Extractable aluminium , iron and manganese in mineral soils II Extractability by oxalate and pyrophosphate

Acid ammonium oxalate (Tamm 1922, Schwertmann 1964, McKeague and Day 1966) and 0.1 M Na 4P 2 07 (Aleksandrova 1960, McKeague 1967) are conventionally used for the extraction of aluminium, iron and manganese from soil. Amorphous oxides and metals bound to organic matter in soil are dissolved by these solutions. The use of these reagents, however, may cause some problems in analytical work. Abstract. The extractability of aluminium, iron and manganese by 0.05 M oxalate and pyrophosphate was studied in samples of 23 mineral soils. Dilute extractants were studied because conventional reagents may cause problems in analytical work. The mean values for Al, Fe and Mn extracted by conventional Tamm’s oxalate were 67, 81 and 1.5mmol/kg soil, respectively. On the average, 0.05 M oxalate solutions at pH 2.9 and 4.2 extracted Al, Fe and Mn amounts that were 103, 113 and 87 % and 72, 82 and 83 % of the amounts extractable by Tamm’s oxalate, respectively. Eeach metal released by 0.05 M oxalates correlated closely with that dissolved by Tamm’s oxalate; the r values ranged from 0.967*** to 0.997***. The mean values for Al, Fe and Mn extracted by 0.1 M Na4P 207 and 0.05 M K 4P 207 were 38, 28 and 0.6 and 33, 29 and 0.6 mmol/kg soil, respectively. The amount of each metal extracted by Na 4P 2 07 correlated closely with that released by K 4P 207 ; the r values ranged from o.B7*** to o.97***.

Traditional colorimetric, titrimetric and gravimetric techniques have commonly been superseded by atomic absorption spectrophotometry.Application of this technique may sometimes be hampered by the tendency of the nebulizer and burner slot to clog when solutions of high salt concentration are aspirated.Acid ammonium oxalate has often proved to be especially troublesome in this respect (e.g.Webber et al. 1974, Searly and  Daly 1977, Simmons and Plues-Foster  1977).Difficulties may partially be overcome by dilution of the solutions to be analyzed.Pyrophosphate is difficult to use as an extrac- tant because of soil dispersion.Especially with clay soils, normal filtration through paper is not enough for clarifying extracts (Sheldrick and McKeague 1975, Schuppli et al. 1983).
These difficulties in analytical work should be alleviated by the use of more dilute extrac- tants.A previous paper (Niskanen 1989) showed that the ability of oxalate and pyrophosphate to extract soil aluminium, iron and manganese depends on the pH value, and that the extractability by 0.05 M reagents can reach the same order of magnitude as that by conventional methods.The aim of this paper was to study the possibility of employing 0.05 M oxalate and pyrophosphate solutions instead of the traditional reagents for the extraction of aluminium, iron and manganese from soil.
The soils were air-dried and ground to pass through a 2-mm sieve.The particle-size distri- bution of the inorganic material in the soil was determined by the pipette method (Elonen 1971), the organic carbon content by the Ai.ten wet combustion method (Graham 1948).The soil pH was measured in a soil-0.01M CaCl 2 suspension (1:2.5 v/v) (Ryti 1965).
The soils were extracted by the methods given in Table 2. Na 4 P207 extracts were clarified by the addition of 1 M HCI and then filtration through hard paper, K 4P 2 0 7 ex- Table 1.Soil samples (a = o -2 o 0 -20 cm, b = 20-40 cm, V = virgin soil).tracts by filtration through 0.2-|im membrane filters.Aluminium, iron and manganese in filtrates were determined by atomic absorption spectrophotometry, iron and manganese with air-acetylene and aluminium with N2O--flame.The experiment was carried out in duplicate.

Extractability by oxalate
Aluminium and iron extracted byTamm's oxalate, respectively, were nearly of the same order of magnitude as those extracted by 0.05  M oxalate at pH 2.9, but higher than those extracted by 0.05 M oxalate at pH 4.2 (Tables 3 and 4).In most samples Tamm's oxalate ex- tracted more manganese than that extracted by dilute oxalates (Table 5).The metals ex- tracted by dilute oxalates correlated closely to those extracted by Tamm's oxalate.Regression equations describing the relationship be- tween extractabilities and corresponding linear correlation coefficients were as follows: that extractable by Tamm's oxalate (Table 6).
In this study, pyrophosphate extracts were cleared by precipitation of suspended clay and organic matter with acid or filtration through 0.2-pm filters.The latter method seems to be more certain, as acidification of extracts may dissolve metals from suspended material or, on the other hand, metals may partially re- main in precipitates.
K 4P 2 Oj, used by Bascomb (1968), was used as an extractant because it was thought that peptization of soil may be less than if Na 4 P 2 0 7 were used.This opinion was based on the fact that the hydration sphere of K + -ion is smaller than that of Na + -ion.However, lowering the concentration of pyrophosphate from 0.1 M to 0.05 M hardly reduced dispersion of soil.According to Elonen (1971), 0.05 M concentration of Na 4 P 2 0 7 is high enough for peptization of soil in particle-size analysis.

Conclusion
It seems possible to employ oxalate and pyrophosphate extractants, which are more dilute than conventional reagents.The amounts of metals released by dilute and traditional reagents are closely correlated.In general, extraction methods do not give absolute contents of elements.For example, the amounts extracted increase as the extraction time increases.However, the results obtained with a given method are valuable for soil comparisons.SELOSTUS Kivennäismaiden uuttuva alumiini, rauta ja mangaani Il Uutluvuus oksalaatti-ja pyrofosfaattiliuoksilla

Table 3 .
Extractable aluminium, mmol/kg soil.* *Each soil tested separately.Values marked with the same letter do not deviate, with 5 % risk.

Table 4 .
Extractable iron, mmol/kg soil.* *Each soil tested separately.Values marked with the same letter do not deviate, with 5 % risk.

Table 5 .
Extractable manganese, pmol/kg soil.* *Each soil tested separately.Values marked with the same letter do not deviate, with 5 % risk.