Citation
Measurement of adhesion between calcium oxalate monohydrate and model surfaces using a dynamic wet cell

Material Information

Title:
Measurement of adhesion between calcium oxalate monohydrate and model surfaces using a dynamic wet cell
Creator:
Habeger, Craig F., 1969-
Publication Date:
Language:
English
Physical Description:
xvii, 192 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Adhesion ( jstor )
Aggregation ( jstor )
Calcium ( jstor )
Crystals ( jstor )
Flow velocity ( jstor )
Microfilms ( jstor )
Oxalates ( jstor )
Particle interactions ( jstor )
Permittivity ( jstor )
Streaming ( jstor )
Dissertations, Academic -- Materials Science and Engineering -- UF ( lcsh )
Materials Science and Engineering thesis, Ph.D ( lcsh )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph.D.)--University of Florida, 1997.
Bibliography:
Includes bibliographical references (leaves 182-191).
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Craig F. Habeger.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright [name of dissertation author]. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
028628258 ( ALEPH )
38863553 ( OCLC )
ALS8237 ( NOTIS )

Downloads

This item has the following downloads:


Full Text











MEASUREMENT OF ADHESION
BETWEEN CALCIUM OXALATE MONOHYDRATE AND MODEL SURFACES
USING A DYNAMIC WET CELL












By


CRAIG F. HABEGER


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


1997































This work is dedicated to my loving and supportive wife,


Nicole Habeger,



my parents,

James and Sandra Habeger,



and



my brother and sister, niece and nephew,

Larrin Habeger and Bonnie, Miranda, and Ralph Isaac Testa.

You are all the greatest!















ACKNOWLEDGMENTS


I would like to thank all of the people who made this research possible. Sincere

thanks are extended to Dr. James H. Adair, my committee chair and advisor, for his

guidance and patience. I would like to recognize the other members of my committee,

Drs. Christopher D. Batich, Anthony B. Brennan, and Brij M. Moudgil, from the

Department of Materials Science and Engineering, and Dr. Raymond L. Hackett, from the

Department of Pathology. All of these gentlemen contributed to the work reported herein

and I am grateful for their advice. I would also like to thank Dr. Saeed R. Khan, from the

Department of Pathology, who was not lucky enough to be on my committee but

graciously gave me advice on my research anyway. I appreciate the advice given to me by

Dr. John J. Mecholsky, Jr., from the Department of Materials Science and Engineering,

and his willingness to substitute for Dr. Brennan during my defense. I would like to

acknowledge the National Institutes of Health grant number POG 5P01 DK20586-17

which supported the work reported herein.

I would like to acknowledge the staff of the Major Analytical Instrumentation

Center (MAIC) for valuable advice, discussions, and training. I would also like to thank

Paula Scott, Pat Glenton, and Karen Byer for all of their assistance throughout my

research.

I am grateful to all of my friends, those in Dr. Adair's research group and those

not in the group. When I broke my leg, you showed me what friends really are. Special

thanks are extended to Drs. Robert E. Chodelka, Malanie L. Carasso, Tuo Li and soon to

be Drs. Jeffrey A. Kerchner, Nelson S. Bell, and Paul A. Demkowicz for their fruitful

intellectual conversations but mostly for their lousy card playing ability.










I am indebted to my wife, Nicole, who nursed me through a broken leg, twice, and

many other aches and pains too numerous to develop upon. She pushed me to finish,

however, at times, not hard enough. I would like to thank my entire family for being so

supportive. Finally, but most importantly, I would like to thank my personal Lord and

Savior, Jesus Christ, without whom none of this would be possible.















TABLE OF CONTENTS


ACKNOWLEDGMENT..................................... ........................................................ii

LIST O F TA B LES...................................................................................................... viii

LIST OF FIGURES ....................................................................................................... x

ABSTRACT .................................................................................................................. ...xvi

CHAPTERS

1 INTRODUCTION .................................................................................................... 1
1.1 Introduction....................................................................................................... 1
1.2 Literature Review.............................................................................................. 1
1.3 Development and Characterization of Model COM Particles................................2
1.4 Evaluation of the Calcium Oxalate Monohydrate Hamaker Constant..................3
1.5 Development of the Dynamic Wet Cell and Streaming Potential
M easurem ents .................................................................................................. 3
1.6 Measurement of COM Adhesion to Macromolecular Substrates .........................4
1.7 Conclusions and Future Research ..................................................................4

2 LITERATURE REVIEW........................................................................................... 5
2.1 Introduction....................................................................................................... 5
2.2 Particle Interactions........................................................................................... 7
2.2.1 Electrostatic Repulsion ....................................... ...................................9
2.2.2 A ttraction............................................................................................... 14
2.2.4 Interaction Energy.................................................................................. 15
2.3 Aggregation Mechanisms Among Particles and Particles at Surfaces ...................17
2.3.1 Particle Aggregation in Simple Electrolytes............................................... 19
2.3.2 Secondary Minimum Coagulation........................ ....................................19
2.3.3 Heterocoagulation.................................................................................20
2.3.4 Polymer Bridging Flocculation..................................................................23
2.3.5 Flocculation of COM Particles by Phospholipids and other
Intercellular Substances ...........................................................................26
2.3.6 Adhesion of Particles at Surfaces............................................................26
2.4 The Human Kidney.........................................................................................31
2.5 Characterization Techniques.............................................................................34









2.5.1 Electrokinetic Measurements.............................................................34
2.5.1.1 Electrophoresis............................................ ................................36
2.5.1.2 Streaming Potential.............................. ............... ......................... 36
2.5.2 Scanning Probe M icroscopy .................................. ....................................41
2.5.3 Scanning Electron Microscopy ................................................................43
2.5.4 Fourier Transform Infrared Spectroscopy ...........................................43
2.5.5 Particle Size Determination.................................................................. 45
2.5.6 X-Ray Diffraction................................................................................48

3 DEVELOPMENT AND CHARACTERIZATION OF MODEL COM
PA R TIC L E S .......................................................................................................... 49
3.1 Introduction.....................................................................................................49
3.2 Preparation and Characterization of Calcium Oxalate Monohydrate
Particles under Different Conditions................................................................54
3.2.1 COM particles ("33" particles) without seeds............................................ 54
3.2.2 COM particles with seeds .................................................................................56
3.2.3 COM particles precipitated from homogeneous solution at 900C .............56
3.2.4 Characterization ..................................................................................... 58
3.2.5 Crystal and Atomic Structure Modeling Using the Computer
Programs SHAPE and ATOMS .........................................................60
3.3 Results and Discussion .................................................... ............................. 62
3.3.1 COM particles ("33" particles) without seeds............................................ 66
3.3.2 COM particles ("33" particles) with seeds .............................................. 69
3.3.3 COM particles precipitated from homogeneous solution at 900C ..............71
3.3.4 Atomic structures of COM particles as a function of habit plane for
the high tem perature form ........................................................................71
3.3.5 Characterization of COM crystals...................................................................79
3.4 C onclusions..................................................................................................... 82

4 EVALUATION OF THE CALCIUM OXALATE MOMOHYDRATE
HAMAKER CONSTANT BASED ON STATIC DIELECTRIC
CONSTANT DETERMINATION AND ELECTRONIC POLARIZATION.........85
4.1 Introduction..................................................................................................... 85
4.2 M materials and M ethods............................................................. .................... 89
4.3 Results and Discussion ............................................................. .................... 95
4.4 C onclusions................................................................................................... 103

5 DEVELOPMENT OF THE DYNAMIC WET CELL AND STREAMING
POTENTIAL MEASUREMENTS...................................................................... 106
5.1 Introduction.............................. ...................................................................... 106
5.2 M materials and M ethods......................................................... ...................... 109
5.2.1 Adhesion Measurements using the Dynamic Wet Cell............................109
5.2.2 Streaming Potential Measurement.............................................................. 116









5.3 Results and D discussion ....................................................................................... 122
5.3.1 Dynam ic W et Cell...................................... ............. ...........................122
5.3.2 M odel CO M Particles..............................................................................125
5.3.3 Streaming Potential.............................................................................127
5.3.4 A dhesion ....................................................................................................13 1
5.3.5 Theoretical Modeling of Interactions..........................................................133
5.4 C conclusions .........................................................................................................138

6 MEASUREMENT OF COM ADHESION TO MACROMOLECULAR
SU B STRA TES...................................................................................................... 140
6.1 Introduction................................................................................................... 140
6.2 Materials and Methods...............................................................................143
6.2.1 Particle Synthesis and Characterization...................................................143
6.2.2 Substrate Coating ....................................................................................... 143
6.2.3 Adhesion Measurements....................................................................145
6.2.4 Streaming Potential Measurements......................................................... 146
6.3 Results and D discussion ....................................................................................... 146
6.3.1 Substrate Coverage................................................................................... 146
6.3.2 Zeta Potential Determinations ...............................................................149
6.3.3 Adhesion Measurements....................................................................153
6.3.4 Hydrodynamic Model of the Human Kidney..........................................167
6.4 C conclusions .........................................................................................................170

7 SUMMARY AND FUTURE WORK........................................................................173
7.1 Sum m ary ....................................................................................................... 173
7.2 Future W ork..................................................................................................175

APPENDICES

A DETERMINATION OF THE AREA OF COM CRYSTALLOGRAPHIC
FACES USING EQUIVALENT SPHERICAL DIAMETER................................177

B DETERMINATION OF THE STRESS IN THE KIDNEY AS A FUNCTION
OF FLOW AND DISTANCE FROM THE TUBULE WALL................................ 179

R E F E R E N C E S ................................................................................................................. 182

BIOGRAPHICAL SKETCH........................................................................................ 192















LIST OF TABLES


Table page

3.1. Composition of artificial urine ion solution............................................................ 61

3.2. Values of the lattice parameters of calcium oxalate monohydrate...........................63

3.3. Atomic coordinates of W hewellite..........................................................................65

3.4. Crystal forms and corresponding central distance values used to generate the
theoretical shape of COM crystals.............................................................................68

3.5. Values of the crystallographic data for the high temperature form of
W hew ellite........................................................................................................ 73

3.6. Atomic coordinates for high temperature form of Whewellite................................75

4.1. Index of refraction for COM as a function of optical direction and wavelength.......88

4.2. The composite mixing rules which were evaluated ........................................ ...90

4.3. Molecular structures of the silane coupling agents used to disperse COM in
Eccosil 5019 silicone. .............................................................. .....................92

4.4. Dielectric constant values of COM determined by fitting the mixing rules to
the experim mental data........................................................................................ 98

4.5. Calculated values of the UV characteristic frequency and corresponding
dielectric constants and refractive indices as a function of crystallographic
direction determined from Cauchy plots..........................................................102

4.6. Comparison of A131 calculated for COM using Gregory's approximation vs.
the Tabor-W interton relationship. ................................................................... 105

5.1. A list of the physical parameters necessary to calculate zeta potential from
stream ing potential ................................. ............................................................ 121

6.1. The values of zeta potential of the macromolecular substrates determined
using streaming potential measurements in saturated COM solution.................. 151









6.2. Values of adhesion strength for COM particles adhering to macromolecular
substrates as a function of crystallographic habit plane in COM saturated
so lu tio n ................................................................................................................. 16 5

6.3. Values of adhesion strength for COM particles adhering to macromolecular
substrates as a function of crystallographic habit plane in AUIS......................166

6.4. Human kidney tubule dimensions and volumetric flow rates reported by Kok
and K han ......................................................................................................169















LIST OF FIGURES


Figure page
2.1. Pathways to kidney stone formation................................................. ................. 6

2.2. COM particle aggregate attached to the wall of the proximal tubule in a rat
nephron with one end of the constituent crystals joined together (near the
arrow) and the other end free. ................................... .................................... 8

2.3. The electrical double layer structure illustrating the distribution of ions
surrounding an electrostatically charged particle............................................10

2.4. The effect of electrolyte concentration on the ionic cloud and particle
separation distance in a (A) low ionic strength solution and a (B) high ionic
strength solution.....................................................................................................

2.5. The effect of ioic strength on zeta potential or the electric potential at the
shear plane........................................................................................................ 13

2.6. A schematic diagram illustrating the attractive, repulsive, and total energy
curves for two interacting materials as function of separation distance ...............16

2.7. A schematic representation of six aggregation mechanisms which may
contribute to stone form ation.................................................... .....................18

2.8. Heterocoagulation of COM with HU is predicted based on: (A) zeta potential
determinations, (B) mixing of HU and COM at pCa < 5 where COM and
HU are both negatively charged, and (C) mixing at pCa = 4 where COM and
HU are both oppositely charged........................................................................22

2.9. A schematic representation of patch charge flocculation whereby incomplete
macromolecular coverage may create electrostatic shielding or opposite
charge in the case of a charged polymer. ............................. ...........................24

2.10. Scenario depicting the balance of hydrodynamic forces (Fd) and adhesive
forces (Fa) acting on a COM particle bound to the brush border in the human
nephron .......................................................................................................... 29









2.11. A scanning electron photomicrograph of a tubule cross section in the rat
animal model demonstrating association of a COM crystal with the
basem ent m em brane............................................................................................ 32

2.12. A schematic illustration of a human nephron.........................................................33

2.13. A schematic representation of a tubule wall found in a kidney ..............................35

2.14. A schematic representing charged particle movement in an applied electric field
to determine electrophoretic mobility...............................................................37

2.15. A schematic representation of the flow of ions under an applied hydrodynamic
pressure which generate the potential across the capillary or streaming
potential. ............................................... ............. ..................................................39

2.16. A schematic diagram of the SPM .................................... ...... ...........................42

2.17. Multiple internal reflection within the internal reflection element which is
coupled to the sam ple......................................................................................... 44

2.18. Sampling depth as a function of wavenumber for the KRS-5 internal reflection
element having a 450 incident angle......................................................................46

2.19. A schamatic diagram of a Coulter counter, electrical pulse counting instrument......47

3.1. Possible aggregation mechanisms for particles in the urinary environment ............51

3.2. Schematic representation of processing steps for the preparation of COM
crystals ("33" particles) without seeds...........................................................55

3.3. Schematic representation of processing steps for the preparation of COM
crystals ("33" particles) with seeds.......................................................... ...........57

3.4. Schematic representation of processing steps for the preparation of COM
crystals ("32" particles) with heat treatment at 900C.........................................59

3.5. (A) SEM photomicrograph and (B) and (C) theoretical equilibrium shapes of
the COM crystals precipitated without seeds.(B) The equilibrium shape of
COM crystals based on the crystallographic data of Deganello and Piro and
(C) the equilibrium shape of COM crystals based on the crystallographic
data of Tazzoli and Domeneghetti. .......................................................................64









3.6. Theoretical atomic structures of COM crystals as a function of habit plane:
(A) and (B) are the (010) and (100) planes, respectively, drawn using Cocco
and Sabelli atomic coordinates; (C) and (D) are the (010) and (100) planes,
respectively, drawn using Tazzoli and Domeneghetti atomic coordinates; (E)
and (F) are the (010) and (100) planes, respectively, drawn using Deganello
and Piro atomic coordinates. .............................. .............................................67

3.7. SEM photomicrographs and theoretical equilibrium shapes of seed crystals
and the COM crystals precipitated using seeds:(A) seed crystals, (B)
equilibrium shapes of seed crystals generated with the (001) and (010) faces,
respectively, (C) COM crystals grown using seeds, and (D) equilibrium
shape of COM crystals................................. ................ ............................... 70

3.8. SEM photomicrograph and theoretical equilibrium shapes of the COM
crystals precipitated from homogeneous solution at 900C:(A) COM
crystals, (B) equilibrium shapes of an individual COM crystal and a crystal
with a twin generated on the (001) face, and (C) equilibrium shapes of COM
crystals generated using the (010) face as the dominant face...............................72

3.9. Theoretical atomic high temperature structure (stability range: 318-415 K) of
COM crystals as a function of habit plane, (A) (010) and (B) (101)....................76

3.10 A (top) typical experimental COM x-ray diffraction pattern and (bottom) the
JCPDS file for the mineral W hewellite...............................................................77

3.11. The differential frequency vs. equivalent spherical diameter particle size
distribution of the experimentally produced COM crystals grown without
seeds fit to a log-normal probability distribution.........................................79

3.12. The relative linear dimensions of experimentally produced COM crystals..............80

3.13 Zeta potential as a function of pH for COM in saturated COM solution and
zeta potential at pH 6 for COM in 10% AUIS........................................ ...81

3.14 Contact mode SPM scan of the COM (010) crystallographic face showing the
surface roughness. ............................................................................................83

3.15 Contact mode SPM scan of the COM (101) crystallographic face showing the
surface roughness. ............................................................................................84

4.1. Optical photomicrographs of COM particles dispersed in Eccosil 5019 using
(A) no coupling agent, (B) SMAEPS coupling agent, and (C) GPTMS
coupling agent.................................................................................................. 96

4.2. The composite mixing rules fit to experimental dielectric data. ...............................97









4.3. Cauchy plots for water and COM as a function of optical direction....................101

5.1. (A) A picture and (B) a schematic diagram of the dynamic wet cell developed
to measure adhesion of particulate to surfaces.In (B), structural thru-holes
have been omitted from the drawing to improve clarity....................................108

5.2. (A) A picture and (B) a schematic diagram of the dynamic wet cell supporting
equipm ent................................................................................................... 112

5.3. A time lapse sequence of events during an adhesion experiment.The flow rates
are (A) 12 ml/min, (B) 53 ml/min, and (C) 108ml/min.......................................113

5.4. COM particles of controlled morphology shown in a (A) scanning electron
micrograph, (B) modeled using SHAPEc software showing the two
dominant crystallographic faces, (010) and (101), and (C) demonstrating the
relative crystallographic size ratios.................................................................... 114

5.5. Theoretical (010) and (101) crystallographic planes of calcium oxalate
monohydrate.The Ca2+/C2042" ratio is given above each theoretical atomic
structure. .............................................. ....................................................... 15

5.6. A (A) picture and a (B) schematic diagram of the streaming potential cell.............17

5.7. A diagram of the R-C circuit used to eliminate asymmetry and electrode
polarizations......................... ......................................................................... 118

5.8. (A) A picture and (B) a schematic diagram of the streaming potential
instrument including all of the supporting equipment........................................ 119

5.9. A plot of dial setting versus volumetric flow rate used to calibrate the
pump.The error bars represent the standard deviation of five individual
experim ents....................................................................................................... 123

5.10. A plot of flow rate versus Reynolds number indicating the stable flow inside
the dynamic wet cell at all experimental flow rates................................. ...124

5.11. Experimentally determined zeta potential of COM particles in saturated
calcium oxalate monohydrate solution.Each data point is the mean +/- 95%
confidence interval.The ionic strength of the saturated COM solution was
8x10 6 M ... ............................................................................................... ......... 126

5.12. A plot of pressure drop or driving pressure across the streaming capillary
indicating that the flow is laminar under all experimental flow conditions..........128









5.13. A plot of streaming potential vs. driving pressure for fused quartz in COM
saturated solution showing a linear regression with an r2=0.98 and 95%
confidence lim its............................................................................................129

5.14. The probability of an initial particle on the (010) and on the (101) adhering
versus the applied stress acting on the particle at failure as determined using
the dynamic wet cell.The total number of particles counted that were lying
on the (010) was n=255 and on the (101) was n=380.......................................132

5.15. A scanning probe microscopy image of the fused quartz surface under a COM
saturated solution liquid environment.Also given is the image roughness
statistics which include the ra or mean surface roughness of 0.871 nm ...............135

6.1. A plot of wavenumber vs. absorbance for fused quartz, collagen type I,
fibronectin, MATRIGEL, and PEI produced using ATR-FTIR.All of the
macromolecular substrates were coated on fused quartz...................................147

6.2. A plot of wavenumber vs. absorbance for fused quartz, collagen type I,
fibronectin, MATRIGEL, and PEI produced using ATR-FTIR for the
wavenumbers 2100-3600 cm'1.All of the macromolecular substrates were
coated on fused quartz. ................................... ................................................148

6.3. Zeta potential as a function of pH for COM in saturated COM solution and
zeta potential at a pH of 6 for COM in 10% AUIS..........................................150

6.4. A plot of driving pressure vs. streaming potential for the substrates collagen
type I, fibronectin, MATRIGEL and PEI .....................................................152

6.5. The probability of an initial COM particle on the (010) and the (101) adhering
to fused quartz in AUIS versus the applied stress acting on a particle at
failure as determined using the dynamic wet cell.The total number of
particles counted lying on the (010) was n= 13 and on the (101) was n=84......154

6.6. The probability of an initial COM particle on the (010) and the (101) adhering
to collagen type I in COM saturated solution versus the applied stress
acting on a particle at failure as determined using the dynamic wet cell.The
total number of particles counted lying on the (010) was n=82 and on the
(101) w as n= 133. ........................................................................................... 155

6.7. The probability of an initial COM particle on the (010) and the (101) adhering
to collagen type I in AUIS versus the applied stress acting on a particle at
failure as determined using the dynamic wet cell.The total number of
particles counted lying on the (010) was n=84 and on the (101) was n=72........156









6.8. The probability of an initial COM particle on the (010) and the (10i) adhering
to fibronectin in COM saturated solution versus the applied stress acting on
a particle at failure as determined using the dynamic wet cell.The total
number of particles counted lying on the (010) was n=87 and on the (10i)
w as n= 104............................................ ......................................................... 157

6.9. The probability of an initial COM particle on the (010) and the (101) adhering
to fibronectin in AUIS versus the applied stress acting on a particle at failure
as determined using the dynamic wet cell.The total number of particles
counted lying on the (010) was n=95 and on the (101) was n=143...................158

6.10. The probability of an initial COM particle on the (010) and the (101) adhering
to MATRIGEL in COM saturated solution versus the applied stress acting
on a particle at failure as determined using the dynamic wet cell.The total
number of particles counted lying on the (010) was n=57 and on the (101)
w as n=54 ............................................. ........................................................ 159

6.11. The probability of an initial COM particle on the (010) and the (101) adhering
to PEI in COM saturated solution versus the applied stress acting on a
particle at failure as determined using the dynamic wet cell.The total number
of particles counted lying on the (010) was n=61 and on the (101) was n=69....160

6.12. A bar chart summarizing the adhesion data experimentally determined in COM
saturated solution (i.e., low ionic strength).The numbers above each set of
bars is the value of the zeta potential for each substrate...................................161

6.13. A bar chart summarizing the adhesion data experimentally determined in COM
saturated solution (i.e., low ionic strength).The numbers above each set of
bars is the value of the zeta potential for each substrate...................................163

6.14. COM particles (A) coated during an adhesion experiment using fibronectin and
(B) uncoated COM particles used in an adhesion experiment.Both
photomicrographs were taken after flow ...................................................168

6.15. A plot of the stress on a 5 uim to 8 gm radius model COM particle under the
hydrodynamic conditions found in the different areas in the human
kidney.The shaded region represents the range of experimentally measured
values of adhesion of COM to biological materials................................... 171















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


MEASUREMENT OF ADHESION
BETWEEN CALCIUM OXALATE MONOHYDRATE AND MODEL SURFACES
USING A DYNAMIC WET CELL

By

Craig F. Habeger

December, 1997




Chairman: Dr. James H. Adair
Major Department: Materials Science and Engineering


Calcium oxalate monohydrate (COM) is the primary constituent in kidney stones.

