The Chemistry of Sorption by Black Carbon: The Important Role of Surface Oxides.

The ability of powdered black carbon (BC) to adsorb solutes from aqueous solution has been known for over 3,500 years, as documented by ancient Egyptians, Hippocrates, and others. Carbons from wood and animals were commonly used in the 17^th century for liquid purification, and granular forms of bone char (containing only small percentages of carbon) were widely used in the 18^th century for the decolorization of sugar and other solutions. Activated carbons with highly developed pore structures and very large specific surface areas have been commercially produced since the beginning of the 20^th century and both the powdered and granular forms of these carbons have been studied extensively over the past 50 years, in the context of water and wastewater treatment.  More recently, the uptake of various aqueous solutes by naturally occurring forms of BC in soils and sediments have been explored in different environments, including surface soils, groundwater aquifers, water-saturated aerosols, and river, lake, estuarine and marine sediments. Finally, within the last eight years, the sorption properties of carbon-based nanoparticles, such as fullerenes and carbon nanotubes (CNTs), whose molecular structures are well-defined compared to more traditional forms of BC, have been studied.

For activated carbons and naturally occurring BC (i.e., char, charcoal, soot), the physical and chemical heterogeneity of the sorbents preclude rigorous evaluations of the role that surface chemistry plays in moderating sorption, even for simple, non-polar sorbates that are capable of only physical (Van der Waals) interactions with the sorbent.  Nonetheless, a substantial amount of information can be obtained through experimental sorption studies under a variety of conditions, especially when combined with careful physical and chemical characterization of the bulk sorbent and thoughtful analysis of the data through the application of phenomenological sorption isotherm models.  For more chemically homogeneous surfaces such as those of carbon nanotubes, the surface chemistry can be more fully characterized. This is a consequence not only of the particle's controlled structure, but also of the fact that as particles decrease in size, the proportion of atoms at the surface increases relative to the bulk.

Our own studies of the sorption of organic and inorganic solutes from water onto all of the above forms of BC have revealed strongly nonlinear adsorption isotherms that reflect a wide distribution of site binding energies.  Such results are of course expected for activated carbon and heterogeneous natural sorbents, but the degree of nonlinearity has also found to be very high with CNTs, even for nonpolar organic solutes on relatively pristine carbon surfaces.  For these cases, the complexities of condensation and multi-layer adsorption on physically diverse surfaces may still be an important cause of the heterogeneity in binding energies.

A particular point of emphasis in our research has been to specifically explore the role of surface oxides on the sorption uptake of both hydrophobic organic chemicals such as polycyclic aromatic hydrocarbons (PAHs), chlorinated benzenes (CBs) and divalent metal cations such as zinc and cadmium.  For the PAHs and CBs, we have performed such studies on chars and soot of similar specific surface area but varying surface oxygen content. We have conducted similar studies and also a full suite of metal adsorption studies for one of the natural chars, for a selected activated carbon, and for a full suite of surface modified CNTs. In all of these studies, a suite of BC materials were formed by careful oxidative treatments of original starting materials and characterized by high-resolution surface area measurements and X-ray photoelectron spectroscopy (XPS) for quantification of surface composition.  We have also independently determined the concentration of hydroxyl, carbonyl or carboxylic acid functional groups by means of a suite of chemical derivatization reactions that specifically tag these oxides; surface products contain fluorine atoms that are readily quantified by XPS.    Together, results from these sorption studies tell a consistent story about the role of surface oxidation in decreasing the uptake of hydrophobic organic chemicals from water, while increasing the capacity of the surfaces to adsorb Zn²⁺ and Cd²⁺.  In the latter case, we have been able to mechanistically explain the results by a dual Langmuir isotherm that considers that adsorption is primarily to carboxyl and graphene surface sites, with the relative quantities of these sites determined from the independent XPS data.  The results provide useful information about the importance of carboxyl functional groups in facilitating metal sorption while also providing new understanding of the interaction of metal cations with graphene surfaces.  In our continuing work, these results are being combined with independent studies of the black carbon’s mobility in the environment to provide new insight into the potential role of black carbon not only as a sequestering agent but also as a possible facilitator of contaminant transport and movement.  In all of these regard, oxidation or other processes that modify surface functionality will play a critically important role.