CHEMICAL AND PHYSICAL FACTGRS AFFECTING THE RELATIVE Ammmm a? momma PHOSPHORUS m $0st Thesis far the Gegree 0? Ph. D. MiCHIGAN STATE UNIVERSITY Anthony Shiang-Ru Jua 1955 ..‘- THE-£35 L l B R A R Y Michagm ‘fitatc Univcrmty :5- This is to certify that the thesis entitled Chemical and physical factors affecting the relative availability of inorganic phOSphates in soils presented by Tony 8. R. Juo has been accepted towards fulfillment of the requirements for PhoDo degree in 8011 SCienCE /'-‘\ ’- (‘w///4/fl (’/Le(,.’ {7’ Major professor //f/ ’I' "” /"3 7 "I Date A 4/7 Mi»; x’f' 4'61- 0-169 ABSTRACT CHEMICAL AND PHYSICAL FACTORS AFFECTING THE RELATIVE.AVAILABILITY OF INORGANIC PHOSPHORUS IN SOILS by Anthony Shiang-Ru Juo An evaluation of the chemical and physical proper- ties related to the availability of inorganic P compounds to plants was approached from two aspects: (a) examination of the chemical and physical properties of synthetic phosphate compounds, and (b) the distribution of the various forms of inorganic P in soil particle separates. Synthetic colloidal ferric and aluminum phosphates were crystallized in aqueous medium to form variscite, A1P04'2H20 and strengite, FePO4'2H20. The compounds were identified and characterized by XEray and infrared analysis and electron micrographs. Fe phosphate crystallized at a much faster rate than Al phosphate in acidic aqueous medium under the same temperature and pressure conditions. Varis- cite and strengite which were crystallized at 1050C for 40 l and 1.98 ng—l, days gave specific surface areas of 27.5 ng- respectively. The strengite consisted of single spherical crystals with an average diameter of 6 microns and the varis— cite consisted of reniform crystalline aggregates with an Anthony Shiang—Ru Juo average diameter of 1.5 microns. In quartz-sand cultures, with Sudan grass as the indication crop, the relative availa- bility of P from the synthetic compounds was: colloidal ferric phosphate¢§;colloidal aluminum phosphate >>> variscite > strengite. In studying the distribution of soil inorganic P, four Michigan soil profile samples with different textural classes were selected. Inorganic P in soils :was fractionated into Al-P, Fe-P and Ca—P according to the method of Chang and Jackson. Al—P and Fe-P contents are highest in the surface soils studied and decrease with the increase of depth, while the content of Ca-P generally increases with the increase of depth in the soil profiles studied. Percentage distribution curves of Al-P, Fe-P and Ca-P in the soil profiles show re- lationships in supporting the concept which states that the relative abundance of inorganic phosphates in soileay be used as an index for soil chemical weathering. To determine the distribution of inorganic P in soil particle separates, a proposed procedure of dispersion using NaCl saturation and sonic vibration was developed and re- ported in this thesis. Fractionation of inorganic P for each of the particle separates showed that Al-P and Fe-P was most highly concen- trated in the clay fraction throughout the soil profiles Anthony Shiang-Ru Juo studied. Ca-P tends to be highly distributed in both the silt and clay fractions in the surface horizons of the soils and becomes more highly distributed in the silt than in the clay fraction in the lower horizons of the profiles. The contents of all the three forms of inorganic P in sand are the lowest among the three particle separates. During fixation and transformation of inorganic P, Ca-P, which is of primary origin and is present originally in the sand fraction of the soil, is brokendown chemically and physically and enters into the finer fractions of the soil. But, it still remains in high degree of crystallinity throughout the soil textural fractions. The portion of Ca—P that has been dissolved by the action of H+ ions may repre- cipitate as Al-P and/or Fe-P in the clay fraction. Also, the fixed P in soils is mainly in the form of Al-P and Fe—P and is present in the clay fraction. Al-P and Fe—P in soils can be regarded as available forms of P for plants as long as they remain in colloidal forms. It is concluded that the degree of crystallinity is the most important factor in controlling the relative availa— bility of inorganic P to plants. For soils under acidic and aerobic conditions, the relative availability of the native and fixed forms of inorganic P is in the order of Al-P > Fe-P > Ca-P. Approved by CHEMICAL AND PHYSICAL FACTORS AFFECTING THE RELATIVE AVAILABILITY OF INORGANIC PHOSPHORUS IN SOILS BY Anthony Shiang-Ru Juo A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Soil Science 1966 ACKNOWLEDGMENTS To Dr. B. G. Ellis, I wish to express my profound gratitude for his guidance and encouragement. His ever- enthusiastic and patient way in which he gave his advice has been indispensable for the completion of this work. Sincere appreciation is due to Drs. R. L. Cook, M. M. Mortland, E. C. Doll and H. A. Eick, who as members of the guidance committee, offered helpful suggestions during the course of study. In particular, I wish to thank Dr. E. C. Doll for his valuable suggestions and criticisms in re- gard to the presentation of the manuscript. I also wish to acknowledge with thanks the generous financial assistance of Michigan State University. Special gratitude goes to Dr. and Mrs. R. L. Cook. Their life and work have always been a great inspiration to me throughout the years. My debt to my parents is beyond thanks. ii TABLE OF CONTENTS INTRODUCTION . REVIEW OF LITERATURE 1. Studies on the Native and Fixed Forms of Inorganic Phosphorus in Soils 2. Distribution of Inorganic Phosphorus in Soil Profiles and Particle Separates 3. Aluminum and Iron Phosphates as Sources of Phosphorus for Plants EVALUATION ON THE CHEMICAL AND PHYSICAL FACTORS AF- FECTING THE RELATIVE AVAILABILITY OF SYNTHETIC ALUMINUM AND IRON PHOSPHATES . . 1. Preparation and Characterization of Samples xeray Diffraction Infrared Absorption Analysis 2. Particle—size and Specific Surface Electron Microscopic Observations B.E.T. Surface Area 3. Rate of Crystallization 4. Relative Availability 5. Conclusion DISTRIBUTION OF ALUMINUM, IRON AND CALCIUM PHOS- PHATES IN SOME MICHIGAN SOIL PROFILES AND THEIR PARTICLE SEPARATES . . . . . . . . . . 1. Description of Soil Samples 2. Distribution of Al-P, Fe-P and Ca-P in Soil Profiles 3. A Proposed Procedure for the Separation of Soil Particle—size Fractions for Nutrient Analysis Procedure Efficiency of the Procedure iii Page 15 18 24 24 25 25 31 31 33 36 4O 45 48 48 50 59 6O 63 Page 4. Distribution of Al-P, Fe-P and Ca-P in Soil Particle Separates 65 Percentage Distribution of Al-P, Fe-P and Ca-P in Soil Particle Separates 69 Particle-size Distribution of Al-P, Fe-P and Ca-P in Soils 76 GENERAL DISCUSSION AND SUMMARY . . . . . . . . . . . . 84 BIBLIOGRAPHY . . . . . . . . . . . . . . . . . . . . . 90 iv Table 6a 6b 10 ll 12 LIST OF TABLES Xkray powder diffraction data for synthetic Al and Fe phosphates crystals . . . B.E.T. surface area of synthetic phosphate compounds . . . . . . . . . . . . . Yield and P uptake by Sudan grass after 30 days growth . . . . Total P and inorganic P fractions of soil profile samples Percentage distribution of Al, Fe and Ca-P in soil profiles Comparison of the proposed sonic and the calgon methods of dispersion Change in pH of NaCl-soil suspension before and after dialysis against distilled water Particle-size composition of soil profile samples based on large—scale separation A1-, Fe- and Ca—P contents in soil particle- size fractions . Percentage distribution of inorganic P in sand, silt and clay fractions Particle-size distribution of Al-P in soils Particle-size distribution of Fe-P in soils Particle-size distribution of Ca-P in soils Page 26 35 43 52 53 64 64 66 68 7O 78 79 81 Figure LIST OF FIGURES Infrared absorption spectra of synthetic A1 phosphates: (A) colloidal A1P04, (B) crystalline A1 phosphate (variscite, A1P04°2H20), (C) crystalline A1 phosphate heated at 65°C. for 12 hours. *Bands due to vaseline . . . . . . . . . . . . Infrared absorption spectra of synthetic .Fe phosphate: (a) colloidal FePO4, (b) crystalline Fe phosphate (strengite, FePO4'2H20), (c) crystalline Fe phosphate heated at 65°C for 12 hours, (d) crystal- ‘ line Fe phosphate heated at 130°C for 12 hours. *Bands due to vaseline . . Electron micrographs of synthetic Al and Fe phosphates: Plate 1, newly precipitated colloidal AlPO4, Cr shadowed; Plate 2, col- loidal AlP04 digested at 105°C for 40 days to form variscite, parlodion film; Plate 3, newly precipitated colloidal FeP04, Cr shadowed; Plate 4, colloidal FePO4 digested at 105°C for 40 days to form strengite, parlodion film . . . . . . . . . . . . . Electron micrographs of synthetic strengite: Plate 1, Cr shadowed strengite; Plate 2, parlodion film supported strengite Xkray diffraction patterns of A1 phosphates: (A) newly precipitated colloidal A1P04, (B) colloidal A1P04 digested at 105°C for 72 hours, (C) colloidal A1P04 digested at 105°C for 40 days. *Peaks due to quartz in the ceramic plates . . . . . . . . . . X+ray diffraction patterns of Fe phosphates: (a) newly precipitated colloidal FePO4, (b) colloidal FeP04 digested at 105°C for 12 hours, (c) colloidal FePO4 digested at 105°C for 40 days . . . . . . . . . . . . vi Page 28 3O 32 34 37 39 Figure 10. ll. 12. 13. One month old Sudan grass seedlings with P applied as different P minerals Transformation of soluble P into less avail- able forms Percentage distribution of Al-P, Fe-P and Ca-P in profiles of (A) Pewamo clay loam, (B) Karlin sandy loam . . . Percentage distribution of Al- P, Fe-P and Ca-P in profiles of (A) Onaway loam, (B) Sims clay loam . . . . . . Percentage distribution of Al-P, Fe-P and Ca-P in particle separates of Pewamo pro- file: (a) sand fraction, (b) silt fraction and (c) clay fraction . . . . . Percentage distribution of Al—P, Fe-P and Ca-P in particle separates of Karlin pro- file: (a) sand fraction, (b) silt fraction and (c) clay fraction . . . . . Percentage distribution of Al-P, Fe—P and Ca-P in particle separates of Onaway pro- file: (a) sand fraction, (b) silt fraction and (c) clay fraction . vii Page 41 47 55 56 71 73 75 INTRODUCTION Inorganic phosphorus in soils is believed to exist as sparingly soluble orthophosphates of Al, Fe and Ca. A1 and Fe phosphates are the dominant forms of inorganic P in acid soils while Ca phosphates are most abundant in calcare— ous soils. Ca phosphate compounds, such as apatites, are the primary Pbbearing minerals in soils which are further transformed to A1 phosphate and Fe phosphate during the course of weathering. It is also well known that soluble P fertilizers react rapidly in soils to form insoluble phos- phates bound mainly to A1 and Fe. The reaction is greatly affected by the pH of the system and the sources of cations. Since both the native and fixed forms of inorganic P in soils are present as insoluble compounds, an understand- ing of the factors affecting the availability of the various forms of inorganic P to plants is of great importance. To consider solubility alone would be unsatisfactory. Chemical and physical properties such as particle size, specific sur— face and degree of crystallinity have been brought to at— tention. However, a clear picture has not yet been well established. The objectives of this investigation are: 1. To evaluate the chemical and physical properties such as particle size, specific surface, rate of crystalli- zation and degree of crystallinity of synthetic A1 and Fe phosphates as related to the relative availability of these phosphate compounds to plants. 2. To determine the distribution of various forms of inorganic P in soil particle separates of some Michigan soil profiles. REVIEW OF LITERATURE 1. Studies on Native and Fixed Forms of Inorganic Phosphorus in Soils Soils usually contain from 0.04 to 0.11 percent of P. Inorganic P is the preponderant form of P in mineral soils and exists almost entirely as salts of orthophosphoric acid, among which, the phosphates of Ca, Fe and A1 are be— lieved to be the most predominating forms (60,113). Studies prior to 1953 on compounds formed during soil genesis and as a result of P fertilization in both acid and calcareous soils were reviewed by Olsen (109) and Kurtz (81). The inorganic P content in agricultural soils is so low that attempts to study the nature of soil P compounds are beset with many difficulties. Three different ap- proaches have been commonly employed to study the inorganic P in soils: (a) direct observations by means of Xkray dif- fraction, electron microscope, infrared analysis and/or differential thermal analysis; (b) the application of solu— bility product concepts; and (c) extraction with different acid, alkaline and complexing solutions. by VK at “ t C0 VaI fag kt (a) Direct Observations Several complex Fe and A1 phosphates such as palmer- ite, taranakite and minyulite were identified by Xkray and optical methods from clay minerals, gibbsite, and geothite digested with P solutions by Haseman,_gt_§1. (57,58). By means of electron microscope, Kittrick and Jackson (77) were able to observe the progressive formation of A1 phosphate crystals concurrent with the decomposition of the original kaolinite crystals. They also found that the phosphate treatment diminished the kaolinite Xkray diffraction pattern and that various A1 phosphates were the end products of P in— duced kaolinite decomposition. Based on the results of a sequence of similar studies, Kittrick and Jackson (78,79) at- tempted to arrive at a unified theory of P fixation in soils and the resulting theory was merely an explanation on the basis of precipitation reactions. Beaton, §£_§1. (14) by infra-red absorption analysis, found dicalcium phosphate di- hydrate (CaHPO °2H20) and hydroxyapatite [Ca10(OH)2(PO 4 4’2] to be the soil—fertilizer reaction products in a calcareous soil. With X-ray and petrographic analysis, Lindsay,_g§_al- (93) identified about 30 crystalline P compounds of variable composition as reaction products following the addition of various fertilizer solutions to soils and soil constituents. Due to the complexity of the soil system and the fact that most of the inorganic P compounds occur character— istically in the clay fraction and are difficult, if not impossible to separate by physical methods, the applica- bility of the direct means has been very limited. Further— more, the great preponderance of nonphosphatic material serves to mask any measurable properties of the P compounds found by the various physical methods. The identification of the native, inorganic P com- pounds in soils by means of the above mentioned physical techniques has not yet been reported. (b) Phosphate Equilibria in Soils Stelly and Pierre (137) compared the solubility versus the pH curves of soils with those of known P—bearing minerals and found that alkaline soils usually display a solubility-pH curve similar to that of apatite and acid soils usually display a solubility-pH curve similar to that of Fe-P and/or Al—P. By applying the solubility product principles, Aslying (5) plotted the phosphate potential, (8pCa + pH2P04), against lime potential, (pH - 8pCa) of soils and the solu— bility of calcium phosphates. He found the presence or for- mation of octocalcium phosphate in some calcareous soils to which a large amount of superphosphate had been added. From this he concluded that dicalcium or octocalcium phosphates are formed when Ca phosphates are precipitated under con- dition similar to those existing in soils. But neither is stable-~both are converted to a more basic calcium phosphate, presumably, hydroxyapatite, although the rate of formation of hydroxyapatite in soil may be extremely slow. Withee and Ellis (49) equilibrated two calcareous soils with 0.01 M CaCl and found that the P minerals that 2 controlled the level of P in solution were basic calcium phosphates similar in solubility to octocalcium phosphate, or in other words, the equilibriated SOlutions were super— saturated with respect to hydroxyapatite. ,Rather large' additions of soluble P were necessary to saturate the soils with respect to CaHPO4. Soil P solubility determinations by Clark and Peech (38) showed that neither CaHPO nor hydroxyapatite was 4 present in