Tidskrift/serie: Examensarbete - Sveriges lantbruksuniversitet, Institutionen för markvetenskap, avd. för växtnäringslära
Utgivare: SLU, Institutionen för markvetenskap, avd. för växtnäringslära
Utgivningsår: 1998
Nr/avsnitt: 108
Författare: Strand L.
Titel: Remineralisation and defixation of soil nitrogen
Huvudspråk: Engelska
Målgrupp: Rådgivare

OBS! Fel i texten kan ha uppkommit då dokumentet överfördes från papper.

Remineralisation and defixation of soil nitrogen

Line Strand


  • 1. Introduction
  • 2. The nitrogen cycle
  • 3. Ammonium
  • 3.1 Fixation
  • 3.1.1. Fixation occurs rapidly
  • 3.1.2.pH
  • 3.1.3. Temperature and moisture status
  • 3.1.4. Fractions capable of fixation
  • Fixation mechanisms
  • Fractions providing fixation
  • 3.1.5. K+-interaction
  • 3. 1.6. Slow release
  • 3.1.7. Surface versus non-surface samples
  • 3.2. Nitrification
  • 3.2.1. Optimal conditions
  • 3.2.2. Utilisation of fixed NH4+-N
  • 3.3 DCD
  • 3.4. Ammonia losses
  • 3.5. Immobilisation of NH4+
  • 3.6. Plant availability
  • 3.7. Fixation to organic material
  • 3.8. Ammonification
  • 3.9. Mobilisation and Leaching
  • 4. Nitrate
  • 4.1. Immobilisation of NO3-
  • 4.2. Denitrification
  • 4.3. Leaching
  • 4.4. Plant availability
  • 5. Materials and methods
  • 5.1 Soils
  • 5.2 Experimental plan
  • 5.2.1 Plant trial
  • 5.2.2 PE-bottle trial
  • 5.3 Analysis
  • 6. Results
  • Plant trial
  • PE-bottles
  • 7. Discussion
  • Method discussion
  • Yields
  • Nmin
  • 15N in the soil
  • Biomass N
  • Abstract
  • References
  • 1. Introduction

    This project was conducted as a part of the work to determine an optimum N level in arable soils to decrease the pressure of the environmental strain, which would lead to a better ecological and economic situation for the farmer and consumer. Since plant uptake of applied nitrogen is the prime concern in agricultural practice, it is important to know the magnitude of reactions that utilise the added nitrogen in other ways.

    To determine the optimum fertiliser rate for a field, the N nett mineralisation rate must be known. Mineralisation is here defined as ammonification plus the part of that ammonium that is turned into nitrate through nitrification. Mineralisation is dependent on how active the microbes are in the soil and the material; if it is N-poor no nett mineralisation will occur. Good methods for measuring this do not exist. Another difficulty is the ammonium-fixing capacity of the soils and how stable this fixation is. It is also difficult to measure defixed and remineralised nitrogen separately from one another. The pool of mineralised nitrogen is plant -and biomass available and changes continuously.

    During the growing season, 30-85 % of the added nitrogen could be determined in the crop and in Nmin. The rest can not easily be detected but, according to the literature, parts of this nitrogen are plant available during the next growing season and remain undetected in the soil. This nitrogen might have been fixed by clay particles if added as ammonium or immobilised by the biomass as nitrate or ammonium. Incubation studies have shown greater immobilisation of nitrogen from ammonium than from nitrate forms of nitrogen (Broadbent and Tylor 1962).

    Both defixation and mineralisation of secondary material are slow processes and difficult to measure. For example, the remobilised nitrogen could be reimmobilised before it is detected and the remobilisation rate will be underestimated. Mineralisation and defixation can only be calculated indirectly. Before tracer techniques were available, the making-up of balance sheets for fertilisation N encountered severe analytical difficulties (e.g. Jansson, 1962). In the present study 15N labelled fertilisers were used.

    The two objectives studied concerned the stability of N-fixation and N-immobilisation, and whether defixation of fixed ammonium is affected if more N is added.

    2. The nitrogen cycle

    The universal N cycle is built up of three interdependent partial cycles having one or more common pathway (Campbell 1978; Jansson 1971). These three subcycles may be called the elemental cycle (E), the autotrophic cycle (A), and the heterotrophic cycle (H). They are schematically illustrated in Fig 1.

    Fig. 1: The universal N cycle divided into its three subcycles: the elemental (E), the autotrophic (A) and the heterotrophic (H) (Jansson and Persson, 1982)

    The elemental subcycle (E) includes the connection of biological life to the dominating N pool of the earth, the atmosphere. Its specific N pathway is biological N2 fixation and denitrification under certain environmental conditions (restricted O2 supply). Nitrogen fixation (N2 + 3H2 ⇒ 2NH3) is the result of activity by nonsymbiotic and symbiotic microbes and the symbiotics have an higher fixing capacity (e.g., in clover with 100-400 kg N/ha, compared with nonsymbiotics 5-10 kg N/ha (Mengel and Kirkby, 1987)). Synthetic nitrogen fertilisers are made in the same way and this is a very big part of the world input of N from the atmosphere. Denitrification (NO3 ⇒ NO2 ⇒ NO ⇒ N2O ⇒ N2) is done by facultative aerobic microbes that use NO3- during O2 -shortage. Between 5-50 % of the N can get lost through this process which is promoted by high moisture conditions, neutral pH, high temperatures, nitrate, low rate of O2, and soluble organic matter (Mengel and Kirkby, 1987).

    The autotrophic subcycle (A) includes the activities of green plants, their phototrophic binding of solar energy, and build-up of primary organic N substances. Rao and Rains (1976) proved that NH4+ is preferred to NO3 at pH levels higher than 6.8, and vice versa for pH under 6.8, and at pH 6.8 the plants take up equal quantities if nitrate and ammonium are provided in equal amounts.

    Some species of bacteria can oxidise ammonium to nitrate. This is done in two steps, the first by Nitrosomonas, Nitrosolobus and Nitrosospira; 2NH4+ + 3O2 ⇒ 2HNO2 + 2 H+ + 2H2O and the second by Nitrobacter species; 2HNO2 + O2 ⇒ 2 NO3+ + 2 H+ These species are autotrophic bacteria that use CO2 as C-source and utilise the energy from the oxidation of organic salts. They are obligate aerobes and O2-supply is their limiting factor.

    The heterotrophic subcycle (H) is determined by the activities of heterotrophic microorganisms. The specific ecological characteristic of this cycle is mineralisation, energy dissipation from organic matter, whereby the nitrogenous organic subcycles are converted into NH3 and NH4+. Beck (1983) found a very close relationship between the biomass and the rate of mineralisation (r=0.96). Predominant sources of mineralisation are amino-N and polymers of amino sugars present in the soil microbial biomass. Influx into this pool occurs with the application of organic matter (green manure, straw), organic carbon released by roots, N2 assimilation by leguminous species and inorganic nitrogen (Mengel, 1996). The pH measurement gives a rough measure of the mineralisation status in the soil, because the enzyme activity and processes in the soil are pH dependent and it is known that in acid soils the mineralisation of organic nitrogen is retarded or even blocked (Kuntze and B artels, 1979). Also temperature and soil moisture influence N mineralisation. According to Schnurer and Rosswall (1987), fungi have a higher N mineralisation rate than bacteria. Ammonification occurs only together with an easily digested C-source:

    1. soil organic N ⇒ RNH2 + CO2 + Additional products + energy

    2. RNH2 + H2O ⇒ NH3 + ROH + energy

    N immobilisation is defined as the transformation of inorganic N compound (NH4+, NH3, NO3-, NO2-) into the organic state. Plant uptake and assimilation of inorganic N compounds and N2 fixation by autotrophic and heterotrophic soil organisms are also a variant of immobilisation but are excluded from the definition (Jansson, 1971).

    The function of all three subcycles is dependent on the mineralised N. Partly, but invariably, this mineralised N will be immobilised in the heterotrophic subcycle, partly taken up by plants of the autotrophic subcycle, and partly nitrified and denitrified in the elemental subcycle. The continuous transfer of mineralised N into organic products of synthesis and of remineralised N back into inorganic decay products - underlying the building-up and dying away of the heterotrophic biomass - can be defined as MIT (mineralisation-immobilisation turnover) (Campbell, 1978; Jansson, 1958) The interactions between MIT and the activities of green plants have already been touched on; plants serve as energy sources for heterotrophic microorganisms, and they compete for the mineralisation outflow (Jansson and Persson, 1982). The net effect of MIT - the difference between the two opposing processes of mineralisation and immobilisation - will be dependent on the energy supply to soil micro-organisms. A supply of energy-rich material to the soil (for example, by glucose and straw addition) has increased the difference by lowering plant uptake of tagged N (Jansson and Persson, 1982). Mineralisation and availability to plants of soil N is generally increased following application of fertiliser N, a phenomenon termed as "priming" effect (Westman and Kurtz (1973). Jenkinson et al. (1985) introduced the term "added nitrogen interaction" (ANI) for the so-called "priming effect".

    The ANI may be real if the added N causes a change in the processes that move N into or out of a particular compartment (e.g. increased root proliferation resulting in an increase in the soil volume being explored), and apparent if it is caused by pool substitution or isotopic displacement (Jenkinson et al. 1985). ANI is mainly due to pool substitution which is strongly influenced by the rate at which applied N is immobilised in the soil. Therefore ANI is directly proportional to the rate of N immobilisation. ANI experiments with 15N labelled fertilisers often show that plants given fertiliser N take up more N from the soil than plants not given N.

