Sequential Extraction of River Sediments Impacted by a Recent Coal Fly Ash Slide

ABSTRACT

A four-step sequential extraction method and inductively coupled plasma optical emission spectroscopy (ICP-OES) were used to determine the trace element distribution in river sediments impacted by a recent coal fly ash slide. The extractable, carbonate-bound, iron/manganese oxide-bound, and organic-bound phases were extracted. A large number of trace elements were investigated, but As, Co, Cr, Cu, Ni, and Zn had the best relative standard deviations (RSDs), usually less than 50%. Better reproducibility (less than 16%) was obtained for a standard reference material (SRM) that indicated that sample heterogeneity caused the large RSDs observed for the sediments. Analysis by ICP-OES was challenging because of the high salt concentrations in the extract. Calibration standards were matrix-matched, and some extracts were diluted to decrease the salt concentration. The percentage of extractable elements was smaller at the impacted sites compared to the upstream site; however, the concentrations of extractable elements were larger at the impacted sites. The largest exchangeable concentrations of As, Cr, and Cu were observed at the impacted sites and were 4.6, 1.1, and 5.5 mg/kg, respectively. Contrarily, the largest exchangeable Co and Zn concentrations were observed upstream of the spill, which indicated that these elements were more tightly bound to fly ash compared to uncontaminated sediments. The majority of extractable As was bound to iron and manganese oxides and may be liberated under reducing conditions. Students in a senior-level environmental chemistry course performed most of this work. The entire course was devoted to this research, and students benefited from exploring one problem in great depth.

KEYWORDS:

• coal fly ash
• ICP-OES
• river sediment
• sequential extraction

ACKNOWLEDGMENTS

We thank Dr. Anthony Calamai, Dean of the College of Arts and Sciences, the A.R. Smith Department of Chemistry, and the Biology Department at Appalachian State University for financial support. We also thank Ms. Donna Lisenby, the Upper Watauga RiverKeeper, for travel support and argon to run the ICP-OES. Dr. Dave Neely graciously shared boats and helped to collect samples. The hard work of Yosuke Sakamachi and Daniel Jackson in performing microwave-assisted acid digestions is also greatly appreciated. The ICP-OES used in this work was acquired through an NSF grant (DUE-CCLI-#0633550), and the microwave digester was acquired through a North Carolina Biotechnology Center Education Enhancement grant.

Notes

^a All extractions were performed at ambient temperature unless stated otherwise.

^b Based on Tessier's method.

^c This phase was also referred to as the surface-bound oxide phase in references 26 and 27.

^d In the original Tessier method, the residual fraction was extracted with a 5:1 mixture of HF:HClO₄.

^e Based on the BCR method.

*This matrix was diluted by a factor of 2 for ICP analysis.

The average, standard deviation (s), and percent relative standard deviation (RSD) for three replicate digestions are shown.

Units are mg/kg.

^a Data are from TVA.^{[ 15 ]}

^b Our data from sediments collected at ERM 2.2 are shown for comparison.

“Sum” is the sum of the exchangeable, carbonate-, Fe/Mn oxide-, and organic-bound concentrations. “Unextracted” was calculated by subtracting “Sum” from the total element concentration determined by microwave-assisted acid digestion.

*n = 2, so no standard deviation could be calculated.

Percentages were calculated by dividing concentrations in mg/kg for each step by the total concentration determined by microwave-assisted acid digestion.

*The sum of the extractable fractions was greater than the total concentration obtained by microwave-assisted acid digestion.

The average and range are reported for the four steps of the sequential extraction.

^a The RSD of the organic-bound fraction was 129%, but not included in the average or range.

Units are mg/kg. nd = not detected.