were standardized by titration using sodium thiosulfate. Sodium thiosulfate was standardized against a primary standard of potassium iodate using standard methods (11). Humic Substances. Soil FA (1S102F), soil HA (1S102H), and Suwannee River FA (1R101F) were purchased from the International Humic Substances Society (IHSS) and used without further purification. In addition, a soil HA was isolated from the surface horizon of a Mudwell soil series (an Inceptisol (12)), collected in the vicinity of Mt. Shasta, using the procedure of Schnitzer (9). Ash contents were determined on the extracted humic acid using a dry-ashing method for 6 h at 600 °C (13). Carbon content was measured on the same sample by combustion (1020 °C) using a C/N analyzer (Carlo-Erba). The dissolved organic carbon content of all HS solutions used in kinetic investigations was measured with a Phoenix 8000 Dohrmann analyzer utilizing UV/persulfate oxidation and infrared detection. To estimate the phenolic group content of the HS used in this investigation, total acidity was measured using the barium hydroxide (Ba(OH)2) method (14). Acidity from carboxylic acid groups was measured by using the calcium acetate method, modified by filtering through 0.45 µm filters. 10.1021/es991228t CCC: $19.00
2000 American Chemical Society Published on Web 07/01/2000
Kinetic Methods. All kinetic experiments were conducted in a batch reactor in a thermostated water bath (( 0.1 °C) in the dark with constant stirring. The pH of the solutions was kept constant by additions of NaOH with a Mettler DL21 pH-stat apparatus. Iodide (I-(aq)) activities were monitored using an Orion 94-53 iodide-selective electrode coupled to an Orion 90-02 Ag/AgCl double-junction reference electrode. The progress of the reaction was monitored using triiodide ion, I3-(aq), because of its intense absorbance and rapid equilibrium with I2(aq)
I2(aq) + I-(aq) T I3-(aq)
K ) 729 (15)
(1)
Iodination experiments were conducted under pseudofirst-order conditions where only molecular iodine (I2) concentration limited the rate. In addition to simplifying the rate equation, these conditions reduce the likelihood of multiiodination reactions. A free-radical mechanism was tested by adding an excess of the radical scavenger acrylamide to some kinetic experiments. To keep variations of ionic strength below 10%, all kinetic experiments were performed in 0.1 M sodium perchlorate. In a typical experiment, 50 mL of phenolic compound in 0.1 M NaClO4 solution was added to a batch reactor and allowed to equilibrate to the required temperature. The initial pH was adjusted to the desired value with either HClO4 or NaOH, and the reaction was initiated by adding 1 mL of standardized iodine solution. At fixed intervals, the solution absorbance and potential difference, from the iodide/reference electrodes, were recorded. Experiments with 2-hydroxybenzoate (salicylate) (pKa ) 2.99) and 4-hydroxybenzoate (pKa ) 4.55) were conducted at pH 5.5 and 5.0, respectively. Rates were adjusted for percentages of nonprotonated versus protonated species for 4-hydroxybenzoate using the fraction of ionized to nonionized species. Rates for resorcinol were measured at pH 1.5 and at 5 °C and were extrapolated to 25 °C using an Arrhenius relationship determined for phenol (see below). The rate of reaction of 3-methoxyphenol was measured at pH ) 2. The rates of all other phenolic compounds were measured at 25 °C and pH ) 3.5 using concentrations of about 2.5 × 10-2 M, with [I2] ) 1.25 × 10-4 M and [I-] ) 2.5 × 10-3 M. Separate experiments were