    Microbial immobilisation of N, whether driven by the decomposition of soil organic matter or by the decomposition of plant roots, can lead to pool substitution and is the dominant cause of apparent ANI (Jenkinson et al. 1985).

    A major conclusion is that the "ammonium phase of soil N is an integral part of the internal soil N cycle, subject to continuous consumption and renewal, whereas the nitrate phase becomes a more or less temporary storage pool of surplus inorganic N not needed in the internal N cycle" (Jansson, 1958).

    3. Ammonium

    3.1 Fixation

    For most agricultural soils, from a few to about 10 % of the N in the plough layer occurs as fixed NH4+ On a profile-depth basis, the fixed NH4+-N accounted for nearly 50 % of the total soil N (Young, 1962). Relative and even some absolute increases in fixed NH4+ content with depth seem common, and since the organic matter content of most soils decreases sharply with depth, the percentage of total N as fixed NH4+ becomes proportionally greater. As much as 90 % of the N in some lower soil horizons may occur as fixed NH4+. However, most of the native fixed NH4+, particularly in the lower horizons of residual soils, probably has been inherited from in situ parent material (Young and Aldag, 1982).

    The content of fixed (non-exchangeable) NH4+ in soil depends on the soil's parent material, soil texture, clay content, and the clay's mineral composition (Niederbudde, 1983). In central European loess soils, about 65 % of the non-exchangeable NH4+ is interlayer NH4+ (Nieder et al.,1996).

    Reported values for fixed NH4+ range from 0 to > 1,000 ppm, with the lowest values being typical of sandy surface soils and the highest being typical for clayey subsurface horizons. However, the surface layers of some soils will contain as much as 200 ppm N of fixed NH4+, equivalent to 750 kg N/ha for a plough layer of 25-cm depth and a soil bulk density of 1.5 (Young and Aldag, 1982). Fixed NH4+ may provide a reserve N supply that is released slowly for crop use (Mengel and Scherer, 1981; Mengel et al., 1990; Haas et al., 1993). Kudeyrov (1981) observed seasonal changes in non-exchangeable NH4+ in different soils depending on N fertiliser applications, N uptake by the crops, and microbial activity.

    3.1.1. Fixation occurs rapidly

    Fixation appears to be very fast. Drury and Beauchamp (1991) found that about half of the 15NH4 fixation occurred within 6h after application in both of their soils. The fixed-15NH4 pool appeared to be a slow-release reservoir, with fixed NH4 release being slower than the rate of fixation. Kowalenko (1989) showed that 36 % of the applied NH4+ was immediately fixed by the clays but after 14 d in the field it decreased to less than 1 %. Sowden et al. (1978) found that most of the NH4+ fixation was complete in 2h (43 out of 55 %) with a much slower rate of fixation in the next 3 days.

    Fixation rate, being controlled mainly by ion diffusion, is highest in periods immediately after NH4+ addition and slows down as the equilibrium point is approached (Nommik, 1965).

    The amount of NH4+ fixed has been generally found to increase with an increase in the amounts of NH4+ added (Nommik, 1965; Opuvaribo and Odu, 1974; Sowden et al. 1978; Drury and Beachamp, 1991). Sowden et al. (1978) have shown that the fixing capacities of the soils are not closely related to their contents of native fixed NH4+. There is no relationship between the total N and the fixed N, and again the fixing capacity is not related to the native fixed NH4+.

    3.1.2. pH

    The pH influences the fixation of island, according to Olu Obi et al. (1986), it seems that soil pH rather than OM or clay content are responsible for the fixation of ammonium because fixed nitrogen from ammonium-yielding sources was not related to the organic matter nor the clay content of the soils in his trials. Nommik (1957) showed that there was a tendency for NH4+ fixation to increase with increasing pH of the soils. Soils with pH values lower than 5.5 generally showed very low fixation. This is underlined by Wiklander and Anderson (1959) who demonstrated that the ability of soils to fix NH4+ and K+ was measurably increased by liming.

    The amount of nitrogen retained from ammonium-yielding fertilisers decreases as the pH of experimental soils increases. The average amount of nitrogen retained, which was nearly 10-15 % with acid or near neutral soils, decreased to 5-10 % for neutral to alkaline soils (Olu Obi etal., 1986).

    3.1.3. Temperature and moisture status

    Drury and Beauchamp (1991) reports that low temperatures do not affect the quantity of added 15NH4 fixed, and add that ammonium fixation is relatively insensitive to temperature. Nommik and Vahtras (1982) found that within the temperature interval of 0-60° C, fixation rate increases with increasing temperature. The higher temperature, resulting in the fixation rate increasing, might explain why water is lost. Black and Waring (1972) and Opuwaribo and Odu (1975) found that the fixation of NH4+ increased as much as 3-to 10-fold following air-drying, and Nommik and Vahtras (1982) suggested that since freezing, like drying, removes water from the system, it may influence the amounts of NH4+ fixed.

    In many studies, nitrogen is added with water and then the soil is air-dried. In the field, however, the soil probably does not dry to the same extent naturally. If added NH4+ requires drying to make it stable against microbial activity, then the relatively constant moisture content of the soil throughout this field study allowed effective competition for NH4-N by nitrifiers against clay fixation (Kowalenko 1989).

    Sowden et al. (1978) found that for soils that did fix NH4+, a larger amount but a lower percentage was fixed as the amount of added NH4+ increased. The amount of water in which the NH4+ was added had a slight effect on the percentage fixed, being a little lower as the amount of water increased.

    3.1.4. Fractions capable of fixation Fixation mechanisms

    Young and Aldag (1982) suggest the following explanation of the fixation mechanism; The entrapped NH4+, like K+, fills interstitial lattice voids formed by hexagonal oxygen rings, and it serves to neutralise negative charges arising from isomorphous substitution. This nonexchangeable NH4+ resists removal by neutral salt solutions typically used for extracting exchangeable ions; thus, it is considered generally unavailable to plants and micro-organisms. It is proved that clay and clay loam soils generally contain more fixed NH4+ than silty loams, which in turn contain more than sandy soils. Higashi (1953) showed that accessibility of fixed NH4+ was mainly dependent on the nature of the fixing material. Sowden et al. (1978) found later, after analyses of the clay mineralogy of some soils, a fairly close relationship between venniculitic clay content and NH4+ fixation.

    Nommik and Vahtras (1982) explain the mechanism of fixation as follows; It is now generally accepted that the soil's capacity to fix K+ and NH4+ is related to the presence of clay minerals of the three-layer or 2:1 type. Each unit cell of these minerals consists of an octahedral A1-O-OH sheet "sandwiched" between two tetrahedral Si-O sheets. As a result of isomorphous substitution of A13+ and Si4+ by cations with lower valence during crystallisation, the lattice obtains an excess negative charge, which will be balanced by the cations, e.g. Ca2+, Mg2+, K+, Na+, H3O+, either inside the crystal or outside the structural unit. The magnitude of the layer charge plays a dominant role in determining the strength of bonding in the basal plane. The greater the layer charge, the stronger the interlayer bond. With minerals of high layer charge, such as micas, the bond energy is so great that polar molecules cannot enter between the basal planes. Early investigation revealed that the saturation of vermiculites and degraded illite (under special conditions also montmorillonite) with the NH4+ and K+ leads to collapse of the crystal lattice. The K+ and NH4+ are, thereby, "trapped" between the silica sheets, and are largely withdrawn from exchange reactions.

    The molecules NH4+ and K+ are held harder because the exposed surface (and surfaces between the sheets) of three-layer minerals consists of oxygen ions, arranged hexagonally and the opening within the hexagon is equal to the diameter of an oxygen ion (approx. 2.8 Å). Ions having a diameter of this magnitude (e.g., NH4+ and K+) will fit snugly into these lattice holes and such ions will be held very tightly as they come closer to the negative electrical charges within the crystal (Nommik and Vahtras, 1982).

    Fig. 2: Schematic picture showing the different forms of NH4+ on illite (modified after Wiklander, 1958, and Schachtschabel, 1961).

    Cations in an expanded crystal lattice are readily replaced by cations which leave the lattice in an expanded state, but only with difficulty by cations with contracted lattice. Thus, the interlayer NH4+ in a contracted lattice of vermiculite, illite, or montmorrillonite can be slowly replaced by such cations as Ca2+, Mg2+ and Na2+, but hardly by K+ (Nommik and Vahtras, 1982).

    The differential behaviour of 2:1 type clay minerals depends on whether the main part of the negative charge on the lattice originates from the octahedral Al-layer or from the tetrahedral Si-layer. The force of attraction between the positively charged interlayer cations and the negative charges in the lattice will be greatest when the negative charge results from an isomorphous substitution of A1 for Si in tetrahedral layers, given a shorter distance between the interlayer cations and the negative sites of the lattice. Substitution in tetrahedral Si-sheets accounts for 80 to 90 % of the substitution in vermiculites, about 65 % or more in illites, and < 20 % in montmorillonites (Nommik, 1965). Fractions providing fixation

    Clay minerals of the 1:1 type or kaolin group fix little ammonium because the 7.2 Å interlayer distance cannot be expanded by hydration because of hydrogen bonding. In the 2:1 type of clay, minerals such as montmorillonite, beidellite, vermiculite, illite, and intergradients vary in their ability to fix ammonium (Allison et al., 1953; Bailey, 1942). From the behaviour of these specimen clay minerals one would expect the native fixed NH4+ content and fixing ability to follow the order: illites > vermiculites > smectites > kaolinites. By contrast, vermiculites showed the greatest additional fixing capacity (Young and Aldag, 1982), but some authors consider that the vermiculites have the greatest capacity to fix K+ and NH4+. Illite may or may not fix, depending on the degree of weathering and K+ saturation of the lattice. Montmorillonites do not fix NH4+ under moist conditions. Kaolinites are generally considered as non-fixing minerals (Nommik and Vahtras, 1982).

