Citation
Charge characteristics of selected soils from tropical areas

Material Information

Title:
Charge characteristics of selected soils from tropical areas
Creator:
Adams-Melendez, Melitón José, 1944-
Publication Date:
Language:
English
Physical Description:
xiv, 152 leaves : ill. ; 28 cm.

Subjects

Subjects / Keywords:
Acidity ( jstor )
Adsorption ( jstor )
Ions ( jstor )
Isotherms ( jstor )
pH ( jstor )
Phosphates ( jstor )
Phosphorus ( jstor )
Potreros ( jstor )
Soil science ( jstor )
Soils ( jstor )
Dissertations, Academic -- Soil Science -- UF
Soil Science thesis Ph. D
Soil colloids ( lcsh )
Soils -- Analysis -- Venezuela ( lcsh )
Soils -- Tropics ( lcsh )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis--University of Florida.
Bibliography:
Bibliography: leaves 143-151.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Melitón José Adams-Melendez.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
020663534 ( ALEPH )
04804028 ( OCLC )

Downloads

This item has the following downloads:


Full Text











CHARGE CHARACTERISTICS OF SELECTED SOILS FROM TROPICAL AREAS








By

MELITON JOSE ADAMS-MELENDEZ















A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY

















UNIVERSITY OF FLORIDA 1978

































To my Father, Melit6n Adams, who died May 1978
















ACKNOWLEDGEMENTS



The author wishes to express his deep gratitude to Dr. W.G. Blue, chairman of the supervisory committee, for his guidance, assistance, and patience during the preparation of this dissertation. The author is pleased to extend his sincere acknowledgements to Dr. T.L. Yuan, cochairman, Dr. V. Berkheiser, Dr. O.C. Ruelke, and Dr. A.H. Krezdorn for their participation in the supervisory committee and constructive criticism of this manuscript.

Special recognition is extended to Mr. Jorge Beltran for his invaluable help with the computer programs.

Appreciation is extended to others who contributed their time and knowledge in making this endeavor possible: Dr. P.S.C. Rao, for his help and encouragement during this program; the personnel of Instituto de EdafologFa, Facultad de AgronomTa (U.C.V.) for their help in some laboratory analyses, especially Mr. A. Prada, Mrs. L. Burguera; Mrs. I. Rojas, Mrs. M. AlegrFa, and Mr. W. Zupan; Mr. W.G. Pothier for his help with X-ray patterns interpretation; Mr. J. Miller, fellow graduate student, for those good discussions during this program; Mrs. Rossina Fern5ndez, for her patience in deciphering the handwriting and typing this manuscript; and Consejo CientTfico y Humanfstico de la Universidad Central de Venezuela for its financial support given during this graduate program.

iii










Last, but not least, the author would like to thank his wife, Nieves, and children, Lara and Josg, for their support and understanding throughout the course of this study.




















































iv
















TABLE OF CONTENTS



Page

ACKNOWLEDGEMENTS. . .. ......... . . . . . . . . . . . . . . iii

LIST OF TABLES. . . . . . . . . . . . .. . . . . . . . . . . . . . . vii

LIST OF FIGURES . . . . . . . . . . . . . . . . . . . ............ . xii

ABSTRACT. . ........... . . . . . . ................. . . . xiii

INTRODUCTION. . ................ . . . . . . . . . . 1

LITERATURE REVIEW . . . . . . . . . . . . . . . . .. . . .... .. 3

Soil Characteristics . . . . . . . . . . . . . . . . . . . . . 3
Mineralogical Characteristics . . . . . . . . . . . . . . 3
Soil Acidity Sources. . ............. . . . . 5
Lime Requirement. . ......... . ............ . 7
Potassium Quantity/Intensity Measures . ......... 7 Phosphorus Fractions. . ............... ... 9
Phosphorus Adsorption . ... ........... . . . . 11
Factors affecting adsorption equilibrium. . ..... 12
Thermodynamic parameters associated with adsorption
process . . . . . . . . . . . . . . . . . . . . . . . 16
Surface Charge . . . . . . . . . . . . . . . . . .. . . . . . . 19
Double-layer Models ................... . 19
Gouy-Chapman double-layer model . ... . ...... . 20
Stern double-layer model. . ............ . 21
Point of Zero Net Charge. . ..... ........... . . 22
Effects of Specific Adsorption on the PZNC. . .. ..... 23 Cation Exchange Capacity . ............... . 25
Effect of pH. . . . . . . . . . . . . . . . . . . . . 25
Effect of P adsorption. . . . . . . . . . . . . . . . 26
Crop Response to Lime and P Applications . ......... . 26

MATERIALS AND METHODS . . . . . . . . . . . . . . . . . . . . . . . 28

Soil Physical, Chemical, and Mineralogical Properties. . ... 31
X-ray Diffraction Analysis. . ............ . . 31
Fe-oxides Determination . . . . . . . . . . . . . . . . . 32
Surface Area Determination. . .............. . 32
Soil Acidity. . . . . . . . . . . . . . . . . . . . . . . 33
Potassium Q/I Characteristics . . . . . . . . . . . . . . 34









Page

Exchange Properties. . . . . . . . . . . . . . . . . . . 35
Point of Zero Net Charge . . . . . . . . . . . . . . . 36
Phosphorus Analysis . . . . . . .................. . 36
Calculations of Thermodynamic Parameters Associated with
Adsorption Processes . ....... . . . . . . . . . . 37
Phosphorus Adsorption Isotherms . . . . . . . . . . . . 37
Phosphorus Release . . . . . . . . . . . . . . . . . . 37
Lime and Phosphorus Treatments . . . . . . . . . . . . 38
Laboratory and Greenhouse Experiments . . . . . . . . . 39
Statistical Analyses . . . . . . . . . . . . . . . . . . . 39

RESULTS AND DISCUSSION . . .. . .... .. . . . . . . . . . . . . 41

Soil Characteristics with Emphasis on Surface Charge .... . 41
Clay Mineralogy. . . . . . . . . . . . . . . . . . . . . 45
Point of Zero Net Charge . . . . . . . . . . . . . . . . 46
Phosphorus and Sulfur Fractionation . . . . . . . . . . 46
Negative Surface Charge . ..... ... . . . . . . . . . . 73
Acidity. . . . . . . . . . . . .... . . . . . . . . . 73
Lime Requirement . .. .. . ... . . . . . . . . . . 77
Potassium Q/I Parameters . . . . . . . . . . . . . . . . 79
Positive Surface Charge . . . . . . . . . . . . . . . . . . . 82
Phosphorus Adsorption . . . . . . . . . . . . . . . . 83
Temperature effects . . . . . . . . . . . . . . . 83
Ionic system effects . . . . . . . . . . . . . . . 89
Thermodynamic parameters associated with adsorption
process . . . . . . . . . . . . . . . . . . . . . . 92
Phosphorus Release . . . . . . . . . . . . . . . . . . . 100
Changes in Charge Surface . . . . . . . . . . . . . . . . . . 103
Lime and Phosphorus Effects. . ........ . . . . . 103
Effects of Change of Charge Characteristics on Crop
Production . . . . . . . . . . . . . . . . . . . . . . . 118

SUMMARY AND CONCLUSIONS. . ............... . . . . . 130

APPENDIX I DETAILED CALCULATIONS FOR CaCO3 AND PHOSPHORUS
TREATMENTS ............ . .. .. . . . . . . 132

APPENDIX II DETAILED CALCULATIONS OF THERMODYNAMIC PARAMETERS BY
THE METHOD OF BIGGAR AND CHEUNG (1973). . . . . . . 135

APPENDIX III DETAILED CALCULATIONS OF SURFACE CHARGE. . ...... 139

LITERATURE CITED ................. . . . . . . . . . . . . . 143

BIOGRAPHICAL SKETCH . ....... ........... . . . . . 152




vi
















LIST OF TABLES


TABLES Page

1 PZNC of pure minerals found in soils. . ......... 24

2 Treatment combinations used in the laboratory and greenhouse experiment with the soils used. . ........ . 40

3 Physical and chemical properties of soils examined. . . 42

4 Exchange properties and free oxides of the soils
examined. . ..... .............. ..... . 43

5 Phosphorus fractionation of the virgin soils examined . 47

6 Sulfur fractionation of some of the virgin soils
examined. . . . . . . . . . . . . . . . . . . . .... 48

7 Effect of incubation temperature on phosphorus fractions
in a typic Paleustult from Anzoategui State, Venezuela
(0-30 cm) after applying several solution concentrations
of phosphorus . ..... ......... .... . . . . 50

8 Effect of incubation temperature on phosphorus fractions
in a Grossarenic Psammentic Haplustox from Anzoategui State, Venezuela (All horizon) after applying several
solution concentrations of phosphorus . ........ 51

9 Effect of incubation temperature on phosphorus fractions
in a Grossarenic Psammentic Haplustox from Anzoategui State, Venezuela (A12 horizon) after applying several
solution concentrations of phosphorus . . . ...... 52

10 Effect of incubation temperature on phosphorus fractions
in an Ultic Tropudalf from Tachira State, Venezuela
(0-30 cm) after applying several solution concentrations
of phosphorus . . . . . . . . . . . . . . . . . .. .. 54

11 Effect of incubation temperature on phosphorus fractions
in a Ustic Quartzipsamment from Portuguesa State, Venezuela (0-30 cm) after applying several solution concentrations of phosphorus. . ............... . 55

vii










TABLES Page

12 Effect of incubation temperature on phosphorus fractions
in an Oxic Tropudalf from Miranda State, Venezuela (0-30
cm) after applying several solution concentrations of
phosphorus. . . . . . . . . . . . . . . . . . . . . . . . 56

13 Effect of incubation temperature on phosphorus fractions
in a Tropeptic Haplustox from Carabobo State, Venezuela
(0-30 cm) after applying several solution concentrations
of phosphorus . . . . . . . . . . . . . . . . . . . . . . 57

14 Effect of incubation temperature on phosphorus fractions
in an Udorthentic Pellustert from Portuguesa State, Venezuela (0-30 cm) after applying several solution concentrations of phosphorus. . . . . . . . . . . . . . . . . . 58

15 Effect of incubation temperature on phosphorus fractions
in a Fluaquentic Humitropept from Portuguesa State, Venezuela (0-30 cm) after applying several solution concentrations of phosphorus. . ........ . . . . . . . . 59

16 Effect of incubation temperature on phosphorus fractions
in a Haplargid from Lara State, Venezuela (0-30 cm)
after applying several solution concentrations of phosphorus. . . . . . . . . . . . . . . . . . . . . . . . . . 60

17 Effect of incubation temperature on phosphorus fractions
in an Oxic Haplustalf from Barinas State, Venezulea (030 cm) after applying several solution concentrations of
phosphorus. . . . . . . . . . . . . . . . . . . . . . . . 61

18 Effect of incubation temperature on phosphorus fractions
in a Rhodic Paleudult from Florida, U.S.A. (Al horizon)
after applying several solution concentrations of
phosphorus. . . . . . . . . . . . . . . . . . . . . . . . 62

19 Effect of incubation temperature on phosphorus fractions
in a Rhodic Paleudult from Florida, U.S.A. (B2t horizon)
after applying several solution concentrations of
phosphorus. . . . . . . . . . . . . . . . . . . . . . . . 63

20 Values of pH2PO4 and pH at equilibrium with different
ionic systems after P adsorption for a typic Paleustult from Anzoategui State, Venezuela . . . . . . . . . . 64

21 Values of pH2PO4 and pH at equilibrium with different
ionic systems after P adsorption for an Ultic Tropudalf from Tichira State, Venezuela. . . . . . . . . . . 65

22 Values of pH2P04 and pH at equilibrium with different
ionic systems after P adsorption for an Ustic Quartzipsamment from Portuguesa State,Venezuela . . . . . . . . 66

viii










TABLES Page

23 Values of pH2PO4 and pH at equilibirum with different
ionic systems after P adsorption for an Oxic Tropudalf
from Miranda State, Venezuela. . ........... . 67

24 Values of pH2P04 and pH at equilibrium with different
ionic systems after P adsorption for an Udorthentic
Pellustert from Portuguesa State, Venezuela. . ... . 68

25 Values of pH2PO4 and pH at equilibrium with different
ionic systems after P adsorption for a Tropeptic
Haplustox from Carabobo State, Venezuela . ...... 69

26 Values of pH2P04 and pH at equilibrium with different
ionic systems after P adsorption for a Fluaquentic
Humitropept from Portuguesa State, Venezuela ..... 70

27 Values of pH2PO4 and pH at equilibrium with different
ionic system after P adsorption for a Haplargid from
Lara State, Venezuela. . ............... 71

28 Values of pH2PO4 and pH at equilibrium with different
ionic systems after P adsorption for an Oxic Haplustalf from Barinas State, Venezuela . ......... 72

29 Soil reaction with selected electrolyte and calculated
ApH value for the soils examined . .......... 75

30 Acidity from potentiometric titrations at pH 8.00 and
conductometric titrations with NaOH, Ba(OH)2, and
Na2B407 for the Venezuelan soils examined. . ..... . 76

31 Potassium Q/I parameters and Gapon's selectivity coefficient for the soils examined . .......... 81

32 Phosphorus sorption constants at different temperatures
from the Langmuir equation ............ . . 85

33 Freundlich equations for phosphorus adsorption at
different temperatures for the soils examined .... 87

34 Freundlich equations for phosphorus adsorption at
different ionic systems for the Venezuelan soils
examined . . . . . . . . . . . . . . . . . . . . . . . 91

35 Equations of the relationship between solution phosphorus concentration and reciprocal of time, and the K and AG values calculated from the concentration at
equilibrium for the soils examined . ......... 94


ix









TABLES Page

36 Thermodynamic parameters calculated using Freundlich
equations for phosphorus adsorption reactions for
the soils examined . . . . . . . . . . . . . . . . 96

37 Values of Ko, AGo, AHO, and AS* associated with the
adsorption of phosphorus for the selected soils
calculated by the Biggar and Cheung method (1973) . 97

38 Relationship between the accumulative release P and
the number of washings and linear correlation coefficients for the Venezuelan soils with different
ionic systems . . . . . . . . . . . . . . . . . . . . 101

39 Effect of treatments on electrochemical properties
measured before cropping for the Guanipa I soil . . . 104

40 Effect of treatments on electrochemical properties
measured before cropping for the Guanipa 5 soil . . . 105

41 Effect of treatments on electrochemical properties
measured before cropping for the El Potrero soil. . 106

42 pH and point of zero net charge for CaCO3 and phosphorus treatments in Guanipa I after cropping . . . . 108

43 pH and point of zero net charge for CaCO3 and phosphorus treatments in Guanipa 5 after cropping . . . . 109

44 pH and point of zero net charge for CaC03 and phosphorus treatments in El Potrero after cropping. . . . 110

45 Electrochemical potentials of CaC03 and phosphorus
treatments in soil Guanipa 1 after cropping ..... Ill

46 Electrochemical potentials of CaCO3 and phosphorus
treatments in soil Guanipa 5 after cropping . . . . 112

47 Electrochemical potentials of CaCO3 and phosphorus
treatments in soil El Potrero after cropping. .... 113

48 Charge distribution at the surface for CaC03 and
phosphorus treatments in Guanipa 1 after cropping .. 114

49 Charge distribution at the surface for CaCO3 and
phosphorus treatments in Guanipa 5A after cropping. . 115

50 Charge distribution at the surface for CaCO3 and
phosphorus treatments in El Potrero after cropping. . 116


X









TABLES Page

51 Dry matter yields of sorghum and cation exchange capacity after cropping for soil Guanipa 1 treated with CaC03
and phosphorus . . . . . . . . . . . . . . . . . . . . . 119

52 Dry matter yields of sorghum and cation exchance capacity after cropping for soil Guanipa 5 treated with CaC03
and phosphorus. . ............ . . . . . . . 120

53 Dry matter yields of sorghum and cation exchance capacity after cropping for soil El Potrero treated with CaC03
and phosphorus. . ............... . . . . . 121

54 Effect of cropping on CEC values for Guanipa 5 soil . 122 55 Effect of cropping on CEC values for Guanipa 1 soil . 123 56 Effect of cropping on CEC values for El Protrero soil . 124 57 Change in CEC after cropping for the soils selected . 125

58 Effect of soil, CaC03, P, and interaction CaC03xP on
the variables measured. . ................ 128

59 Prediction equation for pH, go, DM, and CEC to fit a
surface response for the three soils selected . .... 129

























xi
















LIST OF FIGURES


FIGURES Page

1 Relative location of Venezuela in South America. . ... 29

2 Relative location of the soil sampling sites in
Venezuela. . . . . . . . . . . . . . . . . . .. . . . . 30

3 Phosphorus fractions as affected by P concentration
after an ll-day equilibrium period of 25oC . ...... 53

4 Effects of CaC03 applications after 6 months incubation
period . . . . . . . . . . . . . . . . . . . . . . . . . 78

5 Potassium Q/I relationship for El Rocio soil . ..... 80

6 Langmuir adsorption isotherms for San Crist6bal soil in
nonsaturated systems at three temperatures . ...... 84

7 Freundlich adsorption isotherms for P by Guanaguanare,
San Crist6bal, and Red Bay B2t in nonsaturated systems
at three temperatures. . ................ 88

8 Freundlich adsorption isotherms for P by Guataparo and
San Crist6bal in different ionic systems at 250C . . . . 90

9 Relationship between P solution concentration and the
reciprocal of time for Guataparo, San Crist6bal, and
Red Bay B2t soils. . .................. . 95

10 Release of phosphorus from Guanipa 1, San Crist6bal, and
El Rocio soils . . . . . . . . . . . . . . . . . . ... 102















xii










Abstract of Dissertation Presented to the Graduate Council
of the University of Florida in Partial Fulfillment of the Requirements
for the Degree of Doctor of Philosophy


CHARGE CHARACTERISTICS OF SELECTED SOILS FROM TROPICAL AREAS

By

Melit6n Jos6 Adams-Melendez

December 1978

Chairman: Dr. W.G. Blue
Major Department: Soil Science

Ten soils from Venezuela and one from Florida, U.S.A., were investigated by chemical, physical, and mineralogical techniques to elucidate surface reactivity. Also, two Oxisols and one Alfisol were selected to determine the effect of lime and P levels on surface charge. A sorghum crop was planted under greenhouse conditions to investigate the changeof surface charge in each lime- and P-treated soil after cropping. Stern and Gouy-Chapman models for diffuse double layer together with the point of zero net charge (PZNC) and the surface area were used to calculate the charge distribution and the cation exchange capacity (CEC) of the soils for each lime and P treatment.

The following parameters were determined: (a) acidity sources; (b) quantity-intensity K relationships; (c) P fractionation in virgin soils and after P adsorption; (d) P-adsorption isotherms at different temperatures and with different ionic systems; (e) PZNC of virgin soils before and after lime and P treatments; (f) pH before and after cropping; (g) electrical conductivity before cropping; and (h) sorghum dry-matter yields.



xiii










The following parameters were calculated: (a) thermodynamic parameters associated with the adsorption process (AGO, AHO, and AS*) and

(b) charge distribution in double layer and CEC after lime and P treatments.

The results showed that exchangeable Al+3 and dissociation of OH groups from organic matter and silicate are responsible for the acidity in these soils. The soils had low available K, but very high selectivity for K. There was an increase in P adsorption with temperature in all soils. Lime increased P adsorption in two Alfisols and one Oxisol. There was an increase in both Al-P and Fe-P with temperature and P concentration applied. Phosphorus had a highly significant effect on PZNC decrease while P and lime increased CEC due to the combined decrease of PZNC and increased pH. Salt content as indicated by increasing electrical conductivity may have depressed the dry matter yield of sorghum in lime and P treatments. The negative AGO values and positive values of AHe and ASO showed that P adsorption in these soils is a spontaneous endothermic reaction that occurs with an increase of entropy resulting from displacement of water molecules by phosphate ions.
















xiv
















INTRODUCTION



Physico-chemical properties of a soil are determined by the surface characteristics of its organic and inorganic colloidal components. The quantity and type of those components are greatly affected by various soil-forming environmental conditions. Because of the great variations in climate, topography, etc., soils in Venezuela vary from relatively recently formed Entisols and Inceptisols to highly weathered Ultisols and Oxisols. Like other tropical regions, highly weathered soils dominate in Venezuela. These soils are mostly acidic and P deficient. Finding ways of making tropical soils more productive, by correcting soil acidity and making conditions favorable for nutrient retention and uptake by crops are the concern of most soil research in tropical countries. A study of surface phenomena such as proton release, cation selectivity, P adsorption, and the effect of liming and P application on surface charge can help our understanding of the behavior of highly weathered tropical soils with concomitant implications to their management. Of particular significance is the efficient use of P in soils because farmers wish to manage their chemical investment in the most economical manner and this element is one of the most expensive. The phenomenon of P adsorption directly influences the magnitude of available P for crops in highly weathered soils.

The intent and objectivesof the study presented herein were to

elucidate (a) sources of acidity, (b) ion selectivity, (c) adsorption

I




2




and release of P, and (d) effects of CaC03 and P treatments on surface charge (CEC) as related to crop production in selected soils from Venezuela.















LITERATURE REVIEW



Evaluation of soils in relation to classification, agronomic practices, and other purposes is aided by the identification, characterization, and understanding of their properties and behaviors. Soil surface reactions involving source of protons, ionic selectivity, specific adsorption, and responses to chemical treatments have great importance in the understanding of electrochemical behaviors of highly weathered soils.


Soil Characteristics

Mineralogical Characteristics

Physical and chemical properties of any soil are controlled to a very large degree by the minerals in the soil and especially by those constituting the clay fraction. Identification, characterization, and an understanding of properties of the different minerals aid in the evaluation of soils in relation to classification, agronomic practices, and other purposes. Positive identification of mineral species and quantitative estimation of their properties in the soil clay fraction usually require the application of several qualitative and quantitative analyses. X-rays diffraction, CEC, and surface area are among the most common analytical measurements used by soil scientists (Jackson, 1956; Whittig, 1965).

Some soil constituents interfere in the soil mineralogical analysis. In these cases, a chemical pretreatment of the soil is necesary.

3








Pretreatments are designed to have a minimum effect on fractions of the soil other than those for which the treatments are intended. The pretreatments that are most used in soil mineralogical analyses are removal of salts, carbonates, Fe, organic matter and amorphous material, and particle-size fractionation (Jackson, 1956).

Soluble salts and carbonates affect organic matter removal, soil

sample dispersion, and also make it impossible to saturate the exchange complex with a specific cation. Removal of soluble salts and carbonates simplifies X-ray diffraction analysis. Carbonates reduce degree of orientation of particles in slide preparation (Jackson, 1956).

Organic and Fe-oxide cementing agents are frequently removed by

chemical treatment prior to clay separation to obtain dispersion for effective particle-size separation, enhanced parallel orientation of clays, and decreased background and attenuation of X-rays (Zelazny and Carlisle,

1971).

Mehra and Jackson (1960) found that citrate system was the most effective in removal of free Fe-oxides from latosolic soilsand the least destructive of Fe-silicate clays as indicated by least loss in CEC after Fe-oxide removal.

On the other hand, organic matter may cause a high background and prevent parallel orientation in the preparation of the slides for X-ray diffraction studies. Traditionally, organic matter has been oxidized with H202. In order to prevent a strong acid condition from developing which could degrade crystalline minerals, soils have to be buffered around pH 5 in the presence of H202.

Mineralogical analyses are normally performed on specificsize fractions of the soil such as 2 to 0.2 microns. For mineralogical study, concentration of the clay fraction is necessary (Jackson, 1956).




5




Identification and quantitative analysis of the minerals are possible from an X-ray diffraction pattern because each diffraction peak depends upon the atomic arrangement in the crystal and is proportional to the concentration of the mineral in a mixture (Jackson, 1956). The choice of cations for saturation of the exchange sites along with solvation becomes important especially since spacings of expanding clay minerals are made larger with Mg and Ca ions than with K or Na ions (Jackson, 1956).

Soil Acidity Sources

In most of the soil classification studies and in soil characterization for agricultural purposes, the pH value has great importance. Deficiencies of P, Ca, Mg and Mo, and Al and Mn toxicities seem to be the most important effects of soil acidity on plant growth (Lora and Riveros, 1971).

Soil acidity has been defined as a soil system's proton yielding

capacity in going from a given state to a reference state. Soil acidity has long been recognized to involve exchangeable Al (Jackson, 1963). Acid soil systems that are Al-saturated have weak acid characteristics while those that are H-saturated have strong acid characteristcs (Coulter, 1969). But the latter is an unstable state because H ions attack the clay structure and Al is released as shown by Davis et al. (1962) in their study of autotransformation of H-bentonite to Albentonite.

On the other hand, Yuan (1963) studied H and Al ions, and pH

relationships in pure chemical solutions and compared the results with those obtained from several soil systems. He pointed out that pH determined with the usual methods can be considered as a measure of the H ion




6




activity in the system when the liquid and solid phases are in equilibrium. He also indicated that soils with the same H ion activity and different amounts of unhydrolyzed Al ions may have the same pH.

Since Bradfield (1941) pointed out the nature of potentiometric and conductometric titrations in soil systems, these techniques have been very useful in studying the nature of soil acidity. Several authors have studied the factors that affect the results obtained with these techniques. Low (1955) pointed out that potentiometric and conductometric titrations could be used to determine the amount of Al on an acid bentonite. On the other hand, Marshall (1964) showed initial inflections due to H titration, followed by regions of buffering attributable to Al. These inflections and their corresponding indication of soil acidity sources can be verified by bases that titrate specifically H or Al. Yuan (1959) used fluoride to eliminate the interference of Al ions in the H titration.

It was found that Na2BO7titrations gave the permanent negative

charge on monrtmorillonite particles both when this clay contained exchangeable Al and when it did not (Shainberg and Dawson, 1967). They also pointed out that the differences between NaOH and Na2B407 titrations can be made equivalent to H ions formed from proton release by hydroxyl groups, and by selecting an appropriate end point, it can be made approximately equal to this quantity plus the aluminate equivalent of exchangeable Al. They also determined the effect of salt content on the slope of the first segment of the conductometric curves.

Chao and Harward (1962) found that the features of titration curves were independent of the kind of clay and dependent on the nature of the acid clay. Rich (1970) pointed out that the differences in results of




7




both potentiometric and conductometric titrations, and of extractions by neutral salts of acidity from AICl3- treated cation exchangers were due to the fact that neutral salts, especially K salts, tend to induce hydrolysis of Al+3 ions on exchange sites. Dewan and Rich (1970) found that consistent results are obtained when the aluminate reaction is considered for titrations with strong bases.

Lime Requirement

The several methods that have been proposed to determine lime requirement of acid soils have advantages and disadvantages depending on type of soil and intended use of the data. However, it is now generally accepted that the most reliable index for predicting lime requirement of soils of tropical regions is the quantity of exchangeable Al (Reeve and Sumner, 1970a; Kamprath, 1970; Evans and Kamprath, 1970; Amedee and Peech, 1976). However, the incubation method still appears to be as good as any, if the desired pH level is not as high as complete neturalization. Spain et al. (1975) pointed out that liming rates for soils of the tropics often have been overestimated even when based on Al suppression. Pearson (1975) concluded that relatively low rates of lime in the humid tropics are usually adequate for maximum crop production. Potassium Quantity/Intensity Measures

Because of the importance of K in plant nutrition, many investigators have attempted to find a rapid and reliable method for evaluation of K availability in soils. Potassium availability depends on the exchangeable form of this element in the double layer surrounding the exchange complex. The chemical potential of K in the double layer controls the activity of K in solution (Beckett, 1964a). Its chemical potential cannot be measured experimentally and is not directly




8




proportional to the activity of K in the soil solution (San Valentin et al., 1972). Beckett (1964b) proposed the activity ratio aK/(aCa + Mg)2 as a measure of the ruling chemical potential of labile K in a soil. The activity ratio, which is now stated as ARk, has the same value for all solutions in equilibrium with a given soil, and is independent of the ionic strength of the soil solution or the proportion of Ca to Mg (Beckett and Nafady, 1967b). This ratio is related to availability of K and is termed intensity (1) (Beckett, 1964b). The ability of a soil to maintain the activity ratio against depletion by plant roots is governed by the character of the pool of labile K, the rate of release of fixed K, and the diffusion and transport of K ion in the solution (Beckett, 1964c). The relationship between K availability or intensity (1) and the amount of K present (Q) in soils is termed the Q/I relationship (Beckett, 1964a).

The Q/I relationship can be determined by equilibrating a sample of soil with solutions containing a constant amount of CaCl2 and increasing amounts of KCI. The soil either gains or losses K in order to achieve the characteristic ARk of that soil, or remains unchanged if its ARk is the same as that of the equilibrating solution. The characteristic Q/I relationship is formed by plotting the gain or losses of K (AK) against the ARk of the equilibrium solution (San Valentin et al., 1972; Beckett and Nafady, 1967a, 1967b). San Valentin et al. (1972) pointed out that three parameters are derived from a Q/I plot which can characterize the soil K status. When AK=O, the ARk is a measure of the K availability at equilibrium (termed ARk). When AR k=0, the value of K
e
is a measure of the labile or exchangeable K in soils (termed AK*). The slope of the linear portion of the curve gives the Potential Buffering Capacity of the soil for K (termed PBCk ) and is proportional to CEC.









The curved portion of the Q/I plot is also an indication of the number of specific sites for K. The number of specific sites for K (termed K ) is
x
determined from the difference between the intercept of the curved and linear portions of Q/I plot at AR =0.

Forms of the Q/1 relationship remain almost unchanged either by large K additions, a high degree of K fixation, or by depletion of labile and some fixed soil K. However, acidic M CaCl2 solutions may influence the form of Q/I relationship by removing substances blocking portions of the exchange surfaces of the soil clay, increasing the extent of the exchange surface participating in the Q/I exchange equilibria, inducing the collapse of the expanded portion of soil clay, and by decreasing the extent of those exchange surfaces (Lee, 1973).

Moss and Beckett (1971) reviewed the sources of error in the determination of soil K activity ratios by the Q/I procedure. They pointed out that errors may arise in sampling, preparation and storage of soil samples, and recommended analysis of samples as soon as possible after sampling, with only gentle sieving. They also found that exchange between cations and soluble salts or microbially dissolved carbonates during the determination affect the results obtained by this procedure. Phosphorus Fractions

Knowledge of specific chemical forms of phosphates is important in understanding the chemistry of soil P and also has importance in soil genesis and soil fertility. In acid soils, P forms insoluble compounds with Fe and Al while under alkaline conditions, Ca compounds are formed. The particular compounds formed depend on soil properties. Furthermore, the solubility and stability of these compounds affect the availability of P under field conditions.




10




Juo and Ellis (1968) showed that when soluble P is applied to acid

soils, or when Ca-P is dissolved during the process of chemical weathering, the soluble P precipitates rapidly to form colloidal AI-P and Fe-P. The P in these colloidal forms is relatively available to plants initially, but crystallization tends to form hydrated compounds such as variscite and strengite that are less available. Furthermore, Fe-P crystallizes at a more rapid rate than AI-P and this is why AI-P seems to be more available to plants than Fe-P. Singh et al. (1966) studying the availability of forms of residual P in Davidson clay loam found that neutral soils, pH 6.8, favored the availability of AI-P to first-cutting alfalfa and Fe-P to first-, second-, and third-cutting alfalfa. They also pointed out that Al and Fe activities at this pH are low, and as a result, P from Al- and Fe-P becomes more available to plants. Lindsay and Moreno (1960) pointed out that in soil below pH 7, the H2PO4 constitutes the larger fraction of the total P in solution and the logarithmic acitivity function permits a linear relationship with pH. They developed a phase diagram that permits prediction of the H2P04 concentration as controlled by variscite, strengite, fluorapatite, hydroxyapatite, octocalcium phosphate, and dicalcium phosphate solubilities. In that diagram, increase in pH tends to decrease precipitation of Al and Fe compounds, but increase precipitation of Ca forms.

Yuan et al. (1960) studying newly. fixed P forms in acid sandy soils found that the ratio AI-P to Fe-P increased with increasing rates of applied P. They also found that increasing soil drying temperature decreased the percentage of P in AI-P form, but increased that in Fe-P form. Fractionation of inorganic soil phosphates is widely used in soil fertility and genesis studies. These procedures are based on




11




differential solubilities of inorganic phosphates in various extractants. The most used of these methods is that of Chang and Jackson (1957) modified by Petersen and Corey (1969); these procedures differ only in the sequence of extraction of the different P forms, pH of the NH4F solution used to extract Al-P which reduces interference by.Fe, and other details to increase the efficiency of analysis ofalarge number of samples. Phosphorus Adsorption

Adsorption is the process of concentration of liquid or gaseous

material on the surface of a solid. Phosphorus adsorption studies may be divided into those concerned with investigating the nature and mechanism of the sorption system, and those concerned with its quantitative measurement (Larsen, 1967).

Measurement of the size of the P adsorption system can be made by fitting adsorption data to a previously described adsorption isotherm. The Langmuir isotherm has often been used for this purpose.

The Langmuir isotherm has constants which, at least when applied to the adsorption of gases on solids, have quantitative meaning. The linear form of the Langmuir isotherm is c 1 c (1) x/m Kb b '

where x
where = mg P adsorbed/100g soil
m

b = the adsorption maximum, mg/lOOg soil

c = equilibrium P concentration, moles/liter

k = a constant related to the energy of bonding. The Freundlich isotherm has the form x = ac (2)





12




or the linear form

log x = log a + b log c, (3) where x is the amount of P adsorbed per unit weight of soil, c is the concentration of P in solution, and a and b are the constants that vary between soils. Larsen (1967) claimed that this isotherm is purely empirical, and the constants have no physical meaning. However, Adamson (1976) pointed out that the Freundlich equation, unlike the Langmuir, does not become linear at low P concentration but remains convex to the concentration axis. Also, it does not show a saturation or limiting value, but the intercept of log x vs log c gives a measure of the adsorbent capacity and the slope of adsorption intensity.

Factors affecting adsorption equilibrium. Equilibrium adsorption of P studies on a variety of soils and minerals have shown that several factors have consistent influence on the final position of the equilibrium state. Among these factors are (a) concentration of P in the solution, (b) temperature, (c) ionic strength, (d) the pH, and (e) specific cation interactions with the surface and adsorbed species. Noting the apparent independenceof the adsorption isotherm from the overall adsorption capacity of the adsorbent material, Muljadi et al. (1966) proposed employing an empirical splitting of the isotherms into three regions. The regions, designated I, II, and III, were found to correspond to equilibrium solution concentration ranges which may be described roughly as low, medium, and high, respectively. Regions I and II fit a Langmuir adsorption isotherm while region III exhibited a linear dependence on solution concentration. They also found that adsorption in regions I and II increased irreversibly with increasing temperature, while the increase in region III was reversible with respect to temperature. Ryden





13




et al. (1977) later employed this curve splitting technique and reported on the effects of pH, ionic strength, specific cation, and cation adsorption. These authors noted that Ca electrolyte solutions were more effective in facilitating adsorption than Na solutions of the same ionic strength; in conjunction with this finding, the authors reported observing a specific adsorption of Ca which was not reversible to IN KCI. Sodium adsorption was measured together with phosphate adsorption and it was observed that the adsorption of Na corresponded to that of phosphate in regions I and III, but that no Na was adsorbed in region II. They interpreted those results in terms of the adsorption mechanisms in the three regions. Region I was thought to represent the replacement of adsorbed water by phosphate; region II, the displacement of surface hydroxyls; and region III, the association of phosphate ligand with the surface metal which was potentially dependent and less energetic than a specific electrovalent coordination with the metal. The concentration of phosphate in solution was thought to affect both the amount of P adsorbed and the nature of the adsorption process.

Temperature effects on equilibrium adsorption isotherms have been reported for soils and minerals. Muljadi et al. (1966) measured phosphate adsorption on K-kaolinite, gibbsite, and pseudoboehmite at 2, 20, and 400C, respectively. They reported a marked increase in adsorption with temperature for the three equilibrium concentration regions. Region I of K-kaolinite (pH 5) reached a maximum at about 200C; regions II and III showed increases to 400C for all adsorbents. The increase in adsorption in region I was reported to be irreversible with respect










to temperature; region II was slightly reversible for K-kaolinite but irreversible for gibbsite and pseudoboehmite; and region III was found to be completely reversible. Barrow and Shaw (1975a) measured the adsorption of phosphate on soils as a function of time and temperature. They observed an increase in adsorption with temperature from 4 to 420C for contact times up to 100 days. However, they reported that adsorption exhibited a time-temperature dependence which indicated that increasing temperature merely speeded up the adsorption process, rather than shifting the equilibrium position.

Griffin and Jurinak (1974), studying the kinetics of interaction of phosphate with calcite at 0, 11, 23, and 40'C, demonstrated that temperature has a much more pronounced effect on the adsorption rate constant than on the desorption rate constant. These results'indicated that the adsorption process is favored by an increase in temperature and that equilibrium adsorption should increase over the temperature range studied.

The effects of the ionic strength and of selected cations on the adsorption of phosphate have been reported for a group of Hawaiian soils (Ryden et al., 1977a, b) and gibbsite (Helyar et al., 1976a). Ryden and coworkers, observing a linear relationship in plots of the solution phosphate concentration against the reciprocal of time for periods greater than 72 hours, were able to estimate the equilibrium concentration for a series of solutions of differing ionic strengths. The plots converged at equilibrium for low phosphate concentrations, indicating that the ionic strength effect was kinetic. At moderate to high solution concentrations, the ionic-strength effect was found to be absolute and the authors concluded that the adsorption at high





15




solution concentrations was potentially determined. Adsorption from solutions with a CaCl2- supporting electrolyte was greater than from NaCIl solutions; however, the pH of the Ca solutions was lower. The authors, noting that more Ca was adsorbed than Na and that the adsorbed Ca was not completely displaced by IN KCI, proposed that Ca was specifically adsorbed, thus facilitating the adsorption of phosphate. Helyar et al. (1976a) studied the effect of selected cations on the adsorption of phosphate on gibbsite. They maintained a constant pH by controlling the partial pressure of C02 in the solid-solution mixture. Their results for Na, K, and Mg were similar and the observed ionic strength effect was small. Adsorption in the presence of Ca, however, increased out of all proportion to the change in ionic strength for equilibrium concentrations above 1.0 pM, but converged with the adsorption noted for other cations at lower equilibrium concentrations.

