Mechanism of Slow Desorption of Organic Compounds from Sediments: A Study Using Model Sorbents GERARD CORNELISSEN,* ,† PAUL C. M. VAN NOORT, † AND HARRIE A. J. GOVERS ‡ Institute for Inland Water Management and Wastewater Treatment, P.O. Box 17, 8200 AA Lelystad, The Netherlands, and Department of Environmental and Toxicological Chem istry, ARISE, University of Am sterdam , Nieuwe Achtergracht 166, 1018 WV Amsterdam, The Netherlands The desorption kinetics of PCBs and chlorobenzenes have been studied at 5, 20, and 60 °C for model sorbents in which either micropore diffusion (zeolite, montmorillonite, and XAD-8) or organic matrix diffusion/entrapment (rubbery polyacetal and glassy polystyrene) could occur. Also, a sediment was studied whose organic matter (OM) had been completely removed. All sorbents exhibited slow desorption (rate constants (1-5 × 10 -3 h -1 ). The sediment without OM showed significantly smaller slowly desorbing fractions (factor 3-8) than the original sediment (about 6% OM). Sorbent -water distribution ratios of the microporous sorbents and the sediment without OM were 10- 100 times lower than the ones of the original sediment. So, although the presence of both mineral micropores and/ or OM can result in slow desorption behavior of organic compounds from soils and sediments, OM is more important for slow desorption than mineral micropores in sediments with more than about 0.1-0.5% OM . The sorption and desorption parameters measured for the sorbents were compared to the ones measured for sediment. This analysis showed that the observations for XAD-8 (in which slow desorption is assumed to be caused by slow diffusion along hydrophobic pore walls) were most similar to the ones for the sediment, indicating that diffusion through pores in the organic matter or pores coated with organic material play roles in slow desorption. Introduction The desorption oforganicchemicalsfrom soilsand sediments has often been observed to occur in two stages: a rapid stage followed by a stage of much slower release (1-9). Currently there is much interest in the “resistant” fractions because theycan greatlyinfluence the fate and behavior ofsediment- sorbed organic compounds (1), especially with respect to their bioaccumulation (10, 11),biodegradation (11-13),and transport (14-16). It is generally thought that slow desorp- tion is controlled by slow diffusion in sediment aggregates (1, 4, 5, 8, 9, 14-16). Roughly,two kindsofretarded diffusion through the sediment matrixcan be envisioned: (i)diffusion through the organic matter (OM) (1, 4) with possibly a distinction between amorphous and microcrystalline mate- rial, the latter causing relatively strong desorptional limita- tions (17-19), and (ii) diffusion through and along the (hydrophobic) walls of micropores (1, 5, 14-16). The size of such micropores is hypothesized to be of the same order ofmagnitude (nanometer size) as the diffusing compounds. In addition, the entrapment of organic chemicals in “voids” in the OM (similar to those in glassy polymers) has been proposed to account for (de)sorptional retardations (1, 20- 22). These voidshave recentlybeen hypothesized to undergo a physicalrearrangement upon sorption ofa sorbate molecule through which this sorbate becomes more strongly bound (23). On the basis of sediment desorption experiments only, it is difficult to distinguish between processes in pores and in OM, respectively. Therefore, in the present study we describe the slow desorption from (i)microporous materials that show retarded diffusion in mineral pores (the clay mineral montmorillonite and a zeolite) or in pores with hydrophobic walls (the porous resin XAD-8), (ii) organic polymers without pores to mimic retarded diffusion and/or void entrapment in a macromolecular matrix (rubbery polyacetaland glassypolystyrene),and (iii)a sediment whose OM has been removed. Previously, it has been shown that diffusion through polymers (24, 25) and microporous ma- terials (26-28) can be highly retarded. In the present study, the desorption kinetics of three chlorobenzenes and three PCBs from all these materials are measured at three different temperatures (sediment without OM, one temperature) by using Tenax TA beads as infinite sinkfordesorbedcompounds(7). Acomparison willbemade between the model sorbents and a lab-contaminated sedi- ment with respect to desorption rate constants, slowly desorbingfractions,activation enthalpiesofslowdesorption, and the variation of these parameters among the six test compounds used. Methods Chemicals. 1,2,3,4-Tetrachlorobenzene (TeCB), pentachloro- benzene (QCB), hexachlorobenzene (HCB), 2,4,4′-trichloro- biphenyl (PCB-28), 2,3,5,6-tetrachlorobiphenyl (PCB-65), 2,3′,4,4′,5-pentachlorobiphenyl (PCB-118), sodium bicar- bonate (NaHCO3), and sodium persulfate (Na 2S2O8) were obtained from variouscommercialsourcesand used assuch. Tenax TA (177-250 μm), a porous polymer based on 2,6- diphenyl- p -phenylene oxide,wasobtained from Chrompack. Before use, the Tenax TA beads were rinsed with hexane, acetone, and water (each 3 times with 10 mL/g Tenax) and dried overnight at 75 °C. Model Sorbents. Montmorillonite was purchased from Aldrich (>99%; 63-125 μm). Itsorganiccarbon content (OC) was measured by chromatographic element analysis (Carlo Elba NA 1500, Milan, Italy) after heating the sediments to 1100 °C and after removal of carbonates with 0.1 M phosphoric acid. It was determined to be <0.01%. Mont- morillonite corresponds to the formula K[Si7.5Al0.5]Al3.5- Mg0.5O20(OH)4 and consists of two tetrahedral sheets that sandwich an octahedral one. These sheets form platelets; between the platelets are interlayer spacings (pores); these spacings are 1-4 nm wide (29-31). The zeolite Molsieve 13X (250-400 μm) was purchased from Chrompack and used without further treatment. Molsieve 13Xcorresponds to the formula m Na 2O‚Al2O3‚n SiO2‚ *Corresponding author fax: +31-320-249218; e-mail: g.cornelissen@riza.rws.minvenw.nl. † Institute for Inland Water Management and Wastewater Treat- ment. ‡ University of Amsterdam. Environ. Sci. Technol. 1998, 32, 3124-3131 3124 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 32, NO. 20, 1998 S0013-936X(97)00976-0 CCC: $15.00  1998 American Chemical Society Published on Web 08/29/1998