United States Environmental Protection Agency Robert S. Kerr Environmental Research Laboratory Ada OK 74820 Research and Development EPA/600/S2-89/011 Sept. 1989 Project Summary Treatability Potential for EPA Listed Hazardous Wastes in Soil Raymond C. Loehr This study developed comprehen- sive screening data on the treatability in soil of: (a) specific listed haz- ardous organic chemicals, and (b) waste sludge from explosives pro- duction (KO44) and related chem- icals. Laboratory experiments were conducted using two soil types, an acidic soil (Mississippi soil) with less than one percent organic matter, and a slightly basic sandy loam soil (Texas soil) containing 3.25% organic matter. These experiments evaluated the: (a) relative toxicity of the chem- icals and waste, (b) degradation of the chemicals and waste in the soils, (c) adsorption characteristics of the chemicals in the two soils, and (d) toxicity reduction that occurred dur- ing degradation. The major conclusions were: 1. The chemical structure of the compounds evaluated affected their relative toxicity. With chlo- rophenols, the relative toxicity was related to the position of the chlorine group on the phenol ring. The order of relative toxicity was para >meta >ortho. The same order appeared to occur for methylphenols and nitrophenols. The chemical substituted on the phenol ring appeared to have an effect on toxicity. Nitro-substi- tuted phenols appeared to be less toxic than the methyl- or chloro- substituted phenols. Mixing of the chemicals with the soils did not affect the relative toxicity of the chemicals in the two soils. 2. Data characterizing the chemical loss in the soil and in the water soluble fraction (WSF) extracted from the soil as well as the toxicity reduction In the WSF could be represented satisfac- torily by either first or zero order kinetics. In most cases, the data were represented by either kinetic parameter with high correlation coefficients. 3. The rates of chemical loss were higher in the Texas soil. Chloro- phenols with chlorine substituted in the meta position had greater half-lives and lower loss rates. Chemicals with a nitro group substituted in the phenol ring ap- peared to have a lower loss rate. 4. The Freundlich equation de- scribed the adsorption of most of the chemicals with the two soils satisfactorily. The values of the Freundlich constant (Kf) for the chemicals in the two soils were different For the acid extracta- bles, the K, values generally were greater in the Mississippi soil. For the amines and alcohols, the K, values were greater in the Texas soil. 5. The loss of the applied chemical in the soil and in the WSF as well as the reduction of the WSF toxicity were compared for nine of the chemicals. The chemical loss in the WSF was about 1.5 times faster than the chemical loss in the soil. The WSF toxicity de- creased at about the same rate as the WSF chemical concentration. No enhanced mobilization of the applied chemical occurred during degradation. This Project Summary was devel- oped by EPA's Robert S. Kerr Environ- mental Research Laboratory, Ada, OK, to announce key findings of the research project that is fully docu- mented in a separate report of the same title (see Project Report order- ing information at back). ------- Introduction This study was conducted to provide comprehensive screening data on the treatability in soil of: (a) EPA listed hazardous organic chemicals, and (b) a specific hazardous waste and related chemicals. The results provide data that can be used when permitting decisions are made related to: (a) management of spills, (b) remediation of contaminated soils, and (c) the use of land as a waste management alternative. The degradation and partitioning data can be used as input to predictive models that estimate the movement of chemicals in the un- saturated zone of the soil. The models in- tegrate the processes that affect chem- icals in soil (degradation and partitioning) so that an assessment can be made of the extent to which protection of human health and the environment occurs. The understanding that results from the use of such models allows the identification of chemicals and wastes that require control to reduce or eliminate their hazard potential prior to application to soil. Lab- oratory studies were conducted to de- termine: (a) degradation kinetics, (b) sorption, (c) toxicity of the chemicals and waste, and (d) the reduction in toxicity that occurred during degradation. Designated Chemicals and Wastes The chemicals and specific waste that were part of this study are identified hazardous wastes. These chemicals can be expected to be components of many industrial compounds and wastes that