0444815163 Humic Substances

HUMIC SUBSTANCES IN TERRESTRIAL ECOSYSTEMS

This Page Intentionally Left Blank

Humic Substances in Terrestrial EcosystemsEdited by ALESSANDRO PICCOLO Dipartimento di Scienze Chimico-Agrarie Universitb di Napoli "Federico 11" Via Universitb 100 80055 Portici Italy

1996 ELSEVIER Amsterdam Lausanne New York Oxford Shannon Tokyo

ELSEVIER SCIENCE B.V. Sara Burgerhartstraat 25 P.O. Box 21 1, 1000 AE Amsterdam, The Netherlands

ISBN 0-444-81516-3

0 1996 ELSEVIER SCIENCE B.V. All rights reserved.

No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise without the prior written permission of the publisher, Elsevier Science B.V., Copyright & Permissions Department, P.O. Box 521, 1000 AM Amsterdam, The Netherlands. Special regulations for readers in the U.S.A.: This publication has been registered with the Copyright Clearance Center Inc. (CCC), 222 Rosewood Drive, Danvers, MA 01923. Information can be obtained from the CCC about conditions under which photocopies of parts of this publication may be made in the U.S.A. All other copyright questions, including photocopying outside the U.S.A., should be referred to the copyright owner, Elsevier Science B.V., unless otherwise specified. No responsibility is assumed by the publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. This book is printed on acid-free paper. Printed in The Netherlands

PrefaceThe idea for this book stems from a meeting sponsored by the European Union, organized by N. van Breemen, and held in Doorweerth at the end of 1991. At this meeting a large number of European scientists discussed the different issues related to the accumulation and decomposition of organic matter in terrestrial ecosystems. One of the objectives was to gather scientists from various disciplines (biologists, chemists, ecologists, agriculturalists) to pool their different disciplinary approaches and come up with a common perspective for future research on soil organic matter. Despite the great communication effort exerted by the participants, each discipline had the tendency to apply a particular viewpoint to the concept of soil organic matter and its ecological role. It was clear that further harmonization should have been achieved on the definition, functions, and dynamics of humus in terrestrial ecosystems. The term 'soil organic matter' is generally used to represent the organic constituents in soil, excluding undecayed plant and animal tissue, their partial decomposition products, and the soil biomass. Soil organic matter is recognized to be generally comprised of humic substances and of non-humic substances. The latter material includes discrete compounds of known chemical structure such as polysaccharides and sugars, proteins and amino acids, fats, simple organic acids, and so on. On the other hand, humic substances ~ the largest constituents of soil organic matter ~ are heterogeneous compounds which are still undefined in their chemical identity. It is this major fraction that eventually controls soil biological activity and determines the overall ecological functions of organic matter in soils. The important role of soil humic substances in preserving the ecology of our planet is recognized by scientists who see in the most stable part of soil organic matter not only a nutritional reservoir to match the demands of an increasing world population, but also a means of efficiently recycling in soil the growing production of waste biomass in rapidly enlarging urban areas. Humic substances may represent a possible measure to counter the menacing ecological consequences of the greenhouse effect, by functioning as a sink of carbon in the presence of an excessive concentration of atmospheric CO2. Finally, the conservation and enhancement of humic substance content in the soil is perceived by environmental scientists and the general media (Pierre Rognon: Au Maghreb, la rrsistible avancre du desert. In: Le Monde Diplomatique, Fevrier 1995, Paris) as the most incumbent ecological challenge to fight against the socio-economic disaster deriving from existing and progressing soil erosion and soil desertification.

This book thus has the double objective of attempting to give an updated account of the scientific issues relating to the nature and function of humic substances in terrestrial ecosystems, and of addressing concerned ecologists towards an advanced understanding of the fundamental role played by humic substances in the ecological equilibrium. The chapters have been written by specialists in interdisciplinary fields relating to the still unresolved issues of soil humic substances, such as their chemical nature (C. Saiz-Jimenez), their distribution in world soils (F. Andreux), their influence on soil biological activity (H. Insam, P. Nannipieri et al.), their nutritional value (K. Kelley and F.J. Stevenson, Y. Chen, F.J. Zhao et al., J. Magid et al.), and their influence on soil conservation (A. Piccolo). Other chapters are devoted to aspects of growing ecological concern such as the role of humic substances in the physiological stimulation of plant cells (S. Nardi et al.), their impact on the activity of soil organisms (L. Brussaard and N.G. Juma), their influence on the formation and preservation of forest layers (W. Zech et al.), their interaction with inorganic soil particles (J. Cornejo and M.C. Hermosin) and organic chemicals reaching the soil (J. Kozak), their role in biomass recycling (N. Senesi et al.), and their behaviour when dissolved in water (A. Zsolnay). The compilation of these chapters has not been an easy task because of the many commitments of the authors. However, all of them have shown a constant determination in accomplishing this task, and I am obliged and grateful for their participation in this endeavour. My particular appreciation is directed to Elsevier who believed in the project and supported my efforts all the way.

ALESSANDRO PICCOLO February 1996

vii

Contents

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chapter 1. The Chemical Structure of Humic Substances: Recent Advances Cesareo Saiz-Jimenez . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chapter 2. Humus in World Soils Francis Andreux . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chapter 3. Organic Matter Dynamics in Forest Soils of Temperate and Tropical Ecosystems Wolfgang Zech and Georg Guggenberger, with the participation of L. Haumaier, R. Pohhacker, D. Schayer, W. Amelung, A. Miltner, K. Kaiser and F.Ziegler.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chapter 4. Dissolved Humus in Soil Waters Adam Zsolnay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chapter 5. Humus and Soil Conservation Alessandro Piccolo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chapter 6. Microorganisms and Humus in Soils Heribert Insam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chapter 7. Humus and Enzyme Activity . PClo/o Nanniperi, P. Sequi and P. Fusi . . . . . . . . . . . . . . . . . . . . . . Chapter 8. Organisms and Humus in Soils Lijbert Brussaard and N.J. Juma . . . . . . . . . . . . . . . . . . . . . . . . .

v

ix1

45

101

171 225

265293 329

Chapter 9. Biological Activityof Humus Serenella Nardi, G. Concheri and G. Dell 'Agnola . . . . . . . . . . . . . . . . 36 1 Chapter 10. Organic Forms of N in Soil Kenneth R. Kelley and F.J. Stevenson . . . . . . . . . . . . . . . . . . . . . . . 407

Chapter 1 1. Dynamics of Organic Phosphorus in Soils under Natural and Agricultural Ecosystems Jakob Magid, H. Tiessen and L.M. Condron . . . . . . . . . . . . . . . . . . .429 Chapter 12. Soil Organic Sulphur and its Turnover Fang-Jie Zhao, J . Wu and S.P. McGrath . . . . . . . . . . . . . . . . . . . . . 467

viii

Chapter 13. Organic Matter Reactions Involving Micronutrients in Soils and their Effect on Plants Yona Chen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chapter 14. Humic-like Substances in Organic Amendments and Effects on Native Humic Substances Nicola Senesi, T.M. Miano and G. Brunetti . . . . . . . . . . . . . . . . . . . . Chapter 15. Interactions of Humic Substances and Soil Clays Juan Cornejo and M. C. Hermosin . . . . . . . . . . . . . . . . . . . . . . . . Chapter 16. Soil Organic Matter as a Factor Influencing the Fate of Organic Chemicals in the Soil Environment Josef Kozak . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Subject Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

507

531

595

625

665

ix

Contributors

F. Andreux Universit6 de Bourgogne, Centre des Sciences de la Terre 6, Boulevard Gabriel, F-21000 Dijon, France G. Brunetti Istituto di Chimica Agraria, Universit?~ di Bari, Via Amendola 165/A, 70126 Bari, Italy L. Brussaard Department of Terrestrial Ecology and Nature Conservation, Agricultural University, Bornsesteeg 69, 6708 PD Wageningen, The Netherlands Y. Chen Faculty of Agriculture, The Hebrew University of Jerusalem, P.O. Box 12, Rehovot 76100, Israel G. Concheri Dipartimento di Biotecnologie Agrarie, Universith di Padova, Via Gradenigo 6, 1-35131 Padova, Italy L.M. Condron Lincoln University, Departm_entof Soil Science, P.O. Box 84, Canterbury, New Zealand J. Cornejo Instituto de Recursos Naturales y Agrobiologia, C.S.I.C., Apartado 1052, E-41080 Sevilla, Spain G. Dell'Agnola Dipartimento di Territorio e Sistemi Agroforestali, Universit?a di Padova, Via Gradenigo 6,1-35131 Padova, Italy P. Fusi Dipartimento di Scienza del Suolo e Nutrizione della Pianta, Universit~ degli Studi di Firenze, Piazzale delle Cascine 28,1-50144 Firenze, Italy G. Guggenberger Institute of Soil Science, University of Bayreuth, D-95440 Bayreuth, Germany M.C. Hermosin Instituto de Recursos Naturales y Agrobiologia, C.S.I.C., Apartado 1052, E-41080 Sevilla, Spain