Helyar et al. (1976b) proposed that certain cations could facilitate adsorption by forming a stabilizing bridge between two adsorbed phosphate ligands. The ability of the cation to form such a bridge would depend upon its fit into the space between the oxygen of two adjacent phosphates and its ability to coordinate with two oxy-ligands. The authors estimated an oxygen to oxygen distance of approximately

1.0 R for adsorption on gibbsite; cations of approximately this crystal radius and divalent charge would best accommodate the space if they could form coordinate covalent bonds. They selected a number of cations with a range of radii for both mono and divalent species. The adsorption of phosphate was enhanced markedly by Ca, Cd, and Sr which have radii of 0.99, 0.99, and 1.13 2, respectively. No such enhancement was noted for Mg, Zn, Na, or K which have radii of 0.66, 0.74, 0.95, and 1.37 R,




16




respectively. The stabilizing effect of a bridging cation was thought to result from a reduction of the mutual repulsion of the oxygen or hydroxyl groups on adjacent phosphate ligands.

Thermodynamic parameters associated with adsorption process. Thermodynamic parameters calculated from adsorption measurements have been found useful in elucidating the mechanisms involved (Biggar and Cheung, 1973). Biggar and Cheung (1973) calculated the thermodynamic parameters associated with the adsorption process from the variation of the thermodynamic equilibrium constant, Ko, with changes in temperature. Ko for the adsorption reaction is defined as:

Ko =as ys Cs (4) ae ye Ce

where as = activity of the adsorbed solute, ae = activity of the solute in the equilibrium solution, Cs = g of solute adsorbed/ml of solventin contact with the adsorbent surface, Ce = pg of solute/ml of solvent in the equilibrium solution, ys = activity coefficient of the adsorbed solute, and ye = activity coefficient of the solute in the equilibrium solution.

Cs is calculated according to the following equation:


Cs = (pl/Mi) Al (5) S A2
Na (x/m) M2 x 106

where pl = density of solvent (g/ml); MI and M2 = molecular weights (g/mole) of the solvent and the solute, respectively; A1 and A2 = cross-sectional areas (cm2/molecule) of the solvent molecule and the solute molecule, respectively; Na = Avogadro's number (6.02 x 1023 molecules/mole); S = surface area of the adsorbent (cm2/g); and x/m = specific adsorption (Hg/g).




17




The cross-sectional areas (in cm2) of the solvent and solute molecules were estimated from the following equation: A = 1.091 x 10 -1 FMx 024 2/3 (6) Nap

where M and p are, respectively, the molecular weight (g/mole) and the density (g/ml) of the solvent or solute. Biggar and Cheung (1973) assumed that

S A2 (7)
Na (x/m) > M2 x 106 and reduced equation (5) to Cs (pl/Mi) x Ai (8)
S
Na (x/m)

and calculated Cs from this equation. They also assumed that concentration of the solute in the solution approaches zero, and ys and ye approach unity. The authors obtained the values of Ko by plotting In (Cs/Ce) vs Cs and extrapolating to zero Cs. The standard free energy, AG*, was calculated from

AGO = - RT In Ko. (9) The standard enthalpy, HO, is obtained from the integrated form of the van't Hoff equation, In (AH ) (10) Kol R T2 TI The standard entropy, ASo, is obtained from ASO = (AHo - AGO)/T. (11)




18





Biggar and Cheung (1973), noting the linear nature of In Ko vs 1/T, concluded that the mechanics of adsorption were not changed as temperature was changed and that the amount of adsorption was changed because the supply of thermal energy was different.

Rydens et al. (1977 ) used the expression:

G =- RT InK (12) where K is the Langmuir equilibrium constant obtained for a single temperature and has units of reciprocal concentration. Assuming that K exhibits a temperature dependency, the value of the estimated AG will depend upon the measurement temperature of K, and the concentration units used.

Some parameters related to the adsorbent capacity and the intensity of adsorption could be calculated using the Freundlich equation. In a plot of In X vs log c where X = P adsorbed (pg/g), and c the equilibrium concentration (moles/liter), the intercept gives InK that can be used to determine a AG related to the adsorption capacity energy (Adamson, 1976).

Using Ryden et al. (1977a, b) plots of the solution phosphate concentration vs the reciprocal of time to determine the equilibrium concentration byextrapolating at infinite time, an equilibrium constant can be calculated:

[Soil-P] (13)
Keq = [P soil] (13) [P04][Soil]

where [soil-P] is g of P adsorbed per g of soil,[P04] is equilibrium concentration in ig/ml, and [soil] is g of soil/liter of solution. Applying (12), a AG for adsorption can be calculated.





19




Surface Charge

Double-layer Models

Soil electrochemical behavior is determined by the quantity, type, and interactions of colloidal components. Colloids with a constant surface potential have a reversible double layer with the surface charge determined by the nature of the potential-determining ions (PDI) adsorbed on the surface. The electrical double layer at the solid-solution interface is constituted by the surface charge and its countercharge in solution. Solid oxide particles immersed in aqueous electrolyte solution develop surface electrical charges by adsorption or desorption of potential-determining ions (Atkinson et al., 1967). For oxides, the surface charge and surface potential are functions of solution pH, and H+ and OH are referred to as potential-determining ions (Breeuwsma, 1973). The electrochemical behaviors of Oxisols and Alfisols were found to be similar to those exhibited by many metallic oxides (Van Raij and Peech, 1972).

The constant-potential surfaces are associated with oxides and

hydroxides of metals and edges of clay minerals as well as with those clay minerals with minimal ion substitution in the lattice (Tinsley, 1974). These colloids include crystalline and noncrystalline oxides and hydrous oxides of Al, Fe, Ti, Mn, and Si; kaolinite, halloysite; allophane; quartz; and organic matter. The surface potential remains constant, and its magnitude is not affected by the presence of indifferent electrolytes (Stumm and Morgan, 1970).

For constant-potential surfaces, the surface charge is directly

related to the concentration of potential-determining ions in the solution. The surface charge density of minerals of this type varies with pH and salt concentration (Keng and Uehara, 1974). From known surface





20




areas and Point of Zero Net Charge (PZNC) determination of Oxisols, the values for net electrical charge calculated by the application of Gouy-Chapman and Stern models of the double-layer theory were found to be in good agreement with experimental results (Van Raij and Peech, 1972; Adams, 1976).

Gouy-Chapman double-layer model. The Gouy-Chapman model relates the charge density of the surface with the surface potential as follows:

a = (2nEKT/w) Sinh (Zeto/2KT) (14) where a = the surface charge density

n = the concentration of the equilibrium solutions in

number of ions per cm3

e = the dielectric constant of the medium

e = the charge of an electron

K = the Boltzmannconstant

T = the absolute temperature

Z = the valence of the ion to = the surface potential.

In constant-potential systems, change in concentration will

result in a change in surface-charge density (a ). For some constantsurface potential colloids, to can be held constant by maintaining the pH constant (Keng and Uehara, 1974).

The Nernst equation can be used to relate the H ion concentration to the surface potential as follows: RT H+
o = In (15) ZF H+




21





where R = the gas constant

F = the Faraday constant

H+ = the hydrogen ion concentration

H+ = the hydrogen ion concentration at which *o=o.
o
At 25*C the Nernst equation can be reduced to the following: o = 59(PZNC-pH) (16) where PZNC is the pH at the point of zero net charge (pHo).

The Gouy-Chapman model for the double layer has limited quantitative application because of the assumption that ions in solution behave as point charges and can approach the surface without limitation (Tinsley, 1974).

Stern double-layer model. In the Stern double-layer model, charge density can be described by the following equation (Van Raij and Peech, 1972):

0 = 01 + 02

where 01 = Stern layer charge

02 = diffuse layer charge.

The Stern layer charge can be given by


NIZe (17)
01 = (17) I + (Nap/Mn) exp 2eKT


where NI = available spots per cm2 for adsorption of ions

Na = Avogadro's number

M = molecular weight of the solvent

p = solvent density

96 = electrical potential at the boundary between the Stern

layer and the diffuse layer or the Stern potential





22




S = specific adsorption potential.

The diffuse-layer charge is given by the following equation: o2 = (2nEKT/u)1 Sinh (Ze 6/2KT) (18) Since a linear drop in potential across the specific Stern layer is assumed, the surface charge can also be given by the Gauss equation for a molecular condensor:


6 = (fo-p6) (19) where e1 = average dielectric constant

6 = average thickness of double layer.

These equations can be used to calculate the relative distribution

between oa and a2 if values of N1, 1 , and 6 can be assumed (Adams, 1976). Van Raij and Peech (1972) claimed that the only soil properties needed to calculate the net charge are PZNC and surface area. Point of Zero Net Charge

Concentration of potential-determining ions and the net surface charge are obviously pH dependent on constant-potential surfaces. There will be a pH at which the densities of positive and negative charges are equal and the surface possesses no net charge (Tinsley, 1974; Keng and Uehara, 1974).

The acidic and basic properties of a solid that influence the location of PZNC are functions of such variables as cation size and valence, the hydration state of the solid, and the geometrical arrangement of the ions. Parks (1965) pointed out that the broad range of isoelectric points in oxides and hydroxides was explained by differences in hydration state, purity, and cation radii.

The PZNC of some pure minerals found in soils can be observed in

Table 1 (Adams, 1976). It can be seen that dehydration, dehydroxylation,





23




and increased crystallinity result in an increase in the PZNC. However, the presence of anionic impurities of pH-dependent, potential-determining ions will decrease the PZNC. Also, structures or compounds which are ion exchangers and have an intrinsic structural charge can be expected to shift the PZNC away from the calculated value by a variable amount. A
+
positive charge density, a , may be expected to move the PZNC in the basic direction, while negative a should move the PZNC in the acid direction (Parks, 1967).

In Fe and Al oxide systems, the Fe and Al ions are not exchangeable, but can be potential-determining ions. In general, the presence of Fe and Al oxides will tend to increase the PZNC of soil toward higher pH values, while the presence of clay minerals with permanent or structural negative charges tends to shift the PZNC of soil to lower pH values (Van Raij and Peech, 1972).

Effects of Specific Adsorption on the PZNC

When specific adsorption of cations or anions occurs at the PZNC,

the pH shifts toward higher and lower pH values, respectively (Breeuwsma, 1973). An excess of a specifically adsorbed ionic species will remove pH dependence or change the PZNC to that of the impure species (Parks, 1967). Specific adsorption of ions other than the potential-determining species H+ and OH- is related to differences in PZNC between soils and within profiles (Adams, 1976).

Specific adsorption of anions produces a negative surface under otherwise identical conditions. Anions, such as P04-3, which are dehydrated and specifically adsorbed by ferric oxides are considered to form a new potential-determining layer by ligand-exchange reactions




















Table 1. PZNC of pure minerals found in soils.

Minerals PZNC Authors

Si02 2.0 Parks and de Bruyn (1962) Montmorillonite 2.5 Stumm and Morgan (1970) Kaolinite 4.6 Stumm and Morgan (1970) Gibbsite 5.0 Stumm and Morgan (1970) a-AI(OH)3 5.0 Stumm and Morgan (1970) Magnetite 6.5 Stumm and Morgan (1970) a-Fe203 6.7 Stumm and Morgan (1970) Goethite 7.55 Atkinson et al. (1967) Fe(OH)3 amorphous 8.5 Breeuwsma (1973) Hematite 8.45-9.27 Atkinson et al. (1967) a-Al203 9.1 Stumm and Morgan (1970) MgO 12.4 Stumm and Morgan (1970)




25




with OH- or structural H20 in the ferric first coordination shell (Atkinson et al., 1967). When P or other anionic chemicals are added to a soil with variable-charge colloids, the adsorbed ("fixed") anions shift the PZNC to lower pH values (Adams, 1976).

Lowering of PZNC in goethite (Hingston et al., 1968) and in tropical soils (Van Raij and Peech, 1972) as a consequence of specific sorption of anions has been clearly demonstrated. Cation Exchange Capacity

Effect of pH. Change in CEC with change in soil pH has been logically termed pH-dependent CEC. Soil fractions contributing to the degree of pH-dependent CEC were reported to be organic matter and clays which either contain noncrystalline components or have expanding plyllosilicates with hydroxy-Al interlayers (Fiskell and Zelazny, 1971). The cations are held in these systems because of negative charges resulting from ionization of the OH groups attached to Si and Al of broken tetrahedral and octahedral positions and functional groups of organic matter (Wiklander, 1967).

Fiskell and Zelazny (1971) studied the effects of selected soils buffered at several increasing pH values on CEC. The increase in CEC with soil pH was attributed predominantly to the increase in pHdependent charges on organic matter. Likewise, Fiskell et al. (1964) studied the changes in CEC with change in pH in a Leon fine sand. They found an increase in the CEC of this soil almost linearly with increasing pH. Helling et al. (1964) found that the average increase in CEC of the organic fraction of 60 Wisconsin surface soils was correlated with soil organic matter.





26




Effect of P adsorption. The CEC of soils is known to increase on treatment with phosphate (Sawhney, 1974). The increase in CEC has been attributed to the replacement of hydroxyl ions by phosphate ions at the surfacesof clay minerals and sesquioxides. Recently, it has been shown that sorption of anions on sesquioxides with surface at constant potential shifts the PZNC to lower pH (Breeuwsma, 1973), thereby, requiring more cations to balance the additional negative charge producedby specific adsorption of anions. Similarly, phosphate sorption caused large increases in the CEC of tropical soils rich in sesquioxides (Mekaru and Uehara, 1972).

Specific sorption of anions on sesquioxides involves not only the replacement of hydroxyl ions but also of water molecules in octahedral coordination on oxide surfaces (Sawhney, 1974). Thus,soil pH is reduced more where a portion of the surface hydroxyls are protonated to form water molecules,and replacement of the water molecules by an anion increases the CEC. Sawhney (1974) found that as the sorbed phosphate increased, the CEC also increased. He attributed the increase in CEC to the replacement of octahedrally coordinated water molecules on sesquioxides surfaces at pH 5, while at higher pH values, increased CEC was caused by replacement of hydroxyl ions.


Crop Response to Lime and P Applications

It has been pointed out (Blue, 1974) that one of the most important considerations in the nutrition of crops in highly weathered Ultisols and Oxisols is the prevalence of P deficiency. Blue (1974) also stated that these soils have relatively high P retention capacities and, therefore, the efficient use of this element is important. The effects of P





27




and lime on crops have received wide attention. Recently, phosphate adsorption isotherms have been used as a criterion forP applications to tropical soils with high P adsorption capacities.

Fox and Kamprath (1970) found that the effects of large phosphate applications to soils with high phosphate sorption capacities were evident in phosphate sorption isotherms 10 years later. They also used the sorption curves as a basis for fertilizing soils in pot experiments. Reeve and Sumner (1970b) concluded that the wide spectrum of P-fixing abilities of soils indicates that fertilizer recommendations based on direct determination of the amount of P required to raise the status of a soil to a desirable level are likely to be more accurate than those based on an estimate of the amount of available P in the soil per se.

On the other hand, it has been found that lime reduces the adsorption maximum of P in Ultisols by reducing the exchangeable Al (Woodruff and Kamprath, cited by Blue, 1974). However, Amarasiri and Olsen (1973) found that a limed soil had a higher maximum adsorption capacity for P than unlimed soil.

Evans and Kamprath (1970) found that lime increased the growth of corn on mineral soils when Al saturation was greater than 70% and soybean when Al saturation was greater than 30%. Reeve and Sumner (1970a) found that the yield of 'trudan' (Sorghum sudanense) increased with increasing lime and P application rates. They concluded that the increase was due to the reduction of Al toxicity and improvement of P availability.
















MATERIALS AND METHODS



Ten soils from Venezuela were used in this study including two Oxisols, a Grossarenic Psammentic Haplustox (Guanipa 5, Anzoategui State), and a Tropeptic Haplustox (Guataparo, Carabobo State); an Ultisol, Typic Paleustult (Guanipa i, Anzoategui State); three Alfisols, an Ultic Tropudalf (San Crist6bal, Tachira State), an Oxic Haplustalf (Barinas, Barinas State), and an Oxic Tropudalf (Bajo Seco, Miranda State); an Entisol, Ustic Quartzipsamment (El Rocio), an Inceptisol, Fluaquentic Humitropept (Alambre), and a Vertisol, Udorthentic Pellustert (Guanaguanare) from Portuguesa State; and a Aridisol, Haplargid (El Potrero, Lara State). The samples were collected from uncultivated sites (Figs. I and 2). The surface 30 cm of soil was mixed (in all samples) except Guanipa 5 for which the All (0 to 10 cm) and A21 (10 to 30 cm) horizons were analyzed as separate samples. These soils had a pH range of 3.50 to

5.50, organic matter content of 0.50 to 4.50%, and clay content of 7 to

51%.

To compare soil behavior, Red Bay, a Rhodic Paleudult (horizons Al and B2t) from Florida, U.S.A., was also used in some analyses. This soil was formed under subtropical conditions and belonged to a highly weathered order.

All samples were air-dried and ground to pass a 2-mm sieve for laboratory and greenhouse experiments.

28


































VENEZUELA




COLOMBIA BRAZIL Fig. i. Relative location of Venezuela in South America.













LARA(


0 0O GUES
0 ~ANZOATEGUI
BARINA
TA HIRP










OSAMPLIIG SITES












Fig. 2. Relative location of the soil sampling sites in Venezuela.





31




Soil Physical, Chemical, and Mineralogical Properties

In order to characterize these soils, several physical, chemical, and mineralogical analyses were performed.

Except for Red Bay, particle-size distribution of the soil samples was determined by the pipette method. Moisture equivalent was determined by saturating with water and centrifuging at 1,000 rpm (Beaver et al., 1972). Organic carbon was measured by wet digestion with K2Cr2O7H2SO4 (Walkley, 1947).

To prevent interference with X-ray analysis, samples were treated with 30% H202 to eliminate organic matter. Free Fe oxides were removed by Na-dithionite-citrate-bicarbonate (DCB). After these cementing agents were removed, samples were adjusted to pH 10 with Na2CO3 and gently heated to remove any excess of H202. The sand was separated from silt and clay by wet sieving with a 300-mesh sieve using H20 adjusted to pH 10 with Na2CO3. Clay was separated from silt by repeated decantation after centrifuging at 1,000 rpm for 2 minutes in an International No. 2 centrifuge. The residue was considered to be the silt fraction (Zelazny and Qureshi, 1971). X-ray Diffraction Analysis

X-ray diffraction analysis criteria were used to identify the crystalline components of the inorganic colloidal separates. This was performed by orienting approximately 200-mg samples of the appropriate clay suspension on several glass slides and leaving them until air dry. One sample was saturated with Mg, one was Mg-saturated glycerol solvated, and the rest were saturated with K, nonheated, and heated to 100 and 5000C for 4 hours. Samples so prepared were analyzed with a Phillips PW1120-96 X-ray diffractometer with Co radiation and scanning from 2 to 500 2e.




32




Fe-oxides Determination

Since Al- and Fe-oxides contribute to P adsorption and to a significant portion of the constant-potential surfaces in many tropical soils, these were determined with Na-DCB. The Fe and Al in the extracts were analyzed by atomic adsorption spectrophotometry in a Perkin-Elmer 303 Spectrophotometer.

Surface Area Determination

Samples for surface-area determination were prepared by saturating the soil with IN MgCl2. Excess salt was removed with water and the samples were dried at 1050C. The clay was separated by centrifugation and ground to pass a 60-mesh sieve and mixed again with other components (Adams, 1976). Labeled Al dishes were placed in a desiccator over P205 under vacuum for 6 hours to obtain constant weight. Approximately l-g samples of soil were then placed in the dishes and brought to constant weight as before. A soil-absorbate slurry was prepared by placing 3 ml of ethylene glycol monomethyl ether (EGME) into the dried sample. The slurry was mixed and placed in a desiccator containing a CaCI2-EGME solvated slurry to equilibrate for I hour before applying vacuum (Carter et al., 1965). The desiccator was then evacuated at 0.25 mm Hg for I hour. Periodic weights were taken on the soil-adsorbate slurry until constant weight was obtained. The surface area was calculated using the following equation:


Wg (22)
Ws x 0.00286

where A = specific surface in m2/g

Wg = weight of EGME retained by the sample after

equilibration in g




33




Ws = weight of P205 dried sample in g.

The constant was obtained from the assumption that 2.86 x 10 g EGME is required to form a mono-layer on each m2 of sample (Carter et al., 1965). Soil Acidity

Soil acidity was determined by potentiometric and conductometric titrations using the following procedure: 5-g soil samples were placed in 200-ml plastic bottles and 50-ml portions of distilled water containing increasing amounts of 0.05N Ba(OH)2, O.IN NaOH, or 0.1N Na2B407 were added to the bottles. The pH and electrical conductivity were measured after continuous shaking for 48 hours. Extractable acidity was determined by leaching soil samples with IN KCI. A 10-g sample was placed in a 125-ml Erlenmeyer flask to which 100 ml of IN KCI were added. The flask was stoppered and shaken periodically for 2 days. Then, the mixture was filtered through Whatman No. 42 filter paper into a 250-mi volumetric flask and washed with three separate 50-ml portions of IN KCI. One aliquot of 100 ml was titrated with 0.0458N Ba(OH)2 to pH 8.0. A 50-ml aliquot was analyzed for Al by atomic adsorption. Intensity measurements were also done by reading the pH in H20, IN KCI, IN K2S04, 0.02M CaCl2-, 4% NaF at 1:2 soil:solvent suspensions. Differences between salt-pH and H20-pH were calculated as ApH values in which pH (H20) was substracted from the pH (salt). All measurements were made using a Beckman pH meter with a combination electrode.

Components of acidity were estimated from the graphs resulting from plotting pH and EC vs meq of base added. Readings were made at inflection points at pH 8 and 10. In conductometric titration curves, inflections were determined at the intersection of two straight lines drawn along the linear portions at the extremities of the curves. End-point





34




values were multiplied by 0.75 to account for the neutralization factor associated with the formation of aluminate in the Ba(OH)2- treated samples.

Potassium Q/ Characteristics

To determine the Q/I relation of the soil samples, duplicate subsamples of soil were shaken with 50 ml of 0.001M CaC12 with KCI ranging from 0 to 0.002M in 100-ml centrifuge tubes for 24 hours at room temperature. Equilibrium solution was separated from the soil by centrifugation and analyzed for Na and K with an Ependorf flame spectrophotometer. The gain or loss of K was determined by substracting equilibrium K concentrations from initial K concentrations, assuming that the difference was either adsorbed or released from the soil complexes and converted to meq/100g. The graphs were obtained by plotting the gain or loss of K against the activity ratio of K (ARk) present in the equilibrium solutions. Cation concentrations were corrected to their chemical activities by using a modified Debye-Huckel equation assuming that all anions were monovalent and Ca, Mg, K, and Na were the only cations present.

-AZ+Z-VT
log a -= (23)
1 + Ba If

where A = 0.502

Z+ = valence of cation Z- = valence of anion
-8
a = 2.5 x 10- cm

B = 3.9 x 107

a = activity coefficient of electrolytes

I = ionic strength

I = 1/2 Z Ci Z2





35




Ci = concentration of the ion of charge Zi

Potassium selectivity coefficients were calculated by the Gapon equation as follows:

K (sorbed) aK (solution) (24) Kg = Ca + Mg (sorbed) /a(Ca + Mg) (solution)

Soil equilibrated with O.OO1N KCI and O.OO1M CaC12 was extracted three times with NH4NO3 and the combined extract was analyzed for K, Ca, and Mg as before. The results were corrected for the entrained solution by analyzing for Cl with an anion-specific ion electrode and potentiometer. The determined Cl was partitioned between cations in proportion to their concentrations in the equilibrium solution and these calculated amounts were substracted from measured values in the NH4NO3 extracts to obtain the sorbed cations. Cations in solution were determined from the equilibrium solution and their chemical activities were calculated (San ValentTn et al., 1972).

Other parameters were derived from the Q/I graphs: PBCk was calculated from the slope of the linear portion of the curve, the value of ARk from the intercept of the curve when AK = 0. Also, AKO and Kx were
o
determined as outlined by Beckett and Nafady (1967a) and Lee (1973). Exchange Properties

Soil CEC was determined by saturating with IN NH4O0Ac (pH 7.0) washing with C2H50H, exchanging the retained NH4 with IN NaCl, and determining the NH4 by micro-Kjeldahl method (Bremner, 1965). The extract from NH40Ac saturation was analyzed for Na, K, Ca, and Mg as before. Cation exchange capacity also was determined using the Stern model for double layer and point of zero net charge (PZNC) values.





36




Point of Zero Net Charge

This analysis is based on the premise that at PZNC, ionic strength will have no effect on potential, i.e., pH. The basic procedure used was that outlined by Adams (1976) except that only two concentrations of NaCl (1.0 and 0.001N) were used. This procedure involves equilibrating several 3-g soil samples for 3 hours in the two concentrations of NaCI to which various aliquots of NaOH or HCI have been added. After equilibration, the samples were centrifuged, the pH of the solution determined, and compared to the pH of the same solution before adding the soil. From this difference in pH, the milliequivalents of H+ or OH- adsorbed by the soil were determined and plotted against pH for each salt concentration. The intersection of the graphs was taken as the PZNC of the soil. Phosphorus Analysis

After centrifugation in a super-speed centrifuge, P in the supernatant solutions was determined. Phosphorus that disappeared from solution was considered to have been sorbed. Phosphorus sorbed was plotted against P in the supernatant solution. After equilibration, soil samples were saved for P release and P fractionation using the Petersen and Corey (1969) modification of the Chang and Jackson (1957) P fractionation procedure. This is a sequential extraction using NH4CI, NH4F adjusted to pH 8.32, NaOH, DCB, and H2S04, to extract water soluble, Al, Fe, reductant soluble, and Ca forms of P. The organic P was estimated from the difference between P extracted with NaOH followed by H2S04 from samples ignited to 6000C for 4 hours compared with nonignited samples. The total P was calculated by adding all the fractions. The P in this and all subsequent P experiments was determined colorimetrically using ammonium molybdate with ascorbic acid as a reducing agent according





37




to the procedures developed by Watanabe and Olsen (1965). All P determinations were made with a Bauch and Lomb Spectronic 20 spectrophotometer at a wavelength of 880 pm.

Calculations of Thermodynamic Parameters Associated with Adsorption
Processes

Thermodynamic parameters AG, AS, and AH were calculated from the variation of the thermodynamic equilibrium constant Ko (or the thermodynamic distribution coefficient) with changes with temperature. This technique was outlined by Biggar and Cheung (1973) and is described in details in Appendix II. Also, AG was calculated by using Keq from the relationship of concentration and the inverse of time as outlined by Rydens and Syers (1975). Langmuir and Freundlich equations were used to calculate some values of AG related to energy of sorption reactions (Ryden et al., 1977a). All these calculations were performed using a computer, examples of them are shown in Appendix II. Phosphorus Adsorption Isotherms

Data for plotting P adsorption isotherms were obtained by equilibrating 3-g soil samples for 11 days at 5'C, 250C, and 40C in 30 ml of water containing various amounts of KH2PO4. Also, isotherms were obtained in samples saturated with KCI and CaCl2 and washed free of Cl These were performed at 250C using KH2P04 and Ca(H2P04)2, respectively, to keep homoionic systems. Equilibration was carried out in 50-ml plastic centrifuge tubes. Two drops of toluene were added per sample to avoid biological activity. The tubes were shaken in a reciprocal shaker for a 30-minute period daily. Phosphorus Release

Samples from P adsorption isotherms were extracted 10 times with

0.02 M KCI or 0.02 M CaCl2, shaken for 5 minutes and centrifuged for





38




25 minutes with a high speed centrifuge. The KCl was used in samples with and without previous K saturation, while CaC12 was used with the Ca-saturated samples. Phosphorus was analyzed in each extract. The amount was corrected for the initial concentration of the solution left in the soil sample. A sulfur fractionation was performed in the Soil Testing Laboratory of the Soil Department, Facultad de Agronomfa, Venezuela, as outlined by Casanova (1974). This is a sequential extraction using HCI and NaHC03, with and without heat at 5000C. Lime and Phosphorus Treatments

Lime requirements of the soils were determined by incubating 500g samples in plastic pots with amounts of CaC03 equivalent to 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, and 10 metric tons/ha. Distilled water was added to the samples to bring the moisture content to moisture equivalent. The pots were closely covered and placed in a greenhouse over a period of 6 months. After incubation, soil samples were dried and pH was determined in H20. Lime requirement curves were obtained by plotting pH vs amount of CaCO3 applied.

In order to compare lime requirements, IN KCl-exchangeable Al

was determining. The rates of CaC03 selected for laboratory and greenhouse experiments were 0, 1/2, 1, and 2 times the IN KCI-exchangeable A1+3
Al

Phosphorus requirements were calculated from adsorption data obtained from the Langmuir isotherm. The P levels applied to the soil samples were 0, 1/4, 1/2, and I times the adsorption maximum calculated from Langmuir isotherms.





39




Laboratory and Greenhouse Experiments

Samples from nine Venezuelan soils were limed with CaC03 equivalent to 0, 1/2, 1, and 2 times the IN KCI-exchangeable Al +3. After 15 days of incubation, four levels of P were established in each soil by adding P as indicated by the appropriate sorption curve, to reach 0, 1/4, 1/2, and I adsorption maximum (Table 2). Phosphorus was added as KH2PO4 in solution form. The soil was then adjusted with water to moisture equivalent and incubated for 11 days. A portion of sample from each treatment was saved for PZNC and pH determinations.

One-kilogram samples of Guanipa 1, Guanipa 5, and El Potrero soils

were place in pots in a greenhouse and planted to sorghum (Sorghum bicolor L.), 'Hybrid Chaguaramas 3' was maintained by frequent irrigation. Plants were harvested 18 days after seeding. All pots received uniform fertilization with N, K, Mg, and micronutrients. The experimental design was two factors (lime and P) at four levels (1, 2, 3, and 4) with three replications and complete randomization. After harvest, soil samples were airdried, ground to pass a 2-mm sieve, and analyzed for PZNC and pH. Plant material was dried at 600C and weighed to determine yields.


Statistical Analyses


Analyses of variance were performed for greenhouse experimental data by the General Linear Model procedure (Barr et al., 1976) in the factorial design. Duncan's Multiple Range Test was used to compare individual treatments means. Also, a linear regression study was performed for P adsorption data. Special programs were made to calculate the surface charge with Stern and Gouy-Chapman models, the Q/I parameters for K status in the soils, and thermodynamic parameters.




1140

















Table 2. Treatment combinations used in the
laboratory and greenhouse experiment with the soils used.

Phosphorus applied CaC03 applied
Rate Designation 1 2 3 4

Treatment combinations Obt 1 11 21 31 41 b 2 12 22 32 42 b 3 13 23 33 43 lb 4 14 24 34 44


tAdsorption maximum (b) from the appropriate
Langmuir equation.

*0, , 1, and 2 times the IN KCl-exchangeable Al.


















RESULTS AND DISCUSSION



Soil Characteristics with Emphasis on Surface Charge


The soils selected for this study varied considerably in clay and organic matter contents, CEC, exchangeables bases, and acidity sources. Also, they showed a very large variation in free Fe and Al oxide contents, surface area, clay mineralogy, PZNC, and P and S fractions. The K status was quite different for these soils as shown by the Q/1 parameters found with this technique.

Soil organic matter content is a result of biological processes that are controlled by temperature and moisture. The relatively low organic matter contents of these soils was due mainly to the oxidation enhancement of high temperatures. Organic matter colloid translocation was not noted since there was a tendency for lower organic matter contents with depth. The organic matter contents ranged from 0.28 to 4.54% (Table 3). These values are the result of the combined temperature, rainfall, and altitude factors (Westin et al., 1968).

Exchangeable bases in these soils were generally low. Calcium was the predominant ion in all soils (Table 4). Magnesium followed Ca in the exchangeable bases. Potassium values were very low in all soils. Except for San Crist6bal and El Potrero, exchangeable Na was less than

0.1 meq/100g. In general, the highly weathered soils contained less bases than those less weathered. There was a tendency for bases to 41












Table 3. Physical and chemical properties of soils examined.

Soil O.M. Clay Surface Area CEC ECEC PZNC Clay Mineralst
------------ --(m2/100g)- -meq/100g-.. --(m2/lOOg)- -meq/lOOg-Guanipa 1 0.60 6.90 2.7 2.0 1.88 3.25 KA, CH, QZ, Co Guanipa 5A 0.33 4.40 2.9 1.2 0.64 2.85 KA, QZ, IN, CH, HE Guanipa 5B 0.28 6.80 2.8 1.1 0.32 3.20 KA, IN, CH, HE, Co San Crist6bal 1.26 51.00 46.6 23.0 14.10 4.10 VM, KA, QZ, CH, MI El Rocro 2.72 29.40 39.4 8.3 3.65 3.50 IN, KA, QZ, CH, MI, Co Bajo Seco 2.52 22.20 27.0 17.0 7.90 3.00 IL, QZ, MI, KA, Co, CH Guanaguanare 2.49 26.00 54.6 30.4 8.29 3.90 IL, QZ, KA, CH, Co Guataparo 1.37 30.30 25.0 5.6 3.65 3.75 KA, QZ, IN, GI, Co Alambre 4.54 24.70 44.0 17.6 7.55 3.50 IN, QZ, KA, GI, Co, CH El Potrero 3.26 50.60 69.9 11.0 5.40 3.20 KA, QZ, IL, Go Barinas 1.28 17.70 34.3 6.3 2.40 4.60 KA, QZ, IL, Go Red Bay Al 2.50 8.20 13.3 7.0 7.98 3.95 IN, GI, KA, QZ, Co Red Bay B2t 0.60 36.60 25.2 2.2 1.26 4.20 KA, IN, GI, QZ

tThe following abbreviations are used: KA=kaolinite; CH=chlorite; QZ=quartz; VM=vermiculite; IN= intergrade; HE=haematite; Go=goethite; MI=mica; GI=gibbsite; Co=corundum; IL=illite.













Table 4. Exchange properties and free oxides of the soils examined.
Exchangeable
Al Acidity Bases
Soil Fe oxides Al oxides IN KCI Ba(OH)2 Na K Ca Mg
-----------%----------- ------------------meq/100g--------------------Guanipa 1 0.93 0.93 0.021 1.40 0.02 0.08 0.36 0.35 Guanipa 5A 0.75 1.03 0.035 1.30 0.01 0.04 0.35 0.25 Guanipa 58 0.87 0.76 0.040 1.70 0.01 0.02 0.20 0.10 San Crist6bal 3.57 3.45 0.141 10.00 0.82 0.50 7.50 6.10 El Rocio 3.10 2.30 0.600 3.30 0.06 0.15 3.30 0.20 Bajo Seco 3.07 1.90 0.050 9.00 0.01 0.55 6.30 1.03 Guataparo 3.07 1.52 1.300 4.36 0.07 0.25 1.05 0.37 Guanaguanare 3.39 2.60 0.080 6.00 0.06 0.19 5.10 3.00 Alambre 3.46 1.73 3.900 12.00 0.02 0.15 2.40 1.00 El Potrero 4.29 2.90 3.020 13.00 0.80 0.30 1.10 1.00 Barinas 1.82 1.50 0.100 4.95 0.07 0.15 0.82 0.69 Red Bay Al 1.29 0.64 0.550 16.01 0.02 0.26 5.72 1.00 Red Bay B2t 1.79 0.89 0.430 3.40 0.01 0.05 0.50 0.33




44




decrease with depth. These values were expected because of the low pH of all soils. Cation exchange capacity values were variable (Table 3) and higher than effective CEC (ECEC). Guanipa I showed CEC equal to ECEC. Cation exchange values ranged from 1.1 to 30.4 meq/100g. The values of CEC were due to the contribution of organic matter and clay content in these soils. The magnitude of permanent negative charge would be expected to vary according to the degree of tetrahedral substitution in the clay mineral. The amounts of Al and Fe oxides present in addition to the presence of crystalline colloidal species with low pH-dependent charges would considerably influence CEC measurement at pH 7.00.

Free Fe-oxide values ranged from 0.75 to 4.29%. Aluminum oxides ranged from 0.64 to 3.45% (Table 4). These values reflected the stage of soil weathering. These results indicate a loss of Si due to the acidic environment and an enrichment in the Al and Fe oxides. A tendency of these values to decrease with depth indicated that Fe and Al have been released from primary minerals and translocated by percolating water to lower horizons.

Surface-area values ranged from 2.7 to 69.9 m2/100g (Table 3) and were related to organic matter and clay content in these soils. The highest value was obtained for the Aridisol and the lowest for the sandy Ultisol. These soils had increasing surface area with depth. The migration of fine particles in soil formation processes can account for these results (Luque, 1975).

The amount of clay in soils depends upon the soil-forming factors including parent material, relief, time, biotic, and climate. The clay




45




contents in the soils studied ranged from 4.40 to 51.00% (Table 3). Clay content tended to increase with depth, indicating downward movement and accumulation in lower horizons.

Clay Mineralogy

Abundance of kaolinite and quartz is indicative of parent materials in their last weathering stage. Oxisols which represent the extreme of pedogenic weathering have traces of aluminosilicates such as intergrade (Table 3). These soils contain various Fe and Al oxides (gibbsite, hematite, and goethite) as evidenced by 4.83, 3.67, and 4.96 R peaks, respectively, and in some cases, chlorite. Soils Guanipa 5 and Guataparo represent this order. Weathering has been so severe that only resistant minerals (quartz 4.2 and 3.35 2, and corundum 3.47, 2.55, and 2.08 2), and highly resistant layer silicates remain.

Clay fractions with Ultisols were dominated by kaolinite and intergraded phyllosilicates (Guanipa 1 and Red Bay). Kaolinite was identified by first order spacing 7.0 2. A band in 11.6 to 12.6 2 region was assumed to represent intergrade minerals because of their persistence after heating. Guanipa 1 showed some chlorite. The peaks at 13.6 to 14.3 R without expanding at Mg treatment were assumed to be chlorites.

Alfisols (San Crist6bal, Barinas, and Bajo Seco) tended to have a somewhat wide-ranging clay mineral suite. Barinas (Haplustalf) showed predominance of kaolinite, followed by quartz and illite, and some amount of goethite, while Bajo Seco (Tropudalf) had illite, quartz, mica, and some amounts of kaolinite, corundum, and chlorite. Finally, San Crist6bal (Tropudalf) showed vermiculite, in addition to kaolinite,quartz, mica, and chlorite. Vermiculite was concluded to be present by the Mgsaturation 14 to 15 R spacing and a 10.0 2 spacing after heating to 5000C.




