Plant Compounds and Their Turnover and Stabilization as Soil Organic Matter

Gerd Gleixner, Claudia J. Czimczik, Christiane Kramer, Barbara Lühkcr, and Michael W.I. Schmidt

Max Planck Inslinu for

Biogeochemistry, fcna, Germany

1. Introduction 201

2. Pathways of Soil Organic Matter Formation 201

3. Stabilization of Soil Organic Matter 208

4. Turnover of Soil Organic Matter 211

5. Conclusion 213

References 213

1. Introduction

The increase in atmospheric CO, because of fossil fuel emissions has been identified as a major driving force for global climate change. Soil organic matter (SOM) is expected to be an important sink for this carbon (Ciais et al., 1995; Schimel, 1995; Steffen et al., 1998). However, at higher mean temperatures, this sink may act as additional source for C02 if it is accessible to microbial decomposition. To understand these complex interactions between stabilization and decomposition of SOM, it is crucial to investigate not only the turnover and stability, but also the chemical nature of soil organic matter.

Plant biomass, formed by photosynthesis from atmospheric COz, is the first organic substrate in the terrestrial carbon cycle (Fig. 1). The net biomass formation rate is estimated as up to 1.7 GT(1015 g) carbon per year, while the global pools of living biomass and atmospheric carbon amount to 620 and 720 Gt C, respectively. However, plants can store this carbon only temporarily. During decay, biomass is rapidly mineralized by microorganisms and less than 1% of photosynthetically assimilated CO, enters the more stable SOM pool. Despite this low rate, this pool has accumulated 1580 Gt carbon over centuries and millennia. This is more than the sum of the atmospheric and biological carbon pools. So far, the mechanisms and factors controlling the accumulation and rcmobilization of carbon in soils arc only marginally understood.

The following chapter will provide basic biogeochemical knowledge of the formation and decomposition of primary plant biomass initiating SOM formation. Better awareness of these phy-tochcmical and microbial processes is the basis for understanding soil organic matter chemistry and consequently stability. Addition ally, nonbiotic factors and processes, e.g., oxygen partial pressure, water, radiation, and fire, are involved in the formation of SOM. Of particular interest is the formation of black carbon, e.g., charred material remaining from biomass burning and soot, as these compounds are thought to be the most stable fractions of carbon in soils. This present chapter will review current knowledge on the stabilization of organic compounds. The focus will be on the chemical stability of molecules, the interactions of organic molecules with clay or metal (Fe or Al) oxides and hydroxides, and the possibility of biological carbon stabilization. Finally, current knowledge of turnover of SOM is presented.

2. Pathways of Soil Organic Matter Formation

2.1 Formation and Decomposition of Biomass

Carbon turnover in terrestrial ecosystems is mostly linked to biochemical reactions of three types of organisms. Primary biomass is produced by autotrophic organisms, mainly plants. Their biomass is transformed into new but chemically similar secondary biomass of consumers. These are connected by trophic relations in food chains and carbon recycling systems. Nonliving biomass is again mineralized by decomposers to carbon dioxide, water, and minerals. The basic biochemical pathways such as glycolysis, the pentose-phosphate cycle (Calvin cycle), and the Krebs cycle are for all organisms nearly identical. Only a few main biochemical pathways produce metabolites for biomass production, in particular cell walls.

Photosynthesis -CO,

Heterotrophic respiration

Earth Battery Organic Matter

Heterotrophic respiration

FIGURE 1 Major processes, pools, and fluxes involved in the formation of soil organic matter.

Types Compounds Plant Residues



Lignin, Tannins


FIGURE 2 Scheme of biochemical pathways and pools leading to carbohydrates, lignin, lipids, and other metabolites.



Lignin, Tannins

FIGURE 1 Major processes, pools, and fluxes involved in the formation of soil organic matter.

Most important for all organisms is the carbohydrate metabolism, which provides metabolic energy for reproduction and growth. The central part of carbohydrate metabolism is the intermediate C( pool (Fig. 2), where primary assimilates enter and the glycolytic breakdown to energy and C02 starts. This pool also provides precursors for the polymerization of structural (cellulose) and storage (starch) compounds via the C6 pool and for the regeneration of the photosynthetic CO, acceptors, the C5 pool of the Calvin cycle. Other intermediates from the Calvin cycle, e.g., from the C4 pool, and intermediates from the C, pool generate the C6-C3 pool (phenylpropanes). This pool is the starting point for the production of aromatic and phenolic compounds, e.g., lignin.


FIGURE 2 Scheme of biochemical pathways and pools leading to carbohydrates, lignin, lipids, and other metabolites.

The C2 pool, which is also part of glycolytic breakdown, is the starting point for lipid synthesis. In contrast, amino acids have several precursors and they are connected to a range of pools and metabolic pathways. To understand the structural and chemical similarity and possible differences between organisms, the following biochemical groups are described more in detail: carbohydrates, phenylpropanes and their associated derivatives, amino acids, lipids, and the major cell wall constituents.

2.1.1 Carbohydrates

Carbohydrates are the initial carbon and energy source for meta bolism and therefore the most important metabolites for biological life. They cover a broad range of molecules consisting of mainly five (pentose) or six (hexose) carbon atoms, which form oxygen-containing ring structures (Fig. 3). Their degree of polymerization is linked to different cellular and biological functions. Monosaccharides, such as glucose, are soluble sugars of the cell that are directly involved in metabolic reactions. Disaccharides, e.g., sucrose, are often involved in the transport of carbohydrates












FIGURE 3 Chemical structures of important carbohydrates. Glucose (left), cellulose (upper right), and chitin (lower right).

in plants. Most abundant in nature are polysaccharides (Fig. 3), such as cellulose starch, hemicelluloses, and chilin. Cellulose and starch are polymers of glucose, hemicelluloses are a mixture of polymers from other hexoses and pentose units, and chilin is formed from a nitrogen-containing derivative of glucose (Fig. 3). Most of these polymers either form the cellular structure or are used as storage compounds. Polysaccharides are the major structural part of plant and microbial cell walls; in microorganisms they are associated with lipids and proteins. Some carbohydrates are preferentially found in microorganisms, e.g., the hexose fu-cose, while pentoses such as arabinose or xylose are typical constituents of plants.

