SSSAJ Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Rubæk, G.H.
Right arrow Articles by Christensen, B.T.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Rubæk, G.H.
Right arrow Articles by Christensen, B.T.
Agricola
Right arrow Articles by Rubæk, G.H.
Right arrow Articles by Christensen, B.T.
Soil Science Society of America Journal 63:1123-1132 (1999)
© 1999 Soil Science Society of America

DIVISION S-2-SOIL CHEMISTRY

Organic Phosphorus in Soil Size Separates Characterized by Phosphorus-31 Nuclear Magnetic Resonance and Resin Extraction

G.H. Rubæka, G. Guggenbergerb, W. Zechb and B.T. Christensena

a Dep. of Crop Physiology and Soil Science, Danish Inst. of Agricultural Sciences, Research Centre Foulum, P.O. Box 50, DK-8830 Tjele, Denmark
b Univ. of Bayreuth, Inst. of Soil Science and Soil Geography, 95440 Bayreuth, Germany

gitte.rubaek{at}agrsci.dk


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 Materials and methods
 Results and discussion
 Conclusions
 REFERENCES
 
Land use and soil management affect soil organic C in whole soil and size separates, but knowledge of the accompanying soil organic P (Po) is limited. The objectives of this study were (i) to identify the structure of Po in soil size separates by solution 31P-nuclear magnetic resonance (NMR) spectroscopy, (ii) to determine the labile Po pool in the size separates by anion-exchange resin extraction, and (iii) to characterize the labile Po pool. We used soils from two long-term experimental sites, one in Bavaria (under spruce and deciduous forests, permanent grassland, and arable farming) and one in Denmark (with arable rotation and different fertilization strategies — unfertilized, mineral fertilizer, and animal manure). Total Po content increased with decreasing particle size. The dialyzed NaOH extracts of clay were enriched in microbial-derived teichoic acid-P and other diester-P forms compared with silt and sand. Clay from permanently vegetated soil had larger proportions of teichoic acid-P and other diester-P forms and was richer in resin extractable Po than clay from arable soil. There was a linear relationship between the proportion of the 31P-NMR spectra allocated to diester-P (including teichoic acid-P) and resin-Po. Our results suggest that the highly active and easily mineralized soil Po was mainly associated with clay. The larger part of the clay-associated Po was tightly bound and not extractable. Although the composition of this Po remained unknown, it was probably inaccessible to rapid microbial utilization. The composition of NaOH-extractable Po in the clay fraction was influenced to a greater extent by land use than by fertilizer inputs.

Abbreviations: 2AM, animal manure treatment • Ar, arable rotation • CEC, cation-exchange capacity • Df, deciduous forest • Gp, permanent grassland • NMR, nuclear magnetic resonance • 2NPK, inorganic fertilized treatment • Pi, inorganic P • Po, organic P • Pt, total P • Sf, spruce forest • SOM, soil organic matter • 0, unfertilized treatment


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
Materials and methods
 Results and discussion
 Conclusions
 REFERENCES
 
ORGANICALLY BOUND P (Po) accounts for 20 to 90% of total P (Pt) in topsoil (Tate, 1985), but only a small part of Po is readily available for mineralization (Stewart and Tiessen, 1987; Magid et al., 1996). Nevertheless, the labile portion of Po represents an important component in plant P nutrition, both in forest and arable ecosystems.

Various chemical extraction methods have been developed to assess the P-supplying capacity of soil, in particular the pool of plant-available inorganic P (Pi). The sequential extraction procedure of Hedley et al. (1982) and modifications thereof have regularly been used to isolate operational pools of Pi and Po according to differences in lability (Tiessen et al., 1983, 1984; Roberts et al., 1985; Oberson et al., 1993). Rubæk and Sibbesen (1993, 1995) demonstrated the value of ion-exchange resin extraction in estimating labile Pi and Po pools; however, such extraction methods provide little insight into the chemical structure of soil Po. To this end, solution 31P-NMR spectroscopy has successfully revealed structural features of alkali-soluble Pi and Po (Newman and Tate, 1980; Tate and Newman, 1982; Hawkes et al., 1984; Zech et al., 1985; Condron et al., 1985, 1990; Bedrock et al., 1994). Simultaneous 13C- and 31P-NMR spectroscopy on alkali extracts has been used to link P and C transformations (Gressel et al., 1996).

The efficiency of various extractants toward individual P forms varies, and extractants may alter the P forms during the extraction (Cade-Menun and Preston, 1996; Leinweber et al., 1997). Therefore care must be taken when comparing results based on different extraction procedures.

In a previous study (Guggenberger et al., 1996), we combined 31P-NMR spectroscopy of dialyzed NaOH extracts and extraction with bicarbonate-loaded macroporous anion-exchange resins to examine the impact of land use and fertilization on the forms of P in soil, emphasizing the potentially labile Po. We found that the concentration of resin-Po was linearly related to the concentration of diester-P (including teichoic acid-P) in dialyzed NaOH extracts, forms of P that appear to be readily mineralized in soils (Tate and Newman, 1982; Hawkes et al., 1984; Hinedi et al., 1988; Condron et al., 1990). It was concluded that the resin isolates a structurally and functionally reasonably uniform pool of potentially labile soil Po.

