Citation
Dairy Manure-Component Effects on Phosphorus Release from Sandy Soils

Material Information

Title:
Dairy Manure-Component Effects on Phosphorus Release from Sandy Soils
Copyright Date:
2008

Subjects

Subjects / Keywords:
Crystallization ( jstor )
Dairy manure ( jstor )
Incubation ( jstor )
Manure ( jstor )
Minerals ( jstor )
Phosphates ( jstor )
Phosphorus ( jstor )
Soil water ( jstor )
Soils ( jstor )
Torsades de Pointes ( jstor )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Embargo Date:
7/12/2007

Downloads

This item has the following downloads:


Full Text












DAIRY MANURE-COMPONENT EFFECTS ON PHOSPHORUS RELEASE FROM
SANDY SOILS















By

MANOHARDEEP SINGH JOSAN


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

UNIVERSITY OF FLORIDA


2007

































2007 Manohardeep Singh Josan



































To my parents















ACKNOWLEDGMENTS

With all humility and sincerity, I bow before the Almighty for His benevolence and

blessings for the completion of my dissertation.

It is my privilege to express my deep sense of gratitude to Dr. V. D. Nair, Research

Associate Professor, Environmental Chemistry, Soil and Water Science Department,

University of Florida for excellent guidance, invaluable suggestions and perceptive

enthusiasm without which this work would not have its present shape. Her association

and moral encouragement and approach to see life perspectives as they are, throughout

this academic pursuit, would be an invaluable experience of an everlasting value.

My thanks go to my supervisory committee Willie G. Harris, George O'Connor,

Roy D. Rhue, Tom A. Obreza, Luisa A. Dempere, for valuable suggestions and positive

criticism during the course of investigation.

I express my sincere thanks to Willie G. Harris for providing excellent laboratory

conditions, and abundant supply of peanuts.

I am grateful to my wife Syliva Lang-Josan for her untiring and apt help in reading

my manuscript. I am immensely grateful to my parents, brothers and sisters Gagan and

Nimar. Their inspiration, sacrifice, helpful blessing, encouragement, support, and loving

emotions sustained me. I am thankful to Don Mitchell and Ilona Lang for their nurturing

support, and love.

My father is a farmer, has never had a chance to go school. He put all his efforts in

me to fulfill his life's ambitions. He provided me the freedom and confidence that I









needed to succeed in my life. So this degree is a commemoration to his dedications and

sacrifices that he made in his life.
















TABLE OF CONTENTS

page

A C K N O W L E D G M E N T S ................................................................................................. iv

LIST OF TABLES ......... ........ ................................... .......... .... .......... .. ix

LIST OF FIGURES ......... ......................... ...... ........ ............ xi

A B S T R A C T .........1..... ................. ............... ..................................... 14

CHAPTER

1 IN TR OD U CTION ............................................... .. ......................... ..

R ationale and Significance ........................................ ................................. 3
H y p o th e se s .............................. ............................................................. ............... .. 4
Specific Objectives .............. ................. ................................... .4

2 REVIEW OF LITERATURE ......................................................... .............. 6

M anure Application .................. .. .. .. ......... .......... ......... ... ...............
Release of P U sing V various Extractants................................. ................................... 6
Geochemical Models: Forms and Solubility of Phosphorus .......................................7
Solid State Assessments and P Associations............................................................10
Prospective Inhibitors on Ca-P Crystallization...................................................11

3 SOIL AND MANURE CHARACTERIZATION.....................................................13

S o il S a m p lin g ....................................... ...... ..... ..... ....................................... 1 3
Physicochemical Properties of Soils (Soil Characterization)............................14
Particle Size Fractionation and Mineralogical Analysis ....................................15
M anure Sam pling ........... ...... ...................................................................... .. 15
M anure Characterization ......................................................... ........... .... ...... 15
M ineralogical A nalysis................................................ ............................ 16
Q A /Q C for A nalyses ............................................ .. .. .. ...... .......... 16
Statistical A analyses ................................................. ....... .............. 16
Results and D discussion ................. ................ ............. .... ........ .............. 16
Soil Chemical Characterization.................. ........ ......................... 16
S o il M in eralo g y ............. ............................................................. .............. 17
M anure Characterization ................................. ........................... ..... 18









M anure M ineralogy ............................................... ... .. ... .. ........ .... 19
Sum m ary and Conclusions ............................................................ ............... 19

4 ASSOCIATED RELEASES OF PHOSPHORUS, CALCIUM AND
MAGNESIUM IN SOIL SOLUTIONS FROM DAIRY MANURE-AMENDED
S O IL S ................................................................................................................... 2 4

In tro d u ctio n .......................................................................................2 4
M methods and M materials ................ ....... ............................................................. 25
Repeated Water Extractions and Chemical Analyses ....................................25
Soil Leaching Characterization and Chemical Equilibrium Modeling ..............25
R esu lts an d D iscu ssion ..................................................................... ................ .. 2 8
R repeated W ater Extractions ..................................................... .....................28
Soil Leaching and Chemical Equilibrium Modeling..................................29
Sum m ary and C onclu sions .............................................................. .....................3 1

5 RELATIONSHIPS BETWEEN PHOSPHORUS, CALCIUM AND
MAGNESIUM INFERRED FROM SELECTIVE DISSOLUTION.........................39

Intro du action ...................................... ................................................ 3 9
M material and M methods .......................................... ..... ..................... ............... 40
Phosphorus Fractionation: Repeated and Sequential Extractions .....................40
A nalyses of P and M etals ............................................................................. 42
R results and D discussion ...................................................... ........ .. ...... ... ....... .... 43
Release of P, Ca, and Mg in Repeated 1.0 MNH4Cl Extractions....................43
Phosphorus Concentrations in Sequential Extractions.............. ... .............45
Sum m ary and C onclu sions .............................................................. .....................47

6 SOLID STATE ASSESSMENTS: CONFIRMING THE ASSOCIATIONS OF
PHOSPHORUS, CALCIUM AND MAGNESIUM.......................................54

In tro du ctio n ...................................... ................................................ 54
M material and M methods ......................... .... .............................. .. ............... 56
Approach I: X-ray Diffraction of Untreated Clays.........................................56
Approach II: Ashed and Whole Dairy Manure Analyses.............................. 56
Approach III: SEM Imaging and EDS Analyses.................. ..............................57
Approach IV: Electron Microprobe Microanalyses of Whole Silt + Clay of
D airy M anure-am ended Soils ........................................ ....... ............... 58
R results and D iscu ssion .............................. ........................ .. ...... .... ...... ...... 58
Sum m ary and C onclu sions .............................................................. .....................6 1

7 CALCIUM PHOSPHATE CRYSTALINITY AND DAIRY MANURE
COM PONENTS ......................... ........... .. .. ......... ..... ..... 67

Introdu action ...................................... ................................................. 67
M material and M methods ............................................................... ....................68
Particle Size Separations of Dairy Manure-amended Soils..............................68
Carbonate and Organic Matter Removal from the Resulting Clays....................69









Density Separations for Soil Clay M materials ....................................... .......... 69
Preparation of Incubating Solutions ................................................................. 70
Incubation Setup, Monitoring and Solution Analyses...................................71
S olid State A ssessm ents ........................................................... .....................72
X -ray diffraction analyses ........................................ ........................ 72
Energy dispersive spectroscopy analyses.........................................72
Statistical A nalysis................................................... 73
R results and D discussion ...................... ...................... .................... .. .. .......... 73
Effects of Mg, Si and DOC on Ca-P Crystallization................ ..................73
Effects of Solids on Ca-P Crystallization .............. .. ............... ......... .............76
Sum m ary and C onclu sions .............................................................. .....................77

8 SUMMARY AND CONCLUSIONS.................................................................85

APPENDIX

A DETAILS OF LEACHATE CONCENTRATIONS USED FOR V-MINTEQ
A N A L Y S E S ...............................................................................................................9 0

B CORRELATION MATRIX FOR DIFFERENT P FRACTIONS OF DAIRY
MANURE-AMENDED SOILS...................................................... 95

L IST O F R E FE R E N C E S ....................................................................... ... ................... 97

BIOGRAPHICAL SKETCH ............................................................. ............... 107















LIST OF TABLES


Table pge

3-1 Characteristics of active and abandoned manure-amended soils, and minimally
m anure-im pacted soils............ .... .................................................... .. ........ .... 20

3-2 Characteristics of dairy manures collected from four locations in Florida. ...............21

4-1 Cumulative average release of SRP, Ca, Mg and EC in repeated water extractions
of active, abandoned and minimally-impacted soils. ............................................. 32

4-2 Saturation indices (SI) in active and abandoned dairy manure-amended soils..........33

4-3 Percent of observations that are undersaturated, saturated, and supersaturated for
selected minerals based on chemical equilibrium modeling of active and
abandoned dairy column leachates............... ......... ........... ................... 34

5-1 Total dissolved phosphorus (TDP) as a function of Ca+Mg in repeated 1.0 M
NH4C1 extractions (Data from 1st extraction omitted). ..........................................49

5-2 Sequential release of total dissolved phosphorus (TDP) as a function of Ca, Mg,
and Fe in 0.1 MNaOH extractions ...................... .................. 49

5-3 Sequential release of total dissolved phosphorus (TDP) as a function of Ca, Mg,
Fe, and Al in 0.5 M H C1 extractions ............................................. ............... 49

5-4 Sequential release of total dissolved phosphorus (TDP) as a function of Ca, Mg,
Fe, and A l in residual fractions ........................................... ......................... 49

7-1 Average leachate composition of manure-amended soils used for the incubating
so lu tio n s .......................................................................... 7 8

7-2 Incubation treatments to study the effects ofMg, Si, and DOC on Ca-P
crystallization in the presence and absence of manure-derived solids ............... 78

A-i Column leachate pH, EC and P concentrations of leachates............................. 90

A-2 Concentrations of Ca, Mg and dissolved organic carbon (DOC) observed in
column n leachates. ................................................... ................. 9 1

A-3 Concentrations of K, Fe and Al observed in column leachates..............................92









A-4 Concentrations of sulfate, chloride and ammonium observed in column leachates..93

A-5 Concentrations of nitrates, and silcic acid (H4SiO4) observed in column leachates .94

B-l Correlation matrix for different fractions in active dairy manure-amended soils .....95

B-2 Correlation matrix for different fractions for abandoned dairy manure-amended
s o ils ........................................................................... 9 6















LIST OF FIGURES


Figure pge

3-1 X-ray diffraction patterns of clays obtained from active dairy (ACS) manure-
amended soils. HIV = hydroxyinterlayered vermiculite. .......................................22

3-2 X-ray diffraction patterns of clays obtained from abandoned dairy (ABS) manure-
am ended soils. ........................................................................22

3-3 X-ray diffraction patterns of clays obtained from minimally-impacted (MIS) soils..23

3-4 X-ray diffraction pattern of clay obtained from an oven dried dairy manure. ...........23

4-1 Column set-up used for the column leaching study. ........................................ ......35

4-2 Changes in electrical conductivity (EC) (dS m-1) with repeated water extractions.
ACS = active dairy manure-impacted soil; ABS = abandoned dairy manure-
impacted soil; M IS = minimally impacted soil. ...................................... ........... 35

4-3 Changes in Ca concentrations (mg kg-1) with repeated water extractions. ACS =
active dairy manure-amended soil; ABS = abandoned dairy manure-amended
soil; M IS = m inim ally im pacted soil ............................................... ............... 36

4-4 Changes in Mg concentrations (mg kg-1) with repeated water extractions. ACS =
active dairy manure-amended soil; ABS = abandoned dairy manure-amended
soil; M IS = m inim ally im pacted soil ............................................... ............... 36

4-5 Changes in soluble reactive phosphorus (SRP) concentrations (mg kg-1) with
repeated water extractions. ACS = active dairy manure-amended soil; ABS =
abandoned dairy manure-amended soil; MIS = minimally impacted soil...............37

4-6 Relationships between soluble reactive phosphorus (SRP) and Mg and Ca
released during repeated water extractions of active dairy manure-amended
s o ils ..............................................................................3 7

4-7 Relationships between soluble reactive phosphorus (SRP) and Mg and Ca
released during repeated water extractions of abandoned dairy manure-amended
so ils ................................................................................. . 3 8

4-8 Relationships between soluble reactive phosphorus (SRP) and Mg and Ca
released during the column leaching of dairy manure-amended soils. ....................38









5-1 Schematic of repeated and sequential extraction procedure adapted from Nair et
al. (19 9 5 ). ......................................................................... 5 0

5-2 Release of phosphorus using 1.0 MNH4C1 repeated extractions in active and
abandoned dairy manure-amended soils. ACS = active dairy manure-amended
soil; ABS = abandoned dairy manure-amended soil. ...................... ...............50

5-3 Release of calcium using 1.0 MNH4Cl repeated extractions in active and
abandoned dairy manure-amended soils. ACS = active dairy manure-amended
soil; ABS = abandoned dairy manure-amended soil. ...................... ...............51

5-4 Release of magnesium using 1.0 MNH4C1 repeated extractions in active and
abandoned dairy manure-amended soils. ACS = active dairy manure-amended
soil; ABS = abandoned dairy manure-amended soil. ...................... ...............51

5-5 Distribution of P in 1.0 MNH4C1, 0.1 MNaOH, 0.5 MHC1, and residual-P
fractions for an active dairy soil (ACS-2). ..................................... ............... 52

5-6 Distribution of P in 1.0 MNH4C1, 0.1 MNaOH, 0.5 MHC1, and residual-P
fractions for an active dairy soil (ACS-4). ..................................... ............... 52

5-7 Distribution of P in 1.0 MNH4C1, 0.1 MNaOH, 0.5 MHC1, and residual-P
fractions for an abandoned dairy soil (ABS-3) ............................................ ........... 53

5-8 Distribution of P in 1.0 MNH4C1, 0.1 MNaOH, 0.5 MHC1, and residual-P
fractions for an abandoned dairy soil (ABS-4) ............................................ ........... 53

6-1 X-ray diffraction patterns of four active (ACS-1 to ACS-4) and four abandoned
(ABS-1 to ABS-4) dairy manure-amended untreated clays. HIV =
hydroxyinterlayered verm iculite. ........................................ ......................... 62

6-2 X-ray diffraction pattern of < 1.0 mm dried dairy manures showing the presence
of w hew ellite, quartz and calcite ................................................................. ....... 62

6-3 X-ray diffraction patterns of oven dried and ashed dairy manures (3) showing the
presence of Mg-Ca Whitlockite (Mg-Ca phosphate). .............................................63

6-4 Energy dispersive spectrum of an ashed dairy manure showing the high intensity
peaks of C a, M g and P ............................... .... .......... ................ ............. 63

6-5 Energy dispersive dot maps of a dairy manure showing an association of Mg and
P ....................................................................................... . 6 4

6-6 Energy dispersive dot maps (scale 20 [tm) of a dairy manure-amended soil clay
showing an association of P and Ca. .............................................. ............... 64

6-7 Dot image of dairy manure showing the spatial associations of Mg, P, and Ca. .......65









6-8 EDS spectrum of a manure P rich particle obtained at 400X magnification
showing the dominance of M g, P, and Ca......... .............................................. 65

6-9 Relationships between P, Mg and Ca for active (6-9a & 6-9b) and abandoned
dairy (6-9c & 6-9d) manure-amended dry sieved (45km) silt+clay using the
electron probe microanalyses (EPMA). **significant at p<0.01. ..........................66

7-1 P concentrations in the presence of Mg during 20 weeks (wk) of incubation.
Different letters (a and b) indicate statistically significant differences among
median concentrations (p<0.05) observed after an incubation period. HAP =
H ydroxyapatite. ........................................................................79

7-2 XRD pattern of the precipitate in the Mg solution after 20 weeks of incubation.......79

7-3 XRD patterns of precipitates observed after 20 weeks of incubation in control (no
inhibitor), Si, and D O C treatm ents..................................... .......................... 80

7-4 Ca concentrations for control and Mg treatment during 20 weeks of incubation.
Different letters indicate statistically significant differences (p<0.05) of median
Ca concentrations after an incubation period. ................. ............................... 80

7-5 Variations in P concentrations in the presence of Si during 20 weeks of
incubation. Different letters indicate statistically significant differences at
p<0.05. HAP=Hydroxyapatite................................... 81

7-6 P concentrations in the presence of soil DOC during 20 weeks of incubation.
Different letters indicate a significant difference (p<0.05) after the specified
incubation period .................................................................. .. .......... ..... 81

7-7 Changes in DOC concentrations in control and in the presence of solids during the
20 w eek (w k) incubation study. ........................................ .......................... 82

7-8 Effects of clay size fractions on P concentrations during 20 weeks of incubation.
Different letters indicate statistically significant differences atp<0.05 level..........82

7-9 Variations in P concentrations due to the presence of low-density- (float) and
high-density- (sink) clay during 20 weeks of incubation. Different letters
indicate a significant difference (p<0.05) after the specified incubation period......83

7-10 Variations of Ca concentrations due to the presence of low-density- (float) and
high-density- ("sink") clay during 20 weeks of incubation. Different letters
indicate a significant difference (p<0.05) after the specified incubation period......83

7-11 EDS spectrum of the low-density clay showing the dominance of Si....................84

7-12 SEM imaging of low-density clay showing the presence of biogenic silica
("dumbbell" serrated shaped particles in image on left; rod shaped particles in
im age on right). .......................................................................84









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

DAIRY MANURE-COMPONENT EFFECTS ON PHOSPHORUS RELEASE FROM
SANDY SOILS


By

Manohardeep Singh Josan

May 2007

Chair: Vimala D. Nair
Major: Soil and Water Science

Phosphorus (P) in heavily manure-amended soils can be labile even years or

decades after manure input cease. Knowledge of manure-derived components and their

associations with P is pertinent to nutrient management for sandy soils with minimum P

sorbing capacity. The overall objective of this research was to understand the effects of

manure-derived components such as Mg, Si, and dissolved organic carbon (DOC) on P

solubility in manure-amended sandy soils. Soil samples (0-25 cm) from manure-impacted

areas were collected from the Suwannee and Okeechobee Basins of Florida. The soil

release of P, Ca and Mg was studied using repeated water extractions and 1.0 MNH4C1

extractions. Columns of the soils were leached with deionized water and leachate

speciation was modeled using MINTEQ. Solid state assessments of dairy manure and

manure-amended soils were done using x-ray diffraction, scanning electron microscopy,

and elemental microanalysis. The inhibitory effects of Mg, Si, and manure-derived DOC

on Ca-P crystallization were studied by incubating solutions with and without clay-sized

solids for 20 weeks. Repeated water and ammonium chloride extractions and speciation

of column leachates confirm that sparingly-soluble phases of P associated with Mg and









Ca control P release from the manure-amended soils and maintain elevated P

concentrations in soil solutions even years after abandonment of the dairies. Solid state

assessments suggested Mg-P and Ca-P associations in dairy manure and manure-

amended soils. Formation of the most stable Ca-P mineral, hydroxyapatite, was inhibited

by Mg and/or DOC, but not Si, in dairy manure amended soils. Mg-P associations in

manure and manure-amended soils could maintain elevated P solubility, and Mg in soil

solution could inhibit formation of stable forms of Ca-P. Therefore, consideration of Mg

and Ca is necessary to explain the nature of P in manure-amended soils. Preemptive

dietary controls to maximize Ca-P and minimize Mg-P in manure would be a strategy to

reduce P loss from these soils in the future. Application of Al-based water treatment

residuals could minimize the release of P from the manure-amended soils. DOC

inhibition of Ca-P precipitation and competitive effects on P sorption reduce prospects

for stabilizing P reactions in heavily manure-amended soils.














CHAPTER 1
INTRODUCTION

Continuous release of phosphorus (P) in dairy manure-amended sandy soils even

years after manure addition cease (dairy abandonment) poses both a scientific mystery

and an environmental problem (Nair et al., 1995). The USEPA (1996) identified P-

induced eutrophication as the most extensive cause of water quality impairment in the

USA, and the USGS identified agriculture as a major source of P to surface waters

(United States Geological Survey, 1999). Dairy manure accumulation in soils can

increase the potential for P loss to surface waters either via erosion (Sharpley & Smith,

1983) or subsurface drainage (Mansell et al., 1991). The P enrichment can cause both

surface and sub-surface water pollution (Whalen & Chang, 2001). Many soils effectively

retain P, but some sandy soils can be exceptions due to a paucity of P-retaining minerals

(Neller et al., 1951; Ozanne et al., 1961; Gillman, 1973; Burgoa, 1991; Mansell et al.,

1991; Harris et al., 1996; Nair et al., 1998; Novak et al., 2003). Thus, the stability of

manure-derived forms of P is an especially relevant environmental concern in sandy

soils.

Soil environmental factors such as pH, the presence of dissolved completing

species and the kind of phosphate mineral present determine the phosphate activity in

solution (Lindsay, 1979). Dairy manure-amended sandy soils typically contain large

amounts of Ca and P in both solid and solution phases, with accompanying moderately

higher pH (Nair et al., 2003) than the native, non-impacted soils. The conditions

thermodynamically favor the formation of relatively stable Ca-P minerals (Lindsay,









1979), but the release of P from these soils can be greater than predicted from the

solubility of the minerals (Wang et al., 1995). High Mg concentrations in soil solution of

heavily manure-amended soils suggest that Mg, in addition to Ca, could control the

release of P via a sparingly soluble Mg-P phase (Nair et al., 1995; Josan et al., 2005).

Alternatively, Mg can act as an inhibiting cation for Ca-P crystallization on calcium

carbonate by masking adsorption sites, in the presence of high P concentrations (Yadav et

al., 1984). If the latter were the case, abandoned dairy manure-amended soils should

exhibit some Ca-P stabilization after soluble salts (e.g. MgC12, CaC12) are leached (Harris

et al., 1994). Active dairy manure-amended soils are those that currently receive dairy

manure, whereas abandoned dairy manure-amended soils are those soils where dairy

activities have ceased for at least 10 years.

Minerals such as vaterite, whitlockhite, monetite and struvite in poultry and pig

manure possibly control solution P and the majority of Ca and Mg in the soil solutions of

soils amended with poultry manure are completed by dissolved organic matter (Bril &

Salmons, 1990). Density separation of the clay fraction of heavily P fertilized loamy soils

yielded P-rich particles associated with Fe, Al and Ca (Piersenyzki et al., 1990a).

Scanning electron microscopy (SEM) and energy dispersive x-ray spectroscopy (EDS) of

poultry manure suggested that sparingly soluble Ca and Mg -phosphate minerals

controlled soil solution P concentrations (Cooperband & Good, 2002), the presence of

discrete P forms/minerals in dairy manures and dairy manure-amended soils is lacking

(Harris et al., 1995; Cooperband & Good, 2002). X-ray absorption near-edge

spectroscopy (XANES), a non-destructive chemical speciation technique, was used to

assess P minerals and P speciation in "phosphorus-enriched agricultural soils"









(Beauchemin et al., 2003).The authors estimated that most of P was present as

octacalcium phosphate (OCP) (45% of total P) and hydroxyapatite (HAP) (11% of total

P).

Rationale and Significance

Phosphorus in intensively manure-amended soils can be labile years or decades

after abandonment (Nair et al., 1995), and cause P leakage at environmentally-

unacceptable rates. Heavy loading of dairy manure to soils, leads to continuous release of

P, Ca, and Mg both in active and abandoned dairy manure-amended soils (Nair et al,

1995). If a sparingly soluble Mg-P phase is responsible for the continued release of P

from the manure-amended soils, other amendment (e.g. water treatment residuals)

applications may be the only way to stabilize the P (O'Connor & Elliott, 2000) because

all Mg-P phases are relatively soluble (Lindsay, 1979). Alternatively, reduction of P via

dietary management could be a viable strategy to reduce P solubility in the manure (Dou

et al., 2003; Cerosaletti et al., 2004). Additionally, if a Ca-Mg-P phase controls the

release of P from the soils, it could be subject to dissolution. Ruminant feed contains both

Mg and Ca and the Mg and Ca doses are typically exceeding recommended levels

(National Research Council, 2001). During heat stress, increased diet potassium and

magnesium levels are recommended (Beede & Shearer, 1991).

It is important to document if Ca or Mg or both are initially associated with the

high release of P from dairy manure-amended sandy soils. A finding that Mg-P is

prevalent would diminish the prospect of stabilization via slow crystallization and

transformation of P to more stable Ca-P forms. Also, loss of more soluble forms of Ca

after abandonment would diminish the prospect of P released from a Mg-P phase being

precipitated as Ca-P.









Knowledge of manure-derived components and their associations with P is

pertinent to nutrient management, particularly for sandy soils with minimum P sorbing

components. This research probed the forms of Ca-P and Mg-P in dairy manure and dairy

manure-amended soils in hopes of identifying phases responsible for long-term P release

and developing mitigation approaches to reduce P loss. The overall objective was to

understand the role of manure-derived components, specifically Mg, Si, and dissolved

organic carbon (DOC), in maintaining high P solubility in active and abandoned dairy

manure-amended sandy soils. Specific hypotheses and objectives are given below:

Hypotheses

1. Active and abandoned dairy manure-amended soils release comparable amounts of
P because solution P is controlled by sparingly-soluble Mg and/or Ca phosphate
phases that require many years for depletion.

2. Concentrations of P, Ca, and Mg are spatially correlated in solid manure and
manure-amended soil samples.

3. Activities of DOC, Mg and Si in soil solution of manure-amended soils are
sufficient, jointly or separately, to inhibit crystallization of stable Ca-P forms, thus
leading toward high P release from the soils.

4. Noncrystalline Si forms (including biogenic Si in dairy manure) can retain P at
circumneutral pH and high Ca activity because Ca serves as a bridge between the
silicate surface and P.

Specific Objectives

1. Assess the release of P, Ca and Mg in soil solutions of dairy manure-amended soils.

2. Study the associations of P, Ca, and Mg in dairy manure and manure-amended soils
using solid state assessments.

3. Study the effects of Mg, Si and DOC on Ca-P crystallization using average
concentrations of the species found in manure-amended soil leachates.

4. Study the role of "low-density clay" (<2 Mg m-3) and "high-density clay" (> 2 Mg
m-3) from manure-amended soils on Ca-P crystallization.









Chapter 2 discusses previous researches on P release, geochemical modeling, and

solid state assessments in relation to soil and manure environments, and prospective

inhibitors of Ca-P crystallization in soil science and other fields. The soil and manure

sample collection and characterization procedures are discussed in Chapter 3. The first

study (Chapter 4) assesses the release of P, Ca and Mg in soil solution and tested the

hypothesis that abandoned and active dairy manure-amended soils release comparable

amounts of P because solution P is controlled by a sparingly-soluble Mg-P phase or Ca-

Mg-P phase that requires many years for depletion. Testing included repeated water

extractions and a column leaching experiment. A selective dissolution study (Chapter 5)

was done on manure-amended soils. Various inorganic and organic forms of P were

measured and the relationships to dissolved Ca, Mg, Fe, and Al were documented.

Chapter 6 describes the associations of P with Ca and Mg in minimally altered samples of

dairy manure and manure-amended soils, using x-ray diffraction (XRD), scanning

electron microscopy (SEM), energy dispersive spectroscopy (EDS), and electron probe

microanalyses (EPMA). The study was conducted to confirm the associations of P, Ca

and Mg observed in solution P chemistry. Results tested the hypothesis that Ca-P, Mg-P

or Ca-Mg-P phases exist in dairy manure and manure-amended soils, and are spatially

associated at the microscopic level. The results of the column leaching experiment

(Chapter 4) and x-ray diffraction studies (Chapter 6) suggested that the dairy manure-

derived components can inhibit the formation of stable Ca-P mineral phases. Therefore,

the inhibitory effects of dairy manure-derived components on the Ca-P stabilization were

investigated using an incubation study (Chapter 7). The final chapter (Chapter 8)

summarizes the four individual studies and how they relate to each other.














CHAPTER 2
REVIEW OF LITERATURE

Manure Application

In intensive dairy production areas, including the Okeechobee basin in Florida

(FL), more dairy manure is generated than needed to meet crop nitrogen (N) requirements

of available crop land. The long-term application of dairy manure at N-based rates in

such areas has increased levels of P in soils above the crop needs (Sharpley et al., 1996;

Kleinman et al., 2000), and values frequently approach or exceed environmental P

thresholds. Lawsuits have been filed in some animal production area, such as in Eucha-

Spavinaw Basin of Ozarks (DeLaune et al., 2006) to restrict the land applications of

animal wastes thus leaving surplus at the farm. Sims et al. (2000) recommended the

integration of soil P tests into environmentally based agricultural management practices

so that optimum levels of manures can be applied to a piece of agricultural land.

Release of P Using Various Extractants

The release of P from manure-amended soils with successive extractions has been

studied by several researchers, using different soil extractants. Sharpley (1996) observed

that the release of P with 0.01 M CaC12 using Fe strips as the adsorbent decreased

exponentially with successive extractions. A sequential extraction before and after 15

strip P extractions revealed that majority of P released (46%) was inorganic. Nair et al.

(1995) repeatedly extracting manure-amended soils with 1.0 MNH4C1 solution, and

found that z 80% of total P in the surface horizons of these soils was labile P (readily

soluble P) and associated with Ca and Mg-P forms. Authors speculated that P was loosely









bound with Ca and Mg, probably by some weak adsorption mechanism or as poorly

crystalline solids, and as available for sustained leaching under suitable conditions. The

classical use of soil P fractionation of Chang & Jackson (1956) use 1.0 MNNH4Cl to

effectively remove water soluble and loosely bound P during the initial extractions. The

Hedley et al. (1982) P fractionation procedure is broadly adopted to differentiate various

organic and inorganic pools of P. Additionally, NH4C1 extractable P has been defined as

loosely adsorbed or easily available P (Psenner & Pucsko, 1988) or easily soluble P

(Williams et al., 1967). Cooperband & Good (2002) suggested that "sparingly-soluble"

Ca and Mg- P minerals (more soluble than apatite) controlled solution P concentrations in

soils amended with poultry manure but were unable to directly identify the forms. He et

al. (2004) sequentially extracted 0.25 g of dairy manure in 25 mL of deionzed water for 2

hours, and this water extractable P was the largest fraction of total-P. Most of the

estimated P was in the inorganic form (12 to 44% of manure total P), and water

extractable P was better correlated (r2 = 0.62) to total P than organic P (r2 = 0.24). There

have been few efforts to study the associated cations like Ca, Mg, Fe and Al that are

released with P in manure-amended soils (Sharpley et al., 2004; Josan et al., 2005;

Silveira et al., 2006).

Geochemical Models: Forms and Solubility of Phosphorus

Various computer geochemical speciation models can assist in understanding the

forms and P-speciation in waste waters and soil solutions. The most commonly used

models are MINTEQ (Felmy et al., 1984), MINTEQA2 (Allison et al., 1990),

GEOCHEM (Sposito & Mattigod, 1980), and V-MINTEQ (Department of Land and

Water Resources Engineering. 2006, an updated version of MINTEQA2).









All the models assume equilibrium among dissolved species. The assumptions for

the equilibrium for aqueous (dissolved) speciation reactions are likely appropriate,

because most of the interactions occur very rapidly (Pankow & Morgan, 1981). Most of

the ion exchange and adsorption/desorption reactions in well stirred soils systems attain

equilibrium within several hours (Mattigod, 1995). These models are called equilibria or

solution speciation models and incorporate corrections for activity coefficients and

solution complexation reactions and, thus, evaluate the saturation status of the solution

with respect to thermodyamically stable solid phases. Many models assume activity

coefficients depend only on the ionic strength and ignore specific ion effects. Ion

interactions become important above ionic strengths of 0.5 mol L-1, and must be

considered (Mattigod, 1995). Some models, like MINTEQ (Felmy et al., 1984) and

GEOCHEM (Sposito & Mattigod, 1980), also assume that the solution is in equilibrium

with the thermodynamically predicted solid phases. Such models are predictive in nature

and are called solution-solid equilibria models. The models modify the solution

composition, assuming that only the most stable phase (or phases) can occur and that the

solution is always in equilibrium with the solids. Predictive models can also be used as

solution speciation models by disabling the solid phase reactions.

Zhang et al. (2001) used MINTEQA2 to predict the nature of aluminum and iron-P

fractions in sandy soils of Florida. Wavelite, crandallite, variscite and strengite were

predicted to stable in fertilized acid soils, whereas at higher soil pH values, Ca-P minerals

were predicted to control P activities in soil solutions. Sharpley et al. (2004) studied the

effect of long term manure applications on soil Ca-P forms using MINTEQA2. Ion

activity products of soil solutions were calculated using ion activities determined in









extracts of soil reactions for 16 h with 0.01 M CaC12 at 1:5 soil/solution ratio. Long term

(10-15 y) manure-amended soils were dominated by the more soluble crystalline Ca-P

forms tricalcium phosphate and octacalcium phosphates than the less HAP. Shanker &

Bloom (2004) cautioned that geochemical modeling results should be supported by solid

state assessment techniques. They also observed that

Oversaturation alone is merely a prerequisite for mineral formations; non-
equilibrium is the common state in soils and kinetics of precipitation would be an
important factor to dictate whether and to what extent various Ca-, Fe-, or Al-P
would accumulate in the soil.

Hutchison & Hesterberg (2004) studied the effects of dissolved organic carbon

(DOC) as citrate on P dissolution on the soil amended with swine lagoon. With increasing

rates of citrate concentrations, dissolved reactive P (DRP) and total Fe and total Al

concentrations were increased. Geochemical modeling (V-MINTEQ) predicted 69 to

99% of aqueous Fe(III) completed as Fe(citrate)0, and 87 to 100% of Al(III) completed

as Al(citrate)o and Al(citrate)3-. The authors concluded that Fe and Al can be completed

by dissolved organic matter as Al-DOM or Fe-DOM complexes in the system. Silveira et

al. (2006) studied P solubility characteristics in a dairy manure-amended sandy soil under

same conditions. Using V-MINTEQ as a speciation model for leachates, they observed

HPO42 (-50% of total soluble P), and Mg-P [MgHPO4(aq)] and Ca-P (CaPO4 ) complexes

(-30 and 13% of total solubleP, respectively) as the major chemical P species and thus

concluded that both Ca-P and Mg-P mineral phases can be a major factor in controlling

long term P release in manure-amended soils.

Geochemical models can be valuable tools in describing P speciation in various

kinds of soil environments. However, model predictions must be validated with solid

state assessments to confirm the presence of a particular mineral phase or a phase









association of P with other metals at micro scale (Mackay et al., 1986; Brennan &

Lindsay, 1998).

Solid State Assessments and P Associations

Solid state assessments are less destructive than chemical or solution extractions

and can be helpful in providing better description of relationships of P with metals. X-ray

diffraction (XRD) produces constructive interference of coherently scattered x-rays, and

produces diffraction peaks related to spacing of atomic planes in samples and wavelength

of x-rays (Amonette, 2002). X-ray diffraction needs minimal soil preparation; however,

the identification of poorly ordered/short range materials is impossible (Harris et al.,

1994). Scanning electron microscopy (SEM) provides large depth of fields and requires

minimal sample preparation, and particles can be seen at very high resolution (Goldstein

et al., 2003). Pierzynski et al. (1990a) successfully quantified the P minerals formed

under excessively fertilized soils where total P concentrations ranged from 540 mg kg-1 to

8340 mg kg-1. Researches used XRD, scanning electron microscopy (SEM) equipped

with energy dispersive spectroscopy (EDS), and Fourier-transformed infrared

spectroscopy to identify P minerals in concentrated clay size fractions obtained from

excessively fertilized soils. The use of XRD was unsuccessful in identifying P bearing

minerals in the soils. However, SEM identified P-rich particles in the soils with

'detectable quantities' of Al, Si, Ca and Fe that were associated with P. Huang & Shenker

(2004) studied the solid state speciation of P in stabilized sewage sludge using XRD,

SEM, and energy dispersive x-ray spectroscopy (EDXS). The XRD patterns confirmed

the formation of brushite, ferrian variscite, calcite and dolomite in ferrous sulfate

stabilized sludge. The SEM images also showed an elemental spatial correlation of P and

Ca, P and Fe and confirmed the presence of minerals detected in XRD analyses. Using









transmission electron microscopy, Jager et al. (2006) observed the presence of HAP (with

significantly broadened XRD peaks), which was confirmed by the solid state nuclear

magnetic resonance (NMR) technique. The Ca/P ratio of 1.52 estimated by the NMR

studies agreed with Ca/P ratios obtained from chemical analyses of nanocrystalline

hydroxyapatite.

Prospective Inhibitors on Ca-P Crystallization

Magnesium is a biologically essential element and reduces heat stress in ruminants

(Beede & Shearer, 1991). In the presence of high concentrations of grass K, the animal

absorbs less Mg in the rumen and suffers hypomagnesemia (Littledike et al., 1983). As a

result, there is a tendency to feed more Mg as feed supplements. Mg affects the

crystallinity of synthetic apatite (LeGeros et al., 1989; Bigi et al., 1993) and also inhibits

the formation of apatite (LeGeros et al., 1989; Abbona & Franchini-Angela, 1990). An

amorphous phosphate phase, magnesium-whitlockite, was observed in the presence of

Mg (LeGeros & LeGeros, 1984). Martin & Brown (1997) investigated the effect of Mg

on the formation of Ca-deficient HAP [Ca9HPO4(P04)50H, CDHAP]. The progress of

the reaction was determined by isothermal calorimetry. They observed two heat peaks

during the formation of CDHAP in water at 37.40C with 10 mM of Mg concentrations.

High Mg ion concentrations (3.16 M) resulted in the formation of Newberyite

(MgHPO43H20), but no Ca-P mineral phase was observed. Researches concluded that

Mg-P complexes are more likely to inhibit the Ca-P formation than magnesium chloride

complexes. Mg delays (or prevents) the conversion of the amorphous Ca-P phase into a

crystalline P phase (LeGeros et al., 1976). Sahai (2005) modeled the apatite nucleation

using crystallography, and NMR and suggested that an outersphere complex of Mg with

P formed faster than Ca-P inner-sphere complexes, which inhibited the Ca-P interactions









at the active silanol sites. The behavior of Mg was attributed to a greater charge density

for Mg than Ca, which favors greater electrostatic attractions at the active sites. The

effect of manure-derived silica on Ca-P interactions has not been studied. In a dental

study, Damen & Cate (1992) observed decreased induction time of Ca-P precipitation in

the presence of SiO2, which resulted in spontaneous precipitation of calcium phosphate

with a wide range of Ca:P ratios from supersaturated solutions. Addition of organic

amendments to calcareous soils increased P solubility with time more than a single

addition of inorganic phosphate (O'Connor et al., 1986). Inskeep & Silvertooth (1988)

observed that organic acids common to soil environments inhibited HAP precipitation

and concluded that organic acids adsorbed on to crystal seeds acting as nuclei for crystal

growth. Lindsay et al. (1989) suggested that adsorption regulates P retention at low P

concentrations, whereas mineral precipitation controls solubility at high P concentrations.

