INFORMATION TO USERS This material was produced from a microfilm copy of the original document. While the most advanced technological means to photograph and reproduce this document have been used, the quality is heavily dependent upon the quality of die original submitted. The following explanation of techniques is provided to help you understand markings or patterns which may appear on this reproduction. 1. The sign or "target" for pages apparentiy lacking from the document photographed is "Missing Page(s)". If it was possible to obtain the missing page(s) or section, they are spliced into the film along with adjacent pages. This may have necessitated cutting thru an image and duplicating adjacent pages to insure you complete continuity. 2. When an image on the film is obliterated with a large round black mark, it is an indication that the photographer suspected that die copy may have moved during exposure and thus cause a blurred image. You will find a good image of the page in the adjacent frame. 3. When a map, drawing or chart, etc., was part of the material being photographed the photographer followed a definite method in "sectioning" the material. It is customary to begin photoing at the upper left hand corner of a large sheet and to continue photoing from left to right in equal sections with a small overlap. If necessary, sectioning is continued again — beginning below the first row and continuing on until complete. 4. The majority of users indicate that the textual content is of greatest value, however, a somewhat higher quality reproduction could be made from "photographs" if essential to the understanding of the dissertation. Silver prints o f "photographs" may be ordered at additional charge by writing the Order Department, giving the catalog number, title, author and specific pages you wish reproduced. 5. PLEASE NOTE: Some pages may have indistinct print. Filmed as received. Xerox University Microfilms 300 North ZMbRoad Ann Arbor, Michigan 43106 75-14,833 SRISEN, Manoowetaya, 1940ADSORPTION OF PHOSPHORUS BY FIVE MICHIGAN SOILS UNDER ANAEROBIC CONDITIONS. Michigan State University, Ph.D., 1974 Agriculture, plant culture Xerox University Microfilms , Ann Arbor, Michigan 481 OS ADSORPTION QP PHOSPHORUS BY FIVE MICHIGAN SOH A UNDER ANAEROBIC CONDITIONS Manoowetaya Srlsen A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Crop and Soil Sciences 197* ABSTRACT ADSORPTION OF PHOSPHORUS BY FIVE MICHIGAN SOILS UNDER ANAEROBIC CONDITIONS By Kanoowetaya Srisen Chemistry of P in flooded paddy soils, in lakes and in streams is important for rice cultivation in Thailand and for the control of P level in natural waters in the United States as well as other parts of the world. Flooded soils as well as air dried soils can adsorb applied P from dilute solutions. It is known that P availability will increase upon flooding. This study was designed to investigate the behavior of P adsorption by Michigan soils under anaerobic conditions and to examine the Langmulr adsorption isotherm as a method of evaluating P adsorption by flooded soils. Five Michigan soils varying in chemical and physical properties were used for the laboratory study. The anaerobic chamber with 1Q?6 Hg and 9Q/& Ng atmosphere was used to keep the flooded soils under anaerobic conditions. Preliminary studies showed that air dried soils adsorbed more P from 0.01 M CaClg than from a water solution. At high P concentration soils adsorbed more P from the 0.01 H CaClg solution where it was adjusted to pH 6 than when pH was not. The P adsorption reaction in 0.01 M CaClg, for both adjusted and unadjusted pH treatment, reached equilibrium faster than in water solution treatment. Manoowetaya Srisen Phosphorus adsorption isotherm showed that for air dried soils, subsoils adsorbed more P and with greater bonding energy than surface soils. Pine textured soils adsorbed more P than coarse textured ones. For coarse textured soils, subsoils adsorbed more P than surface soils, but for fine textured soils the difference between the amount of P adsorbed by A and B horizons was not significant, Langmuir adsorption isotherms showed that soils have at least two different components with respect to P adsorption, one involving tightly bound and the other involving loosely bound P. In flooded surface soils, the P adsorption behavior changes with the chemical changes upon flooding. The adsorption changes occurred mostly in the loosely P bound portion in the soils. The amount of P adsorbed increased with lower bonding energy of P adsorption upon sub­ mergence. Extractable Fe apparently played an important role in the change in P adsorption after flooding. The P adsorption in the loosely bound portion apparently Is responsible for the increase in the extractable P for flooded soils and may be considered to be of benefit for rice cultivation. For flooded subsoils, the change in P adsorption also occurred in the loosely bound portion, with increases in the amount of P adsorbed and with lower bonding strength. But the change was not as great as that which occurred in surface soils due to the lesser degree of reduction upon flooding* To my parents and Thai farmers. ii ACKNOWLEDGMENTS The author expressed sincere appreciation to his major professor, Dr, Boyd G. Ellis, for his valuable guidance, active assistance and enlightening suggestions throughout the course of his studies and during the preparation of this manuscript. Gratitude is expressed to the members of his guidance committee t Dr. Donald R, Christenson, Dr, Clifford J. Pollard, Dr. Anton E. Erickson, Dr. Charles E. Cress, and Dr. Eugene C. Doll, for their time and effort on my behalf. Appreciation Is also expressed to other memiters of the Crop and Soil Science Department for the many opportunities for Intellectual growth. The author deeply appreciates the financial aid provided by the Thai government, Finally, special thanks is expressed to Rajanee for her continued encouragement and patience. iii TABLE (F CONTENTS Page LIST OP TABLES vi LIST OP FIGURES Vli 1. INTRODUCTION 1 2. REVIEW OP LITERATURE 3 2.1 Soli Chemical Changes After Submergence 3 2.1.1 2.1.2 2.1.3 OxIdation-Heduction Potential and Soil Acidity 3 Manganese and Iron Related to Reducing Condition 14 Phosphorus and Anaerobic Condition 24 2.2 Phosphorus Adsorption Isotherms of Soil 3. 34 2.2.1 Theory of Adsorption Isotherm 35 2.2.2 Application of Adsorption Isotherm to P Adsorption 39 2.2.3 Factors Affecting P Adsorption 42 2.2.4 Correlation between P Adsorption and P Availability 46 2.3 Phosphorus Adsorption by Flooded Soil 48 MATERIALS AND METHODS 53 3.1 Materials 53 3.1.1 Soil Samples 53 3.1.2 Phosphorus Analysis 53 iv i Page 3.2 53 3.1.2*2 SnClg-HgSO^ Method 54 Methods 3.2.1 3.2.2 4. 3.1.2.1 Murphy and Riley or Ascorbic Method 56 A Study of the Methods Used in the Determination of P 56 Phosphorus Adsorption by Flooded Soils 60 3.2.2.1 Anaerobic Condition 60 3.2.2.2 Phosphorus Solution 60 3.2.2.3 Phosphorus AdsorptionIsotherm 61 RESULTS AND 4.1 Preliminary Studies 4.1.1 4.2 DISCUSSIONS 64 64 Study of Contact Time and Effect of Salt Solution 64 4.1.2 Effect of the pH of the Initial P Solution 69 4.1.3 Effect of Contact Time of P Adsorption from 0.01 M CaClg adjusted pH 6 74 Phosphorus Adsorption Isotherms ?6 4.2.1 Phosphorus Adsorption by Air Dried Soils ?6 4.2.2 Phosphorus Adsorption by Flooded Soils 93 4.2.2.1 Phosphorus Adsorption Isotherms for Flooded Surface Soils 93 4.2.2.2 Phosphorus Adsorption Isotherms for Flooded Subsurface Soils 112 5. CONCLUSIONS 128 6. LITERATURECITED 130 7. APPENDIX 138 v LIST OF TABLES Table Page 1. Reduction sequence as related to E h ................ *.... 2. Comparison of two analytical methods for P analysis in P adsorption studies ................................... 3. The pH of Increasing KHgPO^ 70 Langmulr adsorption parameters for air dried surface and subsoils..... 5* 57 concentrations in solutions of 0.01 M CaClg and HgO ................................. 4. 8 .......... .............. 84 Ratio of bonding energy constant and adsorption maximum of A and B horizons...... % 6. Amount of P adsorbed by anaerobic surface soils 7. Langmulr adsorption parameters for anaerobic surface soils .. 102 8. Effect of air leaks in P adsorption by flooded soils...... 9. Amount of P adsorbed by anaerobic subsoils............... 113 10. Langmuir adsorption parameters for anaerobic subsoils.......120 vi .......... 94 Ill LIST OF FIGURES Figure Page 1. F adsorbed calculated from Ascorbic and SnClg-HgSO^ methods for F determination.................... .............. 2. Diagram of anaerobic glove chamber 58 63 3. Effect of contact time, CaClg, and HgO solution on P adsorption by Locke A soil........ ....... 66 4. Effect of contact time, CaClg, and HgO solution onF adsorp­ tion by Hoytvllle A soil 6? 5* Effect of pH of F solution on F adsorption by Hillsdale A soil....................................... ...... 71 6. Effect of pH of F solution on F adsorption by Miami Asoil .. 71 7. Effect of pH of P solution on P adsorption by Hoytvllle A soil .... 72 8. Effect of shaking time on F adsorption from 0.01 H CaClg adjusted to pH 6 ..................................... 75 9. Amount of P adsorbed as a function of (P) by Hillsdale soils. ?8 10. Amount of F adsorbed as a function of (p) by Locke soils .... 78 11. Amount of P adsorbed as a function of (p) by Miami soils .... 80 12, Amount of F adsorbed as a function of (p) by Sims soils 80 13, Amount of F adsorbed as a function of (F) by Hoytvllle soils • 81 14-. Langmulr F adsorption Isothermsfor Hillsdale soils 15. Langmulr F adsorption Isothermsfor Locke soils....... .....86 vii 85 Figure Page 16. Langmulr P adsorption isotherms for Miami soils 17* ..... Langmulr P adsorption isotherms for Sims soils..... ....... 18, Langmulr P adsorption isotherms for Hoytvllle soils....... 87 88 89 19* Amount of F adsorbed by air dried and flooded Hillsdale A soils.......................................... 96 20, Amount of P adsorbed by air dried and flooded Locke A soils . 96 21, Amount of P adsorbed by air dried and flooded Miami A soils , 96 22, Amount of P adsorbed by air dried and flooded Sims A soils 23, Amount of P adsorbed by air dried and flooded Hoytvllle A soils.......... 24, 98 -99 Langmulr P adsorption Isotherms for flooded Hillsdale A soils........................................ 103 25* langmulr P adsorption isotherms for flooded Locke A soils 104 26. Langmulr P adsorption isotherms for flooded Miami A soils 105 27. Langmulr P adsorption Isotherms for flooded Sims soils... 106 28. Langmulr P adsorption isotherms for flooded Hoytvllle A soils. 107 29. Amount of P adsorbed by air dried and flooded Hillsdale B soils .............................................. 115 30. Amount of P adsorbed by air dried and flooded Locke B soils .. 115 31. Amount of P adsorbed by air dried and flooded Miami B soils .. 117 32. Amount of P adsorbed by air dried andflooded Sims B soils ,.. 117 33* Amount of P adsorbed by air dried andflooded Hoytvllle B soils ................ 118 34, Langmulr P adsorption isotherms for flooded Hillsdale B soils. 121 35* Langnuir P adsorption isotherms for flooded Locke B soils ,.,, 122 viii Figure Page 36 . langnuirP adsorption Isotherms for flooded Miami B soils .... 123 37* langmulrP adsorption Isotherms for flooded Sims B soils 38, LangmulrP adsorption isotherms for flooded Hoytvllle B soils. 125 ix 12*f INTRODUCTION Phosphorus has long been recognized as an Important element In plant nutrition. All plants require P for growth and development because P is a constituent of metabolism, and plays a key role In energy conservation and biosynthetic reactions In plants. Nutrient enrichment in natural water, a natural process which may be accelerated by man, is one of the main problems in water ecology. However, P Is most often considered the limiting nutrient for plant growth in lakes and streams (Ellis and Erickson 19691 Byers et al., 1973 a-). Because of the low natural levels of P and the nature of the source of P to the natural waters, prevention or reversal of eutrophicatlon through control Is generally more feasible with P than with other nutrients. Therefore the behavior of P in poddy soils and in lakes and streams are important for rice cultivation and the control of P level in natural waters. Soil has ability to adsorb P from dilute solution. The applications of adsorption isotherm equations to describe the behavior of P adsorp­ tion by soil have been Intensively studied (Olsen and Watanabe, 1957* Shapiro and Pried, 1959* Udo and Uzu, 1972* and Syors et al., 1973 b). A P adsorption equation appears to be useful in describing F availability to plants (Pried and Broeshart, 196?) and the correlation between P adsorption isotherms and plant growth also have been studied (Woodruff 1 and Kamprath, 1965)* The use of F adsorption Isotherms to describe and predict the reduction of F level In sewage effluent by soils has been reported (Ellis and Erickson, 1969)* However, the F adsorption study and Its application was limited to upland and dried soils. The solubility and availability of soil and applied F Increases upon flooding (Shapiro, 1958a and bf Patrick and Mahapatra, 1968; and Fonnaperuma, 1972). But applied F can also be adsorbed by paddy soils, muds, and lake sediments (Mandal and Das, 19701 Hesse, 1962; Harter, 19681 Shukla et al,, 1971). However, few of these studies used the P adsorption Isotherm to describe the behavior of F adsorption by an­ aerobic soils. The adsorption of P by submerged soils In Michigan has not been studied. The first objective of this study was to Investigate the behavior of F adsorption by Michigan soils under anaerobic conditions. The second objective was to determine if the langmulr adsorption Isotherm could be used to predict F adsorption by anaerobic soils. literature:review 2.1 Soil Chemical Changes After Submergence. Submerging or waterlogging a soil creates conditions markedly different from those of a well-drained soil. The most important chemical and electro-chemical changes that take place when a soil is flooded are l) decrease in oxidation-reduction potential (Eh) and change in pH value j 2) reduction of Fe and Mn with the accompanying increase in their solubilities f 3 ) change in availability of P and Si; 4) denitrification; 5) accumulation of ammonia; 6) generation of toxic organic compounds such as fatty acids, mercaptans, sulfides, amines and other unidentified substances; and ?) release of cations such as NH^~, Na+, K+, Ca+2 and Mg+Z from the colloidal complex into soil solution (XRRI, 1963; Ponnamperuma, 1965 )* Some of these chemical changes were reviewed and will be discussed In the following paragraphs. 2.1.1 Oxidation-ReductIon Potential and Soil Acidity. Reduction of the soil is one of the most important changes caused by flooding. ^ In the absence of 02t facultative and obligate anaerobes — O use NO^'t Mn(lV), Fe (Hi), S0^~ , dissimilation products of organic matter, C02, and even H as electron acceptors in their respiration, reducing NO” to Ng, Hn(iv) to Hn(ll), Fe(lll) to Fe(ll), SO^"2 to H^S, CO^ to CH^f N2 to NH^ and H to H^. Anaerobic respiration also produces substances that reduce soil components chemically (Ponnamperuma, 1972), The Intensity of soil reduction can be measured quantitatively as the oxidation-reduction potential (Eh) of the soil. Because of its diagnostic and interpretative value in the study of anaerobic condition 3 of environments, such as waterlogged soils, marsh soils, lake muds, and marine sediments, redox-potential (Eh) has been Intensively measured and Its ecological significance Indicated (Gillespie, 1920| Mortimer, 192**, Pearsall and Mortimer, 19391 Zobell, 19**6( Pierce, 1953f Jefferry, I960* and Aomlne, 1962). However, the high promise this property has for diagnostic and Interpretative purposes In rice production cannot be fully exploited without a clear understanding of redox-potentlal In waterlogged soils (Ponnamperuma and Castro, 196**). Because the redox-potential is systems, it is a mixed potential inmost natural very difficult to understand. The theory of redox- potential, both In pure and mixed systems, equilibrium potentials and nonequilibrium potentials was well explained by Bohn (1971). The redox-potentlal or electron availability affects the oxidation states of H, C, N, 0, S, Mn, Fe, Co and Cu in the systems. Any chemical reaction which Involves exchange of electrons will be Influenced byEh. Ponnamperuma (1972) comprehensively explained the measurement of the state of reduction — both intensity (redox-potentlal) and capacity (total concentration of reduction products). He reviewed and discussed the problems, advantages and disadvantages of Eh measurements, and also reviewed the reducing capacity measurement In natural systems, Including soils. Because the hydrogen Ion concentration affects Eh by direct partici­ pation In oxldatlon-reduction or by Influencing the dissociation of oxidant or reductant, Ponnamperuma and Castro (196**) proposed the Idea of dE/dpH expression to evaluate the reduction of soils. They noted some similarities between the potentials and dE/dpH values of the reduced 5 soils and those of bacterial suspension. Ponnamperuma (1972) noted that cU5/dpH varies with the system. Although most workers use - 59 m.v./pH unit (at 25 C) as the correction factort Patrick (i960 ) found that the experimental values ranged from about -60 for aerobic soils to as steep a slope as -232 for some reduced soils. This un­ certainty makes many potentials corrected to pH ? unreliable. But Ponnamperuma et al. (196?) have shown theoretically and experimentally that the dS/dpH slope for the solutions of reduced ferruginous soils after the peak of water-soluble Fe is -59 m.v./pH. This corrected value is also reported as a good conversion factor if the redoxpotentlal is an equilibrium potential (Aomine, 1962; Bohn, 1968, 1971) and Patrick and Mahapatra, 1968)1 Ehy - Ehx - 60(?-x) ... (1) where x is the pH of the soil suspension, and Eh is in m.v. Practically speaking. Eh is the single electrochemical property that serves to differentiate waterlogged soils from well-drained soils. Ponnamperuma (1972) stated that the low potentials (+200 to -400 m.v.) of submerged soils and sediment reflect this reduced state1 whereas, high potentials (+300 to +800 m.v.) of aerobic media reflect their oxidised conditions. However, Patrick and Mahapatra (1968 ) concluded that aerated soils have characteristic redox-potentlal in the range +400 to +700 m.v. 1 waterlogged soils exhibit potentials from +700 m.v. to as low as -250 to -300 m.v. In well-drained soils the Eh is difficult to reproduce because of a lack of poising (easy to change In Eh) and the range In Eh values is narrow. Expected Eh values corrected to 6 pH 7.0 are shown belowi .... .. I waterlogged soil aerated ooil iPonnamperuma,19?2) waterlogged soil (Patrick and Mahapatra,l9§§^te<* a 1-ado 1-- b— 1— ztta— 1 2 diffusion rates employed in well-drained soil cannot be used in waterlogged soils. Reduction of a submerged soil proceeds roughly In sequential reductions, and can be predicted by thermodynamics, (Ponnamperuma and 7 Castro, 19&*t Turner and Patrick, 1968), Many researchers studied and/or summarised these phenomena, Aomine (1962 ), reviewed the studies on Eh of paddy soils in Japan and noted that there are 7 steps of microbic metabolism in waterlogged soils t l) 0 ^ disappearin ancei 2 ) NO^ reduction! 3 ) Mn formation (all of these steps 19 occurred at the initial Eh of +600 to +500 m.v. )| 4) Fe formation (Eh +500 to +300 m.v,)| 5) sulfide formation and sulfate reduction (Eh 0 to -190 m.v,)j 6 ) Hg formation (-150 to -220 m.v,)| and ?) CH^ formation (-150 to -190 m.v,). This is reflected in the microbial succession — aerobic, facultative anaerobic and strict anaerobic — aftor submerging of the soils. As long as 0g Is present in the soil, other oxidized components of soil are relatively safe from biological and chemical reductions. After 02 has disappeared from a waterlogged soil, the need for electron acceptors by facultative anaerobic and true anaerobic micro-organisms results in the reduction of several oxidized components. If an energy source is available to micro-organisms NO”, higher oxides of Mn, hydrated ferric oxides and S0£ will be reduced (Patrick and Mahapatra, 1968). Og is the first oxidant to be reduced, and it becomes unstable within one day after submergence. The next component to be attacked is NO” but its reduction begins only after 0^ concentration has been reduced to a very low value (Breinner and Shaw, 1958| Greenwood, 1962| and Turner and Patrick, 1 968 ). Nitrate also retards the reduction of other redox components. It will stabilize the potential at +200 m.v. to +4-00 m.v. and retard the production of Mn+2, Fo+2( s”2, «nd CH^ in submerged soils (Ponnamperuma and Castro, 19641 Turner and Patrick, 1968). * Patrick (i960), showed that under controlled redox-potential conditions, NO^ In a waterlogged soil was unstable at Eh of +3*H) m.v. at pH 5.i. Turner and Patrick (1968), found that the Eh at which 0^ disappeared was in the range +320 to +3**0 m.v. (assuming the Eh/pH slope of -60 m.v. per pH unit with EJv, = +225 m.v.). Mn follows NO” in the reduction sequence. Hut its influence is weaker than that of NO” because it is insoluble in water and is used as an electron acceptor in respiration by only a limited number of micro­ organisms. However, added and/or native MnOg retards the decrease in Eh of flooded soils and prevents the accumulation of Fe+2 and other products (Ponnamperuma and Castro, 196*11 Ponnamperuma et al,, 1965). Io The next mineral system in reduction sequence is the Fe (OH)^-Fe system. But its influence on soil reduction is not as obvious as that of NO” or MnO,, due to its low standard potential. However, soils rich in Fe(lll) showed a slower reduction in Eh in the zone -50 to -200 m.v. (Ponnamperuma, 1965)* and the addition of Fe^O^ to paddy soil depressed CH^ production (Asami and Tarki, 1970). The reduction sequence that may be expected is shown in the following tablet Table 1. Reduction sequence as related to Eh.