Generally, carbohydrates are rapidly decomposed, as they are part of energy metabolism. Therefore, in plants cellulose is protected by other compounds against breakdown. Cellulose fibers are surrounded by hemicelluloses (Barton el al., 1999), which are additionally crusted with lignin, which is highly resistant to metabolic breakdown (Paul and Clark, 1996). Additionally the nonenzymatic browning reaction (Maillard reaction) stabilizes carbohydates forming hydroxymethylfurfurals from sugars and amino acids.

2.1.2 Phenylpropanes

Derivatives of the C6-C5 pool are the most important secondary products of organisms. They are involved in the stabilization of tissues, especially lignins, in the chemical communication of plants and in important electron transport processes. Most abundant are lignins in woody plants and derivatives of gallic acids (tannins). Lignin

Besides cellulose, lignin is the most abundant constituent of wood (Killops and Killops, 1993). The production of lignin is specific to terrestrial life, stabilizing plant tissues during growth. It consists of three different alcohols from the C6-C, pool, namely coumaryl alcohol, coniferyl alcohol, and sinapyl alcohol (Fig. 4). Their relative abundance in lignin indicates for different

FIGURE 5 Partial structure of lignin, which is a polyphenol built up from units of phenylpropane derivatives (Fig. 4) by condensation and de-hydrogenation within the plant. Condensation points, lignin precursors via ether, and C-C bonds are indicated. Modified after Killops and Killops (1993).

plant types; e.g., coniferyl alcohol units dominate conifers and sinapyl alcohol is only found in the lignin of deciduous trees. The three-dimensional network of lignin is formed by polymerization of free radicals of the monomers. Mainly ether links and the C,-groups form covalent cross linkages, but additionally more stable C-C-bonds are formed in this nonspecific reaction (Fig. 5).

Lignin is only decomposed by highly specialized organisms, e.g., white rot fungi, called after the residue, which is white cellulose. Lignin decomposers have specific enzymes, namely lignin peroxidase, manganese peroxidase, and laccase, which catalyze the strongest biological oxidations (free oxygen radicals). They degrade the phenol structure to CO,, but the carbon is not used for metabolic reactions (cometabolic breakdown)(Fritsche, 1998). Once lignin is broken into monomers, microorganisms gain access to the protected carbohydrates. Since the use of oxidases for breakdown requires molecular oxygen, lignin is consequently decomposed mainly in terrestrial environments. In marine systems lignin remains undecomposed. It is therefore a biomarker for terrestrial input.

Reaction Ellagic Acid Nrf2

FIGURE 4 Chemical structure of lignin precursors, coumaryl alcohol (left), coniferyl alcohol (middle), and sinapyl alcohol (right).

FIGURE 4 Chemical structure of lignin precursors, coumaryl alcohol (left), coniferyl alcohol (middle), and sinapyl alcohol (right).

FIGURE 6 Chemical structure of tannins: gallic acid (left), ellagic acid (middle), emoldin (right). Tannins and related compounds

Tannins are widespread in nature but they are less abundant than lignin. They are part of the chemical defense and attractant system of plants, which make them less palatable to herbivores. Their chemical composition is used for chemolaxonomic classifications. Tannins are polyhydroxyaromatic acids, especially gallic acid or ellagic acid, which are, like lignin, produced via the C6-C, pool (Fig. 6). In general, these compounds are resistant to microbial attack.

Similar structures and functions are found in anthraquinones, e.g., emoldin. They are found in higher plant tissues, particularly bark, heartwood, and roots, but also in a range of organisms including fungi, lichens, vascular plants, and insects.

2.1.3 Amino Acids

Amino acids are important elements of organisms, because they are substrates for protein synthesis and enzymes. Microorganisms also liberate amino acids as exoenzymes to degrade complex organic matter outside their cells to smaller digestible monomers. Amino acids with the common a-amino-acid structure originate from various metabolic pathways (Fig. 7). Most nitrogen in organisms and in soil organic matter is found as amino groups. In contrast to plant and animal cell walls, amino acids are the major constituents of microbial cell walls. Here they are linked to a carbohydrate structure and form glycoproteins, proteoglycans, and peptidoglycans.

Proteins and enzymes are readily decomposed by proteolytic enzymes that hydrolyze the peptide links. Therefore, enzymes are often protected by secondary glycolizations, which are an integral part of cell communication. As nitrogen is generally a limiting factor for terrestrial ecosystems, most organisms store this restricted element; e.g., some microorganisms store nitrogen in the form of y-amino butyric acid.

suberin, various kinds of terpenoids, e.g., steroids or hopanoids, and tetrapyrrole pigments, e.g., chlorophyll. Lipids are often highly specific biomarkers that are used in taxonomic classifications. Glycerides

Glycerides consist of glycerin, an alcohol from the C, pool, which is esterified with three fatty acids (Fig. 8) to form fats as an energy store. In phospholipids one fatty acid is replaced by phosphoric acid. Phospholipids form membranes that isolate the inner part of cells from the surrounding environment because of their arrangement as a bilayer. The hydrophobic alkyl chains of the fatty acids are directed toward the inner side of the bilayer and the hy-drophilic phosphate ends form the surface of the membrane. Membranes are most important for cellular function and therefore are part of all organisms. The composition of fatty acids in membranes is specific to source organisms and hence is used to describe microbial community structures (Olsson, 1999). Terpenoids

The branched isoprene unit, which is also synthesized from the C2 pool, is the basic structure of terpenoids. Less condensed structures are used as volatile pheromones, e.g., jasmonic acid, menthol, or camphor, or as natural rubber material. More condensed structures such as steroids and hopanoids are part of membranes, influencing their fluidity. They are also highly specific to their source organisms. Best known are cholesterol (in animals and plants), ergosterol (hi fungi), and brassicasterol (in diatoms). Besides cellulose, hopanoids are the most abundant biomolecules.


2.1.4 Lipids

Lipids include a great variety of substances that are all soluble in nonpolar solvents such as hexane or chloroform. They are mostly synthesized from the C2 pool using two different pathways. One pathway produces long-chain molecules such as fatty acids or alcohols, and the other produces branched terpenes. The most important substance classes are glycerides and their constituents, e.g., fatty acids, waxes and related compounds, e.g., cutin and

FIGURE 7 General structure of amino acids (left) (R-groups are different for each amino acid), the amino acid serine (right), amino acids linked to a peptide chain (beyond).

