Supporting Information. The effects of iron on optical properties of dissolved organic matter

Transcription

1 1 1 Supporting Information The effects of iron on optical properties of dissolved organic matter Brett A. Poulin 1, Joseph N. Ryan 1, and George R. Aiken 2 * 1 Department of Civil, Environmental and Architectural Engi neering, University of Colorado Boulder 2 United States Geological Survey, Boulder, CO *Corresponding Author; Tel: (303) ; Fax: (303) ; graiken@usgs.gov Contents 1. EEMs AND ION CONCENTRATIONS OF DOM SAMPLES (Figure S1, Table S1) ESTIMATES OF DOM ELECTRON-DONATING CAPACITY IRON(III)-DOM SOLUTIONS: EQUILIBRATION TIME AND MATRIX EFFECTS (Figure S2) OPTICAL MEASUREMENTS OF DOM RAYLEIGH SCATTERING OF FLUORESCENCE SPECTRA (Figure S3) SUWANNEE RIVER HPOA SOLUTIONS AT ph (Figure S4) UV ABSORPTION SPECTRA OF IRON(III)-DOM SOLUTIONS (Figure S5) IRON(III) ABSORTION COEFFICIENTS (Table S2) EFFECTS OF IRON(III) ON UV ABSORPTION PARAMETERS (Figure S6) EEMs OF FLUORESCENCE QUENCHED BY IRON (Figure S7, S8, S9, S10) ph EFFECTS ON DOM FLUROESCENCE IN THE PRESENCE AND ABSENCE OF IRON (Figure S11, S12, Table S3) REFERENCES... 20

2 EEMs AND BACKGROUND ION AND METAL CONCENTRATIONS OF DOM SAMPLES Excitation-Emission Matrices of Dissolved Organic Matter Samples Figure S1. Figure showing excitation-emission matrices (EEMs) of (a) Everglades F1 water, (b) Everglades F1 HPoA, (c) Williams Lake water, and (d) Suwannee River HPoA at ph Ion and Metal Concentrations Surface water samples and dissolved organic matter (DOM) isolates were analyzed for metals to ensure that the samples did not contain significant levels of metals known to quench or enhance DOM fluorescence. Data reported in Table S1 are for dilute surface water samples. Metal concentrations were well below levels reported to induce changes in DOM fluorescence properties. 1-3 Although metal

3 concentrations for Williams Lake water were not available, historic ( ) metal concentrations of Williams Lake surface waters from the U. S. Geological Survey Shingobee Headwaters Aquatic Ecosystems Program (SHAEP) did not show appreciable levels of Mn (< mg L -1 ), Cu (< 0.03 µg L -1 ), Ni (< 0.04 µg L -1 ), or Zn (< 0.03 µg L -1 ) Table S1. Ion and metal concentrations of surface water samples and DOM isolates of experimental solutions. Constituents Suwannee River Everglades F1 Everglades F1 Williams Lake HPoA (GA) HPoA (FL) water (FL) water (MN) Cl - (mg L -1 ) < 0.02 < NO - 3 (mg L -1 ) 0.11 < 0.05 < SO 2-4 (mg L -1 ) < 0.2 < NH + 4 (mg L -1 ) NA NA < Na + (mg L -1 ) < 0.02 K + (mg L -1 ) < 0.10 < Mg 2+ (mg L -1 ) < 0.02 < Ca 2+ (mg L -1 ) Fe (mg L -1 ) < 0.01 < 0.01 < 0.01 < 0.01 Mn (mg L -1 ) < < N/A Co (µg L -1 ) < 0.04 <0.02 < 0.04 N/A Cu (µg L -1 ) 2.0 < N/A Ni (µg L -1 ) 0.3 < N/A Zn (µg L -1 ) < 1.69 <0.43 < 1.69 N/A Ions determined by ion chromatography; trace and major metal concentrations determined by acidification (1% HNO 3 ) and measured by inductively coupled plasma mass spectrometry (ICP-MS) and inductively coupled plasma optical emission spectrometry (ICP-OES); N/A, not available.

