Can Chlorofluorocarbon Sorption to Black Carbon (Char) Affect

May 14, 2010 - sufficiently large to retard transport and affect groundwater ages obtained with CFCs. Sorption isotherms of CFC-11, -12, and. -113 to ...
0 downloads 0 Views 286KB Size
Environ. Sci. Technol. 2010, 44, 4459–4464

Can Chlorofluorocarbon Sorption to Black Carbon (Char) Affect Groundwater Age Determinations? SUNGWOOK CHOUNG AND RICHELLE M. ALLEN-KING* Department of Geology, 411 Cooke Hall, University at Buffalo, The State University of New York, Buffalo, New York 14260

Received February 24, 2010. Revised manuscript received April 28, 2010. Accepted April 29, 2010.

Although adsorption is not generally considered important in low foc (fraction organic carbon) aquifers, we show that chlorofluorocarbon (CFC) adsorption to black carbon (BC) is sufficiently large to retard transport and affect groundwater ages obtained with CFCs. Sorption isotherms of CFC-11, -12, and -113 to synthetic wood char were nonlinear (Freundlich n ) 0.71-0.94) while humic acid isotherms were linear. Moreover, sorption to char was 10-1000 times greater than to humic acid for all three CFCs at the lowest observed concentrations, Cw/S ∼ 10-8-10-7. We used the observed isotherms for char and humic acid to represent sorption to BC and amorphous organic matter, respectively, in a dual mode model to estimate retardation factors for a low foc aquifer () 0.06% gC g-1). The estimated retardation factors for the char-containing aquifer (presumed BC fraction ) 9% of foc) were ∼6.8-10.6 at Cw/S )10-8 and>5timesthoseestimatedassumingamorphousorganic matter partitioning only. The results indicate that unless CFC adsorption to BC is evaluated in transport, the groundwater age determined may be biased toward older than true ages. The CFC data archived in BC-containing aquifers may contain information about its adsorbent properties that could be useful to predict retardation of other chlorinated organic contaminants.

Introduction Chlorofluorocarbons (CFCs) are anthropogenic organic compounds which have been used in a wide range of industries as refrigerants, aerosol propellants, and foam blowing agents since their development in the 1930s until the Montreal Protocol restricted their production in 1987. As first suggested by Thompson et al. (1), hydrogeologists have used CFC-11 (trichlorofluoromethane), CFC-12 (dichlorodifluoromethane), and CFC-113 (1,1,2-trichloro-1,2,2-trifluoroethane) as environmental tracers for dating groundwater recharge. With relatively distinct increasing atmospheric concentrationsfromapproximatelythe1940suntil∼1994-2001, CFCs are useful for dating young groundwater recharged approximately 10-60 years before present (2, 3). This time scale correlates with increased industrial and agricultural chemical usage so that the CFCs are particularly relevant to anthropogenic groundwater contamination problems. The measured groundwater CFC concentrations are converted to gas phase equivalent concentrations via well established equilibrium partitioning relations and accounting for recharge temperature and elevation. Comparison to the known * Corresponding author e-mail: [email protected]; phone: 716645-4287; fax: 716-645-3999. 10.1021/es100620g