FIGURE 8 Chemical structure of lipids: Triglyceride (upper, glycerin esterified with one phosphate group, and a saturated and an unsaturated fatty acid unit) and terpenoids (lower, ergosterol from fungi (left), brassicasterol from diatoms (middle), and a hopanoid from plants (right).

They were discovered in the late 60s in geological samples (Albrecht and Ourisson, 1969), but are present in low concentrations in almost all organisms.

2.1.5 Cellular Components of Terrestrial Plants

The development of terrestrial life required specialized cellular components to resist atmospheric influences like drought, high oxygen concentrations, or wind. In the last case, the three-dimensional lignin network was built to stabilize cell walls of terrestrial plants. To resist drought, two different strategies evolved. One strategy, used by plants, is to protect the exposed part of outer cells with less permeable, hydrophobic compounds, such as waxes. The other strategy, used by microorganisms, is to use gel-like substances as cell walls and extracellular polymeric substances to retain water.

Waxes, in particular cutin and suberin, are polymerized and cross linked structures of hydroxy fatty acids that are resistant to oxidation and to microbial and enzymatic attack. Cutin is found on the outer surface of plant tissue while suberin is mainly associated with roots and bark of plants. Both contain an even number of carbons in the range from Cu, to C,6. Cutan and suberan are also highly aliphatic polymers lacking ester cross linkages. They are linked by carbohydrate structures to form glycolipids that are integral parts of microbial cell walls (De Leeuw and Largeau, 1993). Microbial cell walls and extracellular polymeric substances

Biosynthetic efforts of organisms, reproduction and growth, arc connected to the synthesis of new cell walls. Outside the cell wall often extracellular polymeric substances similar to these cell wall components produce a "diffusion space" that anchors exoenzymes. While fungi use chitin, glucan, or even cellulose to form their cell walls, bacteria use more complex materials, such as glycolipids, peptidoglycans, proteoglycans, and glycoproteins. Glycolipids consist of carbohydrates and lipids, whereas peptidoglycans, proteoglycans, and glycoprotein consist of amino acid polymers and carbohydrates or chitin. The latter three differ only in their relative composition and cross linkage. They are high-molecular-weight compounds with a rigid, gel-like structure stabilizing the extracellular and intracellular reaction space.

To summarize this biochemical and structural diversity, it becomes obvious that primary biomass of plants is dominated by carbohydrates, e.g., hexoses and pentoses, and lignin. Lipids and amino acids are also present but they are generally less abundant. In contrast, secondary biomass of microorganisms is dominated by carbohydrates, e.g., hexoses and chitin, in combination with lipids and proteins. Living biomass is protected from decay by cellular defense mechanisms using, for example, tannin-like structures, whereas nonliving biomass is metabolized rapidly and similar biochemical compounds are formed at different trophic levels. Only less palatable molecules, e.g., hopanoids, tannins, or antibiotics, are resistant to decay. They are less abundant and no organism is adapted to feed on them because the energetic cost of metabolic breakdown is too high. Often these compounds are mineralized in a cometabolic way using different exoenzymes of various organisms. No energy is provided from this process. Basically, in principle, all organic substances can be broken down by microorganisms. Thus additional processes are needed to stabilize carbon in the SOM pool. These are (a) environmental conditions, (b) fires, and (c) the direct interaction of organic matter with mineral particles in soil.

2.2 Influence of Environmental Conditions on SOM Formation

Environmental factors such as ambient temperature, radiation, and the availability of water, oxygen, and anions and cations influence directly or indirectly the decay of biomass. Mostly, these factors are coupled. In peatland and marshes the abundance of water forms anaerobic conditions under which the metabolism of microorganisms shifts to less energy-efficient fermentations or to nitrate and sulfate reduction and methane production. Under these conditions, the breakdown of aromatic and phenolic substances such as lignin, which requires molecular oxygen, is not possible and nondegraded biomass accumulates. Under humid conditions the availability of water and oxygen is well balanced and the decomposition of organic matter should be high, unless the process of biomass is limited by the nutrient supply of the decomposing microorganisms or by litter quality, e.g., acidic litter of conifers. In tropical regions high temperatures increase the respiration of organisms, which results in higher energy requirements for the basic metabolism. Thus, as the SOM turnover rates are high, SOM does not generally accumulate in tropical ecosystems. Moreover, high amounts of rainfall in combination with deeply weathered soil profiles enable the transport of dissolved organic matter (DOM) to deeper soil horizons from which DOM is transported to rivers.

In arid and semiarid regions, water restricts the production and decay of biomass. Consequently, the turnover of SOM is low but erosion prevents SOM accumulation in these regions. In boreal regions cold winters and hot and dry summers restrict biomass production and decay. Production of biomass is slow but decay is even slower and hence slow accumulation of SOM occurs in upper soil horizons. Radiation directly affects the decay of biomass, forming oxygen radicals from water. This mechanism has been found to be an important factor in the oxidation of DOM in northern peallands (Bertilsson etal, 1999).

2.3 Formation of Black Carbon

Natural and artificial fires (including energy production) are an important factor in the nonbiological breakdown of biomass. Fires occur in almost all ecosystems due to natural lightning, mainly in hot and dry weather, or due to anthropogenic activities, especially landclearings. Organic remains of fires are recalcitrant structures such as charcoal and soot, both often referred to as black carbon (BC). Charcoal is the solid residue of the biomass burned, whereas soot is generated in the gas phase of a fire. The global BC production for the 1980s is estimated at 0.04-0.6 Gt per year from vegetation fires and 0.007-0.024 Gt per year from fossil fuel combustion (Kuhlbusch and Crutzen, 1995). Hence, BC

Initial stage

Smoldering combustion

Flaming combustion

Aerosols ca. 2 % Tar ca. 5-11 % Charcoal ca. 2.3 1 Ash ca. 5-10 %

Aerosols ca. 10 %

Glowing combustion

Drying and Distilling



FIGURE 9 Biomass-C partitioning in different combustion stages of natural fires with an average combustion efficiency of 80wt-%. Values for gas and tar are ased on estimations. (Fearnside el al„ 1993; Kuhlbusch et al„ 1996; Laursen et al„ 1992; Lobert and Warnatz, 1993; Simoneit, 1999.)