    The fine silt fraction (2-5 &b.mu;m) fixes a larger amount (whole soil basis) than the fine clay fraction (< 0.2 &b.mu;m). The coarse clay fraction (0.2-2 &b.mu;m) fixes the most NH4+ added as well as being the fraction containing the most native fixed NH4+. Sand size fractions can contain native fixed ammonium and are capable of fixing a small amount of added ammonium. Measurements of recently fixed NH4+ in various particle sizes covering four time intervals (up to 40 mo) of field weathering under fallow showed that the 0.2-2 &b.mu;m fraction was quantitatively the most important, the < 0.2 &b.mu;m fraction most readily released recently fixed NH4+, and the 2-5 &b.mu;m fraction was the most stable with respect to recently fixed NH4+ (Kowalenko and Ross, 1979). Sowden et al. (1978) found that the amounts of native fixed NH4+ are greater in the < 0.2 &b.mu;m clay than in the 2 &b.mu;m clay.

    3.1.5. K+-interaction

    The depressive effect of K+ on fixation of NH4+ has been the subject of a number of investigations. It has been found that the effect may vary, depending on whether K+ is added simultaneously, prior to, or after the addition of NH4+. When K+ and NH4+ were added simultaneously in equivalent amounts, the soil fixed NH4+ and K+ in the proportion 3.4:1.

    According to Nommik (1965), the preferential fixation of NH4+ was characteristic not only of relatively low concentration levels, but also of rather high NH4+ and K+ addition rates. One interesting and somewhat unexpected consequence of the preferential fixation of NH4+ was that within a narrow concentration interval the simultaneous addition of K+ led to a small but significant increase in the NH4+ fixation.

    When K was added prior to NH4, or vice versa, the fixation of the cation that was added first partially blocked the fixation of the cation that was added subsequently (Beauchamp, 1982). Osborne (1976) found that K+ preaddition reduced NH4+ fixation after 24 h but was ineffective after 72 and 120 h. Nommik (1965) found that a vermiculite-containing clay soil fixed 10-15 % more NH4+ than K+.

    Numerous investigations have demonstrated that cations that do not contract the crystal lattice may influence the capacity of the mineral to fix NH4+. Addition of a cation of high replacing power will lead to a decrease in exchangeable NH4+ and, consequently, to a lower fixation of added NH4+ or, alternatively, to release of some of the already fixed NH4+. The greatest depressive effect on NH4+ fixation is exerted by trivalent cations and by H+ The considerable replacement of native fixed ammonium with added NH4+ observed by the tracer method, implies that ammonium is more efficient than K+ in replacing fixed NH4+ (Kowalenko and Ross, 1979).

    3.1.6. Slow release

    It appears that the fixed NH4 pool acts as a slow-released reservoir after an initial moderately fast release. The rate of fixed 15NH4 release is much slower than the rate of fixation (Drury and Beauchamp, 1991; Kowalenko and Ross, 1979).

    Kowalenko and Cameron (1976) showed data supporting the theory that clay-fixed ammonium was in equilibrium with soluble and exchangeable NH4+. This suggests that release of clay-fixed ammonium occurs when soluble and exchangeable NH4+ are depleted by, for example, nitrification or by plant uptake. This implies that biological processes only indirectly have an effect on release of fixed NH4+ but Green et al. (1994) observed a release of 45 % of the fixed N in 15 days and this and other observations indicated a relatively rapid release of fixed NH4+ during (due to) nitrification.

    3.1.7. Surface versus non-surface samples

    Considering a 30-cm layer with a specific weight of 1.2 kg/dm3, an interlayer NH4+-N content of 200 mg/kg soil corresponds to 720 kg/ha. If this amount were available to plant uptake it would be sufficient to satisfy the N crop requirements for many years (Marzadori et al. 1996). The relative difference between the surface and subsurface samples in their ability to fix added NH4+-N suggested that fewer sites not occupied by native fixed NH4+ are available for fixation of added NH4+ at the surface of the profile. The slightly lower cation exchange capacity of the subsurface sample may be related to the lower organic matter content (Kowalenko and Ross, 1979).

    Axley and Legg (1960) showed an increase in the interlayer NH4+-N during the autumn/winter period. This is a positive factor for two reasons: (1) it is a reserve of N which may become available when needed by plants and (2) it may prevent or reduce nitrification and thus nitrate losses due to leaching (Sowden, 1976).

    3.2. Nitrification

    3.2.1. Optimal conditions

    In soil, two main microbial processes are responsible for the consumption and removal of the NH4+-N: nitrification and assimilation by heterotrophic microflora (Nommik and Vahtras, 1982). Incubation of soil treated with (NH4) 2SO4 at optimum temperatures and moisture contents usually results in an exponential decline in the exchangeable NH4+ concentration with a simultaneous increase in the NO3--N concentration (Kowalenko and Cameron, 1976). The more optimal the conditions are, the more rapidly the nitrification will proceed. A preliminary experiment showed that nitrification of the exchangeable NH4+ was nearly complete after 25 days (Sowden 1976) and Kowalenko (1989) found that nitrification of applied NH4+-N proceeded rapidly with essentially complete conversion of the 120 kg NH4+-N/ha added occurring within 14 d of application.

    The niftrifying bacteria have their optimum temperature at 26° C (Beck, 1983). The often neglected moisture-temperature interaction is very important. Due to this interaction, optimum moisture content for the activity of nitrifying population appeared to be dependent upon temperature (Kowalenko and Cameron, 1976). Campell et al. (1974) conducted a series of laboratory and field incubation studies and found that neither ammonification nor nitrification showed a quantitative relationship to temperature (10-35° C). They indicated that the effect of temperature was most likely involved in the wetting-drying processes which would suggest that the interaction of temperature and water content might be more important than temperature alone.

    A pH of 6.8 is mentioned as being optimal (Mengel and Kirkby, 1987). Trehan (1996) gave a good example of this in his trials where nitrification of 15NH4+-N was slow, as might be expected in a soil with pH 4.2. About 51 % of the labelled mineral N was present as 15NO3- after 20 days. According to Schmidt (1982), the rate of highest nitrification appeared to be mainly dependent on the pH of the soil.

    3.2.2. Utilisation of fixed NH4+-N

    The fixed NH4+-N is probably too tightly bound to be directly available to the nitrifying micro-organisms, but as the soluble and exchangeable NH4+ is depleted by action of the nitrifiers (or immobilisers), desorption of tightly held NH4+-N would be favoured according to the proposed Langmuir kinetic model relationship. Nitrifying bacteria could be intermediaries in making non-exchangeable NH4+ available for crop uptake. However, the desorption of fixed NH4+-N would be much slower than its adsorption (Kowalenko and Cameron, 1976; Drury et al., 1991; Trehan, 1996). An incubation carried out by Bower (1950) revealed that 13-28 % of the non-exchangeable NH4+ was nitrified over a 4-day period. Prolonged incubation did not result in any further release of fixed NH4+.

    Low availability of added NH4+ to nitrifiers in highly fixing Swedish soils has been reported by Nommik (1957) and Jansson (1958). According to Nommik, availability of fixed NH4+ was dependent, among other things, on the degree to which the total fixing capacity of the soil was saturated by NH4+. Availability increased with increasing saturation of the soil's NH4+ -fixing capacity (Nommik and Vahtras, 1982). The percentage of the added NH4+ nitrified during incubation showed an unmistakable tendency to decrease with increasing NH4+-fixing capacity of the soil. In highly fixing soils, only 10-15 % of the added NH4+ was recovered as NO3- during an incubation period of 50 days (Axley and Legg, 1960).

    The low availability of fixed NH4+ to nitrifiers, reported by several workers, may be considered inconsistent with the theory of reversibility of the fixation-defixation reaction. However, it should be borne in mind that, in soil systems, K+ may interfere with defixation of NH4+. The interfering effect of K+ is evidenced by the fact that the availability of fixed NH4+ in K+-free clay minerals, such as vermiculite and montmorillonite, is high and comparable with that of exchangeable NH4+ (Nommik and Vahtras, 1982).

    3.3 DCD

    To include dicyandiamide in a treatment maintains more N in the NH4 form, resulting in greater crop uptake and more microbial immobilisation of fertiliser. Dicyandiamide has the chemical formula NCNHC(=NH)NH2 and has a nitrogen content of 67 %. It is a nitrification inhibitor and, if adding in an amount of 5-10 % of the added N, it is sufficient to inhibit nitrification over a period of six weeks (Ammberger and Gutser, 1978). The effectivity of dicyandiamide is influenced by temperature, moisture and organic matter. The breakdown of dicyandiamide is accelerated at increasing temperature (highest between 30-40° C) and if the organic matter content is high (Vilsmeier and Amberger, 1980).

    3.4. Ammonia losses

    Losses of ammonia occur with pH around 8 (Mengel and Kirkby, 1987). Many authors have been unable to demonstrate losses (e.g. Sowden, 1978), whereas others suggest that a deficit of nitrogen is caused by denitrification or ammonia losses (Hanschmann and Lippold, 1987).