46




The presence of these minerals may have been the result of a more baseenriched environment, and to low rainfall and a distinct dry season (Westin et al., 1968; Zelazny and Calhoun, 1971). Entisols and Inceptisols exhibited inherited minerals such as mica, intergrade layer silicates and chloride in addition to kaolinite and quartz (Table 3). Point of Zero Net Charge

The PZNC values for the soils examined ranged from pH 2.85 to 4.60 (Table 3). The Guanipa I and Barinas soils were the lowest and highest, respectively. These values tended to increase with soil depth and weathering. If a PZNC range of pH 3.5 to 5.0 is characteristic of soils dominated by constant surface-potential colloids as proposed by Keng and Uehara (1974), those soils, except for Guanipa 1, should belong to this group. The lower PZNC values for these soils were due to the presence of organic matter in their surface horizons. It has been shown by Adams (1976) that the narrowed range of PZNC in tropical and temperate regions was due to the interlayered Al and organic matter. Phosphorus and Sulfur Fractionation

Phosphorus fractions as determined by the Peterson and Cory (1969)

method showed dominant P forms to be Fe-P, Al-P, and Ca-P. All the soils, except Guanipa 1 and 5A had lower amounts of AI-P than Fe-P. This may have been due to the lower solubility product of Fe-P. Also, aging may have been responsible because of the high Fe oxide content (Yuan et al., 1960). Soluble P was not detected in these soils. Most of the P was in the organic form (Table 5).

On the other hand, S fractionation (Casanova, 1974) showed that S was mainly in the clay and organic forms (Table 6). These values of S in the clay fraction suggested that these soils have a high capacity to adsorb and exchange phosphate (Casanova, 1974).













Table 5. Phosphorus fractionation of the virgin soils examined.
Soil Organic P Soluble P Al-P Fe-P Prs Ca-P P Total
----------------------------------ppm--Guanipa 1 43.2 Tt 14.6 14.1 T 5.6 75.1 Guanipa 5A 28.1 T 12.2 9.0 T 5.6 56.2 Guanipa 5B 28.1 T 8.5 9.0 T 3.5 47.8 San Crist6bal 197.3 T 31.5 36.0 T 84.7 331.3 El Rocio 422.7 T 98.2 123.8 16.2 70.0 776.8 Bajo Seco 147.9 T 15.2 22.5 T 9.1 178.7 Guataparo 281.8 T 26.8 56.3 T 11.9 369.8 Guanaguanare 380.4 T 47.0 58.5 18.3 44.8 524.9 Alambre 285.0 T 15.9 31.2 4.6 19.6 382.2 El Potrero 335.0 T 21.4 36.0 T 14.0 410.1 Barinas 172.0 T 15.9 19.1 T 11.5 211.0 Red Bay Al 118.7 T 17.1 29.3 T 7.0 176.9 Red Bay B2t 290.1 T 18.0 27.0 T 9.8 343.9


The letter T indicates trace quantities.















Table 6. Sulfur fractionation of some of the virgin soils examined.

Soil Organic S Available S Clay - S Total S
--------------------- --ppm----------------------Guanipa 1 T 86.0 62.0 119.0 San Crist6bal 33.3 54.7 42.21 87.3 El Rocio 13.0 53.3 47.43 66.6 Bajo Seco 1042.0 19.0 57.0 1231.0 Guataparo 76.0 91.0 63.0 170.0 Guanaguanare 40.0 40.0 30.0 80.0 Alambre 40.0 26.6 15.0 66.6 El Potrero 42.0 53.3 48.0 95.3 Barinas 71.0 90.0 51.0 161.0




49




The experiment carried out to determine the temperature of incubation effect on P fractions after an Ii-day equilibrium period showed that most of the P was in the NH4F and NaOH extracts and was assumed to be in the Al and Fe forms, respectively (Tables 7 through 19). Only small amounts of P were found in the NH4CI and H2S04 extracts which represent water-soluble and Ca phosphate forms (Fig. 3). The phosphate applied, even at high rates, was fixed by Al and Fe. This was due to the fact that in acid soils the applied P precipitates or is adsorbed as Fe-P and Al-P (Yuan et al. 1960).

In the Guanipa 5B (A12 horizon), at highest P applied and 400C, the amount of P in the NaOH extract was more than in the NH4F extract (Table 9). This indicated that P in this soil tended to form Fe-P rather than AI-P because of the lower solubility product of Fe phosphate or by a shifting of P from the Al-P form to Fe-P form induced by the higher temperature (Yuan et al., 1960).

The ratios of AI-P to Fe-P broadened with increasing P applied.

It has been pointed out that kaolinite can function in the fixation of P by bonding of P to Al at the broken edges. This could have been the case in soils with kaolinite as the dominant clay mineral (Yuan et al.,

1960).

The pH and pH2P04 of the different ionic systems at 250C are shown in Tables 20 through 28. The pH2PO4 values were very low at high equilibrium P concentrations. This indicated that precipitation of Al and Fe phosphates such as strengite and variscite can occur. An experiment was carried out to observe the effect of high temperature on Al- and Fe-phosphate precipitation in the nonsaturated system with the highest





50















Table 7. Effect of incubation temperature on phosphorus
fractions in a typic Paleustult from Anzoategui State, Venezuela (0-30 cm) after applying several solution concentrations of phosphorus.

P applied Al-P Fe-P Prs Ca-P as KH2P04

-----------------------------ppm----------------------50C

0 11.04 9.41 8.35 1.67
100 56.76 19.98 8.35 1.39 500 104.4 30.14 8.6 3.29 1000 145.7 25.59 7.97 2.46

250C

0 10.88 9.66 15.25 1.84
100 80.62 30.32 6.96 6.39 500 125.75 37.93 11.19 5.68 1000 151.8 37.93 8.6 4.90

400c

0 6.1 9.7 9.88 1.92
100 71.77 22.8 13.81 2.73 500 157.9 26.78 16.65 2.73 1000 276.5 28.82 17.88 2.87




51
















Table 8. Effect of incubation temperature on phosphorus
fractions in a Grossarenic Psammentic Haplustox
from Anzoategui State, Venezuela (All horizon) after applying several solution concentrations
of phosphorus.

P applied Al-P Fe-P Prs Ca-P as KH2PO4
-----------------------------ppm-----------------------50C

0 11.7 14.21 8.64
100 160.4 10.62 8.60 500 326.11 11.21 8.60 1000 380.0 13.97 9.88

250C

0 12.2 9.0 9.80 1.80
100 64.25 20.12 4.83 3.28 500 115.5 23.58 7.34 4.85 1000 140.0 27.10 3.60 4.58

40�c

0 5.51 14.67 12.50 0.37
100 66.97 26.01 16.38 0.78 500 143.61 30.79 17.75 1.26 1000 388.17 34.85 19.27 1.26





52














Table 9. Effect of incubation temperature on phosphorus
fractions in a Grossarenic Psammentic Haplustox
from Anzoategui State, Venezuela (A12 horizon) after applying several solution concentrations
of phosphorus.

P applied Al-P Fe-P Prs Ca-P as KH2P04
----------------------------ppm-----------------------5�C
0 8.85 6.88 8.60
100 64.25 13.17 8.60 0.50 500 144.9 15.01 8.35 0.62 1000 179.0 20.11 8.60 0.88

250C

0 8.5 9.0 4.8 2.2
100 77.17 25.74 4.83 2.31 500 128.7 31.40 4.83 2.51 1000 138.8 23.58 4.83 1.12

400C

0 4.37 18.25 13.16 1.01
100 94.02 301.41 19.00 1.52 500 138.62 315.86 19.27 1.79 1000 186.84 367.0 10.00 1.79

















1000



P applied / as KH2PO4 7 7 (uig//ml)




500-L
SA1-P
S7Fe-P
'3J Ca-P







50 3 5 6 7 8 9 10 11 12 13 SOILS

Fig. 3. Phosphorus fractions as affected by P concentrations after an 11-day equilibrium period of 25'C.
(l=Guanipa 1; 2=Guanipa 5A; 3=Guanipa 5B; 4=San Cristobal; 5=El Rocio; 6=Bajo Seco; 7=Guataparo;
3=Guanaguanare; 9=Alambre; 10=El Potrero; 1l=Barinas; 12=Red Bay Al; and 13=Red Bay B2t.)





54















Table 10. Effect of incubation temperature on phosphorus
fractions in an Ultic Tropudalf from Thchira
State, Venezuela (0-30 cm) after applying
several solution concentrations of phosphorus.

P applied Al-P Fe-P Prs Ca-P as KH2P04
----------------------------ppm-----------------------50C

0 10.4 40.9 14.5 1.84
100 283.7 188.6 25.3 3.40 500 856.1 293.1 31.0 4.00 1000 1179.0 357.1 34.2 9.50

250C

0 26.5 44.70 1.2
100 311.7 255.0 9.25 500 908.8 264.3 15.15 1000 1454.3 372.3 15.15

400C

0 8.94 42.6 1.3 1.76
100 680.4 264.8 44.0 2.6 500 1124.3 294.7 72.8 3.1 1000 2196.3 301.4 84.1 3.7




55















Table 11. Effect of incubation temperature on phosphorus
fractions in a Ustic Quartzipsamment from Portuguesa State, Venezuela (0-30 cm) after
applying several solution concentrations of
phosphorus.

P applied Al-P Fe-P Prs Ca-P as KH2P04
----------------------------ppm-----------------------5"0C

0 28.9 91.0 52.55 2.73
100 342.8 235.79 87.0 12.58 500 788.2 377.1 96.8 10.61 1000 1016.8 782.6 102.3 7.75

250C

0 35.8 41.7 51.9 5.40
100 394.6 612.8 60.6 5.65 500 808.1 420.7 64.6 5.68 1000 1170.7 480.8 64.6 5.68

400C

0 34.5 47.26 89.5 2.9
100 658.4 217.7 132.1 3.15
500 1361.4 329.1 174.2 4.6 1000 1650.6 387.5 175.3 4.5





56














Table 12. Effect of incubation temperature on phosphorus
fractions in an Oxic Tropudalf from Miranda
State, Venezuela (0-30 cm) after applying
several solution concentrations of phosphorus.

P applied Al-P Fe-P Prs Ca-P as KH2PO4
----------------------------ppm------------------------50C

0 15.9 47.1 2.40 2.56
100 582.3 126.4 45.37 5.19 500 756.9 188.6 45.73 6.59 1000 886.0 270.9 52.91 4.59

250C

0 18.2 201.4 8.6 2.7
100 309.0 396.9 64.6 7.4 500 775.5 241.4 79.8 12.2 1000 810.3 204.0 56.6 11.5

400c

0 12.5 39.9 2.8 6.1
100 420.6 338.2 75.0 5.5 500 870.8 347.5 97.8 5.4 1000 938.5 420.6 113.1 5.8




57















Table 13. Effect of incubation temperature on phosphorus
fractions in a Tropeptic Haplustox from Carabobo State, Venezuela (0-30 cm) after applying several solution concentrations of phosphorus.

P applied Al-P Fe-P Prs Ca-P as KH2PO4
-------- -------- -- -----ppm------------------------5*C

0 9.3 1.5 11.2 2.4
100 214.5 47.2 17.6 3.9 500 667.6 83.3 24.9 3.4 1000 956.1 82.3 24.9 2.8

250C

0 5.2 153.0 1.2 2.4
100 325.3 181.9 17.2 4.0 500 676.6 194.6 6.1 3.5 1000 930.0 158.1 8.6 3.2

400C

0 10.6 24.0 19.4 2.6
100 432.0 96.5 32.2 3.6 500 694.5 148.9 37.3 5.5 1000 1175.9 176.4 37.3 16.6





58













Table 14. Effect of incubation temperature on phosphorus fractions in an Udorthentic Pellustert
from Portuguesa State, Venezuela (0-30 cm) after applying several solution concentrations of phosphorus.
P applied Al-P Fe-P Prs Ca-P as KH2PO4
----------------------------ppm-----------------------5SC
0 19.5 17.7 44.3 6.7
100 382.5 97.4 44.3 5.9 500 933.4 132.3 51.0 10.8 1000 1007.0 163.6 71.2 9.5

250C

0 24.7 68.0 23.1 2.1
100 233.0 441.1 129.6 3.7 500 662.0 676.6 129.6 4.0 1000 868.7 511.7 150.9 4.8

400C

0 17.3 53.9 74.8 4.5
100 534.3 490.3 162.0 5.2 500 1253.8 609.2 206.6 5.9 1000 1896.2 688.2 226.5 8.1




59













Table 15. Effect of incubation temperature on phosphorus fractions in a Fluaquentic Humitropept
from Portuguesa State, Venezuela (0-30 cm) after applying several solution concentrations of phosphorus.
P applied Al-P Fe-P Prs Ca-P as K!12P04

--------- ---------- -----ppm------------------------5"C

0 6.6 31.4 33.4 3.7
100 406.8 62.3 49.3 6.8 500 1158.5 61.3 55.1 7.5 1000 712.1 66.5 54.8 11.4

250C

0 15.9 9.8 67.7 2.0
100 456.9 82.4 52.9 4.3 500 1186.2 87.0 59.5 4.1 1000 1247.1 357.2 70.8 4.8

400C

0 26.9 36.2 37.3 1.5
100 971.2 351.4 58.6 1.5 500 2502.9 382.3 80.3 3.5 1000 3014.5 484.6 83.3 3.4




60















Table 16. Effect of incubation temperature on phosphorus
fractions in a Haplargid from Lara State, Venezuela (0-30 cm) after applying several
solution concentrations of phosphorus.

P applied Al-P Fe-P Prs Ca-P as KH2PO4
--------------------------pm------------------------50C

0 9.3 23.4 33.8 3.6
100 345.3 55.9 37.3 87.0 500 997.4 62.3 40.9 87.0 1000 1284.3 59.4 42.3 87.0

250C

0 10.3 57.2 24.9 9.3
100 513.8 185.4 32.1 4.9 500 908.8 277.2 38.1 6.1 1000 1300.1 311.5 51.1 9.0

400c

0 16.4 59.4 37.3 8.7
100 673.0 320.3 41.9 8.7 500 1576.5 333.6 45.7 8.7 1000 2111.6 382.3 56.5 10.5





61














Table 17. Effect of incubation temperature on phosphorus
fractions in an Oxic Haplustalf from Barinas
State, Venezuela (0-30 cm) after applying
several solution concentrations of phosphorus.

P applied Al-P Fe-P Prs Ca-P as KH2PO4
--------------------------ppm------------------------50C

0 6.9 16.2 22.1 3.9
100 262.4 28.0 29.4 6.1 500 696.6 52.2 29.6 6.5 1000 840.9 41.1 30.2 7.3

250C

0 24.6 15.3 29.4 3.1
100 252.6 69.7 35.4 2.7 500 702.9 105.2 35.4 2.5 1000 988.4 123.8 40.9 3.0

400c

0 13.1 24.4 27.2 6.4
100 361.3 118.7 33.8 9.3 500 788.2 171.2 46.6 10.6 1000 1231.1 210.8 48.3 10.6





62















Table 18. Effect of incubation temperature on phosphorus
fractions in a Rhodic Paleudult from Florida,
U.S.A. (Al horizon) after applying several
solution concentrations of phosphorus.

P applied Al-P Fe-P Prs Ca-P as KH2PO4
---------------------------ppm------------------------50C

0 9.7 21.3 17.8 1.6
100 294.3 38.4 23.6 3.0 500 763.4 37.4 30.9 3.7 1000 992.5 72.6 25.7 4.3

250C

0 37.3 27.7 17.8 7.4
100 473.0 110.1 30.2 8.7 500 916.7 152.2 35.7 8.5 1000 1003.2 181.3 34.9 9.7

400c

0 28.6 36.7 16.5 3.9
100 691.5 220.0 35.7 6.4 500 2313.6 270.9 46.6 10.0 1000 2227.3 286.6 54.7 9.8




63















Table 19. Effect of incubation temperature on phosphorus
fractions in a Rhodic Paleudult from Florida,
U.S.A. (B2t horizon) after applying several
solution concentrations of phosphorus.

P applied Al-P Fe-P Prs Ca-P as KH2P04
----------------------------ppm----------------------50C

0 11.3 20.5 32.5 4.0
100 385.0 28.0 33.8 7.6 500 765.8 34.0 35.7 7.7 1000 1145.3 31.7 36.0 7.7

250C

0 21.0 22.6 21.3 5.3
100 410.5 63.5 30.2 5.4 500 871.2 92.6 41.3 4.7 1000 1052.3 110.1 34.1 3.9

400C

0 12.6 31.3 27.9 6.0
100 445.0 220.6 43.9 11.7 500 1121.4 228.8 47.5 11.7 1000 1443.8 244.9 61.5 15.6





64
















Table 20. Values of pH2PO4 and pH at equilibrium with different
ionic systems after P adsorption for a typic Paleustult from Anzoategui State, Venezuela.

P Nonsaturated K saturated Ca saturated added pH2P04 pH pH2P04 pH pH2PO4 pH

-1g/ml10 3.61 6.75 4.41 5.95 3.52 5.40 50 2.84 6.35 3.14 5.85 3.22 5.30 100 2.52 5.85 2.86 5.65 2.92 5.00 200 2.22 5.70 2.61 5.50 2.62 4.55 700 1.68 5.30 1.93 5.10 1.92 3.70 1000 1.55 5.20 1.52 4.80 1.58 3.40





65
















Table 21. Values of pH2P04 and pH at equilibrium with different
ionic systems after P adsorption for an Ultic Tropudalf from Tchira State, Venezuela.
P Nonsaturated K saturated Ca saturated added pH2PO4 pH pH2PO4 pH pH2PO4 pH

-pg/ml10 4.11 6.60 3.67 6.50 3.71 6.15

50 6.05 3.64 6.00 3.64 5.60

100 2.80 5.70 3.08 6.10 3.30 5.30 200 2.40 5.55 2.79 5.90 2.88 5.00 700 1.75 5.20 2.03 5.60 2.02 4.90 1000 1.58 5.10 1.69 5.25 1.67 4.60




66
















Table 22. Values of pH2P04 and pH at equilibrium with different
ionic systems after P adsorption for an Ustic Quartzipsamment from Portuguesa State, Venezuela.

P Nonsaturated K saturated Ca saturated added pH2PO4 pH pH2PO4 pH pH2PO4 pH

-pg/ml10 4.58 6.75 3.94 6.70 3.62 6.10 50 3.55 6.70 3.43 6.30 3.55 6.15 100 3.05 6.50 3.08 6.05 3.40 6.10 200 2.48 5.70 2.82 6.25 3.31 5.45 700 1.75 5.15 2.00 6.90 2.04 4.70 1000 1.58 5.10 1.55 5.35 1.67 4.30





67
















Table 23. Values of pH2PO4 and pH at equilibrium with different
ionic systems after P adsorption for an Oxic Tropudalf
from Miranda State, Venezuela.

P Nonsaturated K saturated Ca saturated added pH2PO4 pH pH2PO4 pH pH2PO4 pH

-pg/ml10 4.26 7.35 3.51 6.85 4.07 6.50 50 3.56 7.25 3.12 6.80 3.53 6.30 100 2,97 7.05 2.94 6.70 3.16 5.80 200 2.46 6.60 2.69 6.40 2.77 5.40 700 1.74 5.60 1.97 5.80 1.98 4.80 1000 1.55 5.45 1.60 5.50 1.68 4.30





68
















Table 24. Values of pH2PC4 and pH at equilibrium with different
ionic systems after P adsorption for an Udorthentic
Pellustert from Portuguesa State, Venezuela.

P Nonsaturated K saturated Ca saturated added pH2PO4 pH pH2P04 pH pH2PO4 pH

-vg/ml10 4.53 7.45 3.93 7.60 3.71 6.90 50 3.95 7.75 3.34 7.20 3.63 6.70 100 3.03 7.70 3.00 7.10 3.23 6.35 200 2.69 6.90 2.72 6.35 2.83 5.80 700 1.82 6.05 2.02 6.00 1.98 4.50 1000 1.61 5.60 1.62 5.60 1.67 4.20





69
















Table 25. Values of pH2P04 and pH at equilibrium with different
ionic systems after P adsorption for a Tropeptic
Haplustox from Carabobo State, Venezuela.
P Nonsaturated K saturated Ca saturated added pH2P04 pH pH2P04 pH pH2P04 pH

-pg/ml10 4.28 6.05 3.86 5.55 3.71 6.00 50 3.12 5.45 3.30 5.55 3.58 5.70 100 2.72 5.40 3.01 5.62 3.21 5.55 200 2.33 5.35 2.81 5.55 2.80 5.25 700 1.71 5.10 1.98 5.20 1.95 4.60 1000 1.54 5.05 1.57 5.05 1.68 3.90




70
















Table 26. Values of pH2P04 and pH at equilibrium with different
ionic systems after P adsorption for a Fluaquentic
Humitropept from Portuguesa State, Venezuela.

P Nonsaturated K saturated Ca saturated added pH2PO4 pH pH2PO4 pH pH2PO4 pH

-Pg/ml10 5.01 6.55 3.78 6.00 3.72 5.60 50 3.85 6.75 3.51 6.00 3.68 5.30 100 3.03 5.55 3.18 5.35 3.50 4.80 200 2.51 4.76 2.88 5.25 3.03 4.50 700 1.78 4.60 2.04 5.30 2.07 4.00 1000 1.59 4.65 1.60 4.90 1.67 3.80





71

















Table 27. Values of pH2P04 and pH at equilibrium with different
ionic system after P adsorption for a Haplargid from
Lara State, Venezuela.

P Nonsaturated K saturated Ca saturated added pH2PO4 pH pH2PO4 pH pH2PO4 pH

-1g/ml10 4.53 4.80 3.58 5.65 3.75 4.80 50 3.38 4.70 3.37 5.45 3.71 4.70 100 2.88 4.30 3.06 5.40 3.35 4.40 200 2.43 4.55 2.81 5.20 2.91 4.30 700 1.75 4.50 1.99 4.80 2.07 3.80 1000 1.58 4.50 1.60 4.65 1.66 3.70





72
















Table 28. Values of pH2PO4 and pH at equilibrium with different
ionic systems after P adsorption for an Oxic Haplustalf from Barinas State, Venezuela.
P Nonsaturated K saturated Ca saturated added pH2PO4 pH pH2PO4 pH pH2PO4 pH

-1g/ml10 4.41 6.75 3.64 6.55 3.75 6.95 50 3.13 6.30 3.40 6.50 3.74 6.85 100 2.70 6.00 3.02 6.30 3.36 6.50 200 2.34 5.85 2.64 6.00 3.27 5.95 700 1.71 5.50 1.95 5.60 2.00 5.00 1000 1.54 5.40 1.57 5.35 1.67 4.65





73




P concentration. Soil samples were autoclaved first for 20 hours at 1200C and then for 40 hours at 1200C. X-ray diffraction patterns showed the presence of strengite 20-hour treatment in some soils but no variscite. For the 40-hour treatment, both variscite and strengite were present, except in the San Crist6bal and El Rocio soils. These minerals were identified by the strong peaks at 16.4, 20.6, and 26 R (Blanchard, 1974). As has been pointed out by Webber (1978), hydrous aluminum oxide plays an important role in binding phosphate in acid soils and clays; variscite formation is expected in acid tropical soils, but amorphous Al phosphate would be expected to persist indefinitely in other acid soils (Webber, 1978).

In these tropical acid soils, high P application may tend to produce precipitation of (AlFe)P04 compounds. This could explain why P is more deficient in tropical acid soils than in acid soils from temperate regions.


Negative Surface Charge


Acidity

To determine the release of protons in these surfaces a complete study of acidity sources was carried out through intensity and extractable acidity measurements.

The pH of a soil suspension is sometimes lowered or increased by the addition of an electrolyte. Electrolytes such as KCI, K2S04, and CaCl2 lower the pH while higher pH values are found with NaF in soil suspension. The sign of the difference between soil pH in KCI, an indifferent electrolyte, and soil pH in water has been related to the sign of the net surface change of soil colloids (Mekaru and Uehara,





74




1972). The difference between pH in KCI and K2S04 has been associated with anion adsorption (Adams, 1976). The pH in NaF has been related to the release of hydroxyl groups in Andosols (Calhoun and Carlisle, 1971).

The measurements of pH showed a wide range of variation from very strongly acid in El Potrero to weakly acid in Guanipa 5A. All pH values in water in the soils examined were decreased with addition of KCI. Except for Guataparo and El Potrero soils, pH values in water were decreased with addition of K2S04 and CaCl2. The pH values in NaF ranged from 8.40 to 10.25. These values tended to be higher in the Oxisols, Ultisols, and Alfisols than in the less weathered soils (Table 29).

The negative values of ApH(KCI-H20) indicated that these soils had a net negative charge. The negative value of ApH(KCI-K2S04) indicated a high capacity of those soils to exchange anions and, therefore, to fix phosphate. The ApH(KCI-K2S04) values were caused by the fact that the amount of hydroxyl ions displaced by sulphate anions was greater than the Al displaced by K ions (Adams, 1976). The very high negative values of ApH(H20-NaF) indicated a large displacement of hydroxyl groups linked to Al and Fe oxides in these soils (Calhoun and Carlisle, 1971).

The values of extractable acidity and Al with IN KCI and those from potentiometric and conductometric titrations are shown in Table 30. The IN KCl-exchangeable Al and acidity titrated with Ba(OH)2 ranged from 0.02 to 3.90 and 1.30 to 16.01 meq/100g, respectively. The higher values of exchangeable acidity compared to exchangeable Al+3 may have been caused by the contribution of polymeric Al and aluminate reactions that occurred during shaking and titration in the Ba(OH)2 (Dewan and Rich, 1970).
For Venezuelan soils potentiometric and conductometric titration

values with NaOH, Ba(OH)2, and Na2B407 are shown in Table 30. At pH 8.00,














Table 29. Soil reaction with selected electrolyte and calculated ApH values for the soils examined.

pH ApH
Soil H20 KCI K2SO4 NaF CaCl2 KCI-H20 KCl-K2S04 H20-NaF

Guanipa 1 5.45 4.36 4.84 8.60 4.64 -1.09 -0.48 -3.15
Guanipa 5A 5.68 4.47 5.01 8.45 4.80 -1.19 -0.54 -2.77
Guanipa 5B 5.58 4.34 5.02 9.10 4.60 -1.24 -0.68 -3.52
San Crist6bal 5.50 3.98 4.53 9.20 4.69 -1.52 -0.55 -3.70
El Rocro 4.50 3.94 4.48 8.95 4.51 -0.56 -0.54 -4.45
Bajo Seco 5.25 4.75 5.07 8.75 5.12 -5.12 -0.32 -3.50
Guanaguanare 5.00 4.55 5.00 8.60 5.03 -0.45 -0.45 -3.60
Guataparo 4.03 3.90 4.43 9.55 4.32 -0.40 -0.53 -5.25
Alambre 4.20 3.57 4.08 9.15 4.13 -0.63 -0.51 -4.95
El Potrero 3.55 3.44 3.95 8.80 3.82 -0.11 -0.51 -5.25
Barinas 5.10 4.37 5.01 9.85 5.03 -0.73 -0.64 -4.75
Red Bay Al 5.30 4.61 5.20 9.75 5.20 -0.69 -0.59 -4.45
Red Bay B2t 5.50 5.00 5.50 10.25 5.40 -0.50 -0.50 -4.75




76













Table 30. Acidity from potentiometric titrations at pH 8.00 and conductometric titrations with NaOH, Ba(OH)2, and Na2B407 for
the Venezuelan soils examined.

Potentiometric titration Conductometric titration Soil at pH 8.00 with with NaOH Ba(OH)2 Na2B407 NaOH Ba(OH)2 Na2B407

---------------------meq/100g-----------------Guanipa 1 2.0 2.6 1.6 22.8 23.8 12.0 Guanipa 5A 2.0 2.8 1.6 27.6 27.6 20.0 Guanipa 5B 1.2 1.6 1.6 30.4 26.1 12.0 San Crist6bal 12.8 6.0 8.0 50.4 39.6 12.0
El Rocio 8.4 10.6 10.6 46.4 34.8 10.4 Bajo Seco 7.2 6.0 6.0 45.6 24.4 7.0 Guanaguanare 11.2 10.0 10.0 38.4 32.7 44.8 Guataparo 7.6 6.4 7.6 28.4 19.8 24.4 Alambre 11.2 13.0 13.0 61.6 31.5 58.0
El Potrero 16.4 14.0 16.0 46.0 41.7 22.8 Barinas 3.6 4.4 4.0 24.0 28.5 24.4





77



the acidity titrated ranged from 1.2 to 16.4 meq/100g in NaOH, from 1.6 to 14.0 in Ba(OH)2, and from 1.6 to 16.0 in Na2B407. The soils with lowest and highest values were Guanipa 58 and El Potrero. The acidity titrated at this point was probably due to dissociation of organic matter groups, SiOH groups, and exchangeable and polymeric Al (Jackson, 1963). The values obtained with Ba(OH)2 and NaOH were similar except for the San Crist6bal soil. This suggested that there was no exchangeable or bonding of H to permanent negative charge in these soils (Shainberg and Dawson, 1967).

As pointed by Shainberg and Dawson (1967), the acidity titrated with Na2B407 is closely related to permanent negative charge (Table 3). The values obtained with conductometric titrations indicated high values of exchangeable Al, probably caused by dissolution of minerals with release of structural Al or reaction of bases with organic acids (Dewan and Rich, 1970). The higher values of extractable acidity shown by the soils compared to IN KCI exchangeable Al is good evidence of Al retention by organic matter. There was no evidence of exchangeable H in potentiometric and conductometric titration curves. The low values of Al in some soils were probably due to the fact that Al was not exchangeable by IN KCI, even though potentiometric and conductometric titrations detected it in those soils.

Lime Requirement

El Potrero and Alambre soils had the highest buffer capacities and

lime requirements (Fig. 4). These soils reached pH5.80 only after application of 10 metric tons of CaC03/ha. Guanipa soil had the lowest buffer capacity because with I metric ton/ha it reached pH 6.60 and with 2 metric tons/ha, pH 7.30. In general, lime requirements were as expected because




78









8

7



5


8


pH 5..






8

7 -

6 - 10

5

4 t
1 2 3 4 5 6 7 8 9 10 Metric ton/ha CaC03








Fig. 4. Effects of CaCO3 applications after 6 months incubation
period. (Guanipa 1=1; San Crist6bal=4; El Rocio=5; Bajo
Seco=6; Guataparo=7; Guanaguanare=8; Alambre=9; El Potrero=1O; and 3arinas=ll.





79




soils differed in amounts of organic matter and clay. From the results, one can calculate the amount of lime needed to adjust these soils to the desired pH. The amounts of CaC03 needed to neutralize exchangeable Al (pH 5.50) varied with soils and in El Potrero soil was approximately 10 metric tons/ha (Fig. 4).

Potassium Q/I Parameters

To investigate ionic selectivity of the negative surface charge in these soils, a study of adsorption-desorption of K related to Ca was conducted. The Beckett Q/I relationship was studied. The parameters found were not related to the position of soil in soil classification systems, but to the minerals predominating in the clay fractions.

The shapes of the curves resulting from Q/I relationships were the same for all soils studied (Fig. 5). Exchangeable K ranged from 0.02 to 0.05 meq/100g (Table 31). The highest exchangeable K values occurred in the less weathered soils. The presence of mica (Table 3) could explain the highest values of exchangeable K. The ARk values ranged from
e
0.0017 in Guanipa to 0.028 in Red Bay Al. It is assumed that those soils with low ARe values may have high K-fixation capacities. Red Bay B2t
e
showed a value of 0.01 meq/100g for AK' while Bajo Seco had 0.3 meq/100g. The high value of AK* for the Bajo Seco soil was probably a consequence of the presence of illite and other K fixing minerals. The highest values of PBCk were found in San Crist6bal and Guanaguanare soils. San Crist6bal had a higher PBCk than Guanaguanare perhaps because of the higher clay content. These two soils had the highest CEC values, the difference in PBCk values between them was probably caused by vermiculite in the San Crist6bal soil and illite in Guanaguanare.









0.7

0.6 0.5 0.4 0.3

AKo 0.2 meq/1 00g
0.1

0
010 20 30



0.2 0.3

0.4



ARKx103 (mole/1 iter) Fig. 5. Potassium Q/I relationship for El Rocro soil.











Table 31. Potassium Q/1 parameters and Gapon's selectivity coefficient for the soils examined.
Adsorbed at
Soil Exchangeable equilibrium k K Ca+Mg K Ca+Mg ARk AK Kx PBCK Kg
e
---------meq/100g--------- (mole/Z) --meq/00g-- (meq/100) (U/mole)
(T/mole) 2
Guanipa 1 0.08 1.80 0.13 1.92 0.0017 0.05 0.09 1.69 1.47 Guanipa 5A 0.04 0.60 0.07 0.97 0.0026 0.03 0.03 1.16 1.45 Guanipa 58 0.02 0.30 0.08 0.58 0.009 0.02 0.03 1.30 2.70 San Crist6bal 0.50 13.60 1.43 13.65 0.003 0.24 0.26 56.40 5.50 El Rocro 0.15 3.50 0.70 4.12 0.002 0.08 0.07 17.18 5.17 Bajo Seco 0.55 7.35 0.87 7.54 0.017 0.30 0.30 14.78 2.97 Guataparo 0.25 1.42 0.46 0.98 0.017 0.16 0.07 6.19 10.15 Guanaguanare 0.19 8.10 0.60 8.58 0.003 0.11 0.09 21.56 2.25 Alambre 0.15 3.40 0.50 3.90 0.0039 0.07 0.05 9.20 3.20 El Potrero 0.30 2.10 0.67 2.50 0.008 0.16 0.04 8.26 6.86 Barinas 0.15 1.51 0.46 2.07 0.005 0.09 0.06 8.20 5.20 Red Bay Al 0.26 6.72 0.50 7.16 0.025 0.17 0.06 14.20 1.58 Red Bay B2t 0.05 0.83 0.46 0.99 0.0012 0.01 0.03 12.40 12.89





82




The K Q/I parameters found in Red Bay tended to decrease with depth. The decrease of K with depth in this soil and in Guanipa 5 was probably caused by fixation by the increasing clay contents in deeper horizons. Decrease of ARk with depth seemed to indicate that available K depended
e
on K adsorbed on organic matter. The AK values indicated that there was little labile K in these soils. Values for PBCk tended to increase with depth in Guanipa 5 probably due to the effect of CaCl2 in increasing pH-dependent CEC by removing substances blocking the exchange surfaces, and increasing the reactive surface participating in the Q/1 exchange equilibria (Lee, 1973).

The values of K, Ca, and Mg shown in Table 31 were used to calculate the Gapon selectivity coefficient (KG) for those soils. The highest value of Ca + Mg adsorbed was shown in El Rocfo soil while the Guataparo soil showed a negative adsorption (desorption) of Ca + Mg. The latter indicated a high selectivity for K. All Gapon selectivity coefficient (KG) values were larger than unity indicating a high selectivity for K, especially Guataparo and Red Bay B2t.

These results indicate that the negatively charged sites for exchange in these soils in their "natural" conditions have higher selectivities for monovalent cations. It also suggests that the charge density is low, and some soil surfaces (San Crist6bal and Bajo Seco) have specific sites to fix ions with ionic radii with sizes similar to K.


Positive Surface Charge


In order to characterize positive surface charge, several studies of.P adsorption and factors affecting it were carried out. Langmuir and Freundlich isotherms were used to determine the behavior of P adsorption in these soils.




83




Phosphorus Adsorption

Correlation coefficients between the adsorbed P (x/m) and P concentration in soil solutions at equilibrium were significant for all soils which permitted Langmuir adsorption maxima to be calculated (Fig. 6). The data showed that at any particular final P concentration, P sorption was largest for Guanaguanare soil and least for Barinas soil (Table 32). However, the overall shape of the isotherms was remarkably similar. Each isotherm was characterized by a large change in the amount of P sorbed over a low P concentration range followed by a more gradual change in the amount of P sorbed with increasing P concentrations.

To facilitate comparison with other physical and chemical processes, K values for each soil were transformed into corresponding free energies for sorption using the relationship AG = -RTlnK. This relationship, which equates K with the equilibrium constant, assumes that the activity coefficients of occupied and unoccupied sites are the same (Ryden et al., 1977a).

Values of AG and b for each temperature obtained for each soil are given in Table 32. An interesting feature of the equilibrium data was the similarity of AG values for a particular temperature. The type and properties of the soils used were only reflected in the magnitude of the sorption maxima (b) obtained for the temperature used.

Data did not fit the Langmuir isotherm for all temperatures. There were deviations from linearity, and because of this, the Freundlich isotherm was used to fit data from electrolyte and temperature-effect experiments.

Temperature effects. The P adsorption isotherms were determined on all soils used in this study at 50, 250, and 400C. The extent of




84











0 50C 1.4 - O 250C a 400



1.2




1.0 x/m
(g/ml)
0.8




0.6




0.4


0 100 200 300 400 500 600 700 800 900 1000 C (pg/ml)











Fig. 6. Langmuir adsorption isotherms for San Crist6bal soil in nonsaturated systems at three temperatures.




















Table 32. Phosphorus sorption constants at different temperatures
from the Langmuir equation.

Soil AGs AG25 AG40"I b5 b25 b40t

------ Kcal/mole-------- -------- mmole/g -------San Crist6bal -2.96 -3.16 -3.37 59.69 75.02 84.89 El Rocro -3.07 -3.53 -3.43 56.59 71.68 82.71 Bajo Seco -2.95 -3.49 -3.85 53.76 56.59 63.25 Guataparo -2.85 -3.29 -3.46 50.56 49.63 49.63 Guanaguanare -3.14 -3.85 -4.29 62.04 80.65 97.74 Alambre -3.29 -3.66 -3.89 70.13 73.31 82.71 El Potrero -3.01 -3.39 -3.53 68.64 70.13 80.65 Barinas -2.89 -3.25 -2.99 44.19 46.75 84.89 Red Bay Al -2.94 -3.47 -3.47 48.88 50.40 57.69 Red Bay B2t -3.10 -3.22 -3.59 46.75 56.59 53.74


Tlhe AG is the free energy of sorption from the Langmuir sorption energy constant and b is the sorption maximum.




