United States Environmental Protection Agency Environmental Monitoring Systems Laboratory Las Vegas, NV 89193-3478 Research and Development EPA/600/S4-89/022 Nov. 1989 Project Summary Performance Testing of Method 1312-QA Support for RCRA Testing T. C. Chiang, C. A. Valkenburg, D. A. Miller, and G. W. Sovocool The question of how to access the risks associated with ground water contamination from soils containing toxic substances or wastes disposed of in a monofill environment is a crit- ical issue for the U.S. Environmental Protection Agency (EPA). A major limitation of using Methods 1310 and 1311 for this purpose is the fact that the sanitary landfill co-disposal scen- ario does not apply to contaminated soils or wastes disposed of In a monofill environment If these meth- ods are used to assess sites for cleanup purposes, the acetic acid leaching fluid could selectively solu- billze toxicants (specifically lead) and incorrectly classify the soil or waste as hazardous when, In fact, no mobilization (leaching) would be ex- pected to occur In the environment The EPA is considering the use of a newly-created synthetic acid precip- itation leach test for soils and wastes (Method 1312) to provide information about the mobility (teachability) of both organic and Inorganic contam- inants present in these materlals.Thls new test method is similar to the TCLP (Method 1311) except that the acetic acid buffer extraction fluid has been replaced by a dilute nitric acld/sulfurlc acid mixture. This acid mixture simulates the nature of the precipitation occurring in the region where the soil sample originated. A pH 4.2 acid precipitation solution is used for extraction of wastes. The purpose of the full report Is to present results obtained from a pre- cision evaluation of and ruggedness test for Method 1312 for soils only. Several different soils were fortified with semi-volatile organlcs, metal salts and volatile organics, and then leached in replicates of 3 or 6 to determine method precision. A rug- gedness evaluation was performed by making minor changes In speci- fied method values to Identify pro- cedural variations requiring careful control. This Project Summary was devel- oped by EPA's Environmental Monitor- Ing Systems Laboratory, Las Vegas, NV, 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 The full report summarizes the quality assurance support provided to the Office of Solid Waste in FY-88 and FY-89 by the Quality Assurance and Methods Devel- opment Division, Environmental Monitor- ing Systems Laboratory-Las Vegas, Office of Research and Development, under D109 QO1, "QA Support for RCRA." The major activity conducted by the EMSL-LV under D109 Q01 was the evaluation of EPA Method 1312, a pro- posed synthetic acid precipitation leach test for soils and wastes. The question of how to assess the risks associated with ground water contam- ination from soils containing toxic sub- stances or wastes disposed of in a monofill environment is a critical issue for the EPA. The large number of samples needing analysis under legislative man- date requires that a leaching procedure be rapid, accurate, reproducible, rugged, ------- and suitable for a variety of matrices. A major limitation of using Methods 1310 and 1311 for this purpose is the fact that the sanitary landfill co-disposal scenario does not apply to contaminated soils or wastes disposed of in a monofill environ- ment. If these methods are used to assess sites for cleanup purposes, the acetic acid leaching fluid could selectively solubilize toxicants (specifi- cally lead) and incorrectly classify the soil or waste as hazardous when, in fact, no mobilization (leaching) would be ex- pected to occur in the environment. The EPA is considering the use of a newly created synthetic acid precipitation leach test for soils and wastes (Method 1312) to provide information about the mobility (teachability) of both organic and inor- ganic contaminants present in these materials. This new test method is similar to the TCLP (Method 1311) except that the acetic acid buffer extraction fluid has been replaced by a