    3.5. Immobilisation of NH4+

    It is found that ammonium is preferentially utilised by micro-organisms, and the tracers and mathematical equations could be used to measure and describe the nett mineralisation-immobilisation method (Paul and Juma, 1981). Neeteson et al. (1986) observed that more than one half of the added NH4+ disappeared and reappeared after 5 weeks. Since the clay minerals in these soils were not found to fix added NH4, they postulated that the immobilisation occurred as an osmoregulation mechanism and then NH4 was remineralised when the microbes died and decomposed. In another trial, incorporation of 15NH4+ into organic matter in unautoclaved soil accounted for 6-13 % of applied N, whereas in autoclaved soil, only 2-7 % was recovered in the organic N pool (Trehan, 1996).

    No consistent relationship between microbial biomass and mineral N has been observed, possibly because an excess uptake of N by microbes may occur only when large amounts of available C substrates are also added (Haas et al., 1993; Francis et al., 1993). When sufficient amounts of organic C are available, soil micro-organisms compete with clay minerals for NH4+-N (Kowalenko, 1981). Numerous trials have been conducted to demonstrate if fixed NH4-N could be used by heterotrophic microbes. Kudeyarov (1991) observed that about 30-60 % of the non-exchangeable NH4+ is potentially available for plants and microbial biomass.

    The remaining 40-70 % is truly fixed. In this study, 15-25 % of the non-exchangeable NH4+ was released during the growing season. This relationship suggests that, apart from plant cover and N fertilisation, the microbial biomass exerts a great influence on the dynamics of non-exchangeable NH4+ In comparison with the nitrifying organisms, the heterotrophic micro-organisms and higher plants have a high K+ requirement, which suggests that the two groups of organisms groups are different as regards their ability to utilise fixed NH4+ in soils. As concerns the heterotrophic soil microflora, their requirements for assimilable N and K+ are highly dependent on the supply of available energy. In soils containing excessive amounts of available energy, exchangeable K+ and NH4+ are generally held at low levels, a condition that is favourable for the release of fixed NH4+ and K+ from clay minerals (Nommik and Vahtras, 1982).

    According to Jansson (1958), the heterotrophic microflora appeared as a weak and ineffective extractant of fixed NH4+, in some cases being equivalent to, and in other cases even weaker than, a 1N KC1 solution. The ability of the microbes to utilise NH4+ is markedly reduced by addition of K+, with only about 10 % of the fixed NH4+ being available (Nommik, 1957).

    Cells do not accumulate high concentrations of NH4+ and NH4+-N is mainly incorporated, by an energy-dependent process, into the amide group of glutamin, from which it enters the major biochemical pathways of the cell. It is thus possible that exocellular 15NH4+ may exchange not only with the small pool of unlabelled NH4+ inside the cell but also, through the operation of enzyme-mediated exchange reactions, with endocellular glutamine or aspargine (Jenkinson et al., 1985).

    Results show that biomass N is a very constant percentage of the total soil N in soils (ranging from about 2-6 % with a mean of 4 %) despite differences in their previous agricultural history (Brookes, 1985b; Paul and Juma, 1981).

    3.6. Plant availability

    The relative stability of recently fixed NH4+ in the various fractions appeared to be different in the presence of a crop. Release of fixed NH4+ from various particle size fractions was slightly different in the presence of plant growth than under fallow conditions (Kowalenko and Ross, 1979). Sowden et al. (1978) also found that cropped areas had lower native fixed NH4+ contents than areas that were in fallow.

    The presence of the root can influence the release of interlayer NH4+-N in two different ways: one direct because the release is favoured by the reduction of N available caused by plant uptake in the soil zone penetrated by roots (Wehrmann and Colewey-Zum Eschenhoff 1986), and the other indirect due to microbial N uptake stimulated by the exudates of plants produced during their growth. The carbon released from the roots serves as an energy source for the normally energy-limited bacteria that can mineralise the organic matter but also use the interlayer NH4+ as a source of N for their growth (Marzadori et al., 1996).

    In spite of there being a chronic N deficit, it is not possible to prove a remineralisation of the fixed N by the plants. This is in contrast with what Mengel and Scherer (1981) found in a trial with an alluvial soil, where they found a significant decrease of non-exchangeable ammonium during the main growing period. Later, Scherer (1984) showed that during the growing season 72 % and 65 %, respectively, were defixed in an alluvial soil and a loess soil. In two soils containing 3.5 and 4.0 me of fixed NH4+/100g, respectively, a maximum of only 10 % of the fixed NH4+ was recovered by barley plants. In a parallel study, from 13 to 28 % of the nonexchangeable NH4+ was released through nitrification, indicating that plants were unable to utilise fixed NH4+ in excess of that made available through nitrification (Bower 1951).

    The general experience in German agricultural practice is that soil N is more available to plants in wet than dry summers. Schachtschabel (1961) suggests that this may be due to easier release of fixed NH4+ from the soil during the wet season. When K+ is added in combination with ((NH4)2SO,, uptake of NH4+-N by the crop decreases markedly (Atanasiu et al., 1968). Marzadori et al. (1996) showed that the concentration of interlayer NH4+-N showed significant variations in the sites where the soil was heavily penetrated by tree roots, whereas in the sites where the soil remained free from tree roots no significant variation was found.

    Nommik (1958) suggests that this difference between the release of fixed NH4+ to higher plants and to microorganisms is due to the greater withdrawal of K from soil solution by plants, resulting in more favourable conditions for release of fixed NH4+. However, only under the most exceptional circumstances will sufficient unlabelled fixed NH4+ be displaced from the soil by added labelled NH4+ to produce a measurable apparent ANI (Broadbent and Nakashima, 1971). The possibility cannot be excluded that different plant species may behave differently (Nommik and Vahtras, 1982).

    3.7. Fixation to organic material

    There is also evidence for an abiotic reaction of the 15NH4+-N with soil organic matter. Nommik and Vahtras (1982) only report that the organic fraction of the soil has the capacity to bind ammonia in non-exchangeable forms. The evidence for an abiotic reaction is, however, not conclusive (Trehan, 1996).

    Initial studies of an Ontario soil showed that considerable proportions of added NH4+ were clay-fixed. Fixation by the organic matter of the soil was negligible (Kowalenko and Cameron, 1976). The fixation of ammonium by the organic fraction of the soil has also been studied by Burge and Broadbent (1961) and Sohn and Peech (1958). They found acidity and high organic matter content of a soil to be responsible for fixation of ammonium, so that it could not be recovered by protracted extraction with NaC1 solution. Ammonium fixation was related to the organic matter content and was a chemical reaction involving the phenolic hydroxyl group. The fixed ammonium was found to be very slowly available to a crop (Olu Obi etal., 1986).

    Another organic fixation that can occur is the fixation of ammonia. Nommik and Vahtras (1982) studied this form of fixation and reported that a number of physical and chemical reactions are responsible for the efficiency of NH3 retention. They range from weak physical sorption of the type of H-bonding to irreversible incorporation of NH3 into the soil organic matter. Physical sorption only occurs when there is a positive pressure of NH3 in the soil atmosphere. As soon as NH3 pressure decreases, equilibrium is displaced and the physically sorbed NH3 returns to the gaseous phase. The initial distribution pattern is, to a large extent, controlled by the moisture content of the soil, depending both on the high solubility of NH3 in water and the low diffusion rate of NH3 in the aqueous phase as compared with the gaseous phase. If this should be a problem when dealing with ammonium fertilisers, the pH in the soil must be around 8.

    3.8. Ammonification

    This process is controlled by microbial activity. The soil N in organic matter (aminoform and heterocyclic N compounds) is mineralised in two steps. First the N is protolysed, amino N is released from organic matter and this amino N is later reduced to NH3, the so-called ammonification (see N-cycle p. 5). In both steps energy is released that is utilised by the heterotrophic microbes. To bring about the reactions, the microbes require organic C as energy source. Other depressing factors are low temperature and deficiency or excess of water (Mengel and Kirkby, 1987). Beck (1983) has found that the optimal temperature for this process is 50° C. By using the constant release rate, Jansson (1963) calculated that the minimum half-life of the residual N would be 15 years.

    3.9. Mobilisation and Leaching

    The mobilisation of fixed NH4+ is dependent on the concentration of NH4+ in the soil solution. Ammonium is adsorbed by clay particles that have a negative charge at their surface and this binding is not as strong as that by fixation. Therefore, as the soil NH4+-concentration decreases the adsorbed NH4+ becomes mobile through diffusion. Trehan (1996), for example, reached the conclusion that 15NH4+-N fixed by clay was released when the supply of labile 15NH4+-N (exchangeable and in solution) to the soil microbes became low. Drury and Beauchamp (1991) also reported that when nitrification depleted extractable NH4+, fixed NH4+ was released.

    Losses of NH4+-N by leaching will be significant only in soils with extremely low CEC (Nommik and Vahtras, 1982). Ammonium is an attractive molecule in the soil and the competition for it is so high that leaching losses never occur under natural conditions.

    4. Nitrate

    4.1. Immobilisation of NO3-

    The NH4+ pool in the soil is markedly preferred to NO3--N by heterotrophic micro-organisms during immobilisation. When NO3- is available but NH4+ missing, the former is utilised (Jansson, 1958). Experiments where glucose has been used as substrate show that net immobilisation is followed by rapid nett mineralisation (Bjarnason, 1987). Usually from one growing season to another only a small portion (< 15 %) of N immobilised in organic forms becomes available to the plants (Stevenson, 1986).