86




adsorption is shown by the data from the Freundlich equation in Table 33 for selected soils. These data showed that although the values of K changed with temperature, the value of (1/n) remained similar. The constant K varied widely among soils. The values of K ranged from 10.2 to 147.9 at 5*C, from 10.7 to 199.7 at 250C, and from 10.7 to 263.1 at 400C. Sandy soils showed the lowest K values. The values of the exponent of the Freundlich isotherm (I/n) were in all cases between 0.33 and 0.50. The linearity of the plots of equation (2) (Fig. 7) implies that a sufficiently wide spectrum of sites existed in the total surface to satisfy the condition that the bonding energy decreased exponentially with increasing site coverage (White and Taylor, 1977). The larger values of 1/n were associated with soils having relatively fewer kinds of adsorption sites. It is possible that low values of I/n could occur with an adsorbent containing more than one type of adsorption site

(Mukhtar, 1976).

An increase in temperature increased adsorption of phosphate in these soils suggesting that greater number of new sites would become available with increasing temperature (Fig. 7 and Table 33). This indicates a strengthening of the attractive forces between the phosphate ions and the soil. One would expect the standard enthalpy associated with the adsorption process to be endothermic. These new sites were presumably formed by breaking some bonds as pointed out by Muljadi et al. (1966). The heat of adsorption may thus be divided into two parts:

(a) that due to the effect of temperature on the equilibrium constant

(K) of the exchange process, and (b) that due to the irreversible increase in the number of adsorption sites, i.e., an endothermic process.




Full Text

PAGE 1

CHARGE CHARACTERISTICS OF SELECTED SOILS FROM TROPICAL AREAS By MELITON JOSE ADAMS -MELENDEZ A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1978

PAGE 2

To my Father, Mel iton Adams , who died May 1978

PAGE 3

ACKNOWLEDGEMENTS The author wishes to express his deep gratitude to Dr. W.G. Blue, chairman of the supervisory committee, for his guidance, assistance, and patience during the preparation of this dissertation. The author is pleased to extend his sincere acknowledgements to Dr. T.L. Yuan, cochairman, Dr. V. Berkheiser, Dr. O.C. Ruelke, and Dr. A.H. Krezdorn for their participation in the supervisory committee and constructive criticism of this manuscript. Special recognition is extended to Mr. Jorge Beltran for his invaluable help with the computer programs. Appreciation is extended to others who contributed their time and knowledge in making this endeavor possible: Dr. P.S.C. Rao, for his help and encouragement during this program; the personnel of instituto de EdafologTa, Facultad de AgronomTa (U.C.V.) for their help in some laboratory analyses, especially Mr. A. Prada, Mrs. L. Burguera; Mrs. I. Rojas, Mrs. M. Alegria, and Mr. W. Zupan; Mr. W.G. Pothier for his help with X-ray patterns interpretation; Mr. J. Miller, fellow graduate student, for those good discussions during this program; Mrs. Rossina Fernandez, for her patience in deciphering the handwriting and typing this manuscript; and Consejo CientTfico y HumanTstico de la Universidad Central de Venezuela for its financial support given during this graduate program. I I I

PAGE 4

Last, but not least, the author would like to thank his wife, Nieves, and children, Lara and Jose, for their support and understanding throughout the course of this study. i V

PAGE 5

TABLE OF CONTENTS Page ACKNOWLEDGEMENTS LIST OF TABLES vii LIST OF FIGURES i ABSTRACT i i INTRODUCTION 1 LITERATURE REVIEW 3 Soil Characteristics 3 Mineralogical Characteristics 3 Soil Acidity Sources 5 Lime Requirement 7 Potassium Quant i ty/ I ntens i ty Measures 7 Phosphorus Fractions 9 Phosphorus Adsorption 11 Factors affecting adsorption equilibrium 12 Thermodynamic parameters associated with adsorption process 16 Surface Charge 19 Double-layer Models 19 Gouy-Chapman double-layer model 20 Stern double-layer model 21 Point of Zero Net Charge 22 Effects of Specific Adsorption on the PZNC. . . 23 Cation Exchange Capacity 25 Effect of pH 25 Effect of P adsorption 26 Crop Response to Lime and P Applications 26 MATERIALS AND METHODS 28 Soil Physical, Chemical, and Mineralogical Properties 31 X-ray Diffraction Analysis 31 Fe-oxides Determination 32 Surface Area Determination 32 Soil Acidity 33 Potassium Q/ I Characteristics 3^ V

PAGE 6

Page Exchange Properties 35 Point of Zero Net Charge 36 Phosphorus Analysis 36 Calculations of Thermodynamic Parameters Associated with Adsorption Processes 37 Phosphorus Adsorption Isotherms 37 Phosphorus Release 37 Lime and Phosphorus Treatments 38 Laboratory and Greenhouse Experiments 39 Statistical Analyses 39 RESULTS AND DISCUSSION 'I Soil Characteristics with Emphasis on Surface Charge ^1 Clay Mineralogy ^5 Point of Zero Net Charge '6 Phosphorus and Sulfur Fractionation ^^6 Negative Surface Charge 73 Acidity 73 Lime Requirement 77 Potassium Q/l Parameters 79 Positive Surface Charge 82 Phosphorus Adsorption 83 Temperature effects 83 Ionic system effects 89 Thermodynamic parameters associated with adsorption process 92 Phosphorus Release • • Changes in Charge Surface 1^3 Lime and Phosphorus Effects 103 Effects of Change of Charge Characteristics on Crop Production ^'8 SUMMARY AND CONCLUSIONS 130 APPENDIX I DETAILED CALCULATIONS FOR CaCOg AND PHOSPHORUS TREATMENTS 132 APPENDIX II DETAILED CALCULATIONS OF THERMODYNAMIC PARAMETERS BY THE METHOD OF BIGGAR AND CHEUNG (1973) 135 APPENDIX III DETAILED CALCULATIONS OF SURFACE CHARGE 139 LITERATURE CITED 1^3 BIOGRAPHICAL SKETCH 152 vi

PAGE 7

LIST OF TABLES TABLES Page 1 PZNC of pure minerals found in soils 2k 2 Treatment combinations used in the laboratory and greenhouse experiment with the soils used ^0 3 Physical and chemical properties of soils examined. . . k2 h Exchange properties and free oxides of the soils examined k3 5 Phosphorus fractionation of the virgin soils examined . 6 Sulfur fractionation of some of the virgin soils examined 7 Effect of incubation temperature on phosphorus fractions in a typic Paleustult from Anzoategui State, Venezuela (0-30 cm) after applying several solution concentrations of phosphorus 50 8 Effect of incubation temperature on phosphorus fractions in a Grossarenic Psammentic Haplustox from Anzoategui ' State, Venezuela (All horizon) after applying several solution concentrations of phosphorus 51 9 Effect of incubation temperature on phosphorus fractions in a Grossarenic Psammentic Haplustox from Anzoategui State, Venezuela (A12 horizon) after applying several solution concentrations of phosphorus . . 52 10 Effect of incubation temperature on phosphorus fractions in an Ultic Tropudalf from Tachira State, Venezuela (0-30 cm) after applying several solution concentrations of phosphorus 5^ 11 Effect of incubation temperature on phosphorus fractions in a Ustic Quartz i psamment from Portuguesa State, Venezuela (0-30 cm) after applying several solution concentrations of phosphorus 55 V i i

PAGE 8

TABLES Page 12 Effect of incubation temperature on phosphorus fractions in an Oxic Tropudalf from Miranda State, Venezuela (O-3O cm) after applying several solution concentrations of phosphorus 56 13 Effect of incubation temperature on phosphorus fractions in a Tropeptic Haplustox from Carabobo State, Venezuela (0-30 cm) after applying several solution concentrations of phosphorus 57 14 Effect of incubation temperature on phosphorus fractions in an Udorthentic Pellustert from Portuguese State, Venezuela (0-30 cm) after applying several solution concentrations of phosphorus 58 15 Effect of incubation temperature on phosphorus fractions in a Fluaquentic Humitropept from Portuguesa State, Venezuela (0-30 cm) after applying several solution concentrations of phosphorus 59 16 Effect of incubation temperature on phosphorus fractions in a Haplargid from Lara State, Venezuela (O-3O cm) after applying several solution concentrations of phosphorus ^0 17 Effect of incubation temperature on phosphorus fractions in an Oxic Haplustalf from Barinas State, Venezulea (030 cm) after applying several solution concentrations of phosphorus ^1 18 Effect of incubation temperature on phosphorus fractions in a Rhodic Paleudult from Florida, U.S.A. (Al horizon) after applying several solution concentrations of phosphorus 19 Effect of incubation temperature on phosphorus fractions in a Rhodic Paleudult from Florida, U.S.A. (B2t horizon) after applying several solution concentrations of phosphorus 63 20 Values of pH2P0it and pH at equilibrium with different ionic systems after P adsorption for a typic Paleustult from Anzoategui State, Venezuela 6A 21 Values of pH2P0tj and pH at equilibrium with different ionic systems after P adsorption for an Ultic Tropudalf fromTachira State, Venezuela 65 22 Values of PH2PO11 and pH at equilibrium with different ionic systems after P adsorption for an Ustic Quartzipsamment from Portuguesa State, Venezuel a 66 V i i i

PAGE 9

-4 TABLES Page 23 Values of pH2P0i+ and pH at equilibirum with different ionic systems after P adsorption for an Oxic Tropudalf from Miranda State, Venezuela 67 2k Values of pH2P0ij and pH at equilibrium with different ionic systems after P adsorption for an Udorthentic Pellustert from Portuguesa State, Venezuela 68 25 Values of pH2P0t4 and pH at equilibrium with different ^ ionic systems after P adsorption for a Tropeptic Haplustox from Carabobo State, Venezuela 69 26 Values of pH2P0t4 and pH at equilibrium with different ionic systems after P adsorption for a Fluaquentic Humitropept from Portuguesa State, Venezuela 70 27 Values of PH2PO11 and pH at equilibrium with different ionic system after P adsorption for a Haplargid from Lara State, Venezuela 71 28 Values of PH2PO4 and pH at equilibrium with different ionic systems after P adsorption for an Oxic Haplustalf from Barinas State, Venezuela 72 29 Soil reaction with selected electrolyte and calculated ApH value for the soils examined 75 30 Acidity from potent iometr ic titrations at pH 8.00 and conductometr ic titrations with NaOH, Ba(0H)2, and Na2Bi^07 for the Venezuelan soils examined 76 31 Potassium Q/i parameters and Gapon's selectivity coefficient for the soils examined 8I 32 Phosphorus sorption constants at different temperatures from the Langmuir equation 85 33 Freundl ich equations for phosphorus adsorption at different temperatures for the soils examined 87 3A Freundl ich equations for phosphorus adsorption at different ionic systems for the Venezuelan soils examined 91 35 Equations of the relationship between solution phosphorus concentration and reciprocal of time, and the K and AG values calculated from the concentration at equilibrium for the soils examined 9^ i X

PAGE 10

TABLES Page 36 Thermodynamic parameters calculated using Freundlich equations for phosphorus adsorption reactions for the soils examined 96 37 Values of K° , AG°, AH°, and AS° associated with the adsorption of phosphorus for the selected soils calculated by the Biggar and Cheung method (1973) • 97 38 Relationship between the accumulative release P and the number of washings and linear correlation coefficients for the Venezuelan soils with different ionic systems ^0' 39 Effect of treatments on electrochemical properties measured before cropping for the Guanipa 1 soil . . . 104 kO Effect of treatments on electrochemical properties measured before cropping for the Guanipa 5 soil . . . 105 k] Effect of treatments on electrochemical properties measured before cropping for the El Potrero soil. . . IO6 42 pH and point of zero net charge for CaCOs and phosphorus treatments in Guanipa 1 after cropping .... IO8 43 pH and point of zero net charge for CaCOs and phosphorus treatments in Guanipa 5 after cropping .... 109 kk pH and point of zero net charge for CaCOs and phosphorus treatments in El Potrero after cropping. ... 110 kS Electrochemical potentials of CaCOa and phosphorus treatments in soil Guanipa 1 after cropping Ill 46 Electrochemical potentials of CaC03 and phosphorus treatments in soil Guanipa 5 after cropping 112 47 Electrochemical potentials of CaCOs and phosphorus treatments in soil El Potrero after cropping II3 48 Charge distribution at the surface for CaCOs and phosphorus treatments in Guanipa 1 after cropping . . 114 49 Charge distribution at the surface for CaCOs and phosphorus treatments in Guanipa 5A after cropping. . 115 50 Charge distribution at the surface for CaCOs and phosphorus treatments in El Potrero after cropping. . II6 X

PAGE 11

TABLES Page 51 Dry matter yields of sorghum and cation exchange capacity after cropping for soil Guanipa 1 treated with CaC03 and phosphorus 119 52 Dry matter yields of sorghum and cation exchance capacity after cropping for soil Guanipa 5 treated with CaCOs and phosphorus 120 53 Dry matter yields of sorghum and cation exchance capacity after cropping for soil El Potrero treated with CaC03 and phosphorus 121 5A Effect of cropping on CEC values for Guanipa 5 soil . . 122 55 Effect of cropping on CEC values for Guanipa 1 soil . . 123 56 Effect of cropping on CEC values for El Protrero soil . 12^ 57 Change in CEC after cropping for the soils selected . . 125 58 Effect of soil, CaCOs, P, and interaction CaCOaxP on the variables measured 128 59 Prediction equation for pH, iio, DM, and CEC to fit a surface response for the three soils selected 129 xi

PAGE 12

LIST OF FIGURES FIGURES Page 1 Relative location of Venezuela in South America 29 2 Relative location of the soil sampling sites in Venezuela 30 3 Phosphorus fractions as affected by P concentration after an 11-day equilibrium period of ZB^C 53 k Effects of CaC03 applications after 6 months incubation period 78 5 Potassium Q/ 1 relationship for El RocTo soil 80 6 Langmuir adsorption isotherms for San Cristobal soil in nonsaturated systems at three temperatures . . . 7 Freundl ich adsorption isotherms for P by Guanaguanare , San Cristobal, and Red Bay B2t in nonsaturated systems at three temperatures 88 8 Freundl ich adsorption isotherms for P by Guataparo and San Cristobal in different ionic systems at 25°C .... 90 9 Relationship between P solution concentration and the reciprocal of time for Guataparo, San Cristobal, and Red Bay B2t soils 95 10 Release of phosphorus from Guanipa 1, San Cristobal, and El RocTo soils 102 xi i

PAGE 13

Abstract of Dissertation Presented to the Graduate Council of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy CHARGE CHARACTERISTICS OF SELECTED SOILS FROM TROPICAL AREAS By Meliton Jose AdamsMe lendez December 1978 Chairman: Dr. W.G. Blue Major Department: Soil Science Ten soils from Venezuela and one from Florida, U.S.A., were investigated by chemical, physical, and mineralogical techniques to elucidate surface reactivity. Also, two Oxisols and one Alfisol were selected to determine the effect of lime and P levels on surface charge. A sorghum crop was planted under greenhouse conditions to investigate the change of surface charge in each limeand P-treated soil after cropping. Stern and Gouy-Chapman models for diffuse double layer together with the point of zero net charge (PZNC) and the surface area were used to calculate the charge distribution and the cation exchange capacity (CEC) of the soils for each lime and P treatment. The following parameters were determined: (a) acidity sources; (b) quantity-intensity K relationships; (c) P fractionation in virgin soils and after P adsorption; (d) P-adsorption isotherms at different temperatures and with different ionic systems; (e) PZNC of virgin soils before and after lime and P treatments; (f) pH before and after cropping; (g) electrical conductivity before cropping; and (h) sorghum dry-matter yields. X i i i

PAGE 14

The following parameters were calculated: (a) thermodynamic parameters associated with the adsorption process (AG°, AH°, and AS°) and (b) charge distribution in double layer and CEC after lime and P treatments. +3 The results showed that exchangeable Al and dissociation of OH groups from organic matter and silicate are responsible for the acidity in these soils. The soils had low available K, but very high selectivity for K. There was an increase in P adsorption with temperature in all soils. Lime increased P adsorption in two Alfisols and one Oxisol. There was an increase in both Al-P and Fe-P with temperature and P concentration applied. Phosphorus had a highly significant effect on PZNC decrease while P and lime increased CEC due to the combined decrease of PZNC and increased pH. Salt content as indicated by increasing electrical conductivity may have depressed the dry matter yield of sorghum in lime and P treatments. The negative AG° values and positive values of AH° and AS° showed that P adsorption in these soils is a spontaneous endothermic reaction that occurs with an increase of entropy resulting from displacement of water molecules by phosphate ions. xiv

PAGE 15

INTRODUCTION Physico-chemical properties of a soil are determined by the surface characteristics of its organic and inorganic colloidal components. The quantity and type of those components are greatly affected by various soil-forming environmental conditions. Because of the great variations in climate, topography, etc., soils in Venezuela vary from relatively recently formed Entisols and Inceptisols to highly weathered Ultisols and Oxisols. Like other tropical regions, highly weathered soils dominate in Venezuela. These soils are mostly acidic and P deficient. Finding ways of making tropical soils more productive, by correcting soil acidity and making conditions favorable for nutrient retention and uptake by crops are the concern of most soil research in tropical countries. A study of surface phenomena such as proton release, cation selectivity, P adsorption, and the effect of liming and P application on surface charge can help our understanding of the behavior of highly weathered tropical soils with concomitant implications to their management. Of particular significance is the efficient use of P in soils because farmers wish to manage their chemical investment in the most economical manner and this element is one of the most expensive. The phenomenon of P adsorption directly influences the magnitude of available P for crops in highly weathered soils. The intent and objectivesof the study presented here i n vjere to elucidate (a) sources of acidity, (b) ion selectivity, (c) adsorption 1

PAGE 16

2 and release of P, and (d) effects of CaC0 3 and P treatments on surface charge (CEC) as related to crop production in selected soils from Venezuel a.

PAGE 17

LITERATURE REVIEW Evaluation of soils in relation to classification, agronomic practices, and other purposes is aided by the identification, characterization, and understanding of their properties and behaviors. Soil surface reactions involving source of protons, ionic selectivity, specific adsorption, and responses to chemical treatments have great importance in the understanding of electrochemical behaviors of highly weathered soi 1 s. Soil Characteristics Mineralog ical Characteristics Physical and chemical properties of any soil are controlled to a very large degree by the minerals in the soil and especially by those constituting the clay fraction. Identification, characterization, and an understanding of properties of the different minerals aid in the evaluation of soils in relation to classification, agronomic practices, and other purposes. Positive identification of mineral species and quantitative estimation of their properties in the soil clay fraction usually require the application of several qualitative and quantitative analyses. X-rays diffraction, GEO, and surface area are among the most common analytical measurements used by soil scientists (Jackson, 1956; Whittig, 1965). Some soil constituents interfere in the soil mi nera 1 og i ca 1 analysis. In these cases, a chemical pretreatment of the soil is necesary. 3

PAGE 18

Pretreatments are designed to have a minimum effect on fractions of the soil other than those for which the treatments are intended. The pretreatments that are most used in soil mineralogical analyses are removal of salts, carbonates, Fe, organic matter and amorphous material, and part^icle-size fractionation (Jackson, 1956). Soluble salts and carbonates affect organic matter removal, soil sample dispersion, and also make it impossible to saturate the exchange complex with a specific cation. Removal of soluble salts and carbonates simplifies X-ray diffraction analysis. Carbonates reduce degree of orientation of particles in slide preparation (Jackson, 1956). Organic and Fe-oxide cementing agents are frequently removed by chemical treatment prior to clay separation to obtain dispersion for effective particle-size separation, enhanced parallel orientation of clays, and decreased background and attenuation of X-rays (Zelazny and Carlisle, 1971). Mehra and Jackson (i960) found that citrate system was the most effective in removal of free Fe-oxides from latosolic soils and the least destructive of Fe-silicate clays as indicated by least loss in CEC after Fe-oxide removal. On the other hand, organic matter may cause a high background and prevent parallel orientation in the preparation of the slides for X-ray diffraction studies. Traditionally, organic matter has been oxidized with H202. In order to prevent a strong acid condition from developing which could degrade crystalline minerals, soils have to be buffered around pH 5 in the presence of H2O2. Mineralogical analyses are normally performed on specific size fractions of the soil such as 2 to 0.2 microns. For mineralogical study, concentration of the clay fraction is necessary (Jackson, 1956).

PAGE 19

5 Identification and quantitative analysis of the minerals are possible from an X-ray diffraction pattern because each diffraction peak depends upon the atomic arrangement in the crystal and is proportional to the concentration of the mineral in a mixture (Jackson, 1956). The choice of cations for saturation of the exchange sites along with solvation becomes important especially since spacings of expanding clay minerals are made larger with Mg and Ca ions than with K or Na ions (Jackson, 1956). Soil Acidity Sources In most of the soil classification studies and in soil characterization for agricultural purposes, the pH value has great importance. Deficiencies of P, Ca, Mg and Mo, and Al and Mn toxicities seem to be the most important effects of soil acidity on plant growth (Lora and Riveros, 1971). Soil acidity has been defined as a soil system's proton yielding capacity in going from a given state to a reference state. Soil acidity has long been recognized to involve exchangeable Al (Jackson, 1963). Acid soil systems that are Al-saturated have weak acid characteristics while those that are H-saturated have strong acid character istcs (Coulter, 1969). But the latter is an unstable state because H ions attack the clay structure and Al is released as shown by Davis et al. (1962) in their study of autotransformat ion of H-bentonite to Albenton i te. On the other hand, Yuan (I963) studied H and Al ions, and pH relationships in pure chemical solutions and compared the results with those obtained from several soil systems. He pointed out that pH determined with the usual methods can be considered as a measure of the H ion

PAGE 20

6 activity !n the system when the liquid and solid phases are in equilibrium. He also indicated that soils with the same H ion activity and different amounts of unhydrolyzed Al ions may have the same pH. Since Bradfield (19^0 pointed out the nature of potent iometr ic and conductometr ic titrations in soil systems, these techniques have been very useful in studying the nature of soil acidity. Several authors have studied the factors that affect the results obtained with these techniques. Low (1955) pointed out that potent iometr ic and conductometr ic titrations could be used to determine the amount of Al on an acid bentonite. On the other hand, Marshall i]S(>h) shov;ed initial inflections due to H titration, followed by regions of buffering attributable to Al . These inflections and their corresponding indication of soil acidity sources can be verified by bases that titrate specifically H or Al . Yuan (1959) used fluoride to eliminate the interference of Al ions in the H titration. It was found that Na2B407 t i t rat ions gave the permanent negative charge on monrtmori 1 lonite particles both when this clay contained exchangeable Al and when it did not (Shainberg and Dawson, I967). They also pointed out that the differences between NaOH and Ha^^O; titrations can be made equivalent to H ions formed from proton release by hydroxyl groups, and by selecting an appropriate end point, it can be made approximately equal to this quantity plus the aluminate equivalent of exchangeable Al . They also determined the effect of salt content on the slope of the first segment of the conductometr ic curves. Chao and Harwcrd (I962) found that the features of titration curves were independent of the kind of clay and dependent on the nature of the acid clay. Rich (1970) pointed out that the differences in results of

PAGE 21

7 both potent iometric and conductometr ic titrations, and of extractions by neutral salts of acidity from AICI3treated cation exchangers were due to the fact that neutral salts, especially K salts, tend to induce hydrolysis of Al"^^ ions on exchange sites. Dewan and Rich (1970) found that consistent results are obtained when the aluminate reaction is considered for titrations with strong bases. Lime Requirement The several methods that have been proposed to determine lime requirement of acid soils have advantages and disadvantages depending on type of soil and intended use of the data. However, it is now generally accepted that the most reliable index for predicting lime requirement of soils of tropical regions is the quantity of exchangeable Al (Reeve and Sumner, 1970a; Kamprath, 1970; Evans and Kamprath, 1970; Amedee and Peech, 1976). However, the incubation method still appears to be as good as any, if the desired pH level is not as high as complete netural izat ion . Spain et al . (1975) pointed out that liming rates for soils of the tropics often have been overestimated even when based on Al suppression. Pearson (1975) concluded that relatively low rates of lime in the humid tropics are usually adequate for maximum crop production. Potassium Quant i ty/ intens i ty Measures Because of the importance of K in plant nutrition, many investigators have attempted to find a rapid and reliable method for evaluation of K availability in soils. Potassium availability depends on the exchangeable form of this element in the double layer surrounding the exchange complex. The chemical potential of K in the double layer controls the activity of K in solution (Beckett, ]SGka) . Its chemical potential cannot be measured experimentally and is not directly

PAGE 22

8 proportional to the activity of K in the soil solution (San Valentin et al 1972), Beckett (l96Ab) proposed the activity ratio aK/(aCa + Mg) ^ as a measure of the ruling chemical potential of labile K in a soil. The acti vity ratio, which is now stated as AR*^, has the same value for all solutions in equilibrium with a given soil, and is independent of the ionic strength of the soil solution or the proportion of Ca to Mg (Beckett and Nafady, 1967b). This ratio is related to availability of K and is termed intensity (l) (Beckett, 196'tb). The ability of a soil to maintain the activity ratio against depletion by plant roots is governed by the character of the pool of labile K, the rate of release of fixed K, and the diffusion and transport of K ion in the solution (Beckett, 196'*c). The relationship between K availability or intensity (l) and the amount of K present (Q) in soils is termed the Q/l relationship (Beckett, 1964a). The Q/l relationship can be determined by equilibrating a sample of soil with solutions containing a constant amount of CaCl2 and increasing amounts of KCl . The soil either gains or losses K in order to achieve the characteristic AR*^ of that soil, or remains unchanged if its AR*^ is the same as that of the equilibrating solution. The characteristic Q/l relationship is formed by plotting the gain or losses of k / t K (AK) against the AR of the equilibrium solution (San Valentin et al., 1972; Beckett and Nafady, 1967a, 1967b). San Valentin et al. (1972) pointed out that three parameters are derived from a Q/l plot which can characterize the soil K status. When AK=0, the AR is a measure of the k k K availability at equilibrium (termed AR^) . When AR =0, the value of K is a measure of the labile or exchangeable K in soils (termed AK°) . The slope of the linear portion of the curve gives the Potential Buffering Capacity of the soil for K (termed PBC ) and is proportional to CEC.

PAGE 23

9 The curved portion of the Q/l plot Is also an indication of the number of specific sites for K. The number of specific sites for K (termed K^) is determined from the difference between the intercept of the curved and linear portions of Q/l plot at AR =0. Forms of the Q/l relationship remain almost unchanged either by large K additions, a high degree of K fixation, or by depletion of labile and some fixed soil K. However, acidic M CaClz solutions may influence the form of Q/l relationship by removing substances blocking portions of the exchange surfaces of the soil clay, increasing the extent of the exchange surface participating in the Q/l exchange equilibria, inducing the collapse of the expanded portion of soil clay, and by decreasing the extent of those exchange surfaces (Lee, 1973). Moss and Beckett (1970 reviewed the sources of error in the determination of soil K activity ratios by the Q/l procedure. They pointed out that errors may arise in sampling, preparation and storage of soil samples, and recommended analysis of samples as soon as possible after sampling, with only gentle sieving. They also found that exchange between cations and soluble salts or microbial ly dissolved carbonates during the determination affect the results obtained by this procedure. Phosphorus Fractions Knowledge of specific chemical forms of phosphates is important in understanding the chemistry of soil P and also has importance in soil genesis and soil fertility. In acid soils, P forms insoluble compounds with Fe and Al while under alkaline conditions, Ca compounds are formed. The particular compounds formed depend on soil properties. Furthermore, the solubility and stability of these compounds affect the availability of P under field conditions.

PAGE 24

10 Juo and Ellis (1968) showed that when soluble P is applied to acid soils, or when Ca-P is dissolved during the process of chemical weathering, the soluble P precipitates rapidly to form colloidal Al-P and Fe-P. The P in these colloidal forms is relatively available to plants initially, but crystallization tends to form hydrated compounds such as variscite and strengite that are less available. Furthermore, Fe-P crystallizes at a more rapid rate than Al-P and this is why Al-P seems to be more available to plants than Fe-P. Singh et al. (1966) studying the availability of forms of residual P in Davidson clay loam found that neutral soils, pH 6.8, favored the availability of Al-P to first-cutting alfalfa and Fe-P to first-, second-, and third-cutting alfalfa. They also pointed out that Al and Fe activities at this pH are low, and as a result, P from Aland Fe-P becomes more available to plants. Lindsay and Moreno (i960) pointed out that in soil below pH 7, the H2P0it constitutes the larger fraction of the total P in solution and the logarithmic acitivfty function permits a linear relationship with pH. They developed a phase diagram that permits prediction of the HaPOit concentration as controlled by variscite, strengite, f 1 uorapat i te , hydroxyapat i te, octocalcium phosphate, and dicalcium phosphate solubilities. In that diagram, increase in pH tends to decrease precipitation of Al and Fe compounds, but increase precipitation of Ca forms. Yuan et al. (I96O) studying newly fixed P forms in acid sandy soils found that the ratio Al-P to Fe-P increased with increasing rates of applied P. They also found that increasing soil drying temperature decreased the percentage of P in Al-P form, but increased that in Fe-P form. Fractionation of inorganic soil phosphates is widely used in soil fertility and genesis studies. These procedures are based on

PAGE 25

n differential solubilities of inorganic phosphates in various extractants. The most used of these methods is that of Chang and Jackson (1957) modified by Petersen and Corey (1969); these procedures differ only in the sequence of extraction of the different P forms, pH of the NHi^F solution used to extract Al-P which reduces interference by. Fe, and other details to increase the efficiency of analysis of a large number of samples. Phosphorus Adsorption Adsorption is the process of concentration of liquid or gaseous material on the surface of a solid. Phosphorus adsorption studies may be divided into those concerned with investigating the nature and mechanism of the sorption system, and those concerned with its quantitative measurement (Larsen, I967). Measurement of the size of the P adsorption system can be made by fitting adsorption data to a previously described adsorption isotherm. The Langmuir isotherm has often been used for this purpose. The Langmuir isotherm has constants which, at least when applied to the adsorption of gases on solids, have quantitative meaning. The linear form of the Langmuir isotherm is c c (1) x/m Kb where x mg P adsorbed/1 OOg soil m b the adsorption maximum, mg/lOOg soil c equilibrium P concentration, moles/liter k a constant related to the energy of bonding. The Freundlich isotherm has the forrr b (2) X = ac

PAGE 26

12 or the 1 inear form log X = log a + b log c, (3) where x is the amount of P adsorbed per unit weight of soil, c Is the concentration of P in solution, and a and b are the constants that vary between soils. Larsen (196?) claimed that this isotherm is purely empirical, and the constants have no physical meaning. However, Adamson (1976) pointed out that the Freundl ich equation, unlike the Langmuir, does not become linear at low P concentration but remains convex to the concentration axis. Also, it does not show a saturation or limiting value, but the intercept of log x vs log c gives a measure of the adsorbent capacity and the slope of adsorption intensity. Factors affecting adsorption equilibrium . Equilibrium adsorption of P studies on a variety of soils and minerals have shown that several factors have consistent influence on the final position of the equilibrium state. Among these factors are (a) concentration of P in the solution, (b) temperature, (c) ionic strength, (d) the pH, and (e) specific cation interactions with the surface and adsorbed species. Noting the apparent i ndependence of the adsorption isotherm from the overall adsorption capacity of the adsorbent material, Muljadi et al. (1966) proposed employing an empirical splitting of the isotherms into three regions. The regions, designated I, II, and III, were found to correspond to equilibrium solution concentration ranges which may be described roughly as low, medium, and high, respectively. Regions I and II fit a Langmuir adsorption isotherm while region III exhibited a linear dependence on solution concentration. They also found that adsorption in regions I and II increased irreversibly with increasing temperature, while the increase in region III was reversible with respect to temperature. Ryden

PAGE 27

13 et al. (1977) later employed this curve splitting technique and reported on the effects of pH, ionic strength, specific cation, and cation adsorption. These authors noted that Ca electrolyte solutions were more effective in facilitating adsorption than Na solutions of the same ionic strength; in conjunct ion wi th this finding, the authors reported observing a specific adsorption of Ca which was not reversible to IN KCl . Sodium adsorption was measured together with phosphate adsorption and it was observed that the adsorption of Na corresponded to that of phosphate in regions I and III, but that no Na was adsorbed in region II. They interpreted those results in terms of the adsorption mechanisms in the three regions. Region I was thought to represent the replacement of adsorbed water by phosphate; region II, the displacement of surface hydroxyls; and region III, the association of phosphate ligand with the surface metal which was potentially dependent and less energetic than a specific electrovalent coordination with the metal. The concentration of phosphate in solution was thought to affect both the amount of P adsorbed and the nature of the adsorption process. Temperature effects on equilibrium adsorption isotherms have been reported for soils and minerals. Muljadi et al. (I966) measured phosphate adsorption on K-kaolinite, gibbsite, and pseudoboehmi te at 2, 20, and kO°C, respectively. They reported a marked increase in adsorption with temperature for the three equilibrium concentration regions. Region I of K-kaolinite (pH 5) reached a maximum at about 20°C; regions II and III showed increases to ^0°C for all adsorbents. The increase in adsorption in region I was reported to be irreversible with respect

PAGE 28

to temperature; region II was slightly reversible for K-kaolinite but irreversible for gibbsite and pseudoboehmi te; and region III was found to be completely reversible. Barrow and Shaw (1975a) measured the adsorption of phosphate on soils as a function of time and temperature. They observed an increase in adsorption with temperature from h to A2°C for contact times up to 100 days. However, they reported that adsorption exhibited a time-temperature dependence which indicated that increasing temperature merely speeded up the adsorption process, rather than shifting the equilibrium position. Griffin and Jurinak (197^), studying the kinetics of interaction of phosphate with calcite at 0, 11, 23, and ^0°C, demonstrated that temperature has a much more pronounced effect on the adsorption rate constant than on the desorption rate constant. These resul ts ' ind i cated that the adsorption process is favored by an increase in temperature and that equilibrium adsorption should increase over the temperature range studied. The effects of the ionic strength and of selected cations on the adsorption of phosphate have been reported for a group of Hawaiian soils (Ryden et al., 1977a, b) and gibbsite (Helyar et al., 1976a). Ryden and coworkers, observing a linear relationship in plots of the solution phosphate concentration against the reciprocal of time for periods greater than 72 hours, were able to estimate the equilibrium concentration for a series of solutions of differing ionic strengths. The plots converged at equilibrium for low phosphate concentrations, indicating that the ionic strength effect was kinetic. At moderate to high solution concentrations, the ionic-strength effect was found to be absolute and the authors concluded that the adsorption at high

PAGE 29

15 solution concentrations was potentially determined. Adsorption from solutions with a CaCl2supporting electrolyte was greater than from NaCl solutions; however, the pH of the Ca solutions was lower. The authors, noting that more Ca was adsorbed than Na and that the adsorbed Ca was not completely displaced by IN KCl , proposed that Ca was specifically adsorbed, thus facilitating the adsorption of phosphate. Helyar et al. (1976a) studied the effect of selected cations on the adsorption of phosphate on gibbsite. They maintained a constant pH by controlling the partial pressure of CO2 in the solid-solution mixture. Their results for Na, K, and Mg were similar and the observed ionic strength effect was small. Adsorption in the presence of Ca, however, increased out of all proportion to the change in ionic strength for equilibrium concentrations above 1.0 pM, but converged with the adsorption noted for other cations at lower equilibrium concentrations. Helyar et al. (I976b) proposed that certain cations could facilitate adsorption by forming a stabilizing bridge between two adsorbed phosphate ligands. The ability of the cation to form such a bridge would depend upon its fit into the space between the oxygen of two adjacent phosphates and its ability to coordinate with two oxy-ligands. The authors estimated an oxygen to oxygen distance of approximately 1.0 % for adsorption on gibbsite; cations of approximately this crystal radius and divalent charge would best accommodate the space if they could form coordinate covalent bonds. They selected a number of cations with a range of radii for both mono and divalent species. The adsorption of phosphate was enhanced markedly by Ca, Cd, and Sr which have radii of 0.99, 0.99, and 1.13 8, respectively. No such enhancement was noted for Mg, Zn, Na, or K which have radii of 0.66, 0.7^, 0.95, and 1.37 8,

PAGE 30

16 respectively. The stabilizing effect of a bridging cation was thought to result from a reduction of the mutual repulsion of the oxygen or hydroxyl groups on adjacent phosphate ligands. Thermodynamic parameters associated with adsorption process . Thermodynamic parameters calculated from adsorption measurements have been found useful in elucidating the mechanisms involved (Biggar and Cheung, 1973). Biggar and Cheung (1973) calculated the thermodynamic parameters associated with the adsorption process from the variation of the thermodynamic equilibrium constant, Ko, with changes in temperature. Ko for the adsorption reaction is defined as: ae ye Ce where as = activity of the adsorbed solute, ae = activity of the solute in the equilibrium solution, Cs =yg of solute adsorbed/ml of solvent in contact with the adsorbent surface, Ce = yg of solute/ml of solvent in the equilibrium solution, ys = activity coefficient of the adsorbed solute, and ye = activity coefficient of the solute in the equilibrium solution. Cs is calculated according to the following equation: Cs = ^(Pi/^^^ (5) S A2 Na (x/m) " M2 X 10& where pj = density of solvent (g/ml); and M2 = molecular weights (g/mole) of the solvent and the solute, respectively, and A2 = cross-sectional areas (cm^/molecul e) of the solvent molecule and the solute molecule, respectively; Na = Avogadro's number (6.02 x 10^3 molecules/mole); S = surface area of the adsorbent (cm^/g) ; and x/m = specific adsorption (pg/g) .

