Physical control on CCl 4 and CHCl 3 desorption from artificially contaminated and aged sediments with supercritical carbon dioxide Nikolla P. Qafoku * , Lirong Zhong, Christopher J. Thompson, Chongxuan Liu, Bruce W. Arey, Alex Mitroshkov, Robert G. Riley Pacific Northwest National Laboratory, P.O. Box 999, MSIN: K8-96, Richland, WA 99352, USA article info Article history: Received 16 August 2008 Received in revised form 1 October 2008 Accepted 3 October 2008 Available online 29 November 2008 Keywords: Carbon tetrachloride Chloroform Adsorption/desorption Supercritical carbon dioxide Sediment Contaminants abstract The long-term interactions of carbon tetrachloride (CCl 4 ) and chloroform (CHCl 3 ) with sediments that are low in organic matter (OM) are not well studied. In this study, CCl 4 and CHCl 3 were mixed with super- critical carbon dioxide (CO 2 ) and loaded onto columns packed with two sediments with low OM and dif- ferent textures, to establish contamination and achieve expedited artificial aging. The columns were subsequently leached with a simulated groundwater under hydraulically saturated conditions. Scanning electron microscopy was used to inspect the morphology of sediment single particles, determine the degree of particle association in aggregates and qualitatively estimate porosity and the possible diffu- sional pathways that might affect the overall contaminant desorption rates. Results demonstrated that most of contaminant inventories were rapidly released in the first pore volume of effluent, although a small portion of contaminants’ total mass exhibited time-dependent desorption. The calculated K d values of CCl 4 or CHCl 3 partition were negligibly small. Both contaminants had similar transport behavior which was simulated well with a distributed (multiple)-rate (DR) statistical model. The model accounted for the apparent contaminant mass transfer through diffusional pathways of different lengths, towards the advective pores. The distribution of contaminant mass between equilibrium and kinetic fractions, the dis- tribution of the individual rate constants, and the average rate constants calculated with the parameters of the c-distribution function (b and g) of the DR model, were sediment (texture) dependent. This indi- cated that contaminant desorption during the late stage of leaching was driven by concentration gradi- ents (i.e., diffusion) within sediment matrix porosity. Ó 2008 Elsevier Ltd. All rights reserved. 1. Introduction Chlorinated hydrocarbon compounds are major constituents contaminating the vadose zone and groundwater at many sites throughout the United States of America. One example is the Department of Energy’s Hanford Site, where thousands of liters of CCl 4 and associated organic co-contaminants were discharged over a period of 20 years (from the 1950s through the 1970s), leading to extensive vadose zone and groundwater contamination. Information on sorption behavior of CCl 4 and CHCl 3 is crucial to make realistic predictions of the fate and mobility of these contam- inants in natural subsurface systems. Previous research have demonstrated that distribution coeffi- cient (K d ) values of low molecular weight organic contaminants, such as CCl 4 and CHCl 3 , were strongly and positively correlated with the organic matter content in soils (Duffy et al., 1997). Although in water-saturated systems, adsorption of nonpolar com- pounds to soil minerals can be blocked due to the strong dipole interactions between water and mineral surfaces (Chiou et al., 1983), sorption increases in the presence of soil organic matter (OM) and water, due to stronger OM: contaminant interactions compared to OM: water interactions (Chiou, 1989; Duffy et al., 1997). On the other hand, previous research has also demonstrated that the K d : soil organic matter relationship might not be success- fully used to predict the K d and the sorption extent in sediments poor in organic matter (Zhao et al., 1999). Because estimating sorp- tion capacity based on OM content may lead to greater errors than performing isotherms sorption experiments (Zhao et al., 2005), experiments are needed to investigate the behavior of both CCl 4 and CHCl 3 in sediments with low organic matter content. Recent studies suggested that a migration resistant fraction of contaminant total mass may be present in aged sediments, which slowly releases the contaminant into the aqueous phase (Riley et al., 2005). The rearrangement of sorbed species towards more energetically favorable sites (surface diffusion) may occur during aging. In addition, contaminant molecules may slowly move into sites within the soil matrix that are not readily accessed by even the smallest microorganisms (Alexander, 2000). Under these conditions, degradation is almost impossible and contamination 0045-6535/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2008.10.033 * Corresponding author. Tel.: +1 509 375 4364; fax: +1 509 375 6954. E-mail address: nik.qafoku@pnl.gov (N.P. Qafoku). Chemosphere 74 (2009) 494–500 Contents lists available at ScienceDirect Chemosphere journal homepage: www.elsevier.com/locate/chemosphere