dilute nitric acid/ sulfuric acid mixture. This acid mixture is either pH 4.2 or 5.0, which simulates the nature of the precipitation occurring where the soil sample originated. A pH 4.2 acid precipitation solution is used for extraction of wastes. The full report summarizes results ob- tained from a precision evaluation of and ruggedness test for Method 1312 for soils only. Several different soils were fortified with semi-volatile organic compounds, metal salts and volatile organic com- pounds, and then leached in replicates of 3 or 6 analyses to determine method pre- cision. Minor changes were made in specified method values, and a rug- gedness evaluation was performed to identify procedural variations requiring careful control. Experimental Design Extractabfe Compounds Procedure Two different types of soil were used in the single laboratory (EMSL-LV, LESC) precision and ruggedness evaluation of the Method 1312 protocol: an eastern soil with high organic content and a western soil with low organic content (sandy type). These soils were first screened for background level and then fortified with selected TCLP target compounds at levels suitable for regu- latory purposes. Refer to Table 2 for the amount of each extractable compound typically spiked into 100 grams of soil. The spiked soil samples were leached according to the procedure described in Method 1312. Triplicate aliquots of the eastern soil were" leached using four different extraction fluids (pH 3.2, 4.0, 5.0, 6.1) to determine if the pH of the leaching fluid has a significant effect on either method precision or recovery. Triplicate aliquots of the western soil were spiked at two different analyte concentration levels, leached, and analyzed to obtain data on the matrix sensitivity of the method as well as its dynamic range. A ruggedness evaluation was designed to determine the sensitivity of Method 1312 to modest departures from the leaching protocol which can be expected during routine application of the protocol. The ruggedness evaluation of Method 1312 for both volatile and non-volatile species was performed by following the test procedure of Youden and Steiner (Statistical Manual of the AOAC, 1975) which is designed to determine the level of significance for n variables using just n +1 different measurements (in this case seven variables were chosen with eight experiments). For the semi-volatile com- pounds and the metals, the western soil was used for the ruggedness evaluation. Volatile Compounds Procedure The single laboratory precision and ruggedness evaluation of Method 1312 for volatile organic compounds was per- formed by Acurex and Midwest Research Institute (MRI) as a single parallel labora- tory study. Two contaminated soils orig- inating in the western (Soil 1) and eastern (Soil 2) U.S. from Superfund sites and two soils prepared by combining a clean soil from Hayward, California with two different municipal sludges in the labora- tory (identified as California Urban Soils 3 and 4) were used for the precision evalu- ation of Method 1312. These soil samples were analyzed to measure organic and inorganic contaminants in order to estab- lish appropriate spiking levels for the fortification mixture containing 27 Method 1312 volatile target analytes. The fortified soils were leached in the ZHE and analyzed according to the procedure in Method 1312. The California Model Urban Soils (Soils 3 and 4) were prepared by separately mixing clean Hayward, California soil (dried overnight at 120°C) with two municipal sludges from San Francisco Bay area sites. They were prepared to simulate a worst case soil that might be tested by Method 1312. Aliquots of 25 grams of Soils 3 and 4 were placed into a ZHE, then spiked directly with the ZHE piston up using 240 pL of fortification solution containing 200 pg/mL of 27 vola- tile analytes (Table 3). The ZHE was assembled as quickly as possible ar then was placed in a refrigerator at 4" for 1 hour prior to the addition of leachir fluid. Following this equilibration perio the sample was leached. The soil lead ate was transferred to an evacuate Tedtar