Determination of physicochemical properties of tetrabromobisphenol A

Aqueous solubility (Sw), 1-octanol/water partition coefficient (Kow), and vapor pressure of the nonionic form of 2,2',6,6'-tetrabromo-4,4'-isopropylidenediphenol (tetrabromobisphenol A or TBBP-A) were measured. From this, enthalpies of solution and vaporization were estimated. Furthermore, enthalpy of fusion and melting point were measured to estimate subcooled liquid vapor pressure, the infinite dilution activity coefficient, and Henry's law constant. Since TBBP-A is expected to exit in both ionic and nonionic forms at near neutral pH, pH effects on physicochemical properties were also examined. Because of the ionization of TBBP-A, Sw increased by five orders of magnitude, while Kow decreased by eight orders of magnitude. Furthermore, an analytical model based on mass balance and dissociation of TBBP-A was applied to represent the pH dependence.     

Vapor pressures,aqueous solubilities,and Henry's law constants of some brominated flame retardants

The subcooled liquid vapor pressures (P0(L),25S) and aqueous solubilities (Sw,25s) were determined and Henry's law constants (H25s) were estimated for a number of brominated flame retardants (BFRs) at 25 degrees C. The established methods of the gas chromatography-retention time and generator column techniques were used to experimentally determine P0(L),25 and Sw,25 for hexabromobenzene and a series of brominated diphenyl ether (BDE) congeners. The H25 was estimated as the ratio of P0(L)25 to the subcooled liquid aqueous solubility. Values of PL0(L),25 obtained ranged from 0.000000282 Pa (BDE-190) to 0.259 Pa (BDE-3); Sw,25 ranged from 0.00000087 g/L (BDE-153 and BDE-154) to 0.00013 g/L (BDE-15); and H25 ranged from 0.0074 Pa m3/mol (BDE-183) to 21 Pa m3/mol (BDE-15). An increase in the bromine content of polybrominated diphenyl ether congeners resulted in significant decreases Of P0(L),25, Sw25, and H25. A simple four-compartment equilibrium distribution model suggested that the majority of BFRs being released into the environment would reside in soil and sediment and have localized distributions. The model also suggested that lower brominated congeners tend to be somewhat more mobile. Degradative debromination reactions that yield these congeners would mobilize them environmentally, and ultimately affect the fate and distribution of BFRs.     

Acute aquatic toxicity of alkyl phenol ethoxylates

The recently derived log Kow (octanol/water partition coefficient in logarithmic form) increment for a nonterminal oxyethylene unit was used to calculate a quantitative structure-activity relationships for literature data on the acute crustacean toxicity of polyoxyethylene surfactants. The resulting log Kow regression parameters are between the corresponding values for nonpolar and polar narcosis, which supports an interpretation of the surfactants' aquatic toxicity on the basis of another distinct mode of action. Furthermore, a comparison with calculated water solubility data indicates that for log Kow greater than 5 an aquatic toxicity decrease due to a solubility limit is expected, which gets support from two other sets on toxicity data of nonyl phenol polyethoxylates.     

Use of polynomial expressions to describe the bioconcentration of hydrophobic chemicals by fish

For the bioconcentration of hydrophobic chemicals by fish, relationships have been previously established between uptake rate constants (k1) and the octanol/water partition coefficient (Kow), and also between the clearance rate constant (k2) and Kow. These have been refined and extended on the basis of data for chlorinated hydrocarbons, and closely related compounds including polychlorinated dibenzodioxins, that covered a wider range of hydrophobicity (2.5 less than log Kow less than 9.5). This has allowed the development of new relationships between log Kow and various factors, including the bioconcentration factor (as log KB), equilibrium time (as log teq), and maximum biotic concentration (as log CB), which include extremely hydrophobic compounds previously not taken into account. The shape of the curves generated by these equations are in qualitative agreement with theoretical prediction and are described by polynomial expressions which are generally approximately linear over the more limited range of log Kow values used to develop previous relationships. The influences of factors such as hydrophobicity, aqueous solubility, molecular weight, lipid solubility, and also exposure time were considered. Decreasing lipid solubilities of extremely hydrophobic chemicals were found to result in increasing clearance rate constants, as well decreasing equilibrium times and bioconcentration factors.     