In this study, we pursue this line of research by combining 31P-NMR and resin analyses with physical fractionation of soil according to particle size. Separation of soil into size classes of primary organomineral complexes has proven to be a useful tool in the study of soil organic matter (SOM) structure and function (Christensen, 1996). For example, solid state and solution 13C-NMR spectroscopy of size separates have shown how land use and soil management affect the chemical quality of SOM in situ, that is while the organic matter is still associated with its natural mineral soil components (Oades et al., 1988; Baldock et al., 1992; Preston et al., 1994; Guggenberger et al., 1995; Randall et al., 1995). The C species within sand-sized separates are primarily plant-derived, while microbial metabolites dominate in the clay-sized fraction (Guggenberger et al., 1994, 1995). Much less is known about the sources and behavior of different P forms in primary size separates.

Tiessen and Stewart (1983) and Sibbesen (1995) reported a decrease in the C/Po ratio with decreasing soil particle size. Sequential extractions of P based on Hedley types of procedures have shown that the lability of Pi and Po varies with particle size (Tiessen et al., 1983; Agbenin and Tiessen, 1995; Leinweber et al., 1997), and that slope position and cultivation affect the various P pools. Generally, finer-sized separates are enriched in labile Pi and Po, but little is known about the chemical structure of Po in differently sized separates and how land management influences the forms of Po. Our objectives were (i) to identify the structure of Po in differently sized organomineral complexes by 31P-NMR spectroscopy, (ii) to determine the labile Po pool in the size separates by a macroporous anion-exchange resin, and (iii) to link these results in order to characterize the labile Po in soils exposed to different land use and management practices.


   Materials and methods
TOP
 ABSTRACT
 INTRODUCTION
 Materials and methods
Results and discussion
 Conclusions
 REFERENCES
 
Soil samples from two long-term experimental sites, one in Bavaria under four different land uses (spruce and deciduous forests, permanent grassland, and arable) and one in Denmark with three different fertilization strategies in an arable rotation (unfertilized, mineral fertilizer, and animal manure) were used.

The Taubenberg Site
The Taubenberg site is in the pre-alpine region of Bavaria, Germany (43°53'N, 11°48'E). The soil is a silty loam developed on calcareous glacial outwash covered by loess. Plots otherwise similar are under spruce forest (Sf), mixed deciduous forest (Df), permanent grassland (Gp), and arable rotation (Ar). Until 1905 the plots had similar vegetation; in the 17th century the natural mixed montane beech/fir forest (Fagus sylvatic L. and Abies alba L. as dominating trees) was cleared and an arable rotation (4–6 yr) with intermittent grass (2 yr) was established. Norway spruce [Picea abies (L.) Karsten] was planted in 1905 and mixed deciduous forest in 1956 on a former spruce-covered area. The permanent grass, laid out in 1956 on a previously arable site, receives animal manure and mineral fertilizer. The all-arable rotation, introduced in 1905, receives mineral fertilizers at a rate recommended for optimum crop yields; the types of fertilizers and amounts of nutrients applied to the fertilized plots have not been recorded.

Additions of manure and fertilizers to the Gp and Ar soils have increased the pH, Ca and Mg contents, the cation-exchange capacity (CEC), and base saturation (Table 1) . Less favorable soil properties are found on the Sf plot. Manganese mobilization in the A(e) horizon indicates the onset of podzolization (Brückner et al., 1987). The spruce plot is classified as a Dystric Eurtochrept, whereas the Df, Gp, and Ar soils are Typic Eutrochrepts (Soil Survey Staff, 1997). Brückner et al. (1987) and Guggenberger et al. (1994) provide more details on site characteristics and land-use history.


View this table:
[in this window]
[in a new window]
  		Table 1 Chemical properties of whole soils from the Taubenberg and Askov sites (Guggenberger et al., 1996)

 
Field replications of these plots were not available; therefore, the surface horizons of four profiles within each plot were sampled and pooled in October 1994.