enter the soil from spills and inadequately sealed impoundments (pits, ponds and lagoons) and as part of wastes applied to operating land treatment units. The chemicals that were evaluated are iden- tified in Table 1. The specific hazardous waste, and chemicals related to that waste, that were evaluated are noted in Table 2. Samples of the explosives waste sludge (K044) and the chemicals TNT, RDX, and HMX were obtained with the help of the U.S. Army Toxic and Haz- ardous Materials Agency (USATHAMA). A sample of wastewater treatment sludge resulting from the manufacture and processing of explosives was obtained from the Holston Army Ammunition Plant with the assistance of USATHAMA. This material was stored at 4°C until required for analysis and use. Soils The intent of this study was to provide comprehensive screening data on the treatability of specific chemicals and a hazardous waste in soil. The charac- teristics of the soil will affect the degradation, sorption, and treatment po- tential and two soils with different charac- teristics were used. One was an acid soil with a low organic content and the other was a basic soil with a higher organic content and cation exchange capacity (CEC). The acid soil, obtained from an area near Wiggins, Mississippi, was sup- plied by researchers at Mississippi State University and was referred to as Mis- sissippi soil. The basic soil was obtained from an area near Austin, Texas, that, to the knowledge of the personnel of this project, had not been exposed to in- dustrial chemicals or wastes. This soil was referred to as Texas soil. Relative Toxicity and Chemical Loading The relative toxicity tests that were conducted were not intended to provide information on toxicity from a human health or safety or from an environmental standpoint. Rather, these tests were used as a relative toxicity screening method. Such tests also can be used to identify the relative toxicity reduction that occurs when chemicals and waste are managed by the land treatment process. Although no single bioassay procedure can provide a comprehensive toxicity evaluation of a chemical, a valid toxicity screening test can provide information about the relative toxicity of a compound and can help predict non-inhibitory chemical application rates. The Microtox* system is a relatively simple, rapid and inexpensive test and was used as the toxicity screening method in this project to determine: (a) the relative toxicity of the chemicals and wastes, (b) the non- inhibitory chemical and waste loadings used in the degradation studies, and (c) the toxicity reduction that occurred in the respective studies. The use of the Microtox® procedure to screen and predict the treatability potential of waste in soil has been evaluated and found to be satisfactory. Degradation Studies These experiments were conducted to determine the removal kinetics of the designated chemicals and wastes in soil. Several of the chemicals could no evaluated because of chemical reac in the soils that made analytical detei impossible. These were Diphenylan m-Phenylenediamine and Thiophenol Biodegradation is believed to be most important removal mechanism organic compounds in soil systems. degradation of organics is accompli: in a series of biochemical reacti through which a parent compouni changed or transformed to organic inorganic end products. Complete de dation is the term used to describe process whereby constituents are mi alized to inorganic end products, inc ing carbon dioxide, water, and inorg nitrogen, phosphorus, and sulfur c pounds. Aerobic soil bacteria possess ability to biochemically catalyze oxidation of organic compounds. For reason, and because the zone of in poration at land treatment sites genei is aerobic, the protocol used in this st allowed aerobic conditions and aerc biodegradation reactions to occur. The primary goal of biodegrada testing is to obtain an overall estimati the rate at which a compound will I degrade in a soil environment. While compounds appear in the environmen pure form, a common approach studying removal rates of organic c< pounds has been to evaluate indivic compounds. Although this approach f vides an understanding of the rerm rates for specific compounds, it is rec nized that during actual land treatm chemicals normally are applied as n lures. Interactions between compounds a mixture within the soil matrix may p mote or inhibit their removal from s The noted chemicals (Tables 1 and were evaluated as individual compoui using standard laboratory microcos and protocol. In this study, no distinction was mi between specific loss mechanisms. I moval rates can be due to biodegrai tion, chemical degradation, hydroly; photolysis