Dana Zimmer, Rhena Schumann

Question:

It has to be estimated, which weight of sample material is necessary and in which range the P concentration is in the extract. Th estimation results from estimated P concentrations in environmental samples (see chapter 1) and a standard digestion method.

Known requirements:

►    An estimated P concentration range of the sample has to be known (see chapter 1): e.g. in bone char 100…150 g P kg-1. In the following formula it is calculated with 100 g P to make it easier. Calculation for 150 g P are analogue.

►    A standard procedure including supposed weigh-in, dilution and so on is selected. Bone char has an organic matrix and can therefore be digested according to a standard method for plant material. Normally, plant material is processed as in the following:

         ► Weigh in 0.1 g sample
         ► Digest with 5 ml conc. HNO₃ and 3 ml 30 % H₂O₂
         ► Fill to 100 ml with ultra-pure water
         ► Measure P at ICP-OES
         ► The middle standard has a concentration of 10 mg P per litre
         ► the P concentration in the extract should be in the range of the middle standard of calibration line and/or not     exceed highest standard by a factor of 10

Stepwise procedure to estimate P concentration in the extract according to the standard procedure:

► the estimated P concentration in the sample (100…150 g P per 1000 g) is converted to the standard weigh-in by a ratio equation:

conversion results in the following formula:

In 0.1 g bone char there are 0.01 to 0.015 g P.

► This amount of P in the weigh-in mass corresponds to the P amount in the extract after microwave digestion (from this weigh-in). If after the microwave digestion the extract with HNO₃ and H₂O₂ is filled to 100 ml with ultra-pure water, between 0.01 to 0.015 g P can be expected in the 100 ml extract.

► This P amount in 100 ml extract is converted by the ratio equation to P concentrations per litre:

conversion results in the following formula:

► The extract has a P concentration between 0.1 and 0.15 g P per litre. This corresponds to a concentration of 100 to 150 mg P per litre in the extract.

Comparison of the estimated P concentration to the standards:

► Comparison
    ► P concentration of the middle standard: 10 mg P per litre
    ► P concentration of the extract 100 to 150 mg P per litre
    ► For a standard weigh-in of 0,1 g bone char and filling to 100 ml the P concentration in the extract is 10 to 15 times as high as the middle standard.
► Conclusion
    ► P concentration in the extract should be by a fifth to tenth lower!
► Generally, there are 2 opportunities to achieve this:
    ► Decreasing the weigh-in to a tenth: weigh-in only 0.01 g bone char instead of 0.1 g
    ► Dilution of extracts by a factor of 5 to 10

Evaluation of pros and cons of both opportunities

Lower weigh-in:

► Advantages

► Less sample material is necessary.
► No further dilution of extracts is necessary, which might cause dilution errors.
► Less extraction agents are necessary.

► Disadvantages

► If the material can statically charge, lower weigh-in can be problematic.
► For heterogenic material lower weigh-in can increase standard deviation. Either the material has to be homogenised (e.g. milling) or the number of weight-ins has to be increased.
► If other elements than P shall be measured, their concentration could fall below the detection limit if the weigh-in decreased.

5- to 10-times dilution

► Advantages

► No/less problems, which can be caused by lower weigh-in (statically charge, heterogeneity)
► Other necessary elements are in the measurement range or by different dilution the ideal concentration ranges of different elements can be achieved.

► Disadvantages

► Dilution error
► Higher amounts of chemicals are needed.

Decision with weighting of individual points (most important first)

► A lower weigh-in should be considered, if: 1. the material is relatively homogenic, 2. the material is not statically charged during weighing-in, 3. No other elements have to be determined in the extract and 4. less sample material is available.
► A dilution by factor 5 to 10 should be selected, if: 1. other elements have to be determined, 2. the material is heterogenic, 3. static charges could cause problems during weighing-in and 4. sufficient material is available.

Suggestions for adjustment of the digestion method for bone char

The decrease of weigh-in by a factor of ten decreases, at a standard volume of 100 ml, the P concentration in the extract (from 100 to 150 mg l-1) to 10 to 15 mg P per litre and less chemicals are necessary for the extract (Tab. 4.1.1-1). If the volume is filled not to 100 but to 50 ml, the P concentration increased to 20 to 30 mg P per litre. This is still in the measurement range for the ICP-OES, and enables simultaneously to determine other elements. If trace elements such as Cd, Cu and Zn have to be determined, the end-volume can be decreased to 20 to 25 ml. Under theses circumstances the P concentration increased to 40 to 60 mg P per litre. That means that a further dilution of the extracts could be necessary for P measurement.

For citation: Zimmer D, Schumann R (year of download) Chapter 4.1.1 Estimation of Weight (Version 1.0) in Zimmer D, Baumann K, Berthold M, Schumann R: Handbook on the Selection of Methods for Digestion and Determination of Total Phosphorus in Environmental Samples. DOI: 10.12754/misc-2020-0001

download chapter (PDF)

Rhena Schumann, Maximilian Berthold, Dana Zimmer

Suitability

Phosphorus in sediments consists of numerous, quite different available fractions. Phosphate in the interstitial water is easily available. However, this fraction is quantitatively of minor importance. Phosphorus is hardly available if it is organically bound, for example in biomass or in detritus. Some phosphate salts, which only can be dissolved under anoxic conditions, phosphate ions being adsorbed to clay minerals and other inorganic particles are difficult to access. (Berthold et al. 2018, Nausch 1981).

Large amounts of bound phosphates are removed under acidic conditions (HCl, HNO₃ or H₂SO₄) and by adding strong oxidising agents (H₂O₂ or persulfate). The usage of ashes instead of dry matter often improves the digestion yield. If there are higher iron concentrations in the sediments, discolouration can complicate neutralisation. Other slightly soluble inorganics salts can complicate P determination as well. Phosphate ions react in acidic solutions with molybdate to molybdenum blue. Molybdenum blue is quantified photometrically. With phosphate relatively insoluble compounds (such as metal ions, iron, calcium and aluminium) can be determined in the same sediment extract.

High phosphate concentrations strengthen eutrophication processes because in the often anoxic sediment a lot of phosphate is mobile and can be re-dissolved into the water column as well. TP is a basic parameter, demonstrating the total load with phosphorus. However, almost no conclusions can be drawn from TP concentrations for the phytoplankton availability of P.

Necessary sediment parameters:

►  Determine water content and loss of ignition of the sediment sample!

►  For a reference to area and volume the dry matter density is necessary. Alternatively, the dry matter density can be estimated from a correlation between water content and dry matter density (Berthold et al. 2018).

Protocol

Preparation of samples:

►  Sediments and soils have to be sieved < 2 mm. The fraction < 2 mm (fine soil) is used for further analysis.

►  Grinding: The reproducibility is improved. However, grinding affects the availability of element in soils and sediments. Recommendations for mills, time of grinding and the energy input are extremely diverse and have to be adapted to the specific samples.

►  Ashing: The advantage is the better availability of organically bound phosphate (organic substances are burned). However, inorganic compounds can be transformed. For example, normally yellow-brown goethite, a Fe(III)-oxid, can be transformed at 500 °C to reddish hematite changing the binding of phosphate (Derie et al. 1976, Prasad et al. 2006). Poorly soluble salts such as Ca₃(PO₄)₂ remain mainly unavailable.

Procedure:

Further processing / neutralisation:

Samples do not have to be neutralised, if they do not have to be filtered (fine-grinded material), be measured manually immediately (not in Continuous Flow Analyser) and no retention samples have to be stored in PE-tubes (see below).

► without neutralisation:

► Cool down and transfer solution completely into a 50 ml volumetric flask and fill with ultra-pure water to the calibration mark.

► with neutralisation (work under a laboratory hood):

ATTENTION: At high Fe concentrations (sample discolours red after ashing, Fig. 4.2.1-1) Fe can flocculate as yellowish flakes by addition of NaOH! Take care for re-dissolution of flakes by HCl addition!

► with filtration:

Samples with high iron concentration (visible as red discolouration after ashing, Fig. 4.2.1-1) can disturb titration by the development of iron flakes in the solution by addition of NaOH. Normally, iron flakes re-dissolve by addition of acid during neutralisation, but this has to be checked. Otherwise, turbidity by flakes and binding of P to the iron flakes can affect photometric P measurement (increase turbidity, removal of measurable P from solution).

Further processing:

If extinction is ≥ 0.8 (in a 5 cm cuvette), discard sample and begin anew with a diluted sample! Empirical dilution for ashed sediment samples:

Calculations:

Hints for quality management for acidic persulfate digestion:

Further hints for quality management see chapter 6!

Chemicals:

available soon

Dana Zimmer, Sebastian Marcus Strauch, Rhena Schumann

All works take place under the fume hood! Check if the available fume hood is suitable for HClO₄!

► Day 1: Preparation and HClO₄ digestion

► Put on protective clothing (gloves, apron, glasses).
► Weigh in 250 mg dried fish sample in Teflon vessels (Heinrichs et al. 1986).
► To oxidize the organic substance 3 ml conc. HNO₃ are added and the samples are heated to 60 °C on a hotplate for 1 hour.
► Subsequently, 3 ml conc. HClO₄ are added and the Teflon vessels are closed.
► The vessels are heated in the oven to 185 °C for 12 hours.

► Day 2: Quantitative transfer for the determination of elements

► Afterwards, the vessels are opened carefully and
► the acids are evaporated at 185 °C on the heating plate until the samples are almost dry.
► Subsequently, the samples are evaporated 3 times with 2 ml 1 + 1 HCl.
► Add 5 ml 2 vol. % HNO₃ to the samples and handle for 1 h at 60 °C.
► After cooling down, the sample is transferred into a 50 ml centrifuge tube and
► rinsed with 2 vol. % HNO₃ from the Teflon vessel and filled with             2 vol. % HNO₃ up to 50 ml in the centrifuge tube.
► The determination of element took place at ICP-OES at IOW (iCAP 7400 Duo, Thermo Fisher Scientific).

Reference

Heinrichs, H, Brumsack, HJ, Loftfield, N, König, N (1986) Verbessertes Druckaufschlußsystem für biologische und anorganische Materialien. J Plant Nutr Soil Sci 149, 350-353, DOI: 10.1002/jpln.19861490313

For citation: Zimmer D, Strauch SM, Schumann R (year of download) Chapter 4.3.3 Digestion with HClO₄: Fish Meat and Fish Bones (Version 1.0) in Zimmer D, Baumann K, Berthold M, Schumann R: Handbook on the Selection of Methods for Digestion and Determination of Total Phosphorus in Environmental Samples. DOI: 10.12754/misc-2020-0001

download chapter (PDF)

Dana Zimmer, Rhena Schumann

When there are increased iron concentrations in soil and sediment samples the sample turns red after ashing, since the present yellow-brownish ferrihydrite resp. goethite dehydrates between 500 and 600 °C and converts to red hematite (Derie et al. 1976, Prasad et al. 2006, Schwertmann 1959). If, after sample digestion with acid persulfate or HCl, the sample is alkalized by ammonia (colour change of nitrophenol from colourless to yellow, subsequently neutralised with HCl, colour change to colourless), the iron precipitates as yellow flakes (probably ferryhydrate) after addition of ammonia (Fig. 4.4.1, Schwertmann et al. 2000, p. 73 ff.). Normally, theses flakes dissolve after addition of HCl. Nevertheless, sample extracts have to be checked for small flakes after reaching the point of colour change (from yellow to colourless). If necessary, 1 or 2 drops of HCl have to be added to dissolve flakes completely. It must be ensured that no flake exists any longer, since iron is a strong sorbent for phosphorus. For this reason, phosphate concentrations could be underestimated by photometric P determination if P is precipitated and sedimented with Fe flakes. Furthermore, such flakes would block the hose system of the ICP-OES.