PAGE 31

17 The cross-sectional areas (in cm^) of the solvent and solute molecules were estimated from the following equation: ^'^ (6) where M and p are, respectively, the molecular weight (g/mole) and the density (g/ml) of the solvent or solute. Biggar and Cheung (1973) assumed that Na (x/m) ^ M2 X 1 06 and reduced equation (5) to Cs = (P^/'^-^ ^ (8) Na (x/m) and calculated Cs from this equation. They also assumed that concentration of the solute in the solution approaches zero, and ys and ye approach unity. The authors obtained the values of Ko by plotting In (Cs/Ce) vs Cs and extrapolating to zero Cs. The standard free energy, AG°, was calculated from AG° = RT In Ko. (9) The standard enthal py, H" , is obtained from the integrated form of the van't Hoff equation, , Ko2 AH° , 1 1 X The standard entropy, AS°, is obtained from AS° = (AH° AG°)/T. (11) (7)

PAGE 32

18 BIggar and Cheung (1973), noting the linear nature of In Ko vs 1/T, concluded that the mechanics of adsorpt ion were not changed as temperature was changed and that the amount of adsorption was changed because the supply of thermal energy was different. Rydens et al. (1977 ) used the expression: G = RT InK (12) where K is the Langmuir equilibrium constant obtained for a single temperature and has units of reciprocal concentration. Assuming that K exhibits a temperature dependency, the value of the estimated AG will depend upon the measurement temperature of K, and the concentration units used. Some parameters related to the adsorbent capacity and the intensity of adsorption could be calculated using the Freundl ich equation. In a plot of In X vs log c where X = P adsorbed (yg/g), and c the equilibrium concentration (moles/liter), the intercept gives InK that can be. used to determine a AG related to the adsorption capacity energy (Adamson, 1976). Using Ryden et al. (1977a, b) plots of the solution phosphate concentration vs the reciprocal of time to determine the equilibrium concentration by extrapolat ing at infinite time, an equilibrium constant can be calculated: [Soil-P] (^:,^ = lPOj[SoilJ where [soil-P] is yg of P adsorbed per g of soil,[POit] is equilibrium concent rat ion in yg/ml , and [soi 1 ] is g of soi 1 /I iter of sol ut ion. Applying (12), a AG for adsorption can be calculated.

PAGE 33

19 Surface Charge Double-layer Models Soil electrochemical behavior is determined by the quantity, type, and interactions of colloidal components. Colloids with a constant surface potential have a reversible double layer with the surface charge determined by the nature of the potent ial -determining ions (PDl) adsorbed on the surface. The electrical double layer at the solid-solution interface is constituted by the surface charge and its countercharge in solution. Solid oxide particles immersed in aqueous electrolyte solution develop surface electrical charges by adsorption or desorption of potential-determining ions (Atkinson et al., 1967). For oxides, the surface charge and surface potential are functions of solution pH, and and OH are referred to as potent ial -determining ions (Breeuwsma, 1973). The electrochemical behaviors of Oxisols and Alfisols were found to be similar to those exhibited by many metallic oxides (Van Raij and Peech, 1972). The constant-potential surfaces are associated with oxides and hydroxides of metals and edges of clay minerals as well as with those clay minerals with minimal ion substitution in the lattice (Tinsley, 197'*). These colloids include crystalline and noncrystalline oxides and hydrous oxides of Al , Fe, Ti, Mn, and Si; kaolinite, halloysite; allophane; quartz; and organic matter. The surface potential remains constant, and its magnitude is not affected by the presence of indifferent electrolytes (Stumm and Morgan, 1970). For constant-potential surfaces, the surface charge is directly related to the concentration of potent ia 1 -determin ing ions in the solution. The surface charge density of minerals of this type varies with pH and salt concentration (Keng and Uehara, 197^). From known surface

PAGE 34

20 areas and Point of Zero Net Charge (PZNC) determination of Oxisols, the values for net electrical charge calculated by the application of Gouy-Chapman and Stern models of the double-layer theory were found to be in good agreement with experimental results (Van Raij and Peech, 1972; Adams, 1976) . Gouy-Chapman double-layer model . The Gouy-Chapman model relates the charge density of the surface with the surface potential as fol lows: a = (2neKT/TT)^ Sinh (Zei{;o/2KT) where a = the surface charge density n = the concentration of the equilibrium solutions in number of ions per cm^ e = the dielectric constant of the medium e = the charge of an electron K = the Bol tzmann constant T = the absolute temperature Z = the valence of the ion }po = the surface potential. In constant-potential systems, change in concentration will result in a change in surface-charge density (a ). For some constantsurface potential colloids, t(jo can be held constant by maintaining the pH constant (Keng and Uehara, 197'+)The Nernst equation can be used to relate the ion concentration to the surface potential as follows:

PAGE 35

21 where R = the gas constant F = the Faraday constant = the hydrogen ion concentration H"*^ = the hydrogen ion concentration at which i|;o=o. o At 25°C the Nernst equation can be reduced to the following: if)0 = 59(PZNC-pH) (16) where PZNC is the pH at the point of zero net charge (pHo) . The Gouy-Chapman model for the double layer has limited quantitative application because of the assumption that ions in solution behave as point charges and can approach the surface without limitation (Tinsley, 197^). Stern double-layer model . In the Stern double-layer model, charge density can be described by the following equation (Van Raij and Peech, 1972): 0=01 + 02 where oj = Stern layer charge 02 = diffuse layer charge. The Stern layer charge can be given by _ N|Ze 01 = . . (17) 1 + (Nap/Mn) exp L J where Nj = available spots per cm^ for adsorption of ions Na = Avogadro's number M = molecular weight of the solvent p = solvent density i}/6 = electrical potential at the boundary between the Stern layer and the diffuse layer or the Stern potential

PAGE 36

22 $ = specific adsorption potential. The diffuse-layer cfiarge is given by the following equation: 02 = (2neKT/TT)^^ Sinh (Zei|^6/2KT) (I8) Since a linear drop in potential across the specific Stern layer is assumed, the surface charge can also be given by the Gauss equation for a molecular condensor: 6 = i^o-yliS) (19) where = average dielectric constant 6 = average thickness of double layer. These equations can be used to calculate the relative distribution between and 02 if values of N]^, , and 6 can be assumed (Adams, 1976). Van Raij and Peech (1972) claimed that the only soil properties needed to calculate the net charge are PZNC and surface area. Point of Zero Net Charge Concentration of potent ial -determi n i ng ions and the net surface charge are obviously pH dependent on constant-potential surfaces. There will be a pH at which the densities of positive and negative charges are equal and the surface possesses no net charge (Tinsley, 197^^; Keng and Uehara, 197't). The acidic and basic properties of a solid that influence the location of PZNC are functions of such variables as cation size and valence, the hydration state of the solid, and the geometrical arrangement of the ions. Parks (1965) pointed out that the broad range of isoelectric points in oxides and hydroxides was explained by differences in hydration state, purity, and cation radii. The PZNC of some pure minerals found in soils can be observed in Table 1 (Adams, 1976). It can be seen that dehydration, dehydroxyl at ion ,

PAGE 37

23 and increased crystal 1 i n i ty result in an increase in the PZNC. However, the presence of anionic impurities of pH-dependent , potent ia I -determin ing ions will decrease the PZNC. Also, structures or compounds which are ion exchangers and have an intrinsic structural charge can be expected to shift the PZNC away from the calculated value by a variable amount. A positive charge density, a^, may be expected to move the PZNC in the basic direction, while negative a should move the PZNC in the acid direction (Parks, 1967). In Fe and Al oxide systems, the Fe and Al ions are not exchangeable, but can be potent ial -determin ing ions. In general, the presence of Fe and Al oxides will tend to increase the PZNC of soil toward higher pH values, while the presence of clay minerals with permanent or structural negative charges tends to shift the PZNC of soil to lower pH values (Van Rai j and Peech, 1972) . Effects of Specific Adsorption on the PZNC When specific adsorption of cations or anions occurs at the PZNC, the pH shifts toward higher and lower pH values, respectively (Breeuwsma, 1973). An excess of a specifically adsorbed ionic species will remove pH dependence or change the PZNC to that of the impure species (Parks, 19^7). Specific adsorption of ions other than the potent ial -determin i ng species H+ and OHis related to differences in PZNC between soils and within profiles (Adams, 1976)Specific adsorption of anions produces a negative surface under otherwise identical conditions. Anions, such as PO4 ^, which are dehydrated and specifically adsorbed by ferric oxides are considered to form a new potent ia 1 -determ i n i ng layer by 1 igand-exchange reactions

PAGE 38

24 Table 1. PZNC of pure minerals found in soils. Minerals PZNC Authors Si02 Montmor i 1 Ion ite Kaol in ite Gibbsite a-A1 (OH) 3 Magnet ite a-Fe203 Goethite Fe(OH) 3 amorphous Hemat ite a-Al 2O3 MgO 2.0 Parks and de Bruyn (1962) 2.5 Stumm and Morgan (1970) 4.6 Stumm and Morgan (1970) 5.0 Stumm and Morgan (1970) 5.0 Stumm and Morgan (1970) 6.5 Stumm and Morgan (1970) 6.7 Stumm and Morgan (1970) 7.55 Atkinson et al . (1967) 8.5 Breeuwsma (1973) 8.45-9.27 Atkinson et al . (1967) 9.1 Stumm and Morgan (1970) 12.4 Stumm and Morgan (1970)

PAGE 39

25 with OHor structural H2O in the ferric first coordination shell (Atkinson et al., I967). V/hen P or other anionic chemicals are added to a soil with variable-charge colloids, the adsorbed ("fixed") anions shift the PZNC to lower pH values (Adams, 1976). Lowering of PZNC in goethite (Hingston et al., I968) and in tropical soils (Van Raij and Peech, 1972) as a consequence of specific sorption of anions has been clearly demonstrated. Cation Exchange Capacity Effect of pH . Change in CEC with change in soil pH has been logically termed pH-dependent CEC. Soil fractions contributing to the degree of pH-dependent CEC were reported to be organic matter and clays which either contain noncrystalline components or have expanding plyl losi 1 icates with hydroxy-Al interlayers (Fiskell and Zelazny, 1971). The cations are held in these systems because of negative charges resulting from ionization of the OH groups attached to Si and Al of broken tetrahedral and octahedral positions and functional groups of organic matter (Wiklander, 1967). Fiskell and Zelazny (1971) studied the effects of selected soils buffered at several increasing pH values on CEC. The increase in CEC with soil pH was attributed predominantly to the increase in pHdependent charges on organic matter. Likewise, Fiskell et al. (1964) studied the changes in CEC with change in pH in a Leon fine sand. They found an increase in the CEC of this soil almost linearly with increasing pH. Helling et al. (196'<) found that the average increase in CEC of the organic fraction of 60 Wisconsin surface soils was correlated with soil organic matter.

PAGE 40

26 Effect of P adsorption . The CEC of soils is known to increase on treatment with phosphate (Sawhney, 197^). The increase in CEC has been attributed to the replacement of hydroxy! ions by phosphate ions at the surfaces of clay minerals and sesqu ioxides. Recently, it has been shown that sorption of anions on sesquioxides with surface at constant potential shifts the PZNC to lower pH (Breeuwsma, 1 973), thereby, requ i r ing more cations to balance the additional negative charge produced by specific adsorption of anions. Similarly, phosphate sorption caused large increases in the CEC of tropical soils rich in sesquioxides (Mekaru and Uehara, 1972). Specific sorption of anions on sesquioxides involves not only the replacement of hydroxyl ions but also of water molecules in octahedral coordination on oxide surfaces (Sawhney, 197^). Thus, soil pH is reduced more where a portion of the surface hydroxyl s are protonated to form water molecules, and replacement of the water molecules by an anion increases the CEC. Sawhney (197^) found that as the sorbed phosphate increased, the CEC also increased. He attributed the increase in CEC to the replacement of octahedral ly coordinated water molecules on sesquioxides surfaces at pH 5, while at higher pH values, increased CEC was caused by replacement of hydroxyl ions. Crop Response to Lime and P Applications It has been pointed out (Blue, 197^) that one of the most important considerations in the nutrition of crops in highly weathered Ultisols and Oxisols is the prevalence of P deficiency. Blue (1974) also stated that these soils have relatively high P retention capacities and, therefore, the efficient use of this element is important. The effects of P

PAGE 41

27 and lime on crops have received wide attention. Recently, phosphate adsorption isotherms have been used as a cr i ter ion f or P applications to tropical soils with high P adsorption capacities. Fox and Kamprath (1970) found that the effects of large phosphate applications to soils with high phosphate sorption capacities were evident in phosphate sorption isotherms 10 years later. They also used the sorption curves as a basis for fertilizing soils in pot experiments. Reeve and Sumner (1970b) concluded that the wide spectrum of P-fixing abilities of soils indicates that fertilizer recommendations based on direct determination of the amount of P required to raise the status of a soil to a desirable level are likely to be more accurate than those based on an estimate of the amount of available P in the soil per se. On the other hand, it has been found that lime reduces the adsorption maximum of P in Ultisols by reducing the exchangeable Al (Woodruff and Kamprath, cited by Blue, 197^). However, Amarasiri and Olsen (1973) found that a limed soil had a higher maximum adsorption capacity for P than unl imed soil. Evans and Kamprath (1970) found that lime increased the growth of corn on mineral soils when Al saturation was greater than 70^ and soybean when Al saturation was greater than 30%. Reeve and Sumner (1970a) found that the yield of 'trudan' ( Sorghum sudanense) increased with increasing lime and P application rates. They concluded that the increase was due to the reduction of Al toxicity and improvement of P ava i labi 1 i ty.

PAGE 42

MATERIALS AND METHODS Ten soils from Venezuela were used in this study including two Oxisols, a Grossarenic Psammentic Haplustox (Guanipa 5, Anzoategui State), and a Tropeptic Haplustox (Guataparo, Carabobo State); an Ultisol, Typic Paleustult (Guanipa 1, Anzoategui State); three Alfisols, an Ultic Tropudalf (San Cristobal, Tachira State), an Oxic Haplustalf (Barinas, Barinas State), and an Oxic Tropudalf (Bajo Seco, Miranda State); an Entisol, Ustic Quartzipsamment (El Roc'o) , an Inceptisol, Fluaquentic Humitropept (Alambre), and a Vertisol, Udorthentic Pellustert (Guanaguanare) from Portuguesa State; and a Aridisol, Haplargid (El Potrero, Lara State). The samples were collected from uncultivated sites (Figs. 1 and 2). The surface 30 cm of soil was mixed (in all samples) except Guanipa 5 for which the All (O to 10 cm) and A21 (10 to 30 cm) horizons were analyzed as separate samples. These soils had a pH range of 3-50 to 5.50, organic matter content of 0.50 to ^.50^, and clay content of 7 to sn. To compare soil behavior, Red Bay, a Rhodic Paleudult (horizons Al and B2t) from Florida, U.S.A., was also used in some analyses. This soil was formed under subtropical conditions and belonged to a highly weathered order. All samples were air-dried and ground to pass a 2-mm sieve for laboratory and greenhouse experiments. 28

PAGE 43

29

PAGE 45

31 Soil Physical, Chemical, and Mineralogical Properties In order to characterize these soils, several physical, chemical, and mineralogical analyses were performed. Except for Red Bay, particle-size distribution of the soil samples was determined by the pipette method. Moisture equivalent was determined by saturating with water and centrifuging at 1,000 rpm (Beaver et al., 1972). Organic carbon was measured by wet digestion with K2Cr207H2SO1, (Walkley, 19^7). To prevent interference with X-ray analysis, samples were treated with 30^ H2O2 to eliminate organic matter. Free Fe oxides were removed by Na-dithionite-citrate-bicarbonate (DCS). After these cementing agents were removed, samples were adjusted to pH 10 with Na2C03 and gently heated to remove any excess of H2O2. The sand was separated from silt and clay by wet sieving with a 300-mesh sieve using H2O adjusted to pH 10 with Na2C03. Clay was separated from silt by repeated decantation after centrifuging at 1,000 rpm for 2 minutes in an International No. 2 centrifuge. The residue was considered to be the silt fraction (Zelazny and Qureshi, 1971). X-ray Diffraction Analysis X-ray diffraction analysis criteria were used to identify the crystalline components of the inorganic colloidal separates. This was performed by orienting approximately 200-mg samples of the appropriate clay suspension on several glass slides and leaving them until air dry. One sample was saturated with Mg, one was Mg-saturated glycerol solvated, and the rest were saturated with K, nonheated, and heated to 100 and 500°C for A hours. Samples so prepared were analyzed with a Phillips PWl 120-96 X-ray d i ff Tactometer with Co radia-tion and scanning from 2 to 50° 26.

PAGE 46

32 Fe-oxides Determination Since Aland Fe-oxides contribute to P adsorption and to a significant portion of the constant-potential surfaces in many tropical soils, these were determined with Na-DCB. The Fe and Al in the extracts were analyzed by atomic adsorption spectrophotometry in a Perkin-Elmer 303 Spectrophotometer . Surface Area Determination Samples for surface-area determination were prepared by saturating the soil with IN MgCl2. Excess salt was removed with water and the samples were dried at 105°C. The clay was separated by centr i f ugat ion and ground to pass a 60-mesh sieve and mixed again with other components (Adams, 1976), Labeled Al dishes were placed in a desiccator over P2O5 under vacuum for 6 hours to obtain constant weight. Approximately 1-g samples of soil were then placed in the dishes and brought to constant weight as before. A sol 1 -absorbate slurry was prepared by placing 3 ml of ethylene glycol monomethyl ether (EGME) into the dried sample. The slurry was mixed and placed in a desiccator containing a CaCl2-EGME solvated slurry to equilibrate for 1 hour before applying vacuum (Carter et al., 1965). The desiccator was then evacuated at 0.25 mm Hg for 1 hour. Periodic weights were taken on the so i 1 -adsorbate slurry until constant weight was obtained. The surface area vMas calculated using the following equation: A = Wg (22) V/s X 0.00286 where A = specific surface in m Vg Vi/g = weight of EGME retained by the sample after equ i 1 i brat ion in g

PAGE 47

33 . Ws = weight of P2O5 dried sample in g. The constant was obtained from the assumption that 2.86 x 10 g EGME is required to form a mono-layer on each m^ of sample (Carter et al., I965). So i 1 Ac i d i ty Soil acidity was determined by potent iometr ic and conductometr ic titrations using the following procedure: 5-g soil samples were placed in 200-ml plastic bottles and 50-ml portions of distilled water containing increasing amounts of 0.05N Ba(0H)2, O.IN NaOH, or O.IN NazB^Oy were added to the bottles. The pH and electrical conductivity were measured after continuous shaking for hB hours. Extractable acidity was determined by leaching soil samples with IN KCl . A 10-g sample was placed in a 125-ml Erlenmeyer flask to which 100 ml of IN KCl were added. The flask was stoppered and shaken periodically for 2 days. Then, the mixture was filtered through V^hatman No. A2 filter paper into a 250-ml volumetric flask and washed with three separate 50-ml portions of IN KCl. One aliquot of 100 ml was titrated with O.O^^SSN Ba(0H)2 to pH 8.0. A 50-ml aliquot was analyzed for Al by atomic adsorption. Intensity measurements were also done by reading the pH in H2O, IN KCl, IN K2S04, 0.02M CaC^-, h% NaF at 1:2 soil:solvent suspensions. Differences between salt-pH and H2O-PH were calculated as ApH values in which pH (H2O) was substracted from the pH (salt). All measurements were made using a Beckman pH meter with a combination electrode. Components of acidity were estimated from the graphs resulting from plotting pH and EC vs meq of base added. Readings were made at inflection points at pH 8 and 10. In conductometr ic titration curves, inflections v;ere determined at the intersection of tvjo straight lines drawn along the linear portions at the extremities of the curves. End-point

PAGE 48

3^ values were multiplied by 0.75 to account for the neutralization factor associated with the formation of aluminate in the Ba(0H)2treated samples. Potassium Q/l Characteristics To determine the Q/l relation of the soil samples, duplicate subsamples of soil were shaken with 50 ml of O.OOIM CaCl2 with KCl ranging from 0 to 0.002M in lOO-ml centrifuge tubes for 2k hours at room temperature. Equilibrium solution was separated from the soil by centrifugation and analyzed for Na and K with an Ependorf flame spectrophotometer. The gain or loss of K was determined by substracting equilibrium K concentrations from initial K concentrations, assuming that the difference was either adsorbed or released from the soil complexes and converted to meq/lOOg. The graphs were obtained by plotting the gain or loss of K against the activity ratio of K (AR*^) present in the equilibrium solutions. Cation concentrations were corrected to their chemical activities by using a modified Debye-Huckel equation assuming that all anions were monovalent and Ca, Mg, K, and Na were the only cations present. log a = -AZ+Z-/r 1 + Ba /r (23) where A = 0.502 Z+ = valence of cation Z= valence of anion -8 a = 2.5 X 10 cm B = 3.9 X 10^ a = activity coefficient of electrolytes I = ionic strength I = 1/2 E Ci Z2

PAGE 49

35 Ci = concentration of the ion of charge Zi Potassium selectivity coefficients were calculated by the Gapon equation as follows: = K (sorbed) gK (solution) . {2k) ^ Ca + Mg (sorbed) ' /alCa + (solutionT Soil equilibrated with O.OOIN KCl and O.OOIM CaCl2 was extracted three times with NH^N03 and the combined extract was analyzed for K, Ca, and Mg as before. The results viere corrected for the entrained solution by analyzing for CI with an an ion-spec i f ic ion electrode and potentiometer. The determined CI was partitioned between cations in proportion to their concentrations in the equilibrium solution and these calculated amounts were substracted from measured values in the NH^N03 extracts to obtain the sorbed cations. Cations in solution were determined from the equilibrium solution and their chemical activities were calculated (San ValentTn et al . , 1972) . Other parameters v;ere derived from the Q/l graphs: PBC was calculated from the slope of the linear portion of the curve, the value of AR^ from the intercept of the curve when AK = 0. Also, AK° and Kx were determined as outlined by Beckett and Nafady (1967a) and Lee (1973)' Exchange Properties Soil CEC was determined by saturating with IN NHi^OAc (pH 7.0) washing with C2H5OH, exchanging the retained UH^ with IN NaCl , and determining the NHi^ by micro-Kjel dahl method (Bremner, 1965). The extract from NHi^OAc saturation was analyzed for Na, K, Ca, and Mg as before. Cation exchange capacity also was determined using the Stern model for double layer and point of zero net charge (PZNC) values.

PAGE 50

36 Point of Zero Net Charge This analysis is based on the premise that at PZNC, ionic strength will have no effect on potential, i.e., pH. The basic procedure used was that outlined by Adams (1976) except that only two concentrations of NaCl (1.0 and O.OOIN) were used. This procedure involves equilibrating several 3-g soil samples for 3 hours in the two concentrations of NaCl to which various al iquots of NaOH or HCl have been added. After equilibration, the samples were centrifuged, the pH of the solution determined, and compared to the pH of the same solution before adding the soil. From this difference in pH, the mi 1 1 iequ i val ents of h"^ or OH adsorbed by the soil were determined and plotted against pH for each salt concentration. The Intersection of the graphs was taken as the PZNC of the soil. Phosphorus Analysis After centrifugation in a super-speed centrifuge, P in the supernatant solutions was determined. Phosphorus that disappeared from solution was considered to have been sorbed. Phosphorus sorbed was plotted against P in the supernatant solution. After equilibration, soil samples were saved for P release and P fractionation using the Petersen and Corey (I969) modification of the Chang and Jackson (1957) P fractionation procedure. This is a sequential extraction using NHi^Cl , NHt^F adjusted to pH 8.32, NaOH, DCB, and H2S0it, to extract water soluble, Al , Fe, reductant soluble, and Ca forms of P. The organic P was estimated from the difference between P extracted with NaOH followed by H2S04 from samples ignited to 600°C for k hours compared with nonignited samples. The total P was calculated by adding all the fractions. The P in this and all subsequent P experiments was determined col or imet r ica 1 1 y using ammonium molybdate with ascorbic acid as a reducing agent according

PAGE 51

37 I to the procedures developed by Watanabe and 01 sen (1965). All P determinations were made with a Bauch and Lomb Spectronic 20 spectrophotometer at a wavelength of 880 ym. Calculations of Thermodynamic Parameters Associated with Adsorption Processes Thermodynamic parameters AG, AS, and AH were calculated from the variation of the thermodynamic equilibrium constant Ko (or the thermodynamic distribution coefficient) with changes with temperature. This technique was outlined by Biggar and Cheung (1973) and is described in details in Appendix II. Also, AG was calculated by using Keq from the relationship of concentration and the inverse of time as outlined by Rydens and Syers (1975). Langmuir and Freundl ich equations were used to calculate some values of AG related to energy of sorption reactions (Ryden et al . , 1977a). All these calculations were performed using a computer, examples of them are shown in Appendix II. Phosphorus Adsorption Isotherms Data for plotting P adsorption isotherms were obtained by equilibrating 3-g soil samples for 11 days at 5°C, 25°C, and kO°C in 30 ml of water containing various amounts of KH2P0i^. Also, isotherms were obtained in samples saturated with KCl and CaCl2 and washed free of CI . These were performed at 25°C using KH2P0tt and Ca(H2P04)2, respectively, to keep homoionic systems. Equilibration was carried out in 50-ml plastic centrifuge tubes. Two drops of toluene were added per sample to avoid biological activity. The tubes were shaken in a reciprocal shaker for a 30-minute period daily. Phosphorus Release Samples from P adsorption isotherms were extracted 10 times with 0.02 M KCl or 0.02 M CaCl2, shaken for 5 minutes and centrifuged for

PAGE 52

38 25 minutes with a high speed centrifuge. The KCl was used in samples with and without previous K saturation, while CaCla was used with the Ca-saturated samples. Phosphorus was analyzed in each extract. The amount was corrected for the initial concentration of the solution left in the soil sample. A sulfur fractionation was performed in the Soil Testing Laboratory of the Soil Department, Facultad de AgronomTa, Venezuela, as outlined by Casanova {\37^) . This is a sequential extraction using HCl and NaHCOs, with and without heat at 500°C. Lime and Phosphorus Treatments Lime requirements of the soils were determined by incubating 500g samples in plastic pots with amounts of CaCOg equivalent to 0, 1, 2, 3, ^, 5. 6, 7, 8, 9, and 10 metric tons/ha. Distilled water was added to the samples to bring the moisture content to moisture equivalent. The pots were closely covered and placed in a greenhouse over a period of 6 months. After incubation, soil samples were dried and pH was determined in H2O. Lime requirement curves were obtained by plotting pH vs amount of CaCOg applied. +3 In order to compare lime requirements, IN KCl -exchangeable Al was determining. The rates of CaCOs selected for laboratory and greenhouse experiments were 0, 1/2, 1, and 2 times the IN KCl -exchangeabl e Phosphorus requirements were calculated from adsorption data obtained from the Langmuir isotherm. The P levels applied to the soil samples were 0, ]/h, 1/2, and 1 times the adsorption maximum calculated from Langmuir isotherms.

PAGE 53

39 Laboratory and Greenhouse Experiments Samples from nine Venezuelan soils were limed with CaCOs equivalent to 0, 1/2, 1 , and 2 times the IN KCl -exchangeable Al"^^ After 15 days of incubation, four levels of P were established in each soil by adding P as indicated by the appropriate sorption curve, to reach 0, 1/^, 1/2, and 1 adsorption maximum (Table 2), Phosphorus was added as KH2P0it in solution form. The soil was then adjusted with water to moisture equivalent and incubated for 11 days. A portion of sample from each treatment was saved for PZNC and pH determinations. One-kilogram samples of Guanipa 1, Guanipa 5, and El Potrero soils were place in pots in a greenhouse and planted to sorghum ( Sorghum bicolor L.), 'Hybrid Chaguaramas 3' was maintained by frequent irrigation. Plants were harvested l8 days after seeding. All pots received uniform fertilization with N, K, Mg, and micronutr ients . The experimental design was two factors (lime and P) at four levels (l , 2, 3, and k) with three replications and complete randomization. After harvest, soil samples were airdried, ground to pass a 2-mm sieve, and analyzed for PZNC and pH. Plant material was dried at 60°C and weighed to determine yields. Statistical Analyses Analyses of variance were performed for greenhouse experimental data by the General Linear Model procedure (Barr et al., 1976) in the factorial design. Duncan's Multiple Range Test was used to compare individual treatments means. Also, a linear regression study was performed for P adsorption data. Special programs were made to calculate the surface charge with Stern and Gouy-Chapman models, the Q/ I parameters for K status in the soils, and thermodynamic parameters.

PAGE 54

Table 2. Treatment combinations used in the laboratory and greenhouse experiment with the soils used. Phosphorus appi ied CaCOa appl ie dt Rate Designation ~] 2 3 h Treatment combinations Ob"^ 1 11 21 31 k] 2 12 22 32 h2 hh 3 13 23 33 h3 lb k ]k 2k 3A hh TAdsorption maximum (b) from the appropriate Langmuir equation. *0, h, 1 , and 2 times the IN KCl -exchangeabl e Al .

PAGE 55

RESULTS AND DISCUSSION Soil Characteristics with Emphasis on Surface Charge The soils selected for this study varied considerably in clay and organic matter contents, CEC, exchangeabl es bases, and acidity sources. Also, they showed a very large variation in free Fe and Al oxide contents, surface area, clay mineralogy, PZNC, and P and S fractions. The K status was quite different for these soils as shown by the Q/l parameters found with this technique. Soil organic matter content is a result of biological processes that are controlled by temperature and moisture. The relatively low organic matter contents of these soils was due mainly to the oxidation enhancement of high temperatures. Organic matter colloid translocation was not noted since there was a tendency for lower organic matter contents with depth. The organic matter contents ranged from 0.28 to ^.5>^% (Table 3). These values are the result of the combined temperature, rainfall, and altitude factors (Westin et al., I968). Exchangeable bases in these soils were generally low. Calcium was the predominant ion in all soils (Table k) . Magnesium followed Ca in the exchangeable bases. Potassium values were very low in all soils. Except for San Cristobal and El Potrero, exchangeable Na was less than 0.1 meq/lOOg. In general, the highly weathered soils contained less bases than those less weathered. There was a tendency for bases to

PAGE 56

o 3r LUO oooo o t_> CD o o 03:ijJ3:3: i.^ — i2 — V LALAOOOOOLTvOOOLAO oo-TCMOunocr>LAi-r\oooo\£> O CM 0fv-\0-3-vuO0ca0cs| E Ol CM — — OA CO 0 LA vO r~. CM (U II CM CA r— 4-> to c TO 0 D) TO 0 i 0 -300 0 -3CM 0 CA vD CM v£> 0 1 in 1 vD -3<^ — CM ^0-30 CO vD c LTV CM CM CM OA CM LA OA 0 1 (U 1 u M TO 1 •U 0 CO CO v£> CM CM cr» r~-dvo CO 0 0 > TO 1 vO CO CM CM LA -T CA LA CNl CN) LA vi) E 1 u 0) 1 0 00 — CM csl CM — -3OA CM 0 JD TO JD ^ TO II 1 1 O) 3: C (C >. "o Dl CL QCL »0 CO C71 CL U t-J TO TO TO M1.._ 0 0 TO TO -Q 0 c CO CQ 0) c c c DC 0 a. (U •»-> ro (D TO C • — 1 TO TO TO !_ -0 X) SI c n D 3 TO TO 13 3 ^ TO h(J} CD C3 CO LU CD t3 < LU CO 0£ CC 43 O 'i • 0) 4^ > — N — +J U •» TO E 3 3 CTTJ II C M 3 cy lO o
PAGE 57

1 LA LA O O O rA o O o a^ O ro CD 1 CO CNI — CM O rA O O O vO O CA z: 1 1 1 O O O v£> O — O rA o — o 1 1 1 vD LA O O O O LA O o o CM CM O TO 1 rACACsl LTVOArOO • — -JCO LA o 1 1 o O O rA >X) •— LA cs o LA O • 0) 1 T3 01 1 (U TO 1 C CO 1 CO o u\ LA LA CA LA o LA vD LA E 1 1 o O O LA — LA M*-> 1 o o o o o O v£) O O O LA o O 3: 1 -3CA O OA O rA O O o CTi o -cr -a o 1 V) ^ — t — — O rA cr\ vO CM rA-3\0 rA (U o 1 TJ •— < CO 1 X 1 1 o 1 (U 1 1 (U 1 LA O — O O O O O O o o o u o 1 OA -3_T O LA O OO O CM o LA ro M< 1 o O O — vO O rA O CA O LA -cr TO 1 1 o o o o o O — O OA CA O o o C TO (/) 0) •— (/) 1 M (U 1 iT3 1 CA rAvX> LA O O cvl O rA O o -aCA LA ^ CX3 Q_ O o ! o — O rA — ^ CM CM o o 1Cl < ! d) CD c TO s: o (/) X (U T3 cA LA r-. r-~. o ~ cn r^oo LA — O O rA -sCM CO CM 1 — X • o o O O rA rA rA rA rA rA -3TO hTO (U XI J_ 4-" < QQ \0 TO O — CM LA LA -M O O C L. < CO in O O ITO i) TO TO TO — 0) TO 3 > Q. Q. Q. 1O OO Q. D> 4-1 TO ro TO — — O O TO TO J3 O C CO CO c C C OC o +-> c E CL TO TO TO C —1 TO TO TO u -a TD o rj 3 D TO — TO 3 3 TO 0) 1) to 13 CD CJ l/l LU OQ O O < UJ CO

PAGE 58

decrease with depth. These values were expected because of the low pH of all soils. Cation exchange capacity values were variable (Table 3) and higher than effective CEC (ECEC). Guanipa 1 showed CEC equal to ECEC. Cation exchange values ranged from 1.1 to 30. k meq/lOOg. The values of CEC were due to the contribution of organic matter and clay content in these soils. The magnitude of permanent negative charge would be expected to vary according to the degree of tetrahedral substitution in the clay mineral. The amounts of Al and Fe oxides present in addition to the presence of crystalline colloidal species with low pH-dependent charges would considerably influence CEC measurement at pH 7.00. Free Fe-oxide values ranged from 0.75 to k.2S%. Aluminum oxides ranged from 0.64 to 3.^5% (Table k) . These values reflected the stage of soil weathering. These results indicate a loss of Si due to the acidic environment and an enrichment in the Al and Fe oxides. A tendency of these values to decrease with depth indicated that Fe and Al have been released from primary minerals and translocated by percolating water to lower horizons. Surface-area values ranged from 2.7 to 69.9 m^/lOOg (Table 3) and were related to organic matter and clay content in these soils. The highest value was obtained for the Aridisol and the lowest for the sandy Ultisol. These soils had increasing surface area with depth. The migration of fine particles in soil formation processes can account for these results (Luque, 1975). The amount of clay in soils depends upon the soil -forming factors including parent material, relief, time, biotic, and climate. The clay

PAGE 59

45 contents in the soils studied ranged from ^.kO to 51.00?^ (Table 3). Clay content tended to increase with depth, indicating downward movement and accumulation in lower horizons. Clay Mineralogy Abundance of kaolinite and quartz is indicative of parent materials in their last weathering stage. Oxisols which represent the extreme of pedogenic weathering have traces of aluminosi licates such as intergrade (Table 3). These soils contain various Fe and Al oxides (gibbsite, hematite, and goethite) as evi denced by 't.83, 3.6?, and k.SS % peaks , respectively, and in some cases, chlorite. Soils Guanipa 5 and Guataparo represent this order. Weathering has been so severe that only resistant minerals (quartz k.2 and 3.35 8, and corundum 3-^17, 2.55, and 2.08 %) , and highly resistant layer silicates remain. Clay fractions with Ultisols were dominated by kaolinite and intergraded phyl los i 1 icates (Guanipa 1 and Red Bay). Kaolinite was identified by first order spacing 7-0 ^. A band in 11.6 to 12.6 % region was assumed to represent intergrade minerals because of their persistence after heating Guanipa 1 showed some chlorite. The peaks at 13-6 to 1^.3 8 without expanding at Mg treatment were assumed to be chlorites. Alfisols (San Cristobal, Barinas, and Bajo Seco) tended to have a somewhat wide-ranging clay mineral suite. Barinas (Haplustalf) showed predominance of kaolinite, followed by quartz and illite, and some amount of goethite, while Bajo Seco (Tropudalf) had illite, quartz, mica, and some amounts of kaolinite, corundum, and chlorite. Finally, San Cristobal (Tropudalf) showed vermiculite, in addition to kaol i n ite, quartz , mica, and chlorite. Vermiculite was concluded to be present by the Mgsaturation ]h to 15 8 spacing and a 10.0 K spacing after heating to 500°C.

PAGE 60

46 The presence of these minerals may have been the result of a more baseenriched environment, and to low rainfall and a distinct dry season (Westin et al . , I968; Zelazny and Calhoun, 1971)Entisols and Inceptisols exhibited inherited minerals such as mica, intergrade layer silicates and chloride in addition to kaolinite and quartz (Table 3). Point of Zero Net Charge The PZNC values for the soils examined ranged from pH 2.85 to k.GO (Table 3). The Guanipa 1 and Barinas soils were the lowest and highest, respectively. These values tended to increase with soil depth and weathering. If a PZNC range of pH 3.5 to 5-0 is characteristic of soils dominated by constant surface-potential colloids as proposed by Keng and Uehara (197^*), those soils, except for Guanipa 1, should belong to this group. The lower PZNC values for these soils were due to the presence of organic matter in their surface horizons. It has been shown by Adams (1976) that the narrowed range of PZNC in tropical and temperate regions was due to the interlayered AI and organic matter. Phosphorus and Sulfur Fractionation Phosphorus fractions as determined by the Peterson and Cory (I969) method showed dominant P forms to be Fe-P, Al-P, and Ca-P. All the soils except Guanipa 1 and 5A had lower amounts of Al-P than Fe-P. This may have been due to the lower solubility product of Fe-P. Also, aging may have been responsible because of the high Fe oxide content (Yuan et al., i960). Soluble P was not detected in these soils. Most of the P v/as in the organic form (Table 5) • On the other hand, S fractionation (Casanova, 197't) showed that S was mainly in the clay and organic forms (Table 6). These values of S in the clay fraction suggested that these soils have a high capacity to adsorb and exchange phosphate (Casanova, 197^).

