* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Structural characterisation of groundwater hydrophobic
Survey
Document related concepts
Transcript
Organic Geochemistry Organic Geochemistry 36 (2005) 385–397 www.elsevier.com/locate/orggeochem Structural characterisation of groundwater hydrophobic acids isolated from the Tomago Sand Beds, Australia Cameron McIntyre a a,1 , Christopher McRae Alessandro Piccolo b a,* , Barry D. Batts a, Division of Environmental and Life Sciences, Department of Chemistry, Macquarie University, Sydney, NSW 2109, Australia b Dipartimento di Scienze del Suolo, della Pianta e dellÕAmbiente (DISSPA) Sezione di Scienze Chimico-Agrarie, Università di Napoli ‘‘Federico II’’, Via Università 100, 80055 Portici, Italy Received 22 March 2004; accepted 7 October 2004 (returned to author for revision 28 June 2004) Available online 8 December 2004 Abstract Groundwater hydrophobic acids isolated from the Tomago Sand Beds, Australia have been structurally characterised using multiple analytical techniques. Temporal variability was observed during the study period (1996–2003) and data for two representative samples isolated in 1996 and 2002 are presented. The sample isolated in 2002 was a highly aromatic fulvic acid with branched aliphatic functional structures and high oxygen content. It differed from structural concepts which suggest that groundwater hydrophobic acids isolated from organically lean aquifers are predominantly aliphatic in nature, with a low oxygen content resulting from extended degradation processes. Data are consistent with the hydrophobic acids having inputs from both surface recharge and mineralised sedimentary organic carbon. The sample isolated in 1996 was significantly more aliphatic, indicating that a change in microbial activity/community has occurred at the site. 2004 Elsevier Ltd. All rights reserved. 1. Introduction Groundwater is an important source of potable water worldwide and its contamination with organic and inorganic compounds is a major concern. The dissolved organic carbon (DOC) in groundwater contains a significant amount of humic substances (HSs) derived from the breakdown of biogenic matter (Thurman, 1985a,b). HSs bind and transport beneficial and pollut* Corresponding author. Tel.: +612 9850 8288; fax: +612 9850 8313. E-mail address: [email protected] (C. McRae). 1 Current address: CSIRO Petroleum, P.O. Box 136, North Ryde, NSW 1670, Australia. ant species such as metal ions and organic compounds in a range of aquatic and terrestrial environments (Koopal et al., 2001), and in groundwaters they have been shown to increase the solubility of pesticides and form mutagenic compounds during the chlorination of drinking water (Johnson-Logan et al., 1992; Koivusalo and Vartiaanen, 1997). They have a significant geochemical role and understanding the structural features of HSs has been identified as a basic requirement for comparison of samples, understanding their chemistry and the determination of their environmental and ecological importance (Frimmel and Abbt-Braun, 1999; Nikolaou and Lekkas, 2001). As an example, Leenheer et al. (2001) recently characterised the dissolved organic matter (DOM) and its reactivity towards chlorine in reclaimed 0146-6380/$ - see front matter 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.orggeochem.2004.10.004 386 C. McIntyre et al. / Organic Geochemistry 36 (2005) 385–397 and receiving groundwater of Los Angeles County, CA to assess its safety for use as drinking water. The structural features of HSs in groundwater have received relatively little attention compared with samples from other aquatic and terrestrial environments. Thurman (1985a) provided a review of the first major study of groundwater HSs which established their chemical, physical and spectroscopic properties, and several unique characteristics were found compared to surface water HSs. The content of DOC in groundwater was generally low (>5 mg L1) and the isolated HSs contained low amounts of humic acid, aromatics and oxygenated functional groups. The exception was when the source was located in organic rich sediments containing kerogen, where DOC, humic acid and aromaticity were exceptionally high. Several other studies have been conducted on the specific characterisation of groundwater HSs or have included groundwater samples in broader aquatic HS characterisations (Wassenaar et al., 1990; Kim et al., 1990; Gron et al., 1996; Artinger et al., 2000; Alberts et al., 1992; Pettersson et al., 1989, 1994). These confirm that groundwater DOC is structurally unique as a result of extended exposure to adsorption and degradation processes. Adsorption during infiltration and movement tends to remove the humic acids, lowering DOC and aromaticity, while degradation results in the loss of oxygenated functional groups. Groundwater HSs have also been characterised as part of studies on landfill leachates, contaminated sites, the mineralisation of sedimentary organic carbon (SOC), drinking water studies and DOC budgets (Zwiener et al., 1999; Wong et al., 2002; Wassenaar et al., 1991; Thorn and Aiken, 1998; Cronan and Aiken, 1985; Leinweber et al., 2001; Kumke et al., 1999; Christensen et al., 1998; Calace et al., 2001; Backau et al., 2000; Nanny and Ratasuk, 2002). Given that the much of the literature does not deal specifically with background structural characterisations or Australian sites, data for these areas remain somewhat limited and inconsistent. In particular, the variability in the chemical, physical and spectroscopic techniques applied makes inter-study comparisons difficult. The Tomago Sand Beds near Sydney, Australia are a remnant dune system and shallow sandy aquifer and have been mined for mineral sands and used to supply drinking water. The groundwater geochemistry was radically altered during mining activities, leading to the potential contamination of water supplies for nearby populations with species such as iron and arsenic. DOC levels were also observed to increase with mining activities and it was thought that organic acids may be involved in the release and transportation of the inorganic contaminants. The objective of this study was to conduct a comprehensive background structural characterisation of the groundwater hydrophobic acids isolated from groundwater at the Tomago Sand Beds and com- pare the findings with other studies. The hypothesis was that the hydrophobic acids at this Australian site are structurally unique and have a significant role in the groundwater geochemistry. Structural changes in the hydrophobic acids were observed as the study progressed from 1996 to 2003 and thus data are presented for two representative samples isolated in 1996 (GW96) and 2002 (GW02). 