TABLE 1 Emissions from Biomass Combustion*

Initial Drying and Distilling




Water (ILO)

Carbon Monoxide (CO)

Carbon dioxide (CO:)

Carbon dioxide (CO,)

Methane (CH4)

Ethyne (C:H:)


Non-methane hydrocarbons

Carbon monoxide (CO)

(NMHC, mainly monounsaturated Ci_7)

Nitric oxide (NO)


Poiycyclic aromatic hydrocarbons (PAHs)

Nitrous oxide (N,0)


Ammonia (Nil,)

Nitrogen (N,)

Hydrogen cyanide (HCN)

Cyanogen (NCCN)

Acetonitrile (CH.CN)

Sulfur dioxide (S02)

Cyanogen (NCCN)


Amines, heterocycles, amino acids

(40 wt % Soot)

Methyl chloride (CH,C1)

Sulfur compounds


Aerosols (4 wt % Soot)

"The highest number of compounds is produced within the smoldering stage, whereas the highest amount of (CHNS-) emissions derives from the flaming stage of a fire (modified after Lobert and Warnatz, 1993 #5967).

"The highest number of compounds is produced within the smoldering stage, whereas the highest amount of (CHNS-) emissions derives from the flaming stage of a fire (modified after Lobert and Warnatz, 1993 #5967).

is found in the soil of almost all ecosystems (Goldberg, 1985; Schmidt and Noack, 2000; Skjemstad et ai, 1996).

The combustion process itself is rather complex, yielding mainly volatile products (50-100%), e.g., gases and aerosols; to a lower extent solids remain (0-50%), e.g., ash, charcoal, and tar. The composition and yield of these products is determined by fuel properties and the combustion process itself, e.g., temperature, oxygen concentration, and stage of combustion. The whole process can be divided into three main stages (Fig. 9). During the initial stage (up to 127°C) the fuel is dried by distillation. Water and other high-volatile compounds, mainly lipids, are lost (Table 1). In the second stage, BC is produced under smoldering or flaming conditions depending on temperature and oxygen concentration. In the final stage at temperatures above 500°C and high oxygen concentrations, BC is burned again to C02 and CO.

Smoldering conditions (below 300°C, 5-15 vol. % O,) produce mainly aerosols and gases. The latter are a mixture of molecules of low molecular weight at various oxidation stages (Table 1), whereas aerosols are more complex molecules that are often highly specific to the type of fuel (Simoneit, 1999)(Table 2). They can be unaltered fuel constituents that are released by steam-stripping or ther-modesorption (mainly from lipids) or thermally less altered pyrol-ysis products (mainly from carbohydrates and lignin). The corresponding processes are dehydration and oxidation leading to depolymerization and fragmentation reactions. Additionally molecules, e.g., less condensed polycylic aromatic hydrocarbons (PAH), dioxins, or soot are de novo synthesized (Lobert and Warnatz, 1993; Simoneit, 1999).

Flaming conditions (>300°C, >15 vol. % 02) form fully oxidized gases (Table 1). High-molecular-weight substances of the fuel

TABLE 2 Organic Aerosols from Biomass Burning*

Compound Group

Plant Source

Product properties



Thermally altered

(e.g., levoglucosan)



Thermally altered

Amino acids, amines, heterocycles


Thermally altered

(1/3 of fuel N is emittited as particle)


Epicuticular waxes



Epicuticular waxes/lipids

Thermally altered

//-Aikanoic acids

Internal lipid substances



Epicuticular waxes



Gymnosperm resins, waxes

Natural/thermally altered


Angiosperm waxes, gums

Natural/thermally altered


Internal lipid substances

Natural/thermally altered

Wax esters

Lipid membranes, waxes


Triterpenoid esters

Internal lipid substances


Poiycyclic aromatic

Multiple sources

Thermally altered

Hydrocarbons (e.g., retene)



Multiple sources

Thermally altered

'Lobert and Warnatz, 1993; Simoneit, 1999.

'Lobert and Warnatz, 1993; Simoneit, 1999.

Pi raw raw

FIGURE 10 Cross section of partially burned particles from the Cretaceous-Tertiary boundary layer, as seen by reflected light microscopy (Kruge et til., 1994). Particles show typical plant xylem cell structures.

are progressively broken down to intermediate-molecular-weight tar products by free-radical-controlled pyrolysis. The tar fraction provides the energy for the fire. It is a mixture of low-volatile pyrolysis products mainly from lignin and lipids. Depending on temperature and oxygen supply, they are further cracked, become volatilized, and are fully oxidized. Also, thermodesorption of thermally nonaltered tar products continues. In the flame, molecular rearrangements with free radicals forming soot are maximized.

The solid residue from the second stage of vegetation fires is black carbon, e.g., charcoal and soot. Charcoal is the remains of the solid fuel phase and often still holds the morphological properties of the biomass burned (Fig. 10). Its yield depends mainly on the lignin content of the fuel. From 26% to 39% of lignin char is produced. Initially higher condensed lignins from hardwood lead to higher charcoal production (Jakab ct ai, 1997; Wiedemann et al., 1988). Charcoal is mainly produced under flaming conditions (Kuhlbusch and Crutzen, 1995).

In contrast, soot is synthesized de novo within the flame. Basic soot structures are multilayers of highly condensed PAHs (Fig. 11 a, b). These multilayers are either randomly oriented, or well ordered, forming three-dimensional "onion" structures (Figs 11c, lid). The initial reaction forming aromatic soot structures involves free CH and CH2 radicals and intermediate reactive C(H, molecules. Reactions leading to further growth of the soot molecule are still rather speculative (Lobert and Warnatz, 1993). The same building reactions are described for less condensed PAFIs and fullerenes. However, like charcoal, formation, soot formation occurs mainly under flaming combustion, while smoldering leads to the production of smaller, less condensed PAHs.

BC was assumed to be stable on geological time scales, as charcoal particles of similar particle size were found at various depths of 65 X 106-year-old marine sediments (Herring, 1985). Moreover, BC was found to resist various oxidation procedures, e.g., wet-chemical or thermal treatment (Kuhlbusch, 1995). Flowever, recent carboji and oxygen isotopic studies suggest that BC degrades in soils and well-oxygenated marine sediments takes less than a cen-

Soot Structure

FIGURE 11 Soot structure as (a) produced in the laboratory (Sergides et ni, 1987), forming (b) basic structural units of 3-4 layers (Heidenreich et al., 1968), (c) randomly oriented basic structural units shown as a 2-dimensional schematic diagram, (d) onion-type particle with several condensation seeds (Heidenreich et ai, 1968).