performed to measure the rate order for each component. Molecular iodine (I2(aq)) was used at two different concentrations (10-4 and 10-5 M) and plotted using the first-order integrated rate equation to determine its reaction order. In all kinetic experiments I2(aq) concentrations were constant, but limiting, and pH, initial concentration of phenolic compound, and I-(aq) concentration were varied. The kinetic experiments using humic and fulvic acids were performed similarly except for the following changes. The variation of the rates of reaction with respect to the concentration of the humic substance itself was not investigated. Absorbance from the humic substances at 351 nm was set to zero prior to adding I2(aq). Solutions of HS at certain pH values were prepared in advance and allowed to stand until the pH remained constant, due to the slow equilibration properties of humic substances (16). To account for all iodine in the kinetic system, some samples were allowed to react completely and subsequently extracted using solutions of 0.02 M sodium dihydrogen phosphate, 0.1 M sodium hydroxide, and boiling 2 M sodium hydroxide (17). Only the last treatment was able to recover essentially all of the iodine retained in the kinetic experiments. Product Analysis. Analyses for products of the reaction between the substituted phenolic compounds and I2(aq) were performed using gas chromatography coupled to a mass spectrometer (GC/MS) with a Varian 3400 GC using a 30 m DB-1701 column (ID ) 0.25 mm, 0.25 µm film) and Finnegan MAT ion-trap MS. Product was obtained from selected kinetic
TABLE 1. Hammett Substituent Values (Σiσi) and Rate Data for the Substituted Phenolsf compound
∑σ+ a
kobs (M s-1)
log(kobs/koobsb)
phenol m-cresol p-cresol guaiacol m-methoxyphenol resorcinold salicylatee p-hydroxybenzoatee
-0.41c -0.67 -0.48 -0.74 -1.06 -1.26 -0.95 -0.51
(5.6 ( 0.5) × 10-8 (1.9 ( 0.2) × 10-7 (4.6 ( 0.6) × 10-8 (1.0 ( 0.1) × 10-7 (5.5 ( 1) × 10-6 (3.8 ( 0.6) × 10-5 (5.7 ( 2) × 10-11 (1.3 ( 0.2) × 10-8
0 0.611 -0.004 0.337 2.074 2.923 -2.910 -0.548
a Data collected from ref 23. b ko obs is the overall rate coefficient of phenol iodination (5.6 × 10-8 M s-1). c Data from ref 24. d Rate extrapolated from data collected at 5 °C (see text). e These compounds existed in their deprotonated forms at the experimental pH. f All values at 25.0 °C unless otherwise indicated.
FIGURE 1. Plots of I3-(aq) absorbance against time during reaction of I2(aq) with m-cresol as a function of a; pH, b; I-(aq) concentration, and c; m-cresol concentration. First-order plots of triiodide (I3-) against time during reaction of I2(aq) with m-cresol as a function of d; pH, e; I-(aq) concentration, and f; m-cresol concentration. Conditions were T ) 25 °C, [I2] ) 1.23 × 10-4 M, [NaClO4] ) 0.098 M and unless otherwise indicated pH ) 3.5, [I-] ) 2.50 × 10-3 M and [m-cresol] ) 2.44 × 10-3 M. runs of the substituted phenolic compounds by extracting the reaction solution with three 100 mL washes of methylene chloride which was then washed once with a 50 mL solution of 1 mM sodium thiosulfate. After solvent removal, the residue was redissolved in 99+% methylene chloride in preparation for GC/MS analysis. Extraction of hydroxy-substituted benzoic acids included an acidification step to ensure that these compounds were extracted into the organic phase. It was also necessary to derivatize these polar acids to their trimethylsilyl analogues prior to GC using an excess of N,Obis(trimethylsilyl) acetamide (BSA) at room temperature with 10 min of low power sonication (18).