PAGE 61

vOvDLnr^O — CTiOOvDOLAOCO oo • — a•— — — CM POO vO CO -a— ooocounrauntNO-— r<->o -3-cr>cr»vorocM>x>co— vi3crvcr>r-^ — (v-vcMCNlLALrvraco-— CMCNI \£)CMLrvlACMCMC0OCri-3-0^— o -crcMoo— coLrwDr^uA'— LTyr^oo .— — roo^-— cM-a— CM — — •— "T-l-HI-l-h-H-l-l-HI-HlcM — — r<-\i^crvoo-3-ooor~-.— racacoi^cMr^— ouALncMcoo -^CMcMCr\csl-:fCOOOCOcor^— or-\ CM c-v CM OA CM TO (U JD lM < CO VD TO O CM un M o O c i_ < DD (/) o o TO 0) TO TO V — TO >Q. Q. QV_ o to DOl 1_ •M TO TO TO o o TO TO JD o C CQ c C c q: o +J C E TO TO TO c ' — » TO TO TO 1_ XJ -o rj TO TO TO (U 0) C3 (/> UJ D3 O CD < UJ CD CC a:

PAGE 62

OrovOOOOvOr^O CT>r^vD<— OOv£)Ur\r— — CO \0 mi^oOvO CTvvD — tM r— — rr\ otM-a-oooooo cMcNir-^r^raoLAoo — Ot — POOOOOf^O v£>-3oacr\r— OsO raO cnooooooo I— c^racNvDOOCNJ — r^ — -ar^-:}-^^ o X) I. %o fO o •t-J o O C u in ^2 o u TO (U a) (D 0) 1in D. i_ o C/) Q. C7) 1_ 0) o o (D fD J3 o c C en o !-> C E Q(U c •— > fD fD 03 1_ ro nj D 03 t3 (/> UJ 00 C3 CJ3 < UJ CO

PAGE 63

The experiment carried out to determine the temperature of incubation effect on P fractions after an 1 1 -day equilibrium period showed that most of the P was in the NH4F and NaOH extracts and was assumed to be in the Al and Fe forms, respectively (Tables 7 through 19). Only small amounts of P were found in the NHi^Cl and H2S0i^ extracts which represent water-soluble and Ca phosphate forms (Fig. 3). The phosphate applied, even at high rates, was fixed by Al and Fe. This was due to the fact that in acid soils the applied P precipitates or is adsorbed as Fe-P and Al-P (Yuan et al . I96O) . In the Guanipa 5B (A12 horizon), at highest P applied and ^0°C, the amount of P in the NaOH extract was more than in the NHi^F extract (Table 9). This indicated that P in this soil tended to form Fe-P rather than Al-P because of the lower solubility product of Fe phosphate or by a shifting of P from the Al-P form to Fe-P form induced by the higher temperature (Yuan et al., 19^0). The ratios of Al-P to Fe-P broadened with increasing P applied. It has been pointed out that kaolinite can function in the fixation of P by bonding of P to Al at the broken edges. This could have been the case in soils with kaolinite as the dominant clay mineral (Yuan et al., i960) . The pH and pH2P0i4 of the different ionic systems at 25°C are shov^n in Tables 20 through 28. The pH2P0i^ values were very low at high equilibrium P concentrations. This indicated that precipitation of Al and Fe phosphates such as strengite and variscite can occur. An experiment was carried out to observe the effect of high temperature on Aland Fe-phosphate precipitation in the nonsaturated system with the highest

PAGE 64

Table 7Effect of incubation temperature on phosphorus fractions In a typic Paleustult from Anzoategui State, Venezuela (O-3O cm) after applying several solution concentrations of phosphorus. P appl ied Al-P Fe-P Prs Ca-P as KH2PO4 ppm 5°C 0 11 .Oi» 9.^1 8.35 1 .67 100 56.76 19.98 8.35 1.39 500 30.14 8.6 3.29 1000 1^5.7 25.59 7.97 lAG 25°C 0 10.88 9.66 15.25 1.8i» 100 80.62 30.32 6.96 6.39 500 125.75 37.93 11.19 5.68 1000 151 .8 37.93 8.6 A. 90 ko°c 0 6.1 9.7 9.88 1 .92 100 71.77 22.8 13.81 2.73 500 157.9 26.78 16.65 2.73 1000 276.5 28.82 17.88 2.87

PAGE 65

Table 8. Effect of incubation temperature on phosphorus fractions in a Grossarenic Psammentic Haplustox from Anzoategui State, Venezuela (All horizon) after applying several solution concentrations of phosphorus. P applied Al-P Fe-P Prs Ca-P as KH2P0ij 0 11.7 14.21 8.64 100 160. A 10.62 8.60 500 326.11 1 1.21 8.60 1000 380.0 13.97 9.88 25°C 0 12.2 9.0 9.80 1 .80 100 6i».25 20. 12 4.83 3.28 500 115.5 23.58 7.34 4.85 1000 l^tO.O 27.10 3.60 4.58 AO°C 0 5.51 14.67 12.50 0.37 100 66.97 26.01 16.38 0.78 500 143.61 30.79 17.75 1.26 1000 388.17 34.85 19.27 1 .26

PAGE 66

Table 9Effect of incubation temperature on phosphorus fractions in a Grossarenic Psammentic Haplustox from Anzoategui State, Venezuela (A12 horizon) after applying several solution concentrations of phosphorus. P applied Al-P Fe-P Prs Ca-P as KH2P0it •ppm0 100 500 1000 8.85 ek.25 179.0 5°C 6.88 13.17 15.01 20.1 1 8.60 8.60 8.35 8.60 0.50 0.62 0.88 0 100 500 1000 8.5 77.17 128.7 138.8 25°C 9.0 2S.7k 31 .40 23.58 AO°C A. 83 A. 83 4.83 2.2 2.31 2.51 1.12 0 100 500 1000 4.37 94.02 138.62 186.84 18.25 301 .41 315.86 367.0 13.16 19.00 19.27 10.00 1 .01 1.52 1.79 1 .79

PAGE 67

53 n r r r r i D_ D. e_ I I I — o < iL O HDB o o o o o LA CTv CO o t o O O LTv n D. a u~ *^ — » 0 c • D *-> "O C3 CNJ O II CD — O TO CL O CO o E O XI — oc O li •—1 r<^ — O — CO — II tJ 3 V0 C cr (D 0) •rfO o < -o o 1 cr >— 01 in in (D c — c o >— .O V+j TO VQ II 4-> — C II r_ o C O O c2 I . 1_ a o >..Dre — (U t3 II M II o O .— K•H< fO LTl (D J_ (/) O X) o o. E — ro c — Q < n II C3 cn u II It^ TO C i/i ra TO 3 Q. 3 1— • cn O C TO x: Q c DD TO IT, CD 3 O II t3 JC ^ II cl. — CO c o -O -30) o --^ — Q. ^ ^ CM E D. 3: --s, DD1 TO :a a, TO

PAGE 68

Table 10. Effect of incubation temperature on phosphorus fractions in an Ultic Tropudalf from Tachira State, Venezuela (O-3O cm) after applying several solution concentrations of phosphorus. P applied Al-P Fe-P Prs Ca-P as KHzPOij ppm 5°C 0 10. i» 40.9 14.5 1.84 100 283.7 188.6 25.3 3.40 500 856.1 293.1 31.0 4.00 1000 1179.0 357.1 34.2 9.50 25°C 0 26.5 44.70 1.2 100 311.7 255.0 9.25 500 908.8 264.3 15.15 1000 1^5^*. 3 372.3 15.15 40°c 0 8.94 42.6 1.3 1 .76 100 680. A 264.8 44.0 2.6 500 11 24. 3 294.7 72.8 3.1 1000 2196.3 301.4 84.1 3.7

PAGE 69

Table 11. Effect of incubation temperature on phosphorus fractions in a Ustic Quartz i psamment from Portuguesa State, Venezuela (O-3O cm) after applying several solution concentrations of phosphorus . P appl ied as KHoPOi^ Al-P Fe-P Prs Ca-P ppm 5°C 0 28.9 91.0 52.55 2.73 100 342.8 235.79 87.0 12.58 500 788.2 377.1 96.8 10.61 1000 1016.8 782.6 102.3 7.75 25°C 0 35.8 41.7 51.9 5.40 100 394.6 612.8 60.6 5.65 500 808.1 420.7 64.6 5.68 1000 1170.7 480.8 64.6 5.68 40°C 0 34.5 47.26 89.5 2.9 100 658.4 217.7 132.1 3.15 500 1361 .4 329.1 174.2 4.6 1000 1650.6 387.5 175.3 4.5

PAGE 70

Table 12. Effect of incubation temperature on phosphorus fractions in an Oxic Tropudalf from Miranda State, Venezuela (O-3O cm) after applying several solution concentrations of phosphorus. P appl ied as KH2P0^ Al-P Fe-P Prs Ca-P ppm 5''C 0 15.9 klA 2.40 2.56 100 582.3 126. if 45.37 5.19 500 756.9 188.6 45.73 6.59 1000 886.0 270. s 52.91 4.59 25°C 0 18.2 201 .4 8.6 2.7 100 309.0 396.9 64.6 7.4 500 775.5 2i4l A 79.8 12.2 1000 810.3 204.0 56.6 11.5 A0°C 0 12.5 39.9 2.8 6.1 100 i»20.6 338.2 75.0 5.5 500 870.8 3^*7.5 97.8 5.4 1000 938.5 '»20.6 113.1 5.8

PAGE 71

Table 13Effect of incubation temperature on phosphorus fractions in a Tropeptic Haplustox from Carabobo State, Venezuela (O-3O cm) after applying several solution concentrations of phosphorus. P applied Al-P Fe-P Prs Ca-P as KH2PO4 ppm 0 9.3 1.5 n .2 2.4 100 21^.5 47.2 17.6 3.9 500 667.6 83.3 24.9 3.4 1000 956.1 82.3 24.9 2.8 25°C 0 5.2 153.0 1 .2 2.4 100 325.3 181 .9 17.2 4.0 500 676.6 194.6 6.1 3.5 1000 930.0 158.1 8.6 3.2 40°C 0 10.6 24.0 19.4 2.6 100 432.0 96.5 32.2 3.6 500 694.5 148.9 37.3 5.5 1000 1175.9 176.4 37.3 16.6

PAGE 72

Table lA. Effect of incubation temperature on phosphorus fractions in an Udorthentic Pellustert from Portuguesa State, Venezuela (O-3O cm) after applying several solution concentrations of phosphorus. P appl ied Al-P Fe-P Prs Ca-P as KH2P0^ ppm 5'C 0 19.5 17.7 6.7 100 382.5 97.^ hk.l 5.9 500 933.^ 132.3 51 .0 10.8 1000 1007.0 163.6 71.2 9.5 25'"C 0 Ik.l 68.0 23. 1 2.1 100 233.0 hh\.\ 129.6 3.7 500 662.0 676.6 129.6 1000 868.7 511.7 150.9 Ao°c 0 17.3 53.9 74.8 U.5 100 53^.3 ^90.3 162.0 5.2 500 1253.8 609.2 206.6 5.9 1000 1896.2 688.2 226.5 8.1

PAGE 73

Table 15. Effect of incubation temperature on phosphorus fractions in a Fluaquentic Humitropept from Portuguesa State, Venezuela (O-3O cm) after applying several solution concentrations of phosphorus. P applied A)-P Fe-P Prs Ca-P as KHzPOtt ppm 5°C 0 6.6 31.4 33.4 3.7 100 406.8 62.3 49.3 6.8 500 1 158.5 61 .3 55.1 7.5 1000 712. 1 66.5 54.8 11.4 25°C 0 15.9 9.8 67.7 2.0 100 456.9 82.4 52.9 4.3 500 1 186.2 87.0 59.5 4.1 1000 1247.1 357.2 70.8 4.8 40°C 0 26.9 36.2 37.3 1.5 100 971.2 351 .4 58.6 1.5 500 2502.9 382.3 80.3 3.5 1000 3014.5 484.6 83.3 3.4

PAGE 74

60 Table 16. Effect of incubation temperature on phosphorus fractions in a Haplargid from Lara State, Venezuela (O-3O cm) after applying several solution concentrations of phosphorus. P applied Al-P Fe-P Prs Ca-P as KH2PO1, 5»C 0 9.3 23. k 33.8 3.6 100 55.9 37.3 87.0 500 997.^ 62.3 AO. 9 87.0 1000 128i*.3 59.'+ A2.3 87.0 25°C 0 10.3 57.2 2k. 3 9.3 100 513.8 185. A 32.1 '».9 500 908.8 277.2 38.1 6.1 1000 1300.1 311 .5 51.1 9.0 ko°c 0 16. i» 59. 't 37.3 8.7 100 673.0 320.3 41.9 8.7 500 1576.5 333.6 '»5.7 8.7 1000 2111.6 382.3 56.5 10.5

PAGE 75

Table 17. Effect of Incubation temperature on phosphorus fractions in an Oxic Haplustalf from Barinas State, Venezuela (O-3O cm) after applying several solution concentrations of phosphorus. P appl ied as KH2PO4 Al-P Fe-P Prs Ca-P 5°C 0 6.9 16.2 22.1 3.9 100 262. if 28.0 23. k 6.1 500 696.6 52.2 29.6 6.5 1000 8AO.9 30.2 7.3 25°C 0 2k. 6 15.3 29.'* 3.1 100 252.6 69.7 35.h 2.7 500 702.9 105.2 35. A 2.5 1000 988. i» 123.8 40.9 3.0 ko°c 0 13.1 2h.k 27.2 S.k 100 361.3 1 18.7 33.8 9.3 500 788.2 171 .2 10.6 1000 1231 .1 210.8 10.6

PAGE 76

Table 18. Effect of incubation temperature on phosphorus f ract ions in a Rhodic Paleudult from Flor ida , U.S.A. (Al horizon) after applying several solution concentrations of phosphorus P appl ied Al-P Fe-P Prs Ca-P as KH2PO4 5°C 0 9.7 21.3 17.8 1.6 100 29^.3 38. i» 23.6 3.0 500 763.^ 30.9 3.7 1000 992.5 72.6 25.7 it. 3 25°C 0 37.3 27.7 17.8 7.h 100 '»73.0 110.1 30.2 8.7 500 916.7 152.2 35.7 8.5 1000 1003.2 181 .3 3^.9 9.7 ko°c 0 28.6 36.7 16.5 3.9 100 691.5 220.0 35.7 e.k 500 2313.6 270.9 ke.6 10.0 1000 2227.3 286.6 5^.7 9.8

PAGE 77

Table 19. Effect of incubation temperature on phosphorus fractions in a Rhodic Paleudult from Florida, U.S.A. (B2t horizon) after applying several solution concentrations of phosphorus. P appl ied as KH2P0it Al-P Fe-P Prs Ca-P ppm 5°C 0 11.3 20.5 32.5 4.0 100 385.0 28.0 33.8 7.6 500 765.8 3^.0 35.7 7.7 1000 1H5.3 31.7 36.0 7.7 25°C 0 21 .0 22.6 21.3 5.3 100 k]0.5 63.5 30.2 5.4 500 871 .2 92.6 41.3 4.7 1000 1052.3 110.1 3it.l 3.9 0 12.6 31.3 27.9 6.0 100 220.6 ^3.9 11.7 500 1121 .k 228.8 47.5 11.7 1000 li»it3.8 2hh.S 61 .5 15.6

PAGE 78

6i> Table 20. Values of pH2P0i^ and pH at equilibrium with different ionic systems after P adsorption for a typic Paleustult f rom Anzoategu i State, Venezuela. P Nonsaturated K saturated Ca saturated added pHjPO^ pH pHgPO^ pH pHzPOit pH -yg/ml10 3.61 6.75 ii.Al 5.95 3.52 50 2M 6.35 3.1A 5.85 3.22 5.30 100 2.52 5.85 2.86 5.65 2.92 5.00 200 2.22 5.70 2.61 5.50 2.62 A. 55 700 1 .68 5.30 1 .93 5.10 1 .92 3.70 1000 1.55 5.20 1.52 A. 80 I .58 3.^0

PAGE 79

65 Table 21 . Val ues of pH2P0i^ and pH at equilibrium with different ionic systems after P adsorption for an Ultic Tropudalf from Tachira State, Venezuela. P Nonsaturated K saturated Ca saturated added pHzPOit pH pH2P0^ pH PH2PO4 pH -yg/ml 10 k.]] 6.60 3.67 6.50 3.71 6.15 50 6.05 3.6i» 6.00 3.64 5.60 100 2.80 5.70 3.08 6.10 3.30 5.30 200 l.liO 5.55 2.79 5.90 2.88 5.00 700 1.75 5.20 2.03 5.60 2.02 h.SO 1000 1 .58 5.10 1 .69 5.25 1.67 A. 60

PAGE 80

Table 22. Values of PH2PO4 and pH at equilibrium with different ionic systems after P adsorption for an Ustic Quartzipsamment from Portuguesa State, Venezuela. P Nonsaturated K saturated Ca saturated added PH2PO:, pH pHjPOi^ pH pH2P0i^ pH pg/ml 10 A. 58 6.75 3.9A 6.70 3.62 6.10 50 3.55 6.70 3. '3 6.30 3.55 6.15 100 3.05 6.50 3.08 6.05 3.^*0 6.10 200 2.1»8 5.70 2.82 6.25 3.31 5.hS 700 1.75 5.15 2.00 6.90 2. Oh A. 70 1000 1 .58 5.10 1.55 5.35 1.67 4.30

PAGE 81

67 Table 23. Values of PH2PO1+ and pH at equilibrium with different ionic systems after P adsorption for an Oxic Tropudalf from Miranda State, Venezuela. P Nonsaturated K saturated Ca saturated added pH2P0it pH PH2PO1+ pH pH2P0[t pH -yg/ml 10 4.26 7.35 3.51 6.85 4.07 6.50 50 3.56 7.25 3.12 6.80 3.53 6.30 100 2,97 7.05 2.94 6.70 3.16 5.80 200 2.46 6.60 2.69 6.i)0 2.77 5. AO 700 1.74 5.60 1.97 5.80 1 .98 A. 80 1000 1.55 S.hS 1 .60 5.50 1.68 A. 30

PAGE 82

68 Table 2^. Values of pH2pCit and pH at equilibrium with different ionic systems after P adsorption for an Udorthentic Pellustert from Portuguesa State, Venezuela. P Nonsaturated K saturated Ca saturated added pH2P0t^ pH pH2P0it pH pHzPO^ pH -yg/ml10 ^.53 7.45 3.93 7.60 3.71 6.90 50 3.95 7.75 3.3't 7.20 3.63 6.70 100 3.03 7.70 3.00 7.10 3.23 6.35 200 2.69 6.90 2.72 6.35 2.83 5.80 700 1 .82 6.05 2.02 6.00 1 .98 A. 50 1000 1 .61 5.60 1.62 5.60 1.67 '».20

PAGE 83

69 Table 25. Values of PH2PO4 and pH at equilibrium with different ionic systems after P adsorption for a Tropeptic Haplustox from Carabobo State, Venezuela. P Nonsaturated K saturated Ca saturated added PH2P04 pH pH2P0it pH PH2PO,, pH -pg/ml 10 A. 28 6.05 3.86 5.55 3.71 6.00 50 3.12 5.^5 3.30 5.55 3.58 5.70 100 2.72 5.^0 3.01 5.62 3.21 5.55 200 2.33 5.35 2.81 5.55 2.80 5.25 700 1.71 5.10 1 .98 5.20 1.95 i».60 1000 1.5A 5.05 1.57 5.05 1.68 3.90

PAGE 84

Table 26. Values of pH2P0tt and pH at equilibrium with different ionic systems after P adsorption for a Fluaquentic Humitropept from Portuguesa State, Venezuela. P Nonsaturated K saturated Ca saturated added PH2P04 pH pHp.POtt pH pHaPOit pH -pg/ml10 5.01 6.55 3.78 6.00 3.72 5.60 50 3.85 6.75 3.51 6.00 3.68 5.30 100 3.03 5.55 3.18 5.35 3.50 '+.80 200 2.51 A. 76 2.88 5.25 3.03 A. 50 700 1.78 i».6o 2.0k 5.30 2.07 ^.00 1000 1.59 k.es 1 .60 A. 90 1 .67 3.80

PAGE 85

71 Table 27. Values of pH2P0tj and pH at equilibrium with different ionic system after P adsorption for a Haplargid from Lara State, Venezuela. P Nonsaturated K saturated Ca saturated added pHjPOi, pH pH2P0it pH pH2P0ij pH -pg/ml 10 ^.53 'k80 3.58 5.65 3.75 '.80 50 3.38 A. 70 3.37 5.^5 3.71 A. 70 100 2.88 it. 30 3.06 S.^iO 3.35 li.hO 200 2.i»3 A. 55 2.81 5.20 2.91 4.30 700 1.75 k.so 1.99 k.80 2.07 3.80 1000 1.58 A. 50 1.60 i».65 1.66 3. 70

PAGE 86

72 1 Table 28. Values of pHjPOi^ and pH at equilibrium with different ionic systems after P adsorption for an Oxic Haplustalf from B arinas State, Venezuela. P Nonsaturated K saturated Ca saturated added PH2PO1, pH pHgPOi, pH pHaPO^ pH -yg/ml10 k.h] 6.75 3.64 6.55 3.75 6.95 50 3.13 6.30 3.40 6.50 3.74 6.85 100 2.70 6.00 3.02 6.30 3.36 6.50 200 2.3^» 5.85 2.64 6.00 3.27 5.95 700 1.71 5.50 1.95 5.60 2.00 5.00 1000 ].5^ 5.^0 1 .57 5.35 1 .67 4.65

PAGE 87

73 P concentration. Soil samples were autoclaved first for 20 hours at i20**C and then for hO hours at 120°C. X-ray diffraction patterns showed the presence of strengite 20-hour treatment in some soils but no variscite. For the 'tO-hour treatment, both variscite and strengite v/ere present, except in the San Crist5bal and El Roc'o soils. These minerals were identified by the strong peaks at l6.i», 20.6, and 26 K (Blanchard, 197^). As has been pointed out by V/ebber (1978), hydrous aluminum oxide plays an important role in binding phosphate in acid soils and clays; variscite formation is expected in acid tropical soils, but amorphous Al phosphate would be expected to persist indefinitely in other acid soils (Webber, 1978). In these tropical acid soils, high P application may tend to produce precipitation of (AlFe)P04 compounds. This could explain why P is more deficient in tropical acid soils than in acid soils from temperate regions. Negative Surface Charge Ac id i ty To determine the release of protons in these surfaces a complete study of acidity sources was carried out through intensity and extractable acidity measurements. The pH of a soil suspension is sometimes lowered or increased by the addition of an electrolyte. Electrolytes such as KCl, K2SO1+, and CaCl2 lower the pH while higher pH values are found with NaF in soil suspension. The sign of the difference between soil pH in KCl, an indifferent electrolyte, and soil pH in water has been related to the sign of the net surface change of soil colloids (Mekaru and Uehara,

PAGE 88

7^ 1972). The difference between pH In KCl and K2S0^ has been associated with anion adsorption (Adams, 1976). The pH in NaF has been related to the release of hydroxyl groups in Andosols (Calhoun and Carlisle, 1971). The measurements of pH showed a wide range of variation from very strongly acid in El Potrero to weakly acid in Guanipa 5A. All pH values in water in the soils examined were decreased with addition of KCl. Except for Guataparo and El Potrero soils, pH values in water were decreased with addition of K2SO4 and CaCl2. The pH values in NaF ranged from 8.^10 to 10.25. These values tended to be higher in the Oxisols, Ultisols, and Alfisols than in the less weathered soils (Table 29). The negative values of ApH(KCl-H20) indicated that these soils had a net negative charge. The negative value of ApH(KCl -K2S0^) indicated a high capacity of those soils to exchange anions and, therefore, to fix phosphate. The ApH (KCl -K2S04) values were caused by the fact that the amount of hydroxyl ions displaced by sulphate anions was greater than the Al displaced by K ions (Adams, 1976). The very high negative values of ApH(H20-NaF) indicated a large displacement of hydroxyl groups linked to Al and Fe oxides in these soils (Calhoun and Carlisle, 1971). The values of extractable acidity and Al with IN KCl and those from potent iometr ic and conductometr ic titrations are shov/n in Table 30. The IN KCl -exchangeable Al and acidity titrated with Ba(0H)2 ranged from 0.02 to 3.90 and 1.30 to I6.OI meq/lOOg, respectively. The higher values of + 3 exchangeable acidity compared to exchangeable Al may have been caused by the contribution of polymeric Al and aluminate reactions that occurred during shaking and titration in the Ba(0H)2 (Dewan and Rich, 1970). For Venezuelan soils potent iomet r i c and conductometr ic titration values with NaOH, Ba(0H)2, and Na2Bit07 are shown in Table 30. At pH 8.00,

PAGE 89

n: Q. < -dO I o Jo to 1^ o 3: LAr^csjOLAOOLrvLALnLALrvLrv — I — LAr~»-s-Lr\vDcMcricNjr^-3-r--. I I I I I I I I I I I I I oo-a-ooLn-s-cjLrvm-— — -:j-tr>o -3LTWD LALAtv^-aLaur\LAvO LOLn ooooooooooooo I I I I I I I I I I I I I Cr\0^-3CM\£> < CO o CM LA LA St CO an ro er < CO (D fD (U fD 0) L1/1 >DD. Qu o CO Ol D. 1+-1 fD fD fD o O fD fD JD o C CO 00 C C C oc O C +-> E fO (D nj c 1 (D fD fD 1_ TO D o 3 (D D Z3 fD (U Q) lo CJ CO LU CO C3 C3 < LlJ a:

PAGE 90

76 Table 30. Acidity from potent iometr Ic titrations at pH 8.00 and conductometric titrations with NaOH, Ba(0H)2, and Na2B;^07 for the Venezuelan soils examined. Potent iometr ic titration Conductometr ic t i tration Soil at pH 8.00 with with NaOH Ba{0H)2 Na2Bi+07 NaOH Ba(0H)2 Na2Bi+07 meq/1 OOg Guanipa 1 2, ,0 2, .6 1 , .6 22 .8 23 .8 12, ,0 Guanipa 5A 2, .0 2, .8 1 , .6 27 .6 27 .6 20. .0 Guanipa 5B 1, .2 1 , .6 1. ,6 30 .4 26 .1 12. .0 San Cristobal 12, .8 6, .0 8, .0 50 .4 39 .6 12, .0 El Roc To 8, .A 10, ,6 10, .6 i*6 A 34 .8 10. .4 Bajo Seco 7. .2 6, .0 6, .0 k5 .6 24 .4 7. .0 Guanaguanare 11 . ,2 10, .0 10, .0 38 .4 32 .7 44, .8 Guataparo 7. .6 6, .i» 7. .6 28 .4 19 .8 24, .4 Alambre 11 , .2 13. .0 13. .0 61 .6 31 .5 58, .0 El Potrero 16, .k 14, .0 16, .0 46 .0 41 .7 22, .8 Barinas 3. .6 i». k. .0 Zk .0 28 .5 24, .4

PAGE 91

77 the acidity titrated ranged from 1.2 to 16.4 meq/lOOg in NaOH, from 1.6 to 1^.0 in Ba(0H)2, and from 1.6 to 16.0 in Na2Btt07. The soils with lowest and highest values were Guanipa 5B and El Potrero. The acidity titrated at this point was probably due to dissociation of organic matter groups, SiOH groups, and exchangeable and polymeric Al (Jackson, 1963). The values obtained with Ba(0H)2 and NaOH were similar except for the San Cristobal soil. This suggested that there was no exchangeable or bonding of H to permanent negative charge in these soils (Shainberg and Dawson, I967). As pointed by Shainberg and Dawson (I967), the acidity titrated with Na2B^07 is closely related to permanent negative charge (Table 3). The values obtained with conductometr ic titrations indicated high values of exchangeable Al , probably caused by dissolution of minerals with release of structural Al or reaction of bases with organic acids (Dewan and Rich, 1970). The higher values of extractable acidity shown by the soils compared to IN KCl exchangeable Al is good evidence of Al retention by organic matter. There was no evidence of exchangeable H in potent iometr ic and conductometr ic titration curves. The low values of Al in some soils were probably due to the fact that Al was not exchangeable by IN KCl, even though potent iometr ic and conductometr ic titrations detected it in those soils. Lime Requirement El Potrero and Alambre soils had the highest buffer capacities and lime requirements (Fig. i») . These soils reached pH5.80 only after application of 10 metric tons of CaC03/ha. Guanipa soil had the lowest buffer capacity because with 1 metric ton/ha it reached pH 6.6O and with 2 metric tons/ha, pH 7-30. In general, lime requirements were as expected because

PAGE 92

78 1 Fig. ^. Effects of CaCOs applications after 6 months incubation period. (Guanipa 1=1; San Crist6bal=4; El Rocro=5; Bajo Seco=6; Guataparo=7; Guanaguanare=8; Alambre=9; El Potrero=10; and 3arinas=ll.

PAGE 93

79 soils differed in amounts of organic matter and clay. From the results, one can calculate the amount of lime needed to adjust these soils to the desired pH. The amounts of CaC03 needed to neutralize exchangeable Al (pH 5.50) varied with soils and in El Potrero soil was approximately 10 metric tons/ha (Fig. 't) . Potassium Q/l Parameters To investigate ionic selectivity of the negative surface charge in these soils, a study of adsorpt ion-desorpt ion of K related to Ca was conducted. The Beckett Q/l relationship was studied. The parameters found were not related to the position of soil in soil classification systems, but to the minerals predominating in the clay fractions. The shapes of the curves resulting from Q/l relationships were the same for all soils studied (Fig. 5). Exchangeable K ranged from 0.02 to 0.05 meq/lOOg (Table 31). The highest exchangeable K values occurred in the less weathered soils. The presence of mica (Table 3) could explain the highest values of exchangeable K. The AR^ values ranged from 0.0017 in Guanipa to 0.028 in Red Bay Al . It is assumed that those soils k with low AR^ values may have high K-fixation capacities. Red Bay B2t showed a value of 0.01 meq/lOOg for aK° while Bajo Seco had 0.3 meq/lOOg. The high value of AK° for the Bajo Seco soil was probably a consequence of the presence of illite and other K fixing minerals. The highest values of PBC were found in San Cristobal and Guanaguanare soils. San Cristobal k had a higher PBC than Guanaguanare perhaps because of the higher clay content. These two soils had the highest CEC values, the difference in PBC values between them was probably caused by vermiculite in the San Crist5bal soil and illite in Guanaguanare.

PAGE 94

80

PAGE 95

a. X o cr: < t-J E CD D + o kjQ <_) JO L. o ds nb < (U 0) CD geab Ca+I c O Ex o to (1) r^LTvoor^r^iALAOvDOoocTi E ^r— cMLnLncsiocMrovDLA.— CNi cn\Doocococr>vDOv£>ooc O — E — ^ vo r-^ ^ vo La — — cn CD CO -arsl I I o o Cnoor^O r^O CTi LA -3" vO vO OOOCSIOCAOOOOOOO ooooooooooooo cr V E mrACM-3-oOOvD — r-^vDO^r-".-— OOOCMOCO — — O-— O-— O ooooooooooooo 0) o E r^vD cn CM — cMcr>0Acsir-~r~-.rA0A0OLrvuA'— OOOOO — — OOOOCslO ooooooooooooo OOOOOOOOOOOOO cMr^courvcsi-a-oooDooi^vjOcr\ o~>cr>LnvD'-Lricr\ur\crvi-AO^ cr» — oom-a-r^ooomcMcvir^o ror^corAor^>~Door^\oo^ — -00-d-r^CO-3-\DurvvO^LA-^ o ooo — ooooooooo o C" ni OOOOOLACMOOO-— CNirO E covDf'^vnLAca-a-'— -a--— Lnr^oo I r— OOcar^r^-— OOrO(N'— vOO co-3-csiouni-nLrva>Lr\oLAvDur\ OOOt-A-— LTVCN — — ra-— Oslo ooooooooooooo fD a) V•t-j < £0 fD O CM LA LA St o CO ro c fD er < CO TO \— Q) fD D L. W1 >>Q. Q. Q. 1O 00 D. cn i_ •M fD fD fD o o fD fD XI O C QQ CO c c C O *-i c E D_ fD (0 fD c fD fD fD 1_ "O XI fD fD fD (U (U C3 CD t3 LiJ OQ O O < LU q:

PAGE 96

82 The K O./l parameters found in Red Bay tended to decrease with depth. The decrease of K with depth in this soil and in Guanipa 5 was probably caused by fixation by the increasing clay contents in deeper horizons. Decrease of AR with depth seemed to indicate that available K depended e on K adsorbed on organic matter. The AK° values indicated that there k was little labile K in these soils. Values for PBC tended to increase with depth in Guanipa 5 probably due to the effect of CaCl2 in increasing pH-dependent CEC by removing substances blocking the exchange surfaces, and increasing the reactive surface participating in the Q/ 1 exchange equilibria (Lee, 1973). The values of K, Ca, and Mg shown in Table 31 were used to calculate the Gapon selectivity coefficient (KG) for those soils. The highest value of Ca + Mg adsorbed was shown in El Rocfo soil while the Guataparo soil showed a negative adsorption (desorption) of Ca + Mg. The latter indicated a high selectivity for K. All Gapon selectivity coefficient (KG) values were larger than unity indicating a high selectivity for K, especially Guataparo and Red Bay B2t. These results indicate that the negatively charged sites for exchange in these soils in their "natural" conditions have higher selectivities for monovalent cations. It also suggests that the charge density is low, and some soil surfaces (San Cristobal and Bajo Seco) have specific sites to fix ions with ionic radii with sizes similar to K. Positive Surface Charge In order to characterize positive surface charge, several studies of P adsorption and factors affecting it were carried out. Langmuir and Freundl ich isotherms were used to determine the behavior of P adsorption in these soils.

PAGE 97

83 Phosphorus Adsorption Correlation coefficients between the adsorbed P (x/tn) and P concentration in soil solutions at equi 1 ibri urn were significant for all soils which permitted Langmuir adsorption maxima to be calculated (Fig. 6). The data showed that at any particular final P concentration, P sorption was largest for Guanaguanare soil and least for Barinas soil (Table 32). However, the overall shape of the isotherms was remarkably similar. Each isotherm was characterized by a large change in the amount of P sorbed over a low P concentration range followed by a more gradual change in the amount of P sorbed with increasing P concentrations. To facilitate comparison with other physical and chemical processes, K values for each soil were transformed into corresponding free energies for sorption using the relationship AG = -RTlnK. This relationship, which equates K with the equilibrium constant, assumes that the activity coefficients of occupied and unoccupied sites are the same (Ryden et al . , 1977a). Values of AG and b for each temperature obtained for each soil are given in Table 32. An interesting feature of the equilibrium data was the similarity of AG values for a particular temperature. The type and properties of the soils used were only reflected in the magnitude of the sorption maxima (b) obtained for the temperature used. Data did not fit the Langmuir isotherm for all temperatures. There were deviations from linearity, and because of this, the Freundl ich isotherm was used to fit data from electrolyte and temperature-effect experi ments. Temperature effects. The P adsorption isotherms were determined on all soils used in this study at 5°, 25°, and hO°C. The extent of

PAGE 98

8A t I Fig. 6. Langmuir adsorption isotherms for San Cristobal soil in nonsaturated systems at three temperatures.

PAGE 99

85 Table 32. Phosphorus sorption constants at different temperatures from the Langmuir equation. Soil AGs AG25 AGijot ^5 625 b^Qf Kcal/mole mmole/g San Cristobal -2.96 -3. .16 -3 .37 59. 69 75 .02 84.89 E 1 Roc To -3.07 -3. .53 -3 .43 56. 59 71 .68 82.71 Bajo Seco -2.95 -3. .49 -3 .85 53. 76 56 .59 63.25 Guataparo -2.85 -3. .29 -3 .46 50. 56 49 .63 49.63 Guanaguanare -3.14 -3. .85 -4 .29 62. 04 80 .65 97.74 Alambre -3.29 -3. ,66 -3 .89 70. 13 73 .31 82.71 El Potrero -3.01 -3. .39 -3 .53 68. 64 70 .13 80.65 Barinas -2.89 -3. .25 -2 .99 44. 19 46 .75 84.89 Red Bay A1 -2.94 -3. .47 -3 .47 48. 88 50 .40 57.69 Red Bay B2t -3.10 -3. .22 -3 .59 46. 75 56 .59 53.74 tlhe AG is the free energy of sorption from the Langmuir sorption energy constant and b is the sorption maximum.

PAGE 100

86 adsorption is shown by the data from the Freundl ich equation in Table 33 for selected soils. These data showed that although the values of K changed with temperature, the value of (l/n) remained similar. The constant K varied widely among soils. The values of K ranged from 10.2 to 1^)7.9 at 5°C, from 10.7 to 199.7 at 25°C, and from 10.7 to 263.1 at kO°C. Sandy soils showed the lowest K values. The values of the exponent of the Freundl ich isotherm (l/n) were in all cases between 0.33 and 0.50. The linearity of the plots of equation (2) (Fig. 7) implies that a sufficiently wide spectrum of sites existed in the total surface to satisfy the condition that the bonding energy decreased exponentially with increasing site coverage (White and Taylor, 1977). The larger values of l/n were associated with soils having relatively fewer kinds of adsorption sites. It is possible that low values of I/n could occur with an adsorbent containing more than one type of adsorption site (Mukhtar, 1976). An increase in temperature increased adsorption of phosphate in these soils suggesting that greater number of new sites would become available with increasing temperature (Fig. 7 and Table 33). This indicates a strengthening of the attractive forces between the phosphate ions and the soil. One would expect the standard enthalpy associated with the adsorption process to be endothermic. These new sites were presumably formed by breaking some bonds as pointed out by Muljadi et al. (1966). The heat of adsorption may thus be divided into two parts: (a) that due to the effect of temperature on the equilibrium constant (K) of the exchange process, and (b) that due to the irreversible increase in the number of adsorption sites, i.e., an endothermic process.

PAGE 101

o 0) 13 4-1 c 0> l_ 0) ItM— "O 4-1 c o 4-> O. o (/I in i_ o f~ Q. m O r~ D. u o M(/) c o 4-1 (D 3 CT 0) jr o X) V o c c 3 i 0) TO 1_ X U(U DC O -aLA LTV CM a: -QXl Q) CD— C D.cr4-' > X cr\cr\cr\cr>OAcnj — — LnvOcr\0^oocnvO-3ooovoraLar^i — 3--Tcor-^vD-:rr<-vOLnr-^o (0 JQ u 4-1 < CQ \o ro o CNJ St v2 CO ro an er < 00 ro ro ro 0) ro 3 (U i_ in > >. D. Q. Q. 1_ o to Q. Cn v_ ro ro ro o o ro ro -Q o c CO CQ c c c q; o 4-) c E ro ro ro c • — » ro ro ro i. TJ TJ o 3 3 ro ro 3 3 ro (U (U to o O O to LU CO O O < UJ CQ a: a: >it — ro J3 O i_ Q. 14O > LTV 4-1 ro >^ 4-> c ro u i^ c in *l Q Q H) 1 1 4-» in Q) E 4-» ro m u> 0) C f— ro lU '~ 4-J > j» u 3 21 u in CI in ro > o o \3 O c >, x: x> 4-1 o i Q) Dl in X) a> 3 3 •— > ro 01 > ro

PAGE 102

88 C (yg/ml) Fig. 7. Freundlich adsorption isotherms for P by Guanaguanare , San Cristobal, and Red Bay B2t in nonsaturated systems at three temperatures.