bag. Each bag was used to t several VOA vials. These vials wei capped and then immediately stored 4°C until they were analyzed. The leac ate was analyzed for volatile compoum by GC/MS (Method 8260). Soils 3 and were leached in replicates of 6 analyst to obtain additional data to evalua precision. For the Method 1312 rugge ness test for volatile compounds usir the ZHE, the clean Hayward, Californ soil was used. Refer to Tables 6 and 7 f the variables and fortification levels. Results and Discussion Extractable Compounds Precision The precision of Method 1312 w; evaluated by measuring the repeatabili of recovery of 14 semi-volatile organ compounds, lead and cadmium. Table presents the data for an eastern soil, ar Table 2 that for a western soil. Tt recovery determinations were made t fortifying soil samples prior to using ti leaching procedure. The reported vai ability is a combination of that from ti leaching test (Method 1312) and th from the analytical methods (Metho< 3250/8270 for organics and Method 60; for inorganics). The variability of tl organic analytical methods can be es mated from the RSDs (relative standa deviations) of the organic surrogate thus, they are reported with the samp data. The organic surrogates added the eastern soil leachate had recover!) greater than 60 percent and RSDs lei than 9 percent while the recoveries f the semi-volatile analytes varied from G to 75 percent and the RSOs for analyti with reasonable recoveries ranged from to 12 percent. The RSDs of the recovc ies of most compounds reported in Tab 1 are less than 10 percent. The precis!) of the Method 1312 recoveries for me compounds at all four pH values similar to those of the GC/MS surrogat and is better than the precision obtain* using Method 1311. Large (greater th, 15 percent) RSDs were observed for fo compounds. Three of these, 1, dichlorobenzene, 2,4-dimethylphenol, ai 2,4-dinitrophenol, present analytical diffi ulties due to their volatility or reactivi The fourth compound, hexachlorobe zene, had very low recovery and its ve ------- Table 1. Method 7372 Precision Results from Eastern Soil' Extraction Fluid pH 3.2 FORTIFIED ANALYTES bis(2-chloroethyl)-ether 2-Chlorophenol 1 -4,Dichlorobenzene 1 -2,Dichlorobemene 2-Methylphenol Nitrobenzene 2,4-Dimeftylphenol Hexachlorobutadiene Acenaphthene 2, 4-Dinitrophenol 2,4-Dinitrotoluene Hexachlorobenzene y-BHC P-BHC METALS Lead Cadmium SURROGATES (in Leachate) 2-Fluorophenol ds-Phenol ds-Nitrobenzene 2-Fluorobiphenyl 2,4,5-Tribromophenol d14-p-Terphenyl Avg. % flee. 75.6 61.4 15.8 11.5 47.6 72.9 12.3 1.2 4.9 60.4 57.5 0.1 3.46 5.3 3.2*> 55.7 62.0 77.0 64.5 62.0 68.8 89.3 %flSO 9.5 8.8 5.3 20.2 9.9 2.3 9.7 3.5 1.0 16.3 3.9 42.5 9.4 8.5 3.6 7.5 5.3 6.2 3.2 3.7 7.6 8.6 4.0 Avg. % flee. 80.2 62.5 77.2 11.3 47.3 80.4 13.7 1.5 5.7 68.9" 60.4 0.2 3.1 5.4 1.4 38.7 65.8 83.3 71.8 70.1 71.6 97.9 %RSD 12.5 6.8 12.3 8.0 7.7 10.0 18.4 12.9 8.1 6.1 5.4 12.0 16.3 13.3 4.3 2.3 6.5 5.4 8.3 8.4 5.1 6.7 5.0 Avg. % flee. 69.4 52. 1 16.0 9.9 40.4 66.9 8.4b 1.2 5.2 56.7 52.7 0.1 3.6 5.5 1.3 33.5 58.2 77.7 65.5 59.7 58.7 80.5 %flSD 5.7 9.7 70.7 72.7 7.6 4.3 0.4 5.8 6.6 70.4 5.6 43.3 77.7 2.9 37.4 78.8 2.5 7.5 4.4 4.0 6.6 2.8 6.0 Avg. % flee. 77.7 57.2 75.3 77.0 43.5 65.5 7.8 7.2 4.7 46.8 49.8 0.7 2.8" 4.8 7.5 30.3 59.3 70.7 62.5 60.5 56.4 77.7 % flSD 6.0 9.5 72.3 27.5 72.6 4.3 3.8 73.4 70.5 73.7 3.3 0.0 72.7 2.4 76.7 76.8 4.7 2.9 7.5 6.9 3.2 6.8 Method 1311 Avg. % flee. 56.6 42.5 7.9 8.2 40.8 45.2 3.8 0.3 7.7 20.8 27.7 0.0 5.0 4.0 6.3" 37.4 % flSD 74.7 75.5 77.7 5.8 74.4 72.8 27.5 22.4 70.9 70.9 70.8 22.3 78.2 0.0 7.5 'Triplicate analyses. bDuplicate analyses: one value was rejected as an outlier at the 90% confidence level using the Dixon Q test. large RSD, in part, is the result of measurements made near the quanti- tation limit of Method 8270. In general, semi-volatile analyte recov- eries were lower and the RSDs were higher for the western soil than for the eastern soil. This appears to be related to the extraction and measurements steps since the RSDs of the organic surrogates added to the soil leachate were also higher for the western