FIGURE 11 Soot structure as (a) produced in the laboratory (Sergides et ni, 1987), forming (b) basic structural units of 3-4 layers (Heidenreich et al., 1968), (c) randomly oriented basic structural units shown as a 2-dimensional schematic diagram, (d) onion-type particle with several condensation seeds (Heidenreich et ai, 1968).

tury (Bird et ai, 1999; Middleburg et al, 1999). Free-radical mechanisms, e.g., photochemical (Ogren and Charlson, 1983) or microbial cometabolic breakdown (Shneour, 1966; Winkler, 1985), are proposed to be responsible for BC degradation. Corresponding breakdown products such as benzenepolycarboxylic acids have been found in various soils (Glaser et al., 1998; Hayatsu et al., 1979). Flowever, both the observed long-term stability of BC and the proposed degradation reactions are still poorly understood.

3. Stabilization of Soil Organic Matter

Formation and degradation of plant biomass, through biological and thermal processes, produce molecules differing in their intrinsic or chemical stability. More stable compounds are potentially preserved in the genesis of SOM, whereas others are transformed into biomass again. Additionally, the interaction of SOM with the soil matrix, e.g., mineral particles and metal oxides and hydroxides, may stabilize carbon in this pool. The formation of stable aggregates forming closed environments, or the adsorption of molecules on inner and outer surfaces of clay minerals may reduce the effect of exoenzymes. Flowever, adsorption of organic matter will not be in the focus of this chapter. However, the current knowledge of the importance of individual stabilization mechanisms are still limited.

2.4 Chemical Stability of Molecules

Chemical stability of molecules is often determined by physical and biological parameters. Physically, molecules are only destabilized when the activation energy needed for bond breaking is available. As a rough estimate, this energy can be derived from the heat of combustion or corresponding bond energy, indicating that double and triple bonds are most stable, followed by homopolar C-C and C-II bonds. Ileteropolar C-O and C-N bonds are most unstable. Aromatic and phenolic systems are further stabilized by resonance phenomena of translocated electrons. In biological systems the required activation energy is lowered by specific enzymes that catalyze the breakdown of molecules. For this purpose, the "active center" of the enzymes is formed in the geometry of the "transition state" between the two reaction stages. Further reduction in the activation energy for biological bond breaking is reached using sequences of enzymatic steps for bond breaking. However, this procedure needs a whole set of enzymes that are usually available only for the most common natural products. Additionally, a specific molecular environment is needed for enzymes to catalyze reactions. This implies that enzymes need to get access to their substrates. Consequently, the decay rate for polymeric substances is often slower than that for single molecules.

These interactions can be illustrated using 14C-labeled monomeric and polymeric compounds (Fig. 12)(Azam et al., 1985). Carbohydrates, lipids, and proteins are degraded rapidly. Most organisms have the complete set of enzymes to degrade these major metabolic products completely and to produce metabolic energy and metabolites. Also, monomeric lignin constituents mineralize rapidly. Even if carbon was labeled in the more stable aromatic ring systems, most carbon was recovered as respired C02. Only small amounts of radioactivity were found in the microbial biomass, indicating cometabolic breakdown of ring systems by exoenzymes (Azam et al., 1985). In contrast, the stability of lignin dramatically increased when polymeric substances were



Incubation time [weeks]

FIGURE 12 Decomposition of l4C-labeled lignin analog. From Haider and Martin (1975) in Paul and Clark (1996).

Incubation time [weeks]

FIGURE 12 Decomposition of l4C-labeled lignin analog. From Haider and Martin (1975) in Paul and Clark (1996).

used (Fig. 12). Obviously, the polymeric structure prevents degradation by exoenzymes.

Similar results were obtained for the degradation of BC. Here it is suggested that the resistance to microbial or photochemical degradation depends on the condensed and disordered molecular structure (Almendros and Dorado, 1999). Mainly, the high degree of internal cross linkages stabilizes black carbon.

The long-term stability of natural polymers can be assessed from geological samples (Table 3). Besides aromatic ring structures of lignin and tannin, the polyaromatic systems of steranes

TABLE 3 Occurrence of Presently Known Biomacromolecules and Their Potential for Survival during Sedimentation and Diagenesis*



Preservation potential


Vascular plants, some fungi

-/ +


Arthropods, copcpods, crustacea, fungi, algae



Vascular plants

+ + + +


Vascular plants, algae

+ + +/+ + + +


Vascular plants

+ + + +


Vascular plants

+/+ +


All organisms

-/ +


Plants, algae, eubacteria

+ /+ +


Cram-negative eubactcria

+ +

'Modified after Tegelaar (1989) in de Leeuw and Largeau, (1993)

'Modified after Tegelaar (1989) in de Leeuw and Largeau, (1993)

and hopanes are found to be resistant to biodégradation. The intrinsic stability of aromatic systems enables their preservation potential. In contrast, the cross linked structures of suberans and cu-tans reduce the biological breakdown. Beta oxidation, the usual mechanism of breakdown of these compounds, is blocked by the cross link in beta position.

It is obvious that two main factors control the stability of organic molecules. First, the intrinsic stability of organic molecules stabilizes aromatic substances and lipids. Second, the cross linkage between biomolecules inhibits the interaction of enzymes. Only reactions forming small radicals are able to break bonds of cross linked structures and release breakdown products.

3.2 Stabilization of SOM by Interactions with the Soil Matrix

Another major mechanism stabilizing SOM is the interaction between SOM and clay particles and metal oxides and hydroxides. Evidently, SOM content correlates to clay and metal oxide and hydroxide content. Furthermore, turnover rates of easily decomposable compounds are much higher in aerobic fermenters than in soils (Van Veen and Paul, 1981), and marine sediments (Keil et al, 1994). At least the thermal disruption of soil aggregates followed by rewetting increases carbon and nitrogen mineralization rates (Gregorich et al, 1989). However, this effect is mainly assigned to microbial carbon and nitrogen (Magid et al, 1999). Thus, stabilization of SOM may occur via formation of closed environments (aggregates) and via sorption of SOM to the mineral matrix (primary particles).