PAGE 103

89 Ionic system effects . These soils were selected to contain different proportions of particular components and to have different properties with respect to P sorption. The soils differed appreciably in their ability to sorb P from Ca-, K-, and nonsaturated systems as shown by the isotherms (Fig. 8 and Table 3^). The composition and concentration of supporting media had marked effects on the amounts of P sorbed. For Guanipa 1, Guataparo, Guanaguanare , Alambre, and Barinas soils, the P adsorbed followed the order Ca > K > nonsaturated, while San Cristobal, Bajo Seco, and El Potrero soils showed K > Ca > nonsaturated. The sequence of isotherms for El Rocfo soil appeared to be somewhat different because the Ca-saturated system adsorbed more P than the nonsaturated and this more than the K-saturated system. The difference in the P sorption isotherms obtained with the three supporting media demonstrated an effect of cation species on P sorption.The high P adsorption from the K-system is related to values of AK° and Kx in Q/l curves shown by San Cristobal, Bajo Seco, and El Potrero soils. The average pH values of supernatant solutions of the various supporting media following an 11-day sorption period were higher in nonsaturated than in K-saturated and this higher than Ca-saturated system in the Guanipa 1, El RocTo, Bajo Seco, Guanaguanare, and Alambre soils while San Cristobal, Guataparo, and El Potrero soils showed a sequence K > non> Ca-saturated system. Furthermore, for the Barinas soil the order of decreasing pH values of the supernatant solutions of various supporting media were the same as the order of decrease in P sorption from respective support media (Table 3^). This soil was the one with largest difference between pHKjSOit and pHKCl; this ApH value has been considered to be related to the adsorption of anions in soils (Adams, 1976).

PAGE 104

90 O K None 8000 ^ Ca Adsorbed P 1 1 1 1 1 I I I concentration (mg/lOOg) C (yg/ml) Fig. 8. Freund) ich adsorption isotherms for P by Guataparo and San Cristobal in different ionic systems at 25°C.

PAGE 105

a: X3 I/I I (0 o M CC o i) 4-) (D u 4-1 in T3 (D 1ZJ 4J fC c o cr>cr»oa^tM.— c^cno cnoo r^cncnmoocooo ooooooooo o-ico r^cricrir-.-3laoo ooooooooo ^LTVCMCMr^ra — vDLA cr>Lr»cr\u-\oravOcMr^ CM CM CM (v^ CM CM ra-T fot3>+O)-— I— c a.»CM — raCTi-dl^cooo-aLAcncnco cncTvcrvoo en ooooooooo r^vx> cMco r»-\r — 3r«-»r— ooo — ooooo r^— <7^u^fM — uTvo-aCO Lntv^o CTioo LAcao CM — r^O'— — CslcM^ <0O O o C (/) o u 0) (0 o 03 fa X) o c c q: O •t-> c E TO c 03 03 03 u o D m D 03 CO C3 CO CO C3 O < LU CO 4-1 o 4-" c 1/1 0) o XJ a 03 u 03 c 4-J 03 4-> Q£ 03 03 flJ CL E 03 4-> (/I D 2: 4-" i/i >t: XI s c XI 3 03 O 5 O > XI 'o XI u03 cn l/l XI 3 o • — 1 1. in c 03 j:: >4-' 4-» i i/i XJ 03 03 XI o 03 L. > Q.

PAGE 106

92 It has been shown (Helyar et al . , 1976a) that increasing CaCl2 concentration increases P adsorption. Calcium increases adsorption because it complexes with adjacent adsorbed phosphate ions, reducing the repulsion force between them. The higher effect of Ca than K on P adsorption in some of the soils suggested a specific association between Ca and the adsorbed P. In soils which contain exchangeable Ca, additions of any neutral salt displace Ca into solution. The increased Ca concentration in solution could then lead to increased P adsorption by the hydrous oxides as a result of the specific effect. Consistent with this are reports showing that the relative effect of the common cations in increasing P adsorption by soils increases in the order Na > K > Mg > Ca. This same order would be expected if the increased P adsorption were due to increased Ca added to or displaced into the solution phase (Helyar et al . , 1976a). It has been suggested (Helyar et al., 1976b) that two adjacent adsorbed P ions could form a stable surface complex with a cation if it were divalent, if it were able to form a coordinate bond with two pairs of resonant 0.5 charge oxygen atoms in one plane, and if it had a radius of about 1 ^. The higher adsorption of P in K-systems in some soils was probably due to the high capacity of those soils to fix K and a probable occlusion or adsorption of K phosphate, e.g., (K^H2P0i^) into an amorphous or semicrystalline region of the clay surface as pointed out by Barrow and Shaw (1975b). Thermodynamic parameters associated with adsorption process . Thermodynamic parameters are thought to be related to bonding energies and can be useful in determining mechanisms of adsorption. Enthalpy (AH) is heat content and is an extensive state property. It represents the amount of heat evolved in the adsorption of a definitive quantity of solute on an

PAGE 107

93 adsorbentThe bonding energy is stronger as the value of H is more negative. The symbol G is a property of the state and is called free energy of the system or Gibbs free energy. The value AG gives an idea about equilibrium conditions. If AG = 0, the system is at equilibrium, while if AG is negative the process tends to proceed. A positive value of AG means that the reaction tends to reverse. Also, another quantity called the entropy change (aS) as been found to affect the equilibrium constant of a reaction. The entropy change is an extensive property of the system. If entropy is less than zero, the cycle of any reaction will be reversible (Adamson, 1976; Biggar and Cheung, 1973; Mukhtar, 1976). Thermodynamfcs of P adsorption were calculated using Freundl ich equation for the sorption of P, the equilibrium constants obtained from the relationship between solution P concentration and reciprocal of time, and Biggar and Cheung (1973) methods. The values of AG, AH, and AS are shown in Tables 35, 36, and 37. The limitations of these calculations are that no acceptable method of defining and measuring the activity coefficient (used in G calculations) in adsorbed liquid phase has been devel oped . The AG values calculated using the K value obtained from plots of concentration of P solutions against the reciprocal of the time at 25°C are shown in Table 35 and Fig. 9The AG values were negative for all soils and ranged from -1.39 to -3.52 Kcal/mole. The values of AG obtained were all negative ranging from -1.29 to -3-^5 Kcal/mole for temperatures ranging from 5° to ^0°C (Table 36). These values correlated well with those found from the equilibrium concentration-reciprocal time relationship at 25°C. The AH and AS values were positive. The

PAGE 108

1 E 0) TO O < — U iO 4-1 He C < (U -o 1 — O (U C -M O O n) o — 3 n — • I(0 o o c (/) o o f— — «a 4-> o D -O in — C o m ^ +J c 0) Q) u (U ^ O 3 +J M(U -o E D • — cx 1. JD c — O -M cr c a. 4-1 0) uo 1_ o D— x: UJ 4-1 M LA «) JQ ID O 1CO OLAO r-^ooo rocnuA-a-oooooo OOCSCMCS'— — — CsICMf— — CM I I I I I I I I I I I I I ^CM^O-— CACM. — OCAtMCALA LACAUACMOO — rAvO-3-OOOCM -a-LArACMCACMCSI^rAfAvD CMtNlu^CNjrA— '-OO — OO-3-rA oo CAOOOOCOCO CAOOOOCOCOCOOO ooooooooooooo 1 4-1 4-' 1 1 +-> 4-> 4-> 4-> •M 1 1 4-1 4-> 1 4-1 1 1 4-> 4-J +-I O O O •— LA CNl OO -3cA cr> CO vo o CM — (Ti cn — -dCA CM LA LA LA r~. cn 1 — v£> LA VO + + + + + + + + + + + + + -dLA LA vD — 00 1/1 o CO ro an er < CO TO 03 (D — \— 0) ro ZJ (U in >>O. CL QLo oo Q. CD L. 4-1 ro ro ro o ro ro X> o c CO CO C C C a: o 4-1 c E Qu to TO ro c ' — i ro ro ro i_ -o -o 3 ro ro D ro 0) Q) C3 O (/) LU CO CD CD < UJ DO a:

PAGE 109

50 © RED BAY B2t SAN CRISTOBAL A GUATAPARO Fig. 9. Relationship betv^een P solution concentration and the reciprocal of time for Guataparo, San Cristobal, and Red Bay B2t soils.

PAGE 110

96 o 1 1 Jto 1 <] 1 1 1 1 o >«i o in 0) (/> c eg "o o < E \ 4-1 ro ro 3 u cr 0) • 1 T5 in 1 0) to 1 o C <3 1 E 1 •a ro o (U c Jo o o o D 4-1 (J ro ro [ < D l_ ! Q I/) 1 o XI in ro i < ro c 3 >T3 O OL. (/) 0) o -C a. Iq TO o 1in r^r^\DcrioocsiocNjoooo-a-oo 0~\C0 urvLOvD CM roCTiO — vOvDvD — CMCN'— — cs — — — — OvOOOOO"— vXJvDr^-— — oo — oovo LAr^LTvt^r^ — ooo crvcnupi a~>oo ununvD c<\ a^ o — \d \o \0 — CMCSl-— >— CM — — — — o^— r^cxDr^LTvcor^vD-c— — — oo r~--3-vr>-a--3-oo LACPvLrv < ca \o ro o CM LA LA St o CO ro an er < CQ ro ro ro \— (U ro 3 (1) >. D. QQ. u o in Q. Ol 14-> ro ro ro o o ro ro XI o c CO CQ c c c oc O 4-1 c E Qro ro ro c • — t ro ro ro 1TD -o 3 3 3 ro ro 3 3 ro 0) o t3 O i/i LU CQ C3 < LU CD cc: o -aT5 c ro LA CM (U i3 4-1 ro u
PAGE 111

o 1 0 o 1 [ 1 o o o o < ur\ 'o rsi E o 0 LA < CM O a LTV o LA CM O o LA (vj-crcvjooo-crcN)ooooo-aOOOOOLACMrA-— LA— CTvvD -3-v£>COOcMLAr^vO — CMO CM — — CM rAOOoA-a-OCMOOt^CM-aLAP^CALALArACALA — CX)vO — — — CM coocrivDcMcMoor^LA-3r^oo cri-3— rA-— LA — cnvo -3-vDr^OCMLAr-~v£> — cso CM — — CM ^vO' — ^X>^OcA<^^rACM— CO CMOOCA-a-CACM-a-r^'— vO >— — — LAfM — — — OrALA t^OO OcOvO-3--arALAcAOO cMcMvocnoo-3-r^rACNicr>r~ooooooooooo — rA-3-vDoOvO cr\rACA — r-» — CM -3c>A^ rAsO CM CM-a--3-rACM — — CM — CM — OA— — CACM r^vOLA — OOOO^OOLA-3-CM — Or^LA^ — LA-3-CM-d--3— — — — — CM — CM' — — CM — O ra
PAGE 112

98 values of Ko, AG°, AH°, and AS° (Table 37) associated with the adsorption were calculated, respectively, according to equations (5), (9), (10), and (11) (Biggar and Cheung, 1973). Values of AH° were calculated from plots of In K vs 1/T. The linear nature of the plot indicated that the mechanism of adsorption did not change as temperature vias changed. The amount of adsorption was changed because the supply of thermal energy was different. The AG° values v;ere in the range of -0.03 to -0.98 Kcal/ mole between temperatures 5° and hO°C. The negative values of AG° indicated adsorption processes which were expected because of the reported fixation of P in these soils. The AG values are related to the amount of P adsorbed. It can be observed in Tables 35, 3^, and 37 that soils with higher values of AG (more negative) adsorbed more phosphate. Even though the values in Table 35 have the limitation mentioned before, it was clear that the lower the equilibrium concentration the more negative the AG values. If one examines the thermodynamic parameters as contained in Table 37, one notices that AH° and AS° for the adsorption process with the different soils ranged from 0.12 to 5-68 Kcal/mole and 1.17 to 20. cal/mole degree. The drastic differences in AH° and AS° values indicated different adsorption mechanisms. Also, change in free energy is related to adsorption energy; the more negative the AG° the greater is thought to be the bonding strength. Thus, it is correlated with mechanisms of adsorption (Mukhtar, 1976). The positive values of AS in phosphate may be the result of replacing more ordered water molecules which may be adsorbed on the surface in a less orderly fashion (more entropy), and consequently gives a positive entropy change. A positive entropy change is also commonly observed in formation of a complex where a ligand displaces several

PAGE 113

99 water molecules from around a central ion (Muljadi et al., I966). The positive entropy change observed for the phosphate adsorption may be attributed to a disordering of water molecules around the adsorbing site and the adsorbate. Muljadi et al. (I966) suggested that the adsorption of one mole of phosphate corresponded to the liberation of about 2 moles of water. The high positive value of AS in these soils may be related to the P fixation on Fe oxides. Bajo Seco, Guanaguanare , and Red Bay soils showed the highest values of AS (Table 37) and from P fractionation data (Tables 7 through 19), these soils showed a tendency to fix more P as Fe than as Al . This P fixation by Fe and also the contribution of Al compounds increases the amount of water displaced from the soil surface (Muljadi et al., I966). These results agree with Ryden et al. (1977a) who proposed a mechanism for three regions. During P adsorption the surface becomes negatively charged because P adsorption involves potential determining mechanisms. The portion of surface active in P adsorption may be represented as a series of single coordinated -OH groups, some of which may be protonated. The proportion of -OH2 to -OH is determined, among other factors, by the pH and PZNC of the surface (Ryden et al . , 1977a) . It is clear that there is more than one type of interaction site on the soil surfaces. The interaction process of the phosphate ions with the soil usually involves more than one mechanism depending on the physical and chemical properties of the soil, its major clay minerals constituents, the experimental condition, and the concentration of P (Mukhtar, 1976). However, a single mechanism may dominate. The above results suggested that in these tropical soils phosphate ions are bonded to positive charge in Al and Fe compounds by displacement of -OH2 and -OH groups and that in some soils (Guanaguanare,

PAGE 114

100 ] Bajo Seco, and Red Bay), the contribution of Fe oxides to P adsorption was larger than in the others. Phosphorus Release In Table 38 and Fig. 10 the values of P release after adsorption equilibrium are shown as related to the number of washings for the Venezuelan soils at the highest P application. Values of b^ and a from the equation Y = b^ aX are shown, where Y = the accumulative P desorbed or released, and X = the number of washing, b^ = the intercept, and a = the slope of the straight line, when In Y is plotted against X. The data show that the amount of P released was quite different in each soil and the order of P release in the different ionic system varied with soils. In Barinas and Guanaguanare soils, the sequence of P released was K> Ca> nonsaturated system, while in Alambre and Guataparo soils, it was Ca> non> K-saturated. On the other hand. El Potrero and Bajo Seco showed a non> Ca> K-saturated system sequence. The sequence in San Cristobal and El Roc To soils was non> K> Casaturated system, while Guanipa shov/ed a very different sequence, Knon> Ca-saturated system. The P released tended to be the same in each additional washing. The linear nature of the plot of the In Y vs X indicated that the release of P becomes asymptotic with the number of washings. This behavior of soils with respect to P liberation is related to the strength of adsorption and the specific effect of each cation. Those soils with m i nera 1 og i ca 1 composition that increase K adsorption like Alambre, Guataparo, Bajo Seco, and El Potrero showed a low release ' of P into the K system. While those soils where the P was associated with Ca showed a low release into Ca systems. In Guanaguanare and

PAGE 115

c o •— 4-1 i_ I. o o u fO V c •— — -D c 15) c •— 1/1 £5 S «»w 1/) E U (U 0) 4-> if) E >D in C o <1> ^ c +-> r\ 1 T) C 4J TO C (U Cl 0) D <+V <4(A TO T3 i) 4-1 i 0) (/) > at soi 3 c E TO 3 O Qi U 3 TO N a) c (U 4-1 > c 0) (U x: (1) 4-1 5 4-> L. 0 o X) MD. (/) 4-1 x: c i/i (U c o o 4-1 uTO ^(U OJ o a: o -o (1) o t/l T3 TO X) 0) JD o (/) X> TO XJ 0) -.— cOLTvOCr^rALAO r^r^cAcsicx) lAoAcMoo CA' — CM.— CMCVICVJ.— OCMCMCA-^.— LAO-(T\ a\ 1_ 4-> TO o o TO TO XI o C C a: o 4-) c E TO c — 1 TO TO TO 1_ D TO TO 3 3 TO 13 to 1x1 ca O t3 < UJ ca

PAGE 116

102 Fig. 10. Release of phosphorus from Guanipa 1, San Cr i stoba 1, and El Roc To so i 1 s .

PAGE 117

103 Barlnas soils, saturation with K and Ca increased P release which may have been caused by reduction in P bonding strength to Al and Fe (Table 38). Changes in Charge Surface The increase of negative charge in soils with lime is known and has been attributed to ionization of organic matter groups and OH groups attached to Si of broken tetrahedral positions (Wiklander, 19^7) • On the other hand, the increase in negative charge with P adsorption has been attributed to the replacement of hydroxy] ions by phosphate at the surface of clay minerals and sesquioxides (Sawhney, ]37h) . Lime and P effects on CEC together with cropping effect on this "new" CEC are discussed in the following paragraphs. Lime and Phosphorus Effects Guanipa 1, Guanipa 5, and El Potrero soils were selected to determine the effects of combination of P and CaCOs on soil reactivity before and after cropping. Sorghum ( Sorghum bicolor L.) was chosen to measure the effects of CaCOs and P on dry matter yields and at the same time to observe the effect of cropping on changes in pH, PZNC, and CEC of soils treated with CaCOs and P. The effects of CaCOa and P treatments on pH, PZNC, and CEC before cropping were as expected (Tables 39, ^0, and h]) . There was a highly significant effect of CaCOs on pH increase due to the neutralization of exchangeable Al . On the other hand, the adsorption of P decreased the PZNC due to the neutralization of positive charge and to the reversion of charge by the negative charge of the phosphate anions. Calcium

PAGE 118

Table 39Effect of treatments on electrochemical properties measured before cropping for the Guanipa 1 soi 1 . Treatment pH PZNC EC CEC mmhos/cm meq/1 OOg 12 6.20b'^ 3.70ab 0.83c 5.64c 13 6.28b 3.60b 1 .OBabc 6.46b 6.35b 3.55b 1 .44abc 7.03b 22 6.25b 3.80ab 0.88abc 5.41c 23 6.50a 3.50b 1 .06abc 8.02ab 24 6.50a 2.30c 1 .30a be 17.65ab 32 6.28b 3.50b 0.63bc 6.94b 33 6.35b 3.40b 1.33abc 7.77ab 3't 6.60a 3.30b 1 .Olabc 9.70a k2 6.25b 3.50b 0.80abc 6.79b 43 6.40b 4.00a 1 .06abc 5.1 8c 44 6.60a 4.10a 1 .l6a 5.64c %a 1 ues differ j udged within rows followed by the same letter significantly at 5% level if probability by Duncan's Multiple Range test. do not as

PAGE 119

Table kO. Effect of treatments on electrochemical properties measured before cropping for the Guanipa 5 so i 1 . Treatment pH PZNC EC CEC mmhos/cm meq/1 OOg 12 5.60ab"^ 3.90a 0.'»8c 2.15c 13 5.50ab 3.65b 0.59c 2.65c 14 5.80ab 3.60b 0.95b 3.99c 22 5.68ab A. 00a 0.55c 2.09c 23 5.55ab 3.70b 0.64c 2.65c 2k 5.90a 2.50c 1 .05b 9.62a 32 5.53ab 3.70b 0.40c 2.58c 33 5.98a 3.60a 0.75bc h.Jkh 34 6.05ab 3.50b 1 .25a 5.46b i»2 5.53ab 3.70b 0.53c 2.58c 43 5.60ab 3.50b 0.59c 3.59c 5.98a 3.90a 1 .50a 3.51c Values within rows followed by the same letter do not differ significantly at SI level if probability as judged by Duncan's Multiple Range test.

PAGE 120

Table Al . Effect of treatmentson electrochemical properties measured before cropping for the El Potrero soi 1 , Treatment PH PZNC EC CEC mmhos/cm meq/1 OOg 12 3.20 b 0.78c 38.27 ci 13 4.70 d 3.00 b 1 .00 c 55.87c }k k.SSd 3.00 b 1 .81 ab 78.06 b 22 h.88d 3.40 b 1 .06 40. 13a 23 5.03d 3.15 b 1 .49 b 71 .44 be Zh 5.28 c 2.90c 2.44 a 122.85a 32 5.20c 3.25 b 1.13c 78.06 b 33 5.25c 3.80ab 1 .56 b 38.27 d 3^ 5.55 be 3.80ab 2.13a 59.97c k2 6.20 a 3.50ab 1. 14 c 158. 15 a 43 5.95 b 4.20 a 2.75a 59.97 c 6.18 a 4.30a 2.38 a 71 .44 be Values within rows follov;ed by the same letter do not differ significantly at 5% level if probability as judged by Duncan's Multiple Range test.

PAGE 121

107 carbonate treatments at high levels Increased PZNC. This was probably caused by precipitation of calcium phosphate and also to the effect of Ca on the exchange complex. Divalent specifically adsorbed cations increase the PZNC of colloids with constant potential surface and, therefore, decrease CEC (Breewsman, 1973; Adams, 1976). In these soils, the combined effect of decreased PZNC and increased pH brought about an increase in CEC as calculated by the Van Raij and Peech method (1972). The effect of CaCOs and P treatments before cropping were not so marked as after cropping perhaps due to the shorter time of contact and also to an enhancement of reactions by the crop. The effect of combination levels of P and CaCOs on pH, PZNC, GouyChapman and Stern potentials, and the distribution of charge on the surface for the selected soils after cropping were calculated using Van Raij and Peech procedure (1972) (Tables hi through 50). The variables examined showed differences among soils. The pH increased with increasing levels of CaCOs and P. The values of PZNC were changed significantly by increasing levels of P. The PZNC decreased with P levels at low levels of CaCOa, but increased at high levels. The effect of P on pH values may have been caused by exchange of phosphate with OH groups (Rajan, 1978). The effect of P on PZNC has been attributed to the increase in negative charge. The increase of PZNC with phosphate application at higher levels of CaCOs was probably due to a precipitation of Ca phosphate compounds. The Stern potential is the electric potential at the boundary between the Stern layer and the diffuse layer. The Gouy-Chapman potential is the electric potential at the boundary between diffuse layer and bulk solution. The values of Stern potential are used to calculate the charge

PAGE 122

Table A2. pH and point of zero net charge for CaC03 and phosphorus treatments in Guanipal after cropping. Treatment combination pH PZNC 11 5.52def'*' A. 05a 12 5.50def 3. 'Ob 13 5.73bcde 3.35b ]h 5.97abc 3.'40ab 21 5.52def 3.65a 22 5.68cde 3.35b 23 5.71cde 3.10b 2h 6.23a 3.35ab 31 5.'»2ef 3.30a 32 5.68cde 3.55b 33 6.00abc 3.90b 3'* 6.07ab 3.65ab 5.22f 3.80a h2 5.65cde 3.00b k3 5.60ed 3.70b kh 5.80bcd 3.55ab Values within rows followed by the same letter do not differ significantly at 5% level of probability as judged by Duncan's Multiple Range test.

PAGE 123

Table k}. pH and point of zero net charge for CaCOg and phosphorus treatments in Guan i pa 5 after cropping . Treatment combi nat ion pH PZNC 11 5.37c"^ 3.90a 12 5.32c 3.60b 13 5.52c 3.60b ]h 6.72a 3.50ab 21 5.^7c 3.90a 22 5.^0c 3.25b 23 5.60bc 3.35b 2h 6.87a 3.90ab 31 5.62bc 3.87a 32 5.37c 3.15b 33 5.87b 3.55b 3'^ 6.80a 3.75ab i»l 5.^3c 3.60a hi 5.38c 3.55b h3> 5.A8c 3.35b hh 6.70a 3.95ab Values within rows followed by the same letter do not differ significantly at 5% level of probability as judged by Duncan's Multiple Range test.

PAGE 124

no Table ^'f. pH and point of zero net charge for CaCOa and phosphorus treatments in El Potrero after cropping. Treatment combinat ion pH PZNC 1 1 '4.25h'^ 3.50a 12 A.yof 3.05b 13 '.70f 3.00b ]h 5.00e 2.60b 21 4.i»8g 3.60a 22 5.07de 3.10b 23 5.12de 3.25b 2^ 5Mc 3.55ab 31 k.llf 3.50a 32 5.19de 3.0!,b 33 5.23d 3.65b 3k 5.75b 3.80ab 5.72b k.30a kl 5.98a i|.20a A3 5.97a 6.02a 3.55ab Values within rows followed by the same letter do not differ signifcantly at 5% level of probability as judged by Duncan's Multiple Range test.

PAGE 125

Table hS. Electrochemical potentials of CaC03 and phosphorus treatments in soil Guanipa 1 after cropping . Treatment GC Stern combination potential potential Vol ts 11 0.087g 0.086 12 0.]2h 0. 123 13 0. I4lbcd 0.139 ]k 0. ISIabc 0.150 21 0.110 0. 109 22 0.138bcd 0.137 23 0. IS'tab 0. 152 2k 0. 170a 0.168 31 0.125edf 0. 12^ 32 0. 126edf 0.125 33 0. 12i
PAGE 126

Table ^6. Electrochemical potentials of CaC03 and phosphorus treatments in soil Guanipa 5 after cropping. Treatment GC Stern combination potential potential Vol ts 1 1 0.106 O.IOScdef'*' 12 0. 100 O.lOOef 13 0.1 13 0. 1 12bcdef ]k 0. 170 0.168a 21 0.093 0.092f 22 0.127 0.126bcde 23 0.133 0.132bc 2k 0.175 0.173a 31 0.103 0.103def 32 0.131 0.130cbd 33 0.137 0.135b 3h 0.180 0.177a k] 0.108 0. lOSbcdef kl 0.108 0. 108bcdef 0. 126 0.125bcde hk 0. 164 0.l62a Values within rows followed by the same letter do not differ significantly at 5% level of probability as judged by Duncan's Multiple Range test.

PAGE 127

Table h7 . Electrochemical potentials of CaCOs and phosphorus treatments in soil El Potrero after cropping. Treatment GC Stern combination potential potential Volts 11 0.044 0.044h 12 0.097 0.097e 13 0. 100 0. lOOde 14 0.142 0.l40a 21 0.052 0.052h 22 0.1 16 0.115bc 23 0.1 10 0. 109cd 2k 0.111 O.llOcd 31 0.075 0.074g 32 0.126 0.125b 33 0.093 0.093 34 0.115 0.1l4bc 41 0.084 0.083fg 42 0.105 0.105cde 43 0.148 0.147a 44 0.146 0.l44a Values within rows followed by the same letter do not differ significantly at S% level of probability as judged by Duncan's Multiple Range test.

PAGE 128

Table 48. Charge distribution at the surface for CaCOs and phosphorus treatments in Guanipa 1 after cropp ing . Treatment comb i nat ion 11 0.93 12 2.32 13 2.97 14 3.33 21 1.75 22 2.88 23 3.hS 2i» 3.83 31 2.36 32 2.40 33 2.31 34 3.05 41 0.83 k2 3.42 43 1 .83 2.69 02 a 0.61 0.62fg 1.27 1 .29dc 1 .80 1 .85bcd 2.21 2.28abc 0.98 0.99efg 1 .67 1 .71bcd 2.26 2.33ab 3.11 3.24a 1 .32 1 .34de 1.34 1 .37de 1.31 1 .34de 1.85 1 .90bcd 0.57 0.57g 2.52 2.63ab 1 .01 1 .03 1.52 1 .55cde Values within rows followed by the same letter do not differ significantly at 5% level of probability as judged by Duncan's Multiple Range test.

PAGE 129

115 Table kS. Charge distribution at the surface for CaC03 and phosphorus treatments in Guanipa 5A after cropping, Treatment combination 02 cf meq/1 OOg 1 1 1 .80 1.1^ 1 . 16bcde 12 1.^7 0.86 0.87de 13 2.02 1.13 1 .ISbcde ]k 3.96 3.60 3.79a 21 1.19 0.7^ 0.75e 22 2.63 \.k5 I .i*8bcd 23 2.88 ].6k 1 .68bc 2k k.lk 3.66 3.83a 31 1.62 0.93 0.95cde 32 2.80 1.57 1 .60bcd 33 3.00 1 .8^* 1 .89b 3'» A. 32 3.97 i*.l8a h] 1.79 1 .01 1 .02bcde m 1.78 1 .00 1 .02bcde k3 2.58 1 .42 1 .ASbcde kk 3.94 3.0k 3.17a Values within rows followed by the same letter do not differ significantly at 5% level of probability as judged by Duncan; s Multiple Range test.

PAGE 130

Table 50. Charge distribution at the surface for CaCOs and phosphorus treatments in El Potrero after cropping. Treatment combination Oj 02 a meq/1 OOg 11 5.58 5.88 5.91h 12 32.65 19.57 19.78def 13 35.2i( 20.70 20.93def \k 78.71 A6.3? hj. hGa 21 7.50 7.26 7.30h 22 28.25 28.67bc 23 ks.u 25. 18 25.52cd 24 i*6.l8 25.70 26.05cd 31 16.28 12.20 12.29gh 32 62. 7^* 35.10b 33 29.07 18.00 I8.l8efg 3A 50.^5 27.91 28.33bc 21 .90 1^.80 1^.92fg k2 i»0.01 22.82 23.10cde 8^.95 52.85 5^. 33a 80.i»6 52. A2 54.01a Values within rows followed by the same letter do not differ significantly at 5% level of probability as judged by Duncan;s Multiple Range test.

PAGE 131

117 in Stern layer and diffuse layer as outlined by Van Raij and Peech (1972), while the Gouy-Chapman potential is used to calculate the charge on the surface (equation 1^). This model presents difficulties and is not adequate to describe realistically the relationship betv/een the surface charge and the surface potential at high surface potential values. It was used in this study because the potential surface values were low, and this model could apply to these soil surfaces. Gouy-Chapman and Stern potentials, increased with increasing P levels and these effects were highly significant. Calculations of CEC and the charge in and 02 are somewhat complicated and equations (l6), (17), (l8), and (19) were used. The value taken for was lO^^cm ^. The value of was 0.15 8 which was found by Van Raij and Peech (1972) to give the best fit of their experimental results; specific adsorption of ions was assumed to be negligible, thus $ = 0. These substitutions finally yielded four equations with four unknowns. The solution was found by using the computer program described in Appendix 111. Even though there was an increase in the Stern-layer charge (oj) and diffuse-layer charge (02), the effects of CaCOs levels were not significant. Munns and Fox (1977), working with an Oxisol, found that the CEC (sum of exchangeable cations) increased three fold as the pH increased, and this additional charge was almost equivalent to the amount of CaC03 reacted. On the other hand, P levels had a highly significant effect on increased charge. The Gouy-Chapman model showed values similar to the Stern model. It seems that the potential values on these surface are low enough to allow the Gouy-Chapman model to apply. The combination of increasing pH and decreasing PZNC resulted in an increase in CEC in all soils.

PAGE 132

118 The changes in soil CEC with increasing pH have been attributed to the pH-dependent CEC. Soil reactions contributing to this CEC are organic matter and clays with noncrystalline components and interlayered Al (Fiskell and Zelazny, 1971). In El Potrero soil, this increase should be greater than in the other two soils because of higher organic matter and clay contents. Increased CEC with phosphate ions can be attributed to the replacement of hydroxyl ions by phosphate ions on the surfaces of clay minerals and sesquioxides in soils with pH 5 to 6. In soils, vj\th lower pH values, a portion of the surface hydroxyl s are protonated to form water molecules. The replacement of water molecules by phosphate ions also increases CEC (Breewsma, 1973; Mekaru and Uehara, 1972; Sawhney, 197^). The increasing levels of P had a highly significant effect on electrical conductivity (EC). The values of EC increased with increasing P. In El Potrero soil, the EC reached values greater than 2 mmhos/cm at 25*'C in the saturation extract; this is considered the limit of salinity tolerance for some crops (Hasan et al., 1970). Effects of Change of Charge Characteristics on Crop Production There were significant differences among soils in sorghum dry matter yields (Tables 51, 52, and 53). El Potrero had higher yields than Guanipa 1 and Guanipa 5. The latter soils did not differ in their effect on dry matter yield. Levels of CaCOs did not significantly affect dry matter yield. But, yields were increased significantly by increasing levels of applied P. On the other hand, dry matter production decreased with increasing salinity levels in soils. The decrease in dry matter production probably resulted from the decreasing availability of soil water and decreased uptake of nutrients by the crop , espec i a 1 1 y Ca and Mg associated with increasing salinity (Hasan et al., 1970).

PAGE 133

Table 51. Dry matter yields of sorghum and cation exchange capacity after cropping for soil Guanipa 1 treated with CaCOs and phosphorus. P appl ied as KH2P0it CaC03 appl ied 3 1 2 3 k 1 2 3 k Oven-dry plant material, g/pot 1 12b 3.23a 3. 'la 2.97a 1 .5^fg 3.59dc 4.77bcd 5.55abc 1 .02b 3.28a 3.25a 3.10a 0.92b 3.01a 2.9^*3 3.08a CEC, meq/lOOg 2.72efg /».55bcd 5.71ab 6.9^*3 3.68de 3.7'
PAGE 134

120 Table 52. Dry matter yields of sorghum and cation exchange capacity after cropping for soil Guanipa 5 treated with CaC03 and phosphorus. P appl ied as KHzPOit CaCOs appl ied 1 2 3 h Oven -dry plant material, g/pot 1 l.OSe"^ 1.05e 0.87ef 1 .Ole 2 3.87ab h.]3a 3.7^ab 3.73ab 3 2.86bcd 2.97cd 2.62d 3.30bc it 0.33fg O.lhfg 0.35fg 0.50efg CEC, meq/lOOg 1 2.9^bcde 1.93e 2.56bcde 2.79bcde 2 2.33de it.OSbcd ^.37bcd 2.79bcde 3 3.15bcde A.52bc ^i.Shh 'i.Olbcde h 7.56a 7.89a 8.29a 6.98a Values within subtables follov;ed by the same letter do not differ significantly at S% level of probability as judged by Duncan's Multiple Range Test.

PAGE 135

Table 53Dry matter yields of sorghum and cation exchange capacity after cropping for soil El Potrero treated v/ith CaCOa and phosphorus, P applied CaCOa applied as KH2PO4 i 2 3 5~ Open-dry plant material, g/pot 1 O.SSe"*" 0.83e 0.82e 0.76e 2 2.50d 3.76abc A.02ab 3.71abc 3 3.23bcd 3.69abc h.lGa k.]Sa k 2.3kcd 3.29bcd 3.17cd 3.12cd CEC, meq/lOOg 1 ll.i»7i ]h.76\ 28.48gh 36.69fg 2 52.22def 79.69bc 97.21b 62.83cde 3 55.9^def 70.29cd i*7.08efg 137. 80a k 125.03a 71.88cd 78.36bc 132.87a Values within subtables followed by the same letter do not differ significantly at 5% level of probability as judged by Duncan's Multiple Range test.

PAGE 136

Table 5^. Effect of cropping on CEC values for Guanipa 5 soi 1 , CEC Treatment Before After combinations cropping cropping Difference 12 5.64 2.33 3.31 b 13 4.46 3.15 1.31 c ]h 7.03 7.55 -0.52d 22 5.41 4.08 1.33c 23 8.02 4.52 3.50b 2k 17.65 7.89 9.76a 32 6.93 4.37 2.56bc 33 7.76 4.84 2. 92 be 3h 9.70 8.28 1.41 c 42 6.79 2.79 4.00 b 43 5.18 4.00 1 .I8c 44 5.64 6.98 -1 .34c Values followed by the same letter do not differ significantly at 5% level of probability as judged by Duncan's Multiple Range test.

PAGE 137

Table 55. Effect of cropping on CEC values for Guanipa 1 so i 1 . CEC Treatment Before After combinations cropping cropping Difference 12 2.15 3.59 13 2.65 it. 77 -2.12b ]h 3.99 5.5^* -1.55c 22 2.09 ^1.55 -2.'»6b 23 2.65 5.70 -3.05a 2h 9.62 6.9'» -2.68b 32 2.58 3.7'» -1 . I6c 33 h.7h 3.62 1 . 12c 3k 5. he k.SO 0.56d h2 2.58 S.Sh -3.35a A3 3.59 2.Sh 0.75d 3.51 k.20 -0.69d Values followed by the same letter do not differ significantly at 5% level of probability as judged by Duncan's Multiple Range test.

PAGE 138

Table 56. Effect of cropping on CEC values for El Potrero so i 1 . CEC Treatment Before After combinations cropping cropping Difference meq/1 OOg 12 38.27 52.22 -13.95b 13 55.87 55.93 0.06a ]k 78.06 125.08 -46.973 22 ^0. 13 79.69 -39.56a 23 l\.kk 70.29 + 1 . 15e 2i» 122.85 71.87 +50.98f 32 78.06 97.21 -19. 15b 33 38.27 ^7.07 8.80c 3'f 59.97 78.35 -18.38b 158.15 62.86 +95.32g ^3 59.97 137.79 -77.82a kk 71. ^'t 132.88 -6l.4Aa Values followed by the same letter do not differ significantly at 5% level of probability as judged by Duncan's Multiple Range test.

