Investigation of seawater and terrestrial humic substances with carbon-13 and proton nuclear magnetic resonance

Geochimica et Cosmochimica

Acta. 1976,

Vol. 40, pp. II09 to

I114. Pergamon Press. Printed in Great Britain

Investigation of seawater and terrestrial humic substances with carbon-13 and proton nuclear magnetic resonance* DANIEL H. STUERMERt and JAMESR. PAYNE~ Woods Hole Oceanographic Institution, Woods Hole, Massachusetts 02543, U.S.A. (Received

13 June

1975; accepted

in revisrd form

25 Frhruary

1976)

Abstract-Seawater and terrestrial fulvic acids have been investigated with carbon-13 and proton nuclear magnetic resonance and by i.r. spectroscopy. The structural differences observed seem to result mainly from the low abundance of aromatic precursors (i.e. lignin) in seawater as compared to the terrestrial environment.

INTRODUCTION

HUMIC substances constitute between 6 and 30% of the total organic matter in seawater (STUERMER, 1975; STUERMERand HARVEY, 1974; KERR and QUINN, 1975; KHAYLOV,1968; KALLE, 1966; SIEBURTHand JENSEN,1968). Studies of these humic substances in seawater and those in lacustrine and marine sediments (NISSENBAUM and KAPLAN, 1972; RASHIDand KING, 1971; ISHIWATARI,1971) provide evidence that the marine and lacustrine environments produce humic substances which differ from those in terrestrial environments (SCHNITZER and KHAN, 1972; KONONOVA,1966). Defining these differences in terms the effects of different organic carbon sources and environmental parameters on the production of humic substances. This paper presents carbon-13 and proton nuclear magnetic resonance ( 13C-NMR; ‘H-NMR) spectra, i.r. absorption spectra and elemental composition data on Sargasso Sea surface water fulvic acid. The data are compared with existing data and 13C-NMR spectra obtained on terrestrial fulvic acid and interpreted to provide structural comparison between terrestrial and seawater fulvic acids.

EXPERIMENTAL The marine fulvic acid sample was isolated from seawater collected in the northwestern Sargasso Sea at stations listed in Table 1. The method used is reported in detail elsewhere (STUERMER, 1975). Briefly, organic matter was adsorbed from pH 2 seawater on Amberlite XAD-2 resin (ROHM & HAAS, 1972; STUERMER and HARVEY, 1974; RILEY and TAYLOR, 1969; MANT~URA and RILEY, 1975). The resin was rinsed with distilled water to remove salt,

and humic substances were eluted with 0.5 N NH,OH. Following lyophilization, fulvic acid was dissolved in 0.01 N HCl. extracted with CH,C12 to remove lipids, and recovered from the aqueous solution by lyophilization. The soil fulvic acid was isolated from the B,, horizon of a podzol (BUCKMAN and BRADY, 1960) in Falmouth, Massachusetts, U.S.A. The sample was extracted with 0.4 N NaOH under a nitrogen atmosphere for 24 hr. The extract was acidified to pH 2.0 and filtered through Reeve Angel Grade 934 AH glass filter. The filtrate (pH 2; 0.4 M NaCl) was passed through Amberlite XAD-2 resin and fulvic acid was recovered from the resin following the same procedure as used for the seawater sample. A sample of the seawater fulvic acid (49.8 mg) was suspended in dry methanol and methylated with diazomethane in CH,Cl, for 30 min. After drying under a stream of nitrogen gas, the methylene chloride soluble products were recovered and the residual material was again treated with diazomethane. The methylation of the residual material was repeated twice, yielding 51% of the fulvic acid (25.3 mg) soluble in CH,Cl,. Analyses for carbon, hydrogen, nitrogen and ash were carried out by Galbraith Laboratories, Inc. (Knoxville, Tennessee). Oxygen was calculated by difference. Infra-red (i.r.) spectra were obtained on a Perkin-Elmer model 337 grating i.r. spectrophotometer with a beam condensor. Continuous wave ‘H-NMR spectra were obtained at 35°C with a HitachiiPerkin-Elmer R-20B 6OMHz NMR instrument for the seawater fulvic acid and a Varian A-60 NMR instrument for the methylated seawater fulvic acid. The i3C-NMR spectra were obtained on a Biuker HFX-90-18 NMR spectrometer interfaced with a Digilab FTS/NMR-3 computer. This spectrometer has been modified to achieve multi-nuclei capability (TRAFICANTE, SIMMS and MULCAY, 1974) and was operated in the Fourer transform mode with a fluorine field/frequency lock and full proton decoupling (1.2 kHz bandwidth). A 10 mm (o.d.) coaxial NMR tube with an internal tube of 5 mm (o.d.) Table 1. Station data-279 mg of fulvic acid was isolated from the combined surface water from the stations listed