    As might be expected from plant uptake data, there is a reverse relationship between the rate of N application and the percent of fertiliser N remaining in the soil. This results from the fact that at low N rates a relatively large proportion of the fertiliser N is immobilised in the roots and in soil micro-organisms. At higher rates the utilisation of N by micro-organisms is limited by the supply of readily available energy material (Legg and Allison, 1967). Energy supply, however, is not the only prerequisite. Since the N-free energy rich materials do not provide N to the growing microbes, the microbes draw on the inorganic N pool; if this pool is depleted, they will suffer from N deficiency, and their activities will be limited (Jansson and Persson, 1982). In a soil not provided with liberal amounts of easily decomposable organic material poor in N, the biologically available ammonium fertiliser N will be found in the microbial population and in the crop plants, whereas the nitrate fertiliser N will mainly be found in the latter (Jansson, 1962).

    4.2. Denitrification

    In similarity to the problem with NH3-losses it is difficult to prove N-losses due to denitrification. Many authors (Jansson, 1962; Wickramasinghe, 1985; Kowalenko 1989; Green et al., 1994) suggest denitrification as the cause of small deficits in their balance sheets or dismiss losses with assumptions like "loss of N by denitrification was doubtful, as aerobic conditions prevailed in the system" (Kowalenko and Cameron, 1976). To calculate the real denitrification losses it is necessary to measure gaseous losses like N2, N2O and NO, but nitrification rather than denitrification is the source of NO emitted from the soil. N2O (nitrous oxide), on the other hand, is a product of both nitrification and denitrification (Skiba et al., 1993). The magnitude of the NO and N2O emissions from a soil is dependent on the soil available NH4+ and NO3- concentrations, as well as on other factors such as soil temperature, soil water content and soil aeration (Skiba et al., 1993).

    Dilz and Woldendorp (1960) found that denitrification losses are particularly high when abundant living roots are present. It is supposed that root exudates stimulate denitrifiers. Other promoting conditions are high soil moisture that reduces aeration, neutral pH and high temperature.

    4.3. Leaching

    Adsorption of NO3--N to organic matter and clay occurs only at extremely low values but some authors (Kinjo et al., 1971) consider that adsorption might have contributed to some of the resistance of N to leaching. In contrast to fixed NH4+, the distribution of NO3- in profiles is not correlated with a specific soil characteristic, such as clay content or particle size, but usually follows the water regime, except in soils with a significant anion exchange or other sorption sites where mobility of NO3- could be lessened appreciably (Black and Waring, 1976 a, b; Espinoza et al., 1975). As long as the soil is not covered by vegetation, leaching is a problem. This was demonstrated by Kowalenko (1989), where the tracer confirmed non-tracer observation that applied N did not leach beyond the rooting zone (45 cm) during the growing season, despite the above-average precipitation that occurred, but that all residual NO3--N was leached over the winter.

    4.4. Plant availability

    N-uptake by plants is temperature dependent, rates of uptake being depressed by lower temperatures. Clarkson and Wamer (1979) found that when two ions are supplied to equal concentrations to ryegrass, NH4+ is absorbed more readily than NO3- at lower temperatures. NH4-N uptake takes place best in a neutral medium and is depressed as pH falls. The converse is true for NO3- absorption and at pH 6.8 both are absorbed at equal rates (Rao and Rains, 1976). In moist well-drained soils, NH4+ is usually nitrified rapidly to NO,. Nitrate, a very mobile nutrient, is carried readily to plant roots by mass flow but under wet conditions it is subject to loss due leaching and denitrification.

    5. Materials and Methods

    5.1 Soils

    Three soils of different parent material, one sandy soil from Dülmen, a loess soil (silt) from Bonn and a loess soil (silt) from Hannover, were used. They were collected in November from the 0- to 15-cm depth of cultivated fields and were sieved (6 mm).

    5.2 Experimental plan

    Two trials, one trial with and one without plants, were done and in both the soils got five different treatments with four replicates each.

    5.2.1 Plant trial

    Mischerlich pots were used in this trial and each pot was filled with 5 kg dw soil. Each pot got basic fertilisers for ryegrass, 14.4 g Thomaskali (8 K-15 P-6 Mg) (equivalent to 26.7 mg K/ 100g soil) and 0.8 g Fetrilon, which was mixed in as dry powder. At the same time, all pots got NaH2PO4 equivalent to 2.51g per pot, solved in aqua demin (25 ml) and the relevant nitrogen solution. For each of the three soils, there were four replicates for each treatment and there were totally four harvests.

    The treatments were:

    1. 1. A solution of Ca(NO3):, equivalent to 300 mg N per pot and 15g glucose (equivalent to 20g C against 1 g N) solved in 100 ml aqua demin.
    2. 2. A water solution of NH4Cl equivalent to 300 mg N per pot and didin (10 % didin of the applied N weight).
    3. 3. No nitrogen fertiliser.

    The 15N-enrichment was approximately 10 atom % in each of the two N-sources. All the pots were watered to 60 % of max. WHC. After this they were placed in a greenhouse with a main temperature of 18° C.

    Table 1: Treatments of the plant trial


    Number of



    NH4-N after 1st
    harvest (mg)


    didin (&b.mu;l)




































    The pots were sampled after week 1, 5, 7 and 9. After one week the harvested soils were analysed for Nmin and Ntot). The other pots were sown with ryegrass, cultivar Liberta (Lolium multiflorum).

    At the second, third and fourth harvests the grass was analysed for Ntot content. The soil was analysed for Ntot and Nmin. At the second harvest (week 5), the grass was cut off on all the pots and analysed for N-content, but at weeks 7 and 9 only the pots according to the plan were harvested. In the same week two series of each soil ammonium replication were supplied with a second dose of ammonium, 150 mg/pot and 300 mg/pot, respectively, and this time with non-labelled NH4+. These pots were harvested at the last harvest (week 9).

    Table 2: Scheme of analysis

    Sample date

    Soil Analysis



    1(w 1)

    Nmin, Ntot (incl. 15N)


    2 (w 5)


    Ntot (incl. 15N)

    3 (w 7)



    4 (w 9)



    5.2.2 PE-bottle trial

    The same procedure was done with 216 PE-bottles (1000 ml) with the differences that every bottle contained 100 g soil dw after application and did not get any basic fertiliser for ryegrass.

    Before the bottles were filled, the soils were mixed with KCl (equivalent to 43,2 mg K/ 100 g soil) except from four bottles of each soil, which did not receive KCl but the same NH4Cl amount.

    To get the same water solution for all three soils, water was added one hour before N-application, so that all soils would have a WHC of 60 %.

    The treatments were:

    1. 1. Bottles with no N added.
    2. 2. Bottles with labelled Ca (NO3)2 (6,0 mg N per bottle) solved in 5 ml aqua demin with 300 mg glucose (equivalent to 20g C against 1 g N).
    3. 3. Bottles with labelled NH4Cl (6,0 mg N per bottle) solved in 5 ml aqua demin with didin (10 % didin of the applied N weight).

    Table 3: PE-bottle trial


    of pots




    NH4-N after
    2nd harvest


    didin (&b.mu;l)

















































    The 15N-enrichment was approximately 10 atom % in each of the two N-sources. To keep the bottles at 60 % WHC during the experiment, they were covered with pierced parafilm and watered when needed. The bottles were harvested totally 5 times, after 1, 2, 5, 7 and 9 weeks. At the first harvest the bottles without KCl and those with K+ added were compered to see if there was any difference in fixation rate between them. All bottles were analysed for Nmin and biomass-N.

    After five weeks, two series from each of the soils treated with ammonium were given extra ammonium, 3.0 mg/PE-bottle and 6.0 mg, respectively, this time with unlabelled ammonium, and they were harvested at week 9.

    5.3 Analysis

    The Nmin analysis was mostly done the same day and 50 g soil was shaken with 1N K2SO4 and the solutions were analysed colorimetically on an auto-analyser (Technicon) for NH4-N, NO3--, NO3-N and soluble organic N according to the Alliance instruments method. The soils for Ntot-analysis was air-dried, sieved in a 1 mm sieve and sent to Kiel where they were analysed for Ntot, according to the dry combustion method on an element-analyser (NA1500, Carlo Erba Instruments, Hofheim/Taunus). For the determination of the 15N fraction a N-analyser-Isotope-masspectrometer (ANCA-MS, Robprep-CN + Tracermass, Europe Scientific Ltd., Crewe) was used. The same analytical method was used for the plant analysis of 15N but it was carried out in Berlin.

    All PE-bottles and the harvest from the first week's Mischerlich pots were analysed for biomass-N through the fumigation-extraction (FE) method by Brooks et al. (1985b). According to his calculations KN= 0.54. For this analysis, the soil was stored frozen. All plants were dried and ground before being Ntot tested with the Kjeldahl analysis with selenium as catalyt.

    Table 4: Particle size (%)



    fine silt



    fine sand























    The untreated soils were texture-defined by the hydrometer method described by ISIRC (1993) and revised by the Department of Soil Science, SLU, Uppsala (1996). The pH was measured in 0.01 M CaCl2. A Ctot- and Ntot-analysis was done with the dry-combustion method (Svensk standard, SS-ISO 10694) and analysed on a Leco CNS-2000. Nfix capacity was determined by adding (NH4)2SO4 and KCl, shaking and thereafter analyse on an Auto-analyser according to M. Schneiders, ACl, Bonn.