soil. The eastern (high organic content clay) and western (low organic content sand) soils are considerably different and thus the matrix sensitivity suggested is reasonable and is consistent with previous TCLP work. In Table 2, the surrogate RSOs vary from 7 to 63 percent and the semi-volatile ana- lyte RSDs range from 6 to 55 percent for the western soil. Thus, the analytical vari- ability of Methods 3520/8270 is compa- rable to the total variability of the leaching procedure. The precision data reported for the western soil is in general agree- ment with that of a prior precision valuation of the TCLP (Method 1311) for extractable components. In the TCLP precision study, RSDs for replicate leach- ings were usually less than 30 percent. The precision results for the eastern soil are much better than those of the previous TCLP study. The recovery of lead from both the eastern and the western soils was very low and the precision was poor. The large RSDs are, in part, the result of the fact that analytical measurements were made near the quantitation limit of Method 6020 (ICP/MS). Cadmium had reasonable re- covery (30 to 56 percent) from the eastern soil and its replicate leaching had recovery RSDs less than 20 percent. Cadmium recovery was only 4 to 9 per- cent from the western soil. The precision (variability) of the triplicate teachings of cadmium was much greater for the western soil (63 percent RSD versus 2.3 percent RSD at pH 4.0). Thus, cadmium showed a greater sensitivity to soil type in these experiments than did lead. Both cadmium and lead recoveries are sensitive to the pH of the leaching fluid. Lead recovery is significantly greater when the pH 3.0 leaching fluid is used. although all of the lead recovery values are low (1.3 to 3.2 percent). Cadmium recovery is also greater when low pH leaching fluid is used, but it varies over a greater range than lead (30 to 56 percent). The method precision for both metals is worst when the recoveries are lowest. Interestingly, lead recovery was low and only moderately higher when Method 1311 was used to leach the fortified eastern soil; this is in contrast to the results obtained in a recent inter- laboratory study that showed dramatically higher lead recoveries with Method 1311. The reason for this difference is not known, but points out the need for further study on the sensitivity of leaching methods to soil type as well as to the metal species presented in the sample. In contrast to results for the metals, the pH of the leaching fluid has little effect on either the recovery or the precision (vari- ability) of replicate teachings for semi- volatile organic compounds. Periodic monitoring of the pH of the fluid during the leaching showed that the pH of the fluid changed negligibly after the first ------- Method 1312 Precision Results from Western Soil Amount Spiked (VQ) pH = 4.2 High Spiking Level pH *5.0 Low Spiking Level6 Avg. % flee." %flSD Avg. flec.« FORTIFIED ANALYTES bis(2-chloroe1hyl)-ether 1040 45.1 13.7 59.2 14.2 2-Chlorophenol 1620 58.9 28.6 32.4 54.9 1,4-Dichlorobenzene 2000 12.8 11.8 13.6 34.6 1,2-Dichlorobenzene 8920 15.0 6.0 17.0 28.4 2-Methytphenol 3940 40.5 12.2 28.6 32.6 Nitrobenzene 1010 36.1 14.6 45.2 21.3 2,4-Dimethylphenol 1460 3.6 23.3 1.2 87.6 Hexachlorobutadiene 6300 2.5 17.0 4.5 22.8 Acenaphthene 3640 22.2 20.6 8.4* 7.7 2,4-Dinitrophenol 1300 20.5 - 1.8" 15.7 2,4-Dinitrotoluene 1900 61.6 30.1 30.8 54.4 Hexachlorobenzene 1840 0.2 41.8 0.8 173.2 y-BHC 7440 2J.2 23.8<> 76.6 55.2 0-BHC 640 t3.2 33.7 T0.2 5f.7 METALS Lead Cadmium SURROGATES (In Leachate) 2-Fluorophenol ds-Phenol d5-Nitrobenzene 2-Fluorobiphenyl 2.4.5-Tribromophenol d,4-p-Terphenyl 5000 1000 200 200 100 100 200 100 0.3 4.4 65.1 93.5 41.4 44.0 68.0 100.1 27.0 63.0 7.5 10.4 52.4 42.1 10.3 8.1 0.2 9.1 34.4 51.7 46.4 36.8 50.7 87.7 51.7 71.3 60.6 62.7 10.6 15.6 57.1 13.6 •Triplicate analyses. ^Duplicate analyses; one value was rejected as an outlier at the 90% confidence level using the Dixon 0 test. cThe low spiking level was 0.20 times the high spiking level. hour of the 18-hour leaching period. The final leachate pH of