^ Component Detectability TS^ undetectable m.v. 330 °2 NO’ ♦i Mn Fe scv2 detectable 1/ Ponnamperum. (1972) n II It undetectable 2 2 0 20 0 12 0 -1 5 0 9 Studies of redox systems In waterlogged soils have been conducted in two ways: 1) the soil Is waterlogged and changes In Eh and concentration of reducible substances are determined* or z) the Eh of the waterlogged soil is controlled at a constant potential and the concentration of reducible substances are determined. According to Patrick (1966) one of the biggest limitations of the use of Eh for studying reactions in waterlogged soils has been the difficulty of maintaining the Eh of the soil at a particular value throughout an experiment. apparatus used for controlling the redox-potential. He designed an This system allows waterlogged soils to become highly reduced (as low as -200 to -280 m.v.). This controlled Eh method has been used to study many redox systems in waterlogged soils. (Patrick, I960* Patrick, 1964; Turner and Patrick, 1968* Cornell and Patrick, 1968* Gotoh and Patrick, 1972, 197*0. The details of their studies on the Influence of Eh and/or pH upon Fe, Hn, and P transformation will be discussed in the Fe, Mn and P sections. Ponnamperuma and his colleages studied the redox system by measuring Eh and reducible materials with time of submergence (XRRX, 1963* 196b, 1965» 1966* Ponnamperuma, 1 965 ). The quantitative treatment of redoxequllibria has been formulated. The iron hydroxide redox system has been intensively studied and described by Ponnamperuma et al. (1967). The manganese oxide redox system also has been studied and explained by Ponnamperuma et al. (1969). The pH value of waterlogged soils tends to become neutral. When acid soils ore submerged the pH increases* whereas that of calcareous and alkaline soils decrease, (IRRI, 1963* IRRI, 196b* Mahapatra, 1968* and Ponnamperuma, 1972). The tendency and the magnitude of pH change depends upon soil characteristics, such as the initial soil pH, active Fe and Mn con­ tents, organic matter content and cation exchange capacity (C.E.C.) of the soils, IRRI (1963) reported that In acid soils (pH *K^-5.8) high In active Fe contents, pH values decreased shortly for a few days after flooding and then Increased sharply to a maxima hetween 6,5 to 7.0 within 20 days of submergence. The slightly acid soils, which were low In active Fe but high In active Mn content, with Initial pH values of 6,2-6.6, Increased somewhat more slowly In pH to a steady maxima of 7.0-7,2. A soil with an active Mn content of more than 2,000 ppm gave a maximum pH of 7.5* The neutral and slightly alkaline soils showed slight decreases In pH a day after submergence, then resumed and maintained almost their initial values. Another acid soil, low In both active Fe and Mn but high in organic matter, showed more fluctuations in pH than the others and the pH Increased with the mean maximum pH value less than 6,5* Puttaswamygowda and Pratt (1973), demonstrated that pH of sodlc soils was decreased by the addition of straw and CaClg with accompanying increases in Ca , Na , Mg concentrations. Submerged anaerobic conditions have the largest effect on promoting these reactions. However, Olomu et al, (1973)* found that pH of soil solution extracted from calcareous soils (pH 7.^ - 7.9) changed little upon flooding. soils were high In C.E.C. and high In organic matter contents. These But coarse-textured soils exhibited relatively large changes In their pH values. It appears that the buffering capacity of the soils influenced the relative changes of soil pH upon flooding. 11 Lining the soil has no effect on pH change due to submergence, Mukhopadhyay et al. (1967)* found that with an acid soli with pH 4,4, the pH decxeased within the first 1? days and then increased and reached stability near pH 7.0 about 62 days after flooding. Heavily lined soil (pH changed to be 7.7 Initially), showed a decrease in Its pH value and stabilised at pH 7.0 within the same period of submergence. Ponnamperuma (1972) concluded that when an aerobic soil Is submerged, its pH decreases during the first few days, reaches a minimum, and then Increases to a very stable value of 6.7 - 7,2 a few weeks later. The overall effect of flooding Is to Increase the pH of acid soils and to depress the pH of sodic and calcareous soils. Draining and exposure to air reverses the pH changes in paddy soils, mangrove swamps, lake muds and bogs, and anaerobic soils, (Ponnamperuma, 1972). The mechanism of pH change Including quantitative Interpretation of pH change has been Intensively studied (IRRI, 1964). it was con­ cluded that the Increase In pH of an acid soil upon flooding Is caused by reduction of Fe and Mn; pH throughout submergence In calcareous soils Is regulated by the partial pressure of CO^ and the pH value of the soil solution of reduced soli lies between 6.5 and 7,0 regardless of initial pH and other properties of the soil. Ponnamperuma et al, (1966 a) concluded that the pH of alkali and calcareous soils decreased, while those of acid soils increased to a fairly stable value of 6.7 to 7.2 twelve weeks after flooding. At this time the solution pH values 12 were 6.5 to 7*0, The pH of the soli solutions were lower than those of the soil suspensions for the alkali and calcareous soils from the start of submergence, and for the acid soils after soil reduction. The increase In soil solution pH of acid soils was quantitatively related to the potential of the Fe (OH)^- Fe+2 system, while the decrease In pH of the alkali and calcareous soils was defined by the partial pressure of COg through the NagCO^-HgO-COg and GaGO^-HgO-COg systems, respectively. The pH of reduced-acid soils high In Fe appears to be determined by the Fe^OH^-HgO-COg system (Ponnamperuma et al., 1966 a,b| Ponnamperuma et al., 1967} and Ponnamperuma, 1972). Ponnamperuma (1972) noted that the decrease In pH shortly after submergence Is probably due to the accumulation of COg produced by respiration of aerobic bacteria, because COg depressed the pH even of acid soils. The subsequent Increase In pH of acid soils Is due to the reduction. The pH of submerged calcareous and sodic soils are lower than those of aerobic soils because of the accumulation of COg. The pH of alkaline soils Is also sensitive to COg changes as shown by Ponnamperuma et al. (1966 a). The important reduction reactions that occur in nature Involve the consumption of H ions. This means a decrease In acidity or net gain In OH* concentration (Ponnamperuma, 1972). Since most soils contain more Fo(in) oxide hydrates than any other oxidants, the Increase in pH of acid soils Is largely due to the reduction of Fe, (Ponnamperuma et al., 196?). — —2 The pH change Influences OH , C0^ ,3 —2 t P and 31 equilibria In submerged soils. These equilibria regulate the precipitation and 13 dissolution of solids, the sorption and desorption of ions, and concentrations of such ions or substances as A1+3 , Fe+2 , H^S » H^CO^, and undissociated organic acids. The details of these phenomena were dis­ cussed by* Fonnaraperuma (1972), 1^ 2.1.2 Manganese and Iron Related to Reducing Condition. When axeated soils axe submerged, Mn reduction from insoluble oxldzed forms to b lightly reduced forms occurs. Together with dis- appearance of 0^ and NO^, an increase in Mn+2 in the soil solution ” and on the cation exchange complex is one of the first measurable effects of reducing conditions caused by waterlogging (Patrick and Turner, 1968). Submerged soil solutions were different from aerated soil sol­ utions in that they contained high concentrations of Fe and Mn present as protobicarbonates. Submerged soil solutions were also high in Ca and Mg and contained H^S and other sulfides (Robinson, 1930). It was noted that the high concentration of Fe, Mn, Ca and Mg was caused directly by microbiological action on the organic matter which produced CO^ and other gases. This CO^ was mainly responsible for holding Fe, Mn, Ca and Mg in the solution. In the absence of organic matter the solubility of Fe, Mn, Ca and Mg was not increased under submerged soil conditions. It also found that toxic concentrations of Fe+2 and sul­ fides developed In a few days after submergence. Toxic concentrations of Mn developed somewhat more slowly. The main transformations of Mn In submerged soils are the reduction »p of Mn(lV) oxides to Mn (II), (resulting In increased solution Mn ), precipitation of MnCO^ and oxidation of Mn+2 diffusing or moving by mass flow to oxygenated interfaces in the soil (Ponnamperuma, 1972). There were marked differences among the soils in the pattern of Mn reduction. Soils with a high content of active Mn, regardless of pH and organic matter content showed a rapid. Increase in Mn+2 15 concentration during the first 3 0 days of waterlogging and than decline slowly thereafter. Soils low in active Mn increases slowly in Mn+2 concentration and reached maximum values which were very much lower than in soils with higher Mn content. The overriding factor that determined the kinetics of Mn reduction was the active Mn content of the soil (XRRI, 1963). IRRI (196*0, reported that within 1 to 3 weeks of flooding almost all of the EDTA-dithlonate extractable Mn present in the soils, except those low in organic matter, is reduced. They concluded that the reduction is both chemical and biological and precedes the reduction of Fe. The kinetics of water-soluble Mn+2 reflects the Influence of soil properties on the transformation of Mn. Patrick and Turner (1968), noted that Mn exists In at least four forms in waterlogged soils* 1) water soluble; 2) exchangeable* 3) easily reducible; and 4) residual Mn or that remaining after extraction of the other three forms. The first two forms are largely Mn+2 while the easily reducible fraction is composed of higher oxides of Mn. In addition to these forms, chelated Mn may also be an important form in some soils. Studying anaerobic silty clay soil with low organic matter, they found that with 25 days of submergence as the content of reducible Mn decreased the content of exchangabls Mn increased. These two fractions accounted for most of the total Mn present in the soil. Water-soluble Mn did not increase until after several weeks of waterlogging in that soil and accounted for only a very small fraction of the total soil Mn. Residual Mn was little affected by flooding. They also found that with controlled Eh of waterlogged soil samples, the abrupt changes occurred In 16 distribution of Mn between the easily reducible and the exchangeable fraction at a potential of about +400 m.v. Beginning at this potential there was a pronounced conversion of easily reducible Mn to exchangeable forms as the soil became more reduced.. The transition of Mn from easily reducible to the exchangeable form was greatest over the range +400 to +300 m.v, and was essentially completed at +200 m.v. Solution Mn also Increased In concentration at a potential below +400 m.v,* but not to the extent of the exchangeable fraction. Residual Mn was not affected by changes in Eh. For paddy soils* Gotoh and Yanashita (1966 ), reported that there was a rapid increase in Mn+2 content In the early stage of flooding) after that successive rises and falls In Mn+2 content were observed. The rapid Increase in Mn+2 content was promoted by the addition of organic matter. They concluded that the oxides of Mn comprising KnO^i Mn^O^, and Mn^O^ are easily reducible under strong reducing conditions upon waterlogging soils. The effects of organic matter level* time of Incubation and moisture content on extractable Fe and active Mn In acidic* neutral, calcareous and saline-alkaline soils were also reported, (Takkar* 1969). They found that Fe and Mn availabilities were Influenced by soil temper­ ature. On incubation at 60 percent water holding capacity* Fe Increased greatly after 35 days and Mn decreased after 7 days. These changes were accompanied with Increasing organic matter and with time in most soils except in acidic soils In which Mn was found to increase after 69 days. Under waterlogged conditions, marked increases of Fe and decreases of Mn were noticed after 7 days. The lower amount of Fe and the time lag In 1? Its release In calcareous soils were due to the presence of Iron oxide In highly crystalline forms In these soils. High content of free Fe and low pH resulted In higher amounts of extractable Fe In acid soils. Under waterlogged and high organic matter conditions a decrease in Eh and a high concentration of Fe+2 were observed. As reported before, soil pH tended to become neutral. Most flooded soils contain sufficient water-soluble Mn for growth of rice and Mn toxicity has not been shown to occur In flooded soils. But native or added MnO^ retards Fe+3 reduction and counteracts the adverse effect of excess Fe+2 and other reduction products. Yuan and Ponnamperuma (1966) found that MnO^ retarded the decrease In Eh of the soils. In spite of the slight pH Increase, it generally depressed the concentration of water-soluble Fe+2 and oxldlzable organic matter and Increased the concentration of water-soluble Mn+2 in soils. Nhung and Ponnamporuiua (1966) found that MnO^ at 1.0 percent markedly retarded the fall In Eh and reduction of S0^ -2 , slightly Increased pH, depressed the concentration of A1+3 and Fe+2 and oxldlzable materials, and markedly Increased the concentration of Mn+2 In the soil solution. Temperature also affects the kinetics of Fe+2 and Mn+2 In low-land rice soils, Cho and Ponnanperuma (1971), found that Fe+2 and Mn+2 Increased after flooding, reached peak values, and then declined. In the 38 > 20 C and 30 C regimes the peaks were early, short and sharp, while in the 20 C and 15 * 31 ^ 18 C regimes they were late, high, and broad. The transformation of Mn and Fe upon flooding were studied by Kandal (l961). He reported that Mn transformation was somewhat different 18 from that of Fe, It enters Into exchangeable complexes and also appeared In soluble form much earlier than Fe. The reducible Mn, comprising higher oxides such as MnOg, Mn^O^, and shoved a decrease for some time after the soil had been waterlogged and then began to Increase to a level that was more than at the start. Iron and Mn In the organic complex fraction was not affected to any appreciable extent by the flooding condition with or without the addition of organic natter. Ponnanperuma, et al. (1969)# noted that the chemistry of manganese oxides is very complex since i 1) Mn can exist In several oxidation states i 2) it forms non-stolchiometrie oxides with the metal In mixed valences statesi 3) higher oxides exist In several crystalline or pseudo crystalline states; and 4) the oxides form co-precipitates, solid solution and perhaps superstructures with iron oxides. Therefore the application of thermodynamics to the study of Mn systems In flooded soil is very difficult. They formulated the redox-equilibria in terms of pE, (Where pE - -log (e~) “ Eh/0.059 at 25 c), of the MnO^, Mn^O^, and Mn^O^ systems. In studies to test these formulas, they found that the observed data sup­ ported the hypothesis that the manganese oxides Involved In redox-equili­ bria in soli that undergo seasonal oxldatlon-reduction axe complex nonstoichiometric oxides of variable composition whose apparent standards free energy of formation axe considerably less than those of their Ideal compounds. In addition to reduction processes soil pH also has a marked effect on the activity of soil Mn, with a low pH readily bringing Mn into soil solution under aerobic conditions. A combination of low pH and restricted soil aeration, such as occurs In poorly-drained acid 19 soils t may sometimes result in the release of a large amount of Mn (Gotoh and Patrick, 1972). The distribution of different forms of Mn In a water-logged soil over a wide range of controlled Eh-pH conditions was studied. The results show that at pH 5 almost all of the soil Mn was converted from easily reducible to water-soluble plus exchangeable fraction even at an Eh as high as +500 m.v. In sharp contrast, at pH levels between 6 and 8 most of the conversion took place at a relatively lower Eh of +200 to +300 m.v. When the water-soluble plus exchangeable fractions were further divided Into two components, low pH and low Eh were found to Increase water-soluble Mn at the expense of exchangeable forms. Cation-exchange reactions were apparently Important In regulating the equilibria between water-soluble and exchangeable Mn, The observed pMn/pH slope and Eq were much lower at pH 6 to 8 than they calculated for pure equilibrium systems. They concluded that Eh and pH of flooded soils provide general control of Mn transformation which probably include both chemical and biological processes, the complexity of which makes it difficult to explain the reaction in terms of a simple chemical system. The reduction of Fe has important chemical consequencesi l) the concentration of water-soluble Fe increases| 2) pH increases! 3) cations are displaced from exchange sites % the solubility of P and 3i increases} and 5) new minerals are formed (Ponnamperuma, 19?2), He also summarized that reduction of Fe is a consequence of the anaerobic metabolism of bacteria and appears to be chiefly a chemical reduotion by bacterial metabolites, although direct reduction coupled with respiration may be involved. The kinetics of Fe(Il) follow roughly an asymptotic course. Five to 50 percent of the free iron oxides present in a soil may be 20 reduced within a few weeks of submergence depending on temperature, organic natter content, and the crystallinlty of oocldes. The lower the degree of crystallinlty, the higher Is the reduction percentage, Soil characteristics Influence the kinetics of Fe+2 In the soils, XRBI (l963 )t reported that, after flooding, Fe solubility Increased, but only a small amount of the reduced Fe was In the soil solution. The bulk of It was In the solid, phase as hydroxides, +2 carbonates and sulfides or exchange Fe . Clayey, latosollc, rice soils with high active Fe and organic matter contents (and also low pH) showed steep Increases In reduced Fe concentration within 30 days of flooding. The concentration reached maximum values of about 10,000 to 15*000 ppm and remained more or less constant. Soils with lower organic matter or active Fe contents, regardless of pH and texture, attained lower maxima, (6,000 to 9*000 ppm) and the rate of reduction Is slower than the first group. Soils low In Fe Irrespective of pH and organic matter content reach still lower maxima (500 to 4,000 ppm) at an even slower rate (Fonnanperuma, 1965). He suggested that the reduction of Fe in flooded soils Is favored by* l) the absence of substances at a higher level of oxidation, such as N0^ and MnO^i 2) the presence of readily decomposable organic matter| and 3) a good supply of active Fe, The pH of soil Is relatively less Important. The distribution of Fe+2 between the solid and solution phases Is controlled by pH, organic matter content, C.E.C. of the soils, and duration of submergence (Fonnamperuma, 19&5)* The amount of Fe+2 In the flooded soil solution also depended on 21 soil properties (XRRX, 1965) Ponnamperuna, 1965 ). Soil characteristics Influenced the kinetics of water-soluble Fe+2 more drastically than that of total Fe 1*2. Acid soils high in organic natter and Fe build * up Mater-soluble Fe+2 concentrations as high as 600 ppn within 1 to 3 weeks of subnergence and showed a steep roughly exponential decrease to levels of 50 to 100 ppn which persist for several months. Soils high in organic natter but low In Fe gave high concentrations that persist for several months. +2 In neutral and calcareous soils Fe rarely exceeds 20 ppm (Ponnamperuma, 1972). Gotoh and Yamashlta- (1966), found that the Increases in the amount of Mn , Fe and S- with consequent decrease in Eh was enhanced by the level of organic matter. Drainage caused a sharp decrease In the amount of reduced compounds, especially Fe+2 . This decrease Is favored by the addition of organic matter which leads to an Increase In soil poxes. They also suggested that Eh in paddy soils which lies between +200 and -200 m.v. would be controlled largely by the Fe system. Patrick (196**) found that Fe+^ predominated In the soil when Eh was above +200 m.v. The maximum concentration of Fe+3 extracted from soil was approximately 200 ppm and was fairly constant at Eh between +200 to +500 m.v. Fe+3 was detected only In trace amounts at an Eh of 0 m.v. or lower. The Fe+2 predominated at Eh of +200 m.v. and lower. The amount of extractable Fe+2 Increased with a decrease In Eh and ranged from a trace amount at +200 m.v. to 4,300 ppm at -200 m.v. One of the soil properties which has a narked effeot on Fe behavior 22 is Eh. Even though Eh of many natural systems is a mixed potential. (Bohn, 1971)* however, Hem (i960) found that the behavior of Fe in natural systems is to a large extent predicted by fluctuation in Eh and pH in the environment. While the physico-chemical relation to flooded soils are rather complex, chemists in the past two decades have correlated the formation of Fe compounds with Eh and pH (Eriksson, 1952* Jeffery, I960* Ponnamperuma et al., 1967* Collins and Buol, 1970* and Bohn, 1971). Most of these studies of Fe behavior have been restricted to a narrow range of Eh and pH. Gotoh and Patrick (197*0, conducted the studies over a wide range of closely controlled redox-potential and pH conditions and found that the increase in water-soluble and exchangeable Fe was favored by a decrease in both redox-potential and pH. The critical Eh values for Fe reduction and consequent dissolution was about +300 m.v. at pH 6, +100 m.v. at pH 7 and -100 m.v. at pH 8. While at pH 5 appreeialbe reduction occurred at +300 m.v. The distribution of water-soluble and exchangeable Fe fractions was highly pH dependent, with a decrease in pH at a given Eh increasing the relative amount of Fe+2 in the soil solution at the expense of that on the exchange complex. A thermo­ dynamic approach to the equilibria between solid phase ferric oxyhydroxide and a water-soluble species of Fe, (Fe+2 ), indicated that it +2 was largely governed by the Fe -Fe(0H)^ system in which the ferric oxyhydroxldes was a mixture of geothite and amorphous material. Because flooded soil is a very complex system, the quantitative treatment of redox-equilibria in waterlogged soil is very difficult. Ponnamperuma et al. (196?), discussed this difficulty. They formu­ lated the Fe (ill) - Fe(n) hydroxides equlllbira, and found that the 23 test data were closely related to hypothetic equations. The Increase in concentration of water-soluble Fe can. In most soils, be related to the Eh and pH of the Fe+2-Fe(OH)^ system. The details of these subjects were well explained by Ponnamperuma (1972), Ponnamperuma et al. (196?), and Gotoh and Patrick (l97^). The dynamics of the concentration of Fe I +2 (and Mn ) affects the concentrations of other cations. This is clear from the similarity of the kinetics of (Ca+2 + Mg+2 + Na+2 + K+ + NhJ) to that of (Fe+2 + Mn ). These dynamics were well discussed by Ponnamperuma (1972). 