INVESTIGATION OF METHODS FOR Fe , Mn AND Zn SOLUBILITY IN THE SMONITZAS OF SERBIA

The methods for estimating Fe, Mn and Zn solubility in smonitzas over Serbia (ten localities) used as fields and meadows, were studied in this paper. The total contents of these elements were found to be nearly equal to their mean values in the world soils. Regarding their availability, investigated smonitzas were found to be well supplied with manganese and, for the majority of samples, with iron. But, deficiency of iron in the soil were found in 25% and zinc in 65% of analyzed samples, after a more sensitive DTPA method had been used. A certain number of positive or negative correlations between the total and the available contents of the elements studied and some basic properties of the smonitzas were obtained, which is in agreement with the behavior of these elements in the soil.

organic-mineral, liquid and gaseous phases, being characterized with some physical, chemical and biological properties (K a b a t a -P e n d i a s and P e n d i a s ,1989).
The origin of microelements in the soil is mainly geo-chemical (originating from parent substrate) and partly anthropogenous (various sources of pollution).The soil solution contains considerably small amounts of the microelements, which are mostly adsorbed on the colloid particles of soil, occluded on oxides of iron and manganese, and are found to be bound in various compounds of organic and inorganic origin.
The dynamics of the microelements in the soil is, therefore, dependent on how strong these elements are bound over the solid phase of the soil.The weakest bond existed with exchangeable adsorbed microelements and the strongest one with those bound in the structure of the primary and secondary minerals.Thus, the extraction of the microelements from the soil seems to be highly varying and depends on the strength of the extractant.By means of the appropriate extractant, total, pseudo-total and the available content of the individual microelements in the soil, could be determined (A l l o w a y , 1995).The total content is most often determined by destructing samples with the mixture of acids: HNO 3 , HF and HClO 4 .Through this procedure, all the forms of the elements present in the soil were converted into a solution.However, less complex and, therefore, more often used are the methods for determining of the pseudo total contents of the microelements, that is, the methods with the application of concentrated HNO 3 and H 2 O 2 or "aqua regia" (HCl +HNO 3 ).Strong mineral acids and their mixtures do not solve silicates and silicate complexes, but are strong enough to extract the microelements that are not bound with silicate forms.
A range of extractants is used to determine microelement availability in a soil.Mineral acids, most often 0.1 M HCl were also used for such extraction.However, the use of chelate reagents, such as EDTA and DTPA (R i s s e r and B a k e r , 1990) is being on increase.Critical levels of contents, i.e. their boundary values dependent on the extractant used, were determined for the microelements that are most significant to mineral nutrition (Fe, Mn, Zn and Cu).
Microelements have a manifold role in the plant life.Since the microelements may directly or indirectly affect or take part in the physiological processes of the plants, a detailed study was conducted on their solubility and availability in smonitza.

Materials and Method
The investigations were conducted on the smonitza type of soil taken from ten different localities in Serbia.Soil samples had been taken from the field and meadow ecosystem at the depth of 0 to 20 cm (acc.to map 1: 200000) after which the samples were dried (up to air dry state) and crashed in the porcelain mortar up to particles 1 mm.
The basic physical and chemical properties of soil were determined using standard methods.The following analyses of the soil were made: acidity in the suspension with water and 1M KCl, with the ratio of soil/solution 1: 2.5, humus content using humus method after Kotzmann, available P and K, content using Al method after Egner-Riehm, CEC, using the method with 1 M NH 4 OAc, pH 7, whereas the mechanical fractions of clay were determined using pipette method.
The contents of the total microelements content (Fe, Mn and Zn) were determined by AAS, after digestion of the soil samples, that had previously been heated (2h) at 450 o C, with the mixture of acids (HF, HNO 3 and HClO 4 ) in Pt dishes (U r e , 1995).
"Pseudo-total" contents of the microelements were determined by means of atomic absorption spectrometry, after the acidic digestion with HNO 3 and H 2 O 2 .
The content of the available microelement soil samples was determined, using AAS method, after the extraction with 0.1 M HCl and 0.005 M DTPA.