the soils or clay system studied. The solubility data of Al-P obtained in their experiment indicated the presence of variscite, A1P04'2H20 in the soils studied. In a later study, Lindsay,_§t_§1. (87) treated acid soils with Ca(H2PO and found that after 18 months of aging the 4),. soils yielded extracts that were nearly in equilibrium with variscite, indicating that variscite was the final reaction product and was governing the P concentration in these acid soils. In a more detailed study on Al-P ion products in solution equilibrated with an acid, P—deficient soil, Taylor and Gurney (141) showed that the P status of the undisturbed soil is compatible with the existence of variscite in the soil. When soil is acidified, subsequent changes in compo- sition of the solution reflect the dissolution of A1(OH)3 In U1 Si Ma Al: 1. and decay of clay minerals. Phosphate additions to the acidi- fied soil are rapidly precipitated, and the final P concen— tration is less than would be supported by variscite under more acid conditions. Chakravarti and Talibudeen (24) examined P equilibria in 54 acid soils from Britain and India and found that the equilibrium concentration in British soils was governed by P residues less basic than variscite and strengite. In Indian soils, both variscite and strengite are effective in con- trolling phosphate concentration over the pH range of 3.8- 5.7. Wada (147) found that P equilibria in some ando soils, red-yellow podzolic soils and alluvial soils in Japan were governed by Fe and Al compounds from pH 4 to 7. Below pH 5.2 P activity was apparently controlled by variscite and gibbsite or possibly with strengite and Fe(OH)3. From pH 5.2 to 7.0, these compounds still persisted but alkaline hydrolysis occurred at the surface, resulting in occlusion of the P in Fe and Al oxides. This effect, together with uncertainty in estimation of unhydrolyzed metal cations, makes the use of solubility product principles difficult. Similar studies were conducted by Hernando_gt_al. (61) and Mare (101). Based upon known solubility data of the various Ca, Al, and Fe phosphate compounds that may form in soils and of other inorganic phases of the soil that furnish ions participating in soil P reactions, Lindsay and Moreno (92) presented the activity isotherms for AlPO4'2H20 (variscite), FePO4-2H20 (strengite), Ca10(PO4)6F2 (fluorapatite), CalO(PO4)6(OH)2 (hydroxypaptite), Ca4H(PO '3H20 (octo— 4)3 calcium phosphate), and CaHPO4'2H20 (dicalcium phosphate di— hydrate) on a single solubility diagram in which a function of P activity in solution was plotted against pH. This generalized plot is considered to be useful for assessing relative stabilities of these P compounds and for predicting their transformations in soils upon the application of ferti- lizer or lime. In a sequence of studies on the nature of the re- actions of Ca(H2PO4)2'H20 in soils, Lindsay, §£_§l. (88,89, 90,91) also found the formation of complex P compounds simi— lar to those reported by Haseman (57,58). But they pointed out that these complex phosphates are unstable in soil sus— pension where the equilibrium activities of component ions are much lower than those in equilibrium with the complex phosphates. They also stated that in arable soils, especial- ly when leached in the presence of growing plants, the com— plex phosphates very soon revert to phosphates whose equi- librium ion activities are of the same order as those for the variscite—strengite group. Wild (148) on the basis of solubility product princi- ple, calculated the P concentrations to be expected in the presence of variscite and various calcium phosphates at different pH values and compared his results with the P concentrations in several soil extracts as by other workers. In the pH range 4.5 to 6.0 none of the phosphates considered could explain the level of P found in solution. Solubility product concept has been widely used by many workers in recent years in an attempt to identify the inorganic P compounds in soils. But, due to the complex nature of the soil and the limitations of the solubility pro— duct principle itself in describing precipitation and dis— solution phenomena, the results obtained by many workers may show the possibility of the existence of certain phosphate compounds in soils, but do not show explicately which form of the phosphates will control the concentration in the soil solution-solid system. Taylor and Gurney (141) pointed out that there are several reactions that may reduce the P concentration to the desired level in the soil solution; therefore, the finding of Al-P ion products similar to the solubility product of variscite is not a satisfactory criterion for the existence of the mineral in soils. Larsen and Court (83) examined the solubility data of a great number of British soils with or without P appli- cations with respect to the solubilities of calcium phos- phates assumed to form in soils. A scattered distribution of the points on the solubility diagram was Obtained. Although some of the points fall near the lines for a pure compounds, 10 the over-all distribution suggested that either no definite forms of calcium phosphates can be inferred to exist in soils or that the apparent solubility relationships are being in— fluenced by other factors. Measurements of the chemical potential of P in soil suspensions and difficulties arising in applying solubility criteria for determining the nature of P compounds in soils have been investigated by many workers (15,47,82,83,84,85, 117,152,153). However, the application of these criteria to experimental determinations of soil P solubility has not been very successful. The results for some soils obey the relationships for pure compounds while others do not, and the proposition that these can be attributed to substances of intermediate composition is not wholly satisfactory (83). (c) Chemical Extractions Chang and Jackson (25) developed a soil P fraction— ation procedure by using different extraction solutions which enables fractionation of the soil inorganic P into discrete chemical forms, namely Al-P, Fe-P and Ca-P. The method has greatly stimulated interest in study fixation, transformation, and availability of the inorganic P in soils. By using this procedure, they fractionated a number of soils and concluded that the distribution of soil inorganic P was a measurement of the degree of chemical weathering, the ll weathering sequence being Ca-P, Al-P, Fe-P and occluded P (27). Chu and Chang (36) analyzed a great number of acid and calcareous soils of Taiwan and found that there are three different phosphate distribution patterns in these soils, namely, Fe-P dominating, Ca-P dominating, and Fe-P and Ca-P dominating. Al—P was not found to be a dominating form in any of the paddy soils studied. In fractionating P from soils representing different stages of maturity in India. Geol and Agarwal (52) found that total P and organic P de- creased with soil maturity. Inorganic P tends to be bound to Fe and Al in mature soils which are responsive to P ferti- lizers, and in the form of Ca-P in immature soils which do not respond to P application. Kaila (74) fractionated 363 mineral soil samples from Finland and showed that sandy soils were richer in NH4F-soluble P (Al-P) and poorer in acid soluble P (Ca-P) than the other soils were, but alkali- soluble P (Fe-P) was independent in soil texture. The pre- dominance of acid-soluble P in the organic fraction, together with low amounts of reductant soluble P, pointed to a low de— gree of weathering of the soils studied. The high contents of the NH F and alkali—soluble P in the surface sample of 4 cultivated soils as compared with samples from virgin soils or subsoils may be due to application of P fertilizer and a higher degree of weathering. They also pointed out that variations in the NH4F and alkali-soluble P fractions were n1 12 associated with variations in active Fe and Al rather than with a direct effect of soil pH. Serbanescu and Blanaru (130) found that degraded podzols, podzolized brown soils and acid brown mountain soils were low in mobile Al phosphates and brown forest soils were dominated by Ca phosphates. Sen Gupta and Cornfield (128,129) studied P in cal— careous soils and found that the amounts of different in— organic P fractions decreased in the order, "inert P," apa- tites, non-apatite, Ca phosphates, A1 phosphates, Fe phosphates and easily replaceable P. None of the P forms significantly correlated with CaCO3 content in the soils studied. It is generally known that Al and Fe phosphates are abundant in acid soils and Ca phosphate in alkaline or calcareous soils (36,32,23,41,81,60,109). Numerous works have been published on the fate of water-soluble P added to soils and transformations of the various forms of inorganic P in soils. Cecconi (22) stated that fixation of P from P solution in contact with non- calcareous soils is mainly due to the Fe and A1 present. Yuan §t_gl. (155) found that in three acid soils, over 80 percent of the added P was retained by the soils as A1 and Fe phosphates. Prolonged alternate wetting and drying reduced the percentage of P in Al-P form, and increased the percent— age distribution in Fe-P form. This transformation from Al—P to Fe-P may be due to the lower solubility product of the latter. 13 In working with a number of acid and calcareous soils from Taiwan, Chang and Chu (30) found that upon addition of soluble P to the soils, fixation was largely as Al-P with less as Fe-P and Ca-P after three days incubation at field capacity. During 100 days incubation under the same con— dition, the amount of Fe-P increased at the expense of Al—P and Ca-P. Under flooded condition for 100 days, Fe-P became the dominant form of P fixed in all soils studied. Similar results have been reported by Volk and McLean (146), and Kaila (75). Chai and Caldwell (23) reported that the P— fixing capacity of soils increased with departure from a pH near 7.0. Mackenzie and Amer (97) studying acid soils found that the Al—P fraction increased at both high and low rate of P applications, Fe-P increased only at the higher rate of P addition and Ca-P did not increase at either'.rate. .In the alkali soils tested, Ca-P increased at both rates of added P. After removing Fe by a biological—reduction method which did not remove Al, Bromfield (20) found that the P—soprtion of three acid soils was greatly reduced. But when Al was also removed the P—sorption capacity' further decreased. Forms of inorganic P in soils after long-term con— tinued fertilization have been investigated by many workers. Manning and Salomon (99) found that application from 65 years of superphosphate to a brown podzolic soil resulted in a large accumulation of Al-P and less accumulation of Fe—P. Application of rock phosphate increased Ca—P fraction. (n l4 Occluded and organic P fractions were not greatly affected by various levels or sources of P fertilizers. In paddy soils, it was shown that the superphosphate applied to a soil over 31 years is mostly retained in the form of Fe-P, with Al-P next, and that least is retained as Ca-P (30). Liming did not significantly change this distribution pattern. With Loess soils, Schachtschabel and Heinemann (122) found that the Ca-P fraction increased slightly up to pH 6.5, rising sharply as soon as CaCO3 occurred in the soil. Fe—P fraction decreased with increasing pH. Neither Al-P nor oc- cluded P correlated with pH in the soils studied. In studying a number of acid soils, Wright and Peech (154) pointed out that the results from fractionation would support the conclusion obtained from the solubility studies that some crystalline phosphate minerals of the variscite- strengite isomorphous series was the ultimate reaction pro— duct of applied P in acid soils studied. The fractionation method of Chang and Jackson has been criticized especially with regard to the differentiation between Fe—P and Al-P (46). Also, more soluble Ca phosphates such as CaHPO4-2H20 or octocalcium phosphate may exist in soils under heavy P fertilization. These P compounds are readily attacked by NH F (88,89,90,9l,104). It was also 4 pointed out that freshly formed phosphates may have different solubility from aged crystalline compounds (86,155). 15 Chang (29) summarized all the criticisms and modifi— cations and systematized the sequence of extraction in the most up—to-date form. The fractionation procedure may not be as precise as desired. But there is no doubt, even with its shortcomings, the use of this method has broadened our knowledge of the P chemistry in soils. 2. Distribution of Inorganic Phosphorus in Soil Profiles and Particle-size Fractions Winters and Simonson (151) reported that in soil profiles developed on uniform parent materials, the minimum P percentage usually occurs in the lower A or upper B horizon. This apparently results from the combined action of P absorption by plants and leaching. Bauwin and Tyner (13) found the profile distribution of "extractable P" (organic P + acid soluble P + "adsorbed P") in some gray- borwn podzols, brunizems, and planozols to be C horizon < A horizon < B horizon. Conversely, the relative abundance of the nonextractable P was B > A > C. A significant increase in the non-extractable P content of B horizons was found to occur with increasing soil maturity. By dispersing the soil mechanically in water, Williams and Saunder (150) studied the distribution of organic P and inorganic P in particle-size fractions of some Scottish soils and reported that more than 85 percent of the organic P 16 was predominating in the fine sand fractions and appeared to be largely Ca-bound. Hoyos and Garcia (64,65) found that total P in the soil particle-size fractions increased as particle-size decreased, the largest difference between silt and clay fraction. Organic P was highest in clay fraction. Most forms of P decreased with depth in all fractions. Bates_gt_§1. (12) found that in a Nigerian forest soil, in- Organic P was fairly constant down the profile. Only the surface soil contained appreciable amounts of P soluble in NH4F. Considerable amounts of P were extractable from all horizons by 0.1 N NaOH. The clay fraction of the topsoil contained a large amount of P soluble in NaOH. Goel and Agarwal (53) found that total and organic P contents de- creased with the maturity of some Indian soils and the highest P concentrations were found in the clay fraction and the lowest in the sand fractions. By using a microscopic technique, Shipp and Matelshi (133) showed that in some Nebraska soil profiles, the content of apatite increased with depth. The various forms of inorganic P contents in a soil profile were first shown by Chang and Jackson (27) to be a measure of the degree of soil chemical weathering. Their concept is that as weathering increases there is a shift in the relative abundance of the inorganic phosphates from Ca-P toward Al-P, Fe-P and occluded P. Based on this concept, 17 Hawkins and Kunze (59) found that in nine profiles of Grumusols from Texas, the distribution percentages for Ca—P, Al-P and Fe-P served as sensitive indicators of the weather- ing environment. Kaila (73) reported that podzolized virgin soils were low in all forms of inorganic P in the upper layers, especially in the AZ-horizon while their enrichment layers contained fairly high amounts of P bound to Fe and Al. The largest part of inorganic P at depth below 30-40 cm. was bound to Ca in all soils studied, except in the cultivated soil in which alkali soluble P bound to Fe predominates at all depths. Recently, the distribution of inorganic P in soil particle-size fractions has also been studied. Scheffer_gt_al. (125) found that in the mechanically sepa- rated soil particle-size fractions, the proportion of Ca-P markedly increased with increasing particle—size and reached maximum values in the sand fraction, while a reverse, though less marked, trend in respect of Fe—P and Al-P was observed. Henley and Murphy (56) analyzed the contents of Al-P, Fe-P, Ca—P, organic P and total P in sand, silt and clay fractions of 24 Irish surface soil samples and suggested that the Al-P and Fe—P in the clay fraction may be an important controlling factor in the relative availability of inorganic P to plants. 