2b 2.1.3 Phosphorus and Anaerobic Condition. P like N, can exist in valence states from +5 to -3. Phosphate in anaerobic media can be reduced to form phosphite, hypophosphite and phosphine. But the main transformation of P in anaerobic soils is the mobilization of orthophosphate Ions (Ponnamperuma, 1972). Many studies have shown the Increase in solubility and availability of P caused by flooding soils (Gasser, 195^1 Shapiro, 1958 a, hf Mitsui, 1960i Chiang, 1963 at and Mahapatra, 1968)* The amount and the pattern of solubility of native P increase upon flooding depends on soil characteristics, including amount and kind of P compounds, organic matter content, texture and acidity of soil. It was reported (IRRI, 196*0 that the increase in available P on flooding was highest and the subsequent decline least in coarse textured calcareous soils low in active Fe and medium in organic matter. Soils high in organic matter and low in active Fe showed moderate increases in water-soluble P| strongly acid soils, high in active Fe showed moderate increases followed by a quick decline to levels even below the initial, Ponnamperuma (l965 ) reported that the largest increase in Morgan's extractable P was observed in neutral soils with a moderate content of organic matter. On the other hand for soils low In organic matter, P concentration in the soil solution hardly increased. It was shown (IRRI, 1966) that the increase in P availability was highest and most stable in sandy soils high in organic matter and less in acid soils low in organic matter. Recently Ponnamperuma (1972) concluded that the P concentration in solution increased on flooding and the peak values were highest in sandy soils low in Fe, moderate in acid sandy soil3 25 low in Fe, still smaller in nearly neutral clays and smailsst in acid clays high in Fe, When soils are submerged, not only native P but also P added prior to flooding increase in solubility and availability. This Increase is enhanced by organic natter and/or lime application, Gasser (1956) noted that waterlogging acid soils caused an increase in the dilute acidsoluble PO^. Addition of glucose, potassium-citrate, or green manure +2 increased both Fe and dilute acid-soluble PO^, He concluded that the Increase in P solubility is due to the reduction of ferric phosphate to ferrous phosphate. He proposed that in an acid soil with high Fe con­ tent, ferric phosphate should be considered as a source of P to rice, Mitsui (I960) stated that the Increase in P availability by submergence is mainly due to the combination of PO^ with Fe+2 instead of Fe ^ as in upland soils. In anaerobic conditions, solubility of P on acid soils proceeds by ordinary solution, while an alkaline soils hydrolysis of Fe-phosphate, a kind of exchange between PO^ OH", occurs first. Flooding soils caused an Increase in the availability of native P, applied F and N to rice. Yield of rice, P and N uptake on both lowland and upland soils were Increased by flooding (Shapiro, 1953 a). It was noted that submergence increased the availability of native P, but had no apparent effect on that of native N. Applied P and N were both utilized more effectively under flooding conditions. The Increase in availability of P could be due to reduction conditions or hydrolysis or both. Broeshart, et al. (19&5) found that In rice soils with absence of freeCaCO^, the available P increased upon submergence. The availability of fertilizer F was similar under lowland and upland conditions but that 26 the efficiency of use is greater under flooded conditions. Terman et al. (1970) conducted a series of green-house pot experi­ ments with flooded rice on a soil which was low in available P for upland crops. They found that a marked yield response of rice was obtained, but the maximum yields were obtained at much lower rates of applied P than was true for most upland crops. Applied P decreased with liming of the soils and with increasing levels of acid-soluble soil P. Granular, water-soluble P sources were most effective. The P in Fe-P form was more available than that in Al-P form in flooded soils. Both forms were more available in fine particles than as granules and in colloidal form than as fine crystals, Chakravarti and Ghoshal (1968) mixed two acid soils with 5Q& dried grass and added Fe-P, Al-P, and Fe-P plus Al-P. After the samples were flooded and Incubated for 53 days, the amount of P released was found to be greatest in the Fe-P treatment, followed by Fe-P plus Al-P, Al-P and control in decreasing order. Shapiro (1958 b) found that availability of natural P and applied synthetic Fe-P and Al-P increased in acid soils upon flooding. Applica­ tion of P-free cellulose increased the availability of native and applied P. He noted that the increase in P availability came from both Fe and A1-phosphate but that Fe-phosphate was affected to a much greater degree. The Increase in available Fe-phosphate was due to the reduction brought about by the addition of organic matter and flooding. The increase in A 1-phosphate availability was due to the chelation reaction. His data showed that the availability of applied amorphous Fe and Al-phosphate were greater than that of the crystallized ones in both flooded and unflooded soil conditions. Transformations of P after flooding soils have been studied by 27 many workers. Furukawa and Kawaguchi (1969) found that organic P decreased up to 21 percent upon submerging paddy soils samples for 2 weeks at *K) C. This decrease correlated well with an Increase In Bray2 extractable P and was attributed to mineralisation of organic P, This increased mineralisation was thought to be due to the enhanced solubility of Fe and/or Al salts of inositol hexaphosphoric acid, the prominent organic P constituent in the soil. Mandal (196*+) tested the effects of starch and lime on the trans­ formation of P in waterlogged soils and found that submerging a soil, only slightly increased acetic-acid soluble P and slightly decreased FeP with no change in Ca-P and Al-P fractions. But when starch was added, a considerable increase in acetic-acid soluble P and a decrease in the Ca-P fraction occurred. He stated that the release of large quantities of CO^ formed during starch decomposition may have caused tricalciim phosphate to convert to more soluble dl-and mono-calcium phsophates. Although a large amount of Fe was formed, the Fe phosphate fraction showed no decrease and Al-P also showed no change. With the addition of lime, Fe-P decreased considerably and Al-P decreased slightly. This decrease might be due to the hydrolysis of Fe and Al-P due to liming. Ca-P was increased appreciably, the effect seemed to be a conversion from Fe-P to Ca-P upon flooding. He suggested that in acid soils having most J.T of their inorganic P in the Fe form, liming followed by the addition of organic matter will result in a few days in an increase in P under water­ logged conditions. Hahapatra and Patrick (1969) showed that in air dried soils, the more acid soil had a greater fraction of Al- and Fe-P while Ca-P was 28 highest In the less acid soils. After flooding these soils for 2 months transformation of P occurred. The Al-P fraction Increased except In finer textured soils. Fe-P also Increased In most of the soils except onei and generally increased more than the Al-P fraction, Ca-P did not change much as a result of flooding, while the xeductant soluble Fe-P was greatly decreased. They suggested that the Fe-P Increase was due to the conversion from reductant soluble Fe-P and other Inorganic fractions. They also studied the transformation of P added as A1P0^.2I^0( FePO^.SH^O and CaCHgPOj^Jg.H^O under optimium moisture and waterlogged conditions, and found that after 2 months of Incubation regardless of the source of added P, P In the soli was present as Al and Fe-P. Little of the added P was present as Ca-P. They concluded that waterlogged soils definitely accelerated the conversion of soluble P to the Fe-fraction. They also found that after flooding the soils for 55 days, the amount of Bray-2 extractable P increases an average of 53 percent. Mul­ tiple regression analysis show that water soluble and loosely bound P, Al- and Ca-P were highly correlated with the amount of phosphate extracted In air dry samples, while Fe-, Ca- and Al- phosphate apparently account for most of the phosphate extracted under waterlogging condition. The relationship between Eh of flooded soils and P availability and solubility also have been studied. Savant and Ellis (196*0 found that Eh decreased rapidly within 15 days of submergence In acid soils and within 30 days In calcareous soils. this period In both soils. Organic natter enhanced the Eh drop during But at field capacity (F.C) the Eh of the soils was not affected by the addition of organic matter. However, Eh drop was affected by organic matter at two times F.C. on acid soils. 29 At F.C., changes In availability of native F varied less than changes in availability of applied P. The over all availability of both native and applied F as a function of tine aret P available at F.C. <<* at 2F.C submerged condition In acid soils. But in calcareous soils, the avail­ ability of applied P fluctuated widely and soil moisture Inconsistently affected P availability. Increases in availability of native P were more closely correlated with a decrease in Eh than was the increase In avail­ ability of applied P. A negative linear relationship between available P and Eh was distinct in acid soils. But the relationship was not evident in the calcareous soils. Hayes and Phillips (1958) reported that the Eh has little or no effect upon the level of F in the soil solution. They feel that the bio­ logical system was the controlling factor and that the Inorganic chemical system was secondary in importance. After waterlogging a series of soil samples and administering antibiotics to quench the biological system, 10 out of 12 of the soil samples showed an increase in P under reducing con­ ditions ; but in h out of 10 the difference was less than 5%. They con­ cluded that their results indicated the minor influence of Inorganic reducing system upon P levels in solution. Chiang (1968) noted that soils rich in organic matter showed rapid lowering of Eh and increased formation of organic acids and gases upon waterlogging. Disintegration of soil aggregates and lowered Eh values caused an increase in F solubility. Fe-rich soils. The effects were more significant in H^S and the organic acids, especially acetic acid, increased F solubility in Fe and Al rich soils, whereas CO^ increased Ca-P solubility. 30 Under controlled Eh level of submerged soli, Patrick (1964) found +3 +2 that extractable Fe and Fe Here very sensitive to changes in Eh of the soil. Fe+3 predominated at Eh above +200 m.v. and extractable Fe was mostly Fe+2 below +200 m.v. At the same time he found that the extractable P Increased more than three fold between Eh +200 and -200 m.v. The sharp break In the P release curve at +200 m.v. is the same point at which Fe+3 begins to be reduced, indicated that the conversion of P to an extractable P form Is dependent upon the reduction of Fe+^ compounds in the soil. The effect of Eh and pH on the reduction of strongite (FeP0^.2H20) In a pure system to more soluble compounds has been studied by Williams and Patrick (1971). Under controlled Eh and pH conditions they found that at pH 4.0, Eh values of less than +100 m.v, have a marked effect on the solubility of strenglte and reaches rapid dissolution at Eh about -240 m.v, .it higher pH values (> 5.°), however, even with considerably lower Eh value (i.e. -240 m.v.), rapid dissolution of strenglte did not take place. This suggests that a critical H concentration is required before added electrons chemically reduce Fe-P to more soluble reduced forms. They concluded that the results are in contrast with those described for a flooded soil system (Patrick, 1964) where the amount of reductant-soluble P Increase rapidly at Eh of less than -200 m.v. and within the range of pH 6 to 7. This difference may Indicate the presence in soils of P compounds other than FeP0^.2H20 which are susceptible to reducing condition or may simply reflect the well known ability of various soil micro-organisms to overcome the energy barrier necessary to change the chemical state of sparingly soluble Fe compounds. 31 Patrick et al. (1973) investigated the effects of Eh and pH on the reduction of FeFO^.SH^O in flooded soils. They controlled Eh of the soil suspension (2 5 g of soil and 80 g of HgO) at +3 0 0 , +100, -100 and -2 5 0 m.v. in combination with pH values of were incubated for 3 days. 5, 6, 7and 8, respectively. The samples eg Then fine FePO^^f^O was added and additional incubation continued for 7 to 10 days. They found that both Eh and pH had a great effect on the dissolution of FeFO^^HgO in flooded soils. At the highest conbination of Eh and pH (pH 8, Eh + 300 m.v.) there Is no dissol­ ution of FePO^^HgO. At all pH levels, a decreasein Eh caused an increase in the amount of extractable Fe and F0^. The amount of both were greatest at the lowest pH values (pH 5)* Eh showed similar trendt at lowest Eh (-250 m.v.), a decrease in pH caused a large increase in Fe and PO^ diss­ olved, but at +300 m.v. pH has much less effect on the amount of Fe and P dissolved. They concluded that FeFO^^HgO partially dissolves under reducing conditions such as exist in flooded soils deprived of Og. A closed re­ lationship existed between the amount of FePO^. 2^0 dissolved and both Eh and pH of the soil suspension. There is a marked interaction between acidity and Eh, with FePO^.ZH^O reduction and solubilization being more pronounced under conditions of low pH combined with low Eh. The dissolution of complex ferric phosphates (CaFe^H^CPO^)^. 8^0 and KFe^Kjj^FOi^g.ifHgO) under controlled Eh and pH was also studied by Williams and Patrick (1973). In a pure chemical system study they found that calcium ferric phosphate (CFP) dissolved incongruently to release Ca and P in a molar ratio of Ii2 and solid FePO^ was formed. At pH 4, Ca, Fe, and F in solution increased with each successive lowering of Eh. 32 Between pH 5 and 6» 'the amount dissolved remained constant until an Eh less than -200 m.v. was reached. Below an Eh of -200 a.v. further P was released due to FePO^ dissolution. Potassium ferric phosphate (KFP) crystals dissolved rapidly with complete release of K. At pH 4 an increase in P was accompanied by an increase in soluble Fe at all Eh values. At pH 5 to 7 the amount of P in solution remained constant except for a slight increase at pH 5 and Eh -300 m.v. The P concentration corresponded closely to total dissolution of the KFP crystals, leaving an insoluble FePO^ residue. They concluded that the dissolution of CFP was retarded probably due to the precipitation of FePO^. The solubility of precipitated FePO^ differed slightly from that observed with crystalline FePO^.BH^O and at pH 6 and 7# Fe+^ hydroxides apparently entered into the reaction. The mechanism of the Increase in P solubility upon flooding in acid soils was summarised ast l) hydrolysis of Fe(lll) and A 1-phosphate j 2) release of F held by anion exchange on clay and hydrous oxides of Fe(lll) and All and 3) reduction of Fe(lll) to Fe(ll) with liberation of sorbed and chemically bound P. The first two reactions are due to increases in alkalinity upon soil reduction. In alkaline soils an in­ crease in solubility of P is caused by the increase in acidity of these soils upon flooding (for the solubility of hydroxyl apatite increases as pH decreases), (Ponnamperuma, 1972). Patrick and Mahapatra (1968) and Patrick et al. (1973) explained the mechanism of P liberation on flood soils ast 1) the reduction of highly insoluble Fe+3 to the more soluble noncrystalline formi 2) the displacement of P0^ from Fe+3 and Al-P by organic anions produced i 33 in anaerobic flooded soils; and 3) the hydrolysis of Pe+^ as a result of the almost inevitable increase in the soil pH which results when an acid soil is flooded. In addition to these major mechanisms, the release of occluded P by reduction of hydrated ferric oxide coating, hydrolysis of A1-phosphate and anion (phosphate) exchange between clay and organic anions may also occur. 3^ 2,2 Phosphorus Adsorption Isothenna of Soil • Applied F In soils is retained and Immobilized by P fixation pro­ cesses. Two mechanisms, adsorption and precipitation, are considered to be responsible for this fixation. Precipitation of P occurs when phosphate ions in the solution react with components of the soil such as Al, Fe and Ca to form slightly soluble crystallized components.The reaction occurs when P concentration Is sufficiently high and the solubility products are higher than the solubility product constant of the respective P compounds. But P can also be adsorbed by reactive sites existing on the surface of soil particles. Adsorption differs from precipitation in that one component of the reaction is al­ ready a solid, (Ellis, 1973). The extent of F adsorption from dilute solution by soil depends on the nature of soil, P solution and temperature. At a given concentration of P the amount of P adsorbed increases with an increase in temperature. Therefore a P adsorption isotherm should be used for P adsorption study. The effect of temperature on F adsorption (by Kaolinite) was demonstrated by Low and Black (1950). Phosphorous adsorption by soils and clay minerals has been Inves­ tigated, and adsorption isotherms have been used to describe the exper­ imental data, (Kurtz et al,, 19^1 Low and Black, 1950; Olsen and tfatanabe, 1957* Shapiro and Fried, 1959; Weir and Soper, 1962; Woodruff and Kamprath, 1965; Udo and Uzu, 1972; and Syers et al., 1973 b). 35 2,2.1 Theory of Adsorption Isotherm . The atoms or molecules axe said to be adsorbed on the solid surface and It Is distinguished from penetration of a component throughout the body of a second, called absorption. The distinction between the two processes is not always clear-cut and the word sorption is sometimes used, but whan the process is essentially surface effects, the word adsorption is preferred, (Barrow, 1966), Adsorption occurs on the solid surface because of the attraction forces of atoms or molecules in the surface of the solid. The potential energy of a surface plus molecules decreases as the molecules approach the surface. When a molecule approaches the surface it is attracted and this leads to an increase in kinetic energy, but the molecule will bounce away from the surface unless some of its momentum in the component normal to the surface is lost so that the molecule is trapped into a potential well. The adsorption process continues until the free surface energy of the system due to imbalance of surface forces, has been reduced to a minimum value. It is convenient to classify adsorption behaviors into physical adsorption and chemisorption. Physical adsorption is due to the oper­ ation of forces between the solid surface and the adsorbsto molecules, (similar to Van de Vaals forces between the molecules). This adsorption is reversible. 300 The energies of adsoprtion evolved are of the order of to 3000 J mol-^ (this corresponds to a heat of adsorption less than about 10 K cal/mole), and the amount of material adsorbed may correspond to several monolayers. Chemisorption is the result of much stronger binding forces comparable with those leading to the formation of chemical compounds. i 36 Such adsorption nay be regarded as formation of a surface compound. The energies range from 40 to 400 k j mol*”* (The heat of adsorption is greater than about 20 k cal/mole). At loner temperatures, chemisorption Is seldom reversible and no more than a monolayer nay be adsorbed (Barrow, 1966; Moore, 1972). Two commonly used adsorption isotherms to quantitatively describe adsorption of phosphate ions from solution onto soil particles are Freundlich Isotherm and Langmuir Isotherm, A) Freundlich Isotherm. x/m - kCn where* (2) x/m - the quantity of ions adsorbed per unit weight of adsorber, C = the equilibrium concentration of the adsorbate after adsorption occurred, k and - a constant, n = a constant This equation can be determined by using a log - log plot, since the following equation is linear. log x/m ™ log k + n log G (3) Thus, n is the slope of the line and k is the antilogarithm of the intercept. According to the Freundlich equation, the amount of adsorption increases indefinitely with increasing concentration. This empirical equation does not predict a maximum quantity of adsorption, which has been considered a limitation. B) Langmuir Adsorption Isotherm. This theory was developed by Langmuir (1918) to explain gas adsorption onto solid surfaces. There are many assumptions In this theory. However, 37 Olsen and Watsnabe (1957) have suggested a aethod for application of Langmuir Adsorption Isotherm to the study of P adsorption from solution "by soils, A rigorous* theoretical development for application of this equation to adsorption of ions from solution has not been presentedt thus a simplified derivation for P adsorption by soils is given below. The following assumptions and notation are madei^ 1. It is assumed that each adsorption site had anequal probability of adsorption of F from solution, I.e. this requires a constant free energy of adsorption. 