United States Environmental Protection Agency Environmental Monitoring and Support Laboratory Cincinnati OH 45268 Research and Development EPA/600/S4-85/011 Mar. 1985 ŁEPA Project Summary Evaluation of Methods for the Determination of Total Organic Halide in Water and Waste T. F. Cole Various methods were evaluated for determining the total organic halides (TOX) in groundwater and in waste oil samples. From three inorganic halide species generation approaches and three inorganic halide determinative techniques evaluated for groundwater analyses, one combined approach, which used Schoeniger flask oxidation with colorimetric chloride determina- tion, was chosen for laboratory valida- tion and method detection limit (MDL) studies. Groundwater samples were also analyzed for TOX using EPA Meth- od 9022, "TOX by Neutron Activation Analysis." The Schoeniger flask/colorimetric chloride and neutron activation analysis methods gave TOX results for spiked groundwaters comparable to those ob- tained using Interim Method 450.1 at halide levels above O.2 mg/L. Oil sample analysis using the sodium biphenyl reagent and a colorimetric chloride method was found to be unsat- isfactory for the analysis of various oils spiked with PCBs due to interferences in the colorimetric determinative step which resulted in recoveries greater than 100 percent. This Project Summary was developed by EPA's Environmental Monitoring and Support Laboratory. Cincinnati. OH, to announce key findings of the research project that is fully document- ed in a separate report of the same title (see Project Report ordering informa- tion at back). Introduction This program is a continuation of EPA Contract No. 68-03-2984, "Surrogate Methods for Priority Pollutants in Waste- water." The objective of the work con- ducted under this research program was to evaluate additional direct and indirect methods for TOX analysis. Indirect methods evaluated included Parr bomb oxidation, Schoeniger flask oxidation, and sodium biphenyl dehydro- halogenation as species-generation ap- proaches for conversion of organic halide to inorganic halide. The determinative techniques evaluated for the inorganic halide produced included ion chromatog- raphy, chloride ion selective electrode, and a colorimetric halide procedure. In all cases, the sample preparation procedure used was the charcoal adsorption tech- nique from Interim Method 450.1. The primary thrust of this portion of the program was to select the best species generation approach and best determi- native technique, and to combine these approaches into a TOX method for use as a less costly alternative to the microcolor- imetric determinative technique of Inter- im Method 450.1. Neutron activation analysis using the sample preparation procedure from Inter- im Method 450.1 was the only direct approach evaluated. This method was chosen for its ability to distinguish be- tween the various halogen species at parts-per-million levels. In addition to these methods for the determination of TOX in aqueous samp- les, the sodium biphenyl reagent ap- proach was evaluated for the analysis of TOX in transformer oil samples. The determinative technique for these eval- uations was the colorimetric chloride method used for the aqueous sample analyses. ------- Analytical Methods Evaluation Evaluation of Oxidation Approach Three oxidation approaches were eval- uated for the conversion of organic chlo- rine to chloride ion: Schoeniger flask combustion, Parr bomb combustion, and sodium biphenyl dehydrohalogenation. The reference determinative technique for the first two comparisons was ion chromatography (1C). For the sodium biphenyl approach, ion chromatography was not suitable because of interferences caused by high nitrate concentrations, and the ASTM D-512 colorimetric chlor- ide procedure was used. Schoeniger Flask Procedure The Schoeniger flask procedure was evaluated for halide blank levels due to glassware, absorbing solution, sample wrapping papers, and glassware cleaning methods Recovery of halide spiked onto gran- ulated activated charcoal (GAC) and ad- sorbed from aqueous solutions onto GAC was also evaluated. In all spiking exper- iments a three component spiking solu- tion containing tetrachloroethylene, 2,4,6-trichlorophenol and bromoform was used. The Schoeniger flask procedure was found to be superior to both the Parr bomb and the sodium biphenyl reagent proce- dures (described below) in terms of halide recovery and halide blank levels, and was used in conjunction with the ASTM D- 512 colorimetric method for chloride ion for method validation. Parr Bomb Oxidation Procedure The Parr bomb oxidation procedure was evaluated for halide blank levels from apparatus and reagents, recovery of halide spiked onto GAC, and effect of absorption time after sample ignition. The Parr bomb procedure was unsuitable for halide analysis because of incomplete combustion of the GAC and associated poor halide recovery. Sodium Biphenyl Reagent Procedure The sodium biphenyl reagent procedure was evaluated for halide recovery for standards in methanol and for standards spiked onto GAC and in reagent water. Blank halide levels associated with the reagents and glassware were also invest- igated. The sodium biphenyl procedure was found to be unsuitable because of high halide blank levels associated with the reagent; also, the extraction procedure was found to be too difficult for routine use due to the presence of the GAC at the interface between the organic and aqueous phases in the separatory funnel. Evaluation of Determinative Technique Three determinative techniques were evaluated: ion chromatography, chloride ion specific electrode (ISE), andthe ASTM D-512 colorimetric chloride procedure. Ion Chromatography 1C was evaluated using a Dionex S2 column over two chloride concentration ranges: 0.1 to 0.35/ug/mL and 2.0 to 10.0 /ug/mL In addition, the linearity of re- sponse for chloride and bromide over the range of 2 to 10 A/g/mL and the separation of six common anions was evaluated with an Si column. The chloride response on both columns and the bromide response on the Si column were linear over the ranges tested. Due to peak broadening, the detection limit for bromide, estimated at 0.5 fjQ/mL, was much higher than that for chloride (approximately 0.05 //g/L). In the evaluation of anion separation, bro- mide and nitrate ions were only partially resolved. The high nitrate level associated with the GAC was found to obscure small amounts of bromide and to render 1C ineffective for halide speciation in GAC digests from the Schoeniger flask oxida- tion procedure. Ion Specific Electrode The ISE determinative technique was initially evaluated for linearity of response over a range of 1 to 100 /yg/mL, and the detection limit for both chloride and bromide was found to be approximately 1 /ug/mL. ISE evaluations were stopped when it was determined that extremely high recoveries (>240 percent), presum- ably due to some interference in the GAC digest, were observed in spiked GAC samples prepared using the Schoeniger flask oxidation procedure. ASTM D-512 Colorimetric Chloride Procedure The ASTM colorimetric procedure was evaluated for reagent blank levels, chlo- ride and bromide response, and the effect of nitrate interference. The response for both chloride and bromide was linear from 1 to 4 /ug/mL. The estimated detec- tion limit for both halideswasfoundtobe approximately 0.5 /vg/mL, and nitrate appeared to have no effect on response Blank and spiked charcoal samples were prepared by the Schoeniger flask oxidation technique and analyzed by the colorimetric procedure. Comparison of the results of these samples with similar results obtained with ISE and 1C showed that ISE was least effective for halide determination, while 1C and the color- imetric method were essentially similar. The colorimetric procedure was chosen for the determinative technique for vali- dation since the 1C method was not capable of halide speciation with the GAC matrix and also due to the less sophfs- ticated and less costly nature of the colorimetric method. Method Validation The MDL of the Schoeniger flask/color- imetric halide procedure was determined by the analyses of spiked and unspiked replicates of reagent water according to the standard EPA protocol. A detection limit of 0.1 5 mg/L was calculated with an average recovery of 57 percent. For method validation, three identical sets of samples were prepared and ana- lyzed by the Schoeniger flask/colorimet- ric procedure, EPA Method 9022 (neutron activation analysis), and Interim Method 450.1. A summary of recovery results obtained in these analyses is provided in Table 1. Agreement between the methods was good for samples which contain halide concentrations above approximate- ly 0.2 mg/L. Below this value, the halide blank of