18 3. Aluminum and Iron Phosphates as Sources of Phosphorus for Plants Having assessed the role of Fe-P and Al-P in soils, it now remains to be seen how satisfactorily these compounds are in supplying P for plant growth. A general, but thorough review on Al-P and Fe-P in soils has been published by Smith (136). (a) The solubility of Aluminum and Iron Phosphates The solubility of Al and Fe~ phosphates is important to an understanding of the factors that control the availa- bility of P fertilizer in soils. Huffman 2E_§l- (69), taking strengite as unity, found that the relative rate of disso- lution in water of colloidal FePO4 was 30. It was postulated that the great chemical similarity between strengite and colloidal phosphate was masked by the larger particle size and consequently slower rate of solution of strengite. In a latter paper, Huffman and Taylor (70), reviewing the sub— ject of the behavior of water—soluble P in soils, gave the relative initial rates of solution of phosphates in water in terms of weight of P per unit area per unit time. Strengite was again taken as unity, colloidal FePO colloidal AlPO 4) 4) and variscite had the values of 1.3-2.4, 3-10, and 10—20 re- spectively. Fujiwara (50) found that the solubility of the crystalline form of Al-P and Fe-P is far less than that of their amorphous forms. 19 Solubility products of the various synthetic Al-P and Fe-P have been reported by many workers (26,88,45,39,78, 7, 143). The reported szp values are 22.5 for variscite, AlPO '2H20 (143) and 35.0 for strengite based on the 4 formula Fe(OH)2H2PO4 (26). The value becomes 29.0 if the calculation is based on the correct formula FePO4-2H20. (b) Availability of Synthetic Aluminum and Iron Phosphates to Plants Some green-house experiments have shown that a sub- stantial proportion of the P required by plants canlae sup— plied by pure Fe-P and/or Al-P (139,144). But on the other hand, synthetic Al-P and Fe-P have also been reported to be relatively poor sources of P for plants (42,121). Fujiwara (50) found that synthetic Al-P and Fe-P were equally avail- able to paddy rice, but Al-P was a better source of P than Fe-P for upland crops, such as barley. Comparison of the availabilities of the synthetic Al-P and Fe—P have also been investigated by Bartholomew §t_§l. (10), Marais (100), Sannikova (119), Taylor, §£_2l° (139), Lindsay and DeMent (94), Taylor, §E_2i- (142), and Pirkl.gt_al. (112). Colloidal AlPO and FePO Ca-Fe phosphate, and K 4) 4) and NH4 taranakites, all of which are believed to be formed by the action of acid solutions of fertilizers upon soil, were found to be relatively good sources of P for plants. However, variscite and strengite were almost completely un- available to plants in acid soils (139,94,112). 20 Taylor,_et_§l. (142) pointed out that the availa— bility of the applied amorphous Al-P appeared to be con— trolled by its rate of hydrolysis rather than its equilibrium solubility. In calcareous soils, amorphous AlPO4 and varis- cite were about as good sources of P for corn as Ca(H2PO4)2. The particle size of the Al—P seemed to be the dominant factor in their availability. Huffman,_gt_§l. (69) found that the uptake of P by the plant from strengite and colloi— dal Fe-P was closely parallel to the rate of dissolution of these compounds in water. (c) Availability of Soil Aluminum and Iron Bound Phosphates to Plants Fried and Shapiro (48,49) evaluated the availability of native soil phosphorus on two factors: the amount and concentration of P in the soil solution (intensity factor) and the ability to renew the P in soil solution (capacity factor). Chemical tests haVe been widely used to measure the "available" soil P. Chang and Juo (32) studied the relation— ships between available P determined by the various con- ventional chemical tests and the forms of inorganic P in soils and found Al-P and Fe-P to be the main sources of available P in the lowland, rice soils in Taiwan. Similar studies have been done by Susuki, §t_al. (138) on some Michigan soils. They found that the available phosphorus was highly correlated with the Al phosphate fraction in soils. 21 In pot experiments, significant correlations have been found between A-value P and Fe-P in paddy soils (28,33). Payne and Hanna (111) showed that the P content in millet tops was chiefly correlated with the Al-P fraction of the three acid soils studied. Measurements in the uptake of P by plants from the various forms of inorganic P in soils have been studied. Hanley (55) grew Perennial ryegrass, white-clover and Agrostis tenuis on six soils derived from calcareous and non— calcareous parent material. The inorganic P was separated into six fractions after cropping. The Al—P extracted by NH4F was preferred by crops, but Fe—P was important in some soils. Calcium phosphate contributed little and reductant- soluble P and occluded P were not available at all. In similar studies, Melton (103) found that the change in Al-P during cropping was highly correlated with the initial Al-P content of the soils studied. Smith (135) found that on an acid red soil, Al-P was the preferred source of P for wheat. Alban, gt_al. (2) showed that the up- take of P during cropping period was significantly corre- lated with the decrease of NH4F-soluble P in the soil. In a paddy soil studied by Basak and Bhattacharya (11), it contains 847 #/A of Fe—P and Al-P which represented 47 percent of the total P at planting time in the unmanured soil. It decreased gradually to 424 #/A at post harvest time, and then increased to 521 #/A by the planting time of 22 next season. Scheffer,.gt_§l. (126) found that the uptake of total P'by plants was high from < Z‘p and 2-6’p fractions, and decreased with increasing particle size, only to increase again from the sand fraction. The uptake Of Fe-P and Al-P was highest from the < 2 p and 2-6’p fractions. (d) Factors Controlling the Availability of Iron and Aluminum Phosphates Degree of crystallinity and specific surfaces have been considered as important factors controlling the availa- bility of the applied and native P compounds in soils. Al- though many of these facts have been mentioned in previous sections, a systematic and conclusive picture has not yet been established. Cecconi (22) pointed out that newly formed phosphates are generally amorphous and possess a high ex— change capacity-—the latter, together with the solubility, progressively decreasing with the loss of amorphous charac- teristics during aging. Chiang, §t_§l. (33) suggested that under water logged conditions in soils, strengite was first changed to more soluble ferrous phosphate and the P ions were then reprecipitated by active sesquioxide to form amor- phous ferric phosphate which is a source of available P for plants. The effect of microbial activities on P mobilization in.soil has been investigated (l8,l9,42,63,98,131,l32)- Changes in redox potential and pH in rice paddy soils caused by microbial activities have been observed. There 23 was a marked increase in pH, available P, and Fe and a de- crease in Eh as the time of soil kept at flooding conditions increased. DiffiCultly soluble Al-P and Fe-P became more soluble by being hydrolyzed or reduced (34,35,120). Bromfield (19) found that Fe-PO4 became more soluble when incubated under water in the presence of air or N2, while AlPO4 was not dissolved under the same conditions. Ojima and Kawaguchi (107) reported that ferric and basic ferric phosphates dissolved completely in water after saturation with H28 and subsequent aeration, and Al-P was not affected. Liming an acid soil generally increases the avail- able P. This may be due to the hydrolysis of Al-P and Fe-P, increase in microbial activities, and reducing the activity of the A13+ in the soil solution as a result of liming (87, 98,115). Effect of P uptake related to plant species has been studied (106). It was also found that air drying generally increased the easily soluble P content of non-calcareous soils (6). Mack and Barber (95,96) found that soil incu— bated at —25°C for nine months released more P when leached with water than soil incubated at 2.70C. They suggested that the preconditioning temperature changes either the types of P compounds or the surface area of those present in the soil. EVALUATION ON THE CHEMICAL AND PHYSICAL FACTORS AFFECTING THE RELATIVE AVAILA- BILITY OF SYNTHETIC ALUMINUM AND IRON PHOSPHATES 1. Preparation and Characterization of Samples Pure colloidal AlPO land variscite were pre- 4 pared according to the method described by Deming and Cate (41). Colloidal AlPO4 was prepared by reacting 16 g. of Al pellets in 240 ml of 50 percent H3PO4 while heating on a water bath. Crystalline AlPO4 was prepared by digesting the colloidal material in one liter of distilled water at 105°C in an autoclave. The pH of the suspension was adjusted daily to 3.2 (measured at 25°C) by adding 0.02 N H3PO4 or distilled water. After 40 days of digestion, the completion of crystallization was observed under a microscope. The crystallized material was then air dried at room temperature. Colloidal FePO was synthesized by the same procedure as for 4 the colloidal AlPO using iron powder and, 35 percent H202 4 was added to complete the oxidation of the system to ferric ion. Crystallization of FePO4 was carried out at the same temperature, pressure and pH conditions and the same di- gestion period as for AlPO4. The synthetic phosphate 24 25 materials were characterized by infrared absorption analysis, Xkray diffraction and electron microsc0py. xeray Diffraction Xkray diffraction of the compounds was conducted using the powder technique. A thin layer of the sample was deposited on a porous ceramic plate, which was rotated with respect to an Xkray beam produced by copper ratiation. The reflections were recorded with a scanningwgoniometerylitilizing a Geiger—Muller counter tube in conjugation with a scale- ratio meter with an automatic recorder. A Philips electronic powder diffraction camera‘was also employed in other studies. The interplaner spacings were calculated using a wave length of 1.5418 2 (Cu K—oKaverage). The radiation was nickle filtered. The colloidal AlPO4 and FePO4 were amorphous to X— ray diffraction. The crystalline Al and Fe phosphates pre- pared under the same environmental conditions were identi- fied by thay powder diffraction to be variscite, AlPO4-2H20, and strengite, FePO4'2H20, respectively (Table 1). Infrared Absorption Analysis For infrared analysis, samples were examined as fine powders mulled in vaseline between salt windows and recorded on a Beckman IRr7 Spectrometer. The films were mounted at right angles to the infrared beam and scanned from 600 to 4000 -1 cm . 26 Table 1. Xkray powder diffraction data for synthetic Al and Fe phosphates crystals. 1 ——‘ r Syn. A1 phosphate Syn. Fe phosphate (Variscite, AlPO4-2H20) (Strengite, FePO4-2H20) d (3) 1* d (3) 1* 5.40 vs 5.43 s 4.85 s 4.96 w 4.31 vs 4.91 m 3.91 s 4.36 vs 3.64 w 3.97 m 3.27 m 3.91 w 3.08 vs 3.64 w 2.93 ms 3.27 w 2.80 ms 3.11 s 2.64 m 3.00 m 2.47 m 2.92 m 2.54 w 2.53 m 2.44 w *Intensity: vs = very strong, 5 = strong, ms = medium strong, m = medium, w = weak. 27 The infrared absorption spectra of the colloidal and crystalline A1 phosphates within the frequency range of 600 [to 4000 cm.1 are given in Figure l. The crystalline Al phosphate was identified as variscite by comparing its spectra with the reported I.R. spectra for natural variscite (4,40). According to Corbridge and Lowe (40), the strong ab- sorption between 1000 and 1200 cm"1 in the variscite spectra represents the ionic P90 stretching band of neutral ortho- l in the phosphate. The two strong bands at 1050 and 1075 cm’ PLO stretching region are characteristic Of variscite. The medium bands at 935 and 1160 cm—1 represent other modes of p-o stretching. The broad band near 1600 cm"1 is due to OH bending of the hydrated water and the absorption at 3580 cm.-1 is due to the OH stretching of hydrates. This sharp band at 3580 cm_1 indicates that the water of crystallization of variscite is probably loosely held through hydrogen bonding with the oxygen of the phosphate anion in the structure. After heating the variscite sample at 65°C for 12 hours, the hydrated water bands of variscite at 1600 and 3580 cm-1 disappeared and the anhydrous form of the phosphate was obtained (Figure l—C). The spectra also shows that the P—O stretching band shifted to higher frequencies upon dehydra- tion. The PhO stretching bands occurred at lower frequencies in variscite as compared to its dehydrated form or the colloidal form. This is because of hydrogen bonding of the 28 .ocflaommkr on mop mos—mm... .muson NH How .OOmo on wound: oumnmmonm Hg ocflaamumwuo AOV .AONmN.¢OmH¢ .ouflomflum>v oumnawonm ad ocflaamummuo Amv .sOmH¢ Hmcfloaaoo Amv ”monogamonm Hg UHuonucmm mo muuoomm COHDQHOQO commumcH .H ousmflm 000 00;. 00. 80 80. 00.. 80. 00». 89. 8a. 80. 8: . 8.. 80. OODN 8Nn 80m 000' q a d 1 A a s o s is > b b n h p p r p L OW. one: om. . 0WD 08. cm: 00“. 8a. 00'. 80. 00.. 8h. 80. 000. 000“ OONn coon oooc Antone... :3 52:99 760 .mumizz w><3 29 hydrated water molecules to the oxygen in the phosphate anion (40). Infrared spectra of the synthetic ferric phosphates are given in Figure 2. The Fe phosphate which was crystal— lized under the same conditions as variscite gave spectra (Figure 2-b) which shows strong P—O stretching bands between 900 and 1200 cm-1. The water deformation band is near 1600 cm_1 and the sharp OH stretching band at 3560 cm—1 also indi- cates the loosely held water of crystallization. The spectra of the crystalline Fe phosphate was similar to that obtained by Corbridge and Lowe (40) for FePO4'2H20, but differences in the shape of the P—O stretching bands were noted when com- pared with the spectra reported by Arlidge, gt_al. (4) for strengite. The P—O bands given in Figure 2-b indicate that neutral orthophosphate ion containing PO 3- 4 is the only ion present in the crystal. Heating the crystalline Fe phosphate at 65°C for 12 hours resulted in partial loss of the hydrated water as shown in Figure 2—c. Hydrated water bands at 1600 and 3560 cm:1 still persisted after heating but the intensities were con— siderably less. After further heating at 130°C for 12 hours, the crystalline Fe phosphate was converted to an anhydrous material (Figure 2-d). The loss of hydrated water from strengite at a higher temperature than from variscite may be because of the larger particle size of strengite, or because the hydrated water was more strongly held in the strengite structure.r 3O .ocaaomm> on map mocmmt .mnson NH How OOOMH um poumoc oumnmmonm om ocaaamummno AUV .mnson NH now 0 0mm um woumon oumzmmocm om ocaaamummuo loo .AONmm. scoot .ousmsosuov outsmmoso om ocaaamummuo any .vOmom Hmofloaaou Amv .moumnmmonm mm UHuonuc>m mo muuoomm coaumHOQO poumumcH .N ousmflm 80 003 00. 0°. 08. 8.. 80. 80. 80. 8a. 8.. 8». 8.. 80. SON sun 80m 000v p p p h p p p b p > p h 8. 02 oo- 8. 8o. co: 8.: 8n. 3! 8o. 82 8: 8... 8... 730053 :8 guav TED .mmmzaz u><3 31 2. Particle Size and Specific Surface Electron Microscopic Observations By means of electron microscope, the particle size and the shape of the synthetic colloidal and crystalline phosphates were observed using the parlodion film and the metal vapor shadowing techniques. Chromium shadowed electron micrographs of colloidal AlPO4 and colloidal FePO4 show that the individual particles of the colloidal FePO4 are smaller than the particles of AlPO4 (Figure 3, Plates 1 and 3); however, both are less than a half micron in average diameter. Since the hydrated water in the variscite and strengite crystals is loosely held in the structure (40), the crystalline surfaces may be damaged during the process of metal vapor casting under high vacuum. Therefore, electron micrographs of the two crystalline ma— terials were taken using the parlodion film technique in which the sample film is prepared under room temperature and atmospheric pressure. Much larger particles were formed after the two colloidal phosphates were crystallized at'105O C in water to form variscite and strengite (Figure 3, Plates 2 and 4). The variscite crystals shown in Figure 3, Plate 2 are reniform aggregates with an average diameter of about 1.5 microns. The strengite crystallized under the same con- ditions seem to form large, single, spherical crystals Plate 1 Plate 2 Figure 3. 