The Askov Site
The Askov Long-Term Experiments on Animal Manure and Mineral Fertilizers are at Lermarken, Denmark (55°28'N, 09°07'E). The soil is a light sandy loam developed on Weichselian glacial deposits (Typic Hapludalf). The site carries a four-course rotation of winter wheat (Triticum aestivum L.), sugar beet (Beta vulgaris L.), spring barley (Hordeum vulgare L.), and a grass–clover mixture composed of alfalfa (Medicago sativa L.), alsike clover (Trifolium hybridum L.), birdsfoot trefoil (Lotus corniculatus L.), ryegrass (Lolium perenne L.), fescue (Festuca pratentis Huds.), and timothy (Phleum pratense L.), with seeding rates of 10, 3, 3, 5, 5, and 2 kg ha-1, respectively. The experiment includes unfertilized (0, since 1893), mineral fertilized (2NPK, since 1923), and fertilized with animal manure (2AM, since 1923) treatments. The 2NPK and the 2AM treatments receive equivalent dressings of total N, P, and K; the annual mean of crop rotation being 200 kg N ha-1, 39 kg P ha-1, and 184 kg K ha-1 since 1973. Animal manure has been cattle slurry since 1973. The treatments have introduced differences in exchangeable Mg and K, and in total P (Table 1). The 2NPK and 2AM treatments have increased CEC by 11 and 17%, respectively. Christensen and Johnston (1997) have summarized additional experimental results.

Bulk soil was sampled in April 1994 from the 0- to 15-cm depth (Ap horizon) in the center of the 0, 2NPK, and 2AM treatments in the B2 field, one of four blocks included in the Askov long-term experiments.

Particle-Size Separation
Soil samples were air-dried, visible remnants of roots removed and the soil sieved to <2 mm before separation. Particle-size separation was carried out according to Christensen (1985). Briefly, 50 g of soil was dispersed ultrasonically (300 W for 15 min) in 150 mL of water with a probe-type disintegrator (Model 1510, B. Braun Labsonic, Melsungen, Germany). Clay-sized (<2 µm) and silt-sized (2–20 µm) fractions were obtained by repeated gravitational sedimentation in water, the sand (20–2000 µm) being recovered as the sediment left after isolation of clay and silt. Clay particles in decanted suspensions were flocculated by addition of CaCl2 and concentrated by centrifugation. Results are expressed on a 40°C dry basis.

Chemical Analyses
Total P in whole soils and size separates was determined after digestion of finely ground subsamples in a mixture of concentrated H2SO4 and HClO4 at 250°C. Inorganic P was determined after extraction with 6 M H2SO4 at 70°C for 30 min. Organic P is the difference between Pt and Pi.

Subsamples of 25 g (clay and silt), 45 g (Taubenberg sand), or 200 g (Askov sand) were extracted with 0.1 M NaOH under a N2 atmosphere for 24 h, using a sample/extractant ratio of 1:4 (w/v). After centrifuging and removal of the supernatant, the pellet was dispersed ultrasonically in 0.5 M NaOH for 10 min (same amounts as in the first extraction) using a probe-type disintegrator (Model W 185F, Heat Systems-Ultrasonics, New York), the energy dissipated being 270 J mL-1. The suspension was centrifuged and the supernatant combined with the corresponding supernatant from the first extraction. The combined supernatant was dialyzed (Servapor, MWCO 3000, SERVA Electrophoresis GmbH, Heidelberg, Germany) and then freeze-dried. This material is referred to as dialyzed NaOH extract. Inorganic P in the dialyzed NaOH extract was determined in duplicate by dissolving 50 to 120 mg in 45 mL of 1 M NaOH. The mixture was acidified with 45 mL of 4 M H2SO4, filtered, and Pi determined in the filtrate. For determination of Pt in NaOH extracts, 50 to 100 mg of the freeze-dried, dialyzed NaOH extract was digested at 250°C in a mixture of concentrated H2SO4 and HClO4 as described by Rubæk and Sibbesen (1993) for resin extracts.

Separates and whole soils were analyzed with the resin extraction method of Rubæk and Sibbesen (1993). Five milliliters of macroporous, strongly basic anion-exchange resin (Lewatite MP500A/WS, Bayer AG, Leverkusen, Germany) loaded with bicarbonate was enclosed in a mesh bag and placed in a 50-mL centrifuge tube along with 40 mL of water and 0.5 g of clay, 0.8 g of silt, or 20 g of sand. The tubes were shaken end-over-end for 17 h at 25°C. The resin bags were removed, washed with water to remove soil particles, and allowed to drain for 15 min. The resin was then eluted batchwise with two times 20 mL of 1 M NaCl (each for 30 min) and the eluates were pooled for subsequent analyses of P and C. Taubenberg sand was not analyzed due to lack of material. The Askov sand contained <5 mg total P kg-1.

Total C in size separates, whole soils, and freeze-dried dialyzed NaOH extracts was determined by dry combustion using either a LECO CNS 1000 (LECO Corporation, St. Joseph, MI) or an Elementar Vario EL CHNS analyzer (Elementar GmbH, Hanau, Germany). Total C in acidified resin eluates was determined with a Shimadzu TOC 5000A (Shimadzu Scientific Instruments, Columbia, MD).