*Woods Hole Oceanographic Institution Contribution No. 3583. t Present address: Institute of Geophysics and Planetary Physics, Universitv of California at Los Angeles. Los I Angeles, California 90024, U.S.A. $ Present address: Bodega Marine Laboratory, University of California at Berkeley, P.O. Box 247, Bodega Bay, California 94923, U.S.A. 1109

station Location

Yolune(liters)

Salinity

(O/so)

TmpeMt”M PC

1

D. H. STUEKMEK

1110

and J. R. PAW CARBONYL

diameter was used for the 13C-NMR experiments. The fulvie acid samples were placed in the cavity between the tubes. and a saturated aqueous KF lock solution (containing 2% dioxane internal standard) was placed in the internal tube. The spectra were obtained at 44°C. and chemical shifts are reported in ppm downfield from TMS.

AROMATIC

POLYHYDROXYL

ALIPHATIC

A--

Immediately before each ‘“C-NMR study, the fulvic acid sample was dissolved in 0.01 N HCI and centrifuged (12,000 rev/min. 20 min, 0°C) to remove any suspended material. The molecular weight range of the seawater fulvic acid was determined to be 200-5000 with 81% less than 1500 (STIJERMER and HARVEY, 1974). A molecular weight range of 175-3600 was determined by several methods for a podzol fulvic acid from the B,, horizon by S~HNITZER and SKINNER (1968); a molecular weight range of 50&10,000 was indicated for a B,, horizon podzol fulvic acid by LEVESQUE(1972). The soil fulvic acid used in this study is assumed to have a molecular weight range between 175 and 10.000.

RESULTS The elemental composition and ash contents of the seawater and terrestrial fulvic acid samples are presented in Table 2. The major differences are the relatively lower oxygen content, higher nitrogen content, and the higher H/C ratio of the seawater sample. The 13C-NMR spectra of the seawater and terrestrial fulvic acid samples are presented in Figs. 1 and 2, respectively. The broad resonances between 170 spectra of both the and 185 ppm in the 13C-NMR _ seawater and terrestrial fulvic acid samples are characteristic of ester, carboxylic acid and amide groups. The broad resonances in the 60-100 ppm range of both the seawater and terrestrial spectra are characteristic of carbon singly bonded to oxygen or nitrogen. The sharp resonance at 67.4ppm in both spectra is from the internal standard, dioxane. Significant differences exist between the seawater and terrestrial fulvic acid spectra. The seawater fulvic acid spectrum shows more predominant resonances in the 10-60ppm region characteristic of aliphatic moieties. Resonances in the 1l&l60 ppm region of the spectra characteristic of aromatic, heteroaromatic and olefinic constituents are slightly more predominant in the terrestrial fulvic acid spectrum. The ‘H-NMR spectra of the seawater fulvic acid and of the methylated seawater fulvic acid are shown in Figs. 3 and 4, respectively. The fulvic acid NMR spectrum shows broad resonances in the 1.0-1.7 ppm range characteristic of aliphatic protons, in the 1.7-2.5 ppm range characteristic of protons on carbons adjacent to functional groups such as carbonyls, aromatics or double bonds, and in the 7.3-7.9 ppm region characteristic of aromatic protons. The relative areas of these resonances are 15: lO:l, respectively.