H. Insam Universit~it Innsbruck, Institut ffir Mikrobiologie, Technikerstrasse 25, A-6020 Innsbruck, Austria N.G. Juma Department of Renewable Resources, University of Alberta, Faculty of Agriculture and Forestry and Home Economics, 4-42 Earth Science Building, Edmonton, Canada T6G 2E3 K.R. Kelley Agricultural Research and Practices, TVA Environmental Research Center, P.O. Box 1010, Muscle Shoals, AL 35660-1010, USA J. Kozak Department of Soil Science, University of Prague, Prague, Czech Republic J. Magid The Royal Veterinary and Agricultural University, Department of Agricultural Sciences, Section of Soil, Water and Plant Nutrition, Thorvaldsensvej 40, DK-1871 Frederiksberg C, Copenhagen, Denmark S.P. McGrath Soil Science Department, Institute of Arable Crops Research, Rothamsted Experimental Station, Harpenden, Herts. AL5 2JQ, UK T.M. Miano Istituto di Chimica Agraria, Universit?~ di Bari, Via Amendola, 165/A, 1-70126 Bari, Italy P. Nannipieri Dipartimento di Scienza del Suolo e Nutrizione della Pianta, Universitb. degli Studi di Firenze, Piazzale delle Cascine 28, 1-50144 Firenze, Italy S. Nardi Dipartimento di Biotecnologie Agrarie, Universit~ di Padova, Via Gradenigo 6, 1-35131 Padova, Italy A. Piccolo Dipartimento di Scienze Chimico-Agrarie, UniversitY. di Napoli "Federico II", Via Universit?a 100, 1-80055 Portici, Italy C. Saiz-Jimenez Instituto de Recursos Naturales y Agrobiologia, C.S.I.C., Apartado 1052, E-41080 Sevilla, Spain N. Senesi Istituto di Chimica Agraria, Universith di Bari, Via Amendola, 165/A, 1-70126 Bari, Italy

P. Sequi Istituto Sperimentale per la Nutrizione delle Piante, Via della Navicella 2, 1-00184 Roma, Italy F.J. Stevenson Department of Natural Resources and Environmental Sciences, University of Illinois, Urbana, IL 61801, USA H. Tiessen University of Saskatchewan, Department of Soil Science, Saskatoon, S7N 0W0, Canada J. Wu Department of Agricultural and Environmental Science, Newcastle University, Newcastle upon Tyne NEI 7RU, UK W. Zech Institute of Soil Science, University of Bayreuth, D-95440 Bayreuth, Germany F.J. Zhao Soil Science Department, Institute of Arable Crops Research, Rothamsted Experimental Station, Harpenden, Herts. AL5 2JQ, UK A. Zsolnay GSF Forschungszentrum ffir Umwelt und Gesundheit, Institut ffir Boden6kologie, D-85758 Neuherberg bei Mfinchen, Germany

This Page Intentionally Left Blank

Humic Substances in Terrestrial Ecosystems Edited by A. Piccolo 9 1996 Elsevier Science B.V. All rights reserved.

Chapter 1

The Chemical Structure of Humic Substances: Recent AdvancesC. SAIZ-JIMENEZ

INTRODUCTION The chemical nature of humic substances is perhaps one of the most frequently discussed and recurrent topics in science. The study of these macromolecules has been a challenge to the ingenuity of scientists for more than 200 years, and in spite of the application of almost all available analytical instrumentation developed over the last four decades, knowledge of their nature and composition is still limited. It has been reported that most of the degradative methods used so far in humus chemistry have a strong impact on the original building blocks and subsequently lead to their significant alteration. Therefore, these analytical approaches are of limited value as the reaction products only partially reflect the structures of the building blocks and not so much their linkages. The interpretations could possibly be incorrect, as in many instances the naturally occurring units can be altered before or after their release from the macromolecular structure (Saiz-Jimenez, 1992a). The chemical degradation approach has previously been discussed at length (Hayes and Swift, 1978; Norwood, 1988) and will not be the subject of this chapter. Advances or new insights into the structure of humic substances have inevitably developed together with the application of analytical techniques. In a search for reliable methods capable of solving the intriguing aspects of humus chemistry, attention was focused on analytical pyrolysis, which has been applied to a variety of plant materials and soil organic matter fractions over the last fifteen years. Analytical pyrolysis is considered to be a small-scale thermal degradation method which is very useful for the chemical characterization of materials from their pyrolysis products. This technique usually involves an integrated pyrolysis-analysis system which is carefully controlled to produce reproducible results and which uses small (ng-~tg range) amounts of samples (Irwin, 1979). In a previous paper (Saiz-Jimenez, 1992a), an attempt was made to summarize the most important milestones in pyrolysis studies of humic substances, and, at the same time, to demonstrate how the existing theories on the chemical structure of

2

C. Saiz-Jimenez

humic substances, from the sixties to the eighties, have decisively influenced structural data derived from pyrolysis, reducing or changing their impact. Even nowadays, pyrolysis data of humic substances are not correctly interpreted, and misunderstandings often occur in pyrolysis papers. The technique, widely used over the last decade, can cause drastic modification of the original building blocks, which may lead to incorrect conclusions on structure. In fact, the most biased tendency is to consider pyrolysis products as building blocks of the macromolecule, the same mistake inferred many years ago for oxidative degradation. The purpose of this chapter is twofold. Firstly, to demonstrate that pyrolysis products cannot be considered as representative building blocks, due to the considerable thermal reactions and rearrangements produced in the process, which therefore limit the usefulness of the technique. Secondly, it provides recent information on a novel pyrolysis method: simultaneous pyrolysis/methylation, which is able to alleviate limitations imposed by the technique, in order to take pyrolysis data back to the genuine chemical structure from which they originate. To achieve this, an attempt is made to define the most probable precursors of humic substances, followed by a review on pyrolysis of the most important humic components.

PRECURSORS OF HUMIC SUBSTANCES Soil organic matter is divided into nonhumic and humic substances. Nonhumic substances include those with still-recognizable chemical characteristics (e.g. polysaccharides, proteins, nucleic acids, lipids, etc.), while humic substances are regarded as transformed materials which have lost the chemical characteristics of their precursors. However, on an operative basis, it is difficult to distinguish between nonhumic and humic substances as, once extracted from soil, humic fractions can be exhaustively purified from admixtures by different procedures (solvent precipitation, resin adsorption, column or gel fractionation, acid hydrolysis, etc.). In order to properly discuss the chemical composition of humic substances it is important to investigate the possible contribution of the main precursors of the humic macromolecule.

Abiotic synthesis of humic substancesIt is generally accepted that kerogen is derived from biologically synthesized organic matter through a series of predominantly microbially-mediated reactions, collectively called diagenesis. The differences between the chemical constitution of kerogen and the extant biomass, mainly consisting of cellulose, lignin, proteins and lipids, led to the conclusion that the main components of kerogen were newly formed during diagenesis. Accordingly, kerogen was considered to be the result of a series of consecutive random polymerization and condensation reactions of

The chemical structure o f humic substances: recent advances

3

lipids with sugars and amino acids. In parallel, in terrestrial environments the synthesis of humic acids has sometimes been considered as the result of similar condensation and polymerization reactions of sugars with amino acids or proteins. It has been suggested that the Maillard or browning reaction is a viable mechanism for humic acid formation in aquatic ecosystems with phytoplanktonderived amino acid and carbohydrate as precursors. The reaction is initiated by the formation of a Schiff base between the carbonyl of a sugar and the nitrogen of an amino acid or ammonia. The resulting N-substituted derivative then undergoes a complex series of dehydration, rearrangement, and condensation reactions to produce both simple fragmentation products and structurally complex brown nitrogenous polymers (Hedges, 1988). In terrestrial environments melanoidin formation would involve the condensation or repolymerization of reactive small organic molecules which have been generated by essentially complete microbial breakdown of bonds between structural units in the original polysaccharides and proteins, two of the most abundant constituents of all living organisms. However, microbial degradation and rapid turnover of these biomacromolecules and their units would prevent condensation and browning reactions. Furthermore, such reactions would imply that single, easily degradable molecules are not used for microbial metabolism, which is difficult to accept in a highly competitive microbial ecosystem such as soil. Ikan et al. (1986) considered that on the basis of ~3CNMR spectra, melanoidins have a remarkable resemblance to some humic acids and hence it is suggested that the Maillard reaction (of sugars and amino acids) plays a more significant role in the formation of structure of humic substances than previously thought. However, such similarities have been questioned (Hedges, 1988), and evidence based upon spectroscopic methods rather than upon chemical degradation methods can lead to misinterpretations. According to Hedges (1988), conclusive molecular-level evidence that melanoidins exist naturally has yet to be presented and will be a challenge to provide as most of the degradative reactions commonly used for structural analysis of such polymers can themselves form melanoidin-like material from polysaccharide and protein precursors as a reaction side-product. This is corroborated by the formation of a melanoidin upon 6 N HC1 hydrolysis of chitin (unpublished data). It appears that hydrolysis releases the N-acetyl groups, as monitored by pyrolysismass spectrometry, and denotes the decreasing/elimination of peaks at m/z 59 (acetamide), 60 (acetic acid), 73 (propionamide), 109 (hydroxypicoline), 125 (3-acetamidofuran), 137 (3-acetoxypyridine), 151 (methyl-N-acetyl-2-pyridone), and 167 (trianhydro-2-acetamido-2-deoxyglucose), and induces condensation and polymerization of glucose units. Furthermore, some compounds appeared to be increased, as shown by peaks at m/z 110, 126, 142 and 150, which could probably be assigned to carbohydrate pyrolysis products (e.g. 110: methylfuraldehyde, 126: levoglucosenone/methylhydroxypyranones, 142: methyl-dihydroxypyranones). It is clear that in addition to removal of N-acetyl groups, hydrolysis and heating

4

c. Saiz-Jimenez

induce changes in the structure of polysaccharides, which are not completely hydrolysed, leading to brown polymer formation. This should be taken into account when hydrolysing humic materials rich in polysaccharides. Furthermore, melanoidins originating from xylose-lysine (molecular ratio 4:1), both at pH 3 and 9, yield very similar pyrolysis-mass spectra, indicating that brown polymers are formed in spite of very different pH conditions, which were thought to have considerable influence on the developing polymer. The formation of melanoidins from polysaccharides, at acid pH, is important as acid hydrolysis is commonly used in purification processes.Bacteria and fungi