Phosphorus-31 Nuclear Magnetic Resonance Spectroscopy
For solution 31P-NMR analysis, 150 mg of freeze-dried, dialyzed NaOH extract was dissolved in 3 mL of 0.5 M NaOD and then centrifuged at 19600 g for 15 min. The supernatant was transferred to 10-mm NMR tubes and analyzed the same day. A 500-MHz NMR spectrometer (Bruker, Karlsruhe, Germany, Model AM 500) operating at 202 MHz was used to give 31P spectra without 1H-decoupling after collection of 8000 to 48000 scans. Additional recording conditions were temperature, 20°C; frequency range, 50 kHz; pulse angle, 90°; pulse delay, 0.2 s; and acquisition time, 0.1 s. Peak positions on spectra were relative to 85% orthophosphoric acid and peak assignments were according to Newman and Tate (1980) and Condron et al. (1990). Peak areas were obtained by electronic integration. The inorganic orthophosphate and orthophosphate monoester signals were separated using a boundary determined from the valley between the two signals to the baseline, as suggested by Dai et al. (1996).

Statistics
To examine the relation between chemical species (diester-P including teichoic acid-P) obtained by solution 31P NMR spectroscopy and resin-Po, we carried out an analysis of functional relation as outlined by Webster (1989). For resin-Po error variances, s2, were estimated from the measured data. For the 31P-NMR measurements we assumed a population variance of 5% (Preston, 1987).


   Results and discussion
TOP
 ABSTRACT
 INTRODUCTION
 Materials and methods
 Results and discussion
Conclusions
 REFERENCES
 
This study emphasizes the analytical potential of combining different chemical methods with physical fractionation to evaluate differences in the nature of soil Po. To obtain comparable samples across the contrasting land uses, we retrieved soil from the mineral A horizon, which could be diagnosed in all plots. We minimized interference from fragile litter components by removing visible roots and litter fragments, and by sieving soils (<2 mm). We focused on organomineral complexes, and previously reported results on whole soils (Guggenberger et al., 1996) are incorporated only when relevant for discussion of data on size separates. Generally, soils from the Taubenberg site were, in comparison with Askov soils, higher in CEC and in contents of total P, total and resin-extractable C, and dithionite-citrate-bicarbonate soluble Fe and Al (Table 1). However, the soil chemical properties reflect not only differences in land use and fertilization, but also differences in geological origin and in the climatic regime of the two sites.

Total Phosphorus and Carbon in Size Separates
Clay-sized separates accounted for 21 to 25%, silt for 35 to 45%, and sand for 34 to 41% of the soil dry weight in the Taubenberg soils. At Askov, clay accounted for 9%, silt for 11%, and sand for 80% (Table 2) . The size distributions of organomineral complexes are in accordance with results of standard textural analyses (data not shown), indicating a complete dispersion of the soils under study. Recovery of soil solids was between 98 and 102%. In general the concentration of total C increased with decreasing particle size, the sand-sized separate of the Taubenberg arable soil (Ar) being much lower in C than sand from permanently vegetated soils (Sf, Df, Gp). The C content in sand from unfertilized Askov soil was below the detection limit. The recovery of whole soil C in size separates varied from 81 to 88%, indicating that some C was lost during the fractionation procedure. The distribution of C across size separates and the effects of cultivation accords with previous studies (Christensen, 1996).


View this table:
[in this window]
[in a new window]
  		Table 2 The distribution of soil dry weight across size separates, and concentrations of total C, total P, and organic P in sand, silt, and clay

 
Total P content increased with decreasing particle size, which confirms results of Tiessen et al. (1983), Sibbesen (1995), and Leinweber et al. (1997). The differences between silt and clay Pt were much larger than for C. In the forested plots, 68 to 72% of the clay and silt P was found as Po, whereas Po in sand accounted for half of the Pt. Clay and silt isolated from fertilized plots at Taubenberg (Gp, Ar) and from the unfertilized plot at Askov (0) also had {approx}50% of the Pt in Po forms, whereas Po in clay and silt from 2NPK and 2AM accounted for 30 to 35% of the Pt. Between 80 and 90% of whole soil Pt was recovered in size separates. For Askov soils the lack of recovery arose from almost equal losses of Pi and Po, whereas the lack of recovery in Taubenberg soil was almost exclusively due to loss of Po. Although clay made up <10% of the Askov soil weight, it held 46 to 47% of the Pt. At Taubenberg, clay accounted for 20 to 25% of the soil weight and 50% of the Pt in the arable soil. In permanently vegetated soils, clay held less of the Pt (30–40%). Like Pt, the C and Po content increased with decreasing particle size. The Po content varied very little within each site and size separate, whereas the proportion of total P in organic form varied considerably, reflecting that depletion and enrichment of soil P mainly occurs in soil Pi. The only exception was the arable size separates from the Taubenberg site, which was depleted in organic P compared with the corresponding separates from permanently vegetated Taubenberg soils.