4 ESTIMATES OF DOM ELECTRON-DONATING CAPACITY The electron-donating capacities of Suwannee River HPoA and DOM in Williams Lake water were estimated using recently published values for International Humic Substances Society (IHSS) isolates. 5 The electron-donating capacities of DOM samples were measured at an applied E h = V. 5 Although the E h of experimental solutions was not measured, it is anticipated to be lower than V, which would result in a decrease in the thermodynamic driving force for electron transfer from DOM to iron(iii). The 2.5 mgc L -1 Suwannee River HPoA sample used in this study encompasses both humic and fulvic acid. The electron-donating capacities of 2.5 mgc L -1 Suwannee River humic (4.2 µmol e- mgc -1 ) and fulvic acid (3.2 µmol e- mgc -1 ) were estimated to be 1.1 and 0.8 µmol e- (ph 7, E h V), respectively. Pony Lake fulvic acid reference was used to estimate the electron-donating capacity for Williams Lake water (4.5 mgc L -1 ) since both samples were obtained from microbial sources and have lower molecular weights and lower aromatic carbon contents. When assuming that Pony Lake fulvic acid comprised 100% of Williams Lake water DOC, the electron-donating capacity of 4.5 mgc L -1 Pony Lake fulvic acid (1.2 µmol e- mgc -1 ) was estimated to be 0.5 µmol e-. The electron equivalents of observed levels of iron(iii) reduction were then compared to estimates of DOM electron-donating capacity. Iron(III) reduction in Suwannee River HPoA systems varied from mg L -1 depending on the experimental iron concentration. This range is equivalent to µmol e-. The estimated electron-donating capacities of Suwannee River humic and fulvic acid were always greater than eight-fold and six-fold, respectively, the electron equivalence of reduced iron(iii). A maximum of 0.02 mg L -1 iron(iii) reduction (0.036 µmol e- ) was observed in the presence of Williams Lake water, which was equivalent to ~7% of the DOM electron-donating capacity assuming 100% of Williams Lake water DOC was Pony Lake fulvic acid. When assuming that 50% Williams Lake water DOC was composed of Pony Lake fulvic acid, the electron-donating capacity exceeded electron equivalence seven-fold. Under these scenarios the electron-donating capacities of samples always exceeded the electron equivalence of reduced iron(iii). Therefore, the oxidation of DOM due to changes in iron oxidation state are considered negligible.

5 IRON(III)-DOM SOLUTIONS: EQUILIBRATION TIME AND MATRIX EFFECTS No significant changes in the absorption spectra ( 2% at wavelengths above 220 nm) of Suwannee River HPoA (2.5 mgc L -1 ) equilibrated with 1.5 mg L -1 iron(iii) were observed between 4 and 96 h at ph 6.7 (Figure S2). Additional iron solubility tests were performed on identical 24 h-equilibrated solutions. Solutions were filtered through a 0.45 µm Supor Membrane syringe filter (Pall Corporation) and analyzed for total iron, [DOC], SUVA 254, and fluorescence properties. No significant changes ( 1%) in analyte concentrations or DOM spectrometric measurements were observed. An operational cutoff of 0.45 µm was chosen due to its wide acceptance for the collection of field samples. Samples containing Suwannee River HPoA fluorescence between ph were conducted in the presence of 1 mm NaClO 4. With the exception of ph 2 treatments, ionic strengths of all solutions were within 10% of each other (Visual MINTEQ version 3.0). 6 Further, adjustment of solution ionic strength to 1 mm (equivalent to the ionic strength of ph treatments) and 10 mm (equivalent to the ionic strength of ph 2 treatment) with NaClO 4 did not alter Suwannee River HPoA spectroscopic properties Figure S2. Figure showing ultraviolet and visible (UV-Vis) light absorption spectra of Suwannee River HPoA in the presence of 1.5 mg L -1 iron(iii) equilibrated between h.

