Environ. Sci. Technol. 2004, 38, 3574-3580
Effect of Sorbate Planarity on Environmental Black Carbon Sorption G E R A R D C O R N E L I S S E N , * ,† MARIE ELMQUIST,† INGA GROTH,‡ AND O ¨ RJAN GUSTAFSSON† Institute of Applied Environmental Research (ITM), Stockholm University, 10691 Stockholm, Sweden, and KTH-Royal Institute of Technology, Chemical Technology, 10044 Stockholm, Sweden
Soot and charcoal, collectively termed “black carbon” or BC, can exhibit extremely strong sorption of many hydrophobic organic compounds. In order to include BC sorption in fate models, it is important to know BC nanopore surface areas. In addition, it is useful to know for which compounds BC sorption can be expected to be important. By nitrogen adsorption measurements at ultralow pressures on sediment that was strongly enriched in BC by HF treatment and/or chemothermal oxidation at 375 °C, we found that environmental BC has nanoporosity in the 50% sorbed to BC in the nanogram per liter range) than for nonplanar ones (50% of total sorption in the nanogram per liter range (10, 14). The importance of BC for overall sorption levels off at higher concentrations because of the strong nonlinearity of the BC sorption process (6, 7, 10). For correct modeling of BC adsorption, it is important to know the specific surface area (SSA) of the BC. BC sorption is probably a surface-site adsorption phenomenon; some debate still exists whether this adsorption is best interpreted with BC surface area (7, 15, 16) or BC nanoporosity volume (17, 18). The two existing models that estimate BC sorption have the BC SSA as an important input parameter (15, 16). Previous studies on planar (non-ortho) and nonplanar (mono- or di-ortho substituted) PCBs revealed that sorption to pure BC (diesel soot) is ∼10 times stronger for planar PCBs than for nonplanar ones of the same hydrophobicity (8, 11). This indicates that BC may only need to be included in sorption models for planar compounds. However, this has not been confirmed for environmental BC. Further, sorption to pure BC of planar and nonplanar compounds has only been compared at one single concentration in these previous studies. Studying adsorption isotherms over a large concentration interval for planar and nonplanar compounds could give additional information on the mechanism of BC sorption as well as on the way BC sorption should be modeled. The objectives of the present study were 2-fold: (i) assess the SSA of environmental BC and (ii) study the differences in BC sorption strength between planar and nonplanar sorbates. To investigate the nanopore SSA of the environmental BC in sediment, we measured nitrogen adsorption at very low relative pressures for original sediment as well as sediment treated in various ways. We interpreted the experimental isotherms using models based on density functional theory (DFT). In this way, the SSA contribution of angstrom-sized pores can be assessed as a function of pore size. This gives insight in both the distribution of SSA over various nanopore sizes and the total SSA of the nanopores. To investigate the importance of BC sorption for planar and nonplanar compounds, we used four different chemicals with the same KOW of 104.6 ( 0.1 (19): planar anthracene (ANT), phenanthrene (PHE), and 4-chlorobiphenyl (4-PCB) as well as nonplanar 2,2′-dichlorobiphenyl (2,2′-PCB). For these compounds, we studied TOC sorption in unchanged sediment as well as BC sorption in sediment from which the nonpyrogenic OM had been removed by combustion at 375 °C. To investigate the effect of planarity on environmental BC isotherm linearity, we measured sorption isotherms over a wide concentration range (0.3-10,000 ng/L). With the results of the nanopore SSA and the insight in differences in environmental BC sorption behavior between planar and nonplanar compounds, better BC-inclusive modeling of hydrophobic compounds can be achieved.