32 Plate 3 Plate 4 Electron micrographs of synthetic A1 and Fe phos- phates: Plate 1, newly precipitated colloidal AlPO4, Cr shadowed; Plate 2, colloidal AlPO di— gested at 105°C for 40 days to form varisci e, parlodion film; Plate 3, newly precipitated col— loidal FePO , Cr shadowed; Plate 4, colloidal FePO4 digested at 105°C for 40 days to form strengite, parlodion film. 33 (Figure 3, Plate 4 and Figure 4, Plate 1) with an average diameter of about 6 microns. The high degree of crystal- linity and the distinctive crystalline surfaces are also illustrated (Figure 4, Plate 2). B.E.T. Surface Area Surface area of the four P materials was measured by N2-adsorption at -l95°C with a Perkin-Elmer-Shell sorptometer. Samples were degassed in a flow of He gas at room temperature for 48 hours before measurements. Surface area of the sample was then calculated according to the B.E.T. equation: V(p°-p) z Iii—5 + WE‘ELCl-gg Where, p is the equilibrium vapor pressure,,pO the satur- ation vapor pressure, V the volume of the gas adsorbed, Vm the volume adsorbed for a monomolecular layer and c is a constant for a given system. By plotting p/V(po-p) against p/po, a straight line should be obtained of slope (c-l)/Vmc and intercept 1/Vmc. Thus Vfi and c can both be determined. If the area per gaseous molecule is known, the surface area is then easily calculated (102). Specific surfaces of the four phosphate samples measured by the B.E.T. method (Table 2) agree well with electron microscopic observations. The surface area of 27.5 ng—l colloidal Fe phosphate decreases sharply to 1.98 ng-l after crystallization to strengite. The greater surface 34 Plate 1 Plate 2 Figure 4. Electron micrographs of synthetic strengite: Plate 1, Cr shadowed strengite; Plate 2, parlodion film supported strengite. 35 'Table 2. B.E.T. surface area of synthetic phosphate compounds. Material Surface Area Notes 2 —l m 9 Colloidal Al-phosphate 10.5 Freshly precipitated Colloidal Fe—phosphate 27.5 Freshly precipitated Crystallized Al-phosphate 24.9 Crystallized at (variscite) 105°C, 40 days Crystallized Fe—phosphate 1.98 Crystallized at (strengite) 105°C, 40 days 36 area of the variscite (24.9 ng-l) particles as compared to 29-1) is apparently due to the colloidal Al phosphate (10.5 m irregular aggregated crystalline surfaces of variscite as shown in the electron microscopic observations (Figure 3, Plate 2). These results suggest that specific surface and degree of crystallinity are the important factors that govern the relative availability of A1 and Fe-phosphates in acid soils. 3. Rate of Crystallization To determine the rate crystallization of colloidal AlPO4 and FePO4 in aqueous medium, a set of colloidal AlPO4 water suspensions, each consisting of 5 g. of one material and 500 ml of distilled water, were digested in an autoclave at 105°C. The pH of the suspensions was adjusted to 3.2 daily with 0.02 N H3PO4 or distilled water. Samples were re- moved after 12 hours, 72 hours, and 40 days, and were air dried for Xéray analysis for crystallinity. Another set of colloidal AlPO4 and colloidal FePO4 water suspensions were crystallized at 35°C in a constant temperature incubator under atmospheric pressure. The pH of the suspensions were also adjusted to 3.2 daily. After 9 months, these samples were air dried for Xkray diffraction analysis and microscopic observations. As shown in Figure 5, after digesting the colloidal AlPO4 for 72 hours, the material still remained amorphous to 37 3035 19:5 43:3 5403 l 1 1 l l l L l l 1 1 J 40 38 36 34 32 30 28 26 24 22 20 I8 I6 :4 DEGREES 2 9 Figure 5. Xkray diffraction patterns of A1 phosphates: (A) newly precipitated colloidal AlPO4, (B) colloidal A1P04 digested at 105° C for 72 hours, (C) col- loidal AlPO4 digested at 1050 C for 40 days. *Peaks due to quartz in the ceramic plates. 38 Xéray diffraction. But after 40 days of digestion, an XEray diffraction pattern for pure variscite, A1P04'2H20 was obtained. By employing the same technique to study the crystal- lization of colloidal FePO4 under the same temperature and pressure conditions as that for the AlPO it was found that 4, the FePO4 crystallized at a much faster rate than the AlPO4. Microscopic observations indicated that the crystallization was almost complete after 12 hours of digestion. Xkray dif— fraction of the material gave a pattern of pure strengite, FePO4'2H20 (Figure 6) at that time. Crystallization of colloidal FePO in water at 35°C 4 was also observed under electron microscope after 9 months of incubation. However, the material was still amorphous to Xeray. AlPO4 remained essentially colloidal throughout the 9 months incubation at 35°C. The above results indicate that FePO4 crystallizes at a much faster rate than AlPO4 under the same enrivonmental conditions. This may also be true during P fixation and transformation in soils under natural conditions. In addition to temperature and pressure, the rate of crystallization of Al and Fe phosphates in aqueous media may also be greatly influenced by the pH of the medium and the presence of other ionic species. Further investigations are needed with regard to these points. 39 o o o 3.” A 4.36 A 5.43A L 1 L L l l l l L 1 40 38 36 34 32 30 28 26 24 22 DEGREES 2 e L. 20 l8 l6 I4 Figure 6. X—ray diffraction patterns of Fe phosphates: (a) newly precipitated colloidal FePO4, (b) colloidal FePO4 digested at 105 C for 12 hours, (c) col— loidal FePO4 digested at 105°C for 40 days. 4O 4. Relative Availability A quartz-sand culture experiment with two repli— cations was conducted in a growth chamber by using Sudan grass seedlings as the indication crop, to compare the rela- tive availability of the colloidal AlPO colloidal FePO 4’ 4’ variscite and strengite as sources of P for plant. The latter two minerals were synthesized under the same temper— ature, pressure and digesting period. The two colloidal phosphates were applied at the levels of 80 and 200 mg P per pot and the two crystalline phosphates were applied at the levels of 80 mg P per pot. A control was included which re— ceived no P. The phosphates were mixed thoroughly with 400 g of acid-washed fine quartz—sand, and the mixture was placed in a card—board paper cup. Four Sudan grass seedlings, two- Weeks old, were transplanted into each plot. Hoagland solu- tion minus P and/or distilled water were added through a side tube attached inside the cup as required. The moisture con- tent of the quartz—sand was maintained at its field moisture capacity. Growth differences due to treatments were noted during the incubation period of 30 days as shown in Figure 7. After 30 days, the above-ground portions of plants of each pot were clipped, dried at 65°C, weighed, and ground. The P content was determined by dry ashing in the presence of an alcoholic solution of Mg(NO3)2 and the P dissolved in di— lute HCl solution was then determined colorimetrically (71). 41 Figure 7. One month old Sudan grass seedlings with P ap- plied as different P minerals. 42 The yield of tops and the P uptake indicate that colloidal AlPO4 and FePO4 were about equally available and were much better sources of P for the plant than their crystallized forms (Table 3). The P in the crystalline ferric phosphate form (strengite)~seemed to be completely unavailable during the 30 days of cropping. The P in the form of crystalline Al phosphate (variscite) seemed to be slightly more available than strengite although both were shown to be very poor sources of P for the Sudan grass seed— lings in the slightly acidic sand-cultural medium. As also indicated in Table 3, higher rates of col- loidal AlPO4 and FePO4 (500 ppm P or 200 mg P per pot) in- creased the P uptake significantly. However, increasing the rate of P applications did not effectively increase yield, probably because of some other limiting factors. Results obtained in these experiments agree well with those reported previously by Taylor,_gt_§l. (139,142) and Lindsay and DeMent (94). In their greenhouse experi- ments, availability of various Al and Fe phosphates were com— pared by growing corn on P—deficient acid soils using Ca(H2PO4)2 as a standard P source. After three successive crops of corn grown on a acid soil of pH 4.8, the total P uptake from colloidal AlPO was considerably higher than that 4 from colloidal FePO4 (139). This may be because of the faster rate of crystallization of FePO4- In other words, during the time required for three successive crops of corn, 43 Table 3. Yield and P uptake by Sudan grass after 30 days growth. Source of P Rate of Yield of P Uptake Application Tops in Tops mg P/Pot g Dry wt/Pot mg P/Pot No P O 0.13 0.08 Crystallized Fe-P 80 0.15 0.08 Crystallized Al—P 80 0.23 0.14 Colloidal Fe-P 80 0.47 0.64 Colloidal Fe-P 200 0.55 0.85 Colloidal Al-P 80 0.49 0.60 Colloidal Al-P 200 0.47 1.00 LSD (0.01) 0.12 0.26 LSD (0.05) 0.08 0.17 44 a considerable portion of the colloidal FePO4 may crystallize under the acidic conditions and concurrently increase in‘ lattice energy and particle-size while specific surface is greatly reduced. Meantime, Al phosphate may still remain highly colloidal during the entire cropping period. V The effect of pH on the availability of Fe and A1 phosphates is also critical. The rate of crystallization of colloidal AlPO4 and FePO4 in aqueous media may be greatly in— fluenced by the pH of the media. The fact that on a calcar- eous soil with pH 8.5 colloidal AlPO4 and variscite as re- ported by Taylor, §t_al. (142) to be equally available clearly suggested that a rapid hydrolytic reaction of the A1 phosphate compounds at alkaline pH had taken place which readily dis- solved the Al—P compounds. Under this circumstance, the hydrolytic reaction greatly overwhelms the properties such as specific surface, particle size and degree of crystal- linity which are the important factors that control the rela— tive availability of Al-P and Fe—P under acidic environments. Soil Al—P fraction determined by Chang and Jackson's procedure (25) has also been reported by many workers (55, 103,135,2) to be a more preferable form of P for upland crops than the Fe-P fraction. The findings seemed to sug- gest that the Fe-P fraction may exist in higher degree of crystallinity than the Al-P fraction in the acid soils studied. It is the degree of crystallinity, not the specific 45 surface or particle size that is the factor most responsible for controlling the availability of Al—P and Fe—P in acid soils. 5. Conclusion Based on the results obtained in this investigation and those reported by other investigators, chemical and physical factors controlling the relative availability of Al-P and Fe-P during P fixation and transformation in acid upland soils are summarized and presented diagrammatically in Figure 8. When applying soluble P to acid upland soils, or, when soil Ca-P is dissolved during the process of chemical weathering, the soluble P is believed to precipitate rapidly and FePO . Because of their smaller 4 4 particle sizes, greater surface area and amorphous structure, to form colloidal AlPO the colloidal form of AlPO and FePO4 are readily available 4 to plants. But these colloidal phosphates tend to crystal- lize readily to form hydrated compounds such as variscite and strengite which are much less available for plant growth. Thereby, as shown in the diagram, during the first stage of P fixation, the rate of crystallization is the factor cone trolling the relative availability of these two forms of phosphates in acid soils. As time goes on, crystallization takes place under favorable environmental conditions. Since FePO crystallizes 4 46 at a much faster rate than AlPO4, at the second stage the de— gree of crystallinity, or in other words, the relative pro— portion of the colloidal form to the crystalline form is the factor controlling the availability of Al-P and Fe—P in acid soils. Most of the arable, acid soils in the upland cropping region should fall into this category. In acid upland soils, under natural conditions, the crystallization of AlPO4 may take many years while that of FePO may only require a few 4 years. These assumptions may be employed to explain the findings reported by many workers that in acid upland soils, the native Al-P fraction seemed always to be more available to plants than the Fe—P fraction. In the final stage, when both forms of phosphates have completely crystallized, specific surface and crystal structure may become more important in controlling the avail- ability of the phosphates, although both are very poor sources of P for plants in acidic, well-aerated soils. In calcareous soils and lowland soils, such as rice paddies, different mechanisms involving hydrolytic reactions and changes in redox potentials in the soil have to be considered. 47 .mEMOm oHQMHHm>m mmwa ODCH m oHQSHOw mo coaumEH0mmcwHE axon. 31.3325 D 3.9. 933.38 § .m ousmflm :2 .\_\. 4 a I u... AlzIIISoS \ sAlalllemEo zo:AImu \ Amt 44 DISTRIBUTION OF ALUMINUM, IRON AND CALCIUM PHOSPHATES IN SOME MICHIGAN SOIL PRO- FILES AND THEIR PARTICLE SEPARATES Rate of crystallization has been shown in the pre- vious part of this investigation to be the most important factor in controlling the relative availability of Al—P and Fe-P‘with synthetic systems. However, with extended periods of time, as with natural soils, the degree of crystallinity and growth of crystals or crystal size of P minerals in soils may also be important. It is of great importance, therefore, to study the distribution of P minerals within textural size fractions of natural soils. Four Michigan soils were selected to use in the de— velopment of methods that will allow for such a study. These soils were selected to give a range in textural classes and to allow for correlation with P field studies being con— ducted by other persons. 1. Description of Soil Samples The four Michigan soil profile samples used for this investigation have been extensively cropped, therefore, the lField experiments are being conducted by E. C. Doll, D. R. Christenson, R. P. White and G. R. Rinkenberger. 48 49 surface horizons of the soils have been disturbed. Descrip- tions of the soils are stated as follows: (1) Pewamo clay loam 3. Characteristics: It is a poorly drained, Humic Gley soil in the Gray Brown Podzolic Region. It developed in calcareous silty clay or clay loam till. Sample location: Shirkey Farm, St. Clair County, Columbus Township, Memphis, Michigan. T5N9R15E NW corner of Section 6. Sampling date: Fall, 1964. Previous P fertilization: Received 300# 12—12-12 (36# PZOS/A) in 1962. No P was applied in 1963 and 1964. (2) Karlin sandy loam 3. Characteristics: The soil is a well-drained Podzol developed in loamy fine sand to fine sandy loam over- lying sand. The A2 horizon is absent in this culti- vated soil and the Ap rests directly on the B2hir horizon. Sample location: Estelle Farm, Antrim County, Star Township, Michigan. T 30 Ns-R 5 W, Section 1. Sampling date: Fall, 1964. Previous P fertilization: Received 800# 5—10-20 (80# PZOS/A) in 1963, 200# 5-10—20 (20# PZOS/A) in 1964. 50 (3) Onaway loam a. Characteristics: It is a well drained soil with a Podzol upper sequum and a Gray-Wooded lower sequum. It developed in calcareous loam or silt loam till. b. Sample location: Misiak Farm, Presque Isle County, Posen Township, Michigan. T 33 N - R 6 E, Section 24. c. Sampling date: Fall, 1964. d. Previous P fertilization: Received 200# 0—54-0 (108# P205) in 1963. No P was applied in 1964. (4) Sims clay loam a. Characteristics: It is a poorly drained Humic-Gley soil developed in calcareous clay loam till. b. Sample location: Ferden Farm, Saginaw County, Chesaning Township, Michigan. T 9 N - R 3 E, Section 33. c. Sampling date: Fall, 1964. d. Previous P fertilization: No P has been applied for 9 years. 2. Distribution of Al—P, Fe-P and Ca-P in Soil Profiles The amounts of Al—P, Fe—P and Ca-P in the soil samples were determined by the procedure of Chang and Jackson (25, 29,31). Total P content in soil was determined by the method of sodium carbonate fusion as stated in Jackson's "Soil Chemical Analysis" (71). 