COM crystals were synthesized in the laboratory and characterized. Computer

calculations of particle shape have been reconciled to observed shapes of COM crystals

experimentally synthesized under various conditions. Comparison between the

theoretical atomic structures generated by computer calculations are consistent with

previously reported atomic layering sequences.

Composite mixing rules were used to deconvolute the dielectric constant of COM

from a COM/silicone composite. Utilizing the Lichtenecker dielectric mixing model, the

value of the static dielectric constant of COM was determined to be 28.9. Optical and

dielectric data were then used in the Tabor-Winterton relationship to calculate the

Hamaker constant, Ai31, of COM particles interacting in water. The A131 for COM as a









function of crystallographic habit also was examined. The mean value of A131 for COM

was calculated to be 13.7x10'21 J at 370C in an aqueous environment.

A hydrodynamic method for measuring the adhesion of particles to a surface has

been designed for use in the study of kidney stone disease and other pathological

biomineralization phenomena. The hydrodynamic force required to displace a particle

adhering to a fused quartz substrate was calculated via the Poiseuille equation. The

strength necessary to remove 50% of the COM particles adhering to the substrate on the

(010) and (10 ) crystallographic surfaces are 81 and 170 Pa, respectively. The

previously determined Hamaker constant and measured values of zeta potential were used

to calculate the energy of interaction between a COM particle and the fused quartz

substrate which was found to be comparable to experimentally measured values, provided

the separation distance was on the order of 20 nm.

Using the instrument and technique developed, the adhesion of COM to

biologically and non-biologically relevant materials was measured in COM saturated

solution and in an artificial urine ion solution. The biologically significant materials were

the proteins collagen type I, fibronectin, and MATRIGEL, a mixture of basement

membrane proteins. The non-biologically relevant material was polyethyleneimine a

positively charged macromolecule used as a control. MATRIGEL and the positive

control, polyethyleneimine, exhibited the highest adhesion to COM crystals.















CHAPTER 1
INTRODUCTION


1.1 Introduction

Kidney stones effect hundreds of thousands of people each year. In 1991,

895,467 visits were made to doctors for both kidney and ureter stones [NIH97] and in

1993, kidney and ureter stones resulted in approximately 302,00 hospitalizations [Cla95].

The direct and indirect cost of kidney and ureter stones in 1993 was approximately $1.83

billion [Nat95]. Prevention of kidney stones is the focus of much research; however,

before prevention can occur the mechanisms of stone formation must be understood.




1.2 Literature Review

Chapter 2 reviews the pertinent kidney stone literature and gives background on

techniques utilized in the work reported herein. Kidney stone disease has been described

as an opportunistic disease which can rely on many causative mechanisms acting in

concert [Fin78a, Fin84]. One of those possible mechanisms is the growth of an adherent

particle, called a fixed particle, in the lumen of a nephron. Such an adherent particle can

subsequently grow by secondary nucleation and growth and/or aggregation of crystallites

to a size large enough to occlude a renal tubule, which is the basis of the fixed stone

mechanism of kidney stone development. Finlayson and Reid [Fin78b] calculated the

time for a COM crystal to grow to a large enough size to cause an occlusion and

determined that the crystal would not have time enough to grow to such a size.









Therefore, crystal attachment is a necessary process in kidney stone formation [Man91,

Man94].

Renal injury has also been implicated in fixed stone formation [Gil79, Kha84,

Man91]. Khan [Kha82, Kha95a] has discussed that after attachment of a crystal to the

epithelial cell surface, injury to the cell may or may not occur.

A few different results may ensue due to crystal attachment to the epithelial cell.

The cell may die and be washed away in the urine flow taking with it the crystal. The cell

may envelope the crystal and continue to function normally [Lie92, Lie93, Lie94] with no

further growth of the crystal. However, if epithelial cell injury has occurred, basement

membrane proteins may be accessible to adhere to COM crystals by sloughing of the cells

leaving behind the exposed underlying architecture, the basement membrane. The

basement membrane is a layer of proteins that underlies the epithelial sheet and will be

discussed subsequently in more detail. Depending on the extent of injury, other proteins

common to the extracellular matrix found below the basement membrane may also be

accessible for attachment to COM crystals.

Researchers have shown the attachment of COM crystals to epithelial cells and

fibrous proteins characteristic of those found in the basement membrane and extracellular

matrix, yet the importance of these individual materials in kidney stone formation with

respect to COM crystal adhesion is not known. The necessity to determine the

important materials most responsible for COM crystal adhesion will allow researchers to

focus their interests to those more important materials in the search for a possible cause

and therefore relief from kidney stone formation.


1.3 Development and Characterization of Model COM Particles

Chapter 3 demonstrates the laboratory synthesis of COM particles having a

desired size and crystallographic shape to be used in adhesion experiments. The optimal

particle shape is spherical; however, COM has a monoclinic crystal structure and









spherical, crystalline particles cannot be produced. Therefore well-defined shaped,

rectangular particles having a narrow size distribution were produced to limit errors

associated with calculations. The experimentally produced particles were modeled using

computer software to determine the crystallographic face indices and the theoretical

atomic structure of two dominant crystallographic faces. The particles were characterized

using electrophoresis and atomic force microscopy. The zeta potential values for the

particles were determined using electrophoresis in a low ionic strength solution, COM

saturated solution, and a high ionic strength solution, artificial urine ion solution. Finally,

the surface roughness of each of the two dominant crystal faces were measured in

saturated COM solution using contact mode Atomic Force Microscopy.


1.4 Evaluation of the Calcium Oxalate Monohydrate Hamaker Constant


Chapter 4 involves the determination of the Hamaker constant, a measure of the

van der Waals interaction forces, for COM. Attractive van der Waals forces are very

important in the adhesion process and a reliable value of the Hamaker constant is

necessary to determine these forces. The COM particles developed in Chapter 3 were

mixed with a silicone material to produce a composite. The dielectric properties of the

composite were measured and composite mixing rules were used to deconvolute the

dielectric constant of COM from the COM/silicone composite. Optical and dielectric

data were then used to calculate the Hamaker constant, A131, of COM particles interacting

in water.


1.5 Development of the Dynamic Wet Cell and Streaming Potential Measurements

Chapter 5 describes the development of a hydrodynamic method for measuring

the adhesion of particles to a surface for use in the study of kidney stone disease and

other pathological biomineralization phenomena. By using hydrodynamic flow to remove

particles from a model surface, the strength with which particles adhere to a surface can









be measured. The hydrodynamic force required to displace a particle is calculated via the

Poiseuille equation using the dynamic wet cell dimensions and the fluid flow rate. A

model surface consisting of fused, spectroscopic grade quartz was used in the

development of the apparatus. Also developed was an instrument to measure the

electrostatic potential, called zeta potential, at the surface of a flat substrate in solution.

The previously determined Hamaker constant and measured values of zeta potential were

used to calculate the energy of interaction between a COM particle and the fused quartz

substrate. The theoretical values of adhesive pressure were compared to the measured

values of adhesion.


1.6 Measurement of COM Adhesion to Macromolecular Substrates


Chapter 6 investigates the adhesion of COM to both biologically relevant and

non-biologically relevant materials. The adhesion measurements were measured in COM

saturated solution and in an artificial urine ion solution, using the instrument and

technique developed in this research. The use of two different ionic strength solutions

allows interpretation of the types of material-material interactions taking place (i.e., van

der Waals, electrostatic, chemical) to be made. The biologically significant materials were

the proteins collagen type I, fibronectin, and MATRIGEL, a mixture of basement

membrane proteins. The non-biologically relevant material was polyethyleneimine, PEI, a

positively charged macromolecule used as a control.


1.7 Conclusions and Future Research

Chapter 7 summarizes the conclusions of the current research and presents

suggestions for future work.















CHAPTER 2
LITERATURE REVIEW


2.1 Introduction

Kidney stone disease has been called an opportunistic disease because it is

thought that kidney stone disease takes an unfortunate, simultaneous convergence of

factors that may be achieved by several pathways for a stone to form [Con90, Fin78a,

Fin84, Rob86]. Thus, injury, supersaturation (particularly hyperoxaluria), nucleation and

growth mechanisms all play a role in contributing to stone formation as demonstrated in

Figure 2.1. Even so, urolithiasis is idiopathic with no one risk factor providing an obvious

source of stones in chronic stone former. However, both in vivo and in vitro evidence

indicates that the presence of particles is a necessary but insufficient condition for stone

formation. Most non-stone forming individuals, at certain times, have crystalluria yet do

not have a stone incident. For a stone to be created from freely flowing particles, either a

particle must grow to a large enough size to occlude the tubule or aggregation of multiple

particles must occur. Furthermore, if aggregation is responsible for stone formation, the

attractive energy or energies holding the primary particles together in an aggregate must be

great enough to withstand the hydrodynamic shear forces due to the flow of fluid through

the nephron.

The possibility that the growth of a large crystal leads to a stone has been shown

to be unlikely by several teams of investigators [Fin78b, Kok94]. Finlayson and Reid

[Fin78b] predicted that there was not enough time for a freely flowing particle to grow to

a size sufficient to occlude the tubule based upon the hydrodynamics in the nephron and

measured rates of crystal growth for calcium oxalate monohydrate. It was also suggested
















SCatalysis


Cel Aggregation
-cell/ 2
Damage
a F1


C Aggregation


Figure 2.1. Pathways to kidney stone formation.









by Finlayson and Reid that aggregation of freely flowing particles was not likely due to

the relatively low concentration of particles within the nephron at any given point in

time. However, a recent analysis of the hydrodynamic aspects of transit time through the

nephron by Kok and Khan [Kok94], based on better estimates of the diameters of

different regions in the nephron, indicated that aggregation of freely flowing particles is

possible. Regardless, attachment of calcium oxalate particles to the epithelial nephron

wall has been demonstrated in a rat animal model [Fin84, Fin78b], as shown in Figure 2.2.

Thus, the concept of fixed particles at the epithelial wall has become a fundamental

principle in the development of kidney stones. In more recent work [Kha84, Man87,

Man91, Rie88], it has been demonstrated that particle attachment can be induced by

epithelial wall damage in which intercellular species, such as phospholipids, may play a

role.

The objective of this chapter is to review the dynamics of aggregation for particles

and interactions of the particles and/or aggregates with biological surfaces such as the

basement membrane or interluminal wall of the nephron. The nature of the forces among

particles and surfaces will be discussed with respect to the hydrodynamic flow scheme

found in the nephron; however, particle-particle interactions must first be discussed.


2.2 Particle Interactions

Particles in solution are constantly moving. The movement of the smaller

particles (<1 Igm) is due to Brownian motion, also called perikinetic motion. The

movement of larger particles (>5 jgm), defined as orthokinetic motion, is due to gravity

and convection currents. The constant motion in solution causes the particles to interact

with one another ultimately resulting in either stabilization or aggregation. When

repulsive forces dominate the particle-particle interaction, stabilization is the result. The

opposite is the case if attractive forces dominate the particle-particle interaction, called

aggregation. These forces arise from an electrostatic potential on the surface of the



















































Figure 2.2. COM particle aggregate attached to the wall of the proximal tubule in a rat
nephron with one end of the constituent crystals joined together (near the arrow) and the
other end free (bar = 5 gm). Photomicrograph by S.R. Khan [Kha91].









particles and from an inherent dipole interaction of atoms and molecules within one

particle interacting with another particle.


2.2.1 Electrostatic Repulsion

The surface potential, 'Fo, is the potential difference between the solid surface and

the bulk solution. Ions which alter the surface potential are called potential determining

ions. Curreri et al. [Cur79] have hypothesized that Ca2+ and C2042- specifically adsorb to

the COM surface and, thus, are termed the potential determining ions for COM.

The Gouy-Chapman electrical double layer model describes the excess ions

present on the surface of the solid phase and the distribution of ionic charge of opposite

sign in the solution phase surrounding the electrically charged surface which assures

electroneutrality. A schematic diagram of the Gouy-Chapman model is shown in Figure

2.3. The double layer model consists of an inner layer, often called the Ster layer, which

describes ions adsorbed into the compact or inner region of the double layer, and the outer

or diffuse layer [Ove52]. Theoretical analysis of the double layer shows that the charge

density in an aqueous solution decreases rapidly with increasing distance from the solid

surface, and the electrostatic potential for low potentials is given by the Debye-Htickel

approximation [Hun81],


V = ,o exp(-xt), (2.1)


where Vy is the potential as a function of distance into the solution, K is called the Debye-

Htickel parameter (1/length), and x is the distance from the solid surface.

Based on the electrostatic model for aggregation, in the simplest case, the repulsive

contributions between particles in solution arise from the interaction of ionic clouds

surrounding the particles, as demonstrated by Figure 2.4. The extent of the ionic cloud is

governed by the solution ionic strength described by the simplified equation





















p < tStern Layer

S1\ Gouy-Chapman (Diffuse) Layer
o +0
0 ) 0
E)
W) E) Eo oD
+ 0000 0

Distance from Surface










Figure 2.3. The electrical double layer structure illustrating the distribution of ions
surrounding an electrostatically charged particle.


























+
d

(A)






+ ++


.: ""++.y
++




d

(B)






Figure 2.4. The effect of electrolyte concentration on the ionic cloud and particle
separation distance in a (A) low ionic strength solution and a (B) high ionic strength
solution.
solution.












C = kT (2.2)



where e is the elemental charge, nf, is the number of ions of type i per unit volume far

from the surface, z, is the valence of ions of type i, e is the dielectric permittivity of the

solvent, k is the Boltzmann constant, and T is the temperature in Kelvin. The Debye-

Hiickel simplification assumes that the surface potential or stern potential is low (e.g.,

less than 25 mV). The quantity 1/K is referred to as the Debye length and indicates the

thickness of the double layer. The Debye length is dependent only on temperature,

valence of the electrolyte ions, and electrolyte concentration. Solution ionic strength, I, is

given by


I= (cz2)z (2.3)



where c, is the solution concentration (M). As the ionic strength of the solution increases,

the surface charge decay into the solution occurs more rapidly due to the inability of the

solution to support charge. That is, the diffuse double layer becomes compressed due to

the increased concentration of ions in solution as demonstrated in Figure 2.5. The

increased concentration of ions near the charged particle surface causes the electrical

charge to be neutralized by the ions of opposite charge, counterions, in solution. Figure

2.4 schematically describes the ionic cloud surrounding a charged particle in low and high

ionic strength environments. The variable d in Figure 2.4 is the distance of separation for

the two charged particles. At low ionic strength, the ionic cloud is large and electrostatic

repulsion prevent particle-particle contact. At higher ionic strength the particles can
















I SHEAR PLANE


I


IIi


<12< 13


INCREASING IONIC
STRENGTH


SURFACE DISTANCE INTO THE SOLUTION
















Figure 2.5. The effect of ioic strength on zeta potential or the electric potential at the
shear plane.









come into much closer proximity due to collapse of the surrounding ionic cloud. If the

collapse is significant, aggregation may occur.


2.2.2 Attraction


The tendency towards aggregation can result from oppositely charged electrostatic

potentials or from van der Waals forces which are present at all times, irrespective of

solution conditions. The attractive van der Waals forces are composed of multiple

intermolecular interactions between the ions, molecules, and electrons that make up the

particles interacting across a dielectric medium such as water [Hou80, Hun93, Ber90].

The mutual attraction between interacting particles arise from harmonic oscillations at the

molecular, atomic, or subatomic level. Three primary sources of such intermolecular

interactions exist depending upon the nature of the interacting species. Some of the more

important specific dispersion interactions are known as Keesom, Debye, and London

interactions [Isr92]. Keesom forces are due to molecular dipole-dipole interactions in the

particles. Debye interactions occur when molecular dipoles in one particle induce

electronic polarization in the other interacting particle. As such, Keesom and Debye

interactions occur only if a material has one or more dipoles present within its structure.

In contrast, London interactions are more ubiquitous because London interactions are due

to mutual electronic polarization with all atoms, or electron cloud density shift, of

adjacent atoms.

The range of the force due to dipole interactions between two atoms or molecules

is a nanometer or less and varies as 1/r6. However, the interactions are to some degree

additive in that they effect all neighboring atoms or molecules in a particle. Particle-

particle van der Waals forces interact over a much longer range, 1/r2.









2.2.4 Interaction Energy


The Derjaguin and Landau [Der41] and Verway and Overbeek [Ver48] (DLVO)

theory considered colloid stability in terms of the electrical double layer and van der

Waals forces. Figure 2.6 is a plot of interaction energy as a function of separation

distance, which is a potential energy diagram for two interacting materials across a

medium. Positive interaction energy refers to repulsive energy and negative interaction

energy refers to attractive energy. In the diagram both the attractive and repulsive

portions are plotted with respect to separation distance of the two interacting materials.

The third interaction line is the sum of the attractive and repulsive curves, the total

interaction energy curve. Labeled on the total interaction energy curve in Figure 2.6 are

the primary and secondary minimum, and the energy barrier to aggregation.

The primary minimum is the potential energy well that occurs when the two

interacting materials are in contact. The main attractive forces are due to van der Waals

forces. When two interacting particles are at a separation distance such that their

associated interaction energy is in the primary minimum, the attractive energies dominate

the total interaction energy and the two materials cannot be separated. The two particles

are said to be thermodynamically unstable with respect to dispersion.

The secondary energy minimum occurs at larger separation distances, beyond the

energy barrier. When two interacting materials are in the secondary energy minimum,

they form a weak aggregate and may be dispersed with the addition of energy. The

secondary minimum is thermodynamically metastable with respect to aggregation.

The potential energy barrier to aggregation occurs between 1 to 4 nm [Isr92] and is

due to strong electronic repulsion between the ionic clouds of the interacting particles and

is dependent on the surface potential. If the surface potential is high, then the potential

energy barrier will be high; and, if the surface potential is low, the potential energy barrier

will be low. As the electrolyte concentration of the medium increases, the energy barrier
























>1 -roientai energy 5arner

C
w




SSecondary Minimum
Total Energy

? Attactive Energy




Separation Distance











Figure 2.6. A schematic diagram illustrating the attractive, repulsive, and total energy
curves for two interacting materials as a function of separation distance.









decreases and aggregation becomes more and more likely to occur. Finally, at the critical

coagulation concentration, the concentration of electrolyte at which the repulsive energy

barrier falls below zero, aggregation occurs.


2.3 Aggregation Mechanisms Among Particles and Particles at Surfaces

Numerous studies [Fin78a, Fin84, Fin78b, Rie88, Ada81, Coe91, Deg91, Edy86,

Edy87a, Har86a, Har86b, Kok90, Rob85, Rya81a, Rya81b, Rya81c, Rya84, Rya86,

Scu86a, Scu86b, Wie87] have shown that the aggregation and/or adhesion of COM

particles potentially can lead to the ultrastructures composing kidney stones. However,

the actual mechanisms) of aggregation within the biophysical environment of the human

kidney have not been examined in detail. There are at least six aggregation mechanisms, as

shown in Figure 2.7, acting alone or in concert which may contribute to stone formation

[Ada95].

To establish dispersion techniques which may prevent stone formation, first it is

necessary to determine the most important aggregation mechanisms with respect to

interparticle strength. A number of possible mechanisms exist for COM aggregation

within the human kidney. In addition to minimum double layer interactions, other

interactions include secondary minimum coagulation, heterocoagulation, polymer bridging

flocculation, aggregation by secondary nucleation and growth, and immiscible amphiphilic

molecule flocculation. Hydrodynamic factors also need to be considered, including the

size of the tubule and the shear rate associated with fluid flow within the nephron, as

discussed by Finlayson and Reid [Fin78b], and Kok and Khan [Kok94]. Furthermore,

the concentration of the COM particles performs a role whether one is considering fixed

stone disease or aggregation of the particles in a freely flowing state. Finally, the flow

patterns associated with peristaltic compared to continuous flow within the nephron have

not been addressed in past studies but should be to realistically assess aggregate formation






























E-
Q- .-


t 4.


-I


0

E
E0
o u


U.



8


4-
0
8
o
6-5



.RO


o

00
It
c.


+


+ 0


1*









in the kidney. Each of the mechanisms will be discussed with respect to their likelihood

in the human nephron.


2.3.1 Particle Aggregation in Simple Electrolytes


The aggregation of COM particles has traditionally been attributed to the absence of

charge on COM particle surfaces and/or the high ionic strength of urine. However, neither

of these aggregation mechanisms accommodates the observation that COM particles

observed in vivo invariably have organic matter on their surfaces, as shown in Figure 2.2.

Generally, particles in aqueous solution have a surface charge created by a combination of

several different charging mechanisms. In the case of COM, Curreri et al. [Cur79a,

Cur79b, Cur87] showed that surface charge is due to incongruent dissolution of the

constituent Ca2+ and C2042-. These and other species from solution then can adsorb (in

their hydrated form) into an adsorbed ion layer known as the Stern layer. The

electroneutrality of the system, composed of the surface and Ster charges and

surrounding solution, is achieved by the charge in the diffuse cloud of ions that only are

attracted electrostatically toward the surface.


2.3.2 Secondary Minimum Coagulation


As shown in Figure 2.7, a minimum exists at relatively large distances of

separation in the interaction energy for COM particles. As the ionic strength increases,

the magnitude of this secondary minimum increases. The secondary minimum in the

interaction energy curve is a consequence of the longer range of the van der Waals

attractive forces than the electrostatic repulsive interactions [Hou80, Son72]. Thus, even

when the charge at the surfaces of the interacting particles is great enough to produce an

energy barrier at intermediate separation distances, secondary minimum coagulation may

take place because there is no energy barrier for this mechanism of aggregation. The

strength of the interparticle bonds for secondary minimum interactions has been evaluated









theoretically and experimentally by Chan and Halle [Cha84]. It was demonstrated that

the mean lifetime for secondary minimum aggregates increased with increasing ionic

strength of the suspension containing the model spherical polystyrene particles.

Adair [Ada81] showed that COM suspensions composed of relatively coarse

primary particles (-5 gtm equivalent spherical diameter) aggregate over a wide range of

solution and surface charge conditions. Secondary minimum aggregation was implicated in

conditions where primary minimum aggregation was minimized because of low ionic

strength and relatively high zeta potential. However, the results were ambiguous because

only the thermodynamic aspects of the coagulation process were addressed in Adair's

study. The strength of the proposed secondary minimum interaction was not determined

by analyzing the hydrodynamic shear forces required to promote breakup of the

aggregates. Aggregate bond strength measurements as a function of particle size were

suggested since the magnitude of the secondary minimum increases as a function of the

radii of the interacting particles. The increased likelihood of secondary minimum

coagulation with increasing particle size may have important implications to stone disease

since it has been reported by Robertson [Rob69] that stone forming individuals have

particles significantly larger in size than non-stone former.


2.3.3 Heterocoagulation

Heterocoagulation is the aggregation among particles of different materials. We are

not aware of any investigators that have addressed this potential mechanism for stone

formation. However, there have been a number of investigators [Bar78, Der54, Kuo80,

Mat81] within the colloid chemistry community that have developed the theoretical and

practical framework for a study of heterocoagulation. The basis for heterocoagulation is

the difference in surface charge polarity of particles comprised of different materials.