Results and Discussion
The analyses have shown that the samples taken from 10 localities in Serbia were considerably differing regarding their physical and chemical properties (tab.1).The reaction of soil was found to range in a fairly wide interval of pH values from acid to weakly alkaline reaction (from 4.60 to 7.04 in n KCl).The content of humus was considerably varying, ranging from 2.00 to 5.59% (weakly humus to humus soils), being unaffected by land use.In addition, the content of available phosphorous also ranged in a wide interval.Thus, on the smonitzas under meadow soil, the content of available P ranged from 0.8 to 17.8 mg/100g, whereas on those under field, its content ranged from 0.6 to 28 mg/100g.According to the content of the available P, the smonitzas in question are low to medium supplied with this element.In contrast to the available phosphorous, the soils were found to be rich in potassium (19.0-53.5 mg/100g).In addition, these soils indicated a high cation exchange capacity and, by mechanical composition, could be defined as medium to heavy clay soils.The above-mentioned average values of the investigated properties proved to be very much the same for the field and for the meadow smonitzas.
The content of total Fe indicated nearly the same values for meadow and field soils, i.e. 3.6% and 3.7%, respectively (tab.2).The values obtained are in the range from 5.000 to 50.000 mg/kg and can be considered to be normal for the soils (L y o n et al., 1982).While determining pseudo total content of Fe using the extraction with HNO 3 (U r e , 1990), the extracted Fe amounted to 84.9% in the field and to 83.4% in the meadow soil (tab.3), of its total content.The data suggest that 15 and 17% of the total Fe in the soil are bound in the structure of the primary and secondary minerals.Hence, the method applied does not seem to be a very suitable one for determination of geochemical originates Fe.After using both methods (HCl and DTPA) for extracting available Fe forms, the content of soluble Fe was found to be extremely low, compared to the total one, which, when extracted in 0.1 M HCl averaged 53.5 mg/kg or 0.15% of its total content for a field, and 60.6 mg/kg or 0.18% of its total content for a meadow soil.These values are twice as much than those obtained with DTPA extraction (0.07% for the field and 0.09% for the meadow smonitza).According to the boundary values of the available microelement forms in 0.1 M HCl obtained by A n k e r m a n (1977), the samples have a high level of available Fe.However, according to the content of the available Fe in 0.005 M DTPA, two groups of the soils can be classified: those with low Fe content with the values ranging from 7 to 10 mg/kg Fe and, those with a good supply with Fe (from 22 to 86 mg/kg), which suggests a higher selectivity of this determination method.
The mean value of the total Mn content in the meadow and field smonitzas amounted to 882 mg/kg and 927 mg/kg, respectively.However, B o w e n (1979) found out that normal soil contained 1000 mg/kg Mn (within the interval from 20 to 10000 mg/kg), on the average, while F r e e d m a n and H u t c h i n s o n (1981) established 850 mg/kg Mn to be the mean value (from 100 to 4000 mg/kg).
The extraction of the pseudo total forms of Mn with HNO 3 was found to be the highest in relation to the other two elements, amounting to over 90% of its total content.Therefore, it may be concluded that the method with concentrated HNO 3 and H 2 O 2 ensured highly real total contents, considering that the values ranged from 81.9 to 99.6% for field and from 85.7 to 102% for meadow smonitza.The average amounts of the extracted Mn with different solvents from the meadow and field smonitzas' samples were found to be similar (tab.2).However, the amount of the extracted Mn was considerably differing with the application of the extraction with differing solvents.Thus, Mn amount was doubled when extracted with solution 0.1 M HCl, compared to that extracted with DTPA solution.According to A n k e r m a n (1977), boundary values of the content of the available microelement forms in 0.1 M HCl indicated that the soils concerned were highly rich in manganese.Also, L i n d s a y and N o r w e l l (1978) established that the soils, containing more than 1 ppm DTPA extracted Mn, could be considered as well supplied with this element.In our case, the smonitza samples indicated even a higher Mn level and, therefore, are well supplied with Mn available to the plants.
The content of the total Zn (extracted with mixture HNO 3 -HF -HClO 4 ) averaged 82.0 mg/kg on both, field and meadow soils, with the value being somewhat higher than the average Zn content in the world soils (60.0 kg/kg, D a v i e s and J o n e s , 1988).However, when compared with maximally permissible content in a soil (300 mg/kg, B o w e n , 1979) this content is considerably lower.In the individual samples of the soil under study, the content of Zn was found to range from 55.0 to 119.0 mg/kg in the meadow and from 60.0 to 117.0 kg/kg in the field soil (tab.2).
The results have shown that for the total contents of Zn, mixture of HNO 3 +H 2 O 2 indicated a lower efficiency than mixture of HF+HNO 3 +HClO 4 did (tab.3).On the average, nitric acid extracted 79.4% of its total content in the smonitzas under field and, 77.1% of its total content in those under meadow.Variations between relative values coincided with those between the total Zn contents in the soil samples.The greatest amounts of Zn were mostly extracted from the samples with Zn highest total content.Therefore, it may be concluded that about 20% Zn exists in crystal structures of the silicates, which are the forms of metals that have the highest resistance to digestion (U r e , 1995).
The content of the available Zn varied in dependence of the extractant.Thus, the mean value of the available Zn amounted to 3.4 mg/kg, using 0.1 M HCl and 0.8 mg/kg with 0.005 M DTPA for field, and 4.0 mg/kg (extraction in 0.1 M HCl) and 1.1 mg/kg (in 0.005 M DTPA) for meadow smonitzas.According to A n k e r m a n (1977), on the basis of the boundary values of the contents of the available microelement forms in 0.1 M HCl, the soils indicated a medium content of the available Zn. On the other hand, according to the content of Zn in 0.005 M DTPA, i.e. according to L i n d s a y and N o r w e l l (1978), two groups of smonitzas existed: one with low Zn content, with values ranging from 0.5 to 0.8 mg/kg Zn and one with the medium ZN content (from 1.0 to 2.2 mg/kg).The extraction with 0.1 M HCl was four times stronger than that with DTPA solution.Thus, the results have shown that 0.1 M HCl led to the release of the metals that were comparatively strongly bound in the solid phase of the soil (R i s s e r and B a k e r , 1990).A higher extractive efficiency of 0.1 M HCl than that of DTPA was also observed before (B r o w n et al., 1971: H w a n g et al., 1988: J a k o v l j e v i ć et al., 1997).
The dependences between the individual microelement forms and the established soil properties can be seen in the Tab 4.
Chemical and physical properties of the soil were found to affect the total and the available Fe amounts in the field soil samples (tab 4).The medium correlation existed between the total and pseudo total Fe content and % clay and % silt + clay.The amounts of Fe, extracted with 0.005 M DTPA, indicated a highly strong negative correlation with pH (R =-0.97 xxx ), CEC (R=-0.90xxx ), clay (R=-0.76xx ) and a positive one with silt content in the soil (R=0.72 x ).
The content of the total and pseudo total Fe in the samples from meadow indicated the medium and strong dependence on pH (R = 0.61-0.73),CEC T a b .4 While extracting Mn with 0.1 M HCl and 0.005 M DTPA in the samples from field, the only correlation coefficient was obtained between soil pH and DTPA -available Mn.However, the contents of the individual Mn forms in the meadow soil samples were non-dependent on the soil properties.
Statistically significant correlation coefficients of the samples from the field smonitza existed between the total, pseudo total content of Zn and the mechanical fractions of clay and silt+clay.The available Zn content was non-dependent on the soil properties.
When meadow soil samples are concerned, much more correlation coefficients between total Zn content and soil properties were found.Significant correlation between available Zn content were found only for soil pH, when Zn have been extracted with 0.1 N HCl.

