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


     

Published online 1 May 2008
Published in Soil Sci Soc Am J 72:786-790 (2008)
DOI: 10.2136/sssaj2007.0178N
© 2008 Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
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 Pagel, H.
Right arrow Articles by Kaupenjohann, M.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Pagel, H.
Right arrow Articles by Kaupenjohann, M.
Agricola
Right arrow Articles by Pagel, H.
Right arrow Articles by Kaupenjohann, M.
Related Collections
Right arrow Phosphorus
Right arrow Colloids
Right arrow Nutrients

SOIL CHEMISTRY NOTE

Total Phosphorus Determination in Colloid-Containing Soil Solutions by Enhanced Persulfate Digestion

Holger Pagela, Katrin Ilgb, Jan Siemensb,* and Martin Kaupenjohannb

a Institute of Soil Science and Land Evaluation, Biogeophysics Section, Univ. of Hohenheim, D-70593 Stuttgart, Germany
b Department of Soil Science, Institute of Ecology, Berlin Univ. of Technology, Salzufer 11-12, D-10587 Berlin, Germany

* Corresponding author (jan.siemens{at}tu-berlin.de).


   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
It is unclear whether persulfate digestions completely liberate P associated with mineral colloids in soil solutions and aqueous soil extracts. We tested a modified persulfate digestion using aqueous soil extracts and suspensions of goethite, gibbsite, illite, and montmorillonite that were exposed to 24 µmol P L–1 as orthophosphate (ortho-P) or myo-inositol hexaphosphate (IHP). Digestion was performed by autoclaving (121°C, 60 min) 5 mL of extract or suspension after the addition of 1 mL of solution, which contained 150 mmol L–1 K2O8S2 and 180 mmol L–1 H2SO4. Subsequently, 0.7 mL of 188 mmol L–1 ascorbic acid was added and the sample was heated in a water bath at 95°C for 60 min. For the samples where P was sorbed onto goethite and gibbsite, recoveries of total P were 97.2 ± 3.7% (ortho-P) and 102 ± 3.7% (IHP). For the samples where P was sorbed onto clay minerals, average total P recoveries were 99.2 ± 5.1% (ortho-P) and 106 ± 3.0% (IHP). No differences between total P concentrations measured in persulfate-digested soil extracts or with an additional digestion with HF and HNO3 were detected.

Abbreviations: IHP, myo-inositol hexaphosphate • ortho-P, ortho-phosphate • PE, polyethylene


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
SUBSURFACE LOSSES OF P FROM AGRICULTURAL soils are among the main sources for excessive P inputs into water bodies, leading to the eutrophication of surface waters (Sims et al., 1998). Colloidal P may account for up to 80% of total P in soil water samples (Heathwaite et al., 2005).

In noncalcareous soils, the main sorbents for P are primarily Fe and Al (hydr)oxides and, to a lesser extent, clay minerals (Brady, 1990, p. 358–361). These soil constituents and organic matter represent major sources of mobile colloids found in soil solutions (Hayes and Bolt, 1991). Concentrations of P associated with colloids are commonly determined as the difference between total P and dissolved P concentrations (Ilg et al., 2005). Thus, this approach requires that the P bound to colloids be completely recovered by the methods that are used to determine total P concentrations in soil solutions and aqueous soil extracts.

Acid persulfate digestions are widely used to determine total P or total dissolved P in natural waters and soil solutions (Maher and Woo, 1998; Worsfold et al., 2005). Compared with other conventional methods, such as digestion with HClO4 or H2O2, persulfate digestions are easier to handle and show better precision (Maher and Woo, 1998). For organically bound P, complete recovery of P has been reported using acid persulfate digestion (Rowland and Haygarth, 1997). Since the acid persulfate digestion uses concentrated H2SO4, it can be expected that the method also partly releases P that is associated with mineral colloids. There is little research, however, addressing the efficiency of the persulfate digestion method regarding P bound to soil colloids. O'Connor and Syers (1975) ascribed an incomplete recovery of P with a standard acid persulfate digestion method (USEPA, 1983) to the imperfect release of P that was occluded in Fe (hydr)oxides.