2. Site description A detailed description of the study area can be found in Prosser and Roseby (1995) and a brief summary is given here. The Tomago sand beds are located 16 km north of Newcastle, Australia. They are characterised by a dune system of reworked Pleistocene marine sands that have developed podzol soil profiles since stabilisation (12,000 BP). Vegetation is dominated by dry sclerophyll forests with low lying areas containing swamps. The sands are dominated by quartz and an extensive aquifer underlies the beds, with the water table depth ranging from 1 to 5 m. The podzol soils are predominantly humic in nature at low relief and iron/iron-humic at high relief. The depth of the B horizon varies with the water table and ranges from 0.5 to 3 m. The sands contain economically viable quantities of heavy minerals and these were mined for 30 years (production has ceased). The groundwater is used a source of potable water by the local water corporation for nearby populations and potential contamination resulting from mining has been an ongoing issue of concern. A monitoring programme for the groundwater is ongoing. 3. Experimental 3.1. Background data Background data for the study site were obtained by personal communication with Peter R. Wieland, Department of Chemistry, Macquarie University, Sydney. The data were extracted from the records of the groundwater monitoring programme at RZ Mines Pty Ltd, Newcastle. Additional data obtained from Peter R. Wieland were part of a separate research study of the site. 3.2. Samples, sampling methodology extraction procedures Samples were taken from a control piezometer located at UTM 56H 372940 1371020 (RZ Mines Pty Ltd identification: location 3502, piezometer number 2). The sampling site was a control one unaffected by mining activities, with piezometers set at depths of 3.6, C. McIntyre et al. / Organic Geochemistry 36 (2005) 385–397 6.5, 10.4 and 13.9 m below ground level. The piezometer used was at 6.5 m and was selected because of its high DOC and hydrogen sulfide content. It samples at least 4 m below the groundwater level (GWL) and below the B horizon of the soil. Standard Suwannee River fulvic acid (SRFA) was obtained from the International Humic Substances Society (IHSS; St. Paul, MN, USA). Standard SRFA was chosen as it has been well characterised (Averett et al., 1994) and was considered to be a suitable reference material for the characterisation of the groundwater hydrophobic acids in this study. Elemental data for SRFA were supplied by the IHSS with the sample. Representative samples were collected on 12th October 1996 (3 · 10 L) and 18th March 2002 (150 L). These samples highlight the structural changes observed over the study period (1996–2003) and data for other samples are not included for brevity. Samples were pumped mechanically and pressure-filtered with nitrogen through a 0.45 lm cellulose acetate membrane (Advantech MFS, USA) into glass (1996) and HDPE (2002) containers. The samples were acidified to pH 2 with concentrated HCl and transported to the laboratory. The hydrophobic acid fraction was isolated by adsorption on to a column of XAD-7 resin using a procedure based on that of Thurman and Malcolm (1981) and detailed in McIntyre et al. (1997). Elution of the acids was with 0.01 M KOH and cation exchange was effected with Amberlite IR-120 resin. The resulting solution was freeze dried without separation of the humic acids in order to obtain the whole hydrophobic acid fraction. 3.3. Dissolved organic carbon, elemental analysis and isotopic composition Non-purgeable organic carbon (NPOC) was determined on filtered and acidified samples using a Shimadzu TOC 5000A total organic carbon analyzer. Samples were sparged with CO2-free oxygen for 2 min prior to analysis. The instrument was calibrated over an appropriate range using potassium hydrogen phthalate; % DOC as hydrophobic acids was calculated as the mass of isolated hydrophobic acids L1 of groundwater · % C/DOC content of the groundwater. Humic acid content was determined gravimetrically by dissolving a known weight of hydrophobic acids in water and precipitating the humic acids at pH 1 with concentrated HCl. The humic acids were filtered off, dried, weighed and % wt determined. Elemental microanalysis was performed by The Campbell Microanalytical Laboratory (Department of Chemistry, University of Otago, NZ). C, H, N, S, O (dry ash free) were determined for GW02. C, H, N, S (dry ash free) were determined for GW96. Samples were dried for 24 h at 60 C under vacuum prior to elemental microanalysis. Reported elemental compositions of 75 387 groundwater fulvic acids were averaged for comparison (Wassenaar et al., 1990; Thurman, 1985a; Kim et al., 1990; Frimmel and Abbt-Braun, 1999; Artinger et al., 2000; Alberts et al., 1992; Pettersson et al., 1994). Stable isotope analysis for sulfur and carbon were performed at the Stable Isotope Science Laboratory, Department of Physics and Astronomy, University of Calgary, Canada. 3.4. Carboxyl group content Carboxyl group content was determined by titration to pH 8 using the method of Christensen et al. (1998). Titrations were conducted using a Radiometer Copenhagen TTA80 titration assembly, PHM82 standard pH meter and a GK2401C pH electrode. Carboxylic acid content was not determined for GW96 due to limited sample quantity. Phenol content was not determined as titration has been shown to give variable results above pH 8 due to hydrolysis reactions (Averett et al., 1994). 3.5. Infra-red spectroscopy Infra-red (IR) spectra were recorded on a Perkin– Elmer Paragon 1000PC FT-IR Spectrometer. Samples were dried at 60 C under vacuum for 24 h and analyzed as 1% w/w KBr discs. Spectra were recorded over the range 4000–400 cm1 using 16 scans at a resolution of 2 cm1 and were smoothed after acquisition. 3.6. 1H and 13 C NMR spectroscopy 1 H Spectra were recorded on a Bruker DPX400–400 MHz spectrometer using a WATERGATE pulse sequence for suppression of the water resonance (Liu et al., 1998). Samples were dissolved in D2O and 10,240 transients were collected with a spectral width of 5,995 Hz, an acquisition time of 2.73 s and pulse delay of 2 s. The solvent peak at 4.6 ppm was irradiated so protons in this region were not observed. Peak assignments were based on the literature (Wilson, 1987; Malcolm, 1990; Leenheer et al., 1998). Conditions used for Solid-State Cross Polarisation Magic Angle Spinning (CPMAS) 13C NMR spectroscopy are detailed in Leenheer et al. (2003). Insufficient sample prevented a spectrum for GW96 being recorded. Peak assignments were based on the literature (Malcolm, 1990; Nanny and Ratasuk, 2002). 