Microorganisms synthesize a variety of materials including polysaccharides, proteins, nucleic acids, carotenoids, nucleic acids, lipids, etc. Most of these are easily metabolized upon death and autolysis of the organisms, and enter the carbon cycle. Whilst some of the biomacromolecules appear to be degraded relatively rapidly in nature, some refractory materials escape degradation and accumulate in the environment. Of all macromolecules synthesized by microorganisms, perhaps the most recalcitrant is melanin, which was consequently considered as a possible precursor of humic substances (Haider et al., 1975). Nicolaus (1968) reported that melanins, a class of pigment widespread in the animal and vegetal kingdoms, may be classified into eumelanins, phaeomelanins and allomelanins. Whereas eumelanins and phaeomelanins are found in the animal kingdom and are represented by the pigment giving colour to human black and red hair, allomelanins occur in microorganisms and plants. It is interesting to note that generally, melanins based on the tyrosine unit through dihydroxyphenylalanine (DOPA) formation are eumelanins and phaeomelanins. Allomelanins, consequently, are black pigments formed from different, usually nitrogen-free precursors. It seems that the term melanin has frequently been indiscriminately used to describe bacterial pigments which in some cases fail to qualify as melanins (Nicolaus, 1968). The formation of black pigments by bacteria in culture media containing aromatic precursors is a well-known process. Synthesis of black pigments by Streptomyces in culture media containing tyrosine was reported many years ago (Mencher and Heim, 1962). It appears that tyrosinase, an enzyme necessary for oxidation and polymerization of aromatic amino acid precursors, occurs in streptomycetes. However, from a physiological point of view, these black materials cannot be considered as natural pigments but as artifacts produced by an excess of aromatics in culture media, which, of course, are not the conditions existing in nature. A DOPA melanin is formed by the oxidation of tyrosine by tyrosinase. Because the amino acid tyrosine occurs universally in living organisms and the enzyme tyrosinase is well documented in a few fungi, there has been a tendency to conclude that intracellular melanins are also DOPA melanins (Bell and Wheeler,

The chemical structure of humic substances: recent advances

5

1986). For Ellis and Griffiths (1974) the melanins of some soil fungi (e.g. Epicoccum nigrum) appeared to be DOPA melanin. However, the Ehrlich's reagent test used for evidencing indole, a characteristic marker of such a type of melanin, is nonspecific as phenols also react positively. Bell and Wheeler (1986) considered that some melanins originally identified as DOPA melanins by using nonspecific tests have since been shown to be dihydroxynaphthalene (DHN) melanins. A few examples of fungi-producing heterogeneous melanin (which could not be classified as DOPA, catechol, or DHN melanins) are Aspergillus niger, Aspergiilus sydowi, Coprinus spp., Eurotium echinulatum, Hendersonula toruloidea, etc. (Nicolaus, 1968; Haider and Martin, 1970; Martin et al., 1972; Saiz-Jimenez, 1983; Saiz-Jimenez et al., 1975). Even Epicoccum purpurascens (= E. nigrum) melanin was a heterogeneous melanin (Haider and Martin, 1967), although Ellis and Griffiths's statements suggest the contrary. Characterization of most fungal intra- and extracellular melanins by means of chemical and thermal degradative methods showed content of a wide variety of chemically recognized macromolecular materials, including polysaccharides, proteins, nucleic acids, lipids, and phenol derivatives (Saiz-Jimenez, 1994b; Martin et al., 1974; Saiz-Jimenez et al., 1979), which are well-known components of the fungal cells. It is possible that minor amounts of a real melanin can be identified in these polymers. In fact, the Aspergillus niger melanin contains considerable amounts of polysaccharides, not encountered in other melanins (Saiz-Jimenez et al., 1979). Stachybotrys chartarum (= S. atra) melanin exhibits dominant series of alkadienylcyclohexenes upon pyrolysis (Saiz-Jimenez, 1994b), and Eurotium echinulatum melanin shows some differences when extracted from culture media or mycelia, as represented in the increase in amino acid pyrolysis products (unpublished data). These results lead us to consider the so-called melanins as a mixture of every class of macromolecular materials existing in the fungal cell or released to the medium upon autolysis. Alkaline extraction and/or acid precipitation could favour artificial aggregation, physical adsorption, hydrogen bonding, etc. of the extracted materials. This process imitates the extraction of humic substances from soil and, in the same way, the end product is a mixture of free and macromolecular components. It should be noted that the production of such macromolecular materials is not homogeneous for all the fungi studied. In fact, each fungus produces different materials as the cellular components and metabolites involved are dissimilar. Consequently, a few fungi can synthesize extracellular phenolic polymers or heterogeneous melanins under laboratory conditions, and these are difficult to compare with intracellular DHN melanins from Ascomycetes or catechol melanins from Basidiomycetes (Bell and Wheeler, 1986). Similarities between soil humic acids and fungal melanins were proposed on the basis of IR spectra, amino acids released upon acid hydrolysis and phenols recovered after Na-amalgam reduction (Martin et al., 1967). These similarities

6

C. Sa&-Jimenez

seem difficult to accept according to further studies (Haider and Martin, 1967; Martin et al., 1967, 1972; Schnitzer et al., 1973; Saiz-Jimenez, 1983), in which fungal melanins were demonstrated to be different from each other and to contain high amounts of complex but chemically defined structures, which cannot be collectively identified as melanins or humic substances. Similarities can only be accepted by the fact that both types of materials contain polysaccharides, proteins, lipids, etc., which can easily be removed (Saiz-Jimenez et al., 1979). Concerning the possible participation of fungal macromolecular components in soil humus there is always the possibility of extracting microbial materials (including melanins) from soils with alkaline agents, and these, operationally defined, are considered to be humic substances. Finally, it can be noted that the chemical nature of fungal melanins was, in the past, an object of controversy. Schnitzer et al. (1973) suggested on the basis of permanganate oxidation that the melanins produced by Aspergillus niger, Epicoccum nigrum and Stachybotrys chartarum (= S. atra) were complex material containing aliphatic and aromatic structures, only some of which were phenolic. Martin et al. (1974) reported that reductive degradation of fungal melanins yielded between 10 and 60% of ether-soluble substances, and from these 4-32% were phenolic compounds. It seems clear that (1) no conclusions can be drawn from the application of a single, usually very drastic degradative method, (2) that different fungi (even the same species but different strain) can produce different macromolecular materials and/or melanins, and (3) that a discussion on structural similarities and dissimilarities cannot be based on single chemical data obtained from different degradative methods and samples.

Phototrophic microorganismsPhototrophic microorganisms (cyanobacteria and algae), common in stones and wet soil surfaces, grow as a mixed community intimately entangled with its extracellular products, generally polysaccharides. This characteristic growth, socalled microbial films, consists of a layer of microbes held in a gelatinous or slimy matrix of extracellular polymer. The synthesized extracellular polysaccharides (50-90% of biofilms) composed largely of mannans, glucans, uronic acids and associated glycoproteins and other heteropolymers, serve to hold the cells together and to irreversibly bind surfaces. Microbial films may also contain significant amounts of adsorbed inorganic materials derived from the soil (quartz, calcium carbonate, clays, etc.) and detritus (dead cells, microbial by-products). This biofilm supports the growth of other heterotrophic organisms such as bacteria or fungi, which can live at the expense of the extracellular organic matter synthesized by living cyanobacteria and algae or from dead cells. In this way an important organic carbon input is provided for the ecosystem and the mixed populations constitute a microcosmos with varied activities, the final result being that recalcitrant organic materials or residues remain after a rapid


7

microbial turnover of the organic matter. The refractory materials from phototrophic microorganisms, collectively called algaenan (Tegelaar et al., 1989), are resistant non-hydrolysable macromolecular structures present in cell walls of a number of algae, and could in all probability accumulate in soils and sediments upon senescence and microbial decomposition of degradable structures (SaizJimenez, 1992b). Zelibor et al. (1988), studying mixed and pure cultures of green algae and their decomposed residues (under laboratory conditions) confirmed the presence of chemically refractory components. The refractory fraction ranged from 33% in aerobic to 44% in anaerobic cultures and is composed of insoluble paraffinic carbon materials, which are resistant to chemical and bacterial degradation. It appears that these green algae materials are usually associated with cell wall components. It has been reported that all algal strains producing ketocarotenoids also form sporopollenin-like polymers which are known for their extraordinary resistance to chemical and biological agents. These polymers are deposited in the outer cell wall layer and can reach up to 10% of the total algal biomass (Largeau et al., 1984). However, it has been clearly demonstrated that the structure of the resistant biopolymers consists chiefly of long, unbranched saturated hydrocarbons chains, probably linked by ether bridges, and accordingly they cannot be considered as sporopollenin (Chalansonnet et al., 1988). The insoluble, refractory materials from cyanobacteria have been considered as the basic building blocks for kerogen (Philp and Calvin, 1977) and humin in sediment of aquatic origin (Hatcher et al., 1985).Lichens and bryophytes

Lichens produce a large number of intra- and extracellular products. The latter are almost all water-insoluble crystals deposited on the surface of the fungal hyphae, usually oxalates, phenolic derivatives and anthraquinones. Most of the intracellular compounds isolated from lichens are non-specific and occur not only in free-living fungi and algae but also in vascular plants. Polyols, monosaccharides, polysaccharides and other low-molecular-weight carbohydrates are abundant in lichens. Proteins, amino acids, carotenoids and vitamins are also known to be present in lichens (Hale, 1983). Due to restricted or spotting distribution of lichens in bare rocks (saxicolous communities), life span (50-100 years or more) and antimicrobial activity of secondary products, it is difficult to consider lichen structures from saxicolous communities as important humus precursors. However, lichen substances and oxalic acid can form complexes with cations (calcium, magnesium, iron, aluminium, etc.) leading to rock weathering. Possibly, soil organic matter contribution is restricted to soil-inhabiting lichen communities which, in turn, are sparsely distributed and highly vulnerable to environmental changes induced by the surrounding vegetation, provoking senescence and death.