 2010 American Chemical Society

Published on Web 05/14/2010

atmospheric concentration record yields the apparent age of groundwater (recharge date) (2-4). Additional corrections for excess air in recharge water and, in some circumstances, factors that affect unsaturated zone transport must also be taken into account (2, 3). Correct groundwater dates must account for retardation due to sorption. In general, conservative behavior (e.g., no sorption) or linear partitioning is assumed for CFC transport (5-7) because these chemicals are only moderately hydrophobic (Supporting Information, Table S1). In low foc (organic carbon fraction) aquifers, the assumption of essentially unretarded CFC transport is supported by low sorption distribution coefficients (Kd) estimated by an empirical approach, that is, Kd ) Koc,ref foc. The reference organic carbon-water partition coefficient (Koc,ref) can be estimated from an appropriate linear free energy relationship (LFER) using another chemical property such as the octanol-water partition coefficient (Kow) (8). Research shows that black carbons (BC) such as char, charcoal, and soot can exhibit nonlinear apolar and monopolar compound sorption with Koc,C as much as 3 orders of magnitude greater than Koc,ref for chlorinated compounds, such as TCE, PCE, and PCBs (9, 10). The Koc,C is defined as the carbon-normalized equilibrium Kd corresponding to a particular aqueous concentration, Cw (Koc,C ) Kd,C/foc ) [q/Cw]/ foc, where q is the sorbed concentration). Nonlinear sorption to BC has been attributed to mechanisms such as site-limited surface accumulation and pore filling (9-12). Differences between the reference and BC Koc values are greatest at very low dissolved concentrations (9-11). Formed through incomplete combustion and/or pyrolysis in forest fires, BC resists weathering such that it can be found in sediments deposited since the late-Devonian (13, 14). For example, Karapanagioti et al. (15) reported charcoal as one of the BC forms identified petrographically in aquifer sediments. We posit that the widespread distribution of BC in sediments combined with the very low concentrations of CFCs that are of interest in the environment (pg L-1 or pg kg-1) (16) create a situation in which sorption is likely to retard CFC tracer transport in some aquifers with a concomitant increase in apparent groundwater ages, and is likely to be observable in some low foc aquifers. This study evaluated the potential for sorption to BC to bias groundwater recharge dates obtained using CFC tracers. The specific objectives were to characterize CFC-BC sorption behavior and to place these measurements in context through estimating the effect of BC on transport in a low foc aquifer. The first objective was addressed using batch experiments to contrast CFC sorption to char and humic acid over a moderately low concentration range (Cw,atm , Cw,exp , S). The Cw,atm and Cw,exp are aqueous concentrations in equilibrium with the present day atmosphere (Cw,atm/S ∼ 10-12-10-9) or used in our experiments (Cw,exp/S ∼ 10-8-10-3), respectively, and S is aqueous solubility. We used synthetic wood char as a representative BC and commercial humic acid to represent amorphous organic matter (AOM) for the purposes of characterizing sorption. Because this is an initial study to focus on the potential for impact, we conducted experiments at aqueous concentrations well above those of interest in groundwater to minimize the impact of background. The second objective was addressed through applying the measured isotherm results to demonstrate the potential effect of BC on CFC transport in a low foc aquifer. VOL. 44, NO. 12, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4459

Experimental Section Materials. Standard solutions of CFC-11, -12, and -113 in methanol were purchased from Supelco (Bellefonte, PA). Working solutions were prepared by dilution of the received CFC solutions in methanol. More information is in the Supporting Information, including CFC physical and chemical properties (Supporting Information, Table S1). A synthetic wood char was produced in the laboratory. Western Red Cedar xylem (heartwood) was cut to approximately 1 cm3 blocks and dried at 105 °C for 24 h. Small batches of xylem blocks were treated in a tube furnace with a constant nitrogen flow rate of 0.1 L min-1. Following the approach of (17), the xylem blocks were pyrolized at 600 °C as a final temperature for 90 min following a temperature ramp of 1000 °C h-1. The char was pulverized with a pestle and mortar, sieved to isolate the 75-150 µm size fraction (100 to 200-mesh), and stored in a desiccator at room temperature prior to characterization and sorption experiments. Commercial humic acid by Fluka was purchased from Sigma-Aldrich (St. Louis, MO). Sorbents Characterization. The foc of the synthetic wood char and humic acid were determined by combustion at 950 °C under pure oxygen with quantification of CO2 by coulometric detection (CM 5012, UIC Inc., IL). The average of triplicate foc measurements were 0.48 gC g-1 for humic acid and 0.88 gC g-1 for wood char. Wood char was sent to Huffman Laboratories (Golden, CO) for elemental analysis and to Quantachrome Instruments (Boynton Beach, FL) for density, surface area, and pore volume determinations (methods elaborated in Supporting Information). The char has high surface area of 483 m2 g-1 and micropore (