6 OPTICAL MEASUREMENTS OF DOM UV-Vis absorption spectra from nm were obtained using a UV-Vis spectrophotometer (Agilent Technologies, model 8453) and a 1 cm quartz cuvette. The UV-Vis spectrophotometer was blanked with DI water prior to data collection. Decadic absorbance values were converted to absorption coefficients as follows: (S1) where is the absorption coefficient (cm -1 ), is the absorbance, and is the path length (cm). SUVA 254 values were calculated by dividing the by [DOC] and are reported in units of L mgc 1 m The E 2 :E 3 ratio is defined as the ratio of to and shows an inverse correlation with DOM molecular size. 8 For measurements of spectral slope, Naperian absorption coefficients were determined using the following: (S2) where is the Naperian absorption coefficient (cm -1 ), is the absorbance, and is the path length (cm). The spectral slope was calculated by fitting the exponential equation S3 to the Naperian absorption spectra between wavelengths of nm (S ) and nm (S ): (S3) where is the absorption coefficient at the specified wavelength, is the absorption coefficient at the reference wavelength, and S is the slope fitting parameter. 9 The spectral slope ratio (S R ) was defined as S : S and has been inversely correlated to DOM molecular weight. 10 Fluorescence EEMs were collected by scanning samples between excitation nm at 5 nm intervals and emission nm at 2 nm intervals using a Jobin Yvon Horiba Fluoromax 3 fluorometer. Fluorescence spectra, reported in Raman units (RU), were corrected for UV-Vis absorption, DI blank, and Raman scatter. EEM subtractions were performed using MATLAB (version R2008b) with average EEMs from experimental replicates in the presence and absence of iron. EEMs of fluorescence quenched (EEM FQ ) only included values that exceeded the error in replicate EEM measurements. The fluorescence index (FI) was defined as the ratio of emission intensities measured at 470 and 520 nm with excitation at 370 nm. 11 The redox index (RI) was determined as follows:

7 7 119 (S4) where is the sum of the reduced quinone components (SQ1, SQ2, SQ3, HQ) and is the sum of the oxidized quinone components (Q1, Q2, Q3) according to assignments proposed by Cory and McKnight. 11 A greater RI has been proposed to indicate DOM of lower redox potential. 12 The loading for the reduced and oxidized components were generated from the 13-component PARAFAC model. 11

8 RAYLEIGH SCATTERING OF FLUORESCENCE SPECTRA Figure S3. Figure showing the relative changes in Rayleigh scattering (unitless) measured at excitation wavelength 350 nm emission wavelength 350 nm in (a) iron(ii) and (b) iron(iii) experiments with increasing iron:doc concentration ratio (mg Fe mg -1 C ) at ph 6.7. Panel (c) shows changes in Rayleigh scattering (RU) of Suwannee River HPoA between ph in the presence and absence of iron. Data points in plot a and b are mean values of experimental duplicates with error bars indicating the high and low values.

9 SUWANNEE RIVER HPOA SOLUTION AT ph Figure S4. Figure showing (a) distribution of iron(ii) and iron(iii) in the absence of carbonate, (b) decadic absorption coefficients at λ = 254 nm, and (c) the overall fluorescence intensity (OFI) for solutions containing 0.68 mg L -1 total iron added as iron(ii) in the presence of 2.5 mgc L 1 Suwannee River HPoA between ph In plot b, absorption coefficients at λ = 254 nm of iron containing solutions were corrected for quantified iron(iii) using the absorption coefficient in Table S2.

10 UV ABSORPTION SPECTRA OF IRON(III)-DOM SOLUTIONS Figure S5. Figure showing ultraviolet and visible (UV-Vis) light absorption spectra of Suwannee River HPoA (ph 6.7) in the presence and absence of (a) 0.75 mg L -1 added iron(iii) and (b) 0.68 mg L -1 iron(iii) generated from iron(ii) oxidation. Panel (c) shows absorption spectra of added 0.75 mg L -1 iron(iii) collected in the presence and absence of DOM and panel (d) shows absorption spectra of 0.68 mg L -1 iron(iii) generated from iron(ii) oxidation in the presence of DOM. 150

11 IRON(III) ABSORTION COEFFICIENTS Table S2. Iron(III) absorption coefficients (ε λ ) at wavelengths of 254, 280, 350, and 400 nm; R 2 and p values are from linear regression analysis. Wavelength (nm) ε λ (L mg -1 cm -1 ) n R 2 p value x < x < x < x < 0.001

12 EFFECTS OF IRON(III) ON UV ABSORPTION PARAMETERS Our finding that iron(iii) reduces spectral slopes is consistent with previous studies showing a reduction in spectral slopes with increasing total iron 13 and iron(iii). 14,15 More pronounced decreases in S R and E 2 :E 3 ratios observed in Williams Lake water samples (Figure S6) illustrate that water samples containing primarily chromophore-poor, aliphatic DOM are more susceptible to interference by iron(iii). Similar results were observed by Xiao et al. 15 when comparing change in S R between Suwannee River humic and fulvic acids and Pony Lake fulvic acid Figure S6. Figure showing negative relationships between (a) spectral slope ratios (S R ) and (b) E 2 :E 3 ratios ( : ) with increasing iron(iii) concentrations. Data points are mean values of experimental duplicates with error bars indicating the high and low values.

