JOURNAL OF THE SCIENTIFIC AGRICULTURAL SOCIETY OFFINLAND Maataloustieteellinen Aikakauskirja 287 Vol. 54:287-296, 1982 Chloride and sulphate solutions as extractants for soil P I Effect of ionic species and ionic strength on P desorp- tion HELINÄ HARTIKAINEN and MARKKU YLI-HALLA Department of Agricultural Chemistry, University of Helsinki, 00710 Helsinki 71 Abstract. The effect of Cl and SO 2 ', anions as well as that of the ionic strength on the desorption of soil P were studied in 102 mineral soil samples by extracting them with KCI and K2 SO, solutions at ionic strengths of 0.025 and 0.1. The quantities of salt solubleand water soluble P in the soils were compared. Both sulphate solutions extracted more P in every single sample than either of the chloride solutions. On the other hand, the material could be divided into three groups according to the position of water in the order of extraction efficiency. In the group W>S>Cl, consis'ing of 53 samples, water was the most effective extractant, in the group S>W>Cl of 37 samples water was less effective than sulphate solutions but more effective than the chloride solutions, and in 12 samples (S>Cl>W) water was the least effective, less effective than even chloride. The groups classified according to P extractability did not deviate from each other in terms of soil texture,pH or org. C %, but the salt solutions tended to be the more effective extractants the poorer theP status of the soil was. The salt soluble and water soluble P which are closely related to each other correlated with the same factors. Exclusive of the group of 12 samples (S>Cl>W), the highest values of correlation coefficient were found for NH,F-P/Al, i.e. the molar ratio of NH,F soluble P (CHANG and JACKSON's method) to oxalate extractable A 1 (r=0.89 ’-0.93'“). The absolute differences between amounts of P dissolved in KCI and K2 S0 4 solutions of the same initial ionic strength were the greater the more water soluble P the samples contained (r=0.58 -0.94 "). An increase in ionic strength tended to depress the extractability of P in both salt solutions. Therefore the ligand exchange between sulphate and phosphate or hydroxyl was regarded unprobable. A theory of the extraction mechanism of sulphate was presented. The greater extraction efficiency of sulphate was assumed to be caused by the exchange reactions with H 2O groups which affect the ionic strength in a solution and the electric condition near the surface. Introduction The ionic composition and concentration of a particular ion in soil solution may vary markedly during the growth season. The phenomenon is well-known in cations, but it holds good also for anions. The competition between anions in soil is maintained by changes in the ionic strength and https://www.c-info.fi/en/info/?token=S6BzS_i58pIE6Tzb.BNKyGlBIVxufWToEItor0Q.mSEP6iuTX50dGSbZ9bjZ1klfKaRUB_TPk1bbyFwqFdWA2ZiVXiQcv_fXUnakRfdgyai1NGzvSzVmnCT80XCBVAhduKxOSF2fmZ1ywAR9KwjM9NKUNXGy_wyp_6kNBByga4hCFWyedYc4RCBvTt3OmrmQTE_KjdcaqWKdpYD3OF6KgB1ydBK-R_Sf4Y65ck0L52U4gZbji6sPag 288 anion composition in soil solution which, in turn, are affected e.g. by the fertilization, irrigation and rain as well as water and nutrient uptake of plants. Thus, external factors may have some effect on the solubility of soil P. The desorption of sorbed anions from soil tends to be depressed by applica- tion of so-called indifferent anions (RYDEN and SYERS 1975) and enhanced by application of anions with sorption tendency (DEMOLON and BASTISSE 1934, HINGSTON et al. 1967, etc.). The extent of displacement is dependent on the sorption affinity of the anions involved and their activity ratios in a solution (e.g. HARWARD and REISENAUER 1966) and the anion saturation on the surface of oxides providing sorption sites (e.g. HARTIKAINEN 1979). Also the nature of sorption components may be of importance, but mostly only the hydrated oxides of Al and Fe only are practically responsible