References

Derie R, Ghodsi M, Calvo-Roche C (1976) DTA study of the dehydration of synthetic goethite αFeOOH. J Thermal Anal 9: 435-440, DOI: 10.1007/BF01909409

Prasad, PSR, Shiva Prasadad, K, Krishna Chaitanya, V, Babua EVSSK, Sreedhar, B, Ramana Murthy, S (2006) In situ FTIR study on the dehydration of natural goethite. J Asian Earth Sci 27, 503-511, DOI: 10.1016/j.jseaes.2005.05.005

Schwertmann, U (1959) Die fraktionierte Extraktion der freien Eisenoxyde in Boden, ihre mineralogischen Formen und ihre Entstehungsweisen. J Plant Nutr Soil Sci 84, 194-204, DOI: 10.1002/jpln.19590840131

Schwertmann U, Cornell R M (2000) Pure Goethite from Fe III Systems. In Schwertmann and Cornell Iron oxides in the laboratory. Wiley VCH, 2nd ed., Weinheim, DOI: 10.1002/9783527613229.ch05

For citation: Zimmer D, Schumann R (year of download) Chapter 4.4.1 Neutralisation for Increased Iron Concentrations (Version 1.0) in Zimmer D, Baumann K, Berthold M, Schumann R: Handbook on the Selection of Methods for Digestion and Determination of Total Phosphorus in Environmental Samples. DOI: 10.12754/misc-2020-0001

download chapter (PDF)

available soon

Dana Zimmer, Karen Baumann

Principle and suitability of sequential P fractionation

In the soil, phosphorus is bound to different soil components and can therefore be mobilized and bioavailable in different ways. There are various inorganic and organic P fractions in the soil P pool, which can be regarded as labile, moderately labile, relatively insoluble and stable (long-term availability) P pools from the point of view of P plant availability. Various fractionation methods have been developed to differentiate between these P-forms. Most sequential P fractionations first extract a "weakly bound" fraction with a salt solution (e.g. NH₄Cl), followed by an extraction of Fe- and Al-bound P with an alkaline extractant (e.g. NaOH) and finally an acidic extraction (e.g. HCl) to extract Ca-bound P (Condron and Newman, 2011). In addition, a distinction is made in some cases between organic and inorganic P in the individual fractions by means of P determination using molybdenum blue (MB) and ICP-OES/-MS. If P in the extracts is determined using both methods, the P concentration using ICP-OES or -MS is interpreted as total P (P_t) and that using MB as inorganic P (P_i) and the difference between the two as organic P (P_o). Since the acidic environment of the MB reagent converts an unknown proportion of the labile organic P to phosphate and thus overestimates the proportion of P_i or, alternatively, unknown proportions of non-reactive inorganic P are present, which leads to an underestimation of P_i (e.g. Cade-Menun and Liu 2013, Condron and Newman 2011), the term molybdate-reactive P and non-reactive P is the better term (Haygarth and Sharpley 2000, Felgentreu et al. 2018).
One of the most common sequential P fractionations is the fractionation according to Hedley et al. (1982) or Thiessen and Moir (1993) (Alamgir and Marschner 2013 a, b). The modified Hedley fractionation, as carried out in the Agronomy and Soil Science working groups of the Faculty of Agricultural and Environmental Sciences (University of Rostock), comprises the following extraction steps in sequence (F1) water-anion resin, (F2) NaHCO₃, (F3) NaOH and (F4) HCl or H₂SO₄.

Note:
► This P fractionation is normally used for agricultural soils but is generally suitable for terrestrial mineral soils. If it is applied to semi-terrestrial (e.g. gleys), semi-sub-hydric (e.g. mudflats) and sub-hydric (e.g. gyttja) soils, bogs, marine sediments or substrates such as manure, the results must be interpreted with even greater caution, as these substrates may have pH and Eh values (see Chapter 2.3) and binding partners for P (e.g. concentration of organic matter) that differ greatly from terrestrial soils.

Interpretation of the results

In this sequential P-fractionation, the fractions can generally be interpreted as follows, although it should be noted that, as with all sequential fractionations, the fractions are operationally defined and do not correspond 100% to the interpretations (Bacon and Davidson 2008).

► F1: Resin-P (labile P): exchangeable P, superficially sorbed, readily available to plants, reflects the removal of phosphate from the extraction water by the anion exchange resin, the removal by the plant roots (compared to a cold-water extract, where a solubility equilibrium between the soil sample and the extraction water is established more quickly).
► F2: NaHCO₃-P (labile P), easily mineralizable, plant-available P (simulates root respiration: formation of HCO₃^- from CO₂ release)
► F3: NaOH-P is moderately labile P and therefore available in the medium or long term, NaOH-P is considered to be P bound to Al-Fe or humic substances.
► F4: H₂SO₄ P: P bound in Ca or carbonate
► F5: residual P = total P (TP from aqua regia extract of the soil sample and ICP-OES measurement) minus the sum of fractions F1...F4 (P from ICP-OES measurements); or determine TP in the extraction residue, only long-term available P

The ICP-OES measures total P in the respective extract, while the MB method can be used to measure the reactive phosphate P and thus approximately the inorganic P content (P_i) in the extract. The difference between the two measurements gives approximately the organic P content (P_o). It should be noted that the use of HCl and H₂SO₄ in the course of extraction can already convert parts of organic P compounds into free phosphate-P. Therefore, the terms molybdate-reactive P and non-reactive P should generally be used instead of P_i and P_o (Cade-Menun and Liu 2013, Haygarth and Sharpley, 2000).

The photometric P determination with the molybdenum blue method is only possible on colorless, undimmed extracts! Particularly in soil samples with high concentrations of organic matter (e.g. peat), the extracts are often dark-colored (especially the NaOH extract); this means that P determination with MB is not useful. An attempt can be made to dilute the extracts accordingly so that the extracts lighten in color.

Protocol for sequential P fractionation

Material and chemicals required for 24 samples + 2 blank values + solutions for ICP-OES standards

► Sufficient solutions must be prepared for the extractions themselves and for preparing the standards for ICP and, if necessary, MB measurement.
► If several runs of sequential P fractionation or a higher number of samples are planned, correspondingly larger quantities of chemicals should be prepared in order to use the same solutions for all extracts and for the standards for the calibration lines.

Preparation of the resin strips

► Anion exchanger membrane BDH #55164 2S, cut into 12 strips of 6 x 2 cm each
► Storage in ultrapure water (UW) in the refrigerator
► Prepare 2 L 0.5 M NaHCO₃ and fill into two 1-liter beakers
► Place resin strip in first beaker for 1 h, transfer to second beaker with tweezers for 1 h
► Wash resin strips 3 times by placing them in beakers with UW (move with tweezers, place tweezers in UW before use)
► Storage in UW in the refrigerator (24 h before use, after preparation with HCO₃^-)

Preparation of chemicals

2 liters 1 M HCl (washing of the resin strips) for F1

► Fill 2-liter flask to approx. 1.7 L with UW, add 166 ml 37 % HCl
► After cooling, fill up to 2 liters with UW

5 liters 0,5 M NaHCO3 (pH 8,5) for F2

► Add 210 g NaHCO₃ to a 5 L flask and fill up to approx. 4 liters with UW
► Adjust the pH value with 1 M NaOH (approx. 50 to 100 ml required)
► Fill up to 5 liters with UW

1 liter 1 M NaOH for pH adjustment

► Fill 1-liter flask with approx. 700 ml UW, add 40 g NaOH pellets, fill incompletely with UW
► Allow to cool, fill up to 1 liter with UW

3 liters 0,1 M NaOH for F3: Prepare 1 liter + 2 liters (if no 3-liter flask is available)

► Fill 1-liter flask to approx. 700 ml with UW, add 4 g NaOH pellets, after cooling fill to 1 liter with UW
► Fill 2-liter flask to approx. 1.5 L with UW, add 8 g NaOH pellets, after cooling fill to 2 liters with UW

3 liters 1 M H₂SO₄ for F4: Prepare 1 liter + 2 liters (if no 3-liter flask is available)

► Fill a 1-liter flask with approx. 700 ml UW and add 55 ml H₂SO₄ (95-97 %)
► Fill up to 1 L with UW the next day after cooling down
► Fill a 2-liter flask with approx. 1.5 liters of RW and add 110 ml of H₂SO₄ (95-97 %),
► Fill up to 2 L with UW the next day after cooling down

Sample preparation:

► Drying soil samples (see chapters 2.4 and 3.1)
► Sieve soil samples <2 mm and use the <2 mm fraction (fine soil)
► Determine total element concentrations in a subsample (e.g. using aqua regia extract, see chapter 4.1.2)
► Sequential extraction must be started on Monday so that the fourth fraction is ready on Friday
► The sample can also be weighed in the previous week.
► Prepare the anion exchange resin (see above: Preparation of the resin strips)

Procedure:

► Prepare at least 3 replicates per soil sample and at least 1 blank value per 10 extraction samples.
► Weigh 0.5 g of fine soil into 50 ml centrifuge tubes
► For F1: Add 30 ml ultrapure water (UW) and a strip of anion exchange resin, shake in an overhead shaker for 18 hours (start at approx. 2 pm)
► Remove resin strips with tweezers, rinse adhering soil particles with UW (spray bottle) back into the centrifuge tube
► Wash P from resin strips with max. 45 ml 1 M HCl via funnel with filter (P-free) in 50 ml volumetric flask
► Place the resin strips in beakers with UW, later place in the refrigerator
► Fill the graduated flask with 1 M HCl to 50 ml (F1)
► Fill aliquots into ICP tubes ((1.) Determine Pt of F1 on the ICP and if necessary (2.) P_i photometrically, difference = P_o)
► For F2: Add 30 ml 0.5 M NaHCO₃ to the soil sample, mix briefly and shake in an overhead shaker for 18 h (start approx. 2 pm)
► Centrifuge at 2500 x g for 20 min
► Filter the supernatant into a 100 ml volumetric flask (funnel + filter)
► For washing, add another 30 ml of 0.5 M NaHCO₃ to the soil sample, mix by hand and centrifuge at 2500 x g for 20 min
► Add the supernatant to the volumetric flask (combine the filtrates) and make up to 100 ml with 0.5 M NaHCO₃, shake well
► Fill 10 ml of the extract into an Erlmeyer flask and slowly (!) add 1 ml of conc. HCl to destroy HCO₃^- for the ICP measurement!
► Leave the Erlmeyer flask under the fume cupboard overnight for outgassing and add 9 ml UW to the sample (for ICP measurement) in the Erlmeyer flask the next day (fraction F2 labile P_t at the ICP)
► if necessary, determine labile Pi photometrically (second tube) with MB, difference to P_t = P_o); do not destroy this sample with HCl
► For F3: Add 30 ml 0.1 M NaOH to the soil sample in the centrifuge tube, shake in an overhead shaker for 18 h (start approx. 2 pm)
► Centrifuge at 2500 x g for 20 min
► Filter the supernatant into a 100 ml volumetric flask
► For washing, add another 30 ml of 0.1 M NaOH to the soil sample, mix by hand and centrifuge again at 2500 x g for 20 min
► Also filter the supernatant into the volumetric flask and make up to 100 ml with 0.1 M NaOH (F3)
► Fill aliquots into ICP tubes ((1.) Determine Pt in NaOH (F3) on the ICP and if necessary (2.) P_i photometrically, difference = P_o)
► For F4: Add 30 ml 1 M H₂SO₄ to the soil sample in the centrifuge tube under the fume cupboard, shake in an overhead shaker for 18 h (start approx. 2 pm)
► Filter the extract into a 100 ml flask. Since H₂SO₄ vapors corrode the centrifuge, do not centrifuge!
► Fill the flask with 1 M H2SO4 to 100 ml (F4)
► Fill aliquots into ICP vessels ((1.) Determine Pt in H₂SO₄ on the ICP and if necessary (2.) P_i photometrically, difference = Po)
► For F5: either dry the extraction residue and determine the total element concentrations in it using aqua regia and ICP-OES or subtract the sum of fractions F1 to F4 from the total element concentration (e.g. in the aqua regia extract) of the untreated soil samples.