The interaction of organic matter with free mineral particles (sand, silt, clay) may form micro and macroaggregates. Microaggregates (<250 ¡xm in diameter) are the basic structural units in soils that are neither disrupted by water nor affected by agricultural practices. Microaggregates may form macroaggregates, larger than 250 /am. Both micro and macroaggregates contain primary particles, organic matter, and pores of different sizes (Tisdall and Oades, 1982). The mechanical stability of aggregates is determined mainly by their contents of microbial biomass and water-extractable carbohydrates (Haynes, 2000).

Primary particles can be separated using ultrasonic dispersion followed by either gravity or density separation (Amelung and Zech, 1999; Schmidt et al., 1999; Turchenek and Oades, 1979). Large, light particles are assumed to represent remaining plant biomass, whereas small, dense particles would represent highly degraded material and microbial remains (Christensen, 1996; Turchenek and Oades, 1979). This is supported by several independent observations on content and composition of SOM by particle size fractions, mainly of A horizons (Fig. 13). Concentrations of C and N increase with decreasing particle size. More than 50% of total soil C and N are found in the clay fractions and more than 90% in the combined clay-and-silt fraction (Christensen, 1996). Concurrently, C/N ratios decrease from values typical of plants (C/N ~ 40) in the sand fraction to values typical of microorganisms (C/N ~ 10) in the clay fraction (Gregorich et al, 1989). The amount of hydrolyzeable N (10-40% of total N) can mainly be attributed to amino acids and amino sugars (Stevenson, 1982) whereas in the insoluble remainder, in addition to amino functions (Knicker et al, 1997), heterocyclic N compounds were detected (Leinweber and Schulten, 1998). However, the stabilization and the sources of N-containing compounds in soils are only poorly understood. Sand-size particles are dominated by polysaccharides and nonde-graded lignins from plant residues, confirmed by bulk chemical carbon functionality (Baldock et al, 1997; Mahieu et al, 1999) and molecular markers (Guggenberger et al, 1994; 1995; Hedges et al, 1988; Oades et al, 1987; Schulten and Leinweber, 1991; Turchenek and Oades, 1979), whereas lipids are scarcely detected in this fraction. Silt-size fractions are dominated by degraded lignins, whereas plant waxes, microbial lipids, and carbohydrates dominate clay fractions (Fig. 13).

Microbial availability of organic matter for decomposition can be limited by organomineral interactions such as adsorption onto clay particles or complexation with polyvalent cations (Oades et al, 1988; Sollins et al, 1996). The incorporation of cationic amides into interlayers of clay minerals (Huang and Schnitzer, 1986), or the formation of highly persistent microbial spores may be involved in this stabilization (Danielson et al, 2000; Kanzawa et al, 1995). In alisols, SOM forms organominerals associated with clay minerals, whereas in podzols organic matter is complexed by iron. Generally, clay contents are positively correlated with SOM concentrations when other factors such as vegetation, climate, and hydrology are similar (Davidson, 1995). Recent research, however, seems to indicate the existence of a distinct protective capacity, characteristic of individual soils (Hassink et al, 1997; Hassink and Whitmore, 1997c). Some volcanic soils may have a greater stabilizing influence on organic matter than predicted from their clay contents (Parfitt et al, 1997). These observations may be explained by the presence of allophane and ferrihydrite, both of which have a large specific surface capable of adsorbing organic molecules.

3.3 Biological Stabilization of Organic Matter in Soils

Summarizing the presented results on the genesis and stabilization of SOM, it is possible to develop a conceptual model of SOM turnover including the microbial lifecycle (Fig. 14). Coarse N-de-pleted litter added to soils will be broken down by shredders, e.g., woodlice or earthworms, into smaller particles. The main result of the process is an increase of litter surface for inoculation with microorganisms, which transform cellulose and lignin into easily decomposable and N-containing microbial biomass. The inoculation takes place in the guts of these animals. In nature these inoculated feces are often "eaten" a second time to get access to the transformed food. Termites and ants, for example, have "fungal gardens" to digest biomass. Microorganisms are not able to incorporate particles into their cells directly. Only small molecules such as amino acids or sugars can diffuse into their cells. This implies that macromolecules are digested outside their cells using exoenzymes

Organics/ Living



Lignin Lignin degradation products


Microbial lipids



Nematodes Protozoa Root hairs Pollen

Fungal Plant cell hyphae remains


Microbial remains

Viruses m]

FIGURE 13 Stabilization of carbon in biological life cycles.

and that these enzymes stay within diffusion distance. Therefore, after substrate contact, microorganisms produce sticky carbohydrates (alginates, extracellular polymeric substances) to allow close contact between exoenzymes and substrate within this diffusion space (Fig. 14). Additionally this "glue" forms stable aggregates with soil minerals which exclude other microorganisms from this environment (Fig. 14). Inside the aggregates organic matter will be digested by the aggregate-forming organisms using a set of exoenzymes, e.g., cellulases, proteases, lyases. Under oxidative conditions, nonspecific oxidases are additionally able to degrade most compounds using small oxygen radicals. After substrate depletion, the carbohydrates of the diffusion space are again incorporated into the cell and highly persistent spores are formed. After spore formation, aggregates are destabilized due to changing geometry. The constituents of the aggregates are rebound as free primary particles, adsorbed spores, and recalcitrant organic matter (Fig. 14).

This conceptual model explains the existing experimental evidence. Coarse particles (sand size) from disrupted aggregates are mainly nondegraded plant remains, e.g., cellulose and lignin. As cellulose degrades more rapidly than lignin, smaller sand-size particles are relatively enriched in partially degraded lignin. In clay-size particles mainly microbial cells and spores (lipid, protein, and carbohydrate) are found. Moreover, this would suggest that en richments of nitrogen in smaller particles are microbial remains. Only highly crosslinked structures and intrinsically stable substances have the potential to survive this process. These results suggest that SOM can be stabilized biologically: active protection of carbon from decay by cellular defense mechanism in combination with storage of carbon in soil food webs are the suggested mechanisms. In order to understand and proove the underlying processes, the turnover of carbon at the molecular level using applicable tracers has to be studied.

4. Turnover of Soil Organic Matter

Appropriate tracers to investigate SOM turnover rates are l4C and l3C, the two naturally occurring isotopes of l2C. The radioactive l4C atom is continuously formed in the atmosphere by solar radiation and from the remaining 14C in organic compounds their age can be estimated. The mean natural abundance of ,3C is constant; however, small variations in the l3C/l2C ratio identify sources and processes involved in the formation of organic molecules. The best-known examples are the isotopic difference between "heavy" C4 plants and "light" C, plants and the isotopic enrichment of food chains.