Results and Discussion Results for m-cresol, which were representative of the compounds studied (Table 1), are presented in the following section. None of the compounds showed significant change in reaction rate in the presence of the radical scavenger acrylamide. Iodination Kinetics of the Substituted Phenolic Compounds. The change in I3-(aq) absorbance during reaction with m-cresol, as a function of pH, I-(aq) concentration and m-cresol concentration, is shown in Figure 1a-c. The absorbance decrease is least rapid at low pH values and becomes VOL. 34, NO. 15, 2000 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3181
Using this method we find the order of reaction with respect to [H+] was -0.97 ( 0.05 (Figure 2a). This value corresponds to b in eq 3 and is assumed to be indistinguishable from -1. The slope of the rate with respect to m-cresol concentration was found to be 0.88 ( 0.13 (Figure 2c). Figure 2b shows the rate dependence of the iodination of m-cresol on the I-(aq) concentration, which gives a rate order (c in eq 3) with respect to I-(aq) concentration of -1.58 ( 0.10. A similar fractional rate order with respect to I-(aq) concentration was found for all of the other substituted phenols as well (data not shown). Grovenstein et al. (19) pointed out that an adjustment to the rate coefficient obtained from Figure 2b is necessary to account for changes in the equilibrium between I-(aq) and I3-(aq) at low concentrations of I-(aq). A simple adjustment has the form
kadj ) k′ ×
FIGURE 2. Plots of log k′ against a; pH, b; logarithm of I-(aq) concentration, and c; logarithm of m-cresol concentration for the reaction between I2(aq) and m-cresol. Conditions were T ) 25 °C, [I2] ) 1.23 × 10-4 M, [NaClO4] ) 0.098 M, [m-cresol] ) 2.44 × 10-3 M (a and b), [I-] ) 2.50 × 10-3 M (a and c) and pH ) 3.5 (b and c). more rapid as the pH increases in the interval 3.0 e pH e 4.5 (Figure 1a). In contrast, the reaction rate decreases as the I-(aq) concentration increases in the interval [I-] ) 1.27 × 10-3 to [I-] ) 1.44 × 10-2 M (Figure 1b). The rate of triiodide absorbance decay increases as the concentration of m-cresol is increased from 4.67 × 10-3 to 7.42 × 10-2 M (Figure 1c). A pseudo-order technique was used to simplify the experimental kinetics. The rate of reaction can be written in terms of I3-(aq) concentration
rate ) -
d[I2]t ) k′ [I3-] dt
(2)
where k′ is the pseudo-first-order rate coefficient (s-1), and [I2]t represents the stoichiometric concentration of iodine, as would be determined from thiosulfate titration (includes species I2(aq), I3-(aq), and HOI(aq)). The absorbance data shown in Figure 1a-c were transformed using the integrated first-order rate equation to obtain the linear plots in Figure 1d-f. The negative values of the slopes of these linear plots represent the pseudo-first-order rate coefficients. After determining the reactant species involved in the reaction, eq 2 can be rewritten as follows
rate ) kobs × [I3-]a × [H+]b × [I-]c × [m-cresol]d (3) where the superscripts a, b, c and d represent the orders of reaction with respect to triiodide, hydrogen, iodide, and m-cresol concentrations, respectively, and kobs is the overall rate coefficient (with units M1-(a+b+c+d) s-1). The rate orders for the H+, I-, and phenol concentrations were found by, equating eqs 2 and 3, taking the logarithm of both sides of the resulting equation and plotting log k′ against log [H+], log [I-], and log [m-cresol] (see Figure 2). 3182
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 34, NO. 15, 2000
[I3-]o [I3-]i
(4)
and arises because the concentration of I-(aq) imperfectly correlates with the concentration of I3-(aq) at low total concentrations of I-(aq) (eq 1). In eq 4, kadj is an adjusted pseudo-first-order rate coefficient, [I3-]o is the triiodide concentration at the highest I-(aq) concentration (1.44 × 10-2 M), and [I3-]i is the triiodide concentration at each I-(aq) concentration used. The data used to adjust k′ were found by extrapolating the concentrations of I2(aq) and I3-(aq) back to 0% reaction using the data for I3-(aq) (Figure 1b) and I2(aq) (data not shown). The reaction rate order with respect to I-(aq), after accounting for I3-(aq) formed at high I-(aq) concentrations, was -1.87 ( 0.14 (Figure 2b), which was assumed to be indistinguishable from a rate order of -2 because small amounts of iodine may be lost due to iodine hydrolysis and hypoiodous acid (HOI) disproportionation. Similar experimental determinations of the rate orders were made for the other phenolic compounds. Within the experimental uncertainties, the rate expression for all the simple substituted phenolic compounds, except salicylate, was of the form
rate ) kobs
[phenolic compound]1[I3-]1 [H+]1[I-]2
(5)
where kobs in units of M2 s-1 is the overall rate coefficient and phenolic compound refers to each of the compounds listed in Table 1, except salicylate (see below). The rate expression in eq 5 can be converted to a form in which the iodine concentration is expressed in terms of I2(aq) to the first power and the I-(aq) rate order is changed from -2 to -1 by dividing the rate expression in eq 5 by the triiodide equilibrium expression (eq 1). This latter form is used to calculate the rate coefficients in this study, which adjusts the units of kobs to M s-1. The reaction rate coefficients between I2(aq) and a reaction series of eight phenolic compounds varied by over 6 ordersof-magnitude (Table 1). The fastest iodination rate was with resorcinol, and the slowest reaction was with salicylate. The reaction between I2(aq) and resorcinol in this study was too fast at 25 °C to measure via the batch technique that was used for the other compounds. Therefore, its rate of reaction with I2(aq) at 25 °C was extrapolated from its rate at 5 °C, using an integrated form of the Arrhenius rate law (ln(k298) ) -(∆H‡/R)(1/298-1/278) + ln(k278)) established for phenol. The activation energy for the iodination of phenol was found to be 68 ( 1 kJ mol-1 (data not shown) and relates to the enthalpy of activation through the relationship, Ea ) RT + ∆H‡ (20).