for an anion retention observed. ' . ■ purpose of the present study was to illustrate the effectiveness of vari .us electrolyte solutions as P extractants. In addition to the action of ionic strength and anion species in a given solution, also the importance of the ? factors in different soils was investigated, comparing the extractability of soil P by KCI and K 2S04 solutions at two ionic strengths. The results were assumed to give some information about the consequences of the recent development in Finland, where the significance of chloride as a fertilizer component has increased at the expence of sulphate. Materials and methods The material consisted of 102 mineral soil samples; 17 heavy clay soils, 51 coarser clay soils and 34 non-clay soils. Exclusive of two heavy clay soil samples no more available, the material was the same as that presented in an earlier paper by HARTIKAINEN (1982). Soil P was extracted by two KCI and K 2S0 4 solutions of ionic strengths of 0.025 and 0.1, the KCI concentration being 0.025 M and 0.1 M and the K 2S0 4 concentration 0.0083 M and 0.0333 M, respectively. According to MOORE (1974, p. 443), solutions equal in ionic strength involve equal electrostatic forces. KOENIGS et al. (1981) state that the ionic strength in plough layer is about 0.1 immediately after fertilization. The soil - solution ratio was 1:60 (w/v) and desorption time 18 hours. The soil suspensions were centrifuged and filtered through a membrane filter of 0.2 pore size. The pH of the extracts was measured and the solutions were analyzed for P by the molybdenum blue method (ANON. 1969) and for Ca by an atomic absoption spectrophotometer (acetylene - nitrogen oxidule flame). Results Table 1 shows the relative amounts of P dissolved from various textural soil classes by KCI and K 2S04 solutions of ionic strengths of 0.025 and 0.1. By way of comparison, the quantities of water soluble P (analyzed by Table 1. Relative extractability of soil P in different solutions. Means with the confidence limits at the 95 per cent level. Soil class H 2O KCI K2 SO, I = 0.025 I = 0.1 I = 0.025 I = 0.1 Heavy clays Coarser clays Non-clay soils 100 68+33 56±31 89+38 80±34 100 104 ±4l 91+35 134±15 123±42 100 102± 18 98±21 142±27 140±30 HARTIKAINEN 1982) were denoted by 100. Comparison of means of P quantities extracted by salt solutions and water soluble reserves showed that from heavy clay soils the salt solutions dissolved less P, but from the other soil groups at least as much or even more than did water. A closer examina- tion disclosed, however, coaser clay as well as non-clay soil samples, where P was considerably more extracted by water than by either salt solution. When comparing only salt solutions as extractants for soil P, both sulphate solutions were found more effective than the chloride solutions. In 88 samples the order of solubility of P in various extractants was: K 2S0 4(I = 0.025)>K250 4(1 = O.I)>KCI (I = 0.025)>KC1 (I = 0.1) This was the average order in the whole soil material, but there were 14 samples from which either of the more concentrated salt solutions seemed to dissolve slightly more P than the less concentrated solution. The results remained, however, statistically uncertain. Because it was essential in this study to find out the soil properties which control the effectiveness of various salt solutions as extractants for soil P, the soil sample material was divided into three categories according to the solubility of P in different solutions. The criterion of classification was a difference of 0.1 mg/kg between amounts of P dissolved. The extractability sequence was expressed by the following abbreviations: W (= water), S (= sulphate solution) and Cl (= chloride solution). Thus, in the group termed W > S > Cl, consisting of 53 soil samples, P was most effectively desorbed by water. In 12 soil samples, termed S > Cl > W, water was the most ineffective extractant. In the group of 