Rough Drawing Microbes Gas

C\ Bacteria

\ Plant residue

Fungal and


bacterial spores

Microbial exudates


Fungal hyphae


Clay minerals

FIGURE 14 Elemental and molecular characteristics of different particle-size fractions of soil organic matter and size relation to organic matter.

FIGURE 14 Elemental and molecular characteristics of different particle-size fractions of soil organic matter and size relation to organic matter.

The l4C age of SOM in depth profiles of different soil types indicates that SOM in deeper horizons can reach mean ages between 1000 and 15,000 years (Bol ct al„ 1999; Jenkinson et al, 1999). In peat even l4C ages of 40,000 years were determined (Zimov et al., 1997). However, these ages are only mean ages as in SOM an unknown proportion of old and continuously added new carbon is measured simultaneously (Wang et al., 1996) and consequently, even in upper horizons, recalcitrant matter can be found. So far neither in bulk chemical nor in physical fractions have substances substantially older than this mean value been identified (Balesdent and Guillet, 1987; Wang et al, 1996). Recently, compound-specific l4C ages indicated for the first time that terrestrial biomarkers (lipids) are ten times older than bulk organic matter in marine environments (Eglinton et al, 1997). In general, mean ages of SOM are highest in both wet-and-cold and dry-and-hot ecosystems (see above) having high or low carbon accumulation rates, respectively, and low turnover rates. In contrast, tropical rainforests with high turnover rates have the lowest mean ages.

Additionally, the l4C signal, introduced by atmospheric thermonuclear bomb tests at the end of the 60s, can be used to investigate turnover rates of SOM. This signal is often found in the upper

5 cm of wet-and-cold soils or dry-and-hot soils, indicating that the carbon is still present after 30 years. In ecosystems with high turnover rates this peak often appears in deeper horizons. Using modeling approaches, the distribution of the l4C signal over the profile suggests that tropical soil consists mainly of SOM with a mean residence time below 10 years (Trumbore, 1993), which is in good agreement with mean residence times of 6 years determined for dead trees in rainforests (Chambers et al, 2000). For soils of temperate climates, SOM pools of different mean residence times (10, 100, 1000 years) are used to model the l4C distribution.

Carbon turnover rates can be estimated using natural labeling techniques with ' 'C in combination with vegetation shifts from "light" C, plants to "heavy" C4 plants. (Balesdent and Guillet, 1987). Coarse particles that are mainly fresh litter have mean residence times between 0.5 and 20 years, whereas the carbon in the clay fraction has mean residence times of about 60-80 years (Balesdent, 1996). Recently, the direct determination of molecular turnover rates using pyrolysis products was applied to SOM after vegetation change (Gleixner et al, 1999). This technique indicated mean residence times between 9 and 220 years for individual pyrolysis products for the first time. Most intriguing was the fact that some of the more resistant pyrolysis products were derived from proteins. This in fact supports the possibility of SOM stabilization in the form of microbial carbon, either actively protected or adsorbed on metal oxides.

5. Conclusion

The turnover and stability of SOM depends mainly on environmental and biological parameters. Either biomass production or decomposition rates are affected. Additionally, soil matrix and litter quality and fire frequencies stabilize carbon in soils. From the presented results it is obvious that ecosystems have different mechanisms for stabilizing SOM, which lead to different chemistries of the stable compounds. For a better understanding of SOM in the terrestrial carbon cycle and to identify the "missing carbon sink," some major points have to be considered:

1. The content of SOM depends mainly on four functions: (a) biomass input, (b) decomposition rate, (c) retention capacity, and (d) carbon output. All these functions arc controlled by environmental and biological parameters.

2. The chemical type of stable carbon is specific to each ecosystem. Therefore, isotopic tracers are more appropriate to understand turnover and stability of SOM.

3. Retention of carbon as microbial biomass in combination with "active" protection as "biological carbon stabilization" may be an important factor controlling carbon accumulation in soils.

To identify the corresponding processes and mechanisms we will need:

1. to investigate compound specific mean residence times of stable compounds and biomarkers,

2. to develop new soil carbon models that are able to model the molecular turnover of L1C and l4C.

The combined information will give new insight into soil carbon turnover and will help to understand and to quantify ecosystem-specific retention mechanisms for carbon. Additionally, this information may identify the carbon sink capacities of soils.


Albrecht, P. and Ourisson, G. (1969). Terpene alcohol isolation from oil shale. Science 163, 1192-t 193. Almendros, G. and Dorado, J. (1999). Molecular characteristics related to the biodegradability of humic acid preparations. Eur. J. Soil Sei. 50, 227-236.

Amelung, W. and Zech, W. (1999). Minimisation of organic matter disruption during particle-size fractionation of grassland epipedons. Geo-dermn 92, 73-85.

Azam, E, Haider, K. and Malik, K.A. (1985). Transformation of carbon-

14-labeled plant components in soil in relation to immobilization and remineralization of nitrogen-15 fertilizer. Plant anil Soil 86, 15-26.

Baldock, J. A., Oades, J. M„ Nelson, P. N„ Skene, T. M„ Golchin, A., and Clarke, P. (1997). Assessing the extent of decomposition of natural organic materials using solid-state K,C NMR spectroscopy. Atist. J. Soil Res. 35, 1061-1083.

Balesdent, J. (1996). The significance of organic separates to carbon dynamics and its modelling in some cultivated soils. Eur. J. Soil Sei. 47, 485-493.

Balesdent, J. A. M. and Guillet, B. (1987). Natural 13C abundance as a tracer for studies of soil organic matter dynamics. Soil Biol. Biochem. 19, 25-30.

Barton, S. D„ Nakanishi, K„ and Meth-Cohn, O. (1999). "Comprehensive Natural Products Chemistry," 1st ed. Elsevier, Amsterdam.

Bertilsson, S., Stepanauskas, R., Cuadros-Hansson, R., Graneli, W., Wikner, J., and Tranvik, L. (1999). Photochemically induced changes in bioavailable carbon and nitrogen pools in a boreal watershed. Aquatic Microbial Ecol. 19, 47-56.

Bird, M. I., Moyo, C„ Veenendaal, E. M„ Lloyd, J„ and Frost, P. (1999). Stability of elemental carbon in a savanna soil. Global Biogeochem. Cycles 13, 923-932.

