Environ. Sci. Technol. 2009, 43, 4864–4870
Inferring Black Carbon Concentrations in Particulate Organic Matter by Observing Pyrene Fluorescence Losses D. XANAT FLORES-CERVANTES,† CHRISTOPHER M. REDDY,‡ AND P H I L I P M . G S C H W E N D * ,† R.M. Parsons Laboratory, MIT 48-413, Department of Civil and Environmental Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139, and Department of Marine Chemistry and Geochemistry, Woods Hole Oceanographic Institution, Woods Hole, Massachusetts, 02543
Received January 6, 2009. Revised manuscript received May 7, 2009. Accepted May 11, 2009.
Black carbon (BC), the soot and char formed during incomplete combustion of fossil and biomass fuels, is ubiquitous, participates in diverse environmental processes, and has adverse effects on human health. However, uncertainty persists regarding how accurately the present measurement methods quantify total BC or even defined subportions of the BC continuum. Hence, we sought to improve this situation by developing a new, low-sample manipulation methodology that does not require any oxidative or pyrolytic treatments but rather differentiates BC from other non-BC organic carbon (OC) using its sorbent properties. The procedure, referred to as the pyrene fluorescence loss (PFL) method, infers BC concentrations in particulate organic matter (POM) by observing the decrease in fluorescence from pyrene spiked into aqueous POM suspensions. The method was first tested using diverse materials previously utilized in an international BC method intercomparison study, and then its effectiveness (e.g., sensitivity and geochemical reasonableness) was tested by applying it to sediment and seawater POM samples collected from a coastal area downwind of important BC sources. Parallel evaluation of BC, using the PFL method and CTO-375 procedure, suggested we can characterize the predominant BC in a given sample as (i) thermally recalcitrant and highly sorptive per mass (e.g., soot), (ii) thermally labile and highly sorptive per mass (e.g., char), or (iii) thermally recalcitrant but not highly sorptive (e.g., lignite coal).
Introduction Research on black carbon (BC), a term given to a range of thermally altered carbonaceous materials produced by the incomplete combustion of fossil fuels and biomass, is diversifying as new observations suggest that BC has potential impacts on human health (1, 2), climate change (3, 4), global carbon cycling (5), and pollutant fates (6). These diverse roles have led to a wide variety of analytical methodologies, * Corresponding author phone: (617)253-1638; fax: (617)258-8850; e-mail:
[email protected]. † Massachusetts Institute of Technology. ‡ Woods Hole Oceanographic Institution. 4864
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 13, 2009
terminologies, and conceptual approaches, with various scientific communities interested in different characteristics. For example, most atmospheric scientists are interested in the “blackness” of aerosols and the resultant effect on the radiative heat balance (3, 7), while oceanographers and soil scientists are interested in the refractory properties of BC in soils and sediments influencing its role in carbon cycling (5). Concurrently, environmental chemists and engineers are investigating the impact of BC on contaminant mobility and bioavailability. Studies since the late 1970s and early 1980s have suggested that absorption into biogenic and diagenetic organic matter is a key process controlling the fate and effects of hydrophobic organic pollutants (8, 9), and this process has been described using a linear sorption model Kd ) fTOCKTOC
(1)
where Kd is the solid-water distribution coefficient (L/kgsolid), and this parameter is estimated using the product of the total organic carbon (TOC) fraction (fTOC, kgTOC/kgsediment) and the TOC-normalized distribution coefficient (KTOC, L/kgTOC). However, some observations found polycyclic aromatic hydrocarbons (PAHs) sorbed to sediments more than predicted using this model (10-12). Consequently, recent studies (13, 14) refined this sorption model by adding an adsorption term to eq 1 that is the product of the BC fraction (fBC, kgBC/ kgsediment), the BC-normalized distribution coefficient (KBC, (µg/kgBC/(µg/L)n), and a function of Cw, the truly dissolved concentration (µg/L) to the (n - 1) power, where n is the Freundlich exponent n-1 Kd ) fOCKOC + fBCKBCCw
(2)
In this expression, OC reflects the difference, TOC - BC. Many investigations have now demonstrated the importance of the second term on the right-hand side of eq 2 with respect to the sorption of hydrophobic compounds like PAHs by many natural heterogeneous solids (6, 11-15). The main objective of this effort was to develop a new method for measuring BC in environmental samples by taking advantage of the prominent role of BC as a sorbent for PAHs. To this end, we sought to quantify pyrene fluorescence loss (PFL) in pyrene-spiked aqueous solutions to which materials potentially containing BC sorbents were added. We made the assumption that pyrene would sorb in a manner as previously described by Accardi-Dey and Gschwend (14) with KOC ) 104.7, KBC ) 106.25, and n ) 0.62. Then together with measures of the TOC (OC + BC), we tested this approach using a diverse group of BC reference materials previously used in a method intercomparison (16). Finally, we used the PFL method on a natural sediment and particulate organic matter (POM) filtered from seawater samples to demonstrate the effectiveness (e.g., sensitivity and geochemical reasonableness) of this approach for examining the presence of BC in real world environments.
Materials and Methods Reference Materials. Twelve reference materials, previously utilized in an international BC method intercomparison study (16), were examined with our new method. This suite of samples included (1) three laboratory-produced BC-rich materials and a BC-free material (hexane soot, wood char, grass char, and melanoidin), and (2) eight environmental matrices (Suwannee River dissolved organic matter (DOM), shale, lignite coal, bituminous coal, urban dust, German soil, Australian soil, and marine sediments). The chemothermal 10.1021/es900043c CCC: $40.75
2009 American Chemical Society
Published on Web 05/29/2009
oxidation method (CTO-375) (6) results reported here for these 12 reference materials correspond to those previously published (16). South Dorchester Bay (SDB) Sediment. Grab samples of surface sediments were collected from a littoral site in South Dorchester Bay (SDB) near Quincy, MA (42° 18.0′ N, 71° 1.8′ W) in December 1999 and have previously been shown to contain BC using CTO-375 (14). The samples were stored in ziplock bags at 4 °C in dark. Seawater Sample Collection. Particulate organic matter (POM) was collected from five stations in the Gulf of Maine (GoM) (Figure S1 of the Supporting Information) during April 2004. At each station, between 100 and 500 L of shallow (