37 samples, S > W > Cl, the extraction ability of water was poorer than that of sulphate solutions but better than that of chloride solutions. The coarser clay and non-clay soil samples were evenly distributed in the different categories, but there was only one heavy clay soil sample in the groups S > W > Cl and S > Cl > W. Most samples in the group S > Cl > W originated from virgin soils or subsoils. The means (with the confidence limits at the 95 per cent level) of some soil characteristics in these soil classes are listed in Table 2 and the means of P amounts dissolved by various solutions in Table 3. By way of comparison, also the data on water soluble P are presented. The soil classes were equal in means of pH and the differences between the contents of organic carbon were not statistically significant. On the contrary, the soil group S > Cl > W seemed to be poorer in secondary 5 289 290 Table 2. Soil properties in various groups classified according to extractability sequence of P. Soil group W>S>Cl S > W > Cl S>Cl>W clay % 47±6 32±4 32+12 pH 5.2±0.1 5.2+0.2 5.0±0.5 org. C % 4.8±0.7 3.6±0.4 3.1±1.1 NH4F-Pppm») 131±26 95±24 63±42 NaOH-Pppm*) 258±30 181±22 152±79 Oxal.extr. A 1 mmol/kg 70±8 57±7 71 ±23 Oxal.extr. Fe mmol/kg 80±7 63±7 64+14 NH,F-P/A1 (102 ) 7.1± 1.7 5.5+1.3 2.4±0.8 NaOH-P/Fe (102 ) 10.6±1.2 9.6±1.2 7.9±3.8 *) = analyzed by a modified CHANG and JACKSON’s (1957) fractionation method. Table 3. Amounts of P (ppm) dissolved by various solutions from soil groups of different extractability sequence. Extractant Soil group S > W > ClW > S > Cl S > Cl > w h 2o KCI (I = 0.025) KCI (I = 0.1) K2 SO, (I = 0.025) K2 S0 4 (I = 0.1) 15.615.7 9.513.6 10.914.9 8.413.3 9.514.5 7.813.1 13.215.4 10.613.7 12.315.1 10.013.7 0.8±0.4 1.6±0.5 1.6±0.5 2,3±0.8 2.3±0.8 phosphates with a markedly lower molar ratio NH4F-P/A1 than the other groups. In this group the sulphate solutions seemed to desorb about three times and the chloride solutions about twice as much P as did water. In the class W > S > Cl, on the other hand, the chloride solutions dissolved on an average 60—70 % and the suphate solutions about 80—85 % of the water soluble P. The effect of ionic species and ionic strength of the extractant on the desorption of P was studied by the t-test for the means of differences between the paired measurement data (STEEL and TORRIE 1960). As can be seen from the t-values presented in Table 4, the differences between P amounts dissolved in the sulphate and chloride solutions were statistically significant in all soil groups, and so were also the differences between P quantities dissolved by water and all the salt solutions. In addition, exclusive of the soil group S > Cl > W, an increase in ionic strength tended to depress the solubility of P. It should be mentioned that in the group W > S > Cl containing on an average most abundantly water soluble P, the differences between KCI soluble and K2S0 4 soluble P were significantly more marked at the ionic strength of 0.1 than at 0.025. This was due to an increase in the ionic strength reducing more strongly the extraction ability of the chloride solu- tion than that of sulphate solution. 291 Table 4. t-values for the means of differences between quantities of P extracted by various solutions. Difference Soil group S > W > ClW > S > Cl 53 S > Cl > w 1237 KCI fO.O2Jj-K2S0 4 (0.025) KCI (0.1)- K2 S0 4 (0.1) KCI (0.025)- KCI (0.1) K2S0 4 (0.023j—K2 S0 4 (0.1) -8.49'” -7.95'” -6.19'" -4.79“’ -8.37”' -7.20"' -5.71'” -5.24"' —3.99‘" -4.24”' -0.72n ‘ —o.37n ‘ n.s. = not significant In order to get further information about the extraction activity of various salt solutions, the pH of the extracts was measured and the equilib- rium solutions were analyzed for Ca exchanged by K from the soil to the extractant. It was observed that the filtrates of higher ionic strength con- tained more abundantly exchanged Ca, but there were no differences be- tween the Ca