    Table 5: Basic soil data


    0.01 M


    % C

    % N

    weight %

    Nfix capacity
    dry soil
    (mg/kg dw)

    Nfix capacity
    wet soil
    (mg/kg dw)

























    6. Results

    Plant trial

    Looking at the yield results, one can discern a clear pattern. In all three soils there was a difference in N-content between AC and CN. The N-content in CN and non-fertilised soil is almost the same while in the AC-treatment it is 200 mg higher. The difference between AC and CN was smallest in the Bonn soil.

    Diagram 1: Ntot in the Dülmen, Bonn and Hannover soils.

    Normally the yield measured in g dw/pot shows that the second harvest is lower than the first but there are some exceptions and they are always from the second harvest in the week 9 and from the CN-treatment such as in the Dülmen soil with first cut 6.91g and second 8.48g, the Bonn soil with 7.44g and 9.63, respectively, and the Hannover soil with 5.94g and 6.83g. The same pattern is found in the Hannover AC treatment but is more marginal; 9.33g compared with 10.33g. In all pots, the percentage N-amount was lower in the second harvest.

    Table 6: Yield and 15N in the plants in the first cut (week 5, the pot trial).


    dw (g)

    Ntot (mg)

    New excess

    15N tot (mg)

    Dülmen CN










    Bonn CN










    Hannover CN










    Looking at the Ntot amounts from the first cut (table 6) it can be seen throughout that amounts are twice as large in the AC-treatments than in the CN-treatments. The N-rate is highest in the Bonn soil, both in AC and CN-treatment, but compared with the Hannover and Dülmen soils the difference is biggest in the CN-treatment. But if one looks at the 15N content at the same harvest the view is different. The Dülmen soil has the highest labelled N-rate followed by the Hannover and Bonn soils.

    In the Nmin test (NH4+ and NO3 in the soil), another pattern can be seen. After one week the Nmin amount was almost the same in the unfertilised treatment as at the start, but in the nitrate+glucose (CN) step nothing was left and in the AC-treatments between 200-250 mg N remain. From week five and later only very small amounts are found. The Bonn soil has the highest original amount of Nmin followed by the Dülmen and Hannover soils.

    Diagram 2: Nmin in the Dülmen, Bonn and Hannover soils.

    Diagram 2: Nmin in the Dülmen, Bonn and Hannover soils.

    The Ntot levels in the pots are rather similar in the Dülmen and Bonn soils but in the Hannover soil the values are one-third higher. The 15N values for the CN-treatment looks good but in the AC-treatment something has happened. The values are 30-36 mg too low per pot in the Dülmen and Hannover soils.

    Table 7: 15N in the soil after one week (pot trial)


    N tot in pot (g)

    15N in pot (mg)


    Dülmen CN








    Bonn CN








    Hannover CN








    Looking at the nett change, a nett loss of N is found after the first week in both CN and AC treatments in all three soils, but it is highest m the Bonn soil. After the fifth week there is a nett profit, that increases for each week. It is highest in the AC-treatments and the Hannover soil has the highest profit in all treatments. In the AC+1/2 and AC+1 the profit gets lower the more N that was added.

    Diagram 3: Nett change in the Dülmen, Bonn and Hannover soils calculated as follows: (Nmin at start + fertiliser) - (Plant N + Nmin) - (N0Nmin + N0Plant) = Nett change

    In the nett mineralisation diagram for the unfertilised treatment it can be seen that the mineralisation rate increases with the time. After the first week (without plants) less nitrogen was found compared with the untreated soils (diagram 4), but after five weeks the mineralisation has increased and continues until week nine.

    Diagram 4: Nett mineralisation from all three soils calculated as: Nmin at start - (Plant N + Nmin) = nett mineralisation

    The results from the Biomass N cannot be used as exact values, but they give a hint of the differences between the soils and the difference between the three treatments. It is obvious that less NH4+ was immobilised compared with NO3-. The standarddeviation is rather large, especially in the 0 and AC-treatments.

    Table 8: Biomass N measured with the fumigation-extraction method after one week


    Biomass N

    % of 300 mg N




























    *Only one replication was done from the AC-treatment, whereas three replications were done for the others.


    In the Dülmen soil m the CN-treatment, the Nmin value was about 2 mg after one week. With time it increased slowly to about 4 mg. The Nmin from the non-fertilised soil remained rather stable throughout the experiment. Another noticeable change was that the amount of NO3 increased with time in all AC-treatments and at the end almost the whole Nmin-pool consist of nitrate. In the Bonn soil the Nmin pool in the CN treatment increases faster, but no further than 3.5 mg. The Nmin store in the AC treatment is almost constant but from the first week and subsequently the pool contain NO3- only. In the AC+1/2and AC+1, more Nmin is found than fertiliser added plus starter Nmin found, more than 2 mg, and everything is in NO3- form.

    In the Hannover soil, the Nmin in the non-fertilised treatment increases slowly and the same happens with the CN treatment but faster. In the AC treatment, there is also a slighter increase with 1 mg N totally and in the AC+1/2 and AC+1 Nmin increases with over 2 mg compared with what was added plus the starter Nmin. Everything in the Nmin pool is kept as NO3- in the AC and CN treatments.

    Diagram 5: Nmin in PE-bottle trial in the Dülmen, Bonn and Hannover soils.

    The biomass N shows a negative value for AC and AC-K after one week. This means that the biomass N-value was smaller than the Nmin for the same time of soil sampling. The phenomenon is the same in the Bonn soil. The only stable trend that could be found in all three soils is that the biomass N in the non-fertilised treatments got higher for every week that passed. The amount of N in the biomass is larger in the Hannover soil from the start and the change of N in the biomass is smallest in the Dülmen soil.

    Diagram 6: Biomass N measured with the fumigation-extraction method in soils from Dülmen, Bonn and Hannover.

    7. Discussion

    Method discussion

    If one should be absolutely correct one should calculate with the nitrogen added through didin. Didin contains 67 % N and in this case constitutes about 6 % of the total added N.


    The yield results indicate that in the CN treatment all the newly added nitrogen was immobilised or has, in other ways, left the system so that the plants only have the mineralised N available. The N-contents in the non-fertilised and CN yields are the same in all three soils (between 200-300 mg N compared with 400-550 in the AC-treatment). The immobilisation idea is verified by 15N analysis of the plants that shows that CN-yield contains very little labelled N. There must have been a minor mineralisation of the newly immobilised nitrate because almost no nitrate was left according to the Nmin analysis. In the AC treatment we could still find large amounts of NH4+ or NO3- when the plants were sown, so all the labelled N found later in the plants does not come from mineralised 15N. The 15N-rate is higher in the AC-treatment in the Dülmen soil and this probably is due to the fact that the Nmin-rate was lowest there from the start so dilution effect was rather low and more of the 15N-molecules were taken up by the plants. None of the Nmin-trials were tested for 15N and, consequently it is impossible to tell how much of the 15N that comes from the Nmin fraction or from mineralisation/defixation.

    More Ntot in plants is found in the CN- and non-fertilised treatments in the Bonn soil compared with the two other soils. This is probably an effect of the higher Nmin-pool from the start. The soil from Bonn was the only one that had been in fallow over the summer.

    The difference in N-content in the plants between weeks 7 and 9 is very small and this indicates that the N-pool in the soil is depleted and the mineralisation is rather small. The increase is big only in the AC-treatment and this might have to do with increased mineralisation or defixation. In the Hannover soil, the roots were more widespread and the whole soil volume was penetrated more by roots compared with the Dülmen soil where only the upper half of the soil volume was covered by roots. There were less roots in the Bonn soil but they penetrated the whole soil volume.


    Nmin values show that the pool of mobilisable nitrate became depleted after the first week and probably everything is immobilised due to the glucose. The ammonium becomes depleted from week five and subsequently. This is surely due to the plants preferring NH4+ to nitrate. If all the nitrate was immobilised, the plants in these CN-pots somehow got nitrogen and obviously there must have been a mineralisation. Diagram 1 supports the idea that all the immobilised labelled nitrogen together with the glucose remains immobilised during the whole experiment and the plants get only nitrogen that was mineralised from old material like the non-fertilised step. Zagal and Persson (1994) found in an experiment where NO3--N was added with high amounts of glucose (four different rates), that NO3- was rapidly immobilised and by day 3 no mineral N was detectable in any of the treatments with the exception of the highest N-treatment. But in the two higher N-treatments they showed a rapid mineralisation and after 82 days of incubation 32-42 % of immobilised N had become mineralised. More biomass was formed in the N-rich treatments, which will result in more microbial C being available for humus formation. During the decomposition of glucose, nitrate was rapidly immobilised and the immobilisation increased with increased N addition suggesting that at low N-level the microflora could use more N if it was supplied. In a similar trial, Voroney and Paul (1984) showed that the nett mineralisation starts to occur between 7 and 21 days.

    In two cases the didin did not work as wanted, i.e. the Bonn and Hannover soils. After the first week in the B-AC treatment, were 6.98 ppm NH4+ and 6.06 ppm NO3- found in the soil compared with 0.19 and 4.19 ppm at the start. In the H-AC treatment the values were 0.40 ppm NH4+ and 11.38 ppm NO3- compared with 0.20 and 15 at the start. The amount of nitrate is dramatically increased, especially in Hannover soil even if a nitrification inhibitor was used. This probably has to do with the slightly higher C-level in the Hannover soils (Vilsmeier and Amberger, 1980) or the moisture status. The Bonn and Hannover soils retained water better than the Dülmen soil.

