For total trace metals, a 0.5 g subsample of dried sediment was placed in a Teflon vessel and digested 

 with nitric acid, hydrochloric acid, and hydrogen peroxide for analysis of total mercury. A second 0.5 

 g portion of dried sediment was treated with nitric, perchloric, and hydrofluoric acids to prepare a 

 digestate suitable for total metals determination. This latter digestate was diluted with 5.7% nitric acid 

 prior to Zeeman furnace or flow injection AAS. A final portion of dried sediment was placed in a 

 Coulometrics total carbon apparatus and combusted in pure oxygen for the determination of total or- 

 ganic carbon. 



Various instrumental approaches were used for the determination of elements in the total recoverable 

 extractions, as well as the SEM fraction. Aluminum and iron in SEM extracts and aluminum, chro- 

 mium, copper, iron, and zinc in the total sediment digestates were determined by inductively coupled 

 plasma spectroscopy (ICP). Zinc in SEM extracts was determined by flame atomic absorption. Ar- 

 senic and selenium in SEM extracts and total sediment digestates were determined by flow injection 

 hydride generation atomic spectroscopy. Mercury was determined on total recoverable digestates by 

 flow injection cold vapor AAS. All remaining analytes (cadmium, chromium, copper, lead, nickel, 

 silver, antimony, and tin in SEM extracts, and cadmium, lead, nickel, silver, antimony, and tin in total 

 sediment digestates) were determined by Zeeman furnace AAS. 



For the analyses of organic compounds, the sediments were dried, homogenized, and extracted accord- 

 ing to NBS Midwest Science Center procedures. Different sample aliquot sizes were extracted for 

 each class of compounds. In all cases the appropriate internal standards were spiked into the sample 

 before extraction. Sediment samples were mixed with anhydrous sodium sulfate and column-extracted 

 with methylene chloride (MeCl). 



For the organochlorine pesticides, sample extracts were injected onto an automated, high-performance 

 gel permeation chromatography (HPGPC) system that was eluted with 80/20 hexane/MeCl. The col- 

 lected portion then went through serial fractionation on Florisil and silica gel columns. One of the 

 resultant three fractions (the first fraction of the silica gel) was treated for sulfur with acid-activated 

 copper. The three fractions were then analyzed by GC/ECD on two different phase 30-m columns, 

 DB-1 (methyl silicone) and OV-17 (50% phenyl-50% methylsilicone). All GC analyses were cool on- 

 column injections. 



For the polynuclear aromatic hydrocarbons (PAHs), the sample extracts were taken through a potas- 

 sium silicate (KS) cleanup and the HPGPC system. Extracts were treated for sulfur with acid activated 

 copper. The extracts went through a second KS cleanup and then were fractionated on a silver nitrate 

 treated benzenesulfonic acid cartridge which separated chlorinated aromatics from the PAHs. The 

 PAH fractions were analyzed by GC/MS on a quadrapole system in full scan mode. The column was a 

 60-m DB-5 (5% phenyl-95% methylsilicone). Compounds were determined by comparison of peak 

 retention times to those of a standard and by checking the mass spectra. The concentrations of 12 low 

 molecular weight (2- and 3-ring) PAHs and 12 high molecular weight (4- and 5-ring) PAHs were 

 quantified. Recoveries were determined by deuterated internal standard spikes. 



Samples extracted for polychlorinated biphenyls (PCB congeners, mono-ortho, and non-ortho) and 

 polychlorinated dibenzodioxins and dibenzofurans (PCDDs and PCDFs) were taken through two stages 

 of reactive column cleanup followed by HPGPC. After GPC cleanup the extracts were fractionated on 

 an automated C-18/PX-21 carbon column system. Four fractions were collected from the carbon col- 

 umn corresponding to congener PCBs (Fl), mono-ortho PCBs (F2), non-ortho (F3) and PCDD/PCDFs 

 (F4). 



26 