51 Total P, Al-P, Fe-P and Ca-P contents and pH of the soil profiles are presented in Table 4. The pH of the Pewamo and Karlin profiles are acidic throughout except for the change to neutral in the Cl horizon of the Pewamo profile. The pH of the Onaway profile varies with depth from neutral to slightly alkaline and for the Sims profile, it varies from slightly acidic to slightly alkaline from surface to subsoil. The higher contents of total P in all surface soils are apparently due to the higher organic P contents in the surface horizons. The occluded form of Al and Fe-P was not determined. The contents of P soluble in neutral NH4C1 (water soluble P) are negligible in all cases. As shown in Table 4, Ca-P is the predominating form of inorganic P in the profiles of Pewamo, Onaway and Sims. In the case of Karlin, the three forms of inorganic P are about equally distributed throughout the profile except for the surface 9 inches which is highest in Al—P due to recent P application. In all cases, the contents of Al—P and Fe-P are higher in the surface horizons. The Ca—P content increases with depth for the two calcareous soils (Onaway and Sims). With Pewamo and Karlin soils, the same trend holds in the lower part of the profile, but the surface 10 inches layer is considerably higher in Ca-P than its underlying horizon. The percentage distribution of Al-P, Fe-P and Ca-P in the soil profiles was calculated and presented in Table 5 52 .OHumu Hmum3\aflom Haas ms 66H 66H as ea moose Hes 6.6 66-66 60 66 66 66H 66H os 6 moths 666 H.e emI6H 6666 06 66 666 66H 66 66 tomes 666 6.6 6Hn6 mean no 66 H66 66H 66 so moose 666 0.6 6:6 as we . 6266 oos 66H 66H 0 6 homes 66H 6.6 omues 666 66 6s 66H 66H 6 as 6.6 666 6.6 6HI6H ms 66 H6 666 66 NH om 6.H mO6 6.6 6HI6 seem 06 mm 66H 66 6H 6H 6.6 666 6.6 6-6 me am as Hma 66 66 6s s.a 6H6 6.6 euo as 66 >m3mco H6 66 66 6s om moose 66H 6.6 66:66 No 66 66 66 om 6H om domes 66 6.6 6muo~ Ho 06 6m 66 as am am momma 66H «.6 66-6 sanmm am 66 6H6 66 66 60H 6.6 6H6 0.6 6.6 mm mm cHHHmM H6 on 66H 06 om moose O66 6.6 66:66 Ho.6mmm we 66 66 o6 66 6 domes 666 6.6 eNI6H 6666 6H 66 Has 66 66 as 60666 666 6.6 6HIoH mmsm 0H 66 now was 66 mm 6.6 666 6.6 6HI6 was he 66 How mos 66 66 towns 6H6 6.6 6.0 as 6H OEm30m X Emu IIIIIIIIIIIIIIII EQQ IIIIIIIIIIIIIIIII .CH 6 H6606 656 6.60 not muse mansaom mo coauomum Houmz m tum spawn. conauom .02 .. Hmeoe H666 m I Amo + mm + H5 mcofluomum m UHCMmAOCH .monEmm oaamoum Hflom mo mCOHuomHm m UHCMmHOCH tam m Hmuoa .v magma 53 Table 5. Percentage distribution of Al, Fe and Ca—P in soil profiles.* Soil No. Horizon Al-P Fe-P Ca—P % % % Pewamo la Ap 16.9 32.3 50.7 lb A12 15.5 30.4 54.1 1c B12g 7.8 25.5 66.7 1d B22g 8.3 39.6 52.1 1e B22g,Cl 9.5 19.1 71.4 Karlin 2a Ap 48.4 26.0 25.6 2b B2hir 35.6 40.7 23.7 2c D1 35.7 28.6 35.7 2d D2 30.3 28.8 40.9 Onaway 3a Ap 24.9 20.4 54.7 3b A2 13.1 15.6 71.3 3c Bhir 15.6 9.4 75.0 3d A2 7.7 5.6 86.7 3e B2t 1.4 0 98.6 Sims 4a Ap 20.3 23.8 55.8 4b B2lg 17.0 20.6 62.3 4c B22g 6.3 7.0 86.7 4d Cg 5.3 5.9 88.8 ppm *Calculated as: P of Al or Fe or Ca-P in soil Sum of(A1+Fe+Ca)—P in ppm P in soil 54 and Figures 9 and 10. The percentage distributions of Ca—P and Fe-P for the Pewamo soil profile are essentially of mirror images of each other--Fe-P decreases with depth and Ca-P increases with depth. Al—P never exceeds 17 percent of the three fractions and remains relatively constant through- out the B and C horizons. The relative distribution of in— organic P in a soil profile has been shown by Chang and Jackson (27) to be a measure of the degree of soil chemical weathering. Their concept is that as weathering progresses there is a shift in relative abundance of the inorganic phos- phates from Ca—P toward Al-P, Fe-P and occluded P. The Ca-P is believed to be of primary origin and exists in the form of apatites. According to this concept, the percentage distributions of inorganic P for the Pewamo soil profile indi— cates that weathering within the profile has been relatively uniform. The remarkably high percentage of Fe—P and corre- spondingly low percentage of Ca-P in the B22g horizon (19— 27") is apparently due to the strongly reduced condition in this horizon of this poorly drained, Humic-Gley soil which brings Fe as ferrous ion into the solution phase and at the same time results in a more acidic environment that favors the dissolution of Ca-P. Consequently, when the soil is brought to a more oxidized environment, the P in the solution phase precipitates instantaneously as Fe—P with the ferric ion which has been oxidized from the ferrous form. The con- tent of Al-P in the Pewamo soil indicates that under this 55 -—0--a— Fe-P (A) Ca-P ‘- -‘-‘ T H O r \ n\.~ .—‘ x ‘5 /“ 10-19 ' Depth (inch) ./ 19-27 7 f-——-.——---“ N O‘\ 40 60 80 100 27-42 % of total P (A1 + Fe + Ca) ? KO 1 I N O I 1 20—29 Depth (inch) J l 4 29-37 . 60 80 100 % of total P (A1 + Fe + Ca) Figure 9. Percentage distribution of Al-P, Fe—P and Ca-P in profiles of (A) Pewamo clay loam, (B) Karlin sandy loam. Figure 10. 56 -—---— Al—P (A) 0-7 F“ / 1" A 7_9 __ (a 6 I c [I :1 9—16 — /’} fi [/ 8‘ 16—17 H,’ D ’I p 17-30 L ‘ ‘ 1 . 20 4O 6O 80 100 '% of total P (Al + Fe + Ca) (B) 0‘9 {- It”! A I n .- 2 9-16 _ ;] .H 7/ V / 1: 16-27 _ f/ 4.) I Q g. 3 :! 27—34 . . . . _ I 60 80 100 % of total P (Al + Fe + Ca) Percentage distribution of Al-P, Ca-P in profiles of (A) Onaway loam, Sims clay loam. Fe—P and (B) r—Fe-P Ca—P 57 environment, it serves either as a transitional phase for the decreasing Ca-P and increasing Fe—P fractions (27,59) or as an initial phase of P fixation in the surface soil as a result of fertilization (30,97,118). The distributions of the Ca—P and Fe-P for the Karlin loamy sand soil from 9 to 37 inches of depth show a mirror image relationship similar to that of the Pewamo clay loam profile although Fe—P is the dominant fraction in the 9 to 20 inches layer indicating a relatively greater degree of transformation of inorganic P within this horizon. Al—P re- mains essentially constant throughout the profile between 9 to 37 inches of depth and the difference between Al-P and Fe-P is less than for the Pewamo soil. But, in the top nine inches, Al—P is a predominating fraction which is easily ac- counted for by the recent P application to the soil (97,118). In the calcareous Onaway soil profile, which is com— prised of a Podzol upper sequum and a Gray-WOOded lower sequum, the percentage distributions of Ca-P and Fe-P also show a mirror image relationship. The Al—P decreases with depth and the difference between Al-P and Fe-P is also less. However, the percentage of Al-P exceeds that of Fe-P in the surface (0-7 inches) layer which may be attributed mainly to the heavy P fertilization during the previous year. The same relationship observed in the lower sequum of the profile may indicate a lower degree of weathering. Ca-P is the most abundant form among the three forms of P throughout the 58 profile and it increases from 55 to 99 percent from surface to 30 inches of depth. Neither Al—P nor Fe-P exceeds 25 per- cent. These relationships indicate a much lesser amount of weathering for the Onaway profile as compared with the pro- files of Pewamo and Karlin providing the distribution of in— organic P does serve as an indicator of chemical weathering. The distribution curves of Ca-P and Fe-P for the Sims clay loam profile are also mirror images of each other. Again, Ca—P, the predominating component, increases with depth, while the content of Fe-P never exceeds 24 percent and decreases with depth. However, the Al-P and Ca-P distri- bution curves also show mirror image relationship between each.other; nevertheless, the content of Al-P never exceeds that of Fe-P throughout the profile. The inorganic P distribution curves of the four soil profiles studied seem to indicate the progressive dissolution of Ca-P, followed by the subsequent precipitation as Al-P and/or Fe—P in soils during the course of weathering. The degree of chemical weathering is generally greater for the acid soils than that for the calcareous soils with regard to the relative abundance of the inorganic P forms. As also shown in the distribution curves, Al-P seems to serve as a transitional phase for the decreasing Ca—P and increasing Fe—P fractions within the profile. The relationship is par— ticularly pronounced in the case of Pewamo soil profile. 59 3. A Proposed Procedure for the Separation of Soil Particle—size Fractions for Nutrients Analysis Methods developed by Soil Physicists (16) to dis- perse soil samples prior to mechanical analysis generally in— clude acid treatment (final pH about 4.0) and neutralization with NaOH to pH about 9.5. It is likely that the acid treat- ment will dissolve Ca-P, particularly that in the clay fraction, and the dissolved P may reprecipitate as Al—P and/ or Fe—P. The alkaline treatment undoubtedly results in the dissolution of Fe-P and Al-P. Therefore, the original distri- bution of the soil P compounds is greatly disturbed. The use of Na2CO3 as a dispersing agent will again hydrolyze Fe-P in soils. Also, the use of sodium hexametaphosphate (Calgon) Will not be satisfactory because of the content of P in the compound and the high pH value of the solution. Mechanically shaking the soil in distilled water for an extended period of time also does not seem to be satisfactory, particularly with soils of heavy texture or with high organic matter content. For the analysis of inorganic P and/or other mineral nutrients contents in soil particle separates, it seems that none of the commonly used dispersion procedures would satis— factorily serve the purpose of obtaining a complete separation of the soil particles without causing drastic disturbance of the original distribution of P and other mineral nutrients 60 in the solid phase of the soil. Therefore, a proposed pro- cedure involving NaCl treatment and sonic vibration was de- velOped during this investigation. Procedure Sample pretreatment.-—Hand-sieve the air—dried soil sample through a 2 mm sieve. Use a mortar and rubber-tipped pestle to break the aggregate which are larger than 2 mm. Dispersion of the soil sample.--To 50 g of soil(passed 2 mm sieve) in a 250-ml beaker, add 150 m1 of 5 percent NaCl solution. After 20 minutes soaking, disperse the suspension with the supersonic dispersion apparatus (Bronwill Biosonik) for 2-5 minutes. After standing overnight, remove the ex- cess NaCl solution first by decantation or by the suction filtration apparatus (16). After most of the excess NaCl solution has been removed, add an additional 100 m1 of dis- tilled water and transfer the suspension into a dialysis bag. Wash the beaker thoroughly in order to transfer all the soil particles by using an additional 100 m1 of distilled water. A few drops of sulphanilamide is then added to the suspension to suppress growth of microorganisms. Dialyze the suspension for 5 to 10 days by putting each five bags in a 20-1iter con— tainer filled with distilled water. The water is changed daily. Transfer the suspension back into a 600-ml beaker. Further disperse 100 mls increments of the soil suspension by the supersonic disperser for 2-5 minutes. 61 Separation of clay from the silt and sand fractions bygsiphoning (l6).--Transfer the dispersed soil suspension completely into a l-liter graduate cylinder. Add distilled water to bring up to mark that gives a 5 percent suspension. Allow it to settle under constant temperature. Record the temperature of the suspension and determine from Stokes' equation the time required for a Z’p particle to fall through a definite height, equal to 80 percent of the depth of the suspension. Insert a siphon tube to the predetermined depth at the proper time, and siphon the suspension into a sample jar, proceeding very slowly to reduce disturbance. Continue the siphoning until the supernatant liquid has been removed to the desired depth. Remove the siphon, add distilled water again to the original level, mix the suspension thoroughly as before, and repeat the sedimentation and siphoning. While the second period sedimentation is in pro- gress, use the filtration apparatus (16) to concentrate the clay that has been collected in the sample jar. When a co- herent layer of the clay has formed on the surface of the filter candle, disconnect the vacuum and apply air pressure to the interior of the filter by squeezing the rubber bulb attached to it. Continue the filtration until no free water remains. Collect all the siphonings of the clay fraction in the same jar, and repeat the operation until the yield be- comes negligible. The clay is then air dried, ground and stored. 62 Separation of silt and sand fractions by wet-sieving (l6).--Inspect the 300-mesh (47p) sieve for cleanliness and mechanical condition. Moisten both sides with water and place it in a wide-mouth funnel which is supported by a ring supporter over a 1-1iter cylinder. Without shaking or swirl- ing the mixture, pour the suspended portion of the sample in- to the sieve. Add more water to the residue in the cylinder. Swirl the mixture and allow to settle for 2 minutes. Decant the suspended portion into the sieve as before. Repeat the operation several times until most of the fine material has been transferred. Tilt the cylinder downward over the sieve. Direct a jet of water upward into it, sweeping the soil particles downward into the sieve by the force of the effluent stream. Do not rub the screen at any time. When the trans- fer has been completed, agitate the residue on the sieve with a jet of water and obtain as complete a separation as possi— ble. Air dry the sand fraction. Dry sieve the sand into further fractions if needed. Remove the water in the silt fraction by vacuum filtration. The silt is then air dried, ground and stored. The air—dried samples of the soil particle-size fractions are then ready for P fractionation and other mineral nutrients analysis. Notes: (a). In light of the rise of pH during di- alysis, complete removal of the excess Na and Cl ions from the Na—saturated soil system by dialysis is not necessary; 63 however, the concentration of the salt in the solution phase of the system should be low enough to maintain the soil in a well dispersed condition. (b) Excessive dialysis against distilled water will cause Donnan hydrolysis which may pro- duce reflocculation of the system (102). (c) Time of treat— ment under sonic vibration should be carefully controlled in order to prevent undesired breakdown of soil particles into finer ones. (d) The procedure may not be found satisfactory for soils high in CaCO3 or sesquioxides. Efficiency of the Procedure Efficiency of the proposed procedure was compared with that of the common procedure of sodium hexametaphosphate (Calgon) dispersion by using the Hydrometer method for soil mechanical analysis (17). Results are listed in Table 6a. Mechanical analysis of six soil samples showed that the two dispersion procedures were in good agreement and the pro- posed procedure of dispersion can thus be regarded as com— parable to Calgon dispersion. The change in pH of the NaCl-soil suspension before and after dialysis was also measured and is presented in Table 6b. The final pH values of the four acid soils are near neutral, while for the two alkaline soils, the pH has increased about 0.8 unit after removal of the Cl ions from the system. The results indicate that the change in pH will not be sufficient to cause hydrolysis or dissolution of the 64 Table 6a. Comparison of the proposed sonic and the calgon methods of diapersion.