Environ. Sci. Techno/. 1995, 29, 2735-2740
Decomposition of Tetrachloro-1,&benzoquinone (p-Chloranil) in Aqueous Solution D A V I D H . SARR,+ C H I K O M A K A Z U N G A , M. JUDITH CHARLES, JAMES G. PAVLOVICH,t AND MICHAEL D. AITKEN' Department of Environmental Sciences and Engineering, CB #7400, University of North Carolina, Chapel Hill, North Carolina 27599- 7400
p-Chloranil (2,3,5,6-tetrachloro-2,5-cyclohexadienelf4-dione; tetrachloro-1,4-benzoquinone) has been observed as an oxidation product in processes used .to oxidize pentachlorophenol (PCP), has known biocidal properties, and has been implicated in genotoxic effects associated with PCP. Chloranil undergoes displacement of chloride by hydroxide under highly alkaline conditions, but no previous work on chloranil decomposition has been conducted at environmentally relevant pH. Electrospray mass spectrometry was used in this study to confirm the two-step hydrolysis of chloranil to yield chloranilic acid (2,5-dichloro-3,6dihydroxy-1,4-benzoquinone), and the kinetics of each step were quantified as a function of pH. The halflife of chloranil at pH 7 is estimated to be slightly over 1 h, while that of its first hydrolysis product (trichlorohydroxyquinone) is about 21 d. Chloranil also reacts with hydrogen peroxide in a pH-dependent manner at rates substantially greater than the rate of spontaneous hydrolysis. The low yield of chloranilic acid from this reaction suggests that other, as yet unidentified, products are formed. Chloranilic acid has lower acute toxicity (as measured in the Microtox screening assay) than does chloranil, so that promoting the accelerated hydrolysis of chloranil may be advantageous in waste treatment or remediation processes in which it is formed.
Introduction p-Chloranil (Figure 1) has been observed as a significant product from the oxidation of pentachlorophenol (PCP) when mediated by phenol-oxidizingenzymes in vitro (131, by fungal metabolism (4, 5) and by Ti02-assisted * Author to whom correspondence should be addressed; e-mail address:
[email protected]; telephone: (919) 966-1481; fax: (919) 966-7911. t Present address: Montgomery-Watson,Inc.,560 Herndon Pkwy., Herndon, VA 22070. Present address: Department of Chemistry, University of California, Santa Barbara, CA 93106.
*
0013-936W95/0929-2735$09.00/0
Q 1995 American Chemical Society
Q p-chloranil
TCHQ
0 chloranilic acid
FIGURE 1. Structures of pchloranil, trichlorohydroxyquinone,and chloranilic acid. The reported p& for TrCHQ is 1.1 (13,and the reported values for chloranilic acid range from 0.6 to 2.0 for p 4 and from 2.4 to 3.2 for p/(2 (14, 15).
photolysis (6'). In the past, p-chloranil was used as a fungicide and algicide under the trade name Spergon (7). It is known to form protein adducts both in vitro and in vivo and has been implicated in the genotoxicity associated with PCP (8). While there is much interest in developing technologies to remediate PCP-contaminated soils and groundwater (9, IO),the formation of toxic byproducts such as chloranil must be considered and, ifpossible, overcome. Work reported in the 1960s (11, 12) indicated that chloranil is hydrolyzed in strong alkaline solution to yield chloranilic acid (2,5-dichloro-3,6-dihydroxy-1,4-benzoquinone; Figure 1) through an intermediate designated trichlorohydroxyquinone (TrCHQ; Figure 1). Under milder conditions in aqueous solution, the colorless chloranil can be observed to develop a purple color (21, suggestingthat decomposition reactions may occur in systems of environmental interest. The purposes of the current studywere to identify TrCHQ and chloranilic acid as products of chloranil hydrolysis at near-neutral to slightly acidic pH, to quantlfy the kinetics of chloranil and TrCHQ decomposition, and to evaluate the potential acute aquatic toxicity of chloranil and its decomposition products using the Microtox screening assay. Because chloranil is formed from PCP in peroxidase-catalyzed oxidation reactions (2, 31, we also examined the influence of hydrogen peroxide on chloranil decomposition.
