Supporting Information. Properties of Fe-Organic Matter Associations via Coprecipitation versus Adsorption. 1. Department of Plant and Soil Sciences

Transcription

1 Supporting Information Properties of Fe-Organic Matter Associations via Coprecipitation versus Adsorption Chunmei Chen 1*, James J. Dynes 2, Jian Wang 2, and Donald L Sparks 1 1. Department of Plant and Soil Sciences Delaware Environmental Institute University of Delaware, Newark, DE, USA Canadian Light Source Inc. University of Saskatchewan, Saskatoon, SK, Canada S7N X4 Number of Pages: 29 Number of Figures: 17 Number of Tables: 3 Corresponding Author *Phone: (32) Fax: (32) cmchen@udel.edu S1

2 Materials and Methods Dissolved Organic Matter Dissolved organic matter used in the adsorption and coprecipitation experiments was extracted from field-fresh samples of a forest floor layer by mixing organic materials with deionized water for 24 h. The dominant tree species were beech (Fagus grandifolia), red and black oak (Quercus rubra and Q. velutina), and tulip poplar (Liriodendron tulipifera). The suspension was stirred periodically, centrifuged at 2 g for 2 min and subsequently isolated by pressure filtration through.2-µm polysulfone membrane filters. The choice of a.2-µm filter instead of traditionally used.45-µm is to exclude the possible presence of small particles in the extracted DOM, since the small particles in the extracted DOM solution might affect the evaluation of the properties of adsorption and coprecipitation samples. The extraction was carried out just before the adsorption and coprecipitation experiments. Ferrihydrite The poorly crystalline structure of pure ferrihydrite was confirmed by X-ray diffraction (SI Figure S2). The product had a specific surface area (SSA) of 28.3 m 2 g -1, and micro- and mesopore volumes of 38.6 and 47.2 mm 3 g -1, respectively (Table 1). Adsorption and Coprecipitation Experiments In order to check whether the original DOM solution precipitates while ph increases, we adjusted the originally extracted DOM solution with 75 mg C L -1 to ph 7 by adding.1 M NaOH solution under vigourous stirring (~65 rpm). No DOM precipitation was observed. S2

3 The adsorption of DOM onto ferrihydrite was also conducted at ph 4 in addition to ph 7 while coprecipitation was performed at ph 7, as we want to compare our results with a previous similar study by Eusterhues et al., 1 who conducted adsorption and coprecipitation experiments at ph ~4.2 and 7, respectively. They found that adsorption of DOM took place at ph of ~4.5, at which most coprecipitates formed, based on a sudden increase in turbidity at this ph. In order to be consistent with the adsorption and coprecipitation study, both adsorption and coprecipitation suspensions were shaken in the dark at 6 rpm for 24 hours. No obvious particle alteration or disaggregation was observed after shaking. The possibility of DOM-vessel interaction and/or microbial degradation of DOM was excluded by verifying the TOC of a DOM solution not reacted with ferrihydrite. The DOC concentration without added ferrihydrite did not change during a similar equilibration period. The adsorption and coprecipitaion samples were prepared for C and Fe content (freeze-dried), SSA and porosity (freeze-dried), XRD (freeze-dried), FTIR (freeze-dried), Fe EXAFS (freeze-dried and STXM-NEXAFS (moist) analysis. Additionally the precipitated insoluble Fe(III)-organic complexes reference sample was synthesized for Fe EXAFS analysis. The samples were stored in the 4 C refrigerator before analysis to prevent microbial growth. The moist samples for STXM-NEXAFS analysis were freshly prepared before the analysis, as we observed microbial growth after storing the moist samples for a week before analysis. S3