t”““““““““““1 200

150 PPM

2. Elemental

samplesoui-ce XC seawater TWPeSWi3.l

50.0 46.7

%H 6.8 4.6

composition %N

40 (by

(ash-free

difference)

50

TMS

Fig. 1. Fully proton-decoupled 22.6MHz Fourier transform 13C-NMR spectrum of Sargasso Sea fulvic acid in 0.01 N HCl (48 mg/ml) at 44°C. Chemical shifts are given in ppm downfield from tetramethylsilane (2% dioxane in-

ternal standard, d, at 67.4 ppm). This spectrum is the result of I1 3,789 pulses using a 90” pulse and a total pulse delay and acquisition time of 0.74 sec. The large peak at 5.2ppm IS due to HDO in the solvent from hydrogen exchange. The methanol internal standard is observed at 3.45 ppm. The ‘H-NMR spectrum of the methylated fulvic acid shows similar characteristics in the 1.0-2.5 ppm region. However, a new resonance is observed between 3.4 and 3.9 ppm from protons on methoxy groups of methyl esters. The minor aromatic resonance is not observed in this spectrum, perhaps because of the smaller sample size or the lower resolution of the Varian instrument. The resonance at 5.7 ppm is from a small amount of CHDCl, in the solvent. The area ratio of the resonances between 1.0 and 2.5 ppm and between 3.4 and 3.9 ppm is 4.8-1.0. POLYHYOROXYL

AROMATIC

CARBO#YL

ALIPHATIC

r-\/---n--\-

/

d

I I

III’ 200

I

III

II

1 100

150 PPM

Table

100 FROM

FROM

II

1

I

I

t

I”

50

TMS

basis)

9: Ash

H,C Ratio

6.4

36.9

3.37

1.61

0.5

44.3

2.35

1.15

Fig. 2. Fully proton-decoupled 22.6 MHz Fourier Transform ‘%Z-NMR spectrum of terrestrial fulvic acid in 0.01 N HCl (44mg/ml) at 44°C. Chemical shifts are given as in Fig. I This spectrum is the result of 88,307 pulses using

a 90” pulse and a total pulse delay and acquisition time at 0.74 sec.

Investigation

I

I

IO.0

I

I

9.0

I

of seawater

I

I

6.0

I

7.0

I

and terrestrial

I

6.0

I

PPM

FROM

I

I

5.0

4.0

humic

I

substances

I

I

I

1111

I

2.0

3.0

I

I

1.0

I 0.0

TMS

Fig. 3. Continuous wave 60MHz ‘H-NMR spectrum of Sargasso Sea fulvic acid in D,O (lOOmg/ml) at 35°C. Chemical shifts are given in ppm downfield from tetramethylsilane (methanol internal standard at 3.45 ppm)

I

I

10.0

I

I

9.0

I

I

8.0

I 7.0

I

I

I

6.0

I

I

5.0 PPM

FROM

Ii 4.0

I

I

3.0

I 2.0

I

I 1.0

I

I 0.0

TMS

Fig. 4. Continuous wave 60 MHz ‘H-NMR spectrum of methylated Sargasso Sea fulvic acid in CD,Cl, (6Omg/mI) at 35°C. Chemical shifts given in ppm downfield from tetramethylsilane (TMS internal standard). Microns

Microns

0

Frequency,

Fig. 5. Infra-red

spectra

cm-’

of Sargasso Sea fulvic acid in a KBr pellet (a) and CH,Cl,-soluble Sargasso Sea fulvic acid between NaCl plates (neat) (b).

methylated

D. H.

III2

STLEKMER

Methylation of the marine fulvic acid results in a 2&30cm~’ shift to higher frequency in the carbonyl absorption and a reduction and sharpening of O-H stretching absorption in the i.r. spectra shown in Fig. 5.