Bol, R. A., Harkness, D. D„ Huang, Y„ and Howard, D. M. (1999). The influence of soil processes on carbon isotope distribution and turnover in the British uplands. Eur. J. Soil Sci. 50, 41 - 51.

Chambers, J. Q., Higuchi, N., Schimel, J. P., Ferreira, L. V., and Melack, J. M. (2000). Decomposition and carbon cycling of dead trees in tropical forests of central Amazon. Oecologia 122, 380-388.

Christensen, B. T. (1996). "Carbon in Primary and Secondary Organomin-eral Complexes", pp. 97-165. Springer, New York.

Ciais, P., Tans, P. P., Trolier, M„ White, J. W. C., and Francey, R. J. (1995). A large Northern Hemisphere terrestrial CO, sink indicated by the C-13/C-12 ratio of atmospheric CO,. Science 269, 1098-1102.

Danielson, J. W„ Zuroski, K. E„ Twohy, C., Thompson, R. D„ Bell, E„ and McClure, F. (2000). Recovery and sporicidal resistance of various B. subtilis spore preparations on porcelain penicylinders compared with results from AOAC test methods. /. AOAC International 83, 145-155.

Davidson, E. R. (1995). What are the physical, chemical and biological processe that control the formation and degradation of nonliving organic matter? In "Role of Nonliving Organic Matter in the Earth's Carbon Cycle, Dahlem Workshop Report 16." (R. G. Zepp and C. Sonntag, eds.), pp. 305-324. Wiley, Chichester.

De Leeuw, J. R- and Largeau, C. (1993). A review of macromolecular organic compounds that comprise living organisms and their role in kerogen, coal, and petroleum formation. In "Organic Geochemistry" (M. H. Engel and S. A. Macko, Eds.), pp. 23-72. Plenum Press, New York.

Eglinton, T. I., Benitez-Nelson, B. C., Pearson, A., McNichol, A. P., Bauer, J. E., and Druffel, E. R. M. (1997). Variability in radiocarbon ages of individual organic compounds from marine sediments. Science 277, 796-799.

Fearnside, P. M., Leal, N. J., and Fernandes, F. M. (1993). Rainforest burning and the global carbon budget: Biomass, combustion efficiency, and charcoal formation in the Brazilian Amazon. /. Geophys. Res. 98, 16733-16743.

Fritsche, W. (1998). "Umwelt-Mikrobiologie." Fischer, Jena. Ulm

Glaser, B., Haumaier, L., Guggenberger, G., and Zech, W. (1998). Black carbon in soils: The use of benzenecarboxylic acids as specific markers. Org. Geochem. 29, 811-819.

Gleixner, G„ Bol, R., and Balesdent, J. (1999). Molecular insight into soil carbon turnover. Rapid Coimnun. Mass Spectrom. 13, 1278- 1283.

Goldberg, E. D. 1985. "Black Carbon in the Environment." Wiley, New York.

Gregorich, E. G., Kachanoski, R. G., and Voroney, R. P. (1989). Carbon mineralization in soil size fractions after various amounts of aggregate disruption. J. Soil Sci. 40, 649-659.

Guggenberger, G., Christensen, B. T., and Zech, W. (1994). Land-use effects on the composition of organic matter in particle-size separates of soil. I. Lignin and carbohydrate signature. Eur. J. Soil Sci. 45, 449-458.

Guggenberger, G., Zech, W„ and Thomas, R. J. (1995). Lignin and carbohydrate alteration in particle-size separates of an oxisol under tropical pastures following native savanna. Soil Biol. Biochem. 27, 1629-1638.

Haider, Iv. and Martin, J. P. (1975). Decomposition of specifically carbon-14 labelled benzoic and cinnamic acid derivatives in soil. Soil Sci. Soc. Am. Proc. 39, 657-662.

Hassink, J„ Matus, F. J., Chenu, C, and Dalenberg, J. W. (1997). Interactions between soil biota, soil organic matter, and soil structure. In: "Soil Ecology in Sustainable Agricultural Systems." (L. Brussard and R. Fer-rera-Cerrato, Eds.), pp. 15-35. CRC Press, Boca Raton, FL.

Hassink, J. and Whitmore, A. P. (1997). A Model of the physical protection of soils. Soil Sci. Soc. Am. J., 61(1), 131-139.

Hayatsu, R„ Winans, R. E„ McBeth, R. L„ Scott, R. G„ Moore, L. P., and Studier, M. H. (1979). Lignin-like polymers in coals. Nature 278, 41-43.

Haynes, R. J. (2000). Interactions between soil organic matter status, cropping history, method of quantification and sample pretreatment and their effects on measured aggregate stability. Biol. Eertil. Soils 30, 270-275.

Hedges, J. I., Blanchette, R. A., Weliky, K„ and Devol, A. II. (1988). Effects of fungal degradation on the CuO oxidation products of lignin: A controlled laboratory study. Geochim. Cosmochim. Acta 52, 2717-2726.

Heidenreich, R. D„ Hess, W. M. and Ban, L. L. (1968). A test object and criteria for high resolution electron microscopy. /. Appl. Crystollogr. 1, 1-19.

Herring, J. R. (1985). Charcoal fluxes into sediments of the North Pacific Ocean: The Cenozoic record of burning. Geophys. Monogr. 32, 419-442.

Huang, P. M. and Schnitzer, M. (1986). "Interaction of Soil Minerals with Natural Organics and Microbes. Soil Science Society of America, Madison, WI.

Jakab, E„ Faix, O., and Till, F. (1997). Thermal decomposition of milled wood lignins studied by thermogravimetry/mass spectroscopy. /. Anal. Appl. Pyroly. 40-41, 171 - 186.

Jenkinson, D. S., Meredith, J., Kinyamario, J. I., Warren, G. P., Wong, M. T. F„ Harkness, D. D„ Bol, R„ and Coleman, K. (1999). Estimating net primary production from measurements made on soil organic-matter. Ecology 80,2762-2773.

Kanzawa, Y„ Harada, A., Takeuchi, M„ Yokota, A., and Harada, T. (1995). Bacillus curdlanolyticus sp. Nov and Bacillus kobensis sp. Nov, which hy-drolyze resistant curdlan. Int. }. Systematic Bacterial

0 0

Post a comment