Thus, a positively charged COM particle may be electrostatically (as well as through the

van der Waals interactions) attracted to a negatively charged hydroxyapatite (HAP)









particle. However, even if the surface charge on particles of dissimilar materials are the

same, the van der Waals attractive forces may be strong enough to promote

heterocoagulation (as well as homocoagulation).

Within the human nephron a variety of potential combinations exist that may lead

to heterocoagulation. These interactions include, COM-HAP, COM-HU (uric acid),

COM-NH4U (ammonium urate), COM-COD (calcium oxalate dihydrate), COM-E.coli

and other bacteria, and COM with various macromolecules. Heterocoagulation will be

reversible only when the sign of the surface charge is made the same for all particles and is

sufficient to overcome van der Waals interactions. The sign of the surface charge and

corresponding solution conditions that will promote or inhibit heterocoagulation need to

be evaluated for each system. It has been predicted that epitaxy of the high temperature

form of COM and HAP is unlikely because of incoherent crystal structures [Man81].

However, heterocoagulation can explain the presence of HAP with COM in a stone.

Heterocoagulation has not been addressed for material systems relevant to

urolithiasis. Preliminary experiments have been conducted in our laboratory to determine

the conditions in which particles composed of various materials will be likely to

heterocoagulate. An obvious starting point in this initial evaluation is to examine the

effect of the polarity of the zeta potential for the various materials as a function of

relevant solution conditions. The zeta potential data for COM and HU are summarized

in Figure 2.8(A) as a function of Ca2+ and C2042- concentration. The COM data are from

Curreri et al. [Cur79a] and incorporate the solubility product for COM in the Ca2+

concentrations. The uric acid data from Adair et al. [Ada88] are given as a function of

Ca2+ or C2042-. COM has a point of zero charge (pzc) at pCa = 5.2 (Ca2+ = 6.3x10-6 M)

with COM having negatively charged surfaces above this pCa and positively charged

surfaces at higher Ca2+ concentrations. Adair et al. [Ada88] showed that uric acid is

negatively charged over a wide range of pH and Ca2+ and C2042- concentrations.












40

20

0

-20

-40


i-C

-7 -6 4 4 2 -1
10 10 10 10 10 10 10
CONCENTRATION OF Ca (M)


(B)













(C)


Figure 2.8. Heterocoagulation of COM with HU is predicted based on: (A) zeta potential
determinations, (B) mixing of HU and COM at pCa < 5 where COM and HU are both
negatively charged, and (C) mixing at pCa = 4 where COM and HU are both oppositely
charged (bar = 20 gm) [Ada95].


M COM
* Uric Acid Ca


I


-

-









Thus, mixing COM and HU particles where pCa is less than 5.2 should promote

heterocoagulation between the negatively charged HU and the positively charged COM.

As shown in Figure 2.8(B) and 2.8(C), fine COM particles produced by the dimethyl

oxalate decomposition adhere to the larger, prismatic HU particles when pCa=4. When

COM and HU particles in saturated COM solutions are mixed with pCa=5, the fine

COM particles have a greater affinity toward one another than the HU particles and

heterocoagulation does not occur, as shown in Figure 2.8(C).


2.3.4 Polymer Bridging Flocculation


This mechanism takes place when there is insufficient polymer (or

macromolecule) for full surface coverage on particles. This is the only aggregation

mechanism that explains some ultrastructural observations of Boyce, Khan et al., and

others [Boy68, Kha83a, Kha83b, Kha87, Mey82, Mey71, Pri86] on the role of matrix

macromolecules in the microstructure and ultrastructure of the mature stone. Maximum

flocculation takes place when one-half of the surface of a particle is covered by polymer

[Son72, Hun86], as shown schematically in Figure 2.9. However, flocculation takes place

anywhere between about one-tenth surface coverage to greater than 75 percent coverage.

We are not aware of any studies relevant to urolithiasis that have addressed the role of

flocculation in detail. Kok et al. [Kok90], Finlayson [Fin78a], and Robertson and

Peacock [Rob85] have discussed this mechanism with other possible mechanisms of

aggregation; but, the research emphasis has been on the prevention of aggregates by

employing large concentrations of macromolecules or polymers [Coe91, Deg91, Rya81a,

Rya84, Scu86a, Scu86b, Lan88, Lea77]. This mechanism can also explain the conflicting

reports of inhibition versus promotion [Edy86, Cam89, Gro90]. Thus, macromolecular

species such as uropontin, Tamm-Horsfall mucoprotein, and nephrocalcin may play a

dual role: at sufficiently low concentration, aggregation is promoted through flocculation




















































Figure 2.9. A schematic representation of patch charge flocculation whereby incomplete
macromolecular coverage may create electrostatic shielding or opposite charge in the case
of a charged polymer.









while at higher surface coverage, dispersion of particles is achieved through the protective

colloidal effect of the macromolecular coating [Hou80, Son72].

Although there have been limited studies on flocculation with respect to

urolithiasis, there have been a number of investigations on floc formation and

hydrodynamic breakup because of its importance in wastewater treatment and other

technologies [Dit82, Eis85, Gre81, Gre85, Ray87]. These studies provide a basis for

evaluating the role of flocculation as an aggregation mechanism for COM and other

relevant particles in urine. For example, an excellent starting point is to determine the

macromolecular or polymer dosage for relevant urinary species required to achieve the

maximum flocculation. The critical flocculation concentrations (CFC) for a particular

macromolecule will indicate whether this mechanism is likely by comparison with its

concentration range in urine.

In preliminary studies, we have examined a flocculant commonly used in mineral

recovery. Polyethyleneimine (PEI) is a positively charged highly branched polymer

molecule used by Pelton and Allen [Pel84] to produce positive charge on glass surfaces in

their particle adhesion studies. It has been shown in preliminary studies, using the

apparatus developed by Eisenlauer and Horn [Eis85], to evaluate the aggregation of freely

flowing particles in suspension. This device has the advantage that flow rate and mode of

flow (i.e., continuous versus peristaltic) can be varied. The commercial analogue to

Eisenlauer and Horn's device, known as a photometric dispersion analyzer (Rank

Brothers, Cambridge, UK), has been used extensively to monitor the flocculation of model

systems. Initial experiments have demonstrated that fine COM particles flocculate at

intermediate dosages of PEI. The relative degree of flocculation varies as a function of

charge (as dictated by the concentrations of Ca2+ and C2042).









2.3.5 Flocculation of COM Particles by Phospholipids and other Intercellular Substances


Work by Khan, and Mandel and co-workers [Kha88, Man94] indicates that

interaction of COM particles with phospholipids at either the epithelial surface of cells or

with phospholipids liberated into the bulk solution by cell injury are important features

in the formation of aggregates in the interluminal channel of the nephron. Flocculation by

sparingly soluble, amphiphilic molecules such as the phospholipids forming a major

component of the epithelial cell membrane have not been studied with respect to

aggregation, but Mandel et al. [Man94] clearly show that adhesion of COM particles is

important. The ability of phospholipids to adsorb to the surfaces of sparingly soluble

inorganic particles and surfaces is well established as are the forces arising from the

interaction of phospholipid monolayers on mica and similar surfaces based on force

balance work by Israelachvili and others [Isr92]. Thus, one would expect phospholipids

to demonstrate an effect similar to macromolecules or polymers capable of promoting

flocculation. However, the interparticle strength of particles flocculated in either freely

flowing suspensions or at surfaces containing phospholipids (i.e., cell membranes) would

be expected to depend on the concentration of phospholipids and the efficiency of

adsorption of such species to COM surfaces.


2.3.6 Adhesion of Particles at Surfaces


Adhesion is an important mechanism in the formation of fixed kidney stones and

is dependent on the adhering materials and the suspending medium. One of the first

theories of adhesion between solid particles and surfaces was given by Krupp [Kru67],

where he defined three classes of interactions:


1. Class I interactions include long range attractive interactions resulting from
van der Waals forces and electrostatic forces.









2. Class II interactions are given by short range attractive interactions such as
chemical bonds and hydrogen bonds.

3. Class III interactions involve interfacial reactions occurring at elevated
temperatures including sintering effects, diffusive mixing, and mutual
dissolution and alloying.


Class III interactions may also be important at lower temperatures for polymer or

macromolecular diffusive mixing. Other important forces also exist that were not directly

addressed by Krupp. These forces are solvation forces, structural forces, or in a water

medium called hydration forces. These short range forces involve ordering of the solvent

medium to cause either attraction or repulsion between particles depending on the

hydrophobic/hydrophilic nature of the solids in suspension [Hor90, Isr92].

All of the aforementioned interactions may apply during adhesion; however,

Krupp [Kru67] believed that van der Waals forces and electrostatic forces are the

dominating long range attractive forces between adhering materials. Krupp thought that

only under ultra-high vacuum or extremely pure systems would primary chemical bond

formation take place due to the saturation of bonding sites by contaminants under

ambient conditions. Kallay et al. [Kal87] was also of the opinion that van der Waals

interactions and electrostatic interactions were dominating, but they furthermore

suspected that short range repulsion played a role in the total interaction energy of

adhering bodies, citing solvation forces, as did Israelachvili [Isr92] and also electron cloud

repulsion. These short range forces were said to act at separation distances on the order

of the distance of closest approach of the two surfaces. Theoretical models for both long

range interactions [Der41, Ver48, Hog66, Wil93] and short range [Isr92, Kal87]

interactions are available for varying surface geometry interactions and will be discussed

later.

The study of adhering materials is fundamental and very broad in application.

The adhesion of many materials by several researchers has been documented [Vis76]. For









example, the adhesion of Fe203 to a glass substrate and a glass substrate covered by a

layer of gelatin [Ryd95], red blood cells to glass [Moh74], human fibroblasts to glass

[van92], submicrometer particles to silicon [Bus93], and bacteria to glass [Bus92] have all

been measured. As observed, applications of the adhesion measurement range widely.

Just as a number of applications are in need of the adhesion measurement, a

number of techniques can be used to measure the adhesion of solid particulate material to

solid surfaces. In past studies researchers have used many techniques to evaluate

adhesion [Cor66, Zim82]. These techniques include a rotating disc [Kri94], a packed

column [Kal87], a centrifuge method [Kor60], a vibrating method [Der61], a surface force

apparatus [Isr78], and a hydrodynamic method that utilizes parallel plates [Pel84, Pel82]

to name only a few. The most recent technique for measuring the force with which a

particle adheres to a surface utilizes the scanning probe microscope (SPM) [Duc91].

Other adhesion measuring techniques exist and are used but are best suited for specific

material systems, much like the techniques mentioned above. Further discussion will be

limited to only the hydrodynamic parallel-plate and adhesion measuring technique.

Another approach in evaluating the bond strength of particles adhering to a surface

is to determine the hydrodynamic shear force required to remove particles. Pelton

[Pel84], Busscher [Bus84], Matijevic [Mat80, Mat81], Owens [Owe87], and others

[Ols78] have used this approach to evaluate thermodynamic models for the attachment of

particles to surfaces. The balance of forces proposed for a COM particle attached to the
epithelium is shown schematically in Figure 2.10. The shear stress at the wall (Tw) is

given by [Owe87, Esk68]


dP
rw = b (2.4)
dl


where Tw is the shear stress at the wall, P is the hydrostatic pressure drop across the

conduit of length, 1, and wall separation, b.
































Epithelial
Brush Border
in the Human
Nephron


COM Particle
FLOW

Fd F
{ gM


Figure 2.10. Scenario depicting the balance of hydrodynamic forces (Fd) and adhesive
forces (Fa) acting on a COM particle bound to the brush border in the human nephron.









Determination of the adhesion strength for particles adhering at surfaces is of

fundamental importance in deducing whether the fixed particle mechanism for stone

formation proposed by Finlayson and Reid [Fin78b] is reasonable within the

hydrodynamic system of the nephron. Kok and Khan [Kok92] recently re-evaluated the

hydrodynamics in the kidney and determined that it is not possible to have a stone

occurrence with out the aid of adhesion.

Riese, Mandel, and coworkers and others [Man94, Rie92, Rie88, Yam96] have

shown in vitro attachment of COM crystals to inner medullary collecting duct (IMCD)

epithelial cells of the rat animal model in the static case (i.e., without the presence of

flow). They also proposed that perturbations in the cell membrane structure with a loss

in membrane polarity can enhance crystal attachment [Rie92]. Lieske et al. [Lie95,

Lie96a] has also measured adhesion of COM to MDCK cells and to monkey renal

epithelial cells in a similar manner to Reise et al [Rie88]. They determined that cell

anionic sites can be blocked by specific cations [Lie96a] and the positive sites on a COM

crystal may be blocked by specific anions [Lie95] thereby minimizing adhesion. Bigalow

et al. [Big97] recently demonstrated that COM crystal attachment to IMCD cells was

effected by the cell membrane fluidity. Changes in temperature, cholesterol content, and

cell culture time which increase cell membrane fluidity also increase the ability of COM

crystals to bind to the membranes leading the researchers to conclude that a long range

arrangement in the membrane is created to match the COM crystal structure [Big97]. In

vivo crystal attachment to epithelial cells in the rat animal has been demonstrated by

Khan et al. [Kha82]. Renal injury has also been implicated in fixed stone formation

[Man91, Gil79, Kha84]. Khan [Kha82, Kha95a] has discussed that after attachment of a

crystal to the epithelial cell surface that injury to the cell may occur but may not always

occur.

A few different results may ensue crystal attachment to the epithelial cell. If the

cell dies, the cell and attached crystal may be washed away in the urine flow. The cell









may envelope the crystal and continue to function normally [Lie92, Lie93, Lie94, Koh96]

with no further growth of the crystal. However, if epithelial cell injury has occurred,

basement membrane proteins may be accessible to adhere to COM crystals by sloughing

of the cells leaving behind the exposed underlying architecture, the basement membrane.

The basement membrane is a layer of proteins that underlies the epithelial sheet and will

be discussed subsequently in more detail.

Depending on the extent of injury, other proteins common to the extracellular

matrix found below the basement membrane may also be accessible for attachment to

COM crystals. Khan et al. [Kha84] demonstrated the association of crystals to fibrillar

macromolecular structures by scanning electron microscopy (SEM). Khan [Kha95a] has

also shown crystals passing through the basement membrane into the extracellular space

near the papillary tip as shown in Figure 2.11.

Researchers have shown the attachment of COM crystals to epithelial cells and

fibrous proteins characteristic of those found in the basement membrane and extracellular

matrix, yet the importance of these individual materials in kidney stone formation with

respect to COM crystal adhesion is not known. The necessity to determine the

important materials most responsible for COM crystal adhesion will allow researchers to

focus their interests to those more important materials in the search for a possible cause

and therefore relief from kidney stone formation.


2.4 The Human Kidney

The human kidney contains approximately 1 million nephrons which absorb

nutrients and water back into the body after being filtered out of the blood by the renal

glomeruli [Ham74]. The nephrons consist of tubules through which the waste is

collected. As the waste products travel further down the length of the nephron, the waste

becomes concentrated and supersaturated conditions may exist allowing precipitation of

crystallites, such as COM, to occur. Figure 2.12 shows a nephron.









32









u


0

o

Cd

0
0
c:
-4-3





o
0









0
CO
-o













C!
a
0



ci

S0
o

*M

0


I-
4-,






o v
o











4-





I U
||0
1NI4.
4.5 -i
iu -c
^3













proximal
convoluted
tubule


collecting
tubule /


Figure 2.12. A schematic illustration of a human nephron [Ham74].









The renal tubules are lined with epithelium. Underlying the epithelial cells is a

structure called the basement membrane which is a continuous thin mat of specialized

extracellular matrix that functions as a support, a molecular sieve, and a cell regulator

[Yur90].

The basement membrane consists of the basal lamina and the lamina reticularis.

The basal lamina is also divided into two different sections, the lamina rara and the lamina

densa. The basal lamina is essentially a mat of collagen type-IV with specific molecules

on each side of the mat that help it bind to adjacent cells or matrix materials [Alb89].

Although the composition of the basal laminae varies from tissue to tissue, one of the

molecules always found in the basal lamina is laminin. Another molecule often found in

the basement membrane, in particular the lamina densa [Ino94], is fibronectin. The

structure below the basement membrane is the extracellular matrix, which is primarily

made up of fibrous proteins in a hydrated polysaccharide gel [Alb89]. Collagen type I is

a fibril forming collagen found in skin, tendon, bone, intestine, uterus, and surrounding

organs [Kuc92]. Figure 2.13 is a schematic representation of the renal epithelial cells,

basement membrane, and extracellular matrix showing there relative positions in the tubule

wall.


2.5 Characterization Techniques



2.5.1 Electrokinetic Measurements


Charge characterization at the solid-solution interface is very important when

attempting to discern mechanisms of adhesion. As described in section 2.2.1 of this

chapter, charged materials when immersed in water are surrounded by strongly adsorbed

ions, called the Stem layer, and by a gradient ionic cloud, called the diffuse part of the

double layer, as shown in Figure 2.3. Electrokinetic measurements are useful in

determining the electrostatic potential, often called the zeta potential, i, at the boundary





















































Figure 2.13. A schematic representation of a tubule wall found in a kidney.









between the strongly adsorbed layer and the beginning of the diffuse double layer as

schematically depicted in Figure 2.5. Electrokinetics is the measurement of the movement

of one phase with respect to another phase in which a charged boundary between the two

exists. Two types of electrokinetic measurements, electrophoresis and streaming

potential, are described below.

2.5.1.1 Electrophoresis

Colloidal particles immersed in water can move under an applied electric field as a

result of the charge on the particle surface as shown in Figure 2.14. The electrophoretic

mobility, uE, is defined as the particle velocity per unit static electric field. The mobility

can be determined by measuring the velocity of a particle under an applied electric field.

The mobility is related to the electrical potential at the shear plane surrounding the

particle. The shear plane is defined as the boundary between the bulk solution, where the

ions are free to move, and the inner layer of strongly adsorbed ions which move with the

particle under an applied electric field. The corresponding zeta potential, C, is calculated

from the experimentally determined electrophoretic mobility, UE, according to the

Smoluchowski equation [Hun81]


= tuE (2.5)
DE,



where 77 is the viscosity of the solution, D is the dielectric constant of the

solution, Eo is the permittivity of free space. A number of techniques exist to determine

the electrophoretic mobility under an applied electric field such as a simple optical

method, electrophoretic light scattering method, and electroacoustics to name a few.

2.5.1.2 Streaming Potential

The charge at the surface or the more commonly accepted charge at the Stern

layer, the zeta potential, can be determined by measuring the streaming potential














-'Illl --



+ -
<4 -


4-0


Figure 2.14. A schematic representing charged particle movement in an applied electric
field to determine electrophoretic mobility.


Zb:;









associated with a surface. Streaming potential is the potential generated when a fluid is

forced through a capillary. The hydrodynamic pressure forces the mobile charges in the

double layer in the direction of flow. The accumulation of charge at one end of the

capillary creates an electrical potential across the capillary as demonstrated in Figure 2.15.

In the case of glass or fused quarts, the charge on the surface is negative; therefore, the

mobile charges will be primarily positive causing a positive current in the direction of

flow. The streaming potential can be measured with a high impedance voltmeter as a

function of pressure. If the potential is measured as a function of pressure, the

Smoluchowski equation can be used to calculate the zeta potential [Hun81]


4r=,E' (2.6)
eP


If the usual mixed units are used, the equation is


( 0.1 O (poise) x (~ cm) x 100 E,(mV)
8.854 x10-'2(C )X p(dyn/ 2)



where and E, are the zeta and streaming potentials, respectively, both measured in mV,

P is the pressure drop across the capillary in cm Hg, the conductivity of the elution, ,, is
in units of l-'cm-', and e is the dielectric constant of the elution.

Equation (2.6) is only valid for solutions in which all or almost all of the current

generated due to streaming is carried through the bulk liquid. For solutions having ionic

concentrations less than 10-3 M, the need to account for surface conduction is important

[Rut47]. At low electrolyte concentrations, a large part of the current may be carried

through the double layers near the capillary walls because of the higher charge density in

that region. Equation (2.6) becomes [Hun88]















vDoube Lar (lon
,..\Double Layer Ions (",


0


Neutral ElectrolyteO 0 () ,Accumulation of
0 G 0 0,Olons S C
0OQI OC
0 G Q)" (0 Flow Direction-m 0














Figure 2.15. A schematic representation of the flow of ions under an applied
hydrodynamic pressure which generate the potential across the capillary or streaming
potential.


)









S4,c(A +2A,/r)E,
P = (2.7)



where the term 22s/r is the specific surface conductivity and r is the capillary radius or

the distance between plates for a flat plate system. The variable A, is the conductance of

a square section of material of unit area and constant thickness measured in Q-1. Briggs

[Bri28] suggested a simpler procedure for correcting for surface conduction. The

procedure involves measuring the resistance of the liquid in the capillary at low ionic

concentration, Rexp, and compare with the value of resistance expected from

measurements at high ionic concentration, RcaIc, where surface conduction can be expected

to be negligible. Equation (2.7) becomes


= 4rE R (2.8)
EP Rexp


The general effect of accounting for surface conduction will give an increase in magnitude

of the calculated value of zeta potential at low ionic concentration conditions.

Ball and Fuerstenau [Bal73] performed an extensive review of the streaming

potential literature and found that the measurement should be performed over a

sufficiently wide range of pressures to obtain an accurate estimate of the slope, EJP, the

obtained slope should be linear, and the intercept of the slope should be zero. Ball and

Fuerstenau [Bal73] and Hunter [Hun88, Hun93] have all stressed the importance of the

linear dependence of the streaming potential, Es, on the applied pressure, P. The linearity

of the plot of streaming potential vs. applied pressure is a necessary first step to

determine experimental reliability; however, unless surface conduction is accounted for,

linearity does not ensure accuracy of the resulting zeta potential [Hun88]. The finite

intercept of the slope is an indication of an asymmetry potential or rest potential which is

said to be a function of the electrodes, including their preparation, treatment, and cleaning,









and of the electrolyte and its concentration [Bal73]. No electrode system, Pt, Au,

Ag/AgCI, Ag/AgI, or calomel, has been found to eliminate the asymmetry potential

completely.

Many authors have developed methods for eliminating asymmetry or rest

potentials [Hun62, Mar30, Bul35, Hor77]. Hunter and Alexander [Hun62] found that the

rest potential could be nullified when the liquid was stationary. Martin and Gortner

[Mar30] and Bull [Bul35] avoided an asymmetry potential by working at high applied

pressures where the asymmetry potential is negligible relative to the streaming potential.

Horn and Onoda [Hor77] devised a resistance-capacitance (R-C) circuit to store the

asymmetry potential in a large capacitor and subtract the potential from the streaming

potential when flow begins.


2.5.2 Scanning Probe Microscopy


Scanning probe microscopy (SPM) offers a versatile range of techniques which

can be employed to acquire information about a material surface. Topographical

information can be acquired using Contact Mode Atomic Force Microscopy (AFM),

TappingModeTM AFM, and Non-contact AFM. Other techniques can also acquire

topographical information.