2. (p) “ activity of P in solution (mole/liter). 3. x/m - mg P adsorbed per 100 grams of soil, 4. b - the maxinium amount of P that will be adsorbed by a given soil (mg/100 g soil). However, these assumptions are somewhat different from that of the original Langmuir theory for gas adsorption on the surface of solid. The rate of P adsorption will be proportional to(p)and to the sites yet unoccupied by P (b - x/m). Thust Rate of adsorption “ k^ (P)(b-x/m) (4) The rate of dissociation (desorption) will be proportional to the occupied sites. Thusj Rate of dissociation “ kg (x/m) At equilibrium, both rates must be equal. (5) Thus, equating (4) and (5). kg (x/m) - R^PXb-x/ra) (6) x/m k2+k1(P) i+(k1/k2)(p) (7) 1 Taken From Ellis and Erickson (1969); and Ellis and Knesek (1972), 38 (8) wheret and kg “ the proportionalconstants K - kj/kgi is sometimes calledthe adsorption coefficient. It is related to bonding energy of f adsorption on soil, (Olsen and tfatanabe, 1957). This is the common form of the Langmulr equation. The typical curve is obtained by plotting x/m versus (p). The limit of x/n as (p) approaches infinity is b, the adsorption maximum. However, this is not a particularly useful form of the equation from which to obtain b and K. The linear form can be obtained by rearranging equation (8). Thust (9) This form of the equation lends itself well to evaluation by computer methods in that the best fit of the linear relationship may be obtained by standard least squares method. The adsorption maximum, b, is the reciprocal of the slope of the regression line, and K may be obtained as the slope divided by the intercept. Using the equation -F° ** RT in K, the standard free energy of P ad­ sorption may be calculated, which constitutes a measure of the strength of adsorption bond. Another useful form of the Langmulr equation may be obtained as followsI 1. Let be the fraction of possible sites occupied by P. 2. Other symbols and assumptions axe as stated above. Then* (10) 39 and equation (6) divided by b, givesi or k. (P) (1 - x^m) - k- (x/m) 1 b (11) k^PKl-e) -kgfr (12) K " (13) ~ tPMl - ») Since K Is known from fitting experimental data into equation (?), equation (13) can be used to predict the concentration of P in solution at any level of P adsorption up to the adsorption maximum. Tangingir theory does not describe the mechanism of adsorption, but the model is based on many assumptions including the chemical adsorption type. However, Low and Black (1950) demonstrated, by using In 1C “ -AH° + constant, that the nature of P adsorption reaction (by Kaolinit©) RT did not change and the standard heat of reaction H ) was positive, therefore P is cheadsorption. The reactions that are responsible for adsorption from solution have been discussed by Giles (1959). They nay be classified as; l) non-polar Van de Vaals attractioni 2) formation of H bond; 3) Ion-exchange; and 4-) covalent bond formations. 2.2.2 Application of Adsorption Isotherm to P Adsorption. Experimental data of P adsorption by soil has been reported to con­ form to Freundlieh equation (Kurtz et al,, 19^6) and by kaollnite (lew and Black, 1950; Russell and Low, 1954), at high concentration of P solution (up to 50 ppm). Kurtz et al. (19^) indicated that at the low concentration of P (less than 1 ppm) their data dit not conform to the Freundllch equation. Langmulr equation has been widely used to describe P adsorption by soil and clay minerals. Ellis and Knezek (1972), pointed out that this 40 equation was derived by assuming constant free energy of adsorption (no interaction between adsorbed ions and a surface is uniform), but this situation would not be expected to occur in nature. They noted that the heat of adsorption, the nest important factor in the constant K, should be expected to decrease with increasing surface coverage. But the inter­ action of ions already adsorbed with the ions being adsorbed would in­ crease the heat of adsorption with increasing coverage. The two changes would in part compensate, tending to give a constant heat of adsorption. They indicated that even though the Langmulr equation Is still theor­ etically sound, one should avoid extrapolation of the regression beyond experimental data. Deviations from Langmuir isotherm also axe expected if precipitation of the ion being adsorbed occurs. Precipitation is indicated by a rapid Increase in x/m with a small change in equilibrium in solution. Many workers found their experimental data on P adsorption by soils conformed to tto Langmulr Isotherm at low equilibrium P concentration (Olsen and Vatanabe, 1957* Weir and Soper, 1962} Bennie and McKercher, 1959, etc.). They also found that the deviations from a single linear Langmulr relationship at higher equilibrium p concentration (above approx­ imately 15 mg/ml, or ppm P which is equivalent to about 48,45 x 10 ^ mole/ liter.) Udo and Ussu, (1972 ) reported that a straight line was obtained from Langmuir equation at (P) up to 2.5 to 5,5 ppm, for tropical Nigerian soils. Gunary, (1970) reported that deviations were found with P concentrations above 5 to 5.5 ppm. Recently Syers et al, (1973 found that there are two straight lines for the Langmuir plot for P concentration up to 14 ppm. The first straight line was for low P (up to 1.5 to 3*15 ppm). Deviations were found for P concentrations higher than these, hut these deviations also yielded a straight line which conformed to Langmuir Isotherm. They postulated that two populations of sites exist which have different affinities for P. Thoy concluded that the sites in part I, at low (P) had a very much higher binding energy constant (K) than in part II, at high (p ). They rewrote the Langmuir equation for two population of sites ast x/m - k V (p) + KZXb1X ^ 1 + KI(P) 1 + KII(P) where t superscript I and II refer to part I and II of the population sites. For clay minerals Cole et al. (1953) found that P adsorption on the surface of CaCO^ yielded a straight line for a Langmulr equation up to 9,6 ppm P, Higher P concentrations caused precipitation. Muljadl et al. (1966) found that there are two or more straight lines for P adsorption by Kaolinite, gibbslte and pseudoboehmlte, Hsu and Rennie (1 9 6 2 ) found that P adsorption gave a straight line up to 12 ppm P for amorphous Al hydroxide. The deviation also yielded a straight line which conformed to the Langmulr relationship. Even though the Langmuir Isotherm does not always accurately describe P adsorption by soil, as pointed out by Olsen and Watanabe (1957)* & major advantage of the Langmuir equation is that it is possible to calculate an adsorption maximum and a relative bonding energy term for P sorption. Agreement of P adsorption data with Langmulr Isotherm does not necessarily imply any specific adsorption or reaction mechanism. However, If the data obeys the Langmulr Isotherm the adsorption mei-mm (b) and a constant h2 related to binding energy of the soil for P axe very useful to describe the behavior of adsorbed F. Ellis and Erickson (1969)» used Langmulr Isotherm successfully to describe and predict the reduction of F level In sewage effluent by soils In Michigan. Their study mas very useful in evaluating a sewage treatment system for eliminating F from municipal waste by spreading It on soils. 2.2.3 Factors Affecting P Adsorption. Phosphorus adsorption from dilute solution by soil depends not only on temperature but also on soil characteristics such as soil acidity, amount of Al, Fe and their oxides, amount and kind of clay minerals, specific surface area or texture, organic matter content, and Ca In the soil and P solution. For F solution, factors Including salt effect, anion competition and pH of the solution may affect P adsorption. Kurtz et al. (l9*+6) reported the amount of P adsorbed In the presence of KC1 in the solution Is greater than that of water solution. Barrow (1972) found that at a given concen­ tration of phosphate, sulfate or molybdate, the adsorptionof each res­ pective anion Increased as creased. Ca concentration In the Initialsolution In­ Therefore salt concentration In P solution should be kept the same as the ionic strength of the natural soil solution. Phosphorus adsorptionIs not affected by soil solution ratio (Kurts et al,, 19^6) Shapiro and Fried, 1959). Shapiro and Fried (1959) found that the characteristics of P adsorption by equilibrium method Is not dependent on soiltP solution ratio of 1*5 to li2000. Phosphorus adsorption Is also affected by other anions contained in 43 the F solution and pH of the solution. Adsorbed F on the surface of soil can be replaced by other anions. The replacing ability at a neutral pH are In the order fluoride > oxalate > citrate} bicarbonate} borate> acetate > thlocyanate> sulfate} chloride. The ability of soil to adsorb acid 1st oxalate} citrate} phosphate} sulfate (Kurtz et al., 19W). Barrow (1970) reported that as the pH of the soil decreased from 6 to 4 (in 0.01 M CaCLj solution), the ratio of sulfatesphosphate adsorbed Increased. Adsorption was measured at concentrations of 20 and 0.2 ppm for sulfate and phosphate, respectively. The P adsorption by amorphous Al hydroxide from buffer P solution (buffer with acetate) Increased as pH of the solution increased (Hsu and Bennie, 1962). Black (1942) also found that pH of the P solution affected the amount of P adsorbed by kaolinite, bentonite, illlie and soil. Soil characteristics have greatly affected P adsorption. Kurtz et al. (19^6) discussed that P adsorption might take place at a point where 0H~, silicate or organic anions could be exchanged, or at a point where Al or Fe are exposed on the surface. If OH- was present where Al and Pe occurred at lattice edges, adsorption at these points would be an exchange of OH** and phosphate ions. Black (1942) explained that fixing power of the clay fraction was attributed to reaction of phosphate with Fe and Al present as exchangeable cations, as hydrous oxides or in solution, and to reaction with Ca. Russell and Low (1954) found that kaolinite saturated with exchangeable Al adsorbed about ten times more P than that of control. Kaolinite treated with oxalate or flouride and then washed to remove Al markedly reduced the amount of P adsorbed. However, oxalate treated and washed kaolinite adsorbed P from solution greater than that from F plus m oxalate solution. They concluded that exchangeable Al+3 or hydrous oxides were responsible for F adsorption. In the application of Langmulr equation, Olsen and Watanabe (1957) found that the adsorption maximum, (b), was closely related to the specific surface area of soils. Acid soils adsorbed more F per unit surface area and held P with a greater bonding energy (indicated by the constant K) than alkaline soils. They Indicated that acid soils are weathered more than alkaline soils, and have more weathering products such as the hydrous Fe and Al oxides, which may be responsible for F adsorption. Udo and Uzu (1972) reported that tropical acid soils adsorbed a large amount of P added to them. The capacity of F adsorption generally dep­ ends on degree of weathering. The F adsorption was positively correlated with sesquioxides and exchangeable forms of Al and Fe as well as clay con­ tent and was negatively correlated with pH of soils. The citrate-dithlonite and the oxalate extractable oxides were of equal Importance In P adsor­ ption, but the role of Al was more Important than that of Fe. They sug­ gested that sesquioxides (probably amorphous form) were reactive in adsorb­ ing F. Clay also provides adsorption sites for P. It Is likely that clay forms a part of complex gel which consists of hydrate along with small amounts of organic matter, A 1^0^ and associated Si(OH)^ and F, which Is considered as a major site for F adsorption. The F adsorption Increases as pH of soil decreases which may be due to the effect of pH on the exchangeable Al which is strongly correlated with F adsorption. Ellis and Truog (1955) also found that mantmcxr illonite would not fix F after all of Al and Fe were removed. Hsu (196*0 stated that the adsorption of P by a slight acid soil from dilute P solution at pH 7 is due to amorphous Al and Fe oxides In soil. 45 +3 They discussed -that activity of Al 4*3 and Fe in solution Is United by pH, and becomes negligible at pH 5 or above. But the surface reactivity of amorphous Al and Fe oxides have no auch limitation. Hazier (1969) stated that organic matter provides the major site for P adsorption. Bennie and McKercher (1959) also noted that organic natter Is considered to be equally as Important as Inorganic colloids In P adsorbing capacity of four Canada soils. They also found that Ca Is not closely related to the P adsorption maximum. Whereas, Udo and Uzu (1972) found no correlation between organic matter and Nigerian soils. However, they suggested that organic natter could be part of the complex gel responsblle for P adsorption. Ellis (1973) stated that the adsorption maximum, (b), depended on soil texture and horizon. Sandy soils have lower P adsorption maximum than heavy texture soils. The A horizon adsorbed less P than the B horizon In soils. This Is due to the accumulation of Fe In subsoil, Syers et al. (1971) found that P adsorption from dilute solution by tropical soils was correlated with exchangeable Al, with oxalate and citrate-dlthloulte, bicarbonate, Fe and Al and with clay content. Sor­ ption of added P was always better correlated with Al parameters than with Fe parameters. It Is difficult to Identify the soil parameters most responsible for P sorption because they are themselves correlated to a considerable degree. Fried and Bean (1955) reported that the retention of P by soil probably Involves exchange of phosphate from solution with an­ ions on the surface of the particle. Involved In this adsorption. Hydrous oxides of Fe and Al are The presence of oatloos such as Ca, Na, and H, affect the extent of this retention. Shapiro and Fried (1959) stated that P can be adsorbed on the surface of clay, surface of Fe and 1*6 Al oxides or hydroxides and CaCO^, and Indirectly adsorbed by means of cation bridge to the clay. Adsorbed P is either hydrolytically dis­ placed or displaced by another anion. 2.2.4 Correlation Between P Adsorption and P Availability. The availability of P is related with the reactive surfaces of soil material. Olsen (1953) pointed out that the amount of surface P as measured by P exchange is closely related to plant available P. Fried and Shapiro (195&) and Olsen and Watanabe (1957)* showed a close agreement between the adsorbed P based on Langmulr Isotherm and surface P measured by P ^ equilibrium. Olsen and Watanabe (1957) also pointed out that the range in value covered by P adsorption maximum will lie within the range of P fertiliser addition to soils. Phosphorus adsorption maximum as measured by Langmuir isotherm and its relationship to plant availability was studied by Woodruff and Kamprath (1965). They found that conversion of exchangeable Al to Al(OH)^ remarkedly reduced P adsorption maximum. The growth of plants was related to the degree of saturation of the P adsorption maximum. The soils with a large F adsorption maximum did not require as high a P saturation as those with a low P adsorption maximum. The soils with a higher P adsorption maximum apparently are able to supply sufficient P for growth at a lower saturation than the soils with a low P adsorption maximum. The maximum yields on the limed soils were obtained when P was greater than 1 x 10 ”^ moles/liter. Much higher P was needed on the soils with a high saturation of exchangeable Al to give maximum yieldsj however, the yields were lower than on the limed soils. Fried and Broesshart (1967) noted that there are three methods of describing soil P system In relation to solution concentration! l) phosphate potential, 2) solubility diagram, 3) adsorption equation. The adsorption equation appears to be the most useful description for plant nutrition. U6 2.3 Phosphorus Adsorption tar Flooded Sollc, Even though the availability of native P and added P before flooding increases upon submergence, when water-soluble F was applied to water­ logged soils , rapid removal of P from the solution occurred (De Datta et al., 1966| Mandel and Das, 1970). Acid, lowland rice soils generally contained large amounts of hydrated ferric oxides, which are highly reactive In removing P from solu­ tion. Besides, such soils, under waterlogging conditions, also contain higher quantities of soluble Fe (1RRI, 1963} Mandal, 196*0, which can rapidly precipitate the added water-soluble F from the soil solution (XRRI, 196*1-). Fixation of added P occurs in both acid and calcareous paddy soils. Chlang (1963 b) reported that the fixation of P compounds added to acid paddy soils was generally very rapid and completed within 5 - 9 days j then there was a slight increase of soluble P which in the later stages declined to nearly the same extent as in the untreated soil samples. In alkaline soils, fixation was generally completed in about 13 days after P application} and the soluble P present was in greater amounts than that of acid soils. Williams and Simpson (1963)* studied the effect of waterlogging in clover cultivation on pasture soils and found that short periods of flooding (1-2 days) produced anaerobic conditions that caused a decrease in P availability and an Increase in sorption capacity, particularly in disturbed (mixed) soil. This reduced P availability applied to both native and added P. De Datta et al. (1966), showed that the disappearance of added P in solution in equilibration with acid and neutral lowland rice soils is 1 49 very rapid. For two acid soils with pH of 5.1 and 6.0, none of the added P remained in the solution after 4 days of equilibration. The mechanism of P fixation also was observed. Chians (1963 b) noted that the chief forma of P fixed upon the addition of soluble P are Al-P and Fe-P in acid soils and Ca-P in alkaline soils. De l>atta et al, (1966 ) reported that the amount of F remaining in the solution within a period of 4 days from application was lowest for those acid latosollc soils containing kaolin type minerals and slightly higher in soil con­ taining predominantly montmorillonite. In calcareous soil (pH 7.6 ), the added F remaining in solution was unaltered during 4 days of equilibrium. Williams and Simpson (1966)* stated that the decrease In P avail­ ability upon waterlogging was due to the effects of certain reactions during the waterlogging treatment. Reducible metals such as Mn could have interfered; the nature of the sorption sites may have been altered due to Fe reduction causing P to be more tightly bound; or some of the P sorbed could have been occuluded by repreclpitation of Fe upon restoration of aerobic conditions. They felt that the increase in soil P availability upon waterlogging must be due to the presence of easily reducible ferric phosphate. Soils low in ferric phoshpate would not be expected to increase in P availability upon submergence. Kandal and Das (1970), studied the transformation of applied watersoluble P in three acidic lowland rice soils by incubating the soil samples with distilled water for 7 days and then adding P solution (as KHgPO^, 20 mg P®r !00 gm of soil). Each week after P application, they analyzed for P in both soil and solution and found that the amount of added P remaining in the solution in equilibrium with the soil declined sharply in all three soils. But the rate at which it declined was found 50 to be influenced more by the Fe^O^ and active Fe content than by the pH of the soils, They also reported that the solid-bound F fraction In the treated soils showed practically no Increase over the control. The A1-phosphate fraction In the treated soli, showed a marked increase over the untreated sample in all three soils. The magnitude of increase appeared to be influenced by A 1^0^ content rather than the pH of the soils. The maximum fixation as Al-phosphate occurred within the first week of F application. The transformation of added F into Fe-phosphate fraction indicated that the magnitude of F fixation was affected by Fe^O^ and active Fe content* not by the initial pH of the soils. The rate of fix­ ation of applied P as Fe-phosphate was progressive with time of incubation in the soil which was rich in the amount of Fe^O^ and active Fe. Mud and Lake sediment have the ability to adsorb P from Introduced solution* (Hesse* 1962) Harter* 1968| and Shukla et al. * 19?i). Hesse (1962) found that the fixation of F by fresh and dried mud is extremely rapid. Equilibrium was almost attained within 30 minutes and considerable fixation had occurred within 5 minutes. The difference in amount of F fixed in 30 minutes and 30 days of contact time was very small. However* Harter (1968) found that with low concentrations of F* the dried sediment was nearly identical to mud in its adsorption character­ istics. He suggested that sediment can be dried at 110 C without much change in F adsorption ability. But then more P was added the dried sediment adsorbed less F than did the fresh mud. The dried sediment did not wet easily and its adsorption capacity was likely decreased due to inaccessibility of a portion of reactive sites. The F-free sediment wetted more easily* and its adsorption Isotherm followed by that of fresh sedl- 51 ment throughout the range of P concentration used. Like air dried soil, there are many factors affecting P adsorption by lake sediment. Shukla et al. (1971) found that In general, noncalcarecus sediments adsorbed more added P than calcareous sediments. They stated that CaCO^ was not the major component responsible for sorption of P by the calcareous sediments. The proportion of added F sorbed was closely related to the amount of oxalate soluble Fe and to the levels of native total Inorganic P in the sediments. They suggested that the best single criterion indicating P adsorption In both non- and calcareous sediments was the amount of oxalate-extractable Fe presumed to be derived from amorphous Fe oxides, and proposed that a gel complex of hydrated Fe oxide which contains small amounts of A1^0^, Si(OH)^ and organic matter, is the main P sorbing components of sediments. Form of adsorbed P by mud and lake sediments also have been Investi­ gated. Hesse (1962) found that when Ca-F was added to mud at 20 ppm P, the Al-P after an initial Increase, decreased within 30 days to its original value. He explained that the Al bounded P decreased due to a transfer to Fe-P. According to Harter (1968), when 0.1 mg P or less was added to lake sediments, NH^F extracted a tightly bound form of P, probably occurring as Al-P. However, when more than 0.1 mg P was added, additional P in a more loosely bounded form was adsorbed Into NH^F extractable fraction. The loosely bounded F appeared to be Independent of Al content and could be removed by successive water extractions. Phosphorus Isotherms of lake sediments studied by Harter (1968) showed that total adsorption as measured by loss from solution and by 52 extraction with 0.5 NH^F followed Langmuir Isothern at low concentrations of F hut at high F concentration the Isothern curve deviated from the conventional Langmulr Isothern, However, adsorbed F recovered by extraction with 0,1 N NaOH followed Langmuir Isothern throughout the F concentration used. MA.TERIAIS AND METHODS 3.1 Materials. 