4 Synthesis of the precipitated insoluble Fe(III)-organic complexes For preparing the precipitated insoluble Fe(III)-organic complexes sample,.2 g Fe(NO 3 ) 3 9H 2 O was mixed with 2ml of a 75 mg C L -1 DOM solution, generating a C/Fe molar of 25.. The ph of this suspension is ~2.6. Such a low ph was chosen to avoid the precipitation of hydrous Fe oxides. After 1 h of shaking the suspensions at 6 rpm, the suspensions were centrifuged at 2 g for 3 minutes. The settled material was washed twice to remove the remainder of the equilibrium solution, and then freezedried. Ultraviolet visible Spectroscopy To characterize the fractionation process of DOM in the course of adsorption and coprecipitation with ferrihydrite, the initial DOM solution and the filtrates, after adsorption and coprecipitation, were analyzed by UV-vis spectroscopy with a Shimadzu UV-Probe PC251 UV Vis spectrophotometer (Columbia, MD). Prior to measurement, the OC content of these solutions was diluted with DI water to 1 mg L -1 to ensure comparability of all OM samples. Absorbances were recorded at 254, 28 and 365 nm. The specific UV absorbance at 28 nm, calculated as the ratio between absorption intensity at 28 nm and the OC concentration of the sample, is an index of aromaticity 2. Scheel et al. 3 found a linear relationship between UV absorbance at 28 nm and the aromatic C content as revealed by 13 C nuclear magnetic resonance spectroscopy. The ratio of absorbance at 254 nm to that at 365 nm (E 2 /E 3 ) serves as an index of the relative molecular size of a material and was calculated based on the respective absorption values at 254 and 365 nm. 4 The E 2 /E 3 ratio can provide a proxy of molar mass, with low values generally associated with high molecular weights. 5, 6 S4

5 Fe(II) analysis For determination of the Fe(II) content, ~ 2 mg of freeze-dried samples were dissolved in 5 ml of 37% HCl and subsequently made up to a final volume of 3 ml with 1% (v/v). Ferrous Fe was determined by the Ferrozine method 7. However, no Fe(II) was detectable in the adsorption and coprecipitate samples. X-ray Diffraction Freeze-dried pure ferrihydrite and coprecipitates were spread on glass slides and analyzed by X-ray diffraction (CuKα radiation, k= Å) with a Bruker D8 Discover diffractometer. The measurements ranged from 5 to 8 2θ, the step size was.5 2θ, and the acquisition time was 3s. Iron K-edge Extended X-ray Absorption Fine Structure (EXAFS) Spectroscopy Samples were mounted as thin powders between Kapton tape for EXAFS analysis. Mounted samples were placed in a He-purged sample chamber at room temperature and two to six replicate spectra were collected using a Lytle detector with energy calibrated against an Fe() foil (7112 ev). X-ray energy resolution was maintained by a crystal Si(22) and Si(111) monochromator at SSRL and NSLS, respectively. The monochromator was detuned by 5% of the maximum intensity in order to reduce the higher harmonic content in the beam. FTIR Analysis Spectra resolution was 2 cm -1. An automatic baseline correction and normalization were applied to each spectrum. To obtain spectra of OM associated with mineral phases, the baseline-corrected and normalized spectrum of pure ferrhydrite was subtracted from the spectra of the adsorption complexes and coprecipitates. S5

6 Scanning Transmission X-ray Microscopy Coupled with Near Edge X-ray Absorption Fine Structure (STXM-NEXAFS) Spectroscopy Samples for STXM-NEXAFS analysis were prepared by depositing 1-2 µl of an aqueous suspension of freshly prepared adsorption and coprecipitation samples onto a Si 3 N 4 window (75 nm thick), which was then air-dried. The samples were analyzed at STXM spectromicroscopy beamline 1ID-1 at the Canadian Light Source (CLS), a 2.9 GeV third-generation synchrotron source 8. The beamline is equipped with an APPLE II type elliptically polarizing undulator (EPU), which provides circularly polarized light (up to 1 ev) and linearly polarized light (13-25 ev). All measurements were performed using a 25 nm Fresnel zone plate, which provided a maximum spatial resolution of ~3 nm and the samples were kept under 1/6 atm of He during data collection. The samples were raster scanned with synchronized detection of transmitted X-rays to generate images. The energy scale of C1s spectra was calibrated by using the CO 2 absorption peak (C 3p) at ev by measuring CO 2 gas spectra before or after the measurements. 9 The carbon image sequences were aligned and then converted to optical density (OD=ln(I /I)) by using the transmitted flux through the sample (I) and the incident flux measured in an empty region adjacent to the analyzed particles (I ), to correct for the absorbance by the silicon nitride membranes. According to Lambert Beer s law, the optical density (OD) is directly proportional to the amount (thickness) of the individual compound in the X-ray beam path. The energies for obtaining the elemental distribution maps (on-resonance minus off-resonance) for C and Fe are listed in Table S1. The principle component analysis (PCA) and cluster analysis (CA) suggested that there were no significant spatial variations of C NEXAFS spectra for the adsorption S6