However, Kim et al. (2005) observed that formation of HAP from amorphous calcium

phosphate is an internal rearrangement process rather than a dissolution-precipitation

process. Long term addition of dairy manure to land not only increases P concentrations

but Ca, Mg, Si and DOC also. The effect of these manure-derived components on Ca-P

interactions needs to be addressed.














CHAPTER 3
SOIL AND MANURE CHARACTERIZATION

Soil Sampling

Soil samples were collected from manure-amended soils at four active (ACS-1 to

ACS-4) and five abandoned dairies (ABS-1 to ABS-5), and four minimally-impacted

soils (MIS-1 to MIS-4) from the Suwannee River Basin and Lake Okeechobee Basin of

Florida. The active dairy manure-amended soils currently receive dairy manure, whereas

the abandoned dairy manure-amended soils no longer receive high levels of dairy manure

daily. Abandoned dairy sites were formerly heavily manure-amended and received high

manure loads similar to the active dairy sites. Years of abandonment ranged from 12-32

years. Soils were collected by tile spade to a depth of 25 cm, or to the bottom of the Ap

horizon (whichever was shallower), from representative locations within the high-

intensity areas. Composited Ap horizon soil samples were also taken at each site using a

4.0 cm diameter soil auger for bulk density calculations. Representative soil profiles (to a

two-meter depth) were sampled to include Bt/Bh-horizons (when present) for

classification and characterization purposes. Parent materials for all soils were sandy

marine sediments. Slopes were < 2% and drainage classes were estimated to be poorly- to

somewhat poorly drained. Active sites were bare or nearly bare, whereas abandoned sites

were grassed. Two of the active sites were on Spodosols and two were on Ultisols. All

abandoned sites were on Spodosols. One active site (ACS-1) and one abandoned site

(ABS-1) had fill material greater than 25-cm thick, which obviated classification for the

purposes of this study. Samples from one active (ACS-4) and one abandoned dairy site









(ABS-2) were supplied by other researchers. Soil profiles descriptions were not

confirmed for the two sites, but both sites were located in areas dominated by Spodosols.

Soil samples were collected to approximate depths of 25 cm for active and 15 cm for

abandoned dairies. All samples were either dried shortly after collection or stored moist

under refrigeration. Soils were air dried and crushed to pass 2-mm sieve before use. It

also helped in potential screening out CaCO3 material applied as fill material. All

minimally-impacted (MIS 1- 4) soil samples were obtained from other researchers

(Graetz et al., 1999).

Physicochemical Properties of Soils (Soil Characterization)

The soil pH, which is an indicator of soil reaction, was measured with a glass

calomel electrode assembly using 1:2 soil water suspensions. Soil electrical conductivity

(EC, dS m-) was determined in 1:2 soil:water ratio with the help of a conductivity

bridge. Total P and metals i.e. Ca, Mg, Fe, Al, Na, and K in the soils were determined by

the procedure outlined by Anderson (1974). One gram of soil was weighed into a 50 mL

glass beaker and placed in a muffle furnace at 350C for an hour. The furnace

temperature was raised to 550C and soil was ignited at this temperature for 2 h. The

furnace was allowed to cool for overnight. Few water drops were used to moist the ash

and 20 ml of 6.0 MHC1 was added with a graduated cylinder and allowed the solution to

evaporate slowly on a hot plate (80C). Additionally 2.25 ml of 6.0 MHC1 was added to

the digested sample to dislodge the residue and solution was transferred quantitatively

(Whatman # 41 filter paper) in to a 50 ml volumetric flask by washing the beaker several

times with smaller amounts of DI water. The filtrate was collected in the same flask each

along with rinsing the sides of the filter paper into the flask. Phosphorus was analyzed by









ascorbic acid colorimetry (Murphy & Riley, 1962) (U.S. EPA, 1993; method 365.1). All

metals were analyzed by atomic absorption spectrophotometry.

Particle Size Fractionation and Mineralogical Analysis

Air-dried soil samples (50 g each) were treated with bleach (10% sodium hypo-

chlorite, adjusted to pH 9.5), at a 1:20 soil:bleach ratio, overnight to oxidize organic

matter (Lavkulich & Wiens, 1970). The supernatants were siphoned off and the soils

transferred to 250-mL centrifuge bottles for washing (3 times) with 1.0 MNaCl to

remove entrained bleach. Samples were then washed with deionized (DI) water (3-4

times) to remove salt, until the supernatant appeared turbid. Deionized water adjusted to

pH =9.5 with Na2CO3 was added to promote dispersion. Sand was collected by sieving,

and clay and silt by centrifugation (Whittig & Allardice, 1986). Oriented mounts for clay

were prepared for X-ray diffraction (XRD) by depositing 250 mg of clay as a suspension

onto a porous ceramic tile under suction. Clay was saturated on the tiles with Mg and

solvated with glycerol following an initial XRD scan. Silt was mounted as an oriented

dry powder on a low-background quartz crystal mount.

Manure Sampling

Fresh manure samples (- 20% solids) were collected from four dairies at different

locations in Florida using 5 gallons polyvinyl buckets. The manure samples were dried at

65C and then ground to pass a 2mm stainless steel sieve.

Manure Characterization

The manure pH, EC, total P and total metals i.e. Ca, Mg, Fe and Al were

determined as outlined in the soil characterization section.









Mineralogical Analysis

Manure samples were treated with 10% hypochlorite adjusted to pH 9.5 using large

plastic beakers. The manure to bleach ratio was 1:50 (manure-amended soil:bleach ratio

was 1:20), as manure contains more organic material than the soils. After treatment to

reduce organic the sample were wet-sieved (45 [tM) to separate the sand from silt and

clay size fractions. Silt and clay fractions were separated using centrifugation (Whittig &

Allardice, 1986). Resulted materials i.e. both silt and clay fractions were scanned for

mineral characterization as explained in the soil characterization section.

QA/QC for Analyses

To assure data quality for P and metal analyses for soil and manure samples proper

QA/QC procedures were adopted and included 1 blank, 1 replicate, 1 spike, and 1

certified standard after every 15 samples run. All extractions were performed on triplicate

samples.

Statistical Analyses

Pair wise comparison by ANOVA of P and metal concentrations of active,

abandoned and minimally-impacted soils were done. Mean separations were done by the

Waller-Duncan procedure at 5% level of significance. Computations were performed

using SAS Institute software (SAS, 2001).

Results and Discussion

Soil Chemical Characterization

Active dairy soils had higher pH values (7.1 7.9) than the pH of abandoned dairy

manure-amended soils (6.0 7.2) (Table 3-1). Higher pH values in manure-amended

soils are due to high inputs of Ca and Mg from dairy manure (Kingery et al., 1994; Nair

et al., 1995; Iyamuremye et al., 1996; Eghball, 2002) and the buffering effects of added









bicarbonates, organic acids with carboxyl and phenolic hydroxyl groups (Sharpley &

Moyer, 2000; Whalen et al., 2000). The minimally-impacted soils had significantly (p <

0.05) lower pH values (3.8 5.7) than manure-amended soils. Manure-amended soils had

significantly greater EC values than minimally-impacted soils (Table 3-1) attributable to

the large amounts of Na, Ca, Mg and other salts in the added manure. Average Ca and

Mg concentrations for active (Ca = 7086 mg kg-1, Mg = 1292 mg kg-1) and abandoned

dairy (Ca = 11082 mg kg-1, Mg = 603 mg kg-1) soils were exceeded than the average Ca

and Mg concentrations of minimally-impacted soils (Ca = 301 mg kg-1, Mg = 25 mg kg-

1). One of the abandoned dairy soils (ABS-1) had an extremely high total Ca

concentration (>35,000 mg kg-1), probably as a result of lime added as a fill material to

the soil. Total P concentrations were similar for active and abandoned dairy soils

(average 2000 mg kg-1), but much greater than minimally-impacted soils (average ~

100 mg kg-1). Manure-amended soils and minimally-impacted soils had similar total Al

and Fe concentrations, which suggested that the long term addition of dairy manure did

not change Al and Fe concentrations significantly (p > 0.05) because both Al and Fe are

not a major dietary constituent of dairy animals (NRC, 2001).

Soil Mineralogy

Quartz was the dominant mineral in the sand, silt and clay size fractions of all soils

except for the one abandoned dairy sample (ABS-1) presumably influenced by fill

material (Figures 3-1, and 3-2), in which calcite (CaCO3) was dominant in the silt and

clay. The silt fraction of an active dairy site, also influenced by fill, had high calcite as

well. Calcite was detected in all samples and was probably derived from either manure or

amendments used to stabilize the soil for heavy animal traffic.









Other minerals present in minor to moderate amounts included kaolinite (two

abandoned dairies and one active dairy), smectite (one abandoned dairy), and hydroxyl-

interlayered minerals (three active dairies). No phosphate minerals were directly

identified in the samples analyzed via XRD. Pierzynski et al. (1990) and Harris et al.

(1994) were also unable to identify distinct P minerals in excessively fertilized and dairy

manure-impacted soils, respectively. Either the phosphate phases are noncrystalline or

mineral concentrations are too low for detection (<1%) without further preconcentration

(e.g., selective dissolution, density separation, etc.), or both. The presence of a broad

"amorphous hump" on clay XRD plots, observed between 16-20 20 (Figure 3-1),

suggested the presence of appreciable noncrystalline material, probably biogenic silica

(Harris et al., 1994) derived from plant phytoliths used as forage to dairy animals.

Minimally-impacted soils exhibited the same XRD peaks observed in manure-amended

soils, except the presence of hump related to amorphous biogenic Si (Figure 3-3).

Manure Characterization

Manures pH values (Table 3-2) exceeded soil pH (Table 3-1) values, ranging from

8.2 to 8.6. Manure total P concentrations ranged from 5965 to 6137 mg kg-1 on a dry

weight basis. Kleinman et al. (2005) surveyed the composition of 68 dairy manure

samples and reported average total P concentration of 6900 mg kg-1. Manure samples

used in this study contained similar averaged (5965 mg kg-1) total P concentrations. The

Ca and Mg concentrations ranged from 9899 to 13503 mg kg-1 and 2808 to 4281 mg kg-',

respectively and represent significance of P, Ca and Mg. Both Al and Fe concentrations

in manure were minimal. On average, manure samples contained 77% organic matter on

a dry weight basis, mainly from plant phytoliths used as fodder. Thus, long-term









applications of dairy-manure not only buildup P concentrations in soils, but also result in

accumulation of Ca, Mg, and very fine digested plant materials.

Manure Mineralogy

The XRD patterns of clays of four dairy manures from four locations yielded

similar results. Irrespective of the management practices, the same kind of mineralogical

components were observed (Figure 3-4). All samples dominated by a hump (16-20, 20),

which is believed to be biogenic silica derived from plant phytoliths. No P-bearing

minerals were detected in the manure samples. Quartz was detected in manure samples,

likely due to using sand as bedding material. Calcite was also observed in dairy manures.

Summary and Conclusions

Dairy manures contain higher amounts of P, Ca and Mg than dairy manure-

amended soils. The long term addition of dairy manure significantly altered soil chemical

properties and nutrient concentrations compared to minimally-impacted (native) soils.

Both active and abandoned manure-amended soils had higher pH values than native soils,

accompanied by higher electrical conductivity values. There was a significant P buildup

in manure-amended soils. The native P retention capacity of sandy soils is low (Mansell

et al., 1991), and is almost certainly exceeded by the levels of P loading characteristic of

high-intensity areas near dairy barns (Nair et al., 1995 and 1998). Calcium phosphate

minerals are predicted to be stable under these conditions using chemical equilibrium

modeling (Wang et al., 1995), where as no P-bearing minerals were detected in manure-

amended soils. Pre-concentrations of clay size fractions and minimally alteration of soil

samples can be helpful in identifying P-mineral phases associated in the soil.









Table 3-1. Characteristics of active and abandoned manure-amended soils, and
minimally manure-impacted soils.


Sample ID pH


aACS 1
ACS 2
ACS 3
ACS 4
Mean
SD
bABS 1
ABS 2
ABS 3
ABS 4
ABS 5
Mean
SD
CMIS 1
MIS 2
MIS 3
MIS 4
Mean
dSD


7.9
7.1
7.5
7.6
7.5ae
0.3
6.8
7.0
6.4
6.0
7.2
6.7a
0.5
5.7
6.6
3.8
5.0
5.3b
1.2


fEC
dS m-
0.61
0.68
0.63
0.62
0.64 a
0.03
0.53
0.34
0.35
0.44
0.82
0.50a
0.20
0.21
0.20
0.11
0.07
0.15b
0.07


Total


Total Total


Total


Total


P Ca Mg Fe Al
----------------------------------mg kg -----------
1752 8786 629 845 3184
3117 5909 2166 1231 4014
3251 8452 1600 1161 2691
1215 5197 774 435 311
2334 a 7086 a 1292a 918a 2550a
1007 1799 723 363 1590
2796 35549 1063 869 1860
1944 4000 584 820 1237
2485 6708 374 194 1184
1001 3256 271 358 1528
2230 5897 722 347 308
2091a 11082a 603b 518a 1223a
686 13748 312 306 578
58 52 23 1055 2854
63 642 41 194 136
193 446 25 153 122
101 62 15 318 1439
104b 301b 25c 430a 1138a
63 292 11 423 1300


aACS = active dairy manure-amended soil
bABS = abandoned dairy manure-amended soil
cMIS = minimally-impacted soil.
dSD= standard deviation
eMean values of soil parameters within active, abandoned and minimally impacted soils
followed by the same letter in a column are not significantly different (p > 0.05).
fElectrical conductivity (EC) 1:2 soil: water ratio









Table 3-2. Characteristics of dairy manures collected from four locations in Florida.
Sa Organic Eb Total Total Total Total Total
ample p Mattera P Ca Mg Fe Al


% dS m ------
Manure-1 8.2 62 0.92 6071 13
Manure-2 8.5 83 0.98 6137 11
Manure-3 8.6 84 0.91 5670 13
Manure-4 8.5 79 0.87 5983 c
Mean 8.5 77 0.92 5965 11
cSD 0.2 10 0.05 207 1
organic matter determined by loss on ignition
bEC = Electrical conductivity; 1:2 manure: water ratio
cSD = standard deviation


-----mg kg ----------------- ---


052
310
503
)899
941
658


2808
4281
3888
3054
3508
693


203
268
147
125
186
64

































Two-Thela (de a


Figure 3-1. X-ray diffraction patterns of clays obtained from active dairy (ACS) manure-
amended soils. HIV = hydroxyinterlayered vermiculite.


TwG-Stheta (deg) -


Figure 3-2. X-ray diffraction patterns of clays obtained from abandoned dairy (ABS)
manure-amended soils.


























OW 3.9746 i
I 1.l746 Kaolinite





I I










soils.







Quartz





Calcit
1 = 300?


obtained from minimally-impacted (MIS)


Figure 3-4. X-ray diffraction pattern of clay obtained from an oven dried dairy manure.














CHAPTER 4
ASSOCIATED RELEASES OF PHOSPHORUS, CALCIUM AND MAGNESIUM IN
SOIL SOLUTIONS FROM DAIRY MANURE-AMENDED SOILS

Introduction

Continuous release of P from dairy manure-amended soils enriches adjacent water

bodies with P (Graetz & Nair, 1995). Long term addition of manure to soils can alter the

chemical and physical characteristics (Nair et al., 1995; Eghball, 2002; Josan et al., 2005;

Silveira et al., 2006). Generally, the fate of P in soils is controlled by the inherent soil

components; however, excessive manure applications can alter the nature and fate of P

forms by the dairy manure-derived components. Several studies of P in dairy manure-

amended soils have been conducted. Nair et al. (1995) studied the forms of P in manure-

amended soils of south Florida and found that 70% of total P in surface soil was

associated with Ca-Mg. Results presented earlier (Chapter 3) showed that the addition of

dairy manure in soils increased total Ca, and total Mg concentrations in soils. Therefore,

we hypothesized that abandoned and active dairy manure-amended soils release

comparable amounts of P because solution P is controlled by a sparingly-soluble Mg-P

and/or Ca-Mg-P phase that requires many years for depletion. The hypothesis was tested

in two studies. The first study consisted of repeated water extractions (wide soil:solution

ratio and minimum soluble salt). The second study was a column leaching experiment,

implemented to provide a narrower soil:solution ratio as compared to sequential

extractions. The narrower soil:solution ratio was expected to more closely approximate









equilibrium conditions and, therefore, cation and anion concentrations useful to chemical

modeling ofP speciation in the leachates.

Methods and Materials

Repeated Water Extractions and Chemical Analyses

Soil (20 g) was repeatedly extracted with 200-mL of deionized water (DI) eight

times. The soil suspensions were initially shaken for 5 min, and repeated for

progressively longer intervals (0, 0.5, 3, 6, 12, 24, 36 and 48 h), for a total of 8, in

successive extractions. After each extraction, the samples were centrifuged at 738 x g for

5 min. The supernatants were collected by decanting and filtered 0.45 [im filter. All

extractions were conducted at room temperature (25C). The collected filtrates were

analyzed for pH, electrical conductivity (EC), soluble reactive phosphorus (SRP), and

total dissolved Ca, Mg, Na, K, Fe and Al.

All P determinations were carried out on a UV-visible recording spectrophotometer

at 880 nm wave-length via a molybdate-blue colorimetric procedure (Murphy & Riley,

1962) (U.S. EPA, 1993; method 365.1). The filtrates were analyzed for metals by atomic

absorption spectroscopy. Total inorganic carbon and total carbon was determined in all

the extracted solutions using a carbon analyzer (TOC-5050A, Shimadzu) (method 5310A,

1992). Total organic carbon in the solutions was determined by difference.

Soil Leaching Characterization and Chemical Equilibrium Modeling

Leachates from surface samples of four active and four abandoned dairy-manure

impacted soils were collected using a column approach (Figure 4-1). Ten acrylic columns

were constructed with a stopper at one end fitted with glass tube and glass wool. Air-

dried forms of the 10 soils (4 active dairy and 4 abandoned dairy manure-amended soils,

including two replications from each soil type) were packed in columns to a height of 30









cm and a density of 1.2 g cm-3. The internal diameter of the columns was 5 cm, so the

volume occupied by the soil to a depth of 30 cm was (3.1416 2.5 2.5 30) = 589

cm-3. Therefore the weight of soil in each column was 589 x 1.2 = 707 g. Soil was

adjusted to a moisture content of 25% (air dry basis) by very slowly adding 283 mL of DI

water. This amount was calculated as x 707 g = 0.25 x, where "x" = final weight after

water added sufficient to equal 25% of the final soil weight. This amount of water would

constitute about 88 % of the pore volume. Columns were subsequently leached by slowly

adding 283 mL of DI water over a 2-hour period, 2.35 mL min-1, and allowed to drain for

at least 16 hours (overnight). Leachates were collected in 500-mL beakers, using plastic

wrap to keep out dust, and transferred to scintillation vials for chemical analyses. A

portion of leachates was transferred to 20 mL plastic vials and kept frozen for backup

analyses. Leachates were analyzed for metals (Ca, Mg, Na, K, Fe, Si and Al) by

inductively coupled plasma-atomic emission spectroscopy (ICP-AES) using EPA method

200.7. Soluble reactive phosphorus (SRP) concentrations were measured using the

ascorbic acid colorimetry (U.S. EPA, 1993; method 365.1). Chlorides were determined

using EPA method 325.2, nitrates by automated colorimetry with the use of ALPKEM

Auto-analyzer (U.S. EPA, 1993; method 353.2), ammonium by the semi-automated

colorimetry method (U.S. EPA, 1993; method 350.1), and sulfateby ion chromatography

with separator AS-14 (DIONEX) (U.S. EPA, 1993; method 300.0). The pH of leachates

was determined and the ionic strength ([t) calculated from EC measurements (Griffin &

Jurinak, 1973):

/ = 0.013 xEC (1)









Dissolved organic carbon was determined by TOC-5050A, Shimadzu (method

5310A, 1992). Interferences from the inorganic carbon in leachates were first removed by

sparging with CO2 -free gas after acidification of the sample (Sharp & Peltzer, 1993).

Visual-MINTEQ version 2.51 (Department of Land & Water Resources

Engineering, 2004) was used as a chemical equilibrium model for speciation calculations

and solubility equilibrium indices for leachates. The model was chosen over other

existing models because of its wide applicability in soil science, its windows-based data

input (rather than DOS mode), its extensive thermodynamic database for the P species to

be modeled, and its ease of addition and modification of data codes (Mattigod, 1995).

The model was set to a charge balance of 30% i.e. the program would be terminated if

charge balance exceeded more than 30%. The activity corrections were calculated by

Davis equation (Davis, 1962) using Davis 'b' parameter 0.3. The equation is given below

1/2
logy, =-AZi2(- /2 0.3/) (2)
1+ pI/2

where y, is the activity coefficient of species 'i'and represents the ratio of the activity of

an ion to its concentrations (c,) (Lindsay, 1979), A is a temperature-dependent constant,

with a water value = 0.509 at 250C, Z, is the valence of the ion 'i' and P is the ionic

strength define as


p =I c Z2 (3)

Oversaturated solids were not allowed to precipitate, excluding the infinite solids, finite

solids or possible solids. Gaussian model for dissolved organic matter (DOM) was

selected to take into the account of the complexation of metals by dissolved organic

matter. Both pH-dependency and competition among multiple components that bind with









DOM are considered in this model (Dobbs et al., 1989a,b). This model is easy to use, as it

required dissolved organic carbon as input, as compared other two metal-humic

complexation models available in MINTEQ.

Results and Discussion

Repeated Water Extractions

The EC values of repeated water extractions were greatest (p<0.01) for active dairy

soils (Table 4-1), intermediate for abandoned dairy soils, and least for minimally-

impacted soils, with no overlap (Figure 4-2). The trend reflects the salts in manure and

partial depletion via leaching upon abandonment and cessation of manure loading.

Calcium removed by repeated extractions followed the same trend as for EC with respect

to site groupings (active>abandoned >minimally-impacted), but differed in that there was

no decline with successive extractions for active dairies (Figure 4-3). Active dairy

manure-amended soils had more (p<0.01) Ca in the eighth extractions (62.6 to 95.3 mg

Ca kg-1 soil) than abandoned dairy soils (32.9 to 62.0 mg Ca kg-1 soil) (Table 4-1). The

release of Mg and P, in contrast to Ca, was similar (p<0.01) for active and abandoned

dairy soils; however, like Ca, there was less (p<0.01) Mg and P release for the minimally-

impacted soils (Figures 4-4 and 4-5). The data are not consistent with the concept of a

highly-soluble phase of Mg and P that is depleted with abandonment. Data are consistent

with the concept that Mg and P exist mainly in sparingly-soluble form(s), which would

require a long time for depletion. The correlation between Mg and P release was much

stronger than the correlation between Ca and P release for both active (Figure 4-6) and

abandoned dairy (Figure 4-7) soils, suggesting a Mg-P phase in the manure-amended

soils. The r2 value of Mg with P changed from 0.68 to 0.71 for the active dairies and from

0.62 to 0.75 for the abandoned dairies when the first two extracts (most heavily









influenced by soluble salts, including short equilibration time) were removed from the

regression equation. Thus the P released from high intensity area dairy soils during

repeated water extractions was more closely associated with Mg than Ca release. Release

of Mg and P was similar in both active and abandoned dairy sites. The data are consistent

with observations that old abandoned sites remain a source of P. Release of Ca was less

for abandoned dairies than active dairies.

Soil Leaching and Chemical Equilibrium Modeling

Concentrations of ions column leachates were much greater than in the repeated

extractions (mean SRP = 33 mg L-1, Ca = 87 mg L-1, and Mg = 54 mg L-1 in the leachates

vs SRP = 4 mg L-1, Ca = 7 mg L-1, and Mg = 3 mg L-1 for repeated water extractions),

because of the much narrower soil:solution ratio in the column study (Detailed results of

column leachates are presented in Appendix A). The leachates, in contrast to repeated

extractions, showed a negative relationship between SRP and Ca and Mg concentrations

(Figure 4-8). The trend can be attributed to a common ion effect in the leachates arising

from the high concentrations of salts (consistent with high EC in the initial leachates).

The use of V-MINTEQ speciation model calculated the saturation index (SI) values of

different P minerals. The index represents relative concentrations of ions, and is defined

as the difference between the log of the ion activity product (IAP) and the log of the

solubility product (Ksp) for a particular solid i.e.

SI = log IAP log Ksp (4)

Both IAP and Ksp are calculated the same way, and the only difference is that IAP is

ratio of the activity of products to reactants measured in soil solutions, whereas the Ksp is

the ratio of activity of products to reactants that will be present in soil solution at









equilibrium with a specific mineral (Essington, 2003). SI values were interpreted using

the criteria outlined by Bohn & Bohn (1987). According to this criteria if the SI values

are -1
respect to that P-mineral. It can also be concluded that this particular P-mineral is present

and can control the solubility, provided equilibrium conditions exist. For SI values <-1,

lechate is regarded as undersaturated with respect to the mineral, and the mineral is in

disequilibrium and would dissolve. For SI values > 0, the leachate is considered as

supersaturated with respect to the P mineral and if the mineral is present, it can

precipitate. Chemical modeling (Table 4-2 & 4-3) indicated that all leachates were

supersaturated with respect to all but the most soluble Ca-P minerals (monetite and

brushite), whereas all leachates were either undersaturated or near saturation with respect

to all Mg-P minerals. The data are consistent with the idea of a sparingly-soluble Mg-P

and Ca-P phase controlling P release from the manure-amended soils and that the soild

phases would maintain elevated P concentrations in soil solutions even years after

abandonment of the dairies. Sharpley et al. (2004) conducted a study on 20 yr old

manure-amended silty loam soils. Addition of dairy manure altered the Ca-P chemistry of

the soils from hydroxyapaptite to tricalcium phosphate and octacalcium phosphate

reaction products. Sharpley et al. (2004) concluded that P release from the soils is

controlled by these Ca-P forms. However, Sharpley et al. (2004) did not consider the

possible role of Mg in controlling P release from the soils. Sylveira et al. (2006) used

manure-amended sandy soils of the Okeechobee basin, in a small column leachate study

that last for 36 weeks and resulted in 35 pore volume equivalents of leaching. The

researched concluded that both Ca-P and Mg-P minerals control P release from these









soils. Sylveira et al. (2006) recommended using (Al or Fe-based) water treatment

residuals (WTR) to minimize the release of P from the soils. The addition of WTR

changes the P release behavior of the soils to adsorption-desorption mechanisms instead

of simple dissolution of sparingly soluble P forms associated with Ca and Mg.

Summary and Conclusions

During repeated water extractions, release of P from high intensity area dairy

manure-amended soils was more closely associated with the release of Mg than Ca. In

addition, release of Mg and P in repeated water extractions was similar for both active

and abandoned dairy manure-amended soil samples. Active dairy soils released more Ca

than abandoned dairy soils and the EC of the extracts was also higher for active dairy

soils. Column leachate data suggested that leachates obtained from this study were

supersaturated with respect to the most stable Ca-P minerals and were near saturation

with respect to soluble Ca-P minerals. The leachates were either undersaturated or near

saturation for all Mg-P minerals considered. In effect, there appears to be no significant

highly-soluble phase of Mg and P that becomes depleted with abandonment. Therefore,

we concluded that the release of P from manure-amended soils is associated with the

release of both Mg and Ca. The solubility of P in sandy soils amended with dairy manure

likely controlled by sparingly soluble Mg-P and Ca-P phases, which are expected to

continue to release P for a very long time, probably several decades. Abandoned dairies

in this study were 12 to 32 years old.









Table 4-1. Cumulative average release of SRP, Ca, Mg and EC in repeated water
extractions of active, abandoned and minimally-impacted soils.
Soil Type N dEC fSRP Ca Mg
(dS m-) -----------mg kg-----------
aACS 4 el.lla 495a 691a 271a
bABS 5 0.47b 441a 477b 251a
CMIS 4 0.13c 0.3b 80.lc 5.34b
aACS = active dairy manure-impacted soil
bABS = abandoned dairy manure-impacted soil
cMIS = minimally-impacted soil
dEC = electrical conductivity
eMean values of soil parameters within active, abandoned and minimally-impacted soils
followed by the same letter in a column are not significantly different using Waller-
Duncan procedure at 5% level of significance (p < 0.05)
fSRP = Soluble reactive phosphorus











Table 4-2. Saturation indices (SI) in active and abandoned dairy manure-amended soils.


Struvite Farringtonite Newberyite Monetite Brushite Whitlockite OCPc HApd


Sample ID


--------------------Mg-P minerals------------


---------------------------Ca-P minerals-------------------


---------------------------------SI


ACSa 1.1b -2.52
ACS 1.3 -1.26
ACS 1.7 -1.70
ACS 2.1 -0.77
ACS 2.3 -0.89
ACS 2.7 -1.09
ACS 3.1 -0.43
ACS 3.3 -0.09
ACS 3.7 -0.70
ACS 4.1 -0.21
ACS 4.3 -0.28
ACS 4.7 -0.42
ABS 1.1 -1.52
ABS 1.3 -1.94
ABS 1.7 -1.53
ABS 2.1 -0.96
ABS 2.3 -0.88
ABS 2.7 -2.07
ABS 3.1 -1.27
ABS 3.3 -1.63
ABS 3.7 -2.26
ABS 4.1 -1.60
ABS 4.3 -1.47
ABS 4.7 -1.93
aACS = active dairy


-3.54
-2.95
-2.87
-1.22
-0.99
-1.19
-0.84
-0.18
-1.04
-0.85
-1.03
-1.39
-3.65
-3.65
11.95
-3.15
-1.39
-1.12
-2.12
-1.68
-3.05
-3.45
-3.82
-3.79


log IAP
-2.28
-1.65
-1.36
-0.99
-0.51
-0.54
-1.28
-0.87
-0.31
-1.09
-0.60
-0.61
-1.85
-0.99
-1.02
-0.96
-0.37
-0.53
-1.48
-0.86
-9.10
-1.15
-0.67
-0.85


-log Kspo-----------------


-1.31
-0.54
-0.23
-0.38
0.11
0.15
-0.82
0.14
0.29
-0.36
0.17
0.29
-0.42
0.37
0.40
-0.12
0.55
0.46
-0.16
0.54
0.51
0.10
0.66
0.49


manure-impacted soil; and ABS


-1.59
-0.82
-0.51
-0.66
-0.17
-0.13
-1.10
-0.14
0.01
-0.64
-0.11
0.01
-0.70
0.09
0.12
-0.40
0.27
0.18
-0.45
0.26
0.23
-0.18
0.38
0.21


1.70
2.70
2.87
2.96
3.20
3.22
2.89
3.06
3.09
3.67
3.62
3.64
2.95
4.09
3.46
3.48
4.00
3.18
4.60
3.46
2.77
2.66
3.06
2.59


abandoned dairy manure-


impacted soil
bl.1, 1.3, and 1.7 corresponds to 1st, 3rd, and 7th leaching events of soil.
cOCP = Octacalcium phosphate {Ca4H(PO4)3.3H20}
dHAP = Hydroxyapatite Ca5(PO4)3(OH)
Struvite: NH4MgPO46H20
Farringtonite: {Mg3(PO4)2}
Newberyite: (MgHPO4.3H20)
Monetite: (CaHPO4)
Brushite: (CaHP04.2H20)
Ca-whitlockite: {Ca3(PO4)2} (P)


-0.73
1.05
1.52
1.47
2.20
2.26
0.96
2.08
2.27
2.23
2.68
2.82
1.41
3.35
2.74
2.24
3.44
2.52
3.32
2.88
2.16
1.62
2.61
1.96


10.48
11.72
11.73
12.06
12.07
12.06
12.36
11.74
11.66
13.52
12.84
12.77
12.09
13.58
12.28
12.84
13.22
11.66
15.12
12.16
10.80
10.91
11.22
10.45










Table 4-3. Percent of observations that are undersaturated, saturated, and supersaturated
for selected minerals based on chemical equilibrium modeling of active and
abandoned dairy column leachates.

SoilsMinerals Struvite Farringtonite Newberyite Monetite Brushite
Soils/Minerals
NH4MgPO4 6H20 Mg3(PO4)2 MgHPO4 3H20 CaHPO4 CaHPO4 2H20
Undersaturated
Active dairies 32 57 32 4 11
Abandoned dairies 89 100 21 0 0
Saturated
Active dairies 68 43 68 32 64
Abandoned dairies 11 0 79 11 21
Supersaturated
Active dairies 0 0 0 64 25
Abandoned dairies 0 0 0 89 79

aBased on comparison of computed saturation index (SI) from ion activity product (IAP)
and solubility product (Ksp) (Bohn & Bohn, 1987).






































Figure 4-1. Column set-up used for the column leaching study.





0.35 -


S.
4"'



~~-a-~ -8 -8

*----------------- :-;:-:-~-.------------rrr--- I I Irii


I I I I


1 2 3 4 5
Sequential extractions


6 7 8


Figure 4-2. Changes in electrical conductivity (EC) (dS m-') with repeated water
extractions. ACS = active dairy manure-impacted soil; ABS = abandoned
dairy manure-impacted soil; MIS = minimally impacted soil.


0.30 -

0.25 -

0.20 -

0.15 -

0.10 -

0.05 -

0.00


- ACS-1
- ACS-2
- ACS-3
- ACS-4
-- ABS-1
-- ABS-2
- ABS-3
-x-ABS-4
- ABS-5
-- MIS-1
- MIS-2
- K- MIS-3
- MIS-4


I I I












160 -

140 -

120 -

100 -

80 -

60 -

40

20 -

0 -


1 2 3 4 5
Sequential extractions


6 7 8


Figure 4-3. Changes in Ca concentrations (mg kg-1) with repeated water extractions.
ACS = active dairy manure-amended soil; ABS = abandoned dairy manure-
amended soil; MIS = minimally impacted soil.


100
90 -
90


70 -
80




50 -
70

60

50

40

30
20

10

0


--ACS-1
- ACS-2
- ACS-3
- ACS-4
--ABS-1
---ABS-2
- ABS-3
-- ABS-4
- ABS-5
- MIS-1
-- MIS-2
-- MIS-3
-- MIS-4


- --


1 2 3 4 5 6 7 8


Sequential extractions

Figure 4-4. Changes in Mg concentrations (mg kg-1) with repeated water extractions.
ACS = active dairy manure-amended soil; ABS = abandoned dairy manure-
amended soil; MIS = minimally impacted soil.


---'- _- -
N. -- ~ 4B-- -
*- .
S-- .. -..
....- ..- --V 1
-- - --
- :1~~~-*~-.;. -


- ACS-1
- ACS-2
- ACS-3
-- ACS-4
-ABS-1
- ABS-2
- ABS-3
-x-ABS-4
--ABS-5
-- MIS-1
- MIS-2
- MIS-3
- *- MIS-4


\--------- ---- -











160

140 --ACS-1
-'-ACS-2
120 ---ACS-3
1--0-ACS-4
S100- --ABS-1
-'-ABS-2
80 < .<-... -- ABS-3

S60 -- -- --ABS-4
S----o-ABS-5
40 i MIS-1
4 *MIS-2
20 ---- -*- MIS-3
*- MIS-4

1 2 3 4 5 6 7 8
Extraction number


Figure 4-5. Changes in soluble reactive phosphorus (SRP) concentrations (mg kg-1) with
repeated water extractions. ACS = active dairy manure-amended soil; ABS =
abandoned dairy manure-amended soil; MIS = minimally impacted soil.



160 y= 1.98x- 3.81 Mg Ca y = -0.19x + 79.4
r2 = 0.68 (Mg) r = 0.01(Ca)
140 o .
120-
S100 x
S80 *
60 -* *4 x
40
20 *
0
0 20 40 60 80 100 120 140

Ca or Mg (mg kg 1)




Figure 4-6. Relationships between soluble reactive phosphorus (SRP) and Mg and Ca
released during repeated water extractions of active dairy manure-amended
soils.














y= 1.19x + 10.83
r2 = 0.62 (Mg)


y = 0.45x + 21.0
r2 = 0.20(Ca)


20 >'


0 20 40 60 80 100 120 140
Ca or Mg (mg kg-')


Figure 4-7. Relationships between soluble reactive phosphorus (SRP) and Mg and Ca
released during repeated water extractions of abandoned dairy manure-
amended soils.


P andCa P andMg
P and Ca P and Mg


100 200 300


SRP = -0.19Mg + 48.13
2
r = 0.20

SRP = -0.14Ca + 50.67
2
r = 0.27






700


Ca or Mg (mg L )

Figure 4-8. Relationships between soluble reactive phosphorus (SRP) and Mg and Ca
released during the column leaching of dairy manure-amended soils.


140

120

100

S80

60
vy


m
x
m
~ :m m /iu~














CHAPTER 5
RELATIONSHIPS BETWEEN PHOSPHORUS, CALCIUM AND MAGNESIUM
INFERRED FROM SELECTIVE DISSOLUTION

Introduction

Repeated water extractions can provide valuable information about the release

pattern of different P phases that may exist in manures or manure-amended soils. The use

of 8 repeated water extractions to study the release of P from the soils was demonstrated

in Chapter 4. The data suggested that water can extract P indefinitely from the dairy

manure-amended soils. Graetz & Nair (1995) used 10 extractions with DI water to extract

P from dairy manure-amended soils. From 2 to 18% of total soil P was extracted, and the

authors concluded that this "labile-P" will solubilize in several years.

The P concentrations in the 8th repeated water extraction used in this study ranged

from 20 mg kg-1 to 80 mg kg-1, and similar concentrations were expected to be released

indefinitely. In another experiment, Silveira et al. (2006) used 40 repeated water

extractions to deplete P concentrations to undetectable levels from dairy manure-

amended soils. Such experiments are time consuming. Using a higher ionic strength

solution, for example 1.0 M ammonium chloride (NH4C1) (Chang & Juo, 1963; Hieltjes

& Lijklema, 1980) or 1.0Mpotassium chloride (KC1) (Reddy et al., 1998), can

simultaneously extract both "soluble and exchangeable P" using fewer number of

extractions.

In addition to repeated extractions, sequential extraction techniques are used widely

to obtain information of various P forms in soils, manures, deposited sediments (Chang &









Juo 1963; Martin et al., 1987; Magnus et al., 1988; He et al., 2005). Sequential extraction

procedures use chemical reagents with varying degrees of chemical strengths. The first

step in these procedures is to extract or dissolve easily soluble components or forms of

nutrients relatively available to plants (Chang & Juo, 1963). Sequential extraction

procedures have been used extensively to make inferences about the fate of P. The

continuous release of P by sequential extractions with NH4C1 had been demonstrated for

dairy manure-amended soils (Graetz & Nair, 1995; Nair et al., 1995) and for the lake

sediments (Petterson & Istvanovics, 1988). Fewer authors have reported the concomitant

release of Ca, Mg, Fe and Al along with P for different kinds of sequential extraction

studies (Nair et al., 2003; Sharpley et al., 2004; Silveira et al., 2006). Generally, the first

two extractions efficiently dissolve CaCO3 during successive extractions by 1.0 MNH4C1

(Hieltjes & Lijklema, 1980).