concentrations in KCI and K2S04 extracts of the same initial ionic strength. The pH of the K2S0 4 extracts were 0.05 0.40 and 0.00—0.35 pH units higher than that of the KCI extracts at ionic strengths of 0.1 and 0.025, respectively. The differences were statistically significant in all soil groups. An increase in the ionic strength lowered the pH level slightly, the decrease of pH being greater in KCI than in K 2S04 extracts. The relationship between the salt soluble P and soil characteristics was investigated by the correlation analysis. The quantities of P desorbed by both electrolytes were closely related to the amount of water soluble P, the correlation coefficients varying in the groups W > S > Cl and S > W > Cl from r = 0.98 to r = 0.998 ,in the group S>Cl > W from r = 0.74 to r = 0.85 . As anticipated, the quantities of P dissolved by all these extract- ants were correlated with the same soil properties. In the groups W > S > Cl and S > W > Cl, the highest values of correlation coefficients were for the molar ratio NH4F-P/A1 (r = 0.89 —0.92 ’), but in the group S>Cl > W for the ratio NaOH-P/Fe (r = 0.60 —0.77 ). When the effect of NaOH-P/ Fe was eliminated in the first two groups, the values of the partial correlation coefficients for relations between salt soluble P and NH4F-P/A1 decreased only somewhat (r = 0.75 —0.85 ). In the group S > Cl > W, on the contrary, the partial correlation analyses showed the extractability of P not to be related to these molar ratios. In this soil category the salt soluble P rather seemed to be related to NaOH-P (r = 0.59 —0.79 ’), but when the influence of NH 4F-P was eliminated the values of partial correlation coefficients were lowered to r = 0.43n s -0.66\ The other soil properties studied seemed not to correlate with KCI or K 2S04 extractable P in soils. Comparison of the salt solutions showed that the absolute differences between the amounts of P dissolved by KCI and K 2S0 4 tended to be the greater the more water soluble P the soil contained. The correlation coeffi- cients for the relation between that difference and water soluble P ranged r = 0.84 —0.94 and r = 0.58 —0.87 at ionic strengths of 0.1 and 0.025, 292 respectively, the lowest values of r being found in the soil group S > Cl > W. In the soil group W > S > Cl, an increase in ionic strength tended, to some extent, to increase the absolute differences but not in the other groups. It should be mentioned also that the differences between the pH values of the extracts seemed not to be associated with the differences in P quantities dissolved by KCI and K2S04 solutions. Discussion According to the prevalent notion, water is normally able to desorb more P from soils than solutions of neutral salts (e.g. LEHR and WESEMAEL 1952, CLARK and PEECH 1962, STÄHLBERG 1980). In the present material consist- ing of 102 ordinary Finnish soils, however, this was the case in 53 samples only. In fact, the soil samples could be divided into three groups differing in the sequence of extractability of soil P in water and salt solutions. As found earlier e.g. by KURTZ et al. (1946) and STÄHLBERG (1980), also the salt solutions differed in their ability to desorb P: both sulphate solutions extracted P more abundantly than either one of the KCI solutions. KCI and K2 S04 soluble P as well as the absolute differences between P quantities dissolved by these extractants seemed to be closely related to the amounts of water soluble P in soils. Because in this soil material the molar ratio NH4F-P/Al is found to be the first-rate factor controlling the level of water soluble P (HARTIKAINEN 1982), the same factor could be expected to correlate closely also with the amounts of salt soluble P. This suggests that an increase in P coverage on oxide surfaces (a decrease in bonding strength) improves P extractability in electrolyte solutions, too. However, the dis- parities between the soil samples in the sequence of P