Notes:

► If F5 is determined in the extraction residue, the sum of the P concentrations of F1 to F5 should theoretically correspond to the total element concentration in the untreated soil sample after aqua regia extract. However, it is possible that the sum of the 5 fractions is greater than the total element concentration. This is caused by the fact that aqua regia extracts only provide so-called pseudo-total concentrations (silicates are not broken down) and sequential P fractionation may release higher proportions of P. Therefore, F5 is usually calculated as the difference between the total P concentration and the sum of F1 to F4 as residual P.
► Only aliquots of the fractions are required for P determination. The retained samples of the extracts should be frozen until the end of all analyses if repeat measurements are necessary.
► The alkalis and acids must be added under the fume cupboard!
► The appropriate protective clothing must be worn, especially when working with H₂SO₄.

Sequential P fractionation is carried out in the working groups Soil Science and Agronomy (both at the Faculty of Agricultural and Environmental Sciences at the University of Rostock).

References

Bacon JR, Davidson CM (2008) Is there a future for sequential chemical extraction? Analyst, 133, 25–46, DOI: 10.1039/b711896a

Cade-Menun B and Liu CW (2013) Solution Phosphorus-31 Nuclear Magnetic Resonance Spectroscopy of soils from 2005 to 2013: A review of sample preparation and experimental parameters. Soil Sci. Soc. Am. J. 78, 19–37, DOI: 10.2136/sssaj2013.05.0187dgs

Condron LM, Newman S (2011) Revisiting the fundamentals of phosphorus fractionation of sediments and soils. J Soils Sediments 11, 830–840, DOI: 10.1007/s11368-011-0363-2

Felgentreu L, Nausch G, Bitschowsky F, Nausch M, Schulz-Bull D (2018) Colorimetric chemical differentiation and detection of phosphorus in eutrophic and high particulate waters: advantages of a new monitoring approach. Frontiers in Marine Science 5, article 212, DOI: 10.3389/fmars.2018.00212

Haygarth PM, Sharpley AN (2000) Terminology for phosphorus transfer. Environ. Qual. 29, 10–1, DOI: 10.2134/jeq2000.00472425002900010002x

Turner BL, Cade-Menun BJ, Condron LM, Newman S (2005) Extraction of soil organic phosphorus. Talanta 66: 294–306, DOI: 10.1016/j.talanta.2004.11.012

For citation: Zimmer D, Baumann K (year of download) Chapter 4.5 Methods for the determination of P fractions or binding forms in soil samples (Version 1.0) in Zimmer D, Baumann K, Berthold M, Schumann R: Handbook on the Selection of Methods for Digestion and Determination of Total Phosphorus in Environmental Samples. DOI: 10.12754/misc-2020-0001

download chapter (PDF)

Dana Zimmer, Karen Baumann

In soil, phosphorus is bound to various soil components and only a small proportion is dissolved in the soil solution. This means that in different soils P is differently mobilisable and bioavailable. The proportion of bioavailable or plant-available P therefore depends not only on how high the P concentration in the soil is, but also on how P is supplied from the soil particles into the soil solution and is thus available to plants at all (Abdu 2006, Zheng and Zhang 2012). In particular, the proportion of nutrients such as P available during a growing season is important for agricultural crops and their fertilization according to their requirements. Therefore, when determining the potentially bio- or plant-available P, not only the current P concentration in the soil solution, but also the P solubility or P replenishment from the soil solid phase must be taken into account (e.g. Abdu 2006, Zheng and Zhang 2012). Various extraction methods have been developed since the first half of the 20th century to estimate the fertilizer requirements for the most important nutrients such as P, K and Mg; later, so-called ion-sink methods were added. The methods for evaluating the bio- or plant-available P attempt to simulate the conditions around the plant roots (e.g. the release of low-molecular weight organic acids by the plant root). In general, a distinction is made between chemical extractions and ion-sink methods. All of these methods have their specific advantages and disadvantages. The chemical extracts in particular were often developed for special soils with special properties, on which they can be used very effectively; however, they can only be used to a limited extent on other soils (Abdu 2006). The pH value and the concentration of carbonate play a particularly important role here (e.g. Schüller 1969, Zheng and Zhang 2012). The advantage of ion-sink methods lies in their universal applicability, as they are largely independent of soil properties and, similar to plant roots, mimic P removal. In addition, the chemical properties of the soil are not changed and the anion exchange resins in particular can be reused without losing their extraction capacity (Abdu 2006, Zheng and Zhang 2012).
Some methods for estimating plant-available phosphorus are presented below.

4.5.2.1 DL and CAL extract

Principle and suitability of the extracts

In Germany, double lactate (DL) and calcium lactate (CAL) extracts are used in agriculture to estimate the proportion of phosphorus, potassium and magnesium that is potentially available to plants. This makes it possible to estimate the amount of fertilizer to be applied. Only P is discussed in the following. In other countries, other extracts are used (e.g. with NaHCO3) to estimate the phosphorus that is potentially available to plants and can be used as fertilizer. The CAL extract according to Schüller (1969) is used in Germany for calcareous soil samples and the DL extract according to Riehm (1948) for lime-free soil samples. The specifications of the "Landesuntersuchungs- und Forschungsanstalten (LUFen)" as to whether a DL or CAL extract is to be used in agriculture depend on the federal state. The DL extract consists of a calcium lactate solution with a pH value of 3.8. The CAL extract is a buffered solution of calcium acetate, calcium lactate and acetic acid with a pH value of 4.1 (VDLUFA 2015).
The following section was taken from Schüller (1969), van Laak et al. (2018) and VDLUFA (2012): Since apatitic phosphates do not or hardly dissolve in neutral to alkaline soils, in contrast to acidic soils, and thus have hardly any P fertilization effect, the extraction agent should not or only slightly dissolve the apatites in neutral to alkaline soils. Both DL and CAL dissolve monocalcium and dicalcium phosphates. However, due to the higher Ca activity of the CAL extract, in contrast to the DL extract, the poorly soluble Ca phosphates such as apatites are not dissolved with the CAL extract. Therefore, in contrast to the DL extract, an overestimation of P availability can be avoided with the CAL extract on neutral and alkaline soil. On acidic soils, however, the CAL extract may incompletely dissolve P in the presence of apatitic phosphates. Due to the higher buffering capacity of the CAL extract compared to the DL extract (increase in the pH value of the extraction solution for carbonate-containing soil samples), the CAL extract can be used for CaCO3 concentrations of up to 8 or up to 15 %, respectively. To make DL and CAL extracts more comparable, van Laak et al. (2018) developed pH-dependent equations to convert P concentrations between DL and CAL extracts.

Interpretation of the results

For both the DL extract and the CAL extract, so-called content classes ("Gehaltsklassen") A (very low content) to E (very high content) are de-fined for the P concentrations (as well as Mg and K) (VDLUFA 2015, Table 4.5.2-1). The guideline values for the P concentrations for the content classes can also be found in Table 4.5.2-1 (VDLUFA 2015). The fertilizer recommendations are derived from these content classes. The specific fer-tilizer recommendation depends on the type of crop and the possible and targeted yield on the respective soil. In the VDLUFA regulation, the P con-centration is traditionally determined using molybdenum blue on a photo-meter. However, according to VDLUFA (2015), it is also possible to deter-mine P using ICP-OES (in parallel with Mg and K), as carried out in the working group Soil Science. It generally provides comparable results. It is assumed that the determination using MB and ICP-OES provides approximately the same results because DL and CAL are mainly used to extract inorganic phosphate and not organically bound phosphate. However, this should be adjusted specifically for the soil to be extracted.
Note: In organic farming, content class B is often aimed for, so that the fertilizer recommendations are correspondingly lower.

Note from VDLUA (2015): Vertically rotating shaking machines have become widely established in routine operation. The optimum rotational speed depends on the diameter of the rotor. Soil and extraction solution must be mixed reliably, but centrifugal effects caused by excessive rotational speeds must be avoided.

Sample preparation for DL and CAL extract:

► Drying soil samples (see chapters 2.4 and 3.1)
► Sieve soil samples <2 mm and use this <2 mm fraction (fine soil)
► Determine total element concentrations in a subsample (e.g. using aqua regia extract, see chapter 4.1.2)

Protocol for the DL extraction

Material and chemicals required for the DL extract

DL storage solution

► Place 120 g of calcium lactate (C₆H₁₀CaO₆ 5H₂O) in a 1-liter graduated flask and pour approx. 0.8 L of boiling ultrapure water (UW) over it and stir until everything is dissolved.
► Add 40 ml of 10 M HCl to the still warm solution and fill up to 1 L with UW after cooling down.
DL solution for use (prepare fresh daily)
► Fill 500 ml of the DL stock solution with UW to 10 L or fill 50 ml to 1 liter
► Important: pH control 3.6

Procedure DL extraction:

► Prepare at least 3 replicates per soil sample and at least 1 blank value per 10 extraction samples.
► Normally, 4 g of soil is weighed into 250 ml shake flasks and mixed with 200 ml DL working solution; however, 0.6 g of soil can also be weighed into 50 ml centrifuge tubes and mixed with 30 ml working solution; in this case, pay more attention to the homogeneity of the sample
► Shake overhead for 90 minutes at approx. 20 rotations per minute
► Filter DL extract through P-free filters in Erlmeyer flasks
► If no P-free filters are used, discard the first 10 to 30 ml of the filtrate (rinse the filters), replace the Erlmeyer flasks that have collected the filtrate or place the funnels with the filters on new Erlmeyer flasks after the first ml of filtrate and filter the rest
► Transfer the aliquot to ICP tubes and store in the refrigerator or freeze until measurement (if the extracts must be stored for >1 day)
► Determination of the elements K (766.490 nm), Mg (285.213 nm) and P (213.617 and/or 214.914 nm) by ICP-OES and/or P photometrically with molybdenum blue, see below

Protocol for the CAL extraction

Material and chemicals required for the CAL extract

CAL stock solution

► Dissolve 77 g Ca-lactate pentahydrate (C₆H₁₀CaO₆ * 5H₂O) in approx. 300 ml boiling RW and 39.5 g Ca-acetate (Ca(CH₃COO)₂ * H₂O) (or 35.5 g Ca(CH₃COO)₂ anhydrous) in approx. 300 ml RW (heat if necessary)
► Combine both solutions in a 1-liter volumetric flask and add 89.5 ml 100 % acetic acid p.a.
► Fill the mixture with UW in the graduated flask to 1 L.

CAL usage solution:

► Fill up 1 L of the stock solution with UW in the graduated flask to 5 L
(= 0.1 M Ca-lactate; 0.1 M Ca-acetate; 0.3 M acetic acid; pH 4.1)

Procedure CAL extraction:

► Prepare at least 3 replicates per sample and at least 1 blank value per 10 samples.
► Normally, 5 g of soil is weighed into 300 ml shake flasks and mixed with 100 ml CAL working solution; however, 1.5 g of soil can also be weighed into 50 ml centrifuge tubes and mixed with 30 ml working solution; in doing so, pay more attention to the homogeneity of the sample
► Shake overhead for 90 minutes at approx. 20 rotations per minute
► Filter CAL extract through P-free filters in Erlmeyer flasks
► If no P-free filters are used, discard the first 10 to 30 ml of the filtrate (rinse the filters), replace the Erlmeyer flasks that have collected the filtrate or place the funnels with the filters on new Erlmeyer flasks after the first ml of filtrate and filter the rest
► Transfer the aliquot to ICP tubes or similar and store in the refrigerator or freeze until measurement (if storage is required for >1 day)
► Determination of the elements K (766.490 nm), Mg (285.213 nm) and P (213.617 and/or 214.914 nm) by ICP-OES and/or P photometrically with molybdenum blue, see below

P determination in DL and CAL extract with molybdenum blue (according to Murphy and Riley, 1962)

Photometric P determination is only possible on colorless, non-turbid extracts!