and volatilization. The che icals that were evaluated did not have high volatilization potential and volatili tion was not considered an important moval mechanism in these degradat experiments. The rate of degradation was expe mentally determined by measuring I difference between the amount of co pound initially added to a soil and tl which was recovered after specified tii ------- Table 1. Chemicals Compound Acid Extractables Phenol o-Cresol p-Cresol m-Cresol 2-Chlorophenol 3-Chlorophenol 4-Chlorophenol 2,3-Dichlorophenol 2, 4 -Dichlorophenol 2,5-Dichlorophenol 2, 6-Dichlorophenol 3,4-Dichlorophenol 2,4,5-Trichlorophenol 2,4, 6-Trichlorophenol Pentachlorophenol 2, 4 -Dimethyl phenol 2-Methyt-4-Chlorophenol 3-Methyl-4-Chlorophenol 3-MeV>yl-6-Chlorophenol p-Nitrophenol 2,4-Dinitrophenol 4, 6-Dinitro-o-Cresol Thiophenol Amines Diphenylamine m -Phenylenediamine Toluenediamine Brucine Alcohols Isobutyl alcohol Allyl alcohol Propargyl alcohol 1-Butanol 2, 3-Dichloropropanol Methanol Other Carbon disutfide 2-Nitropropane Thiourea that were Evaluated in this Study Formula C6H60 C7HgO C7H80 C7H80 CgH5CIO CeH5CIO CgH5CIO CeH4CI20 CeH4C/20 C6H4C/20 CeH4CI20 CeH4CI20 C6H3CI30 CeHyClsO CgHClsO CgH100 C-fH^IO C7H7CIO C7H7C/0 CeHgNOj CeH^Os CTH^OS CeHgS C12H,,N CeHeA/2 C7/VWH2>2 ^23^26^204 C4H,00 CaHgO C3H40 C4H,00 C3H6C/2 CH4O CS2 0^7^02 CH4W2S EPA Hazardous Waste Number U188 U052 U052 U052 U048 NOS NOS NOS U081 NOS U082 NOS U230 U231 U242 U101 NOS U039 NOS U170 P048 P048 U014 X016 X017 U221 P018 U140 POOS P102 U031 X006 U154 P022 U171 U219 ------- Table 2. The Hazardous Waste and Related Chemicals that were Evaluated Specific Hazardous Waste K044 - Wastewater treatment sludge from the manufacturing and processing of explosives Explosive and Munitions Manufacturing Chemicals Compound Formula EPA Hazardous Waste Number 2,4-Dinitrotoluene 2,6-Dinitrotoluene TNT (2,4,6-Trinitrotoluene) RDX + HMX+ + C7H6A/204 C7H6N204 C7H5N306 C3H6A/606 C^gNgOg U105 U106 - - - RDX = Hexahydrotrinitrotriazine + HMX = Cyclotetramethylenetetramtramine intervals. A plot of the disappearance of a constituent originally present in the chemical/soil mixture versus treatment time provided the following: (a) the type of reaction (generally zero or first order), (b) the reaction rate constants for the zero or first order reactions, and (c) the half-life (t1/2) time of each constituent of concern. The soil used: (a) had not had previous exposure to industrial chemicals or wastes, and (b) did not receive any pretreatment such as soil amendments or specially acclimated biological cultures prior to these experiments. The naturally occurring soil microbial consortium was responsible for the bioremediated re- moval of the chemicals. Chemical mass loadings were deter- mined as part of the toxicity screening evaluations and ensured that the loadings at which the chemicals were applied did not inhibit soil microbial activity. Soil pH was not adjusted nor were supplementary organic substrates used. Adsorption Experiments The persistence of hazardous organic compounds in soils is related to reactions that affect the transport and fate of such chemicals. One of the most important reactions is adsorption. Adsorption is the process by which ions or molecules present in one phase tend to concentrate at a surface or interface. The tendency of organic molecules to adsorb on soil is determined by the physical and chemical characteristics of the chemical compound and the soil to which it is added. The two driving forces for adsorption are the lyophobic (solvent-disliking) character of a solute relative to a particular solvent, and the affinity of the solute for the solid, such as electrical attraction. Adsorption is the major retention mechanism for most organic and inor- ganic compounds in soils. As a result, the leaching potential of a chemical in soil is, in general, proportional to the magnitude of the adsorption (partitioning) coefficient of that chemical in a soil. The adsorption potential of a chemical is governed by the properties of both the soil and the chemical. Important properties of the chemical that affect adsorption include: (a) chemical structure, (b) acidity of basicity of the molecule (pKa or pKb), (c) water solubility, (d) permanent charge, (e) polarity, and (f) molecule size. At equilibrium, the solute remaining in solution is in dynamic equilibrium with that of the soil surface. At this point, there is a defined distribution of solute between the liquid and solid phases. The preferred form for depicting this distribution is to express the quantity qe (amount of solute sorbed per unit weight of solid sorbent) as a function of the equilibrium solution concentration (Ce) at a fixed temperature. An expression of this type is an adsorption isotherm. An adsorption equation that has be used widely for solid-liquid systems the Freundlich equation: Qe = Kf C,"" where qe is the equilibrium distribut coefficient (mg of chemical/gm of j sorbent), Ce is the equilibrium chemi concentration (mg/liter of solvent), and and 1/n are constants The constant, is related to the capacity or affinity of I adsorbent and the exponential term, 1 is an indicator of the intensity, or how I capacity of the adsorbent varies with 1 equilibrium solute concentration. T Freundlich isotherm has had success describing sorption behavior of organ and the adsorption data generated in tl study were evaluated by this equation. Toxicity Reduction A major objective of this study was provide comprehensive screening d< on the treatability of specific orgar chemicals in soil. Hazardous constituer that enter the soil are to be detoxified immobilized. When a chemical is add to the soil, it is transformed into otr products through chemical and biologic reactions with or without complete detc ification and immobilization. Measuri the loss of the parent compound does r assure that complete detoxification a immobilization occurs. Intermediate d radation products, which may be ------- mobile and/or toxic than the parent compound, may be generated as the parent compound degrades. Additional information on the transformation and/or detoxification of a chemical is necessary to establish that the loss of the parent compound leads to the complete detoxifi- cation of the chemical or waste. Such information can be obtained using either chemical or bioassay analyses. The reduction of toxicity that occurred in selected degradation studies was eval- uated by determining the toxicity of the water soluble fraction (WSF) of the chem- ical/soil mixture at the same sampling intervals used to obtain the degradation data. The chemical compounds that can be extracted with water represent the potentially leachable fraction of the chemical or any intermediate chemical detoxification products. The WSF of the chemical poses the greatest threat to groundwater contamination. Hence, evaluating the loss of the potentially leachable fraction of a chemical is important. The concentration of the parent chemical in the WSF also was de- termined. This concentration was ex- pressed in terms of quantity of chemical that was water extractable per kg of the soil. To put the toxicity reduction data in perspective, the WSF toxicity reductions, he WSF chemical concentration reduc- tions and the soil chemical concentration reductions were compared. Conclusions Specific conclusions based on the results of the project include: 1. The Microtox0 biological assay rep- resents an appropriate method with which to evaluate the EC50 toxicity of a chemical or waste. 2. Comparison of the EC50 data indi- cated that: (a) the alcohols were less toxic than the acid extractable com- pounds, and (b) within chemical categories, there were considerable differences in relative toxicity. 3. The chemical structure of the com- pounds evaluated affected the rela- tive toxicity of a compound. With chlorophenols, the relative toxicity was related to the substitution posi- tion of the chlorine group on the phenol ring. The order of relative toxicity was para>meta>ortho. The EC50 data suggested that the same order occurred for methylphenols and nitrophenols. 4. The chemical that was substituted on the phenol ring appeared to have an effect on toxicity. Nitro-substi- tuted phenols, even when substi- tuted in the para position, appeared to less toxic than the methyl- or chloro-substituted phenols. 5. When the chemicals were mixed with two different soils, and the EC50 value of the water soluble fraction (WSF) of the soil mixtures was measured, the values also indicated that chemicals with the chlorine in the para position had the greater toxicity. Mixing of the chemicals with the soils did not affect the relative toxicity of the chemicals in the two soils. 6. In general, the acceptable non- inhibitory chemical loading rates for the Mississippi soil were lower than those for the Texas soil. There was no consistent pattern for the differences. 7. The chemical or waste loading pro- cedure (described in Table 9, Sec- tion 4 of the full report) resulted in chemical loadings that did not inhibit the non-acclimated organisms in the laboratory microcosms, except in one case (4,6-Dinitro-o-Cresol). This procedure provided a good estimate of initial, acceptable chemical loadings that can be used in lab- oratory degradation studies. 