extractability in various solutions imply the activity and effectiveness of a particular extractant likely to be dependent on soil characteristics. An increase in the ionic strength depressed the desorption of P, but enhanced the exchange of cations to the solution. Equal amounts of Ca2+ were exchanged from the soils by KCI and K2S04 extractants of the same ionic strength, even if KCI solutions contained 1.5 times more K + than the corresponding K2S0 4 solutions. This shows the ionic strength to be a decisive factor in exchange reactions. Secondly, it demonstrates that the electrostatic forces in the extractants were of the same magnitude and, thus, the difference in the desorption ability was caused by the disparity in the nature of the anions. The desorption ability of an anion can be decided from its sorption tendency. The retention of phosphate by hydrated oxides of A 1 and Fe is considered to occur by the so-called ligand exchange. The theory of acids and bases presented by PEARSON (1966) is applied when explaining this reaction in soils (AURA 1980): the central ions Al3+ and Fe3+ are hard Lewis acids preferring to associate with hard bases with low polarizability and high electronegativity (e.g. F - and 02~). The pKa-value of an acid corresponding to the oxyanion shows the ability of anion oxygen to bind a proton (a hard 293 Lewis acid) and also the nucleophilic strength on which the binding of oxygen by the central ion primarily is dependent. Thus, on the basis of the greater basicity, the SO 2' ion can be concluded to have a greater sorption tendency than the Cl~ ion. Also the pH values of the extracts suggest some disparity in the action of the salt solutions. On the basis of Ca 2+ exchange, it can be concluded that equal amounts of H+ ions as well would have been exchanged from the soil by both electrolyte solutions at the same ionic strength. The pH in the K2S0 4 extracts was, however, always somewhat higher than that in KCI extracts and an increase in ionic strength lowered pH more in KCI than in K 2S0 4 extracts. Also CHAO et al. (1965) found similar differences in pH in KCI and K 2S0 4 soil filtrates in spite of the salt solutions compared being of the same normality, not of the same ionic strength. They assumed the differences in pH values to be due to the greater ability of sulphate to replace OH~ ions from the soil. If this occurs, the OH'exchanged does not originate from the oxide surface. According to KINGSTON et al. (1972) and RAJAN (1979), sulphate belongs to anions which are more strongly bound to the oxides than is water, but cannot change the structure of oxides, i.e. they are not firmly bound by oxides above the point of zero charge (pzc). Also the pKa value corresponding to the OH“ ion is so high, as compared to that corresponding to SO2- , that this specific ligand exchange is questionable. P desorption can further be assumed to improve when the concentration or activity of an ion, capable of specific ligand exchange, in the extract increases. Because an increase in ionic strength (i.e. sulphate concentration) depressed the P desorption, a direct ligand exchange with phosphate hardly explains the greater desorptive power of sulphate as compared to that of chloride. On the basis of the facts stated above, some contributory factors improv- ing P desorption seem to be active in the K 2S0 4 extraction. Further investiga- tions are needed, but some theoretical aspects can be presented. The pH in the extracts being a result of reactions occurring in the treatment, it gives reason to suppose that the higher pH in the K2S04 extracts indicates an enhanced H+ adsorption rather than the desorption of OHA This would mean that part of H+ ions exchanged from the soil would have reacted with S04 ~, forming HSOJA If this occurs, it can be expected to have an additional consequence. Provided the quantities of cations exchanged by KCI and K 2S04 solutions are equal, the