Transport of Single-Walled Carbon Nanotubes in Porous Media: Filtration Mechanisms and Reversibility DEB P. JAISI, †,‡ NAVID B. SALEH, † RUTH E. BLAKE, ‡ AND MENACHEM ELIMELECH* ,† Department of Chemical Engineering, Environmental Engineering Program, Yale University, New Haven, Connecticut 06520-8286, and Department of Geology and Geophysics, Yale University, New Haven, Connecticut 06520 8109 Received June 14, 2008. Revised manuscript received September 1, 2008. Accepted September 8, 2008. Deposition of nanomaterials onto surfaces is a key process governing their transport, fate, and reactivity in aquatic systems. We evaluated the transport and deposition behavior of carboxyl functionalized single-walled carbon nanotubes (SWNTs) in a well-defined porous medium composed of clean quartz sand over a range of solution chemistries. Our results show that increasing solution ionic strength or addition of calcium ions result in increased SWNT deposition (filtration). This observation is consistent with conventional colloid deposition theories, thereby suggesting that physicochemical filtration plays an important role in SWNT transport. However, the relatively insignificant change of SWNT filtration at low ionic strengths ( e3.0 mM KCl) and the incomplete breakthrough of SWNTs in deionized water (C/C 0 ) 0.90) indicate that physical straining also plays a role in the capture of SWNTs within the packed sand column. It is proposed that SWNT shape and structure, particularly the very large aspect ratio and its highly bundled (aggregated) state in aqueous solutions, contribute considerably to straining in flow through porous media. We conclude that both physicochemical filtration and straining play a role at low ( <3.0 mM) ionic strength, while physicochemical filtration is the dominant mechanism of SWNT filtration at higher ionic strengths. Our results further show that deposited SWNTs are mobilized (released) from the quartz sand upon introduction of low ionic strength solution following deposition experiments with monovalent salt (KCl). In contrast, SWNTs deposited in the presence of calcium ions were not released upon introduction of low ionic strength solution to the packed column, even when humic acid was present in solution during SWNT deposition. Introduction Carbon nanotubes (CNTs) are a class of nanomaterials composed entirely of carbon. They are the long and slender form of fullerenes, discovered a few years after the spherical Buckminsterfullerene in 1985 (1). CNTs have several unique size- and structure-dependent optical, electronic, magnetic, thermal, chemical, and mechanical properties (2, 3). These exceptional properties prompt the use of CNTs in numerous technological applications, such as fabrication of supercon- ductors, optical and storage devices, fuel cells, sensors, and catalysts (4). CNTs are expected to share a significant portion of the projected over $1 trillion nanotechnology market by 2010 (5). Such widespread uses and applications envisioned for CNTs along with the lack of regulations for disposal leave a gap for possible release into the environment (6). CNTs released into soils and sediments may eventually find their way into groundwater, reservoirs, and river systems and thereby enter into the food chains of living organisms (5). Furthermore, CNTs are one of the least biodegradable manmade materials (7), are insoluble in water in pristine form (7), and are toxic to human cells and bacteria (8, 9). These concerns demand a thorough understanding of the fate and mobility of CNTs, including the mechanisms of filtration (deposition) and remobilization in subsurface porous media. This information, in turn, will be useful for developing regulatory guidelines for disposal facilities such as repositories and landfills. The factors governing the transport and filtration (depo- sition) of colloidal particles such as silica, latex, and iron oxide (10-14) as well as bacteria and other microbial pathogens (15, 16) in porous media have been studied extensively. However, despite the large number of publica- tions on the synthesis and characterization of carbon-based nanoparticles, such as fullerenes, there are very few studies on their deposition and transport behavior (17, 18). Moreover, there have been no systematic studies so far on the transport and filtration of CNTs in porous media. In this paper, we investigate the transport behavior of single- walled carbon nanotubes (SWNTs) in a well-characterized satu- rated porous medium. A laboratory-scale column packed with cleaned quartz sand is used to study the retention of SWNTs under a wide range of repulsive (unfavorable) electrostatic conditions. Solution chemistry is varied to include different environmentally relevant conditionssmonovalent and di- valent salts and natural organic mattersthereby providing insights into the mechanisms controlling the removal and remobilization of SWNTs. Results of these experiments suggest that the mobility of SWNTs is effectively limited by their irregular shape and high aspect ratio. Materials and Methods Preparation of SWNTs. Carboxyl (sCOOH) functionalized SWNTs were purchased from Cheap Tubes Inc. (Brattleboro, VT). The SWNT suspension (1.0 g/L) was ultrasonicated (Misonix 3000, Misonix Inc., Farmingdale, NY) twice in deionized water for a cycle of 1 h each time. The dispersed SWNTs from the second cycle were split evenly into different vials and served as stock solution for the subsequent column experiments. The concentration of the SWNT stock solution (0.2 mL sample) was measured gravimetrically using an ultramicrobalance (Mettler Toledo MX5) with a repeatability of 2.5 μg. The sonicated SWNTs were used for the transport, electrophoretic mobility, and aggregation experiments. A fresh SWNT stock was used for each transport experiment. The stock was sonicated first for 10 min. The time gap between sonication and electrolyte addition and the SWNT injection into the column was kept constant (10 min each) in all experiments. Similarly, the time allowed to elapse between SWNT preparation, electrophoretic mobility, and particle aggregation measurements was kept close to that of the column experiments. Before column experiments, SWNT * Corresponding author phone: (203) 432-2789; e-mail: menachem.elimelech@yale.edu. † Department of Chemical Engineering. ‡ Department of Geology and Geophysics. Environ. Sci. Technol. 2008, 42, 8317–8323 10.1021/es801641v CCC: $40.75  2008 American Chemical Society VOL. 42, NO. 22, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 8317 Published on Web 10/22/2008