7 complexes and coprecipitates. In addition, there was no significant difference in the C spectra among three selected regions. Therefore, the C NEXAFS spectrum (Figure 3) for each sample was obtained from the averaged spectra from three selected regions. The spectra were baseline corrected and normalized. Results and Discussion DOM Fractionation: E 2 /E 3 and Specific UV Adsorption at 28 nm The OM remaining in solution after adsorption and coprecipitation was characterized by E 2 /E 3 ratios and specific UV adsorption at 28 nm. The extent of molecular weight fractionation appears to be greatly influenced by initial C/Fe ratios Figure S5a). The E 2 /E 3 was observed to decrease with increasing initial C/Fe ratios, remaining relatively constant at initial molar C/Fe ratios > 5 for all the systems (Figure S5a). Also, a larger portion of the OC was removed from solution at lower initial C/Fe ratios (Figure S6). This indicates that there was a preferential uptake of highermolecular-weight compounds at low C/Fe ratios, consistent with previous observations The extent of molecular weight fractionation appears to be increasing with increasing OC removal from solution and therefore with decreasing initial C/Fe ratios. Highmolecular-weight compounds were most enriched in the adsorption complexes and coprecipitates formed at low C/Fe ratios. At the comparable OC removal rate, the extent of increase in the E 2 /E 3 ratio after adsorption and coprecipitation followed the trend: coprecipitation at ph 7> adsorption at ph 4> adsorption at ph 7. This indicates that highmolecular-weight organic compounds are more selectively taken up by coprecipitation than adsorption and are more preferentially adsorbed at lower ph. S7

8 At initial molar C/Fe ratios <5, where larger portions of OC were removed from solution, the SUVA 28 decreased in the DOM solutions from 2.2 to ~.2-.3 after adsorption and coprecipitation (Figure S5b and S6), indicating preferential adsorption and coprecipitaton of more complex and aromatic moieties, which matches well with previous studies (e.g., Chorover and Amistadi, 21) 1. The decrease in specific UV absorbance after coprecipitation and adsorption became more pronounced with increasing C removal from solution and therefore with decreasing C/Fe ratios, suggesting that aromatic C was most enriched in the adsorption complexes and coprecipitates formed at low C/Fe ratios. At a comparable OC removal rate, coprecipitation at ph 7 resulted in a similar extent of reduction of SUVA 28 to adsorption at ph 4, while adsorption at ph 4 reduced SUVA 28 to a larger extent (on average, 24-54%) than at ph 7 (on average, 9-3%). This reflects a more selective adsorption of aromatic compounds at lower ph. S8

9 Table S1. Energies to map C and Fe distribution using image difference maps. Edge Off-resonance (ev) On-resonance (ev) C 1s Fe 2p S9

10 Table S2. Peak assignments for FTIR spectra of organic matter samples according to Gu et al., Swift, 15 Chorover and Amistadi, 1 Fu and Quan, 16 Artz et al., 17 Heckman et al., 18 Oren and Chefetz 12 (and references therein). Peak position (cm -1 ) Assignment 173 Symmetric C=O stretch of esters Asymmetric COO - stretch 152 Aromatic C=C stretching 146 CH 2 scissoring 14 Symmetric COO - stretch COO-metal stretch 127 O H stretching of phenolic OH 113 C O stretch of carbohydrates 18 C O stretch of carbohydrates 15 C O stretch of carbohydrates S1

11 Table S3. FTIR peak areas of the symmetric COO - band. Unreacted DOM Adsorption complexes Molar C/Fe ratio Peak position (cm -1 ) Peak Area Coprecipitates S11

12 Adsorbed and coprecipitated C (mg g -1 ) Adsorption, ph 4 Adsorption, ph 7 Coprecipitation, ph Initial molar C:Fe ratio Fig.S1. C contents in the adsorption complexes and coprecipitates as a function of the initial molar C/Fe ratio. S12

13 .49 nm.26 nm pure ferrihydrite.15 nm C/Fe = 1.5 Intensity C/Fe = 2.8 C/Fe = 5.6 C/Fe = θ (CuKα) Fig.S2. XRD diffractograms of the ferrihydrite used for preparation of adsorption complexes and the coprecipitates with different molar C/Fe ratios. S13

14 3 25 SSA (m 2 g -1 ) micropore volume mesopore volume total pore volume Pore Volume (cm 3 g -1 ) Fig.S3. Correlation between specific surface area and pore volumes. S14

15 Fig.S4. Iron k 3 -weighted EXAFS spectra of pure ferrihydrite (dotted line) and adsorption complexes with molar C/Fe ratio of 4.5 (solid line). S15