all the eastern soil samples was nearly the same (pH 4.57 to pH 4.64) for the pH 4, 5 and 6 fluids and was thus unaffected by the initial pH of the leaching fluid. The final leachate pH was low (pH 4.14) when pH 3.2 fluid was used. This might be partly responsible for the 'higher recovery observed for the metals with low pH leaching fluid. Volatile Compounds Precision The precision of Method 1312 was evaluated by measuring the repeatability of recovery of 27 volatile organic com- pounds using four different types of soil. A summary of the precision data for these compounds is presented in Table 3. Soil 1 was collected at a Superfund site west of the Mississippi River and Soil 2 came from a Superfund site in the eastern United States. Excluding isobu- tanol, a polar water soluble compound with known purging difficulties, the average recoveries for the target analytes ranged from 10 percent to 85 percent for Soil 1 and ranged from 7 percent to 89 percent for Soil 2. Twenty-two of the 26 (85 percent) analytes for Soil 1 and 17 of 26 (65 percent) analytes for Soil 2 had RSDs less than 20 percent. Only four analytes had RSDs greater than 50 per- cent; these analytes present significant analytical difficulties during the purge- and-trap GC/MS analysis (Method 8260). Vinyl chloride and the Freons are extremely volatile and are not trapped efficiently; acrylonitrito Is water soluble and does not purge well. In general, re- coveries were lower and RSDs were higher from the eastern soil than from the western soil. This matrix sensitivity ap- pears most pronounced for carbon tetra- chloride, ethylbenzene. tetrachloroethene, Freon 13 (trichlorofluoromethane) and Freon 113 (1,1.2-trichloro-trifluoroethane). In general, replicate teachings of the sludge-contaminated soils (Soils 3 and 4 in Table 3) exhibited greater variability (larger RSD) than those of the Superfund soils (Soils 1 and 2). For Soil 3, 20 of the 27 volatile analytes had RSD values be- tween 28 percent and 42 percent and 5 of the target analytes had RSDs great than 50 percent. Ethyl acetate displa) highly variable recoveries ranging from to 50 percent, its lack of precisic probably reflects the known difficult!* associated with purging polar con pounds from water. The sludge samp used to prepare Soil 3 contained aceton (120 ppm) which resulted in an apparei high recovery (116 percent) for aceton For Soil 4, 19 of 27 volatile analytes he RSDs between 24 and 41 percent ar only two compounds, both polar, eth acetate and acetone, had RSD value greater than 50 percent. Three volatile surrogate compounc were spiked into the leachates of Soils 4 just prior to analysis by Method 826 Surrogate recoveries were consistent high (greater than 90 percent with RSC less than 6 percent) and indicated th the purge-and-trap and GC/MS systerr used in the volatile analyses were pe forming satisfactorily. The Method 131 precision data reported for Soils 1 and is in general agreement with that ri ported in a previous precision evaluatic of the TCLP (Method 1311). The RSC reported in the TCLP study general ranged from 3 percent to 20 percent fi three different types of waste; volati compound recoveries were found to t matrix and compound dependent. Ruggedness The minor procedural variations use in the AOAC type ruggedness evaluatic of Method 1312 for semi-volatile cor pounds and metals are listed in Table Two levels of each experimental corn tion are assigned capital and lower ca letters and are varied in the mann shown in the matrix given in Table 4. Tl group differences calculated from tl recovery results are shown in Table 5. columns Va, Vc, and Vd, nearly all groi difference values have the same sig generally this indicates possible signi cance of the column variable. Usually ai difference which is more than twice tJ standard deviation of the analytical met ods is significant and should be furth studied. Most values in Column Va a negative, which implies that the extra table compounds had greater recove with pH 5.0 leaching fluid than with f 4.2 fluid. However, since all the d ferences in Column V, pf Table 6 a less