formation of some HSOj" decreases the ionic strength of sulphate solutions as compared to that of chloride solutions which, in turn, may favour the desorption of P from soil. This process is not possible in the chloride solution. H2S04 is, however, a strong acid likely to be completely dissociated at the pH range prevailing in the soil. Thus, it seems more obvious that the higher pH of K 2S0 4 extracts would result from an enhanced adsorption of H+ ions by HP0 4 “ ions, these being more abundantly dissolved in the sulphate solutions. The greater desorption power of sulphate, on the other hand, may be caused by its ability to replace H2O groups from the surface of oxides (KINGSTON et al. 1972, RAJAN 1979). This exchange reaction can 294 favour P desorption by lowering the ionic strength in the solution or/ and by weakening the bonding strength of P due to the increasing negative charge near the surface. Consequently, it can be concluded that the replacement power of sulphate increases with a decrease in the P coverage and in the soil pH because both these factors contribute to an increased portion of H2O groups on the oxide surface. If this hypothesis presented above is valid, it could explain, at least to some extent, the different extractability sequence in the soils. The P replace- ment power of sulphate tended to increase as compared to that of water and chloride solutions when the content of secondary P as well as the P coverage decreased. Nevertheless, further studies are needed to elucidate the signifi- cance of cation exchange reactions as well as that of the ligand exchange between sulphate ions in a solution and H2O groups on the oxide surface in P desorption. Plants need plenty of P during the early growth season. However, the ionic strength of the soil solution is then highest and the P concentration therefore lowest (WIKLANDER and ANDERSSON 1972, KOENIGS et al. 1981). According to ERIKSSON (1940), yields may even decrease because of the depressing effect of KCI fertilization on the solubility of soil P. If sulphate enhances P concentration also in soil solution, where the solution - soil ratio is much below one, it would be advantageous to give potassium fertilization using sulphate salt instead of chloride salt. The weight and cost of fertilizer containing a certain mass of potassium would, however, increase. Further studies are needed to confirm the advantages or disadvantages of these contradicting points of view. Acknowledgement. The authors wish to thank the Maj and Tor Nessling Foundation for supporting this study financially. References ANON. 1969. Juoma- ja talousveden tutkimusmenetelmät. Elintarviketutkijain Seura. 169 p. Helsinki. AURA, E. 1980. Oxygen as an exchangeable ligand in soil. J. Scient. Agric. Soc. Finl. 52: 34—44. CHANG, S. C. & JACKSON, M. L. 1957. Fractionation of soil phosphorus. Soil Sci. 84: 133-144. CHAO, T. T., HARWARD, M. E. & FANG, S. C. 1965. Exchange reactions between hydroxyl and sulphate in soils. Soil Sci. 99: 104—108. CLARK, J. S. & PEECH, M. 1960. Influence of neutral salts on the phosphate ion concentration in soil solution. Soil Sei. Soc. Amer. Proc. 24: 346—348. DEMOLON, A. & BASTISSE, E. 1934. Contribution ä I'etude de la mecanique chimique des anions dans le sol. Ann. Agronomiques 4; 53 76. ERIKSSON, S. 1940. Über die Einwirkung der Kalidiingung auf die Festlegung der Phosphorsäure im Boden. Ann. Agric. Coll. Sweden 8: 87—130. HARTIKAINEN, H. 1979. Phosphorus and its reactions in terrestrial soils and lake sediments. J. Scient. Agric. Soc. Eini. 51: 537—624. 1982. Water soluble phosphorus in Finnish mineral soils and its dependence on soil properties. J. Scient. Agric. Soc. Eini. 54; 89—98. HARWARD, M. E. & REISENAUER, H. M. 1966. Reactions and movement of inorganic soil sulphur. Soil Sei, 101: 326-335. 295 HINGSTON, F. J., ATKINSON, R. J., POSNER, A. M, & QUIRK, J. P. 1967. Specific adsorption of anions. Nature 215: 1459—1461. POSNER, A. M. & QUIRK, J. P. 1972. Anion adsorption by goethite and gibbsite. 