13 EXCITATION-EMISSION MATRICES OF FLOURESCENCE QUENCHED BY IRON Figure S7. Figure showing excitation-emission matrices (EEMs) of Everglades F1 water with increasing iron(iii) (a d) and EEMs of fluorescence quenched (EEM FQ ) by iron(iii) (e g) at ph 6.7. The iron:doc concentration ratio is specified on each plot. Fluorescence intensities for each panel are normalized in Raman units (RU).

14 Figure S8. Figure showing excitation-emission matrices (EEMs) of Williams Lake water and EEMs of fluorescence quenched (EEM FQ ) by iron(ii) (left panel; a g) and iron(iii) (right panel; h-n) at ph 6.7. The iron:doc concentration ratio is specified on each plot. Fluorescence intensities for each panel are normalized in Raman units (RU).

15 Figure S9. Figure showing excitation-emission matrices (EEMs) of Everglades F1 HPoA and EEMs of fluorescence quenched (EEM FQ ) by iron(ii) (left panel; a g) and iron(iii) (right panel; h-n) at ph 6.7. The iron:doc concentration ratio is specified on each plot. Fluorescence intensities for each panel are normalized in Raman units (RU)

16 Figure S10. Figure showing excitation-emission matrices (EEMs) of Suwannee River HPoA and EEMs of fluorescence quenched (EEM FQ ) by iron(ii) (left panel; a g) and iron(iii) (right panel; h-n) at ph 6.7. The iron:doc concentration ratio is specified on each plot. Fluorescence intensities for each panel are normalized in Raman units (RU).

17 ph EFFECTS ON DOM FLUROESCENCE IN THE PRESENCE AND ABSENCE OF IRON Figure S11. Figure showing excitation-emission matrices (EEMs) of Suwannee River HPoA in the absence of iron between ph (left panel; a-d) and EEMs of fluorescence reduction (EEM FR ) (right panel; e-g) at the specified ph. Fluorescence intensities for each panel are normalized in Raman units (RU).

18 Figure S12. Figure showing changes in the fluorescence index (FI), defined here as the ratio of emission intensities at λ = 470 and λ = 520 nm at excitation λ = 370 nm, 11 of Suwannee River HPoA in the absence and presence of 0.68 mg L -1 iron added as iron(ii) between ph

19 19 Table S3. PARAFAC model component distributions of Suwannee River HPoA in the presence and absence of iron obtained using two different models; 11,16 values are averages of experimental duplicates. Difference in component distributions between experimental duplicates did not exceed 0.7% and 1.3% in the 13- and 7-component models in all samples. ph 5.0 samples i n the presence of iron omitted due to deviation from target ph. 13-Component Model 11 Excitation Maxima (nm) Emission Maxima (nm) Control: Suwannee River HPoA (no iron) Suwannee River HPoA mg L -1 iron ph 6.7 ph 6 ph 5.5 ph 5 ph 4 ph 3 ph 2 ph 6.7 ph 6 ph 5.5 ph 4 ph 3 ph 2 C % 12% 13% 13% 13% 15% 14% 11% 12% 13% 15% 15% 14% C2 (Q2) % 16% 16% 16% 15% 16% 16% 18% 17% 17% 16% 16% 16% C % 1% 1% 1% 1% 1% 2% 2% 1% 1% 1% 1% 2% C4 (HQ) % 27% 27% 27% 26% 24% 24% 26% 25% 24% 23% 24% 24% C5 (SQ1) % 10% 10% 10% 9% 9% 9% 9% 9% 8% 8% 9% 9% C % 10% 11% 11% 11% 10% 10% 7% 9% 10% 10% 10% 10% C7 (SQ2) % 4% 3% 3% 2% 3% 2% 7% 5% 4% 4% 3% 2% C8 (trp-like) 280 < 350 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 1% C9 (SQ3) 265, % 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% C % 6% 6% 6% 5% 5% 4% 5% 6% 6% 6% 5% 4% C11 (Q1) % 8% 8% 8% 8% 8% 8% 8% 8% 9% 9% 8% 8% C12 (Q3) % 4% 4% 4% 4% 5% 6% 5% 5% 5% 5% 5% 6% C13 (tyr-like) % 2% 2% 2% 3% 3% 3% 2% 3% 3% 3% 3% 4% 7-Component Model 16 Excitation Maxima (nm) Emission Maxima (nm) ph 6.7 ph 6 ph 5.5 ph 5 ph 4 ph 3 ph 2 ph 6.7 ph 6 ph 5.5 ph 4 ph 3 ph 2 C1 < % 29% 29% 30% 30% 31% 28% 25% 28% 28% 29% 27% 25% C2 320, < % 32% 32% 35% 33% 34% 30% 27% 30% 23% 31% 30% 27% C3 270, % 24% 24% 19% 22% 22% 25% 28% 23% 23% 21% 21% 28% C4 305, < % 7% 7% 8% 6% 6% 7% 8% 8% 8% 9% 9% 8% C5 270 < 350 2% 2% 2% 4% 2% 2% 2% 2% 2% 2% 2% 2% 2% C6 275, % 5% 5% 2% 5% 5% 4% 4% 4% 4% 5% 5% 4% C7 250, % 2% 2% 2% 1% 1% 4% 7% 4% 4% 4% 5% 7%