than twice the analyte standa deviations calculated from the RSDs ai recoveries given for the western soil Table 2, the Va group differences a insignificant and careful control leaching fluid pH does not appear to I ------- ~«ftte 3. Method 1312 Precision Results on Volatile Compounds Soil No. 1 Soil No. 2 Soil No. 3 •Triplicate analyses. hSix replicate analyses. «Rve replicate analyses. Soil No. 4 Compound Name Acetone Acrylonitrile Benzene n-Butyl alcohol (1-Butanol) Carbon disulfide Carbon tetrachloride Chlorobenzene Chloroform 1.2-Dichtoroethane 1,1-Dichloroethene Ethyl acetate Ethylbenzene Ethyl ether Isobutanol (4-Methyl-i -propanol) Methylene chloride Methyl ethyl ketone (2-Butanone) Methyl isobutyl ketone 1, 1,1,2-Tetrachloroethane 1, 1,2,2-Tetrachloroethane TetracHoroethene Toluene 1,1,1-Trichloroethane 1, 1,2-Trichloroethane Trichloroethene Trichlorofluoromethane 1,1,2-Trichlorotrifluoroethane Vinyl chloride Avg. % Rec.» 44.0 52.5 47.8 55.5 27.4 40.6 64.4 67.3 73.4 37.4 76.4 56.2 48.0 0.0 47.5 56.7 87.7 69.0 85.3 45.7 59.2 47.2 76.2 54.5 20.7 78.7 70.7 %RSD 12.4 68.4 8.29 2.91 16.4 18.6 6.76 8.04 4.59 14.5 9.65 9.22 16.4 NA 30.3 5.94 10.3 6.73 7.04 12.7 8.06 16.0 5.72 77.7 24.5 26.7 20.3 Avg. % flec.« 43.8 50.5 34.8 49.2 72.9 22.3 47.5 54.8 68.7 22.9 75.4 23.2 55.7 0.0 42.2 67.9 68.9 47.7 58.9 75.2 49.3 33.8 67.3 39.4 72.6 6.95 7.77 %RSD 2.25 70.0 76.3 74.6 49.5 29.7 73.7 76.4 77.3 39.3 4.02 77.5 9.72 NA 42.9 3.94 2.99 11.3 4.15 17.4 70.5 22.8 8.43 79.5 60.7 58.0 72.8 Avg. % flee.* 776.0 49.3 49.8 65.5 36.5 36.2 44.2 67.8 58.3 32.0 23.0 37.5 37.3 67.8 52.0 73.7 58.3 50.8 64.0 26.2 45.7 i 40.7 67.7 38.8 28.5 27.5 25.0 %flSD 77.5 44.9 36.7 37.2 57.5 47.4 32.0 29.7 33.3 54.4 779.8 36.7 37.2 37.7 37.4 37.3 32.6 37.5 25.7 44.0 35.2 40.6 28.0 40.9 34.0 67.8 67.0 Avg. % flec.c 27.3 57.8 33.4 73.0 27.3 24.0 33.0 45.8 47.2 76.8 77.0 27.2 42.0 76.0 37.3 40.6 39.8 36.8 53.6 78.6 37.4 26.2 46.4 25.6 79.8 75.3 77.8 %RSD 71.4 4.6 41.1 13.9 31.5 34.0 24.9 38.6 37.8 26.4 115.5 28.6 77.6 72.2 76.6 39.0 40.3 23.8 75.8 24.2 37.2 38.8 25.4 34.7 33.9 24.8 25.4 Tattle 4. Variables Selected for Method 1312 Ruggedness Test on Semi-Volatile Compounds Parameter Type Rug 1 Rug 2 Rug 3 Rug 4 Rug 5 Rug 6 Rug 7 Rug 8 pH Extraction time Particle size Extractor Liquid/Solid Ratio Temperature filter A *4.2pH a = S.OpH A.a B,b C.c D.d Ee F,f G,g A B C D E F G B- 18 hr b = 16 hr E = ratio = 20 (2000 mL HjOHOOg) e * ratio = 16 (1600 mL H^/IOOg) A B c D e f g F * f = A A a b b B C c C d d d E e e f F F g G g C * not reduced c * reduced (grinded) Ambient approx. 77'F (25'C) 60 - 65'F (16-18'C) a B c d e f G a b C D e f G a b c D f F g D * standard vessel d =bottle G * g " one filter two filters ------- Table 5. Ruggedness Test Results Fortified Analytes V, bis (2-Chloroethyl)ether 2-Chlorophenol 1,4-Dichtorobenzene 1,2-Dichlorobemene 2-Methytphenol Nitrobenzene 2,4-Dimethylphenol Hexachlorobutadiene Acenaphthene 2,4-Dlnitrophenol 2,4-Dinitrotoluene Hexachlorobenzene y-BHC 0-BHC Lead Cadmium -2.70 -10.92 -2.52 -4.40 -4.20 -1.72 1.00 -2.25 -3.70 -22.35 -4.90 0.00 -4.68 -2.15 0.20 -3.6 - Group Differences v* vc 1.75 -4.77 1.77 1.90 -0.05 2.87 0.25 -0.60 2.30 -6.15 0.40 0.05 -1.88 -6.35 -0.20 -0.80 -2.20 -8.08 -1.07 -2.70 -7.90 -3.03 -3.55 0.70 -2.05 -24.70 -2.75 0.10 -0.97 6.30 0.0 7.7 for Semi-Volatile Test Compounds Vd V. V, vg 6.40 0.23 4.88 4.80 2.40 4.22 0.95 1.00 7.05 5.90 1.65 -0.05 2.68 -5.60 -0.55 -1.85 -0.35 -5.88 -0.67 -0.30 2.05 1.03 0.15 0.50 -2.10 -21.95 2.10 -0.00 3.47 8.55 -0.10 -5.42 2.90 -3.93 -1.08 -1.50 0.40 1.23 1.30 2.25 5.50 -0.15 4.00 0.15 3.97 4.05 0.10 -2.08 -1.55 2.27 -2.77 -4.90 0.15 -0.82 0.30 -0.45 -2.25 5.00 -4.75 0.05 -0.97 1.20 -0.10 -3.30 a critical parameter for Method 1312. A similar comparison of analyte method precision with the mostly small dif- ferences given in Columns Vc and Vd leads to the same conclusion for the experimental parameters, particle size reduction and extractor vessel type. The other four variables, extraction time, liquid solid ratio, extraction temperature, and number of filters used, do not appear to affect the performance