* Soil Particle-size Composition (USDA) N°' 3°11 Sonic Dispersed Calgon Dispersed Sand Silt Clay Sand Silt Clay ________ %-______ ______-_%,______ 1 Pewamo, surface 36.0 31.7 32.2 34.8 31.7 33.5 2 Pewamo, subsoil 30.0 30.3 39.7 30.0 29.1 40.9 3 Karlin, surface 83.2 12.3 4.5 82.0 12.5 5.5 4 Karlin, subsoil 93.2 4.5 2.3 93.2 3.7 3.0 5 Onaway, surface 61.2 30.6 8.2 61.2 29.4 9.4 6 Onaway, subsoil 59.4 25.1 15.6 59.4 23.8 16.8 *Particle-size composition was determined by the Hydrometer Method. Table 6b. Change in pH of NaCl-soil suspension before and after dialysis against distilled water. Soil Reaction Soil No. Before Dialysis After Dialysis ___________________ pH________________-_ 1 5.8 6.8 2 5.4 7.0 3 5.6 6.6 4 4.5 6.6 5 7.1 7.9 6 7.6 8.4 65 inorganic P. However, excessive dialysis does bring the pH of the acid soil systems to the vicinity of 8.5 and under this circumstance, considerable hydrolysis of Fe-P may occur. Therefore, the pH of the dialyzed soil suspension should be carefully controlled in order to achieve high dispersion with the least disturbance of the original P distribution of the soil. The proposed procedure of dispersion is considered to be a satisfactory method for the purpose of studying the particle-size distribution of inorganic P as well as other water-insoluble mineral nutrients in the soil on the basis of the following advantages: (a) It does not cause drastic change in pH of the system. (b) The homogeneous sonic vibration treatment has less tendency to cause the mechanical breakdown of the original soil particles than might occur when a motor mixer is used. 4. Distribution of Al-P, Fe-P and Ca—P in Soil Particle Separates Particle-size fractions of the soil samples were separated according to the proposed procedures of dispersion and separation. Particle—size composition of the soil samples based on large-scale separation is given in Table 7. Al-P, Fe—P and Ca-P contents in the soil particle separates were determined according to the method of Chang and 66 Table 7. Particle-size composition of soil profile samples based on large-scale separation. Particle-size Soil Depth Composition* No. Horizon in. Texture Sand Silt Clay __________ %k—-------— Pewamo 1a Ap 0—6 20.6 42.9 37.1 CL 1b A12 6-10 20.0 44.6 35.4 CL 1c B129 10-19 18.2 41.5 40.3 SiC 1d B22g 19-27 18.8 40.7 40.5 SiC 1e B22g,Cl 27-42 18.4 41.1 40.5 SiC Karlin 2b B2hir 9-20 78.9 17.0 4.1 LS 2c D1 20-29 78.6 16.9 4.5 LS 2d D2 29-37 87.3 4.9 7.9 S Onaway 3a Ap 0—7 39.2 43.2 17.6 L 3b A2 7—9 38.5 36.5 25.0 L 3c Bhir 9-16 40.9 32.3 26.8 L 3d A2 16-17 71.1 21.0 7.9 LS 3e B2t 17-30 53.0 33.7 13.2 SL Sims 4c B229 16—27 36.4 30.5 33.1 CL 4d Cg 27—34 34.9 30.9 32.2 CL *(1) According to USDA classification of soil particles. (2) Calculated from the actual weights of the sand and silt fractions obtained in the large-scale separation. The clay content is obtained by subtraction. (3) % air-dried basis. 67 .Jackson (25) and the results are given in Table 8. Due to time presence of certain organic compounds and/or microbial activity in the Karlin and Sims surface soils, the dialyzing loags disintegrated and these surface soil samples were lost. For all soil profile samples studied, Al-P and Fe-P seem to be highly concentrated in the clay fraction; whereas, the Ca—P has a tendency to be equally distributed in the silt and clay fractions in the upper horizons and higher in the silt fraction than in the clay in the lower horizons indicating that Ca-P in soil is of primary origin. The sand fraction contains less inorganic P than other separates. The contents of Al-P, Fe-P and Ca—P in each of the particle separates throughout the soil profile are in the following orders: in the sand fraction, the order is Ca—P > Fe-P > Al—P for Pewamo, Onaway and Sims soil pro— files; and Ca-P’kFe—PQ Al-P for the Karlin profile. In the silt fraction, the order is Ca-P >> Fe-P > Al-P for the Pewamo profile and Sims subsoil; Fe-PfiQrAl-P2Q5Ca-P for Karlin; and Ca-P >> Al-P > Fe-P for the Onaway profile. In the clay fraction, the order is Ca-P > Fe-P5%=A1-P for Pewamo, Onaway and Sims profiles. It becomes Fe-P > Al-szz Ca—P in the case of Karlin. The above results indicate differences in the degree of weathering of inorganic P within the soil profiles. The fact that all three forms of inorganic P are more concen- trated in the finer fractions of the soil suggests that 68 66H 666 66 6H 6 6 66 6 6 6o 66 o6H H66 66 6 6 6 66 6 6 666m 66 mEHm 666 666 66 o H 6 6 6 6 666 66 666 H66 66 6H 6 6 66 66 6 6m 66 666 H6H H6 6H 6 6 66 66 oH 6666 06 6H6 o6 66 H6 6 6 66 66 6 64 66 666 66H 66 66 6H 6 66 66 6 as 66 mm3mco 66H 66 66 66 66 6H 66 66 6H 66 66 66 66 6H 66H H6 6 66 66 6H H6 06 66 66 6H 666 66 6 66 66 6H 6Hs6m s6 CHHHMM 66H 666 66 66 6H H6 66 6 6H Ho.6666 6H 66 H6 H6 66 66 66 6H 6 6 6H66 6H 66 66H 66 66 6H 6H 6H 6 6 66Hm oH H6H 66H 66 66 H6 66 66 66 OH 6H6 6H 66H 66H 66 66 66 66 H6 66 6H 66 6H OEm3om 6 6H6 6HH6 6666 66Ho 6HH6 6666 66Ho 6HH6 6666 :oNHuom .oz “IMO mlmrm QIHAN HHOW COHuomnw oNHmIoHUHuHmm CH m 5mm .6coHuomHm omHmIoHoHpumm H606 cH mucoucoo mImo 6cm Iom .IHm .m oHQmB 69 :specific surface and/or particle size are not likely to be the factors controlling the relative availability of the 'various forms of inorganic P in soils. To obtain a clearer picture of the distribution of the various forms of inorganic P in the soil, the percentage distribution of Al—P, Fe-P and Ca-P in sand, silt and clay, and the distribution of Al—P, Fe-P and Ca-P within soil separates were calculated. Percentage Distributions of Al-P, Fe-P and Ca-P in soil Particle Separates The percentage distribution of Al-P, Fe-P and Ca-P in sand, silt and clay fractions was calculated by: ppm P of Al-P (or Fe-P or Ca-P) in a certain particle—size fraction ppm P of (Al + Fe + Ca)-P in the same particle-size fraction x 100. The results are listed in Table 9 and Figures 11, 12 and 13. For the Pewamo clay loam, the percentage distri— bution curves (Figure 11) of Ca-P and Fe-P show excellent mirror image relationships in the sand and silt fractions. The Al—P remains fairly constant throughout the profile in the sand faction, but it decreases somewhat with depth in the silt fraction. Ca-P predominates in this soil and the Al-P is the least abundant form. The difference between the Fe-P and Al-P in the sand fraction is greater than that in the silt fraction indicating the particle-size x Hm Ema CH CoHuomHm mNHmlmHoHuumm 6566 0CD CH mImO 6C6 Imm .IHm mo Esmv OOH Hm 8mm CH CoHuomum mNHmIoHUHuHmmICHmuHoo m CH m UHCmmHOCH ooHCu 6C6 mo oCOH ”66 66666660. 7O 6.66 6.6 6.6H 6.66 6.6 6.6 6.66 6.6 6.6 66 66 H.66 o.6 6.6H 6.66 6.6 6.6 6.66 6.6 6.6 6666 06 mEHm 6.66 o 6.6 H.66 6.o 6.H H.66 6.6 6.6 666 66 o.H6 6.6 6.6H 6.H6 H.H 6.6 6.66 6.6 6.6 64 66 6.66 6.6 6.6H 6.66 6.6 6.6H 6.66 6.6 o.6H 6666 06 6.66 6.6 H.6H 6.66 6.6 6.H6 6.66 6.6H 6.6H 66 66 6.66 6.6H 6.66 6.66 H.6H 6.66 6.66 6.6H o.6H 66 66 mmBmCO 6.66 H.66 6.66 H.66 6.66 6.66 6.66 6.66 6.H6 66 66 H.H6 6.66 6.66 6.66 6.66 6.66 o.H6 6.66 6.66 H6 06 H.6H 6.66 H.66 6.66 6.66 H.66 H.66 6.66 6.66 6Hs66 66 CHHHNVH 6.66 H.66 H.66 H.66 6.6 6.6 6.66 H.66 o.HH Ho.6666 6H 6.66 H.66 6.6H 6.66 6.66 6.6 6.66 6.H6 H.6 . 6666 6H 6.66 6.H6 H.6H 6.66 6.6H 6.6 6.66 6.6H o.6 66H6 6H 6.66 o.o6 6.6H H.H6 o.6H 6.6H 6.66 6.6H 6.6 6H6 6H 6.66 6.66 6.6H 6.66 6.6H 6.6H 6.66 6.66 6.6H 66 6H OEm3wnH .6 .6 .6 .6 .6 .6 .6 .6 .6 6-66 6-66 6-H6 6-60 6-66 6-H6 _ 6-66 6-66 6-H6 soNHsom .02 COHUOmHh >MHU COHuOmHh #Hflm — COHHUMHR Uflmm HHOW 6.6C0Huomnm mmHo UCm DHHm .tme CH m UHCmmHOCH mo COHusQHHumHt ommDCoouom .m oHQmB 71 .COHpomum mmHo HOV 6C6 COHuomum pHHm Any .COHuomHm pCmm Amy "oHHmoum OEmBmm mo moumnmmom oHoHuumm CH mlmo 6C6 mlom .mIH< mo COHusQHHume mmmucoouwm .HH onsmHm Hmo+om+Hmv 6 H6606 60 X Hmo+wm+H4v m Hmuou mo 6 Hmo+om+H¢9 m Hmuou mo 6 90H om om 0v ON OOH bm oo 0% ON OOH Om om _O¢ ON All a — H — q 6 \ss — _ - — o\a\fl A Nfilhm / o\ \ o\ a ., /. - .\ 1. - fix ,1 I 66 6H s, .. /. H. / _ ./ _ / H; / . w u I v..- . . - 6HIoH . _ T _ n : H... _ . 6 . I _. L l I 6HI6 o N a" N N .\ H b. x . \ . L .“ L H H L QIO E 3v 2: nHImU .766 I .l..I 6.H6 IIII I, Depth (inch) 72 relationship between the two phosphate compounds. In the clay fraction of the Pewamo soil, the distribution curves of Ca—P and Fe-P show mirror image relationships from the sur— face to 27 inches of depth but the difference between the two curves is relatively smaller than those for the sand and silt fractions. The Al-P remains constant and is the least abundant form of P from 0 to 27 inches of depth. However, in the layer of 27 to 42 inches, Al-P exceeds Fe-P and be- comes the second most abundant form of P in the clay fraction. This change of status may indicate that the chemi- cal weathering and transformation of the inorganic P remains in a more transitional stage in the clay fraction than in the surface horizons of the Pewamo soil. In the case of Karlin loamy sand (Figure 12), Fe-P is the least abundant form in the sand fraction, and remains constant in the three horizons, while Al-P decreases with depth and Ca-P increases with depth. Ca-P predominates in the lower horizons. The differences in the three P distri- bution curves in the sand fraction is probably not signifi- cant since the contents of Al-P, Fe-P and Ca-P (Table 8) are so low that experimental errors and the sensitivity of the analytical procedure may influence the shape of the curves. In the silt fraction, the relative abundance of the three forms of P is Fe-P > Al-P > Ca-P in the B2hir horizon (9—20 inches), Fe-Px Al—P/A’Ca-P in the D1 horizon (20—29 inches), and Ca- QFe-P > Al-P in the D2 horizon (29-34 inches). In 73 .CoHuome mmHo HOV tCm CoHuomum 6HH6 HHV .CoHuomum tcmm Hwy "oHHmonm CHHme mo moumummmm oHoHunmm CH mImO 6C6 mlom .mIHd mo COHuanupmHU ommuCoouom .NH ousmHm H6o+66+H> Al-P > Fe-P for sand, silt and clay fractions throughout the profile (Figure 13).In all three soil particle fractions,both Al-P'and Fe-P distribution curves are essentially mirror images of the Ca-P curves, although the best relationship is seen between Al-P and Ca-P. The difference between the Al-P and Fe—P curves is greater for the clay and silt fractions than for the sand fraction. All these relationships indicate that within the calcarous Onaway soil profile, the degree of in- organic phosphate weathering and transformation is less than that in the Pewamo and Karlin soil profiles, and that a greater degree of P transformation has occurred in the finer fractions than in the coarse fraction within the surface horizons. In the lower horizons of Sims clay loam, Ca-P is the dominating form of P in all three soil particle fractions. The differences between Al-P and Fe—P contents are very 75 .cofluomnm mmao on Ucm cofluomum uaflm AQV mo mmumummom maofluumm CH mlmo cam mnom 663.6635 6 66666 .66 x qcoauomnm comm Amv "waflwoun mmBmco .mlam mo coflpsnflupmflo mmmucoouom .MH ousmflm 60.66635 6 66666 66 .6 6686.635 6 66666 66.6. 66 66 66 66 666 66 66 66 om 666 66 66 66 6m _ . . x . . . . . _ , 6 . _ . . 66-: \ ~ s m m \ \. \H .d .. s 6 . .1 FHIGH . m 6 ~ .\ s .- \ V N .. .. 6 3 .\- 66-6 . ~ \ . . . _ \ \ . w . x. . . \. w . - a. - 6-6 x \x _. _ ’a r r _ — g a L BIO 36 A66 on mImU 6:66 l.|.l 6:66:11: Depth (inch) 76 small in the sand and silt fractions while the Al-P content is greater than that of Fe-P in the clay fraction. The percentage distribution curves of Al-P, Fe-P and Ca—P in the soil particle separates are essentially the differentiated forms of the curves obtained from the whole soil and generally in good agreement with each other in re— vealing the relationships of inorganic P weathering. Particle-size Distribution of .AlzP, Fe—P and Ca—P in soils In order to evaluate the inorganic P content in each of the particle separates together with the soil textural composition, the mg of P per 1,000 g of soil was calculated by the following formula: mg P per 1,000 g soil = x - y / 100 'where, x is the content of P in ppm in a particle-size fraction, and y is the percent content of a particle-size fraction of the soil. This value is essentially the same as the "F-value" introduced by Williams and Saunders (150). The percentage particle-size composition obtained from the large-scale separation (Table 7) was used for the calculation. The relative distribution of each form of P within the particle separates was calculated as follows: mg M—P in a given_particle-size fraction x 100 Total of mg of MrP in all particle-size fractions where, M—P is Al—P, or Fe—P, or Ca-P. 77 The particle-size distribution of Al-P in soils is given in Table 10. In the Pewamo profile, which has a texture of clay loam in the surface 10 inches and of silt clay loam in the lower horizons, more than 50 percent of the Al—P is present in the clay fraction of the soil throughout the profile, and less than 12 percent of the Al—P is in the sand fraction. The distribution of Al—P in sand, silt and clay fractions of the soil remains fairly constant with depth from surface to 27 inches. But, in the 27 to 42 inches layer, more than 80 percent of the Al—P is present in the clay fraction. The Karlin soil, a loamy sand topsoil changing to sand in the D2 horizon, has more than 50 percent of the Al-P in the sand fraction, with the remainder equally distributed in the silt and clay fractions in the B2 and D1 horizons. In the calcareous Onaway profile with a loamy textured top soil and a sandy textured subsoil, the distri- bution of Al—P is highest in clay, medium in silt and lowest in sand in upper Podzol sequum (0-16 inches). In the lower sequum (a Gray-WOoded profile), Al-P becomes equally dis- tributed in all particle-size fractions. In the lower portion of the Sims profile, a clay loam, more than 65 percent of the Al—P is distributed in the clay fraction. The particle-size distribution of Fe—P is given in Table ll. In the Pewamo profile, the Fe—P is the highest in 78 .HHOm CH mIH¢ Hmpou mo Rs 6.66 6.66 6.66 6.66 6.66 6.6 6.6 mo 66 6.66 6.66 6.66 6.66 6.6 6.6 6.6 6666 66 6666 6.66 6.66 6.66 6.6 6.6 6.6 6.6 666 mm 6.66 6.66 6.66 6.66 6.6 6.6 6.6 66 66 6.66 6.66 6.66 6.66 6.66 6.6 6.6 6666 66 6.66 6.66 6.66 6.66 6.66 6.6 6.6 66 66 6.66 6.66 6.6 6.66 6.66 6.66 6.6 66 66 5636660 6.66 6.66 6.66 6.66 6.6 6.6 6.66 «6 6m 6.66 6.66 6.66 6.66 6.6 6.6 6.66 66 66 6.66 6.66 6.66 6.66 6.6 6.6 6.66 66666 66 CHHHMVH 6.66 6.66 6.6 6.66 6.66 6.6 6.6 66.6666 66 6.66 6.66 6.66 6.66 6.6 6.6 6.6 6666 66 6.66 6.66 6.66 6.6 6.6 6.6 6.6 6.666 66 6.66 6.66 6.6 6.66 6.66 6.66 6.6 666 66 6.66 6.66 6.6 6.66 6.66 6.66 6.6 66 66 OEMme uuuuuuuuu guuuuunnu nunn6666 m 6666 \ 6 manuu: 6666 6666 6666 666 6666 6666 6:66 6666666 .62 6666 66066656666666 x. .66666 66 6-66 66 666666666666 6666-66666666 .66 66666 .. J In I: -66 @6666 79 .6606 CH mlmh 66606 HO $6 6.66 6.66 6.66 6.6 6.6 6.6 6.6 66 66 6.66 6.66 6.66 6.6 6.6 6.6 6.6 6666 66 6666 6 6.66 6.66 6.6 6 6.6 6.6 666 6m 6.66 6.66 6.66 6.6 6.6 6.6 6.6 66 66 6.66 6.66 6.66 6.6 6.6 6.6 6.6 6666 66 6.66 6.66 6.66 6.66 6.6 6.6 6.6 66 pm 6.66 6.66 6.66 6.66 6.66 6.6 6.6 66 6m hmBMGO 6.6m 6.66 6.66 6.66 6.6 6.6 6.6 66 66 6.66 6.66 6.66 6.66 6.6 6.6 6.6 66 66 6.66 6.66 6.66 6.66 6.6 6.6 6.6 66666 66 CHHHNM 6.66 6.66 6.66 6.66 6.66 6.6 6.6 66.6666 66 6.66 6.66 6.66 6.66 6.6 6.6 6.6 6666 66 6.66 6.66 6.66 6.66 6.6 6.6 6.6 6666 66 6.66 6.66 6.66 6.66 6.66 6.66 6.6 666 66 6.66 6.66 6.66 6.66 6.66 6.66 6.6 66 66 086366 uuuuuuuuu 6uuuuuuu: uuus6666 6 6666 \ 6 menus: >660 666m osmm Esm >660 666m Ucmm 6066603 .02 660m 6666666666666 6 .66606 C6 mlmm mo COHuonflnumHU 666616606666m .HH $6969 80 the clay fraction, intermediate in the silt fraction and lowest in the sand fraction throughout the profile. The distribution pattern changes little with depth. These re- lationships are similar to that of Al—P for this soil. In the Karlin profile, the Fe—P is equally dis— tributed in the sand, silt and clay fractions throughout the profile. Fe-P is highest in the clay fraction in the surface Ap horizon of Onaway soil and lowest in the sand fraction. It remains highest in the clay fraction but becomes equally distributed in the silt and sand fractions in the A2 and B2hir horizons of the Podzol upper sequum of the profile. The content of Fe-P is so low in the lower sequum of the pro- file that the percent distribution becomes less meaningful. In the Sims subsoil, the content of Fe-P is also low, however, it appears that the distribution of Fe-P is highest in clay and lowest in sand. The particle—size distribution of Ca—P in soils is given in Table 12. The sand fraction contains less than 14 percent of the Ca-P throughout the Pewamo profile. The rest of it is equally distributed in the silt and clay fraction in the surface horizons and becomes much higher in the silt than in the clay fraction in the lower horizons. In the Karlin profile, about 60 percent of the Ca-P is present in the sand fraction throughout the profile. 11‘ ”1+I-rfiwlfiy+u..