3.1.1 Soil Samples . Five Michigan soils were selected to give a variation in chemical properties such as pH, organic matter content, amount of Fe and Mn, and physical characteristics including texture ranging from sandy loam to clay loam. Both surface soil (A horizon) and subsoil (B horizon) were collected from each soil series. Soil samples were air-dried, ground and passed through a 1 m.m, sieve. The soil characteristics are shown in Table l.a. of the Appendix. 3.1.2 Phosphorus Analysis. 3.1.2.1 Murphy and Riley or Ascorbic Method. Murphy and Riley (1962) proposed use of a single reagent for deter­ mining P in natural water. The reagent and procedure of this method are as followsi Reagentsi (1) Reagent A. Dissolve 12 g of ammonium molybdate in 250 ml of distilled water. Antimony potassium tartrate 0.2908 g is dissolved in 100 ml of distilled water. In 1,000 ml of 5 N H^SO^ (lh8 ml of concentrated, 36 N HgSO^ is carefully added in to about 700 ml of distilled water, cooled to room temperature and then made to volume with distilled water), both dissolved reagents are added, mixed thoroughly and made to 2 ,0 0 0 ml with distilled water. Reagent A is stored in pyxex glass bottle In a dark and cool compartment at ^ C. (2) Reagent B. Dissolve 1.056 g of ascorbic acid in 200 ml of 53 54 reagent A and nix. This working reagent is light yellow In color and should be prepared, as required since it does not keep for more than 24hours. Procedure i Pipette aliquots containing 1 to 40^jg of orthophosphate into 50-ml volumetric flasks. Add distilled water to make the volume to 40 ml, and then add 8 ml of reagent B. Hake to volume with distilled water and mix thoroughly. After 10 minutes and within 24 hours, the color Is read on an Evelyn photoelectric colorimeter at 660 mu. A standard P curve is prepared In the same manner as the samples, with P concentration ranging from 0 to 1 ppm. In this method, the reagent contains H^O^, ammonium molybdate, asc­ orbic acid and antimony potassium tartrate in a single solution reacts with orthophosphate to produce a blue-purple complex compound within 10 minutes. Antimony accelerates color development and the color remains stable for at least 24 hours. The complex obeys Beer's law up to a P concentration of 2 ppm. The sensitivity of this procedure is comparable with that of the -HgSO^ method for P determination in sea water (Murphy and Riley, 1962), and water extract (Watanabe and Olsen, 1965). 3.1.2.2 SnCl^-JUSOfj Method. Chiorostannous-reduced molybdophosphoric blue color method, in a sulfuric acid system was widely used for determining P In the solution samples (Jackson, 1965). The reagents and procedure of this method are as followst Reagents1 (l) Solution I. Dissolve 25 g ammonium molybdate in 200 ml of dis­ tilled water, and warm at 60 0, The reagent is filtered if necessary. 55 (2) Solution IX. Add 275 ml of concentrated HgSO^ carefully to 425 ml of HgO. The solution Is cooled to room temperature and Solution I added slowly. Then dilute the solution to one liter and protect It from light. (3) Chlorostannous acid reagent. 25 g of SnClg.PHgO Is dissolved in 50 ml of concentrated HC1. The solution is added to 500 ml of dis­ tilled water and then make volume to 1 liter with distilled water. +2 final solution Is 1M Sn The . Procedure! 25 ml of the test solution (the final concentration of P ranges from 0 to .8 ppm)i is placed in a 5 ® ml of volumetric flask, and 2 drops of 2 , 4-dinitrophenol Indicator Is added, and pH Is adjusted with 2 N HgSO^ or 4 N Na^CO^ until one drop of 4 M Ua^CO^ produces a faint yellow color (pH about 3). Two ml of solution II is added and the samples are diluted nearly to volume with distilled water and thoroughly nixed. Then 0.2 ml of chlorostannous solution is added to develop the color and the sample is diluted to 50 ml with distilled water. After 5 minutes, but not later than 10 minutes the color Is read photometrically on an Evelyn photo­ electric colorimeter at 660 mu. Calibrate the method using a standard P in the sane manner as the sample with P concentration ranges from 0 to 1 pjaa P. This method has a very high sensitivity per unit of P content, pro­ viding a working range from 0.02 to 0.6 of P. It provides for noninterference of SI in solution up to 200 ppn, Fe+2 up to 100 ppm, Fe+3 up to 2 ppm, Ti up to 20 ppm, Ca and Mg up to 500 ppm or more, NO^ up to 100 ppm, F up to 5 ppm, Cl up to 250 ppm, up to 1000 ppm — but It includes arsenate in chemical equivalence to P (Jackson, 19$5). 56 3,2 Methods. 3.2.1 A Study of the Methods Used In the Determination of P. The SnClg-HgSO^ method was widely used for P determination In water sample and soli extract because of Its sensitivity. However, this method Is not convenient when working with several samples at the same time. The pH of the samples must be adjusted to pH 3 before color development and the color must be read between 5 and 10 minutes after mixing with all of the reagents. The Ascorbic method, on the other hand is easier to handle, with no pH adjustment and allows for working with several samples at the same time due to the very stable color complex. This method also uses only a single reagent solution. Both methods axe sentitive In the lower range of P concentration of 0.02 ppm to 0.6 ppm. The SnClg-HgSO^ method is the standard method for determining P In the solution In this low range of concentration. This preliminary study was designed to test P determination In P adsorption Isotherm experiments, comparing Ascorbic and SnClg-HgSO^ methods, 50 ml of 0.01 M CaClg solution, containing KH^PO^ at a concentration ranges from 0 to 30 ppm, was added to 5 fS of prepared soil samples In 125 ml Erylenmeyer flasks. The samples were shaken on & rotary shaker machine at 200 rpm for 5 hours at room temperature. The samples were then filtered. The concentration of P In the filtrates were determined by both methods. The amount of P adsorbed (x/m) were calculated by substractlng the P remaining In the filtrate from P In the Initial solution. The data of this study Is shown In Table 2, The results of this study are shown in Figure 1, The amount of P 57 Table 2. Comparison of two analytical methods for P analysis In P adsorption studies. Hillsdale A 0,0 3.0 6,0 9.0 12.0 15.0 20.0 Equilibrium ^ P Ascorbic SnClg-HgSO^ P adsorbed ^ _ (x/m) Ascorbic SnClg-HgSO^ o o Initial P i Soils 3 0 .0 Hoytville A 1/ 0.0 3.0 6.0 9.0 0 .0 6 0.12 — 0.65 1.82 3.20 5.30 7.55 11.2 18.4 0 .7 0 2.35 4.18 5.80 6.70 7.40 8.80 11.6 1.78 3.18 5.35 7.68 11.2 19.9 1 2 .0 0.29 0.70 1.35 15.0 2 .1 2 2 0 .0 3 .6 2 0.12 0.25 0.55 0.95 1.^5 2.45 4,88 3 0 .0 8 .6 2 8 .9 2 o.o6 0 .1 5 average of 2 replications. — 2.85 5.72 8.30 1 0 .6 12.9 16,4 21.4 — 2 .3 0 4.22 5.82 6.65 7.32 8.75 10.0 — 2.75 5.50 8 .0 5 10.5 12.7 15.1 2 1 .1 58 20 Adsorbed P (x/a In ag/100 g soil) o 15 O 10 Hoytvllle A x © Hillsdale A 5? * SnClg.HgSO^ © O l 10 Ascorbic acid X 15 20 25 30 Initial P concentration (ppra) Figure 1, P adsorbed, calculated form Ascorbic and SnGl^-H-S0. methods for P determination. 59 adsorbed which was obtained by both methods of F determination is closely agreement, especially at low concentration of P. At higher P concentration, the amount of P adsorbed obtained from the Ascorbic method deviated somowhat from that obtained by the SnClg-HgSO^ method. For Hoytlvlle A a deviation occurred at 20 ppm P of initial solution but the higher concentration of 30 ppm of initial solution, gave little deviation. For Hillsdale A, the deviation occurs at 30 ppm of the Initial solution. It could be concluded that at low concentration of P in the ini­ tial solution the Ascorbic method gives comparable results to that of SnClg-HgSO^ method. Since the Ascorbic method is mors convenient to work with, the P determination by this method is selected to use in the study of P adsorption in this thesis. 60 3.2.2 Phosphorous Adsorption fry Flooded Soils. 3.2.2.1 Anaerobic Condition. An anaerobic glove chamber (Coy manufacturing Co.» Ann Arbor, Mich.) was used to keep soil samples tinder anaerobic conditions. This anaerobic system was described by Arank! et al. (1969). The atmosphere is composed of 10 percent Hg and 90 percent Kg. Oxygen gas, if present, is reduced by means of Palladium catalyst diffusion box system by the reactioni 2Hg + 0g Pd ^ 2HgO (1*0 Silica gel was used to remove HgO generated by the Pd catalyst and Hydrogen sulfide absorbent, (dissolve ,5 g AggSO^ in 100 ml hot distilled water, and add 1 ml N HgSO^ plus 100 ml glycerol) was used to remove HgS gas which may be produced by bacteria since HgS gas will poison the Pd catalyst. This anaerobic chamber reduced 0g to less than 5 ppm. 3.2.2.2 Phosphorus Solution. The P solutions used in this study were prepared as follows* (1) CaClg solution. 11.099 g of anhydrous CaClg (analytical grade) was dis­ solved in distilled water to obtain a stock solution of 0.1 M CaClg. Working solutions of 0,01 M CaClg were obtained by dilution of this stock solution with distilled water. (2) Phosphorus in 0.01 M CaClg (p H 6 ). 2 .6 3 6 2 g of analytical grade KHgPO^ was dissolved in 0.01 M CaClg to obtain a 600 ppm P stock solution. The working P solutions were prepared by diluting the stock solution with 0.01 M CaClg. These solutions contained F concentrations ranging from 0 to 300 ppm P. 61 The P solutions were adjusted to pH 6 , using a Beckman pH meter, by addition of .02 M Ca(0H)g. The adjusted P solutions were stored over­ night and the pH values measured and readjusted to pH 6 if necessary. Because the P solutions were adjusted to pH 6 , P in the solutions was measured by the Murphy and Riley method to obtain the P concentration in each solution. Solutions containing 34.5, 62.0, 8 8 .6 , 107.2, 147.0, 167.5. 214,5. 294.0 ppm P and 35.0, 63.0, 89.2, 96.25, 117.13, 170.0, 242.1 ppm P were used for surface (horizon A) and subsoils (horizon B) respectively. 3.2.2,3 Phosphorus Adsorption Isotherm. (1) Air dried soils. Five g soil samples were placed in 100 ml plastic centrifuges tubes and 45 ml 0.01 CaClg (adjusted to pH 6 ) and 5 ml of adjusted pH P solutions were added to five initial P concentration varying from 3 to 30 ppm. The samples were shaken on a wrist-action shaker for 24 hours, centrifuged at 2 ,2 0 0 rpn for 15 minutes, filtered and the filtrates analyzed for F concentration by Murphy and Riley method. The amounts of adsorbed P, (x/m), were calculated. (2) Anaerobic soils. 40 ml of 0 .0 1 CaClg (adjusted to pH 6 ) were added to each set of soil samples in 100 ml plastic centrifuge tubes. A set of F solutions was prepared as follows1 5 ml of 0,01 CaClg (adjusted to pH 6 ) In 25 ml Erylenmeyer flask were mixed with adjusted pH P solution (see above). Both soil samples and the set of P solutions were placed in the entry lock portion of the anaerobic glove chamber (see Figure 2). After closing the outside door of the entry lock, the entry lock was evacuated and then 62 flushed with two times, then evacuated and flushed with a gas mix­ ture, (10 percent Hg and 90 percent Ng). The samples and the set of F solutions were transferred from the entry lock into the glove chamber, and closed with porafllm papers. The samples were Incubated In the glove chamber for 1, 3« 7* days. After incubation, the prepared F solutions were added to the soil samples in the glove chamber, and the tubes were closed with paxafllm papers. The tubes were transferred via entry lock and were shaken, centrifuged, filtered, and the filtrates were analyzed for P concentration in the sane manner as air dried soils. 63 in **' GAS lMIX SB* SHELF CAT. BOX DOOR •S' STANDING ROOii Figure 2. Diagram of anaerobic glove chamber, RESULTS AMD DISCUSSION The first portion of this study was designed to determine the optimum tine of contact, salt concentration and pH of P solution for the best data for the Langnulr Adsorption Isotherm. Many conflicting reports In the literature have made these preliminary experiments essential, 4,1 Preliminary Studies. 4,1.1 Study of Contact Time and Effect of Salt Solution. The typical time P adsorption curve Indicates a fast reaction ini­ tially, followed by a reaction which proceeds at a slow and almost constant rate. Researchers have attempted to adjust the contact time to where the rate of P adsorption and desorption are at equilibrium. Practically they attempt to get the time period to where the rate of P change in solution is small. The contact period required to reach equilibrium depends on many factors Including soil properties, P solution and rate of shaking. contact period ranged from a few hours to 3 days or more. The Syers et al, (1973 b) used 72 hours, Olsen and Watanabe (1957) used 24 hours, Shapiro and Fried (1959) used 18 hours for shaking, while Hsu (1964) and Rennie and McKercher (1959) allowed a contact period of 6 hours. The P solution used In adsorption studies varied from researcher to researcher. Some workers used P in salt solutions such as 0.01 M CaCL-, (Ellis and Erickson, 1969* Gunary, 19?0j and Udo and Uzu, 1972) or 0.1M NaCL (Syers et al,, 1973 b). However, P in water solution also has been used, (Olsen and Watanabe, 1957* Hsu, 1964), The purpose of having salt In the solution is to keep the ionic strength of the suspension the 64 sane as that of the n&tuxal soil solution. The following study was designed to determine the appropriate contact period and the effect of 0,01 H on P adsorption. The study was conducted in 0.01 H CaClg and HgO solution containing 6 and 12 ppm P. and 48 and 72 hours. The shaking times were 3, 5» 12, 24, The data of this study axe shown In Table l.b. of the Appendix. The results of this study axe shown In Figures 3 and 4. The amount of P adsorbed by both Hoytvllle A and Locke A soils In 0.01 M CaCl^ P solution Is greater than that of P In water solution at each time interval. The reaction rate Is fast Initially and is followed by a slow rate and app­ arently approaches equilibrium with time of shaking for all treatments. However, rate of P adsorption In salt treatments was initially faster than that of water treatments at all P concentrations, and eventually approached equilibrium at about 24 hours of contact time. But In ^ 0 treatment, the contact period of 24 hours was insufficient to reach equili­ brium. For the Hoytvllle A the contact period for equilibrium F adsorption from HgO solution wan about 48 hours, while Locke A soil can adsorb more P and the contact period of this soil is apparently longer than 72 hours. When solution was added to soil, the ionic strength in the soil solution was dilute and the ions in liquid and solid ^hase were not In equilibrium. It takes time to allow the ions In the soil which is a heterogeneous system to reach an equilibrium state. When we Introduce P Into this system, the P adsorption reaction is also dependent cm the strength of other ions species. For the system in which the native Ionic state Is in equilibrium or near equilibrium, the P adsorption reaction would reach equilibrium faster than In the other system. Therefore the P adsorption In salt solution reached equilibrium faster than that In 66 10 to 03 •0 ■< 12 pp« P In 0,01 M CaClg 12 " " H2° 6 " " 0,01 H CaClg 6 " •• H20 48 72 Shaking Tine (hours) Figure 3. Effect of contact time, CaCl? and ht,0 solution on P adsorption fcy Locke a 3oil. P Ada orbed (x/a In ag/100 g soil) 67 -o 10 . tgy-o — cr 03 5 12 ppm In 0.01 M CaCl, 0 Shaking Tine (hours) Figure U. Effect of contact time, CaClp and HgO solution on P adsorption by Hoytvllle A soil. 68 HgO solution. In addition to the P directly adsorbed on the surface of clay material, P can he adsorbed indirectly by means of cation bridge to clay or other soli materials (Shapiro and Pried, 1959). This mechanism of P adsorption could explain the greater and faster P adsorption in 0.01 M GaClg solution them In water treatment. From this study, it could be concluded that 24 hours Is the appro­ priate contact period for P adsorption isotherms using P in 0.01 M CaCL, with shaking the soil suspension on a wrist-action shaker. The advantage of 0.01 M CaCl^ for the P adsorption study in the Michigan Soils is not only the faster equilibrium adsorption, but also a clear solution is obtained for P analysis after centrifugation at 2,200 rpm (1500 HOP), With P in water solution treatments, the suspension need further centrifugation at the faster speed of 6100 rpm (4500 appropriate clear solution for P determination. rgf ) to set the 69 4.1.2 Effect of the JJH of the Initial P Solution. Host studies of P adsorption isotherms have teen carried out either at uncontrolled pH values (Syers et al. ( 1973 hr Weir and Sopert 1962i and Kurtz et al., 1946) at a buffered pH (Hsu and Rennie. 1962| Bache, 1964)| or adjusted to a certain pH initially without buffering (Olsen and Watanabe, 1957I Rennie and McKercher, 1959? Woodruff and Kamprath, 1965). Phosphorus adsorption Is not a single mechanlsn or type of reaction — it is partially anion exchange because P can be replaced by other anions such as flouride, oxalate, citrate, bicarbonate, borate, acetate, thlocyanate, sulfate and chloride (Kurtz et al., 1946). therefore the P solution with pH controlled by a buffer appears not to be satisfactory because anions used in the buffer are likely to compete with phosphate ions for reactive sites, complicating the interpretation of the results. However, the pH of P solution in 0.01 M CaCl^ is not constant init­ ially. As the P concentration increases the pH of 0.01 M CaCX, solution decreases (Table 3). Thus, the effect of pH of P solution an P adsorption must be studied. The studies were conducted in 0.01 M CaGl^ solution with pH adjusted to 6 (with 0,02 H CafaH^) or the same solution with pH unadjusted. amount of adsorbed P, (x/m), was calculated. The The data from this experiment is shown in Table i.e. of the Appendix. The amounts of P adsorbed as a function of P equilibrium, at low concentrations of P of both treatments are equal in all three soils (See Figures 5» 6 and 7). For higher P concentrations, the amount of P adsorbed at pH 6 became somewhat greater at each P Increment. The pattern of adsorp­ tion curves in both treatments are the same — starting with a fast ad­ sorption rate, followed by a slow rate and approaching the maximum adsorption 70 Table 3. The pH of increasing KHgFO^ 0.01 M CaCl2 and HgO, concentrations in solutions of P concentration PH 0.01 M CaCl2 V 4' — ppm— 0.0 3 0 .0 6 0 .0 9 0 .0 120.0 1 5 0 .0 200.0 1/ deionized distilled water. 5.45 5.10 4.85 4.65 4.55 4.62 4.62 6 .3 0 5.85 5.30 5.20 5.10 5 0 Equilibrium P Cone* (in XlQ \ole/l) P Adsorbed (x/m in mg/100 g soil) Equilibrium P Cone* (in X10 aole/1) Effect of pH of P solution on P adsorption by Miami A soil. 1 Sample adjusted to pH 6 2 Saaple unadjusted Figure 6. 1 Saaple adjusted to pH 6 2 Sample unadjusted Effect of pH of P eolution on P adsorption by Hillsdale A soil 10 Figure 5. H *4 15 72 o m ttf> o o 20 - * 8io ■8 ■< A« 1 Sample adjusted to pH 6 2 Saaple unadjusted 10 20 30 U0 SO Equilibrium F Cone* (in XI0 mole/l) Figure 7 Effect of pH of F solution on P adsorption by Hoytvllle A soli. 73 as P Increased. The amount of F adsorbed from the P solution adjusted to pH at 6 at higher F concentration was somewhat greater than that of the uncontrolled treatment. This may be due to pH effects or the amount of Ca in solution. According to Black (19^2) the pH of solution affected the amount of F adsorption by clay minerals such as kaolinite, hydrated halloyslte, illlte and bentonite and by soil, each of which had a range of pH of final sol­ ution for itsmaximum adsorption. He found that finely ground kaolinite showed a maximum adsorption at a final pH of 3 to 4. For another kao­ linite clay the maximum adsorption occurred at pH about 6 to 7 with 1 and 10 ppm F solutions. At 1 ppm P, the maximum adsorption by Cecil clay in which free Fe oxide was removed occurred at a final pH of 6 to 7 with smaller quantities being adsorbed at lower or higher pH values. The adsorption from a solution concentration of 10 ppm F showed two distinct maxima, one in the range of pH 6 to 7 and the other at pH 3.2. However, for untreated Cecil clay, at 1 ppm P concentration the maximum adsorption occur at about pH 4.5* These studies show that the F maximum adsorption as a function of pH of the solution depends on the soil properties in­ cluding kind and quantities of free oxides and type and size of clay mineral and F concentration. However, in adjusted pH treatment the amount of Ca+2 in the F solution was somewhat greater than that of uncontrolled treatment. This may more or less account for the increase In the amount of F adsorption in pH adjusted treatment. Barrow (1972) found that the amount of F adsorbed increased as Ca concentration in F solution increased. 