8

C. Saiz-Jimenez

It is interesting to note that in the Great Plains of the United States, peat-like deposits of lichen origin have been found (Stevenson, 1982). Until now no algaenan have been reported present in lichen structures, although biomacromolecules with high preservation potential have been found in the phycobiont and it has been stated that some lichen phycobionts contain sporopollenin-like polymers (Krnig and Peveling, 1984). Anyway, the relatively high content of phenolic acid derivatives and quinones (up to 10% of dry thallus) in lichen structures represent a pool of active di- and triphenols and certainly play a role in the reaction and condensation of aromatic compounds in soils. Bryophytes, as opposed to lichens, often form extensive green carpets on wet soils, and represent a relatively important biomass input, contributing to soil organic matter. The main chemical constituents of bryophytes are usually those found in plants, including polysaccharides, proteins, lipids, carotenoids, terpenoids, flavonoids, although the presence of lignin is doubtful. Due to antibiotic production, developed for competition with other organisms, bacteria are not very active and breakdown of moss tissues is not rapid. On the contrary, actinomycetes and fungi (Penicillium, Cladosporium, Geotrichum) are important in the decomposition of dying moss leaves, attacking cell wall tissue of Sphagnum bogs and carpets. It has been reported that the cuticular components of mosses were similar to those of vascular plants. The question arises as to whether mosses also contain aliphatic biopolymers, previously encountered in plants (Nip et al., 1986a). In addition, the occurrence of flavonoids and cinnamic acid derivatives in bryophytes could represent an input to the pool of aromatic compounds in soils. Mosses are usually associated with cyanobacteria and green algae. Heterotrophic bacteria grow at the expense of phototrophic microorganisms whilst a broad range of fungi parasitize mosses. This means that conditions are appropriate for distinct microbial populations to breakdown the pool of complex macromolecules from different origins. In this way, algaenan and lignin-like materials, present in at least some mosses (Chopra and Kumra, 1988), could be preserved in certain soils. In cold, wet regions, humus is apparently formed from mosses (Stevenson, 1982). Peat formation is a process in which, under anaerobic conditions, the rate of decomposition is very slow and the preservation potential of all macromolecules is increased.

Vascular plantsThe formation of soil humus has been traditionally considered to be the result of decomposition of available plant materials by microorganisms, followed by resynthesis of microbial components and accumulation of selectively preserved materials, resistant to decomposition. The degradation of organic matter in soils is not complete and many factors (climate, type of vegetation, soil pH, parent rock, etc.) can influence the microbial breakdown of organic carbon, thus avoiding an essentially complete decomposition of many plant structures.


9

TABLE 1 Biomacromolecules present in organisms and their potential for surviving in terrestrial environments~ Biomacromolecules Polysaccharides Proteins Nucleic acids Waxes Resins, ambers Tannins/phenolic polymers Melanins Lignins Sporopollenin Algaenan Cutan Suberan Occurrence All organisms All organisms All organisms Vascular plants Vascular plants Vascular plants All organisms Vascular plants Vascular plants Algae Vascular plants Vascular plants Preservation potential2 -/+ -/+ -/+ +/++ ++/+++ ++/+++ +++/++++ +++/++++ ++++ ++++ ++++ ++++

l Modified after Tegelaar et al. (1989). 2 Preservation ranges from - (extensive degradation) to ++++ (preservation).

It is far from doubtful that alkaline extraction of soils recovers a wide variety of plant components, ranging from single compounds involved in the rapid turnover of organic matter, more or less transformed materials participating in medium-term degradation processes, to recalcitrant, long-term accumulated plant components. Thus, the most recalcitrant biomacromolecules, usually non-hydrolysable and insoluble structures, accumulate in soils and are preserved from degradation. The survival potential of most biomacromolecules is reported in Table 1. Over the last few years plant refractory materials have been found in kerogens (Tegelaar, 1990), and the survival of these macromolecules following microbial degradation and diagenesis processes have been demonstrated (Logan et al., 1993). One example is the highly aliphatic macromolecules encountered in plant cuticles and barks (Nip et al., 1986b; Tegelaar, 1990), which were also present in humic acids, and evidenced after acid hydrolysis and/or oxidations (Saiz-Jimenez and de Leeuw, 1987a). Lignin biodegradation, considered the major contributing process to humus formation, was overestimated in the past and should probably be replaced by a more balanced contribution of plant materials other than lignins. Although there is direct evidence of the transformation of lignin in soils, it is hard to accept that, invariably, refractory lignin components can be selectively preserved. Most probably, this preservation is not directly related to the recalcitrance of selected building blocks, but to environmentally adverse conditions for continuous microbial activity.

10

C. Saiz-Jimenez

Furthermore, aliphatic macromolecular components have been found in cell walls from some phototrophic microorganisms (Largeau et al., 1984; Zelibor et al., 1988), which makes it quite clear that in addition to lignin and some phenolic derivatives, which tend to accumulate in certain terrestrial environments, other plant and microbial materials could be selectively preserved and incorporated into or co-extracted from humic substances. Interestingly, the resistant biomacromolecules encountered in the humic molecule of unpolluted soils (Saiz-Jimenez and de Leeuw, 1987a,b), have also been considered as major contributors to kerogens (De Leeuw et al., 1991) and are structures commonly identified in algae, pollen, plant cuticles, periderm tissues, etc., all of which are expected to be present in soil surfaces. Therefore, it appears that similarly resistant biomacromolecules could be selectively preserved and form part of the refractory components encountered in soil humus. Contrary to the opinions of past decades, most of the biomacromolecules with high preservation potential are highly aliphatic in nature.

Atmospheric depositionInnumerable chemicals are emitted directly or indirectly into the atmosphere by man's activities. These pollutants are distributed throughout the environment and the atmosphere is recognized as a major route for worldwide dispersion. The two mechanisms by which pollutants are transferred to soils are dry and wet deposition. Dry deposition proceeds without the aid of precipitation and involves the direct transfer of gases and particulates to the Earth's surface. Wet deposition, on the other hand, encompasses all processes by which airborne pollutants are transferred to the Earth's surface in an aqueous form (i.e. rain, snow, or fog). In recent years considerable attention has been paid to the presence of organic compounds in rain and snow samples (Kawamura and Kaplan, 1986), aerosols (Simoneit et al., 1988), and particulates (Yokouchi and Ambe, 1986) over remote, rural and urban areas. Deposition of atmospheric pollutants in urban building stones has also been thoroughly investigated (Saiz-Jimenez, 1991, 1993). Until now, no research has been carried out on the input of atmospheric organic pollutants to soils, and deposition of organic compounds to terrestrial ecosystems has traditionally been a topic ignored in soil organic matter studies. However, there is no reason for disregarding this, especially when this input is of importance in the Northern Hemisphere and airborne materials are consequently transported from heavily polluted to rural areas. As an example, it has been reported that over 40% of the direct primary emissions of organic aerosols into the Los Angeles atmosphere are contributed by anthropogenic air pollution sources. The total fine aerosol organic carbon emissions within an 80 x 80 km heavily urbanized area was estimated to be 29.8 Tm/day (Hildemann et al., 1991). Natural fires, controlled waste burning, and residential wood combustion cause distillation and pyrolysis of plant materials, and contribute significant amounts of


11

TABLE 2 Classes of compounds identified in aerosols and particulate matter CompoundsAlkanes

Range~C 7 -C40

Compounds Alkylnaphthoic acids Alkylphenanthroic acids Alkylcyclohexanes Diterpenoid hydrocarbons Triterpenoid hydrocarbons Tricyclic terpane hydrocarbons Steranes and diasteranes Unresolved hydrocarbons PAH Oxygen-PAH

RangeC 11-C13 C15-C17

Isoprenoid hydrocarbons Isoprenoid ketones Alkan-2-ones Alkanols Fatty acids Hydroxy fatty acids t~,o3-Dicarboxylicacids Alkylbenzoic acids Alkylbenzenedioic acids

C10-C20C 10---C20

Clo--C32Cll-C28 C1-C34 C~0-C26C 2 -C26

C 9---C29 C27-C35 C27-C35C19-C29 C27-C29

C14-C31

C7 -C 9 C8-Cm

Clo-C24Clo-C16

Range denotes number of carbon atoms in the compounds. carbonaceous aerosols and polycyclic aromatic hydrocarbons (PAH) into the atmosphere. Plumes from forest fires in Canada were detected up to 5000 km away by satellite imagery (Chung and Le, 1984). Hites et al. (1980) analysed fifty soils and sediments from all over the globe and found that the presence of complex mixtures of PAH in the environment is worldwide. The presence in soils of a wide range of alkyl PAH from forest and prairie fires was specifically demonstrated (Youngblood and Blumer, 1975). The distribution of PAH over the remotest areas of the globe adds further evidence for the long-distance transport of carbonaceous particles and adsorbed compounds, and to a global background level of PAH due to natural combustion such as forest fires. Organic species present in aerosols and particulate matter are complex mixtures of many classes of compounds. The major classes of compounds identified in aerosols and particulate matter (Simoneit, 1985, 1986; Kawamura et al., 1985; Kawamura and Kaplan, 1987; Saiz-Jimenez, 1993, 1994b) have been summarized in Table 2. In addition, Fig. 1 shows the TIC chromatogram of a diesel soot extract, and Table 3 some of the major compounds. The major classes of compounds present in the diesel soot are n-alkanes, n-fatty acids, ~,03-dicarboxylic acids, alkylcyclohexanes, alkylbenzenes, naphthalenes, fluorenes, and phenanthrenes, alkylbenzoic acids, alkylnaphthoic acids, and PAH. Also, an unresolved complex mixture of cyclic and branched hydrocarbons or 'hump' was observed. Environmental soot, which is a mixture of various forms of particulate carbon with organic tar and refractory inorganic materials, is an important part of aerosol emissions. It has been found that typical aerosols contain 10 to 30% total carbon. Of this fraction, 20-50% is carbon, less than 5% is carbonate, and the remainder

12TABLE 3 Compounds identified in a diesel soot extract Peak 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 Compound Vinylnaphthalene Methylisobenzofurandione n-Undecanoic acid ~ o~,o)-Octanedioic acid Methylbiphenyl n-Pentadecane Dibenzofuran n-Dodecanoic acid o~,co-Nonanedioic acid Fluorene C a Alkylnaphthalene n-Hexadecane n-Tridecanoic acid Methyldibenzofuran C 4 Alkylnaphthalene Branched hydrocarbon Branched hydrocarbon n-Decylbenzene Methylnonylbenzene Methylfluorene n-Heptadecane Peak 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 Compound

C. Saiz-Jimenez

Pristane n-Tetradecanoic acid C 5 Alkylnaphthalene Octylcyclohexane Anthracene Phenanthrene n-Octadecane Phytane n-Pentadecanoic acid Di-iso-butyl phthalate Branched hydrocarbon n-Nonadecane n-Hexadecanoic acid Dimethyldibenzothiophene Dibutyl phthalate n-Hexadecanoic acid n-Eicosane n-Heptadecanoic acid n-Heneicosane n-Octadecanoic acid n-Docosane

z All acids appeared as methyl esters, except compound 37. Methylation of acid groups was produced during the extraction and/or evaporation procedure.