20 REFERENCES (1) Senesi, N. Molecular and quantitative aspects of the chemistry of fulvic acid and its interactions with metal ions and organic chemicals: Part II. The fluorescence spectroscopy approach. Anal. Chim. Acta 1990, 232, (2) Cabaniss, S. E. Synchronous fluorescence spectra of metal-fulvic acid complexes. Environ. Sci. & Technol. 1992, 26, (3) Plaza, C.; Brunetti, G.; Senesi, N.; Polo, A. Molecular and quantitative analysis of metal ion binding to humic acids from sewage sludge and sludge-amended soils by fluorescence spectroscopy. Environ. Sci. & Technol. 2006, 40, (4) Stets, E., U.S. Geological Survey, Water Resources Discipline, Personal Communication, January (5) Aeschbacher, M.; Graf, C.; Schwarzenback, R. P.; Sander, M. Antioxidant properti es of humic substances. Environ. Sci. & Technol. 2012, 46, (6) Gustafsson, J. P. Visual MINTEQ version 3.0, Stockholm, Sweden, (7) Weishaar, J. L.; Aiken, G. R.; Bergamaschi, B. A.; Fram, M. S.; Fujii, R.; Mopper, K. Evaluation of specific ultraviolet absorbance as an indicator of the chemical composition and reactivity of dissolved organic carbon. Environ. Sci. & Technol. 2003, 37, (8) De Haan, H.; De Boer, T. Applicability of light absorbance and fluorescence as measures of concentration and molecular size of dissolved organic carbon in humic Lake Tjeukemeer. Water Res. 1987, 21, (9) Green, S. A.; Blough, N. V. Optical absorption and fluorescence properties of chromophoric dissolved organic matter in natural waters. Limnol. Oceanogr. 1994, 39, (10)Helms, J. R.; Stubbins, A.; Ritchie, J. D.; Minor, E. C. Absorption spectral slopes and slope ratios as indicators of molecular weight, source, and photobleaching of chromophoric dissolved organic matter. Limnol. Oceanogr. 2008, 53, (11) Cory, R. M.; McKnight D. M. Fluorescence spectroscopy reveals ubiquitous presence of oxidized and reduced quinones in dissolved organic matter. Environ. Sci. & Technol. 2005, 39, (12) Miller, M. P.; McKnight, D. M.; Cory, R. M.; Williams, M. W.; Runkel, R. L. Hyporheic Exchange and Fulvic Acid Redox Reactions in an Alpine Stream/Wetland Ecosystem, Colorado Front Range. Environ. Sci. & Technol. 2006, 40, (13) Maloney, K. O.; Morris, D. P.; Moses, C. O.; Osburn, C. L. The role of iron and dissolved organic carbon in the absorption of ultraviolet radiation in humic lake water. Biogeochemistry 2005, 75, (14) Pullin, M. J.; Anthony, C.; Maurice, P. A. Effects of iron on the molecular weight distribution, light absorption, and fluorescence properties of natural organic matter. Environ. Eng. Sci. 2007, 24, (15) Xiao, Y. H.; Sara-Aho, T.; Hartikainen, H.; Vähätalo, A. V. Contribution of ferric iron to light absorption by chromophoric dissolved organic matter. Limnol. Oceanogr. 2013, 58, (16) Cawley, K. M.; Butler, K. D.; Aiken, G. R.; Larsen, L. G.; Huntington, T. G.; McKnight, D. M. Identifying fluorescent pulp mill effluent in the Gulf of Maine and its watershed. Mar. Pollut. Bull. 2012, 64,