3.7. High-pressure size exclusion chromatography High-pressure size exclusion chromatography (HPSEC) is an established method in the analysis of HSs for determination of molecular weight and size (Cozzolino et al., 2001; Conte and Piccolo, 1999). Analysis was performed using a Shimadzu LC10AD pump, a 388 C. McIntyre et al. / Organic Geochemistry 36 (2005) 385–397 Perking–Elmer LC295 UV–Vis detector and a 600 · 7.8 mm Polysep GFC P 3000 column with a Polysep P 3000 guard column (Phenomenex). The eluent was 0.05 M phosphate buffer at pH 7 and the analytes were dissolved in the eluent at a concentration of 0.2 g L1. Standards used for calibration were polystyrenesulfonates (LabService Analytica s.r.l., Italy) with molecular weights of 6.78, 16.8, 32 and 130 kDa. A calibration curve relating molecular weight to elution time was obtained by plotting the natural logarithm of the molecular weights [ln (Mw)] of the polystyrenesulfonate standards versus time. Linear regression was used to obtained the relationship between ln (Mw) and time. The injection volume was 100 lL and spectra were recorded at 250 nm for standards and 280 nm for samples. The different wavelengths have been shown to have a minimal effect (OÕLoughlin and Chin, 2001). Spectra were acquired using Perking–Elmer PE Turbochrom 4.0.1 software. Data were processed manually by exporting the chromatograms as time versus signal ASCII data files into Microsoft Excel and only time and signal data corresponding to the major peak of interest were retained. For each time slice a corresponding molecular weight, Mi, was calculated using the linear regression equation fitted to the calibration data. The weight fraction, wi, of analyte in any time slice was calculated by dividing the detector voltage response (baseline subtracted) by the sum of detector voltage P responses for each analyte peak (i.e., wi = intensityi/ intensityi). M w was obtained from the P sum of the products of wi and Mi (i.e., M w ¼ wi M i ). 1=M n was obtained P from the sum of the ratios of wi and Mi (i.e., M n ¼ X i M i , where Xi is the mole fraction of Pmolecules of molecular weight mass Mi, thus M n ¼ 1= ðwi =M i Þ). 3.8. Vapour pressure osmometry Vapour pressure osmometry (VPO) was performed using a Westcor, Vapro 5520 vapor pressure osmometer. Samples were analyzed using the method of Aiken and Malcolm (1987) and benzoin was the calibration standard. 3.9. UV–Vis spectroscopy UV absorbance was determined on a Varian, Cary 300 Bio UV–Vis Spectrophotometer using the procedure of Chen et al. (1977). Absorbance values were recorded at 365 and 250 nm for calculation of E2/E3 ratios and at 280 nm for calculation of molar absorbance. 3.10. Thermochemolysis with TMAH Offline tetramethylammonium hydroxide (TMAH) thermochemolysis was performed using the method of del Rio et al. (1998). The internal standard was eicosane at a concentration of 25 ng L1. Analysis was performed using a Fisons Instruments GC8000 gas chromatograph coupled to a MD800 mass selective detector. The oven was held at 40 C for 2 min, then ramped at 10 C min1 to 100 C, then ramped at 4 C min1 to 300 C and held for 2 min. Mass spectra were obtained by scanning from 35 to 500 Da in 1 s. The capillary column was a 50 m BP1 one (SGE, Australia) with a 0.5 lm film thickness. Compounds were identified by comparison of the spectra with a NIST 92 library. 3.11. Electrospray ionisation mass spectrometry Electrospray ionisation mass spectrometry (ESI-MS) was performed using the methods of McIntyre et al. (1997, 2002). Samples were dissolved in 75:25 methanol:water at a concentration of 1 mg mL1. Conditions for the acquisition of spectra are given in McIntyre et al. (2002). Neutral loss spectra were obtained with the same parameters used for product ion spectra but with a mass range of 100–700 Da, a scan time of 3 s and quadrupole resolution settings to achieve unit mass resolution. 4. Results and discussion 4.1. Background data The background data for the piezometer are summarised in Table 1. The pH of the groundwater is acidic and is controlled by the oxidation of sulfur and/or sulfide to sulfate in the oxic zone, with the low content of clay minerals minimising buffering of the system. Increasing sulfate and decreasing iron and sulfide levels with time indicate a change in microbial activity/composition associated with dissolution of pyrite and reduction of sulfate (Massmann et al., 2003). The pH has decreased with time, consistent with a reduction in bicarbonate production associated with microbial mineralisation of SOC during sulfate reduction and acid production from pyrite dissolution. DOC concentration has declined with time, indicating a reduction in the microbial mineralisation of SOC or reduced DOC recharge from surface environments. The decrease in concentration of iron and sulfur is consistent with the former (Backau et al., 2000). Oxidation/reduction potential (ORP) values indicate a reducing environment, although the periodic low levels of dissolved oxygen (DO) reflect the sampling depth being proximal to the water table. Conductivity (Cond.) is constant and indicates that the spear has not been affected by mining activities. Table 1 is incomplete as data were derived from several sources. Trends at this piezometer were observed in adjacent control piezometers set at different depths (data C. McIntyre et al. / Organic Geochemistry 36 (2005) 385–397 389 Table 1 Background geochemical data for groundwater located at the Tomago Sand Beds, Australia Date GWL (m) pH Temperature (C) Total Fe (mg/L) Sulfate (mg/L) Sulfide (mg/L) Total S (mg/L) NPOC (mg/L) ORP (mV) DO (mg/L) Cond. (mS/cm) 09/04/96 08/08/97 01/14/98 01/17/98 05/05/98 06/17/98 08/19/98 01/14/99 05/24/99 06/01/99 06/09/99 03/23/00 05/04/00 01/04/01 05/23/01 07/25/01 11/12/01 06/06/02 07/09/02 1.76 5.3 5.5 5.4 5.3 5.1 5.3 5.3 5.3 5.4 5.1 5.1 5.5 4.8 5.2 5.0 4.9 5.3 4.9 4.7 – – – – 19.6 19.2 – – – 20.6 20.2 – 21.2 – 18.9 15.4 – 19.3 17.1 2.1 1.4 2.6 2.2 1.8 2.2 2.0 1.4 1.5 1.7 0.5 1.4 1.9 1.0 0.7 – 0.8 0.8 0.6 – 9.0 6.9 1.7 – – 1.3 3.0 4.2 – – 7.6 – 9.6 80 32 11 30 25 – – – – 5.7 5.5 – – – 3.5 3.7 – 3.9 – 4.1 0.6 – 0.3 0.6 – – – – 340 128 – – – 68 132 – 11 – 40 – – 27 10 – – – – 8.2 7.0 – – – – – – – – – – – 4.5 2.2 – – – – 36 26 – – – 25 35 – 109 – 47 220 – 43 75 – – – – 0.16 0.27 – – – 0.08 0.11 – 0.13 – 0.18 0.24 – 0.40 0.22 – 0.16 0.20 0.19 0.16 0.15 0.19 0.15 0.13 0.10 0.10 0.17 0.16 0.20 0.16 0.11 0.19 0.15 0.14 1.31 0.80 1.82 1.75 0.10 0.32 0.23 0.89 0.30 1.16 1.43 1.18 1.60 0.80 0.31 0.45 not shown), confirmed their validity. All values were within normal variation for the site which has been monitored for 30 years. Variation in the parameters could not be correlated with rainfall and season but a major influence on the system is the high lateral flow rate of groundwater (1 m week1) which results in the rapid dispersal of ground water and surface recharge. The background data indicate this is a typical groundwater system. 