12

I

7

I

i2500

2

38 29 26 36

1700

2100

2900

3300

3700

Scan numbers

Fig. 1. Diesel soot extract TIC chromatogram. Peak identifications in Table 3.


13

is composed of adsorbed organic compounds (Simoneit and Mazurek, 1981; Griest and Tomkins, 1984). Yu and Hites (1981) identified alkylphenanthrenes (C~-C4) in diesel exhaust particulates together with alkylfluorenes and phenylnaphthalenes. Similar compounds and alkylnaphthalenes were reported by Lee et al. (1977) during the combustion of coal, wood and kerosene. It has been determined that approximately 90% of the PAH emissions in the United States are due to coal combustion processes. The PAH isolated from combustion products of coal are well represented in airborne particulate matter. Kunen et al. (1976) investigated the insoluble polymer-like carbonaceous portion of particulate matter by pyrolysis. The pyrolysate was composed of alkanes, and alkenes. The most remarkable fact is that benzene, toluene, and styrene were dominant, and alkylbenzenes and naphthalenes were also identified. In another work, Mukai and Ambe (1986) isolated a sample of brown material from airborne particulate matter from a rural area of Japan. This material, constituting up to 3% of total carbon, showed the solubility behaviour of humic acids. Agricultural burning was considered as the primary source of the high molecular weight acidic material. Kumada (1983) considered that burning plant materials can produce humic acids very similar to those extracted from soils, on the basis of absorption spectra. However, this is not a relevant technique for structural studies. Also, it must be taken into account that burning of slash and wood produces a tar or distillate of aromatic compounds from lignin, furan derivatives from cellulose, and a carbonaceous residue or carbon. Hawthorne et al. (1988) identified more than 70 organic compounds, extracted from wood smoke particulates, from which 28 were methoxyphenols, also found in flash pyrolysis of lignin. The carbonaceous particles are very sorptive, exhibit high specific surface areas able to strongly sorb PAH, and contain high levels of organic matter. In addition, the ability of fulvic and humic acids to bind anthracene, pyrene and perylene was demonstrated (Schlautman and Morgan, 1993). Therefore it is not surprising that carbon produced in fuel and wood combustion, as well as humic substances, acts as a nucleating agent for adsorption of organic compounds in soils.Possible contribution of precursors to humic substances

In the light of the discussion on the different precursors of humic substances, an attempt to summarize the main conclusions is presented in Table 4. It appears that melanoidins can be considered as artifacts produced exclusively during thermal heating of polysaccharides alone or together with N-containing compounds, which is not a common situation in soil. Formation of melanoidins in soils under thermic stress (e.g. forest fires, controlled waste burning) might be considered, but in these cases, most probably, polysaccharides and proteins are charred and do not react as the browning reaction usually requires water, which

14

C. Saiz-Jimenez

TABLE 4 Most probable aromatic and aliphatic precursors involved in the formation of soil humic substances Precursors Melanoidins Microbial melanins Microbial and lower plant phenols Polyphenols and lignins Algaenan Sporopollenin Cutan/suberan Pollutants Remarks Questionable, the reaction between carbohydrate and amino acids is very unlikely in soils Possible, at least some common soil microorganisms produce cell wall melanins and extracellular pigments Possible, microbial phenols, lichen orcinol derivatives, moss flavonoids, etc. could play a role in polymerization reactions Possible, these molecules are distributed worldwide Possible, phototrophic microorganisms produce this aliphatic macromolecule with high preservation potential Possible, pollen structures are very resistant and can accumulate in soils Possible, vascular plants produce substantial amounts of cutan/suberancontaining structures which are preserved from microbial degradation Possible, soil is a sink for aerosols and particulate matter. The latter contains a wide variety of compounds entrapped in a carbonaceous matrix

favours the process and increases the reaction rate. Anyway, conclusive evidence has yet to be presented as no characteristic melanoidin building blocks have so far been identified in humic molecules. Melanins are macromolecules as complex as humic acids. The macromolecules collectively called melanins (on an operational basis) are a mixture of materials contained in the cell, in the same way as humic substances are mixtures of materials contained in soil. It is easy to understand the differences between both types of macromolecules due to their diverse origins. Indeed, melanins are as different from each other as humic acids from different soils can be. Phototrophic microorganisms, lichens, bryophytes, and vascular plants produce a variety of biomacromolecules with different preservation potential. Due to their distribution in soils, it appears that recalcitrant materials from phototrophic microorganisms can survive to a greater extent than those of lichens and bryophytes. Highly aliphatic plant macromolecules appear to be well preserved, and lignins and plant melanins to a lesser extent. The survival of biomacromolecules seems to play a key role in soil organic matter accumulation processes, as has been deduced from some humic acids studied. The preservation potential depends on the chemical nature of the macromolecules, the microbial population activity, and the environmental conditions in which the macromolecules are immersed.


15

Air pollution and deposition of organic chemicals in soils is a process of considerable importance in most developed countries and should not be neglected. Pollutants are distributed in soils and sediments worldwide, and their presence in humic substances has been stressed in several reports. Forest fires, controlled waste burning and domestic wood combustion produce very high amounts of PAH, tar, and carbonaceous particles. The activated carbon generated not only under wood, but also coal and fuel combustion, could act, in soils, as a nucleating agent for organic chemicals. This is a very interesting hypothesis which deserves further investigation. The distribution of alkylbenzenes, naphthalenes and phenanthrenes in some soils and humic acids is highly suggestive of such a contribution, if they do not represent pyrolysis artifacts produced in the laboratory. The use of agricultural practices (addition of composts, manures, sewage sludges, etc.) modifies the nature of soil humic substances, reflecting the input of alien organic matter (Saiz-Jimenez, 1994b). This is exemplified by a humic acid extracted from sewage sludges, which pyrolysate is constituted by high amounts of isomeric alkylbenzenes in the range CHj-C~3; the same mixture was identified upon pyrolysis of commercial detergents. It seems that the definition of which are the real humic and nonhumic components in a humic molecule would be a milestone in deciphering the chemical structure of humic substances. Until now only recognizable microbial, plant and pollutants compounds, with a well established origin, have been identified. No significant structural units which could be related to melanoidins, microbial melanins, etc. have been identified in degradative studies, but only compounds related to plant and pollutants. The finding of a typical humic structure or building block, different from those of the precursor materials, if such a structure exists, would consolidate our present uncertainty when dealing with humic substances.

COMPONENTS OF HUMIC SUBSTANCES The chemical composition of humic fractions, and particularly fulvic acids, is governed by extraction, fractionation and purification procedures, and humic fractions with different chemical compositions can be extracted from the same soils, according to the procedure employed (Saiz-Jimenez and de Leeuw, 1986a). This heterogeneity is a reflection of the various inputs of organic matter to different soils and its different solubility behaviour, and is one of the factors which incited scientists to propose standard procedures aiming at the reproduction of the chemical characteristics of humic fractions. Unfortunately, it must be accepted that the direct incorporation of biomacromolecules into humic fractions is an extremely underemphasized aspect of the extraction of humic substances. In fact, certain types of biomacromolecules (uronic acids, proteins, tannins, lignins, melanins, etc.) or their slightly microbially modified degradation products have the same solubility behaviour, are

16

C. Saiz-Jimenez

incorporated into humic extracts, and are thus collectively called humic substances, on an operational basis. These materials, which belong to well defined classes of organic compounds, comprise a major portion of the humic substances found in many soils. This implies that biomacromolecules with low preservation potential together with refractory biomacromolecules have a higher probability of becoming part of the humic molecule, assuming that they naturally have, or can be transformed to have, the required solubility behaviour (Ertel et al., 1988). However, not only biomacromolecules but also pollutants worldwide distributed such as phthalates, pesticides, carbonaceous particles originated upon combustion and/or their adsorbed PAH and alkyl homologs, alkylaromatics, etc., can be incorporated into humic extracts. In this respect, fast and complete binding of PAH (Schlautman and Morgan, 1993), and retention of hydrophobic organic compounds by humic acids were reported (Khan and Schnitzer, 1972). This process could be of particular importance in highly polluted areas such as Northeastern United States and neighbouring Canada, central Europe, etc. Therefore, in order to decipher the chemical structure of humic substances it is crucial to understand which really are their components, a basic question not yet fully answered. Accordingly, in the light of current knowledge, the bulk of the humic molecule can be considered as a mixture of every class of plant, microbial materials, and pollutants deposited or buried in soils, and can be extracted by strong alkaline agents. It has been suggested that only a small fraction (10-20%) of the humic molecule can be typically accounted for as recognizable chemicals. This low percentage has led to the general assumption that humic substances must be the product of extensive microbial degradation which has altered much of the initial biochemical signature (Hedges, 1988). However, this amount seems very low. In fact, Saiz-Jimenez and de Leeuw (1987a) stated that about 70% of some humic acids were released as acid hydrolysable, solvent extractable materials, which mainly included polysaccharides, proteins, lignins, and lipids. The residual humic acid accounted for mainly aliphatic biopolymers, as demonstrated by analytical pyrolysis. Most of the molecular components in humic fractions can be chemically recognized and a plant and/or microbial origin assigned to it (Table 5), with the exception of a few pollutants. Interestingly, an important part of these components conforms to the biomacromolecules preserved in terrestrial ecosystems (Table 1). This does not extend to the lipids, mostly free, solvent-extractable compounds, which are widely distributed in the environment. No defined chemical structures, exclusively related to a humic molecule, can be found in any of the humic fractions, which makes chemical characterization of the humic structure itself extremely difficult. In addition, whether such a high variety of compounds present in soils are chemically bonded to each other, and in which way, or whether they are physically bonded, is a question not yet solved. In