Extractable Phosphorus and Carbon
Soil extraction may aim at isolating well-defined, operational pools of specific compounds, e.g., the proportion of P which is readily available to microorganisms or plants, or the proportion of a compound which is bound to the soil matrix by specific mechanisms. The resin method adopted in this study belongs to this type of extraction approach, as it aims at extracting only the labile soil Pi and Po, which are minor parts of the total soil Pi and Po. It has also been suggested that NaOH extraction isolates operational pools of soil organic and inorganic P in studies of long-term changes in the composition of soil P (Cross and Schlesinger, 1995; Rubæk and Sibbesen, 1995). For solution 31P-NMR studies of soil Po, NaOH is routinely used for extraction, and the aim typically is to extract a large and representative portion of total soil Po. However, Cade-Menun and Preston (1996) and Leinweber et al. (1997) found that NaOH could be selective toward different forms of soil Po and that NaOH may modify the extracted Po; therefore, other extraction procedures have been investigated. For example, Cade-Menun and Preston (1996) and Dai et al. (1996) used a mixture of EDTA (ethylenediaminetetraacetic acid) and NAOH for extraction, and found that it released a larger proportion of the Po. In this study, we have used NaOH combined with ultrasonification followed by a second extraction with NaOH and subsequent dialysis to isolate Po for 31P-NMR spectroscopy. Direct analysis of the resin extracts with 31P-NMR was not possible due to the low concentration of Po in the eluates and concomitant high concentrations of NaCl.

Organic P accounted for 88 to 96% of the Pt in dialyzed NaOH extracts, indicating that the dialysis was reasonably efficient in removing Pi from the crude extracts of clay and silt separates (Table 3) . The dialysis was less effective for sand-sized separates where 75 to 89% of the Pt was present as Po. Although some low-molecular Po may have been lost during dialysis, this step was necessary to concentrate Po in the NaOH extracts prior to 31P-NMR analyses. Drying under a stream of N2 as proposed by Newman and Tate (1980) was not applicable due to the low concentration of Po in our extracts.


View this table:
[in this window]
[in a new window]
  		Table 3 Total C and organic P (Po) content in dialyzed NaOH extracts of Taubenberg and Askov size separates

 
The NaOH extracted 28 to 50% of the C associated with silt and Taubenberg sand, whereas the extractability of clay C was much smaller (Table 3). Thus, only 5 to 8% of total C in Taubenberg clay and 10 to 16% in Askov clay, could be found in the dialyzed NaOH extracts. A similar pattern was observed for NaOH-extractable Po, where clay-sized separates were low in extractable Po while silt was much higher. As for C, Taubenberg clay released less Po than Askov clay. The extractability of Po associated with silt was 34 to 35% for Askov and 42 to 77% for Taubenberg. The extractability of total C and Po in silt was rather similar, except in the two forest soils (Sf, Df) where Po was more extractable in NaOH than C.

The NaOH extractability of C deviates from results obtained in previous studies of size separates from Taubenberg (Guggenberger et al., 1995), in which the extraction was made with NaOH in combination with sodium pyrophosphate (Na4P2O7). The pyrophosphate in this mixture forms complexes with the Fe and Al hydroxides whereby more SOM is released. The NaOH–Na4P2O7 mixture therefore resulted in a 9.4 to 21.3% recovery of C in humic acids. Most likely this mixture isolated a SOM fraction that differed in composition from the one isolated with the extraction used in this study.

The recoveries of Po from the clay were even lower than that of C (1.3–7.7% of total Po). Leinweber et al. (1997) extracted between 9 and 60% of total P with NaOH, and stated that extraction efficiency declined with decreasing particle size. Tiessen et al. (1983) obtained higher Po recoveries with NaOH when the Hedley fractionation procedure was applied to clay-sized separates. However, this procedure operates with sequential extraction, including NaOH as only one of the extractants, and a solution/soil ratio 15 times larger than ours.

The recoveries of Po from clay in this study may have been lowered due to the particle-size separation procedure itself, since the CaCl2 that was added to flocculate the clay may have affected the P sorption complex. Helyar et al. (1976) and Barrow et al. (1980) have shown increased sorption of inorganic P in the presence of Ca2+. We cannot exclude the possibility that some Po forms react similarly. Sibbesen (1995) found that anion-exchange resin recovered more inorganic P in clay-sized separates of Danish arable soils than it did in silt and sand separates (10–22% of Pi in the clay, 5–7% in silt, and 2–3% in sand). He used a size-separation procedure in which flocculation of the clay with CaCl2 was substituted with evaporation of the excess water at 40°C. We extracted 5 to 15% of the Pi with resins from the clay and 7 to 18% from the silt separates (Table 4) , and our recovery of Pi from clay was only occasionally larger than from silt. This may be due to differences in soil origin and land use of our soils compared with those of Sibbesen (1995), but it may also be due to Ca2+-induced modifications of the P sorption in our clays.