Validation of CHEMEST, an on-line system for the estimation of chemical properties

CHEMEST is an on-line system for estimating physical/chemical properties important in the assessment of environmental fate of chemicals. Measured values of boiling point, melting point, vapor pressure, water solubility, and soil organic carbon partition coefficient (Koc) were compared to values estimated by CHEMEST, for more than 170 chemicals. The chemicals were composed of seven classes: alcohols, aldehydes and ketones, amines, carbamates, esters, ethers, and phenols. Mean errors for boiling and melting point were 22.6 and 36.2 degrees C, respectively, with relatively little variability among classes. Carbamates were an exception, accuracy being much lower for both properties. Vapor pressures were quite accurately estimated, measured and estimated values generally differing by less than a factor of two. Mean error factors for water solubility and Koc were somewhat higher, but nevertheless better than the order of magnitude accuracy commonly assumed. Solubility was very inaccurately estimated for the phenols, however. The results indicate that, for the chemicals and properties studied, estimated values were sufficiently accurate for screening-level fate assessment. TSTCHEM, a revised version of CHEMEST, out-performed the latter with respect to the accuracy of solubility and vapor pressure estimates.     

Performance of on-line chemical property estimation methods with TSCA premanufacture notice chemicals.

CHEMEST is an interactive on-line system for estimating chemical properties important in environmental fate assessment. In 1988 we described an extensive validation study of CHEMEST. That study showed that for existing chemicals, most properties could be estimated with accuracy sufficient for screening-level assessment. Reported here are the results of a study of system performance with "new" chemicals, for which Premanufacture Notices (PMNs) must be submitted under the Toxic Substances Control Act. Measured values of key properties were retrieved for more than 300 PMN chemicals having discrete structures, which represented the majority of notices with submitted data that were received by the U.S. Environmental Protection Agency during the period 1979-1990. The measured values were compared to estimates from AUTOCHEM, an automated version of CHEMEST. Errors were generally greater than those for existing chemicals, as expected. Water solubility and the octanol/water partition coefficient (Kow) were estimated with acceptable accuracy, but the results for melting point and boiling point indicate that additional effort is needed to improve the estimation methods.     

Predicting the toxicity of neat and weathered crude oil: toxic potential and the toxicity of saturated mixtures

The toxicity of oils can be understood using the concept of toxic potential, or the toxicity of each individual component of the oil at the water solubility of that component. Using the target lipid model to describe the toxicity and the observed relationship of the solubility of oil components to log (Kow), it is demonstrated that components with lower log (Kow) have greater toxic potential than those with higher log (Kow). Weathering removes the lower-log (Kow) chemicals with greater toxic potential, leaving the higher-log (Kow) chemicals with lower toxic potential. The replacement of more toxically potent compounds with less toxically potent compounds lowers the toxicity of the aqueous phase in equilibrium with the oil. Observations confirm that weathering lowers the toxicity of oil. The idea that weathering increases toxicity is based on the erroneous use of the total petroleum hydrocarbons or the total polycyclic aromatic hydrocarbons (PAHs) concentration as if either were a single chemical that can be used to gauge the toxicity of a mixture, regardless of its makeup. The toxicity of the individual PAHs that comprise the mixture varies. Converting the concentrations to toxic units (TUs) normalizes the differences in toxicity. A concentration of one TU resulting from the PAHs in the mixture implies toxicity regardless of the specific PAHs that are present. However, it is impossible to judge whether 1 microg/L of total PAHs is toxic without knowing the PAHs in the mixture. The use of toxic potential and TUs eliminates this confusion, puts the chemicals on the same footing, and allows an intuitive understanding of the effects of weathering.     

Brominated Indoles and Phenols in Marine Sediment and Water Extracts from the North and Baltic Seas–Concentrations and Effects

This work presents results from analytical as well as ecotoxicologic investigations of sediment and water samples from the North and Baltic Seas. A bioassay-directed procedure was used to investigate cause–effect relationships between observed effects in acute laboratory bioassays (luminescent bacteria assay with Vibrio fischeri and embryo test with Danio rerio) and analyte concentrations in extracted samples. Brominated phenols and indoles—including 4-bromophenol; 2,4-dibromophenol; 4- and 6-bromoindole; 3,4-, 4,6-, and 3,6-dibromoindole; and tribrominated compounds—were identified in partly remarkable concentrations (up to 40,000 ng g⁻¹ total organic carbon TOC for 4-bromophenol) in North Sea sediment samples and water samples (913 ng L⁻¹ 3,6-dibromoindole) from the German Bight. The toxicity of some of the identified brominated substances was low, with median effect concentration levels (EC₅₀) ranging from 0.08 to 21.7 mg/L for V. fischeri and 4.3 to 46.3 mg/L for D. rerio. Comparison of the concentrations of analytes with ECs showed a toxicity contribution of brominated phenols and indoles to overall toxicity of the fraction. In the case of one water sample from the German Bight, brominated phenols and indoles accounted for the observed toxicity. Brominated phenols and indoles, which are assumed to be of biogenic origin, have rarely been discussed so far in the context of ecotoxicologic effects in marine ecosystems.     