Reagents:

For 1 liter of detection reagent A

► Pour approx. 500 ml of UW into a 1-liter flask
► Dissolve 6 g of ammonium heptamolybdate (NH₄)₆Mo₇O₂₄ x 4H₂O in it
► Add 74 ml conc. H₂SO₄, allow to cool slightly
► Dissolve 0.149 g potassium antimony oxide tartrate K(SbO)C₄H₄O₆ x 0,5 H₂O (KAT) in it
► As the KAT dissolves poorly, add the KAT when the solution is still warm and, if necessary, place in an ultrasonic bath and swirl repeatedly until completely dissolved.
► Add more UW to slightly below the mark, allow to cool to room temperature and fill to 1 liter with UW.

For 2 liters of detection reagent A:

12 g ammonium heptamolybdate, 148 ml conc. H₂SO₄ and 0.298 g KAT are required.

For 250 ml detection reagent B:

► Add 1.32 g ascorbic acid to a 200 ml volumetric flask
► Add reagent A and swirl until the ascorbic acid has dissolved
► Fill up to 200 ml with reagent A

For 500 ml detection reagent B:

► Add 2.64 g ascorbic acid to a 500 ml volumetric flask
► Add reagent A and swirl until the ascorbic acid has dissolved
► Fill up to 500 ml with reagent A

Note:

Reagent B must be prepared fresh daily, as it cannot be kept for longer than 24 hours. For a 25 ml volumetric flask, 4 ml of detection reagent B is required per sample/standard. Depending on the amount of detection reagent B required, the weight of ascorbic acid and the volume must be adjusted.

Standard series approach for photometric P determination:

Target concentration for the P stock solution: prepare 10 mg P per liter in a 100 ml flask

► To prepare the calibration standards (Table 4.5.2-2), fill the required volume of stock solution into a 25 ml or 50 ml flask and fill up to half full with UW. (Different pipette sizes are required!)
► Add 4 ml (in a 25 ml flask) or 8 ml (50 ml flask) of the detection reagent
► Fill up to 25 or 50 ml with UW
► Wait 30 min and measure absorbances at 882 nm
► Set the standard S0 in the photometer to "zero", observe the manufacturer's instructions for the photometer
► Calculate calibration line

Note:
If the absorbances of most samples are below the standard S2, additional standards must be inserted between S0 and S2 (Tab. 4.5.2-3). Appropriate μl pipettes are required for this.

Photometric determination of the P concentration in the samples with molybdenum blue

► Pipette 1 to max. 10 ml (or 0.5 to max. 5 ml) of the filtrate into each 25 ml (or 50 ml) volumetric flask and then fill up to max. half full with UW
► Add 4 ml (or 8 ml) of detection reagent and make up to 25 (or 50 ml) with UW
► After a reaction time of 30 min, measure at 882 nm, leaving standard S0 at "zero". Observe the photometer manufacturer's specifications
► The blank values are measured in the same way as the samples and the mean value of the absorbance values of the blank values is subtracted from those of the samples
► Calculate the P concentration in the samples using a calibration line

Note:
The sample quantity required for the molybdate reagent depends on the DL or CAL extractable P concentration in the soil. For soil samples from Ap horizons, try 2.5 ml of sample filled up to 25 ml!
The DL and CAL extracts can be carried out in the Soil Science and Agronomy working groups (both Faculty of Agricultural and Environmental Sciences, University of Rostock).

4.5.2.2 NaHCO₃ extract

Principle and suitability of the extract

In general, there is a balance between the following species in water:

CO₂ (g) ⟷ CO₂ (aq) + H₂O ⟷ H₂CO₃ (aq) ⟷ H⁺(aq) + HCO₃^- (aq) ⟷ H⁺(aq)+ CO₃^2-(aq)

If NaHCO₃ is added to the soil solution, it decomposes in the water to Na⁺ and HCO₃^-. In alkaline soil, HCO₃^- reacts with the Ca²⁺ in the soil solution and precipitates as CaCO₃ (Olsen et al. 1954, Soinne 2009). The equilibrium is shifted to the right-hand side of the equation. This allows the P previously bound to Ca to be extracted. In acidic soils, in which P is mainly bound to Al- and Fe-(hydroxides), HCO₃^- reacts with the H⁺ to form H₂CO₃ and further and further to the left to form CO₂, which can also bubble out (CO₂(g)) and is thus removed from the equilibrium. All ions from the above equilibrium reaction compete with P for the binding sites on the pedogenic oxides (Hartikainen, 1981), so that P also dissolve from these. Due to the CO₂ bubbling out, no constant equilibrium can be established between the species. Since no equilibrium is established during extraction and the extracted P concentrations can therefore fluctuate, exact compliance with shaking time, temperature, etc. must be ensured for comparability (Miller et al. 2002). The extract according to Olsen et al. (1954) is used in some states in the USA, for example, to estimate the plant-available P content (Wuenscher et al. 2015). It is preferably used for calcareous soils, but can also be used for non-calcareous soils (Wuenscher et al. 2015).

Protocol

Sample preparation:

► Drying soil samples (see chapters 2.4 and 3.1)
► Sieve soil samples <2 mm and use this <2 mm fraction (fine soil)
► Determine total element concentrations in a subsample (e.g. using aqua regia extract, see chapter 4.1.2)

Preparation of chemicals

1 liter 0,5 M NaHCO₃ (pH 8,5)

► Add 42 g NaHCO3 to 1 L flask and fill up to approx. 800 ml with UW
► Adjust the pH value with 1 M NaOH (approx. 10 to 20 ml required)
► Fill up to 1 liter with UW

1 liter 1 M NaOH for pH adjustment

► Fill 1-liter flask with approx. 700 ml RW, add 40 g NaOH pellets, fill incompletely with UW
► After the pellets have completely dissolved and cooled down to room temperature, fill up to 1 liter with UW

Procedure:

► Weigh 1.00 g of air-dry soil into a 50 ml centrifuge tube
► Add 20 ml of 0.5 M NaHCO₃ solution
► Shake for 30 min
► Centrifuge for 10 min at 3500 x g
► Filter into Erlenmeyer flasks
► Possibly fill 20 ml for photometric molybdenum blue measurement (see chapter on sequential P fractionation F2, no destruction of HCO^- with HCl)
► Acidify 10 ml sample for ICP measurement with HCl (destruction HCO^-):
► Add 10 ml sample + 1 ml conc. HCl to Erlenmeyer flask, leave to stand overnight (outgassing CO₂)
► Add 9 ml UW the next day (dilution factor: 2)
► Filling for P determination on ICP-OES, wavelength: 213.617 nm and/or 214.914 nm

This extraction can be carried out in the Soil Science and Agronomy working groups (both Faculty of Agricultural and Environmental Sciences, University of Rostock).

4.5.2.3 Water extracts

Principle and suitability of the extracts

Water extracts simulate the P pool in the soil that can potentially be mobi-lized by rain events. This P pool can also be regarded as directly available to plants. However, water extracts largely disregard the subsequent supply of the labilely bound P, which goes into solution as soon as the dissolved P has been absorbed by plants. A distinction can be made between cold and hot water extracts. In addition to the mineral elements of interest, the hot water extract can also be used to extract the easily mobilizable or convertible organic soil substance (OBS), i.e. also C and N (Leinweber et al. 1995).

Sample preparation for the water extracts:

► Drying soil samples (see chapters 2.4 and 3.1) and
► Sieve soil samples <2 mm and use this <2 mm fraction (fine soil)
► If naturally moist soil (e.g. rhizosphere soil) is to be used, at least <5 mm should be sieved and stones, roots etc. removed
► Determine total element concentrations in a subsample (e.g. using aqua regia extract, see chapter 4.1.2)

Protocol cold water extraction (soil + water ratio: 1 + 10)

Procedure:

► Weigh 2.00 g of air-dry fine soil (<2 mm) into 50 ml centrifuge tubes
► Add 20 ml ultrapure water
► Shake over head for 60 min
► Centrifuge at 3000 x g for 10 min
► Filter the supernatant through a P-free filter into a 100 ml Erlenmeyer flask
► Acidify with 20 μl conc. HCl to pH 2-3 to prevent element precipita-tion and possible microbial changes ("slime")
► When acidified, the samples should be able to be stored in the refri-gerator for several days. However, as the elements have always been determined promptly up to now, there is no experience of storage life. For longer storage, the extracts should be frozen.
► Element determination on ICP-OES (select e.g. Al, Ca, Cd, Cu, Fe, K, Mg, Mn, Ni, P, Pb, Zn)

Protocol hot water extract for the simultaneous determination of the easily convertible OBS

Procedure:

► Weigh (10) or 25 to 30 g of air-dry fine soil (<2 mm sieved) into a 250 ml glass flask (the heating block offers 12 places, prepare at least one of them as a blank)
► Add a few boiling stones and add 50 to 60 ml of water
► Note the exact weight and volume of water
► Place the glass bulb in the heating block
► Place the reflux condenser on top, heat the suspension in the heating block and boil vigorously for 60 minutes so that soil and water are in-tensively swirled through
► Cool the sample quickly (place containers in cold water)
► Add 3 drops of 2.5 M CaCl₂ solution as a filtration aid and filter through P-free pleated filters into 100 ml Erlenmeyer flasks
► Determination of C and N concentrations in liquid extract as for C_mic
► Determination of P (and possibly other elements) on ICP-OES
► If a solid is required for the determination of organic C-compounds using mass spectrometry (MS), the extract should be freeze-dried.

Both water extracts can be carried out in the Soil Science working group (Faculty of Agricultural and Environmental Sciences, University of Rostock).

4.5.2.4 Extraction with low-molecular weight organic acids

Principle and suitability of the extracts

In the soil, only small amounts of P are present as dissolved phosphate in the soil solution; the largest amounts are bound more or less strongly to the soil particles. Extraction with low molecular weight organic acids (LMWOA) simulates the exudation of root exudates from plants and thus the dissolution of potentially available nutrients in the rhizosphere. Feng et al. (2005) describe the extraction with LMWOA for plant-available heavy metals, while Wei et al. (2010) used citric acid solution of similar molarity for the extraction of plant-available P.
If small-scale hot spots in the soil, such as rhizosphere soil, are examined with these extracts, the use of naturally moist samples is recommended in order to prevent microbial conversion and oxidative changes during drying (see chapter 2.4). This also prevents drying-induced lysis of microbial cells in the microorganism-rich hotspots such as the rhizosphere, which would cause an unwanted release of microbial P (Sparling et al. 1985, Srivastava 1998, Turner and Haygarth 2003). To prevent microbial activity and potential biodegradation of the organic acids, a biocide (e.g. chloroform) can be added (Wei et al. 2010). In addition, the prepared solution must be consumed on the same day to prevent degradation of the organic acids.
The LMWOA extracts have not yet been carried out in the Soil Science laboratories (Faculty of Agricultural and Environmental Sciences, University of Rostock), but could potentially be tested there.