of Method 1312. Since none of the seven variables tested is a critical parameter, Method 1312 appears rugged for the leaching of semi-volatile compounds and metals. These results concur with a previous ruggedness evaluation that demonstrated the TCLP to be "fairly rug- ged" for the semi-volatile organic ana- lytes which were unaffected by variables B, D. and E in Table 4 in addition to the parameters: (a) headspace amount, (b) medium acidity, (c) acid washing of filter, and (d) filter type. It is interesting to note that the EPA had previously intended to investigate extraction temperature but was unable to do so due to a lack of the laboratory equipment necessary to vary the temperature. This study addressed this issue and determined that extraction temperature is not a critical method parameter for the leaching of semi-vola- tile organic compounds and metals. The variables chosen for evaluation in the ruggedness test of Method 1312 for volatile organics are listed in Tables 6 and 7. Variables C and G are not method parameters, per se; they were chosen to determine the effect of analyte concen- tration upon compound recovery and to determine the effect of buffering the leaching fluid. The group differences calculated from the recovery results are given in Table 8. Since the differences in columns Va, Vd, and Ve are generally small (absolute value less than 5) and of random sign, the variables leaching fluid pH, leaching fluid liquid/solid ratio, and extraction time, do not exhibit an ob- servable effect on analyte recoveries and, thus, are not critical parameters for Method 1312. Due to the large magni- tude of many values and the general uniformity of value sign, the variables associated with Columns Vb, Vc. and Vg may be significant. Particle size reduc- tion (Column Vf) generally resulted in increased recoveries for the more highly volatile compounds. Smaller soil particle size probably decreases compound vola- tility loss during the soil spiking step by facilitating adsorption of the fortified compounds by increasing the surface area of the soil particles. As grinding a soil sample would increase the loss of environmentally incorporated analytes, particle size reduction should not be considered a critical method parameter in the leaching of real soil samples. Addition of 0.1 M acetate buffer (Column Vg) adversely affected recoveries of vir- tually all the volatile compounds: studied. This effect is probably the result, in large part, of chromatographic analysis diffi- culties caused by the loading of acetic acid onto the capillary GC column and/or into the purge and trap system used in Method 8260. Fortifying the soil at lower concentration (i.e., 1 ppm versus 4 pprr yielded better overall recoveries (Columi Vc) for the more highly volatile con- pounds. This higher recovery may b related to limited analyte solubility in th purge vessel and/or could result fror less compound loss during the so spiking step. Interestingly, the onl experimental parameter of importanc appears to be the type of ZHE used. Th Millipore ZHE gave higher recovery tha the Associated Design ZHE for nearly a the organic compounds. This result i surprising given our experience wit leakage difficulties associated with use c the Millipore ZHE and is in contrast wit the results of a previous ruggednes evaluation of the TCLP for volatil' compounds. In that study, the onl critical parameter identified was the typi of ZHE; the Millipore ZHE producei lower analyte recoveries than thi Associated Design ZHE and had notabli leakage problems. The reason for thi: major discrepancy is not known but i may be related to operator experience with the different ZHE devices in thi different laboratories. Recommendations Method 1312 is suitable for thi characterization of soil samples. How ever, additional information on the per formance of the method as a model fo the mobility of toxicants in the environ ment is required. It is recommended tha studies be conducted to: measure thi mobility of different lead- and mercury containing compounds in the soil; comp are the mobility of toxicants in soi