C o n c l u s i o n
On the basis of research results, the following conclusions may be drawn: 1.The contents of the microelements (total and pseudo total) were found to be similar in the samples from field and meadow and to have approximately equal levels as the averages for the "world soils".
2. The extraction of the microelements with NHO 3 and H 2 O 2 for Mn brought about the values similar to those for its real total contents (about 90%), whereas the lower ones were evidenced for Zn and Fe (around 80%).
3. The extraction with solution 0.005 M DTPA brought about lower levels of the available microelements (Zn, Mn and Fe) compared to the levels with solution 0.1 M HCl.The amounts of Mn and Fe were doubled and those of Zn even four times higher, when extracted with 0.1 M HCl, compared with those extracted with 0.005 M DTPA.
4. The levels of the microelements extracted in 0.1 M HCl and 0.005 M DTPA were found to be higher in the samples from meadow than those from field, excepting soluble Mn in 0.1 M HCl, which indicated similar values.
5. The boundary values of DTPA method indicated that Mn existed in high, Fe and Zn in the prevalently high amounts in the smonitzas studied, whereas some samples were discovered to be deficient in Zn (65%) and Fe (25%).
6.A certain number of significant correlations existed between the contents of the microelements (total and available) and soil properties.Positive correlations were found between the total and available microelements contents in relation to those of clay, silt CEC and soil pH.For Fe extracted in DTPA numerous and very significant correlation coefficients were found.U radu su proučavane metode za ocenu rastvorljivosti Fe, Mn i Zn u smonicama Srbije sa deset lokaliteta.Uzorci zemljišta su uzeti sa njivskog i livadskog ekosistema sa dubine od 0 do 20 cm (prema karti 1: 200000).Osnovne fizičke i hemijske osobine zemljišta određene su standardnim metodama.
. -Correlation coefficients among total and available contents of microelements and some soil properties