Contact Mode AFM measures topography by sliding the probe tip across the

sample surface in air or in fluid. A laser is focused on the back side of the probe tip also

known as the cantilever as shown in Figure 2.16. The laser reflects from the cantilever to

a mirror onto a photodetector. The sample is mounted on the piezoelectric scanner which

rasters the sample under the tip and surface interactions between the sample and tip cause

the cantilever to deflect. Any motion of the cantilever is registered by the photodetector.

The position of the laser spot is determined by the electronic circuitry which generates a

voltage difference between the photodiode segments and a topographical image is the

result.















Mirrors


Photodiode


Figure 2.16. A schematic diagram of the SPM.


Laser









2.5.3 Scanning Electron Microscopy


Scanning electron microscopy (SEM) is a useful tool to analyze surface

topography of materials. A focused electron beam is rastered across the sample exciting

electrons in the atoms of which the sample is composed. As a result, both secondary and

backscattered electrons are produced from the sample. A detector measures the intensity

of the electrons versus position and displays this information on a cathode ray tube.

Secondary electron contrast is due to the dependence on electron yield on the topography

and the depth of secondary electron emission is about 100 A. Because the sample must

be electrical conductive to be analyzed using SEM, the samples need to be sputter coated

with Au/Pd.


2.5.4 Fourier Transform Infrared Spectroscopy


Infrared spectroscopy (IR) is a useful tool to study polymers and organic

materials as well as inorganic materials. Fourier transform infrared spectroscopy (FT-IR)

uses the principles of interferometry to study bond vibration in materials.

Electromagnetic radiation in the infrared region (7.8x10-5 to 0.1 cm) is passed through or

reflected off the sample. The radiation excites molecular bonds to higher vibrational

levels, absorbing energy. The absorbed energy corresponds to particular vibrational

frequencies characteristic of a molecule or molecular group. Therefore, the IR technique is

useful in determining unknown materials, determining bond orientation, and quantitative

and qualitative analysis of bond types [Mar86].

Attenuated total reflectance (ATR) spectroscopy also called internal reflection

spectroscopy (IRS) may be the most widely utilized adaptation of IR to study inorganic

material surfaces. ATR is performed by coupling the incident electromagnetic radiation

into an IR transparent crystal of high refractive index, as shown in Figure 2.17. When the

beam reaches the interface between the crystal and the materials to be analyzed, it is



























SAMPLE


























Figure 2.17. Multiple internal reflection within the internal reflection element which is
coupled to the sample.









internally reflected. However, 100% internal reflection does not occur and some part of

the wave travels into the sample material to a depth governed by the wavelength of

radiation, the IR crystal, and the refractive index properties of the sample [Knu85].

Figure 2.18 is a plot of wavenumber vs. sampling depth for a thallium bromide-thallium

iodide (KRS-5) IR crystal having a 450 incident angle coupled to a sample assumed to

have an index of refraction of 1.5. As can be seen from Figure 2.18, the sampling depth

increases as function of increasing wavelength (decreasing wavenumber) over the mid-

infrared region.


2.5.5 Particle Size Determination


Many techniques exist to measure particle size. These techniques include

microscopy, sedimentation methods, electrical pulse counting, light scattering methods,

hydrodynamic methods, and electroacoustics. Each of these techniques has its advantages

and limitations, which may be dependent on size, density, optical properties, or other

parameters.

Electrical pulse counting is able to count the number of particles in a known

amount of solution by drawing the suspension through a very small orifice that has an

electrode on either side of it as shown schematically in Figure 2.19 [Hun93]. When a

particle passes through the orifice, interference with the current flowing between the two

electrodes occurs and the resistance changes. The number of changes in current and the

magnitude of change are recorded. Because the change in current is proportional to the

volume of the particle passing through the orifice, the result is a value of particle size

when calibrated against a dispersion of known particle size. The number of current

changes acts as a particle counter generating a particle size distribution.











































I.O


~1.20
a.

O.W
z08


0.40


0
4000 3200 2400 1600 800
WAVENUMBERS (cm'1)


















Figure 2.18. Sampling depth as a function of wavenumber for the KRS-5 internal
reflection element having a 450 incident angle [Knu85].























































Figure 2.19. A schamatic diagram of a Coulter counter, electrical pulse counting
instrument [Hun93].









2.5.6 X-Ray Diffraction


X-Ray diffraction (XRD) is a tool used to investigate the arrangement of atoms in

a crystal. XRD is useful for identifying compounds in crystalline materials, determining

crystallographic information of metals, ceramics, and polymers, and performing

quantitative phase analysis. A monochromatic or near monochromatic beam of x-rays is

focused on a specimen. The incident x-rays are diffracted by the lattice atoms in the

crystal according to Bragg's Law [Cul78]



n = 2dsin9, (2.9)


where n is the order number, A is the wavelength of the incident x-rays, d is the

interplanar spacing, and 0 is half of the angle of diffraction. A scintillation detector is

moved through some angle at a given rate and the number of counts or intensity is

recorded The resulting pattern is a plot of x-ray intensity vs. angle (20) measured with

respect to the incident beam. Because the crystal structure is periodic, the in-phase

scattered electrons create an increase in intensity.















CHAPTER 3
DEVELOPMENT AND CHARACTERIZATION OF MODEL COM PARTICLES


3.1 Introduction

Calcium oxalate monohydrate (COM) is the most thermodynamically stable form

of calcium oxalate and is a major component of human calcium oxalate calculi [Pri47,

Deg81a]. Stone formation in the urinary tract takes place opportunistically with an

unfortunate, simultaneous convergence of factors that may be achieved by several

pathways for a stone to form [Bro92, Ada95]. Many theories have been proposed to

explain the critical factors) responsible for the formation of urinary stones [Bro92,

Ada95, Fin78a, Fin84, Rob86]. Kidney stone disease may take place as a consequence of

injury, supersaturation (particularly hyperoxaluria), nucleation, and growth mechanisms,

which may include adhesion on epithelial cell surfaces and aggregation. All of the

previously mentioned factors may play a role in contributing to stone formation.

Nonetheless, stone formation remains an idiopathic disease with no one risk factor

providing an obvious cause for stones in most cases. However, both the in vivo and

in vitro evidence indicates that the presence of particles is a necessary but insufficient

condition for stone formation. From this perspective, the formation and presence of

crystallites may be a typical renal function directed at concentrating oxalate for more

effective, irreversible elimination from the body.

The abnormal stone-forming condition results from the aggregation and further

growth of such pre-existing crystallites [Fle78, Rya81a, Ada81, Deg91, Edy86, Edy87a,

Har86a, Har86b, Kok90, Rob85, Scu86a, Scu86b]. Indeed, it has been found that while

the solid phase in the urine of normal individuals is made up mainly of individual crystals,









large aggregates are often found in stone former [Fle78]. Therefore, both crystal growth

and aggregation are generally regarded as important steps in the formation of calcium

oxalate renal stones [Fle78, Rya81a]. Numerous studies have shown that the aggregation

and/or adhesion of COM particles may potentially lead to the ultra-structures typical of

kidney stones [Ada81, Deg91, Edy86, Edy87a, Har86a, Har86b, Kok90, Rob85, Scu86a,

Scu86b]. Kok et al. [Kok90] reported that precipitation alone of COM particles cannot

account for the large size particles necessary to cause blockage of the urinary tract.

Aggregation is therefore the only possible mechanism whereby a stone may reach a size

as large as 0.64 cm, the maximum size that can pass through an average adult male urethra

[Fin78b].

Aggregation is the process whereby crystals bind one to another to produce large

clusters. The actual mechanisms) of aggregation within the biophysical environment of

the human kidney is uncertain; however, Adair et al. [Ada95] have classified six

aggregation mechanisms acting alone or in concert which may contribute to stone

formation: (a) coagulation in simple electrolyte solutions; (b) secondary minimum

coagulation; (c) heterocoagulation of particles with different surface charge polarity; (d)

polymer bridging flocculation; (e) secondary nucleation and growth; and (f) flocculation

via immiscible amphiphilic molecules. The previously mentioned mechanisms, which are

summarized in Figure 3.1, stress the importance of knowing the environmental conditions

in the urinary system for formation of calcium oxalate calculi. For example, citrate can

retard the crystallization of stone-forming calcium salts by two broad means [Cha91].

First, citrate complexes calcium and reduces the ionic calcium concentration available for

precipitation in urine, thereby lowering the supersaturation. Second, citrate can directly

inhibit crystallization of both calcium oxalate and calcium phosphate at the growing

crystal-solution interface. Thus, citrate has been shown to inhibit spontaneous

precipitation of calcium oxalate and to retard agglomeration of pre-formed calcium oxalate

crystals. Citrate is believed to also have a modest inhibitory effect against crystal growth



















0

0 0<
c o E,


O )C
8 0 E 3 0

E o





a. 0 o.
8 >



cu g. c

II






C C 4.
U) c




800
+ I 2 I




a e 0


S&E
E : c|
c o



o .o o


So5 0 )- 3
-n c '5
0


Co m
0 0 .
r d =r+o .


1.:.









of calcium oxalate. Cody and Cody [Cod94] reported habit modification of COM

crystals grown with citrate. Interpenetrating crystals grown in distilled water without

citrate have sharply-angled tips and are flattened parallel to the (010) face whereas

crystals grown with citrate have rounded tips and are flattened on the (101) face.

Few papers have dealt with the theoretical equilibrium shapes of COM crystals

synthesized under different conditions or the atomic structure at specific faces [Deg91,

Cod94]. Furthermore, little, if anything, is known at the atomic-molecular level about the
interaction between COM particles and important species such Ca2+, C2042, citrate,

dicitrate, and urea, which are common in the urinary system and known to play a role in

stone disease. Deganello's work [Deg91] is an exception to the lack of data on the

interaction between some species and COM crystals. Adsorption of nephrocalcin, a

protein found by Nakagawa and co-workers [Nak83, Nak87] in human urine, even at the

lowest concentration tested, affects the habit while also inhibiting the growth of COM.

The nephrocalcin protein promotes the preferential development of the (101) faces to

such an extent that the length to width ratio of the crystals decreases by approximately a

factor of three. In the course of this process, the apical planes eventually disappear and

the size of the crystals decrease. Diminution of crystal size eventually becomes extreme

when nephrocalcin reaches a concentration between 1 to 2x 10"5 M.

Mandel [Man94] reported that the molecular surface structures on the prominent

crystal growth faces of COM are very important in determining potential long-range

molecular bonding interactions between the crystals and the lipid-rich regions in the inner

medullary collecting duct (IMCD) cell plasma membrane. Mandel [Man94] also reported

on epitaxial matching calculations for the high temperature form of COM, calcium oxalate

dihydrate (COD), hydroxyapatite (HA), and uric acid (UA) crystal lattices against the

dimensional repeating lattice of phospholipid headgroup structures. The Mandel study

suggests that COD-COM epitaxy may be important relative to COD-lipid interactions in

crystal attachment and stone development. Therefore, the molecular modeling between









specific surfaces of the COM crystal and these additives provide important insight for

specific adsorption sites on COM surfaces, and the growth and aggregation mechanisms

of COM crystals in the urinary system. Other research suggesting the crystallographic

importance of COM was performed by Habeger et al. [Hab97]. A crystallographic

dependence of COM crystals on adhesion to fused quartz glass substrates was observed.

A variation in electrostatic charge as a function of crystallographic habit on the theoretical

COM surface has been used to justify the differences in adhesive strength observed

[Hab97].

In the current work, morphological forms of COM crystals synthesized in

different conditions have been studied by using the commercial computer programs,

SHAPE and ATOMS0. The computer program SHAPED has been used to determine

the face indices of the synthesized COM crystals and generate the equilibrium shape as a

function of central distance. The computer program ATOMS0 has been used to generate

the surface atomic structure of COM crystals as a function of habit plane.

Certain characteristics in a particle system are desirable to rigorously study

theoretical growth and aggregation, and adhesion mechanisms. Ideally, the particles

should be spherical. If spheres cannot be produced, a crystallographically definable

morphology should be produced so that the habit planes can be identified. The

theoretical study of growth and aggregation depends on well-defined particle

morphologies. Unfortunately, COM has a monoclinic crystal structure whose inherent

asymmetry does not lend itself to the formation of spherical particles. Therefore, the

individual COM particles should be at least uniform with narrow particle size and shape

distributions. Several different particle size ranges is also desirable. Thus, the objective

in the current work is to prepare COM particles by several different methods followed

by careful characterization to deduce the habit planes and atomic surface structure.

Lastly, characterization of the particles having the most consistent COM morphology

will be conducted.









3.2 Preparation and Characterization of Calcium Oxalate Monohydrate Particles under
Different Conditions

Reagent grade chemicals were used in all precipitation studies without further
purification. All solutions were prepared with deionized water (specific conductivity less

than 10 gmho.cm'') which was provided by a Milli-Q high purity system (Millipore

Corporation, Bedford, MA) and were passed through a 0.2 [jn filter. For identification

purposes, precipitated COM particles were designated by the negative logarithms of the

initial Ca+2 and C204-2 molar concentrations, respectively (e.g., particles prepared from

10-3 M Ca+2 and 10- M C204-2 were designated "33" particles).


3.2.1 COM particles ("33" particles) without seeds

Reagent grade potassium oxalate monohydrate (K2C204.H20) (Fisher Scientific
Inc., Fair Lawn, NJ) was used to prepare stock 1 molar (M) K2C204 solution at 250C.

After equilibration, stock solutions were passed through a 0.22 gim filter (MAGNA,

nylon, supported, plain, MSI, Westboro, MA). Reagent grade calcium chloride dihydrate

(CaC12.2H20) (Fisher Scientific Inc., Fair Lawn, NJ) was used to prepare 0.01 M CaC12

solution at 250C. The CaC12 solution was prepared the day of use to prevent CaCO3

formation through reaction with CO2 from the atmosphere. Diluted solutions of K2C204

and CaCl2 solutions were volumetrically prepared from the stock solutions.

A schematic representation of the processing steps for the preparation of COM
crystals ("33" particles) without seeds is given in Figure 3.2. Precipitation studies were

performed by mixing equal volumes (10 liter (I)) of 10-3 M CaC12 and 10-3 M K2C204

solutions at 250C. Crystals of COM were aged for at least 24 hours without stirring.
Solutions containing precipitated crystals were 0.22 gm filtered. The filtrate was washed

with a solution saturated with COM to prevent dissolution of the precipitate during
washing. After washing, the recovered powders were freeze dried (Freeze Drier 4.5,

Labconco Corp., Kansas City, MO) and stored in a desiccator.

















10-3 M CaCI2 Solution (10)


10-3 M CaCI2 Solution (10)


Aging
(for at least 24 hours without stirring)




Recovery of COM particles
(concentrate the particles in the centrifuge)


Washing & Filtering
(washing by saturated COM solution with
filtering through 0.22 gm)


Freeze-Drying
(store the particles in a desiccator)




Characterization
(XRD, SEM, and particle size analysis)


Figure 3.2. Schematic representation of processing steps for the preparation of COM
crystals ("33" particles) without seeds.











3.2.2 COM particles with seeds

Seed crystals ("00" particles) of COM were prepared by mixing equal volumes

(50 ml) of 1 M CaCl2 and 1 M K2C204 solutions at 250C. These seed crystals were aged

for at least 24 hours without stirring. Crystals were concentrated by centrifugation and

then washed with saturated COM solution. After washing, the recovered powders were

freeze dried and stored in a desiccator. The seed suspension was prepared by dispersing

0.1 g of seed crystals in 5 ml isopropanol for a crystal growth experiment. The

dispersion was treated by ultrasonication for 3 minutes to break up agglomerates.

A processing schematic for the preparation of COM crystals ("33" particles) with

seeds is given in Figure 3.3. Precipitation studies were performed by mixing equal

volumes (500 ml) of 10-3 M CaCl2 and 10-3 M K2C204 solutions at 250C with various

amounts of seed crystals. Crystals of COM were aged for at least 24 hours without

stirring. Crystals were filtered 0.22 rm and then washed with a solution saturated with

COM. The recovered powders were freeze dried and stored in a desiccator.


3.2.3 COM particles precipitated from homogeneous solution at 900C

Gordon, Salutsky, and Willard have reported a recipe for the production of COM

by the thermal decomposition of dimethyl oxalate [Gor59]. The goal of this section was

to duplicate this procedure and analyze the nature of the particles. Ammonium acetate

(CH3COONH4) (Fisher Scientific Inc., Fair Lawn, NJ) and acetic acid (CH3COOH)

(Fisher Scientific Inc., Fair Lawn, NJ) were used to prepare stock 2.5 M CH3COONH4

(aq) and 2.5 M CH3COOH (aq) solution at 250C. A buffer solution at pH 2.7 was
prepared by mixing equal volumes (500 ml) of the 2.5 M CH3COONH4 (aq) and

CH3COOH (aq) stock solutions at 250C. After equilibration, stock solutions were

passed through a 0.22 gim filter.






































Recovery of COM particles
(concentrate the particles in the centrifuge)


Washing & Filtering
(washing by saturated COM solution with
filtering through 0.22 rnm)


Characterization
(XRD, SEM, and particle size analysis)


Figure 3.3. Schematic representation of processing steps for the preparation of COM
crystals ("33" particles) with seeds.









A schematic representation of processing steps for the preparation of COM

crystals aged at 900C is given in Figure 3.4. Reagent grade CaC12*2H20 was used to

prepare 0.01 M CaCI2 solution at 25C. The CaC12 solution was prepared the day of use

to prevent CaCO3 formation through CO2 from the atmosphere. The dilute CaCl2

solution were volumetrically prepared to obtain 2.5xl0-3 M CaC12 solution. The pH of

the dilute CaC12 solution was adjusted to 4.7 by adding 0.01 M HCI solution. After

equilibration, the dilute CaC12 solution was filtered (0.22 Wm). The 2.5x10"3 M CaCl2

solution (150 ml) was rapidly mixed with buffer solution (100 ml) in a flask and 10 g of

dimethyl oxalate (CH30COCOOCH3) (Fisher Scientific Inc., Fair Lawn, NJ) was added.

The flask was tightly closed and heated at 900C for 1 hour in an oven to equilibrate. The

holding time at 900C was 2.5 hours. After thermal treatment, the solution was rapidly

cooled to room temperature in an ice bath. Crystals were collected via centrifugation

followed by washing with saturated COM solution previously passed through a 0.22 gtm

filter. After washing, the recovered particles were freeze dried and stored in a desiccator.


3.2.4 Characterization


The dried, recovered powders were analyzed for phase composition using x-ray

diffraction (XRD) (APD3720, CuKa, fine tube, 40kV-20mA, Philips Electronics,
Mahwah, NJ) over a 20 range from 10-70 at rate of 2.4'/min.

The morphology of the synthesized crystals were observed using scanning

electron microscopy (SEM) (JSM 6400, JEOL, Boston, MA).

Particle size analysis of typical COM particles without seeds was performed

using an electrical sensing zone technique (ELZONE 80XY, 95 gtm aperture, Particle Data

Incorporated, Elhurst, IL).

Zeta potential determination was performed using Rank Particle

Microelectrophoresis (Apparatus Mark II, Rank Brothers, Cambridge, ENGLAND) as a











0.01M CaCI2 Solution (100ml)

---- Dilutions


2.5 10-3 M CaCI2 Solution (150m)


Aj ustmeat of pH with HCI
(pH = 4.7)


Filtering (0.22gpm)


2.5M CH3COONH4 Solution (50ml)

2.5M CH3COOH
Solution (50ml)

Buffer Solution (100ml)


Filtering (0.22pm)

ZZ


Precipitation
(heat treatment in an oven at 90C
heating time: Ihour
holding time: 2.5 hours)


Cooling
(rapidly cooled to room temperature
by using an ice bath)


Recovery of COM particles
(concentrate the particles in the centrifuge)


Washing & Filtering
(washing by saturated COM solution with
filtering through 0.22pm)


Freeze-Drying
(store the particles in a dessicator)


Characterizat ion
(XRD, SEM, and particle size analysis)


Figure 3.4. Schematic representation of processing steps for the preparation of COM
crystals ("32" particles) with heat treatment at 900C.


I









function of pH in COM saturated solution and in an artificial urine ion solution (AUIS)

buffered to pH 6.0 at 370C. The composition of the AUIS is given in Table 3.1.

Scanning probe microscopy (SPM) (MultiModeTM, Digital Instruments, Santa

Barbara, CA) was performed on the two major crystallographic surfaces of COM to

determine surface roughness using a Nanoscope III controller. The J-type scanner was

operated in contact mode using a Si3N4 cantilever having a spring constant of 0.12 N/m in

saturated COM solution at 250C.


3.2.5 Crystal and Atomic Structure Modeling Using the Computer Programs SHAPE
and ATOMSc


Based on crystallographic data including space groups and lattice parameters, the

computer program SHAPED (Macintosh version 4.0, SHAPE Software, Kingsport TN)

was used to determine the face indices of the COM particles synthesized under various

conditions and to generate the equilibrium shape of the crystals as a function of the

central distance [Coc61]. The central distance is the perpendicular distance from the

center of the crystal to the faces of the corresponding form. The greater the distance, the

less prominent the form (the smaller the area of the faces of that form in the final shape).

Form factors in the program are the least number of face indices needed to generate the

desired equilibrium shape. Twins of COM particles were drawn using the "twin option"

and "epitaxy option" in SHAPE.

The theoretical atomic structure of COM was displayed using the computer

program ATOMS (Macintosh version 2.0, SHAPE Software, Kingsport TN). The slice

option of the computer program ATOMS was used in the process of analyzing and

predicting surface structure as a function of habit plane. To draw an individual surface

atomic structure using this option, it is necessary to enter the crystal class, the space

group, the corresponding unit cell parameters, and the face indices and atomic coordinates.

From the crystal class and space group, the program determines what symmetry





















Table 3.1. Composition of artificial urine ion
solution

Compound Solution
Concentration (M)
NaCI 0.10554
NaHPO4*2H20 0.03654
Na3C6H507-2H20 0.00321
MgSO4 0.0038476
Na2SO4 0.016952
KC1 0.06374
CaC12-2H20 0.001
Na2C204 0.00131
NH4Cl 0.03632
NH40H 0.00062
NaOH 0.005









operators to use in the calculations. In addition to the indices of the face and input atoms,

several other parameters were specified. The thickness of the slice is in fractions of the d-

spacing. Thus, atoms present to a certain depth of unit cell were drawn with the "slice

option" as a function of habit plane based on the atomic coordinates of the high and low

temperature structures.


3.3 Results and Discussion

First, differing reports regarding lattice parameters for COM need to be reconciled.

Table 3.2 summarizes crystallographic data proposed by various authors [Deg81a,

Taz80, Coc61, Coc62, Deg80]. From Table 3.2, the unit cell parameters reported by

Cocco [Coc61] and Tozzoli and Domeneghetti [Taz80] can be seen to have similar values

in contrast to the unit cell parameters of the a and c axes reported by Deganello and Piro

[Deg8la]. The equilibrium structures of COM were generated the crystallographic data

proposed by Deganello and Piro [Deg81a] and Tozzoli and Domeneghetti [Taz80], as

shown in Figure 3.5. Based on comparisons made between the theoretical equilibrium

shape and the morphological form of the experimentally derived crystals, the equilibrium

shape of COM crystals modeled using the lattice parameters of Deganello and Piro

[Deg81a] can be more easily reconciled to the morphology of the experimentally

synthesized COM crystals.