—VWM mn6vla606666m .NH @6668 81 .6606 CH mlmo 66606 mo *6 6.66 6.66 6.66 6.666 6.66 6.66 6.66 66 66 6.66 6.66 6.66 6.666 6.66 6.66 6.66 6666 66 6666 6.66 6.66 6.66 6.666 6.66 6.66 6.66 666 6m 6.66 6.66 6.66 6.666 6.66 6.66 6.66 66 66 6.66 6.66 6.66 6.666 6.66 6.66 6.66 6666 66 6.66 6.66 6.66 6.66 6.66 6.66 6.66 66 66 6.66 6.66 6.66 6.666 6.66 6.66 6.66 66 66 AMBMGO 6.66 6.66 6.66 6.66 6.6 6.6 6.66 66 66 6.66 6.66 6.66 6.66 6.6 6.6 6.66 66 66 6.66 6.66 6.66 6.66 6.6 6.6 6.66 66666 66 CHHHNX 6.66 6.66 6.6 6.666 6.66 6.66 6.66 66.6666 66 6.66 6.66 6.66 6.66 6.66 6.66 6.6 6666 66 6.66 6.66 6.66 6.66 6.66 6.66 6.66 6666 66 6.66 6.66 6.66 6.666 6.66 6.66 6.66 666 66 6.66 6.66 6.66 6.666 6.66 6.66 6.66 66 66 08636m uuuuuuuuu 6uuunuunu nunn6666 6 6666 \ 6 menuuu >660 666m camm Esm >660 666m Ucmm CONHHOm .oz 660m 6666666666666 6 .66666 66 6-66 66 666666666666 6666-66666666 .NH GHQMB 82 There is more Ca-P in the silt than in the clay in the B2hir and D1 horizons and the reverse is true for the D2 horizon. In the Onaway profile, the distribution of Ca-P is similar to that in the Pewamo profile. In the surface hori- zons, Ca—P is highest in the clay fraction, intermediate in the silt fraction and lowest in the sand fraction; while in the A2 horizon of the lower sequum (a Gray-WOOded profile), Ca-P is equally high in the silt and sand fractions and low- est in the clay fraction. In the B2t horizon, Ca-P is high— est in the silt and lowest in the sand fraction. In the Sims subsoil, Ca-P is highest in the silt fraction, intermediate in the clay and lowest in the sand fraction. In considering the P status of a soil from the fer- tility point of view, the texture of the soil becomes im- portant. Since the more available forms of P are present in the clay fraction of the soil, therefore, a soil of heavy texture contains greater amounts of available P than a soil of light texture regardless of their total P contents. The sum of mg P per 1,000 g of soil in each particle separate for a particular form of P is generally in good agreement with the content of that form of inorganic P de- termined directly from P fractionation of the soil (Table 5). Errors may arise from the use of large-scale particle-size separation data for the calculation. Nevertheless, the 83 comparison of the two sets of data did not indicate any sig— nificant loss of P during the processes of dispersion and separation. Neither did any significant breakdown of the occluded P during the sonic vibration treatment appear to have occurred. GENERAL DISCUSSION AND SUMMARY Inorganic P minerals in arable soils are of two origins: (a) primary Pbbearing minerals, such as apatites; and (b) insoluble P compounds formed as a result of P fertilization, such as crystalline and colloidal phosphates of Al and Fe. During the course of soil chemical weathering under acidic condition and intensive vegetation, the primary Pb bearing minerals which are present in the coarse fraction of the soil are broken down into smaller particles by chemical, biological and mechanical forces, and thus are transfered into the finer fractions of the soil. The portion of P that is brought into the solution phase by the action of H+ ions may precipitate as colloidal AlPO4 and colloidal FePO4 in the clay fraction of the soil. The colloidal phosphates then crystallize to form variscite (AlP04°2H20) and strengite (FePO ‘2H20) or related compounds and in an extended period 4 of time with favorable environmental conditions, the com- pounds may enter into the coarser fractions of the soil as a result of crystal growth. For this reason, the relative abundance of the various forms of inorganic P in the soil has been considered as an index of the degree of chemical weathering of the soil. 84 85 It is well known that soluble P compounds added to soils as fertilizer are rapidly changed into insoluble forms. The process is commonly called fixation. Fixed forms of P are believed to be colloidal phosphates of Al and Fe and are present in the clay fraction of the soil. Again, they may crystallize to form variscite and strengite and grow into larger crystals in an extended period of time. The extent of both the P transformation and fix- ation reactions in soils is greatly dependent upon factors, such as, the concentration of H+ ion in the solution, sources of Al and Fe, and temperature and moisture conditions of the soil. It has been shown in the first part of this investi— and colloidal FePO gation that colloidal AlPO are much 4 4 better sources of P for the plant than variscite and streng- ite. The two colloidal phosphates are equally available and the two crystalline phosphates are almost completely unavail- able to the plant. Colloidal AlPO crystallizes at a much 4 slower rate than the colloidal FePO4 in aqueous system under the same temperature and pressure conditions. Based upon these findings, the colloidal form of Al—P and Fe-P in the clay fraction formed during the processes of P fixation and transformation can be regarded as readily available form of P for plant growth. Therefore, the "fixed form" and the "un— available form" of P are not unanimous terms. In other words, the fixed forms of P in soils may be available for 86 plant growth as long as they remain colloidal. Under this circumstance, the rate of crystallization of the fixed forms of P in the soil becomes the most important factor affecting their relative availability to plants. To confirm the conclusion derived from the study of the synthetic phosphate systems, an investigation of the actual distribution of the various forms of inorganic P in the soil particle separates was carried out. Since most of the commonly used dispersion procedures for soil mechanical analysis cannot be used satisfactorily for this purpose, a method was developed and reported in this thesis which may be used to separate the textural fractions from the soil samples without alteration of the P minerals in soils. Four Michigan soil profile samples were used for this study and the results can be concluded as follows: (a) The contents of all the three forms of inorganic P, namely, Al-P, Fe—P and Ca-P in the soil particle separates are in the order of clay > silt > sand for all surface soils studied. This indicates that particle size and/or specific surfaCe may not be the critical factors in controlling the relative availability of the various forms of inorganic P in soils. Therefore, it is the degree of crystallinity of the various inorganic P compounds that is most responsible for their relative availability. (b) In the lower horizons of the soil profiles studied, Ca-P tends to be most highly concentrated in the silt 87 fraction, intermediate in the clay fraction and lowest in the sand fraction. The result provides information in sup- porting the conception that Ca-P in soils is of primary origin. (c) Since clay fraction contains the major portion of the inorganic P in the agricultural soils, soil texture be- comes important when the P status of a soil is considered from the fertility point of view. The amorphous or colloidal form of Al-P and Fe-P which is readily available to plants, is included in the portion of the inorganic P which is present in the clay fraction. It has been reported by Scheffer, et al. (125) who worked on a group of surface soils from Germany, and by Hanely and Murphy (56) who studied a number of grassland surface soils from Ireland, that Ca-P in soils is highly concentrated in the sand fraction and decreases with the de— crease of soil particle size. However, the results from this investigation do not provide evidence to support their conclusion. Two reasons may account for this discrepancy. FirSt, the methods they used for the textural separation of the soil samples may dissolve certain P minerals in the soil, particularly, in~the«clay’fraction.A Second, the degree of . inorganic P weathering may be greater in the acid soils used in this study. The four soil profile samples used in this study were developed under a more acidic environment and have been under 88 intensive cropping operations. Thus, a great portion of the Ca-P which exists as primary minerals such as apatites in the sand fraction should have been transferred into the silt and clay fractions as a result of some chemical and physical forces during the process of weathering, such as dissolution and mechanical breakdown. A major part of the dissolved P reprecipitates as Al-P and Fe-P and the reactions mainly oc— cur in the clay fraction. Meanwhile, the added soluble P fertilizer is also fixed in the forms of Al-P and Fe-P in the clay fraction. The fact that all the three forms of inorganic P are most concentrated in the clay fraction in the surface soils studied strongly support the conclusion from the first part of this thesis that it is the degree of crystallinity, not the specific surface or particle size that is the most im- portant factor affecting the relative availability of the various forms of inorganic P in soils. Ca—P, due to its pri— mary origin, exists as highly crystalline minerals throughout the textural fractions of the soil. Thus, it is the least available form of inorganic P in soils under slightly acidic to alkaline conditions. Al-P, due to its relatively slow rate of crystallization in aqueous medium, remains highly colloidal in the clay fraction, and thus, it is the most available form of inorganic P among the three. The relative availability of Fe-P in acid soils, lies between Ca—P and Al-P. Owing to its relatively faster rate of crystallization 89 in aqueous medium, it exists in a higher degree of crystal- linity in the soil as compared with that of Al-P, although the major portion of the Fe-P still remains in the clay fraction. lO. BIBLIOGRAPHY Aderikhin, P. C. and Volkova, G. S. 1962. Phosphate absorption by individual mechanical soil fractions. Nauch. Kokl. vyssh. shkoly. Bid. Nauki. No. 4:196-201. Alban, L. A., Vacharotayan, S. and Jackson, T. L. 1964. Phosphorus availability in reddish brown later— itic soils. I. Laboratory studies. Agron. J. 56:555- 558. Anderson, W. R., Stringham, B. and Whelan, J. A. 1962. Secondary phosphates from Bingham, Utah. Am. Mineralo— gist. 47:1303-1309. Arlidge, E. 2., Farmer, V. C., Mitchell, B. D. and Mitchell, W. A. 1963. Infrared, X-ray and thermal analysis of some Al and Fe phosphates. J. Applied Chem. (London) 13:17-27. Aslyng, H. C. 1954. The lime and phosphate potentials of soils; the solubility and availability of phosphates. Yearbook 1954. Royal Veterinary and Agri. College, COpenhagen, Denmark. Babayan, G. B. and Gasparyan, O. B. 1962. Effect of drying soil samples on their content of easily soluble phosphoric acid. Izv. Akad. Nauk. armyan. SSR. Biol. Nauki. 15:75—80. Bache, B. W. 1963. Aluminum and iron phosphate studies relating to soils. I. Solution and hydro- lysis of variscite and strengite. J. Soil Sci. 14:113- 123. Bache, B. W. 1964. Aluminum and iron phosphate studies relating to soils. II. Reactions between phosphate and hydrous oxides. J. Soil Sci. 15:109-116. Ballif, J. L. 1965. Suspension criteria in mechanical analysis. Sci. Sol. 1:15-32. Bartholomew, R. P. and Jacob, K. D. 1933. Availa— bility of iron, aluminum, and other phosphates. J. Assoc. Off. Agr. Chem. 16:598-611. 90 ll. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 91 Basak, M. N. and Bhattacharya, R. 1962. Phosphate transformation in rice soil. Soil Sci. 94:258—262. Bates, J. A. R. and Baker, T. C. N. 1960. Studies on a Nigerian forest soil. II. The distribution of phos- phorus in the profile and in various soil fractions. J. Soil Sci. 11:257-265. Bauwin, G. R. and Tyner, E. H. 1957. The distri— bution of non-extractable phosphorus in some gray- brown podzolic brunizem, and Planosol soil profiles. Soil Sci. Soc. Am. Proc. 21:245-250. Beaton, J. D., Charlton, T. L. and Speer, R. 1963. Identification of soil-fertilizer reaction products in a calcareous Saskatchewan soil by infrared absorption analysis. Nature Lond. 197:1329-1330. Beckett, P. H. T. and White, R. E. 1964. Studies on phosphate potentials of soils. Part III. Some proper- ties of the pool of labile inorganic phosphate. Plant and Soil 21:253-282. Black, C. A. (Ed.). 1964. Methods of Soil Analysis. Part I. pp. 562—577. ASA Monograph No. 9. Bouyoucos, G. J. 1951. A recalibration of the hydro— meter method for making mechanical analysis of soils. Agron. J. 43:434-438. Bradley, D. B. and Sieling, D. H. 1953. Effects of organic anions and sugars on phosphate precipitation by iron and aluminum as influenced by pH. Soil Sci. 76:175-179. Bromfield, S. M. 1960. Some factors affecting the solubility of phosphates during the microbial decompo- sition of plant material. Aust. J. Agric. Res. 2:304— 316. Bromfield, S. M. 1964. (Relative contribution of Fe and A1 phosphates sorption by acid surface soils. Nature (London) 201:321—322. Cate, C. V., Huffman, E. D. and Deming, M. E. 1959. Preparation of crystalline ferric phosphate. Soil Sci. 88:130-132. Cecconi, S. 1960. Investigations on the phosphate en— richment of soil. I. an-calcareous soils. Agro— chimica 5:68-96. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 92 Chai Moo Cho and Caldwell, A. C. 1959. Forms of phos- phorus and fixation in soils. Soil Sci. Soc. Am. Proc. 23:458-460. Chakravarti, S. N. and Talibudeen, O. 1962. Phos- phate equilibria in acid soils. J. Soil Sci. 13:231— 240. Chang, S. C. and Jackson, M. L. 1957. Fractionation of soil phosphorus. Soil Sci. 84:133—144. Chang, S. C. and Jackson, M. L. 1957. Solubility pro— duct of iron phosphate. Soil Sci. Soc. Am. Proc. 21: 265-269. Chang, S. C. and Jackson, M. L. 1958. Soil phos— phorus fractions in some representative soils. J. Soil Sci. 9:109-119. Chang, S. C. and Chu, W. K. 1959. Radioactive assay of the availability of soil phosphorus in lowland rice soils. 3rth Japan conf. Radioisotopes,Tokyo: pP-936'941- Chang, S. C. 1962. Modification of fractionation of soil phosphorus. Agr. Chem. Bull. National Taiwan Univ. 1962 issue. pp. 28—31. Chang, S. C. and Chu, W. K. 1961. Fate of soluble phosphate applied to soils. J. Soil Sci. 12:286-293. Chang, S. C. and Liao, F. H. 1962. Separation of aluminum phosphate from iron phosphate in soils. Sci. 136:386. Chang, S. C. and Juo, S. R. 1963. Available phosphorus in relation to forms of phosphorus in soils. Soil Sci. 95:91-96. Chiang, P. F., Lu, R. K. and Ku, I. C. 1963. The con— tent of iron phosphates in paddy soils of southern China and their significance in the phosphorus nutri- tion of the rice plants. Acta. Pedol. Sin. China 11: 361-369. Chiang, C. T. 1963. A study on the availability and forms of phosphorus in paddy soils. I. The interre- lationship between available phosphorus and soil pH, Eh. Soil Fert. Taiwan. No. 61. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 93 Chiang, C. 1963. Changes of the form of fixed phos- phate and its availability in paddy soils. J. Sci. SOilo TOKYO. 38:18-22. Chu, W. K. and Chang, S. C. 1960. Forms of phosphorus in the soils of Taiwan. J. Agri. Assoc. China 30:1-12. Clark, J. S. 1955. Solubility criteria for the ex- istence of hydroxyapatite. Cand. J. Chem. 33:1696- 1700. Clark, J. S. and Peech, M. 1955. Solibility criteria F for the existence of calcium and aluminum phosphates - in soils. Soil Sci. Soc. Am. Proc. 19:171-174. Cole, C. V. and Jackson, M. L. 1951. Solubility equi- librium constant of dihydroxy aluminum dihydrogen phos- phate relating to a mechanism of phosphate fixation in soils. Soil Sci. Soc. Am. Proc. 15:84-89. Corbridge, D. C. E. and Lowe, E. J. 1954. I. R. Spectra of some inorganic phosphorus compounds. Chem. Soc. Jour. 493-502. Dahnke,‘W. C. and Malcolm, J. L. 1964. Phosphorus fractions in selected soil profiles of El Salvador as related to their development. Soil Sci. 98:33-38. Datton, J. D., Russell, G. C. and Siding, D. H. 1952. Effect of organic matter on P availability. Soil Sci. 73:173-181. Deming, M. E. and Cate, W. E. 1963. Preparation of Variscite. Soil Sci. 95:206-208. Edwards, A. P. and Brenner, J. M. 1964. Use of sonic vibration for separation of soil particles. Canad. J. Soil Sci. 44:366. Egan, E. P. Jr., Wakefield, Z. T. and Tuff, B. B. 1961. Low temperature heat capacity, entropy and heat of formation of crystalline and colloidal ferric phos- phate dihydrate. J. Phys. Chem. 65:1265-1270. Fife, C. V. 1959. An evaluation of ammonium fluoride as a selective extractant for aluminum bound soil phos— phate. Soil Sci.96:112—120. Fordham, A. W. 1963. The measurement of chemical po— tential of phosphate in soil suspensions. Aust. J. Soil Res. 1:144-156. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 94 Fried, M. and Shapiro, R. E. 1956. Phosphate supply patterns of various soils. Soil Sci. Soc. Amer. Proc. 20:471-475. Fried, M. and Shapiro, R. E. 1960. Soil—plant re- lations in phosphorus uptake. Soil Sci. 90:69—76. Fujiwara, A. 1950. Studies on the constitutions and manural effects of scarcely soluble phosphates. Part I-IV. Tohoku J. of Agr. Res. 1:129-149 and 223—231. Glenn, R. C. 1959. Flow sheet for soil phosphate fractionation. Agron. Abstracts. Nov. 16—20. Madison, Wisc. Goel, K. N. and Agarwal, R. R. 1959. Forms of soil phosphorus in genetically related soils of Kanpur in Indian Gangetic Alluvium. J. Indian Soc. Soil Sci. 7: 155-161. Goel, K. N. and Agarwal, R. R. 1960. Total and or- ganic P in different size fractions in genetically re- lated soils of Kanpur in the Indian Gangetic Alluvium. J. Indian Soc. Soil Sci. 8:17-22. Grigg: J. L. 1965. Inorganic phosphorus fractions in South Island soils and their solubility in commonly used extracting solutions. New Zealand J. of Agr. Res. 8(2):313-326. Hanley, K. 1962. Soil phosphorus forms and their availability to plants. Irish J. Agric. Res. 1:192- 193. Hanley, P. K. and Murphy, M. 1966. Phosphorus forms in particle size separates. Agron. Abstracts. ASA 1966 Annual Meetings. Stillwater, Oklahoma. Haseman, J. F., Brown, E. H. and Whitt, C. D. 1950. Some reactions of phosphate with clays and hydrous oxides of Fe and A1. Soil Sci. 70:257-271. Haseman, J. F., Lehr, I. R. and Smith, J. P. 1951. Mineralogical character of some iron and aluminum phos— phates containing potassium and ammonium. Soil Sci. Soc. Am. Proc. 15:76-84. Hawkins, R. H. and Kunze, G. W. 1965. Phosphate fractions in some Texas Grumusols and their relation to soil weathering and available phosphorus. Soil Sci. Soc. Am. Proc. 29:650-656. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 95 Hemwall, J. B. 1957. The fixation of phosphorus by soils. Adv. in Agron. 9:95-112. Hernando, V. and Losada, M. 1959. Study of the forms of phosphorus to be found in their dynamic equilibrium in the soil. Agrochemica 3:317—335. Hill, W. L., Caro, J. H. and Wieczorek, G. A. 1954. Surface area of natural and processed phosphates. J. Agr. & Food Chem. 2:1273-1277. Hirte, W. 1963. The effect of microorganisms on g phosphate mobilization in lOWbmoor soils. Zbl. Bakt. 116:485-501. Hoyos, A. and Garcia, R. 1959. Distribution of the different types of phosphorus in the soil fractions of various profiles. Agrochemica 3:191-210. Hoyos, A. and Garcia, R. 1963. Distribution of differ- ent forms of phosphorus in the mechanical fractions of some natural soils. An. Edafol. Agrobiol. 22:413-451. Hsu, P. H. and Jackson, M. L. 1960. .Inorganic phos- phate transformations by chemical weathering in soils as influenced by pH. Soil Sci. 90(1):16-24. Hsu, P.H. 1964. Adsorption of phosphate by aluminum and iron in soils. Soil Sci. Soc. Am. Proc. 28(4):474- 478. Hsu, P. H. 1965. Fixation of phosphate by aluminum and iron in acidic soils. Soil Sci. 99(6):398—402. Huffman, E. 0., Cate, W. E. and Deming, M. E. 1960. Rates and mechanisms of dissolution of some ferric phosphates. Soil Sci. 90:8-15. Huffman, E. O. and Taylor, A. W. 1963. Behavior of water—soluble phosphate in soils. J. Agr. Food Chem. 11:181—187. Jackson, M. L. 1958. Soil Chemical Analysis. Prentice Hall. Johnston, H. W. 1959. The solubilization of insoluble phosphates: 5. The action of some organic acids on iron and aluminum phosphates. New Zealand J. Sci. 2: 215—218. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 96 Kaila, A. 1963. Phosphorus conditions at various depths in some mineral soils. Maataloust. Aikak. 35: 69-79. Kaila, A. 1964. Fractionations of inorganic phos- phorus in finish mineral soils. Maataloust. Aikak. 36:1-13. Kaila, A. 1965. The fate of water—soluble phosphate applied to some mineral soils. Maataloust. Aikakausk. 37:104—115. Kanwar. J. S. and Grewal, J. S. 1959. Forms of phos— phorus in Punjab soils. J. Indian Soc. Soil Sci. 7: 135-142. Kittrick, J. A. and Jackson, M. L. 1954. Electron microscope observations of the formation of aluminum phosphate crystals with kaolinite as the source of aluminum. Science 120:508—509. Kittrick, J. A. and Jackson, M. L. 1955. Rate of phosphate reaction with soil minerals and the electron microscope observations on the reaction mechanism. Soil Sci. Soc. Am. Proc. 19:292—295. Kittrick, J. A. and Jackson, M. L. 1956. Electron- microscope observations of the reaction of phosphate with minerals, leading to a unified theory of phosphate fixation in soils. J. Soil Sci. 7:81-89. Kojima, M. 1963. Infra—red spectra of artificial iron minerals. J. of Science of Soil and Manure (Japan) 34:252-254. Kurtz, L. T. 1953. Inorganic phosphorus in acid and neutral soils. "Soil and Fertilizer Phosphorus." Chap. III. Agron. monographs Vol. IV. Academ. Press. Larsen, 5. and Court, M. N. 1960. Soil phosphate solubility. Nature London. 189:164—165. Larsen, S. and Court, M. N. 1960. Chemical potentials of phosphate ions in soil solutions. 7th Int. Cong. Soil Sci. Madison, Wisc. 1960. Larsen, S. and Widdowson, A. W. 1964. Effect of soil solution ratiO' on determining the chemical potentials of phosphate in soil solution. Nature (London) 203: 942. 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 97 Larsen, 8. 1965. The influence of CaC12 concentra— tion on the determination of lime and phosphate po— tentials of soils. J. Soil Sci. 16:275-278. Laverty, J. C. and McLean, E. O. 1961. Factors af- fecting yields and uptake of phosphorus by different crops. 3. Kinds of phosphate-native, applied and formed.. Soil Sci. 91:166-171. Lindsay, W. L., Peech, M. and Clark, J. S. 1959. Solubility criteria for the existence of variscite in soils. Soil Sci. Soc. Am. Proc. 23(1):12-18. Lindsay, W. L. and Stephenson, H. F. 1959. Nature of the reactions of monocalcium phosphate monohydrate in soils. Soil Sci. Soc. Am. Proc. 23(1):12-18. Lindsay, W. L. and Stephenson, H. F. 1959. Nature of the reactions of" monoca'ldium phosphate monohydrate in soils. II. Dissolution and precipitation reactions involving Fe, Al, Mn and Ca. Soil Sci. Soc. Am. Proc. 23:18-21. 6 Lindsay, W. L., Lehr, J. R. and Stephenson, H. F. 1959. Nature of the reactions of monocalcium phosphate monohydrate in soils. III. Studies with metastable triple—point solution. Soil Sci. Soc. Am. Proc. 23: 342—345. Lindsay, W. L. and Stephenson, H. F. 1959. Nature of the reactions of monocalcium phosphate monohydrate in soils. IV. .Repeated reactions with metastable triple- point solution. Soil Sci. Soc. Am. Proc. 23:4406445. Lindsay, W. L. and Moreno, E. C. 1960. Phosphate phase equilibria in soils. Soil Sci. Soc. Am. Proc. 24:177—182. Lindsay, W. L., Frazier, A. W. and Stephenson, H. F. 1962. Identification of reaction products from phos- phate fertilizers in soils. Soil Sci. Soc. Am. Proc. 26:446—452. Lindsay, W. L. anngeMent, J. D. 1961. Effectiveness of some iron phosphates as sources of phosphorus for plants. Plant and Soil 14:118-126. Mack, A. R. and Barber, 8. A. 1960. Influence of temperature and moisture on soil phosphorus. I. Ef— fect on soil phosphorus fractions. Soil Sci. Soc. Am. Proc. 24:381-385. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 107. 98 Mack, A. R. and Barber, S. A. 1960. Influence of temperature and moisture on soil phosphorus. II. Ef- fect prior to and during cropping on soil phOSphorus availability for millet. Soil Sci. Soc. Am. Proc. 24: 482-485. Mackenzie, A. F. and Amer, S. A. 1964. Reactions of iron, aluminum and calcium phosphates in six Ontario soils. Plant and Soil 21:17-25. Mandal, L. N. 1964. Effect of Time, Starch and Lime on the transformation of inorganic phosphorus in a water-logged rice soil. Soil Sci. 97:127-132. Manning, P. B. and Salomon, M. 1965. Forms of phos— phorus in soil after long continued fertilization. Soil Sci. Soc. Am. Proc. 29:421-423. Marais, J. S. 1922. The comparative agricultural value of insoluble mineral phosphates of aluminum, iron and calcium. Soil Sci. 13:355-410. Mare, Le. P. 1960. Observations on the phosphate po— tential of some tropical soils. Trans. 7th Intern. Congr. Soil Sci. 3:600. Marshall, C. E. 1964. The Physical Chemistry and Mineralogy of Soils. Vol. I. Soil Minerals. John Wiley & Sons. Melton, J. R. 1964. Availability of aluminum, iron and calcium phosphates in soils. M. S. Thesis, Michigan State University. Moreno, E. C., Brown, W. E. and Osborn, G. 1960. Solu— bility of dicalcium phosphate dihydrate in aqueous sys— tems. Soil Sci. Soc. Am. Proc. 24:94—98. Moreno, E. C., Brown, W. E. and Osborn, G. 1960. Sta- bility of dicalcium phosphate dihydrate in aqueous so— lutions and solubility of octocalcium phosphate. Soil Sci. Soc. Am. Proc. 24:99-102. Nye, P. H. and Foster, W. N. M. 1958. A study of the mechanism of soil-phosphate uptake in relation to plant species. Plant and Soil 9:338—352. Ojima, T. and Kawaguchi, K. 1959. The reaction be- tween difficultly soluble phosphate and hydrogen sul- phide. Bull. Res. Inst. Food. Sci. Kyoto Univ. 22:59- 67. -u" -.:- — w-.— 108. 109. 110. 111. 112. 113. 114. 115. 116. 117. 118. 99 Okuda, A. and Kawasaki, T. 1962. Fixation and re- lease of phosphate by and from soils. Soil Sci. and Plant Nutr. (Japan) 8:211-218. Olsen, S. R. 1953. Inorganic phosphorus in alkaline and calcareous soils. "Soil and Fertilizer Phosphor— us," Chap. IV. Agronomy Monographs. Vol. IV. Academ. Press. Omori, K. and Seki, T. 1960. Infra-red study of some phosphate minerals. Japanese Assn. of.Mineralogists, Petrologists and Economic Geologists J.V. 44 pp. 13- 17. Payne, H. and Hanna, W. J. 1965. Correlations among soil phosphorus fractions, extractable phosphorus and plant content of phosphorus. J. of Agr. and Food Chem. 13:322—326. Pirkl, J., Nbvozamsky, I. and Jedlicka, J. 1965. Aluminum phosphate as a source of phosphorus to plants. Ust. Ved. Inf. MZLVH Rostl. Vyroba 38:937-944 (Cz. R. E.) Polynov, B. B. The Cycle of Weathering (USSR). 1937. Translated into English by A. Muir. (London). Pustinger, J. V. Jr., Cave, W. T. and Nielsen, M. L. 1959. Infra-red spectra of inorganic phosphorus com— pounds spectrochimica Acta 15:909-925. Rai, A. K., Prasad, C. R. and Mandal, S. C. 1963. Ef— fect of liming on availability of phosphate in an acidic red loam soil. J. Indian Soc. Soil Sci. 11: 137—140. Rathje, W. 1960. On the mutual conversion of calcium- bound phosphoric acid and aluminum- and ferric-iron~ bound phosphoric acid in soil. Plant and Soil 13:159- 165. Salmon, R. C. 1965.' Changes in phosphate potential on rewetting arid-dry soil. Nature 205:316. Sanchez, C. 1965. Evaluation of certain chemical transformations of soluble fertilizer P applied to three Michigan soils. Ph.D. Thesis. Michigan State Univ. -243: :n r...- i 119. 120. 121. 122. 123. 124. 125. 126. 127. 128. 129. 100 Sannikova, N. M. 1938. Iron and aluminum phosphates as sources of phosphoric acid for plants. Trans. Sci. Inst. Fertilizers Insectorfungicides (USSR) 141:70- 76. Savant, N. K. and Ellis, R. Jr. 1964. Changes in re- dox potential and phosphorus availability in submerged soils. Soil Sci. 98:388-394. Scarseth, G. D. and Tidmore, J. W. 1934. The fix- ation of phosphates by soil colloids. J. Am. Soc. Agron. 26:138-151. Schachtschabel, P. and Heinemann, G. 1964. Relation between P-forms and pH of loess soils. Z. Pflanz. Dung. Bodenkunde 105:1—3. Scheffer, F. and Schulz, H. G. 1955. Preparation and properties of some iron phosphates. A contribution to phosphate fixation in soils. Z. Pflanz. Dung Boden- kunde 70:141-164. Scheffer, F., Ulbrich, B. and Hiestermann, P. 1956. Hydrothermol synthesis of FePO4-2H20 (Strengite). Z. Pflanz. Dung. Bodenkunde 75:135-143. Scheffer, F., Kloke, A. and Hempler, K. 1960. The forms of phosphorus in soil and their distribution be- tween the particle-size fractions. Z. Pflanz. Dung. Bodenkunde 91:240-252. Scheffer, F., Kloke, A. and Hempler, K. 1961. The uptake of phosphate forms from the soil and particle- size fractions by well rooted plants. Z. Pflanz. Dung. Bodenkunde 92:193—201. Schofield, R. K. 1949. Thermodynamic potentials of soil constitutents Ann. Rep. Rothamsted. Exp. Sta. 1949, p. 29. Sen Gupta, M. B6 and Cornfield, A. H. 1962. Phosphor- us in calcareous soils. I. The inorganic phosphate fractions and their relation to the amount of calcium carbonate present. J. Sci. Food Agric. 13:652-655. Sen Gupta, M. B. and Cornfield, A. H. 1963. Phosphor— us in calcareous soil. IV. Nature of and factors in— fluencing the fixation of added phosphate. J. Sci. Food. Agric. 14:873-877. 130. 131. 132. 133. 134. 135. 136. 137. 138. 139. 140. 141. 101 Serbanescu, I. and Blanaru, V. 1963. A study of the different forms of phosphorus in certain soils of Romania and some aspects of phosphorus fixation. An. Inst. Cent. Cerc. Agric. 30A:63-78. Shapiro, R. E. 1958. Effect of flooding on availa- bility of phosphorus and nitrogen. Soil Sci. 85:190- 197. Shapiro, R. E. 1958. Effect of organic matter and flooding on availability of soil and synthetic phos- phates. Soil Sci. 86:267-272. Shipp: R. F. and Matelshi, R. P. 1960. A microscopic determination of apatite and a study of phosphorus in some Nebraska soil profiles. Soil Sci. Soc. Am. Proc. 24:450-452. Singh, R. N. and Martens, D. C. 1965. Plant availa- bility of residual soil phosphorus compounds in David— son clay loam. Agron. Abstracts. ASA 1966 Annual Meetings. Columbus, Ohio. Smith, A. N. 1965. The supply of soluble phosphorus to the wheat plant from inorganic soil phosphorus. Plant and Soil 22:314-316. Smith, A. N. 1965. Aluminum and iron phosphates in soils. J. Aust. Inst. Agric. Sci. 31:110-126. Stelly, M. and Pierre, W. H. 1943. Forms of inorganic phosphorus in the C-horizon of some Iowa soils. Soil Sci. Soc. Am. Proc. 7:139-147. Susuki, A., Lawton, K. and Doll, E. C. 1963. Phos- phorus uptake and soil tests as related to forms of phosphorus in some Michigan soils. Soil Sci. Soc. Am. Proc. 27:401-403. Taylor, A. W., Gurney, E. L. and Lindsay, W. L. 1960. An evaluation of some iron and aluminum phosphates as sources of phosphorus for plants. Soil Sci. 90:25-31. Taylor, A. W. and Gurney, E. L. 1962. Solubility of amorphous aluminum phosphate. Soil Sci. 93:241-245. Taylor, A. W. and Gurney, E. L. 1962. Phosphate equi— libria in an acid soil. J. Soil Sci. 13:187-197. 142. 143. 144. 145. 146. 147. 148. 149. 150. 151. 152. 153. 102 Taylor, A. W., Lindsay, W. L., Huffman, E. O. and Gurney, E. L. 1963. Potassium and ammonium tarana- kites, amorphous Al phosphate and variscite as sources of P for plants. Soil Sci. Soc. Am. Proc. 27:148-151. Taylor, A. W. and Gurney, E. L. 1964. Solubility of variscite. Soil Sci. 98:9-13. TruOg, E. 1916. The utilization of phosphates by agricultural crops, including a new theory regarding the feeding power of plants. Wis. Agr. Expt. Sta. Bull. 41. Ulrich, B. 1963. Reactions of inorganic phosphates in relation to the lime potential. Z. Pflanz. Dung. Bodenkunde 100:97-102. Volk, V. V. and McLean, E. O. 1963. The fate of ap- plied phosphorus in four Ohio soils. Soil Sci. Soc. Am. Proc. 27:53-58. Wada, K. 1964. Phosphate equilibria in arable soils different in soil type and management. Soil Sci. and plant nutr. Japan 10:1-8. Wild, A. 1954. The concentration of phosphate in the soil solution. Trans. 5th. Intern. Congr. Soil Sci. Leopoldville. 2:500—504. Withee, L. V. and Ellis, R. Jr. 1965. Change of phosphate potentials of clacareous soils on adding phosphorus. Soil Sci. Soc. Am. Proc. 29:511-513. Williams, E. G. and Saunders, W. M. H. 1956. Distri- bution of phosphorus in profiles and particle-size fractions of some Scottish soils. J. Soil Sci. 7:90- 108. Winters, E. and Simonson, R. W. 1951. The subsoil. Adv. Agron. 3:1—92. White, R. E. and Beckett, P. H. T. 1964. Studies on the phosphate potentials of soils. Part I. The measurement of phosphate potential. Plant and Soil. 20:1-16. White, R. E. 1964. Studies on the phosphate potentials of soils. Part II. .Microbial effects. Plant and Soil 20:184-193. 154. 155. 103 Wright, B. C. and Peech, M. 1960. Characterization of phosphate reaction products in acid soils by the application of solubility criteria. Soil Sci. 90:32- 43. Yuan, T. L., Robertson, W. K. and Neller, J. R. 1960. Forms of newly fixed phosphorus in three acid sandy soils. Soil Sci. Soc. Am. Proc. 24:447-450. 03 42 4025 KITWHILEMMIflWlltfiIWflilMifilflWs