A comparison of the fish bioconcentration factors for brominated flame retardants with their nonbrominated analogues

Flame retardants (FR) play a significant role in reducing the flammability of many consumer products. On a volume basis, approximately 25% of the FRs in use today utilize bromine as the active flame-retarding moiety. Their applications are those requiring high FR performance or in resins needing an FR that is active during the gas phase. Laboratory fish bioconcentration factors (BCFs) for 11 brominated FRs (BFRs) or their components were compared with those for their nonbrominated analogues. Bioconcentration, defined here as a BCF of greater than 1,000, was not observed in those brominated molecules examined with molecular weights of 700 or greater. These included the decabromodiphenyl oxide and octabromodiphenyl oxide commercial products, ethane 1,2-bis(pentabromophenyl), ethylene bis-tetrabromophthalimide, and decabromobiphenyl. Tetrabromobisphenol A, with a molecular weight of less than 700, also did not bioconcentrate. This likely relates to the ease with which it is metabolized and eliminated. Within the BFR class of polybrominated diphenyl oxides/ethers, the BCFs for those congeners with molecular weights of between approximately 450 and 700 varied with the number and position of the bromine atoms. The BCF of hexabromocyclododecane appeared to be related to its cyclododecane substructure, not to its bromine content. Bioconcentration was not a characteristic feature of the BFRs examined here.     

间接分光光度法测定空气中酚

建立了一种间接分光光度法测定作业场所空气中酚的方法。在Ｈ２ＳＯ４介质中,Ｂｒ＾－ＢｒＯ３＾－反应生成Ｂｒ２,Ｂｒ２与酚定量反应,剩余的溴使甲基橙褪色。酚浓度在０．０１－０．４２ｍｇ／Ｌ范围内符合比尔定律,空气中最低检出浓度为０．０６ｍｇ／ｍ＾３,变异系数为２．０％－８．０％,采样效率在９７％以上,该方法简单、快速、灵敏。     

Physicochemical and toxicological studies on 4-chloro-3,5-dinitrobenzoic acid in aqueous solutions

Physicochemical characterization of hazardous compounds often is required for the development of structure-reactivity correlations. Physical, chemical, and toxicological properties of target pollutants require determination for an efficient application of wastewater treatments. In the present work, we chose a chloro-nitro-aromatic derivative (4-chloro-3,5-dinitrobenzoic acid [CDNBA]), as a model compound on which to perform physicochemical and toxicological studies. Several properties of CDNBA are not available in the literature, although many aromatic nitro-compounds are considered hazardous materials. Measurements of solubility in water, acid dissociation constant, and kinetic parameters for the nucleophilic substitution of chlorine atom in alkaline media are reported. We also performed cytotoxicity studies of CDNBA and ultraviolet-irradiated CDNBA solutions. From the analysis of CDNBA solubility in water at different temperatures, an enthalpy of solution of 23.2 +/- 2.5 kJ/mol was found. The study of the acid dissociation constant Kc by using conductivity measurements and the modified Gran's method yielded values for the equilibrium constant Ka of 2.36 x 10(-3) and 2.26 x 10(-3), respectively. The bimolecular rate constant for the reaction of CDNB- and hydroxyl ion (HO) measured at room temperature and 0.1 M of ionic strength was 5.92/M x s, and the activation energy for this process was 70.7 +/- 3.4 kJ/mol. Cytotoxicity assays with aqueous suspensions of Tetrahymena pyriformis showed lethal effects due to the pH change induced by CDNBA. On the other hand, in buffered solutions, a value of 104.47 microM was observed for the median effective concentration, that is, the concentration of CDNBA at which the proliferation was restricted to one half of the blank. Irradiation of CDNBA solutions increased the toxicity, suggesting the formation of intermediate products with higher cytotoxicity effects.     

Toxic Effects of Brominated Indoles and Phenols on Zebrafish Embryos

Organobromine compounds in the marine environment have been the focus of growing attention in past years. In contrast to anthropogenic brominated flame retardants, other brominated compounds are produced naturally, e.g., by common polychaete worms and algae. Brominated phenols and indoles assumed to be of biogenic origin have been detected in water and sediment extracts from the German Bight. These substances as well as some of their isomers have been tested with the zebrafish embryo test and were found to cause lethal as well as nonlethal malformations. The zebrafish test was able to detect a log K_OW–related toxicity for bromophenols, suggesting nonpolar narcosis as a major mode of action. Different effect patterns could be observed for brominated indoles and bromophenols. The comparison of effective concentrations in the zebrafish embryo test with the concentrations determined in water samples suggests the possibility that brominated indoles may affect early life stages of marine fish species in the North Sea.     

