Anaerobic sediment potential acidification and metal release risk assessment by chemical characterization and batch resuspension experiments

Background, Aim and Scope. Sediments act as a sink for toxic substances (heavy metals, organic pollutants) and, consequently, dredged materials often contain pollutants which are above safe limits. In polluted anaerobic sediments, the presence of sulphides and redox potential changes creates a favorable condition for sulphide oxidation to sulphate, resulting in potential toxic metal release. The oxidation reaction is catalyzed by several microorganisms. Some clean up measures, such as dredging, can initiate the process. The aim of the present work is to assess the acidification and metal release risk in the event of sediment dredging and also to compare two different acid base account techniques with the resuspension results. The oxidation mechanism by means of inoculation with an Acidithiobacillus ferrooxidans strain was also evaluated. Methodology. The sediments were chemically characterized (pH; organic oxidizable carbon; acid volatile sulphides; total sulphur; moisture; Cr, Cu and Zn aqua regia contents). A metal sequential extraction procedure (Community Bureau of Reference, BCR technique) was applied to calculate the Acid Producing Potential (APP) and Acid Consuming Capacity (ACC) of the sediment samples through Fe, Ca2+, and SO42- measurements. The acid base account was also performed by the Sobek methodology (Acid producing potential -AP- calculated with total sulphur and neutralization potential -NP- by titration of the remaining acid after a reaction period with the sample). Fresh sediments were placed in agitated shake flasks and samples were taken at different times to evaluate pH, SO42- and Cr, Cu, Zn and Fe2+, concentration. Some of the systems were inoculated with an Acidithiobacillus ferrooxidans strain to assess the biological catalysis on sulphide oxidation. Results. Sediment chemical characterization showed high organic matter content (5.4-10.6%), total sulphur (0.36-0.86%) and equivalent CaCO3, percentages (4.5-8%). pH was neutral-alkaline for all of the samples. AVS content was high except for sample 5. The acid base account obtained with the two methods gave different results for the acid generating risk of the samples. A decrease of 0.4 to 3.1 pH units was measured in the agitated shake flasks. In all of the systems, sulphate concentration increased (2,100-2,200 mg L-1 to 2,500-3,000 mg L-1), and positively correlated with the initial total sulphur content of the samples in the inoculated flasks. Cu and Cr in solution were not detected in most of the sampling occasions (< 0.5 mg Cu L-1 and < 0.5 mg Cr L-1). Zn reached high concentrations (up to 11.8 mg L-1). For every system-except sediment 1 - the lowest pH registered was similar in comparison to inoculated and control systems. The inoculation effect was mostly evidenced in the systems by a higher sulphate release rate compared to the control systems. Discussion. The BCR method categorized all of the samples as potentially acid generating material. The Sobek method using NPR (NP/AP) criteria classified sample 3 as a possible acid generator and samples 1, 2 and 5 with a low acid generation potential. Despite this, all the samples acidified the media in the kinetic tests in at least one of the conditions employed in this work. It would seem that NPR and NNP (NP-AP) risk classification criteria should not be directly used with anaerobic sediments. Appropriate classification levels for sediments should be developed considering the different sulphide reactivity between rock and sediments. Sediment oxidation can cause acidification, which is partially explained by sulphide oxidation. In the samples studied, we found a positive correlation between sulphate increase in solution after oxidation and total sulphur content in the inoculated systems. Significant amounts of Zn could be released to solution while Cr and Cu remained insoluble despite the pH decrease observed. The low Cu and Cr mobility could be explained by the very low solubility of their hydroxides and high affinity for organic matter and iron oxides/hydroxides that might form during sediment oxidation. Dredged sediment management and disposal should be carefully planned. Conclusions. All of the sediment samples lowered the pH media in the laboratory batch resuspension experiments. However, both risk classification criteria (NNP, NPR) from Sobek acid base account were not able to predict the samples' behavior as accurately as the BCR derived base account. The inoculation effect was mostly associated with a higher sulphate release and not to a lower pH due to acid base equilibrium. Recommendations and Perspectives. Appropriate risk classification levels for sediments should be developed considering the different sulphide reactivity between rock and sediments.