FIGURE 3. Linear free energy relationship between the rate coefficients for iodination of various phenolic compounds and their modified Hammett substituent values, ∑σ+. Data from Table 1. Error bars are not shown for clarity. The salicylate point, 0, is not included in the regression. Products of Phenolic Compound Iodination. The product of iodination of phenol, under the experimental conditions, was shown to be 2-iodophenol by GC/MS, using iodinated reference compounds (data not shown). From these GC/MS spectra, we observed that iodine was an early fragmentation product, usually fragmenting as a solitary iodine atom at 127 m/e. Due to the unavailability of appropriate reference compounds the exact position of the iodine on the substituted phenolic compounds is unknown. Iodination was assumed to have occurred if a molecular ion matching the iodinated product existed in the mass spectrum, and if an iodine atom was evident as a fragmentation product in the mass spectrum. Similar fragmentation patterns for the other substituted phenolic compounds indicate that iodinated phenolic compounds form in all cases. Hammett LFER. A linear free energy relationship (LFER) for the rates of the iodinated phenolic compounds used in this study was constructed using the Hammett σ+ electrophilic substituent constant (21). Values of σ+ are required when a substituent is capable of a strong resonance interaction with a positively charged reaction center in electrophilic reactions. Resonance interactions are possible when substituents are bonded to the aromatic ring through atoms that have p-orbitals (22, 23). Brown and Okamoto (21) developed substituent constants (σ+) for electrophilic reactions based on hydrolysis rates of meta- and para-substituted 2-chloro2-phenylpropanes. Applied to this study, the form of the LFER is
log
( ) kobs
koobs
∑σ
)F
i
+ i
(6)
where kobs and koobs are the rates of a substituted and unsubstituted phenol, respectively, F is a reaction dependent constant, and ∑σ+ is defined as the sum of the σ+ constants for each substituent at its specific ring position. The sum total of the Hammett σ+ substituent constants and the experimental rates are listed in Table 1 and are plotted in Figure 3. Preferred sites for iodination on phenol depend on the combined resonance and inductive effects of the ring substituents. Substitution is expected to occur in a number of instances at the position ortho to the hydroxyl group, due to the strong ortho/para directing effect of the hydroxyl group. Typically the electronic interactions of ortho substituents are difficult to isolate from steric interactions (22). Tribble
FIGURE 4. Plot of I3- absorbance of IHSS soil HA during reaction with I2(aq) as a function of a; pH, and b; I-(aq) concentration. Variation of rate coefficients (k′) for reaction of IHSS soil HA with I2(aq) for c; pH, and d; I-(aq) concentration. Conditions were T ) 25 °C, [I2] ) 1.23 × 10-4 M and [NaClO4] ) 0.098 M and unless otherwise indicated pH ) 5.75 and [I-] ) 2.46 × 10-3 M. and Traynham (24), however, found excellent correlations between chemical shifts of ortho and para substituents in a number of substituted phenols in DMSO solution. The ortho substituent chemical shifts also showed a correlation with the resonance pathway Hammett substituent constants, σ+. Their value of ortho σ+ for iodine substitutions in the ortho position to the hydroxyl group, as predicted from the directing effects of the substituents, has been used. The substituted phenolic compounds