16 (a) 18 E 2 /E Adsorption, ph4 Adsorption, ph7 Coprecipitation, ph7 (b) Initial molar C/Fe ratio SUVA 28 (L mgc -1 m -1 ) Initial molar C/Fe ratio Fig.S5 Changes in E 2 /E 3 ratio and specific UV adsorption at 28 nm of the organic matter remaining in solution followed by adsorption at ph 4 and ph 7 and coprecipitation at ph 7, as related to the initial molar C/Fe ratio. Error bars indicate the standard error of the mean of three replicates. S16

17 8 OC removed (%) Adsorption, ph4 Adsorption, ph7 Coprecipitation, ph Initial molar C/Fe ratio Fig.S6. The portion of organic carbon taken up from solution by coprecipitation and adsorption as a function of initial molar C/Fe ratio. Fig.S7. Color indications of Red-Green-Blue (RGB) composite maps S17

18 a.9 1 µm b 1.5 Fig.S8. STXM elemental distribution maps of (a) carbon and (b) iron for the adsorption complexes with a molar C/Fe ratio of The µm gray scale indicates thickness in optical density (OD). S18

19 a b µm Fig.S9. STXM elemental distribution maps of (a) carbon and (b) iron for the coprecipitates with a molar C/Fe ratio of 1.5. The gray scale indicates thickness in optical density. S19

20 a b µm Fig.S1. STXM elemental distribution maps of (a) carbon and (b) iron for the adsorption complexes with a molar C/Fe ratio of 4.5. The gray scale indicates thickness in optical density. S2

21 a b µm Fig.S11. STXM elemental distribution maps of (a) carbon and (b) iron for the coprecipitates with a molar C/Fe ratio of 5.6. The gray scale indicates thickness in optical density. S21

22 Fig.S12. Correlation plots of thickness (OD) values of C and Fe obtained from STXM elemental distribution maps of the (a) the adsorption complexes with a molar C/Fe ratio of 1.5, (b) coprecipitates with a molar C/Fe ratio of 1.5, (c) the adsorption complexes with a molar C/Fe ratio of 4.5, and (d) coprecipitates with a molar C/Fe ratio of 5.6. Correlation coefficients (R 2 ) are provided on the plots. S22

23 Additional elemental distribution and color-coded composite maps and correlation plots of carbon and iron from STXM technique. a.8 b µm d c Fig.S13. STXM elemental distribution maps of (a) carbon and (b) iron for the adsorption complexes with a molar C/Fe ratio of 1.5. The gray scale indicates thickness in optical density. (c) Color-coded composite maps of (a) carbon and (b) iron (Carbon, blue; Iron, red). (d) Correlation plots of thickness (OD) values of (a) carbon and (b) iron obtained from STXM elemental distribution maps. Correlation coefficients (R 2 ) are provided on the plots. S23

24 a.8 b 2. 2 µm d c Fig.S14. STXM elemental distribution maps of (a) carbon and (b) iron for the coprecipitates with a molar C/Fe ratio of 1.5. The gray scale indicates thickness in optical density. (c) Color-coded composite maps of (a) carbon and (b) iron (Carbon, blue; Iron, red). (d) Correlation plots of thickness (OD) values of (a) carbon and (b) iron obtained from STXM elemental distribution maps. Correlation coefficients (R 2 ) are provided on the plots. S24

25 a 1.5 b µm d c Fig.S15. STXM elemental distribution maps of (a) carbon and (b) iron for the adsorption complexes with a molar C/Fe ratio of 4.5. The gray scale indicates thickness in optical density. (c) Color-coded composite maps of (a) carbon and (b) iron (Carbon, blue; Iron, red). (d) Correlation plots of thickness (OD) values of (a) carbon and (b) iron obtained from STXM elemental distribution maps. Correlation coefficients (R 2 ) are provided on the plots. S25

26 a b µm c d Fig.S16. STXM elemental distribution maps of (a) carbon and (b) iron for the coprecipitates with a molar C/Fe ratio of 5.6. The gray scale indicates thickness in optical density. (c) Color-coded composite maps of (a) carbon and (b) iron (Carbon, blue; Iron, red). (d) Correlation plots of thickness (OD) values of (a) carbon and (b) iron obtained from STXM elemental distribution maps. Correlation coefficients (R 2 ) are provided on the plots. S26