Materials and Methods Chemicals. p-Chloranil(99%)and chloranilic acid (99%) were purchased from Sigma Chemical Co. (St. Louis, MO). Stock solutions of each compound were prepared gravimetrically in methanol and stored at 4 "C. Hydrogen peroxide stock was prepared by dilution of 30% reagentgrade hydrogen peroxide and was calibrated by titration against potassium permanganate standard in 1 N H2S04. Buffers (sodium tartrate for pH 3-5, sodium or potassium phosphate for pH 6-9, and ammonium hydroxide for pH 10)were prepared from reagent-grade acids and salts. Ethyl acetate was reagent-grade, and all other solvents were highpressure liquid chromatography (HPLC) or spectrophotometric grade. I80-Labeled water (95 atom % l80)was purchased from Aldrich Chemical Co. (Milwaukee, WI). Determination of Molar Extinction Coefficients. The molar extinction coefficients at 530 nm for both chloranilic acid and TrCHQ in water were determined by measuring absorbance as a function of concentration using a Hitachi U2000 double-beam spectrophotometer. TrCHQ was prepared by mixing known concentrations of reagent chloranil in pH 8.5 buffer and waiting until a stable value ofASs0was reached. Complete loss of chloranil under these conditions was confirmed by HPLC, and stoichiometric
VOL. 29. NO. 11, 1995 I ENVIRONMENTAL SCIENCE & TECHNOLOGY
2731
conversion to TrCHQ was assumed based on the appearance of a single peak in HPLC analysis, the area of which correlated linearly with initial chloranil concentration (1.2 = 0.99, n = 4). The estimated E530 of TrCHQ was 1.41 x lo3 L mol-' cm-'(1.2 = 0.98; n = 51, which agrees well with the reported value of 1.45 x lo3 at 535 nm ( I ] ) , and was unaffected by pH over the range 6.9-8.5 (the range used in kinetic studies). The measured 6530 for chloranilic acid was 176 L mol-' cm-l at pH 7.0, increasing slightly as pH decreased and reaching a value of 227 L mol-' cm-l at pH 4.1.
Analysis of Decomposition Products. Chloranil solutions (100 mL at 30 yM) were prepared in buffers ranging from pH 7 to pH 10 and allowed to decompose for 24 h at room temperature (approximately 23 "C). Each solution was acidified with HC1 to pH 1 and then extracted with two successive 10d v o l u m e s of ethyl acetate. The ethyl acetate was evaporated under a stream of nitrogen gas, and the extracted material was redissolved in 90% methanol/ 10% water (v:v) to a final volume of 1 mL immediately prior to electrospray mass spectrometry (ESIMS) analysis. A chloranilic acid standard (3mM)was prepared in 90%methanol/ 10%water. Chloranil standards (1 mM) were prepared in 90% acetonitrile/lO% water and in 90% acetonitrile/lO% Hzl*O. ESlMS was performed by flow injection of each solution at a rate of 3yL/min. Mass spectra were obtained in the negative ion mode on a Finnigan 4000 quadrupole mass analyzer retro-fitted with an Analytica of Branford, Inc. electrospray source. Nitrogen was used as the drying gas and was heated to 200 "C, and SFs was used as the sheath gas. Except for the chloranil standards, the inlet capillary was held at ground, and the cylindrical, backing, and capillary electrodes were held at 2500,2800, and 3000 V, respectively. For the chloranil standards, the inlet capillary was held at -2500 to -3500 V, and the cylindrical and backing electrodes were held at ground. The mass analyzer was scanned from 10 to 500 amu, and the data were collected on a Technivent Vector 2 data system. Chloranil was also analyzed by full-scan (50-650 amu) electron-impact ionization analysis on aVG 70 SEQ hybrid mass spectrometer, after introducing chloranil into the mass spectrometer using a direct-insertion probe. HPLC analysis of chloranil and chloranilic acid was performedwith a Waters 600E system and Millennium data collection and operating software. Chloranil was analyzed under isocratic conditions using a CIScolumn with amobile phase (1mL/min) consisting of 60:40 methanol/water (v:v) plus 0.04% trifluoroacetic acid and UV detection at 254 nm. Chloranilic acid was analyzed by C18isocratic HPLC with detection at 320 nm, using a mobile phase of 30:70 methanol/buffered