View this table:
[in this window]
[in a new window]
  		Table 4 Resin extractable C, inorganic P (Pi), and organic P (Po) in clay and silt size separates from Taubenberg and Askov soils

 
The resins took out 4 to 6% of C in Taubenberg clay and 2% in Askov clay (Table 4). Carbon in silt was less extractable with resin than C in clay. Silt from Taubenberg and Askov showed similar extractability. Resin-extractable Po followed the same pattern as seen for C; the resin removed 4.5 to 5.0% and 3.1 to 3.2% of the Po in Taubenberg and Askov clay, respectively, and 1.9 to 3.3% of the silt Po. Of the resin-extractable Pt, Pi was the dominating form in all samples, except for clay isolated from Taubenberg soils under forest. The C/Po ratio of the resin-extracted SOM was lower for arable soils than for permanently vegetated soils, but values for corresponding clay and silt samples were similar.

The clay separates were in general much richer in Po than silt and sand separates, but only a small part could be extracted with the methods used in this study. Although the extractabilities from clay may have been affected by size separation, large proportions of the clay-bound soil Po are, under natural conditions, closely associated with the clay particles and therefore expected to be rather inaccessible to extraction as well as mineralization by microorganisms.

Resin-extractable C in whole soils was 1.8 to 2.9 times higher than we found with our previous procedure (Guggenberger et al., 1996), in which a Xertex Dohrman DC 180 (Rosemount Analytical, Santa Clara, CA) was employed for C determinations in the extracts. Because Cl- ions interfere with the measurement of C on the Xertex Dohrman DC 180, it was necessary to substitute the two by 20 mL of 1.0 M NaCl solution for elution of the resins with two by 40 mL of 0.5 M K2SO4. Our study used a Shimadzu TOC 5000A carbon analyzer, which is not affected by Cl-. The eluent was therefore based on NaCl. Substituting the 1.0 M NaCl with 0.5 M K2SO4 did not affect the resin Pi and Po significantly (P < 0.05, data not shown), but it may have affected the resin C. The measurement of dissolved organic C on the Shimadzu TOC 5000A is based on catalyzed high temperature oxidation, while the Xertex Dohrman DC 180 relies on persulphate oxidation promoted by ultraviolet radiation at low temperature. The latter type of oxidation often results in underestimation of total organic C (Wangersky, 1993). Thus, the different principles of determination of soluble C incorporated into these instruments may also have affected the results on C contents in resin extracts.

Phosphorus-31 Nuclear Magnetic Resonance Spectroscopy
In the 31P NMR spectra of dialyzed NaOH extracts, soil orthophospate monoesters (monoester-P), resonated at 3.0 to 6.1 ppm (Fig. 1 and 2) . Monoester-P (e.g., inositol phosphates, sugar phosphates, and mono-nucleotides) was the dominating form of Po in dialyzed NaOH extracts of the sand and silt separates (Fig. 3 and 4) , accounting for 51 to 71% of the spectral area obtained for sand and 60 to 78% for silt. The proportion of Po in clay assigned to monoester-P was considerably lower. For arable soils (0, 2NPK, 2AM, and Ar), monoester-P dialyzed NaOH extracts of clay comprised 41 to 49% of the spectral area, while clay from soils under permanent vegetation had less monoester-P (25–34% of the total area). Monoester-P includes high proportions of inositol hexakis- and pentakisphosphates primarily of plant origin (Magid et al., 1996). This is consistent with the fact that sand-associated organic matter consists primarily of particulate organic matter (Christensen, 1996), with a chemical composition resembling that of plant litter (Guggenberger et al., 1995, 1996).



View larger version (20K):
[in this window]
[in a new window]
  		Fig. 1 Solution 31P-nuclear magnetic resonance spectra of dialyzed NaOH extracts obtained from the sand, silt, and clay separates of the Taubenberg soil under spruce forest (Sf), deciduous forest (Df), permanent grassland (Gp), and arable rotation (Ar)

 


View larger version (16K):
[in this window]
[in a new window]
  		Fig. 2 Solution 31P-nuclear magnetic resonance spectra of dialyzed NaOH extracts obtained from the sand silt and clay separates of the Askov soil under unfertilized (0), inorganic fertilized (2NPK), and animal manure (2AM) treatments

 


View larger version (31K):
[in this window]
[in a new window]
  		Fig. 3 Percentage of the whole 31P-nuclear magnetic resonance spectral area allocated to each particular P form in the size separates of the Taubenberg soils. Apart from these quantified P forms in the figure, traces of inorganic orthophosphate were detected in all clay separates, traces of phosphonates were detected in spruce forest (Sf) and deciduous forest (Df) clay, and a trace of pyrophosphate was found in permanent grassland (Gp) clay. Ar is arable rotation

 


View larger version (27K):
[in this window]
[in a new window]
  		Fig. 4 Percentage of the whole 31P-nuclear magnetic resonance spectral area allocated to each particular P form in the size separates of the Askov soils. Apart from these quantified P forms, traces of phosphonates were found in the unfertilized (0) and inorganic fertilized (NPK) treatment clays, and traces of inorganic orthophosphate were found in NPK and animal manure (AM) treatment clays

 
Orthophosphate diesters in the range -2.0 to 2.8 ppm (Fig. 1 and 2) were subdivided into teichoic acid-P (0.8–2.8 ppm), which is exclusively found in prokaryotic cell walls, and into other diester-P species (-2.0–0.8), such as phospholipids and DNA (Condron et al., 1990; Newman and Tate, 1980). Teichoic acid-P and other forms of diester-P were much more abundant in clay than in silt and sand (Fig. 3 and 4). For Askov clay, 17 to 21% of the Po was assigned to teichoic acid-P and 25 to 33% to other forms of diester-P. The corresponding values for the two coarser size separates were 6 to 11% teichoic acid-P and 7 to 11% other diester-P. Diester-P contents in the Taubenberg clay varied much more than Askov clay, the Taubenberg clay having 20 to 51% of the Po in teichoic acid-P and 15 to 46% in diester-P. Thus land use had a more significant impact on these forms of clay Po than fertilization.

Phosphonates (compounds with a direct C–P bond) resonating at 18.9 to 19.2 ppm contributed <8% to the spectral area. Orthophosphate signals at 6.1 to 6.7 ppm (Fig. 1 and 2) typically made up 9 to 15% of the spectral area for sand and silt, exceptions being silts from 2NPK and 2AM for which orthophosphate covered 4 to 5% (Fig. 3 and 4). Pyrophosphate resonating from -3.5 to -5.5 ppm was not detected in dialyzed NaOH extracts of clay, and very small amounts were present in extracts of silt and sand. The values for orthophosphate and pyrophosphate determined by 31P-NMR spectroscopy are in accordance with the values for Pi as determined by chemical analyses (Table 3). Less than 1% of the spectral area obtained for clay was assigned to orthophosphate.

The ratio between monoester-P and diester-P plus teichoic acid-P revealed a marked difference in the composition of the NaOH-extractable Po associated with clay and that associated with the coarser separates. On average, there was four times more monoester-P than diester-P plus teichoic acid-P in the Po from sand and silt (Table 5) , while these two groups of P made up equal proportions of the Po extracted from arable clay separates. In clay from the two forest soils, diester-P plus teichoic acid-P accounted for twice as much Po as the monoester-P. Except for the 2AM soil, silt had higher monoester-P to diester-P ratios than both sand and clay, indicating that chemically more recalcitrant P forms (i.e., monoester-P, Hinedi et al., 1988) were enriched in the extracts from silt-sized separates. Guggenberger et al. (1995, 1994) showed that silt-associated organic matter compared with clay-bound organic matter was poorer in N and in carbohydrates (both plant- and microbial-derived) and relatively rich in C-substituted aromatics. Christensen (1987) and Christensen and Olesen (1998) found that the decomposability of C and N associated with silt was lower than that of C and N in clay. Together, these studies indicated that silt-associated organic matter was rather low in organically bound nutrients and dominated by recalcitrant Po and SOM components.


View this table:
[in this window]
[in a new window]
  		Table 5 The ratio between 31P-nuclear magnetic resonance spectral areas of the NaOH extracts allocated to monoester-P and the sum of diester-P, and teichoic acid-P

 
Diester-P was selectively enriched within extracts of clay-sized separates, with teichoic acid-P representing a considerable fraction of the diester-P. Teichoic acids are acidic polysaccharides attached to the cell wall of Gram-positive bacteria (Brock and Madigan, 1988). Bacterial activity is higher near and at clay surfaces (Ladd et al., 1993) leading to accumulation of bacterial remains in the clay-sized separates. Sorption of teichoic acids to clay may be high due to the negative charge of this cell-wall constituent. The accumulation of teichoic acid-P in clay-sized separates is corroborated by the results of Zhang et al. (1998) showing that muramic acid, another exclusive cell-wall constituent of eubacteria, was concentrated in the clay fraction.

The Po in resin extracts was linearly related to the proportion of Po assigned to teichoic acid-P plus diester-P (Fig. 5) . The data separated into three distinct groups. Low resin-Po was found in the silt-sized separates of all soils, which also have a rather low proportion of diester-P, including teichoic acid-P. The clay separates from the arable soils showed intermediate levels of resin-Po and of diester-P plus teichoic acid-P, while levels of resin-Po, diester-P and teichoic acid-P were high in clay from permanently vegetated soils. For whole soils the proportion of diester-P plus teichoic acid-P was almost constant, and the relation between resin-Po and the proportion of teichoic acid-P plus other diester-P forms in the NaOH extracts was weak (Fig. 5). However for the whole soils, a close functional relation between resin-Po and teichoic acid-P plus diester-P was found when the presence of teichoic acid-P and other diester-P forms was converted to milligrams per kilogram (Guggenberger et al., 1996). Such a relation could not be established for diester P forms converted to milligrams per kilogram and resin Po from the size separates.