74 4.1.3 Effect of Contact Time on P Adsorption from 0.01 M CaClp at p Because 24 hours was selected as the appropriate contact period for P adsorption from 0.01 M CaClU, solution, this study was designed to determine the contact time for P adsorption from P solution in 0.01 M CaCl^ which was adjusted to pH 6 with 0.02 M Ca(0H)g. The experiment was conducted in 0.01 M CaClg (pH 6) containing 6.2 and 14,7 ppm P. The shaking times were J, 5, 10, 16, 24, 48 and 72 hours. The data is shown in Table l.d. of the Appendix. The results are summarized in Figure 8. As expected the P adsorption curve is a function of time as was the case In the previous study with P solution in which pH was not adjusted. The rate of P adsorption was first fast, followed by a slower rate and proceeded to reach the equilibrium eventually. At the lower concentration of P (6.2 ppm) the equilibrium was reached in 24 hours of contact time in both Sims A and Locke A soils. But at the higher P concentration (14.7 ppm P) the shaking time of 24 hours approached equilibrium, but equilibrium was not reached until about 48 hours in both soils. It could be concluded that the appropriate contact period for P adsorption by the soils under study, from P solution in 0,01 M CaGl^ which was adjusted pH at 6, is about 24 hours. This contact period of 24 hours will be used in the P adsorption Isotherm study for anaerobic soils in the next section. 75 O 1 Sins A 14.? ppn Locke A ” 51ns A 6.2 ppn Locke A ** *# Shaking Tine (hours) Figure 8. Effect of shaking tine on P adsorption from 0.01 M CaCl0 adjusted to pH 6. 4.2 Phosphorus Adsorption Isotherm. 4.2.1 Phosphorus Adsorption by Air Dried Solis. The amount of P adsorption by soils (x/ra) as a function of P equilibrium concentration, (P), is shown in Figures 9 to 13. The amount of adsorbed P increased as (P) increased. The incre­ ment of P adsorbed, |A (x/m) j. decreased as (p) increased, and x/m apparently approached the P adsorption maximum, (b). For example, when (P) Increased from 2,5 x 10”"* to 5 x 10”"* mole/l, the increase in ad­ sorbed P,|!a(x/m)j was about 2.2 and 1.8 mg P/100 g for Hillsdale B and A horizons. When (P) increased from 5 x 10”"* to 7.5 x 10”"* mole/l, (x/m) became about 1,0 and 1.4 ng p/100 g for the same soil samples. This indicated the typical curve of P adsorption Isotherm. The amount of P adsorbed, (x/m), at any given (P) by subsoils is greater than that of surface soils. For example, at 5 x 10”"* mole (P)/ 1, the amount of P adsorbed by soils in B horizon of Hillsdale, Locke, Miami, Sims and Hoytvllle, are 15*2, 14.2, 13, 10, and 17*5 mg/100 g soil, compared with 6, 6.7, 5.8, 8.5, and 1 3 .0 mg/100 g soils in the respective A horizons. However, with coarse textured soils, (Hillsdale, Locke, and Miami) the amount of P adsorbed by the B horizon is much greater than that adsorbed in the surface, i.e. at 5 x 10""* mole(p)/ 1, x/m of subsoils were approximately 2.5, 2,1, 2,2 times of the sur­ face ones. The differences are 9.2, 7.6, and 6.2 mg/100 g soil res­ pectively. But with the fine textured soils, Sims and Hoytvllle, the difference in the amount of adsorbed P between the two horizons is not so pronounced, i.e. at 5 x 10 "* mole (P)/l the differences axe about 1.5 and 4.5 mg/100 gm soil. Sims B and Hoytvllle B adsorbed 1.2 and 1.4 times more P than the A horizons at the same (p) values. 77 Figure 9. Amount of P adsorbed by Hillsdale soils as a function of (p). Figure 10. Amount of P adsorbed by Locke soils as a function of (P). 78 to 15 Hillsdale B to Hillsdale A 10 u ■< 20 30 {JP Equilibrium P Cone* (in XlO_vnolo/l) 20 o 15 '§* 10 Locke B eke A 20 30 U0 Equilibrium P cone.(in cone XI0 “^ mole/l) 79 Figure 11. Amount of P adsorbed by Miami soils as a function of (p). Figure 12. Amount of P adsorbed by Sims soils as a function of (p). 80 iaml B o Miami A 20 30 - Equilibrium P Cone, (in X10“'>molo/l) bO Sims B Sims A ■< Equilibrium P cone .(in X10“5 mole/l) 81 Hoytvllle fi oytville A 10 Adsorbed P (x/m in ag/100 g Soil) 20 20 Equilibrium P Cono. (in XlO"^aole/l) Figure 1 3 , Amount of P adsorbed Toy Hoytvllle soils as a function of 82 The amount of P adsorbed in the subsoils is greater than that adsorbed by surface soils because of the accumulation of Fe and clay min­ erals in the B horizon, Ellis and Erickson (1969) also reported that B horizons have the ability to adsorb more P than A horizons. The results also showed that fine textured soils adsorbed more P than coarse textured ones. Sims A and Hoytvllle A soils with clay loam texture, adsorbed more P than Hillsdale A, Locke A and Miami A, (sandy loam and loam texture). At 5 x 10”^ mole(p)/l, Sims A and Hoytvllle A adsorbed about 8,5 and 13.0 mg P/lOO g soil, whereas Hillsdale A, Locke A and Miami adsorbed about 6, 6.7 and 5*8 mg P/lOO g soil respectively. For subsoils at the same values of (P), Hoytvllle adsorbed 17.5 wg P/100 g soil (Hoytvllle B), while Hillsdale B, Locke Bf and Miami B ad­ sorbed only about 15.2, 1^.3 and 13 mg P/lOO g soil, However, the Sims B adsorbed only about 10 mg P/lOO g soil. This may be due to a lower accumulation of Fe and clay in this soil than in Hillsdale B, Locke B, and Miami B or it may be due to some other factors. Olsen and Watanabe (1957) found a close relationship between adsoprtlve capacity and the specific surface area of soils. Black (l9^2) reported that ground kaolinite adsorbed more P than the unground miner­ al, Syers et al, (1971) and Udo and Uzu (1972) found that soil ad­ sorption capacity and the amount of clay content were correlated. Because for a given amount of soil, fine textured soil has a greater number of reactive sites than coarse textured soil, clay soils will adsorb more P than sandy soils. Figures 9-13 show that almost all of the P in the low initial P solution was adsorbed by subsoil, indicated by the steep increase in x/m as P increased. But for surface soils the amount of x/m increased as a 83 function of (P) was lowor. This would Indicate that there axe more sites for P adsorption In B horizons than in A horizons, Langmuir adsorption Isotherm for alx dried soils are shown In Figures 14 to 18. A plot of (p)/(x/m) versus (P) gave a straight line at low P concen­ trations — up to about 5 x 10**5 mole (P)/l for surface soils except Hoytvllle A (at about 2.5 x 10“^ mole (P)/l) and up to 1 x 10**^ to 5 x 10“5 mole (P)/l for subsoils. At higher (p) the deviation also yield another straight line for levels up to 10 x 10"^ to 30 x 10“~* mole (P)/l. That Langmuir P adsorption data yielded two straight lines for P levels of about 14 ppm or about 45 x l6~“* mole (P)/l was also found by Syers et al. (1973 b). They noted that there are two components of soils which are very different with respect to P adsorption. The constant K, related to bonding energy# and b( the adsorption maximum of the two parts were calculated from the slopes and the Inter­ cepts of both straight lines. The value of K and b were shown on Table 4. The (K value for the first straight line) values vary from about 0.2 to 0.7 x 10^ l/mole for surface soils, and range from 0.86 to 8,8 x 1CT* l/mole for subsoils. While K2 (corresponds to the second straight line) vary from about 0.07 to 0.22 x 10^ l/mole for surface soils and about 0.15 "to 0.77 x 10^ l/mole for subsoils. For all soils, , is greater than K£ (See Table 4) which Indicated that P was adsorbed in part X more tightly than by part II, since the greater the K value the greater the energy of P adsorption on the surface of soil particles. However, the difference between this bonding energy constants (K^ and Kg) is greater in subsoils with K^/K^ ratios varying from about 5.8 to 15*5, (See Table 4). The K^/fcg ratios of surface soils 84 Table 4. Langmuir adsorption parameters for air dried surface and subsoils. Part X Soil Hillsdale A B Locke A B Miami A B Sims A B Hoytvllle A B Ki Part II K2 b2 *1^2 V bl Eb x 10'^ Tag/100 g soil x 10-5 mg/100 g soil .205 8.33 12.9 12.0 .087 .769 17.9 20.0 2.35 10.8 1.4 1.6 30.8 .354 5.92 10.5 11.2 .126 .694 15.2 18.0 2.80 8.53 1.5 1.6 25.7 29.2 .259 4.28 10.7 11.7 .073 .431 17.9 19.3 3.53 9.94 1.7 1.7 28.5 .389 .857 12.9 11.7 .135 .149 20.0 21.0 2.88 5.76 1.6 1.8 32.9 32.7 .706 8.88 15.0 11.2 .227 .578 26.7 3.36 15.5 1.8 2.6 41.7 40,2 2 9 .0 mg/lO( g soil 3 2 .0 3 1 .0 85 o □ * (<0 Figure langnrulr P adsorption isotherms for Hillsdale soils. 86 v« U-\ JO* * (£) o Figure 15. ^angmulr P adsorption Isotherms for Locke soils 87 Vt UN •CM CM CM ** (w/X Figure 16* Langmuir F adsorption isotherms for Miami soils* 88 CO CO . >o CM o N Figure 17. o o* Langmulr P adsorption isotherms for Sims soils. 89 < (q X I__ m o O tn o ow Figure 18, Langmuir P adsorption Isotherm for Hoytvllle soils. 90 range from about 2 to 3.5. However, Byers et al. (1973 t) found that values range from 33 to 91 for their soils. Bond strength indicated by K values of both regions showed that subsoils adsorbed F more strongly than do surface soils. In coarse tex- tured soils (Hillsdale, Locke and Miami), the difference between K values of A and B horizons In both regions is greater, which is indicated by K ^ a value (Table 5). These values (Kg and KA ) range from about 16 to 40 in part I and about 5.5 to 8.8 for part II, For fine textured soils, 31ms and Hoytville, the YL^/K^ values are about 2 and 12 for region I and about 1 and 2.6 for region II respectively. The b^ (adsorption maximum in region I) values vary from 11 to 15 mg P/lOO g soil for A horizon and range from 11 to 13 mg P/100 g soil for B horizon. While bg values (adsorption maximum in region II) vary from about 15 to 26.7 Bg P/lOO g soil and about 18 to 29 mg P/100 g soil for horizon A and B respectively. In region I, the adsorption maximum indicated byb^ values, was sim­ ilar for sub and surface soils, except for Hoytville. For example, b^ values are about 10.5 and 11.25 mg P/lOO g soil for Locke A and B, and about 10.7 and 11.7 mg P/lOO g soil for Miami A and B respectively. But Hoytville A and B have b^ values about 15 and 11 mg P/lOO g soil. The adsorption naxlsua of region II, b2, of subsoils Is greater than that of surface soils, except Sims. For example, for Locke soil the adsorption maximum of region II are about 15*2 and 18.0 mg p/100 g soil for horizon A and S respectively. While Sims A and S, the bg values are about 20 and 21 mg p/l00 g soil. Figures 14 to 18 showed that at any given (p) the amount of P adsorbed by subsoils is greater than that of surface soils, indicated 91 Table 5. Ratio of banding energy constant and adsorption maximum of A and B horizons. Bart II Bart I Soil V*A V bA V^A V bA Hillsdale 40.7 .93 8.8 1.12 Locke 16.8 1.17 5.5 1.18 Miami 16.5 1.09 5.9 1.0? 2.2 .91 1.1 1.05 12.0 .75 2.6 1.09 Sims Hoytville 92 by the lower P values. However, the total amount of P adsorption x/m maximum, (Xb values In Table 4), which were calculated from Langmulr straight lines, showed that In corase textured soils, (Hillsdale, Locke and Miami), the B horizon has the ability to adsorb more F than surface soil, e.g. 30,8 and 32, 2 5 .7 and 29.25, 28.6 and 31.00 for Hillsdale, Locke and Miami, A and B horizons respectively. This indicated that B horizon of coarse textured soils have the ability to adsorb more P from solution than A horizons. The greater ability is due to the accumulation of clay and Pe In B horizon. But In fine textured soils I.e. Sims and Hoytville, the total maximum adsorption are about 32.9 and 32.7 for Sims A and B and 41.7 and 40.2 mg p/100 g soil for Hoytville A and B horizon respectively. Thi3 shows that the ultimate ability to adsrob P from the solution for A and B horizons of these two soils are not different even though Figure 12, 13 showed that the B horizon of these two soils adsorbed more P than the A horizon at any given (P) under study. For fine textured soil the difference in clay content between the surface and subsoil Is not as much for coarse textured soils. The first straight line of Langmulr plotting may correspond to the steep part of the curve x/m versus (P) in. Figure 9 to 1 3 if This part of the curve corresponded to the part of the soil that adsorbed P more tight­ ly than that of the second part of the curve. 93 **■.2.2 Phosphorus Adsorption by Flooded Soils. **.2.2,1 Phosphorus Adsorption Isotherms for Flooded Surface 3oils. After flooding soil, an increase in extractable P has been found (pannamperuma, 1972). For the five Michigan soils under study, Srisen (1971) found that the amount of Flouride (Bray's), Sodium bicarbonate (Olsen's) and diluted H^SO^ (Troug's) extractable P also increased upon flooding1 (Table I.e. in Appendix). This would suggest that the P adsorption behavior of these soils would change due to flooding and the tendency to decrease P adsorption could be anticipated since the solubility and the amount of extractable P increased. Phosphorus adsorption by flooded and anaerobic surface soils did change upon flooding, (Table 6 and Figures 19 to 23.) But as shown in Table 6 the amount of P adsorbed (x/m) as a function of initial P con­ centration increased upon flooding, especially for Hillsdale A, Miami A, and Sims A soils. For Locke A, the amount of P adsorbed after flooding is somewhat smaller than that of the air dried soil at low P concentration, but at higher P concentration (at 21.*+5 and 29.** ppm P initially), the amount of P adsorbed by flooded soils become greater. On the other hand, the amount of P adsorbed by the Hoytville A horizon decreased upon submergence. These results are shown in Figures 19 to 23. Phosphorus adsorption by flooded soils and by lake sediment has been studied by many workers who found that the disappearance of P from the sol­ ution by submerged soils was related to active Fe and Fe^O^ content of the soils (chlang, 19&3 Be Datta et al,, 1966* and Mandal and Das, 19?0j Shukla et al., 1971) Shukla et al. (1971) also reported that noncalcareous lake sediment adsorbed more P than calcareous ones, and 94 Table 6. Amount of P adsorbed by anaerobic surface soils*, (P) Initial Soil 0 day Hillsdale A Locke Miami 31ms A A A Hoytville A PPa 3.^5 6.20 8.86 10.7 14.7 16.7 21,4 29.4 3.45 6.20 8.86 10.7 14.7 16.7 21.4 29.4 3.45 6.20 8.86 10.7 14.7 16.7 21.4 29.4 3.45 6.20 8.86 10.7 14.7 16.7 21.4 29.4 3.45 6.20 8.86 10.7 14.7 16.7 21.4 29.4 1 average of 3 replications. 3.00 5.01 6.93 a.i 10.1 11.1 12.7 15.3 3.07 5.32 7.07 7.97 9.96 10.7 12.0 13.4 2.99 4.97 6.71 7.62 9.87 10.5 12.2 14.6 3.14 5.56 7.68 8.89 11.5 12.6 14.9 17.9 3.33 5.91 8.34 9.91 13.4 14.9 18.4 2 3 .2 Amount of P adsorbed (m/m) Incubation Time 1 day 3 days g soil— — — — 3.U 2.95 5.26 5.57 7.49 7.72 0.74 8.89 11.2 U.3 12.6 12.0 14.2 14.7 18.0 17.8 2.80 3.06 5.11 4.95 6.67 6.59 7.42 7.70 10.0 9.33 10.6 10.3 12.0 12.0 15.2 15.7 3.12 3.11 5.61 5.59 7 .6 1 7.72 8.74 8.73 11.2 11.2 12.6 12.3 13.8 14.3 1 7 .6 18.3 3.28 3.30 5.81 5.95 8.38 8.17 9.94 9.67 12.8 13.3 14.4 15.1 17.7 18.3 2 3 .8 22.7 3.08 3.24 5.62 5.57 7.80 8.09 9.26 9.85 12.4 12.5 14.0 1 3 .8 16.8 16,5 22.2 21.9 7 days 3.04 5.12 7.24 8.15 10,6 ~ 14.1 19.4 2.88 4.88 6.75 6.87 8.95 — 12.5 19.9 3.21 5.82 7.52 8.37 11,1 — 17.5 20.7 3.27 5.89 8.48 10.0 13.3 - 18.5 24.4 2.95 5.38 7.82 9.03 12.2 13.2 17.2 2 3 .0 95 Figure 19. Amount of P adsorbed by air dried and flooded Hillsdale A soils. Figure 20, Amount of P adsorbed by air dried and flooded Locke A soils. Adsorbed F (x/m in mg/100 g soil) 96 3 days air dried Equilibrium P Cone. (in X10 -We/1) Adsorbed F (x/m in mg/100 g soil) Figtir* 19. 3 days air dried Equilibrium P Cone, (in XlCT^mole/l) Figure 20. 97 Figure 21. Amount of P adsorbed by air dried and flooded Miami A soils. Figure 22. Amount of P adsorbed by air dried and flooded Sims A soils. 98 o W O •H o 3 days ■aH a > 10 lr dried 1 a < 5 - Equilibrium P Cone* (In XlO mole/l) a 2° S to i 3 days air dried o d ■< 5 10 15 _ 20 Equilibrium P Cone, (in XlOmole/l) 99 20 air dried •H days >10 y 10 Equilibrium F Gone Figure 23. Amount of P adsorbed by air dried and flooded Hoytville A soils. 100 oxalated soluble P played an Important role in P adsorption by the lake sediments. De Datta et al. (1966) reported that there is no change in P concentration in solution during 4 days equilibrium with calcareous soils. Greenhouse studies showed that the amount of IN NHjjGAc extractable Fe of Michigan soils under study increased considerably upon flooding (see Table l.f. in Appendix), and reached the maxima about 3 to 6 weeks after flooding (Srisen, 1971). For Hillsdale A , Miami A and Sims A, the amount of Fe increased greatly within 3 weeks of submergence, while for Locke A and Hoytville A, the amount of extractable Fe was also increased but the increase was muoh less within 3 weeks of submergence and reached the maxima slower at about 6 weeks after flooding. But in all five surface soils the extractable P increased greatly within 3 weeks of submergence, (see Table I.e., in Appendix). The Increase in extractable Fe and extractable F upon flooding the soils should change the ability of soils to adsorb P after flooding. However, the amount of extractable Fe appears to play a more important role In this case. For Hillsdale A, Miami A and Sims A, the increase in extractable Fe within 3 weeks of submergence may cause increases in P adsorption by these soils upon flooding. l However, for Locke A and Hoyt- ville A with much lower extractable Fe within 3 weeks of flooding, (19.7, 12.0 and 33.7 ppm for those three soils compared with 1.0** and 2.73 ppm Fe for Locke A and Hoytville A), the amount of P adsorbed decreased a little in flooded treatments. This nay be due to a greater Increase in soluble P than could be adsorbed upon flooding within 7 days of incubation. 101 Therefore the P that remained in solution after shaking with the soils In flooded treatment wus greater than that of air dried soils. The adsorption of P by flooded soil would differ from that of air dried soil In this aspect. For air dried soils, the number of P ad­ sorption sites does not change and the soil adsorbs added P from the solution. For flooded soils the number of P adsorption sites Is more or less changed due to the chemical changes under anaerobic condition, e.g. extractable Fe and P Increase, and flooded soil adsorbs not only applied P from the solution, but it adsorbs native P that Is released Into sol­ ution during chemical changes upon flooding. As concluded by Ponnamperuma (1972) acid clay high In Fe gave a low Increase in solution P while sandy soils low In Fe gave the most Increase in P In the solution of flooded soils. This would Indicate the Importance of Fe in removal of P from solution, some of which is fixed by adsorption reaction and the other portion is precipitated, Langmulr adsorption isotherms for flooded surfaced soilsare shown in Figures 24 to 28, and P Langmulr adsorption data showed In Table l.g. to 1. In Appendix, The Figures show that the longer the time of In­ cubation the greater Is the deviation from a straight line. The bonding energy constant, K, and the adsorption maximum, b, were calculated and are given In Table 7, The strength of F adsorption in part I, Indicated by values, changed a little upon flooding. The tendency to decrease of the energy of adsorption in this part of the soil is shown by Hillsdale A, Locke A, and Hoytville A, However, for Miami A and Sims A soils the Increased a little upon flooding. values 102 Table 7. Langmilr adsorption parameters for anaerobic surface soils. Incubation time Soil Hillsdale A Part I Part II *2 b2 Ki bi days x 10Jt l/mole rag/lOO g soil x 10-“ l/mole 0 1 3 7 2.01*7 2.392 2.231 1.880 12.9 15.8 15.** 1**.0 0.8?2 0.962 0.871 0.663 0 1 3 7 3.535 2.200 2.592 2 .6 6 1 10.5 10.8 10.7 8.9 1.262 0,7**8 0.833 0.367 0 1 10.7 11.8 i**.l 10.7 0.73** 1.2**2 3 7 2.593 **.**53 3.283 **.96** 0 l 3 7 3.889 **.629 **.167 **.3**7 0 1 3 7 7.**07 **.687 7.778 **.167 £b ag/ioo g soil- 2 2 .9 30.8 37.5 36.7 36.9 1 5 .2 2 7 .** 16.9 17.1 21.7 27.7 0.777 0.**59 17.9 19.2 2**.3 32.5 28.6 31.0 38.** **3.2 12.9 16.0 20.0 20.0 1.351 0.998 0.961 0.788 20.0 32.9 37.1 **3.8 32.9 **8.9 57.1 63.8 15.0 13.3 1**.3 13.3 2 .2 0 6 26.7 2 5 .** 25.