Protocols

LMWOA extract modified according to Feng et al. (2005)

LMWOA solution approach according to Feng et al. 2005

► Total concentration of organic acids: 10 mM consisting of: acetic, lactic, citric, malic and formic acid in the ratio: 4:2:1:1:1
► 10 mM = 9/9

Preparation for 1-liter LMWO solution:

► acetic acid   4,44 mM   266,62 mg
► lactic acid   2,22 mM   199,98 mg
► citric acid   1,11 mM   213,25 mg
► malic acid   1,11 mM   148,84 mg
► formic acid   1,11 mM   51,09 mg

Procedure for extraction

► Weigh 2.00 g of moist rhizosphere soil or lutro soil <2 mm into 50 ml centrifuge tubes
► Add 20 ml LMWOA solution plus 2 drops of chloroform (Wei et al. 2010) as a biocide (work under the fume cupboard)
► Shake overhead for 16 hours at approx. 20 rotations per minute
► Centrifuge at 3000 x g for 30 min
► Pipette 5 ml of the supernatant into 10 ml graduated test tubes, fill them up with 2 % HNO₃

Citric acid, oxalic acid or maleic acid extract according to Wei et al. (2010)

According to Wei et al. 2010, extraction with citric acid was most effective for tropical and subtropical soils.

Approach of the solutions

1 mM Citric acid solution

► molar mass: 192,124 g mol^-1
► Weigh-in weight for 1 liter of 1mM citric acid: 192 mg

1 mM Oxalic acid solution

► molar mass: 90,04 g·mol⁻¹ (anhydrous) 126,07 g·mol⁻¹ (dihydrate)
► Weigh-in weight for 1 liter 1 mM oxalic acid: 90 mg (anhydrous) or 126 mg (dihydrate)
1 mM Maleic acid solution
► molar mass: 116,072 g mol^-1
► Weigh-in weight for 1 liter of 1 mM maleic acid: 116 mg

Procedure for extraction with LMWOAs

► Weigh 4.00 g lutro soil <2mm into 50 ml centrifuge tubes
► Add 40 ml citric acid or oxalic acid or maleic acid solution plus 2 drops of chloroform as a biocide (work under the fume cupboard)
► Shake over head for 24 hours
► Centrifuge at 3000 x g for 30 min
► Filter through P-free filters

Determination of element concentrations

► Measurement of the elements of interest using ICP-OES
► If necessary, measure Pi using molybdenum blue

4.5.2.5 Ion-sink methods - anion exchange resin

Ion sink methods include extracts with anion exchange resin and Fe-coated strips or filter papers (Myers et al. 2005). These methods simulate the conditions in the rhizosphere, i. e. the plant roots (or in this case the exchangers) absorb the P present in the soil solution. Following the solubility equilibrium, P is then desorbed from the soil particles and the dissolved P is taken up again by the roots or the exchanger. These methods have the advantage that, in contrast to chemical extraction agents, they can be used independently of the soil properties, do not chemically alter the soil and, in the case of anion exchange resin, can be reused several times (50 to 500 times) (Saggar et al. 1990, Schoenau and Huang 1991).

Notes on the extracts with anion exchange resin

Extractions with anion exchange resins can be carried out either as so-called (1) batch extracts (with resin membranes (anion exchange membranes = AEM) or resin beads), with the (2) miscible displacement technique (Abdu 2006) or (3) with the placement of the exchange resins directly into the undisturbed soil (e.g. Quian and Schoenau 2002). For batch extracts, soil and ultrapure water are mixed in a wide ratio and the exchanger resin is added for shaking. The exchanger resins can be pretreated with either HCl or bicarbonate (Abdu 2006, Sibbesen 1978, Quian and Schoenau 2002). Phosphates are then exchanged for anions of the membrane. Sibbesen (1978) proved that HCO₃^- resins are more suitable than Cl^- resins, as the plants accumulate HCO₃^- in the rhizo-sphere, which leads to an increase in the pH value in the rhizosphere in acidic to neutral soils and to a reduction in pH in alkaline soils. If Cl^- resin is used, the Cl^- can accumulate in the solution and inhibit the exchange reaction (Myers et al. 2005). This method can therefore not be used in saline soils.
Based on sequential P fractionation, the procedure for P extraction with anion exchange resin as a membrane in the HCO₃^- form is described below. As there is no other experience with these extraction methods in the working groups, reference is made to the literature.

Protocol for extraction with strips of anion exchange resin

Preparation of the resin strips (see also Myers et al. 2005, Saggar et al. 1990)

► Anion exchange membrane BDH #55164 2S (see also sequential P fractionation), cut into 12 strips of 6 x 2 cm each
► Prepare 2 L 0.5 M NaHCO3 and fill into two 1-liter beakers
► Place the resin strips in the first beaker for 1 h, transfer to the second beaker with tweezers and incubate again for 1 h
► Wash resin strips 3 times by placing them in beakers with UW (move with tweezers, place tweezers in UW before use))
► Storage in UW in the refrigerator (24 hours before use, after preparation with HCO₃^-)

Sample preparation:

► Drying soil samples (see chapters 2.4 and 3.1)
► Sieve soil samples <2 mm and use this <2 mm fraction (fine soil)
► Determine total element concentrations in a subsample (e.g. using aqua regia extract, see chapter 4.1.2)

Procedure:

► Add 0.5 g soil, 30 ml ultrapure water (UW) and a strip of anion ex-change resin, shake upside down for 18 hours
► Remove resin strips with tweezers, rinse adhering soil particles with UW (spray bottle) back into the centrifuge tube
► Wash resin strips with max. 45 ml 1 M HCl via funnel with P-free filter in 50 ml volumetric flask
► Place the resin strips back into beakers with UW, later carry out another pre-treatment for bicarbonate form before storing in the refrigerator again
► Fill the graduated flask with 1 M HCl to 50 ml
► Fill aliquots (10 to 20 ml) for P determination, determine resin Pt on the ICP, if necessary, Pi using MB (see chapter 4.5.1 Sequential P fractionation)

Prepare 0.5 M NaHCO₃

Molar mass NaHCO₃: 84.007g mol^-1
Add 42 g NaHCO₃ to a 1-liter volumetric flask and fill up with UW

As this is a fraction of sequential P fractionation, extraction with anion exchange resin can be carried out in the Soil Science and Agronomy working groups at the Faculty of Agricultural and Environmental Sciences at the University of Rostock.

Myers et al. (1999) and (2005) deviate from the above approach as follows:
After soil extraction and rinsing of the soil particles with ultrapure water:
► Place the resin membranes in a 125 ml wide-neck bottle and add 50 ml of 0.5 M HCl to remove P from the resin strips
► The sealed bottles are shaken on a horizontal shaker (horizontal and end-to-end on a reciprocating shaker) at 125 to 130 rpm for 90 min
► Remove the resin strips and determine the P concentration in the HCl solution

Schoenau and Huang (1991) deviate as follows:
Once the soil has been washed off the resin membranes, they are placed in 50 ml centrifuge tubes and 30 ml 0.5 M HCl is added.
The resin membranes are shaken for 1 h on the horizontal shaker.

Pretreatment Anion exchange resin, if the Cl- form is to be used, according to Myers et al. (2005):
► Store resin membranes in UW for 24 h before use
► Prepare 250 ml of 1.0 M KCl, pour into a beaker and soak the resin membranes for 30 min
► Repeat the process in a second beaker,
► then rinse the resin strips with UW before storing them in UW

References

Abdu, N. 2006. Soil-phosphorus extraction methodologies: A review. Afr. J. Agric. Res. 1, 159-161

Bacon JR, Davidson CM (2008) Is there a future for sequential chemical extraction? Analyst, 133, 25–46, DOI: 10.1039/b711896a

Cade-Menun B and Liu CW (2013) Solution Phosphorus-31 Nuclear Magnetic Resonance Spectroscopy of soils from 2005 to 2013: A review of sample preparation and experimental parameters. Soil Sci. Soc. Am. J. 78, 19–37, DOI: 10.2136/sssaj2013.05.0187dgs

Condron LM, Newman S (2011) Revisiting the fundamentals of phosphorus fractionation of sediments and soils. J Soils Sediments 11, 830–840, DOI: 10.1007/s11368-011-0363-2

Felgentreu L, Nausch G, Bitschowsky F, Nausch M, Schulz-Bull D (2018) Colorimetric chemical differentiation and detection of phosphorus in eutrophic and high particulate waters: advantages of a new monitoring approach. Frontiers in Marine Science 5, article 212, DOI: 10.3389/fmars.2018.00212

Feng, M-H, Shan, X-Q, Zhang, S, Wen, B (2005) A comparison of the rhizosphere-based method with DTPA, EDTA, CaCl2, and NaNO3 extraction methods for prediction of bioavailability of metals in soil to barley. Environmental Pollution 137, 231-240

Hartikainen, H. 1981. Effect of decreasing acidity on the extractability of inorganic soil phosphorus. Journal of the Scientific Agricultural Society of Finland 53, 16–26.

Haygarth PM, Sharpley AN (2000) Terminology for phosphorus transfer. Environ. Qual. 29, 10–1, DOI: 10.2134/jeq2000.00472425002900010002x

Leinweber, P., Schulten, H.-R., Körschens, M. (1995) Hot water extracted organic matter: chemical composition and temporal variations in a long-term field experiment. Biol Fertil Soils 20, 17-23

Miller, R.O. and Horneck, D. (2002) Bicarbonate (Olsen) Phosphorus troubleshooting. http://www.naptprogram.org/files/napt/publications/method-papers/2002-bicarbonate-phosphorus-p-troublehooting.pdf

Murphy J. and Riley, J. P. (1962) A modified single solution method for the determination of phosphate in natural waters. Anal. Chim. Acta 27: 31-36

Myers RG, Sharpley AN, Thien SJ, Pierzynski GM (2005). Ion-Sink phosphorus extraction methods applied on 24 soils from the continental USA. Soil Sci. Soc. Am. J. 69, 511- 521

Myers, R.G., Thien, S.J., Pierzynski, G.M. (1999) Using an ion sink to extract microbial phosphorus from soil. Soil Sci. Soc. Am J. 63:1229–1237, DOI: 10.2136/sssaj1999.6351229x

Ogwada RA, Sparks DL (1986). Kinetics of ion exchange on clay minerals and soils: 1. Evaluation of methods. Soil Sci. Soc. Am. J. 50, 1158-1162.

Olsen, SR, Cole, CV, Watanave, FS, Dean, LA (1954) Estimation of Available Phosphorus in Soils by Extraction with Sodium Bicarbonate; USDA Circ. 939, USDA, Government Printing Office: Washington, DC

Qian, P. and Schoenau, J. J. 2002. Practical applications of ion exchange resins in agricultural and environmental soil research. Can. J. Soil Sci. 82, 9–21.