View larger version (16K):
[in this window]
[in a new window]
  		Fig. 5 Functional relations between concentrations of Po extracted by resin (X) and the percentage of spectral area allocated to diester-P including teichoic acid-P (Y) for whole soils and for clay and silt separates. The full lines show the functional relations with the gradient b and the dotted lines are obtained by linear regression of Y on X (gradient = bYX) and X on Y (gradient = 1/bXY), respectively. For whole soils, the gradient b was 0.133, bYX was 0.092, and 1/bXY was 0.391; the correlation coefficient r2 was 0.235. For the size separates, the gradient b was 0.974, bYX was 0.862 and 1/bXY was 0.928 and the correlation coefficient r2 was 0.929. Land-use abbreviations are: Ar, arable rotation; Df, deciduous forest; Gp, permanent grassland; and Sf, spruce forest. Fertilizer-treatment abbreviations are: 0, unfertilized treatment; 2AM, animal manure treatment; 2NPK, inorganic fertilized treatment

 
Although the source and the composition of the nonextracted P species in the clay-sized separates remains unknown, the data suggest that this fraction is crucial in microbial P cycling in soil. Considering the predominantly microbial origin of C in the clay-sized separates (Christensen, 1996), we assume that also the nonextracted Po species represent products of microbial resynthesis. However, association with Al or Fe oxides and covalent bindings to minerals may render the Po inaccessible to subsequent microbial mineralization (Hedges and Oades, 1997) and only a small part of the clay Po probably participates in the short-term Po turnover. We assume that resin-Po and NaOH-soluble diester-P represent this active Po fraction.


   Conclusions
TOP
 ABSTRACT
 INTRODUCTION
 Materials and methods
 Results and discussion
 Conclusions
REFERENCES
 
Most soil Po was found in the clay fraction, but only a small fraction was recovered in the resin and dialyzed NaOH extracts. A large part of the clay Po might be rather tightly bound and scarcely available for mineralization, but the highly active and easily mineralizable soil Po was also mainly associated with the clay separates (high proportion of diester-P and rich in resin-Po). Silt and sand were relatively poor in Po and consisted mainly of more recalcitrant forms of Po (low proportion of diester-P, and low in resin-Po).

Fertilization had only minor effects on soil Po in the size separates. In contrast, size separates of permanently vegetated soils differed from the arable soils, indicating a significant effect of land use on soil Po forms and lability.


   ACKNOWLEDGMENTS
 
We thank K.-E. Rehfuess and F.D. Wimmer for their support in sampling the Taubenberg soil and are highly indebted to Ludwig Haumaier for recording the 31P NMR spectra. The Danish Research Academy and the Commission of the European Union financially supported this study.

Received for publication June 22, 1998.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 Materials and methods
 Results and discussion
 Conclusions
 REFERENCES
 




This article has been cited by other articles:


Home page
 Vadose Zone JHome page
K. Schelde, L. W. de Jonge, C. Kjaergaard, M. Laegdsmand, and G. H. Rubaek
Effects of Manure Application and Plowing on Transport of Colloids and Phosphorus to Tile Drains
Vadose Zone J., March 8, 2006; 5(1): 445 - 458.
[Abstract] [Full Text] [PDF]

Home page
 Soil Sci.Home page
A. P. Mallarino and A. M. Atia
Correlation of a Resin Membrane Soil Phosphorus Test with Corn Yield and Routine Soil Tests
Soil Sci. Soc. Am. J., January 1, 2005; 69(1): 266 - 272.
[Abstract] [Full Text] [PDF]

Home page
 J. Environ. Qual.Home page
Z. He, T. S. Griffin, and C. W. Honeycutt
Enzymatic Hydrolysis of Organic Phosphorus in Swine Manure and Soil
J. Environ. Qual., January 1, 2004; 33(1): 367 - 372.
[Abstract] [Full Text] [PDF]

Home page
 J. Environ. Qual.Home page
B.J. Cade-Menun, C. W. Liu, R. Nunlist, and J. G. McColl
Soil and Litter Phosphorus-31 Nuclear Magnetic Resonance Spectroscopy: Extractants, Metals, and Phosphorus Relaxation Times
J. Environ. Qual., March 1, 2002; 31(2): 457 - 465.
[Abstract] [Full Text] [PDF]

Home page
 J. Environ. Qual.Home page
Z. He and C. W. Honeycutt
Enzymatic Characterization of Organic Phosphorus in Animal Manure
J. Environ. Qual., September 1, 2001; 30(5): 1685 - 1692.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Rubæk, G.H.
Right arrow Articles by Christensen, B.T.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Rubæk, G.H.
Right arrow Articles by Christensen, B.T.
Agricola
Right arrow Articles by Rubæk, G.H.
Right arrow Articles by Christensen, B.T.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Crop Science
Journal of Natural Resources
and Life Sciences Education		 Vadose Zone Journal
Journal of Plant Registrations Journal of
Environmental Quality		 The Plant Genome