Optimizing the removal of carbon phases in soils and sediments for sequential chemical extractions by coulometry† Michael K. Schultz,*a Steven R. Biegalski,b Kenneth G. W. Inna, Lee Yu,b William C. Burnett,c Joylene L. W. Thomasa and Gregory E. Smithc aNational Institute of Standards and Technology, Ionizing Radiation Division, Mail Stop 8462, 100 Bureau Drive, Gaithersburg, MD 20899-8462, USA bNational Institute of Standards and Technology, Analytical Chemistry Division, Building 222/Room A125, Gaithersburg, MD 20899, USA cDepartment of Oceanography, Florida State University, Tallahassee, FL 32306-3048, USA. E-mail: burnett@ocean.fsu.edu Received 19th January 1999, Accepted 3rd March 1999 We have developed a coulometric technique to optimize the removal of the carbonate and organic fractions for sequential chemical extractions of soils and sediments. The coulometric system facilitates optimizing these two fractions by direct real-time measurement of carbon dioxide (CO 2 ) evolved during the removal of these two fractions. Further analyses by ICP-MS and alpha-spectrometry aided in interpreting the results of coulometry experiments. The effects of time, temperature, ionic strength and pH were investigated. The sensitivity of the coulometric reaction vessel/detection system was sufficient even at very low total carbon content ( <0.1 mol kg-1 ). The efficiency of the system is estimated to be 96% with a standard deviation of 8%. Experiments were carried out using NIST Standard Reference Materials 4357 Ocean Sediment (OS ), 2704 Buffalo River Sediment (BRS ), and pure calcium carbonate. Carbonate minerals were dissolved selectively using an ammonium acetate–acetic acid buffer. Organic matter was then oxidized to CO 2 using hydrogen peroxide ( H 2 O 2 ) in nitric acid. The carbonate fraction was completely dissolved within 120 min under all conditions examined ( literature suggests up to 8 h). For the OS standard, the oxidation of organic matter self-perpetuates between 45 and 50 °C, a factor of two less than commonly suggested, while organic carbon in the BRS standard required 80 °C for the reaction to proceed to completion. For complete oxidation of organic matter, we find that at least three additions of H 2 O 2 are required ( popular methods suggest one or two). Perhaps the most confounding challenge to developing Introduction reliable sequential extraction methods is optimizing the selec- Sequential chemical extractions are used often to partition soil tivity of the extractions. The ideal sequential extraction proto- and sediment samples into operationally defined fractions. The col maximizes the selective destruction of each target phase in techniques are designed to obtain indirect evidence of the primary geochemical host phases of metals, radionuclides, and other parameters. The sequential extraction approach is applied not only to study stable and radioactive-metal contami- nants in the environment,1–5 but also to investigate natural cycling of metals6–9 and the geochemical partitioning of phos- phate.10 Variations in the experimental conditions for sequen- tial extraction techniques are numerous: choice of reagents, reagent concentrations, reaction temperatures and reaction times vary widely. However, certain fundamental procedures are employed consistently to apply such methods. In general, a soil or sediment is subjected to a series of chemical treat- ments, each designed to attach a unique geochemical phase of the sample. In each step, the sample and reagent are shaken or agitated in some way (at a specified temperature) for some period of time. Following this reaction period, the solid and aqueous phases are separated by centrifugation and/or fil- tration. The residual solid material is reserved for the next reaction step in the sequence and the fluid phase is analyzed for the analyte(s) of interest (Fig. 1). †Present address: General Engineering Laboratories, Radio- chemistry Division, 2040 Savage Road, Charleston, SC, USA. Fig. 1 Illustration of a single sequential extraction step. E-mail: mks@mail.gel.com. J. Environ. Monit. , 1999, 1, 183–190 183