Rennert, T. (2019) Wet-chemical extractions to characterise pedogenic Al and Fe species – a critical review. Soil Res 57, 1–16, DOI: 10.1071/SR18299

Riehm, H. (1948) Arbeitsvorschrift zur Bestimmung der Phosphorsäure und des Kaliums nach Lactatverfahren. Z. Pflanzenernährung. Düng. Bodenk. 40, 152-156, DOI: 10.1002/jpln.19480400206

Saggar, S, Hedley, M.J., White, R.E. (1990) A simplified resin membrane technique for extracting phosphorus from soils. Fertilizer Research 24: 173-180, DOI: 10.1007/BF01073586

Schoenau, J.J., W.Z. Huang. (1991) Anion-exchange membrane, water, and sodium bicarbonate extractions as soil tests for phosphorus. Commun. Soil Sci. Plant Anal. 22,465–492, DOI: 10.1080/00103629109368432

Schüller, H. (1969) Die CAL-Methode, eine neue Methode zur Bestimmung des pflanzenverfügbaren Phosphates in Boden. Zeitschrift für Pflanzenernährung und Bodenkunde 123, 48-63, DOI: 10.1002/jpln.19691230106

Soinne, H (2009) Extraction methods in soil phosphorus characterisation - Limitations and applications. Diss. Department of Applied Chemistry and Microbiology, University of Helsinki

Sparling G.P., Whale K.N., Ramsay A.J. (1985): Quantifying the contribution from the soil microbial biomass to the extractable P levels of fresh and air-dried soils. Austral J Soil Res 23, 613–621, DOI: 10.1071/SR9850613

Srivastava, S.C. (1998) Microbial contribution to extractable N and P after air-drying of dry tropical soils. Biol Fertil Soils 26, 31–34

Turner BL, Cade-Menun BJ, Condron LM, Newman S (2005) Extraction of soil organic phosphorus. Talanta 66: 294–306, DOI: 10.1016/j.talanta.2004.11.012

Turner B.L., Haygarth P.M. (2003) Changes in bicarbonate-extractable inorganic and organic phosphorus by drying pasture soils. Soil Science Society of America Journal, 67, 344–350

van Laak, M., Klingenberg, U., Peiter, E., Reitz, T., Zimmer, D., Buczko, U. (2018) The equivalence of the Calcium-Acetate-Lactate and Double-Lactate extraction methods to assess soil phosphorus fertility. J Plant Nutr Soil Sci 181 (5), 795-801, DOI: 10.1002/jpln.201700366

VDLUFA (2015) A 6.2.1.1 Bestimmung von Phosphor und Kalium im Calcium–Acetat–Lactat–Auszug.

Wei, L, Chen, C, Xu, Z (2010) Citric acid enhances the mobilization of organic phosphorus in subtropical and tropical forest soils. Biol Fertil Soils 46, 765–769

Zheng, ZM, Zhang, TQ (2011) Soil phosphorus tests and transformation analysis to quantify plant availability: A review. In Whalen, JK (Ed.) (2012) Soil fertility improvement and integrated nutrient management – A global perspective. InTech Janeza Trdine 9, 51000 Rijeka, Croatia, ISBN 978-953-307-945-5, pp. 19-36

For citation: Zimmer D, Baumann K (year of download) Chapter 4.5 Methods for the determination of P fractions or binding forms in soil samples (Version 1.0) in Zimmer D, Baumann K, Berthold M, Schumann R: Handbook on the Selection of Methods for Digestion and Determination of Total Phosphorus in Environmental Samples. DOI: 10.12754/misc-2020-0001

download chapter (PDF)

Dana Zimmer, Karen Baumann

In soil, phosphorus is bound to various soil components and only a small proportion is dissolved in the soil solution. This means that P can be mobilized and bioavailable differently in different soils. Especially at acidic pH values, P in the soil is strongly bound to pedogenic Fe and Al (hydr)oxides (abbreviated as oxides in the following; partly Mn oxides). Since well and poorly crystalline Fe oxides in particular differ in their ability to bind P, extracts are used to estimate the Fe crystallinity in order to estimate the binding of P to these Fe oxides. As with all extracts, it should be noted that they are operationally defined extracts and extract not only the target substances, but in some cases also other binding forms or the extraction of the target substance is incomplete (Rennert 2019 and references therein). If these extracts are carried out in the successive horizons of a soil profile, they can, to a limited extent, provide indications of transformation processes in the soil. This is especially true for soils with Fe displacement processes such as podzols.

In general, P is more strongly adsorbed and less desorbed on poorly crystalline, especially Fe oxides (e.g. ferrihydrite) than on better crystalline Fe and Al oxides (e.g. gibbsite) (Gypser et al. 2018). The crystallinity of the pedogenic oxides can be affected by the sampling and drying of the samples (see also chapter 2.4) and can thus also have an effect on the results of the extractions. A drying temperature of 60 °C causes poorly crystalline Fe oxides to age, i.e. their crystallinity increases (Landa and Gast 1973). Ferrihydrite is transformed into more crystalline goethite and hematite at temperatures starting at 50 °C (Das et al. 2011), which can consequently influence P binding and P extractability. Soil samples should therefore be dried at lower temperatures, especially for these extractions.

It is assumed that, in the dark, the oxalate extract (Schwertmann 1964, Landa and Gast 1973, Miller et al. 1986) extracts mainly poorly crystalline Fe (and P bound to it) and the dithionite extract (according to Mehra and Jackson 1960; DCB) also extracts better crystalline Fe oxides. The dithionite extract according to Mehra and Jackson (1960) is used in various forms, but has been criticized for poor hematite extraction. Deb (1950) found that the oxalate extract in sunlight extracted Fe concentrations comparable to those of the DCB extract. Reyes and Torrent (1997) suggest an extract of 0.05 M ascorbic acid with 0.2 M citrate (pH 6, 16 h) to extract Fe from poorly crystalline oxides.

Based on the historical development, there are different names for poorly crystalline pedogenic oxides, e.g. active or reactive oxides, amorphous oxides, non-crystalline oxides (Rennert 2019 and references therein). Since there is no order in amorphous or non-crystalline material, even in the immediate atomic environment, Rennert (2019) therefore proposes to use the term "short-range ordered" as equivalent to "poorly crystalline" pedogenic oxides; however, this is not equivalent to "non-crystalline" or "amorphous". It should be noted that there is no sharp distinction between good and bad crystalline oxides in the soil, but rather a continuum. In this chapter, the terms bad(er) and good (better) crystalline are used to differentiate between them.

Note: If the extracts are carried out, the total element concentrations (Al, Fe, Mn and P) in the soil sample must also be determined. When calculating the P sorption capacity and P saturation, please note that the concentrations must be calculated in mol instead of g! When interpreting the results, it should again be noted that the extracts not only extract the target compounds, but also extract them incompletely or extract non-target compounds (Rennert 2019) and thus the target phases can be under- or overestimated.

The following parameters can be calculated from the oxalate extract, the dithionite extract and the total element concentration:

P sorption capacity = PSC in mmol kg^-1

Degree of P saturation of the pedogenic oxides in % (degree of P saturation = DPS)

Note: In some publications, Mn_ox is also added to Al_ox and Fe_ox! This should be taken into consideration, particularly in the case of elevated Mn concentrations in the soil.

The following fractions can be calculated to estimate the different crystalline proportions and thus the binding capacity of the Fe oxides:

Proportion of poorly crystalline Fe oxides in total Fe =

Proportion of poorly crystalline Fe oxides in pedogenic Fe oxides =

Proportion of pedogenic Fe oxides in total Fe =

Proportion of well crystalline Fe oxides in total Fe =

Abbreviations:

ox = Element concentration in the oxalate extract
d = Element concentration in the dithionite extract
t = Total element concentration (total)

Similarly, the P concentrations in the corresponding fractions can be calculated from the P concentrations of the extracts. The extracts can be carried out in the working group Soil Science (Faculty of Agricultural and Environmental Sciences, University of Rostock).

4.5.3.1 Oxalate extract for extracting the poorly crystalline Fe and Al oxides and the P

Principle and suitability of the extract

It is assumed that the oxalate extract (Schwertmann 1964, Landa and Gast 1973, Schwertmann 1973, Miller et al. 1986) mainly extracts poorly crystalline Fe (and P bound to it) in the dark. In the oxalate extract, oxalic acid and ammonium oxalate form a buffer at pH 3. At pH values <3.5, oxide surfaces are initially protonated before oxalate is subsequently adsorbed and complexed Al³⁺ and Fe³⁺ ions are released (Rennert 2019). Moreover, organically bound Fe, Al and Mn are also extracted. Depending on the crystallinity, Fe in the extract can thus come from dissolved Fe, organically bound Fe compounds and the poorly crystalline ferrihydrite, but also from less crystalline goethite and hematite as well as lepidocrocite; Al in the extract can come from e.g. hydroxy interlayers, less crystalline Al oxides and aluminosilicates as well as from allophanes (Rennert 2019).
According to Parfitt and Childs (1988), the oxalate extract is one of the best methods for estimating the proportion of poorly crystalline Fe oxides, especially ferryhydrite, even if Fe is also extracted from some Al layered silicates (allophanes, imogolites). In sunlight, the oxalate extract extracts Fe concentrations comparable to those of the DCB extract (Deb 1950). Extraction times between 1 and 5 hours were tested for the oxalate extract according to Schwertmann; the setting of 2 hours shaking time is arbitrary (Schwertmann 1964). If the concentrations of Fe are still considerably too high for measurement on the ICP despite 12-fold dilution (e.g. in the presence of turf iron ore, see DIN 19684-6), the sample weight can be reduced (e.g. 0.5 g) or the extracts must be diluted more for the measurement.

Interpretation of the results

The oxalate extract should not be considered on its own, but at least in relation to the total element concentration. P can also be calculated in the same way as Fe (e.g. proportion of oxalate-extracted P in dithionite or total P, see equations above for Fe). The results for P can be used to calculate the P sorption capacity (PSC) and the proportion of P saturation of the pedogenic oxides (DPS) (see equations above).

Comparative values for orientation: DPS values determined in humus-rich topsoil in soil profiles in the state of Mecklenburg-Western Pomerania for the DBG conference 2013 (Ahl & Leinweber 2013).

► in the forest: DPS by 10 %
► "Normal" arable topsoil DPS 20...30 %
► DPS values >30 % were found with heavy manure fertilization or massive input of other P-rich biomass; in the case of soils, this could also be an indication of anthropogenic influence in the past
► DPS values of >90 % or even >100 % were found on former gravel mining sites, which were massively fertilized with liquid manure for recultivation in GDR times and then used as cattle pasture after reseeding, as well as in the fEx horizon of a (former) hortisol in the monastery garden in Rostock! It is assumed here that parts of the P bound to the OBS were also extracted with oxalate.

Protocol for oxalate extraction

Sample preparation:

► Drying soil samples (see chapters 2.4, 3.1)
► Sieve soil samples <2 mm and use this <2 mm fraction (fine soil)
► Determine total element concentrations in a subsample (e.g. using aqua regia extract, see chapter 4.1.2)

Reagents:

Either both solutions can be prepared separately and then combined or the oxalate buffer can be prepared in one step. To save chemicals (DIN 19684-6), the second option is preferable.

(1) Weighing for 1 liter oxalate buffer (according to DIN) (separate preparation)

► Solution 1: Weigh 28.42 g Di-ammonium oxalate monohydrate ((NH₄)₂C₂O₄ * H₂O)) into a 1-liter flask and fill up with ultrapure water
► Solution 2: Weigh 25.21 g oxalic acid dihydrate C₂H₂O₄ * H₂O into a 1-liter flask and fill up with ultrapure water
► Mix 700 ml solution 1 (corresponds to 19.89 g ammonium oxalate) with 535 ml solution 2 (corresponds to 13.48 g oxalic acid) and check the pH value of the extraction solution, it should be between pH 3.0 (Schwertmann 1964) and 3.2 (DIN ISO 19684-6), adjust with ammonia solution if necessary

Note: The pH value is usually around 3.0 and around 100 ml of ammonia solution is required to adjust the pH to 3.2. The pH 3.0 should therefore be used.