8. Both zero and first order kinetics provided adequate representation of the data. For most of the chemicals, the data could be fit to either kinetics with high correlation coef- ficients. 9. The rates of chemical loss were higher in the Texas soil than in the Mississippi soil. There did not ap- pear to be any pattern to the differ- ences in rates in the two soils. 10. Chlorophenols with the chlorine sub- stituted in the meta position had greater half-lives and therefore lower chemical loss rates. This was par- ticularly evident with the mono-, di-, and trichlorophenols in the Texas soil. 11. Chemicals that had a nitro group substituted on the phenol ring ap- peared to have a lower loss rate. 12. The Freundlich equation described the adsorption of the chemicals on the two soils satisfactorily, with high correlative coefficients, except for a few chemicals. 13. The range of chemical concentra- tions evaluated ranged from the low mg/l concentrations to near or at saturation concentrations, and for most chemicals covered two to three orders of magnitude. For these con- centration ranges, a linear ad- sorption relationship, i.e., n = 1, did not occur. 14. The values of the Freundlich con- stant (Kf) for the chemicals in the two soils were different. For the acid extractables, the Kf values gen- erally were greater in the Texas soil which had the higher pH and the greater organic carbon content. For the amines and alcohols, the K, values were greater in the Mis- sissippi soil, which had the lower pH and the lower organic carbon content. 15. Two loading rates, the Texas soil, and nine chemicals (phenol and eight chlorinated phenols) were used in this study. Both first and zero order kinetics satisfactorily fit the water soluble fraction (WSF) chemical loss data and the toxicity reduction data. 16. The higher chemical loading rates resulted in higher chemical concen- trations in the WSF and higher WSF toxicities at the beginning of the experiments. 17. The higher chemical loading rates generally resulted in slower chemical losses (higher half lives) and slower toxicity reduction. However, at both loading rates for each chemical, the chemicals were degraded and the toxicity was reduced. No differences due to the loading rates were apparent in zero order kinetics. 18. The loss of the chemicals in the WSF was about 1.5 times faster than the loss of the chemical in the soil. 19. The WSF toxicity for each chemical decreased as the soil chemical and the WSF chemical concentrations decreased. 20. The WSF toxicity decreased at about the same rate as the WSF chemical concentration when the data for all nine chemicals were compared. 21. No enhanced mobilization of the ap- plied chemicals occurred as the deg- radation and detoxification occurred. 22. No water soluble toxic products ap- peared to be formed as the chem- icals were degraded in the soil. 23. The Freundlich equation described the sorption of 2,4- and 2,6-Dinitro- toluene in the two soils satisfactorily. It did not do so for TNT, RDX, or HMX. ------- 24. No loss of 2,4-Dinitrotoluene occur- red over a 47-day study even though the loading rate used was de- termined to be acceptable using pro- cedures discussed in Section 4 of the full report. Degradation loss rates were obtained for 2,6-Dinitrotoluene and TNT. First order kinetics were a better representation for TNT than were zero order kinetics. 25. The half life of TNT in the Mississippi soil was shorter, and the loss faster, than in the Texas soil. No difference in the loss rates in the two soils for 2,6-Dinitrotoluene was apparent. 26. The sludge resulting from the manufacture and processing of ex- plosives contained: (a) high concen- trations of nitrogen and COD, (b) concentrations generally less than 10 mg/l for heavy metals, and (c) no TNT, RDX or HMX. 27. The munitions sludge had a high toxicity as measured by the Microtox® procedure. The constitu- ents causing the relative toxicity were in the soluble phase of the sludge. ------- ------- Raymond C. Loehr is with the University of Texas at Austin, Austin, TX 78712. Scott G. Hullng is the EPA Project Officer (see below). The complete report, entitled "Treatability Potential for EPA Listed Hazardous Wastes in Soil," (Order No. PB 89-166 581/AS; Cost: $21.95, subject to change) will be available only from: National Technical Information Service 5285 Port Royal Road Springfield, VA22161 Telephone: 703-487-4650 The EPA Project Officer can be contacted at: Robert S. Kerr Environmental Research Laboratory U.S. Environmental Protection Agency Ada, OK 74820 United States Environmental Protection Agency Center for Environmental Research Information Cincinnati OH 45268 -., I Official Business Penalty for Private Use $300 EPA/600/S2-89/011 885 8R(NCH IL CHICAGO ------- |