used in this study included several different functional groups with varying electron-donating/withdrawing capabilities. The exact position of substitution is influenced by the strongest directing effect present on the aromatic ring. Hydroxyl (-OH) substituents are strongly activating and ortho- and paradirecting. The carboxylate (-COO-), methyl (-CH3), and methoxy (-OCH3) groups are also electron donating and ortho- and para-directing. The carboxylic acid group (-COOH) is a meta-directing deactivator but existed in significant amounts only with the p-hydroxybenzoate compound. The expected site of substitution of iodine will depend on the sum total of effects caused by all the various substituents. The precise position of substitution is not known, but it was assumed in calculating ∑σ+ values that the expected products, predicted from the activating and directing effects of each substituent, were similar to the actual products. The LFER suggests that there is a high degree of correlation between iodination rate coefficients and the Hammett Σσ+ parameter of these phenolic compounds. No correlation was observed with a LFER created using iodination rates and acid dissociation constants of either the hydroxyl or carboxylate-substituted compounds. The LFER covers a range of over at least 3 orders of magnitude and generally shows that hydroxy- and methoxy-substituted phenols had the fastest rates, followed by methyl and carboxylate-substituted phenols. This trend closely matches the strength of activation provided by the substituent group. Rate Law of Salicylate. Salicylate, having a carboxylate group ortho to the hydroxyl group, did not obey the same iodination rate law as the other substituted phenol compounds. The rate expression for salicylate was similar to eq 5 except that the order of reaction with respect to the salicylate VOL. 34, NO. 15, 2000 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3183
TABLE 2. Dissolved Organic Carbon (DOC) Concentrations, Elemental Compositions, and Acidity Data of the Humic and Fulvic Acid Kinetic Solutions humic substance
DOC (mg C/L)
carbon contenta (%)
ash content (%)
COOH acidity (mol/kg)
phenolic acidity (mol/kg)
IHSS soil HA IHSS soil FA ext. soil HA IHSS SRFA
19.0 ( 0.9 21.9 ( 0.3 8.7 ( 0.4 23.3 ( 0.7
58.13b 50.12b 45.4 ( 0.1 53.04b
0.88b 1.00b 41.8 ( 0.5 0.98b
4.9 ( 0.3 6.4 ( 0.7 2.3 ( 0.5 9.7 ( 1.6
1.9 ( 0.6 1.7 ( 0.3 6.5 ( 3.5 0.20 ( 0.06
a All carbon contents presented on an ash-free basis. Society (IHSS).
b
Carbon and ash contents were obtained from the International Humics Substances
TABLE 3. Rate Data for the Humic Substancesd humic substance
k′ (s-1)
k′[H+][I-] (M2 s-1)
[aryl-OH]a (M)
IHSS soil HA IHSS soil FA extracted soil HA IHSS SRFA
5.4 × 10-5 ( 3 × 10-5 3.3 × 10-5 ( 3 × 10-5 1.6 × 10-5 ( 2 × 10-5 1.5 × 10-4 ( 4 × 10-5
6.4 × 10-13 ( 6 × 10-13 2.9 × 10-13 ( 2 × 10-13 1.2 × 10-13 ( 2 × 10-13 1.8 × 10-12 ( 1 × 10-12
3.17 × 10-6 ( 1 × 10-6 3.79 × 10-6 ( 7 × 10-7 6.35 × 10-6 ( 4 × 10-6 4.48 × 10-7 ( 1 × 10-7
kobs (M s-1)
log(kobs/koobs)b
2.02 × 10-7 ( 2 × 10-7 0.55 ( 0.47 7.65 × 10-8 ( 7 × 10-8 0.13 ( 0.43 1.89 × 10-8 ( 4 × 10-8 -0.47 ( 0.96 4.02 × 10-6 ( 3 × 10-6 1.85 ( 0.36
predictedc Σiσi+ -0.68 -0.56 -0.41 -1.02
a Phenolic functional group concentrations were determined using eq 7 and data in Table 3. b ko obs is the rate coefficient of phenol iodination (5.6 × 10-8 M s-1) found in Table 1. c Σisi+ values are based on the equation of the line determined for the LFER in Figure 3. d All values at 25.0 °C.