(2) Preparation of 1 liter of oxalate buffer extraction solution in one batch

► Weigh 16.11 g di-ammonium oxalate monohydrate (NH₄)₂C₂O₄ * H₂O and 10.92 g oxalic acid dihydrate C₂H₂O₄ * H₂O in a 1-liter flask
► Fill the flask up to 700 to 800 ml with ultrapure water and stir on the magnetic stirrer until the salts have dissolved
► Fill up to approx. 900 ml with ultrapure water, check pH value see above and adjust to 3.0 if necessary (see note above), fill up to 1 liter with ultrapure water

Procedure:

► Prepare at least 3 replicates per soil sample and at least 1 blank per 10 extraction samples.
► Weigh 1.5 g fine soil (applies to 0.1 to 5 g Fe kg^-1 soil) into 50 ml centrifuge tubes or 5.00 g fine soil into 200 to 250 ml plastic bottles (see DIN)
► Add 30 ml or 100 ml oxalate buffer, respectively
► Shake overhead in the dark for 2 hours
► Centrifuge at 1500 x g for 10 min
► Filter the extract through dry P-free filters
► Dilution: at least factor 12 (1 part sample plus 11 parts ultrapure water, dilution must be tested, at higher Fe concentrations Fe can flocculate)
► Transfer aliquot to ICP tubes and store in the refrigerator or freezer until measurement (if stored for >1 day)
►
Determination of elements using ICP-OES (Al_ox 396.153 nm, Fe_ox 238.204 nm, Mn_ox 257.610 nm and P_ox 214.914 or 213.617 nm)

The oxalate extracts can be carried out in the working groups Soil Science and Agronomy (both Faculty of Agricultural and Environmental Sciences, University of Rostock).

Alternative extraction to the oxalate extract for the estimation of poorly crystalline Fe and Al oxides according to Reyes and Torrent (1997):

According to Reyes and Torrent (1997), the oxalate extract also extracts Fe and Al from the two Al silicates allophane (Al₂O₃·(SiO₂)_1.3-2·(H₂O)_2.5-3 [1]) and imogolite (Al₂SiO₃(OH)₄ [2]), while a solution of citrate + ascorbate (C-A extract) does not extract this Al and Fe. This means that an oxalate extract overestimates the proportion of pedogenic, poorly crystalline Al and Fe oxides if higher proportions of these phyllosilicates are present in the soil (Reyes and Torrent 1997). It is not possible to distinguish whether the additional Fe extracted with oxalate (compared to the C-A extract) originates directly from the allophane Fe or from the Fe oxides included in the allophanes (Reyes and Torrent 1997). The C-A extract can therefore be used to extract the poorly crystalline Al and Fe oxides in a more targeted manner. However, according to Reyes and Torrent (1997), if organic soil matter is present, a citrate extract is recommended in addition to the C-A extract, as the C-A extract also extracts parts of the Al and Fe bound to the organic soil matter, while the citrate extract preferentially extracts Al and Fe bound to the organic soil matter. The difference between the element concentrations (Al, Fe, P) from the C-A extract (poorly crystalline + organically bound) and the citrate extract (organically bound) then gives the element content from the poorly crystalline oxides.

Procedure for the extraction of poorly crystalline Al and Fe oxides according to Reyes and Torrent (1997):

There is no indication of the sample weight. As a comparison was made with the oxalate extract in the publication, a similar sample weight (0.5 g to 1.5 g soil) is assumed. Reyes and Torrent (1997) recommend that no more than 0.25 mmol (=15 mg) Fe per liter should be extracted with the C-A extract, i.e. the sample weight may have to be adjusted for the respective soil.

Protocol Citrate-ascorbate extract (C-A solution)

Reagents for the citrate-ascorbate solution = C-A solution (1 liter)

tri-sodium citrate dihydrate C₆H₅Na₃O₇ · 2H₂O molar mass: 294.10 g mol^-1

► Weigh 58.82 g tri-sodium citrate dihydrate (C₆H₅Na₃O₇ · 2H₂O) into a 1 liter wide-necked volumetric flask
► Add approx. 800 ml ultrapure water and stir on the magnetic stirrer and attach the pH meter
► Add solid ascorbic acid (approx. 7.6...8.4 g for 1 L) while stirring until a pH value of 6 is reached (between 0.38 and 0.42 g ascorbic acid is required per 50 mL solution)
► The solution then has an ascorbic acid concentration of between 0.043 and 0.048 M, up to a molarity of 0.05 M this has no effect on the extraction
► Fill up to 1 liter with ultrapure water

Extraction procedure

► Shake the sample in 50 mL of a 0.2 M Na-citrate-0.05 M Na-ascorbate solution (C-A solution; pH 6) in a maximum 60 mL centrifuge tube for 16 hours
► Centrifuge (1500 x g), filter through P-free filters
► Measurement of Al, Fe, (Mn) and P on ICP-OES

Notes on C-A extract (Reyes and Torrent 1997):

► The air space in the centrifuge tube should not be larger than 15 ml to prevent oxidation of the extraction agent.
► The extraction solution is well buffered and only small pH changes were observed in the extract.
► Up to 50 mg CaCO3 equivalent per 50 ml solution had no effect on the amount of extracted Fe. For samples with higher carbonate concentra-tions, 1 mmol citric acid per mmol CaCO₃ should be added so that the final pH value is 6 ±0,1.
► For deviations of <0.1 pH units, no effects of carbonate on the amount of extracted Fe were detected.

Protocol citrate extract

Reagents for the 0.2 M citrate solution (1 liter)

► Weigh 58.82 g tri-sodium citrate dihydrate C₆H₅Na₃O₇ · 2H₂O into a 1 liter wide-necked volumetric flask
► Add approx. 800 ml ultrapure water and stir on the magnetic stirrer and attach the pH meter
► Adjust to a pH value of 6 with 1 M KOH
► Fill up to the calibration mark with ultrapure water

Preparation 1 liter 1 M KOH

Molar mass KOH 56.1056 g mol^-1

► Pour approx. 800 to 900 ml ultrapure water into a volumetric flask
► Carefully add 56.11 g KOH (solid). Be careful, the solution will heat up!
► After cooling down to room temperature, fill the volumetric flask with ultrapure water to 1 liter

Extraction procedure

► Shake the sample in 50 mL of a 0.2 M Na-citrate solution (pH 6) in a maximum 60 mL centrifuge tube for 16 h (the sample weight must be identical to that in the C-A extract)
► Centrifuge at 100 x g, filter through P-free filters
► Measurement of Al, Fe and P concentrations on the ICP-OES.

Notes:

► It is essential to clarify in advance for both extracts whether the solutions can be measured on the ICP-OES at the given Na concentrations or whether dilution is necessary!
► When preparing the reagents, always work under the fume hood and wear protective clothing, as you are working with acids and alkalis.

The extracts according to Reyes and Torrent (1997) have not yet been carried out in the working groups Soil Science and Agronomy (University of Rostock) but could be requested.

4.5.3.2 Dithionite extract for the extraction of the more crystalline Al and Fe oxides and the P

Principle and suitability of the extract

The dithionite extract, which is used to extract poorly and well crystalline iron oxides, dates back to Mehra and Jackson (1960). The dithionite extraction according to Mehra and Jackson (1960) was converted into a DIN: DIN ISO 12782-2. In this DIN the principle is explained as follows (gray font): The extraction principle is mainly based on the reduction of Fe(III) phases to more soluble Fe(II) phases and on the tendency of the chemicals to form complexes for the extraction of iron from crystalline materials (Dijkstra et al. 2005). The amount of crystalline Fe-(hydr)oxides is determined by extraction with dithionite minus the amount of amorphous Fe-(hydr)oxides determined by extraction with ascorbate according to ISO 12782-1 and other possible reactive iron phases determined by extraction with dithionite which are important in certain materials, such as volatile sulphides (AVS) and silicate under acidic conditions.

The extract according to Mehra and Jackson (1960) is also known as dithionite citrate bicarbonate extract, DCB for short. According to Varadachari et al. (2006), however, the extraction is incomplete for hematites, for example, and contains some further contradictions. According to Varadachari et al. (2006), the modified dithionite extract is therefore a dithionite carbonate oxalate extract, in short DCO. The extract according to Varadachari et al. (2006) has a greater quantitative efficiency than the original DCB extract (Rennert 2019). According to Rennert (2019), an extract of ascorbic acid, due to the ability of ascorbic acid to reduce Fe(III) well crystalline oxides such as hematite and goethite, alone or in combination with oxalate, could also serve as a substitute for the DCB extract according to Mehra and Jackson (1960).

Sample preparation for all variants of the dithionite extract:

► Dry soil samples (see chapter 2.4, 3.1)
► Sieve soil samples <2 mm and use this <2 mm fraction (fine soil)
► Determine total element concentrations in a subsample (e.g. using aqua regia extract, see chapter 4.1.2)

Protocol for the dithionite extract (DCB) according to Mehra and Jackson (1960)

Preparation of chemicals:

0.3 M Na-citrate solution (C₆H₅Na₃O₄ 2 H₂O):
► Weigh 88 g of Na citrate into a 1-liter volumetric flask
► Fill up to the calibration mark with ultrapure water
1 M NaHCO₃ solution:
► Weigh 84 g NaHCO3 into a 1-liter volumetric flask
► Fill up to the calibration mark with ultrapure water
For a saturated NaCl solution:
► Add NaCl to water until it precipitates

Extraction procedure:

► Weigh 1 g of lutro fine soil into 50 ml centrifuge tubes or glass tubes (for heating block)
► Add 20 ml of 0.3 M Na citrate and 5 ml of 1 M NaHCO₃
► Heat to 75 to <80 °C in a water bath under the fume cupboard, eye protection!
► Add 1 g solid Na₂S₂O₄ (Na-dithionit) (Na-dithionite) while stirring vigo-rously (glass rod) and heat for a further 5 min, stirring frequently
► Add 1 g of solid Na₂S₂O₄ (Na-dithionite) again and heat for a further 10 min, stirring frequently
► The sample should now be gray (not red, yellow or brown), otherwise add Na-dithionite again!
► Remove samples from the water bath and allow to cool slightly (sample begins to precipitate)
► Centrifuge samples for 10 min at approx. 1100 x g (try out)
► If the sample does not flocculate after centrifugation, add 1 ml of saturated NaCl and centrifuge again
► Decant the supernatant using a P-free filter
► Dilution 1 : 10 for ICP-OES measurement

Protocol for the dithionite extract (DCO extract) according to Varadachari et al. (2006)

Note on dithionite extract (DCO) according to Varadachari et al. (2006):

The extraction has not yet been carried out in the laboratories of the working group Soil Science (Faculty of Agricultural and Environmental Sciences, University of Rostock), but, generally, could be tested there. Extraction must be carried out under the fume cupboard! Depending on the size of the water bath, up to 5 samples can probably be extracted at once. Due to the centrifuge, an even number of samples must be selected, or empty samples must be weighed with water in advance and placed in the centrifuge!

Sample weighing: The sample quantity depends on the Fe₂O₃ content. The specified extraction agent quantities apply to a maximum of 50 mg Fe₂O₃. 50 mg Fe₂O₃ corresponds to 35 mg Fe, which may be present in the weighed sample. In Cambisols (according to IUSS Working Group WRB (2022); Braunerde according to Ad-hoc-AG Boden (2005) and similar soil types on glacial till the Fe concentrations range between 6 and 21 g kg^-1 (Zimmer und Leinweber 2013). In podzols on sand, Fe concentrations varied between <1 g and 2.5 g Fe kg^-1 (Baum et al. 2013). The unit g per kg is equivalent to unit mg per g. At 21 mg Fe g^-1, 1.5 g of soil could therefore be weighed in without exceeding the 35 mg per extraction. At lower total Fe concentrations, e.g. 6 mg g^-1, the sample weight could theoretically be increased to 5 g. At higher Fe concentrations in the soil samples, the sample weight must be reduced.

Varadachari et al. (2006) list among others the following weaknesses of the DCB extract according to Mehra and Jackson (1960) and Aguilera and Jackson (1953):

► Several studies by other authors show that better crystalline Fe oxides, especially hematite, are only incompletely removed with the DCB extract; in some cases, even more Fe is extracted by the oxalate extract than by the DCB extract.
► There is no evidence that the propagated pH value of 7.3 and the reaction temperature of 80-90 °C are reasonable.
► The authors do not state that dithionite solutions are unstable and must be prepared fresh daily.
► The statements on the pH values for the precipitation of FeS contradict each other.
► The post-extraction washes were repeated too infrequently to ensure that the reduced Fe was actually completely present in the centrifuged supernatant/extract.
► The published X-ray diffractograms before and after the reaction cannot prove complete hematite extraction, as typical lines for hematite also appear in the diffractograms after DCB extraction.