In this study, both repeated and sequential extractions were used to document the

release of P, as well as the release of Ca, Mg, Fe and Al. This work is based on a study

conducted by Nair et al. (2003). Authors used 1.0 MNH4Cl to study P release

characterization of manure and manure-amended soils. Ammonium chloride was tested

for its ability to selectively extract the Mg-P phase first; however, this does not preclude

NH4C1 extraction of Ca-P particularly in the first few extractions. It was hypothesized

that the repeated use of NH4C1 can extract P and associated cations preferentially Ca and

Mg.

Material and Methods

Phosphorus Fractionation: Repeated and Sequential Extractions

For this study, two soil samples from active dairies (ACS-2 & ACS-4), and two

samples from abandoned dairies (ABS-3 and ABS-4) were selected (Table 3-1) as these









samples consisted relatively lower amounts of CaCO3. The soil phosphorus (P)

fractionation scheme used 1.0 MNH4C1, 17 h for 0.1 MNaOH, and 24 h for 0.5 MHC1

solutions for sequential P extractions (Hieltjes & Lijklema, 1980) and modified by Nair et

al. (1995). A 1:10 soil/solution ratio was used for all extractions. All extractions were

carried out at 298 K. In pilot study, ten repeated NH4C1 extractions were sufficient to

reduce Mg concentrations to undetectable (< 1.0 mg L-1). Thus soil samples were

extracted with 1.0 MNH4Cl adjusted to pH 7.0 (using 0.5 MKOH) from 1 to 10 times

before being extracted with 0.1 MNaOH, and finally with 0.5 MHC1. For example, a soil

sample was extracted once with NH4C1 before being extracted by NaOH and HC1, and

another was extracted twice, another three times, another four times and so forth, up to

ten times with NH4C1, before conducting the subsequent extractions with NaOH and HCI

(Figure 5-1).

The NaOH fraction is expected to extract Fe+Al associated P and the HC1 fraction,

Ca+Mg associated P (Chang & Jackson, 1957; Williams et al., 1967; Hieltjes &

Lijklema, 1980; Hedley et al., 1982; Psenner & Pucsko, 1988). Soil samples were shaken

end to end in 30 mL polypropylene centrifuge tubes for 2 h for 1 MNH4C1 extractions,

17 h for 0.1 MNaOH and 24 h for 0.5 MHC1 extractions. The suspensions were

centrifuged for 15 min @ 3620 x g and then filtered (0.45 [m) filters to obtain solutions

for analyses. Following the sequential extractions, soil samples were ashed and digested

with 6 MHC1 to obtain the residual P fraction (Andersen, 1976). Residual P fraction

represents the stable P forms, and is often considered to be organically bound (Hedley et

al., 1982). Entrained solution after each extraction was determined by weight and









corrections were applied during the final calculations of P, and metal concentrations in

each fraction.

Analyses of P and Metals

The 1.0 MNH4C1, 0.1 MNaOH, and 0.5 MHC1 extracts were analyzed for both

inorganic-P (Pi) and total dissolved P (TDP). The TDP was determined by potassium

persulfate (K2S208) digestion method (EPA method 365.1, 1993). The potassium

persulfate used in this study was a certified ACS standard (Fisher Catalog No. P282-500).

The TDP of the 1.0 MNH4C1, 0.1 MNaOH and 0.5 MHC1 extracted solutions was

determined by adding 1 mL of 5 MH2SO4 + 0.3g of K2S208 to 5 mL aliquots of each

filtered extract in digestion tubes. Samples were digested by placing digestion tubes on a

digestion block at 125-1500C for 2-3 h so that a 0.5 mL of solution remained. The

digestion tubes were then covered with glass digestion caps and temperature raised to

380C for 3-4 h. Digested samples were cooled, diluted with 10 mL of DI water and

vortexed to ensure through mixing. The solutions were then stored in 20 mL scintillation

plastic vials at room temperature for analyses. A series of P standards were also digested

to obtain similar matrix effects for colorimetric determinations. Proper QA/QC protocol

was adopted and included a blank, a duplicate, a certified QC after every 15 samples

during each digestion run.

Inorganic-P was determined from solutions obtained by filtering (0.45[tM)

undigested centrifuged supernatant. Phosphorus (both TDP and Pi) was analyzed using

ascorbic acid colorimetery (Murphy & Riley, 1962) (U.S. EPA, 1993; method 365.1).

The difference between TDP and Pi yielded the organic-P pool (Po). Concentrations of

Ca, Mg, Fe, and Al were determined in both digested and undigested samples by atomic

absorption spectrophotometry.









The residual fraction obtained after the 0.5 MHC1 extraction was transferred to

labeled glass beakers and was digested with 6 MHC1 after heating in a muffle furnace for

2 h at 623 K and then raising the temperature to 823 K for 3 h (Anderson, 1974). Both P

and metals including Ca, Mg, Fe and Al were analyzed in the residual fraction of digested

samples.

To assure data quality for P and metal analyses proper QA/QC procedures were

adopted and included 1 blank, 1 replicate, 1 spike, and 1 certified P standard after every

15 samples run. For atomic absorption analyses 1 blank, 1 replicate, 1 spike and 1

standard were analyzed after every 15 samples. The standard curve for each metal was

calibrated by resloping after 15 samples.

Results and Discussion

Release of P, Ca, and Mg in Repeated 1.0 M NH4CI Extractions

The use of 1.0 MNH4Cl extracted more P, Ca and Mg compared to repeated water

extractions. The release of P from active dairy soils declined more sharply with repeated

NH4C1 extractions than the abandoned dairy manure-amended soils, and then leveled off

(Figure 5-2). For active dairy soils the first three extractions released an average of 65%

of total summed P obtained in 10 repeated NH4C1 extractions. In abandoned dairy soils,

an average of 40% of total summed P was released. The data suggested that P in

abandoned dairy soils became stabilized with time and had leached from soils following

abandonment. The soils released comparable amounts of P using repeated water

extractions and P concentrations did not show a declining trend (Figure 4-1). For repeated

NH4C1 extractions active dairy soils behaved differently than the abandoned dairy soils.

The NH4C1 extractable P has also been recognized as labile or loosely bound-P (Chang &

Juo, 1963; Hieltjes & Lijklema, 1980; Graetz & Nair, 1995; Nair et al., 1995; Nair et al.,









2005). During repeated NH4C1 extractions, both Ca and Mg concentrations declined for

all soils. ACS-2 soil (Figure 5-3) released significantly greater amounts of Ca compared

to the other three soils. Release of greater Ca concentrations in ACS-2 soil sample after

10 repeated NH4C1 extractions was attributed to the presence of high amounts of CaCO3

in the soil (confirmed by X-ray diffraction Figure 3-2), which was dissolved by 1.0 M

NH4C1 (Hieltjes & Lijklema, 1980). A soil with high amounts of CaCO3 is expected to

release Ca for long periods. The Mg concentrations in the first two repeated extractions

declined sharply and then leveled off after five extractions for all soils (Figure 5-4).

The use of NH4C1 exhaustively extracted both Mg and P in the initial repeated

extractions of active dairy soils, which suggests a probable sparingly soluble Mg-P phase.

Omitting data for the 1st extraction, the remaining data were subjected to regression

analyses both for the active and abandoned dairy soils (Table 5-1).

Data suggest that the release of P was a linear function of Ca+Mg release for the

active dairy soils, whereas the release of P is a logarithmic function of Ca+Mg release for

the abandoned dairy soils. The correlation coefficients for soils ACS-2 and ACS-4 were

0.946 and 0.996, respectively, showing there was little unexplained variance in the

regression. The release of both Ca and Mg with each other were also significantly

correlated (r = 0.855 for ACS-2 and r = 0.989 for ACS-4), consistent with a specific pool

of Ca-Mg-P phase(s) in the soils, responsible for a continuous P release. The suspected

mineral phase could be amorphous whitlockite (a partially Mg substituted calcium

orthophosphate). The presence of high amount of DOC can impede the crystal growth of

whitlockite, acting as a buffer for maintaining this sparingly soluble phase in amorphous

form (Inskeep & Silvertooth, 1988).









Phosphorus Concentrations in Sequential Extractions

Forms of P extracted by 1.0 MNH4C1 (Pi and Po), 0.1 MNaOH (Pi and Po), 0.5 M

HC1 (Pi and Po), and in the residual P pool for the two active (Figures 5-5 and 5-6) and

two abandoned dairies (Figures 5-7 and 5-8) are depicted. Both NH4Cl-Pi and NH4Cl-Po

forms increased for all soils with repeated NH4Cl extractions. The NH4Cl-Pi was

significantly greater than the NH4Cl-Po in all soils, and Pi fraction dominated all

extractions.

Nair et al. (1995); Gale et al. (2000); Sharpley et al. (2004); and Silveira et al.

(2006) reported similar dominance of inorganic-P over organic-P in soils following

longterm amendments. The cumulative increase of P with repeated NH4C1 extractions can

be attributed to the loosely bound P forms in dairy manure-amended soils, and suggests

that one or two NH4C1 extractions would not be enough to study P behavior of these

soils. The percentage release of P from ACS-2 soil with repeated NH4C1 extractions was

significantly smaller than from the ACS-4 soil. For example the percentage NH4Cl-Pi

released for ACS-2 soil (Figure 5-5) during 1st, 5th and 10th extractions was 9%, 20%, and

25%, respectively, whereas for ACS-4 soil (Figure 5-6) the release was 25%, 64%, and

75%, respectively.

The relative low P release in NH4C1 repeated extractions for ACS-2 soil can be

attributed to the resorption of P by the presence of relatively high amounts of CaCO3

applied as a fill material to stabilize land for proper dairy operations. Accordingly, the

NaOH-Pi fraction for ACS-2 soil (Figure 5-5) exceeded the NaOH-Pi fraction of ACS-4

soil (Figure 5-6). In the abandoned dairy soils the percentage NH4Cl-Pi released for ABS-

3 soil (Figure 5-7) during the 1st, 5th, and 10th extractions was 6%, 27%, and 39%,









respectively and 8%, 25%, and 44% for ABS-4 soil (Figure 5-8). Both abandoned dairy

soils released P more gradually than the active dairy soils.

Active dairy manure-amended soils (ACS-2 & ACS-4) had less P in NaOH fraction

than the abandoned dairy manure-amended soils (ABS-3 & ABS-4). Overall, the NaOH-

Po fraction of abandoned dairy soils was greater than the NaOH-Po fraction of active

dairy soils, which suggested immobilization of P with time or mineralization of organic P

to inorganic forms that were extracted by NH4C1. Nair et al. (1995) reported similar

trends. The HC1-P pool appeared to be depleted in both types of dairies (Figure 5-5 to 5-

8) following repeated NH4C1 extractions. Abandoned dairy soils also had significantly

greater HCl-Po pools than active dairy soils. This fraction deserves further study with

similar manure application histories. He et al. (2006) demonstrated that HC1 fraction

extracted from animal manure contains organic-P fractions in addition to inorganic-P.

Our data suggest that with abandonment, P was stabilized not only into NaOH-Po forms

but also as HCl-Po forms. The residual fraction extractable P was significantly correlated

with Ca, Al and Fe for active dairy soils. For abandoned dairy soils the extractable P in

residual fraction was significantly associated with Ca and Al but not with Fe (Table 5-4).

The residual fraction generally dominated by recalcitrant organic-P material (Hedley et

al., 1982), as observed in this study, and Ca, Fe and Al can be occluded with Po.

The associated release of P and Ca, Mg, Fe and Al to NaOH and HC1 fractions

(Table 5-2 & 5-3) were also determined. NaOH reportedly extracts Al-Fe P (Chang &

Jackson, 1957; Williams et al., 1967; Hieltjes & Lijklema, 1980; Psenner et al., 1985;

Nair et al., 1995; Toor et al., 2006); however there have been no confirmatory analyses of

this form in dairy manure-amended soils. Dairy manure contains large amounts of Ca and









Mg, and continuous manure application for years can alter the chemical forms of P

initially associated with Fe and Al (Sharpley et al., 2004). In this study, NaOH

extractable P and Ca were significantly correlated in both active and abandoned dairy

manure-amended soils (Table 5-2) Correlation coefficients of P with metals for various P

fractions extracted from active and abandoned dairy soils are given in Appendix B. For

these heavily manure-amended soils, the strong relationship between the P and Ca of

NaOH extractions might be an artifact of the 'left over' Ca after incomplete NH4C1

extractions. The HC1-P pool is mainly associated with Ca and Mg, if determined after one

NH4C1 extraction followed by NaOH extraction, thus subject to dissolution with repeated

NH4C1 extractions. However, the dissolution of sparingly soluble Ca-P, Mg-P and/or Ca-

Mg-P solids with repeated NH4C1 extractions, the release of P in HC1 extractions was

significantly correlated only with Ca (Table 5-3). While working on dairy manure-

amended soils Nair et al. (1995 and 2003), authors suggested that Ca-P and Mg-P

associations are not stable P forms and can be released upon the onset of next rainfall

event. Sharpley et al. (1996) used Fe-P strips to extract P continuously from manure-

amended soils and found out that P released more rapidly from manure-amended soils

than from the soils with minimal manure impact.

Summary and Conclusions

The repeated extraction of manure-amended soils with NH4C1 suggested that

readily-soluble P is associated with Mg. After depleting the Mg, further P release was

highly correlated with Ca. The release of P in dairy manure-amended soils, using NH4C1

repeated extractions, is controlled by Ca and Mg-P phases, which are sparingly soluble.

Active dairy soils release more P in the first two NH4C1 extractions than abandoned dairy

soils. Most of P released in NaOH and HC1 extraction was associated with Ca









concentrations. Data also suggested that repeated NH4C1 extractions mineralized the HC1-

P pools that may not be available for release with fewer numbers (8) of repeated water

extractions. Phosphorus in residual fraction was almost constant during the sequential

extractions, and the P appeared to be associated with Ca, Fe and Al. We conclude that the

initial high release of P from dairy manure-amended soils may be due to the presence of

Ca-P and Mg-P phase/s, which will take years to deplete. However, with time the release

of P from these soils will likely be controlled by a relatively less soluble Ca-P phase

provided there is no further manure application.









Table 5-1. Total dissolved phosphorus (TDP) as a function of Ca+Mg in repeated 1.0 M
NH4C1 extractions (Data from 1st extraction omitted).
Soil Type Soil ID N Regression Equation r2
SaACS-2 9 TDP = 0.107(Ca+Mg) 0.912 0.895t
Active
ACS-4 9 TDP = 0.246(Ca+Mg) 0.226 0.993tt
bABS-3 9 TDP = 0.042(Ca+Mg) + 2.3242 0.5361
Aba d ABS-4 9 TDP = 0.047(Ca+Mg) + 0.831 0.604O
ABS-3 9 TDP = 0.624Ln(Ca+Mg) + 1.488 0.703"
ABS-4 9 TDP = 0.421Ln(Ca+Mg) + 0.523 0.793tt
aACS, active dairy manure-amended soil
bABS, abandoned dairy manure-amended soil.
Regression coefficients were significant atp < 0.01 (tt) andp < 0.05 (f) as determined
by LSD


Table 5-2. Sequential release of total dissolved phosphorus (TDP) as a function of Ca,
Mg, and Fe in 0.1 MNaOH extractions
Soil Type N Regression Equation r2
Active 20 TDP = 1.597Catt -2.589Mg + 0.112A1- 1.519Fe + 1.796 0.751W
Abandoned 20 TDP = 1.068xCatt 0.199Mg + 0.091A1 0.276Fe + 8.108 0.644A
Regression coefficients were significant atp < 0.01 (tt) andp < 0.05 (f) as determined
by LSD


Table 5-3. Sequential release of total dissolved phosphorus (TDP) as a function of Ca,
Mg, Fe, and Al in 0.5 MHC1 extractions
Soil Type N Regression Equation r2
Active 2 TDP = 0.217Ca1 + 3.36Mg 0.028A1 -3.120Fe 0.776
0 0.025(CaxMg) + 1.322
Abandoned 2 TDP = 0.752CaT + 0.752Mg 0.116A1+ 0.954t
0 1.831Fe 0.016(CaxMg)- 1.930
Regression coefficients were significant atp < 0.01 (tt) andp < 0.05 (f) as determined
by LSD

Table 5-4. Sequential release of total dissolved phosphorus (TDP) as a function of Ca,
Mg, Fe, and Al in residual fractions
Soil Type N Regression Equation r2
Active 2 TDP = 0.181Ca 0.142Mg -0.021Al~ -
0 0.493Fet 0.08(CaxMg) + 0.920
Abandoned 2 TDP = 0.604Catt + 0.242Mg 0.008A1 + 0.98
0 0.572Fe 0.090(CaxMg) -1.930
Regression coefficients were significant atp < 0.01 (tt) andp < 0.05 (f) as determined
by LSD

























Labile -P ]

Total dissolved P (TDP)
Inorganic-P (Pi)
Organi-P (Po) TDP-
Pi


[ HC-P
17-hshaking 2"mL M
20 mL 0.5 MHCI
I 24-h shaking




| FefAl-P I Ca/Mg -P

TDP TDP
Po= I Po=
(TDP-Pi) (TDP-Pi)


Figure 5-1. Schematic of repeated and sequential extraction procedure adapted from Nair
et al. (1995).


-ACS-2
--ACS-4
- ABS-3
-- ABS-4


100 0- --- -.

50


.L. '-


1 2 3 4 5 6 7 8 9 10
Extraction number


Figure 5-2. Release of phosphorus using 1.0 MNH4Cl repeated extractions in active and
abandoned dairy manure-amended soils. ACS = active dairy manure-amended
soil; ABS = abandoned dairy manure-amended soil.


Mainly organic -P











3000


2500


2000


1500 *


1000 .
S.


-==ACS-2

- ACS-4

- ABS-3

--- ABS-4


* M


1 2 3 4 5 6 7 8 9 10
Extraction number


Figure 5-3. Release of calcium using 1.0 MNH4C1 repeated extractions in active and
abandoned dairy manure-amended soils. ACS = active dairy manure-amended
soil; ABS = abandoned dairy manure-amended soil.



1000- ACS-2
"--- ACS-4
ACS-4
800 -" -ABS-3
-A- ABS-4


1 2 3 4 5 6 7 8 9 10
Extraction number


Figure 5-4. Release of magnesium using 1.0 MNH4Cl repeated extractions in active and
abandoned dairy manure-amended soils. ACS = active dairy manure-amended
soil; ABS = abandoned dairy manure-amended soil.




















P
distribution


B Residual-P

U HCI-Po

E HCI-Pi

* NaOH-Po

m NaOH-Pi

1 NH4CI-Po

V NH4C-Pi


1 2 3 4 5 6
Extraction number


7 8 9 10


Figure 5-5. Distribution of P in 1.0 MNH4C1, 0.1 MNaOH, 0.5 MHC1, and residual-P
fractions for an active dairy soil (ACS-2).


100%


80% -

P
distribution 60%-


40% -


20%


0%-


1 2 3 4 5 6
Extraction number


7 8 9 10


Figure 5-6. Distribution of P in 1.OMNH4C1, 0.1 MNaOH, 0.5MHC1, and residual-P
fractions for an active dairy soil (ACS-4).


100%


80%


60% -


40% -


20%


0% -


0 Residual-P

* HCI-Po

* HCI-Pi

* NaOH-Po

0] NaOH-Pi

N NH4CI-Po

a NH4C-Pi



















P
distribution


E Residual-P

* HCI-Po

O HCI-Pi

B NaOH-Po

[I NaOH-Pi

! NH4CI-Po

2 NH4CI-P


1 2 3 4 5 6 7 8 9 10
Extraction number


Figure 5-7. Distribution of P in 1.0MNH4C1, 0.1 MNaOH, 0.5MHC1, and residual-P
fractions for an abandoned dairy soil (ABS-3).


100%


80%


P
distribution


E Residual-P

* HCI-Po

[ HCI-Pi

2 NaOH-Po

U NaOH-Pi

2 NH4CI-Po

E NH4CI-Pi


1 2 3 4 5 6 7 8 9 10
Extraction number


Figure 5-8. Distribution of P in 1.OMNH4C1, 0.1 MNaOH, 0.5MHC1, and residual-P
fractions for an abandoned dairy soil (ABS-4).


100%


80%


60%


40%














CHAPTER 6
SOLID STATE ASSESSMENTS: CONFIRMING THE ASSOCIATIONS OF
PHOSPHORUS, CALCIUM AND MAGNESIUM

Introduction

Repeated (water and NH4C1) and sequential (NH4C1 NaOH HC1) extractions

used in previous chapters revealed information about the nature of P forms in dairy

manure-amended soils and, hence, form stability under given conditions. Sequential or

repeated extractions have inherent limitations, which were outlined by Harris (2002): "1)

overlap in solubility of various P forms for water or other extractants, 2) time

dependency, 3) loading rate (e.g., extractant/soil ratio) dependency, 4) readsorption of P

forms by CaCO3, and consequent misallocation to P "targets" by subsequent extractions

(as observed in NaOH extractions in the 5th Chapter), 5) variations in solubility of

targeted forms from soil to soil, 6) presence of buffers in some soils that can reduce the

effectiveness of an extractant, and 7) potential hydrolysis of organic-P forms which can

lead to an overestimation of inorganic-P forms.

Therefore, associated release of P with Ca and Mg of dairy manure-amended soils

does not confirm phase associations. Direct assessments using solid state analytical tools,

such as X-ray diffraction (XRD), energy dispersive spectroscopy (EDS), scanning

electron microscopy (SEM), and X-ray absorption near edge spectroscopy (XANES) are

necessary to confirm P-solid phases. Pierzynski et al. (1990a) conducted density

separation of clays obtained from excessively fertilized (inorganic) soils to concentrate P

for subsequent XRD, SEM, and EDS analyses to quantify "P rich particles". Phosphorus









was associated with Al and Si (Pierzynski et al., 1990b). Beauchemin et al. (2003)

performed sequential extractions on soils that received animal manure for 25 years.

Various Ca-P minerals dominated P forms in the B-horizon of an acidic loamy soil (pH

5.5) as confirmed by XANES analyses. About 45% of total P was present as octacalcium

phosphate (OCP) and 11% of total P as hydroxyapatite (HAP). There has been limited

solid phase assessments of P associations with Mg and Ca for dairy manure-amended

soils (Cooperband & Good, 2002; He et al., 2003; Nair et al., 2003; Sharpley et al.,

2004).

The objective of this work was to confirm the P associations with Ca and Mg

suggested in the repeated water and NH4C1 extraction experiments. The findings were

also expected to aid in explaining the P speciation results observed in a column leaching

study (Chapter 4). Four approaches were used to study the associations of P with Ca, Mg,

Fe, and Al in dairy manures and dairy manure-amended soils at the microscopic levels.

The first approach involved the use of XRD analyses of clays obtained from dairy

manure-amended soils. The second approach specifically dealt with dairy manures, and

addressed the hypothesis that a sparingly soluble phase of Ca-Mg-P present in dairy

manures can be crystallized to a well defined P crystalline phase when subjected to high

temperature (550C). In the third approach, the clays obtained in the first approach and

the ashed dairy manures in second approach were examined under SEM for dot map

images and the spatial associations of P, Ca, and Mg were evaluated. In addition to

spatial associations, semi-quantitative analyses of metals of "P-rich particles" were also

performed. The fourth approach involved the use of electron probe microanalyses

(EPMA) of dry sieved silt+clay fractions of dairy manure-amended soils.









Material and Methods

Approach I: X-ray Diffraction of Untreated Clays

Air-dried samples of dairy manure-amended soils (50 g each) and dairy manures

(100 g each) were Na-saturated by several washings with IMNaCl and centrifuging at

728 x g after each washing to remove supernatant. Excess salt was then removed by

rinsing with DI water using the same centrifugation procedure. The final rinse was

determined as the first rinse in which the supernatant appeared turbid. The Na-saturated

samples were wet-sieved to separate the sand from silt and clay. Silt and clay were

separated by mixing with DI water and subsequent centrifugation (Whiting & Allardice,

1986). Clay was collected in decanted supernatant, until a clear supernatant was

achieved. The silt in the bottom of centrifuge bottle was dried and stored. Clay

suspensions were flocculated using 1.0 MNaCl. Clay was transferred to 0.45-[tm filters

and washed with 25 mL of DI water to remove salt. The salt-free clay was transferred

from the filter to a glass slide using a rubber policeman, allowed to dry, and then gently

crushed to a powder and stored in glass scintillation vials.

Oriented mounts for clay were prepared for XRD analyses by depositing

approximately 250 mg of clay as a suspension onto a porous ceramic tile under suction.

Tiles were dried in a glass desiccator and then scanned with CuKa radiation, with the

tube energized at 35 kV and 20 mA current. Scans were conducted at 20/ minute over a

20 range of 2 to 600. Minerals were identified following criteria outlined by Whittig &

Allardice (1986).

Approach II: Ashed and Whole Dairy Manure Analyses

Dairy manures collected from different dairies (Chapter 2) were dried and ground

to 2mm with an automated grinder. Five gram samples of manure were ashed in triplicate









in a muffle furnace at 550C for 5 h. The ashed samples were ground using a porcelain

agar mortar and then transferred to glass vials and capped. Additionally, 2mm-ground

whole dairy manures were passed through a 1-mm-mesh stainless steel sieve. The

materials (manure and ash) were prepared for XRD by gently loading the powder into an

XRD mount in manner that minimized preferred orientation (Harris et al., 1994). Mounts

were scanned using the same XRD methodology as outlined in the approach I.

Approach III: SEM Imaging and EDS Analyses

For successful SEM and EDS analyses, samples need to be conductive in nature, to

minimize electron accumulation at the surface of the sample. Such accumulation

("charging") prevents optimal resolution. Clays (4 active and 4 abandoned) and ashed

manures obtained in the first and second approach respectively were dispersed

ultrasonically and mounted on carbon stubs. The samples were then subjected to

evaporative carbon coating to enhance conductivity and minimize charging. Mounts were

then subjected to a scanning electron microscope (JOEL JSM-6400) equipped with an

Oxford model number 6506 EDS system for the X-ray analysis. The SEM images were

obtained by operating at 15 kV beam voltage and 60tA probe current. Images were taken

at a series of magnifications while adjusting contrast and brightness. Energy dispersive

spectroscopy was performed at a beam voltage of 15 kV. This analysis was performed at

three different locations on each sample to account for compositional variations.

Elemental compositional analyses were performed using the x-ray analysis and SEM

quantitative routine of the Oxford Link ISIS software. The software incorporates atomic

number, x-ray absorption, and fluorescence (ZAF factor) correction factors while

performing semi-quantitative analysis (Goldstein et al., 2003).









Approach IV: Electron Microprobe Microanalyses of Whole Silt + Clay of Dairy
Manure-amended Soils

Three active (ACS-2 -ACS-4) and two abandoned (ABS 1- ABS2) dairy soil

samples were subjected to electron microprobe (JEOL Superprobe-733, operating at 15

kV). Silt+clay fractions were obtained by dry sieving soils through 45KM sieve using a

mechanical shaker for 15 min. One centimeter diameter disks (3 disks for each sample) of

sieved material were prepared by subjecting these samples to a pressure of 34474 kPa

(5000 PSI) using a hydraulic press. Carbon coating was performed on the silt+clay disks

as explained in approach III. Microprobe analyses were done on each disk using 100 [m

beam size to quantify P, Ca, Mg, Al and Si concentrations. Standards used to calibrate the

microprobe, which were served for QA/QC, were calcite for Ca, apatite for P, dolomite

for Mg, and quartz for Si. Additionally Ca PET crystal (Pentaerythritol) and P,Si,Mg,Na

-TAP crystal (Thallium acid Phthalate) were used as monochromators.

The technique has analytical accuracy of 1-2% by weight, but can be exploited up

to 0.1 1% by carefully extracting characteristics x-ray peaks and considering minimum

peak interference (Goldstein et al., 2003). The relatively large beam size was selected to

minimize auto correlations that can arise from local surface imperfections.

Results and Discussion

Both active and abandoned dairy clays had similar mineralogy (Figure 6-1). X-ray

diffractograms of untreated clays (Figure 6-1) were dominated by quartz, with some

kaolinite and traces of hydroxyinterlayered vermiculite (HIV). No P-bearing mineral was

detected. An amorphous hump was observed from 22-26 20, which reflected the presence

of biogenic Si, derived from plant phytoliths present in dairy manure. Either the P-

bearing minerals were amorphous in nature or XRD was not sensitive enough to detect









crystalline phases due to small amounts. Pierzynski et al. (1990a) and Harris et al. (1994)

used similar approaches and were unable to identify P bearing minerals in highly

fertilized soils and in dairy manure-amended soils, respectively. Dairy manure samples

(<1.0 mm) contained whewellite (CaC204.H20; calcium oxalate), calcite (CaC03), and

quartz as shown by XRD analyses (Figure 6-2). The presence of calcite means that there

is a significant amount of Ca not in the form of a soluble salt or associated with P in a

discrete phase. Quartz is common in dairy manure of grazing cows, probably due to

ingestion of small amounts of surface soil material. In ashed dairy manure samples, both

calcite and quartz were present, but no whewellite (Figure 6-3) because the latter

decomposed to CO2 and H20 at 823 K. The Ca-Mg-P mineral whitlockite (Figure 6-3)

was tentatively identified in ashed samples from XRD data. Formation could be

attributed to the re-crystallization of semi- crystalline/amorphous Ca-P or Mg-P or a Ca-

Mg-P phase present in dairy manures. All dairy manures, collected from three locations,

showed similar possible P phase crystallization at 550C. EDS analyses of the ashed

samples showed high intensity peaks of Mg, Ca and P (Figure 6-4), consistent with a Mg-

Ca-P phase, which might be amorphous in nature prior to heating, sparingly soluble, and

undetectable by XRD analyses. Spatial associations of both Mg-P and Ca-P in manure

(Figures 6-5 & 6-6) were also observed in SEM/EDS dot map images and spectra,

confirming that Mg-P is a manure component rather than a component formed in the soil

environment after the application of dairy manures.

This finding could be helpful in formulating new diets to reduce manure-P

solubility by adjusting Ca:Mg ratios in addition to P for dairy animals (Daniel et al.,

2006). Many authors advocate a reduction of P in dairy diets to reduce the P solubility









(Wu et al., 2001; Dou et al., 2003; Toor et al., 2005). However, P interacts within the

animal gut with Mg and Ca (Herrera et al., 2006) and by decreasing the Mg to the

optimal levels in dairy diets, Mg-P formation can be minimized and Ca-P interaction can

be enhanced (Herrera et al., 2006). Minimized Mg-P formation could be beneficial as it

would reduce P solubility and high P leaching potential (Lindsay, 1979) as all Mg-P

minerals are relatively soluble than Ca-P minerals. The P-rich particles of dairy manure

clays, observed under SEM (Figure 6-7), also showed the dominance of Mg, Ca and P as

observed in EDS spectrum (Figure 6-8). There were 24 samples of manure were observed

under SEM, and 46% of the samples showed the spatial associations of P with Ca and Mg

both.

Microprobe analyses of silt+clay samples showed that active dairy silt+clay

samples had greater average concentrations of P (0.38%) and Mg (0.64%) than the

abandoned dairy samples (P = 0.31% and Mg = 0.25 %) (Figure 6-9). This parallel

decline since abandonment suggests that both Mg and P were depleted from the soil with

time, and is consistent with the elements being released via the dissolution of a common

phase. Release of Mg and P was closely associated in repeated water extractions (Figures

4-6 & 4-7), and did not differ between active and abandoned dairies as did Ca, which was

released to a lesser extent in abandoned dairies. Active and abandoned dairies had similar

Ca concentrations (2.30.5%), which were about an order of magnitude greater than Mg

and P concentrations. Both Ca and Mg were significantly (p<0.01) correlated spatially

with P in abandoned dairy silt +clay fractions (Figures 6-9c and 6-9d). For active dairy

samples, only the Mg and P relationship was statistically significant (p<0.01). A possible









explanation for poor correlation of Ca and P in active dairy manure-amended silt+clay

fractions can be due to the presence of CaC03 applied as a fill material.

Summary and Conclusions

Solid state techniques confirmed spatial associations ofP with both Ca and Mg in

dairy manure and manure-amended soils. Results were reasonably consistent with release

trends for the elements observed in repeated water (Chapter 4) and NH4C1 extractions

(Chapter 5), as well as with what could be concluded about mineral equilibria from

chemical speciation of column leachates (Chapter 4). Implications are that Mg is

associated with P in manure and manure-amended soils to at least an equivalent extent as

Ca. However, specific minerals were not identified and uncertainty remains as to

crystallinity of P phases as well as to whether P is associated with Ca and Mg separately

or within a common Ca-Mg-P phase (e.g., whitlockite) that was tentatively identified in

ashed manure samples.








62







dd== d=332B8
25001 4,4C22







2000 ,'









4 (f31;103





1 2... .3 t
1 15 20 25 3 5
d0d=440 26 (1=35















Two Theta (deg)


Figure 6-1. X-ray diffraction patterns of four active (ACS-1 to ACS-4) and four
abandoned (ABS-1 to ABS-4) dairy manure-amended untreated clays. HIV

hydroxyinterlayered vermiculite.





d=3 6598 -d=30196



















Two-Theta (deg)


Figure 6-2. X-ray diffraction pattern of < 1.0 mm dried dairy manures showing the
presence of whewellite, quartz and calcite.
presence of whewellite, quartz and calcite.


































30
Two-Theta (deg)


Figure 6-3. X-ray diffraction patterns of oven dried and ashed dairy manures (3) showing
the presence of Mg-Ca Whitlockite (Mg-Ca phosphate).



Counts


Energy (keV)


Figure 6-4. Energy dispersive spectrum of an ashed dairy manure showing the high
intensity peaks of Ca, Mg and P.


























Figure 6-5. Energy dispersive dot maps of a dairy manure showing an association of Mg
and P.


Figure 6-6. Energy dispersive dot maps (scale 20 [tm) of a dairy manure-amended soil
clay showing an association of P and Ca.










0 IMaKal.. 19 I


Figure 6-7. Dot image of dairy manure showing the spatial associations of Mg, P, and Ca.


6
-Ern I, 4kov)


Figure 6-8. EDS spectrum of a manure P rich particle obtained at 400X magnification
showing the dominance of Mg, P, and Ca.












0.6 -


v-1.05x- 0.29
r2 0.(X7



*


0.0 0.2 0.4 0.6

Figure 6-9a Mg (On)


0.8 1.0


= 3.21x-0.50
r2 = 0.78*



y 4


0.0 0.1 0.2 0.3 0.4


Figure 6.9c


v= 0.09x + 0.18
rz = 0.23



.t


Mg (o)


v0 23x- 0.22
Cr 0.89**
^


0.0 1.0 2.0 3.0 4.0 5.0 0.0

Figure 6-9b Ca (%) Figure 6-9d


1.0 2.0
Ca (0%)


Figure 6-9. Relationships between P, Mg and Ca for active (6-9a & 6-9b) and abandoned
dairy (6-9c & 6-9d) manure-amended dry sieved (45gm) silt+clay using the
electron probe microanalyses (EPMA). **significant at p<0.05.


3.0 4.0


III














CHAPTER 7
CALCIUM PHOSPHATE CRYSTALINITY AND DAIRY MANURE COMPONENTS

Introduction

Phosphate chemistry in dairy manure-amended sandy soils may be influenced by

the presence of Mg, biogenic silica (derived from plant phytoliths), and manure-derived

organic matter and its myriad of organic acids (Inskeep & Silvertooth, 1988). Manure-

amended soils systems are typically in a state of disequilibrium (Silveira et al., 2006), and

even the high pH accompanied by high Ca concentrations do not effectively stabilize Ca-

P forms (Harris et al., 1994; Josan et al., 2005). Apatite and metastable crystalline Ca

phosphates (e.g., octacalcium phosphate OCP, tricalcium phosphate) are not detectible by

XRD (at least, not in the bulk clay) even after years of manure addition despite high total

P concentrations (Harris et al., 1994). Density gradient centrifugation has been used

successfully on soil clay fractions (Jaynes & Bigham, 1986; Pierzynski et al., 1990a) to

study the different Ca-P minerals formed in the soil environment. Pierzynski et al.

(1990a) partitioned clays of heavily fertilized soils into three density fractions (<2.2, 2.2

to 2.5, and > 2.5 Mg m-3) using non aqueous heavy density liquid containing

polyvinylpyrolidone and tetrabromoethane.

This study addressed potential inhibitory effects of manure-derived components on

Ca-P crystallization and the fate of P in manure-amended soils, particularly soils with

minimal native P retention capacity. Results relate to the question: Why are abandoned

dairy manure-amended soils continue to leach high amounts of P after years of









abandonment under the conditions of high pH and high Ca concentrations? It was

hypothesized that:

1. Activities of Mg, Si, and DOC in soil solutions of manure-amended soils are
sufficient, jointly or separately, to inhibit crystallization of stable Ca-P forms, thus
allowing high P release from these soils.

2. Noncrystalline Si forms (including biogenic Si in dairy manure) can retain P at
circumneutral pH and high Ca activity. This hypothesis is based on the idea that Ca
can serve as a bridge between the silicate surface and P.

The hypotheses were tested by the following objectives:

1. To study the effects of Mg, Si and DOC on Ca-P crystallization using average
inorganic species concentrations found in manure-amended soil leachates.

2. To investigate the effects of "low-density clay" (<2 Mg m-3) and "high-density
clay" (> 2 Mg m-3) of manure-amended soils on Ca-P crystallization.

Material and Methods

The study required isolating soil clays, treatment to remove carbonates and organic

matter, and separation into two density fractions. Procedures for accomplishing the

objectives are given in the following sections.

Particle Size Separations of Dairy Manure-amended Soils

The method used to separate particle sizes in soils is a slight modification of the

method developed by Whittig & Allardice (1986). Air-dried samples (50 g each) were

Na-saturated by several washings with 1.0 MNaCl in 250-mL centrifuge bottles. Samples

were shaken for 5 min on a reciprocal shaker for each washing. The supernatant was

decanted after centrifugation at 738 x g (2000 rpm). Excess Na was rinsed with deionized

(DI) water by centrifugation and decantation. The final rinse was determined as the first

rinse that appeared turbid. The Na-saturated sample was wet-sieved (45 tm) to separate

the sand from silt and clay. Silt and clay were separated by centrifugation (Whitting &

Allardice, 1986), using DI water and decanting supernatant as clay, repeated until a clear









supernatant was achieved. Silt collected in the bottom of centrifuge bottle was dried and

stored. Clay suspensions were flocculated using 1.0 MNaCl. Clay was collected from

suspensions by 0.45-[tm filtration and washed with at least 25 mL of double DI water to

remove excess salt. Salt-free clay was transferred from the filter to a glass slide using a

rubber policeman, allowed to dry, gently crushed to a powder, and stored.