PAGE 139

Table 57. Change in CEC after cropping for the soils sel ected . Treatment Guanipa 1 12 -\.hh 13 -2.12 ]h -1.55 22 -2.^*6 23 -3.05 2h -2.63 32 -1 .16 33 +1.12 3'* +0.56 hi -3.35 A3 +0.75 +0.69 Guanipa 5A El Potrero meq/1 OOg 3.37 -13.95 1.31 0.06 -0.52 -A6.97 1.33 -39.56 3.50 + 1.15 9.76 +50.98 2.56 -19.15 2.92 8.80 l.i»l -18.38 A. 00 +95.32 1.18 -77.82 -1.3^1 -e].kk

PAGE 140

126 The most important feature of crop effect was the comparison of CEC values after cropping with those before cropping. In Tables 5^ through 57, differences in CEC obtained by substracting CEC after cropping from that before cropping are shown. There were no significant differences within these values among the three soils. The difference in CEC for Guanipa 1 soil was positive and not significant. These differences were probably due to a release of P and to decreased pH as consequence of Ca uptake by the crop. In this soil, treatments 1^ and kk with the highest level of P and the lowest and highest levels of Ca, respectively, showed an increase in CEC following cropping. This was probably due to an increase of P fixation with time at the high P concentration. Guanipa 5 soil showed negative differences of CEC except for treatment combinations 33, 3^, and ^3The negative values can be explained as before to an increase in P fixation with time The positive values may also have been due to the formation of calcium phosphate compounds that reduced PZNC because of neutralization of negative charges. As pointed out by Helyar et al . (1976a), the application of CaC03 should reduce the availability of P in soils in which phosphate is controlled by adsorption on minerals which behave like gibbsite. They may include kaolinite as well as X-ray amorphous and crystalline hydrous oxides of Fe and Al . El Potrero soil which showed a different tendency may have been due to the higher contents of organic matter, clay, and Al and Fe oxides. Three treatment combinations (23, 2k, and m) showed positive values while other treatments showed negative values This decrease in CEC was related to the increase of PZNC caused by precipitation of Ca phosphates in this soil. This suggested that, even though the effect of CaCOs on PZNC was not significant, it had a specifi effect on P adsorption and on PZNC.

PAGE 141

127 In general, there seemed to be an effect of cropping in decreasing the "new" CEC obtained with CaCOs and P treatments. The treatment combinations which increased dry matter production v/ere those that showed less CEC (Tables 5^ through 57). This may have been due to the uptake of P and Ca by plants. Uptake of P by the crop leaves positive charge on the surface that canceled negative charge and decreased CEC. The effects of soils, CaCOs, P, and the interaction of CaCO^ with P on the variables measured are summarized in Table 58. Only P and soils showed highly significant effects on the characteristics measured. Lime, as expected, had a highly significant effect on pH increase; the interaction CaCOsxP had no significant effect on these parameters. There were no differences in PZNC among soils. Prediction equations for the variables pH, ii;o, (Gouy-Chapman potential), EC, and DM were obtained (Table 59). These equations were calculated by treating the variables as continuous to fit a response surface. The data showed that the linear and quadratic effects of P were significant for all the soils and variables measured, except on the pH of Guanipa 1 and El Potrero soils. These equations can be used to define a new experimental region using treatment combinations to determine increase in CEC without increasing EC to high levels. For the three soils selected in this experiment, the highest dry matter yields were obtained with different treatment combinations. The highest dry matter yield and the least change in CEC after cropping were found in treatment combinations 43, 22, and 33 for Guanipa 1, Guanipa 5, and El Potrero, respectively.

PAGE 142

Table 58. Effect of soil, CaCOs, P, and interaction CaCOsxP on the variables measured . Variable Soil CaCOa P CaCOsxP pH NS PZNC NS NS NS ijjo NS NS CM NS NS EC NS NS CEC NS NS

PAGE 143

" j" -;; -i; -;c CsJ CM Csl J fO "3 fO O O O C-) LlJ O (U LTN Lf\ 4—' VU — ' — -J — J O CO c 0) TO ^— Q} (/) 1 1 T 1 It. (/) — -:; O O t/) LTS U— Q_ CU Q(L) fU 0) D. Q. V\D •— O f~ C < — o Lr\ ' — I+-> • > • • O — O vO O O f~ 4-* 1 1 + 1 o Vo 4-1 O to ro <4TO TO TO TO <_> O CJ CJ cr — O CO LA O O LA CM O O O O + 111 C\l CNJ CM csl TO TO TO TO <_J o o o CM cs -dCO o o o o o ' ao o o o I I + I Da. d. Dvo o -3rA CA sD -3^ O LA CM O O CA csl + + + + TO TO TO TO O O O (_> o cr» CTi o CO CSJ v£) Cv-\ CA O ^ CO o o o — + + I + ^ Jrai o ^ oo ^ o ^ o II II II II 3: o 3: Lij Q. -9O O 1_ o o 00 csl Csl csl CM Q. DCL. Q. OA CO LA rA O LA vD o o CO -dO O O \D I I I I CM CM C-1 CM TO TO TO TO O O C_) o -3C» O vIJ -3o o ^ cn d o o r>» + + I + CL CL a. Q. CTV LAt^OO OA O O LA O O LA VO LA + + + + TO TO TO TO CJ <_)<_) o LA CA O — O O o o 00 CO -3O CA — CA CM I I vO CO — CO O CA O II II 3: o CL OA -3CA -3CO I II II o SZ LU

PAGE 144

SUMMARY AND CONCLUSIONS Ten soils from Venezuela and one from Florida, U.S.A., were investigated by chemical, physical, and mineralog ical techniques to elucidate surface reactivity. Also, three of the 10 Venezuelan soils were selected to determine the effect of lime and P levels on the surface charge. A sorghum crop was planted under greenhouse conditions to investigate the change of surface charge after cropping for each CaC03 and P treatment. Stern and Gouy-Chapman models for diffuse double layer together with the PZNC and the surface area were used to calculate the charge distribution and the CEC of the soils for each CaCOa and P treatment. The following parameters were determined: (a) acidity sources; (b) quantity-intensity K parameters; (c) P fractionation in virgin soils and after P adsorption; (d) P adsorption isotherms at different temperatures and v/ith different ionic systems; (e) PZNC of virgin soils and after CaCOs and P treatments; (f) pH before and after cropping; (g) electrical conductivity before cropping; and (h) dry matter yields. The following parameters were calculated: (a) thermodynamic parameters associated with the adsorption process (AG°, AH°, AS°) and (b) charge distribution in double layer, and CEC after CaCOs and P treatments. +3 Acidity measurements showed that exchangeable Al and dissociation of OH groups from organic matter and silicates were responsible for the release of protons in these soils. The results of the Q/ I technique and Gapon's equation indicated that the soils had a high selectivity for K 130

PAGE 145

131 and low available K. Increasing temperature increased P adsorption in all soils. Study of ionic systems showed that some soils had a specific effect of CaCOs on P adsorption. Fractionation of soil P indicated that there was an increase of Al-P and Fe-P with temperature and quantity of P applied. Phosphorus decreased PZMC while P and CaCOs increased CEC through decreased PZNC and increased pH. Large applications of P and CaC03 resulted in electrical conductivity that may have depressed dry matter yields of sorghum. Calculations of thermodynamic paramters showed negative AG° values and positive values of AH° and AS°. +3 From these results, it can be concluded that (1) exchangeable Al and OH dissociation are the sources of acidity in these soils; (2) temperature increased P adsorption in all soils; (3) incubation temperature increase Al-P and Fe-P; {h) in some soils CaCOs had a specific effect on P adsorption; (5) P adsorption was a spontaneous endothermic reaction with an increase of entropy due to displacement of water molecules from the surface of phosphate ions; (6) the combined effect of increasing pH with lime and decreasing PZNC with applied P increased CEC; (7) the "new" CEC was not permanent but changed with cropping probably due to uptake of P and Ca by plants; (8) to obtain high yields of dry matter, the maximum increase in CEC must be sacrified; and (9) intermediate levels of CaC03 and P should be applied to these soils.

PAGE 146

APPENDIX I DETAILED CALCULATIONS FOR CaCOa AND PHOSPHORUS TREATMENTS

PAGE 147

CaC03 From IN KCl -exchangeabl e Al (Table k) , El Potrero soil contained 3.020 meq Al/lOOg, then lOOg : 3.020 meq Al as 1kg : X, and X = 30.20 meq Al 1 meq of CaCOa will neutralize 1 meq of Al 30.20 meq CaCOa are needed to neutralize the exchangeable Al in 1 kg of soi 1 , so 30.20 meq x 50 mg of CaCOa/meq = 1.51g 1kg of soil needs 1.51g CaCOa to neutralize the exchangeable Al . Levels of CaCOa for 1kg of El Potrero soil are +3 0 Al = Og CaCOa 1/2 Al^^ = 0.755g CaCOa 1 Al"^^ = 1 .51g CaCOa 2 Al"^^ = 3.02g CaCOa. Phosphorus To find parameters for the linear form of the Langmuir equation, data from Fig. 6 v/ere used. The Langmuir equation is written as fol lows : c „ 1 ^ c (1) x/m Kb b 133

PAGE 148

13A The regression equation can be written as where y = a + bx y = ^ ' " = lb ' = • • The computer regression analysis when no supporting electrolyte was used showed values for San Cristobal soil as follows: y = 0.65 + 0.004x then b = q ^qI^ = 250 mg of P/lOOg soil = 2,500 ppm P. These levels were Ob = 0 ppm P lAb = 625 ppm P l/2b = 1 ,250 ppm P lb = 2,500 ppm P from KH2PO1, {11.11% P) b 2,500 o 625 mg P 27^^.8 mg KH2PO1, IT = -IT= PP'" ^ = 1 kg soil = 1 kg soil Ob = Og of KH2P04 lAb = 2.7g of KH2P04 l/2b = 5.590g of KH2PO1+ lb =11 . I80g of KH2PO4 The same calculations were made for each soil.

PAGE 149

APPENDIX 1 I DETAILED CALCULATIONS OF THERMODYNAMIC PARAMETERS BY THE METHOD OF BIGGAR AND CHEUNG (1973)

PAGE 150

The data of P adsorption on Guanipa 1 soil at three temperatures are employed here to illustrate the detailed calculations: Equation: Cs = S Na(x/m) (pi/Mi)A L. A2 (5) M2 X 106 was used to calculate the amount of solute adsorbed per milliliter of solvent in contact with the adsorbent surface (Cs) . To calculate Cs, the cross sectional areas of the solvent (aJ and the solute (A2) are required and can be calculated by equation: A = 1 .091 X 10 •16 M X 102^ Nap 2/3 (6) The cross-sectional area of the solvent (in this case, water) was calculated by putting the value M, Na, and p^ in the above equation. A = 1 .091 X 10 is f" 18 X [6.02 X A = 1 .091 X 10 1^ (29.90) 102t 1023 X 1. ,0.667 2/3 A = 10.522 X 10"^^ cm2/molecule. The cross-sectional area of the solute molecule (in this case, KHgPO^) can be calculated similarly as follows: A = 1 .091 X 10 -16 r 136.13 X 102*^ I |_6.02 X 102^ X .233»J 2/3 A = 1.091 X 10"^^ (96.55)-^^^ A = 22.92 X 10"^^ cm2 /molecule. 136

PAGE 151

137 Using: Pl = ] g/ml P2 = 2.338 g/ml Ml = 18 g/mole M2 = 136.13 g/mole S = 270 cm^/g Na = 6.02 X 1023 and the following computer program: DATA TODOS; INPUT PA 1-5 1 PSN 6-9 1 SA2 10-12 1 SOIL 13-1/4 TEMP 15-16; PAD = PA PSN PADS = PAD/(SA-.'.-100) ; Cs = 3519.02i».':PADS; Yl = Cs/PSN; Yl = IN (Yl); PROC SORT; BY SOIL TEMP; PROC GLM; BY SOIL TEMP; MODEL LYl = Cs; Where: PA = P concentration of solution added in yg/ml Ce = PSN = concentration of P at equilibrium in yg/ml SA = survace area m^/lOOg PAD = x/m = P adsorbed yg/g The print out gave the intercept for each soil and temperature, and that is the thermodynamic constant K for each temperature.

PAGE 152

138 The free energy change was calculated by equation ( 9) AG° =-RTlnKo AG^ = 1 .987 X 278 In 1.17 AG§ = 552.39 X 0.16 AG5 = -86.73 cal/mole AG5 = -0.09 Kcal/mole Enthalpy (AH°)was calculated by equation (lo) AH° =-R In ^ AH" = -1 .987 (-2i(86.l) In 1 .30 AH° = ^939.91 X 0.25 AH° = ]2k0 cal/mole AH° = 1 .24 Kcal/mole The entropy change (AS°) was calculated by equation () 1 ) as: AS°= (AH°AG°)/T (12'tO cal/mole (-90 cal/mole) AS5 = 27B AS| = = h.78 cal/mole Ko The other methods used to calculate AG°are less reliable so the same equations were used; the difference is the K used and the limitation of its values.

PAGE 153

APPENDIX I I I DETAILED CALCULATIONS OF SURFACE CHARGE

PAGE 154

The distribution of charge was calculated using the procedure proposed by Van Raij and Peech (1972) and using the following computer pro gram. DATA POST; INPUT SOIL 1-2 Ca 3 P ^ PH 5-7 2 ZPC 8-10 2 PO 11-15 5 DM I6-I8 2 REP 19 TRT IF SOIL = 1 THEN SA = 2.9; IF SOIL = 2 THEN SA = 2.7; IF SOIL = 3 THEN SA = 69-9; Nl = IE + 15; Z = 1 ; E = 1 .6E-19; Na = 6.02E + 23; . MN = 18; RO = 1; D = 89E 12; K = 1 .38E 23; T = 298; ED = 6e 9; PI = 3. 1^*16; GA = .9E 8; N = 30. IE + 19; F = (Z.vEvcP0)/(2>'^K^'=T) ; CONS = (2v:D-;.-K-.vT/Pl)-.v-.v.5; 1^40

PAGE 155

SIGMA = CONS^^N^v^v.SvcSINHCF) ; PDIF = ^'VPh'cGAASIGMA/ED; PDEL = PO PDIF; Fl = (Z5*cEV.-pDEL)/(2--VK-.vT) ; SIGMA 2 = C0NS'VN-v>v.5-.vsiNH (l); RAT = (NA'VR0)/(MN'VN ); Ml = -2V:F1; SFl = SINH(F1 ) ; SIGMA 1 = (Nl'VZ.'iE)/(l + RAT'^EXP(MI) ; EXPMI = EXP (Ml); CHARGE = SIGMA 1 + SIGMA 2; CEC = CHARGE '"^ 10360 '--SA; SIGMA I = SIGMA 1 'O 0360'VSA; SIGMA 0 = SIGMA 2-.'=l OSSO-'^SA; SIGMA GC = SIGMA^'0 0360-.VSA; PZNC = ZPC; PROC SORT; BY SOIL TRT; PROC MEANS; BY SOIL TRT; VAR SIGMA I SIGMA 0 SIGMA GC; OUTPUT OUT=AVG MEAN=SIGMA I SIGMA 0 SIGMA GC; Where: Po = i|jo So i 1 1 = Guan i pa 5 Soil 2 = Guan i pa 1 Soil 3 = El Potrero e = e Na = Ha

PAGE 156

Mn = Molecular weight of solvent Ro = p D = E K = Boltzman constant ED = = average dielectric constant ( p I = IT GA = 6 = thickness of double layer (0.9 N = concentration number of ion/cm^ PDEC = ^6 Sigma I = inner layer charge Sigma 0 = Stern layer charge Sigma GC = Gouy-Chapman

PAGE 157

LITERATURE CITED Adams, M. 1973. Efectos del suministro de calcio y magnesio solubles a plantas de manT ( Arachis hypogaea L.) en suelos arenosos de la Mesa de Guanipa. Univ. Central de Venezuela, Maracay, Venezuela, pp. ^5. Adams, M. 1976. Interpretive analyses of potent iometr ic titrations of soils and colloidal component. Master Thesis. Soil Department. University of Florida, Gainesville. Adamson, H.W. 1976. Physical Chemistry of Surface. 3rd Ed. John Wiley and Sons. New York. 698 pp. Amarasiri, S.L. and S.R. Olsen. 1973. Liming as related to solubility of P and plant growth in an acid tropical soil. Soil Sci. Soc. Am. Proc. 37:716-721. Amedee, G. and M. Peech. 1976. The significance of KCl -extractable Al(lll) as an index to lime requirement of soils of humid tropics. Soil Sci. 121:227-233. Atkinson, R.V., A.M. Posner, and J. P. Quirk. I967. Adsorption of potent ial -determin ing ions at the ferric oxide-aqueous electrolyte interface. J. Phys. Chem. 71:550-558. Barr, A.J., J.H, Goodnight, J. P. Sail, and J.T. Helwig. 1976. A User's Guide to SAS. SAS Institute, Inc. Raleigh, N.C. Barrow, N.J. and T.C. Shaw. 1975a. The slow reactions between soil and anions: 2. Effect of time and temperature on the decrease in phosphate concentration in the soil solution. Soil Sci. 119:167-177. Barrow, N.J. and T.C. Shaw. 1975b. The slow reactions between soil and anions: 3The effects of time and temperature on the decrease in isotopically exchangeable phosphate. Soil Sci. 119: 190-197. Baver, L.D., W.H. Wardner, and W.R. Gardner. 1972. Soil Physics. 4th Ed. John V/iley and Sons. New York. hS8 pp. Beckett, P. 196'4a. Potassium exchange equilibria in soil: Specific adsorption sites for potassium. Soil Sci. 97:376-383-

PAGE 158

Beckett, P. ]36kh. Studies on soil potassium. I. Conformation of the ratio law: Measurement of potassium potential. J. Soil Sci. 15:1-8. Beckett, P. 196^c. Studies on soil potassium. 11. The "immediate" Q/l relation of labile potassium in the soil. J. Soil Sci. 15: 9-23. Beckett, P.H.T. and C.R. Clement. 1973K-actlvity ratios and the uptake of potassium by ryegrass in the field. J. Soil Sci. 2k: 82-93. Beckett, P.H.T. and M.H.M. Nafady. 1967a. Studies on soil potassium Vl. The effect of K-fixation and release on the form of the K-Ca+Mg exchange isotherm. J. Soil Sci. 18:2^*^-262. Beckett, P.H.T. and M.H.M. Nafady. 1967b. Potassium-calcium exchange equilibria in soils: The location of non-specific (Gapon) and specific exc hange sites. J. Soil Sci. 18:263-281. Biggar, J.W. and M.W. Cheung. 1973Adsorption of Piclorem (4-Amino3,5,6-Tr ichloropicol inic Acid) on Panoche, Ephrata, and Palouse soils: A thermodynamic approach to the adsorption mechanism. J. Soil Sci. Soc. Am. 37:863-868. Blanchard, I.N. 197't. X-ray power diffraction data for vjavellite. Florida Sci . 37:1-^*. Blue, W.G. \S7h. Management of Ultisols and Oxisols. Soil and Crop Sci. Soc. Fla. Proc. 33:126-132. Bradfield, R. 19^1Calcium in the soil. I. Physico-chemical relations. Soil Sci. Soc. Am. Proc. 6:8-l6. Breeuwsma, A. 1973. Adsorption of ions on hematite (a-Fe203) . A colloid-chemical study. Mededelingen LandBouWhoGeschol 1 Wagen ingen -Net her land. 73: 1 -1 23Bremner, J.M. 1965. Inorganic forms of nitrogen. j_n C. A. Black (ed.), Methods of Soil Analysis. Part 2. Am. Soc. Agron. (Madison, Wise.) Agron. 9:1179-1237. Calhoun, F.G. and V.W. Carlisle. 1971. Infrared spectra of selected Colombian Andosols. Soil and Crop Sc. Soc. Fla. Proc. 31:157" 161. Carter, D.L., M.D. Hetlman, and C.L. Gonzalez. 1965The ethylene glycol monoethyl ether technique for determining soil surface area. Soil Sci. 1 00 : 409-'* 1 3 • Casanova, E. 197'*. Determinacion del estado de azufre en los suelos de los Llanos Altos Occidentales de Venezuela. Ill Reunion Nacional de la Ciencia del Suelo. Merida, Venezuela, pp. 22.

PAGE 159

1^*5 Chang, S.C. and M.L. Jackson. 1957Fractionation of soil phosphorus. Soil Sci. 8k:]33-'lhh. Chang, S.C. and M.L. Jackson. 1958. Soil phosphorus fractions in some representative soils. J. Soil Sci. 9:109-119Chao, T.T. and M.E, Harward. 1962. Nature of acid clays and relationships to ion rations in equilibrium solutions. Soil Sci. 93:2^46-253. Coulter, B.S. I969. The chemistry of hydrogen and aluminum ions in soils, clay minerals, and resins. Soils and Fertilizers. 32: 215-223. Davis. L.E., R. Turner, and L.D. V/hittig. I962. Some studies of the autotransformat ion of H-bentonite to Al -benton i te . Soil Sci. Soc. Am. Proc. 26:'i^]-kh3. Dewan, H.C. and C.I. Rich. 1970. Titration of acid soils. Soil Sci. Soc. Am. Proc. 3'»:38-4it. Douglas, L.A., and F. Fiessinger. 1971. Degradation of clay minerals by H2O2 treatments to oxidize organic matter. Clays and Clay Miijerals. 19:67-68. Espinosa, W. , R.G. Gast, and R.S. Adams. 1975Charge characteristics and nitrate retention by two Andepts from south-central Chile. Soil Sci. Soc. Am. Proc. 39:8^2-846. Evans, C.E. and E.J. Kamprath. 1970. Lime response as related to percent Al saturation, solution Al , and organic matter content. Soil Sci. Soc. Am. Proc. 3^:893-896. Fiskell, J.G.A., S.J. Locascio, H.L. Breland, and T.L. Yuan. 1964. Effects of soil acidity and liming of Leon fine sand on the exchange properties and on watermelons as indicator plants. Soil Crop Sci. Soc. Fla. Proc. 24:52-63. Fiskell, J.G.A. and L.W. Zelazny. 1971. Acidic properties of some Florida soils. I. pH-dependent cation exchange. Soil Crop Sci. Soc. Fla. Proc. 31:149-154. Fox, R.L. and E.J. Kamprath. 1970. Phosphate sorption isotherms for evaluating the phosphate requirements of soils. Soil Sci. Soc. Am, Proc. 34:902-906. Griffin, R.A. and J.J. Jurinak. 1974. Kinetics of the phosphate interaction with calcite. Soil Sci. Soc. Am. Proc. 38:75-79Grim, R.E. I96B. Clay Mineralogy. 2nd Ed. McGraw-Hill Book Co. New York.

PAGE 160

1^6 Hassan, N.A.K., J.V. Drew, D. Knudren, and R.A. Olson. 1S70. Influence of soil salinity on production of dry matter and uptake and distribution of nutrients in barley and corn. I. Barley. Agron. J. 62:43-^5. Helling, C.S., D.L. Chesters, and R.B. Corey. 196^Contribution of organic matter and clay to soil cation-exchange capacity as effected by the pH of the saturating solution. Soil Sci. Soc. Am. Proc. 28:517-520. Helyar, K.R., D.N. Munns, and R.G. Burau. 1976a. Adsortion of phosphate by gibbsite. I. Effects of neutral chloride salts of calcium, magnesium, sodium, and potassium. J. Soil Sci. 27:307~315. Helyar, K.R., D.N. Munns, and R.G. Burau. 1976b. Adsortion of phosphate by gibbsite. II. Formation of a surface complex involving divalent cations. J. Soil Sci. 27:315-32^. Hingston, F.J., A.M. Posner, and J. P. 0.uirl<. 1972. Anion adsorption by goethite and gibbsite. I. The role of the proton in determining adsorption envelopes. J. Soil Sci. 23:177-192. Hingston, F.J., A.M. Posner, and J. P. Quirk. 197^*. Anion adsorption by goethite and gibbsite. II. Desorption of anions from hydrous oxide surfaces. J. Soil Sci. 25:16-26. Hsu, P.H. 1965. Fixation of phosphate by aluminum and iron in acidic soils. Soil Sci. 99:398-^*02. Hsu, P.H. and T.F. Bates. 196'f. Fixation of hydroxy-al uminum polymers by vermiculite. Soil Sci. Soc. Am. Proc. 28:763-769Jackson, M.L. 1956. Soil Chemical Analysis-advanced Course. Published by the author, Dept. of Soils, University of Wisconsin, Madison. Jackson, M.L. I963. Aluminum bonding in soils: A unifying principle in soil science. Soil Sci. Soc. Am. Proc. 27:1-10. Juo, A.S.R. and B.C. Ellis. I968. Chemical and physical properties of iron and aluminum phosphates and their relation to phosphorus availability. Soil Sci. Soc. Am. Proc. 32:216-221. Kamprath, E.J. 1970. Exchangeable aluminum as a criterion for liming leached mineral soils. Soil Sci. Soc. Am. Proc. 3^:252-254. Keng, J.C.W. and G. Uehara. 197^. Chemistry, mineralogy, and taxonomy of Oxisols and Ultisols. Soil Crop Sci. Soc. Fla. Proc. 33:119-126. Khasawneh, F.E. and F. Adams. 1967. Effect of dilution on calcium and potassium contents of soils solutions. Soil Sci. Soc. Am. Proc. 31 :172-176.

PAGE 161

1^7 Kissel, D.E. and G.V/. Thomas. 1969. Conduct imetr ic titrations vjith Ca(0H)2 to estimate the neutral salt replaceable and total acidity. Soil Sci. 108:177-179. Kuo, S. and E.G. Lotse. 1972. Kinetics of phosphate adsorption by calcium carbonate and Ca-kaol i n i te . Soil Sci. Soc. Am. Proc. 36: 725-729. Larsen, S. 1967. Soil phosphorus. Advances in Agronomy. 19:151-210. Lee, R. 1973The K/Ca Q/l relationship and preferential adsorption sites for potassium. New Zeland Soil Bureau Scientific Report 11. Sk pp. Lindsay, W.L. and B.C. Moreno. I96O. Phosphate phase equilibria in soils. Soil Sci. Soc. Am. Proc. 24:177-182. Lora, R. and G. Riveros. 1971. Problemas f i s iolog icos en suelos acidos. Suelos Ecuator iales. Sociedad Colombiana de la Ciencia del Suelo. 3(1) :2't-A2. Low, P.E. 1955. The role of aluminum in the titration of bentonite. Soil Sci. Soc. Am. Proc. 19:135-139. Luque, 0. 1975. Genesis de tres ordenes de suelos en los llanos orientales de Venezuela. V Congreso Latino Americano de la Ciencia del Suelo. Medellfn, Colombia. pp. 19. Marshall, C.E. 196^*. The physical chemistry and mineralogy of soils. I. Soil materials. John V/iley £ Sons. New York. pp. 211-259. Mehra, O.P. and M.L. Jackson. I96O. Iron oxide removal from soils and clays by a dithionite-citrate system buffered with sodium bicarbonate. Clays and Clay Minerals. 7:317-327Mekaru, T. and G. Uehara. 1972. Anion adsorption in ferruginous tropical soils. Soil Sci. Soc. Am. Proc. 36:296-300. Moss, P. and P.H.T. Beckett. 1971. Sources of error in the determination of soil potassium activity ratios by the Q/l procedure. J. Soil Sci. 22:514-536. Mukhtar, M. 1976. Desorption of adsorbed ametrym and diuron from soils and soil components in relation to rates, mechanism, and energy of adsorption reactions. Ph.D. Thesis. Department of Agronomy and Soil Sci. University of Hawaii. Muljadi, D. , A.M. Posner, and J. P. Quirk. I966. The mechanism of phosphate adsorption by kaolinite, gibbsite, and pseudoboehmi te. III. The effect of temperature on adsorption. J. Soil Sci. 17:238-247. Munns, D.N. and R.L. Fox. 1977. Stabilization of calcium surface charge variation in an Oxisol. Soil Sci. Soc. Am. J. 41:682-686.

PAGE 162

]h8 Olsen, S.R. and L.A. Dean. I965. Phosphorus. \r^C.k. Black (ed.), Methods of Soil Analysis. Part 2. Agronomy 9Amer. Soc. of Agronomy, Inc., Madison, Wisconsin. Olsen, S.R. and R.W. Watanabe. 1957Amethod to determine a phosphorus adsorption maximum for soils as measured by the Langmuir isotherm. Soil Sci. Soc. Am. Proc. 2\:]hh-]h3. Parks, G.A. I965. The isoelectric points of solid oxides, solid hydroxides, and aqueous hydroxy complex systems. Chem. Rev. 65: 177-198. Parks, G.A. 1967. Aqueous surface chemistry of oxides and complex oxide minerals. _ln^ R. 1 . Gould (ed.) Adv. in Chem. Series 67: 121-160. Parks, G.A. and L.P. de Bruyn. 1962. The zero point of charge of oxides. J. Phys. Chem. 66:967-972. Pearson, R.W. 1975. Soil acidity and liming in the humid tropics. Cornell Int. Agric. Bull. 30. New York State College of Agriculture and Life Sciences. Cornell University. Ithaca, New York. Petersen, G.W. and R.B. Corey. 1969A modified Chang and Jackson procedure for routine fractionation of inorganic soil phosphorus. Soil Sci. Soc. Am. Proc. 30:563-565Rajan, S.S.S. 1978. Sulfate adsorbed on hydrous alumina, 1 igands displaced, and changes in surface charge. Soil Sci. Soc. Am. J. 42:39-^^. Reeve, N.G. and M.E. Sumner. 1970a. Lime requirements of Natal Oxisols based on exchangeable aluminum. Soil Sci. Soc. Am. Proc. 3^:595-598. Reeve, N.G. and M.E. Sumner. 1970a. Effects of aluminum toxicity and phosphorus fixation on crop grov/th onOxisols in Natal. Soil Sci. Soc. Am. Proc. 3^:263-267. Rich, C.I. 196^4. Effect of cation size and pH on potassium exchange in Nason soil. Soil Sci. 98:100-106. Rich, C.I. 1968. Hydroxy interlayers in expansible layer silicates. Clays and Clay Minerals. 16:15-30. Rich, C.I. 1970. Conductometr ic and potent iometr ic titration of exchangeable aluminum. Soil Sci. Soc. Am. Proc. 3'<:31-38. Rich, C.I. and W.R. Black. 196^4. Potassium affected by cation size, pH, and mineral structure. Soil Sci. 97:38^-390. Ryden, J.C., J.R. McLanghlin, and J.K. Syers. 1977a. Mechanism of phosphate sorption by soils and hydrous ferric oxide gel. J. Soil Sci. 28:72-92.

PAGE 163

1^9 Ryden, J.C. and J.K. Syers. 1977. Desorptlon and isotopic exchange relationships of phosphate sorbed by soils and hydrous ferric oxide gel. J. Soil Sci. 28:596-609. Ryden, J.C, J.K. Syers, and J. P. McLanghl in. 1977b. Effects of ionic strength on chemisorpt ion and potent ial -determining sorption of phosphate by soils. J. Soil Sci. 28:62-71. San ValentTn, CO., L.W. Zelazny, and V/.K. Robertson. 1972. Potassium exchange characteristics of a Rhodic Paleudult. Soil and Crop Sci. Soc. Fla. Proc. 32:128-132. Sawhney, B.L. 197^. Charge characteristics of soils as affected by phosphate sorption. Soil Sci. Soc. Am. Proc. 38:159-160. Shaninberg, I. and J.E. Dawson. 1967Titration of H-clay suspensions with salt solutions. Soil Sci, Soc. Am. Proc. 31:619-626, Shaked, D. and A. Banin. 197^. Effect of ionic strength on ion activity in soils: 2. Measured potassium activity in soils at varying ionic strengths. Soil Sci. 117:200-20'*. Shelton, J.E. and N.T. Coleman. I968. Inorganic phosphorus fractions and their relationships to residual value of large applications of phosphorus on high phosphorus fixing soils. Soil Sci. Soc. Am. Proc. 32:91-9'<. Singh, R.N., D.C Martens, and S.S. Abenshain. I966. Plant availability and form of residual phosphorus in Davidson clay loam. Soil Sci. Soc. Am. Proc. 30:617-620. Spain, J.M., CA. Francis, R.W. Howeler, and F. Calvo. 1975. Differential species and varietal tolerance to soil acidity. In E. Bornemisza and A. Alvarado (eds.), Soil Management in Tropical America. North Carolina State University, Raleigh. Stumm, W. and J.J. Morgan. 1970. Aquatic Chemistry. An introduction emphasizing chemical equilibria in natural waters. John Wiley and Sons. New York. Taylor, R.W. and B.C. Ellis. 1978. A mechanism of phosphate adsorption on soil and anion exchange resin surfaces. Soil Sci. Soc. Am. J. A2:432-436. Thomas, CW. I96O. Effects of electrolyte imbibition upon cationexchange behavior of soils. Soil Sci. Soc. Am. Proc. 24:329-332. Thomas, J.M. and W.J. Thomas. I967. Introduction to the principles of heterogeneous catalysis. Academic Press, New York. Tinsley, R.L. 197't. Surface chemistry of calcium and phosphorus retention in selected acid tropical soils from the Republic of Viet Nam. Ph.D. Thesis. University of Florida, Gainesville.

PAGE 164

I 3W Van Raij, B. and M. Peech. 1972. Electrochemical properties of some Oxisols and Alfisols of the tropics. Soil Sci. Soc. Am. Proc. 36:132-136. Veith, J. A. and G. Sposito. 1977. On the use of the Langmuir equation in the interpretation of "adsorption" phenomena. Soil Sci. Soc. Am. J. 41:697-703. Walkley, A. 19^*7. A critical examination of a rapid method for determining organic carbon in soils. Effect of variation in digestion conditions and of inorganic soil constituents. Soil Sci. 63: 251-26'j. Watanabe, F.S. and S.R. Olsen. 1965. Test of an ascorbic acid method for determining phosphorus in water and MaHCOa extracts from soil. Soil Sci. Soc. Am. Proc. 29:677-678. Webber, M.D. 1978. Effects of temperature and time on hydroxy aluminum phosphate montmor i 1 1on i te complex. Soil Sci. 125:107-112. Westin, F.C., J. Avilan, A. Bustamanta, and M. Marifio. 1968. Characteristics of some Venezuelan soils. Soil Sci. 105:92-102. White, R.E. and A.W. Taylor. 1977. Reaction of soluble phosphate with acid soils. J. Soil Sci. 28:3l'*-328. Whittig, L.D. 1965. X-ray diffraction techniques for mineral identification and mineralog ical composition. Jn^ C.A. Black (ed.) Methods of Soil Analysis. Part I. Am. Soc. of Agron. Inc. Madison, Wisconsin. Agron. 9:671-698. Wiklander, L. I96O. Influence of liming on adsorption of cation in soils. Int. Congr. Soil Sci., Trans. 7th. Madison, Wisconsin. 11:283.291. Yuan, T.L. 1 963 . Some relationships among hydrogen, aluminum, and soil pH in solution and soil systems. Soil Sci. 95:155-163. Yuan, T.L. 1959. Determination of exchangeable hydrogen in soils by a titration method. Soil Sci. 88:l6't-l66. Yuan, T.L. 1966. Characteristics of surface and spodic horizons of some Spodosols. Soil Crop Sci. Soc. Fla. Proc. 26:163-17^. Yuan, T.L., N. Gammon, and R.G. Leighty. 1967. Relative contribution of organic and clay fractions to cation exchange capacity of sandy soils from several soil groups. Soil Sci. 104:123-128. Yuan, T.L., W.K. Robertson, and J.R. Neller. I96O. Forms of newly fixed phosphorus in three acid sandy soils. Soil Sci. Soc. Am. Proc. 24:447-450.

PAGE 165

151 Zelazny, L.W. and F.G. Calhoun. 1971. Mineralogy and associated properties of tropical and temperate soils in the V/estern Hemisphere. Soil Crop Sci. Soc. Fla. Proc. 31:179-189. Zelazny, L.W. and V.W. Carlisle. 1971. Mineralogy of Florida Aerie Haplaquods. Soil and Crop. Sci. Soc. Fla. Proc. 31:1^1-165. Zelazny, L.W. and J.G.A. Fiskell, 1971. Acidic properties of some Florida soils. II. Exchangeable and titrable acidity. Soil and Crop Sci. Soc. Fla. Proc. 31:1^9-15^. Zelazny, L.W. and J.N. Qureshi. 1973Chemical pret reatments on clay separation and mineralog ical analyses of selected Florida soils. Soil and Crop Sci. Soc. Fla. Proc. 32:117-121.

PAGE 166

BIOGRAPHICAL SKETCH Meliton Jose Adams-Melendez was born 29 February ]3hk, at Urucure, Lara State, Venezuela. He graduated from Liceo "Pedro Emio Coll" in July 1963. In March 196^, he entered Universidad Central de Venezuela to pursue an Ingeniero Agronomo degree, which he received in September 1970, with an orientation in Soil Science. As a student, he aided in teaching basic soil science courses while conducting research in plant physiology. In August 1971, he became an instructor at Universidad Central de Venezuela and was promoted to assistant professor in September 1973. He entered the University of Florida Graduate School in June 197^4, and received the Master of Science degree in agriculture with a major in Soil Science in March 1976. In December 1976, he returned to Venezuela under a special research program. In September 1977, he v;as promoted to aggregate professor in the Universidad Central de Venezuela. He now is a candidate for the Doctor of Philosophy degree in soil science at the University of Florida. Meliton Jose Adams-Melendez is married to the former Nieves Madal i Vargas. They have two children, Lara and Jose. He is a member of Phi Kappa Phi honor society, American Society of Agronomy, Soil Science Society of America, Crop Science Society of America, Sociedad Venezolana de la Ciencia del Suelo, Sociedad Venezolana de Ingenieros Agronomos, Colegio de Ingenieros de Venezuela, and Asociacion de Profesores de la Universidad Central de Venezuela. 152

PAGE 167

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. ci Dr. W.G. Blue, Chairman Professor of Soil Science I certify that i have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Dr. T.L. Yuan Professor of Soil Science I certify that i have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Dr. V.E. Berkheiser Assistant Professor of Soil Science I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Dr. O.C. Ruelke Professor of Agronomy

PAGE 168

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Dr. A.H. Krezdot^n i Professor of Fruit Crops This dissertation was submitted to the Graduate Faculty of the College of Agriculture and to the Graduate Council, and was accepted as partia fulfillment of the requirements for the degree of Doctor of Philosphy. December 1978 r1 Dean, Col'lege of Agriculture^ u Dean, Graduate School


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EY55BD43D_CV27OL INGEST_TIME 2015-03-09T20:22:26Z PACKAGE AA00028869_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES