Supercritical fluid extraction of polychlorinated biphenyls from real world Saint Lawrence River sediments

Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)


Biomedical and Chemical Engineering


Lawrence L. Tavlarides


Supercritical fluid, Polychlorinated biphenyls, St. Lawrence River, Sediments, Remediation

Subject Categories

Chemical Engineering | Complex Fluids | Environmental Engineering


Systematic studies have been conducted based on real world PCB contaminated St. Lawrence River sediments (SLRSs) to investigate both the technical and economic feasibilities of supercritical fluid extraction (SFE) technology on the remediation of PCB contaminated soils/sediments.

Partition equilibrium data between SLRSs and supercritical CO 2 (SC-CO 2 ) with and without 5 mol% methanol have been obtained for both total PCBs and 16 major PCB congeners (or congener mixtures) presented in the sediments. The partition coefficients increase roughly in the order of molar weight increase of PCB congeners. All the isotherms were fitted with a linear model. The model analysis reveals that PCB adsorption onto sediments from supercritical fluids (SCFs) is an exothermic and physical process with the adsorption energy of ∼4 kcal/mol in pure CO 2 and ∼8 kcal/mol in CO 2 /MeOH system.

The laboratory scale desorption experiments reveal that PCB concentration in the sediment can be reduced from 2200 ppm to less than 10 ppm (99.5% extraction efficiency) in 25 minutes of extraction time and to less than 5 ppm (99.77% extraction efficiency) in 60 minutes when CO 2 /5 mol% MeOH is used.

The bench scale studies were conducted in a unit equipped with a 2-liter fixed-bed extractor, a PCB/CO 2 separator, and an online CO 2 recycle system. The results show that PCB concentration in the sediments can be reduced from ∼1840 ppm to less than 5 ppm (more than 99.73% removal efficiency) in 40 minutes and to ∼3 ppm in 60 minutes.

A linear driving force model has been developed and employed in a two-step simulation of both the laboratory and bench scale desorption data. The parameters obtained from the laboratory scale desorption data are also used to predict the bench scale data and give reasonably good results.

A full scale supercritical fluid extraction process followed by supercritical water oxidation of the extracts has been proposed. The base case of the proposed process can treat 36,000 m 3 contaminated soils/sediments every year. The economic analysis of this process reveals that the operation costs are in the range of $142-175/m 3 , which is competitive with other alternative remediation technologies.


Surface provides description only. Full text is available to ProQuest subscribers. Ask your Librarian for assistance.