concentration was 0.5 instead of 1. For this reason, the salicylate data do not fit on the LFER with the other simple substituted phenolic compounds (Figure 3). Under the experimental conditions (pH ) 5.5), salicylic acid would exist predominantly as salicylate (pKa ) 2.99). Salicylate differs from the compounds tested by the strong degree of ring stability conferred by intramolecular hydrogen bonding between the hydroxyl hydrogen and carbonyl oxygen of the carboxylate group. This ring stability may be sufficient to alter the mechanism of iodination. The dependence of the salicylate rate on pH also differed from the other compounds used in the study. The rate was found to be independent of pH below pH 5 (data not shown). Humic Substances. The kinetic measurements between I2(aq) and the humic substances were conducted in a manner similar to the phenol reaction series. Results for the IHSS soil HA are presented as representative for the humic materials studied. Reaction rates for humic substances were much slower than rates for the simple substituted phenolic compounds (compare Figures 1a-c and 4a,b), most likely because of the much lower phenolic group concentration and limited accessibility to these reaction centers. The slower rates of reaction caused a corresponding increase in the relative uncertainty of absorbance values, which was reduced by repeating the experiments in triplicate. Final rates were taken as the difference between the sample rates and the rates of I2(aq) loss in the absence of HS (controls). Changes in rates as a function of pH and I-(aq) concentration showed similar trends as the substituted phenol reaction series. At more basic pH values, the reaction rates increased for all four of the humic substances (Figure 4a). The rates, as a function of I-(aq) concentration, decreased as the I-(aq) concentration increased for all four HS (Figure 4b). The kinetic data in Figure 4 cannot be satisfactorily linearalized over the entire time range by plotting them as first- or second-order reactions or by using parabolic kinetics (t1/2) (data not shown). Plotted using the integrated firstorder rate equation, each of the resulting curves displayed two regions. Near the beginning of the reaction (t ≈ 0 s) the curves were steep, but the reaction rate decreased at later times. These differing rates had no effect on the rate order analysis; the same rate order was observed whether the rapid initial segment of the curve was used, a later portion of the curve, or the entire curve. Similar two-stage reaction rates have also been observed in the interaction between bromine 3184
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 34, NO. 15, 2000
and natural organic matter (25). Figures 4c,d show the rateorder analysis based on the kinetic data for IHSS soil HA, with respect to H+(aq) and I-(aq) concentration. The rate orders are obtained from log-log plots of the pseudo-firstorder rate coefficients against the pH and logarithm of the I-(aq) concentration (Figure 4c,d). Rate order values, with respect to [I-] and [H+], range from -0.24 to -1.24 and -0.30 to -0.72, respectively, for the four HS. This difference indicates that each humic substance studied has a slightly different rate law for iodination. For comparison with the rates determined for the substituted phenols, the rate data for HS were normalized to the phenolic content of the humic substance in the solution, using dissolved organic carbon (DOC) contents and the phenolic content of the humic substances (Table 2)
DOC
mg carbon mg HS × vol L × × L mg carbon -6 10 mol phenolic OH (7) mg HS
where DOC is the dissolved organic carbon concentration (mg/L), vol is the total volume (L) of the solutions used in the kinetic studies, the third term (mg HS/mg carbon) accounts for the carbon content of the humic substance, and the fourth term represents the moles of phenolic functional groups per milligrams of humic substance as determined from acidity titrations. Division of the pseudofirst-order rate coefficient (k′) by eq 7 gives an observed rate coefficient (kobs) that can be compared with the kobs values from the substituted phenols (Table 3). Pseudo-first-order rate coefficients for all four humic substances were converted to overall rate coefficients (kobs) by using the phenolic group concentrations for the humic materials (Table 3) and eq 7. Although the measured rate orders with respect to H+(aq) and I-(aq) concentration were not equal to one, the calculated values of kobs assume that individual species display rate orders of -1. In addition, the interaction was assumed to follow first-order reaction kinetics with respect to the phenolic group concentration, which was not tested. The rate coefficients obtained for the humic substances can be placed on the same rate scale used for the iodination of substituted phenols (Table 3). The values calculated have a range of -0.47 to 1.85 (Table 3), which compares favorably
to the range of values obtained for the phenolic compounds (-0.55 to 2.92) (Table 1). Average Σiσi+ values are proposed for the HS using the rate data in Table 3 and the equation of the regression line determined for the LFER (log(kobs/koobs) ) -3.79(Σiσi+) - 2.01) (Figure 3). Experimental uncertainties assigned to rate coefficients for the HS are relatively large because of imprecision in the kinetic measurements for these samples. The HS with the largest rate coefficient is the IHSS Suwannee River FA, and the smallest rate coefficient belongs to the extracted soil HA. The Suwannee River fulvic acid (SRFA) had a faster rate of reaction with I2(aq) than the soil FA and the other humic substances investigated. Interestingly, the LFER indicates that the normalized iodination rate of the SRFA is similar to the methoxy-substituted simple phenols studied (see Tables 1 and 3). Riverine humic substances contain high proportions of methoxyl carbon due to the presence of aryl methoxyl groups derived from terrestrial lignin sources (26). The higher proportions in aquatic humic substances result because the main diagenetic modification that occurs for terrestrial humic materials is demethylation, particularly of methoxyl groups; these reactions are slowed in aquatic environments. Other studies have also shown the high percentage of methoxyl groups in aquatic humic materials (27). A comparison of the approximate quantity of methoxyl groups between soil and stream and humic and fulvic acids reveals that stream HS, which are about 90% fulvic acids, contains on average 3.4 mol/g OCH3 compared to 0.3-0.8 and 0.3-1.2 mol/g OCH3 for soil HA and FA, respectively (28, 29). The high reactivity of the SRFA compared to the other HS may also be due to its probable small molar mass in comparison to the other humic substances studied. Molecular weights for aqueous Suwannee River FA of 829, 956, 1460, 1150, and 2310 daltons (30-34) have been previously determined using several techniques. This small molar mass compared to humic acids suggests increased accessibility to foreign compounds compared to higher molecular weight humic acids (32). These considerations may, in part, explain its faster reaction rate with I2(aq). The slowest reacting humic substance was the humic acid extracted from the Mudwell soil. This humic acid differs from the others studied in that it was not purified prior to use, and from its measured ash content (42%) must have been substantially associated with mineral matter, most likely iron, given its parent soil (12). This association would likely decrease its rate of reaction with I2(aq) relative to the other HS by blocking reactive sites in the humic acid. However, the reaction rate of this humic acid may reflect more accurately reaction rates of I2(aq) with soil organic matter. A further consideration is that other reaction pathways may exist for iodine interactions with HS. Iodine (I2(aq)) may react at nonaromatic unsaturated sites in HS, and I-, a strong nucleophile, may react with saturated regions. In summary, the rate law for iodination of seven substituted phenols was found to be identical, which enabled the creation of a linear free energy relationship valid over at least 3 orders of magnitude. Similar rate experiments for the iodination of several humic substances established that they had iodination rates, after normalization for phenolic content, which fell within the range of rate coefficients of the substituted phenols, suggesting that they follow a similar reaction pathway. The iodination rate of at least one HS (SRFA) is consistent with the reactivity of its likely predominant aromatic substituent (methoxyl).
Acknowledgments The authors thank the NSF through grants EAR-9626553 and EAR 98-14152 and the Kearney Foundation of Soil Science for partial support of this research.
Literature Cited (1) Robens, E.; Hauschild, J.; Aumann, D. E. J. Environ. Radioact. 1989, 9, 17-29. (2) Stewart, S. P.; Wilkens, B. T. J. Environ. Radioact. 1985, 2, 175182. (3) Rechard, R. P. Executive summary, Sandia NL report, SAND942563/1; 1995. (4) Krauskopf, K. B. Radioactive Waste Disposal and Geology; Chapman and Hall: London, 1988; pp 12-24. (5) Perel’man, A. I. Geochemistry of elements in the supergene zone; Levinson, A. A. Ed.; Keterpress Enterprises: Jerusalem, Israel, 1977; pp 80-88. (6) Raja, M. E.; Babcock, K. L. Soil Sci. 1961, 91, 1-5. (7) Sheppard, M. I.; Thibault, D. H.; McMurry, J.; Smith, P. A. Water, Air, Soil Pollut. 1995, 83, 51-67. (8) Francois, R. Geochim. Cosmochim. Acta 1987, 51, 2417-2427. (9) Schnitzer, M. In Proc. Symp. Soil Organic Matter Studies; Braunschweig, Germany, 6-10 Sept 1976; Int. Atomic Energy Agency: Vienna, 1977; pp 117-131. (10) Kolthoff, I. M.; Belcher, R. Volumetric Analysis; Interscience Publishers: New York, 1957; Vol. III, p 220. (11) Skoog, D. A.; West, D. M. Fundamentals of Analytical Chemistry; Holt, Rinehart and Winston: New York, 1963; pp 458-466. (12) Takahashi, T.; Dahlgren, R.; van Susteren, P. Geoderma 1993, 59, 131-150. (13) Goldin, A. Commun. Soil Sci. Plant Anal. 1987, 18, 1111-1116. (14) Perdue, E. M. In Humic Substances in soil, sediment and water; Aiken, G. R., McKnight, D. M., Wershaw, R. L., MacCarthy, P., Eds.; 1985; pp 493-534. (15) Turner, D. H.; Flynn, G. W.; Sutin, N.; Beitz, J. V. J. Am. Chem. Soc. 1972, 94(5), 1554-1559. (16) Engebretson, R.; von Wandruszka, R. Environ. Sci. Technol. 1998, 32, 488-493. (17) Yu, Z.; Warner, J. A.; Dahlgren, R. A.; Casey, W. H. Geochim. Cosmochim. Acta 1996, 60, 4945-4956. (18) Klebe, J. F.; Finkbeiner, H.; White, D. M. J. Am. Chem. Soc. 1966, 88, 3390-3395. (19) Grovenstein, E., Jr.; Aprahamian, N. S.; Bryan, C. J.; Gnanapragasam, N. S.; Kilby, D. C.; McKelvey, J. M., Jr.; Sullivan, R. J. J. Am. Chem. Soc. 1973, 92, 4261-4270. (20) Bunnett, J. F. In Investigation of rates and mechanisms of reaction; 4th ed.; Bernasconi, C. F., Ed.; Wiley-Interscience: New York, 1986; Techniques of Chemistry, Vol. VI, pp 251-372. (21) Brown, H. C.; Okamoto, Y. J. Am. Chem. Soc. 1958, 80, 49794987. (22) Leffler, J. E.; Grunwald, E. Rates and Equilibria of Organic Reactions; Wiley: New York, 1963; pp 192-210. (23) Carroll, F. A. Perspectives on Structure and Mechanism in Organic Chemistry; Brooks/Cole Publishers: New York, 1997. (24) Tribble, M. T.; Traynham, J. G. J. Am. Chem. Soc. 1969, 91(2), 379-388. (25) Song, R.; Westerhoff, P.; Minear, R. A.; Amy, G. L. In Water Disinfection and Natural Organic Matter; Minear, R. A., Amy, G. L., Eds.; American Chemical Society: Washington, DC, 1996; Chapter 17. (26) Francois, R. Rev. Aquatic Sci. 1990, 3, 41-80. (27) Sohn, M. L. Org. Geochem. 1985, 8(3), 203-206. (28) Malcolm, R. L. In Humic Substances in soil, sediment, and water: Geochemistry, isolation and characterization; Aiken, G. R., McKnight, D. M., Wershaw, R. L., McCarthy, P., Eds.; Wiley & Sons: New York, 1985; pp 181-209. (29) Stevenson, F. J. Humus Chemistry, Genesis, Composition, Reactions, 2nd ed.; Wiley & Sons: New York, 1994; pp 226-235. (30) Aiken, G. R.; Malcolm, R. L. Geochim. Cosmochim. Acta 1987, 51, 2177-2184. (31) Leenheer, J. A.; Brown, G. K.; McCarthy, P.; Cabaniss, S. E. Environ. Sci. Technol. 1998, 32, 2410-2416. (32) Reid, P. M.; Wilkinson, A. E.; Tipping, E.; Jones, M. N. 1990, 54, 131-138. (33) Beckett, R.; Jue, Z.; Giddings, J. C. Environ. Sci. Technol. 1987, 21, 289-293. (34) Chin, Y.; Aiken, G.; O’Loughlin, E. Environ. Sci. Technol. 1994, 28, 1853-1858.
Received for review November 2, 1999. Revised manuscript received May 2, 2000. Accepted May 4, 2000. ES991228T
VOL. 34, NO. 15, 2000 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3185