Methods Sediment. Sediment was sampled from Ketelmeer (KET), a freshwater lake in The Netherlands (52°36′ N, 5°45′ E) and 10.1021/es049862g CCC: $27.50
2004 American Chemical Society Published on Web 05/20/2004
the first major sedimentation area of the river Rhine. Much of the pollution and BC in river Rhine stems from heavy industry and former coal mining along its shores. The 10-50 cm layer was sampled at a water depth of 2.5 m in the year 2000 using a box corer. The characteristics of the currently used batch of KET sediment (10, 14) include dry weight (47.2%), TOC (5.51%), BC (0.72%), TOC:TON atomic ratio (21), ratio between fossil fuel and biomass-derived BC (80: 20, as revealed by 14C dating of the BC), total native PAH and PCB contents (∼40 and ∼0.15 mg/kg dry weight (dw), respectively). Materials. [13C2]-Phenanthene (13C2-PHE), anthracened10 (ANT-d10), 2-PCB (PCB-2), 4-PCB (PCB-3), 2,2′-PCB (PCB4), and solvents were obtained from various commercial sources. The PAHs and PCBs were of >98% purity, and the solvents were Burdick and Jackson glass-distilled quality. Polyoxymethylene (POM) was obtained in 0.5 mm thick sheets from Vink Kunststoffen BV, The Netherlands. Combustion of Sediment. There exist various conceptual and operational definitions of BC (e.g., refs 2 and 20-22). Any BC isolation and quantification method is inherently operational and is, at best, still only able to describe a portion of the pyrogenic carbon continuum. Most methods are based on isolating BC from other OM by either wet chemical and/ or thermal oxidation pretreatments, followed by quantification of the residual reduced carbon as BC (reviewed in refs 2, 3, 20, 22, and 23). The 375 °C chemothermal oxidation (CTO-375) method employed in this work has been thoroughly tested with negative and positive standards (e.g., refs 2 and 20). It is able to provide BC results in sediments that are geochemically consistent because the spatial distributions of pyrogenic PAHs and PCDD/Fs are better described by BC than by bulk TOC (e.g., refs 2, 15, and 24) and the 14C isotopic composition of the CTO-375 isolated BC is identical to that of PAHs but quite distinct from that of TOC in both sediments and aerosols (e.g., ref 25). However, several limitations of the CTO-375 method are also indicated in the literature. For instance, particles with a high relative content of nitrogen may char to artifactually form BC during CTO-375 combustion (20, 26). On the other hand, in a large round-robin test of BC quantification methods, the CTO-375 method returned the lowest BC:TOC ratios (23). For the same reason as the CTO-375 method may be less prone than other methods to induce charring (i.e., the high thermal oxidation strength); it may remove also some less-condensed pyrogenic constituents formed at lower combustion temperatures (e.g., refs 8, 27, and 28). Following described precautions to avoid and test for charring and being aware that the CTO-375 method returns information on the more recalcitrant and condensed end of the pyrogenic particle spectrum, this method has proven to provide useful and system-consistent BC data in may studies. Hence, to “harvest” combusted sediment for this sorption study, dry sediment (10 mg in Ag capsules) was combusted at 375 ( 2 °C during 18 h under a constant air flow of 200 mL/min. A small size of a well-pulverized sample was employed to optimize access of oxygen to the sediments, which is important to prevent charring (2, 20). In this study, we scaled up the combustion method by combusting 36 (10 mg sized) samples in each run and doing several runs to obtain enough material for the SSA analyses and sorption experiments. A BC:BN of 40 (n ) 10) for the KET sediment suggests that the above-mentioned charring of any N-rich OM was negligible in this study. Nanopore SSA of BC. We tried to isolate the environmental BC by first removing the mineral part of the sediment with HF treatment, followed by combustion at 375 °C to remove the non-BC OM. The HF treatment was a modification of Ge´linas et al. (29) and Griffin and Goldberg (1) (Elmquist et al., manuscript in
preparation). Briefly, 2-3 g of KET sediment was exposed to an aqueous HCl solution (1 M; 10 mL) for 1 h to remove inorganic carbonates and thereafter washed three times with 40 mL of distilled water to discharge Ca2+ ions. Then 10 mL of a mixture of HF (10%) and HCl (1 M) was added to dissolve the silicates and allowed to react for >24 h before the supernatant was discharged and a new addition of HF/HCl was made. A total of four additions of HF/HCl were made, and the total HF-sediment reaction time was 7 d. The remaining material was washed, dried (60 °C), and carefully ground. We measured SSA and nanopore size distribution for original sediment, sediment after CTO-375 treatment (OM removed), sediment after HF treatment (∼90% of the minerals removed), and sediment after both HF and CTO-375 treatments (most minerals and OM removed). SSAs of the original and treated sediments ((0.5 g) were determined by N2 adsorption at -196 °C after degassing at 100 °C for >15 h using a Micromeritics ASAP 2010 system equipped with a high vacuum system. All samples were measured under exactly the same conditions. With regular SSA measurements, it is not possible to analyze nanoporosity in the range