Carbonate and Organic Matter Removal from the Resulting Clays

To remove carbonates, 2 g of clay was treated with 100-mL of 1 M sodium acetate

solution, buffered at pH 5.0 with glacial acetic acid (Anderson, 1963; Jackson, 1985), and

heated in a boiling water bath for 30 min with intermittent stirring. The same procedure

was repeated twice more after decanting the clear supernatant. The treated clay was

transferred to a 0.45-[tm filter and washed with DI water. The retentate was then

subjected to organic matter removal. A 5-mL aliquot of sodium acetate buffer and 10 mL

(30%) of hydrogen peroxide (H202) were added to the samples in beakers and allowed to

stand overnight. The following day the samples were placed on a hotplate at 850C and

gently heated for 10 min. The last step was repeated and more H202 was added to obtain

a light colored solid material. The samples were transferred to 0.45-[tm filters, washed

with DI water using suction, dried, and stored in glass scintillation vials for further

analyses.

Density Separations for Soil Clay Materials

A Na polytungstate solution with a density of 2.0 Mg m-3 was prepared and

adjusted to pH 7.0. Half-gram samples of treated carbonatess and OM removed) clay

were placed in 50-mL centrifuge tubes containing 35-mL of the Na polytungstate

solution, shaken on a reciprocal shaker for 30 min, and centrifuged at 3268 x g (4000









rpm) for 2 minutes. The supernatant was decanted, and the dense material was transferred

to a labeled container. The obtained light and dense fractions were again transferred into

the centrifuge tubes containing 35-mL of 2.0 Mg m-3 Na polytungstate solution, sonicated

using an ultrasonic water bath for 2 min and then shaken on a reciprocal shaker for half

an hour. The material was centrifuged again at 4000 rpm for 2 minutes, and the above

process was repeated until no further separation was evident. The accumulated dense-

and light-fraction suspensions were filtered through a 0.45 |tm filter, rinsed 5 times with

10-mL aliquots of DI water, transferred to a glass slide with a rubber policeman, allowed

to air-dry, and transferred to labeled scintillation vials. Mineralogical analyses were

performed on both light and dense fractions prior to use for incubation experiments.

Preparation of Incubating Solutions

Chemically defined solutions mimicking the average inorganic chemical

composition (Table 7-1) in the leachates of manure-amended soils (Chapter 4) were

prepared in double deionized (DDI) water. Four stock solutions were prepared and

adjusted to a final pH of 6.8 by equilibrating with atmospheric CO2 and by adding 0.1 M

HC1 or 0.1 MNaOH solutions. At this pH the precipitate of Ca-P minerals did not take

place immediately; however, it was expected that Ca-P crystallization would take place in

the control, as the ion activity product was supersaturated with respect to hydroxyapatite.

Solution 1 served as a control (Table 7-2) and contained all the same components except

Mg, Si, and DOC. Solution 2 consisted of the control solution plus Mg, but no Si and

DOC. Solution 3 consisted of the control solution plus Si, but no Mg and DOC. Solutions

1, 2 and 3 were prepared from analytical grade reagents and were filtered through 0.45-

Lm filters prior to use. Solution 4 consisted of the control solution plus DOC but no









added Si or Mg. The DOC solution, which served as a base solution for this treatment,

was prepared as follows: 100 g of each of the soil samples were mixed together and

treated to remove salts and carbonates using 200-mL of the 1.0 M sodium acetate

following the methods outlined in the carbonate removal section of this chapter. The

sodium acetate treated material was washed 5 times with 1.0 MNaCl solution using a

centrifuge and four 250-mL bottles at 738 x g (2000 rpm). The residual NaCl salt was

rinsed with DI water via centrifugation until the solutions appeared turbid. The turbid

solution was retained in the 250-mL bottles. All the 250-mL bottles were shaken for 16 h

on a reciprocal shaker. The suspensions were allowed to settle for 2 h, and then filtered

through a 0.45-[tm filter using filtration jars to collect the DOC solution. The DOC

solution was analyzed for Ca, Mg, K, Na, and Si using inductively coupled plasma-

atomic emission spectroscopy (ICP-AES) (EPA method 200.7). Ammonium was

analyzed by the semi-automated colorimetry method (U.S. EPA, 1993; method 350.1);

nitrates by automated colorimetry with the use of an ALPKEM Auto-analyzer (U.S. EPA,

1993; method 353.2), and sulfateby ion chromatography with a separator AS-14

(DIONEX) (U.S. EPA, 1993; method 300.0). Dissolved organic carbon was determined

by TOC-5050A, Shimadzu (method 5310A, 1992). Interferences from the inorganic

carbon were first removed by sparging with CO2 -free gas after acidification of the

sample (Sharp & Peltzer, 1993). The DOC solution was later diluted to DOC

concentrations representative of column leachate DOC concentrations.

Incubation Setup, Monitoring and Solution Analyses

The effects of Mg, Si, and DOC on Ca-P crystallization in the presence and

absence of solids (clay fractions obtained from density separations), were determined









with four replications in the absence of solids. Three replications included the light-clay

fraction (p < 2.0 Mg m-3) and four replications included the dense fraction (p > 2.0 Mg m

3). Throughout the discussion of the incubation study, the word 'solids' will refer to the

clay-sized material derived from the manure-amended soils. 100 mg of solids and 50-mL

of incubating solution (Table 7-2) were placed in 150-mL nalgene bottles and loosely

capped. The containers were stored in the dark and maintained at room temperature

(25C) for 20 weeks. Solutions were monitored on a weekly basis for water levels, and

small amounts of DDI water were added when necessary to compensate for evaporation.

One mL of supernatant was withdrawn at 1, 3, 5, 10 and 20th week intervals, diluted to 8-

mL with DDI water, and filtered through a 0.45-[tm filter. The resulting solutions were

analyzed for Ca and Mg using atomic absorption spectroscopy. Inorganic P

concentrations were measured using ascorbic acid colorimetry (U.S. EPA, 1993; method

365.1). Details of incubation treatments are explained in Table 7-2.

Solid State Assessments

X-ray diffraction analyses

Low- and high-density clay fractions were subjected to x-ray analyses using a side-

packed powder mount and scanned with CuKa radiation energized at 35 kV and 20 mA

current. Scans were conducted at 20/ minute over a 20 range of 2 to 600. The precipitates

obtained after 20 weeks of incubation were collected on filter paper, spread on a low-

backgound quartz crystal XRD mount, air dried, and scanned for x-ray analyses (Chapter

6).

Energy dispersive spectroscopy analyses

Energy dispersive spectroscopy (EDS) was performed at a beam voltage of 15 kV.

The analysis was performed at three locations on each sample so that compositional









variations could be accounted for accurately. Elemental compositional analyses were

performed using the x-ray analysis and SEM quantitative routine of the Link ISIS

software. The software incorporates atomic number, x-ray absorption, and fluorescence

(ZAF factor) correction factors and performs semi-quantitative analysis (Goldstein et al.,

2003).

Statistical Analysis

To test the differences in solution concentrations of P, Ca, Mg, and DOC after the

specified incubation periods, a non-parametric test (Kruskal-Wallis) was used (p<0.05).

Computations were performed in Minitab version 14.0 (Minitab, 2004).

Results and Discussion

Effects of Mg, Si and DOC on Ca-P Crystallization

In the absence of solids, median concentrations of P in the equilibrating solutions

declined significantly, from 68 mg L-1 to 28 mg L-1, for both the control and the Mg

treatment over the 20 weeks of incubation (Figure 7-1). P concentrations were

significantly lower (p<0.05) for the Mg treatment than for the control from the 1st week to

the 10th week of incubation; however after the 20th week, there was no significant

difference between the P concentrations in the Mg treatment and the control. The sharp

decline in the P concentrations in the 1st through 10th week coincided with the formation

of precipitates in both the control and the Mg treatment solutions. The Mg concentrations

remained almost constant (168 + 5 mg L-1) during this incubation study.

Mineralogical analyses of the precipitates revealed that the only crystalline Ca-P

phase precipitate that formed were brushite (CaHPO4.2H20) in the presence of Mg

(Figure 7-2), and hydroxyapatite (Ca5(P04)30H) in the control (Figure 7-3) after the 10th

and 20th week of incubation. Nielsen (1984) observed that cation dehydration plays an









important role in the growth rate of crystals. Magnesium has a dehydration rate that is

3000 times slower than that of a Ca ion, and thus the incorporation of Mg into

hydroxyapatite can be negligible (Martin & Brown, 1997). Neither a Ca-deficient nor a

Mg incorporated non-crystalline Ca-P phase was observed; however, in the presence of

Mg, a simple Ca-P crystalline structure (brushite) with Ca:P 1:1 was precipitated and

detected by XRD, then a more complex structure with Ca:P 5:3 (HAP). Corresponding

declines in Ca concentrations were also observed for the Mg and control treatments

(Figure 7-4).

The data suggest that Mg promoted the precipitation of brushite and as a result the

solution P concentrations were significantly lower (p<0.05) in Mg treatments for the first

5 weeks of incubation (Figure 7-1). In the control, HAP crystallization was delayed,

resulting in higher P concentrations until the 5th week of incubation. The concentrations

of Mg used in this experiment were very high compared to those used by other

researches. This is because most studies on the formation of Ca-P minerals pertain to the

fields of dentistry and analytical chemistry, and such studies typically use lower

concentrations of Mg. Using solutions with low Mg concentrations, which are

continuously stirred or agitated, results in the formation of amorphous Ca-P phases

(Ferguson & McCarty, 1971). In this incubation study, the solutions were made to mimic

actual leachate concentrations of dairy-manure amended soils (which have comparatively

higher Mg concentrations), and the solutions were not stirred; this enabled the formation

of a crystalline, yet relatively soluble, Ca-P mineral phase.

The Si did not inhibit the formation of HAP. The P and Ca concentrations declined

but were similar in both the control and Si treatment solution (Figure 7-5). Silicon









concentrations did not decline significantly (p >0.05) during the incubation study. The

role of elemental Si in the formation of HAP has been studied in dentistry by Damn &

Cate (1992). Silicon behaved as a heterogeneous nucleation substrate that stabilized the

formation of the calcium phosphate nuclei, and reduced the development time of stable

nuclei (critical nuclei). Hidaka et al. (1993) also studied the role of Si silicicc acid) in the

formation of calcium phosphate precipitates and observed no inhibition of Ca-P

formation. However in soil systems, Shariatmadari & Mermut (1999) studied the

phosphate sorption-desorption behavior of silicate clay-calcite systems at various Mg and

Si concentrations (0-15 mg L-1). The researches concluded that addition of Si decreased

the P sorption of calcite, and attributed the effect to Ca-Si ion-pair formation.

In the DOC treatment, P concentrations in the equilibrating solutions after 20

weeks of incubation were significantly greater (p<0.05) than in the control (Figure. 7-6).

During incubation, P concentrations declined from 68 mg L-1 to 58 mg L-1 attributable to

the immobilization ofP by the DOC. Later, this was confirmed by measuring total

dissolved P in stored frozen solution aliquots that had been taken during the 20 weeks of

incubation. Total dissolved P (TDP) values were greater than SRP values, and TDP

concentrations declined during the incubation period, which can be related to the

flocculants observed at the bottom of DOC treatment bottles.

Additionally in the DOC treatment, there also was a significant decline in Ca

concentrations after 20 weeks of incubation, which was most likely due to Ca-DOC

complexation. This was corroborated by analyzing the same frozen aliquots taken from

the incubating solutions for DOC. The decline in DOC concentrations (Figure 7-7) was

also correlated with observations of increasing flocculation of DOC in the solutions









during the incubation period. Assessments of the solids in the DOC treatments, possibly

from the DOC flocculation, did not reveal the formation of Ca-P mineral phases (Figure

7-3). Thus, the presence of DOC completely inhibited the formation of Ca-P mineral

phases. Grossl & Inskeep (1991, 1992) documented the inhibition of both OCP and

tricalcium phosphates precipitation in presence of organic acids like humic, fulvic, citric,

and tannic acids. Inhibition was attributed to the blockage of adsorption sites of CaCO3

by DOC and also stated that presence of DOC favors the formation of brushite, as

opposed to more thermodynamically stable Ca-P phosphates.

MINTEQ analyses of solutions species after 20 weeks of incubation revealed high

ionic activity products. The results suggest that even though the Ca and P concentrations

declined significantly in the incubating solutions over the 20 weeks, the solutions

remained supersaturated. Therefore, the inhibition of the Ca-P minerals formation was

due to the presence of DOC and not to the decreased Ca and P solution concentrations.

Effects of Solids on Ca-P Crystallization

The presence of solids resulted in higher P concentrations in the incubated

solutions than in the absence of solids (Figure 7-8). The decline of P concentrations

during the incubation in the presence of solids might have been due to the P adsorption

onto the clay; no mineral phases of Ca-P were detected by XRD in any of the treatments

containing the solids. Phosphorus and Ca concentrations in final solutions were

significantly lower (p<0.05) for the low-density clay (p<2.0 Mg m-3) than for the dense

fraction (p > 2.0 Mg m-3) (Figure 7-10). The greater sorption of P and Ca by the low-

density clay may relate to its greater surface reactivity arising from either noncrystalline

(biogenically-derived) Si or residual organic matter not removed by H202, and is

consistent with a Ca-bridging mechanism for P retention. Dominance of Si in low-density









fraction was confirmed by EDS (Figure 7-11), and the presence of noncrystalline material

was indicated by a broad XRD peak ("amorphous hump") characteristic of noncrystalline

material. Also, SEM imaging of < 50-tm material verified the presence of appreciable

biogenic Si (Figure 7-12).

Summary and Conclusions

Representative concentrations of Mg in manure-amended soils inhibit the formation

of HAP but not the precipitation of a more soluble Ca-P mineral (brushite). Brushite that

forms locally in a soil matrix would be subject to dissolution in the next rainfall event,

thus favoring sustained P leaching from these soils. In the presence of DOC, Ca-P

crystallization was completely inhibited and no Ca-P mineral was detected. Si had no

inhibitory effect on Ca-P stabilization; and HAP was formed. Lower P and Ca

concentrations were observed for low-density clay relative to high-density clay, possibly

due to greater solids reactivity (as inferred from evidence of a higher proportion of

noncrystalline material) in conjunction with a Ca bridging sorption mechanism. No Ca-P

crystallization took place in the presence of clay size fractions. Generally, solids act as a

nucleation seed, adsorbed P thus provides a surface of Ca-P interaction. Such an effect

was not observed in this study.









Table 7-1. Average leachate composition of manure-amended soils used for the
incubating solutions
Chemical P Ca2+ Mg2+ Si4 DOC Fe2
Species
mg L1 68.0 312 179 23 427 0.60
Chemical A3+ K+ Na+ NH4+ C1- NO3- S042-
Species
mg L1 0.30 375 144 39.0 169 257 345


Table 7-2. Incubation treatments to study the effects ofMg, Si, and DOC on Ca-P
crystallization in the presence and absence of manure-derived solids
Absence of Solids No. of
Replicates
Control Solution I- no potential inhibitor 4
Treatment 1 Solution II-Mg as the only potential inhibitor 4
Treatment 2 Solution III-Si as the only potential inhibitor 4
Treatment 3 Solution IV-DOC as the only potential inhibitor 4
presence of Light Solids
Treatment 4 Solution I-light solids as the only potential inhibitors 3
Treatment 5 Solution II-Mg and light solids as potential inhibitors 3
Treatment 6 Solution III-Si and light solids as potential inhibitors 3
Treatment 7 Solution IV-DOC and light solids as potential inhibitors 3
Presence of Dense Solids
Treatment 8 Solution I-dense solids as the only potential inhibitor 4
Treatment 9 Solution II-Mg and dense solids as potential inhibitors 4
Treatment 10 Solution III-Si and dense solids as potential inhibitors 4
Treatment 11 Solution IV-DOC and dense solids as potential inhibitors 4
aLess number of replications due to low material availability











70 -

60 -

50 -
--
^ 40 -

30 -

20 -

10 -


a a


* Control
* Mg


Iwk 3wk 5wk lOwk 20 wk


Figure 7-1. P concentrations in the presence of Mg during 20 weeks (wk) of incubation.
Different letters (a and b) indicate statistically significant differences among
median concentrations (p<0.05) observed after an incubation period. HAP =
Hydroxyapatite.


Figure 7-2. XRD pattern of the precipitate in the Mg solution after 20 weeks of
incubation.












Mount Peaks









In the presence of DOC
HAP




No Inhibtor

10 20 30 40
Theta(deg)


Figure 7-3. XRD patterns of precipitates observed after 20 weeks of incubation in control
(no inhibitor), Si, and DOC treatments.


350 -

300 -

250 -

200 -

150 -

100

50 -

0-


a a


* Control

*Mg


a h ah


I I I I I I


0 h lwk 3wk 5v lO1wk 20 wk


Figure 7-4. Ca concentrations for control and Mg treatment during 20 weeks of
incubation. Different letters indicate statistically significant differences
(p<0.05) of median Ca concentrations after an incubation period.


700-


CC~Z~











70

60

50


S30-

20


10 -


n a


* Control
SSi


HAP'

In it


I I


I I I


Iwk 3wk 5wk 10wk 20 wk


Figure 7-5. Variations in P concentrations in the presence of Si during 20 weeks of
incubation. Different letters indicate statistically significant differences at
p<0.05. HAP=Hydroxyapatite.


U Control
70 aa
b b b o DOC
60 b

50 a b
a a
', 41)

S30- a

20

10 -

0
0 h 1wk 3wk 5wk 10wk 20 wk

Figure 7-6. P concentrations in the presence of soil DOC during 20 weeks of incubation.
Different letters indicate a significant difference (p<0.05) after the specified
incubation period.











400 -DOC solids

300

200

100

0 wk 3wk wwk 20
0 h lwk 3wk 5wk lOwk 20wk


Figure 7-7. Changes in DOC concentrations in control and in the presence of solids
during the 20 week (wk) incubation study.


70 -
60
50
-G 41)

30

20


10 -


a a


0 +


No Solids
SSolids




b
ab


I I


Iwk 3wk


Figure 7-8. Effects of clay size fractions on P concentrations during 20 weeks of
incubation. Different letters indicate statistically significant differences at
p<0.05 level.


500


-*- DOC control


5wk 102k 20 wk







83









-- -- -- ----





1 Flo at





Number of weelm 10
20


Figure 7-9. Variations in P concentrations due to the presence of low-density- (float) and
high-density- (sink) clay during 20 weeks of incubation. Different letters
indicate a significant difference (p<0.05) after the specified incubation period.









1-1










3
Ca I Jr










Number of wes 20


Figure 7-10. Variations of Ca concentrations due to the presence of low-density- (float)
and high-density- ("sink") clay during 20 weeks of incubation. Different
letters indicate a significant difference (p<0.05) after the specified incubation
period.












Counts



8000-


6000-


4000-
0

2000 Al
I Mg CI Ca Ti
Na P CI K Ti Fe

0 2 4 6 8 10
Energy (keV)


Figure 7-11. EDS spectrum of the low-density clay showing the dominance of Si.


Figure 7-12. SEM imaging of low-density clay showing the presence of biogenic silica
("dumbbell" serrated shaped particles in image on left; rod shaped particles in
image on right).














CHAPTER 8
SUMMARY AND CONCLUSIONS

Concentrated dairies result in accumulation of dairy manure on soils, giving rise to

risks of nutrient losses that can have deleterious effects on water quality. Phosphorus in

intensively-loaded soils can be labile even years or decades after site abandonment,

causing leakage at environmentally-unacceptable rates. Barriers to formation of stable

crystalline Ca phosphates, which soil solution data suggest would be thermodynamically

favored, pose both a scientific mystery and an environmental problem. The first step in

pursuit of a solution is to determine effects of critical soil components on P stability.

Dairy manure provides the most reactive components for most sandy soils. If the

components) inhibiting Ca-P crystallization could be eliminated or disabled, the result

should be more P assimilation and less P loss to surface water via lateral flow to streams,

etc. An understanding of reasons for continuous release of P from manure-amended soils

is critical for adopting nutrient management strategies to reduce the risk of P loss. The

overall objective of this dissertation was to understand the effects of dairy manure-

derived components (Mg, Si and DOC) on P release and Ca-P crystallization in dairy

manure-amended sandy soils. Research encompassed in this dissertation documented

inhibition effects of Mg and DOC on P stabilization. It also provided multiple lines of

evidence that P in dairy manure and dairy-manure-impacted soils is associated with Mg

as well as Ca, even though P has been presumed to exist primarily as Ca-P in dairy

manure. The association with Mg is significant because Mg-P and Ca-Mg-P are




Full Text

PAGE 1

DAIRY MANURE-COMPONENT EFFECTS ON PHOSPHORUS RELEASE FROM SANDY SOILS By MANOHARDEEP SINGH JOSAN A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2007

PAGE 2

2007 Manohardeep Singh Josan

PAGE 3

To my parents

PAGE 4

iv ACKNOWLEDGMENTS W ith all humility and sincerity, I bow before the Almighty for His benevolence and blessings for the comple tion of my dissertation. It is my privilege to express my deep sens e of gratitude to Dr. V. D. Nair, Research Associate Professor, Environmental Chem istry, Soil and Water Science Department, University of Florida for excellent guidan ce, invaluable suggestions and perceptive enthusiasm without which this work would not have its present shape. Her association and moral encouragement and approach to se e life perspectives as they are, throughout this academic pursuit, would be an invalu able experience of an everlasting value. My thanks go to my supervisory committee Willie G. Harris, George OConnor, Roy D. Rhue, Tom A. Obreza, Luisa A. Dempere, for valuable suggestions and positive criticism during the course of investigation. I express my sincere thanks to Willie G. Harris for providing excellent laboratory conditions, and abundant supply of peanuts. I am grateful to my wife Syliva Lang-Josan for her untiring and apt help in reading my manuscript. I am immensely grateful to my parents, brothers and sisters Gagan and Nimar. Their inspiration, sacr ifice, helpful blessing, encour agement, support, and loving emotions sustained me. I am thankful to Don Mitchell and Ilona Lang for their nurturing support, and love. My father is a farmer, has never had a ch ance to go school. He put all his efforts in me to fulfill his lifes ambitions. He provide d me the freedom and confidence that I

PAGE 5

v needed to succeed in my life. So this degr ee is a commemoration to his dedications and sacrifices that he made in his life.

PAGE 6

vi TABLE OF CONTENTS page ACKNOWLEDGMENTS................................................................................................. iv LIST OF TABLES............................................................................................................. ix LIST OF FIGURES...........................................................................................................xi ABSTRACT.......................................................................................................................14 CHAP TER 1 INTRODUCTION........................................................................................................1 Rationale and Significance...........................................................................................3 Hypotheses....................................................................................................................4 Specific Objectives.......................................................................................................4 2 REVIEW OF LITERATURE....................................................................................... 6 Manure Application......................................................................................................6 Release of P Using Various Extractants.......................................................................6 Geochemical Models: Forms and Solubility of Phosphorus........................................ 7 Solid State Assessments and P Associations.............................................................. 10 Prospective Inhibitors on Ca-P Crystallization...........................................................11 3 SOIL AND MANURE CHARACTERIZATION...................................................... 13 Soil Sampling.............................................................................................................. 13 Physicochemical Properties of Soils (Soil Characterization) .............................. 14 Particle Size Fractionation and Mineralogical Analysis ..................................... 15 Manure Sampling........................................................................................................ 15 Manure Characterization.....................................................................................15 Mineralogical Analysis........................................................................................16 QA/QC for Analyses........................................................................................... 16 Statistical Analyses..................................................................................................... 16 Results and Discussion...............................................................................................16 Soil Chemical Characterization........................................................................... 16 Soil Mineralogy...................................................................................................17 Manure Characterization.....................................................................................18

PAGE 7

vii Manure Mineralogy............................................................................................. 19 Summary and Conclusions.........................................................................................19 4 ASSOCIATED RELEASES OF PHOSPHORUS, CALCIUM AND MAGNESI UM IN SOIL SOLUTIONS FROM DAIRY MANURE-AMENDED SOILS.........................................................................................................................24 Introduction................................................................................................................. 24 Methods and Materials...............................................................................................25 Repeated Water Extractions and Chemical Analyses......................................... 25 Soil Leaching Characterization and Chemical Equilibrium Modeling............... 25 Results and Discussion...............................................................................................28 Repeated Water Extractions................................................................................28 Soil Leaching and Chemical Equilibrium Modeling........................................... 29 Summary and Conclusions.........................................................................................31 5 RELATIONSHIPS BETWEEN PHOSPHORUS, CALCIUM AND MAGNESI UM INFERRED FROM SELECTIVE DISSOLUTION......................... 39 Introduction................................................................................................................. 39 Material and Methods.................................................................................................40 Phosphorus Fractionation: Repeat ed and Sequential Extractions ....................... 40 Analyses of P and Metals.................................................................................... 42 Results and Discussion...............................................................................................43 Release of P, Ca, and Mg in Repeated 1.0 M NH4Cl Extractions.......................43 Phosphorus Concentrations in Sequential Extractions........................................ 45 Summary and Conclusions.........................................................................................47 6 SOLID STATE ASSESSMENTS: CONFIRMING THE ASSOCIATIONS OF PHOSPHORUS, CALCIUM AND MAGNESI UM................................................... 54 Introduction................................................................................................................. 54 Material and Methods.................................................................................................56 Approach I: X-ray Diffrac tion of Untreated Clays ..............................................56 Approach II: Ashed and Whole Dairy Manure Analyses.................................... 56 Approach III: SEM Imaging and EDS Analyses.................................................57 Approach IV: Electron Microprobe Microan alyses of Whole Silt + Clay of Dairy Manure-amended Soils..........................................................................58 Results and Discussion...............................................................................................58 Summary and Conclusions.........................................................................................61 7 CALCIUM PHOSPHATE CRYSTALI NITY AND DAIRY MANURE COMPONENTS .........................................................................................................67 Introduction................................................................................................................. 67 Material and Methods.................................................................................................68 Particle Size Separations of Dairy Manure-am ended Soils................................. 68 Carbonate and Organic Matter Removal from the Resulting Clays.................... 69

PAGE 8

viii Density Separations for Soil Clay Materials....................................................... 69 Preparation of Incubating Solutions.................................................................... 70 Incubation Setup, Monitoring and Solution Analyses.........................................71 Solid State Assessments...................................................................................... 72 X-ray diffraction analyses............................................................................ 72 Energy dispersive spectroscopy analyses..................................................... 72 Statistical Analysis...................................................................................................... 73 Results and Discussion...............................................................................................73 Effects of Mg, Si and DOC on Ca-P Crystallization...........................................73 Effects of Solids on Ca-P Crystallization............................................................ 76 Summary and Conclusions.........................................................................................77 8 SUMMARY AND CONCLUSIONS.........................................................................85 APPENDIX A DETAILS OF LEACHATE CONCENTR ATIONS USE D FOR V-MINTEQ ANALYSES............................................................................................................... 90 B CORRELATION MATRIX FOR DIFFERE NT P FRACTIONS OF DAI RY MANURE-AMENDED SOILS..................................................................................95 LIST OF REFERENCES...................................................................................................97 BIOGRAPHICAL SKETCH...........................................................................................107

PAGE 9

ix LIST OF TABLES Table page 3-1 Characteristics of active and abandoned m anure-amended soils, and minimally manure-impacted soils.............................................................................................. 20 3-2 Characteristics of dairy manures colle cted from four locations in Florida. ...............21 4-1 Cumulative average release of SRP, Ca, Mg and EC in repeated water extraction s of active, abandoned and minimally-impacted soils................................................32 4-2 Saturation indices (SI) in active and abandoned dairy m anure-amended soils.......... 33 4-3 Percent of observations that are undersaturated, satura ted, and supersaturated for selec ted mineralsa based on chemical equilibrium modeling of active and abandoned dairy column leachates........................................................................... 34 5-1 Total dissolved phosphorus (TDP) as a function of Ca+Mg in repeated 1.0 M NH4Cl extractions (Data from 1st extraction omitted)............................................. 49 5-2 Sequential release of total dissolved phosphorus (TDP ) as a function of Ca, Mg, and Fe in 0.1 M NaOH extractions ........................................................................... 49 5-3 Sequential release of total dissolved phosphorus (TDP ) as a function of Ca, Mg, Fe, and Al in 0.5 M HCl extractions ........................................................................49 5-4 Sequential release of total dissolved phosphorus (TDP ) as a function of Ca, Mg, Fe, and Al in residual fractions ................................................................................49 7-1 Average leachate composition of manure-amended soils used for the incubating solutions ...................................................................................................................78 7-2 Incubation treatments to study th e effects of Mg, Si, and DOC on Ca-P crystallization in the presence and absence of m anure-derived solids.....................78 A-1 Column leachate pH, EC and P concentrations of leachates..................................... 90 A-2 Concentrations of Ca, Mg and dissolved organic carbon (DOC) observed in colum n leachates...................................................................................................... 91 A-3 Concentrations of K, Fe a nd Al observed in colum n leachates................................. 92

PAGE 10

x A-4 Concentrations of sulfate, chloride and ammonium observed in column leachates.. 93 A-5 Concentrations of ni trates, and silcic acid (H4SiO4) observed in column leachates. 94 B-1 Correlation matrix for different fract ions in active dairy m anure-amended soils.....95 B-2 Correlation matrix for different fr actions for abandoned dairy m anure-amended soils.......................................................................................................................... .96

PAGE 11

xi LIST OF FIGURES Figure page 3-1 X-ray diffraction patterns of clays obtained from active dairy (ACS) manureamended soils. HIV = hydroxyinterlayered vermiculite.......................................... 22 3-2 X-ray diffraction patterns of clays obtained from abandoned dairy (ABS) manureamended soils........................................................................................................... 22 3-3 X-ray diffraction patterns of clays obtai n ed from minimally-impacted (MIS) soils.. 23 3-4 X-ray diffraction pattern of clay obt ained from an oven dried dairy m anure............ 23 4-1 Column set-up used for the column leaching study................................................... 35 4-2 Changes in electrical conductivity (EC) (dS m-1) with repeated water extractions. ACS = active dairy manureimpacted soil; ABS = abandoned dairy manure impacted soil; MIS = minimally impacted soil........................................................ 35 4-3 Changes in Ca concentrations (mg kg-1) with repeated water extractions. ACS = active dairy manureamended soil; AB S = abandoned dairy manureamended soil; MIS = minimally impacted soil........................................................................ 36 4-4 Changes in Mg concentrations (mg kg-1) with repeated water extractions. ACS = active dairy manureamended soil; AB S = abandoned dairy manureamended soil; MIS = minimally impacted soil........................................................................ 36 4-5 Changes in soluble reactive pho sphorus (SR P) concentrations (mg kg-1) with repeated water extractions. ACS = active dairy manureamended soil; ABS = abandoned dairy manureamended soil; MIS = minimally impacted soil............... 37 4-6 Relationships between soluble reactive phosphorus (SRP) and Mg and Ca released during repeated water extractions of active dairy m anureamended soils...........................................................................................................................37 4-7 Relationships between soluble reactive phosphorus (SRP) and Mg and Ca released during repeated water extractions of abandoned dairy m anureamended soils...........................................................................................................................38 4-8 Relationships between soluble reactive phosphorus (SRP) and Mg and Ca released during the co lumn leaching of dairy manure-amended soils..................... 38

PAGE 12

xii 5-1 Schematic of repeated and sequential extraction pro cedure adapted from Nair et al. (1995).................................................................................................................. 50 5-2 Release of phosphorus using 1.0 M NH4Cl repeated extracti ons in active and abandoned dairy manure-amended soils. ACS = active dairy manureamended soil; ABS = abandoned dairy manureamended soil............................................... 50 5-3 Release of calcium using 1.0 M NH4Cl repeated extractions in active and abandoned dairy manure-amended soils. ACS = active dairy manureamended soil; ABS = abandoned dairy manureamended soil............................................... 51 5-4 Release of magnesium using 1.0 M NH4Cl repeated extractions in active and abandoned dairy manure-amended soils. ACS = active dairy manureamended soil; ABS = abandoned dairy manureamended soil............................................... 51 5-5 Distribution of P in 1.0 M NH4Cl, 0.1 M NaOH, 0.5 M HCl, and residual-P fractions for an active dairy soil (ACS-2)................................................................ 52 5-6 Distribution of P in 1.0 M NH4Cl, 0.1 M NaOH, 0.5 M HCl, and residual-P fractions for an active dairy soil (ACS-4)................................................................ 52 5-7 Distribution of P in 1.0 M NH4Cl, 0.1 M NaOH, 0.5 M HCl, and residual-P fractions for an abandoned dairy soil (ABS-3)........................................................53 5-8 Distribution of P in 1.0 M NH4Cl, 0.1 M NaOH, 0.5 M HCl, and residual-P fractions for an abandoned dairy soil (ABS-4)........................................................53 6-1 X-ray diffraction patterns of four act ive (AC S-1 to ACS-4) and four abandoned (ABS-1 to ABS-4) dairy manureamended untreated clays. HIV = hydroxyinterlayered vermiculite..............................................................................62 6-2 X-ray diffraction pattern of < 1.0 mm dried dairy m anures showing the presence of whewellite, quartz and calcite.............................................................................. 62 6-3 X-ray diffraction patterns of oven drie d and ashed dairy m anures (3) showing the presence of Mg-Ca Whitlockite (Mg-Ca phosphate)............................................... 63 6-4 Energy dispersive spectrum of an ashed dairy m anure showing the high intensity peaks of Ca, Mg and P.............................................................................................63 6-5 Energy dispersive dot maps of a dair y m anure showing an association of Mg and P................................................................................................................................64 6-6 Energy dispersive dot maps (scale 20 m ) of a dairy manure-amended soil clay showing an association of P and Ca......................................................................... 64 6-7 Dot image of dairy manure showing the spatial associations of Mg, P, and Ca. ....... 65

PAGE 13

xiii 6-8 EDS spectrum of a manure P rich pa rticle ob tained at 400X magnification showing the dominance of Mg, P, and Ca................................................................ 65 6-9 Relationships between P, Mg and Ca for active (6-9a & 6-9b) and abandoned dairy (6 -9c & 6-9d) manure-amended dry sieved (45m) silt+clay using the electron probe microanalyses (EPMA). **significant at p<0.01............................. 66 7-1 P concentrations in the presence of Mg during 20 weeks (wk) of incubation. Different letters (a and b) indicate statistically s ign ificant differences among median concentrations ( p 0.05) observed after an incubation period. HAP = Hydroxyapatite......................................................................................................... 79 7-2 XRD pattern of the preci pitate in the Mg solution af ter 20 weeks of incubation....... 79 7-3 XRD patterns of precipitates observed after 20 weeks of incubation in control (no inhibitor), Si, and DOC t reatments...........................................................................80 7-4 Ca concentrations for control and Mg treatm ent during 20 weeks of incubation. Different letters indicate statis tically significant differences ( p 0.05) of median Ca concentrations afte r an incubation period........................................................... 80 7-5 Variations in P conc entra tions in the presence of Si during 20 weeks of incubation. Different letter s indicate statistically significant differences at p 0.05. HAP=Hydroxyapatite.................................................................................. 81 7-6 P concentrations in the presence of soil DOC during 20 weeks of incubation. Different letters indicate a significant difference ( p 0.05) after the specified incubation period. .....................................................................................................81 7-7 Changes in DOC concentrations in cont rol and in the presence of solids during the 20 week (wk) incubation study. ...............................................................................82 7-8 Effects of clay size fractions on P c oncentrations during 20 weeks of incubation. Dif ferent letters indicate statis tically significant differences at p 0.05 level.......... 82 7-9 Variations in P concentrations due to the presence of low-density(float) and high-density(sink) clay during 20 w eeks of incubation. Different letters indicate a s ignificant difference ( p 0.05) after the specified incubation period...... 83 7-10 Variations of Ca concentrations due to the presence of low-density(float) and high-density(sink) clay during 20 weeks of incubation. Different letters indicate a s ignificant difference ( p 0.05) after the specified incubation period...... 83 7-11 EDS spectrum of the low-density clay showing the dom inance of Si...................... 84 7-12 SEM imaging of low-density clay showing the presen ce of biogenic silica (dumbbell serrated shaped particles in image on left; rod shaped particles in image on right)......................................................................................................... 84

PAGE 14

14 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy DAIRY MANURE-COMPONENT EFFECTS ON PHOSPHORUS RELEASE FROM SANDY SOILS By Manohardeep Singh Josan May 2007 Chair: Vimala D. Nair Major: Soil and Water Science Phosphorus (P) in heavily manure-amended soils can be labile even years or decades after manure input cease. Knowledge of manure-derived components and their associations with P is per tinent to nutrient management for sandy soils with minimum P sorbing capacity. The overall objective of this research was to understand the effects of manure-derived components such as Mg, Si and dissolved organic carbon (DOC) on P solubility in manure-amended sandy soils. Soil samples (0-25 cm) from manure-impacted areas were collected from the Suwannee and Okeechobee Basins of Florida. The soil release of P, Ca and Mg was studied using repeated water extractions and 1.0 M NH4Cl extractions. Columns of the soils were l eached with deionized water and leachate speciation was modeled using MINTEQ. Solid state assessments of dairy manure and manure-amended soils were done using x-ray diffraction, scanning electron microscopy, and elemental microanalysis. The inhibitory effects of Mg, Si, and manure-derived DOC on Ca-P crystallization were studied by incubating solutions with and without clay-sized solids for 20 weeks. Repeated water and a mmonium chloride extractions and speciation of column leachates confirm that sparinglysoluble phases of P asso ciated with Mg and

PAGE 15

15 Ca control P release from the manure-amended soils and maintain elevated P concentrations in soil solutions even years after abandonment of the dairies. Solid state assessments suggested Mg-P and Ca-P associations in dairy manure and manureamended soils. Formation of the most stable Ca-P mineral, hydroxyapatite, was inhibited by Mg and/or DOC, but not Si, in dairy ma nure amended soils. Mg-P associations in manure and manure-amended soils could maintain elevated P solubility, and Mg in soil solution could inhibit formation of stable form s of Ca-P. Therefore, consideration of Mg and Ca is necessary to explain the nature of P in manure-amended soils. Preemptive dietary controls to maximize Ca-P and minimi ze Mg-P in manure would be a strategy to reduce P loss from these soils in the future. Application of Al-based water treatment residuals could minimize the release of P from the manure-amended soils. DOC inhibition of Ca-P precipitation and competitive effects on P sorption reduce prospects for stabilizing P reactions in heavily manure-amended soils.