4.2. DOC composition The concentration of DOC in the groundwater in 2002 (Table 2) is consistent with previous studies that have reported concentrations in soil solutions which range from 20 to 50 mg L1 in the O/A horizons, decreasing to >3 mg L1 in groundwater as infiltration occurs (Wassenaar et al., 1991; Thurman, 1985a,b; Pettersson et al., 1994). This reduction is generally attributed to extended exposure to adsorption and degradation processes. The DOC concentration in 1996 was considerably higher (Table 2). The background data indicates a higher level of microbial sulfate reduction in 1996 and are consistent with an elevated DOC being the result of microbial dissolution of SOC. Previous studies have reported the hydrophobic acid content of groundwater in organically lean aquifers (% DOC as hydrophobic acids) as 5–58% with a predominance of fulvic acids (Wassenaar et al., 1990; Thurman, 1985a; Gron et al., 1996; Artinger et al., 2000; Pettersson et al., 1994). Data for GW96 are consistent with this; however, for GW02 the hydrophobic acid content is above the upper end of the range. Higher levels of microbial sulfate reduction in an organically lean aquifer have been shown to produce a greater proportion of hydrophilic acids in the DOC (Wassenaar et al., 1991); this explains the lower value for GW96 and suggests that inputs of surface DOC may be more significant for GW02. The hydrophobic acid content as related to DOC content is not consistent with the findings of Wassenaar et al. (1990) and Thurman (1985a) who showed that humate content increases with DOC content in groundwater. This is also attributed to a change in microbial activity/community which has altered the total DOC inputs over the study period. The low humic acid content of both samples is consistent with removal by adsorption and degradation processes as water infiltrates and migrates through the ground and also reflects the organically lean composition of the aquifer. 4.3. Elemental and isotopic composition The average values of elemental composition from previous studies (cf. 3.3, 1 s.d. given in brackets) for carbon content, hydrogen content and H/C ratio were 53.9% (2.1), 4.95% (0.54) and 1.11 (0.12), respectively (n = 75). Oxygen content and the O/C ratio were 37.6% (2.0) and 0.53 (0.5) (n = 72). Nitrogen and sulfur contents were 1.07% (0.35) (n = 73) and 2.97% (0.68) (n = 48). Compared to reported values the oxygen content of GW02 is high. The lower oxygen content of reported samples has been attributed to the removal of carboxyl groups from fulvic acids during microbial and chemical degradation processes (Pettersson et al., 1994). A lower hydrogen content and H/C ratio indicate that GW02 390 C. McIntyre et al. / Organic Geochemistry 36 (2005) 385–397 Table 2 Sample information and analysis results for the hydrophobic acids isolated from groundwater (GW02 and GW96) and standard Suwannee River fulvic acid (SRFA) Parameter 1996/GW96 2002/GW02 SRFA Sampling date Volume collected (L) DOC (mg C L1) Hydrophobic acids isolated (mg) % of DOC as hydrophobic acids Humic acid content (% wt) 12/10/96 10 · 3 10.1 226 36 <5 18/03/02 150 2.2 410 64 <5 – – – – C (%) H (%) O (%) N (%) S (%) Ash (%) H2O (%) H/C O/C d13C (&) d34S (&) 48.2 4.21 – 1.32 – 8.42 – 1.04 – 27.6 9.9 52.3 4.10 41.5 1.20 2.58 2.87 6.62 0.93 0.56 27.6 1.6 52.4 4.31 42.2 0.72 0.44 0.46 8.8 0.97 0.60 27.6 – Carboxyl content (meq g1) Molecular weight (Da) by VPO Mw (Da) by HPSEC Mn (Da) by HPSEC Polydispersity (Mw/Mn) by HPSEC E2/E3 ratio Absorbance at 280 nm (L mol OC1 cm1) – – 1773 1356 1.31 4.5 310 5.5 953 1810 1499 1.21 4.3 530 5.7 859 2438 1972 1.24 4.4 420 has a less aliphatic character. Groundwater fulvic acids are suggested to have a lower aromatic content due to removal of humic acids by adsorption and degradation. In surface environments where these processes are less extensive, elemental compositions of fulvic acids are similar to that presented for SRFA. The C, H, O data for GW02 is closer to that of SRFA and suggests that the sample is relatively immature and/or may still receive some allochthonous DOC. If the DOC is derived in situ, this suggests that it has been oxidised, resulting in a higher oxygen content, or that the character of the sedimentary organic matter is unique. Dating (14C) of the samples and characterisation of the sediments at the site have not been done and thus conclusions with regard to DOC age/origin are not yet possible. The higher sulfur content of GW02 compared to SRFA is characteristic of groundwater environments and is indicative of hydrogen sulfide/colloidal sulfur produced by microbes during sulfate reduction being incorporated into the hydrophobic acids. Average N and S values of reported values are comparable to GW02 and typical for groundwater HSs. The elemental composition of GW96 is significantly different from that of GW02 and the reported values given above. Limited data were obtained due to the limited amount of sample available, but a lower carbon content and a higher H/C ratio suggest that GW96 contains more aliphatic (possibly oxygenated) compounds – than GW02. This is consistent with an addition of low molecular weight organic acids produced during microbial action on aquifer SOC during sulfate reduction, as proposed by Wassenaar et al. (1991), and the background data presented above. The carbon isotope ratio of 27.6& for all three samples (Table 2) indicates that they are most likely derived from higher plant matter (Smith and Epstein, 1971). This is consistent with published data (Wassenaar et al., 1990; Thurman, 1985a). The organic sulfur isotope ratios are consistent with literature values and are consistent with a change in biological activity at the site (Nissenbaum and Kaplan, 1972). GW96 is isotopically heavier, indicating greater depletion in 32S through sulfate reduction by microbes. This is supported by higher sulfide levels observed in the background data prior to 2002. The d 34S values of the dissolved sulfate and sulfide at the site (data not shown) also support this finding and indicate that the microbial community has changed or that microbial activity was occurring at shallower depths and closer to the piezometer, prior to 2002. 4.4. Carboxyl content GW02 has a carboxylic acid content which lies between values reported by Thurman (1985b) for fulvic acids in ground water (5.1–5.5 meq g1) and surface C. McIntyre et al. / Organic Geochemistry 36 (2005) 385–397 391 waters (5.5–6.2 meq g1). It is higher than a calculated mean value of 5.2 meq g1 for groundwater fulvic acids reported by Pettersson et al. (1994) and similar to the 5.7 meq g1 determined for SRFA (Table 2). These data are consistent with the elevated oxygen content determined by elemental analysis. The high carboxyl group content and oxygen content when compared to literature values again suggest that GW02 is immature and has been exposed to a lesser degree of decomposition or to greater oxidation. 4.5. Infrared spectroscopy Infrared spectra are similar to those reported by Kim et al. (1990) and Alberts et al. (1992) for groundwater fulvic acid (Fig. 1) and show a close resemblance to those of hydrophobic acids isolated from streams, lakes and forest floor solutions (Guggenberger et al., 1994; David and Vance, 1991). The data are typical for terrestrially derived aquatic hydrophobic acids, but also highlight the fairly non-specific nature of infrared spectroscopy. The IR spectra indicate the hydrophobic acids are primarily carboxylic (1720, 3400 cm1; C@O, OH stretching) with a low content of humic acids and complexed metal ions (1620 and 1400 cm1). Fig. 2. 1H NMR spectra of hydrophobic acids isolated from groundwater (GW02 and GW96) and the relative distribution of hydrogen atoms (U = phenyl). 4.6. 