Ascorbic acid is an efficient reducing agent for the dissolution of Fe(III) (hydr)oxides (Sulzberger et al., 1989). We extended the acid persulfate digestion by the addition of ascorbic acid, which is easier to handle and less hazardous than other techniques that are conventionally used to determine total element concentrations (e.g., digestions with HF or nitrohydrochloric acid [aqua regia]).

The objective of our study was to test whether an additional digestion with ascorbic acid would allow the determination of total P concentrations in persulfate-digested samples of aqueous solutions that contain mineral colloids.


   MATERIALS AND METHODS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 REFERENCES
 
Enhanced Acid Persulfate Digestion
We modified methods from Koroleff (1983) and Rowland and Haygarth (1997) to determine P in colloidal suspensions. Five milliliters of the suspensions were mixed with 1 mL of 150 mmol L–1 K2O8S2 and 180 mmol L–1 H2SO4 and autoclaved at 121°C for 60 min in 20-mL pressure-resistant capped glass vessels. After the mixtures were cooled to room temperature, we added 0.7 mL of 188 mmol L–1 ascorbic acid and heated the samples in a 95°C water bath for 60 min. We tested whether heating was necessary for a good recovery of P and Fe by omitting the heating step for two goethite suspensions: one to which IHP was added and one without the addition of P (see below for details of test suspensions). Reagents were always freshly prepared, and all vessels were rinsed with 0.1 mol L–1 HNO3 before use.

Model Systems
We used two goethites with different specific surface areas, gibbsite, illite, and montmorillonite as synthetic model compounds. Goethite 1 (80 m2 g–1) was prepared according to Schwertmann et al. (1985) by dissolving 0.1 mol L–1 FeCl3 in 0.7 mol L–1 KOH and storing the solution for 68 d at 4°C. Goethite 2 (120 m2 g–1) was prepared as described by Schwertmann and Cornell (2000, p. 78–82, Method 5.3). All other minerals were purchased: gibbsite from Nabaltec GmbH (Apyral 120E, Schwandorf, Germany), illite from Dr. F. Krantz, Rheinisches Mineralien-Kontor (Bonn, Germany), and montmorillonite from Sigma-Aldrich (St. Louis, MO). We prepared stock suspensions of goethite, gibbsite, and illite by dispersing each mineral in 10 mmol L–1 KNO3. Montmorillonite had to be treated differently, because it coagulated in the KNO3 background electrolyte. We dialyzed montmorillonite against deionized H2O until the water had an electrical conductivity (inoLab pH/Cond, WTW, Weilheim, Germany) below 5 mS m–1 and separated the fraction <2 µm by dispersion in deionized H2O and sedimentation of larger particles.

We used ortho-P (KH2PO4) and IHP (C6H18O24P6·12 Na x H2O) to test the recovery of the proposed method. Inositol phosphates represent the major components of organic P in most soils (Turner et al., 2002) accounting for 10 to 50% of organic P (Stevenson, 1986, p. 258–263). Furthermore, it is considered to be stable against digestion (Kerouel and Aminot, 1996). Stock solutions of ortho-P and IHP were prepared both in 10 mmol L–1 KNO3 and in deionized H2O (for montmorillonite). We adjusted the pH of the stock solutions to 5.5 using diluted HNO3 or NaOH solutions.