    In the PE-trial, it is rather obvious that the didin did not work especially well. In the Dülmen soil NH4+ was transformed more slowly than in the Hannover and Bonn soils, where it did not work even during the first week. It is hard to say if it was too old or if this has to do with the moisture or organic matter content. But the probability that the organic matter content influenced the situation is small because the Hannover soil has the highest C-content of 1.40 % and that is rather low.

    The mineralisation in the Hannover soil is largest during the whole time according to the N0-treatment, and for which there are many explanations, for example the higher C/N rate in the soil or the higher Nmin rate, or that this is not a result of mineralisation but is caused by defixation. Due to the lower loss of nitrogen after the first week and that the immobilisation is lowest in the Dülmen soil in all other weeks, the N0-values also indicate that the soil biomass is less active in the Dülmen soil. In the AC-treatment, all the soils have between 200-250 mg N left after one week. The differences between the Ntot contents in the grass are the same. About 200 mg N more is found in the AC-treatments after the first plant cut, so it appears fairly clear that the unimmobilised/fixed N from the AC-treatment is found in the harvested grass four weeks later. The other 250 mg comes from mineralisation and maybe defixation. Unfortunately, the fixation could not be determined as planned. The nett change indicates that N in the AC-treatments is moderate after one week and after five weeks has returned to the original level so if there was a NH4+-fixation by the clay minerals it was defixed due to the plants. Another factor indicating that fixation was smaller, is the low clay content in all three soils, but the fixation capacity 38-58 mg N/kg dw should be almost enough to fix 300 mg N. The reability of the fixation capacity method should be questioned because the fixation capacity is highest in the Dülmen soil which had the lowest clay and fine silt content (Table 4).

    15N in the soil

    The 15N losses from the soil after one week are difficult to explain. The additions of ammonium have been checked and the same instrument (pipette) was used for the application of NH4+ and NO3- so if it had been incorrectly calibrated the nitrate amounts should also have been wrong, but now they were too high instead of too low. Losses of ammonia could have taken place, but it would be unexpected that the losses were highest in the soil with lowest pH (Dülmen) if that had been the case.

    Biomass N

    The biomass N values are rather unreliable because the method is rather rough. In this study KN= 0.54 according to Brookes et al. (1985), but other authors recommend other KN values (KN=0.3 by Voroney and Paul, 1984). But even if the values are not exact they can be used to be compared with one another and reveal something about the biomass conditions in the soils, e.g. the Hannover soil had a larger biomass then the other soils from the start. All nitrate in the CN-treatments has been immobilised by the biomass as wanted, due to the glucose. The unfertilised and AC-treatments both have the same N-biomass rate so probably not much of the NH4+ has been used by the microbes. It would be interesting to check this in a 15N-test. In the PE-trial, the biomass N is much higher than in the pot-trial when measured as percentage. This is typical because there is no competitive situation here between plants and microorganisms. Another fact that is expected is the lower natural biomass-N value in the soil from Dülmen compared with the two other soils, because the Dülmen soil has the lowest pH, C-tot, a poor structure and also dries out the easiest.


    In this study we investigated mineralisation and defixation of added labelled NO3- and NH4+ in three different arable soils in western Germany. The trial consists of two tests, one with plants (ryegrass) and one without. Nitrogen was added (300 mg/5 kg dw) as ammonium+didin or nitrate+glucose and after incubation for one week the plant pots were sown. In the plant trial the number of replicates was sufficient for 4 sampling dates, and in the fallow trial there were replicates for 5 sampling dates.

    After one week, all nitrate was immobilised and after the first plants harvest between 14-80 mg 15N were found in the plants. Because of too few measurements of labelled N fractions, it is difficult to state how stable the glucose-immobilised N-fraction was, but the results from Nmin, biomass-N and Ntot in plants indicate that the mineralisation was much slower than the immobilisation, as has been found earlier.

    The defixation rate cannot be discussed at present since such information requires that the study must be repeated with soils with higher clay contents. The present study was unable to determine whether any fixation at all had occurred.

    Acknowledgements - This thesis was made possible by Hydro Agris research station at the Centre for Plant Nutrition and Environmental Research, Hanninghof, Germany. I thank Mr Blankenau, Dr Olfs and greenhouse and laboratory staff for all the assistance at Dülmen and Prof. Persson who supervised me back in Sweden.


    Allison, F.E., Roller, E.M., and Doetsch, J.H. (1963). Ammonium fixation and availability in vermiculite. Soil Sci. No 75, pp. 173-180.

    Amberger, A. and Gutser, R. (1978,). Z. Pflanenernährung. Bodenkunde, No 141, pp. 553-556.

    Antanasiu, N., Westphal, A. und Banerjee, A.K. (1968) Studien liber die Wirkung gedüngten Stickstoffs auf Ertrag und N-Aufnahme der Pflanze bei Boden mit verschiedenem Ammoniumfixierungsvermögen. Agrochemica No 12, pp. 120-129.

    Axley, J.H. and Legg, J.O. (1960). Ammonium fixation in soils and the influence of potassium on nitrogen availability from nitrate and ammonium sources. Soil Sci. No 90, pp. 151-156.

    Bailey, Y.A. (1942). Mechanism of potassium fixation by various clays. Ph.D. Thesis, University of Wisconsin, Madison.

    Beauchamp, E.G. (1982). Fixed ammonium and potassium release from two soils. Commun. Soil Sci. Plant Anal. No 13, pp. 927-943.

    Beck, T., (1983). Mineralization of soil nitrogen in laboratory incubation experiments. Z. Pflanzenernähr. Bodenkd. No 146, pp. 243-252.

    Bjarnason, S. (1987). Immobilization and remineralization of ammonium and nitrate after addition of different energy sources to soil. Plant and soil. Vol 97, pp 381-389.

    Black, A.S. and Waring, S A. (1972). Ammonium fixation and availability in some cereal producing soils in Queensland. Aust J. Soil Res. No 10, pp. 197-207.

    Black, A.S. and Waring, S A. (1976 a). Nitrate leaching and adsorption in a krasnozem from Redland bay, Queensland. I. Leaching of banded ammonium nitrate in a horticultural rotation. Aust. J. Soil Res. No 14, pp. 171-180.

    Black, A.S. and Waring, S A. (1976 b). Nitrate leaching and adsorption in a krasnozem from Redland bay, Queensland. II. Soil factors influencing adsorption. Aust. J. Soil Res. No 14, pp. 181-188.

    Bower, C.A. (1951). Availability of ammonium fixed in difficulty exchangeable form by soils of semiarid region. Soil Sci. Soc. Am. Proc. No 15, pp.119-122.

    Bower, C.A. (1950). Fixation of ammonium in difficulty exchangeable form under moist conditions by some soils of semiarid region. Soil Sci. No 70, pp. 375-382.

    Broadbent, F.A. and Nakashima, T. (1971). Effects of added salts on nitrogen mineralization in three Californian soils. Soil Sci. Soc. Am. Proc. No 35, pp. 457-460.

    Broadbent, F.E and Tyler, K.B. (1962). Laboratory and greenhouse investigation of nitrogen immobilization. In Soil Sci. Soc. Proc., pp 459-462

    Brookes, P.C et al. (1985 a). Chloroform fumigation and the release of soil nitrogen; the effects of fumigation time and temperature. In Soil Biol. Biochem. Vol 17, No. 6, pp. 831-835

    Brookes, P.C et al. (1985b). Chloroform fumigation and the release of soil nitrogen: A rapid direct extraction method to measure microbial biomass nitrogen in soil. In Soil biol. biochem. Vol. 17, No. 6, pp. 837-842.

    Burge, W.D. and Broadbent, F.E. (1961). Fixation of ammonia by organic soils. Soil Sci. Soc. Am. Proc. No 25, pp. 199-204.

    Campell, C.A. (1978), Soil organic carbon, nitrogen and fertility. In M.Schnitzer and S.U. Kahn (edi). Soil organic matter. Developments in soil science No 8, pp. 173-271. Elsevier Scientific Publishing Co. New York.

    Campell, C.A. et al. (1974). Effects of growing season soil temperature, moisture and NH4-N on soil nitrogen. Can. J. Microbiol. No 14, pp. 1348-1350.

    Clarkson, D.T. and Warner, AJ., (1979). Relationship between root temperature and the transport of ammonium and nitrate ions by Italian and perennial ryegrass (Lolium multiflorum and Lolium perenne). Plant Physiol. No 64, pp. 557-561.

    Drury, C.F. and Beauchamp, E.G. (1991). Ammonium fixation, release, nitrification and immobilization in high- and low-fixing soils. Soil Science Society of American Journal. 55: 1, pp.125-129.

    Espinoza, W., Gast, R.G. and Adams, R.S. Jr. (19759.Charge characteristics and nitrate retention by two andepts from south-central Chile. Soil Sci. Soc. Am. Proc. No 39, pp. 842-846.

    Francis, D.D et al. (1993). Immobilization and uptake of nitrogen applied to corn as starter fertilizer. In Soil Sci. Soc. Am. J. 57:1023-1026

    Green, C.J. et al. (1994) Release of fixed ammonium during nitrification in soils. Soil science Society of America. 58: 5, pp. 1411-1415.

    Haas E., Nieder, R and Richter, J. (1993). Temporal variability of non-exchangeable ammonium in a loess-derived luvisol. Modeling Geo-Biosphere Processes No 2, pp. 163-171.

    Hanschmann, A. und Lippold, H. (1987). Untersuchungen zur Dungerstickstoff-Umsetzung im Boden unter besondere Berucksichtigung der Fixierung von Ammonium in Tonmineralen. Archiv fur Acker und Pflanzenbau und Bodenkunde. 31:7, pp. 469-475

    ISRIC, International Soil References and Information Center. 1993. Procedures for Soil Analysis, Particle size analysis, p.3:1-3:12.