It is also necessary to mention conflicting papers regarding COM atomic

coordinates. Table 3.3 shows atomic coordinates proposed by various authors [Deg81a,

Taz80, Coc62]. From this table, the atomic coordinates reported by Cocco and Sabelli

[Coc62] and Tozzoli and Domeneghetti [Taz80] have similar positions but only the x

coordinates for carbon(l), carbon(2), oxygen(l), oxygen(3), and calcium(l) are different.

The atomic coordinates reported by Deganello and Piro [Deg8 a] are quite different from

those of the previously mentioned authors. The equilibrium atomic structures of COM

were generated using the three different atomic coordinates. Based on the comparison































Cl e!

,It- C
m 00






0 0 0






-








N -N 0
0








00

00 00 00
0 m











u uo
Cl Cl ON


o o
oo








02
S 00 0
U H

o








o c
SH a
0 8 <
U



































(B) (C)
A D

V A: (Oil)
B: (0 11)
C: (010) c
D D: (Oi)
E: (101)
E F F: (0T) E










Figure 3.5. (A) SEM photomicrograph and (B) and (C) theoretical equilibrium shapes of
the COM crystals precipitated without seeds. (B) The equilibrium shape of COM
crystals based on the crystallographic data of Deganello and Piro [Deg81a] and (C) the
equilibrium shape of COM crystals based on the crystallographic data of Tazzoli and
Domeneghetti [Taz80].














0- Uin o t-(ON 0 ON 00
m m In e r- V SC\cM Inoo 0 o 0 c \


o) N m- -: MN Cq q 0t
06609 9669669050 0
ad d Nddddddd d




C> r o [I e-: c14 M -o r-
6 6do dd0 dd0d0d od dV



O ON en N W) W) o^ r- >n o "t W) C1 4
X^ 0c, rC0 M W) m c (^ ,o oO < ^ W cmn 0
( mM-c en o II Co0vso It o
o666666666 66 66 000



S q C4 C n 0 0 Nt r-- t \C r1- N 00
0 oo0 m cn W) o O a\ CM Wm O o C -
9 M M cn m 0 CM o 0t m It n
dddd dddddddd dd d6 oo0


0 o CN000(100 -C.00N m o oN O 1






N r- t- tn -co tcn m- 0nc W) -
0 >,0000 00000000 00 C CC o't00 \ 0



8 8
rf--. mo> .e ..



0 C5 O\ 5 5C CO\5 -ir 0 6 C5* )
oo^ in 2 5-> r- 0 SO C c,4 4 O \ c 3-, m :700- l


o3 cc o-m r-- A -o .(mCm-m-n (- 0



o N rdd d cddddd dd dd dd



o t
CMO0or~- Cr~-tso sr~-r- \0






Soco 00000000 C 0





a uuou c ocococo ou >> x
m cu c ~ ----------------------------- ------









between the theoretical atomic structures generated using ATOMSc and the atomic

structures from the papers by Deganello and Piro [Deg81a] and Tazzoli and

Domeneghetti [Taz80], the (010) and (100) planes generated with the atomic coordinates

published by Cocco and Sabelli [Coc62] are exactly reconciled to the layering sequence

reported by Deganello and Piro [Deg81a] and Tozzoli and Domeneghetti [Taz80] as

shown in Figure 3.6(A) and (B). However, the (010) and (100) planes generated with the

atomic coordinates by Tozzoli and Domeneghetti [Taz80] are not reconciled to the

layering sequence reported by Deganello and Piro [Deg81a] and Tozzoli and

Domeneghetti [Taz80]. The atomic structures of some oxalate ions are incomplete as

shown in Figure 3.6(C) and (D). Furthermore, the (010) and (100) planes generated with

the atomic coordinates of Deganello and Piro [Deg81a] are completely different from the

layering sequence reported by Deganello and Piro [Deg81a] and Tozzoli and

Domeneghetti [Taz80] as shown in Figure 3.6(E) and (F). Thus, the atomic coordinates

of Cocco and Sabelli [Coc62] were used in computer simulations of atomic structure as a

function of habit plane.


3.3.1 COM particles ("33" particles) without seeds

The COM crystals grown in distilled water without seeds are shown in 3. 5(A).

The interpenetrating crystals have sharply-angled tips and a distinctive shape. Figure

3.5(B) shows the equilibrium shapes of these interpenetrating twins with labeled face

indices. For the twin operation, the individual crystal is reproduced according to the

reflection on (101). The crystallographic form factors and central distances for the

interpenetrating twins used in SHAPED calculations are listed in Table 3.4.

Determinations made from observations of the computer calculation and the

morphological form of the experimentally derived particles, the {010} and { 101} faces

appear to be dominant.


















(A) (B)


O: Hydrogen : Water Oxygen : Carbon : Calium
Figure 3.6. Theoretical atomic structures of COM crystals as a function of habit plane:
(A) and (B) are the (010) and (100) planes, respectively, drawn using Cocco and Sabelli
[Coc62] atomic coordinates; (C) and (D) are the (010) and (100) planes, respectively,
drawn using Tazzoli and Domeneghetti [Taz80] atomic coordinates; (E) and (F) are the
(010) and (100) planes, respectively, drawn using Deganello and Piro [Deg81a] atomic
coordinates.


























0 0 kn)


C,

a.
4-


0

to
,-c
o



4a





7-
0







4-
o
M

U










40

Il


4-
ri
o
0










4--






uO0
(6
E^


o o 0o i


O 2
0-


cS

c ,r >.





-. U


o c)o -o o





to
Uo 4






R C
"2 o

8 0 0






CIS -
el e .)

8 c -







0 o+-e 0 -C
4- <-






So


.5u c
4 .
UU U 0
R &^-Si
Uc < -3 2
o b) o-o'G
rr i u
Css. ^


0 -~ I-
--
00-r









3.3.2 COM particles ("33" particles) with seeds


The seed crystals ("00" particles) precipitated by mixing equal volumes (50 ml) of 1 M

CaC12 and 1 M K2C204 solutions are shown in Figure 3.7(A). The particle size varies

between 100-500 nm. Although the shape and size are not uniform, the crystals are

composed of submicron size particles suitable for seed materials. Figure 3.7(B) shows

the equilibrium shapes of the seed crystals with face indices as calculated by SHAPE

with the crystallographic form factors and central distances for COM seed crystals listed

in Table 3.4. The {10O } faces appear to be dominant based on the computer calculation

and the morphological form of the experimentally derived particles. Also, only the

equilibrium shape generated with the (001) face was reconciled with the morphology of

experimentally prepared seed crystals as shown in Figure 3.7(B).

Figure 3.7(C) shows that the influence of seed particles on the size of COM

crystals grown in distilled water. These interpenetrating crystals produced with 0.1 ml of

seeding suspension have similar shape to the COM crystals grown without seeds, but are

smaller in size. The maximum size of the particles is about 1 nm. Thus, it is

demonstrated that the presence of seed particles may provide a low energy epitaxial

surface in solution to lower the overall surface energy contribution to the nucleation

barrier, increasing nucleation frequency, and reducing the particle size of COM crystals

without changing the particle morphology. Also, the seeding studies suggest the

possibility of controlling the size of the COM crystals ("33" particles) by controlling the

amount of seed materials. From the morphological difference between seed crystals and

COM crystals synthesized with seeds, the effect of seed crystals can be gauged from two

aspects: (1) the presence of seed crystals in the solution system probably lowers the

surface energy barrier of the system to the nucleation and growth of particles, and (2) the

transformation in this case is not only influenced by the number of added seed crystals,

















s~C's
-I-- o
0000e-0




o~~ 4l 1)r




rJ u


00













cn


-ecn
o~t

















to 0
c 1_. I I


(/2





9 E! 4-e c




o u
lii
2~b









but also by forming larger numbers of nuclei due to the seeds and possibly secondary

nucleation [Nyv85, Ran88].


3.3.3 COM particles precipitated from homogeneous solution at 90C

Habit modification of COM crystals grown with structurally specific additives

gives further demonstration of specific preferential binding. The COM crystals which

were made using dimethyl oxalate decomposition from 2.5xl0-3 M CaC12 solution at 900C

are shown in Figure 3.8(A). Habit modification is obvious for the COM crystals grown

by dimethyl oxalate decomposition.

Figure 3.8(B) shows the equilibrium shapes of an individual COM crystal and

contact twin with face indices. The crystallographic form factors and central distances for

COM crystals are listed in Table 3.4. For the contact twins, the individual crystal is

reproduced according to the epitaxial operation in SHAPEc. The host crystal is specified

by the (001) face and the [010] vector and the guest crystal is specified by the (001) face

and the [010 ] vector. The central distance for the epitactic face (001) is 2.0. The {101 }

faces appear to be dominant when comparing the computer calculation to the

morphological form of the experimentally derived particles. Also, the equilibrium shape

generated with the (001) face more closely resembles the experimentally prepared COM

crystals in contrast to the equilibrium shape generated with the (010) face, as shown in

Figure 3.8(B) and (C).


3.3.4 Atomic structures of COM particles as a function of habit plane for the high
temperature form

Equilibrium atomic structures of COM particles were generated as a function of habit

plane to predict the atomic structures of specific surfaces. In whewellite, there exist two

different crystal structures: a high temperature structure (stability range 318-415 K) and a

low temperature structure (stability range 293-318 K) [Deg8lb, Deg80]. Table 3.5 and













I- ci1- -'
o I-
-oo@00
- '-





u


-0008


Kf


a 0

0
I00
7 3


8cl











c)
nat




U o


0 ^
c,

a 0








U0 -*'
" -- E5










0U




0.

0 ca-
0 CO *-
^3"7 s
CoT


V






















Table 3.5. Values of the crystallographic data for the high temperature
form [Deg81b, Deg80] of Whewellite (CaC204 H20)

High Temperature Form
a (A) 9.978(1)
b (A) 7.295(1)
c (A) 6.292(1)
Space Group 12/m
Stability Range (K) 318-425
Ca/Ox Ratio
(010) 0.958
(101) 1.658









Table 3.6 summarize the crystallographic data and atomic coordinates of the high

temperature form. Based on the crystallographic data and atomic coordinates, Figure 3.9

shows the equilibrium atomic structures of specific habit planes for the high temperature

form. Habeger et al. [Hab97] has shown that the ratio of calcium atoms to oxalate

molecules may have an important implication in the adhesion properties of the different

crystal faces in a model environment. The ratio of calcium ions to oxalate ions is 0.958 in

the (010) plane whereas the ratio of calcium ions to oxalate ions is 1.658 in the (101)

plane for the high temperature form. Although the temperature in the human body cannot

produce the high temperature phase of COM, the adsorbtion of ions or molecules may

stabilize the high temperature structure at lower temperatures. The stabilization of the

high temperature structure of COM in the urinary system may have a large implication in

that the (101) has a Ca+2/C204-2>1, making the crystal face positively charged and able to

electrostatically interact with predominately negatively charged biosurfaces found in the

human body.


3.3.5 Characterization of COM crystals


The morphologies reported herein were determined to be phase pure COM by

XRD. A typical XRD pattern is shown in Figure 3.10.

As shown by SEM of the various COM morphologies, the "33" particle

synthesizes without seeds were the most consistent morphology produced. Visually,

they had very few differently shaped particles and a relatively narrow size distribution

relative to the other particle morphologies synthesized. Therefore further

characterization will only involve the "33" COM particles synthesized without seeds and

further mention of COM particles refers only to this particular morphology.

Peakfit software was used to fit a log normal curve to the experimentally produced

particle size distribution. The particle size distribution was determined using an electrical

zone sensing technique which results in an equivalent spherical diameter vs. frequency



















Table 3.6. Atomic coordinates for high temperature
form [Deg81 b] of Whewellite (CaC204 H20)

High Temperature Form
Atom x y z

C(1) -0.0652 0.0 0.3984
C(2) 0.0 0.3937 0.0

0(1) 0.1105 -0.3157 0.1096
0(2) -0.0493 0.0 0.2119
0(3) -0.1783 0.0 0.4454

Ca(l) 0.1903 0.0 0.1751
Ca(2)

W(1) -0.3552 0.0 0.0438

H(1) 0.45 0.0 0.04
H(2) -0.33 0.05 0.2




























S: Water : Oxygen : Carbon


O : Calcium


Figure 3.9. Theoretical atomic high temperature structure [Deg81b] (stability range: 318-
415 K) of COM crystals as a function of habit plane, (A) (010) plane and (B) (10 ).


PDO "3DO* S.3G er

0%
0% 0%1 4 re
y y y8"ae4~
X.o


*^ ii *


0 :Hydrogen




















0.81

0.64

0.49

0.36

0.25

0.16

0.09

0.04

0.01


10.0


100.0
80.0:
60.0:
40.0:
20.0
10.0


C2Ca04.H20


50.0


60.0 70.0


WHEWELLITE, SYN
20- 231


60.0


70.0


Figure 3.10 A (top) typical experimental COM x-ray diffraction pattern and (bottom)
the JCPDS file for the mineral whewellite, CaC204*2H20.


50.0


20.0


30.0


40.0


20.0


.L I


, 1 A I -


30.0


40.0


S I,,









distribution as shown in Figure 3.11. The mean and standard deviation in particle size is

13.04 gmn and 1.27 pm, respectively. The least square regression coefficient of the curve

fit is 0.98.

For purposes of calculation, the relative dimensions of the COM particles were

measured by analyzing multiple particles from SEM photomicrographs. The relative

sizes illustrated in Figure 3.12 were used calculating surface areas of the two major

crystallographic faces given the equivalent spherical diameter, refer to Appendix A.

Zeta potential of the COM crystals was determined as a function of pH in

saturated COM solution as shown in Figure 3.13. Also on Figure 3.13, one data point

representing the zeta potential of COM crystals in 10% AUIS at the buffered pH of 6.0.

The zeta potential values for COM in saturated COM solution and in 10% AUIS

at pH 6 are -32 mV and -21 mV, respectively. As can be seen from Figure 3.13, the lower

ionic strength, 5x10-6 M, COM saturated solution has a higher zeta potential than the

higher ionic strength, 3.71x10-2 M, 10% artificial urine at pH 6. This effect is expected

and is due to compression of the electrical double layer.

As the ionic strength of the electrolyte the COM crystals are suspended in

increases (i.e., 10% AUIS to 50% AUIS to 100% AUIS), the value of zeta potential will

decrease. When the ionic strength becomes high, the measurement of zeta potential using

Rank Particle Microelectrophoresis becomes impossible. Compression of the electrical

double layer is extreme and the potential distribution near the particle surface drops

rapidly as a function of distance into the solvent, Figure 2.5. Therefore the particles do

not move under the applied electric field and measurements of electrophoretic mobility

cannot be made.

Contact mode scanning probe microscopy (SPM) was performed on both the

(010) and (101) crystallographic faces of a COM particle in saturated COM solution at

250C to determine the surface roughness. The r, values or the mean roughness values for



























250


200


150



100



50


Equivalent Spherical Diameter (rnm)













Figure 3.11. The differential frequency vs. equivalent spherical diameter particle size
distribution of the experimentally produced COM crystals grown without seeds fit to a
log-normal probability distribution.


......_.._______ _








































A:B:C=1.0:0.43:0.35


Figure 3.12. The relative linear dimensions of experimentally produced COM crystals.


























-10

-20

-30

-40

-50


-60
3 4 5 6 7 8 9 10
pH














Figure 3.13 Zeta potential as a function of pH for COM in saturated COM solution and
zeta potential at pH 6 for COM in 10% AUIS.


S COM Saturated Solution
& 10% Artificial Urine






A\

s^}i









the COM (010), Figure 3.14, and (101), Figure 3.15, are 91nm and 81 nm, respectively,

over a scan area of 5 nm by 5 glm. The high surface roughness of the COM crystal may

be due to the growth mechanism of the twinned COM crystal. The crystals have a

growth ledge at the twin boundary where the crystal may grow more rapidly on one half

of the twin boundary but not on the other causing the roughness to be large across this

boundary.


3.4 Conclusions

The equilibrium shapes with face indices of calcium oxalate monohydrate (COM)

particles synthesized in different conditions have been successfully generated using the

computer program SHAPEC. Computer calculations of theoretical crystallographic

shapes have been reconciled to the observed shapes of experimentally synthesized COM

particles under various conditions. Also, the seeding studies suggest that it is possible to

control the size of COM crystals ("33" particles) by controlling the amount of seed

materials. The computer program SHAPED was demonstrated to be a useful tool to

determine the face indices of COM crystals. The equilibrium shape of COM crystals

based on the lattice parameters of Deganello and Piro [Deg8 la] is reconciled to the

morphology of the experimentally synthesized COM crystals from the comparison

between the calculated shape and the morphological form of the experimentally derived

particles.

Based on the face indices of the equilibrium shapes and atomic coordinates, the

atomic structures of COM particles as a function of habit plane have also been generated

by using the computer program ATOMSc. From the comparison between the theoretical

atomic structures generated by ATOMS, the (010) and (100) planes generated with the

atomic coordinates published by Cocco and Sabelli [Coc62] are reconciled to the layering

sequence reported by Deganello and Piro [Deg81a] and Tozzoli and Domeneghetti

[Taz80].



























-15.0


10.0


Box Statistics


2 range 507.09 nN
Mean 5.475 nm
Rns CRq) 91.527 nM
Mean roughness (Ra) 31.592 nM
Surface area
Box x dimension 4.638 pU
5.0 Box y dimension 4.374 uM


5-.O 0
15.0 imn


Figure 3.14 Contact mode SPM scan of the COM (010) crystallographic face showing
the surface roughness.


0 5.0 10.0




Full Text

PAGE 1

MEASUREMENT OF ADHESION BETWEEN CALCIUM OXALATE MONOHYDRATE AND MODEL SURFACES USING A DYNAMIC WET CELL By CRAIG F. HABEGER 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 1997

PAGE 2

This work is dedicated to my loving and supportive wife Nicole Habeger my parents James and Sandra Habeger and m y brother and sister niece and nephew Larrin Habeger and Bonnie Miranda and Ralph Isaac Testa You are all the greatest!

PAGE 3

ACKNOWLEDGMENTS I would like to thank all of the people who made this research possible. Sincere thanks are extended to Dr. James H. Adair my committee chair and advisor, for his guidance and patience. I would like to recognize the other members of my committee Drs. Christopher D. Batich Anthony B. Brennan and Brij M. Moudgil from the Department of Materials Science and Engineering and Dr Raymond L. Hackett from the Department of Pathology All of these gentlemen contributed to the work reported herein and I am grateful for their advice. I would also like to thank Dr. Saeed R. Khan from the Department of Pathology who was not lucky enough to be on my committee but graciously gave me advice on my research anyway. I appreciate the advice given to me by Dr. John J. Mecholsky Jr. from the Department of Materials Science and Engineering and his willingness to substitute for Dr. Brennan during my defense I would like to acknowledge the National Institutes of Health grant number POG 5P01 DK20586-l 7 which supported the work reported herein I would like to acknowledge the staff of the Major Analytical Instrumentation Center (MAIC) for valuable advice discussions and training. I would also like to thank Paula Scott Pat Glenton and Karen Byer for all of their assistance throughout my research. I am grateful to all of my friends those in Dr. Adair s research group and those not in the group. When I broke my leg you showed me what friends really are Special thanks are extended to Drs. Robert E. Chodelka, Malanie L. Carasso Tuo Li and soon to be Drs. Jeffrey A. Kerchner, Nelson S. Bell, and Paul A. Demkowicz for their fruitful intellectual conversations but mostly for their lousy card playing ability. lll

PAGE 4

I am indebted to my wife Nicole who nursed me through a broken leg twice and many other aches and pains too numerous to develop upon She pushed me to finish however at times not hard enough I would like to thank my entire family for being so supportive. Finally but most importantly, I would like to thank my personal Lord and Savior Jesus Christ without whom none of this would be possible. IV

PAGE 5

TABLE OF CONTENTS ACKNOWLEDGMENT ... .............. ............................ ....... . ..... . .......................... .... ...... ii LIST OF TABLES .............................. ...... ..................... ........... . ..................................... viii LIST OF FIGURES .............. .............................. .................................. .......... ......... ............ x ABSTRACT .. ................... ....... ........................................................ ..... .......... .............. xvi CHAPTERS INTRODUCTION ..... .... ...... ........................................ ..................... .................. .......... 1 1.1 Introduction ....... .............. ............... ...... ....... ........... .............. ...... .................. .... ... .. 1 1 2 Literature Review .............. .......................................................................... .......... .. 1 1.3 Development and Characterization of Model COM Particles ........... ..................... 2 1 .4 Evaluation of the Calcium Oxalate Mono hydrate Hamaker Constant.. ............ ...... 3 1.5 Development of the Dynamic Wet Cell and Streaming Potential Measurements .... ............ ................................................................. ...... .. ....... ....... 3 1.6 Measurement of COM Adhesion to Macromolecular Substrates ............ ........ ..... .4 1 7 Conclusions and Future Research ................... ........... ....................... ............. ...... .4 2 LITERATURE REVIEW ............................................. ....... ................ ..... ..... .......... ....... 5 2.1 Introduction ... ....... ...... ........ ....... ................................................. . ....... ..... .... ........ 5 2.2 Particle Interactions .......... ................................................................... ................ .... 7 2.2 1 Electrostatic Repulsion .............................................................................. .... 9 2.2.2 Attraction .... .... ............... ............................................................................. 14 2.2.4 Interaction Energy .......................... ............. ............................................... 15 2.3 Aggregation Mechanisms Among Particles and Particles at Surfaces ............ ...... 17 2.3.1 Particle Aggregation in Simple Electrolytes ........... ...................................... 19 2.3.2 Secondary Minimum Coagulation ................................ . ............................. 19 2.3.3 Heterocoagulation .... ..... ....... .................... .... ..... .... .................... ........... . ....... 20 2.3 .4 Polymer Bridging Floccu lation ...................................... ................... ........... 23 2.3 5 Flocculation of COM Particles by Phospholipids and other Intercellular Substances ............................................................................... 26 2 3.6 Adhesion of Particles at Surfaces .............. .... ......... ...................................... 26 2.4 The Human Kidney ....... .................................... .... .... ................. .......................... 31 2.5 Characterization Techniques ....... ........... ...... .............................. ................ .......... 34 V