** 37.8 **1.7 38.8 39.7 51.1 17.9 2 1 .7 2 1 .2 Locke A 2 7 .8 3 0 .6 Miami A Sins A Hoytville A 1.309 1.267 0.588 103 □ o ■ cv o « o oa o Figure 2b. Langmulr P adsorption Isotherms for flooded Hillsdale A soil. lOf* o r» 3 3 ►* £ ■3 ■a 3 H m 0 (D O £ *9 r□ * o 45 u o o " Figure 25. langmulr P adsorption Isotherms for flooded Locke A soil. 105 O I T\ CM o ■ o O •CM Figure 26. Langmulr P adsorption isotherms for flooded Miami A soil. 106 c«l ,o Fleur© 2?. Langmulr P adsorption isotherms for flooded Sims soil. 10? CsJ Figure 28. Langmulr P adsorption isotherms for flooded Hoytville A soil. 108 The adsorption maximum in part I of adsorption, b^, increased in Hillsdale .A, Miami A and Sims A soils. While Locke A soil the adsorption maximum mas relatively constant and Hoytville A soil showed a slight de­ crease in b^ values. In the second part of F adsorption by soils, the bonding energy constant, K2, decreased upon flooding, this would Indicate that soils adsorb F more loosely in this part of the adsorption than in air dried soils. The adsorption became weaker when the incubation time Increased up to 7 days of incubation in this study. The ability to adsorb F, indicated by values, showed that the adsorption maximum In part II increased upon flooding, e.g. from 17.9 to 21 , and from 20 to U-3 mg P/lOO g soil for Hillsdale A and Sims A after 3 days of Incubation. However, for Hoytville A the adsorption maximum did not increase until 7 days after flooding. The Increase in F adsorption maximum and the decrease in bonding energy for F adsorption in this part, associated with the small change in both b and K values in part I, indicated that the change in the behavior of P adsorption by surface soils after flooding is due to the changes in this portion of the adsorption. Generally the ability of the soil to adsorb F increased but with weaker banding energy than those of air dried soil. The total amount of F adsorption maximum, (£b in Table 7)» Increased upon submergence especially for Hillsdale A, Miami A and Sims A soils. However, the total adsorption maximum for Locke A and Hoytville A soils did not increase until after 7 days of incubation. It should be noted that the adsorption capacity of surface soils generally increased upon flooding but the F was held with a lower bonding 109 energy than In air dried soils. This change Is due to the chemical changes In the soil after flooding. In flooded soils, P became more soluble and Is released in to the soil solution and could, be adsorbed on the reactive sites of the soil. It has been found that (greenhouse studies) the amount of P in solution was very small compared to the extractable P for these five soils (Srlsen, 1971). These showed that P was adsorbed after it was released into the solution after flooding for a period of time. This adsorbed P may account for the increase in extractable P after flooding. It should be also noted that the break point in Langmulr curve between the two parts is between P values of 5 to 10 x 10”^ mole P/l (about 1,8 to 3.1 ppm P). This P concentration is somewhat higher than the amount of P in the natural lake water, Syers et al, (1973 a) con­ cluded that the P concentration in anaerobic water at the bottom of the Lakes is about .1 to ,3 ppm P. Srisen (1971) found that after 3 weeks of incubation the concentration of P in water is less than 1 ppm under green­ house studies for Michigan soils. This means that the amount of P re­ mained in water of flooded soils would have the possibility of removal from the solution if the contact time was longer, (e.g, by adsorption, precipitation or microbal consumption). However, under laboratory studies, the concentration of 5 to 10 x 10**^ mole P/l in the solution after shaking would be considered to be the concentration of P that provide loosely bound P by the soils under flooding conditions in this study. The P adsorption reaction occurring in flooded soils should be of benefit to rice cultivation especially in the paddy field with uncontroll­ able water area. Phosphorus in the water which may come from native or 110 applied P would not be lost with drainage water. It can be adsorbed and can be used by rice during the growing season. It should be noted that the disappearance of P from P solution occurred by the two mechanist©* chemical adsorption and precipitation. Precipitation would occur when the concentration of P and the reactive +3 counter cations such as Fe jl.O , Fe u/i * Al 4.o and Ca were sufficiently high and the solubility products axe higher than the solubility product constant of the respective P compounds. Because the solubility of Fe and P in reducing conditions in flooded soil 1b higher than that of air dried soil* the possibility exists that P may be removed from solution by precipitation, especially when an anaerobic soil becomes aerobic. An example of this reaction is the oxidized zone In paddy rice soil (Patrick and Nahapatra, 1968). Evidence of this was found in this study during an initial experiment, When the surface soils were Incubated in anaerobic chamber for 7 days and were shaken with P solution air leaks developed in the stoppers of some of the samples which then were contacted with air atmosphere during shaking periods. The loss of P from the solution in those tubes was much greater than with the other samples suggesting that precipitation occurred. The data on this Investigation is shown In Table 8. Ill Table 8 . Effect of air leaks In P adsorption by flooded soils. F Initial Soil Hillsdale A P at equilibrium replication 1 2 --- -— — — --- --- -----— 3.45 0.58 6.20 1.24 8.86 1 .7 0 10.7 2.80 14.7 4.10 16.7 1.28 21.4 T35 29.4 4.00 average 3 ppm— --- -- ----- 0.42 0.30 0.91 0.27 1.54 1.03 2.33 0740 1.00 O o 2T5o 4.00 1790 7.15 3.00 10.0 0.44 0.81 1.42 1.84 3.03 2.69 5.52 5.67 Locke 3.45 6.20 8.86 *°*7 14.7 0.64 1.31 2.11 1.72 0.73 0.06 1765 0.21 JJ55 5.75 1 6 .7 2 1 .4 1 .0 0 1 .0 0 1750 3.30 C T 7762 8 .8 7 2 .7 8 2 .6 0 4 .0 9 9.50 6.81 0 .2 0 0.40 0.05 0735 0 .2 1 1 .1 1 0755 0*5,5 1765 3733 3.87 2.35 3.36 3.16 HT5 5735 0.17 0.31 0.96 1.04 2.52 1.80 1.88 7.08 0 .1 0 0 .2 1 29.4 0 .5 0 0.40 0.52 0739 0 5 O o 575o 1 .1 6 0.90 2.00 Miami 3.45 20 Sims Hoytville 8.85 10.7 14.7 16.7 21.4 29.4 0.27 0.18 1735 oj# 3^5 0 .6 0 37^ 9 .0 0 3.45 0.14 0 .3 2 6 .2 0 8 .8 6 0 .2 6 0.44 0.07 0 .75 10.7 14.7 16.7 21.4 29.4 3.45 0.40 0 .6 0 1 .1 0 0 .6 0 ^00 479? 4.44 0*45 0 8 0 .7 0 1 .0 1 I.9 7 1.54 3.07 0.55 0 .5 0 0 .5 0 0.92 1 .1 2 0 .9 2 0.47 1 .1 0 0.85 10.7 1.70 1.69 1.9? 14.7 2 .5 0 1.73 0.15 1.46 16.7 3.18 1.70 3754 2.91 21.4 2JQ0 4741 3.50 3.30 29.4________ 3.18_______6.40_____0.70_______ 3.43 1 Underlined values were suspected of having alxleaks. 6 .2 0 8 .8 6 0.44 0*22 0.98 1.70 1 .5 9 0.39 0.77 0.33 Oloo 3?I5 2.80 0.18 0.31 0.29 •112 4'.2,2.2 Phoaphorttfl Adsorption Isotherm for Flooded Subsurface So^p*. Chemical changes also have been found In subsoils after flooding (Srlsen, 1971 )> The amount of extractable F Increased greatly but Has still smaller than that of flooded surface soils (Table i.e. in Appendix). This may be because the extractable F and organic matter con­ tent in air dried subsoils was much smaller than In surface soils (Table l.a. In Appendix), The amount of extractable Fe In flooded subsoils also increased but the increase was much smaller than in flooded surface soils, even In Sins B and Hoytville B soils which have more extractable Fe than that of those surface soils before flooding, (Table l.f. In Appendix). Actually the total amount of Fe in subsoils Is greater than in surface soils through the process of Fe accumulation. The smaller change In the extractable Fe would indicate a lesser degree of reduction in subsoils after flooding. This may be due to more anaerobic metabolism In surface soils than in subsoils associated with higher organic matter content. Phosphorus adsorption behavior In subsoils is also changed upon flooding, (Table 9 and Figure 29 to 33). Table 9 shows the amount of P adsorbed as a function initial F concentration. It shows that at low Initial F concentrations, up to 9,6 ppm F, the amount of F adsorbed did not change upon 7 days of flooding except for Sims B soil which showed a slight Increase. But at higher P concentrations, flooded subsoils adsorbed more F than unflooded ones, especially Locke B, Miami B, Sims B and Hoytville B soils. However, for Hillsdale B soil, the amount of F adsorbed was slightly decreased upon 7 days of submergence. This Is shown In Figures 29 to 33. Air dried Hillsdale B soil contained slightly higher extractable F when compared with other subsoils, (Table l.a. In Appendix). i However, the 113 Table 9* Amount of P adsorbed by anaerobic Soil (P) Initial Hillsdale B ppm 3.50 6.30 8.92 9.63 11.7 1 7 .0 Locke B 24.2 3.50 6.30 8.92 9.63 11.7 1 7 .0 tfiami B 24.2 3.50 6.30 8.92 9.63 11.7 1 7 .0 Sims B Hoytville B 24.2 3.50 6.30 8.92 9.63 11.7 17.0 24.2 3.50 6.30 8 .9 2 9.63 11.7 1 7 .0 24.2 1/ Average of 3 replications. Amount of P adsorbed (x/m) Incubation Time 7 days 0 day 1 day -------- -- mg/1 0 0 g soil— • 3.48 3.48 3.49 6.24 6.23 6.27 8.80 8 .6 6 8.75 9.36 9.37 9.37 1 1 .2 15.* 18.6 3.48 6.23 8.69 9.32 1 1 .0 15.8 17.7 3.47 6.23 8.74 9.25 1 1.1 1 5 .0 17.3 3.4? 6 .2 1 8 .6 6 9.32 1 1.1 14.8 1 7 .2 3.47 6 .2 0 8.69 9.35 1 1 .1 1 5 .8 1 1 .2 1 6 .0 18.7 3.47 20.4 3.48 6.23 8.78 9.33 6 .1 6 8 .6 2 9.24 1 0 .8 1 0 .7 1 1 .2 15.6 17.5 3.32 5.96 8.19 8.52 15.2 15.9 19.7 3.48 6.15 8.92 9 .8 6 13.3 16.5 3.48 6.25 8.83 9.48 U.3 16.2 21.3 1 7 .2 3.38 5 .8 6 8,15 8.57 9.67 13.3 15.2 3.49 6.27 8.84 9.49 11.4 16.3 20.7 9 .2 0 10.9 15.9 21.7 3.49 6.28 8.85 9.51 11,4 16.4 21.3 114 Figure 29. Amount of P adsorbed by air dried and flooded Hillsdale B soils. Figure 3 0 . Amount of P adsorbed by air dried and flooded Locke B soils. g soil) 115 P (x/in in mg/100 air dried Adsorbed days L i 10 15 -5 20 , Equilibrium P conc.(in X10 mole/1) Figure 29. 20 g soil) 7 days Adsorbed P (x/m in mg/100 air dried 5 oL ; --------- rt-------------------- is- " Equilibrium P conc.(in X10 mole/l) Figure 30. 116 Figure 31* Amount of P adsorbed by Miami B soil, air dried and 7 days Incubation. Figure 32. Amount of P adsorbed by Sims B soil, air dried and 7 days incubation. 117 20 7 days to air dried 10 15 c 20 Equilibrium P Cono. (is XI0 ^molo/l) O - 20 7 days air dried 0^ 5* T■ o « 7]I ■ 20 Equilibrium P Cono* (in Xl0”:>B»ole/l) 118 *o 20 7 days air dried 5 -5 TO Equilibrium P eonc.(in X 10 mole/l) Figure 33. Amount of P adsrobed by air dried and flooded Hoytville B soils. 119 amount of extractable Fe of this soil after 3 weeks of submergence was very small. This would account for the slight decreased In P adsorption upon 7 days of flooding. Locke B and Miami B also have a very small ex- tractable Fe content after 3 weeks of flooding but these soils contain lower amounts of extractable P than Hillsdale B In air dried treatment. Whereas Sims B soil in which the increase in P adsorption is greater than the other subsoils after flooding, contains higher extxactable Fe within 3 weeks of submergence. The P langrauir adsorption data for flooded subsoils are shown in Tables l.k to l.m in Appendix, and the plots (P)/(x/m) versus (p) are shown o b Figures 3**- to 38. The bonding energy constant K and the adsorption maximum were calculated and are given in Thble 10. In the part I of P adsorption, the strength of P adsorption and the ability of subsoil to adsorb P, (indicated by K^, and b^ values in Table 10), were relatively unchanged for Hillsdale B, Locke B, and Miami S soils. However, for Sims B both K^and b^ values increased after 7 days of flooding. While Hoytville B, soil adsorbed somewhat more P and with a little lower energy In this part of P adsorption. In part II of P adsorption, flooded soils adsorbed P with a lower bonding energy than air dried soils, (indicated by Kg In Table 10), except for Hoytville B the values increase upon flooding. The adsorption maximum in this part of the soils was also changed upon flooding. For Locke B and Miami B soils, the b^ values Increased a little, whereas Sims B soil, the adsorption maximum Increased consider­ ably, However, for Hillsdale B and Hoytville B soils, the adsorption showed a slight decrease in the adsorption maximum upon flooding. 120 Table 10, Langrauir adsorption parameters for anaerobic subsoils. Soil Hillsdale B » Locke Miami B B Part II K1 bl K2 b2 lb days x 10"5 mg/lOOg soil x 10"5 rag/lOOg soil uig/lOOf soil 0 1 7 8,333 8.392 8,392 12,0 10.8 10.8 .769 .779 .736 20,0 18.3 18.1 0 l 7 5.926 5.758 5.758 11.2 11,6 11,6 .694 .556 .600 18.0 20.0 22,0 0 1 7 4.286 3.897 4.04-0 11.7 11.7 11.2 .431 .405 .458 19.3 19.0 2 3 .0 30.7 34.2 11.7 11.3 13.8 .149 .097 .139 21.0 25.7 40.0 32.7 37.0 53.7 11.2 12.2 13.0 .575 2 9 .0 .8 6 7 23.1 24.4 40.2 35.3 37.4 B 0 1 7 Hoytville B 0 l 7 Sins Part I Incubation time .857 ,884 1.322 8,889 8,000 7.500 .909 3 2 .0 2 9 .2 2 9 .2 2 9 .2 31.6 33.6 3 1 .0 121 CM CM Figure 3*t. Langnulx F adsorption Isotherms for flooded. Hillsdale B soil. 122 o CM m cm o Figure 35. langmulx P adsorption Isotherms for flooded Locke B soil. o a ir dried 0.0 a 2b hours 0.5 7 days □ Figure 3 6 . Langsuir P adsorption 1b at he nos for flooded 1.0 B soil, 12U O o o* Figure 3 7 . Langmulr P adsorption Isotherms fox flooded Sins B soil. 125 JS v\ « © Figure 3 8 . Langnuir P adsorption isotherns for flooded Hoytville B soU. 126 The capacity of subsoils to adsorb P, Indicated by the total adsorp­ tion maximum, (z,b values In Table 10), also reflected a similar change. However, the change is not as pronounced as for surface soils except Sims B. The change In the behavior of P adsorption by subsoils upon 7 days of flooding, both change In bonding strength and the capacity to adsorb F Is similar to that of surface soils, I.e. generally Increase the capacity of adsorption with more loosely bound P to the soil surface, and generally most of the change is in part II of adsorption. However, the degree of change Is not as great as that of surface soils (See Table 7 and 10). Because the change in the capacity of reduction In surface soils associated with high organic matter content, Is greater than that of subsoils, (i.e. greater increase in extractable Fe and Mn etc.), upon flooding, the difference occurs in the P adsorption by the surface and subsoils. For Sims B soil, the amount of organic matter Is somewhat greater than that of other subsoils (in air dried treatment, Table l.a in Appendix). This may account for the faster increase in the extractable Fe content upon 3 weeks of flooding, (Table l.f in Appendix). The higher content of extractable Fe in flooded Sims B soil may account for the much greater P adsorption with in 7 days of incubation. It should be noted that subsoils, both flooded and air dried, adsorbed almost the same quantity of P from the solution at low P concen­ tration. For surface soils the change in the adsorption at low P concen­ tration is not as great as in higher P concentrations. versus (p) of flooded soils little upon flooding. The plots (P)/(x/m) showed that the part I of P adsorption changed Thissuggested that the adsorption sites in this 127 port m y be due to A1 or other sources which did not change upon flooding. But for the part II of P adsorption! the adsorption sites may be due to Fe or the other sources which were Influenced by the reduction process. However* the further studies on the sources of the P adsorption sites in both part could give a better explanation an this matter. The mechanism of P adsorption is not a single type of reaction. As cited by Shapiro and Fried (l959)» P adsorption consists of anion adsorp­ tion an the surface of clay* adsorption on the surface of Fe and A1 oxides or hydroxide and CaCO^ and Indirectly anion adsorption by means of a cation bridge to the clay. Udo and Uzu (1972) noted that clay forms a part of complex gel which consists of hydrated Fe^O^ along with small amounts of organic matter, AlgO^ and associated Si(OH)^ and P. This com­ plex gel is considered to be the main P adsorption site of sediments, (Shukla et al.,1971). However, Harter (1969) noted that organic matter also provides the main site of P adsorption. Therefore clay particle, oxides or hydroxide of Fe, and Al, CaCO^, organic matter and exchangable cations i.e. Fe+2 , Fe+3 , Al4*3, and Ca+2 would provide directly or in­ directly the sources of P adsorption sites In the soils. Some of which were Influenced by reduction reactions, which account for the change in the behavior of P adsorption upon flooding. CONCLUSIONS (1) Soils adsrobed more P from 0.01 M CaCl^ solution than from water solution. The adsorption reaction in 0,01 M CaClg treatment reached equilibrium faster in 24 hours, while for water solution treatment, the adsorption reaction required 48 hours or longer to reach equilibrium. (2) Soils adsorbed more P from 0.01 M CaClg solution adjusted to pH 6 than from unadjusted pH treatment at high P equillbirum con­ centrations. The adsorption reaction in the pH adjusted treatment also reached equilibrium at 24 hours of shaking. (3) Air dried, fine textured soils adsorbed more P than coarse textured ones. Subsoils adsorbed more P than surface soils. How­ ever, for coarse textured soils, subsoils adsorbed more P than did surface soils, but for fine textured soils the difference in the amount of P adsorbed in the A and B horizons was not great. (4) Langmuir adsorption studies showed that there are at least two different portions of the soils with respect to P adsorption, the tightly and loosely bound portions, (5) The behavior of P adsorption changes upon flooding surface soils. The changes mostly occurred in the loosely P bound portion. The amount of adsorbed P increased with a lower bonding energy upon flooding. Extxactable Fe apparently played an important role in the change in P adsorption. (6) The increase in P adsorbed by surface soils after flooding may be responsible for the Increase in the extractable P. Phosphorus adsorption by flooded surface soils, therefore, should be of benefit 128 129 for rice cultivation in lowland or paddy soils, (7) For subsoils, the change also occurred in the loosely P bound portion of the soil upon flooding. The amount of P adsorbed Increased and possessed a lower bond strength than air dried soils. changes were not as great as for surface soils. However, the LITERATURE CITED LITERATURE CITED Aomlne, S . 1962 , A review of research on redox potentials of paddy soils in Japan. Soil Sci, 94* 6-13. Aranki, A ., S.A. Syed, E.B. Kenney, and R. Freter. 1969 . Isolation of anaerobic bacteria from human gingiva and mouse cecum by means of a simplified glove box procedure. Applied Microbiology. 17* 568 -5 7 6 . ASami, T. and Y. Takai. 1970. Behaviour of free iron oxide in paddy soils 1 Part 4. Relationship between reduction of free iron oxide and formation of gases in paddy soils. J. Sci. Soil and Manure (Japan), 41* 46-55* Bacho, B.1if. 1964, Aluminium and iron phosphate studies relating to soils* II. Reactions between phosphate and hydrous oxides. J. Soil Sci. 15* 110-116, Barrow, G ,M, 1966. Physical chemistry. McGraw-Hill. New York. Barrow, N ,J. 1970. Comparison of the adsorption of molybdate, sulfate and phosphate by soils. Soil Sci, 109* 282-288. Barrow, N, J. 1972. Influence of solution concentration of calcium on the adsorption of phosphate, sulfate, and molybdate by soils. Soil Sci. 113* 175-180. Black, C./.. 1942. Phosphate fixation by kaolinite and other clays as affected by pH, phosphate concentration, and time of contact. Soil Sci. Soc. Amer. Proc. 7* 123-134. Bohn, H.L, 1968, Electromotive force of Inert electrodes in soil suspensions. Soil Sci. Soc, Amer, Proc, 32* 211-215. Bohn, H.L, 1971. Bremner, Redox Potentials, Soil Sci., 112* 39-45. .11. and K. Shaw. 1958. Denitrification in soil* II Factors affecting denitrification. J. Agr. Sci. 51* 40-52. Broechart. H., E. Haunold, and M. Fried. 1965. The effect of water conditions and oxidation-reduction status of rice soils on the availability of soil and fertilizer phosphate. Plant and Soil. 23* 305-313. 130 * 131 Ohakravarti, S. and T.K. Ghoshal, 1968, Anaerobic reduction and phosphate solubility, J. Indian Soc, Soil Sci, l6t *+09-413. Chlang, G.T. 1963 a, A study on the availability and form of phosphorus in paddy soils (No l)t The interrelationship between available phosphorus and soil pH, Eh, Soil and Fertilizer in Taiwan. PP 61. Chlang, G.T. 1963 A study on the availability -and form of phosphorus in paddy soil (No 2)i Changes in the state of fixation and availability of phosphorus. Soil and Fertilizer in Taiwan, pp. 62-63, Chiang, G.T. 1968, Availability and form of phosphorus in paddy soils (No V)i Factors affecting the solubility of soil phosphorus. Taiwan Fertilizer Co., Ltd., Res. Bull, No. 30 pp. 11-37. Cho, D.Y. and F.N. Ponnamperuma. 1971. Influence of soil temperature on the chemical kinetics of flooded soils and the growth of rice. Soil Sci. 112* 184-19^. Cole, C.V., S.R, Olsen, and C.0, Scott. 1953. The nature of phosphate sorption by calcium carbonate. Soil Sci. Soc. Amer. Proc, 17* 352-356. Collins, J.F. and S.W. Soul, 1970. Effects of fluctuations in the Eh-pH environment on iron and/or manganese equilibria. Soil Sci. 110* 111-118. Connell, W.E., and W.H. Patrick, 1968, Sulfate reduction in soil* Effects of redox potential and pH. Science. 159* 86-87* Be Datta, 3.K., J.C. Moomaw, V.V. Racho and G.V. Simsiman. 1966, Phosphorus supplying capacity of lowland rice soils. Soil Sci. Soc. Amer. Proc, 30* 613-617* Ellis, B.G. 1973* The soil as a chemical filter, in “Recycling treated municipal wastewater and sludge through forest and cropland." Ed. W.E. Sopper and L.T. Kardos. Pennsylvania State Univ. Press, pp. **6-70. Ellis, B.G., and A.E. Erickson, 1969* Movement and Transformation of various phosphorus compounds in soils. Report to Michigan Water Resources Commission, Ellis, B.C., and B.D. Knezek. 1972. Adsorption reactions of micronutrients in soils. Chapter *+ in nMicronutrients in Agric­ ulture ." Ed. J.J. Moztvexdt, P.M. Giordano, and W.L. Lindsay. Soil Sci. Soc. Amer., Inc, Madison, Wisconsin, pp. 59-75* 132 Ellisf R. and H. Truog. 1955* Phosphate fixation by Montmorillonite, Soil Sci. Soc. Amer. Proc. 19* 451-454. Eriksson, E. 1952. The physlo-chenlcal behaviour of nutrients In soils. J. Soil Sci. 3* 238-250. Fried, M. and H. Broeshart. 196?. The soll-plant system In relation to Inorganic nutrition. Academic Press. New York and London. Fried, M. and L.S. Bean. 1955. Phosphate retention by Iron and Alum­ inium in cation exchanges systems. Soil Sci. Soc. Amer. Proc. 19* 1*1-3-147. Fried, H. and E.R. Shapiro, 1956. Phosphate supply pattern of various soils. Soil Sci. Soc. Amer. Proc. 20* 471-475. Furukawa, H. and K. Kawaguchi. 1969. Contribution of organic phosphorus to the Increase of easily soluble P in waterlogged soil, espec­ ially related to phytic phosphorus (inositol Hexa Phosphate). J. Sci. Soil and Manure (Japan). 40* 141-1*18, Gasser, J K.R. 1956. The effect of anaerobically fermenting rice straw (Oryza satlva) on the mobilization of aluminium, calcium, and iron phosphates. 6th Int. Congr. Soil Sci. Vol. Ct 479-491 ■ Giles, C,: I. 1959. Hydrogen bonding. Bd. D. Hadzi. New York. Permagon press. Gillespie L.J. 1920. Reduction potentials of bacterial cultures and of water-logged soils. Soil Sci. 9* 199-216. Greenwood D.J. 1962. Nitrification and nitrate dissimilation in soil It Method of studying nitrate dissimilation. Plant and Soil. 17t 365-377. Gotoh, S, and W.H. Patrick. 1972, Transformation of manganese In a water­ logged soil as affected by redox potential and pH. Soil Sci. Soc. Amer. Proc. 361 738-742. Gotoh, S. and W.H. Patrick, 1974, Transformation of iron in a water­ logged soil as Influenced by redox potential and pH. Soil Sci. Amer. Proc. 381 66-71. Gotohr S. and K. Yamashlta, 1966. Gxldatlon-xeductlon potential of a paddy soil In situ with special reference to the production of ferrous iron, manganous manganese and sulfide. Soil Sci. Plant Nutrition (Japan), 12« 230-238. Gunary, D 1970. A new adsorption isotherm for phosphate In soil. Soil Sci. 211 72-77. J. 133 Harter, ]l.D. 1968. Adsorption of phosphorus by lake sediment. Soc, Amer. Proc, J2t 514-518. Soil Sci, Harter, ]t.D. 1969. Phosphorus adsorption sites in 3olls, Soil Sci, Soc. Amer. Proc. 33* 630 -632 . Hayes, F ,R. and J.E. Phillips, 1958. Lake water and sediments XV, Radiophosphorus equilibrium with mud, plants and bacteria under oxidized and reduced conditions. Limnology and Oceonography 3t **59-475. Hera, J.D i960 . Restrains on dissolved ferrous iron imposed by bicar­ bonate, redox potential and pH. U.S. Geol. Survey WaterSupply paper 1459-B* 33-55. Hesse, F R. 1962. Phosphorus fixation in mangrove swamp muds. Nature. 