To sorb P to minerals, we mixed 5 mL of ortho-P or IHP stock solution with 5 mL of mineral stock suspension and added 40 mL 10 mmol L–1 KNO3 for all minerals except for montmorillonite, for which we used deionized H2O. The concentration of total P in the suspensions was 24 µmol P L–1 for both P compounds. Although this concentration is above reported average values found in most soil solutions, similarly large P concentrations may be found in highly fertilized soils (Barber, 1984; Beauchemin et al., 1998; Hens and Merckx, 2001). Final mineral concentrations in the suspensions were 20 mg L–1 (goethite), 100 mg L–1 (gibbsite), 250 mg L–1 (illite) and 90 mg L–1 (montmorillonite). These values are comparable with typical particle concentrations in soil solutions, which are in the range of 10 to 100 mg L–1 but can reach levels of up to several hundred milligrams per liter (Kaplan et al., 1993; Schelde et al., 2002; de Jonge et al., 2004). The suspensions were shaken end over end for 24 h at 10 rpm in 60-mL polyethylene (PE) bottles. Blank samples either without minerals or without P were run to check if a sorption of P to the PE bottles occurred, or if there was a release of P from the minerals. After shaking, we ultracentrifuged an aliquot of each sample at 150,000 x g for 40 min at 25°C to remove all particles >10 nm (Optima L-90k Ultracentrifuge, Beckmann Coulter, Krefeld, Germany) for the determination of dissolved P concentrations. In batch samples, the average particle size (High Performance Particle Sizer HPP 5001, Malvern Instruments, Malvern, UK) was determined before digestion.

We visually observed that goethite colloids were partly adsorbed to the walls of the PE bottles. Therefore, Fe recoveries could not be calculated based on the nominal concentration of Fe, but a reference method for Fe determination was used to quantify the target Fe concentration. To this end, we additionally digested aliquots of goethite samples according to Zeien and Brümmer (1989): to 7-mL samples we added 2 mL of solution containing 0.2 mol L–1 oxalic acid and 0.1 mol L–1 ascorbic acid and heated the samples in a 95°C water bath for 60 min. The stock suspensions of both goethites were digested with the same procedure. In the case of the gibbsite stock suspension, we applied a standard method (USEPA, 1994, Method 3015) for digestion: we added 0.5 mL of concentrated HNO3 to a 4.5-mL sample and heated the sample for 20 min at 170°C in a microwave oven (Mars, CEM, Kamp-Lintfort, Germany).

We measured P concentrations according to the method of Murphy and Riley (1962) using a flow injection system (Skalar, Erkelenz, Germany). The detection limit was 1.5 µmol P L–1. Concentrations of Fe in digested goethite samples were measured photometrically at a wavelength of 562 nm (Specord photometer, Jena, Germany) using the method of Dominik and Kaupenjohann (2000) with a detection limit of 0.4 µmol Fe L–1. We determined Al concentrations in gibbsite samples by flame atomic absorption spectrometry (Model 1100B, PerkinElmer, Wellesley, MA) with a detection limit of 37 µmol Al L–1.

As an additional measure of digestion efficiency, we quantified the optical density in all batch samples before and after applying the proposed digestion. The optical density is a dimensionless measure of the colloid concentration and was determined as the absorbance at a wavelength of 525 nm according to Kretzschmar et al. (1997), with a detection limit of an absorbance of 0.008 (Specord photometer, Jena, Germany).

If determinations of ortho-P concentrations in persulfate-digested samples are to be performed according to the method of Murphy and Riley (1962) manually without a flow injection system, we suggest mixing 0.9-mL samples with 0.1 mL of a coloring reagent. The coloring reagent should contain 80 mL H2SO4 L–1, 9.6 g L–1 ammonium heptamolybdate tetrahydrate ((NH4)6Mo7O24·4 H2O), 0.22 g L–1 potassium antimony(III) oxide tartrate hemihydrate [K(SbO)C4H4O6·0.5 H2O] and 8.45 g L–1 ascorbic acid to meet the optimal range of [H+]/[MoO42–] ratios determined by Drummond and Maher (1995).

To evaluate the proposed method, we determined the proportion of P that was sorbed to minerals (Eq. [1]) and calculated P recoveries after persulfate digestion according to Eq. [2] and [3]:

Formula 1[1]

Formula 2[2]

Formula 3[3]
where Psorb is the concentration of sorbed P (mmol kg–1), Padd is the concentration of added P as determined in samples without minerals (µmol L–1), Pdis is the concentration of dissolved P in the supernatant after ultracentrifugation (µmol L–1), cmin is the concentration of the mineral (g L–1), and Ptot is the concentration of total P after the proposed digestion (µmol L–1).