17

TABLE 5 Major classes of compounds and biomacromolecules identified in humic substances Class of compoundsAliphatic hydrocarbons Alkanes Alkanols Aliphatic acids Fatty acids Hydroxy fatty acids Dicarboxylic acids Alkylaromatics Alkylbenzenes Alkylnaphthalenes Alkylphenols Dialkyl phthalates Aromatic hydrocarbons PAH Aromatic acids Benzenecarboxylic acids Phenolic acids Other hydrocarbons Tocopherols Chlorophylls Terpenoids Steroids Biomacromolecules Waxes Polysaccharides Proteins Lignins Aliphatic macromolecules 2

Possible origin

References ~

Microbial/plant/pollutant Microbial/plant/pollutant

1-4 2, 5

Microbial/plant/pollutant Microbial/plant/pollutant Microbial/plant/pollutant

2-5 2, 5 2

Pollutant/artifact Pollutant/artifact Microbial/plant Pollutant

6-9 6-8 7, 10, 11 7, 12, 13

Pollutant

7, 8, 10

Microbial/plant Microbial/plant

14-16 14-16

Microbial/plant Microbial/plant Microbial/plant Microbial/plant

7 7, 10, 11 7, 11 2, 4, 7

Plant Microbial/plant Microbial/plant Plant Microbial/plant

4 7, 7, 7, 7,

10, 11 11 10, 11, 17 10

References: 1: Schnitzer et al., 1986; 2: Grimalt and Saiz-Jimenez, 1989; 3: Schnitzer and Neyroud, 1975; 4: Schnitzer and Schulten, 1989; 5: Grimalt et al., 1989; 6: Saiz-Jimenez, 1994b; 7: Saiz-Jimenez and de Leeuw, 1987a; 8: Schulten et al., 1991; 9: Saiz-Jimenez, 1994 c; 10: Saiz-Jimenez and de Leeuw, 1987b; 11: Saiz-Jimenez and de Leeuw, 1986a; 12: Khan and Schnitzer, 1972; 13: Schnitzer and Khan, 1972; 14: Schnitzer, 1978; 15: Ogner and Schnitzer, 1971; 16: Saiz-Jimenez, 1994a; 17: Saiz-Jimenez and de Leeuw, 1984a. 2 Highly aliphatic polymers with long alkyl chains (sporopollenin, algaenan, cutan and suberan).

18

C. Saiz-Jimenez

this respect it is noticeable that the structural models proposed over the years shifted from fully hydrogen-bonded (Schnitzer and Khan, 1972) to fully chemically bonded building blocks (Schulten et al., 1991) in twenty years.

PYROLYSIS OF BIOMACROMOLECULES Analytical pyrolysis is not the ideal technique for investigating structural features of complex macromolecular materials, as in many instances, thermal degradation of building blocks is achieved. However, this is probably the best technique employed so far for this purpose. This is due to its facility for investigating macromolecular materials in terms of pyrolysis products. Macromolecular materials are usually recalcitrant to any other direct analytical approach, unless chemical degradation and obtention of more affordable, lower molecular weight products can be accomplished. No chemical degradation, fractionation, purification, and time-consuming derivatization of reaction products are required in pyrolysis studies, which considerably facilitate its analysis. In recent years it has been reported that analytical pyrolysis can provide important clues for understanding the chemical structure of complex macromolecules (Nip et al., 1986a; Boon and de Leeuw, 1987; Saiz-Jimenez and de Leeuw, 1987a; Pouwels et al., 1989; Tegelaar, 1990; Chiavari and Galletti, 1992), but it appears that there is a need to correctly scrutinize the information generated by this technique, as in some cases misleading inferences have been made. In order to achieve this, a brief review is presented of pyrolysis of biomacromolecules, based on literature and unpublished data, and the advantages or disadvantages of the technique are briefly discussed.

PolysaccharidesPyrolysis of plant and soil polysaccharides (Saiz-Jimenez and de Leeuw, 1984a, 1986a; Pouwels et al., 1989) results in complex pyrolysates containing a wide variety of volatile, relatively low molecular weight compounds and some other compounds, mostly anhydrosugars. There is some agreement that thermal decomposition of polysaccharide, exemplified in cellulose, is the result of two competing reactions: a dehydratation to yield anhydrocellulose, and a depolymerization of cellulose to yield primarily levoglucosan, and minor anhydrosugars components. The production of levoglucosan and other anhydrosugars is the first step in the formation of volatiles from the pyrolysis of polysaccharides, and it appears that this reaction is the main process at high temperatures and high heating rates (Pouwels et al., 1989). Several classes of compounds have been identified in polysaccharide pyrolysates. In cellulose they are represented by carbonyl compounds, mainly aldehydes


19

and ketones of different chain length, acids, furans, pyranones, anhydrosugars and phenols, which include phenol, cresol, benzenediol, hydroxybenzaldehyde and dihydroxyacetophenone (Pouwels et al., 1989). Therefore, if one mistakenly considers pyrolysis products as building blocks, one would indicate that "cellulose is a complex polymer made up of phenols, furans, pyranones, and anhydrosugars, to which side chains containing functional groups such as ketones, aldehydes and acids, are attached". Similar pitfalls appear to be widespread in pyrolysis studies of humic substances and these inferences are often made when the difficulties in considering pyrolysis products as original building blocks are not taken into account. Cellulose is a polymer of glucopyranose units linked by [3-1,4 linkages. Obviously, glucopyranose units are not encountered as such in pyrolysis, but as a complex mixture of all kinds of thermally modified products, more or less related to the original building block, as depicted above. This picture can even be complicated, as in the case of soil polysaccharides with glucose, galactose and mannose as main units, from which lignin phenols, fatty acids, and other soil admixtures could not be completely removed during the extraction and purification procedures (Saiz-Jimenez and de Leeuw, 1984a). Therefore, interpretation of the chemical nature of complex biomacromolecules in the light of information provided by pyrolysis data is a difficult task.

Proteins

Pyrolysis of proteins presents problems similar to those encountered in polysaccharides. In this case the range of pyrolysis products is increased as the variety of building units or amino acids is also increased. As opposed to polysaccharide studies, protein or peptide pyrolysis exhibits a high number of unknown compounds, some of which have been recently identified (Boon and de Leeuw, 1987; Chiavari and Galletti, 1992). Relatively low molecular weight pyrolysis products from individual amino acids have been identified in pyrolysates. Simmonds et al. (1972) investigated the products originating from the thermal decomposition of a selected group of aliphatic monoamino-monocarboxylic acids. The primary decomposition is a decarboxylation which yields amine as the major product. Subsequent decomposition or reaction of the amine leads to the formation of nitriles and N-alkyl aldimines as significant secondary products. Boon (1984) identified side chain portions of the main peptide backbone which are specific to individual amino acids, such as indole and methylindole (from tryptophan), toluene, ethylbenzene and benzonitrile (phenylalanine), phenol and cresol (tyrosine), 2-methylbutanal (isoleucine), 3-methylbutanal (leucine), 2-methylpropanal (valine) and methanethiol (methionine). More complex series of pyrolysis products have been identified in pyrolysates of standard polyamino acids (Boon and de Leeuw, 1987),

20

C. Saiz-Jimenez

namely pyrroledione and pyrrolidinedione derivatives, as well as in humic acids (Saiz-Jimenez and de Leeuw, 1986a). They are cyclization products of the aliphatic amino acids alanine, leucine, isoleucine, and valine, all of which appear combined in groups of two in the original proteins. Smith et al. (1988) identified diketopiperazines (DKPs) in the pyrolysates of dipeptides. The DKPs were produced by losing water molecules, and they in turn produced secondary (multistep) products which appear to have resulted from thermal loss of hydrogen, carbon monoxide and propene. They included acetone, imidazole, pyrrole, cyanopyrrole, pyrroline and aliphatic amides. Chiavari and Galleti (1992) provided pyrolysis data of nineteen ~-amino acids. All but glycine, serine, alanine, threonine and histidine produced diagnostic pyrolysis products. Four major thermal degradation pathways were proposed. It is demonstrated, therefore, that in general the chemical structures of pyrolysis products from proteins, peptides or amino acids are very different from those of parent amino acids. Pyrolysis of proteins clearly illustrates the complexity of pyrolysates obtained from macromolecules with many different units.

Nucleic acids

Comparatively little information exists on pyrolysis of nucleic acids with regard to polysaccharides and peptides (Posthumus et al., 1974). Pyrolysis of DNA, at 400~ reveals that major compounds were furan derivatives, including complex furan dimers and trimers (Fig. 2). In addition, pyrroles and some adenine derivatives were tentatively identified. Products introduced during the extraction procedure of commercial DNA were also evident (dialkyl phthalates, alkylphenols used as stabilizers of solvents, etc.). Posthumus et al. (1974) investigated DNA and RNA by pyrolysis-mass spectrometry and pyrolysis-field ionization mass spectrometry. Most of the peaks obtained originated from the carbohydrate moiety of the nucleic acid, which agree with the data herein reported. It was stated that the presence of nitrogen-containing compounds of low molecular weight (HCN, NH 3, nitriles, etc.) suggested a degradation of the base after expulsion from the nucleic acid skeleton.

Lignin

Lignin is perhaps one of the best biomacromolecules to be studied by pyrolysis. Several papers investigated its pyrolysis products in detail together with their significance as biomarkers (Saiz-Jimenez and de Leeuw, 1984b; 1986b). Also, parallel studies on natural and synthetic lignins were carried out. Briefly, pyrolysis of lignins yield a variety of products derived from p-coumaryl, coniferyl or syringyl alcohol units, depending on the type of lignin. Although these precursor


21

8/

6

~U9 11

14

16

15

......

1000 Scan numbers

2000

3000

Fig. 2. TIC chromatogram of DNA pyrolysate. Identification of peaks were 1: furan, 2: methylfuran, 3: imidazole methanol, 4: methoxyfuran, 5: 2,2'-methylenebisfuran, 6: 2-(2-furanylmethyl)-5-methylfuran, 7: 2,2'-oxybis-(methylene)-bisfuran, 8: 5-(2-furanylmethyl)-5-methyl-2(5H)-furanone, 9: 5-(2-furanylmethyl)-2-furanmethanol, 10: 2-(4-methyl-2-furyl)-2-cyclopenten-l-one, 11: 4,6bis(1,1-dimethylethyl)-2-methylphenol, 12: 2,6-bis(1,1-dimethylethyl)-4-ethylphenol, 13: diethyl phthalate, 14: 2,5-bis(2-furanylmethyl)-furan, 15: adeninederivative, 16: N-(2-furanylmethyl)-adenine.

phenols can be found in the pyrolysate, thus representing primary pyrolysis products, many other compounds represent intermediate steps in thermal degradation and probably secondary reactions products. Therefore, from the most simple compounds, such as phenol, guaiacol and 2,6-dimethoxyphenol, in which the propenyl side chain was split off, to the most complex precursor units, the above mentioned alcohols, a range of compounds with thermally modified functionalities, can be readily assigned to lignin phenols. However, low molecular weight compounds resulting from extensive thermal degradation are also found in the pyrolysates (e.g. methane, carbon dioxide, acetone, acetic acid, etc.). In the case of lignin, it is clear that the extent of thermal degradation in the building blocks is related to the pyrolysis temperature. A temperature above 500~ results in extensive degradation and no useful information can be obtained from the aromatic ring functionalities. Figure 3 shows the TIC chromatogram of a kraft lignin pyrolysed at 600~

22

C. Saiz-Jimenez

2

AJl

7

8

13LZJl i1800 2000

800

1000

1200

1400 Scan numbers

1600

Fig. 3. TIC chromatogram of kraft lignin pyrolysate. Peak identifications were 1: phenol, 2: p-cresol, 3: guaiacol, 4: methylguaiacol, 5: 1,2-benzenediol, 6: propylphenol, 7: ethylguaiacol, 8: vinylguaiacol, 9: eugenol, 10: propylguaiacol, 11: cis-isoeugenol, 12: vanillin, 13: trans-isoeugenol, 14: acetoguaiacone, 15: homovanillic acid.

Lipids

Pyrolysis of lipids has scarcely been considered in analytical pyrolysis as most of the lipidic compounds readily evaporate. However, there are some studies indicating that lipids, under pyrolysis, suffer cyclization reactions yielding alkylaromatic compounds. In fact, Hartgers et al. (1991) pyrolyzed the sodium salt of 12-hydroxy-octadecanoic acid which generated a homologous series of alkan-7ones and alken-7-ones with the unsaturation in the (o-position as the major products. However, benzene, toluene and, to a lesser extent, other alkylbenzenes (e.g. 1-phenylalkanes, 2-alkyltoluenes) which can be formed by cyclization and aromatization of a linear chain were also important pyrolysis products. Da Rocha Filho et al. (1993) found that hydrocracking of triglycerides and fatty acids in the presence of catalysts yields alkylcyclohexanes and alkylbenzenes. Although these cyclizations were produced under pressure the process illustrates a possible reactive pathway of fatty acids leading to alkylaromatics. It was suggested that the formation of alkylcyclohexanes and alkylbenzenes might occur with the assistance of already existing double bonds: one unsaturation is necessary for cyclization (cyclohexane formation), two or three unsaturation can lead to


23

cyclohexenes (intermediate state), which can originate aromatics. Alkylcyclohexa;~es and alkylbenzenes were mainly of carbon number equal to or one less than those of the original acids. Alkylbenzenes and alkylcyclohexanes are produced in pyrolysis not only under pressure conditions, but have also been reported in pyrolysis at atmospheric pressure (Traitler and Kratzl, 1980; Alencar et al., 1983). Although it appeared that the pyrolysis products of triglycerides and fatty acids are highly dependent on the nature and amount of catalyst used, Alencar et al. (1983) investigated the pyrolysis of vegetable oils in the absence of catalyst and under atmospheric pressure. Thus, from pyrolysis of oleic acid C~-C9 cyclohexanes, C~-C6 cyclohexenes and C4-C 7 cyclopentenes were obtained. Triglycerides, mainly based on oleic acid, are widely distributed in cyanobacteria, algae, fungi, yeast and plants, and the formation of alkylcyclohexanes is probably related to cyclizations in which the 9-10 double bond of oleic acid plays an important role. Traitler and Kratzl (1980) demonstrated that pyrolysis of tall oil fatty acids (ca. 39% linoleic acid and 40% oleic acid) in the presence of kraft lignin produced alkylbenzenes in the range C4-C8, originally absent from the mixture. Similar compounds were obtained from pyrolysis of pure fatty acids. The authors considered that the generation of alkylbenzenes resulted from the thermal cyclization, aromatization and decarboxylation of long chain fatty acids with the aid of lignin. Alkylbenzenes were also obtained when the mixture was allowed to react at temperatures as low as 160~ The formation of alkylbenzenes from aliphatic precursors, in the conditions used in analytical pyrolysis, was suggested by Saiz-Jimenez (1994b) for explaining the identification of this series of compounds in pyrolysate of humic acids and plant materials. However, a definitive research for understanding the importance of the formation of alkyl cyclic compounds upon pyrolysis of aliphatic precursors was carried out by Saiz-Jimenez (1994c). He proved the formation of alkylbenzenes as artifacts during analytical pyrolysis. Pyrolysis of triglycerides or unsaturated fatty acids in the presence of elemental sulphur yield alkylbenzenes, alkylthiophenes and alkylfurans. There is not only cyclization and aromatization reactions of aliphatic chains with unsaturated bonds, but also incorporation of sulphur and oxygen in the heterocycles. Similar reactions were produced from triglycerides in the presence of kraft lignin, which suggest that not only elemental sulphur but sulphur-containing functional groups can produce cyclization and aromatization of unsaturated fatty acids. Unsaturated fatty acids have been reported in solvent extracts of humic acids (Grimalt and Saiz-Jimenez, 1989), and sulphur content of humic fractions can reach up to 2% (Schnitzer and Khan, 1972), therefore the presence of alkylbenzenes in pyrolysates could be assigned to the formation of artifacts from aliphatic precursors.

24

C. Saiz-Jimenez

PYROLYSIS OF HUMIC SUBSTANCES Considerable efforts have been made over the last decade to understand the chemical structure of humic substances in terms of evolved pyrolysis products. From the various pyrolysis approaches, the most reliable seems to be pyrolysisgas chromatography-mass spectrometry, as pyrolysis products can be separated in the column of the gas chromatograph and identified by mass spectrometry. A comprehensive study of different soil humic fractions was reported by Saiz-Jimenez and de Leeuw (1986a) in which up to 322 compounds were identified in the pyrolysates. More specific studies on soil polysaccharides, fulvic acids and polymaleic acid were done to establish structural relationships between them (SaizJimenez and de Leeuw, 1984a). Detailed studies on the most resistant part of humic acids (hydrolysed or persulphate oxidized residues) revealed that aliphatic biopolymers, similar to those encountered in plant cuticles, could be part of this humic moiety (Saiz-Jimenez and de Leeuw, 1987a). Most of the major classes of compounds and biomacromolecules (or their primary and secondary pyrolysis products) were also apparent in pyrolysis studies of humic substances (Table 5), which indicates that pyrolysis is able to provide general information on complex mixtures of compounds, or materials. However, the weakest point is the transfer of data from pyrolysis to the whole macromolecule, provided that a chemical structure is intended to be established. This is dueto:-

-

-

Extensive thermal degradation of building blocks through secondary reactions, as exemplified in the pyrolysis of polysaccharides, proteins, lignins, etc. Compared to the tar, which is condensed onto the wall tube, and the carbonaceous residue, which remains in the pyrolysis chamber, a relatively low amount of volatile compounds can escape from the pyrolysis unit to the gas chromatograph. Serious limitations in the analytical procedure are experienced due to restrictions in the chromatographic system (peak tailing, column polarity, oven temperature limit depending on the phase employed, etc.). Because of this, it is necessary to approach the subject cautiously bearing in mind the real value of the pyrolysis products and to what extent they can be related to structural units. Taking this into account, the pyrolysis products identified in pyrolysates of humic substances, and particularly of humic acids, can be grouped into several classes, as shown in Table 6. An origin can be readily assigned to each particular class, in the light of data reported for specific biomacromolecules (e.g. Posthumus et al., 1974; Martin et al., 1977, 1979; Pouwels et al., 1989; Chiavari and Galletti, 1992). In addition, possible formation processes can be traced for the majority of products. These are: - combustion, compounds are generated upon burning in the presence of oxygen; - pyrolysis, compounds evolved upon burning in inert atmosphere; - evaporation, free compounds evaporate quickly under pyrolysis.