193* 295-296. Hsu, F.H 1964. Adsorption of phosphate by aluminium and iron in soils. Soil Sci. Soc. Amer. Proc. 281 4?4-478. Hsu, P.H and D.A, Rennie. 1962. Reactions of phosphate in aluminium systems1 X. Adsorption of phosphate by X-ray amorphous "aluminium hydroxides." Canadian J, Soil Sci. 42i 197-209. I.R.R.I. 1963. International RiceResearch Institute, Annual Report. I.R.R.I. 1964. International RiceResearch Institute, Annual Report. X.R.H.J. 1965 . International RiceResearch Institute, Annual Report. X.R.R.I, 1966. International RiceResearch Institute, Annual Report, Jackson, M.L. 1965 . Soil Chemical Analysis. Prentice Hall Inc. 5th ed, Englewood Cliffs, New Jersey. Jefferry, J.W.O. I960. Iron and the Eh of waterlogged soils with par­ ticular reference to paddy. J. Soil Sci. lit 140-148. Kurtz, T, , E.E. Deturk and R.H. Bray. 19+6. Phosphate adsorption by Illinois soils. Soil Sci. 611 111-124. Langmuir I. 1918. The adsorption of gases on plane surfaces of glass, mica and platinum. J. Amer. Chem. Soc. 401 1361-1402. Low, F.F, and C.A. Black. 1950. Reactions of phosphate with kaolinite. Soil sci. 70i273-290. 134 Mahapatra, I.e. 1968. Effect of flooding on soil reactions and mobil­ ization of various nutrients. J. Indian Soc. Soil Sci. l6i 149-153. Mahapatra, I.e. and W.H. Patrick. 1969* Inorganic phosphate trans­ formation in waterlogged soils. Soil Sol. 107i 281-288. Mandal, L.N. 1961. Transformation of iron and manganese in water­ logged rice soils. Soil Sci. 91t 121-126. Mandal, L.N. 1964. Effect of time, starch, and lime on the transformat­ ion of inorganic phosphorus in a water-logged rice soil. Soil Sci. 97* 127-132. Mandal, L.N. and R.K. Das. 1970. Transformation of applied water sol­ uble phosphate in acidic lowland rice soils. Soil Sci. HOi 263-267. Mitsui, S, I960, Inorganic nutrition, fertilisation and soil amelior­ ation for lowland rice. 4th ed. Yokendo Ltd. Tokyo. Moore, W.J. 1972. Physical Chemistry. Prentice-Hall, Cliffs, N.J. Englewood Mortimer, C.H. 1924. The exchange of dissolved substances between mud and water In the lakes. J. Ecol. 30i 147-201. Mukhopadhyay, A., T.R. Fisher and G.E. Smith. 1967. Submergence and liming effects on solli I. Changes in pH, Eh andmanganese uptake by riceplants. Soil Sci. 104« 107-112. Muljadi, D., A.M. Posner, and J.P. Quirk. 1966. The mechanism of phosphate adsorption by kaollnlte, gibbeite, and pseudoboehmitei Part I. The Isotherms and the effect of pH on adsorp­ tion. J. Soil Sci. 17i 212-229. Murphy, J. and J.P. Riley. 1962. A modified single solution method for the determination of phosphate in natural waters. Anal. Chem. Acta. 2?i 31-36. Nhung, M.M. and F.N. Ponn&wperuna. 1966. Effects of calcium carbonate, manganese dioxide, ferric hydroxide and prolonged flooding on chemical and electrochemical changes and growth of rice In a flooded acid sulfate soil. Soil Sci. 102i 29-41, Olsen, S.R. 1953. Inorganic phosphorus in alkaline and calcareous soils. Chapter 4 In "Soil and Fertilizer Phosphorus In Crop Nutrition,'* Agronony monograph. 4i 89-122. Olsen, S.R. and F.S. Watanabe. 1937. A method to determine a phosphorus adsorption maximum of soils as measured by the Langmuir Isotherm. Soil Sci. Soc. Amer. Proc. 211 144-149. 135 Olomu, M.O., G.J. Racz and C.M. Cho. 1973. Effect of flooding on the Eh, pH, and concentrations of Fe and Hn In several Manitoba soils. Soil Sci. Soc. Amer. Proc, 371 220-224. Patrick, V.H. I960. Nitrate reduction rates in a submerged soil as affected by redox-potential. 7th Int. Congr. Soli Sci. Vol. 31 49^-500. Patrick, W.H. 1964. Extractable Iron and phosphorus In a submerged soil at controlled redox-potentials. 8th Int. Congr. Soil Sci. Vol. 2* 59-61. Patrick, W.H. 1966, Apparatus for controlling the oxldatlon-reductlon pot­ ential of waterlogged soils. Nature, 212* 1278-1279. Patrick, W.H. and I.C. Mahapatra. 1968, Transformation and availability to rice of nitrogen and phosphorus In waterlogged soils. Adv. Agron. 20* 323-359. Patrick, W.H, and F.T. Turner. 1968. Effect of redox potential on man­ ganese transformation in waterlogged soil. Nature. 220* 476478. Patrick, W.H., S. Gotoh and B.G. Williams. 1973* Strengite dissolution in flooded soils and sediments. Science. 179* 564-565. Pearsall, W.H., and Mortimer, C.H. 1939. Oxldatlon-reductlon potentials in waterlogged soils, natural waters and muds. J. Ecol, 27* 483-501. Pierce, B.3. 1953. Oxldatlon-reductlon potential and specific conductance of ground water* Their influence on natural forest distribution. Soil Sci. Soc. Amer. Proc, 17* 61-65. Ponnamperuma, F.N. 1965. Dynamic Aspects of flooded soils and the nut­ rition of the rice plant.In,"The mineral nutrition of rice plant.H PP 295-328. John Hopkins Press, Baltimore, Maryland. Ponnamperuma, F.N, 1972. The chemistry of submerged soils. Adv. Agron. 24* 29-96. Ponnamperuma, F.N., and R.U. Castro. 1964. Redox systems in submerged soils. 8th Int. Congr. Soil Sci. Vol. 2t 107-109. Ponnamperuma, F.N., W.L. Yuan and M.T.M. Nhung. 1965. Manganese dioxide as a remedy for a physiological disease of rice associated with reduction of the soil. Nature. 207* 1103-1104. Ponnamperuma, F.N., E. Martinez, and T. Loy. 1966 a. Influence of redox potential and partial pressure of carbon dioxide on pH values and the suspension effect of flooded soils. Soil Sci. 101* 421-431. 136 Ponnairrperuma, F.N., E.M. Tianco and T.A. Loy. 1966 b. Ionic strengths of the solutions of flooded soils and other natural aqueous solutions from specific conductance. Soil Sci. 102* 408-4-13. Ponnamperuma, F.N,, E.M. Tianco and T, Loy. 196?. Redox equilibria in flooded soilst I. The iron hydroxide systems. Soil Sci. 103t 374-382. Ponnamperuia, F.N., T.A, Loy, and E.H. Tianco. 1969. Redox equilibria in flooded soils| II. The manganese oxide systems. Soil Sci. 108i 48-57. Puttaawamygowia, B.S. and P.F, Pratt. 1973. Effects of straw, calcium chloride, and submergence on a sodic soil. Soil Sci. Soc. Amer, Proc. 37* 208-212. Rennie, D.A. and H.B. McKercher. 1959. Adsorption of phosphorus by four Saskatchewan soils. Canadian J. Soil Sci. 39* 64-75* Robinson, O.K. 1930. Some chemical phases of submerged soil conditions. Soil Sci. 30i 197-217. Russell, G.C. and P.F. Low. 195*** Reaction of phosphate with kaolinite in dilute solution. Soil Sci. Soc. Amer. Proc. I8t 22-25. Savant, N.K. and R. Ellis. 1964, Changes in redox potential and phos­ phorus availability in submerged soil. Soil Sci. 96i 388-39**. Shapiro, R.E. 1958 a. Effect of flooding on availability of phosphorus and nitrogen. Soil Sci. 85i 190-197* Shapiro, R.E. 1958 b. Effect of organic matter andflooding an avail­ ability of soil and synthetic phosphates. Soil Sci. 85t 267-272. Shapiro, R.E. and Fried. 1959* Relative release and retentiveness of soil phosphates. Soil Sci. Soc. Amer. Proc. 23i 195-199* Shukla, S.S., J.K. Syers, J.D.H. Williams, D.E. Armstrong, and R.F. Harris. 1971* Sorption of inorganic phosphate by lake sedi­ ments. Soil Sci. Soc. Amer. Proc. 35* 244-249. Srisen, M. 1971* Transformation of phosphates in soils after flooding as measured by chemical extraction and uptake by rice. M.S. Thesis. Michigan State University. Syers, J.K., R.F. Harris and D.E. Armstrong. 1973 a* Phosphate chemistry in lake sediments. J. Environmental Quality. 2i 1-14. Syers, J.K., T.D. Evans, J.D.H. Williams, and J.T. Murdock. 1971* Phosphate sorption parameters of representative soils from Rio Grande Do Sul, Brazil. Soil Sol. 112i 267-275* 137 Syers, J.K., M.G. Brownian, G.W, Smlllle, and R B, Corey, 1973 k. Phosphate sorption by soils evaluated by the Langirulr adsorption equation. Soil Sol, Soc, Amer, Proc. 37* 358-363. Takkar, I’.N. 1969. Effect of organic matter on soil Iron and manganese. Soil Scl. IO81 108-112. Terman, Gi.L., S.E. Allen, and O.P. Engelstad. 1970. Response by paddy rice to rates and sources of applied phosphorus. Agron. J. 6Z1 390-394. Turner, F‘.T. and W.H. Patrick. 1968, Chemical changes in waterlogged soils as a result of oxygen depletion. 9th Int. Congr. Soil Scl. Vol. 4-* 53-65. Udo, E.J. and P.O. Uzu, 1972. Characteristics of phosphorus adsorption by some Nigerian soils. Soil Scl. Soc. Amer. Proc, 36 s 879-833* Watanabe, P.S. and S,R. Olsen. 1965. Test of an ascorbic acid method for determining phosphorus In water and NaHCO^ extracts from soil. Soil Scl. Soc, Amer. Proc. 29i 677-678. Weir, C.C . and R.J. Soper, 1962. Adsorption and exchange studies of phosphorus in some Manitoba soils. Canadian J. Soil Scl. 421 31-42. Williams, B.C. and W.H. Patrick. 1971. Effect of Eh and pH on the diss­ olution of strengite. Nature Physical Scl. 234i 16-1?, Williams, B.G. and W.H. Patrick, 1973. The dissolution of complex ferric phosphates under controlled Eh and pH conditions. Soil Scl. Soc, Amer. Proc. 37* 33-36. Williams, G.H. and J.R. Simpson. 1965. Some effects of cultivation and waterlogging on the availability of phosphorus in pasture soils. Aust. J. Agric. Res. 16* 413-427. Woodruff, J.R. and E.J. Kamprath, 1965. Phosphorus adsorption maximum as measured by the Langmiix isotherm and its relationship to phosphorus availability. Soil Scl, Soc. Amer. Proc. 29* 1*18-150 . Yuan, W.L . and F.N. Ponnamperuma. 1966. Chemical retardation of the reduction of flooded soils and the growth of rice. Plant and Soil. 25i 347-360. Zobeli, C .E. 1946. Studies on redox-potential of marine sediments. Bull. Am. Assoc. Petro. Geol, 30* 477-513* APPENDIX Table 1 a. Physical and chemical properties of air dried Soil Type Texture PH Eh Organic Extractable^ Matter JO# V * Hillsdale A Sandy Loam 5.82 p2/ Extractable^/ Mn Fe Bray's Olsen's Troug's ?> 1.66 7.24 0,?4 57.9 25.9 71.9 5.20 595 660 0.41 6.05 0.44 5.3 6.5 60.8 A Sandy Loam 6.00 536 2.72 1.59 0.15 41,7 23.7 64.3 B 6.29 58? 0.40 0,82 0.29 5.5 37.0 A Loam 6.48 544 2.84 1.45 0.59 3.5 4.8 17.5 B 561 0,48 1.20 0.59 2,6 A Clay Loam 6.52 6.23 3.5 4.1 599 6.77 0.50 0.29 39.5 42.7 113.1 B 6.54 575 0.93 0,82 0.59 2.7 1.7 80,8 Hoytville A Clay Loam 6.50 599 3.16 0.50 0,29 7.5 4.5 51.6 B 6.76 525 0.56 0.76 0,44 1.3 3.2 40.3 B Locke Miami Sims 1/ I N NH^CAc extractable Fe 2/ Saturated CaSO^ extractable Kn }/ a. Bray's P, 0,03 N NH^F in 0.25 M HClj b. Olsen's ?, .5 K NaHCO^ (pH 8 .5)1 c. Troug's Pt .002 N HgSO^ (buffer with (HH4 ) 2 SO^) 4/ From Srisen, M, 1971* M.S. Thesis, Michigan State University. 5.6 Table 1 b. The effect of time, CaCl^ and water solution on P adsorption by air dried soils. Initial P concentration and soil 3 5 P Adsorbedl/.S/ 24 12 48 72 hours-2/ mg P/100 g Soil Hoytville A 1 ppm P in HgO 4.70 4.85 5 .2 5 5.25 5.82 5.75 " In 0.01 M CaCl? 5.76 5.87 5 .8 8 9.40 9.85 3 12 ** In H£0 7.85 8 .2 0 8.76 4 12 " In 0 .0 1 M CaCl2 10.35 1 0 .7 0 10.78 11.00 11.05 5.72 9.90 11.32 2 6 4.55 5 .0 0 6 Locke A 1 6 ppm P in HgO 2.52 2.55 2.76 3.30 3 .6 0 4.18 2 3 4 6 " in 0.01 M GaClg 12 " In HgO 12 M in 0.01 M CaGlg 3.80 4.40 2 .6 0 3 .0 0 7 .0 0 5.05 4.90 8.30 5.27 5.47 6.70 4.75 4.10 7.76 5.30 6.55 9.05 l/ pH of P solution was not adjusted 2j Average of 2 replications, 2/ Shaking Time. 9 .0 0 Table 1 c. Effect of pH of P solution in 0.01 M CaCl_ on P adsorption by soils. Soil _______ Unad.just pH________ Initial BquiliAdsorbed P brium P^/ P ppm Hillsdale A 0 3 6 9 12 15 20 30 Miami Hoytvllle A 0 3 6 9 12 15 20 30 0 3 6 9 12 15 20 30 x 10^ mole /l 1.17 3.89 7.44 12.9 19.7 29.1 **•9.2 1.52 3.94 8.23 12.9 20.0 30.0 52.5 0.40 0.58 2.22 3.39 5.33 10.6 23.8 1/ average of 2 replications. mg/lOO soil _______ Adjust pH________ Initial Equlli- Adsorbed pi/ brium P P ppm x 105 mole/l mg/100 soil 2.64 4.90 6.70 8.00 8.90 11.0 14.7 3.45 6.20 8.85 10.7 14.7 16.7 21.4 29.5 1.47 3.83 6.21 8.41 14.8 18.2 28.2 45.6 2.99 5.01 6.94 8.12 10.0 11.1 12.7 15.3 2.53 4.78 6.45 8.00 8.80 10.7 13.7 3.45 6.20 8.86 10.7 14.7 16.7 21,4 29.4 1.49 3.98 6.94 10.0 15.6 19.9 29.7 47.8 2.99 4.97 6.71 7.65 9.88 10.6 12.2 14.6 2.88 5.82 8.30 10.9 13.3 16.7 22.6 3.45 6.20 8.86 10.7 14.7 16.7 21,4 29.4 0.39 0.95 1.67 2.63 4.15 5.79 9.77 20.0 3.33 5.91 8,34 9.91 13.4 15.0 18.4 2 3 .2 Table Id. Soil P adsorption from 0.01 M CaCl^* pH 6.0 solution as a function of time. Initial P 3 5 10 Locke A hours 48 72 P/100 g £ ppm Sims A Shaking Time 16 24 6.2 4.59 4.85 5.28 5.41 5.61 14.7 9.10 9.76 10.67 11.31 11.68 6.2 4.07 4.34 5.13 5.33 5.41 14.7 7.42 8.15 8.80 10,08 10.88 5.75 5.76 12.54 12.68 5.51 5.53 11.74 11.96 Table 1 e, Quantity of extiactable P, from air dried soils and from soils after 3 weeks of flooding^Soil air dried Troug's P Olsen's P Bray's P flooded air dried flooded air dried Soil water P content flooded 3 weeks of flooding 133.0 0.31 40,4 0.15 0.41 -ppm-------Hillsdale A 57.9 109.2 25.9 41.6 B 5.3 41,0 6.5 12.5 71.9 60.8 A 41.7 80.7 23.7 32.0 64.3 160.0 B 3.5 52.7 5.5 37.0 68.9 A *.8 3.5 17.5 73.2 0.15 0.22 B 2.6 51.3 26.6 7.5 6.0 4.1 4.4 5.6 68.9 0.16 A 39.5 65.6 42.7 55.6 214.0 B 2.7 49.9 1.7 16.0 113.1 80.8 118,0 0.73 0.10 Hoytville A 7.5 17.3 4.5 20.6 51.6 149.0 0.18 B 1.3 49.9 3.2 13.3 40.3 77.5 0.14 Locke Miami Sims 1/ Prom Srisen, M. 1971. MS Thesis Michigan State University. Table 1 f. Saturated CaSO^ extractable Mn, and 1 N NH^C&c extractable Fe from air dried and flooded soils^/. Soil ________ Mn_________ air flooded soils dried 3 weeks ___________Fe___________ air flooded soils dried 3 weeks 6 weeks •ppm Hillsdale A 7.24 1 2 3 .0 0 0.74 19.70 15.30 B 6.05 8.96 0,44 1.04 2.45 A 1.59 8.75 0.29 1.04 13.70 B 0.82 3.40 0.15 0 .2 0 1 .2 2 A 1.45 8 1 .7 0 0.59 1 2 .0 0 9.86 B 1 .2 0 1 2 .6 0 0.59 0 ,2 0 5.13 A 0.50 7.35 0.29 38.70 45.10 B 0.82 9.74 0.59 4.43 7.47 Hoytville A 0 .5 0 4.05 0.29 2.73 55.80 B o.? 6 1 ,6 1 0.44 1 .6 8 5.85 Locke Miami Sims 1/ From Srisen, M. 1971 M.S. Thesis, Michigan State University. Table 1 g. Langmulr P adsorption Isotherm Tor air dried surface soils Initial P Soils Hillsdale A Locke Miami A A ppm 3.45 6.20 8.86 10.7 14.7 16.7 21.4 29.4 Equilibrium (?) ppm 0.45 1.19 1.92 2 .6 0 4.58 5.64 8.73 14.1 0.38 0.88 1.79 2.75 4.72 6.00 9.42 (mole/l)X105 1.47 3.83 6.21 8.41 14.8 18.2 28.2 45.6 3.45 6.20 8.86 10.7 14.7 16.7 21.4 29.4 15.5 1.23 2.84 5.77 8.88 15.2 19.4 30.4 50.1 3.45 6.20 8.86 10.7 14.7 16.7 21.4 29.4 0.46 1.23 2.15 3.10 4,82 6.18 9.21 14,8 1.49 3.96 6.94 10.0 15.6 20.0 29.8 47.8 P adsorbed^/ (x/ra) (P)(x/m) ng/100 g soil 3.00 5.01 6.93 8.12 10.1 11.1 12.7 15.3 X105 0.49 0.76 0.90 1.04 1.46 1.64 2.21 2.98 3.01 5.32 7.07 7.97 9.98 10.7 12.0 13.9 0.40 0.53 0.82 1.11 1.53 1.80 2.53 3.61 2.99 4.97 6.71 0.50 0.80 7 .6 2 1.31 1.58 1.89 2.43 3.2? 9.88 10.6 12.2 14.6 1 .0 3 Table l.g. (cont’d.) Soils Sins Initial P A Hoytville A Equilibrium (P) ppm ppm 3.^5 6,20 8,86 10.7 14,7 16,7 21.4 29.4 o.3l 0.64 1.18 1.83 3.17 4.10 6.52 11.4 3.45 6.20 8.86 10.7 14.7 16.7 21.4 29.4 0.12 0.29 0.52 0.82 1.28 1.79 3.02 6.20 (P)/(*/®) mg/100 g soil XIO5 1.00 2.08 3.81 5.92 10,2 13.2 21.0 37.0 3.14 5.56 7.68 8.89 11.5 12.6 14.9 17.9 0 .3 2 0.39 0.95 1.67 2.64 4.15 5.79 9.77 20.0 3.33 5.91 8.34 9.91 13.4 15.0 18.4 23.2 (mole/1 )X105 1/ Average of 3 replications. P adsorbed^/ (x/m) 0.37 0.50 0.6? 0.89 1.05 1.41 2.05 0,12 0.16 0.20 0 .2 7 0.31 0.39 0.53 0.86 Table 1 h. Langmuir P adsorption isotherms for flooded surface soils, incubated 24 hours. Initial P Soils Hillsdale A Locke A (P)/(x/m) ppm (aole)XlO5 mg/100 g soil XIo5 3.**5 6.20 8.86 10.? 14.7 16.7 21,4 29.4 0.34 0.63 1.14 1.96 3.42 4.17 6.72 11.4 1.09 2.03 3.68 6.40 11.0 13.** 21.7 36.8 3.11 5.57 7.72 8.74 11.3 12,6 14.7 18,0 0.35 0.36 0.48 0.73 0.98 1.06 1.4? 2.04 3.**5 6.20 8.86 10.7 14.7 16.7 21.4 0.65 1.25 2.27 3.30 5.37 6.42 9.39 2.09 4.03 7.32 10.7 17.3 20.7 30.3 44.0 2.80 4.95 6.59 7.42 9.33 10.3 12.1 15.8 0.74 0.81 1.11 1.44 1.86 2.01 2.51 2.80 1 .0 5 3.12 5.61 0.3** 0.3** 0.53 0,7** 1.00 • A (P) P adsorbed-i/ (x/m) ppm CM Miami Equilibrium pi/ 3.45 6.20 8.86 10.7 14.7 16.7 21.4 29.4 1 3 .6 0.33 0.59 1.25 2.00 3.**7 4.40 7.16 11.7 1.92 4.03 6.45 11,2 14.2 23.1 37.9 7 .6 1 8 .7 2 11.2 12.3 14.3 17.6 1.15 1.6l 2.15 Table i h. (cont'd.) Initial P Soils Sims A Hoytville A Equilibrium (P) ppm PPm 3.45 6.20 8.66 10.7 14.7 16.7 21.4 29.4 0.17 0.39 0.69 3.45 6.20 8.86 10.7 14.7 16.7 21.4 29.4 (mole/l )X105 1 .0 6 1.86 2.33 3.74 6 .6 3 0.21 0 .6 2 1.06 1.47 2,16 2.48 4.60 7.12 t/ Average of 3 replications. P adsorbed^/ (x/m) rag/100 g soil 0.54 1.25 2.22 3.42 5.35 7.52 12.0 21.4 3.28 5.81 8.17 9.67 12.8 14.4 17.7 22.8 0.69 2.02 3.44 4.73 6.97 8.00 14.8 3.24 5.57 7.80 9.26 12.5 14.1 16.8 22.3 2 3 .0 (P)/(x/ra) X105 0.16 0.21 0.27 0.35 0 .4 7 0 .5 2 0.68 0.94 0.21 0.36 0.44 0.51 0 .5 6 0.56 0.88 1.03 Table 1 1, Langmuir P adsorption isotherms for flooded surface soils, incubated 3 days. Soils Initial P Hillsdale A ppm (mole/l )X105 mg/1 0 0 g soil 3.45 0 .5 0 2.95 6 .2 0 0.94 1.37 1.84 3.39 4.68 7.18 11.5 1.61 3.05 4.42 5.94 10.9 15.1 1 0 .7 14.7 16.7 21.4 29.4 A 3.45 6 .2 0 8 .8 6 Miami A (P) P adsorbed^/ (x/ra) ppm 8 .8 6 Locke Equilibrium pi/ 0.37 1.09 2.19 3 .0 2 2 3 .2 37.2 5 .2 6 7.49 8.89 11.3 1 2 .1 14.3 17.9 1 .2 0 3 .0 8 3.53 7.05 9.77 5.11 6.67 7.70 10.7 14.7 4,70 1 5 .2 1 0 .0 1 6 .7 6 .0 5 10.7 21.4 29.4 9.38 14.2 19.5 30.3 45.8 3.45 0.34 1 .1 1 6 .2 0 0.61 8 .8 6 1.14 1.99 3.49 4.11 7.58 11.0 1.96 3.67 6.41 10.7 14.7 16.7 21.4 29.4 11.3 13.3 24.5 35.7 (P)/(x/a) XIo5 0.54 0.58 0.59 0.67 0.97 1.25 1 .6 2 2 .0 8 0.39 0.69 1 .0 6 1 2 .1 1.27 1.52 1.83 2.51 1 5 .2 3 .0 1 3.U 5.59 7.72 8.74 0.35 0.35 0.48 0.64 1 1 .2 1 .0 0 1 2 .6 1 .0 5 13.9 18.3 1.76 1.94 Table 1 1. (cont'd.) Soils Sims Initial P A P adsorbed^/ (x/m) ppm ppm (mole/l )X1 0 ^ mg/1 0 0 g soil 3.45 0.15 0.25 0.48 0.78 1.36 0.05 0.81 1.54 2.25 4,40 1 .6 0 5.17 9.99 17.9 3.30 5.95 8.38 9.94 13.3 15.1 18.4 6 .2 0 8 .8 6 1 0 .7 14.7 16.7 21.4 29.4 Hoytville A Equilibrium F ^ (P) 3.45 6 .2 0 8 .8 6 10.7 14.7 16.7 21.4 29.4 3.0? 5.56 0.37 0.58 0.77 0 .8 8 2.25 2.89 4.94 7.43 1/ Average of 3 replications. 1.18 1.89 2,48 2.83 7.25 9.34 16.0 24.0 2 3 .8 3 .0 8 5.62 8.09 9.85 12.4 13.9 16.5 2 2 .0 (F)/(x/m) XIO5 0.15 0.14 0 .1 8 0 .2 5 0.33 0.34 0.54 0.75 0.38 0.34 0.31 0.29 0.58 0.67 0.97 1.09 Table 1 j. Langmulr P adsorption isotherms for flooded surface soils, incubated 7 days. Soils Initial P Hillsdale A Locke Uiami A A ppm 3.45 6.20 8.86 10,7 14.7 16.7 21.4 29.4 3.45 6.20 8.86 10.7 14.7 16.7 21.4 29.4 3.45 6.20 8.86 10.7 14.7 16.7 21.4 29.4 Equilibrium p^/ (P) ppm 0.44 1.08 1 .6 2 2.58 4.05 - (mole/1 )X105 1.41 3.48 5.23 8.32 13.1 - P adsorbed^/ (x/m) (P)/(xA0 mg/100 g soil 3.01 5.12 7.24 XIO5 0.47 0,68 0.72 1.02 1.23 8.15 10.6 14.1 19.4 - 1.68 1.66 7.33 10.0 23.7 32.3 0.57 1.31 2.10 3.85 5.75 9.50 1.84 4.25 6.79 12.43 18.6 28.6 30.7 12,6 19.9 2.28 1.54 0.24 0.37 1.34 2.35 3.50 0.77 1.21 4.33 7.59 11.6 3.21 5.83 7.52 8.37 11.2 0.24 0.21 0.58 0.91 1.01 8 .8 7 - 3.92 8.68 - 12.6 28.0 2,88 4.89 6.76 6.87 8.95 _ - 17.5 20.7 0.64 0.69 1.01 1.81 2,07 - - 0.72 1.35 Table 1 j. (cant'd.) Initial P Soils Sims A Hoytville A Equilibrium 0?) P adsorbed--^ (x/m) (P)/(x/m) ppm ppm (mole/1 )X105 mg/100 g soil XIO5 3.45 6.20 8.86 io.7 14.7 16.7 21.4 29.4 0.18 o.3l 0.40 0 ,6 0 3.27 5.89 8,46 10.0 13.4 0.18 0.17 0.15 0.22 0.33 3.45 6.20 8.86 10.7 14.7 16.7 21.4 29.4 0 .7 0 1.34 - 0.99 1.28 2.25 4.34 - 3.16 4.94 10.2 0.50 0.82 1.04 1 .7 0 2.50 3.51 4.41 6.40 1/ Average of 3 replications. - - 15.9 18.6 34.5 0.55 0.65 1.60 2.66 3.37 5.47 8.07 11.3 14.2 20.7 2.95 5.38 7.82 9.03 12,2 13.2 17.2 23.0 0.54 0.49 0.43 o.6l 0,66 0.86 0.84 0.90 Table 1 k. Langmilr P adsorption isotherms for air dried subsoils. Initial P Hillsdale B B ppm (mole/1 )X105 mg/1 0 0 g soil 3.50 6.30 0 .0 1 8 .9 2 0 .1 2 9 .6 2 3.49 6.27 8.80 9.37 1 7 .0 0.25 0.50 1.47 0.04 0.09 0.40 0.82 24.2 3.50 6.30 8.92 9 .6 2 11.7 17.0 24.2 Miami Sims B B (F) P adsorbed^ (x/m) ppm 11.7 Locke Equilibrium 3.50 6.30 8.92 9.62 11.7 17.0 24.2 3.50 6.30 8.92 9 .6 2 11.7 17.0 24.2 0 .0 3 (P)/(x/m) XIo6 0 .1 2 0,14 0.45 0.87 1.43 3.08 9.76 1 .1 6 1 1 .2 5 .6 2 4.78 18.1 15.5 18.6 0 ,0 2 0,05 0.07 0.23 0.31 0 ,2 2 0 .7 0 2 .2 6 1 1 .0 2.05 1.29 6.50 4.16 15.8 17.7 2 .61 3.47 6.23 8.74 9.25 10.8 15.6 17.5 0,28 0.35 0.76 0.99 2 1 .0 0 .0 3 0 .1 0 0 .0 7 0.22 0.18 0.57 1.20 2.94 0.37 o.9l 1.58 6.70 5.H 21.6 0.17 0.34 0.73 1.10 1.85 3.73 7.72 0.56 1.11 2.37 3.56 5.99 12.0 24.9 3.48 6.23 8 .6 9 0.15 0.35 0.87 9.32 1 .0 6 3.33 5.96 8.19 8.52 9.86 13.3 1 1 .8 0 .6 5 1.30 •fO ro Soils 3.28 12.3 1.68 1.87 2.89 4.18 6.07 9.0? 15.1 Table 1 k. (cont'd.) Soils Hoytville B Initial P Equilibrium (P) ppm ppm (mole/l )X10^ 3.50 6.30 8.92 9.62 11.7 17.0 24.2 0.02 0.05 0.09 0.15 0.36 0.79 2.89 0.06 0.16 0,28 0,48 1/ Average of 3 replications. 1 ,1 6 2.56 9.33 P adsorbed^/ (x/ra) (p)/(x/m) mg/100 g soil XI0^ 3.*l8 6 .2 5 8.83 9.1*8 11,3 16.2 21.3 0.18 0.25 0.45 0.55 1.02 1.58 4.37 Table 1 1. Langnruir P adsorption isotherms for flooded subsoils, Incubated 24 hours. Soils Hillsdale B Locke B Equilibrium Initial P _______ (?) (mole/1 )X105 ppm ppm 0.02 0 .0 6 3.50 0 .0 6 6.30 0.19 8.92 0.17 0,53 9.62 0.82 0 .2 5 0.61 1.97 11.7 17.0 1.11 3.58 24.2 22.4 6.94 3.50 6.30 8.92 9.62 B 3.4? 6.21 8.66 9.32 11.1 15.8 18.7 3.47 6.16 XIo6 0.17 0 .3 0 0.62 0,88 1.78 2.25 12.9 0.27 0.49 0.97 0 .6 3 1 7 .0 1.23 5.53 0.03 0.14 0.30 0.38 0.98 1.80 6.96 0.09 0,44 0.97 1.24 3.16 5.83 22.5 9.24 10.7 15.2 17.2 0.26 0.72 1.13 1.34 2.94 3.83 13.1 3.50 6.30 8.92 0.12 0.43 0.77 9 .6 2 1 .0 6 0,38 1.40 2.48 3.42 6.58 11.8 29.0 3.38 5.87 8.15 8.57 9.67 13.3 15.2 1.12 2.39 3.05 3.99 6.80 8.89 19.1 3.50 6.30 8.92 11.7 17.0 24-,2 B mg/100 g soil 3.48 6.24 8.66 9.37 11.1 15.9 17.3 (?)/(*/») 11.7 9 .6 2 Sims (x/m) 0.09 0.31 0.84 0.99 2.04 3.97 17.8 24-.2 Miami P adsorbed^ 11.7 17.0 24-.2 0.03 0 .0 9 0.26 0.31 2.04 3.67 8.99 8 .6 2 1 .0 6 1.84 2.52 9.56 Table 1 1. (cont*d.) Soils Initial Equilibrium ___________ £ Hoytville B ppm 3.50 6.30 8 .9 2 9.62 11.7 17.0 24.2 P adsorbed^ fel— ppm 0.01 0.03 0.08 0.14 0.33 0.74 3.51 (mole/1 )Xl