1H NMR spectroscopy GW02 has the majority of the observable protons occurring between 1.75 and 3 ppm (Fig. 2), indicating a high content of protons on carbon atoms adjacent to carboxyl groups. The peak at 2.5 ppm indicates a significant proportion of these are in branched aliphatic structures. In unbranched aliphatic structures, protons in this spectral region give a peak at around 2.2 ppm. Although Fig. 1. Infrared spectra of hydrophobic acids isolated from groundwater (GW02 and GW96). protons of methyl groups attached to aromatic rings may also appear in this region, 13C NMR spectra presented in the next section confirm that branched aliphatic structures are dominant in GW02. The spectrum is substantially different from that for hydrophobic acids isolated from a control groundwater, presented by Thorn and Aiken (1998), Well 310), which gave a major peak at 1.1 ppm corresponding to aliphatic methylene groups. It also differs substantially from marine and terrestrial HSs which have been shown to have major peaks in the aliphatic (1 ppm) and carbohydrate (4 ppm) regions (Hatcher et al., 1979; Malcolm, 1990). The spectrum more closely resembles that of surface water fulvic acids isolated from streams and lakes, where resonances near 2 ppm are more prominent (Malcolm, 1990; Ma et al., 2001). In particular, the spectrum shows a similar down field shift of protons adjacent to carboxyl groups as observed in the metal binding fraction of SRFA, which was attributed to clustering of the carboxyl groups (Leenheer et al., 1998). GW96 has a significantly different spectrum to that of GW02, highlighting the change in composition of the hydrophobic acids in the groundwater at the site since 1996 (Fig. 2). More intense peaks at 1.2 and 2.2 ppm indicate a greater content of aliphatic acids, and this is consistent with greater microbial mineralisation of SOC and other data. These may be branched methyl structures rather than linear aliphatic/alicyclic compounds, as resonances at 1.6–1.8 ppm due to protons 392 C. McIntyre et al. / Organic Geochemistry 36 (2005) 385–397 b to carboxylic acids increase only very slightly compared to corresponding peaks in GW02. 4.7. CPMAS 13 C NMR spectroscopy The normal spectrum has a high content of aromatic carbon (130 ppm), which is not consistent with the concept of groundwater humic substances having a low content of aromatic structures due to adsorption of humic acids (Fig. 3). The major aliphatic peak at 43 ppm is indicative of branched aliphatic carbons while the peak at 30 ppm is due to methylene carbons. The low signal at 15–20 ppm due to methyl carbons and the major resonance occurring at 43 ppm indicate that the peak at 2.4 ppm in the 1H NMR can be attributed to protons on branched aliphatic groups adjacent to carboxyl groups, rather than to methyl groups attached to aromatic rings. The small shoulder at 105 ppm and the peak at 76 ppm indicate a low content of anomeric carbon of carbohydrates and C–O linkages, respectively. Shoulders at 115 ppm and 155 ppm indicate lignin-derived aromatics are present in low quantities. Carboxyl carbons (170 ppm) are moderate in content and normal for aquatic fulvic acids (cf. Malcolm, 1990). The low signal level at 58 ppm also indicates methoxyl carbons of lignin Fig. 3. Normal and dipolar-dephased CPMAS 13C NMR spectra of hydrophobic acids isolated from groundwater (GW02) and relative distribution of carbon atoms. and esters have a minimal contribution. The data are consistent with previous sections and indicate predominantly aryl-branched aliphatic structures with a low content of lignin-derived compounds and C–O linkages (carbohydrates). The dipolar dephased spectrum indicates that the majority of aromatic and aliphatic carbon (0–160 ppm) is non-quaternary (Fig. 3). Terminal methyl at 15 ppm, quaternary methine at 48 ppm and tertiary alcohols at 80 ppm are evident while a low proportion of the aromatic carbon is protonated. These structures are hypothesised to be indicative of degraded terpenoid structures, as reported by Leenheer et al. (2003). For this site, selective or incomplete absorption processes in the vadose zone of the soil during water infiltration may have resulted in the enrichment of these types of species. The overlying vegetation (Eucalyptus) at the site is rich in terpenoids and is a ready source of precursor terpenoid molecules. 4.8. UV–Vis spectroscopic, molecular weight and molecular size measurements SRFA has been reported to have a Mn by VPO of 840 Da and a Mn, Mw and polydispersity by HPSEC of 1652 Da, 2290 Da and 1.39 (Aiken and Malcolm, 1987; OÕLoughlin and Chin, 2001). Chin et al. (1994) determined the molar absorbance at 280 nm to be 389 L cm1 (mol OC)1. The results presented for SRFA in this study correlate well with the above data (Table 2). The E2/E3 ratio was measured for comparison of aromaticity and average molecular weights as it has been shown to give a stronger correlation than the E4/E6 ratio (Peuravuori and Pihlaja, 1997). The molecular weight of GW02 determined using VPO is high compared with values for aquatic fulvic acids reported by Aiken and Malcolm (1987) of 500– 950 and 859 Da obtained for SRFA (Table 2). Pettersson et al. (1994) published molecular weights for aquatic fulvic acids determined by HPSEC using polystyrenesulphonate standards. Their results showed that the fulvic acids in deep groundwater were lower in molecular weight than in surface waters and this was attributed to adsorption and degradation processes. The VPO result for GW02 is not consistent with this concept and suggests a lesser degree of exposure to these processes, allowing higher molecular weight compounds to stay in solution. Peuravuori and Pihlaja (1997) have shown that molecular weights determined by HPSEC have a linear relationship to those determined by VPO. HPSEC data for GW02 (Table 2) infer that it has a lower molecular weight than SRFA. However, given the branched, compact nature of the sample as indicated by NMR, this suggests it is more likely to reflect a smaller hydrodynamic size. The polydispersity of GW02 is slightly lower C. McIntyre et al. / Organic Geochemistry 36 (2005) 385–397 then that determined for SRFA and indicates that the sample has a more homogenous composition. This is consistent with the findings of Pettersson et al. (1994) which indicated groundwater fulvic acids were less polydisperse than surface water fulvic acids, which again was attributable to maturity and a greater exposure to adsorption and degradation processes. Molar absorption at 280 nm and E2/E3 ratio have been shown to be related to aromaticity and inversely related to average molecular weight (Peuravuori and Pihlaja, 1997). The results for these parameters are consistent with 13C NMR and VPO data and confirm that GW02 is both more aromatic than and higher in molecular weight than SRFA. The molecular weight for GW96 could not be determined using VPO as in higher ash content resulted its being insoluble in the solvent. Molar absorbance at 280 nm and E2/E3 ratio indicate that the sample is less aromatic than GW02 and SRFA and that it has a lower molecular weight. The polydispersity is higher and indicates that the sample has a less homogenous composition. These data are consistent with the suggestion of greater microbial sulfate reduction and SOC mineralisation at the site in 1996 producing additional low molecular weight aliphatic acids and an increase in DOC. This also infers that the molecular weight determined by 393 HPSEC for GW96 reflects both the lower molecular weight acids produced by microbes as well as a smaller hydrodynamic size. 4.9. Thermochemolysis with TMAH The total ion current (TIC) chromatogram for GW02 (Fig. 4) is very similar to that for SRFA presented by del Rio et al. (1998) and indicates that the sample contains terrestrially derived aquatic DOC. TMAH thermochemolysis products identified in Table 3 are dominated by lignin derived compounds with guaiacyl and syringyl structures (Clifford et al., 1995) which are consistent with the overlying angiosperm vegetation. Products arising from tannin and cutan (peaks 14 and 21) are present and, even though the technique is not very efficient for carbohydrates, some markers can be seen as peaks 13 and 38. The total yield of products is <1.0% assuming a response factor of 1 for all compounds; this is consistent with the low content of these biopolymer classes observed using NMR spectroscopy. The data indicate that there is incomplete adsorption of DOC from surface water recharge. Ratios of vanillyl and syringyl acid:aldehyde are used to indicate lignin degradation (del Rio et al., 1998). Although the presence of vanillyl and syringyl aldehyde could not be confirmed, the methyl esters Fig. 4. TMAH thermochemolysis GC/MS total ion chromatograms for hydrophobic acids isolated from groundwater (GW02 and GW96). Peak identities are given in Table 3; * denotes internal standard eicosane. 394 C. McIntyre et al. / Organic Geochemistry 36 (2005) 385–397 Table 3 Peak identification table for Fig. 4 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 Butanedioic acid, dimethyl ester Methylbutanedioic acid, dimethyl ester C7-Alkenone 2-Methoxy phenol Pentanedioic acid, dimethyl ester 1,2-Dimethoxybenzene 1,3-Dimethoxybenzene Hexanedioic acid, dimethyl ester Dimethoxytoluene 1,2,3-Trimethoxybenzene 3-Methoxybenzoic acid, methyl ester 4-Methoxyacetophenone 1,2,4-Trimethoxybenzene 1,3,5-Trimethoxybenzene 4-Methoxybenzeneacetic acid, methyl ester 1,2,3-Trimethoxytoluene 3,4-Methylenedioxybenzoic acid, methyl ester 4-Methoxypropanoic acid, methyl ester+ Nonanedoic acid, dimethyl ester 3,4-Dimethoxyacetophenone 3,5-Dimethoxybenzoic acid, dimethyl ester 3,4-Dimethoxybenzoic acid, dimethyl ester 3,4-Dimethoxybenzeneacetic acid, methyl ester 1,2,4-Trimethoxyphenylpropene 3,4,5-Trimethoxyacetophenone Hexadecene 3,4,5-Trimethoxybenzoic acid, methyl ester 3,4-Dimethoxyphenylpropanoic acid, methyl ester Tetradecanoic acid, methyl ester 3,4,5-Trimethoxyphenylacetic acid, methyl ester Pentadecanoic acid, methyl ester Hexadecenoic acid, methyl ester Hexadecanoic acid, methyl ester Hexadecanoic acid Octadecenoic acid, methyl ester Octadecanoic acid, methyl ester Nonadecanoic acid, methyl ester Methylated glucose compound Eicosanoic acid, methyl ester Docosenoic acid, methyl ester Docosanoic acid, methyl ester of vanillyl and syringyl benzoic acids (peaks 21, 22, 27) are present in significant amounts. This indicates that the lignin components are highly degraded and this is consistent with the high oxygen content. Nanny and Ratasuk (2002) have suggested that C4–6 n-alkanedioic acids are products of catechol degradation and microbial fermentation processes. These compounds are present (peaks 1, 2, 5, 8) and indicate some microbial activity is still occurring at the site. GW96 has a similar TIC chromatogram to GW02 and to SRFA, indicating predominantly terrestrially derived aquatic DOC (del Rio et al., 1998), but with two significant differences. The C4–6 n-alkanedioic acids are more intense and peaks corresponding to the methyl es- ters of fatty acids (peaks 29, 31–38, 40–41) are prominent. The increased quantity of n-alkanedioic acids is consistent with higher levels of microbial degradation of aromatic compounds (Nanny and Ratasuk, 2002) while the fatty acids are consistent with the mineralisation of sedimentary organic carbon and higher microbial input. Grasset and Amblès (1998) demonstrated that lipids with an even carbon number preference are released after the enzymatic degradation of humin. These findings are consistent with other data suggesting a higher level of microbial sulfate reduction and/or SOC mineralisation in 1996. The similar distribution of ligninderived compounds in both samples suggests they are derived from the same origin. 4.10. Electrospray ionisation mass spectrometry Negative ion ESI-MS mass spectra for GW02, GW96 [Fig. 5(a) and (b)] and SRFA (not shown) are typical for aquatic fulvic acids and there are no major differences, reflecting the non-specific nature of the technique at this level. The spectra show a distribution of ions at even mass from m/z 100–700, with groups of ions spaced every 14 Da. Ions spaced at 2 and 14 Da have been shown for SRFA, using high resolution mass spectrometry, to result from structural differences of two hydrogen atoms and CH2 units such as those found in aliphatic, alicyclic and olefinic structures (Stenson et al., 2002). High-resolution mass data obtained for a sample obtained in 1998 at this site found similar mass differences (unpublished data) and from this finding it is inferred that similar structural entities exist in the samples used in this study. The spectrum for GW02 differs slightly from those of GW96 and SRFA. GW02 contains ions that were less well resolved and with a more uniform intensity, showing that GW02 does contain a different assemblage of compounds, although the exact nature of the compounds the ESI-MS spectrum represents is still unclear (Stenson et al., 2002; McIntyre et al., 2001). The greater resolution between the groups of ions spaced at 14 Da in GW96 is consistent with a higher content of aliphatic acids, as indicated by 1H NMR spectra and elemental analysis. Tandem mass spectrometric experiments have been used to reveal more specific structural features of fulvic acids (McIntyre et al., 1997, 2002). The product ion spectra for both samples indicate the presence of aliphatic polycarboxylic acids [Fig. 5(c) and (d)]. Major fragments spaced at 44 and 18 Da result from loss of CO2 and H2O from carboxylic acids. Lower mass ions such as 59 Da (acetate) and 73 Da (propanoate) are suggestive of fragmentation of aliphatic structures (Leenheer et al., 2003). The product ion spectrum for GW96 contains low mass ions of greater intensity than GW02, consistent with greater aliphaticity. Neutral loss of 88 Da mass spectra (loss of two CO2 molecules) was C. McIntyre et al. / Organic Geochemistry 36 (2005) 385–397 (a) (c) (e) (b) (d) (f) 395 Fig. 5. Negative ion electrospray ionisation spectra for hydrophobic acids isolated from groundwater: (a) GW02 and (b) GW96, and product ion spectra of m/z 351 for: (c) GW02 and (d) GW96, and neutral loss of 88 Da spectra for: (e) GW02 and (f) GW96. used to indicate compounds containing at least two carboxylic acid groups [Fig. 5(e) and (f)]. Based on the product ion spectra of individual ions (data not shown) the neutral loss spectrum shows aromatic polycarboxylic acids, consistent with the products of lignin degradation, at m/z 171, 181, 197, 209 and aliphatic polycarboxylic acids predominantly at m/z 200–400. These data indicate that the samples contain a mixture of aryl aliphatic polycarboxylic acids and lignin-derived compounds. 5. Conclusions Groundwater hydrophobic acids isolated in 2002 from the Tomago Sand Beds contained predominantly fulvic acids with a high content of aromatic and branched aliphatic polycarboxylic structures. The molecular weight and oxygen content are high and similarities to the surface water fulvic acid SRFA were found, suggesting some input from surface DOC still occurs. Low polydispersity and a low content of carbohydrates and lignin-derived compounds indicated that adsorption and degradation processes are a major control on the DOC composition. Although the origin of the hydrophobic acids could not be conclusively determined, it is hypothesised that compounds derived from precursor terpenoid molecules, selectively degraded and adsorbed during infiltration of surface water recharge, may contribute to this fraction. The high lateral flow rate of the system is likely to have minimised the time of exposure of recharge DOC to natural processes. The sample is also interpreted to have at least a minor contribution of compounds from microbially mineralised SOC. The high content of aromatics in the sample is a concern with regards to drinking water supplies as this could produce higher yields of disinfection byproducts during chlorination. The sample is structurally unique, differing from the current concepts of groundwater hydrophobic acids which suggest they should be low molecular weight aliphatic acids depleted in oxygen content, carboxyl groups and aromaticity. The findings show the site is unique with regard to the structure of the hydrophobic acids and that they are likely to have an important geochemical role with regards to drinking water quality. Limited data obtained for the hydrophobic acids isolated in 1996 indicate that sample has a significant content of compounds with structural characteristics different substantially to those of the 2002 sample. Background data and analysis show DOC levels were elevated at that time and suggest that elevated microbial activity associated with sulfate reduction and mineralisation of SOC led to a greater input of lower molecular weight (possibly branched) aliphatic acids. The observed changes in the structural characteristics and composition of the DOC demonstrate that temporal stability of DOC in control locations cannot be assumed, even over relatively short periods of time. The dynamic nature of the system confirms the groundwater DOC has a significant geochemical role. Dating using 14C, metal binding studies and investigation of SOC at site remain to be done. Determination of the DOC origin (SOC or recharge) would greatly assist in the further interpretation of the structural data. 396 C. McIntyre et al. / Organic Geochemistry 36 (2005) 385–397 Acknowledgments We thank Jerry A. Leenheer of the U.S.G.S. for providing the 13C NMR spectrum and assisting with its interpretation, Dr. Pellegrino Conti at the University of Napoli, Italy for providing the HPSEC data, Dr. Sinan Ali and Dr. Rita Holland at Macquarie University, Australia for the use of their vapour pressure osmometer and Dick Holroyde of the Hunter Water Corporation, Australia for supplying the rainfall data we are also grateful to Simon George, José C. del Rio, Ian Bull and an anonymous reviewer for reviews. Associate Editor—I.D. Bull References Aiken, G.R., Malcolm, R.L., 1987. Molecular weight of aquatic fulvic acids by vapor pressure osmometry. Geochimica et Cosmochimica Acta 51, 2177–2184. Alberts, J.J., Filip, Z., Hertkorn, N., 1992. Fulvic and humic acids isolated from groundwater: compositional characteristics and cation binding. Journal of Contaminant Hydrology 11, 317–330. Artinger, R., Backau, G., Geyer, S., Fritz, P., Wolf, M., Kim, J.I., 2000. Characterization of groundwater humic substances: influence of sedimentary organic carbon. Applied Geochemistry 15, 97–166. Averett, R.C., Leenheer, J.A., McKnight, D.M., Thorn, K.A. (Eds.), 1994. Humic substances in the Suwannee River, Georgia: Interactions, properties and proposed structures. US Geological Survey Water-Supply Paper 2373, US Geological Survey, Denver, CO. Backau, G., Artinger, R., Geyer, S., Wolf, M., Fritz, P., Kim, J.I., 2000. Groundwater in-situ generation of aquatic humic and fulvic acids and the mineralization of sedimentary organic carbon. Applied Geochemistry 15, 819–832. Calace, N., Liberatori, A., Petronio, B.M., Pietroletti, M., 2001. Characteristics of different molecular weight fractions of organic matter in landfill leachate and role in soil sorption of heavy metals. Environmental Pollution 113, 331–339. Chen, Y., Senesi, N., Schnitzer, M., 1977. Information provided on humic substances by E4/E6 ratios. Soil Science Society of America Journal 41, 352–358. Chin, Y.-P., Aiken, G.R., OÕLoughlin, E., 1994. Molecular weight, polydispersity and spectroscopic properties, of aquatic humic substances. Environmental Science and Technology 28, 1853–1858. Christensen, J.B., Jensen, D., Gron, C., Filip, Z., Cristensen, T.H., 1998. Characterization of the dissolved organic carbon in landfill leachate-polluted groundwater. Water Research 32, 125–135. Clifford, D.J., Carson, D.M., McKinney, D.E., Bortiatynski, P.G., Hatcher, P.G., 1995. A new rapid technique for the characterization of lignin in vascular plants: thermochemolysis with tetramethylammonium hydroxide (TMAH). Organic Geochemistry 23, 169–175. Conte, P., Piccolo, A., 1999. Conformational arrangement of dissolved humic substances: influence of solution composition on association of humic molecules. Environmental Science and Technology 33, 1682–1690. Cozzolino, A., Conte, P., Piccolo, A., 2001. Conformational changes of humic substances induced by some hydroxy-, keto- and sulfonic acids. Soil Biology and Biochemistry 33, 563–571. Cronan, C.S., Aiken, G.R., 1985. Chemistry and transport of soluble humic substances in forested watersheds of the Adirondack Park, New York. Geochimica et Cosmochimica Acta 49, 1697–1705. David, M.B., Vance, G.F., 1991. Chemical character and origin of organic acids in streams and seepage lakes of central Maine. Biogeochemistry 12, 17–41. del Rio, J.C., McKinney, D.E., Knicker, H., Nanny, M.A., Minard, R.D., Hatcher, P.G., 1998. Structural characterization of bio- and geo-macromolecules by off-line thermochemolysis with tetramethylammonium hydroxide. Journal of Chromatography A 823, 433–448. Frimmel, F.H., Abbt-Braun, G., 1999. Basic characterization of reference NOM from central Europe – similarities and differences. Environment International 25, 191–207. Grasset, L., Ambles, A., 1998. Aliphatic lipids released from a soil humin after enzymatic degradation of cellulose. Organic Geochemistry 29, 893–897. Gron, C., Wassenaar, L., Krog, M., 1996. Origin and structures of groundwater humic substances from three Danish aquifers. Environment International 22, 519–534. Guggenberger, G., Zech, W., Schulten, H.R., 1994. Formation and mobilization pathways of dissolved organic matter: evidence from chemical structural studies of organic matter fractions in acid forest floor solutions. Organic Geochemistry 21, 51–66. Hatcher, P.G., Rowan, R., Mattingly, M.A., 1979. 1H and 13C NMR of marine humic acids. Organic Geochemistry 2, 77–85. Koivusalo, M., Vartiaanen, T., 1997. Drinking water chlorination byproducts and cancer. Reviews on Environmental Health 12, 81–90. Kim, J.I., Backau, G., Li, G.H., Duschner, H., Psarros, N., 1990. Characterization of humic acids and fulvic acids from Gorleben groundwater. FreseniusÕ Journal of Analytical Chemistry 338, 245–252. Koopal, K.K., van Riemsdijk, W.H., Kinniburgh, D.G., 2001. Humic matter and contaminants: general aspects and modeling metal ion behavior. Pure Applied Chemistry 73, 2005–2016. Kumke, M.U., Zwiener, C., Abbt-Braun, G., Frimmel, F.H., 1999. Spectroscopic characterization of fulvic acid fractions of a contaminated groundwater. Acta Hydrochimica Hydrobiologica 27, 409–415. Leenheer, J.A., Brown, G.K., MacCarthy, P., Cabaniss, S.E., 1998. Models of metal binding structures in fulvic acid from the Suwannee River, Georgia. Environmental Science and Technology 32, 2410–2416. Leenheer, J.A., Rostad, C.E., Barber, L.B., Schroeder, R., Anders, R., Davisson, M.L., 2001. Nature and chlorine reactivity of organic constituents from reclaimed water in groundwater, Los Angeles County, California. Environmental Science and Technology 35, 3869–3876. C. McIntyre et al. / Organic Geochemistry 36 (2005) 385–397 Leenheer, J.A., Nanny, M.A., McIntyre, C., 2003. Terpenoids as major precursors of dissolved organic matter in landfill leachates, surface water, and groundwater. Environmental Science and Technology 37, 2323–2331. Leinweber, P., Schulten, H.R., Kalbitz, K., Meissner, R., Jancke, H., 2001. Fulvic acid composition in degraded fenlands. Journal of Plant Nutrition and Soil Science 164, 371–379. Liu, M., Mao, X., Ye, C., Huang, H., Nicholson, J.K., Lindon, J.C., 1998. Improved WATERGATE pulse sequence for solvent suppression in NMR spectroscopy. Journal of Magnetic Resonance 132, 125–129. Ma, H., Allen, H.E., Yin, Y., 2001. Characterization of isolated fractions of dissolved organic matter from natural waters and a wastewater effluent. Water Research 35, 985–996. Malcolm, R.L., 1990. The uniqueness of humic substances in each of soil, stream and marine environments. Analytica Chimica Acta 232, 19–30. Massmann, G., Tichomirowa, M., Merz, C., Pekdeger, A., 2003. Sulfide oxidation and sulfate reduction in a shallow groundwater system (Oderbruch Aquifer, Germany). Journal of Hydrology 278, 231–243. McIntyre, C., Batts, B.D., Jardine, D.R., 1997. Electrospray mass spectrometry of groundwater organic acids. Journal of Mass Spectrometry 32, 328–330. McIntyre, C., Jardine, D., McRae, C., 2001. Electrospray mass spectrometry of aquatic fulvic acids. Rapid Communications in Mass Spectrometry 15, 1974–1975. McIntyre, C., McRae, C., Jardine, D., Batts, B.D., 2002. Identification of compound classes in soil and peat fulvic acids as observed by electrospray ionization tandem mass spectrometry. Rapid Communications in Mass Spectrometry 16, 1604–1609. Nanny, M.A., Ratasuk, N., 2002. Characterization and comparison of hydrophobic neutral and hydrophobic acid dissolved organic carbon isolated from three municipal landfill leachates. Water Research 36, 1572–1584. Nikolaou, A.D., Lekkas, T.D., 2001. The role of natural organic matter during formation of chlorination by-products: a review. Acta Hydrochimica Hydrobiologica 29, 63– 77. Nissenbaum, A., Kaplan, I.R., 1972. Chemical and isotopic evidence for the in situ origin of marine humic substances. Limnology and Oceanography 17, 570–582. OÕLoughlin, E., Chin, Y.-P., 2001. Effect of detector wavelength on the determination of the molecular weight of humic substances by high-pressure size exclusion chromatography. Water Research 35, 333–338. Pettersson, C., Arsenie, I., Ephraim, J., Boren, H., Allard, B., 1989. Properties of fulvic acids from deep groundwaters. The Science of the Total Environment 81/82, 287–296. Pettersson, C., Ephraim, J., Allard, B., 1994. On the composition and properties of humic substances isolated from deep groundwater and surface water. Organic Geochemistry 21, 443–451. 397 Peuravuori, J., Pihlaja, K., 1997. Molecular size and spectroscopic properties of aquatic humic substances. Analytica Chimica Acta 337, 133–149. Prosser, I.P., Roseby, S.J., 1995. A chronosequence of rapid leaching of mixed podzol soil materials following sand mining. Geoderma 64, 297–308. Smith, B.N., Epstein, S., 1971. Two categories of 13C:12C ratios for higher plants. Plant Physiology 47, 380–384. Stenson, A.C., Landing, W.M., Marshall, A.G., Cooper, S.R., 2002. Ionization and fragmentation of humic substances in electrospray ionization Fourier transform-ion cyclotron resonance mass spectrometry. Analytical Chemistry 74, 4397–4409. Thorn, K.A., Aiken, G.R., 1998. Biodegradation of crude oil into nonvolatile organic acids in a contaminated aquifer near Bemidji, Minnesota. Organic Geochemistry 29, 930– 931. Thurman, E.M., Malcolm, R.L., 1981. Preparative isolation of aquatic humic substances. Environmental Science and Technology 15, 463–466. Thurman, E.M., 1985a. Humic substances in groundwater. In: Aiken, G.R., McKnight, D.M., Wershaw, R.L., MacCarthy, P. (Eds.), Humic Substances in Soil, Sediment, and Water: Geochemistry, Isolation, and Characterization. Wiley, New York, pp. 87–103. Thurman, E.M., 1985b. Organic geochemistry of natural waters. Nijhoff and Junk Publishers, Dordrecht, pp. 14– 15, 296. Johnson-Logan, L.R., Broshears, R.E., Klaine, J., 1992. Partitioning behavior and the mobility of chlordane in water. Environmental Science and Technology 26, 2234– 2239. Wassenaar, L., Aravena, R., Fritz, P., Barker, J.F., 1990. Isotopic composition (13C, 14C, 2H) and geochemistry of aquatic humic substances from groundwater. Organic Geochemistry 15, 383–396. Wassenaar, L.I., Aravena, R., Fritz, P., Barker, J.F., 1991. Controls on the transport and carbon isotopic composition of dissolved organic carbon in a shallow groundwater system, Central Ontario, Canada. Chemical Geology 87, 39–57. Wilson, M.A., 1987. NMR Techniques and Applications in Geochemistry and Soil Chemistry. Permagon Press, Oxford. Wong, S., Hanna, J.A., King, S., Carroll, T.J., Eldridge, R.J., Dixon, D.R., Bolto, B.A., Hesse, S., Abbt-Braun, G., Frimmel, F.H., 2002. Fractionation of natural organic matter in drinking water and characterization by 13C crosspolarization magic-angle spinning NMR spectroscopy and size exclusion chromatography. Environmental Science and Technology 36, 3497–3503. Zwiener, C., Kumke, M.U., Abbt-Braun, G., Frimmel, F.H., 1999. Adsorbed and bound residues in fulvic acid fractions of a contaminated groundwater: isolation, chromatographic and spectroscopic characterization. Acta Hydrochimica Hydrobiologica 27, 208–213.