PAGE 6

2 5 1 Electrokinetic M e a s urements ...... . ............................... . ......... ............... .... 34 2.5.1 1 El e ctrophoresis ....... .... .... . .... .... ... .... .......... . . ................. ....... ....... .... 36 2 5.1.2 Streaming Potential.. .... . . ................................... ........ . .. .............. .... 36 2 5.2 Scanning Probe Microscopy .... ... .... ....... .......... ....... .................... ...... . .... .41 2.5 3 Scanning Electron Microscopy .... .. . ............. .. ..... .......... ...... .......... .... .... .43 2 5.4 Fourier Transform Infrared Spectroscopy . . ............ ... .... ......... ........... .. .. .43 2.5.5 Particle Size Determination .... ............. . .... ..................... .................. ...... . .45 2.5.6 X-Ra y Diffraction . .... . .... ..... ........................ ......... ........ . ..... ............. .... .... 48 3 D E VELOPM ENT AND CHARACTERIZATION OF MOD E L COM PARTICLES ....... . ... .. .... ..... ..................... ............. ......... .......... ...... .... ....... ........ ... ... 49 3 1 Introduction ........... .......... . ............................ .... ...... ................. ...................... ... 49 3 .2 Preparation and Characterization of Calcium Oxalate Monohydrate Particles und e r Different Conditions . .. ..... .... .... ......... ....................... ... . .... . ...... 54 3.2.1 COM particles ("33" particles) without seeds .. ..... ........ ..... ................... . ... 54 3.2 2 COM particles with seeds ...... ....... ..... . . ....... ...... ............. ....... . ............... 56 3.2 3 COM p a rticles precipitated from homo ge neous solution at 90 C ....... ...... 56 3.2 4 Charact eriz ation ... ... .. .. . ..... .... ................... .. . .... ...... . ....... . ... ....... .......... 58 3.2.5 Cr ys tal and Atomic Structure Mod e ling Us ing the Computer Pro g ram s SHAP E and A TOMS ....... ........................... ..... .... ... ........... 60 3 3 Results and Discussion ........... ........ ........ .................. . .... .... .................... . ......... 62 3.3.1 COM particles ( "33" particles) without seeds . .... ..... ........ ............... . ....... 66 3 .3.2 COM particles ("33" particles) with seeds ..... . . ............. . . ....... . ........ . .... 69 3 3 .3 COM particles precipitated from homogen e ous solution at 90 C . ......... .. 71 3 3 .4 Atomic structures of COM particles as a function of h a bit plane for the hi g h temperature form . .... .... . . . ..... ...... . .... . . ........... .... ................. ... 71 3.3 5 Charact e riz ation of COM crystals ..... ........... ... . ............ ..... .......... ....... .... . 79 3 4 Conclusions . ........... . . .... .... ...... . . ................................. ........................ . . .... .. ... 82 4 E VAL U ATION O F TH E C A LCIUM OXALAT E MOMOHYDRAT E HAMAKER CONSTANT BASED ON STATIC DIELECTRIC CONST ANT D E T E RMINATION AND EL E CTRONIC POLARIZATION .... ..... 85 4 1 Introduction ... ...... ..... . .......... . ...... .......... . ............. ... .. ........ .. ...... ............. ........ 85 4.2 M a terials a nd M e thods ........ ........ ....... ............. ... ............... ....................... ......... 8 9 4.3 Results and D is cussion ......... ... ...................... ......... . .... ........... .. .. .. ...... .......... ...... 95 4 4 Conclusions .... ......... .... ..... ....... .......... ............... ....... ............. ....... ........ ..... ........ 103 5 D E VELOPM E NT O F TH E DYNAMIC WET C E LL AND S T REAMING POT E NTIAL M EASU R E M E NTS .. ... . ..... ..... ..................... ...... .......... ..... . .... .... . 106 5 1 Introduction .... .... ........ . ............. . ..... ...... .. .......... ..... ............ ...... .... .... ...... ...... 106 5 2 M a terial s and M e thod s ...... .. ....... ....... ..... . ..... ... ....... ......................... ........ . ....... 109 5 .2.1 A dh e sion M e asurements using the D y namic Wet C e ll... .. .. ..... ..... ........ 109 5.2.2 Stre a min g Potential Measurement.. ..... .... ...... ........ . ...... ............. ....... .... 116 V l

PAGE 7

5.3 Results and Discussion .......... ...................... ....... ............. ......... ........... ........ .... 122 5 .3 .1 D y namic Wet Cell ................. ..... ....... ................................................... ..... 122 5.3 2 Model COM Particles .................. .......... ..... ....................... .... ... ................. 125 5.3.3 Streaming Potential. ..... . ...... ... ......................... .......... ....... .... ................. ..... 127 5.3 4 Adhesion ........... ........ ....................... ....... ..... ...................................... ........ 131 5.3 5 Theoretical Modeling ofinteractions .............................. ..................... ...... 133 5.4 Conclusions . ...... . ......................... ..... ... ................................... ..... ...... ............... 138 6 MEASUREMENT OF COM ADHESION TO MACROMOLECULAR SUBSTRA TES ... ........... ...... ................................................ ... ..... ... .... ... ... ....... ......... 140 6 1 Introduction . ....... .... ............................. ............................................ ......... . ....... 140 6.2 Materials and Methods ... ........................ ...... .............. ..................... .... ..... .... ..... 143 6.2.1 Particle Synthesis and Characterization ..................................................... 143 6.2.2 Substrate Coating ........ ..... ................... ..................... .................................. 143 6.2 3 Adhesion Measurements ...................... ............................................ ......... 145 6.2.4 Streaming Potential Measurements ..................... ........ ............................... 146 6.3 Results and Discussion .................................................................. . .... .... ...... .... 146 6.3.1 Substrate Coverage ... ........ ............................. .. .. .............................. .......... 146 6.3 .2 Zeta Potential Determinations .................. ................................................. 149 6 3 .3 Adhesion Measurements .................. .............................. .... ...... .............. .. 153 6 3 .4 Hydrodynamic Model of the Human Kidney ....................... ................... 167 6.4 Conclusions ........... ............... ............. ...... .... ... ........ .............. ... .... .................... . 170 7 SUMMARY AND FUTURE WORK ........ ....... .... .... ...... ....... ...... . ......................... 173 7.1 Summary ......................... ....................... .................... ... .... .................................. 173 7.2 Future Work ...... ... .............................................................................................. 175 APPENDICES A DETERMINATION OF THE AREA OF COM CRYSTALLOGRAPIBC FACES USING EQUIVALENT SPHERICAL DIAMETER .................................. 177 B DETERMINATION OF THE STRESS IN THE KIDNEY AS A FUNCTION OF FLOW AND DISTANCE FROM THE TUBULE W ALL. ............................... 179 REFERENCES ... ..................... ............................................................. ............. .............. 182 BIOGRAPHICAL SKETCH ...... .... ... .... .... ............ .... ........... ..... ............ ... ........ ............. 192 vu

PAGE 8

LIST OF TABLES Table 3.1. Compo s ition o f a rtificial urine ion solution .... ........... ...... ....... ...... .... ... . ......... ... ..... 61 3.2. Values of the lattice parameters of calcium o x alate monohydrate ........... ..... ....... .... 63 3 3 Atomic coordinates of Whewellite ... ....... ........................... ...................... ................ 65 3.4 Crystal forms and corresponding central distance values used to generate the theoretical shape of COM cr y stals ................. .. ......... . ............. .... ........ .... . ... ... .. 68 3.5 Values o f the crystallographic data for the high temperature form of Whewellite .... ..... .... ....... ..... ................ ...... ................... . ............. ................ ... .. ..... 73 3 6 Atomic coordinates for high temperature form ofWhewellite .... ..... ........... . . ......... 75 4 .1. Index of refraction for COM as a function of optical direction and wavelength ....... 88 4 .2. The composite mi x ing rules which were evaluated ........ . ....... ................. .... ........... 90 4 3 Molecular structures of the silane coupling agents used to disperse COM in Eccosil 5019 silicone ...... ................ .... .......... .. ......... ........... ... ...................... ..... 92 4.4. Dielectric constant v alues of COM det e rmined b y fitting the mixin g rules to the ex p e rim e ntal data .............. .... .... .... .... ... ........... .... . .................. . ..... ........... .... 98 4 .5. Calcul at e d values o f the UV characteristic frequency and correspondin g dielectric constants and refractive indices as a function of cr y stallographic direction determined from Cauchy plots ........... ..... .... ............... . .... ......... ..... .... 102 4 .6. Comparison of A131 calculated for COM using Gregory s approximation vs. the TaborWinterton relationship ...... .. ................... . ....... . ....... .. ..... . .... ....... . 105 5 1 A list of the ph y sical parameters necessary to calculate z e t a potential from streaming potential .......... ............ . . . ..... ......... .......... ......... ........ .... ..... ... ...... 1 2 1 6.1. The values of zeta pot e ntial of the macromolecular substrates determined using streaming potential measurements in saturated COM solution .. .. .... .... ..... 151 Vlll

PAGE 9

6.2 Values of adhesion strength for COM particles adhering to macromolecular substrates as a function of crystallographic habit plane in COM saturated solution ..................... ...................................................... .... ...... ...... .................. 165 6.3. Values of adhesion strength for COM particles adhering to macromolecular substrates as a function of crystallographic habit plane in AUIS ...... ................. 166 6.4. Human kidney tubule dimensions and volumetric flow rates reported by Kok and Khan ............. ................................... .................. .... ............................. .......... 169 IX

PAGE 10

LIST OF FIGURES 2.1. Pathw ays to kidne y stone formation ... .. .. ...... ............ ........... .................... ................. 6 2.2. COM particle aggregate attached to the wall of the proximal tubule in a rat nephron with one end of the constituent crystals joined together (near the arrow) and the other end free .......... .... ................... .......... .. ....................... ..... ..... ..... 8 2 .3. The electrical double la y er structure illustrating the distribution of ions surrounding an electrostatically charged particle ........................... ...... .... . . ....... 10 2.4 The effect of electrolyte concentration on the ionic cloud and particle seperation distance in a (A) low ionic strength solution and a (B) high ionic strength solution ... ............. .... ... .. .. .. ... ... ... ..... .......... ....... .................. .... .............. 11 2.5 The effect of ioic strength on zeta potential or the electric potential at the shear plane ...................... ........... ..................... ...................... ........... ......... ...... .... 13 2.6. A schematic diagram illustrating the attractive repulsi v e and total ener gy cur v es for two interacting materials as afunction of seperation distance .... ....... . 16 2 .7. A schem a tic representation of six aggregation mechanisms which may contribute to stone formation ..... ..... ........ ... .... ................. ..... .................. ............ . 18 2.8. Heterocoagulation of COM with HU is predicted based on : (A ) zeta potential determinations (B) mi x ing of HU and COM at pCa 5 where COM and HU are both negatively charged and (C) mixing at pCa = 4 where COM and HU are both oppositel y charged ............................. .................. ... .............. ......... .. 2 2 2.9 A schematic representation of patch charge flocculation whereby incomplete macromolecular coverage may create electrostatic shielding or opposite charge in the case of a charged polymer. ............... .. .. .. .. .. ..... .. .. ............ .... ......... .... 2 4 2 .10. Scenario depicting the balance of hydrodynamic forces ( Fct) and adhesive force s ( Fa) a cting on a COM particle bound to the brush border in the human nephron .......... .... ..... ................. ............. ........ ...... .............................. . . ...... .... ... 2 9 X

PAGE 11

2.11. A scanning electron photomicrograph of a tubule cross section in the rat animal model demonstrating association of a COM crystal with the basement membrane ...................................... ............ ............................................. 32 2.12. A schematic illustration of a human nephron .......... .................... ........ ...................... 33 2 .13. A schematic representation of a tubule wall found in a kidney ..... ..... ..................... 3 5 2.14 A schematic representing charged particle movement in an applied electric field to determine electrophoretic mobility ..... ..................... ........... ........ .......... ....... ..... 37 2.15. A schematic representaion of the flow of ions under an applied hydrodynamic pressure which generate the potential across the capillary or streaming potential. ..................... ......... .................................. ... ...... . ..... ... .... .... ..... ........... 39 2.16 A schematic diagram of the SPM ............. . ..... ........................................ ............... .42 2.17. Multiple internal reflection within the internal reflection element which is coupled to the sample ............................................................................... ...... ....... 44 2.18. Sampling depth as a function of wavenumber for the KRS-5 internal reflection element having a 45 incident angle ....... .................................. .............................. 46 2.19. A schamatic diagram of a Coulter counter electrical pulse counting instrument. .... .47 3 .1. Possible aggregation mechanisms for particles in the urinary environment. ............. 51 3.2. Schematic representation of processing steps for the preparation of COM crystals ("33" particles) without seeds ............................. ............ .............. .... ...... 55 3.3. Schematic representation of processing steps for the preparation of COM crystals ("33" particles) with seeds .............. ........................................................ 57 3 .4. Schematic representation of processing steps for the preparation of COM crystals ( "32" particles) with heat treatment at 90 C ........................................... 59 3 5 (A) SEM photomicrograph and (B) and (C) theoretical equilibrium shapes of the COM crystals precipitated without seeds.(B) The equilibrium shape of COM crystals based on the crystallographic data of Deganello and Piro and (C) the equilibrium shape of COM crystals based on the crystallographic data of Tazzoli and Domeneghetti ............................... ............. ..... ....... .......... .... 64 Xl

PAGE 12

3 .6. Theoretical atomic structures of COM crystals as a function of habit plane: (A) and (B) are the (010) and (100) planes, respectively drawn using Cocco and Sabelli atomic coordinates; (C) and (D) are the (010) and (100) planes, respectively drawn using Tazzoli and Domeneghetti atomic coordinates; (E) and (F) are the (010) and (100) planes, respectively, drawn using Deganello and Piro atomic coordinates ................................................................................... 67 3. 7. SEM photomicrographs and theoretical equilibrium shapes of seed crystals and the COM crystals precipitated using seeds:(A) seed crystals (B) equilibrium shapes of seed crystals generated with the (001) and (010) faces, respectively (C) COM crystals grown using seeds and (D) equilibrium shape of COM crystals ....................................................................................... . 70 3.8 SEM photomicrograph and theoretical equilibrium shapes of the COM crystals precipitated from homogeneous solution at 90C:(A) COM crystals (B) equilibrium shapes of an individual COM crystal and a crystal with a twin generated on the (001) face and (C) equilibrium shapes of COM crystals generated using the (010) face as the dominant face ................................. 72 3.9 Theoretical atomic high temperature structure (stability range: 318-415 K) of COM crystals as a function of habit plane, (A) (010) and (B) (101) .................. .. 76 3.10 A (top) typical experimental COM x-ray diffraction pattern and (bottom) the JCPDS file for the mineral Whewellite ................................................................... 77 3 .11. The differential frequency vs. equivalent spherical diameter particle size distribution of the experimentally produced COM crystals grown without seeds fit to a log-normal probability distribution ...... ................. ............................ 79 3.12. The relative linear dimensions of experimentally produced COM crystals .............. 80 3 .13 Zeta potential as a function of pH for COM in saturated COM solution and zeta potential at pH 6 for COM in 10% AUIS ......... ................................. ........... 81 3.14 Contact mode SPM scan of the COM (010) crystallographic face showing the surface roughness .............. ............ ..... ........ ................... ................. ....... .... ......... 83 3.15 Contact mode SPM scan of the COM (101) crystallographic face showing the surface roughness ............. .................... ......... .... ....................... ............. ...... . ..... 84 4.1. Optical photomicrographs of COM particles dispersed in Eccosil 5019 using (A) no coupling agent, (B) SMAEPS coupling agent, and (C) GPTMS coupling agent. . ....... ...... ............. . . .......... ...... . ........... .................................... .... 96 4.2. The composite mixing rules fit to experimental dielectric data ..... ............................ 97 Xll

PAGE 13

4.3. Cauchy plots for water and COM as a function of optical direction .... ... ............ .101 5 .1. (A) A picture and ( B) a schematic diagram of the dynamic wet cell developed to measure adhes ion of particulate to s ur faces.In ( B ) structura l thru-holes have been omitted from the drawing to improve clarity ..... .... ...... .... ... .... . ...... 108 5 2 (A) A picture and (B) a schematic dia gram of the dynamic wet ce ll supporting equipment. ... ....................... ... .. ..... ....... ..... ....... . ............ ... ... ...... ... ..... ..... ..... ... . 112 5.3. A time lapse sequence of events during an a dh esion experiment.The flow rates are (A) 12 ml/min (B) 53 ml / min, and (C) 108ml/min . . ...... . .... ... ... . . ...... .... 113 5.4. COM particles of controlled morphology s ho wn in a (A) scanning electron micrograph (B) modeled using SHAPE software showing the two dominant crystallographic faces (0 10 ) and ( 101 ) and (C) demonstrating the relative crysta llo graphic size rat io s . . ... ....... ........ ... . . .............. ........ . ....... . . .... 114 5.5. Theoretical (010) and (101) crystallographic planes of calcium oxalate monohydrate The Ca2+;c20 / ratio is given above each theoretical atomic structure ..... . . . ... . .... . . .... .... ... . ... ... . ... ............. ............. . .... .... ......... ............ ... 115 5.6 A (A) picture and a (B) schematic diagram of the streaming potential cell... .... ... . 117 5. 7. A diagram of the R-C circ uit u sed to e limin ate asymmetry and electrode polarizations ... ... .... .... ... . . ... ............. ..... .................. ..... ..... . . . ... ... ............. .... 118 5.8. (A) A picture and (B) a schematic diagram of the streaming potential instrument including all of the supporting equipment. . . . . ....... ......... . .... . . . ... 119 5 .9. A plot of dial setting versus vo lumet ric flow rate used to calibrate the pump.The error bars represent the s tand ar d deviation of five individual experiments ..... ... . ... ... .... . . ...... ......... ... ... ..... .... . ... ... ... .......... ...... .... ..... ..... 123 5.10. A plot of flow rate versus Reynolds number indicating the stable flow inside the dynamic wet cell at all experimenta l flow rates ....... ... ................ . ... . . . . . ... 1 2 4 5 11. Experimentally determined zeta potential of COM particles in saturated calcium oxalate monohydrate so lution Eac h data point is the mean+/95 % confidence interval.The ionic strength of the saturated COM solution was 8xl o-6 M ....... .... . ..... ..... .... .... .......... . . ...... ... ...... . . . ... . .... ... ........ ... ... . . ... . ... 126 5 .12. A plot of pressure drop or driving pressure across the streaming capillary indicating that the flow is laminar under all experimental flow conditions ...... ... 1 2 8 Xlll

PAGE 14

5 .13. A plot of streaming potentia l vs driving pressure for fused quartz in COM saturated solution showing a linear regression with an r2= 0 98 and 95% confidence limits ........ . . ........... .... . . ....... ...... ......... . .... . .... ..... ......... .............. 129 5.14. The probability of an initial particle on the (010) and on the (101) adhering versus the applied stress acting on the particle at failure as determined using the dynamic wet cell. The total number of particles counted that were l y ing on the (010) was n = 255 and on the (101) was n = 380 ............ ..... . . ...... ... .... .... 132 5.15. A scanning probe microscopy image of the fused quartz swface under a COM saturated solution liquid environment.Also given is the ima g e roughness statistics which include the ra or mean surface roughness of 0.871 nm ........... ... .135 6.1. A plot of wavenumber vs absorbance for fused quartz, collagen type I fibronectin MATRIGEL, and PEI produced using ATR-FTIR.All of the macromolecular substrates were coated on fused quartz ..... ... . ....... . . ... ......... . 14 7 6 2 A plot of wa v enumber vs. absorbance for fused quartz collagen type I fibronectin MATRIGEL and PEI produced using ATR-FTIR for the wa v enumbers 2100-3600 cm-1 .All of the macromolecular substrates were coated on fused quartz .... . . ....... ....... ........ .... ..... ............ .......... ........ .. ....... .... ... 148 6 3 Zeta potential as a function of pH for COM in saturated COM solution and zeta potential at a pH of 6 for COM in 10% AUIS ... ..... .......... . .... . .... .... . ..... 150 6.4. A plot of driving pressure vs streaming potential for the substrates collagen type I fibronectin MA TRI GEL and PEI ....... .... ....... . ...... ..... . ....... ..... ....... 152 6.5. The probability of an initial COM particle on the (010) and the ( 101) adhering to fused quartz in AUIS versus the applied stress actin g on a particle a t failure as determined using the d y namic wet cell.The total number of particles counted l y ing on the (010) was n = l 13 and on the (101) was n = 84 ...... 154 6 .6. The probability of an initial COM particle on the (010) and the (101) adhering to collagen type I in COM saturated solution versus the applied stress acting on a particle at failure as determined using the dynamic wet cell.The total number of particles counted lying on the (010) was n = 82 and on the ( IO I) was n= 13 3 ........... .................... . .... ........... .... ....... .... . ... .. ... ..... ......... ...... 15 5 6. 7. The prob a bility of an initial COM particle on the ( 010 ) and the ( 101) a dhering to colla g en type I in AUIS versus the applied stress acting on a particle at failure as determined using the dynamic wet cell. The total number of particles counted lying on the (010) was n = 84 and on the ( 101) was n = 72 ....... 156 xiv

PAGE 15

6.8. The probability of an initial COM particle on the (010) and the (101) adhering to fibronectin in COM saturated solution versus the applied stress acting on a particle at failure as determined using the dynamic wet cell.The total nwnber of particles counted lying on the (010) was n=87 and on the ( 101) was n = l 04 ........ ..... .... .... ............... . ...... ........ . ...... ...................... ..................... .... 157 6.9. The probability of an initial COM particle on the (010) and the (I 01) adhering to fibronectin in AUIS versus the applied stress acting on a particle at failure as determined using the dynamic wet cell. The total ~wnber of particles counted lying on the (010) was n=95 and on the (101) was n=143 ...... ............. 158 6.10 The probability of an initial COM particle on the (010) and the (101) adhering to MATRIGEL in COM saturated solution versus the applied stress acting on a particle at failure as determined using the dynamic wet cell The total nwnber of particles counted lying on the (010) was n = 57 and on the ( I 01) was n=54 .............................................................................................................. 159 6 .11. The probability of an initial COM particle on the (010) and the ( I 01) adhering to PEI in COM saturated solution versus the applied stress acting on a particle at failure as determined using the dynamic wet cell.The total number of particles counted lying on the (010) was n=61 and on the (I 01) was n = 69 .... 160 6.12. A bar chart summarizing the adhesion data experimentally determined in COM saturated solution (i .e., low ionic strength).The numbers above each set of bars is the value of the zeta potential for each substrate ................................. .. .. 161 6 .13. A bar chart summarizing the adhesion data experimentally determined in COM saturated solution (i.e., low ionic strength).The numbers above each set of bars is the value of the zeta potential for each substrate ..................................... I 63 6.14. COM particles (A) coated during an adhesion experiment using fibronectin and (B) uncoated COM particles used in an adhesion experiment.Both photomicrographs were taken after flow ............ .................. ... ..... ..................... 168 6.15. A plot of the stress on a 5 m to 8 m radius model COM particle under the hydrodynamic conditions found in the different areas in the hwnan kidney.The shaded region represents the range of experimentall y measured values of adhesion of COM to biological materials ....... ............ .... ........ . ..... .... ... 171 xv

PAGE 16

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 MEASUREMENT OF ADHESION BETWEEN CALCIUM OXALATE MONO HYDRA TE AND MODEL SURF ACES USING A DYNAMIC WET CELL Chairman: Dr. James H. Adair By Craig F. Habeger December 1997 Major Department: Materials Science and Engineering Calcium oxalate monohydrate (COM) is the primary constituent in kidney stones. COM crystals were synthesized in the l aboratory and characterized. Computer calculations of particle shape have been reconciled to observed shapes of COM cr y stals experimentally synthesized under various conditions. Comparison between the theoretical atomic structures generated by computer calculations are consistent with previously reported atomic layering sequences. Composite mixing rules were used to deconvolute the dielectric constant of COM from a COM / silicone composite. Utilizing the Lichtenecker dielectric mixing model, the value of the static dielectric constant of COM was determined to be 28 9 Optical and dielectric data were then used in the Tabor-Winterton relationship to calculate the Hamaker constant A131, of COM particles interacting in water. The A131 for COM as a XVI

PAGE 17

function of crystallographic habit also was examined. The mean value of A131 for COM was calculated to be 13.7x1021 J at 37 C in an aqueous environment. A hydrodynamic method for measuring the adhesion of particles to a surface has been designed for use in the study of kidney stone disease and other pathological biomineralization phenomena. The hydrodynamic force required to displace a particle adhering to a fused quartz substrate was calculated via the Poiseuille equation The strength necessary to remove 50% of the COM particles adhering to the substrate on the (010) and (1 oi) crystallographic surfaces are 81 and 170 Pa, respectively The previously determined Hamaker constant and measured values of zeta potential were used to calculate the energy of interaction between a COM particle and the fused quartz substrate which was found to be comparable to experimentally measured values provided the separation distance was on the order of 20 nm. Using the instrument and technique developed the adhesion of COM to biologically and non-biologically relevant materials was measured in COM saturated solution and in an artificial urine ion solution. The biologically significant materials were the proteins collagen type I fibronectin and MATRJGEL a mixture of basement membrane proteins. The non-biologically relevant material was polyethyleneimine a positively charged macromolecule used as a control. MA TRI GEL and the positive control polyethyleneirnine exhibited the highest adhesion to COM crystals XVll

PAGE 18

CHAPTER 1 INTRODUCTION 1. 1 Introduction Kidney stones effect hundreds of thousands of people each year. In 1991 895 467 visits were made to doctors for both kidney and ureter stones [NIH97] and in 1993 kidney and ureter stones resulted in approximately 302 00 hospitalizations [Cla95]. The direct and indirect cost of kidney and ureter stones in 1993 was approximately $1.83 billion [Nat95]. Prevention of kidney stones is the focus of much research ; however, before prevention can occur the mechanisms of stone formation must be w1derstood. 1.2 Literature Review Chapter 2 reviews the pertinent kidney stone literature and gives background on techniques utilized in the work reported herein. Kidney stone disease has been described as an opportunistic disease which can rely on many causative mechanisms acting in concert [Fin78a Fin84]. One of those possible mechanisms is the growth of an adherent particle called a fixed particle in the lumen of a nephron. Such an adherent particle can subsequently grow by secondary nucleation and growth and/or aggregation of crystallites to a size large enough to occlude a renal tubule which is the basis of the fixed stone mechanism of kidney stone development. Finlayson and Reid [Fin 78b] calculated the time for a COM crystal to grow to a large enough size to cause an occlusion and determined that the crystal wou ld not have time enough to grow to such a size.