    Jansson, S.L. (1958). Tracer studies on nitrogen transformations in soil with special attention to mineralisation-immobilization relationships. K. Lantbrukshögsk. Ann. 24, pp. 101-361.

    Jansson, S.I. (1971). Use of 15N in studies of soil nitrogen. In A.D. McLaren and J.Skujins (edi) Soil biochemistry II. pp.129-166. Marcel Dekker Inc., New York.

    Jansson, S.L. and Persson, J. (1982). Mineralization and immobilization of soil nitrogen. In Nitrogen in agricultural soils. Vol 22 (FJ. Stevenson, Ed.) pp. 229-252. Am. Soc. Agron., Madison.

    Jansson, S.L., (1963). Balance sheet and residual effects of fertilizer nitrogen in a 6-year study with 15N. Soil Sci. No 95, pp. 31-37.

    Jenkinson, D.S, Fox, R.H and Rayner, J.H. (1985). Interaction between fertilizer nitrogen and soil nitrogen - the so called 'priming' effect. In Journ. Soil Sci. Vol. 36, pp. 425-444.

    Kinjo, T., Pratt, P.F. and Page, A.L. (19 719. Nitrate adsorption. III. Desorption movement and distribution in Andepts. Soil Sci. Soc. Am. proc. No 35, pp. 728-732.

    Knutze, H and Bartels, R. (1979). Nährstoffversorgung und Leistung von Hochmoorgrünland. Landw. Forsch. 31, 1. Sonderh., pp. 208-219.

    Kowalenko C.G and Cameron D.R. (1976). Nitrogen transformation in an incubated soil as affected by combinations of moisture content and temperature and adsorption-fixation of ammonium. Can. Journ. Soil Sci. 56:2, pp. 63-70.

    Kowalenko C.G and Ross G.J. (1979). Studies on the dynamics of "recently" clay-fixed NH4+ using 15N. Can. Journ. Soil Sci. 60:1, pp. 61-70.

    Kowalenko, C.G. (1981). Effect of immobilization on nitrogen transformation in a field 15N experiment. Can. J. Soil Sci. No 61, pp. 387-395.

    Kowalenko, C.G. (1989). Mineral-fixed ammonium in clay- and silt-size fractions of soils incubated with 15N-ammonium sulphate for five years. Can. J. Soil Sci. v 69 (4), pp. 825-833.

    Kudeyarov, V.N. (1981). Mobility of fixed ammonium in soil. In: Clark F.E. and Rosswall, T. (eds). Terrestrial nitrogen cycles. Ecol. Ball (Stockholm) No 33, pp 281-290.

    Legg, J.O and Allison, F.E. (1967). A tracers study of nitrogen balance and residual nitrogen availability with 12 soils. In Soil sci. soc. Amer. proc., Vol 31, pp 403-406

    Marzadori C. et al. (1996). Influence of spatial root distribution, organic nitrogen mineralization and fertilizer application on soil interlayer ammonium. Biol. Fertil, soils, v 23 (4). pp. 368-373.

    Mengel, K. (1996). Turnover in soil and its availability of crops. In Plant and soil 181: 83-93.

    Mengel, K. and Kirkby, A.E., (1987). Principles of plant nutrition. 4:th edition, International Potash Institute (publ.)

    Mengel, K., Horn, D. and Tributh, H. (1990). Availability of interlayer ammonium as related to root vicinity and mineral type. Soil. Sci. No 149, pp. 131-137.

    Mengel. K.and Scherer, H.W. (1981). Release of nonexchangeable (fixed) soil ammonium under field conditions during the growing season. In Soil Science Vol 131 No 4p 226-232.

    Neeteson, J.J.et al. (1986). Dependence of soil mineral n on N-fertilizer application. Plant and Soil. No 91, pp. 417-420.

    Nieder R. et al.(1996). Nitrogen transformation in arable soils of North-West Germany during the cereal growing season. Biology ana fertility of soils, 22:1-2, pp. 179-183.

    Niederbudde, E.A. (1983). Das Tonmineral-ammonium, sein Stellung zu anderen N-bindungsformen sowie Bewertung fur N-immobilisierung und Mobilisierung in Böden. Kali-Briefe (Büntehof) No 16, pp. 365-378.

    Nommik, H. (1957). Fixation and defixation of ammonium in soils. Acta Agric. Scand. No 7, pp 395-436.

    Nommik, H. (1965). Ammonium fixation and other reactions involving a nonenzymatic immobiliozation of mineral nitrogen on soil. In: Bartholomew W.V. and Clark, F.E. (eds), pp 198-258. Soil nitrogen. Agronomy 10. Amer.Soc. Agron. Inc. (Madison)

    Nommik, H. and Vahtras, K. (1982). Retention and fixation of ammonium and ammonia in soils. In Nitrogen in agricultural soils. Vol 22 (F J. Stevensson, Ed.) pp???-???. American society of Agronomy, Madison.

    Olu Obi A. et al. (1986). Crop utilization and soil retention of nitrogen from 15N-labelled urea, calciumnitrate and ammonium sulphate in several Manitoba soils. Can. J. Soil Sci. 66: 4, pp. 661-671.

    Opuwaribo E. and Odu C.T.I. (1975). Fixed ammonium in Nigerian soils. III. Capacity of organic and inorganic soil components to retain added ammonium. Jour. Soil Sci.. 26:4, pp. 358-363.

    Osborne, GJ. (1976). The significance of intercalary ammonium in representative surface and subsoils from southern New South Wales. Aust. J. Soil Res. No 14, pp. 373-380.

    Paul, EA and Juma, N.G. (1981). Mineralization and immobilization of soil nitrogen by microorganisms. In Terrestrial nitrogen cycles, Ecol. Bull. No 33, pp. 179-195.

    Rao, K.P. and Rains, D.W., (1976). Nitrate absorption by barley. Plant Physiol. No 57, pp. 55-58.

    Schachtschabel, P. (1961). Fixierung und Nachlieferung von Kalium- und Ammonium-Ionen. Beurteilung und Bestimmung des Kaliumversorgungsgrades von Boden. Landw. Forschung. Sonderheft No 15, pp. 29-47.

    Scherer, H.W. (1984). Beziehung zwischen dem Stickstoff-Entzug der Pflanzeen und der Abnahme von spezifisch gebundenem NH4-N im Boden. In Pflanzenernährung und Bodenkunde, Weinheim No 147.

    Schmidt E.L. (1982). Nitrification m soil. In: Nitrogen in agricultural Soils. Stevenson, FJ. (edi) pp.253-288.

    Schnürer, J and Rosswall, T. (1987). Mineralization of nitrogen from 15N labeled fungi, soil microbial biomass and roots and its uptake by barley plants. Plant and Soil No 102, pp. 71-78.

    Skiba, U. et al. (1993). Nitrification and denitrification as sources of nitric oxide and nitrous oxide m a sandy loam soil. In Soil Biol. Biochem. Vol 25, No 11. pp. 1527-1536.

    Sohn, J.B., and Peech, M. (19 589. Retention and fixation of ammonia by soils. Soil Sci. No 85, pp. 1-9.

    Sowden, F.J et al. (1978). Native clay-fixed ammonium content and the fixation of added ammonium of some soils of Eastern Canada. Can. Jour. Soil Sci. 58:1, pp. 27-38.

    Sowden, F.J. (1976). Transformations of nitrogen added as ammonium and manure to soil with a high ammonium-fixing capacity under laboratory conditions. Can. Jour. Soil Sci. 56:4, pp. 319-331.

    Stevenson, F.J., (1986) Cykles of soil. John Wiley, New York.

    Svensk standard, SS-ISO 10694.

    Trehan, S.P. (1996). Immobilisation of 15NH4+ in three soils by chemical and biological processes. Soil Biol Biochem. v 28 (8), pp. 1021-1027.

    Wehrmann, J. And Coldewey-Zum Eschenhoff (1986). Distribution of nitrate, exchangeable and non-exchangeable ammonium in the soil-root interface. Plant and Soil. No 91, pp. 421-424.

    Westerman, R.L and Kurtz, L.T (1972).Residual effects of 15N-labelled fertilizers in a field study. In Soil sci. Amer. proc. Vol 36, pp 91-94.

    Wickramasinghe K.N et al. (1985). Transformation of nitrogen fertilizers in soil. Soil Biol. Biochem. 17: 5, pp. 625-630.

    Wiklander, L. and Andersson, E. (1959). Kalkens markeffekt. III. Kemiska undersökningar av ett långvarigt kalkningsförsök på skifte IV vid Lanna. Grundförbättring No 12, pp. 1-40.

    Vilsmeier, K. and Amberger, A. (1980). Z. Pflanzenernährung. Bodenk. No 143, pp. 47-54. Young, J.L (1962). Inorganic nitrogen and carbon:nitrogen ration of some Pacific north-west soils. Soil Sci. No 93, pp.379-404.

    Young, J.L. and Aldag, R.W. (1982). Inorganic forms of nitrogen in soil. In Nitrogen in agricultural soils. Vol 22 (F J. Stevensson, Ed.). American society of Agronomy, Madison.

    Zagal, E. and Persson, J. (1994). Immobilisation and remineralisation of nitrate during glucose decomposition at four rates of nitrogen addition. In Soil Biol. Biochem.Vol. 26, No 10, pp 1313-1321.