Bioaccumulation and trophic transfer of some brominated flame retardants in a Lake Winnipeg (Canada) food web

The extent of bioaccumulation and trophic transfer of brominated diphenyl ether (BDE) congeners, hexabromocyclododecane (HBCD) diastereoisomers (alpha, beta, and gamma), decabromodiphenylethane (DBDPE), and bis(2,4,6-tribromophenoxy)ethane (BTBPE) was examined in a Lake Winnipeg (Canada) food web. Six species of fish, zooplankton, mussels, sediment, and water from the south basin of the lake were selected for study. Significant positive correlations were found between concentrations of total (sigma) polybrominated diphenylethers (PBDEs; p < 0.005), sigmaHBCDs (p < 0.0001), BTBPE (p < 0.0001), and lipid content in fish. Strong positive linear relationships also were observed from individual plots of BDE 47, BDE 209, and DBDPE concentrations (lipid wt) and trophic level (based on delta15N), suggesting that these compounds biomagnify in the Lake Winnipeg food web. Biomagnification factors varied for the chemicals studied. Plots of log bioaccumulation factors for mussel and zooplankton versus log octanol-water partition coefficient (Kow) were similar and suggest that neither mussels nor zooplankton are in equilibrium with the water. Fifteen BDE congeners were consistently detected in water (dissolved phase, n = 3), with BDE 47 having the greatest concentration (17 pg/L). The rank order of compounds in water (arithmetic mean +/- standard error) were sigmaPBDEs (49 +/- 12 pg/ L) > alpha-HBCD (11 +/- 2 pg/L) > BTBPE (1.9 +/- 0.6 pg/L). Concentrations of DPDPE, BDE 209, and beta- and -gamma-HBCD isomers were below their respective method detection limits (MDLs) in water. Total PBDE concentrations in sediment (n = 4) were greater than any other brominated flame retardant examined in the present study and ranged from 1,160 to 1,610 ng/g (dry wt), with BDE 209 contributing roughly 50% of the total. The gamma-HBCD isomer was detected at concentrations of 50 +/- 20 pg/g (dry wt) in sediment, whereas BTBPE and DBDPE were consistently below their respective MDLs in sediment.     

Factors influencing the design of bioaccumulation factor and biota-sediment accumulation factor field studies

A series of modeling simulations was performed to evaluate the underlying factors and principles that drive the uncertainty in measured bioaccumulation factors (BAFs) and biota-sediment accumulation factors (BSAFs) for fish, and to determine which sampling designs minimize those uncertainties. Temporal variability of chemical concentrations in the water column, and the metabolism rate and n-octanol-water partition coefficient (Kow) for the chemical were determined to be the dominant factors that influenced the field-sampling design. Spatial variability of the chemical concentrations, food web structure, and the sediment-water column concentration quotient had a lesser importance upon the overall design. The simulations also demonstrated that collection of composite water samples in comparison to grab water samples resulted in reductions in the uncertainties associated with measured BAFs. Some illustrative sampling design structures for BAF and BSAF measurements based upon the temporal variability of chemical concentrations in the water column and the metabolism rate and Kow for the chemical were developed. These illustrative designs provided a sense of how sampling design structures, that is, the number and spacing over time of sampling events, might be influenced by differences in temporal variabilities, metabolism rates, and Kow. Although the importance of spatial variability was discounted as a major factor in the design process, sampling of water and sediment across the immediate home range of the sampled organism is required for successful measurements because poor spatial coordination of organism, water, or sediment samples will yield BAFs and BSAFs with large and unknown biases.     

Automated high performance liquid chromatography and liquid scintillation counting determination of pesticide mixture octanol/water partition rates

Two novel methods are reported for measuring octanol/water partition rates of pesticides. A liquid scintillation counting (LSC) method was developed for automated monitoring of 14C-labeled pesticides partitioning in biphasic water/octanol cocktail systems with limited success. A high performance liquid chromatography (HPLC) method was developed for automated partition rate monitoring of several constituents in a pesticide mixture, simultaneously. The mean log Kow +/- SD determined from triplicate experimental runs were for: 2,4-D-DMA (2,4-dichlorophenoxyacetic acid dimethylamine), 0.65 +/- .17; Deet (N,N-diethyl-m-toluamide), 2.02 +/- .01; Guthion (O,O-dimethyl-S-(4-oxo-1,2,3-benzotriazin-3(4H)-ylmethyl) phosphorodithioate), 2.43 +/- .03; Methyl-Parathion (O,O-dimethyl-O-(p-nitrophenyl) phosphorothioate), 2.68 +/- .05; and Fenitrothion (O,O-dimethyl O-(4-nitro-m-tolyl) phosphorothioate), 3.16 +/- .03. A strong positive linear correlation (r = .9979) was obtained between log Kow and log k' (log Kow = 2.35 (log k') + 0.63). The advantages that this automated procedure has in comparison with the standard manual shake-flask procedure are discussed.