PAGE 16

1 CHAPTER 1 INTRODUCTION Continuous release of phosphorus (P) in dairy m anure-amended sandy soils even years after manure addition cease (dairy aba ndonment) poses both a scientific mystery and an environmental problem (Nair et al., 1995). The USEPA (1996) identified Pinduced eutrophication as the most extensive cause of water quality impairment in the USA, and the USGS identified agriculture as a major source of P to surface waters (United States Geological Survey, 1999). Dairy manure accumulation in soils can increase the potential for P loss to surface waters either via erosion (Sharpley & Smith, 1983) or subsurface drainage (Mansell et al., 1991). The P enrichment can cause both surface and sub-surface water pollution (Whalen & Chang, 2001). Many soils effectively retain P, but some sandy soils can be exceptions due to a paucity of P-retaining minerals (Neller et al., 1951; Ozanne et al., 1961; Gillman, 1973; Burgoa, 1991; Mansell et al., 1991; Harris et al., 1996; Nair et al., 1998; Novak et al., 2003). Thus, the stability of manure-derived forms of P is an especially relevant environmen tal concern in sandy soils. Soil environmental factors such as pH, the presence of dissolved complexing species and the kind of phosphate mineral pr esent determine the phosphate activity in solution (Lindsay, 1979). Dairy manure-amended sandy soils typically contain large amounts of Ca and P in both solid and solu tion phases, with accompanying moderately higher pH (Nair et al., 2003) than the na tive, non-impacted soils. The conditions thermodynamically favor the formation of relatively stable Ca-P minerals (Lindsay,

PAGE 17

2 1979), but the release of P from these soils can be greater than predicted from the solubility of the minerals (Wang et al., 1995). High Mg concentrations in soil solution of heavily manure-amended soils suggest that Mg, in addition to Ca, could control the release of P via a sparingly soluble Mg-P pha se (Nair et al., 1995; Josan et al., 2005). Alternatively, Mg can act as an inhibiting cation for Ca-P crystallization on calcium carbonate by masking adsorption sites, in the pr esence of high P concentrations (Yadav et al., 1984). If the latter were the case, abandoned dairy ma nure-amended soils should exhibit some Ca-P stabilizati on after soluble salts (e.g. MgCl2, CaCl2) are leached (Harris et al., 1994). Active da iry manure-amended soils are thos e that currently receive dairy manure, whereas abandoned dairy manure-am ended soils are those soils where dairy activities have ceased for at least 10 years. Minerals such as vaterite, whitlockhite, monetite and struvite in poultry and pig manure possibly control solution P and the major ity of Ca and Mg in the soil solutions of soils amended with poultry manure are comp lexed by dissolved or ganic matter (Bril & Salmons, 1990). Density separation of the clay fraction of heavily P fertilized loamy soils yielded P-rich particles associated with Fe, Al and Ca (Piers enyzki et al., 1990a). Scanning electron microscopy (SEM) and energy dispersive x-ray spectroscopy (EDS) of poultry manure suggested that sparingly so luble Ca and Mg -phosphate minerals controlled soil solution P c oncentrations (Cooperband & Good, 2002), the presence of discrete P forms/minerals in dairy manures and dairy manure-amended soils is lacking (Harris et al., 1995; Cooperband & Good, 2002). X-ray absorption near-edge spectroscopy (XANES), a non-dest ructive chemical speciation technique, was used to assess P minerals and P speciation in phosphorus-enriched agricultural soils

PAGE 18

3 (Beauchemin et al., 2003).The authors estimated that most of P was present as octacalcium phosphate (OCP) (45% of total P) and hydroxyapatite ( HAP) (11% of total P). Rationale and Significance Phosphorus in intensively manure-amended soils can be labile years or decades after abandonment (Nair et al., 1995), and cause P leakage at environmentallyunacceptable rates. Heavy loading of dairy manure to soils, leads to continuous release of P, Ca, and Mg both in active and abandoned dairy manure-amended soils (Nair et al, 1995). If a sparingly soluble Mg-P phase is responsible for the c ontinued release of P from the manure-amended soils, other amendment (e.g. water treatment residuals) applications may be the only way to stabi lize the P (OConnor & Elliott, 2000) because all Mg-P phases are relatively soluble (Lindsay, 1979). Alternatively, reduction of P via dietary management could be a viable strate gy to reduce P solubil ity in the manure (Dou et al., 2003; Cerosaletti et al., 2004). Add itionally, if a Ca-Mg-P phase controls the release of P from the soils, it could be subjec t to dissolution. Ruminant feed contains both Mg and Ca and the Mg and Ca doses ar e typically exceeding recommended levels (National Research Council, 2001). During heat stress, in creased diet potassium and magnesium levels are recommended (Beede & Shearer, 1991). It is important to document if Ca or Mg or both are initially associated with the high release of P from dairy manure-ame nded sandy soils. A finding that Mg-P is prevalent would diminish the prospect of stabilization via slow crystallization and transformation of P to more st able Ca-P forms. Also, loss of more soluble forms of Ca after abandonment would diminish the prospect of P released from a Mg-P phase being precipitated as Ca-P.

PAGE 19

4 Knowledge of manure-derived components and their associations with P is pertinent to nutrient management, particular ly for sandy soils with minimum P sorbing components. This research probed the forms of Ca-P and Mg-P in dairy manure and dairy manure-amended soils in hopes of identifying ph ases responsible for long-term P release and developing mitigation approaches to re duce P loss. The overall objective was to understand the role of manure-derived compone nts, specifically Mg, Si, and dissolved organic carbon (DOC), in main taining high P solubility in active and abandoned dairy manure-amended sandy soils. Specific hypothe ses and objectives are given below: Hypotheses 1. Active and abandoned dairy m anure-amended soils release comparable amounts of P because solution P is controlled by sp aringly-soluble Mg and/or Ca phosphate phases that require many years for depletion. 2. Concentrations of P, Ca, and Mg are spatially correlated in solid manure and manure-amended soil samples. 3. Activities of DOC, Mg and Si in soil solution of manure-amended soils are sufficient, jointly or separately, to inhibit crystallization of stable Ca-P forms, thus leading toward high P release from the soils. 4. Noncrystalline Si forms (including biogenic Si in dairy manure) can retain P at circumneutral pH and high Ca activity becau se Ca serves as a bridge between the silicate surface and P. Specific Objectives 1. Assess the release of P, Ca and Mg in so il solutions of dairy m anure-amended soils. 2. Study the associations of P, Ca, and Mg in dairy manure and manure-amended soils using solid state assessments. 3. Study the effects of Mg, Si and DOC on Ca-P crystalliza tion using average concentrations of the species found in manure-amended soil leachates. 4. Study the role of low-d ensity clay (<2 Mg m-3) and high-density clay ( 2 Mg m-3) from manure-amended soils on Ca-P crystallization.

PAGE 20

5 Chapter 2 discusses previous researches on P release, geochemical modeling, and solid state assessments in relation to soil and manure environments, and prospective inhibitors of Ca-P crystallization in soil science and other fields. The soil and manure sample collection and characterization proce dures are discussed in Chapter 3. The first study (Chapter 4) assesses the release of P, Ca and Mg in soil solution and tested the hypothesis that abandoned and active dairy manure-amended soils release comparable amounts of P because solution P is controlled by a sparingly-soluble Mg-P phase or CaMg-P phase that requires many years for depletion. Testing included repeated water extractions and a column leaching experiment. A selective dissolution study (Chapter 5) was done on manure-amended soils. Various inorganic and organic forms of P were measured and the relationships to dissolved Ca, Mg, Fe, and Al were documented. Chapter 6 describes the associa tions of P with Ca and Mg in minimally altered samples of dairy manure and manure-amended soils, us ing x-ray diffraction (XRD), scanning electron microscopy (SEM), energy dispersi ve spectroscopy (EDS), and electron probe microanalyses (EPMA). The study was conducted to confirm the associations of P, Ca and Mg observed in solution P chemistry. Resu lts tested the hypothesis that Ca-P, Mg-P or Ca-Mg-P phases exist in dairy manure and manure-amended soils, and are spatially associated at the microscopic level. The results of the column leaching experiment (Chapter 4) and x-ray diffrac tion studies (Chapter 6) sugge sted that the dairy manurederived components can inhibit the formation of stable Ca-P mineral phases. Therefore, the inhibitory effects of dairy manure-derive d components on the Ca-P stabilization were investigated using an incuba tion study (Chapter 7). The final chapter (Chapter 8) summarizes the four individual studies and how they relate to each other.

PAGE 21

6 CHAPTER 2 REVIEW OF LITERATURE Manure Application In intensive dairy production areas, including the Okeechobee basin in Florida (FL), m ore dairy manure is generated than n eeded to meet crop nitrogen (N) requirements of available crop land. The long-term applicat ion of dairy manure at N-based rates in such areas has increased levels of P in soils above the crop needs (S harpley et al., 1996; Kleinman et al., 2000), and values frequently approach or exce ed environmental P thresholds. Lawsuits have been filed in some animal production area, such as in EuchaSpavinaw Basin of Ozarks (DeLaune et al., 2006) to restrict the land applications of animal wastes thus leaving surplus at the farm. Sims et al. (2000) recommended the integration of soil P tests into environmentally based agricultural management practices so that optimum levels of manures can be applied to a piece of agricultural land. Release of P Using Various Extractants The release of P from manur e-am ended soils with successive extractions has been studied by several researchers, using different soil extract ants. Sharpley (1996) observed that the release of P with 0.01 M CaCl2 using Fe strips as the adsorbent decreased exponentially with successive extractions. A sequential extraction before and after 15 strip P extractions revealed that majority of P released (46%) was i norganic. Nair et al. (1995) repeatedly extracting manure-amended soils with 1.0 M NH4Cl solution, and found that 80% of total P in the surface horizons of these soils was labile P (readily soluble P) and associated with Ca and Mg-P forms. Authors speculated that P was loosely

PAGE 22

7 bound with Ca and Mg, probably by some w eak adsorption mechanism or as poorly crystalline solids, and as available for sust ained leaching under suitable conditions. The classical use of soil P fractionation of Chang & Jackson (1956) use 1.0 M NH4Cl to effectively remove water soluble and loosel y bound P during the initi al extractions. The Hedley et al. (1982) P fractionation procedure is broadly adop ted to differentiate various organic and inorganic pool s of P. Additionally, NH4Cl extractable P has been defined as loosely adsorbed or easily available P (Psenner & Pucsko, 1988) or easily soluble P (Williams et al., 1967). Cooperband & Good (2002) suggested that sparingly-soluble Ca and MgP minerals (more soluble than apat ite) controlled solution P concentrations in soils amended with poultry manure but were unabl e to directly identify the forms. He et al. (2004) sequentially extracte d 0.25 g of dairy manure in 25 mL of deionzed water for 2 hours, and this water extractable P was the largest fraction of total-P. Most of the estimated P was in the inorganic form ( 12 to 44% of manure total P), and water extractable P was better correlated (r2 = 0.62) to total P than organic P (r2 = 0.24). There have been few efforts to study the associated cations like Ca, Mg, Fe and Al that are released with P in manure-amended soils (Sharpley et al., 2004; Josan et al., 2005; Silveira et al., 2006). Geochemical Models: Forms and Solubility of Phosphorus Various computer geochem ical speciation models can assist in understanding the forms and P-speciation in waste waters a nd soil solutions. The most commonly used models are MINTEQ (Felmy et al., 1984), MINTEQA2 (Allison et al., 1990), GEOCHEM (Sposito & Mattigod, 1980), and V-MINTEQ (Department of Land and Water Resources Engineering. 2006, an updated version of MINTEQA2).

PAGE 23

8 All the models assume equilibrium among dissolved species. The assumptions for the equilibrium for aqueous (d issolved) speciation reactions are likely appropriate, because most of the interactions occur ve ry rapidly (Pankow & Morgan, 1981). Most of the ion exchange and adsorption/ desorption reactions in well stirred soils systems attain equilibrium within several hours (Mattigod, 1995). These models are called equilibria or solution speciation models and incorporate corrections for activity coefficients and solution complexation reactions and, thus, eval uate the saturation status of the solution with respect to thermodyamically stable solid phases. Many models assume activity coefficients depend only on th e ionic strength and ignore specific ion effects. Ion interactions become important a bove ionic strengths of 0.5 mol L-1, and must be considered (Mattigod, 1995). Some models like MINTEQ (Felmy et al., 1984) and GEOCHEM (Sposito & Mattigod, 1 980), also assume that the solution is in equilibrium with the thermodynamically predic ted solid phases. Such models are predictive in nature and are called solution-solid equilibria models. The models modify the solution composition, assuming that only the most stable phase (or phases) ca n occur and that the solution is always in equilibrium with the so lids. Predictive models can also be used as solution speciation models by disa bling the solid phase reactions. Zhang et al. (2001) used MI NTEQA2 to predict the natu re of aluminum and iron-P fractions in sandy soils of Florida. Wavelite crandallite, variscite and strengite were predicted to stable in fertilized acid soils, wh ereas at higher soil pH values, Ca-P minerals were predicted to control P activities in soil solutions. Sharpley et al. (2004) studied the effect of long term manure applications on soil Ca-P forms using MINTEQA2. Ion activity products of soil solutions were calcu lated using ion activities determined in

PAGE 24

9 extracts of soil reactions for 16 h with 0.01 M CaCl2 at 1:5 soil/solution ratio. Long term (10-15 y) manure-amended soils were dominat ed by the more soluble crystalline Ca-P forms tricalcium phosphate and octacalcium phosphates than the less HAP. Shanker & Bloom (2004) cautioned that ge ochemical modeling results s hould be supported by solid state assessment techniques. They also observed that Oversaturation alone is merely a prer equisite for mineral formations; nonequilibrium is the common state in soils and kinetics of precipitation would be an important factor to dictate whether and to what extent various Ca-, Fe-, or Al-P would accumulate in the soil. Hutchison & Hesterberg (2004) studied the effects of dissolved organic carbon (DOC) as citrate on P dissolution on the soil amended with swine lagoon. With increasing rates of citrate concentrations, dissolved reactive P (DRP) and total Fe and total Al concentrations were increased. Geochemical modeling (V-MINTEQ ) predicted 69 to 99% of aqueous Fe(III) complexed as Fe(citrate)0, and 87 to 100% of Al(III) complexed as Al(citrate)0 and Al(citrate)3-. The authors concluded that Fe and Al can be complexed by dissolved organic matter as Al-DOM or FeDOM complexes in the system. Silveira et al. (2006) studied P solubility characteristics in a dairy manure-amended sandy soil under same conditions. Using V-MINTEQ as a spec iation model for leachates, they observed HPO4 2 (~50% of total soluble P), and Mg-P [MgHPO4(aq)] and Ca-P (CaPO4 ) complexes (~30 and 13% of total soluble P, respectively) as the major chemical P species and thus concluded that both Ca-P and Mg-P mineral pha ses can be a major factor in controlling long term P release in manure-amended soils. Geochemical models can be valuable tool s in describing P speciation in various kinds of soil environments. However, model predictions must be validated with solid state assessments to confirm the presence of a particular mineral phase or a phase

PAGE 25

10 association of P with other metals at micro scale (Mackay et al., 1986; Brennan & Lindsay, 1998). Solid State Assessments and P Associations Solid state assessm ents are less destructiv e than chemical or solution extractions and can be helpful in providing better descrip tion of relationships of P with metals. X-ray diffraction (XRD) produces constr uctive interference of coherently scattered x-rays, and produces diffraction peaks related to spacing of atomic planes in samples and wavelength of x-rays (Amonette, 2002). X-ray diffracti on needs minimal soil pr eparation; however, the identification of poorly ordered/short range material s is impossible (Harris et al., 1994). Scanning electron microscopy (SEM) prov ides large depth of fields and requires minimal sample preparation, and particles can be seen at very high resolution (Goldstein et al., 2003). Pierzynski et al. (1990a) succes sfully quantified the P minerals formed under excessively fertilized soils where to tal P concentrations ranged from 540 mg kg-1 to 8340 mg kg-1. Researches used XRD, scanning electron microscopy (SEM) equipped with energy dispersive spectroscopy (EDS ), and Fourier-transformed infrared spectroscopy to identify P minerals in concentrated clay size fractions obtained from excessively fertilized soils. The use of XRD was unsuccessful in identifying P bearing minerals in the soils. However, SEM identified P-rich particles in the soils with detectable quantities of Al, Si Ca and Fe that were associ ated with P. Huang & Shenker (2004) studied the solid state speciation of P in stabilized sewage sludge using XRD, SEM, and energy dispersive x-ray spectroscopy (EDXS). The XRD patterns confirmed the formation of brushite, ferrian variscit e, calcite and dolomite in ferrous sulfate stabilized sludge. The SEM images also showed an elemental spatial correlation of P and Ca, P and Fe and confirmed the presence of minerals detected in XRD analyses. Using

PAGE 26

11 transmission electron microscopy, Jager et al (2006) observed the presence of HAP (with significantly broadened XRD peaks), which wa s confirmed by the solid state nuclear magnetic resonance (NMR) technique. The Ca/P ratio of 1.52 estimated by the NMR studies agreed with Ca/P ratios obtained fr om chemical analyses of nanocrystalline hydroxyapatite. Prospective Inhibitors on Ca-P Crystallization Magnesium is a biologically essential elem ent and reduces heat stress in ruminants (Beede & Shearer, 1991). In the presence of high concentrations of grass K, the animal absorbs less Mg in the rumen and suffers hypomagnesemia (Littledike et al., 1983). As a result, there is a tendency to feed more Mg as feed supplements. Mg affects the crystallinity of synthetic apatit e (LeGeros et al., 1989; Bigi et al., 1993) and also inhibits the formation of apatite (LeGeros et al., 1989; Abbona & Franchini-Angela, 1990). An amorphous phosphate phase, magnesium-whitlock ite, was observed in the presence of Mg (LeGeros & LeGeros, 1984). Martin & Brow n (1997) investigated the effect of Mg on the formation of Ca-deficient HAP [Ca9HPO4(PO4)5OH, CDHAP]. The progress of the reaction was determined by isothermal calorimetry. They observed two heat peaks during the formation of CDHAP in water at 37.4oC with 10 mM of Mg concentrations. High Mg ion concentrations (3.16 M) resu lted in the formation of Newberyite (MgHPO43H2O), but no Ca-P mineral phase was obs erved. Researches concluded that Mg-P complexes are more likely to inhibit th e Ca-P formation than magnesium chloride complexes. Mg delays (or prevents) the c onversion of the amorphous Ca-P phase into a crystalline P phase (LeGeros et al., 1976). Sahai (2005) modeled th e apatite nucleation using crystallography, and NMR a nd suggested that an outersp here complex of Mg with P formed faster than Ca-P inner-sphere comple xes, which inhibited the Ca-P interactions

PAGE 27

12 at the active silanol sites. The behavior of Mg was attributed to a greater charge density for Mg than Ca, which favors greater electrostatic attracti ons at the active sites. The effect of manure-derived sili ca on Ca-P interactions has not been studied. In a dental study, Damen & Cate (1992) observed decreased induction time of Ca-P precipitation in the presence of SiO2, which resulted in spontaneous precipitati on of calcium phosphate with a wide range of Ca:P ratios from supersaturated solutions. Addition of organic amendments to calcareous soils increased P solubility with time more than a single addition of inorganic phosphate (OConnor et al., 1986). Inskeep & Silvertooth (1988) observed that organic acids common to soil environments inhibited HAP precipitation and concluded that organic acids adsorbed on to crystal seeds acting as nuclei for crystal growth. Lindsay et al. (1989) suggested that adsorption regulates P retention at low P concentrations, whereas mineral precipitation co ntrols solubility at high P concentrations. However, Kim et al. (2005) observed that formation of HAP from amorphous calcium phosphate is an internal rearrangement pro cess rather than a di ssolution-precipitation process. Long term addition of dairy manure to land not only increases P concentrations but Ca, Mg, Si and DOC also. The effect of these manure-derived components on Ca-P interactions needs to be addressed.

PAGE 28

13 CHAPTER 3 SOIL AND MANURE CHARACTERIZATION Soil Sampling Soil sam ples were collected from manure-a mended soils at four active (ACS-1 to ACS-4) and five abandoned dairies (ABS-1 to ABS-5), and four minimally-impacted soils (MIS-1 to MIS-4) from the Suwannee River Basin and Lake Okeechobee Basin of Florida. The active dairy manur e-amended soils currently receive dairy manure, whereas the abandoned dairy manure-amended soils no longer receive high levels of dairy manure daily. Abandoned dairy sites were formerly heavily manure-amended and received high manure loads similar to the active dairy site s. Years of abandonment ranged from 12-32 years. Soils were collected by tile spade to a depth of 25 cm, or to the bottom of the Ap horizon (whichever was shallower), from re presentative locations within the highintensity areas. Composited Ap horizon soil samp les were also taken at each site using a 4.0 cm diameter soil auger for bulk density cal culations. Representative soil profiles (to a two-meter depth) were sampled to incl ude Bt/Bh-horizons (when present) for classification and characterizat ion purposes. Parent materi als for all soils were sandy marine sediments. Slopes were < 2% and draina ge classes were estimated to be poorlyto somewhat poorly drained. Active sites were ba re or nearly bare, whereas abandoned sites were grassed. Two of the active sites were on Spodosols and two were on Ultisols. All abandoned sites were on Spodosols. One activ e site (ACS-1) and one abandoned site (ABS-1) had fill material greater than 25-cm thick, which obviated classification for the purposes of this study. Samples from one act ive (ACS-4) and one abandoned dairy site

PAGE 29

14 (ABS-2) were supplied by other researchers. Soil profiles descriptions were not confirmed for the two sites, but both sites were located in areas dominated by Spodosols. Soil samples were collected to approximate depths of 25 cm for active and 15 cm for abandoned dairies. All samples were either dr ied shortly after collec tion or stored moist under refrigeration. Soils were air dried and crushed to pass 2-mm sieve before use. It also helped in potential screening out CaCO3 material applied as fill material. All minimally-impacted (MIS 14) soil sample s were obtained from other researchers (Graetz et al., 1999). Physicochemical Properties of Soils (Soil Characteriz ation) The soil pH, which is an indicator of so il reaction, was measured with a glass calomel electrode assembly using 1:2 soil wa ter suspensions. Soil el ectrical conductivity (EC, dS m-1) was determined in 1:2 soil:water ratio with the help of a conductivity bridge. Total P and metals i.e. Ca, Mg, Fe, Al, Na, and K in the soils were determined by the procedure outlined by Anderson (1974). On e gram of soil was weighed into a 50 mL glass beaker and placed in a muffle furnace at 350oC for an hour. The furnace temperature was raised to 550oC and soil was ignited at this temperature for 2 h. The furnace was allowed to cool for overnight. Fe w water drops were used to moist the ash and 20 ml of 6.0 M HCl was added with a graduated cy linder and allowed the solution to evaporate slowly on a hot plate (80oC). Additionally 2.25 ml of 6.0 M HCl was added to the digested sample to dislodge the residue and solution was transf erred quantitatively (Whatman # 41 filter paper) in to a 50 ml vol umetric flask by washing the beaker several times with smaller amounts of DI water. The fi ltrate was collected in the same flask each along with rinsing the sides of the filter pa per into the flask. Phosphorus was analyzed by

PAGE 30

15 ascorbic acid colorimetry (Murphy & Rile y, 1962) (U.S. EPA, 1993; method 365.1). All metals were analyzed by atomic absorption spectrophotometry. Particle Size Fractionation and Mineralogical Analysis Air-dried so il samples (50 g each) were treated with bleach (10% sodium hypochlorite, adjusted to pH 9.5), at a 1:20 soil:bleach ratio, overnight to oxidize organic matter (Lavkulich & Wiens, 1970). The supern atants were siphoned off and the soils transferred to 250-mL centrifuge bo ttles for washing (3 times) with 1.0 M NaCl to remove entrained bleach. Samples were then washed with deionized (DI) water (3-4 times) to remove salt, until the supernatant a ppeared turbid. Deionized water adjusted to pH 9.5 with Na2CO3 was added to promote dispersi on. Sand was collected by sieving, and clay and silt by centrifugation (Whittig & A llardice, 1986). Oriented mounts for clay were prepared for X-ray diffraction (XRD) by depositing 250 mg of clay as a suspension onto a porous ceramic tile under suction. Clay was saturated on the tiles with Mg and solvated with glycerol following an initial XRD scan. Silt was mount ed as an oriented dry powder on a low-backgr ound quartz crystal mount. Manure Sampling Fresh m anure samples (~ 20% solids) were co llected from four dairies at different locations in Florida using 5 gallons polyvinyl buckets. The manure samples were dried at 65oC and then ground to pass a 2mm stainless steel sieve. Manure Characterization The m anure pH, EC, total P and total metals i.e. Ca, Mg, Fe and Al were determined as outlined in the soil characteri zation section.

PAGE 31

16 Mineralogical Analysis Manure sam ples were treated with 10% hypoc hlorite adjusted to pH 9.5 using large plastic beakers. The manure to bleach ratio was 1:50 (manure-amended soil:bleach ratio was 1:20), as manure contains more organic ma terial than the soils. After treatment to reduce organics the sample were wet-sieved (45 M) to separate the sand from silt and clay size fractions. Silt and clay fractions were separated using centrifugation (Whittig & Allardice, 1986). Resulted mate rials i.e. both silt and clay fractions were scanned for mineral characterization as explained in the soil characterization section. QA/QC for Analyses To assure data quality for P and m etal analyses for soil and manure samples proper QA/QC procedures were adopted and include d 1 blank, 1 replicate, 1 spike, and 1 certified standard after every 15 samples run. All extractions were performed on triplicate samples. Statistical Analyses Pair wise comparison by ANOVA of P a nd m etal concentrations of active, abandoned and minimally-impacted soils were done. Mean separations were done by the Waller-Duncan procedure at 5% level of si gnificance. Computations were performed using SAS Institute software (SAS, 2001). Results and Discussion Soil Chemical Characterization Active dairy soils had higher pH values (7.1 7.9) than the pH of abandoned dairy m anure-amended soils (6.0 7.2) (Table 31). Higher pH values in manure-amended soils are due to high inputs of Ca and Mg from dairy manure (Kingery et al., 1994; Nair et al., 1995; Iyamuremye et al., 1996; Eghball, 2002) and the buffering effects of added

PAGE 32

17 bicarbonates, organic acids w ith carboxyl and phenolic hydroxyl groups (Sharpley & Moyer, 2000; Whalen et al., 2000). The mi nimally-impacted soils had significantly ( p < 0.05) lower pH values (3.8 5.7) than manur e-amended soils. Manure-amended soils had significantly greater EC values than minimallyimpacted soils (Table 3-1) attributable to the large amounts of Na, Ca, Mg and other sa lts in the added manure. Average Ca and Mg concentrations for active (Ca = 7086 mg kg-1, Mg = 1292 mg kg-1) and abandoned dairy (Ca = 11082 mg kg-1, Mg = 603 mg kg-1) soils were exceeded than the average Ca and Mg concentrations of minimally-impacted soils (Ca = 301 mg kg-1, Mg = 25 mg kg1). One of the abandoned dairy soils (A BS-1) had an extremely high total Ca concentration (>35,000 mg kg-1), probably as a result of lime added as a fill material to the soil. Total P concentrations were si milar for active and abandoned dairy soils (average ~ 2000 mg kg-1), but much greater than minimally-impacted soils (average ~ 100 mg kg-1). Manure-amended soils and minimally-i mpacted soils had similar total Al and Fe concentrations, which suggested that the long term addition of dairy manure did not change Al and Fe con centrations significantly ( p > 0.05) because both Al and Fe are not a major dietary constituent of dairy animals (NRC, 2001). Soil Mineralogy Quartz was the dom inant mineral in the sand, silt and clay size fractions of all soils except for the one abandoned dairy sample (ABS-1) presumably influenced by fill material (Figures 3-1, and 32), in which calcite (CaCO3) was dominant in the silt and clay. The silt fraction of an active dairy site, also influenced by fill, had high calcite as well. Calcite was detected in all samples a nd was probably derived from either manure or amendments used to stabilize the so il for heavy animal traffic.

PAGE 33

18 Other minerals present in minor to m oderate amounts included kaolinite (two abandoned dairies and one active dairy), sm ectite (one abandoned dairy), and hydroxylinterlayered minerals (three active dairies) No phosphate minerals were directly identified in the samples analyzed via XRD. Pierzynski et al. (1990) and Harris et al. (1994) were also unable to iden tify distinct P minerals in excessively fertilized and dairy manure-impacted soils, respectively. Either th e phosphate phases ar e noncrystalline or mineral concentrations are too low for dete ction (<1%) without further preconcentration (e.g., selective dissolution, density separation, etc.), or both. The presence of a broad amorphous hump on clay XRD pl ots, observed between 16-20 2 (Figure 3-1), suggested the presence of a ppreciable noncrystalline materi al, probably biogenic silica (Harris et al., 1994) derived from plant phyt oliths used as forage to dairy animals. Minimally-impacted soils exhibited the same XRD peaks observed in manure-amended soils, except the presence of hump relate d to amorphous biogenic Si (Figure 3-3). Manure Characterization Manures pH values (Table 3-2) exceeded soil pH (Table 3-1) values, ranging from 8.2 to 8.6. Manure total P concentrati ons ranged from 5965 to 6137 mg kg-1 on a dry weight basis. Kleinman et al. (2005) su rveyed the composition of 68 dairy manure samples and reported average tota l P concentration of 6900 mg kg-1. Manure samples used in this study contained similar averaged (5965 mg kg-1) total P concentrations. The Ca and Mg concentrations ranged from 9899 to 13503 mg kg-1 and 2808 to 4281 mg kg-1, respectively and represent signifi cance of P, Ca and Mg. Both Al and Fe concentrations in manure were minimal. On average, manur e samples contained 77% organic matter on a dry weight basis, mainly from plant phyt oliths used as fodder. Thus, long-term

PAGE 34

19 applications of dairy-manure not only buildup P concentrations in soils, but also result in accumulation of Ca, Mg, and very fine digested plant materials. Manure Mineralogy The XRD pa tterns of clays of four dair y manures from four locations yielded similar results. Irrespective of the manageme nt practices, the same kind of mineralogical components were observed (Figure 3-4). All samples dominated by a hump (16-20, 2 ), which is believed to be biogenic silica derived from plant phytoliths. No P-bearing minerals were detected in the manure samples. Quartz was detected in manure samples, likely due to using sand as bedding material. Calcite was also observed in dairy manures. Summary and Conclusions Dairy m anures contain higher amounts of P, Ca and Mg than dairy manureamended soils. The long term addition of dair y manure significantly altered soil chemical properties and nutrient concentr ations compared to minimally-impacted (native) soils. Both active and abandoned manure-amended soils had higher pH values than native soils, accompanied by higher electrical conductivity values. There was a significant P buildup in manure-amended soils. The native P reten tion capacity of sandy soils is low (Mansell et al., 1991), and is almost certainly exceeded by the levels of P load ing characteristic of high-intensity areas near dairy barns (Nair et al ., 1995 and 1998). Calcium phosphate minerals are predicted to be stable under these conditions using chemical equilibrium modeling (Wang et al., 1995), where as no P-be aring minerals were detected in manureamended soils. Pre-concentrations of clay si ze fractions and minimally alteration of soil samples can be helpful in identifying P-mineral phases associated in the soil.

PAGE 35

20 Table 3-1. Characteristics of active and abandoned manure-amended soils, and minimally manure-impacted soils. Sample ID pH fEC Total P Total Ca Total Mg Total Fe Total Al dS m-1 --------------------------------mg kg-1----------------aACS 1 7.9 0.6117528786629845 3184 ACS 2 7.1 0.683117590921661231 4014 ACS 3 7.5 0.633251845216001161 2691 ACS 4 7.6 0.6212155197774435 311 Mean 7.5ae 0.64 a2334 a7086 a1292a918a 2550a SD 0.3 0.0310071799723363 1590bABS 1 6.8 0.532796355491063869 1860 ABS 2 7.0 0.3419444000584820 1237 ABS 3 6.4 0.3524856708374194 1184 ABS 4 6.0 0.4410013256271358 1528 ABS 5 7.2 0.8222305897722347 308 Mean 6.7a 0.50a2091a11082a603b518a 1223a SD 0.5 0.2068613748312306 578cMIS 1 5.7 0.215852231055 2854 MIS 2 6.6 0.206364241194 136 MIS 3 3.8 0.1119344625153 122 MIS 4 5.0 0.071016215318 1439 Mean 5.3b 0.15b104b301b25c430a 1138adSD 1.2 0.076329211423 1300 aACS = active dairy manure-amended soil bABS = abandoned dairy manure-amended soil cMIS = minimally-impacted soil. dSD= standard deviation eMean values of soil parameters within act ive, abandoned and minimally impacted soils followed by the same letter in a colu mn are not signifi cantly different ( p > 0.05). fElectrical conductivity (EC) 1:2 soil: water ratio

PAGE 36

21 Table 3-2. Characteristics of dairy manures collected from four locations in Florida. Sample pH Organic Mattera ECb Total P Total Ca Total Mg Total Fe Total Al % dS m-1 ---------------------mg kg-1-----------------------Manure-1 8.2 620.926071130522808 203 118 Manure-2 8.5 830.986137113104281 268 667 Manure-3 8.6 840.915670135033888 147 347 Manure-4 8.5 790.87598398993054 125 399 Mean 8.5 77 0.925965119413508 186 383cSD 0.2 10 0.052071658693 64 225 aorganic matter determined by loss on ignition bEC = Electrical conductivity ; 1:2 manure: water ratio cSD = standard deviation

PAGE 37

22 Figure 3-1. X-ray diffraction patterns of clays obtained from active dairy (ACS) manureamended soils. HIV = hydroxyinterlayered vermiculite. Figure 3-2. X-ray diffraction patterns of clays obtained from abandoned dairy (ABS) manure-amended soils.

PAGE 38

23 Figure 3-3. X-ray diffraction patterns of clays obtained fr om minimally-impacted (MIS) soils. Figure 3-4. X-ray diffraction pattern of clay obtained from an oven dried dairy manure.

PAGE 39

24 CHAPTER 4 ASSOCIATED RELEASES OF PHOSPHORUS, CALCIUM AND MAGNESIUM IN SOIL SOL UTIONS FROM DAIRY MANURE-AMENDED SOILS Introduction Continuous release of P from dairy manur e-amended soils enriches adjacent water bodies with P (Graetz & Nair, 1995). Long term addition of manure to soils can alter the chemical and physical characteristics (Nair et al., 1995; Eghball, 2002; Josan et al., 2005; Silveira et al., 2006). Generally, the fate of P in soils is controlled by the inherent soil components; however, excessive manure applica tions can alter the nature and fate of P forms by the dairy manure-derived components. Several studies of P in dairy manureamended soils have been conducted. Nair et al. (1995) studied the forms of P in manureamended soils of south Florida and found that 70% of total P in surface soil was associated with Ca-Mg. Results presented earlie r (Chapter 3) showed that the addition of dairy manure in soils increased total Ca, and to tal Mg concentrations in soils. Therefore, we hypothesized that abandoned and activ e dairy manure-amended soils release comparable amounts of P because solution P is controlled by a sparingly-soluble Mg-P and/or Ca-Mg-P phase that requires many years for deplet ion. The hypothesis was tested in two studies. The first study consisted of repeated water ex tractions (wide soil:solution ratio and minimum soluble salt). The second study was a column leaching experiment, implemented to provide a narrower soil:sol ution ratio as compared to sequential extractions. The narrower soil:solution ratio was expected to more closely approximate

PAGE 40

25 equilibrium conditions and, theref ore, cation and anion concentrations useful to chemical modeling of P speciation in the leachates. Methods and Materials Repeated Water Extractions and Chemical Analyses Soil (20 g) was repeatedly extracted with 200-mL of deionized water (DI) eight tim es. The soil suspensions were initially shaken for 5 min, and repeated for progressively longer intervals (0, 0.5, 3, 6, 12, 24, 36 and 48 h), for a total of 8, in successive extractions. After each extraction, the samples were centrifuged at 738 x g for 5 min. The supernatants were collected by decanting and filtered 0.45 m filter. All extractions were conducted at room temperature (25oC). The collected filtrates were analyzed for pH, electrical conductivity (E C), soluble reactive phosphorus (SRP), and total dissolved Ca, Mg, Na, K, Fe and Al. All P determinations were carried out on a UV-visible recording spectrophotometer at 880 nm wave-length via a molybdate-blu e colorimetric procedure (Murphy & Riley, 1962) (U.S. EPA, 1993; method 365.1). The filtra tes were analyzed for metals by atomic absorption spectroscopy. Total inorganic car bon and total carbon was determined in all the extracted solutions using a carbon analyzer (TOC-5050A, Shimadzu) (method 5310A, 1992). Total organic carbon in the soluti ons was determined by difference. Soil Leaching Characterization a nd Chemical Equilibriu m Modeling Leachates from surface samples of four active and four abandoned dairy-manure impacted soils were collected using a column approach (Figure 4-1). Ten acrylic columns were constructed with a stopper at one end fitted with glass tube and glass wool. Airdried forms of the 10 soils (4 active dair y and 4 abandoned dairy manure-amended soils, including two replications from each soil type) were packed in columns to a height of 30

PAGE 41

26 cm and a density of 1.2 g cm-3. The internal diameter of the columns was 5 cm, so the volume occupied by the soil to a depth of 30 cm was (3.1416 2.5 2.5 30) = 589 cm-3. Therefore the weight of soil in each column was 589 x 1.2 = 707 g. Soil was adjusted to a moisture conten t of 25% (air dry basis) by very slowly adding 283 mL of DI water. This amount was calculated as x 707 g = 0.25 x, where x = final weight after water added sufficient to equal 25% of the fi nal soil weight. This amount of water would constitute about 88 % of the pore volume. Columns were subs equently leached by slowly adding 283 mL of DI water over a 2-hour period, 2.35 mL min-1, and allowed to drain for at least 16 hours (overnight). Leachates were collected in 500-mL beakers, using plastic wrap to keep out dust, and transferred to scintillation vials for chemical analyses. A portion of leachates was transferred to 20 mL plastic vials and kept frozen for backup analyses. Leachates were analyzed for metals (Ca, Mg, Na, K, Fe, Si and Al) by inductively coupled plasma-atomic emissi on spectroscopy (ICP-AES) using EPA method 200.7. Soluble reactive phosphorus (SRP) con centrations were measured using the ascorbic acid colorimetry (U.S. EPA, 1993; method 365.1). Chlorides were determined using EPA method 325.2, nitrates by automated colorimetry with the use of ALPKEM Auto-analyzer (U.S. EPA, 1993; method 353.2), ammonium by the semi-automated colorimetry method (U.S. EPA, 1993; method 350.1), and sulfate by ion chromatography with separator AS-14 (DIONEX) (U.S. EPA, 1993; method 300.0). The pH of leachates was determined and the ionic strength ( ) calculated from EC measurements ( Griffin & Jurinak, 1973): EC 013.0 (1)

PAGE 42

27 Dissolved organic carbon was determined by TOC-5050A, Shimadzu (method 5310A, 1992). Interferences from the inorganic carbon in leachates were first removed by sparging with CO2 -free gas after acidification of the sample (Sharp & Peltzer, 1993). Visual-MINTEQ version 2.51 (Department of Land & Water Resources Engineering, 2004) was used as a chemical e quilibrium model for sp eciation calculations and solubility equilibrium indices for le achates. The model was chosen over other existing models because of its wide applicabil ity in soil science, its windows-based data input (rather than DOS mode), its extensive thermodynamic database for the P species to be modeled, and its ease of addition and m odification of data codes (Mattigod, 1995). The model was set to a charge balance of 30% i.e. the program would be terminated if charge balance exceeded more than 30%. Th e activity corrections were calculated by Davis equation (Davis, 1962) using Davis b parameter 0.3. The equation is given below )3.0 1 ( log2/1 2/1 2 i iAZ (2) where i is the activity coefficient of species i and represents the ratio of the activity of an ion to its concentrations (ci) (Lindsay, 1979), A is a temp erature-dependent constant, with a water value = 0.509 at 25oC, Zi is the valence of the ion i and is the ionic strength define as 22 1iiZc (3) Oversaturated solids were not allowed to precipitate, excluding the infinite solids, finite solids or possible solids. Gaussian mode l for dissolved organic matter (DOM) was selected to take into the account of the complexation of metals by dissolved organic matter. Both pH-dependency and competition among multiple components that bind with

PAGE 43

28 DOM are considered in this model (Dobbs et al ., 1989a,b). This model is easy to use, as it required dissolved organic carbon as input, as compar ed other two metal-humic complexation models available in MINTEQ. Results and Discussion Repeated Water Extractions The EC values of repeated wa ter extractions were greatest ( p <0.01) for active dairy soils (Table 4-1), intermed iate for abandoned dairy soil s, and least for minimallyimpacted soils, with no overlap (Figure 4-2) The trend reflects the salts in manure and partial depletion via leaching upon abandonm ent and cessation of manure loading. Calcium removed by repeated extractions follow ed the same trend as for EC with respect to site groupings (active>abandoned >minimally-impacted), but differe d in that there was no decline with successive extractions for active dairies (Figure 4-3). Active dairy manure-amended soils had more ( p <0.01) Ca in the eighth ex tractions (62.6 to 95.3 mg Ca kg-1 soil) than abandoned dairy so ils (32.9 to 62.0 mg Ca kg-1 soil) (Table 4-1). The release of Mg and P, in contrast to Ca, was similar ( p <0.01) for active and abandoned dairy soils; however, like Ca, there was less ( p <0.01) Mg and P release for the minimallyimpacted soils (Figures 4-4 and 4-5). The data are not consistent w ith the concept of a highly-soluble phase of Mg and P that is depleted with aban donment. Data are consistent with the concept that Mg and P exist mainly in sparingly-soluble form(s), which would require a long time for depletion. The corr elation between Mg and P release was much stronger than the correlation between Ca and P release for both active (Figure 4-6) and abandoned dairy (Figure 4-7) soils, sugges ting a Mg-P phase in the manure-amended soils. The r2 value of Mg with P changed from 0.68 to 0.71 for the active dairies and from 0.62 to 0.75 for the abandoned dairies when the first two extr acts (most heavily

PAGE 44

29 influenced by soluble salts, including short equilibration time) were removed from the regression equation. Thus the P released fr om high intensity area dairy soils during repeated water extractions was more closely a ssociated with Mg than Ca release. Release of Mg and P was similar in both active and aban doned dairy sites. The data are consistent with observations that old abandoned sites rema in a source of P. Release of Ca was less for abandoned dairies than active dairies. Soil Leaching and Chemical Equilibrium Modeling Concentrations of ions column leachates were much greater than in the repeated extractions (mean SRP = 33 mg L-1, Ca = 87 mg L-1, and Mg = 54 mg L-1 in the leachates vs SRP = 4 mg L-1, Ca = 7 mg L-1, and Mg = 3 mg L-1 for repeated water extractions), because of the much narrower soil:solution ra tio in the column study (Detailed results of column leachates are presented in Appendix A) The leachates, in contrast to repeated extractions, showed a negative relationship be tween SRP and Ca and Mg concentrations (Figure 4-8). The trend can be attributed to a common ion effect in the leachates arising from the high concentrations of salts (consiste nt with high EC in the initial leachates). The use of V-MINTEQ speciation model calcu lated the saturation index (SI) values of different P minerals. The index represents rela tive concentrations of ions, and is defined as the difference between the log of the i on activity product (IAP) and the log of the solubility product (Ksp) for a particular solid i.e. Ksp IAPSI log log (4) Both IAP and Ksp are calculated the same wa y, and the only difference is that IAP is ratio of the activity of products to reactants measured in so il solutions, whereas the Ksp is the ratio of activity of products to reactant s that will be present in soil solution at

PAGE 45

30 equilibrium with a specific mineral (Essington, 2003). SI values we re interpreted using the criteria outlined by Bohn & Bohn (1987). Accord ing to this criteria if the SI values are -1 0, the leachate is considered as supersaturated with respect to the P mineral and if the mineral is present, it can precipitate. Chemical modeling (Table 4-2 & 4-3) indicated that all leachates were supersaturated with respect to all but the most soluble Ca-P minerals (monetite and brushite), whereas all leachates were either undersaturated or near saturation with respect to all Mg-P minerals. The data are consistent with the idea of a sparingly-soluble Mg-P and Ca-P phase controlling P release from the manure-amended soils and that the soild phases would maintain elevated P concentra tions in soil solutions even years after abandonment of the dairies. Sharpley et al. (2004) conducted a study on 20 yr old manure-amended silty loam soils. Addition of dairy manure altered the Ca-P chemistry of the soils from hydroxyapaptite to tricalcium phosphate and octacalcium phosphate reaction products. Sharpley et al. (2004) c oncluded that P release from the soils is controlled by these Ca-P forms. However, Sh arpley et al. (2004) did not consider the possible role of Mg in contro lling P release from the soils. Sylveira et al. (2006) used manure-amended sandy soils of the Okeechobee ba sin, in a small column leachate study that last for 36 weeks and resulted in ~ 35 pore volume equivalents of leaching. The researched concluded that both Ca-P and Mg -P minerals control P release from these

PAGE 46

31 soils. Sylveira et al. (2006) recommende d using (Al or Fe-based) water treatment residuals (WTR) to minimize the release of P from the soils. The addition of WTR changes the P release behavior of the soils to adsorption-desorption mechanisms instead of simple dissolution of sparingly solubl e P forms associated with Ca and Mg. Summary and Conclusions During repeated water extractions, rel ease of P from high intensity area dairy manure-amended soils was more closely associat ed with the release of Mg than Ca. In addition, release of Mg and P in repeated water extractions was similar for both active and abandoned dairy manure-amended soil sample s. Active dairy soils released more Ca than abandoned dairy soils and the EC of the extracts was also higher for active dairy soils. Column leachate data suggested that leachates obtained from this study were supersaturated with respect to the most stable Ca-P minerals and were near saturation with respect to soluble Ca-P minerals. The leachates were either undersaturated or near saturation for all Mg-P minerals considered. In effect, there appears to be no significant highly-soluble phase of Mg and P that beco mes depleted with abandonment. Therefore, we concluded that the releas e of P from manure-amended so ils is associated with the release of both Mg and Ca. The solubility of P in sandy soils amended with dairy manure likely controlled by sparingly soluble Mg-P and Ca-P phases, which are expected to continue to release P for a very long time probably several decad es. Abandoned dairies in this study were 12 to 32 years old.

PAGE 47

32 Table 4-1. Cumulative average release of SRP, Ca, Mg and EC in repeated water extractions of active, abandoned and minimally-impacted soils. Soil Type N dEC fSRP Ca Mg (dS m-1) --------------mg kg-1-----------aACS 4 e1.11a495a691a271a bABS 5 0.47b441a477b251a cMIS 4 0.13c0.3b80.1c5.34b aACS = active dairy manureimpacted soil bABS = abandoned dairy manureimpacted soil cMIS = minimallyimpacted soil dEC = electrical conductivity eMean values of soil parameters within act ive, abandoned and minimallyimpacted soils followed by the same letter in a column are not significantly di fferent using WallerDuncan procedure at 5% level of significance ( p 0.05) fSRP = Soluble reactive phosphorus

PAGE 48

33 Table 4-2. Saturation indices (SI) in ac tive and abandoned dairy manure-amended soils. Struvite Farringtonite Newberyite Monetite Brushite Whitlockite OCPc HAPd Sample ID --------------------Mg-P minerals----------------------------------------Ca-P minerals-----------------------------------------------------SI= log IAP log Ksp----------------------------ACSa 1.1b -2.52 -3.54 -2.28 -1.31 -1.59 1.70 -0.73 10.48 ACS 1.3 -1.26 -2.95 -1.65 -0.54 -0.82 2.70 1.05 11.72 ACS 1.7 -1.70 -2.87 -1.36 -0.23 -0.51 2.87 1.52 11.73 ACS 2.1 -0.77 -1.22 -0.99 -0.38 -0.66 2.96 1.47 12.06 ACS 2.3 -0.89 -0.99 -0.51 0.11 -0.17 3.20 2.20 12.07 ACS 2.7 -1.09 -1.19 -0.54 0.15 -0.13 3.22 2.26 12.06 ACS 3.1 -0.43 -0.84 -1.28 -0.82 -1.10 2.89 0.96 12.36 ACS 3.3 -0.09 -0.18 -0.87 0.14 -0.14 3.06 2.08 11.74 ACS 3.7 -0.70 -1.04 -0.31 0.29 0.01 3.09 2.27 11.66 ACS 4.1 -0.21 -0.85 -1.09 -0.36 -0.64 3.67 2.23 13.52 ACS 4.3 -0.28 -1.03 -0.60 0.17 -0.11 3.62 2.68 12.84 ACS 4.7 -0.42 -1.39 -0.61 0.29 0.01 3.64 2.82 12.77 ABS 1.1 -1.52 -3.65 -1.85 -0.42 -0.70 2.95 1.41 12.09 ABS 1.3 -1.94 -3.65 -0.99 0.37 0.09 4.09 3.35 13.58 ABS 1.7 -1.53 -11.95 -1.02 0.40 0.12 3.46 2.74 12.28 ABS 2.1 -0.96 -3.15 -0.96 -0.12 -0.40 3.48 2.24 12.84 ABS 2.3 -0.88 -1.39 -0.37 0.55 0.27 4.00 3.44 13.22 ABS 2.7 -2.07 -1.12 -0.53 0.46 0.18 3.18 2.52 11.66 ABS 3.1 -1.27 -2.12 -1.48 -0.16 -0.45 4.60 3.32 15.12 ABS 3.3 -1.63 -1.68 -0.86 0.54 0.26 3.46 2.88 12.16 ABS 3.7 -2.26 -3.05 -9.10 0.51 0.23 2.77 2.16 10.80 ABS 4.1 -1.60 -3.45 -1.15 0.10 -0.18 2.66 1.62 10.91 ABS 4.3 -1.47 -3.82 -0.67 0.66 0.38 3.06 2.61 11.22 ABS 4.7 -1.93 -3.79 -0.85 0.49 0.21 2.59 1.96 10.45 aACS = active dairy manure-impacted so il; and ABS = abandoned dairy manureimpacted soil b1.1, 1.3, and 1.7 corresponds to 1st, 3rd, and 7th leaching events of soil. cOCP = Octacalcium phosphate {Ca4H(PO4)3.3H2O} dHAP = Hydroxyapatite Ca5(PO4)3(OH) Struvite: NH4MgPO46H2O Farringtonite: {Mg3(PO4)2} Newberyite: (MgHPO4.3H2O) Monetite: (CaHPO4) Brushite: (CaHPO4.2H2O) Ca-whitlockite: {Ca3(PO4)2} ( )

PAGE 49

34 Table 4-3. Percent of observ ations that are undersaturated, saturated, and supersaturated for selected mineralsa based on chemical equilibr ium modeling of active and abandoned dairy column leachates. Soils/Minerals Struvite NH4MgPO4.6H2O Farringtonite Mg3(PO4)2 Newberyite MgHPO4.3H2O Monetite CaHPO4 Brushite CaHPO4.2H2O Undersaturated Active dairies 32 57 32 4 11 Abandoned dairies 89 100 21 0 0 Saturated Active dairies 68 43 68 32 64 Abandoned dairies 11 0 79 11 21 Supersaturated Active dairies 0 0 0 64 25 Abandoned dairies 0 0 0 89 79 aBased on comparison of computed saturation index (SI) from ion activity product (IAP) and solubility product (Ksp) (Bohn & Bohn, 1987).

PAGE 50

35 Figure 4-1. Column set-up used for the column leaching study. 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 12345678 Sequential extractionsEC (dS m-1) ACS-1 ACS-2 ACS-3 ACS-4 ABS-1 ABS-2 ABS-3 ABS-4 ABS-5 MIS-1 MIS-2 MIS-3 MIS-4 Figure 4-2. Changes in elec trical conductivity (EC) (dS m-1) with repeated water extractions. ACS = active dairy manureimpacted soil; ABS = abandoned dairy manureimpacted soil; MIS = minimally impacted soil.

PAGE 51

36 0 20 40 60 80 100 120 140 160 12345678 Sequential extractionsCa (mg kg-1) ACS-1 ACS-2 ACS-3 ACS-4 ABS-1 ABS-2 ABS-3 ABS-4 ABS-5 MIS-1 MIS-2 MIS-3 MIS-4 Figure 4-3. Changes in Ca concentrations (mg kg-1) with repeated water extractions. ACS = active dairy manureamended soil; ABS = abandoned dairy manure amended soil; MIS = minimally impacted soil. 0 10 20 30 40 50 60 70 80 90 100 12345678 Sequential extractionsMg (mg kg-1) ACS-1 ACS-2 ACS-3 ACS-4 ABS-1 ABS-2 ABS-3 ABS-4 ABS-5 MIS-1 MIS-2 MIS-3 MIS-4 Figure 4-4. Changes in Mg concentrations (mg kg-1) with repeated water extractions. ACS = active dairy manureamended soil; ABS = abandoned dairy manure amended soil; MIS = minimally impacted soil.

PAGE 52

37 0 20 40 60 80 100 120 140 16012345678 Extraction numberSRP (mg kg-1) ACS-1 ACS-2 ACS-3 ACS-4 ABS-1 ABS-2 ABS-3 ABS-4 ABS-5 MIS-1 MIS-2 MIS-3 MIS-4 Figure 4-5. Changes in soluble reacti ve phosphorus (SRP) concentrations (mg kg-1) with repeated water extractions. ACS = active dairy manureamended soil; ABS = abandoned dairy manureamended soil; MIS = minimally impacted soil. y = 1.98x 3.81 r2 = 0.68 (Mg) y = -0.19x + 79.4 r2 = 0.01(Ca) 0 20 40 60 80 100 120 140 160 020406080100120140 Ca or Mg (mg kg-1)SRP (mg kg-1) Mg Ca Figure 4-6. Relationships between soluble r eactive phosphorus (SRP) and Mg and Ca released during repeated water extractions of active dairy manureamended soils.

PAGE 53

38 y = 1.19x + 10.83 r2 = 0.62 (Mg) y = 0.45x + 21.0 r2 = 0.20(Ca) 0 20 40 60 80 100 120 140 020406080100120140 Ca or Mg (mg kg-1)SRP (mg kg-1) Mg Ca Figure 4-7. Relationships between soluble r eactive phosphorus (SRP) and Mg and Ca released during repeated water ex tractions of abandoned dairy manure amended soils. SRP = -0.14Ca + 50.67 r2 = 0.27 SRP = -0.19Mg + 48.13 r2 = 0.20 -60 -40 -20 0 20 40 60 80 100 120 140 0100200300400500600700 Ca or Mg (mg L-1)SRP (mg L-1) P and Ca P and Mg Figure 4-8. Relationships between solubl e reactive phosphorus (SRP) and Mg and Ca released during the column leaching of dairy manure-amended soils.

PAGE 54

39 CHAPTER 5 RELATIONSHIPS BETWEEN PHOSPHORUS, CALCIUM AND MAGNESIUM INFERRE D FROM SELECTIVE DISSOLUTION Introduction Repeated water extractions can provide valuable information about the release pattern of different P phases that may exist in manures or manure-amended soils. The use of 8 repeated water extractions to study the re lease of P from the soils was demonstrated in Chapter 4. The data suggested that wate r can extract P indefinitely from the dairy manure-amended soils. Graetz & Nair (1995) used 10 extractions with DI water to extract P from dairy manure-amended soils. From 2 to 18% of total soil P was extracted, and the authors concluded that this labile-P will solubilize in several years. The P concentrations in the 8th repeated water extraction used in this study ranged from 20 mg kg-1 to 80 mg kg-1, and similar concentrations were expected to be released indefinitely. In another experiment, Silvei ra et al. (2006) used 40 repeated water extractions to deplete P concentrations to undetectable levels from dairy manureamended soils. Such experiments are time consuming. Using a higher ionic strength solution, for example 1.0 M ammonium chloride (NH4Cl) (Chang & Juo, 1963; Hieltjes & Lijklema, 1980) or 1.0 M potassium chloride (KCl ) (Reddy et al., 1998), can simultaneously extract both soluble and exchangeable P using fewer number of extractions. In addition to repeated extractions, sequential extraction techniques are used widely to obtain information of various P forms in soils, manures, deposited sediments (Chang &

PAGE 55

40 Juo 1963; Martin et al., 1987; Magnus et al., 1988; He et al., 2005). Sequential extraction procedures use chemical reagents with varyi ng degrees of chemical strengths. The first step in these procedures is to extract or dissolve easily soluble components or forms of nutrients relatively available to plants (Chang & Juo, 1963). Sequential extraction procedures have been used extensively to make inferences about the fate of P. The continuous release of P by se quential extractions with NH4Cl had been demonstrated for dairy manure-amended soils (Graetz & Nair, 199 5; Nair et al., 1995) and for the lake sediments (Petterson & Istvanovics, 1988). Fe wer authors have reported the concomitant release of Ca, Mg, Fe and Al along with P for different kinds of sequential extraction studies (Nair et al., 2003; Sharpl ey et al., 2004; Silveira et al., 2006). Generally, the first two extractions efficiently dissolve CaCO3 during successive extractions by 1.0 M NH4Cl (Hieltjes & Lijklema, 1980). In this study, both repeated and sequential extractions were used to document the release of P, as well as the release of Ca, Mg, Fe and Al. This work is based on a study conducted by Nair et al. (2003). Authors used 1.0 M NH4Cl to study P release characterization of manure and manure-amende d soils. Ammonium chloride was tested for its ability to selectively extract the Mg-P phase first; however, this does not preclude NH4Cl extraction of Ca-P particularly in th e first few extractions. It was hypothesized that the repeated use of NH4Cl can extract P and associated cations preferentially Ca and Mg. Material and Methods Phosphorus Fractionation: Repeated and Sequential Extractions For this study, two soil samples from active dairies (ACS-2 & ACS-4), and two samples from abandoned dairies (ABS-3 and ABS4) were selected (Table 3-1) as these

PAGE 56

41 samples consisted relatively lower amounts of CaCO3. The soil phosphorus (P) fractionation scheme used 1.0 M NH4Cl, 17 h for 0.1 M NaOH, and 24 h for 0.5 M HCl solutions for sequential P extractions (Hieltje s & Lijklema, 1980) and modified by Nair et al. (1995). A 1:10 soil/solution ratio was used for all extrac tions. All extractions were carried out at 298 K. In p ilot study, ten repeated NH4Cl extractions were sufficient to reduce Mg concentrations to undetectable (< 1.0 mg L-1). Thus soil samples were extracted with 1.0 M NH4Cl adjusted to pH 7.0 (using 0.5 M KOH) from 1 to 10 times before being extracted with 0.1 M NaOH, and finally with 0.5 M HCl. For example, a soil sample was extracted once with NH4Cl before being extract ed by NaOH and HCl, and another was extracted twice, another three tim es, another four times and so forth, up to ten times with NH4Cl, before conducting the subsequent extractions with NaOH and HCl (Figure 5-1). The NaOH fraction is expected to extract Fe+Al associated P a nd the HCl fraction, Ca+Mg associated P (Chang & Jackson, 1957; Williams et al., 1967; Hieltjes & Lijklema, 1980; Hedley et al., 1982; Psenner & Pucsko, 1988). Soil samples were shaken end to end in 30 mL polypropylene centrifuge tubes for 2 h for 1 M NH4Cl extractions, 17 h for 0.1 M NaOH and 24 h for 0.5 M HCl extractions. The suspensions were centrifuged for 15 min @ 3620 x g and then filte red (0.45 m) filters to obtain solutions for analyses. Following the sequential extracti ons, soil samples were ashed and digested with 6 M HCl to obtain the residual P fraction (Andersen, 1976). Residual P fraction represents the stable P forms, and is often considered to be organically bound (Hedley et al., 1982). Entrained solution after each ex traction was determined by weight and

PAGE 57

42 corrections were applied during the final calculations of P, and metal concentrations in each fraction. Analyses of P and Metals The 1.0 M NH4Cl, 0.1 M NaOH, and 0.5 M HCl extracts were analyzed for both inorganic-P (Pi) and total dissolved P (T DP). The TDP was determined by potassium persulfate (K2S2O8) digestion method (EPA met hod 365.1, 1993). The potassium persulfate used in this st udy was a certified ACS standard (Fisher Catalog No. P282-500). The TDP of the 1.0 M NH4Cl, 0.1 M NaOH and 0.5 M HCl extracted solutions was determined by adding 1 mL of 5 M H2SO4 + 0.3g of K2S2O8 to 5 mL aliquots of each filtered extract in digestion tubes. Samples were digested by placing digestion tubes on a digestion block at 125-150oC for 2-3 h so that a 0.5 mL of solution remained. The digestion tubes were then covered with gla ss digestion caps and temperature raised to 380oC for 3-4 h. Digested samples were cool ed, diluted with 10 mL of DI water and vortexed to ensure through mixi ng. The solutions were then stored in 20 mL scintillation plastic vials at room temperatur e for analyses. A series of P standards were also digested to obtain similar matrix effects for colorimetric determinations. Proper QA/QC protocol was adopted and included a blank, a duplicat e, a certified QC after every 15 samples during each digestion run. Inorganic-P was determined from solutions obtained by filtering (0.45 M) undigested centrifuged supernatant. Phosphor us (both TDP and Pi) was analyzed using ascorbic acid colorimetery (Murphy & R iley, 1962) (U.S. EPA, 1993; method 365.1). The difference between TDP and Pi yielded the organic-P pool (Po). Concentrations of Ca, Mg, Fe, and Al were determined in bot h digested and undigested samples by atomic absorption spectrophotometry.

PAGE 58

43 The residual fraction obtained after the 0.5 M HCl extraction was transferred to labeled glass beakers an d was digested with 6 M HCl after heating in a muffle furnace for 2 h at 623 K and then raising the temperature to 823 K for 3 h (Anderson, 1974). Both P and metals including Ca, Mg, Fe and Al were an alyzed in the residual fraction of digested samples. To assure data quality for P and metal analyses proper QA/QC procedures were adopted and included 1 blank, 1 replicate, 1 spike, and 1 certified P standard after every 15 samples run. For atomic absorption analys es 1 blank, 1 replicate, 1 spike and 1 standard were analyzed after every 15 samp les. The standard curve for each metal was calibrated by resloping after 15 samples. Results and Discussion Release of P, Ca, and Mg in Repeated 1.0 M NH4Cl Extractions The use of 1.0 M NH4Cl extracted more P, Ca and Mg compared to repeated water extractions. The release of P fr om active dairy soils declined more sharply with repeated NH4Cl extractions than the abandoned dairy ma nure-amended soils, and then leveled off (Figure 5-2). For active dairy soils the first th ree extractions released an average of 65% of total summed P obtained in 10 repeated NH4Cl extractions. In abandoned dairy soils, an average of 40% of total summed P was released. The data suggested that P in abandoned dairy soils became stabilized with time and had leached from soils following abandonment. The soils released comparab le amounts of P using repeated water extractions and P concentrations did not show a declining trend (Figure 4-1). For repeated NH4Cl extractions active dairy so ils behaved differently than the abandoned dairy soils. The NH4Cl extractable P has also been recogniz ed as labile or loosely bound-P (Chang & Juo, 1963; Hieltjes & Lijklema 1980; Graetz & Nair, 1995; Nair et al., 1995; Nair et al.,

PAGE 59

44 2005). During repeated NH4Cl extractions, both Ca and Mg concentrations declined for all soils. ACS-2 soil (Figure 53) released signif icantly greater amounts of Ca compared to the other three soils. Release of greater Ca concentrations in ACS-2 soil sample after 10 repeated NH4Cl extractions was attributed to the presence of high amounts of CaCO3 in the soil (confirmed by X -ray diffraction Figure 3-2), which was dissolved by 1.0 M NH4Cl (Hieltjes & Lijk lema, 1980). A soil with high amounts of CaCO3 is expected to release Ca for long periods. The Mg concentra tions in the first two repeated extractions declined sharply and then leveled off after five extractions for all soils (Figure 5-4). The use of NH4Cl exhaustively extracted both Mg and P in the initial repeated extractions of active dairy soils, which sugge sts a probable sparingly soluble Mg-P phase. Omitting data for the 1st extraction, the remaining data were subjected to regression analyses both for the active and abandoned dairy soils (Table 5-1). Data suggest that the release of P was a linear function of Ca+Mg release for the active dairy soils, whereas the release of P is a logarithmic function of Ca+Mg release for the abandoned dairy soils. The correlation coefficients fo r soils ACS-2 and ACS-4 were 0.946 and 0.996, respectively, showing there was little unexplained variance in the regression. The release of both Ca and Mg with each other were also significantly correlated (r = 0.855 for ACS-2 and r = 0.989 fo r ACS-4), consistent with a specific pool of Ca-Mg-P phase(s) in the soils, responsible for a continuous P re lease. The suspected mineral phase could be amorphous whitlockite (a partially Mg substituted calcium orthophosphate). The presence of high amount of DOC can impede the crystal growth of whitlockite, acting as a buffer for maintaining this sparingly soluble phase in amorphous form (Inskeep & Silvertooth, 1988).

PAGE 60

45 Phosphorus Concentrations in Sequential Extractions Forms of P extracted by 1.0 M NH4Cl (Pi and Po), 0.1 M NaOH (Pi and Po), 0.5 M HCl (Pi and Po), and in the residual P pool for the two active (Figures 5-5 and 5-6) and two abandoned dairies (Figures 5-7 and 5-8) are depicted. Both NH4Cl-Pi and NH4Cl-Po forms increased for all soils with repeated NH4Cl extractions. The NH4Cl-Pi was significantly greater than the NH4Cl-Po in all soils, and Pi fraction dominated all extractions. Nair et al. (1995); Gale et al. (2000); Sharpley et al. (2004); and Silveira et al. (2006) reported similar dominance of inor ganic-P over organic-P in soils following longterm amendments. The cumulative increase of P with repeated NH4Cl extractions can be attributed to the loosely bound P forms in dairy manure-amended soils, and suggests that one or two NH4Cl extractions would not be e nough to study P behavior of these soils. The percentage release of P from ACS-2 soil with repeated NH4Cl extractions was significantly smaller than from the ACS-4 soil. For example the percentage NH4Cl-Pi released for ACS-2 soil (Figure 5-5) during 1st, 5th and 10th extractions was 9%, 20%, and 25%, respectively, whereas for ACS-4 soil (F igure 5-6) the release was 25%, 64%, and 75%, respectively. The relative low P release in NH4Cl repeated extractions for ACS-2 soil can be attributed to the resorption of P by the presence of relatively high amounts of CaCO3 applied as a fill material to stabilize land for proper dairy operations. Accordingly, the NaOH-Pi fraction for ACS-2 soil (Figure 5-5) exceeded the NaOH-Pi fraction of ACS-4 soil (Figure 5-6). In the abandone d dairy soils the percentage NH4Cl-Pi released for ABS3 soil (Figure 5-7) during the 1st, 5th, and 10th extractions was 6%, 27%, and 39%,

PAGE 61

46 respectively and 8%, 25%, and 44% for ABS4 soil (Figure 5-8). Both abandoned dairy soils released P more gradually than the active dairy soils. Active dairy manure-amended soils (ACS-2 & ACS-4) had less P in NaOH fraction than the abandoned dairy manure-amended so ils (ABS-3 & ABS-4). Overall, the NaOHPo fraction of abandoned dairy soils was gr eater than the NaOH-Po fraction of active dairy soils, which suggested immobilization of P with time or mineralization of organic P to inorganic forms that were extracted by NH4Cl. Nair et al. (1995) reported similar trends. The HCl-P pool appeared to be depleted in both types of dairies (Figure 5-5 to 58) following repeated NH4Cl extractions. Abandoned dairy soils also had significantly greater HCl-Po pools than active dairy soils. This fraction deserves further study with similar manure application histories. He et al. (2006) demonstrat ed that HCl fraction extracted from animal manure contains organic-P fractions in addition to inorganic-P. Our data suggest that with abandonment, P was stabilized not only into NaOH-Po forms but also as HCl-Po forms. The residual fract ion extractable P was si gnificantly correlated with Ca, Al and Fe for active dairy soils. Fo r abandoned dairy soils the extractable P in residual fraction was significantly associated wi th Ca and Al but not with Fe (Table 5-4). The residual fraction generally dominated by recalcitrant organic-P material (Hedley et al., 1982), as observed in this study, and Ca, Fe and Al can be occluded with Po. The associated release of P and Ca, Mg, Fe and Al to NaOH and HCl fractions (Table 5-2 & 5-3) were also determined NaOH reportedly extracts Al-Fe P (Chang & Jackson, 1957; Williams et al., 1967; Hiel tjes & Lijklema, 1980; Psenner et al., 1985; Nair et al., 1995; Toor et al., 2006); however th ere have been no confirmatory analyses of this form in dairy manure-amended soils. Da iry manure contains large amounts of Ca and

PAGE 62

47 Mg, and continuous manure application for y ears can alter the chemical forms of P initially associated with Fe and Al (S harpley et al., 2004). In this study, NaOH extractable P and Ca were significantly co rrelated in both active and abandoned dairy manure-amended soils (Table 5-2) Correlation co efficients of P with metals for various P fractions extracted from active and abandoned dairy soils are given in Appendix B. For these heavily manure-amended soils, the str ong relationship between the P and Ca of NaOH extractions might be an artifact of the left over Ca after incomplete NH4Cl extractions. The HCl-P pool is mainly associated with Ca and Mg, if determined after one NH4Cl extraction followed by NaOH extraction, t hus subject to dissolution with repeated NH4Cl extractions. However, the dissolution of sparingly soluble Ca-P, Mg-P and/or CaMg-P solids with repeated NH4Cl extractions, the release of P in HCl extractions was significantly correlated only with Ca (T able 5-3). While working on dairy manureamended soils Nair et al. (1995 and 2003), authors suggested that Ca-P and Mg-P associations are not stable P forms and can be released upon the ons et of next rainfall event. Sharpley et al. (1996) used Fe-P strips to extract P c ontinuously from manureamended soils and found out that P released more rapidly from manure-amended soils than from the soils with minimal manure impact. Summary and Conclusions The repeated extraction of manure-amended soils with NH4Cl suggested that readily-soluble P is associated with Mg. Af ter depleting the Mg, further P release was highly correlated with Ca. Th e release of P in dairy manure-amended soils, using NH4Cl repeated extractions, is controlled by Ca a nd Mg-P phases, which are sparingly soluble. Active dairy soils release more P in the first two NH4Cl extractions than abandoned dairy soils. Most of P released in NaOH and HCl extraction was associated with Ca

PAGE 63

48 concentrations. Data also suggested that repeated NH4Cl extractions mineralized the HClP pools that may not be available for release with fewer numbers (8) of repeated water extractions. Phosphorus in residual fraction was almost constant during the sequential extractions, and the P appeared to be associated with Ca, Fe and Al. We conclude that the initial high release of P from dairy manure-am ended soils may be due to the presence of Ca-P and Mg-P phase/s, which will take years to deplete. However, with time the release of P from these soils will likely be contro lled by a relatively less soluble Ca-P phase provided there is no further manure application.

PAGE 64

49 Table 5-1. Total dissolved phosphorus (TDP) as a function of Ca+Mg in repeated 1.0 M NH4Cl extractions (Data from 1st extraction omitted). Soil Type Soil ID N Regression Equation r2 aACS-2 9 TDP = 0.107(Ca+Mg) 0.912 0.895 Active ACS-4 9 TDP = 0.246(Ca+Mg) 0.226 0.993 bABS-3 9 TDP = 0.042(Ca+Mg) + 2.3242 0.536 ABS-4 9 TDP = 0.047(Ca+Mg) + 0.831 0.604 ABS-3 9 TDP = 0.624Ln(Ca+Mg) + 1.488 0.703 Abandoned ABS-4 9 TDP = 0.421Ln(Ca+Mg) + 0.523 0.793 aACS, active dairy manureamended soil bABS, abandoned dairy manureamended soil. Regression coefficients were significant at p 0.01 () and p 0.05 () as determined by LSD Table 5-2. Sequential release of total di ssolved phosphorus (TDP) as a function of Ca, Mg, and Fe in 0.1 M NaOH extractions Soil Type N Regression Equation r2 Active 20 TDP = 1.597Ca -2.589Mg + 0.112Al 1.519Fe + 1.796 0.751 Abandoned 20 TDP = 1.068xCa 0.199Mg + 0.091Al 0.276Fe + 8.108 0.644 Regression coefficients were significant at p 0.01 () and p 0.05 () as determined by LSD Table 5-3. Sequential release of total di ssolved phosphorus (TDP) as a function of Ca, Mg, Fe, and Al in 0.5 M HCl extractions Soil Type N Regression Equation r2 Active 2 0 TDP = 0.217Ca + 3.36Mg 0.028Al .120Fe 0.025(CaxMg) + 1.322 0.776 Abandoned 2 0 TDP = 0.752Ca + 0.752Mg 0.116Al + 1.831Fe 0.016(CaxMg) 1.930 0.954 Regression coefficients were significant at p 0.01 () and p 0.05 () as determined by LSD Table 5-4. Sequential release of total di ssolved phosphorus (TDP) as a function of Ca, Mg, Fe, and Al in residual fractions Soil Type N Regression Equation r2 Active 2 0 TDP = 0.181Ca 0.142Mg 0.021Al 0.493Fe 0.08(CaxMg) + 0.920 0.937 Abandoned 2 0 TDP = 0.604Ca + 0.242Mg 0.008Al + 0.572Fe 0.090(CaxMg) 1.930 0.982 Regression coefficients were significant at p 0.01 () and p 0.05 () as determined by LSD

PAGE 65

50 Figure 5-1. Schematic of repeated and seque ntial extraction procedure adapted from Nair et al. (1995). 0 50 100 150 200 250 300 350 400 12345678910 Extraction numberSRP (mg kg-1) ACS-2 ACS-4 ABS-3 ABS-4 Figure 5-2. Release of phosphorus using 1.0 M NH4Cl repeated extractions in active and abandoned dairy manure-amended soils. ACS = active dairy manureamended soil; ABS = abandoned dairy manureamended soil.

PAGE 66

51 0 500 1000 1500 2000 2500 3000 12345678910 Extraction numberCa (mg kg-1) ACS-2 ACS-4 ABS-3 ABS-4 Figure 5-3. Release of calcium using 1.0 M NH4Cl repeated extractions in active and abandoned dairy manure-amended soils. ACS = active dairy manureamended soil; ABS = abandoned dairy manureamended soil. 0 200 400 600 800 1000 12345678910 Extraction numberMg (mg kg-1) ACS-2 ACS-4 ABS-3 ABS-4 Figure 5-4. Release of magnesium using 1.0 M NH4Cl repeated extractions in active and abandoned dairy manure-amended soils. ACS = active dairy manureamended soil; ABS = abandoned dairy manureamended soil.

PAGE 67

52 0% 20% 40% 60% 80% 100% 12345678910 Extraction number Residual-P HCl-Po HCl-Pi NaOH-Po NaOH-Pi NH4Cl-Po NH4Cl-Pi Figure 5-5. Distribu tion of P in 1.0 M NH4Cl, 0.1 M NaOH, 0.5 M HCl, and residual-P fractions for an active dairy soil (ACS-2). 0% 20% 40% 60% 80% 100% 12345678910 Extraction number Residual-P HCl-Po HCl-Pi NaOH-Po NaOH-Pi NH4Cl-Po NH4Cl-Pi Figure 5-6. Distri bution of P in 1.0 M NH4Cl, 0.1 M NaOH, 0.5 M HCl, and residual-P fractions for an active dairy soil (ACS-4). P distribution P distribution

PAGE 68

53 0% 20% 40% 60% 80% 100% 12345678910 Extraction number Residual-P HCl-Po HCl-Pi NaOH-Po NaOH-Pi NH4Cl-Po NH4Cl-Pi Figure 5-7. Distri bution of P in 1.0 M NH4Cl, 0.1 M NaOH, 0.5 M HCl, and residual-P fractions for an abandoned dairy soil (ABS-3). 0% 20% 40% 60% 80% 100% 12345678910 Extraction number Residual-P HCl-Po HCl-Pi NaOH-Po NaOH-Pi NH4Cl-Po NH4Cl-Pi Figure 5-8. Distri bution of P in 1.0 M NH4Cl, 0.1 M NaOH, 0.5 M HCl, and residual-P fractions for an abandoned dairy soil (ABS-4). P distribution P distribution

PAGE 69

54 CHAPTER 6 SOLID STATE ASSESSMENTS: CONFIRMING THE ASSOCIATIONS OF PHOS PHORUS, CALCIUM AND MAGNESIUM Introduction Repeated (water and NH4Cl) and sequential (NH4Cl NaOH HCl) extractions used in previous chapters revealed inform ation about the nature of P forms in dairy manure-amended soils and, hence, form stability under given conditions. Sequential or repeated extractions have inherent limitations, which were outlined by Harris (2002): ) overlap in solubility of various P forms for water or other ex tractants, 2) time dependency, 3) loading rate (e.g., extractant/ soil ratio) dependency, 4) readsorption of P forms by CaCO3, and consequent misallocation to P targets by subsequent extractions (as observed in NaOH extractions in the 5th Chapter), 5) variations in solubility of targeted forms from soil to soil, 6) presence of buffers in some soils that can reduce the effectiveness of an extractant, and 7) potenti al hydrolysis of organic-P forms which can lead to an overestimation of inorganic-P forms. Therefore, associated release of P with Ca and Mg of dairy manure-amended soils does not confirm phase associations. Direct assessments using so lid state analytical tools, such as X-ray diffraction (XRD), energy dispersive spectroscopy (EDS), scanning electron microscopy (SEM), and X-ray absorption near edge spect roscopy (XANES) are necessary to confirm P-solid phases. Pi erzynski et al. (1990a) conducted density separation of clays obtained from excessively fe rtilized (inorganic) so ils to concentrate P for subsequent XRD, SEM, and EDS analyses to quantify P rich particles. Phosphorus

PAGE 70

55 was associated with Al and Si (Pierzyns ki et al., 1990b). Beauch emin et al. (2003) performed sequential extracti ons on soils that received animal manure for 25 years. Various Ca-P minerals dominated P forms in the B-horizon of an acidic loamy soil (pH 5.5) as confirmed by XANES analyses. About 45% of total P was present as octacalcium phosphate (OCP) and 11% of total P as hydrox yapatite (HAP). There has been limited solid phase assessments of P associations with Mg and Ca for dairy manure-amended soils (Cooperband & Good, 2002; He et al., 200 3; Nair et al., 2003; Sharpley et al., 2004). The objective of this work was to confirm the P associations with Ca and Mg suggested in the re peated water and NH4Cl extraction experiments. The findings were also expected to aid in explaining the P sp eciation results observed in a column leaching study (Chapter 4). Four approach es were used to study the asso ciations of P with Ca, Mg, Fe, and Al in dairy manures and dairy manur e-amended soils at the microscopic levels. The first approach involved the use of XR D analyses of clays obtained from dairy manure-amended soils. The second approach sp ecifically dealt with dairy manures, and addressed the hypothesis that a sparingly soluble phase of Ca-Mg-P present in dairy manures can be crystallized to a well define d P crystalline phase when subjected to high temperature (550oC). In the third approach, the clays obtained in the first approach and the ashed dairy manures in second approach were examined under SEM for dot map images and the spatial associations of P, Ca, and Mg were evaluated. In addition to spatial associations, semi-quantitative analyses of metals of P-rich particles were also performed. The fourth approach involved the use of electron probe microanalyses (EPMA) of dry sieved silt+clay frac tions of dairy manur e-amended soils.

PAGE 71

56 Material and Methods Approach I: X-ray Diffraction of Untreated Clays Air-dried samples of dairy manure-ame nded soils (50 g each) and dairy manures (100 g each) were Na-saturated by several washings with 1 M NaCl and centrifuging at 728 x g after each washing to remove supernatant. Excess salt was then removed by rinsing with DI water using the same cen trifugation procedure. The final rinse was determined as the first rinse in which the s upernatant appeared turbid. The Na-saturated samples were wet-sieved to separate the sand from silt and clay. Silt and clay were separated by mixing w ith DI water and subsequent cen trifugation (Whiting & Allardice, 1986). Clay was collected in decanted supe rnatant, until a clear supernatant was achieved. The silt in the bottom of cent rifuge bottle was dried and stored. Clay suspensions were flocculated using 1.0 M NaCl. Clay was transferred to 0.45m filters and washed with 25 mL of DI water to remove salt. The salt-free clay was transferred from the filter to a glass slide using a rubbe r policeman, allowed to dry, and then gently crushed to a powder and stored in glass scintillation vials. Oriented mounts for clay were prep ared for XRD analyses by depositing approximately 250 mg of clay as a suspensi on onto a porous ceramic tile under suction. Tiles were dried in a glass desic cator and then scanned with CuK radiation, with the tube energized at 35 kV and 20 mA current. Scans were conducted at 2o/ minute over a 2 range of 2o to 60o. Minerals were identified following criteria outlined by Whittig & Allardice (1986). Approach II: Ashed and Whole Dairy Manure Analyses Dairy manures collected from different dairies (Chapter 2) were dried and ground to 2mm with an automated grinder. Five gram samples of manure were ashed in triplicate

PAGE 72

57 in a muffle furnace at 550oC for 5 h. The ashed samples were ground using a porcelain agar mortar and then transferred to gl ass vials and capped. Additionally, 2mm-ground whole dairy manures were passed through a 1-mm-mesh stainless steel sieve. The materials (manure and ash) were prepared fo r XRD by gently loading the powder into an XRD mount in manner that minimized preferre d orientation (Harris et al., 1994). Mounts were scanned using the same XRD met hodology as outlined in the approach I. Approach III: SEM Imaging and EDS Analyses For successful SEM and EDS analyses, samples need to be conduc tive in nature, to minimize electron accumulation at the surface of the sample. Such accumulation (charging) prevents optimal resolution. Clays (4 active and 4 abandoned) and ashed manures obtained in the first and second approach respectively were dispersed ultrasonically and mounted on carbon stubs. The samples were then subjected to evaporative carbon coating to enhance conducti vity and minimize charging. Mounts were then subjected to a scanning electron mi croscope (JOEL JSM-6400) equipped with an Oxford model number 6506 EDS system for the X-ray analysis. The SEM images were obtained by operating at 15 kV beam voltage and 60A probe current. Images were taken at a series of magnifications while adjusti ng contrast and brightne ss. Energy dispersive spectroscopy was performed at a beam voltage of 15 kV. This analysis was performed at three different locations on each sample to account for compositional variations. Elemental compositional analyses were perf ormed using the x-ra y analysis and SEM quantitative routine of the Oxfo rd Link ISIS software. The software incorporates atomic number, x-ray absorption, and fluorescence (ZAF factor) correction factors while performing semi-quantitative analysis (Goldstein et al., 2003).

PAGE 73

58 Approach IV: Electron Microprobe Microanal yses of Whole Silt + Clay of Dairy Manure-amended Soils Three active (ACS-2 ACS-4) and two abandoned (ABS 1ABS2) dairy soil samples were subjected to electron microprob e (JEOL Superprobe-733, operating at 15 kV). Silt+clay fractions were obtained by dry sieving soils through 45M sieve using a mechanical shaker for 15 min. One centimeter diameter disks (3 disks for each sample) of sieved material were prepared by subjecti ng these samples to a pressure of 34474 kPa (5000 PSI) using a hydraulic press. Carbon coating was performed on the silt+clay disks as explained in approach III. Microprobe analyses were done on each disk using 100 m beam size to quantify P, Ca, Mg, Al and Si c oncentrations. Standards used to calibrate the microprobe, which were served for QA/QC, were calcite for Ca, apatite for P, dolomite for Mg, and quartz for Si. Additionally Ca P ET crystal (Pentaerythritol) and P,Si,Mg,Na -TAP crystal (Thallium acid Phthalate) were used as monochromators. The technique has analytical accuracy of 1-2% by weight, but can be exploited up to 0.1 1% by carefully extracting characte ristics x-ray peaks a nd considering minimum peak interferences (Goldstein et al., 2003). The relatively large beam size was selected to minimize auto correlations that can arise from local surface imperfections. Results and Discussion Both active and abandoned dairy clays had similar mineralogy (Figure 6-1). X-ray diffractograms of untreated clays (Figure 61) were dominated by quartz, with some kaolinite and traces of hydroxyinterlayered vermicul ite (HIV). No P-bearing mineral was detected. An amorphous hump was observed from 22-26 2 which reflected the presence of biogenic Si, derived from plant phytoliths present in dairy manure. Either the Pbearing minerals were amorphous in nature or XRD was not sensitive enough to detect

PAGE 74

59 crystalline phases due to small amounts. Pierz ynski et al. (1990a) and Harris et al. (1994) used similar approaches and were unable to identify P bearin g minerals in highly fertilized soils and in dairy manure-amende d soils, respectively. Dairy manure samples (<1.0 mm) contained whewellite (CaC2O4.H2O; calcium oxalate), calcite (CaCO3), and quartz as shown by XRD analyses (Figure 6-2). The presence of calcite means that there is a significant amount of Ca not in the form of a soluble salt or associated with P in a discrete phase. Quartz is common in dairy manure of grazing cows, probably due to ingestion of small amounts of surface soil ma terial. In ashed dairy manure samples, both calcite and quartz were present, but no whewellite (Figure 6-3) because the latter decomposed to CO2 and H2O at 823 K. The Ca-Mg-P mineral whitlockite (Figure 6-3) was tentatively identified in ashed samp les from XRD data. Formation could be attributed to the re-c rystallization of semicrystalline/a morphous Ca-P or Mg-P or a CaMg-P phase present in dairy manures. All da iry manures, collected from three locations, showed similar possible P pha se crystallization at 550oC. EDS analyses of the ashed samples showed high intensity peaks of Mg, Ca and P (Figure 6-4), consistent with a MgCa-P phase, which might be amorphous in nature prior to heating, sp aringly soluble, and undetectable by XRD analyses. Spatial associ ations of both Mg-P and Ca-P in manure (Figures 6-5 & 6-6) were also observed in SEM/EDS dot map images and spectra, confirming that Mg-P is a manure component ra ther than a component formed in the soil environment after the applic ation of dairy manures. This finding could be helpful in fo rmulating new diets to reduce manure-P solubility by adjusting Ca:M g ratios in addition to P for dairy animals (Daniel et al., 2006). Many authors advocate a reduction of P in dairy diets to reduce the P solubility

PAGE 75

60 (Wu et al., 2001; Dou et al., 2003; Toor et al., 2005). Howeve r, P interacts within the animal gut with Mg and Ca (Herrera et al., 2006) and by decreasing the Mg to the optimal levels in dairy diets, Mg-P formati on can be minimized and Ca-P interaction can be enhanced (Herrera et al ., 2006). Minimized Mg-P formati on could be beneficial as it would reduce P solubility and high P leachi ng potential (Lindsay, 1979) as all Mg-P minerals are relatively soluble than Ca-P mine rals. The P-rich particles of dairy manure clays, observed under SEM (Figure 6-7), also showed the dominance of Mg, Ca and P as observed in EDS spectrum (Figure 6-8). Ther e were 24 samples of manure were observed under SEM, and 46% of the samples showed the spatial associations of P with Ca and Mg both. Microprobe analyses of s ilt+clay samples showed that active dairy silt+clay samples had greater average concentrations of P (0.38%) and Mg (0.64%) than the abandoned dairy samples (P = 0.31% and Mg = 0.25 %) (Figure 6-9). This parallel decline since abandonment suggests that both Mg and P were depleted from the soil with time, and is consistent with the elements be ing released via the dissolution of a common phase. Release of Mg and P was closely associat ed in repeated water extractions (Figures 4-6 & 4-7), and did not differ between active and abandone d dairies as did Ca, which was released to a lesser extent in abandoned da iries. Active and abandoned dairies had similar Ca concentrations (2.3.5%), which were a bout an order of magnit ude greater than Mg and P concentrations. Both Ca and Mg were significantly (p <0.01) correlated spatially with P in abandoned dairy silt +clay fractions (Figures 6-9c and 6-9d). For active dairy samples, only the Mg and P relations hip was statistically significant ( p <0.01). A possible

PAGE 76

61 explanation for poor correlation of Ca and P in active dairy manur e-amended silt+clay fractions can be due to the presence of CaCO3 applied as a fill material. Summary and Conclusions Solid state techniques confirme d spatial associations of P with both Ca and Mg in dairy manure and manure-amended soils. Results were reasonably consistent with release trends for the elements observed in repeated water (Chapter 4) and NH4Cl extractions (Chapter 5), as well as with what could be concluded about mineral equilibria from chemical speciation of column leachates (C hapter 4). Implications are that Mg is associated with P in manure and manure-amended soils to at least an equivalent extent as Ca. However, specific minerals were not identified and uncertainty remains as to crystallinity of P phases as well as to whether P is associated with Ca and Mg separately or within a common Ca-Mg-P pha se (e.g., whitlockite) that wa s tentatively identified in ashed manure samples.

PAGE 77

62 Figure 6-1. X-ray diffracti on patterns of four active (A CS-1 to ACS-4) and four abandoned (ABS-1 to ABS-4) dairy ma nure-amended untreated clays. HIV = hydroxyinterlayered vermiculite. Figure 6-2. X-ray diffracti on pattern of < 1.0 mm dried dairy manures showing the presence of whewellite, quartz and calcite.

PAGE 78

63 Figure 6-3. X-ray diffraction patterns of oven dried and ashe d dairy manures (3) showing the presence of Mg-Ca Whitlockite (Mg-Ca phosphate). Figure 6-4. Energy dispersi ve spectrum of an ashed dairy manure showing the high intensity peaks of Ca, Mg and P.

PAGE 79

64 Figure 6-5. Energy dispersive dot maps of a dairy manure s howing an association of Mg and P. Figure 6-6. Energy dispersi ve dot maps (scale 20 m) of a dairy manure-amended soil clay showing an association of P and Ca.

PAGE 80

65 Figure 6-7. Dot image of dairy manure showing the spatial associations of Mg, P, and Ca. Figure 6-8. EDS spectrum of a manure P rich particle obtained at 400X magnification showing the dominance of Mg, P, and Ca.

PAGE 81

66 Figure 6-9. Relationships between P, Mg and Ca for active (6-9a & 6-9b) and abandoned dairy (6-9c & 6-9d) manure-amended dry sieved (45m) silt+clay using the electron probe microanalyses (EPMA). **significant at p <0.05.

PAGE 82

67 CHAPTER 7 CALCIUM PHOSPHATE CRYSTALINITY AND DAI RY MANURE COMPONENTS Introduction Phosphate chemistry in dairy manure-am ended sandy soils may be influenced by the presence of Mg, biogenic silica (derived from plant phytoliths), and manure-derived organic matter and its myriad of organic acids (Inskeep & Silvertooth, 1988). Manureamended soils systems are typically in a state of disequilibrium (Silvei ra et al., 2006), and even the high pH accompanied by high Ca conc entrations do not eff ectively stabilize CaP forms (Harris et al., 1994; Josan et al., 2005). Apatite and metastable crystalline Ca phosphates (e.g., octacalcium phosphate OCP, tr icalcium phosphate) ar e not detectible by XRD (at least, not in the bulk clay) even af ter years of manure addition despite high total P concentrations (Harris et al., 1994). Density gradient ce ntrifugation has been used successfully on soil clay fractions (Jaynes & Bigham, 1986; Pierzynski et al., 1990a) to study the different Ca-P minerals formed in the soil environment. Pierzynski et al. (1990a) partitioned clays of heavily fertilized so ils into three density fractions (<2.2, 2.2 to 2.5, and > 2.5 Mg m-3) using non aqueous heavy density liquid containing polyvinylpyrolidone and tetrabromoethane. This study addressed potential inhibitory effects of manure-derived components on Ca-P crystallization and the fate of P in manure-amended soils, particularly soils with minimal native P retention cap acity. Results relate to th e question: Why are abandoned dairy manure-amended soils continue to leach high amounts of P after years of

PAGE 83

68 abandonment under the conditions of high pH and high Ca concentrations? It was hypothesized that: 1. Activities of Mg, Si, and DOC in soil solutions of manure-amended soils are sufficient, jointly or separately, to inhibit crystallization of stable Ca-P forms, thus allowing high P release from these soils. 2. Noncrystalline Si forms (i ncluding biogenic Si in dair y manure) can retain P at circumneutral pH and high Ca activity. Th is hypothesis is based on the idea that Ca can serve as a bridge between the silicate surface and P. The hypotheses were tested by the following objectives: 1. To study the effects of Mg, Si and DOC on Ca-P crystallization using average inorganic species concentr ations found in manure-amended soil leachates. 2. To investigate the effects of low-density clay (<2 Mg m-3) and high-density clay ( 2 Mg m-3) of manure-amended soils on Ca-P crystallization. Material and Methods The study required isolating so il clays, treatment to re move carbonates and organic matter, and separation into two density fractions. Procedures for accomplishing the objectives are given in the following sections. Particle Size Separations of Dairy Manure-amended Soils The method used to separate particle sizes in soils is a slight modification of the method developed by Whittig & Allardice (1986). Air-dried samples (50 g each) were Na-saturated by several washings with 1.0 M NaCl in 250-mL centrifuge bottles. Samples were shaken for 5 min on a reciprocal shaker for each washing. The supernatant was decanted after centrifugation at 738 x g (2000 rp m). Excess Na was rinsed with deionized (DI) water by centrifugation and decantation. The final rinse was determined as the first rinse that appeared turbid. The Na-s aturated sample was wet-sieved (45 m) to separate the sand from silt and clay. Silt and clay were separated by centrifugation (Whitting & Allardice, 1986), using DI water and decanting supernatant as clay, repeated until a clear

PAGE 84

69 supernatant was achieved. Silt collected in the bottom of centrifuge bottle was dried and stored. Clay suspensions we re flocculated using 1.0 M NaCl. Clay was collected from suspensions by 0.45m filtration and washed with at least 25 mL of double DI water to remove excess salt. Salt-free clay was transferre d from the filter to a glass slide using a rubber policeman, allowed to dry, gently crushed to a powder, and stored. Carbonate and Organic Matter Removal from the Resulting Clays To remove carbonates, 2 g of clay was treated with 100-mL of 1 M sodium acetate solution, buffered at pH 5.0 with glacial acetic acid (Anderson, 1963; Jackson, 1985), and heated in a boiling water bath for 30 min with intermittent stirring. The same procedure was repeated twice more after decanting th e clear supernatant. The treated clay was transferred to a 0.45m filter and washed with DI water. The retentate was then subjected to organic matter removal. A 5-mL aliquot of sodium acetate buffer and 10 mL (30%) of hydrogen peroxide (H2O2) were added to the samples in beakers and allowed to stand overnight. The following day the sa mples were placed on a hotplate at 85oC and gently heated for 10 min. The last step was repeated and more H2O2 was added to obtain a light colored solid material. The samples were transferred to 0.45m filters, washed with DI water using suction, dried, and st ored in glass scintilla tion vials for further analyses. Density Separations for Soil Clay Materials A Na polytungstate solution with a density of 2.0 Mg m-3 was prepared and adjusted to pH 7.0. Half-gram samples of treated (carbonates and OM removed) clay were placed in 50-mL centrifuge tubes c ontaining 35-mL of the Na polytungstate solution, shaken on a reciprocal shaker for 30 min, and centrifuged at 3268 x g (4000

PAGE 85

70 rpm) for 2 minutes. The supernatant was decanted, and the dense material was transferred to a labeled container. The obt ained light and dense fractions were again transferred into the centrifuge tubes containing 35-mL of 2.0 Mg m-3 Na polytungstate solution, sonicated using an ultrasonic water bath for 2 min and th en shaken on a reciprocal shaker for half an hour. The material was centrifuged again at 4000 rpm for 2 mi nutes, and the above process was repeated until no further separa tion was evident. The accumulated denseand light-fraction suspensions were filtered through a 0.45 m filter, rinsed 5 times with 10-mL aliquots of DI water, transferred to a glass slide with a rubber policeman, allowed to air-dry, and transferred to labeled scinti llation vials. Mineral ogical analyses were performed on both light and dense fractions prior to use for inc ubation experiments. Preparation of Incubating Solutions Chemically defined solutions mimicking the average inorganic chemical composition (Table 7-1) in the leachates of manure-amended soils (Chapter 4) were prepared in double deionized (DDI) water. Four stock so lutions were prepared and adjusted to a final pH of 6.8 by equilibrating with atmospheric CO2 and by adding 0.1 M HCl or 0.1 M NaOH solutions. At this pH the precipi tate of Ca-P minerals did not take place immediately; however, it wa s expected that Ca-P crysta llization would take place in the control, as the ion activit y product was supersaturated with respect to hydroxyapatite. Solution 1 served as a control (Table 7-2) and contained all the same components except Mg, Si, and DOC. Solution 2 consisted of the control solution plus Mg, but no Si and DOC. Solution 3 consisted of the control solu tion plus Si, but no Mg and DOC. Solutions 1, 2 and 3 were prepared from analytical grade reagents and were filtered through 0.45m filters prior to use. Solution 4 consisted of the control solution plus DOC but no

PAGE 86

71 added Si or Mg. The DOC solution, which served as a base solution for this treatment, was prepared as follows: 100 g of each of th e soil samples were mixed together and treated to remove sa lts and carbonates using 200-mL of the 1.0 M sodium acetate following the methods outlined in the carbonate removal section of this chapter. The sodium acetate treated material was washed 5 times with 1.0 M NaCl solution using a centrifuge and four 250-mL bottles at 738 x g (2000 rpm). The residual NaCl salt was rinsed with DI water via centrifugation until the solutions appeared turbid. The turbid solution was retained in the 250-mL bottles. All the 250-mL bottles were shaken for 16 h on a reciprocal shaker. The suspensions were al lowed to settle for 2 h, and then filtered through a 0.45m filter using filtration jars to collect the DOC solution. The DOC solution was analyzed for Ca, Mg, K, Na, and Si using inductively coupled plasmaatomic emission spectroscopy (ICP-AES) (EPA method 200.7). Ammonium was analyzed by the semi-automated colorime try method (U.S. EPA, 1993; method 350.1); nitrates by automated colorimetry with the use of an ALPKEM Auto-analyzer (U.S. EPA, 1993; method 353.2), and sulfate by ion chromatography with a separator AS-14 (DIONEX) (U.S. EPA, 1993; method 300.0). Dissolved organic carbon was determined by TOC-5050A, Shimadzu (method 5310A, 1992). Interferences from the inorganic carbon were first removed by sparging with CO2 -free gas after acidification of the sample (Sharp & Peltzer, 1993). The DOC solution was later diluted to DOC concentrations representative of co lumn leachate DOC concentrations. Incubation Setup, Monitoring and Solution Analyses The effects of Mg, Si, and DOC on Ca-P crystallization in the presence and absence of solids (clay fractions obtained fr om density separations), were determined

PAGE 87

72 with four replications in the absence of so lids. Three replications included the light-clay fraction ( < 2.0 Mg m-3) and four replications in cluded the dense fraction ( > 2.0 Mg m3). Throughout the discussion of the incubation study, the word solids will refer to the clay-sized material derived from the manureamended soils. 100 mg of solids and 50-mL of incubating solution (Table 7-2) were pl aced in 150-mL nalgene bottles and loosely capped. The containers were stored in the dark and maintained at room temperature (25oC) for 20 weeks. Solutions were monitored on a weekly basis for water levels, and small amounts of DDI water were added when necessary to compensate for evaporation. One mL of supernatant was withdrawn at 1, 3, 5, 10 and 20th week intervals, diluted to 8mL with DDI water, and filtered through a 0.45m filter. The resulting solutions were analyzed for Ca and Mg using atomic absorption spectrosc opy. Inorganic P concentrations were measured using ascorb ic acid colorimetry (U.S. EPA, 1993; method 365.1). Details of incubation treatmen ts are explained in Table 7-2. Solid State Assessments X-ray diffraction analyses Lowand high-density clay fr actions were subjected to x -ray analyses using a sidepacked powder mount and scanned with CuK radiation energized at 35 kV and 20 mA current. Scans were conducted at 2o/ minute over a 2 range of 2o to 60o. The precipitates obtained after 20 weeks of inc ubation were collected on filter paper, spread on a lowbackgound quartz crystal XRD mount, air dried, and scanned for x-ray analyses (Chapter 6). Energy dispersive spectroscopy analyses Energy dispersive spectroscopy (EDS) was performed at a beam voltage of 15 kV. The analysis was performed at three locations on each sample so that compositional

PAGE 88

73 variations could be accounted for accurately. Elemental compositional analyses were performed using the x-ray analysis and SEM quantitative routine of the Link ISIS software. The software incorporates atomic number, x-ray absorp tion, and fluorescence (ZAF factor) correction factors and performs se mi-quantitative analysis (Goldstein et al., 2003). Statistical Analysis To test the differences in solution concen trations of P, Ca, Mg, and DOC after the specified incubation periods, a non-parametr ic test (Kruskal-Wallis) was used ( p <0.05). Computations were performed in Minitab version 14.0 (Minitab, 2004). Results and Discussion Effects of Mg, Si and DOC on Ca-P Crystallization In the absence of solids, median concentrations of P in the equilibrating solutions declined significantly, from 68 mg L-1 to 28 mg L-1, for both the control and the Mg treatment over the 20 weeks of incubation (Figure 7-1). P concentrations were significantly lower ( p <0.05) for the Mg treatment than for the control from the 1st week to the 10th week of incubation; however after the 20th week, there was no significant difference between the P concentrations in th e Mg treatment and the control. The sharp decline in the P concentrations in the 1st through 10th week coincided with the formation of precipitates in both the control and the Mg treatment solutions. The Mg concentrations remained almost constant (168 5 mg L-1) during this incubation study. Mineralogical analyses of the precipitates revealed that the only crystalline Ca-P phase precipitate that formed were brushite (CaHPO4.2H2O) in the presence of Mg (Figure 7-2), and hydroxyapatite (Ca5(PO4)3OH) in the control (Figure 7-3) after the 10th and 20th week of incubation. Nielsen (1984) observed that cation dehydration plays an

PAGE 89

74 important role in the growth rate of crysta ls. Magnesium has a dehydr ation rate that is 3000 times slower than that of a Ca ion, and thus the incorporation of Mg into hydroxyapatite can be negligible (Martin & Brown, 1997). Neither a Ca-deficient nor a Mg incorporated non-crystalline Ca-P phase was observed; however, in the presence of Mg, a simple Ca-P crystalline structure (bru shite) with Ca:P 1:1 was precipitated and detected by XRD, then a more complex stru cture with Ca:P 5:3 (HAP). Corresponding declines in Ca concentrations were also observed for the Mg a nd control treatments (Figure 7-4). The data suggest that Mg promoted the prec ipitation of brushite and as a result the solution P concentrations were significantly lower ( p <0.05) in Mg treatments for the first 5 weeks of incubation (Figure 7-1). In the control, HAP crystallization was delayed, resulting in higher P concentrations until the 5th week of incubation. The concentrations of Mg used in this experiment were ve ry high compared to those used by other researches. This is because most studies on th e formation of Ca-P mi nerals pertain to the fields of dentistry and anal ytical chemistry, and such studies typically use lower concentrations of Mg. Usi ng solutions with low Mg concentrations, which are continuously stirred or agita ted, results in the formation of amorphous Ca-P phases (Ferguson & McCarty, 1971). In this incubati on study, the solutions were made to mimic actual leachate concentrations of dairy-manur e amended soils (which have comparatively higher Mg concentrations), and the solutions we re not stirred; this enabled the formation of a crystalline, yet relatively soluble, Ca-P mineral phase. The Si did not inhibit the formation of HAP. The P and Ca concentrations declined but were similar in both the control and Si treatment solution (F igure 7-5). Silicon

PAGE 90

75 concentrations did not decline significantly (p >0.05) during the incubation study. The role of elemental Si in the formation of HAP has been studied in dentistry by Damn & Cate (1992). Silicon behaved as a heterogeneous nucleation substrate that stabilized the formation of the calcium phosphate nuclei, a nd reduced the developm ent time of stable nuclei (critical nuclei). Hidaka et al. (1993) also studied the role of Si (silicic acid) in the formation of calcium phosphate precipit ates and observed no inhibition of Ca-P formation. However in soil systems, Sh ariatmadari & Mermut (1999) studied the phosphate sorption-desorption behavior of silicate clay-calcite system s at various Mg and Si concentrations (0-15 mg L-1). The researches concluded that addition of Si decreased the P sorption of calcite, and attributed the effect to Ca-Si ion-pair formation. In the DOC treatment, P concentrations in the equilibrating solutions after 20 weeks of incubation were significantly greater ( p <0.05) than in the co ntrol (Figure. 7-6). During incubation, P concentra tions declined from 68 mg L-1 to 58 mg L-1 attributable to the immobilization of P by the DOC. Later, this was confirmed by measuring total dissolved P in stored frozen solution aliquot s that had been taken during the 20 weeks of incubation. Total dissolved P (TDP) values were greater than SRP values, and TDP concentrations declined during the incuba tion period, which can be related to the flocculants observed at the bottom of DOC treatment bottles. Additionally in the DOC treatment, there also was a significant decline in Ca concentrations after 20 weeks of incubati on, which was most likely due to Ca-DOC complexation. This was corroborated by analyz ing the same frozen aliquots taken from the incubating solutions for DOC The decline in DOC concen trations (Figure 7-7) was also correlated with observations of incr easing flocculation of DOC in the solutions

PAGE 91

76 during the incubation period. A ssessments of the solids in the DOC treatments, possibly from the DOC flocculation, did not reveal the formation of Ca-P mineral phases (Figure 7-3). Thus, the presence of DOC completely inhibited the formation of Ca-P mineral phases. Grossl & Inskeep (1991, 1992) doc umented the inhibition of both OCP and tricalcium phosphates precipitati on in presence of organic acids like humic, fulvic, citric, and tannic acids. Inhibition was attributed to the blockage of adsorption sites of CaCO3 by DOC and also stated that presence of DOC favors the formation of brushite, as opposed to more thermodynamically stable Ca-P phosphates. MINTEQ analyses of solutions species af ter 20 weeks of incubation revealed high ionic activity products. The results suggest th at even though the Ca and P concentrations declined significantly in the incubating solutions over the 20 weeks, the solutions remained supersaturated. Therefore, the inhi bition of the Ca-P minerals formation was due to the presence of DOC and not to the decreased Ca and P solution concentrations. Effects of Solids on Ca-P Crystallization The presence of solids resulted in hi gher P concentrations in the incubated solutions than in the absence of solids (Figure 7-8). The decline of P concentrations during the incubation in the pr esence of solids might have be en due to the P adsorption onto the clay; no mineral phases of Ca-P were detected by XRD in any of the treatments containing the solids. Phosphorus and Ca c oncentrations in fi nal solutions were significantly lower ( p <0.05) for the low-density clay ( <2.0 Mg m-3) than for the dense fraction ( > 2.0 Mg m-3) (Figure 7-10). The greater so rption of P and Ca by the lowdensity clay may relate to its greater surf ace reactivity arising from either noncrystalline (biogenically-derived) Si or residua l organic matter not removed by H2O2, and is consistent with a Ca-bridging mechanism for P retention. Dominance of Si in low-density

PAGE 92

77 fraction was confirmed by EDS (Figure 7-11), and the presence of nonc rystalline material was indicated by a broad XRD peak (amorphous hump) characteristic of noncrystalline material. Also, SEM imaging of < 50-m mate rial verified the presence of appreciable biogenic Si (Figure 7-12). Summary and Conclusions Representative concentrations of Mg in manure-amended soils inhibit the formation of HAP but not the precipitation of a more sol uble Ca-P mineral (brushite). Brushite that forms locally in a soil matrix would be subject to dissolution in the next rainfall event, thus favoring sustained P l eaching from these soils. In the presence of DOC, Ca-P crystallization was completely inhibited a nd no Ca-P mineral was detected. Si had no inhibitory effect on Ca-P stabilizatio n; and HAP was formed. Lower P and Ca concentrations were observed for low-density clay relative to high-de nsity clay, possibly due to greater solids reactivity (as inferre d from evidence of a higher proportion of noncrystalline material) in conjunction with a Ca bridging sorption mechanism. No Ca-P crystallization took place in th e presence of clay size fractions Generally, solids act as a nucleation seed, adsorbed P thus provides a su rface of Ca-P interaction. Such an effect was not observed in this study.

PAGE 93

78 Table 7-1. Average leachate compositi on of manure-amended soils used for the incubating solutions Chemical Species P Ca2+ Mg2+ Si4+ DOC Fe2+ mg L-1 68.0 312 179 23 427 0.60 Chemical Species Al3+ K+ Na+ NH4 + ClNO3 SO42mg L-1 0.30 375 144 39.0 169 257 345 Table 7-2. Incubation treat ments to study the effects of Mg, Si, and DOC on Ca-P crystallization in the presence a nd absence of manure-derived solids Absence of Solids No. of Replicates Control Treatment 1 Treatment 2 Treatment 3 Solution Ino potential inhibitor Solution II-Mg as the only potential inhibitor Solution III-Si as the on ly potential inhibitor Solution IV-DOC as the only potential inhibitor 4 4 4 4 aPresence of Light Solids Treatment 4 Treatment 5 Treatment 6 Treatment 7 Solution I-light solids as the only potential inhibitors Solution II-Mg and light solids as potential inhibitors Solution III-Si and light solids as potential inhibitors Solution IV-DOC and light solids as potential inhibitors 3 3 3 3 Presence of Dense Solids Treatment 8 Treatment 9 Treatment 10 Treatment 11 Solution I-dense solids as the only potential inhibitor Solution II-Mg and dense solid s as potential inhibitors Solution III-Si and dense solids as potential inhibitors Solution IV-DOC and dense solid s as potential inhibitors 4 4 4 4 aLess number of replications due to low material availability

PAGE 94

79 Figure 7-1. P concentrations in the presence of Mg during 20 weeks (wk) of incubation. Different letters (a and b) indicate statistically sign ificant differences among median concentrations ( p 0.05) observed after an incubation period. HAP = Hydroxyapatite. Figure 7-2. XRD pattern of the precipitat e in the Mg solution after 20 weeks of incubation.

PAGE 95

80 Figure 7-3. XRD patterns of precipitates obser ved after 20 weeks of incubation in control (no inhibitor), Si, and DOC treatments. Figure 7-4. Ca concentrations for cont rol and Mg treatment during 20 weeks of incubation. Different lett ers indicate statistically significant differences ( p 0.05) of median Ca concentratio ns after an incubation period.

PAGE 96

81 Figure 7-5. Variations in P c oncentrations in the presence of Si during 20 weeks of incubation. Different letter s indicate statistically significant differences at p 0.05. HAP=Hydroxyapatite. Figure 7-6. P concentrations in the presence of soil DOC during 20 weeks of incubation. Different letters indicate a significant difference ( p 0.05) after the specified incubation period.

PAGE 97

82 0 100 200 300 400 500 0 h1wk3wk5wk10wk20wkDOC (mg L-1) DOC control DOC solids Figure 7-7. Changes in DOC concentrations in control a nd in the presence of solids during the 20 week (wk) incubation study. Figure 7-8. Effects of clay size fractions on P concentrations during 20 weeks of incubation. Different letter s indicate statistically significant differences at p 0.05 level.

PAGE 98

83 Figure 7-9. Variations in P c oncentrations due to the presen ce of low-density(float) and high-density(sink) clay during 20 weeks of incuba tion. Different letters indicate a significant difference ( p 0.05) after the specifi ed incubation period. Figure 7-10. Variations of Ca concentrations due to the presence of low-density(float) and high-density(sink) clay durin g 20 weeks of incubation. Different letters indicate a significant difference (p 0.05) after the sp ecified incubation period.

PAGE 99

84 Figure 7-11. EDS spectrum of the low-dens ity clay showing the dominance of Si. Figure 7-12. SEM imaging of low-density clay showing the presence of biogenic silica (dumbbell serrated shaped particles in image on left; rod shaped particles in image on right).

PAGE 100

85 CHAPTER 8 SUMMARY AND CONCLUSIONS Concentrated dairies res ult in accumulation of dairy manure on soils, giving rise to risks of nutrient losses that can have deleterious effects on water quality. Phosphorus in intensively-loaded soils can be labile even years or decades after site abandonment, causing leakage at environmentally-unacceptable rates. Barriers to formation of stable crystalline Ca phosphates, which soil solution data suggest would be thermodynamically favored, pose both a scientific mystery and an environmental problem. The first step in pursuit of a solution is to determine effects of critical soil components on P stability. Dairy manure provides the most reactive co mponents for most sandy soils. If the component(s) inhibiting Ca-P crystallization co uld be eliminated or disabled, the result should be more P assimilation and less P loss to surface water via lateral flow to streams, etc. An understanding of reasons for conti nuous release of P from manure-amended soils is critical for adopting nutrient management strategies to reduce the risk of P loss. The overall objective of this di ssertation was to understand th e effects of dairy manurederived components (Mg, Si and DOC) on P re lease and Ca-P crystallization in dairy manure-amended sandy soils. Research encomp assed in this dissertation documented inhibition effects of Mg and DOC on P stabilization. It also provided multiple lines of evidence that P in dairy manure and dairy-manur e-impacted soils is associated with Mg as well as Ca, even though P has been presum ed to exist primarily as Ca-P in dairy manure. The association with Mg is si gnificant because Mg-P and Ca-Mg-P are

PAGE 101

86 sparingly-soluble salts that would not transform to less so luble phases and hence would tend to continue to rel ease P over the long term. Dairy manure-amended soils were categorized into two types: 1) active dairies, operating as dairies at the time of soil sampli ng; and 2) abandoned dairies, not operating for at least 10 years prior to soil sampling. Soil samples were collected from both dairy types and analyzed for basic properties. A dditionally, minimally-impacted soils (never under dairy operations; low P concentrations ) were acquired from previous studies (Chapter 3). Release of P and other ions from the soils was studied using repeated water extractions, column leaching (C hapter 4), and repeated NH4Cl extractions (Chapter 5), coupled with a sequential fractionation procedure. Associated release of P, Mg, and Ca in the extraction assessments was consistent with solid phase associations of these elements that were corroborated by various solid st ate assessments of dairy manure and manureamended soils (Chapter 6). An incubation study was conducted to determine the effects of manure-derived components on Ca-P crystallization (Chapter 7). Manure-amended soils (active and ab andoned) had significantly higher ( p <0.05) pH, and EC values, and concen trations of P, Ca, Mg and DOC than minimally-impacted soils. All soils, manure-amended and minima lly-impacted, were sands and dominated by quartz in all size fractions. Th e clay fractions for all soils contained some secondary phyllosilicates, including kaolinite, hydroxyint erlayered vermiculite, and smectite. Dairy manure samples had high P, Ca, Mg, and DOC concentrations, but small amounts of Fe and Al. Additionally, CaCO3 was present in all manure and manure-amended soils. Longterm addition of dairy manures increased soil pH and EC, and resulted in accumulations of P, Ca, Mg, Si, and DOC.