columns with Method 1312 mobility; am develop specific performance data fo SW-846 methods (i.e., Method 8150 ant Method 8081) when they are used t< analyze Method 1312 leachates. Conclusions Method 1312 is a reasonably ruggec and precise method that can be used to address the mobility of pollutants in soi samples. The performance of Methoc 1312 for leaching organic compound! was very similar to that of Method 131 (TCLP). Method 1312 was less efficien at leaching cadmium and lead than was Method 1311. ------- Table 9. Variables Selected for Method 1312 Ruggedness T Upper Lower Variable case (A) case (a) Extraction fluid pH ZHE apparatus Analyte soil cone.' Extraction time Liquid/solid ratio Particle size reduction Buffer addition % Recovery of analytes 4.1 MP 0.8 ppm 20hrs 22:1 with with 4.3 AD 4 ppm 16hrs 18:1 without without esting for Volatile Compounds Experiment Number 1 A B C D E F G s 2 A B c D e f 9 f 3 A b C d E f 9 u 4 A b C d e F G V 5 a B C d e f 9 w 6 a B c d E f G X 7 a b C D e f G y 8 a b c D E F g z "Fortified soil nominal concentration level. Refer to Table 7 for experiment dependent fortification levels. Abbreviations: MP: Millipore AD: Associated Design ppm: mg/kg Table 7. Soil Fortification Levels for Ruggedness Experiments Variable Combination C E c e C e c e Solid/Liquid Ratio 1:22 1:22 1.18 1:18 Sample grams 22 22 22 22 Leachate mL 484 484 396 396 Spike Level mg/kg 0.88 4.40 0.72 3.60 Max. Leachate Cone, yglkg 40 200 40 200 Table 8. Ruggedness Test Results for Method 1312 Group Differences for Volatile Test Compounds Conditions Compound Name Chloromethane Bromoethane Vinyl chloride Chloroethane Methylene chloride Acetone Carbon disulfide 1 ,1 -Dichloroethene 1 , 1 -Dichloroethane trans-t ,2-DicWoroetr»ene Chloroform 1 ,2-Dichloroethane 2-Butanone 1,1,1 -Trichloroethane Carbon Tetrachloride Bromodichloromethane 1,1,2,2-Tetrachloroethane 1 ,2-Dichloropropane cis- 1 , 3 -Dichloropropene Trichloroethene Dibromochloromethane 1, 1 ,2-Trichloroethane Benzene trans- 1,3-Dichloropropene Bromoform 4-Methyl-2-pentanone Tetrachloroethene Toluene Chlorobenzene Ethyl benzene Styrene v, -6 1 -9 -7 -2 7 -5 -6 0 -5 1 5 8 1 -1 7 3 8 8 1 1 3 3 8 -1 31 1 4 2 4 -1 vb 13 11 11 24 34 76 12 11 15 13 17 3 -3 5 4 3 3 7 3 3 2 1 6 4 2 38 2 4 1 3 -1 Vc 16 20 14 23 69 119 19 13 19 18 29 8 11 7 6 9 5 5 6 4 7 7 9 4 3 -27 2 9 7 6 6 va 0 2 0 -8 -7 40 5 4 6 7 4 0 9 -1 -1 3 -2 2 3 1 1 0 1 1 -2 -20 2 1 2 4 3 v* 0 6 -3 -1 8 30 3 1 6 4 6 1 6 0 1 3 4 -1 5 2 4 4 3 4 5 40 2 3 4 3 2 v, 12 8 11 18 26 87 10 13 a 8 6 3 -1 14 13 6 -1 5 5 6 2 5 6 6 -2 28 3 5 0 4 -6 Vg -18 -12 -11 -38 -45 37 -15 -14 -20 -15 9 -2 20 -6 -6 -1 -2 -10 1 -3 2 -1 -1 -3 -1 58 -1 -3 -1 -5 -4 ------- fabfo 0. (continued) Conditions Compound Name p-Xylene o-Xylene 1 ,4-Dichlorobenzene Trichlorofluoromethane Acrylonttrile n-Butanol Ethyl acetate Ethyl ether Isobutanol 1, 1,2-Trichlorotrifluoroethane (Freon-1 13) 1, 1. 1.2-Tetrachloroethane V. 1 0 -1 -7 10 0 11 7 0 -7 2 vb 2 2 -1 10 13 0 -5 5 0 10 4 vc 5 9 4 9 4 0 8 -2 0 9 6 vd 5 2 4 2 2 0 -12 -2 0 1 -1 ve 1 3 1 -1 17 0 16 5 0 -3 4 v, -1 -1 -6 17 1 0 -20 7 0 15 3 Vg -2 -2 2 -13 -8 0 29 0 0 -12 -4 NOTE: n-Butanol and isobutanol were not recovered in any of the ruggedness experiments. T. C Chiang. C . A. Valkenburg, and D. A. Miller are with Lockheed-ESC, Las Vegas, NV 89119. The EPA author, G. W. Sovocool, (also the EPA Project Officer) is with the Environmental Monitoring Systems Laboratory, Las Vegas, NV 89193- 3478 (see below). The complete report, entitled "Performance Testing of Method 1312-QA Support for RCRA Testing," (Order No. PB 89-224 901 /AS; Cost: $21,95, subject to change) will be available only from: National Technical Information Service 5285 Port Royal Road Springfield, VA 22161 Telephone: 703-487-4650 The EPA Project Officer can be contacted at Environmental Monitoring Systems Laboratory U.S. Environmental Protection Agency Las Vegas, NV 89193-3478 United States Environmental Protection Agency Center for Environmental Research Information Cincinnati OH 45268 Official Business Penalty for Private Use $300 EPA/600/S4-89/022 CHICAGO ------- |