PAGE 19

Therefore crystal attachment is a necessary process in kidney stone formation [Man91 Man94]. Renal injury has also been implicated in fixed stone formation [Gil79 Kha84 Man91]. Khan [Kha82 Kha95a] has discussed that after attachment of a cr y stal to the epithelial cell surface injury to the cell may or may not occur A few different results may ensue due to crystal attachment to the epithelial cell. 2 The cell may die and be washed away in the urine flow taking with it the crystal. The cell may envelope the crystal and continue to function normally [Lie92 Lie93 Lie94] with no further growth of the crystal. However if epithelial cell injury has occurred basement membrane proteins may be accessible to adhere to COM crystals by sloughing of the cells leaving behind the exposed underlying architecture the basement membrane. The basement membrane is a layer of proteins that underlies the epithelial sheet and will be discussed subsequently in more detail. Depending on the extent of injury other proteins common to the extracellular matrix found below the basement membrane ma y also be accessible for attachment to COM crystals Researchers have shown the attachment of COM crystals to epithelial cells and fibrous proteins characteristic of those found in the basement membrane and extracellular matrix, yet the importance of these individual materials in kidney stone formation with respect to COM crystal adhesion is not known The necessity to determine the important materials most responsible for COM crystal adhesion will allow researchers to focus their interests to those more important materials in the search for a possible cause and therefore relief from kidney stone formation. 1.3 Development and Characterization of Model COM Particles Chapter 3 demonstrates the laboratory synthesis of COM particles having a desired size and crystallographic shape to be used in adhesion experiments. The optimal particle shape is spherical ; however COM has a monoclinic crystal structure and

PAGE 20

3 spherical crystalline particles cannot be produced Therefore well -d efined shaped rectangular particles having a narrow size distribution were produced to limit errors associated with calculations The experimentally produced particles were modeled using computer software to determine the crystallographic face indices and the theoretical atomic structure of two dominant crystallographic faces. The particles were characterized using electrophoresis and atomic force microscopy. The zeta potential values for the particles were determined using electrophoresis in a low ionic strength solution COM saturated solution, and a high ionic strength solution artificial urine ion solution. Finally the surface roughness of each of the two dominant crystal faces were measured in saturated COM solution using contact mode Atomic Force Microscopy. 1.4 Evaluation of the Calcium Oxalate Monohydrate Hamaker Constant Chapter 4 involves the determination of the Hamaker constant a measure of the van der Waals interaction forces for COM. Attractive van der Waals forces are very important in the adhesion process and a reliable value of the Hamaker constant is necessary to determine these forces. The COM particles developed in Chapter 3 were mixed with a silicone material to produce a composite. The dielectric properties of the composite were measured and composite mixing rules were used to deconvolute the dielectric constant of COM from the COM / si licone composite. Optical and dielectric data were then used to calculate the Hamaker constant A131, of COM particles interacting in water. 1.5 Development of the Dynamic Wet Cell and Streaming Potential Measurements Chapter 5 describes the development of a hydrodynamic method for measuring the adhesion of particles to a surface for use in the study of kidney stone disease and other pathological biomineralization phenomena. By using hydrodynamic flow to remove particles from a model surface, the strength with which particles adhere to a surface can

PAGE 21

4 be measured. The hydrodynamic force required to displace a particle is calculated via the Poiseuille equation using the dynamic wet cell dimensions and the fluid flow rate. A model surface consisting of fused spectroscopic grade quartz was used in the development of the apparatus Also developed was an instrument to measure the electrostatic potential called zeta potential, at the surface of a flat substrate in solution. The previously determined Hamaker constant and measured values of zeta potential were used to calculate the energy of interaction between a COM particle and the fused quartz substrate. The theoretical values of adhesive pressure were compared to the measured values of adhesion 1.6 Measurement of COM Adhesion to Macromolecular Substrates Chapter 6 investigates the adhesion of COM to both biologically relevant and non-biologically relevant materials. The adhesion measurements were measured in COM saturated solution and in an artificial urine ion solution using the instrument and technique developed in this research. The use of two different ionic strength solutions allows interpretation of the types of material-material interactions taking place (i.e van der Waals electrostatic, chemical) to be made. The biologically significant materials were the proteins collagen type I fibronectin and MA TRI GEL, a mixture of basement membrane proteins. The non-biologically relevant material was polyethyleneimine PEI a positively charged macromolecule used as a control. 1.7 Conclusions and Future Research Chapter 7 summarizes the conclusions of the current research and presents suggestions for future work.

PAGE 22

CHAPTER 2 LITERATURE REVIEW 2.1 Introduction Kidney stone disease has been called an opportunistic disease because it is thought that kidney stone disease takes an unfortunate simultaneous convergence of factors that may be achieved by several pathways for a stone to form [Con90 Fin78a Fin84 Rob86]. Thus, injury supersaturation (particularly hyperoxaluria) nucleation and growth mechanisms all play a role in contributing to stone formation as demonstrated in Figure 2.1. Even so, urolithiasis is idiopathic with no one risk factor providing an obvious source of stones in chronic stone formers However both in vivo and in vitro evidence indicates that the presence of particles is a necessary but insufficient condition for stone formation. Most non-stone forming individuals at certain times have crystalluria yet do not have a stone incident. For a stone to be created from freely flowing particles either a particle must grow to a large enough size to occlude the tubule or aggregation of multiple particles must occur. Furthermore, if aggregation is responsible for stone formation the attractive energy or energies holding the primary particles together in an aggregate must be great enough to withstand the hydrodynamic shear forces due to the flow of fluid through the nephron. The possibility that the growth of a large crystal leads to a stone has been shown to be unlikely by several teams of investigators [Fin78b, Kok94]. Finlayson and Reid [Fin78b] predicted that there was not enough time for a freely flowing particle to grow to a size sufficient to occlude the tubule based upon the hydrodynamics in the nephron and measured rates of crystal growth for calcium oxalate monohydrate. It was also suggested 5

PAGE 23

Agent of Injury Nucleation Catalysis ~ Aggregation l Cell Damage a~ (7 -,,, I Aggregation -~ Figure 2.1. Pathways to kidney stone formation. 6

PAGE 24

7 by Finlayson and Reid that aggregation of freely flowing particles was not likely due to the relatively low concentration of particles within the nephron at any given point in time. However a recent analysis of the hydrodynamic aspects of transit time through the nephron by Kok and Khan [Kok94 ] based on better estimates of the diameters of different regions in the nephron indicated that aggregation of freely flowing particles is possible. Regardless attachment of calcium oxalate particles to the epithelial nephron wall has been demonstrated in a rat animal model [Fin84 Fin78b] as shown in Figure 2 .2. Thus the concept of fixed particles at the epithelial wall has become a fundamental principle in the development of kidney stones. In more recent work [Kha84 Man87 Man 91, Rie8 8], it has been demonstrated that particle attachment can be induced by epithelial wall damage in which intercellular species such as phospholipids may play a role The objective of this chapter is to review the dynamics of aggregation for particles and interactions of the particles and/or aggregates with biological surfaces such as the basement membrane or interluminal wall of the nephron The nature of the forces among particles and surfaces will be discussed with respect to the hydrodynamic flow scheme found in the nephron ; however particle-particle interactions must first be discussed. 2.2 Particle Interactions Particles in solution are constantly moving The movement of the smaller particles ( < 1 m) is due to Brownian motion also called perikinetic motion. The movement of larger particles ( > 5 m) defined as orthokinetic motion is due to gravity and convection currents. The constant motion in solution causes the particles to interact with one another ultimately resulting in either stabilization or aggregation. When repulsive forces dominate the particle-particle interaction stabilization is the result. The opposite is the case if attractive forces dominate the particle-particle interaction called aggregation. These forces arise from an electrostatic potential on the surface of the

PAGE 25

Figure 2.2. COM particle aggregate attached to the wall of the proximal tubule in a rat nephron with one end of the constituent crystals joined together (near the arrow) and the other end free (bar= 5 m) Photomicrograph by S .R. Khan [Kha91]. 8

PAGE 26

particles and from an inherent dipole interaction of atoms and molecules within one particle interacting with another particle. 2.2.1 Electrostatic Repulsion 9 The surface potential 4' 0 is the potential difference between the solid surface and the bulk solution. Ions which alter the surface potential are called potential determining ions Curreri et al. [Cur79] have hypothesized that Ca2+ and C2o / specifically adsorb to the COM surface and thus are termed the potential determining ions for COM. The Gouy-Chapman electrical double layer model describes the excess ions present on the surface of the solid phase and the distribution of ionic charge of opposite sign in the solution phase surrounding the electrically charged surface which assures electroneutrality. A schematic diagram of the Gouy-Chapman model is shown in Figure 2.3 The double layer model consists of an inner layer often called the Stern layer which describes ions adsorbed into the compact or inner region of the double layer and the outer or diffuse layer [Ove52]. Theoretical analysis of the double layer shows that the charge density in an aqueous solution decreases rapidly with increasing distance from the solid surface and the electrostatic potential for low potentials is given by the Debye-Htickel approximation [Hun81 ] 1/f = l/f0 exp(-,a) (2.1) where 1/f is the potential as a function of distance into the solution K is called the Debye Htickel parameter (I / length) and xis the distance from the solid surface. Based on the electrostatic model for aggregation in the simplest case the repulsive contributions between particles in solution arise from the interaction of ionic clouds surrounding the particles as demonstrated by Figure 2.4. The extent of the ionic cloud is governed by the solution ionic strength described by the simplified equation

PAGE 27

q,s-4-,.. ... co q,o .;:::; C Q) -0 0.. () ;:: --() Q) w Gouy-Chapman (Diffuse) Layer Distance from Surface Figure 2.3. The electrical double layer str u cture ill u strating the distribution of ions surrounding an electrostatically charged particle 10

PAGE 28

+ + + + + +++o+ \ + + + + + + + + + + + + + + + .... d (A) + (B) Figure 2.4 The effect of electrolyte concentration on the ionic cloud and particle separation distance in a (A) low ionic strength solution and a (B) high ionic strength solution. 11

PAGE 29

12 (2.2) where e is the elemental charge n~, is the number of ions of type i per unit volume far from the surface zi is the valence of ions of type i Eis the dielectric permittivity of the solvent k is the Boltzmann constant and T is the temperature in Kelvin. The Debye Htickel simplification assumes that the surface potential or stern potential is low (e.g. less than 25 m V). The quantity 1/K is referred to as the Debye length and indicates the thickness of the double layer. The De bye length is dependent only on temperature valence of the electrolyte ions and electrolyte concentration. Solution ionic strength I, is given by I= I(c;z;) I (2 3) where ci is the solution concentration (M) As the ionic strength of the solution increases the surface charge decay into the solution occurs more rapidly due to the inability of the solution to support charge. That is the diffuse double layer becomes compressed due to the increased concentration of ions in solution as demonstrated in Figure 2.5. The increased concentration of ions near the charged particle surface causes the electrical charge to be neutralized by the ions of opposite charge counterions in solution. Figure 2.4 schematically describes the ionic cloud surrounding a charged particle in low and high ionic strength environments The variable d in Figure 2.4 is the distance of separation for the two charged particles. At low ionic strength the ionic cloud is large and electrostatic repulsion prevent particle-particle contact. At higher ionic strength the particles can

PAGE 30

_J 1-z w l-o 0... u a: 1-u w _J w SHEAR PLANE s1 V s2 .. s 3 ---I 1 I I I SURFACE l 1 > ss INCREASING IONIC STRENGTH DISTANCE INTO THE SOLUTION Figure 2 5 The effect of ioic strength on zeta potential or the electric potential at the shear plane 13

PAGE 31

come into much closer proximity due to collapse of the surrounding ionic cloud If the collapse is significant aggregation may occur 2.2.2 Attraction 14 The tendency towards aggregation can result from oppositely charged electrostatic potentials or from van der Waals forces which are present at all times irrespective of solution conditions. The attractive van der Waals forces are composed of multiple intermolecular interactions between the ions molecules and electrons that make up the particles interacting across a dielectric medium such as water [Hou80 Hun93 Ber90]. The mutual attraction between interacting particles arise from harmonic oscillations at the molecular atomic or subatomic level. Three primary sources of such intermolecular interactions exist depending upon the nature of the interacting species. Some of the more important specific dispersion interactions are known as Keesom Debye and London interactions [Isr92]. Keesom forces are due to molecular dipole-dipole interactions in the particles. Debye interactions occur when molecular dipoles in one particle induce electronic polarization in the other interacting particle As such Keesom and Debye interactions occur only if a material has one or more dipoles present within its structure. In contrast London interactions are more ubiquitous because London interactions are due to mutual electronic polarization with all atoms or electron cloud density shift of adjacent atoms. The range of the force due to dipole interactions between two atoms or molecules is a nanometer or less and varies as 1 / r6 However the interactions are to some degree additive in that they effect all neighboring atoms or molecules in a particle. Particle particle van der Waals forces interact over a much longer range 1/r2.

PAGE 32

2.2.4 Interaction Energy The Derj aguin and Landau [Der41] and V erway and Overbeek [V er48] (D L VO) theory considered colloid stability in terms of the electrical double layer and van der Waals forces Figure 2 6 is a plot of interaction energy as a function of separation distance which is a potential energy diagram for two interacting materials across a medium. Positive interaction energy refers to repulsive energy and negative interaction energy refers to attractive energy In the diagram both the attractive and repulsive portions are plotted with respect to separation distance of the two interacting materials. The third interaction line is the sum of the attractive and repulsive curves the total interaction energy curve. Labeled on the total interaction energy curve in Figure 2.6 are the primary and secondary minimum and the energy barrier to aggregation 15 The primary minimum is the potential energy well that occurs when the two interacting materials are in contact. The main attractive forces are due to van der Waals forces. When two interacting particles are at a separation distance such that their associated interaction energy is in the primary minimum the attractive energies dominate the total interaction energy and the two materials cannot be separated The two particles are said to be thermodynamically unstable with respect to dispersion The secondary energy minimum occurs at larger separation distances be y ond the energy barrier. When two interacting materials are in the secondary energy minimum they form a weak aggregate and may be dispersed with the addition of energy. The secondary minimum is thermodynamically metastable with respect to aggregation The potential energy barrier to aggregation occurs between 1 to 4 nm [Isr92] and is due to strong electronic repulsion between the ionic clouds of the interacting particles and is dependent on the surface potential. If the surface potential is high then the potential energy barrier will be high; and if the surface potential is low the potential energy barrier will be low As the electrolyte concentration of the medium increases the energy barrier

PAGE 33

>, 0) ,._ Q) C w C 0 :;::; (..) al ,._ Q) +-' C Potential Energy Barrier Secondary M i nimum Separation Distance Figure 2 .6. A schematic diagram illustrating the attractive repulsive, and total energy curves for two interacting materia l s as a function of seperati on distance 16

PAGE 34

17 decreases and aggregation becomes more and more likely to occur Finally at the critical coagulation concentration the concentration of electrolyte at which the repulsive energy barrier falls below zero aggregation occurs 2.3 Aggregation Mechanisms Among Particles and Particles at Surfaces Numerous studies [Fin78a Fin84, Fin78b, Rie88 Ada81 Coe91 De g91, E dy86 Edy87a Har86a Har86b Kok90 Rob85, Rya81a Rya81 b Rya81c Rya84 Rya86 Scu86a Scu86b Wie87] have shown that the aggregation and/or adhesion of COM particles potentially can lead to the ultrastructures composing kidney stones. However the actual mechanism(s) of aggregation within the biophysical environment of the human kidney have not been examined in detail. There are at least six aggregation mechanisms as shown in Figure 2 7 acting alone or in concert which may contribute to stone formation [Ada95]. To establish dispersion techniques which may prevent stone formation, first it is necessary to determine the most important aggregation mechanisms with respect to interparticle strength A number of possible mechanisms exist for COM aggregation within the human kidne y In addition to minimum double la y er interactions other interactions include secondary minimum coagulation heterocoagulation polymer bridging flocculation aggregation by secondary nucleation and growth and immiscible amphiphilic molecule flocculation. Hydrodynamic factors also need to be considered including the size of the tubule and the shear rate associated with fluid flow within the n e phron as discussed by Finla y son and Reid [Fin78b] and Kok and Khan [Kok94] Furthermore the concentration of the COM particles performs a role whether one is considering fixed stone disease or aggregation of the particles in a freely flowing state. Finally the flow patterns associated with peristaltic compared to continuous flow within the nephron have not been addressed in past studies but should be to realistically assess aggregate formation

PAGE 35

+ .~ Separation Distance Coagulation in Simple Electrolyte Solutions > + Separato n D istance Secondary Minimum Coagulation + (> + C'I l,L cit!< / Electro f{te Concentr aton Separation D istance Heterocoagulation of Particles with Different Surface Charge Polarity Polymer Bridging Flocculation Primary Particle Se c o n dary ~ Parti cles Secondary Nucleation and Growth Flocculation via an Immiscible Amphiphilic Molecule Figure 2.7. A schematic representation of six aggregation mechanisms which may contribute to stone formation (Ada95]. ....... 00

PAGE 36

19 in the kidney. Each of the mechanisms will be discussed with respect to their likelihood in the human nephron. 2.3 .1 Particle Aggregation in Simple Electro lytes The aggregation of COM particles has traditionally been attributed to the absence of charge on COM particle surfaces and / or the high ionic strength of urine However neither of these aggregation mechanisms accommodates the observation that COM particles observed in vivo invariably have organic matter on their surfaces, as shown in Figure 2.2. Generally particles in aqueous solution have a surface charge created by a combination of several different charging mechanisms In the case of COM, Curreri et al. [Cur79a Cur79b, Cur87] showed that surface charge is due to incongruent dissolution of the constituent Ca2+ and C204 2 -. These and other species from solution then can adsorb (in their hydrated form) into an adsorbed ion la yer known as the Stern layer. The electroneutrality of the system composed of the surface and Stern charges and surrounding solution is achieved by the charge in the diffuse cloud of ions that only are attracted electrostatically toward the surface. 2.3.2 Secondary Minimum Coagulation As shown in Figure 2.7 a minimum exists at relatively large distances of separation in the interaction energy for COM particles. As the ionic strength increases the magnitude of this secondary minimum increases. The secondary minimum in the interaction energy curve is a consequence of the longer range of the van der Waals attractive forces than the electrostatic repulsive interactions [Hou80 Son72]. Thus even when the charge at the surfaces of the interacting particles is great enough to produce an energy barrier at intermediate separation distances, secondary minimum coagulation may take place because there is no energy barrier for this mechanism of aggregation. The strength of the interparticle bonds for secondary minimum interactions has been evaluated

PAGE 37

theoretically and experimentally by Chan and Halle [Cha84]. It was demonstrated that the mean lifetime for secondary minimum aggregates increased with increasing ionic strength of the suspension containing the model spherical polystyrene particles. 20 Adair [Ada81] showed that COM suspensions composed of relatively coarse primary particles ( 5 m equivalent spherical diameter) aggregate over a wide range of solution and surface charge conditions. Secondary minimum aggregation was implicated in conditions where primary minimum aggregation was minimized because of low ionic strength and relatively high zeta potential. However the results were ambiguous because only the thermodynamic aspects of the coagulation process were addressed in Adair's study. The strength of the proposed secondary minimum interaction was not determined by analyzing the hydrodynamic shear forces required to promote breakup of the aggregates. Aggregate bond strength measurements as a function of particle size were suggested since the magnitude of the secondary minimum increases as a function of the radii of the interacting particles. The increased likelihood of secondary minimum coagulation with increasing particle size may have important implications to stone disease since it has been reported by Robertson [Rob69] that stone forming individuals have particles significantly larger in size than non-stone formers 2.3 3 Heterocoagulation Heterocoagulation is the aggregation among particles of different materials. We are not aware of any investigators that have addressed this potential mechanism for stone formation. However there have been a number of investigators [Bar78 Der54 Kuo80 Mat81] within the colloid chemistry community that have developed the theoretical and practical framework for a study of heterocoagulation The basis for heterocoagulation is the difference in surface charge polarity of particles comprised of different materials. Thus a positively charged COM particle may be electrostatically (as well as through the van der Waals interactions) attracted to a negatively charged hydroxyapatite (HAP)

PAGE 38

particle. However even if the surface charge on particles of dissimilar materials are the same the van der Waals attractive forces may be strong enough to promote heterocoagulation (as well as homocoagulation). 21 Within the human nephron a variety of potential combinations exist that may lead to heterocoagulation. These interactions include, COM-HAP COM-HU (uric acid) COM-NH4U (ammonium urate) COM-COD (calcium oxalate dihydrate) COM-E coli and other bacteria and COM with various macromolecules. Heterocoagulation will be reversible only when the sign of the smface charge is made the same for all particles and is sufficient to overcome van der Waals interactions. The sign of the surface charge and corresponding solution conditions that will promote or inhibit heterocoagulation need to be evaluated for each system. It has been predicted that epitaxy of the high temperature form of COM and HAP is unlikely because of incoherent crystal structures [Man81]. However heterocoagulation can explain the presence of HAP with COM in a stone. Heterocoagulation has not been addressed for material systems relevant to urolithiasis. Preliminary experiments have been conducted in our laboratory to determine the conditions in which particles composed of various materials will be likely to heterocoagulate. An obvious starting point in this initial evaluation is to examine the effect of the polarity of the zeta potential for the various materials as a function of relevant solution conditions. The zeta potential data for COM and HU are summarized in Figure 2.8(A) as a function of Ca2 + and C2o/concentration. The COM data are from Curreri et al. [Cur79a] and incorporate the solubility product for COM in the Ca2+ concentrations. The uric acid data from Adair et al. [Ada88] are given as a function of Ca2+ or C2o /-. COM has a point of zero charge (pzc) at pCa = 5.2 (Ca2+= 6.3x1Q-6 M) with COM having negatively charged surfaces above this pCa and positively charged surfaces at higher Ca2+ concentrations. Adair et al. [Ada88] showed that uric acid is negatively charged over a wide range of pH and Ca2+ and C2o / concentrations.