PAGE 102

87 Release of P, Ca, and Mg from dairy manure-amended soils was evaluated by a repeated water extraction experiment (emp loying a wide soil: so lution ratio) and a column leaching approach (narrow soil:soluti on ratio, to better simulate equilibrium conditions). The P released in repeated wa ter extractions was si gnificantly correlated ( p <0.05) with the release of Mg both in act ive and abandoned dairy soils. Release of Ca and P was not as strongly correlated as was th e release of Mg and P, possibly due to the presence of other Ca compounds (CaCO3) in addition to Ca-P. Column leachates were supersaturated with respect to the most soluble Ca-P minerals and were either undersaturated, or near saturation, for all Mg -P minerals. The data suggest that P is associated with Mg as well as Ca in manure-amended soils. Possible sparingly soluble phases of Ca-P, Mg-P, or Ca-Mg-P (the mi neral whitlockite in amorphous form) control the release of P. Under natura l environmental conditions, P release from sparingly soluble Mg-P or Ca-Mg-P is likely to continue for a long time. Repeated extractions with 1.0 M NH4Cl extracted greater amounts of P than the repeated water extractions. During the first three NH4Cl extractions, active dairy manureamended soils released greater amounts of P than abandoned dairy manure-amended soils. Thereafter, P concentrations stabili zed and active dairy manure-amended soils behaved like the abandoned dairy soil s. The release of P in repeated NH4Cl extractions was significantly correlated with the releas e of both Ca and Mg, again suggesting that sparingly soluble Ca and Mg-P phases c ontrol P release. Repeated use of NH4Cl depleted P from the HCl pool. The release of P in Na OH extractions was correlated to the release of Ca. This relationship is likely due to the residual effect of Ca-associated P that was carried over from the previous NH4Cl extractions. The release of P in HCl extractions

PAGE 103

88 was associated with Ca concentrations. The re sidual fractions in all soils were unaffected by sequential extractions, and residual P appear ed to be associated with Ca, Fe and Al. The collective extraction data are consistent with the hypothesis th at the initial high release of P from dairy manure-amended soils is due to the presence of sparingly soluble Ca-P and Mg-P phase/s, which will take years to deplete. With time (likely over a period of several decades) the release of P from th e soils will be controlled by a relatively less soluble Ca-P phase, provided there is no further manure application. Solid state techniques confirme d spatial associations of P with both Ca and Mg in dairy manure and manure-amended soils. Results were reasonably consistent with release trends for the elements observed in repeated water (Chapter 4) and NH4Cl extractions (Chapter 5) as well as with what could be in ferred about mineral equi libria from chemical speciation of column leachates (C hapter 4). The data suggest th at Mg is associated with P in manure and manure-amended soils to at l east an equivalent extent as Ca. Despite efforts to concentrate minerals using de nsity and particle size fractionations in conjunction with selective disso lution specific minerals were not identified. Therefore, uncertainty remains as to crystallinity of P phases and to whether P is associated with Ca and Mg separately or within a common Ca-Mg-P phase (e.g., whitlockite) that was identified in ashed manure samples. The incubation experiment confirmed that Mg inhibited the formation of HAP while favoring formation of a more solubl e Ca-P mineral (brushite). The possible explanation for Mg inhibition can be ascribed to its sma ller ionic radius (0.044nm) compared with the Ca radius (0.066nm). Th e smaller Mg ions can prevent development of long-range order necessary for the crystallization of HAP. Dissolved organic C and

PAGE 104

89 clay-sized fractions from manure-amended soils also inhibited Ca-P crystallization, while Si had no inhibitory effect. Results of extraction, solid-state, and incubation assessments all implicated Mg as a factor that could reduce the st ability or potential stabiliza tion of P in manure-impacted soils. There are two possible roles of Mg in preventing P stabilization: (i) the higher solubility of P phases that contain Mg and (ii) the inhibiti ng effect of dissolved Mg on formation of potentially stable Ca-P such as hydroxyapatite. Results of this study indicat e that risk of P release from soils (and impact on adjacent water bodies) could be reduced by minimizing Mg concentrations in manure. One option would be to maximize Ca-P interact ion in animal gastro-intestinal tract by minimizing dietary Mg, without adversely impacting animal health. To date, all research has concentrated on reducing TP in the animal diet to reduce the TP in the manure, which has limitations due to dietary P requirements to maintain animal he alth. This study also suggests that Al-based amendments (e.g., Al -stabilized water trea tment residuals) would mitigate P loss more effectively than Ca-based amendments, since the latter would likely experience the same inhibitory effects of Mg as the manure with respect to Ca-P stabilization. However, the presence of high DOC concentrations may decrease the effectiveness of WTR materials; Fe-based amendment could be evaluated as a way to sequester DOC and reduce its inhibition of P stabilization

PAGE 105

90 APPENDIX A DETAILS OF LEACHATE CONCENTR ATI ONS USED FOR V-MINTEQ ANALYSES Table A-1. Column leachate pH, EC and P concentrations of leachates Leaching Events 1 2 3 4 5 6 7 Soil pH aACS-1 8.33 8.54 8.257.978.298.04 8.02 ACS-2 8.07 7.97 8.118.478.288.19 8.08 ACS-3 8.56 8.16 7.947.947.918.16 7.88 ACS-4 8.47 8.18 8.077.737.598.15 8.11 bABS-1 7.83 7.94 7.917.727.67.69 7.56 ABS-2 8.05 7.85 7.807.697.67.68 7.52 ABS-3 8.55 7.51 7.535.27.437.43 7.26 ABS-4 7.34 7.54 7.177.087.067.25 7.17 EC (dS M-1) ACS-1 2.44 0.89 0.610.430.450.43 0.36 ACS-2 4.78 2.79 1.020.670.760.63 0.52 ACS-3 4.25 2.61 1.020.890.930.71 0.72 ACS-4 3.50 3.80 1.231.081.000.75 0.64 ABS-1 4.02 1.01 0.60.580.510.45 0.42 ABS-2 1.69 0.61 0.340.350.360.28 0.27 ABS-3 1.90 1.13 0.470.430.390.33 0.30 ABS-4 3.05 0.98 0.510.340.400.32 0.33 P (mg L-1) ACS-1 0.5 1.9 3.67.07.013.9 7.0 ACS-2 7.0 29.2 61.748.056.645.5 48.0 ACS-3 3.6 27.5 48.089.081.363.4 75.4 ACS-4 7.0 10.4 30.967.758.3128.3 53.1 ABS-1 1.9 10.4 19.020.732.622.4 20.7 ABS-2 8.7 24.1 49.756.649.742.9 44.6 ABS-3 5.3 12.2 39.539.552.346.3 49.7 ABS-4 10.4 34.4 60.860.860.049.7 49.7 aACS = active dairy manure-amended soil bABS = abandoned dairy manure-amended soil

PAGE 106

91 Table A-2. Concentrations of Ca, Mg a nd dissolved organic car bon (DOC) observed in column leachates. Leaching Events 1 2 3 4 5 6 7 Ca (mg L-1) aACS-1 223.1 91.461.552.852.753.9 50.9 ACS-2 205.9 85.437.930.627.624.9 24.4 ACS-3 150.4 87.546.636.728.126.2 26.0 ACS-4 178.9 190.897.372.454.141.7 35.3 bABS-1 654.0 159.2113.7104.5100.299.6 102.1 ABS-2 162.0 87.770.865.360.954.9 52.3 ABS-3 241.8 159.378.166.159.757.0 55.6 ABS-4 302.7 142.993.161.866.360.2 58.8 Mg (mg L-1) ACS-1 142.1 45.430.026.826.527.5 26.5 ACS-2 317.5 99.039.833.032.930.2 30.0 ACS-3 296.7 141.568.252.943.241.0 41.4 ACS-4 164.0 169.765.544.935.227.6 24.3 ABS-1 167.5 44.432.429.826.425.6 27.5 ABS-2 150.2 69.156.653.047.742.7 39.3 ABS-3 81.4 54.120.517.716.015.1 15.1 ABS-4 109.5 53.928.722.520.918.5 18.6 DOC (mg L-1) ACS-1 469 179119806152 40 ACS-2 398 35026516212397 80 ACS-3 497 394295212142105 86 ACS-4 711 818603423299194 139 ABS-1 463 245166150109141 75 ABS-2 251 143104756462 46 ABS-3 170 1661401188476 61 ABS-4 334 2151441218779 70 aACS = active dairy manure-amended soil bABS = abandoned dairy manure-amended soil

PAGE 107

92 Table A-3. Concentrations of K, Fe and Al observed in column leachates Leaching Events 1 2 3 4 5 6 7 K (mg L-1) aACS-1 392 178 122916750 31 ACS-2 745 474 299245219179 153 ACS-3 711 532 383325274247 222 ACS-4 493 585 397303266220 206 bABS-1 211 96 67614839 37 ABS-2 127 72 42271813 10 ABS-3 125 128 87797261 53 ABS-4 100 85 60473828 23 Fe (mg L-1) ACS-1 0.51 0.41 0.690.740.590.12 0.10 ACS-2 0.29 0.51 0.580.280.340.38 0.39 ACS-3 0.24 0.31 0.390.450.410.12 0.08 ACS-4 0.78 0.96 1.581.671.321.04 0.27 ABS-1 0.15 0.17 0.170.250.150.11 0.09 ABS-2 0.08 0.07 0.060.060.030.05 0.06 ABS-3 0.10 0.16 0.250.370.530.61 0.39 ABS-4 0.25 0.25 0.180.120.140.17 0.13 Al (mg L-1) ACS-1 0.38 0.34 0.280.230.160.15 0.12 ACS-2 0.25 0.49 0.540.400.330.23 0.20 ACS-3 0.10 0.16 0.210.170.150.12 0.10 ACS-4 0.25 0.34 0.460.410.290.22 0.15 ABS-1 0.06 0.09 0.090.090.080.09 0.08 ABS-2 0.10 0.10 0.090.100.100.10 0.11 ABS-3 0.12 0.17 0.260.210.220.19 0.18 ABS-4 0.22 0.32 0.350.270.310.27 0.27 aACS = active dairy manure-amended soil bABS = abandoned dairy manure-amended soil

PAGE 108

93 Table A-4. Concentrations of sulfate, ch loride and ammonium observed in column leachates Leaching Events 1 2 3 4 5 6 7 Sulfate (mg L-1) aACS-1 279.3 18.53.31.51.4cnd nd ACS-2 408.9 85.97.83.92.8nd nd ACS-3 476.7 127.011.04.82.7nd nd ACS-4 175.0 250.816.13.91.8nd nd bABS-1 331.9 42.23.91.70.9nd nd ABS-2 82.3 13.13.32.21.4nd nd ABS-3 95.7 52.55.62.01.2nd nd ABS-4 831.7 268.235.22.51.1nd nd Chloride (mg L-1) ACS-1 69.7 11.87.75.23.33.3 5.4 ACS-2 450.6 85.419.212.68.46.7 5.2 ACS-3 324.2 104.224.015.09.97.7 5.8 ACS-4 155.6 294.847.034.921.615.2 10.1 ABS-1 72.0 12.712.75.85.55.4 4.6 ABS-2 65.5 12.317.24.13.83.7 3.6 ABS-3 30.6 23.68.26.14.94.6 4.4 ABS-4 48.5 28.812.68.89.14.7 4.2 Ammonium (mg L-1) ACS-1 43.3 33.024.519.415.19.1 7.4 ACS-2 27.9 10.85.74.04.04.0 4.0 ACS-3 41.6 31.322.89.116.012.6 9.1 ACS-4 51.8 72.338.219.426.222.8 21.1 ABS-1 59.5 41.632.227.124.519.4 14.3 ABS-2 16.0 7.42.32.31.42.3 1.4 ABS-3 9.1 12.68.37.45.75.7 4.0 ABS-4 29.6 22.817.714.312.69.1 9.1 aACS = active dairy manure-amended soil bABS = abandoned dairy manure-amended soil cnd = not detectable

PAGE 109

94 Table A-5. Concentrations of nitrates, and silcic acid (H4SiO4) observed in column leachates Leaching Events 1 2 3 4 5 6 7 Nitrates (mg L-1) aACS-1 3.7 2.50.70.16.50.5 nd ACS-2 457.4 82.0cnd0.2nd nd nd ACS-3 190.1 59.30.00.1nd 0.2 0.5 ACS-4 25.2 47.90.10.2nd nd nd bABS-1 684.8 4.8ndNd nd 0.1 nd ABS-2 138.9 19.60.2Nd nd nd nd ABS-3 207.1 90.10.3Nd nd nd nd ABS-4 93.5 8.20.3Nd nd nd nd H4SiO4 (mg L-1) ACS-1 33.4 27.324.523.620.721.6 17.0 ACS-2 28.1 23.621.721.320.418.4 16.6 ACS-3 41.1 36.135.832.728.126.0 24.5 ACS-4 37.9 39.340.543.744.442.2 42.9 ABS-1 73.7 75.478.380.774.168.6 68.3 ABS-2 59.8 40.633.429.828.126.5 25.7 ABS-3 59.5 57.048.243.137.835.2 32.7 ABS-4 45.2 37.828.918.021.019.3 18.9 aACS = active dairy manure-amended soil bABS = abandoned dairy manure-amended soil cnd = not detectable

PAGE 110

95 APPENDIX B CORRELATION MATRIX FOR DIFFERENT P FRACTI ONS OF DAIRY MANUREAMENDED SOILS Table B-1. Correlation matrix for different fractions in active dairy manure-amended soils NH4Cl_TDP NH4Cl_CaNH4Cl_MgNH4Cl_DOC NH4Cl_TDP 1.000 NH4Cl_Ca 0.773 1.000 NH4Cl_Mg 0.826 0.855 1.000 NH4Cl_DOC 0.905 0.945 0.931 1.000 NaOH_TDP NaOH_Ca NaOH_Mg NaOH_Fe NaOH_Al NaOH_TDP 1.000 NaOH_Ca 0.812 1.000 NaOH_Mg 0.272 0.546 1.000 NaOH_Fe 0.142 0.245 -0.107 1.000 NaOH_Al 0.424 0.388 0.286 0.331 1 HCl_P HCl_Ca HCl_Al HCl_Mg HCl_Fe HCl_P 1.000 HCl_Ca 0.946 1.000 HCl_Al 0.603 0.581 1.000 HCl_Mg 0.930 0.855 0.505 1.000 HCl_Fe 0.449 0.306 0.578 0.454 1.000 Residual P Ca Al Mg Fe P 1.000 Ca 0.895 1.000 Al 0.723 0.802 1.000 Mg 0.838 0.914 0.950 1.000 Fe 0.890 0.888 0.936 0.972 1.000

PAGE 111

96 Table B-2. Correlation matrix for diffe rent fractions for abandoned dairy manureamended soils NH4Cl_TDP NH4Cl_Ca NH4Cl_Mg NH4Cl_DOC NH4Cl_TDP 1.000 NH4Cl_Ca 0.502 1.000 NH4Cl_Mg 0.420 0.989 1.000 NH4Cl_DOC 0.745 0.924 0.892 1.000 NaOH_TDP NaOH_Ca Na OH_Mg NaOH_Fe NaOH_Al NaOH_TDP 1.000 NaOH_Ca 0.781 1.000 NaOH_Mg 0.344 0.423 1.000 NaOH_Fe -0.157 -0.243 -0.102 1.000 NaOH_Al 0.309 0.178 0.232 0.289 1.000 HCl_P HCl_Ca HCl_Al HCl_Mg HCl_Fe HCl_P 1.000 HCl_Ca 0.840 1.000 HCl_Al -0.058 -0.033 1.000 HCl_Mg 0.618 0.880 0.118 1.000 HCl_Fe -0.737 -0.316 0.320 0.001 1.000 Residual P Ca Al Mg Fe P 1.000 Ca 0.962 1.000 Al -0.468 -0.379 1.000 Mg -0.079 -0.023 -0.039 1.000 Fe 0.955 0.906 -0.499 0.056 1.000

PAGE 112

97 LIST OF REFERENCES Abbona F., and M. Franchini-Angela. 1990. Crystallization of calcium and magnesium phosphates from solutions of low con centration. J. Cry. Gro. 104:661-671. Allison, J.D., D.S. Brown, and K.J. Novo-Gradac. 1990. MINTEQA2/PRODEFA2, a geochemical assessment model for environmen tal systems: Ver. 3.0 users manual. Environ. Res. Lab., USEPA, Athens, GA. Amonettee, J.E. 2002. Methods for determin ation of mineralogy and environmental availability. pp. 153-192. In J.E. Amonette. W.F. Bleam. D.G. Schule. and J.B. Dixon (eds) J.B. Dixon and D.G. Schu ltze (co-eds) Soil Mineralogy with Environmental Applications. Soil Sci. Soc. Amer. Madison, WI. Anderson, J.M. 1974. An ignition method for de termination of total phosphorus in lake sediments. Water Res. 10:329-333. Anderson, J.U. 1963. An improved pretreatment for mineralogical analysis of samples containing organic matter. Cl ays Clay Min. 12: 380-388. Beauchemin, S., D. Hesterberg, J. Chou, M. Beauchemin, R.R. Simard, and D.E. Sayers. 2003. Speciation of phosphorus in phosphor us-enriched agricultu ral soils using xray absorption near-edge structure spect roscopy and chemical fractionation. J Environ. Qual. 32:1809-1819. Beede, D. K., and J. K. Shearer. 1991. Nutrit ional management of da iry cattle during hot weather. Agri-Practice 12:5-7. Bigi A., G. Falini, E. Foresti, A. Ripamonti, M. Gazzano, and N. Roveri. 1993. Magnesium influence on hydroxyapatite crystallization. J. Inorg. Biochem. 49:6978. Bohn, H.L., and R.K. Bohn. 1987. Solid activit ies of trace elements in soils. Soil Sci. 143:398. Brennan, E.W., and W.L. Lindsay. 1998. Reducti on and oxidation effect on the solubility and transformation of iron oxides. Soil Sci. Soc. Am. J. 62:930-937. Burgoa, B. 1991. Distribution of residual soil phosphorus along transe cts for three dairies in Okeechobee Co. Florida Soil Crop. Sci. Soc. Am. Proc. 50:137-144.

PAGE 113

98 Cerosaletti, P.E., D.G. Fox, and L.E. Chase. 2004. Phosphorus reduction through precision feeding of dairy cattl e. J. Dairy Sci. 87:2314-2323. Chang, S. C., and M. L. Jackson, 1957. Fractionation of soil p hosphorus. Soil Sci. 84: 133-144. Chang, S.C., and S.R. Juo. 1963. Available phosphorus in relation to forms of phosphates in soils. Soil Sci. 95:91-96. Copperband R.L., and L.W. Good. 2002. Biogenic phosphate minerals in manure: Implications for phosphorus loss to surf ace waters. Environ. Sci. Technol. 36:50755082. Hererra, D., W.G. Harris, C. Staples, and V.D. Nair. 2006. Dietary control of magnesium and calcium to minimize phosphorus releas e from dairy manure. ASA/SSSA/CSSA Annual Meeting, November 2006, Indi anapolis, IN. CD Publications. Damen, J. J., and J. M. Ten Cate. 1992. Silica-induced precipitation of calcium phosphate in the presence of inhibitors of hydroxyapa tite formation. J. Dent. Res. 71:453-457. DeLaune, P. B., B. E. Haggard, T. C. Daniel and I. Chaubey. 2006. The Eucha/Spavinaw phosphorus index: A court mandated index for litter management J. Soil Water Conserv. 61:96-105. Department of Land and Water Resources E ngineering. 2006. Visu al MINTEQ Version 2.51 [Online]. Available at www.lwr.kth.se/English/OurSoftware/vminteq/ (verified 22 Feb. 2007). Dep. of Land and W ater Resour. Eng., Stockholm. Dobbs, J.C., W. Susetyo, F.E. Knight, M.A. Castles, L.A. Carreira, and L.V. Azarraga. 1989a. Characterization of metal binding si tes in fulvic acids by lanthanide ion probe spectroscopy. Anal. Chem. 61:483-488. Dobbs, J.C., W. Susetyo, L.A. Carreira, a nd L.V. Azarraga. 1989b. Competitive binding of protons and metal ions in humic substances by lanthanide ion probe spectroscopy. Anal. Chem. 61:1519-1524. Dooley, D. O., V.D. Nair, and D. A. Graet z. 1993. Phosphorus release from beef and dairy manures and manure-contaminated soils. p. 29. In Agronomy abstracts. ASA, Madison, WI. Dou, Z., J.D. Ferguson, J. Fiorini, J.D. Toth, S.M. Alexander, L.E. Chase, C.M. Ryan, K.F. Knowlton, R.A. Kohn, A.B. Pe terson, J.T. Sims, and Z. Wu. 2003. Phosphorus feeding levels and critical control points on dairy farms. J. Dairy Sci. 86:3787-3795. Eghball, B. 2002. Soil properties as influe nced by phosphorusand nitrogen-based manure and compost applications. Agron. J. 94:128-135.

PAGE 114

99 Essington, M., E. 2004. Minera l solubility. p. 255-308. In Soil and Water Chemistry. CRC Press, Boca Raton. FL. Ferguson, F.J., and P.L. McCarty. 1971. E ffects of carbonates and magnesium on calcium phosphate precipitation. Environ. Sci. Tech. 5:534-540. Flemy, A.R., D.C. Grivin, and E.A. Jenne. 1984. MINTEQ.-A computer program for calculating aqueous equlibria. NTIS PB 84157148. Gale, P.M., M.D. Mullen, C. Cieslik, D.D. Tyler, B.N. Duck, M. Kirchner, and J. McClure. 2000. Phosphorus distribution a nd availability in response to dairy manure applications. Commun. So il Sci. Plant Anal. 31:553-565. Gillman, G.P. 1973. Studies on some deep sand y soils in Cape York Peninsula, North Queensland: 3. Losses of applied phosphorus and sulphur. Aust. J. Exp. Agric. Animal Husb. 13:418-422. Goldstein, J., Newbury, D.E., Joy, D.C., Ly man, C.E., Echlin, P., Lifshin, E., Sawyer, L.C. and Michael, J.R. 2003. Advanced Scanning Electron Microscopy and X-Ray Microanalysis. Berlin, Springer. Graetz, D.A., and V.D. Nair. 1995. Fate of phosphorus in Florida spodosols contaminated with cattle manure. Ecol. Eng. 5:163-181. Graetz, D.A., V.D. Nair, K.M. Portier, a nd R.L. Voss. 1999. Phosphorus accumulation in manure-impacted Spodosols of Florida. Agr. Ecosyst. Enviorn. 75:31-40. Griffin, R.A., and J.J. Jurinak. 1973. Estima tion of activity coefficients from electrical conductivity of natural aqua tic systems and soil extracts. Soil Sci. 116:26-30. Grossl, P.R., and Inskeep, W.P. 1991. Precip itation of dicalcium phosphate dihydrate in the presence of organic acids. Soil Sci. Soc. Am. J. 55:670-675. Grossl, P.R., and Inskeep, W.P. 1992. Kinetic s of octacalcium phosphate crystal-growth in the presence of organic acids. Ge ochim. Cosmochim. Ac. 56: 1955-1961. Harris, W.G., 2002. Phosphate minerals in soils and sediments. p. 637-665. In J.E. Amonette. W.F. Bleam. D.G. Schule. a nd J.B. Dixon (eds) J.B. Dixon and D.G. Schultze (co-eds.) Soil Mineralogy with E nvironmental Applications. Soil Sci. Soc. Am. Madison, WI. Harris, W.G., H.D. Wang, and K.R. Reddy. 1994. Dairy manure influence on soil and sediment composition: implication for phos phorus retention. J. Environ. Qual. 23:1071-1081. Harris, W.G., R.D. Rhue, G. Kidder, R.C. L ittell, and R.B. Brown. 1996. P retention as related to morphology of sandy coastal plain soils. Soil Sci. Soc. Am. J. 60:15131521.

PAGE 115

100 He, Z., A.M. Fortuna, Z.N. Senwo, I.A. Tazi song, C.W. Honeycutt, and T.S. Griffin. 2006. Hydrochloric fractions in Hedley fractionation may contain inorganic and organic phosphates. Soil Sci. Soc. Am. J. 70:893-899. He, Z., C. Honeycutt, and T. Griffin. 2003. Comparative investigation of sequentially extracted phosphorus fractions in a sandy loam soil and a swine manure. Commun. Soil Sci. Plant Anal. 34:1729-1742. He, Z., T.S. Griffin, and C.W. Honeycutt. 2004. Phosphor us distribution in dairy manures. J Environ Qual 33:1528-1534. Hedley, M.J., J.W.B. Stewart, and B.S. Chauhan. 1982. Changes in inorganic and organic soil phosphorus fractions induced by cult ivation practices and by laboratory incubations. Soil Sci. Soc. Am. J. 46:970-976. Hidaka, S., Y. Okamoto, and K. Abe. 1993. Possi ble regulatory roles of silicic acid, silica and clay minerals in the formation of cal cium phosphate precipitates. Arch. Oral Biol. 38:405-413. Hieltjes, A.H.M., and L. Lijklema. 1980. Fractionation of inorganic phosphates in calcareous sediments. J. Environ. Qual. 9:405-407. Horn V.H.H., G.L. Newton, R.A. Nordstedt, E.C. French, G. Kidder, D.A. Graetz, and C.F. Chambliss. 2003. Dairy manure manageme nt: strategies for recycling nutrients to recover fertilizer value and avoid environmental pollution. Document No. CIR1016, Animal Science Department, Fl orida Cooperative Extension Service, Institute of Food and Agricultural Scienc es, University of Florida. Original publication date November 1990. Reviewed June 2003. Hutchison, K.J., and Hesterberg, D. 2004. Dissolution of phosphate in a phosphorusenriched ultisol as affected by microbial reduction. J. Enviro n. Qual. 33:1793-1802. Inskeep, W.P., and Silvertooth, J.C. 1988. Inhibition of hydroxyapati te precipitation in the presence of fulvic, humic, and tannic acids. Soil Sci. Soc. Am. J. 52: 941-946. Iyamuremye, F., R.P. Dick, and J. Baham. 1996. Organic amendments and soil phosphorus dynamics: I. Phosphorus chemis try and sorption. Soil Sci. 161:426435. Jackson, M.L. 1985. Soil chemical analysis Advanced course. 2nd ed. Madison, WI. Jager C., T. Welzel, W. M. Zaika, and M. Eppl e. 2006. A solid-stat e NMR investigation of the structure of nanocrystalline hy droxyapatite. Magn. Reson. Chem. 44: 573 580. Jaynes, W.F., and J.M. Bigham. 1986. Con centrations of iron oxides from soil clays by density gradient centrifugation. Soil Sci. Soc. Am. J. 50:1633-1639.

PAGE 116

101 Josan, M.S., V.D. Nair, W. G. Harris, and D. Herrera. 2005. Associated release of magnesium and phosphorus from active a nd abandoned dairy soils. J. Environ. Qual. 34:184-191. Joy, D., A.D. Romig, and J. Goldstein. 1986. Principles of analytical electron microscopy. Plenum Publ. New York. Kim, S., H.-S. Ryu, H. Shin, H.S. Jung, and K.S. Hong. 2005. In situ observation of hydroxyapatite nanocrystal formation from amorphous calcium phosphate in calcium-rich solutions. Mater. Chem. Phys. 91:500-506. Kingery, W.L., C.W. Wood, D.P. Delaney, J.C. Williams, and G.L. Mullins. 1994. Impact of long-term land application of br oiler litter on environm entally related soil properties. J. Environ. Qual. 23:139-147. Kleinman, P. J. A., R. B. Bryant, W. S. Reid, A. N. Sharpley, and D. Pimentel. 2000. Using soil phosphorus behavior to identif y environmental thresholds. Soil Sci. 165:943-950. Kleinman, P.J.A., A.M. Wolf, A.N. Sharpl ey, D.B. Beegle, and L.S. Saporito. 2005. Survey of water-extractable phosphorus in livestock manures. Soil Sci. Soc. Am. J. 69:701-708. Lavkulich, L.M., and J.H. Wiens. 1970. Comp arison of organic matter destruction by hydrogen peroxide and sodium hyrpchlorit e and its effects on selected mineral constituents. Soil Sci. Soc. Am. Proc. 34:755-758. LeGeros R.Z., and J.P. LeGeros. 1984. pp. 350-385. In J.O. Nriagu and P.B. Moore (ed.) Phosphate minerals. Springer, Berlin. LeGeros R.Z., M.A. Miravite, G. Quirol gico, and P. Morales. 1976. p. 187-206. In B. Finlayson and W.C. Thomas, Jr. Proc. (ed.) Colloq. in Renal Lithiasis. University of Florida Press, FL. LeGeros, R.Z., G. Daculsi, R. Kijkowska, and B. Kerebel. 1989. The effects of magnesium on the formation of apat ites and whitlockites. pp. 11-19. In. Y. Itokawa and J. Durlach (eds.) Magnesium in H ealth and Disease. John Libbey, London. Lehmann, J., Z. Lan, C. Hyland, S. Sato, D. Solomon, and Q.M. Ketterings. 2005. Longterm dynamics of phosphorus forms and retention in manure-amended soils. Environ. Sci. Technol. 39:6672-6680. Lindsay, W.L. 1979. Chemical Equilibria in Soils. Wiley-Inte rscience, New York. Lindsay, W.L., P.L.G. Vlek, and S.H. Ch ien. 1989. Phosphate minerals. p. 1089-1130. In J.B. Dixon and S.B. Weed (ed.) Mi nerals in Soil Environments. 2nd ed. SSSA Book Series no 5, SSSA, Madison, WI.

PAGE 117

102 Littledike, E., J. Stuedeman, S. Wilkinson, and R. Horst. 1983. Grass tetany syndrome. p. 173. In J. Fontenot, G. Bunce, K. Webb, and V. Allen (ed.). Role of Magenesium in Animal Nutrition. J ohn Lee Pratt Annual Nutrition program. Blacksburg, VA. Mackay, A.D., J.K. Syers, R.W. Tillman, and P.E.H. Gregg. 1986. A simple model describe the dissolution of phosphate rock in soils. Soil Sci. Soc. Am. J. 50:291296. Magnus, E., and L.f. Stefan. 1988. Phosphorus in interstitial water: methods and dynamics. Hydrobiologia. 170:103-132. Makris, K.C., W.G. Harris, G.A. O'C onnor, and T.A. Obreza. 2004. Phosphorus immobilization in micropores of drinkingwater treatment residuals: Implications for long-term stability. Envi ron. Sci. Tech. 38:6590-6596. Mansell, R.S., S.A. Bloom, and B. Burgoa. 1991. Phosphorus transport with water flow in acid, sandy soils. In J. Bear and M.Y. Corapciogl u (ed.) Transport processes in porous media. Kluwer Academic Pub lishers, Dorecht, the Netherlands. Martin R. I., and P. W. Brown. 1997. The effects of magnesium on hydroxyapatite formation in vitro from CaHPO4 and Ca4(PO4)2O at 37.4C. Calcif. Tissue Int. 60:538. Martin, J.M., P. Nirel, and A.J. Thom as. 1987. Sequential extraction techniques: Promises and problems. Marine Chem. 22:313-341. Mattigod S.V. 1995. Chemical equilibrium and reaction models: Applications and future trends. In. R.H. Loeppert, A.P. Schwab, and S. Goldberg (ed) Chemical equilibrium and reaction models Madison, Wisconsin, USA. Minitab. 2004. Minitab Re lease 14.0, Minitab Inc., State College, PA, USA. Murphy, J., and J.P. Riley. 1962. A modified single solution method for determination of phosphate in natural waters. Anal. Chim. Acta 27:31-36. Nair, V.D., D.A. Graetz, and D.O. Dooley 2003. Phosphorus releas e characteristics of manure and manure-impacted soils. Food Agri. Environ. 1:217-223. Nair, V.D., D.A. Graetz, and K.M. Portier. 1995. Forms of phosphorus in soil profiles from dairies of south Florida. So il Sci. Soc. Am. J. 59:1244-1249. Nair, V.D., D.A. Graetz, and K.R. Reddy. 1998. Dairy manure influences on phosphorus retention capacity of Spodosols. J. Environ. Qual. 27:522-527. National Research Council. 2001. Nutrient re quirements of dairy cattle. 7th ed. National Academy Press, Washington, D.C.

PAGE 118

103 Neller, J.R., D.W. Jones, N. Gammon, and R.B. Forbes. 1951. Leaching of fertilizer phosphorus in acid sandy soils as affected by lime. Florida Agric. Exp. Stn. Circ. S-32. Nielsen A.E. 1984. Electrolyte crystal growth mechanisms. J. Cryst. Growth 67:289. Novak, J.M., K.C. Stone, D.W. Watts, and M.H. Johnson. 2003. Dissolved phosphorus transport during storm and base flow condi tions from an agriculturally intensive southeastern coastal plain wa tershed. T. Asae. 46:1355-1363. OConnor, G.A. and H.A. Elliott, 2000. Co-app lication of biosolids and water treatment residuals. Final Report. Florida Dep. of Environ. Protection, Tallahassee, FL. OConnor, G.A., K.L. Knudtsen, and G.A. Connel. 1986. Phosphorus solubility in sludge-amended calcareous soils. J. Envrion. Qual. 15:308-312. Ozanne, P.G., D.J. Kirton, and T.C. Sh aw. 1961. The loss of phosphorus from sandy soils. Aust. J. Agric. Res. 12:409-423. Pankow, J.F., and J.J. Morgan. 1981. Kinetics fo r the aquatic environments. Environ. Sci. Technol. 15:1155-1313. Petterson, K., and V. Istvanovics. 1988. Sedi ment phosphorus in lake balaton forms and mobility. Ergeb. Limnol. 30:25-41. Pierzynski, G.M., T.J. Logan, S.J. Tarian and J.M. Bigham. 1990a Phosphorus chemistry and mineralogy in excessively-f ertilized soils: quantitati ve analysis of phosphorusrich particles. Soil Sci. Soc. Am. Proc. 54:1576-1583. Pierzynski, G.M., T.J. Logan, S.J. Traina, and J.M. Bigham. 1990b. Phosphorus chemistry and mineralogy in excessively fe rtilized soils: Desc ription of phosphorus rich particles. Soil Sci. Soc. Am. J. 54:1583-1589. Psenner, R., and R. Pucsko. 1988. Phosphorus fractionation: Advantages and limits of the method for the study of sediment P origin s and interactions. Ergeb. Limnol. 30:4349. Reddy, K.R., Y. Wang, W. DeBusk, M. Fisher, and S. Newman. 1998. Forms of soil phosphorus in selected hydrol ogic units of the Florida Everglades. Soil Sci. Soc. Am. J. 62:1134-1147. Sahai, N. 2005. Modeling apatite nucleation in the human body and in the geochemical environment. Am. J. Sci. 305:661-672. SAS Institute. 2001. SAS Version 8.1. SAS Inst., Cary, NC.

PAGE 119

104 Shariatmadari, H., and A.R. Mermut. 1999. Magnesium and silicon-induced phosphate desorption in smectite, palygorskite, and sepiolite-calcite systems. Soil Sci. Soc. Am. J. 63:1167-1173. Sharp, J., and E.T. Peltzer. 1993. Proce dures subgroup report. Mar. Chem. 41:37. Sharpley, A. N., T. C. Daniel, J. T. Sims, and D. H. Pote. 1996. Determining environmentally sound soil phosphorus levels. J. Soil Water Conserv 51:160-166. Sharpley, A.N. 1996. Availability of residual phosphorus in manured soils. Soil Sci. Soc. Am. J. 60:1459-1466. Sharpley, A.N., and B. Moyer. 2000. Phosphor us forms in manure and compost and their release during simulated rainfall J. Environ. Qual. 29:1462. Sharpley, A.N., and S.J. Smith. 1983. Distri bution of phosphorus forms in virgin and cultivated soil and po tential erosion basses. Soil Sci. Soc. Am. J. 47:581-586. Sharpley, A.N., M.J. Hedley, E. Sibbeson, A. Hillbricht-Ilkowska, W.A. House, and L. Ryszkowski. 1995. Phosphorus transfers from terrestrial to aquatic ecosystems. p. 171-200. In H. Tiessen (ed.) Phosphorus in th e global environment. John Wiley and Sons, N.Y. Sharpley, A.N., R.W. McDowell, and P.J. A. Kleinman. 2004. Amounts, forms, and solubility of phosphorus in soils receiving manure. Soil Sci. Soc. Am. J. 68:20482057. Shenker, M., and P.R. Bloom. 2005. Comments on amounts, forms, and solubility of phosphorus in soils receiving manure. Soil Sci. Soc. Am. J. 69:1353-1354. Silveira, M.L., M.K. Miyittah, and G.A. O'Connor. 2006. Phosphorus release from a manure-impacted Spodosol: Eff ects of a water treatment residual. J. Environ. Qual. 35:529-541. Sims, J. T., A. C. Edwards, O. F. Schoumans, and R. R. Simard. 2000. Integrating environmental soil phosphorus testing in to environmentally-b ased agricultural management practices. J. Environ. Qual. 29:60-72. Sims, J.T., R.R. Simard, and B.C. Joern. 1998. Phosphorus loss in agricultural drainage: historical perspective and current re search. J. Environ. Qual. 27:277-293. Sposito, G., and S.V. Mattigod. 1980. GE OCHEM: A computer program for the calculations of chemical equilibria in soil solution and other na tural water systems. Kearney Found. Soil Sci., Univ. of California, Riverside. Toor, G. S., J. D. Peak, J. T. Sims, and D. L. Sparks. 2006. Advances in the characterization of phosphorus in organi c wastes: Environmental and agronomic applications. Adv. Agron. 89:1-72.

PAGE 120

105 Toor, G.S., J.T. Sims, and Z. Dou. 2005. Reducing phosphorus in dairy diets improves farm nutrient balances and decreases th e risk of nonpoint pollution of surface and ground waters. Agri. Eco. Environ. 105:401-411. U.S. Geological Survey. 1999. The quality of our nations waters: Nutrients and pesticides. U.S. Geological Survey Circ. 1225. USGS Information Services, Denver, CO. USEPA. 1993. Methods for the determination of inorganic substances in environmental samples. EPA-600/R-93-100, Cincinnati, OH. USEPA. 1996. Environmental indicators of water quality in the United States. EPA 841R-96. U.S. EPA, Office of Water (4503F), U.S. Gov. Printing Office, Washington, DC. Wang, H.D., W.G. Harris, K.R. Reddy, and E. G. Flaig. 1995. Stability of phosphorus forms in dairy-impacted soils under si mulated leaching. Ecol. Eng. 5:209-227. Weiss, W. P. 2004. Macromineral digestion by lactating dairy cows : Factors affecting digestibility of magnesium. J. Dairy Sci. 87:2167-2171. Whalen, J.K., and C. Chang. 2001. Phosphorus accumulation in cultivated soils from long-term annual applications of cattle feedlot manure. J. Environ. Qual. 30:229237. Whalen, J.K., C. Chang, G.W. Clayton, and J.P. Carefoot. 2000. Cattle manure amendments can increase the pH of aci d soils. Soil Sci. Soc. Am. J. 64:962. Whittig, L.D., and W.R. Allardice. 1986. X-ray diffraction techniques for mineral identification and mineralogical composition. p. 331-362. In A. Klute (ed.) Methods of soil analysis. Soil Sci. Soc. Am. Madison, WI. Williams, J.D.H., J.K. Syers, and T.W. Wa lker. 1967. Fractionation of soil inorganic phosphate by a modification of Chang and Jacksons procedure. Soil Sci. Soc. Proc. 31:736-739. Wu, Z., L.D. Satter, A.J. Blohowiak, R.H. Stauffacher and J.H. Wilson. 2001. Milk production, estimated phosphorus excretion, and bone characteristics of dairy cows fed different amounts of phosphorus for two or three years. J. Dairy Sci. 84:1738 1748. Yadav, B.R., K.V. Paliwal, and N.M. Nimgad e. 1984. Effect of magnesium-rich waters on phosphate adsorption by calc ite. Soil Sci. 1984. 138:153-157. Zhang M., A.K. Alva, Y.C. Li, and D.V. Calvert. 2001. Aluminum and iron fractions affecting phosphorus solubility and reactions in selected sandy soils. Soil Sci. 166:940-948

PAGE 121

106 Zhou M., and Yuncong Li. 2001. Phosphorus-sorp tion characteristics of calcareous soils and limestone from the southern Everglades and adjacent farmlands. Soil Sci. Soc. Am. J. 65:1404-1412.

PAGE 122

107 BIOGRAPHICAL SKETCH Manohardeep Singh Josan was born January 12, 1974 in Am ritsar, Punjab, India. He hails from a farming background and growing up at the farm encouraged him to explore the science in agricult ure. In the quest for knowledge, he received his bachelors degree in Agriculture ( Honors.) specialization in Soil Science from Punjab Agricultural University Ludhiana in 1998. He received hi s masters degree from the same department in 2000. Currently he is working toward his doctorate in soil and water science at the University of Florida.