DISCUSSION

From the molecular weight range of the fulvic acid samples and ‘“C-NMR studies of other macromolecules with similar molecular weights [specificially cholesterol (AI.LERHAND c’f ul.. 1971a) and low molecular weight polypeptides (ALLEKHANDand KOMOKOSKI. 1973: DESLA~~RIEKSet ill.. 1974: KI:IM ct (I/.. 1973; ALLERHANIIand OLIIFI~LLX 1973) and proteins (ALLERHANII er ol.. 1971b: GLUWK~ et d., 1972; ALLUWAND cf (II.. 1973; HUNKAPILLER et al.. 1973)], it was predicted that differential saturation of specific carbons would not be a problem with the experimental parameters used in this study (Figs. I and 2). Test spectra of 0.1 M cholesterol in d,-benzene were obtained using the same instrumental conditions. supporting this prediction. Furthermore. compared to protonated or aliphatic carbons. the carbonyl carbons in most macromolecules have relatively longer spin lattice relaxation times and are, therefore. mre subject to saturation (ALLERHANI)and KOMOROSKI,1973). The observation of the predominant carbonyl resonances in the fulvic acid spectra demonstrates that saturation is not a significant problem in these studies. Nuclear Overhauser effects can complicate interpretation of 13C-NMR spectra, particularly if comparisons between peak areas in a given sample are to be made (STOTHEKS. 1972). However, considering the molecular weight ranges of the seawater and terrestrial fulvic acid samples, we believe that similar nuclear Overhauser effects are occurring in both samples (DOIXXU:LL. (‘I crl.. 1972: ALL~RHANI> and OI_I)~IELI), 1973) and thus. the relative peak areas between the spectra of seawater and terrestrial fulvic acid can be compared. Several factors can contribute to the broad and poorly-resolved resonances observed in the marine and terrestrial fulvic acid spectra. These are the extreme molecular complexity of the samples; the presence of paramagnetic metal ions and free radicals (STMLINK and TOLUN. 1967). which can interact with ‘?I nuclei causing line broadening (BECKER, 1969) or changes in chemical shifts (LEVY and NELSON. 1972; STormRs. 1972; HATCH and KREIIXK, 1971): and molecular weight dependent rotational correlation times (DOLXIRI:LLclt al., 1972; ALLERHANLIand HAILSTONI.. 1972; BECKER. 1969). The phenomenon of line broadening due to molecular complexity has been observed in other macromolecules such as proteins and nucicic acids (PAYNE. 1974; LA~UTEKBUR.1970; CHIEV and BRANDTS. 1971 ; FRE.I:DMAN PI al.. 1973: AI~L~RHAXD L’I LI/.. 1973;

GLWHKO

(‘I ul..

KOMOROXI and ALLIRHAND. 1972). However.

1972;

even in

and J. R. PAbNt

proteins with molecular weights as high as 310,000 it is still possible to observe well-resolved resonances from many of the side-chain aliphatic and aromatic carbons (PAYNI.. 1974) because of the limited number of amino acids present as repeating subunits. Therefore. it is quite probable that the lack of any wellresolved peaks in these spectra is mainly attributable to the extreme complexity and non-repeating subunit structure of the samples. although the presence of paramagnetic metals and free radicals may be a contributing factor. Line broadening due to molecular weight dependent rotational correlation times is unlikely considering ‘“C-NMR relaxation studies on molecules with molecular weights similar to the fulvic acid samples studied here (ALLI~RHAND1’1 ~1.. 197la; ALLEKHAND and KOMOKOSKI, 1973: DESLAUKWRS01 (I[.. 1974: K~IM (‘1 cd.. 1973: AI.I.ERHANI) and OLDf-IE:LD. 197.1). The I”(‘-NMR spectrum of seawater fulvic acid shows a much greater abundance of resonances from aliphatic carbons than are present in the spectrum of soil fulvic acid. Conversely. the resonances from aromatic, heteroaromatic and olefinic carbons, are more prominent in the soil fulvic acid spectrum. The more aliphatic character of the seawater fulvic acid is consistent with the higher H/C’ ratio (Table 2). Also, the main features of the ‘H-NMR spectra of seawater fulvic acid (Figs. 3 and 4) are the large broad aliphatic resonances between 1.0 and 2.5 ppm and the lack of large aromatic resonances between 7.0 and 9.0ppm. These ‘H-NMR data are in support of the highly-aliphatic character of seawater fulvic acid indicated by the 13C-NMR data. The less aromatic. more aliphatic character of the seawater fulvic acid may reflect the lack of abundant aromatic precursors in the marine environment. Lignin is believed to be an important component in the formation of soil humic substances (HURST and BU0.s. 1967). Lignin is not abundant in marine plants except for some benthic and marsh grasses (MOORE, 1969: Lto and ~ARC;HOOR&, 1970: GARDNER and MENLEL, 1974). Therefore. this aromatic precursor of terrestrial humic substances is not available in the marine environment (GAIWNER and MENZEL, 1974; H~DG~.s. 1973). The presence of abundant carbonyl oxygen in both fulvic acid samples is demonstrated by the 13C-NMR resonances between I70 and I85 ppm. The lack of significant resonances indicative of aldehydes and ketones in the 190-210 ppm region in both lJC-NMR spectra suggests that the major carbonyl groups in the fulvic acid samples are acids, esters and amides. The presence of carboxylic acids in the seawater fulvic acids is indicated by the shift in the i.r. carbonyl absorbance from 1700 to I730 cm- ’ due to the production of methyl esters upon methylation of the sample. The accompanying reduction and sharpening of the OH stretching absorbance in the i.r. spectrum (3W36OOcm ‘) also suggests the formation of methoxyl groups from the acidic OH groups of acids and

Investigation

of seawater

and terrestrial

perhaps phenols. In addition, the appearance of the new ‘H-NMR resonance between 3.4 and 3.9ppm upon methylation of the fulvic acid (Fig. 4) is evidence of the formation of methyl esters from carboxylic acids. The resonances in the 6(rlOOppm region of the 13C-NMR spectra of both fulvic acid samples suggests the presence of carbons singly bonded to oxygen or nitrogen. The intense i.r. absorption in the 3OW3500 cm- ’ region, which persists after methylation. suggests that these heteroatoms are present mainly as hydroxyl or amino groups. The humic substances in seawater must be formed from the prescursors present such as amino acids, carbohydrates, lipids and pigments. The formation of humic substances by processes analogous to the browning reaction has been suggested by several investigators (KALLE, 1966; NISSENBAUM, 1974; JACKSON, 1975). The incorporation of marine lipids and pigments into the products of the Browning reaction could account for the structural features indicated by the present work. Acknowledgements-The authors thank Dr. D. D. TRAFICANTE and J. A. SIMMS at Massachusetts Institute of Technology for their assistance in collection of the 13C-NMR spectra. We would also like to thank Dr. G. R. HARVEY for his fruitful advice and continued interest in this work. J. R. PAYNE acknowledges the Woods Hole Oceanographic Institution for support as a Postdoctoral Scholar. This project was supported by the National Science Foundation (Doctoral Dissertation Grant GA-36631) and the Arthur Vining Davis Foundations.

REFERENCES ALLERHAND A., CHILDERS R. F. and OLDFIELD E. (1973) Natural abundance carbon-13 nuclear magnetic resonance studies on 20 mm sample tubes. Observations of numerous single-carbon resonances of hen egg-white lysozyme. Biochemistry 12. 1335-1341. ALLERHAND A., DODDRELL D. and KOMOROSKI R. A. (1971a) Natural abundance carbon-13 partially relaxed Fourier transform magnetic resonance spectra of complex molecules. J. Chem. Phys. 55. 189. ALLERHAND A., DODDRELL D., GLUSHKO V., COCHRAN D. W., WENKERT E., LAWSON P. J. and GURD F. R. N. (197lb) Conformation and segmental motion of native and denatured Ribonuclease A in solution. Application of natural-abundance carbon-13 partially relaxed Fourier transform nuclear magnetic resonance. J. Amer. Chem. Sot. 93. 544-546. ALLERHAND A. and HAILSTONE R. K. (1972) Carbon-13 Fourier transform nuclear magnetic resonance. X. Effect of molecular weight on 13C spin lattice relaxation times of polystyrene in solution. J. Chem. Phys. 56. 3718. ALLERHAND A. and KOMOROSKI R. A. (1973) Study of internal rotations in Gramicidin S by means of carbon-13 spin-lattice relaxation measurements. J. Amer. Chem. Sot. 95. 8228-8231. ALLERHAND A. and OLDFIELD E. (1973) Determination of rotational mobilities of backbone and side-chain carbons of poly(y-benzyl L-glutamate) in the helical and randomcoil states from measurements of carbon-13 relaxation times and nuclear Overhauser enhancements. Biochemistry 12. 3428-3433.

humic

substances

1113

BECKER E. D. (1969) High Resolution NMR. Theory and Appkations, pp. l-310. Academic Press. BUCKMAN H. 0. and BRADY N. C. (1960) The Nature and Properties of Soil, 160, p. 309. MacMillan. CHIEN J. C. W. and BRANDT~ J. F. (1971) Natural abundance Fourier transform of carbon-13 nuclear magnetic resonance spectra of lysozyme. Nature New Biol. 230. 209. DESLAURIERSR., SMITH I. C. P. and WALTER R. (1974) Conformational flexibility of the neurohypophyseal hormones oxytocin and lysine-vasopressin. A carbon-13 spin-lattice relaxation study of backbone and side chains. .I. Amer. Chem. Sot. 96. 2289-2291. D~DRELL D., GLUSHKO V. and ALLERHAND A. (1972) Theory of nuclear Overhauser enhancement and 13C-‘H dipolar relaxation in proton-decoupled carbon- I3 nmr spectra of macromolecules. J. Chem. Phys. 56. 3683-3689. FREEDMAN M. H., LYERLA J. R., JR., CHA~KEN I. M. and COHEN J. S. (1973) Carbon-13 nuclear-magneticresonance studies on selected amino acids, peptides, and proteins. Eur. J. Biochem. 32. 215-226. GARDNER W. S. and MENZEL D. W. (1974) Phenolic alkehydes as indicators of terrestrial derived organic matter in the sea. Geochim. Cosmochim. Acta 38. 813-822. GLUSHKO V., LAWSON P. J. and GURD F. R. N. (1972) Conformational states of bovine pancreatic ribonuclease A observed by normal and partially relaxed carbon-13 nuclear magnetic resonance. J. Biol. Chem. 247. 3176-3185. HATCH G. F. and KREILICK R. (1971) 13C-NMR spectra of nitroxide radicals. Chem. Phys. Lett. IO. 49&492. HEDGES J. (1973) Lignin compounds as indicators of terrestrial organic matter in marine sediments. Annual Report to the Director, Geophys. Lab., Carnegie Inst. Wash. Yearb. 58 l-590. HUNKAPILLER M. W., SMALLCOMBES. H.. WHITAKER D. R. and RICHARDS J. H. (1973) Carbon nuclear magnetic resonance studies of the histidine residue in *-lytic protease. Implications for the caralytic mechanism of serine proteases. Biochemistry 12. 47324743. HURST H. M. and BURGES N. (1967) Lignin and humic acids, In Soil Biochemistry, (editors A. D. McLaren and G. H. Peterson), Chapter II. Marcel Dekker. ISHIWATARI R. (1971) Organic polymers in recent sediments--chemical nature and fate in geological environment. D.Sc. Thesis, Tokyo Metropolitan University, Tokyo. JACKSON T. A. (1975) Humic matter in natural waters and sediments. Soil Sci. 119. 5664. KALLE K. (1966) The problem of Gelbstoff in the sea. Oceanogr. Mar. Biol. Ann. Rev. 4. 91-104. KEIM P., VIGNA R. A., MARSHALL R. C. and GURD F. R. N. (1973) Carbon-13 nuclear magnetic resonance of pentapeptides of glycine containing central residues of aliphatic amino acids. J. Biol. Chem. 248. 6104-6113. KERR R. A. and QUINN J. G. (1975) Chemical studies on the dissolved organic matter in seawater. I. Isolation and fraction. Deep-Sea Res. 22. 107-116. KHAYLOV K. M. (1968) Dissolved organic marcomolecules in seawater. Geochem. 111t.Sr 497-503. KONONOVA M. M. (1966) Soil Organic Matter. Pergamon Press. KOMOROSKI R. A. and ALLERHAND A. (1972) Natural abundance carbon-13 Fourier transform nuclear magnetic resonance spectra and spin lattice relaxation times of unfractionated yeast transfer-RNA. Proc. Nat! Acad. Sci. U.S.A. 69. 1804. LAUTERBUR P. C. (1970) “C nuclear magnetic resonance spectra of proteins. Appl. Spectrosc. 24. 450. LEO R. F. and BARGHCXIRNE. S. (1970) Phenolic aldehydes: generation from fossils, woods and carbonaceous sediments from oxidative degradation. Science 168. 582-584.

I114

D. H. STIII:KMLKand J. R. PAIU~

Lt.vt.syr,r. M. (1972) Fluorescence and gel filtration of humic compounds. Soil Sci. 113. 346-353. LI-vu G. C. and NELSON G. L. (1972) Curho,~-I 3 lvuc,lru, Muynetic Rrsormce for Organic Chrnlists. pp. I ~222. Wiley-Interscience. MANT~URA R. F. C. and RIL~‘~.J. P. (1975) The analytical concentration of humic substances from natural waters. 4nul. Chij)f. Acru 76. 97~106. MCK)RE L. R. (1969) In Or(lcl/nc Geochrrnrsrry. (editors G. Eglinton and M. T. J. Murphy). Chapter 22. Springer-Verlag. NISSENBAUMA. (1974) The organic geochemistry of marine and terrestrial humic substances: implications of carbon and hydrogen isotope studies. In ildtwrces 111Orqur~ic Geochemistry, (editors B. Tissot and F. Bienner), pp. 3952. Technip. NISSFNBAUM A. and KAI~LAN 1. R. (1972) Chemical and isotopic evidence for the in situ origin of marine humlc substances. Limnol. Owunogr. 17. 570-5X2. PAYNE J. R. (1974) Probing enzyme-coenzyme intcractlons with vitamin B, analogs and carbon-13 nuclear magnetic resonance. Ph.D. Thesis, University of Wisconsin. Madison. pp. I~-396 RASHIf) M. A. and KI~;(; L. H. (1971) Chemical characterization of fractionated humic acids associated with marine sediments. Chml. &ol. 7. 37-43. RILEY J. P. and TAYLOR D. (1969) The analytical conccntration of traces of dissolved organic materials from

seawater with Amberhtc XAD-I resin. Anul. Chirn. Acta 46. 307 309. ROHM & HAAS (196’)) .4\rlhcrlfrc XAD Mucroreticulur .4h.sorhmt.s. Rohm & Haas. Philadelphia. SCHNITZFR M. and KHAN S. U. (1972) Hurnic Substarves it] ihc En~~inmnwrlr. pp. I 727. Marcel Dekker. SCtfNIx1.K M. and SK1NNf.K S. I. M. (1968) Gel filtration 01‘ fulvic acid, a soil humic compound. In Isotopes and Radiutioil iv Soil Or