The following points were investigated by Varadachari et al. (2006) for the dithionite extract from theoretical considerations and experiments:

► The mechanisms of reduction by Na-dithionite
► The specific structure of the dithionite ion means that it is stable in the dry state but not in the dissolved state and therefore decomposes to elemental S at acidic pH values; it is stable at neutral to alkaline pH values and S is hardly/not precipitates
► Dithionite is a stronger reducing agent under alkaline conditions than under acidic conditions (standard potential in alkaline: E⁰ = 1.12 V; in acidic E⁰ = 0.056 V);
► While dithionite decomposes in aqueous phase, especially under acidic conditions, H₂S and S are formed, which can precipitate with Fe to form largely insoluble FeS
► Metal oxides in the oxidized state such as Fe(III) are largely stable and are only dissolved by changing the oxidation state, e.g. to Fe(II), which requires a strong reducing agent
► In the decomposition of dithionite, the SO₂^- radical is the strongest reducing agent

From this, Varadachari et al. (2006) derive, among other things, the follow-ing conditions for reactions with dithionite:

► The reactions must take place at alkaline pH values, as the reducing power of the dithionite is strongest here and the formation of H₂S is largely avoided.
► The precipitation of FeS by the FeS formed in side reactions, even under alkaline conditions, must be avoided by adding complexing agents.

The following reaction conditions were therefore optimized by Varadachari et al. (2006) (the optimum combination is shown in brackets):
Effects of the complexing agent (oxalate), the temperature (100 °C), the pH value (8), the amount of complexing agent (30 ml for 2 g Na-dithionite), the reaction time (30 min), the amount of dithionite (2 g) and the method of dithionite addition (portion wise).

DCO extraction protocol; an attempt was made to adapt the process to the existing equipment in the working group Soil Science; however, the process has not yet been tested.

► Weigh-in: approx. 1 g soil <2 mm (depending on Fe content, see above)
► Add 30 ml of the oxalate-carbonate solution to a 250 ml Erlmeyer flask, add sample (the depth of the solution should be a maximum of 1.5 cm to ensure contact of the sample with the reductive zone (dithionite) occurring near the surface)
► Place Erlenmeyer flask with suspension in the hot water bath under the fume cupboard.
► As soon as the water bath has reached approximately 100 °C (at least 70 to 80 °C), add 0.4 g Na-dithionite (Na₂S₂O₄) while continuously stirring the sample with a glass rod (when adding the sample and the dithionite, remove the glass rod briefly so that nothing sticks).
► Continue to add Na-dithionite approx. every 5 min until a total of 2 g Na-dithionite (optimum) has been added (no more, otherwise the solution will become acidic)
► Approx. 30 min after the first and 10 min after the last addition of dithionite, remove the Erlenmeyer flask from the water bath and allow to cool to room temperature
► Transfer the suspension completely into 50 ml centrifuge tubes (funnel) (rinse with (try 10 ml at a time?) 1 M NaCl or KCl), centrifuge (3000 x g for 20 min?) and transfer the supernatant through a filter into a graduated flask (100 ml?)
► Wash the sample up to 5 times with 1 M NaCl or KCl (10 to 15 ml each time): shake up each time, centrifuge and decant the supernatant into the volumetric flask each time
► Fill the volumetric flask up to the calibration mark with 1 M NaCl or KCl
► Measurement of Fe, Mn, Al and P in the extract using ICP-OES, compare with oxalate extract (in the dark) to estimate well and poorly crystalline Fe (see formulas above)

Preparation of the reagents:

Oxalate-carbonate buffer solution:

► Weigh 33.3 g Oxalic acid (H₂C₂O₄ · 2H₂O) and 35.3 g anhydrous Na carbonate (Na₂CO₃) into a 1-liter volumetric flask
► Add ultrapure water and heat if necessary to dissolve the chemicals (oxalic acid reacts to form Na-oxalate),
► Check the pH value (it should be 8.05), add oxalic acid or carbonate, if necessary (pH should be >7 in any case),
► Allow the solution to cool and fill up to 1 L

1M NaCl solution:

► Weigh 58.44 g NaCl into a 1-liter flask
► Fill up to 1 liter with ultrapure water

Notes:

► Na-dithionite should be stored in a desiccator, as it easily absorbs water at normal humidity, which causes decomposition of the Na-dithonite and thus leads to a loss of extraction efficiency.
► Washing the reaction suspension: After completion of the reaction, the suspension transferred to a centrifuge tube must be washed at least 5 times and the respective supernatants must be combined after centrifugation. To this, 1 M KCl or NaCl is added repeatedly. Washing also appears to be necessary above all to prevent dispersion of the clay and the deposition of extraction agent residues in the clay residue, i.e. if the extraction residue is also to be collected as pure clay minerals.
► The success of extractions with oxalate or dithionite can be estimated by measuring the crystallinity by X-ray diffraction (XRD) of the Fe oxides before and after extraction in the soil sample (e.g. Pawluk 1972, Landa and Gast 1973, Rennert 2019). Since oxalate should only extract the poorly crystalline Fe oxides and poorly crystalline Fe oxides cannot be detected by XRD, the XRD spectra before and after extraction with oxalate should be identical. Since dithionite (or DCO) should also extract the well-crystalline Fe oxides, the corresponding peaks must be missing in the spectra after extraction compared to the non-extracted samples.
► The dithionite extracts are not part of the routine analysis of the working group Soil Science (Faculty of Agricultural and Environmental Sciences, University of Rostock), but could generally be carried out there on request.
► There were problems with both the KCl and NaCl solution in the ICP-OES during experiments. It is therefore essential to speak to the laboratory technician for the ICP-OES in advance!

References

[1] https://www.mineralienatlas.de/lexikon/index.php/MineralData?mineral=Allophan
[2] https://www.mineralienatlas.de/lexikon/index.php/MineralData?mineral=Imogolit

Ad-hoc-AG Boden (2005) Bodenkundliche Kartieranleitung, 5. Auflage. Hrsg.: Bundesanstalt für Geowissenschaften und Rohstoffe, Hannover, Schweizerbart’scher Verlag, Stuttgart

Aguilera, NH, Jackson, ML (1953) Iron oxide removal from soils and clays. Soil Sci Soc Am Proc 17, 359-364.

Ahl, C, Leinweber, P (2013) Jahrestagung der Deutschen Bodenkundlichen Gesellschaft 7.-12. September 2013, Rostock, Böden – Lebensgrund-lage und Verantwortung. DBG-Mitteilungen 116

Baum, C, Kahle, P, Leinweber, P (2013) G3: Fischland-Darß – Bodenent-wicklung auf pleistozänen und holozänen Sedimenten. Seiten 71-72 In Leinweber, P, Ahl, C (2013) Exkursionsführer der Jahrestagung der Deutschen Bodenkundlichen Gesellschaft 7.-12. September 2013, Rostock „Böden - Lebensgrundlage und Verantwortung“

Carter, CM, Van der Sloot, HA, Cooling, D (2009) pH dependent extraction of soils and soil amendments to understand the factors controlling element mobility - New approach to assess soil and soil amendments. Eur J Soil Sci 60, 622-637

Das, S, Hendry, MJ, Essilfie-Dughan, J (2011) Transformation of Two-line ferrihydrite to goethite and hematite as a function of pH and temperature. Environ Sci Technol 45, 268-275, DOI: 10.1021/es101903y

Deb BC (1950) The estimation of free iron oxides in soils and clays and their removal. J Soil Sci 1, 212–220, DOI: 10.1111/j.1365-2389.1950.tb00733.x

del Campillo MC, Torrent, J (1992) A rapid acid-oxalate extraction procedure for the determination of active Fe-oxide forms in calcareous soils. Z. Pflanzenernähr Bodenk 155, 137-440

DIN ISO 12782-2: 2012 Bodenbeschaffenheit – Parameter zur geochemi-schen Modellierung der Elution und Speziation von Bestandteilen in Böden und Materialien – Teil 2: Extraktion von kristallinen Eisenoxiden und –hydroxiden mittels Dithionit

DIN ISO 19684-6 Bodenuntersuchungen im landwirtschaftlichen Wasserbau Chemische Laboruntersuchungen, Teil 6: Bestimmung des Gehaltes an oxalatextrahierbaren Eisen.

Dijkstra, JJ, Meeussen, JCL, Comans, RNJ (2009) Evaluation of a generic multi-surface sorption model for inorganic soil contaminants. Environ. Sci. Technol. 43, 6196-6201

Gypser, S, Hirsch, F, Schleicher, AM, Freese, D (2018) Impact of crystalline and amorphous iron- and aluminum hydroxides on mechanisms of phosphate adsorption and desorption. J Environ Sci 70, 175-189, DOI: 10.1016/j.jes.2017.12.001

IUSS Working Group WRB (2022) World Reference Base for Soil Resources. International soil classification system for naming soils and creating legends for soil maps. 4th edition. International Union of Soil Sciences (IUSS), Vienna, Austria; https://www.isric.org/explore/wrb

Landa, ER, Gast, RG (1973) Evaluation of crystallinity in hydrated ferric oxides. Clay Clay Miner 2 I, 121-130

Mehra, OP, Jackson, ML (1960) Iron Oxide removal from soils and clays by a dithionite-citrate system buffered with sodium bicarbonate. Clays Clay Miner 7, 317-327

Miller, WP, Zelazny, LW, Martens, DC (1986) Dissolution of synthetic crystalline and noncrystalline iron oxides by organic acids. Geoderma 37, 1-13

Pawluk, S (1972) Measurement of crystalline and amorphous iron removal in soils. Can J Soil Sci 52, 119-123

Parfitt, RL, Childs, CW (1988) Estimation of forms of Fe and Al: A review, and analysis of contrasting soils by dissolution and Moessbauer methods. Aust J Soil Res 26, 121-44

Rennert, T (2019) Wet-chemical extractions to characterise pedogenic Al and Fe species – a critical review. Soil Res 57, 1–16, DOI: 10.1071/SR18299

Reyes I, Torrent J (1997) Citrate-ascorbate as a highly selective extractant for poorly crystalline iron oxides. Soil Sci Soc America J 61, 1647–1654. DOI: 10.2136/sssaj1997.03615995006100060015x

Schwertmann, U (1964) Differenzierung der Eisenoxide des Bodens durch Extraktion mit Ammoniumoxalat-Lösung. Z Pflanzenern/Bodenk 105, 194-202

Schwertmann, U (1973) Use of oxalate for the Fe extraction from soil. Can J Soil Sci 53, 244-246

Varadachari, C, Goswami, G, Ghosh, K (2006) Dissolution of iron oxides. Clay Res 25, 1-19

Zimmer, D und Leinweber, P (2013) G4: Glaziale Serie: Geomorphologie, Böden und Bodenprobleme. Seiten 86-90 In Leinweber, P und Ahl, C (2013) Exkursionsführer der Jahrestagung der Deutschen Bodenkundlichen Gesellschaft 7.-12. September 2013, Rostock "Böden - Lebensgrundlage und Verantwortung"

For citation: Zimmer D, Baumann K (year of download) Chapter 4.5.3 Estimation of P binding to pedogenic oxides in soil by wet chemical methods (Version 1.0) in Zimmer D, Baumann K, Berthold M, Schumann R: Handbook on the Selection of Methods for Digestion and Determination of Total Phosphorus in Environmental Samples. DOI: 10.12754/misc-2020-0001

download chapter (PDF)