1.The role of the proton in determining adsorption envelopes. J. Soil Sci. 23: 177—192. KOENIGS, F. F. R., LEFFELAAR, P. A., BREIMER, T. & VOLLENBROEK, F. A. 1981. The cation- and anion exchange characteristics of soils with a large sesquioxide surface area. Z. Pflanzencrn. Bodenk. 144: 87—100. KURZ, T., DE TURK, E. E. & BRAY, R. 1946. Phosphate adsorption by Illinois soils. Soil Sci. 61; 111-124. LEHR, J. J. & WESEMAEL, J. C. H. van. 1952. The influence of neutral salts on the solubility of soil phosphate. J. Soil Sci. 3; 125—135. MOORE, W. J. 1974. Physical chemistry. 977 p. sth5 th ed. Thetford. PEARSON, R. G. 1966. Acids and bases. Science 151: 172-177. RAJAN, S. S. S. 1979. Adsorption of selenite, phosphate and sulphate on hydrous alumina. J. Soil Sci. 30; 709-718, RYDEN, J. C. & SYERS, J. K. 1975. Rationalization of ionic strength and cation effects on phosphate sorption by soils. J. Soil Sci. 26; 395—406. STEEL, R. G. D. & TORRIE, J. H. 1960. Principles and procedures of statistics. 481 p. New York. STÄHLBERG, S. 1980. A new extraction method for estimation of plant-available P, K and Mg. Acta Agric. Scand. 30; 93 107. WIKLANDER, L. & ANDERSSON, A. 1974. The composition of the soil solmion as influenced by fertilization and nutrient uptake. Geoderma 11: 157—166. Ms received September 17, 1982 SELOSTUS Kloridi- ja sulfaattiliuokset maan fosforin uuttajina I lonilajin ja ionivahvuuden vaikutus P:n desorptioon Helinä Hartikainen ja Markku Yli-Halla Helsingin yliopiston maanviljelyskemian laitos, 00710 Helsinki 71 Tutkimuksessa selvitettiin liuoksen anionien vaikutusta fosforin desorptioon 102 kiven- näismaanäytteessä. Vertailtavien KCI- ja K 2S0 4 -liuosten ionivahvuus säädettiin yhtä suureksi (0.025 ja 0.1), jotta liuoksissa vaikuttavat elektrostaattiset voimat olisivat yhtä suuret. Fosforin lisäksi uutteista määritettiin Ca:n pitoisuus ja pH. Suolaliuoksiin uuttuneita P:n määriä verrattiin vesiliukoisen P:n määriin. Sulfaattiliuokset olivat aina kloridiliuoksia tehokkaampia P:n uuttajia. Sen sijaan veden sijoittuminen tehokkuusjärjestyksessä vaihteli: 53 näytteessä (ryhmä W > S >CI) se uutti eniten, 12 näytteessä (ryhmä S > Cl >W) vähiten ja 37 näytteessä (ryhmä S > W >CI) se oli heikompi uuttaja kuin sulfaattiliuokset mutta tehokkaampi kuin kloridiliuokset. Uuttumisjär- jestyksen mukaan muodostetut ryhmät eivät poikenneet toisistaan lajitekoostumuksen, pH:n ja orgaanisen aineksen pitoisuuden suhteen. Suolaliuoksiin ja veteen desorboituneet P:n määrät korreloivat voimakkaasti keskenään (r = 0.74 *—0.998 ”). Kloridi- ja sulfaattiliukoisen P:n määrä näytti kytkeytyvän kiinteimmin NH4F-liukoisen P:n (määritetty CHANGin ja JACKSONin menetelmällä) ja oksalaattiuuttoisen aluminiumin suhteeseen (r = 0.60* 0.92 ). K 2S04-liuos vaihtoi maasta yhtä paljon kalsiumia kuin ionivahvuudeltaan vastaava KCI-liuos, mutta sen pH pyrki olemaan jonkin 296 verran korkeampi. Nämä tekijät eivät kuitenkaan näyttäneet selittävän suolaliuosten uuttote- hokkuuden eroja. lonivahvuuden kasvaessa fosfaatin desorptio molempiin suolaliuoksiin väheni, minkä vuoksi suora ligandinvaihto sulfaatin ja fosfaatin välillä näyttää olevan epätodennäköistä. Esitetyn teorian mukaan sulfaatin suurempi uuttokyky perustuu vaihtoreaktioihin oksidipin- tojen HjO-ryhmien kanssa, jolloin liuoksen ionivahvuus pienenee ja P;n desorptio kasvaa. Samaan suuntaan saattaa vaikuttaa myös negatiivisen varauksen kasvaminen oksidipinnan läheisyydessä. Vastaavat reaktiot eivät ole todennäköisiä kloridiliuoksissa. Tulokset viittaavat siihen, että maanesteen P:n konsentraatiota voidaan kohottaa ja kasvien P:n saantia edistää käyttämällä moniravinteisten lannoitteiden K:n lähteenä K2S04:a KCl:n asemesta. Lisäksi ne antavat aihetta olettaa, että syntyminen happamien sulfaattimai- den kalkituksessa voi edistää maan P;n desorptiota pienentämällä maanesteen ionivahvuutta.