In the case of goethite samples, we considered ortho-P or IHP that was sorbed to goethite retained in the PE bottles. For this purpose, we substracted Pret from Padd in Eq. [1], [2] and [3]:

Formula 4[4]
where Pret is the concentration of P sorbed to the retained goethite (µmol L–1), Feadd is the concentration of added Fe after oxalate digestion (µmol L–1), and Feref is the concentration of total Fe after oxalate digestion (µmol L–1).

To evaluate the persulfate digestion efficiency for Fe and Al in hydroxides, we calculated their recoveries according to

Formula 5[5]

Formula 6[6]
where Fetot is the concentration of total Fe after the proposed digestion (µmol L–1), Altot is the concentration of total Al after the proposed digestion (µmol L–1), and Aladd is the concentration of added Al after HNO3 digestion (µmol L–1).

Soil Extracts
To test the proposed method for natural soil extracts, we used samples from Ilg et al. (2008). In this experiment, soil samples from the Bw horizon of a Cambisol (Typic Dystrudept) and from the mottles of the Bg horizon of a Gleysol (Typic Endoaquept), both originating from the same catena, were exposed to increasing concentrations of IHP in batch experiments for 24 h. The soil extracts were filtered through 1.2-µm membranes and treated by the proposed method to determine concentrations of colloidal P. All samples were subsequently digested with concentrated HF and HNO3 (4.5 mL sample, 1.5 mL HF, and 0.5 mL HNO3) at 185°C for 20 min in a microwave oven according to Bundesministerium für Ernährung (2005).

Both batch experiments were run in triplicate. Arithmetic means and standard deviations were calculated for all data. Values below the detection limit were set to one-half of the detection limit. Calculated differences below zero were set to zero.


   RESULTS AND DISCUSSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
REFERENCES
 
The average particle sizes for both goethites, gibbsite (except the blank variant), and montmorillonite are in the clay fraction, <2 µm (Table 1 ). In illite samples, coagulation of colloids might be responsible for particle sizes >10 µm.


View this table:
[in this window]
[in a new window]

  		Table 1. Particle size and optical density of colloidal suspensions before and after digestion, recoveries of P, Fe, and Al, and amount of sorbed P in batch suspensions. The data represent mean values of the replicates; standard deviations are given in parentheses.

 
Optical densities below the detection limit of goethite and gibbsite suspensions after the proposed digestion indicate that the proposed method was able to dissolve most goethite and gibbsite colloids. Contrary to goethite and gibbsite, clay minerals were not completely destroyed because the use of HF is necessary for the total solubilization of silica-based materials (Hoenig, 2001).

Among goethites, average recoveries of total P and Fe after the proposed digestion were larger for samples exposed to IHP than for samples exposed to ortho-P (Table 1). Furthermore, sorbed P was completely recovered in goethite samples exposed to IHP, whereas the recovery of sorbed P was <90% for samples exposed to ortho-P. We found the smallest P and Fe recoveries for Goethite 1 exposed to ortho-P. We ascribe this to the smaller specific surface area of Goethite 1 compared with Goethite 2 and to the average particle size, which was larger for Goethite 1 exposed to ortho-P. The larger crystallinity and the formation of microaggregates probably impeded the dissolution of Goethite 1 exposed to ortho-P.

To calculate correct recoveries of P and Fe for goethite suspensions, it was crucial to consider the amount of Fe hydroxides retained on the walls of the PE bottles. For both goethites, the amount of retained Fe was affected by P addition and increased in the order IHP << ortho-P < blank. We found that up to half of the added goethite was retained in PE bottles (data not shown).

Smaller recoveries of P and Fe for the unheated variant of Goethite 2 exposed to IHP compared with the heated variant indicate that heating after the addition of ascorbic acid is essential to achieve a good recovery of P (Table 1).

In all gibbsite samples, total and sorbed P was completely recovered (Table 1). For gibbsite exposed to IHP, we found complete recovery of Al, but in ortho-P and blank variants, the recovery of Al decreased to 70%.