water (12.5mM potassium phosphate, pH 6.8) containing 5 mM of the ion pairing reagent tetrabutylammonium hydrogen sulfate (TBAHS; Aldrich). Both chloranil and chloranilic acid were quantified by external standardization against the reagent-grade chemicals. Qualitative HPLC analysis to verify the existence of decomposition products generated in the presence of methanol (in pH 8.5 buffer) was conducted under isocratic conditions using a mobile phase of 50:50 methanollbuffered water containing TBAHS as described above and visible detection at 530 nm. Under these conditions, chloranilic acid eluted as an unretained peak, but two overlapping peaks were separated from chloranilic acid and from TrCHQ. The absence of these peaks in mixtures from which 2736 1 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 29, NO. 11. 1995
methanol was excluded was determined by adding solid chloranil directly to the pH 8.5 buffer and analyzing the aqueous phase over time by the same HPLC method. The yield of chloranilic acid from chloranil was determined by incubating triplicate samples of 30 yM chloranil in pH 8.5 buffer. After 5 d of incubation, chloranilic acid was quantified by HPLC, and a yield of 75%was calculated. The final absorbance at 530 nm of the 30 yM chloranil solution was 0.003 unit higher than if a 100% yield of chloranilic acid were achieved (the expected A530 of 30 yM chloranilic acid = 0.005), indicating that other products also contributed to absorbance at 530 run. Aqueous DecompositionKinetics. Chloranil solutions (30 yM) were prepared in buffers at various pH, and absorbance at 530 nm was followed with time at room temperature. Chloranil was assumed to be converted to TrCHQ by hydrolysis as follows: C
+ OH- kl.Obs TrCHQ + C1-
(1)
where Cis chloranil and kl,obs is a pseudo-first-order rate coefficient at constant pH. The decomposition of TrCHQ in the presence of methanol was assumed to proceed as two parallel reactions:
+ OHCA + C1TrCHQ + MeOH S.obs other products TrCHQ
(2) (3)
where CAis chloranilic acid, and k2,0bsand k3,0bsare pseudofirst-orderrate coefficientsat agiven pH and fixed methanol concentration for reactions 2 and 3, respectively. The other products are those observed by HPLC in reactions containing methanol introduced with the chloranil stock solution; the methanol concentrations in these reactions were far higher than the initial chloranil concentration and, therefore, could be treated as constant. The two pseudo-firstorder reactions shown as eqs 2 and 3 were combined and expressed as an effective first-order reaction at a fNed pH: TrCHQ
-CA + other products k2,eff
(4)
where k2,enisthe effectivefEst-order rate coefficient (=k2,obs k3,0bs) for the decomposition of TrCHQ at constant pH in the presence of excess methanol. For two first-order reactions in series (reactions 1 and 41, the concentration of the intermediate (TrCHQ) can be expressed as a function of time in a batch system (16):
+
The total concentration of products is given by (16):
[CA+ other products] =
At any time, the absorbance at 530 nm represented the sum of the contributions of TrCHQ, chloranilic acid, and other products, or assuming equivalentminor contributions from chloranilic acid and other products (see above):
By substituting eq 5 for [TrCHQ], eq 6 for the total concentration of products, and the measured extinction coefficients for TrCHQ and chloranilic acid, an equation relating A530 vs time was obtained. This equation was fit to the data using non-linear regression to obtain best estimates for both kl,& and kz,effat each pH value tested. Nonlinear regression was conducted with SYSTAT (SYSTAT, Inc., Evanston, IL). To account for possible influences of the buffer on the observed changes in absorbance, chloranil was incubated at pH 7.6 in the presence of phosphate buffer at concentrations of 1, 10, and 100 mM (triplicate reactions at each concentration), and the absorbance at 530 nm was measured after 65 min. Another set of reactions was run at pH 8.8 in the presence of carbonate buffer or phosphate buffer (triplicate reactions for each buffer), and absorbance was measured after 30 min. The effect of oxygen on the rate of chloranil decomposition was evaluated in side-by-side reactions containing 30 pM chloranil in pH 7 buffer. Triplicate reactions were run in air-saturated buffer, and triplicate reactions were run headspace-free in buffer that had been purged with helium. Optical absorbance at 530 nm was then followed at selected intervals over a 3-h reaction period. Stoichiometryof the Reactionwith Hydrogen Peroxide. The stoichiometryof hydrogen peroxide consumption from its reaction with chloranil was determined by adding peroxide (100pM)to solutions of chloranil at pH 7. Residual peroxide was measured after 10 min or 10 h using an enzymatic assay. The enzymatic assay involved adding a small aliquot of sample to a mixture containing Coprinus macrorhizus peroxidase (CMP; a research gift from Novo NordiskA/S,Bagsvaerd, Denmark) at 1pg/mL and 1.7 mM peroxidase substrate ABTS (2,2’-azinobis[3-ethylbenzothiazoline-6-sulfonic acid], diammonium salt; Sigma) in a pH 6 phosphate buffer. ABTS is converted in this assay to the chromophore ABTS+ (I3 ,the cation radical ofABTS, which can be followed at 405 nm. A stable value of &05 was recorded for each sample, and the peroxide concentration was determined by comparison to a standard curve prepared using known concentrations of peroxide (linear in the range 0-250 pM). Chloride release from the reaction between chloranil and peroxide was determined with a chloride-specific electrode (pHoenix Electrode Co.) connected to an Orion 920A pHImV meter. The ionic strength of each sample was adjusted by adding sodium nitrate to a final concentration of 10 mM. A standard chloride curve was prepared using solutions of reagent-grade sodium chloride. It was not possible to add enough chloranil in a single dose to give chloride concentrations in the desired range (up to 500pM) because chloranil is soluble in water only to about 50 pM. Therefore, chloranil and peroxide were added in pulse doses until the final desired dose of chloranil was achieved. Peroxide was added in excess in all cases. The combined yield of TrCHQ and chloranilic acid was estimated in triplicate samples in which peroxide (100pM) and chloranil(50pM) were mixed at pH 7 for 4 h. The pH of the samplesthen was raised to 10 to hydrolyze any TrCHQ that might have formed. Total chloranilic acid concentration was then determined by HPLC.
Kinetics of Reaction with Hydrogen Peroxide. Chloranil solutions (30pM)were prepared in buffers at different pH and mixed with hydrogen peroxide (120pM). Aliquots of each mixture were treated with catalase to terminate the reaction at specific time intervals and then analyzed by HPLC for residual chloranil concentration. Concentration vs time data were used to estimate rates of chloranil consumption at each pH by nonlinear regression using a first-order decay model. The effect of peroxide concentration on reaction rate was determined at pH 4.1 and a nominal initial chloranil concentration of 50pM. Reaction mixtures were prepared directly in HPLC autosampler vials and injected at intervals programmed into the HPLC controller. Actual initial concentrations of chloranil used for kinetic analysis were those measured in the first injection, which were between 43 and 46pM in all cases. Data from each set of injections correspondingto a given hydrogen peroxide concentration were analyzed according to a first-order decay model. ToxicityAnalysis. The Microtox acute toxicity screening assay was conducted using procedures described in detail elsewhere (18). Chloranil samples were prepared in pH 7 buffer and analyzed immediately or were allowed to decomposefor 3.5 h before beginning the Microtox analysis. Solutions containing chloranil plus an equimolar amount of hydrogen peroxide in pH 7 buffer were mixed for 80 min before analysis. Chloranilic acid and PCP samples were prepared in pH 7 buffer from standard solutions. EC5o values (concentrations corresponding to 50% attenuation in light output by the luminescent test organism) were determined by instrument software (Microbia Corp., Carlsbad, CA).
Results and Discussion Aqueous solutions of p-chloranilwill develop a faint purple color when left standing at near-neutral pH and at room temperature (21,indicating that hydrolysis reactions previously reported to occur under strongly alkaline conditions ( I ] ) might also occur, albeit more slowly, under environmentallyrelevantconditions. Preliminaryefforts to analyze for the presence of TrCHQ and chloranilic acid by gas chromatography/massspectrometry (GCIMS)were unsuccessful because of the high polarity of these compounds. Accordingly, we analyzed decomposed chloranil samples by flow injection ESIMS. The mass spectrum of a sample left at pH 8 for 24 h is shown in Figure 2, along with mass spectra of standards of chloranil and chloranilic acid. The mass spectrum of the chloranil standard in acetonitrile (Figure2a) is characterized by afour-chlorine isotope cluster at rnlz 244 and a three-chlorine isotope cluster at rnlz 225. The three-chlorine cluster evident at mlz 225 presumably is the (M - HI- ion of TrCHQ, since we cannot postulate a fragmentation mechanism for chloranil from which such a cluster would be observed. The appearance of TrCHQ in the chloranil standard was inferred to result from reactions occuring during analysis and not due to contamination of the standard, based on several observations. We did not observe the presence of TrCHQ in the electron-impact mass spectrum of chloranil, and complex patterns of three-chlorine and four-chlorineisotope clusters beginning at rnlz 225 and rnlz 244, respectively, were observed in the ES mass spectra obtained from chloranil dissolved in90%acetonitrile/ 10%Hzl80 (not shown). These patterns were indicative of mixed l60and l80isotopes of the hydroxyl group in TrCHQ and also the carbonyl oxygens VOL. 29, N O . 1 1 , 1 9 9 5 /ENVIRONMENTAL SCIENCE &TECHNOLOGY
2737
100
s:2
1
0.07
0.05
I&
0.01
i
1
b
0.04
0.00
0
0.03
1
2
3
4
5
Time (h)
A A
A
53
0.02 A
180
220
200
240
260
mlz 0
50
100
150
200
250
Time (h) FIGURE 3. Changes in absorbance at 530 nm with time for samples of chloranil incubated at pH 8.0 (@),7.1 (A), or 5.1 (0). Inset shows data for pH 7.1 and pH 8.0 over a shorter time scale.
180
200
220
240
260
240
260
rnlz 225
I
180
200
220
m/z FIGURE 2. ES mass spectra of (a) chloranil; (b) chloranilic acid; (c) extract from chloranil sample allowed to decompose for 24 h at pH 8.
in both TrCHQ and chloranil [suchexchange of an aromatic carbonyl oxygen with that in water has been reported to occur for 2,6-dichloro-1,4-benzoquinone (211. The presence of a hydroxyl oxygen derived from water indicates that chloranil was hydrolyzed partially during analysis. It is possible that, during electrospray, the enrichment of OHions during droplet disintegration and evaporation of the solvent create a basic environment that is sufficient to hydrolyze chloranil to TrCHQ. Because little fragmentation occurs during ESIMS, spectra corresponding to the deprotonated chloranilic acid molecular ion (mlz 207) and to the deprotonated TrCHQ molecular ion (mlz 225) are distinguishable in the decomposed chloranil sample at pH 8 (Figure 2c). The intensity of the TrCHQ (M - HI- ion decreased with increasing pH 2738 rn ENVIRONMENTAL SCIENCE &TECHNOLOGY / VOL. 29, NO.
11, 1995
and corresponded to an increasing intensity of the chloranilic acid (M - HI- ion (not shown), illustrating that the rate of chloranilic acid formation from chloranil increased with increasingpH (allsamples were incubated for the same period of time). The spectra shown in Figure 2c also include a smaller two-chlorine isotope cluster at rnlz 221, which would be consistent with a methyl ether of chloranilic acid or an equivalent compound. The existence of such a product derived from TrCHQ is consistent with HPLC analysis, in which two overlappingpeaks observed in reaction mixtures containing methanol were not observed when solid chloranil was used instead of the methanol stock these peaks might represent isomers resulting from the displacement of chloride by the methoxy group at different positions in TrCHQ. Based on the measured yield of chloranilic acid from the overall hydrolytic decomposition of chloranil, the methoxy-substituted product(s) account for 25% or less of the original chloranil on a molar basis. Kinetics of HydroIyds Reactions. Solutionsof chloranil were prepared in buffers over a pH range from 4 to 8.1, and absorbance at 530 nrn was followed over time. At pH 7 and above, a rapid increase in A530 was observed within 2 h of chloranil addition, followed by a much slower decrease in A530 that was monitored for 220 h, as shown in Figure 3. At acidic pH, the rise in A530occurred very slowly, and no decrease was observed over the 220-h reaction period. No effects of the buffer on reaction kinetics were noted, either by varying the buffer concentration or by substituting carbonate for phosphate. Molecular oxygen also had no effect on the reaction rate. Data similar to that shown in Figure 3 (six pH values from 6.9 to 8.1) were used to estimate pseudo-first-order rate coefficients for hydrolysis of chloranil to TrCHQ ( k l , o b s ) and for subsequent decomposition of TrCHQ to chloranilic acid and other products (k2,eff). Estimates of the rate coefficients as a function of hydroxide ion concentration are shown in Figure 4. The linear fit in each case supports a second-order reaction model involvinghydroxide ion for each step in converting chloranil to chloranilic acid. Hancock etal. (11)proposed that each hydrolysis reaction occurs first by hydroxide addition followed by chloride elimination. Given the strong pH dependence of each step, it is likely that hydroxide addition is the rate-limiting step
2
j0.020
r
5 e
a.
YN
O.O1 0.00
1,
4.0
FIGURE 4. Pseudo-first-order rate coefficients as a function of hydroxide ion concentration for conversion of chloranil to TrCHQ (0)and for subsequent decomposition of TrCHQ (0).Vertical bars represent 95% confidence limits on the estimate for the rate coefficient at each hydroxide ion concentration.
0
, , , ,
, , , ,
4.5
5.5
5.0
6.0
6.5
PH
[OH-] (rM)
0
, ,
1
I
I
10
20
30
pChloranil Added (pM)
FIGURE 5. Stoichiometry of hydrogen peroxide consumed per unit of chloranil reacted after mixing for 10 min (0)or 10 h (0)at pH 7. Linesof bestfit haveslopesof0.88(2=0.995)and1.13(2=0.990), respectively. Vertical bars represent f1 SD from triplicate samples at each chloranil concentration.
in each case. The fit for k2,eff did not pass through the origin, indicating that decomposition of TrCHQ occurred via a second pathway that was independent of pH. The existence of such a pathway was supported by the observation of a methylated derivative of chloranilic acid (or analogous species) in the ES spectra (Figure 2c) and the appearance of two overlapping peaks in HPLC analysis of chloranil mixtures containing methanol. The linear fits to the data shown in Figure 4 were used to estimate second-order rate constants at 23 "C of 6.2 x lo6M-I h-I ($ = 0.980) for conversion of chloranil to TrCHQ and 1.4 x lo4M-' h-' (6= 0.990) for conversion of TrCHQ to chloranilic acid. From the rate constant for the first reaction, the half-life of chloranil at pH 7 is estimated to be only 1.1 h. If the apparent pH-independent decomposition reaction of TrCHQ is ignored, the half-life of this species at pH 7 is estimated to be about 21 d. Reaction with Hydrogen Peroxide. The stoichiometry of the reaction between chloranil and hydrogen peroxide is illustrated in Figure 5 for two different reaction times. After 10 min of reaction, slightlyless than 1 mol of peroxide was consumed per mole of chloranil added (allof the added chloranil was consumed), while a stoichiometric ratio of slightly more than 1:l was observed after 10 h of reaction. Chlorideproduction was measured in a separate experiment
FlGURE6. Pseudo-first-order rate coefficient for reaction of chloranil (30pM initial concentration) and hydrogen peroxide (120pM initial concentration) as a function of pH. Vertical bars represent 95% confidence limits on the estimate for the rate coefficient The dashed line represents the rate of spontaneous hydrolysis of chloranil as a function of pH using the estimated second-order rate constant of 6.2 x 1C M-l h-l.
(not shown); over a 2-h reaction period, slightly more than two chloride ions were released per chloranil molecule (? = 0,998;n = 18). Chloranilicacid did not react with peroxide when incubated for 12 h at pH 7 or pH 4. The rate of reaction between peroxide and chloranil was too fast to measure except at acidic pH. Pseudo-first-order rate coefficients(kobs,H202) for the disappearance of c h l o r d over a pH range of 4.2-6.1 at a fixed peroxide concentration are shown in Figure 6 and are compared to the rate coefficients for chloranil decomposition in the absence of peroxide. Peroxide accelerated the rate of chloranil decomposition by 2 orders of magnitude over the pH range tested. Although the mechanism for peroxide-dependent decomposition of chloranil is unknown at this time, the data shown in Figure 6 seem to fit a pseudo-second-order rate model (first order in both chloranil and in hydroxide ion at fixed peroxide concentration); for the log transformation of the rate data shown in Figure 6, a linear fit was obtained (6= 0.979)with a slope of 11h-' per pH unit. The rate of reaction was approximately half-order with respect to peroxide concentration at pH 4.1 (not shown; slope of log-log plot = 0.47, = 6, 1.2 = 0.92). To determine if the reaction with peroxide led to the same products observed via hydrolysis,chloranilwas mixed with peroxide at pH 7 for 4 h, then the pH was raised to convert any TrCHQ formed to chloranilic acid. The yield of chloranilic acid in this experiment was a small fraction of the starting chloranil concentration and was much lower than the yield obtained in the absence of peroxide (approximately 75701,so that it appears that the reaction with peroxide leads to as yet undetermined products. No additional products have been observable by ES/MS. Acute Toxicity Screening. The Microtox assay has been used as a screening tool to assess the potential acute aquatic toxicity of chemicals, environmental samples, and reaction products derived from waste treatment or remediation processes (18). MicrotoxEC50values of PCP, chloranil, and a chloranil solution decomposed for 3.5 h all were in the range of 1-3 p M . For the chloranil solution incubated at pH 7 for 3.5 h, approximately 90% conversion to TrCHQ would have been expected therefore, it appears that TrCHQ is about as toxic as chloranil. To the contrary, however, chloranilic acid did not elicit a 50% attenuation in light VOL. 29, NO. 11. 1995 /ENVIRONMENTAL SCIENCE &TECHNOLOGY
2739
output in the Microtox assay at concentrations as high as 180 p M . There was little change in toxicity when chloranil was decomposed with hydrogen peroxide. This finding supports the conclusion that products other than chloranilic acid are produced from the reaction between chloranil and peroxide.
Conclusions p-Chloranil is a known toxic compound that has been observed as a significant product in processes used to oxidize pentachlorophenol. In aqueous solution, it undergoes two consecutive hydrolytic dechlorinationreactions to yield chloranilic acid via the intermediate trichlorohydroxyquinone. Each hydrolysis reaction apparently occurs via displacement of chloride by hydroxide, so that the rate of hydrolysis is strongly pH dependent. Reaction rates are described by second-order kinetics, with estimated halflives of chloranil and TrCHQ at pH 7 and 23 "C of 1.1h and 21 d, respectively. Lesser chlorinated quinones, such as 2,6-dichloro-1,4-benzoquinone, have also been observed to undergo color changes in aqueous solution (Z), which may be indicative of decomposition reactions similar to those that occur with chloranil. Hydrolysis reactions involving alkyl halides have been well characterized (191, but the hydrolysis of chlorinated aromatics in environmental systems has not been well studied. Hydrogen peroxide strongly accelerates the rate of decomposition of chloranil, but the low combined yield of TrCHQ and chloranilic acid from this reaction indicates that other products are formed. Peroxide-dependent decomposition pathways may be important in systems where peroxide is either used or produced, as in the peroxidase-catalyzed oxidation of PCP. The final product of hydrolytic chloranil decomposition, chloranilic acid, appears to be significantly less toxic (as defined in the Microtox assay) than the parent compound. This finding may have implications for waste treatment or remediation processes in which chloranil is formed from PCP. For example, it may be possible to expedite the ultimate detoxification of PCP and its oxidation products simply by pH adjustment to maximize the conversion to chloranilic acid in the treatment system itself. Competing reactions involvinghydrogen peroxide must be considered in systems containing peroxide, and such reactions may affect toxicity. The rates of chloranil and TrCHQ hydrolysis measured in this study suggest that conversion to chloranilic acid can occur in aqueous environments over meaningful time
2740
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 29. NO. 11, 1995
scales. However, the potential partitioning behavior of chloranil and TrCHQ, which has not been examined, might influence the rates of decomposition reactions in the environment.