the Schoeniger flask/colorimet- ric halide procedure prevents the deter- mination of TOX in the samples Evaluation of EPA Method 9022 A series of spiked charcoal samples were prepared and analyzed to evaluate the neutron activation analysis procedure (EPA Method 9022) as a direct method for TOX determination. Chloride recoveries obtained ranged from 91 percent for low- level spiked samples to 87 percent for high-level spiked samples. Bromide recov- eries ranged from 90 percent for low- level spiked samples to 88 percent for high-level spiked samples. Evaluation of Waste Oil Methodology Ten PCB-spiked oil samples obtaineo from the Quality Assurance Branch of th€ Environmental Monitoring and Support ------- Table 1 Summary of Recovery Results lor Validation Samples Obtained Using the Schoeniger Flask TOX Method, Interim Method 450 1, and Neutron Activation Analysis Method 9022 Average Halide Amount Found for Given Method, mg/L Sample Identification Reagent water Reagent water Groundwater #/ Groundwater #1 Groundwater # / Groundwater #2 Groundwater #2 Groundwater #2 Spike Level, mg/L Unspiked 0.51 Unspiked 0.51 1.0 Unspiked 5 1 10.2 Schoeniger Flask TOX 0.19 048 0.23 059 13 1.3 4.2 7.9 Interim Method 450.1 0.020 0.43 0.044 0.56 1.3 1 7 4.0 82 Neutron Activation Analysis 0087 0.049 0.12 0.59 15 2.7 7.2 10.1 Average Percent Recovery for Given Method Schoeniger Flask TOX a 56 -- 70 105 58 65 Interim Method 450.1 81 - 94 120 - 45 64 Neutron Activation Analysis .. 78 - 92 135 -- 88 72 "Not applicable. Laboratory at Cincinnati, Ohio, were analyzed for organic chloride content using the sodium biphenyl reagent spec- ies generation method and the ASTM colorimetric chloride determinative. High recoveries ranging from 145 percent to 268 percent were obtained for these oil samples. To evaluate these high recov- eries, an analytical curve study was per- formed by spiking a PCB-f ree transformer oil with Arochlor 1254. Results showed a high positive bias for all samples, al- though this bias was reduced consider- ably with increased sample dilution. Conclusions and Recommendations Based on the results of spiked reagent water and groundwater samples ana- lyzed, the Schoeniger flask oxidation and colorimetric halide TOX procedure is recommended as an alter native to Interim Method 450.1 for use as a TOX screening method for aqueous matrices, with the caveat that this procedure is applicable only to levels of TOX at or above 0.2 mg/L in aqueous matrices. This method pro- vides a significantly less costly alternative for the analysis of samples which contain low parts-per-million levels of organic halogen and gives results comparable to those obtained with Interim Method 450.1. The results of the evaluation of EPA Method 9022, Total Organic Halides by Neutron Activation Analysis, indicate that this direct method also is an ap- propriate technique for TOX analysis and offers the additional advantage of pro- viding individual chlorine and bromine values for the sample at levels equal to TOX detection limits achievable with the rnicrocoulometric determinative tech- nique of Interim Method 450.1. The sodium biphenyl reagent and color- imetric halide determinative technique for analysis of total organic chloride from oil matrices is not suitable as a screening procedure in its present form. Further evaluation of the sodium biphenyl reagent with some alternative determinative tech- nique, such as microcoulometrictitration, is recommended. Perhaps an additional cleanup step for the extract prior to the colorimetric determinative step would establish the usefulness of this technique at those halide levels in the range of from 20 to350jug/g. ------- T F. Cole, A. M. Berry, andR. L. Wilson are withBattelle Columbus Laboratories, Columbus, OH 43201. Thomas A, Pressley is the EPA Protect Officer (see below). The complete report, entitled "Evaluation of Methods for the Determination of Total Organic Halide in Water and Waste." (Order No. PB 85-166 304/AS; Cost: $13.00, 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: Environmental Monitoring and Support Laboratory U.S. Environmental Protection Agency Cincinnati, OH 45268 United States Environmental Protection Agency Center for Environmental Research Information Cincinnati OH 45268 Official Business Penalty for Private Use $300 0000329 PS U S ENVIR PROTECTIOfc SGENCY REGION 5 LIBRARY 118 S OIARBCRN STREET CHICAGO !»- ------- |