In clay mineral suspensions, both P compounds were completely recovered. In contrast to suspensions of metal hydroxides, however, the concentration of sorbed P was very small because primarly crystal edges contribute to the P sorption capacity of clay minerals (Parfitt, 1978). Standard deviations for the recovery of P sorbed to clay minerals were large (Table 1) because large background concentrations of P, released from the pure clay minerals, had to be substracted from P concentrations of samples that were exposed to ortho-P or IHP.

Digestion with HF and HNO3 released no additional P from soil extracts compared with the proposed enhanced persulfate digestion as indicated by the slope of the regression line, which was 1.0 (Fig. 1 ). Hence, P bound to colloids of two sandy subsoils seemed to be recovered completely by the proposed method, which is less harzardous, less time consuming, and easier to handle than the HF digestion.


Figure 1
View larger version (13K):
[in this window]
[in a new window]

  		Fig. 1. Concentrations of total P in 1.2-µm filtered soil extracts from the Bw horizon of a Cambisol (Typic Dystrudept) and the Bg horizon of a Gleysol (Typic Endoaquept) determined with the proposed enhanced persulfate digestion method plotted against concentrations determined after HF digestion. The slope of the regression line equals 1.0.

 
For the interpretation of both batch experiments, it has to be considered that P sorbed after 24 h of equilibration time is mainly reversibly bound on the surface of minerals and soils in a fast reaction, whereas slow reactions like diffusion into pores play a subordinate role (Parfitt, 1978; McGechan and Lewis, 2002). Nevertheless, the batch experiments with synthetic systems showed that Fe and Al (hydr)oxides, which represent major sorbents for P, are largely dissolved by the proposed method. Therefore, residual P bound to these minerals should be released. In contrast, clay minerals are not completely destroyed; however, the good recovery of total P in batch systems with illite and montmorillonite suspensions as well as the good agreement of total P concentrations determined after HF digestion with those determined following the modified persulfate digestion indicate that the incomplete destruction of clay minerals does not impede the quantitative determination of total P.

Our findings suggest that the extended acid persulfate digestion is an appropriate method to determine the concentration of P bound to colloids in aqueous soil extracts and soil solutions. The results of synthetic systems, however, show that inorganic P sorbed to goethite may be slightly underestimated. We cannot conclude from our selection of test suspensions whether the extra ascorbic acid digestion step is needed only for some, most, or all soil solutions. Probably the underestimation of total P concentrations by the common persulfate digestion procedure of Rowland and Haygarth (1997) will be more pronounced for colloid-dominated soil solutions, which are derived from mineral soils with large amounts of Fe and Al oxides or clay minerals, than for soil solutions that contain predominantly organically bound P forms. This means that the magnitude of the bias introduced by incomplete release of P from mineral colloids cannot be generalized but has to be determined for individual sample types. Since the ascorbic acid digestion procedure is simple and time efficient, we recommend the additional digestion step for all colloid-containing soil solution samples.


   ACKNOWLEDGMENTS
 
We thank Claudia Kuntz, Michael Facklam and Nadine Kurowski for their support and technical assistance. We are grateful to Dr. Peter Dominik for technical assistance and inspiring discussions. Furthermore, we would like to thank Alex Toland for language editing This study was funded by the Deutsche Forschungs gemeinschaft (Grant no. KA1139/10-1,-2)


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permission for printing and for reprinting the material contained herein has been obtained by the publisher.

Received for publication May 15, 2007.


   REFERENCES
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 




This article has been cited by other articles:


Home page
 J. Environ. Qual.Home page
J. Siemens, K. Ilg, H. Pagel, and M. Kaupenjohann
Is Colloid-Facilitated Phosphorus Leaching Triggered by Phosphorus Accumulation in Sandy Soils?
J. Environ. Qual., October 23, 2008; 37(6): 2100 - 2107.
[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 Pagel, H.
Right arrow Articles by Kaupenjohann, M.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Pagel, H.
Right arrow Articles by Kaupenjohann, M.
Agricola
Right arrow Articles by Pagel, H.
Right arrow Articles by Kaupenjohann, M.
Related Collections
Right arrow Phosphorus
Right arrow Colloids
Right arrow Nutrients

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