25

TABLE 6 Origin of major pyrolysis products recovered in humic acid pyrolysates Class of compounds Alkanes Fatty acids Nitriles Alkylfurans Alkylpyrroles Alkylpyridines Alkylindoles Methoxyphenols Alkylbenzenes Alkylnaphthalenes Alkylfluorenes Alkylphenanthrenes Process C, P, E2 C, P, E C, P C, P P P P C, P, E C, P, E C, P, E C, P, E C, P, E Origin Fossil fuels, biomass Fossil fuels, biomass Amino acids Sugars, nucleic acids Amino a., nucleic a. Amino acids Amino acids Wood, lignins Fossil fuels, biomass Fossil fuels, biomass Fossil fuels, biomass Fossil fuels, biomass References ~ 1--4 1--4 4-7 8, 9, this chapter 6, 10, this chapter 4, 7 6, 10 11-13 1, 14-17 1, 14 1, 14 1, 14

References: 1: Saiz-Jimenez, 1994b; 2: Saiz-Jimenez and de Leeuw, 1987a; 3: Saiz-Jimenez and de Leeuw, 1986a; 4: Martin et al., 1977; 5: Simmonds et al., 1972; 6: Smith et al., 1988; 7: Martin et al., 1979; 8: Saiz-Jimenez and de Leeuw, 1984a; 9: Pouwels et al., 1989; 10: Chiavari and Galleti, 1992; 11: Hawthorne et al., 1988; 12: Saiz-Jimenez and de Leeuw, 1986b; 13: Saiz-Jimenez and de Leeuw, 1984b; 14: Saiz-Jimenez, 1993; 15: Hartgers et al., 1991; 16: Traitler and Kratzl, 1980; 17: Saiz-Jimenez, 1994c. 2C: combustion, P: pyrolysis, E: evaporation. Probably compounds only listed as pyrolysis products can also be produced upon combustion and evaporate. In fact, lipids (including alkanes, fatty acids, dicarboxylic acids, ketones, hydroxy derivatives, etc.) are synthesized by microorganisms and plants, and can be found as free compounds in soils and soil humic fractions (Schnitzer and Neyroud, 1975; Schnitzer et al., 1986; Grimalt and Saiz-Jimenez, 1989; Grimalt et al., 1989; Schnitzer and Schulten, 1989). Therefore, a major part of them probably represent evaporation products when found in pyrolysates. Most of these compounds are also originated upon combustion of fossil fuel and biomass and can be, in certain areas, the reflection of an input of pollutants to the environment. A third possibility is the thermal breakdown of chains from aliphatic polymers (Nip et al., 1986b). Aliphatic and aromatic nitriles are well-known secondary pyrolysis products of amino acids and proteins (Simmonds et al., 1972), as well as the alkyl pyrroles, pyridines and indoles (Smith et al., 1988; Martin et al., 1977, 1979). Pyrroles were also identified in pyrolysis of nucleic acids. Alkylfurans are common pyrolysis products of carbohydrates and polysaccharides (Saiz-Jimenez and de Leeuw, 1984a; Pouwels et al., 1989). Rearrangements and thermal transformation of original glucopyranose units leading to most simple furans depend on the pyrolysis temperature. A relatively low temperature, as in the case of nucleic acids, provides dimeric and trimeric furyl derivatives.

26

c. Saiz-Jimenez

Methoxylated phenols have been considered to be tracers from wood materials in combustion of biomass (Hawthorne et al., 1988), and pyrolysis of wood and isolated lignins (Saiz-Jimenez and de Leeuw, 1986b). The lignin type can be discriminated through analysis of the methoxylated pyrolysis products. These compounds are also common pyrolysis products in humic acids (Saiz-Jimenez and de Leeuw, 1987a). Alkylbenzenes, naphthalenes, fluorenes, and phenanthrenes are produced upon combustion of biomass or fossil fuels. They are widespread and usually represent pollutants introduced into the environment. Thus, evaporation/pyrolysis of charred brush, or analysis of diesel soot extracts provided a wide variety of these alkylaromatics compounds. In addition, alkylcyclohexanes were identified. Pyrolysis of uncombusted brush stem also yields minor amounts of alkylaromatics, demonstrating that these compounds are related to thermal degradation, as no long chain alkylbenzenes have been reported in plant biomass (Saiz-Jimenez, 1994b). Alkylaromatics were observed fourteen years ago in pyrolysates of humic acids (Martin et al., 1979), and in subsequent analyses (Saiz-Jimenez, 1994b; Saiz-Jimenez and de Leeuw, 1986a, 1987a,b), but akey role in the humic structure was never assigned to them, as their amount was not important. The possible origin of alkylbenzenes identified in environmental samples and humic substances has been discussed in a previous paper (Saiz-Jimenez, 1994b). In the light of existing knowledge, alkylaromatics produced in pyrolysis of humic substances could be considered as artifacts originated upon cyclization and aromarization of aliphatic precursors. Table 7 summarizes the main sources. Recently, alkylaromatics identified in pyrolysates of humic acids were a subject of debate (Schulten et al., 1991; Hempfling and Schulten 1991; Schulten and Schnitzer, 1992; Schulten and Schnitzer, 1993; De Leeuw and Hatcher, 1992). TABLE 7 Origin of alkylbenzenes in soils and humic substances Origin Biosynthetic Anthropogenic CombustionIn situ

Remarks Questionable, most of the microorganismsproducing alkylbenzenes belong to extreme environments, not to soils Possible, present in sewage sludges and detergents, at least their precursors Possible, due to burning of biomass debris as an agricultural practice Possible, present in fuels and its combustion products Possible, worldwide distribution of smoke and particles Possible, production of alkylbenzenes from aliphatic precursor in analytical pyrolysis

Urban/industrial Forest fires Pyrolysis


27

These compounds have been found in pyrolysates of Canadian soil humic acids, where they represent major compounds (Schulten et al., 1991; Schulten and Schnitzer, 1992). However, they represent only minor ones in European soil humic acids (Saiz-Jimenez and de Leeuw, 1987a; Krgel-Knabner et al., 1992). Schulten et al. (1991) proposed that the alkylaromatic compounds found in the pyrolysates of humic acids represent building blocks which are released during low-temperature thermal degradation from an alkylaromatic structural network. Whilst for Schulten et al. (1991) the alkylaromatics are significant structures in humic acids, for De Leeuw and Hatcher (1992) this structural model is incorrect and misleading. These authors considered that the data discussed by Schulten et al. are highly biased and not at all representative of the humic acids analysed. They support such inferences for the following reasons: - The pyrolysis products only represent a minor part of the whole matrix. No oxygen-containing products are identified amongst the pyrolysis products of the humic acids. - The absence of oxygen-containing products may be due to the malfunctioning of the instrument or selection of nonrepresentative soil samples. Overestimation of the alkylaromatics may be caused by the presence of these compounds as such in the original samples. - The pyrolysis data do not conform to previous NMR data which clearly reveals a great deal of carboxyl groups in the humic acid studied. Schulten and Schnitzer (1993) reported a more complete version of the previous basic skeleton, in which oxygen, hydrogen, and nitrogen atoms have been inserted. Oxygen was included as carboxyls, phenolic and alcoholic hydroxyls, carboxylic ester and ethers, and nitrogen in heterocyclic structures and nitriles. This new chemical network is surprising. In fact, polysaccharides and proteins, which according to the authors are considered to be humic acid components for analytical purposes and account for about 20% of the humic acid weight (Schulten and Schnitzer, 1993), are represented in the network not as single carbohydrates or amino acids (and/or polysaccharides or proteins) but as their secondary thermal degradation products. Furan derivatives and benzenemethanol are pyrolysis products of carbohydrates or polysaccharides (Pouwels et al., 1989), or artifacts from pyrolysis (SaizJimenez 1994c), and pyrrole, pyridine, indole, and nitrile derivatives, well-known pyrolysis products of amino acids or proteins (Simmonds et al., 1972; Martin et al., 1977, 1979; Boon, 1984; Boon and de Leeuw, 1987). This is the clearest example of how inappropriate use of pyrolysis products is misleading when it is assumed that they are building blocks. In the same way there is no reason for considering other well-known pyrolysis (or combustion) products as building blocks. The model proposed by Schulten and Schnitzer (1993) is questionable, as the building blocks represented are not natural compounds, but compounds transformed through secondary thermal reactions, or, in other words, a model cannot be proposed on the basis of drastically altered thermal compounds, but on the original ones.-

28

C. Saiz-Jimenez

PYROLYSIS/METHYLATION OF HUMIC SUBSTANCES One of the most intriguing facts in humus chemistry is the finding of benzenecarboxylic acids in fulvic and humic acids upon chemical oxidations (Schnitzer, 1978), and in fulvic acids by column fractionation (Ogner and Schnitzer, 1971), which were not further apparent in other degradative methods, including pyrolysis. The presence of carboxyls groups was, however, evidenced by NMR (Hatcher et al., 1981; Saiz-Jimenez et al., 1986), and, therefore, data obtained from analytical pyrolysis do not conform with NMR data as far as functional groups are concerned. In pyrolysates of fulvic and humic acids, no carboxylic groups are found other than those of a few fatty acids (mainly the C~6 and C~8 members) and rarely benzoic and vanillic acids (Saiz-Jimenez and de Leeuw, 1984a, 1986a, 1987b). The fatty acids are mostly believed to be evaporation (and not pyrolysis) products, as they can be extracted by organic solvents or chromatographically resolved at low temperature pyrolysis (Schnitzer et al., 1986; Saiz-Jimenez and de Leeuw, 1987a; Grimalt and Saiz-Jimenez, 1989; Grimalt et al., 1989). These fatty acids cannot account for the high carboxylic carbon resonances observed in NMR studies. In a few cases these acids were identified in the pyrolysate as methyl esters, which was attributed to methylation produced by the methanol used to suspend the sample and to apply it on the wire (Saiz-Jimenez and de Leeuw, 1984a). It was suspected that conventional chromatographic conditions do not evidence the carboxyl-containing pyrolysis products. This was proved in a previous paper (Saiz-Jimenez, 1993) where solvent extraction v s analytical pyrolysis was applied to environmental samples. In fact, a complex mixture, in which fatty and dicarboxylic acids were the most abundant compounds, was resolved by solvent extraction, subsequent methylation and GC/MS analysis, but a completely different pattern was obtained upon pyrolysis for the same samples, as the series of alkanes and alkenes predominated. It was suggested that carboxylic acids decarboxylated upon pyrolysis, yielding the

0444815163 Humic Substances

Documents

Transcript of 0444815163 Humic Substances