Acknowledgments This work was supported by the National Institute of Environmental Health Sciences (Grant P42ES05948) under its Superfund Basic Research Program. We thank Michele McGinnis for conducting the Microtox analyses, Ten-Yang Yen for assistance with the ES/MS analysis, and Professors Avram Gold and Steve Rappaport for discussions on properties of chloranil.
Literature Cited (1) Konishi, K.; Inoue, Y. Mokuzai Gakkaishi 1972,18 (91,463-469. (2) Hammel, K. E.; Tardone, P. J. Biochemistry 1988,27,6563-6568. (3) Mileski, G. J.; Bumpus, J. A.; Jurek, M. A.; Aust, S. D.Appl. Enuiron. Microbiol. 1988, 54, 2885-2889. (4) Ruckdeschel G.; Renner, G. Appl. Enuiron. Microbiol. 1986,51, 1370-1372. (5) Kremer, S.; Sterner, 0.;Anke, H. Z. Z. Naturforsch. 1992, 47C, 561-566. (6) Mills, G.; Hoffmann,M. R. Environ. Sci. Technol. 1993,27, 16811689. (7) Zweig, G.; Hitt, J. E.; Cho, D. H. J. Agric. Food Chem. 1969, 17, 176- 179. (8) Waidyanatha, S.; McDonald, T. A.; Lin, P.-H.; Rappaport, S. M. Chem. Res. Toxicol. 1994, 7, 463-468. (9) Mueller, J. G.; Chapman, P. J.; Pritchard, P. H. Enuiron. Sci. Technol. 1989, 23, 1197-1201. (10) Approaches for Remediation of Uncontrolled Wood Preserving Sites;US.Environmental Protection Agency: Washington, DC, 1990; EPA/625/7-90/011. (1 1) Hancock, J. W.; Morrell, C. E.; Rhum, D. TetrahedronLett. 1962, 22, 987-988. (12) Bishop, C. A.; Tong, L. K. J. Tetrahedron Lett. 1964, 41, 30433048. (13) Beauchamp,A.; Benoit, R. L. Can.J. Chem. 1966,44,1607-1612. (14) Bode, H.; Eggeling,W.; Steinbrecht, V. Z. Anal. Chim.Acta 1966, 216, 30-36. (15) Cortinez, V.A.; Santagata, I. P.;Marone, C. B. Ann. Quim. 1973, 69, 343-349. (16) h i s , R. Elementary Chemical Reactor Analysis; Butterworths: Boston, 1989; p 100. (17) Putter, J; Becker, R. In Methods ofEnzymatic Analysis, 3rd ed.; Bergmeyer, H., Ed.; Verlag-Chemie: Weinheim, 1983; Vol. 3. (18) Aitken, M. D.; Massey, I. J.; Chen, T.; Heck, P. E. WaterRes. 1994, 28, 1879-1889. (19) Larson, R. A.;Weber, E. J.Reaction Mechanisms inEnvironmenta1 Organic Chemistry;Lewis Publishers: Boca Raton, FL, 1994; p 109.
Receiued f o r review October 5, 1994. Revised manuscript receiued July 7, 1995. Accepted July 7, 1995.@
ES940623A @Abstractpublished in Advance ACS Abstracts, August 15, 1995.