27 2 15 % Fe released 1 5 adsorption complexes, ph 4 adsorption complexes, ph 7 coprecipitates, ph Molar C/Fe ratio of adsorption complexes and coprecipitates Fig.S17 Calculated portion of Fe in the desorption solution expressed as percentage of total Fe in the adsorption complexes and coprecipitates. S27

28 References 1) Eusterhues, K.; Rennert, T.; Knicker, H.; Kögel-Knabner, I.; Totsche, K. U.; Schwertmann, U. Fractionation of Organic Matter Due to Reaction with Ferrihydrite: Coprecipitation versus Adsorption. Environ. Sci. Technol. 211, 45, ) Kalbitz, K.; Schwesig, D.; Schmerwitz, J.; Kaiser, K.; Haumaier, L.; Glaserc, B.; Ellerbrock, R.; Leinweber, P.; Changes in properties of soil-derived dissolved organic matter induced by biodegradation. Soil Biol. Biochem. 23, 35(8), ) Scheel, T.; Dörfler, C.; Kalbitz, K., Precipitation of dissolved organic matter by aluminum stabilizes carbon in acidic forest soils. Soil Sci. Soc. Am. J. 27, 71, ) Polubesova, T.; Chen, Y.; Navon, R.; Chefetz, B. Interactions of hydrophobic fractions of dissolved organic matter with Fe3+- and Cu2+-montmorillonite. Environ. Sci. Technol. 28, 42, ) Peuravuori, J.; Pihlaja, K. Molecular size distribution and spectroscopic properties of aquatic humic substances. Anal. Chim. Acta 1997, 337, ) Guo, M.; Chorover, J. Transport and fractionation of dissolved organic matter in soil columns. Soil Sci. 23, 168, ) Stookey, L. L. Ferrozine--a new spectrophotometric reagent for iron. Anal. Chem. 197, 42(7), ) Kaznatcheev, K. V.; Karunakaran, C.; Lanke, U. D.; Urquhart, S. G.; Obst, M.; Hitchcock, A. P. Soft X-ray spectromicroscopybeamline at the CLS: commissioning results. Nucl. Instrum. Methods Phys. Res. Sect. A-Accel. Spectrom. Dect. Assoc. Equip. 27, 582, ) Ma, Y.; Chen, C.T.; Meigs, G.; Randall, K.; Sette, F. High-resolution K-shell photoabsorption measurements of simple molecules. Phys. Rev. A 1991, 44, ) Chorover, J.; Amistad, M. K. Reaction of forest floor organic matter at goethite, birnessite and smectite surfaces. Geochim. Cosmochim. Acta 21, 65(1), ) Kang, S.; Xing, B.S. Humic Acid Fractionation upon Sequential Adsorption onto Goethite. Langmuir 28, 24, ) Oren, A.; Chefetz, B. Sorptive and desorptive Fractionation of dissolved organic matter by mineral soil matrices. J. Environ. Qual. 212, 41, ) Gu, B.; Schmitt, J.; Chen, Z.; Liang, L; McCarthy, J. F. Adsorption and desorption of natural organic matter on iron oxide: mechanisms and models. Environ. Sci. Technol. 1994, 28, ) Gu, B.; Schmitt, J.; Chen, Z.; Liang, L; McCarthy, J. F. Adsorption and desorption of different organic matter fractions on iron oxide. Geochim. Cosmochim. Acta 1995, 59(2), ) Swift, R. S. Organic matter characterization. In Methods of Soil Analysis: Part 3, Chemical Methods, SSSA Book Series 5 (ed. D. L. Sparks et al.). Soil Science Society of America, Madison, WI, 1996, pp ) Fu, H.; Quan, X.; Complexes of fulvic acid on the surface of hematite, goethite, and akaganeite: FTIR observation. Chemosphere 26, 63, ) Artz, R. R. E.; Chapman, S. J.; Robertson, A. H. J.; Potts, J. M.; Laggoun-Defarge, F.; Gogo, S.; Comont, L.; Disnar, J. R.; Francez, A. J. FTIR spectroscopy can be S28

29 used as a screening tool for organic matter quality in regenerating cutover peat lands. Soil Biol. Biochem. 28, 4, ) Heckman, K.; Vazquez-Ortega, A.; Gao, X.; Chorover, J.; Rasmussen, C. Changes in water extractable organic matter during incubation of forest floor material in the presence of quartz, goethite and gibbsite surfaces. Geochim. Cosmochim. Acta 211, 75, S29