PAGE 39

60 -r----------------, > 40 E 20 j:: z c;:i COM U r ic Acid -Ca w 0-+------rl---+-------l 0 a.
PAGE 40

23 Thus mixing COM and HU particles where pCa is less than 5.2 should promote heterocoagulation between the negatively charged HU and the positively charged COM. As shown in Figure 2.8(B) and 2.8(C) fine COM particles produced by the dimethyl oxalate decomposition adhere to the larger prismatic HU particles when pCa = 4 When COM and HU particles in saturated COM solutions are mixed with pCa = 5 the fine COM particles have a greater affinity toward one another than the HU particles and heterocoagulation does not occur as shown in Figure 2.8(C) 2 3.4 Polymer Bridging Flocculation This mechanism takes place when there is insufficient polymer ( or macromolecule) for full surface coverage on particles This is the only aggregation mechanism that explains some ultrastructural observations of Boyce Khan et al. and others [Boy68 Kha83a Kha83b Kha87 Mey82, Mey71 Pri86] on the role of matrix macromolecules in the microstructure and ultrastructure of the mature stone. Maximum flocculation takes place when one-half of the surface of a particle is covered by polymer [Son72 Hun86] as shown schematically in Figure 2.9. However flocculation takes place anywhere between about one-tenth surface coverage to greater than 75 percent coverage. We are not aware of any studies relevant to urolithiasis that have addressed the role of flocculation in detail. Kok et al. [Kok90) Finlayson [Fin78a] and Robertson and Peacock [Rob85] have discussed this mechanism with other possible mechanisms of aggregation ; but the research emphasis has been on the prevention o f aggregates b y employing large concentrations of macromolecules or polymers [Coe91 Deg91 Rya81 a Rya84 Scu86a Scu86b Lan88 Lea77). This mechanism can also explain the conflicting reports of inhibition versus promotion [Edy 86 Cam89 Gro90). Thus macromolecular species such as uropontin Tamm-Horsfall mucoprotein and nephrocalcin may play a dual role: at su f ficiently low concentration aggregation is promoted through flocculation

PAGE 41

24 Figure 2 9 A schematic representation of patch charge flocculation whereby incomplete macromolecular coverage may create e l ectrostatic shielding or opposite charge in the case of a charged polymer.

PAGE 42

25 while at higher surface coverage dispersion of particles is achieved through the protective colloidal effect of the macromolecular coating [Hou80 Son72]. Although there have been limited studies on flocculation with respect to urolithiasis there have been a number of investigations on floe formation and hydrodynamic breakup because of its importance in wastewater treatment and other technologies [Dit82 Eis85 Gre81 Gre85 Ray87]. These studies provide a basis for evaluating the role of flocculation as an aggregation mechanism for COM and other relevant particles in urine. For example, an excellent starting point is to determine the macromolecular or polymer dosage for relevant urinary species required to achieve the maximum flocculation. The critical flocculation concentrations (CFC) for a particular macromolecule will indicate whether this mechanism is likely by comparison with its concentration range in urine. In preliminary studies we have examined a flocculant commonly used in mineral recovery. Polyethyleneimine (PEI) is a positively charged highly branched polymer molecule used by Pelton and Allen [Pel84] to produce positive charge on glass surfaces in their particle adhesion studies. It has been shown in preliminary studies using the apparatus developed by Eisenlauer and Horn [Eis85] to evaluate the aggregation of freely flowing particles in suspension. This device has the advantage that flow rate and mode of flow (i. e continuous versus peristaltic) can be varied The commercial analogue to Eisenlauer and Hom's device known as a photometric dispersion analyzer (Rank Brothers Cambridge UK) has been used extensively to monitor the flocculation of model systems. Initial experiments have demonstrated that fine COM particles flocculate at intermediate dosages of PEI. The relative degree of flocculation varies as a function of charge (as dictated by the concentrations of Ca2+ and C2o / -).

PAGE 43

26 2 3 .5 Flocculation of COM Particles by Phospholipids and other Intercellular Substances Work by Khan and Mandel and co-workers [Kha88 Man94] indicates that interaction of COM particles with phospholipids at either the epithelial surface of cells or with phospholipids liberated into the bulk solution by cell injury are important features in the formation of aggregates in the interluminal channel of the nephron Flocculation by sparingly soluble amphiphilic molecules such as the phospholipids forming a major component of the epithelial cell membrane have not been studied with respect to aggregation but Mandel et al. [Man94] clearly show that adhesion of COM particles is important. The ability of phospholipids to adsorb to the surfaces of sparingly soluble inorganic particles and surfaces is well established as are the forces arising from the interaction of phospholipid monolayers on mica and similar surfaces based on force balance work by Israelachvili and others [Isr92]. Thus one would expect phospholipids to demonstrate an effect similar to macromolecu l es or polymers capable of promoting flocculation However the interparticle strength of particles flocculated in either freely flowing suspensions or at surfaces containing phospholipids (i .e., cell membranes) would be expected to depend on the concentration of phospholipids and the efficienc y of adsorption of such species to COM surfaces. 2.3 6 Adhesion of Particles at Surfaces Adhesion is an important mechanism in the formation of fixed kidney stones and is dependent on the adhering materials and the suspending medium One of the first theories of adhesion between solid particles and surfaces was given by Krupp [Kru67] where he defined three classes of interactions: 1 Class I interactions include long range attractive interactions resulting from van der Waals forces and electrostatic forces

PAGE 44

2. Class II interactions are given by short range attractive interactions such as chemical bonds and hydrogen bonds. 3. Class III interactions involve interfacial reactions occurring at elevated temperatures including sintering effects diffusive mixing, and mutual dissolution and alloying. 27 Class III interactions may also be important at lower temperatures for polymer or macromolecular diffusive mixing. Other important forces also exist that were not directly addressed by Krupp. These forces are salvation forces, structural forces, or in a water medium called hydration forces. These short range forces involve ordering of the solvent medium to cause either attraction or repulsion between particles depending on the hydrophobic/hydrophilic nature of the solids in suspension [Hor90, Isr92]. All of the aforementioned interactions may apply during adhesion ; however Krupp [Kru67] believed that van der Waals forces and electrostatic forces are the dominating long range attractive forces between adhering materials. Krupp thought that only under ultra-high vacuum or extremely pure systems would primary chemical bond formation take place due to the saturation of bonding sites by contaminants under ambient conditions. Kallay et al. [Kal87] was also of the opinion that van der Waals interactions and electrostatic interactions were dominating but they furthermore suspected that short range repulsion played a role in the total interaction energy of adhering bodies citing salvation forces, as did Israelachvili [Isr92] and also electron cloud repulsion. These short range forces were said to act at separation distances on the order of the distance of closest approach of the two surfaces. Theoretical models for both long range interactions [Der41 Ver48 Hog66 Wil93] and short range [Isr92 Kal87] interactions are available for varying surface geometry interactions and will be discussed later. The study of adhering materials is fundamental and very broad in application. The adhesion of many materials by several researchers has been documented [Vis76]. For

PAGE 45

28 example the adhesion of Fe203 to a glass substrate and a glass substrate covered by a layer of gelatin [Ryd95] red blood cells to glass [Moh74] human fibroblasts to glass [van92] submicrometer particles to silicon [Bus93], and bacteria to glass [Bus92] have all been measured As observed applications of the adhesion measurement range widely Just as a number of applications are in need of the adhesion measurement a number of techniques can be used to measure the adhesion of solid particulate material to solid surfaces In past studies researchers have used many techniques to evaluate adhesion [Cor66 Zim82]. These techniques include a rotating disc [Kri94] a packed column [Kal87] a centrifuge method [Kor60] a vibrating method [Der61 ] a surface force apparatus [Isr78] and a hydrodynamic method that utilizes parallel plates [Pel84 Pel82] to name only a few. The most recent technique for measuring the force with which a particle adheres to a surface utilizes the scanning probe microscope (SPM) [Duc91]. Other adhesion measuring techniques exist and are used but are best suited for specific material systems much like the techniques mentioned above. Further discussion will be limited to only the hydrodynamic parallel-plate and adhesion measuring technique. Another approach in evaluating the bond strength of particles adhering to a surface is to determine the hydrodynamic shear force required to remove particles. Pelton [Pel84], Busscher [Bus84], Matijevic [Mat80 Mat81] Owens [Owe87] and others [Ols78] have used this approach to evaluate thermodynamic models for the attachment of particles to surfaces The balance of forces proposed for a COM particle attached to the epithelium is shown schematically in Figure 2.10. The shear stress at the wall ( -rw) is given by [Owe87 Esk68] dP 'rw =-b dl where 'tw is the shear stress at the wall P is the hydrostatic pressure drop across the conduit of len g th l and wall separation b ( 2.4)

PAGE 46

Epithelial Brush Border { in the Human Nephron Figure 2.10. Scenario depicting the balance of hydrodynamic forces (Fd) and adhesive forces (Fa) acting on a COM particle bound to the brush border in the human nephron. 29

PAGE 47

30 Determination of the adhesion strength for particles adhering at surfaces is of fundamental importance in deducing whether the fixed particle mechanism for stone formation proposed by Finlayson and Reid [Fin78b] is reasonable within the hydrodynamic system of the nephron. Kok and Khan [Kok92] recently re-evaluated the hydrodynamics in the kidney and determined that it is not possible to have a stone occurrence with out the aid of adhesion. Riese, Mandel and coworkers and others [Man94 Rie92 Rie88, Yam96] have shown in vitro attachment of COM crystals to inner medullary collecting duct (IMCD) epithelial cells of the rat animal model in the static case (i.e. without the presence of flow). They also proposed that perturbations in the cell membrane structure with a loss in membrane polarity can enhance crystal attachment [Rie92]. Lieske et al. [Lie95 Lie96a] has also measured adhesion of COM to MOCK cells and to monkey renal epithelial cells in a similar manner to Reise et al [Rie88]. They determined that cell anionic sites can be blocked by specific cations [Lie96a] and the positive sites on a COM crystal may be blocked by specific anions [Lie95] thereby minimizing adhesion. Bigalow et al. [Big97] recently demonstrated that COM crystal attachment to IMCD cells was effected by the cell membrane fluidity. Changes in temperature, cholesterol content and cell culture time which increase cell membrane fluidity also increase the ability of COM crystals to bind to the membranes leading the researchers to conclude that a long range arrangement in the membrane is created to match the COM crystal structure [Big97]. In vivo crystal attachment to epithelial cells in the rat animal has been demonstrated by Khan et al. [Kha82]. Renal injury has also been implicated in fixed stone formation [Man91 Gil79 Kha84]. Khan [Kha82, Kha95a] has discussed that after attachment of a crystal to the epithelial cell surface that injury to the cell may occur but may not always occur A few different results may ensue crystal attachment to the epithelial cell If the cell dies the cell and attached crystal may be washed away in the urine flow. The cell

PAGE 48

31 may envelope the crystal and continue to function normally [Lie92, Lie93, Lie94, Koh96] with no further growth of the crystal. However, if epithelial cell injury has occurred basement membrane proteins may be accessible to adhere to COM crystals by sloughing of the cells leaving behind the exposed underlying architecture the basement membrane. The basement membrane is a layer of proteins that underlies the epithelial sheet and will be discussed subsequently in more detail. Depending on the extent of injury, other proteins common to the extracellular matrix fow1d below the basement membrane may also be accessible for attachment to COM crystals. Khan et al. [Kha84] demonstrated the association of crystals to fibrillar macromolecular structures by scanning electron microscopy (SEM) Khan [Kha95a] has also shown crystals passing through the basement membrane into the extracellular space near the papillary tip as shown in Figure 2.11. Researchers have shown the attachment of COM crystals to epithelial cells and fibrous proteins characteristic of those found in the basement membrane and extracellular matrix yet the importance of these individual materials in kidney stone formation with respect to COM crystal adhesion is not known. The necessity to determine the imp01iant materials most responsible for COM crystal adhesion will allow researchers to focus their interests to those more important materials in the search for a possible cause and therefore relief from kidney stone formation. 2.4 The Human Kidney The human kidney contains approximately 1 million nephrons which absorb nutrients and water back into the body after being filtered out of the blood by the renal glomeruli [Ham74]. The nephrons consist of tubules through which the waste is collected. As the waste products travel further down the length of the nephron the waste becomes concentrated and supersaturated conditions may exist allowing precipitation of crystallites, such as COM to occur. Figure 2 .12 shows a nephron.

PAGE 49

Figure 2 .11. A scanning electron photomicrograph of a tubule cross section in the rat demonstrating association of a COM crystal with the basement membrane. (Courtesy of S .R. Khan). l,.) N

PAGE 50

int~rlohulor o. ascending loop of Henle de.saending loop of Henle proximal convolutczd tubule collecting tubule Figure 2.12. A schematic illustration of a human nephron [Ham74]. 33

PAGE 51

The renal tubules are lined with epithelium Underlying the epithelial cells is a structure called the basement membrane which is a continuous thin mat of specialized extracellular matrix that functions as a support, a molecular sieve and a cell regulator [Yur90]. The basement membrane consists of the basal lamina and the lamina reticularis 34 The basal lamina is also divided into two different sections, the lamina rara and the lamina densa. The basal lamina is essentially a mat of collagen type-IV with specific molecules on each side of the mat that help it bind to adjacent cells or matrix materials [Alb89]. Although the composition of the basal laminae varies from tissue to tissue one of the molecules always found in the basal lamina is laminin. Another molecule often found in the basement membrane in particular the lamina densa [Ino94], is fibronectin. The structure below the basement membrane is the extracellular matrix which is primarily made up of fibrous proteins in a hydrated polysaccharide gel [Alb89]. Collagen type I is a fibril forming collagen found in skin, tendon bone intestine uterus, and surrounding organs [Kuc92]. Figure 2 .13 is a schematic representation of the renal epithelial cells basement membrane and extracellular matrix showing there relative positions in the tubule wall. 2.5 Characterization Techniques 2 5 1 Electrokinetic Measurements Charge characterization at the solid-solution interface is very important when attempting to discern mechanisms of adhesion. As described in section 2.2.1 of this chapter charged materials when immersed in water are surrounded by strongly adsorbed ions called the Stem layer and by a gradient ionic cloud called the diffuse part of the double layer as shown in Figure 2.3. Electrokinetic measurements are useful in determining the electrostatic potential often called the zeta potential s at the boundary

PAGE 52

/ tubule structure plasma membrane of epithelial cell -50 nm [ ~,. "o ~ J basal ] basement ~--...:,-::::::lamina :::-~ -7~ membrane lamina rara lamina densa lamina reticularis Figure 2 .13. A schematic representation of a tubule wall found in a kidney 35

PAGE 53

36 between the strongl y adsorbed la y er and the beginning of the diffuse double layer as schematically depicted in Figure 2 .5. Electrokinetics is the measurement of the movement of one phase with respect to another phase in which a charged boundary between the two exists Two types of electrokinetic measurements electrophoresis and streaming potential are described below. 2.5 .1.1 Electrophoresis Colloidal particles immersed in water can move under an applied electric field as a result of the charge on the particle surface as shown in Figure 2.14. The electrophoretic mobility , is defined as the particle velocity per unit static electric field. The mobility can be determined by measuring the velocity of a particle under an applied electric field. The mobility is related to the electrical potential at the shear plane surrounding the particle. The shear plane is defined as the boundary between the bulk solution where the ions are free to move and the inner layer of strongly adsorbed ions which move with the particle under an applied electric field. The corresponding zeta potential (, is calculated from the experimentally determined electrophoretic mobility, uE, according to the Smoluchowski equation [Hun81] (2.5) where Tl is the viscosity of the solution D is the dielectric constant of the solution 0 is the permittivity of free space A number of techniques exist to determine the electrophoretic mobility under an applied electric field such as a simple optical method electrophoretic light scattering method and electroacoustics to name a few. 2.5.1.2 Streaming Potential The charge at the surface or the more commonly accepted charge at the Stern layer the zeta potential can be determined by measuring the streaming potential

PAGE 54

Figure 2.14 A schematic representing charged particle movement in an applied electric field to determine electrophoretic mobility. 37

PAGE 55

38 associated with a surface Streaming potential is the potential generated when a fluid is forced through a capillary The hydrodynamic pressure forces the mobile charges in the double layer in the direction of flow. The accumulation of charge at one end of the capillary creates an electrical potential across the capillary as demonstrated in Figure 2.15 In the case of glass or fused quarts the charge on the surface is negative ; therefore the mobile charges will be primarily positive causing a positive current in the direction of flow The streaming potential can be measured with a high impedance voltmeter as a function of pressure. If the potential is measured as a function of pressure the Smoluchowski equation can be used to calculate the zeta potential [Hun81] (2.6) If the usual mixed units are used the equation is where t; and E s are the zeta and streaming potentials respectively both measured in m V P is the pressure drop across the capillary in cm Hg the conductivity of the elution A is in units of n-1cm-1 and t: is the dielectric constant of the elution. Equation (2.6) is only valid for solutions in which all or almost all of the current generated due to streaming is carried through the bulk liquid. For solutions having ionic concentrations less than 10-3 M the need to account for surface conduction is important [Rut47]. At low electrolyte concentrations a large part of the current may be carried through the double layers near the capillary walls because of the higher charge density in that region Equation (2.6) becomes [Hun88]

PAGE 56

39 ---------1 00000000 +0 + .. 0 ODouble Layer lons0 0 0 0 0 0 0 0 0 0 Neutral E l ectrolyte -+ 0 + 0Accumulation of Q 0 G) 0 0 Ions 0 0 0 G). O Flow Direction Ill 0 O 0_003 0 0 0 0----~ L I Figure 2.15. A schematic representation of the flow of i ons rmder an app l ied hydrodynamic pressure which generate the potentia l across the capillary or streaming potential.

PAGE 57

S = 4n( A-0 + 2A_1./ r )E s cP 40 (2.7) where the term 2J..s/ r is the specific surface conductivity and r is the capillary radius or the distance between plates for a flat plate system. The variable A s is the conductance of a square section of material of unit area and constant thickness measured in ff 1 Briggs [Bri28] suggested a simpler procedure for correcting for surface conduction. The procedure involves measuring the resistance of the liquid in the capillary at low ionic concentration Rexp, and compare with the value of resistance expected from measurements at high ionic concentration, R:aic, where surface conduction can be expected to be negligible. Equation (2.7) becomes (2.8) The general effect of accounting for surface conduction will give an increase in magnitude of the calculated value of zeta potential at low ionic concentration conditions. Ball and Fuerstenau [Bal73] performed an extensive review of the streaming potential literature and found that the measurement should be performed over a sufficiently wide range of pressures to obtain an accurate estimate of the slope E / P the obtained slope should be linear and the intercept of the slope should be zero. Ball and Fuerstenau [Bal73] and Hunter [Hun88 Hun93] have all stressed the importance of the linear dependence of the streaming potential Es, on the applied pressure P. The linearity of the plot of streaming potential vs. applied pressure is a necessary first step to determine experimental reliability ; however unless surface conduction is accounted for, linearity does not ensure accuracy of the resulting zeta potential [Hun88]. The finite intercept of the slope is an indication of an asymmetry potential or rest potential which is said to be a function of the e lectrodes including their preparation, treatment and cleaning,

PAGE 58

and of the electrolyte and its concentration [Bal73]. No electrode system Pt, Au Ag / AgCl Ag/ Agl or calomel has been found to eliminate the asymmetry potential completely. 41 Many authors have developed methods for eliminating asymmetry or rest potentials [Hun62 Mar30 Bul35 Hor77]. Hunter and Alexander [Hun62] found that the rest potential could be nullified when the liquid was stationary. Martin and Gartner [Mar30] and Bull [Bul35] avoided an asymmetry potential by working at high applied pressures where the asymmetry potentia l is negligible relative to the streaming potential. Horn and Onoda [Hor77] devised a resistance-capacitance (R-C) circuit to store the asymmetry potential in a large capacitor and subtract the potential from the streaming potential when flow begins. 2.5 2 Scanning Probe Microscopy Scanning probe microscopy (SPM) offers a versatile range of techniques which can be employed to acquire information about a material surface. Topographical information can be acquired using Contact Mode Atomic Force Microscopy (AFM) TappingMode AFM and Non contact AFM Other techniques can also acquire topographical information Contact Mode AFM measures topography by sliding the probe tip across the sample surface in air or in fluid. A laser is focused on the back side of the probe tip also known as the cantilever as shown in Figure 2 .16. The laser reflects from the cantilever to a mirror onto a photodetector. The sample is mounted on the piezoelectric scanner which rasters the sample under the tip and surface interactions between the sample and tip cause the cantilever to deflect. Any motion of the cantilever is registered by the photodetector. The position of the laser spot is determined by the electronic circuitry which generates a voltage difference between the photodiode segments and a topographical image is the result.

PAGE 59

42 Photodiode Mirrors Laser Cantilever Piezo tube Feedback Figure 2.16. A schematic diagram of the SPM.

PAGE 60

43 2.5.3 Scanning Electron Microscopy Scanning electron microscopy (SEM) is a useful tool to analyze surface topography of materials. A focused electron beam is rastered across the sample exciting electrons in the atoms of which the sample is composed. As a result both secondary and backscattered electrons are produced from the sample. A detector measures the intensity of the electrons versus position and displays this information on a cathode ray tube. Secondary electron contrast is due to the dependence on electron yield on the topography and the depth of secondary electron emission is about 100 A. Because the sample must be electrical conductive to be analyzed using SEM the samples need to be sputter coated with Au/Pd. 2.5.4 Fourier Transform Infrared Spectroscopy Infrared spectroscopy (IR) is a useful tool to study polymers and organic materials as well as inorganic materials. Fourier transform infrared spectroscopy (FT-IR) uses the principles of interferometry to study bond vibration in materials. Electromagnetic radiation in the infrared region (7 8x10 5 to 0.1 cm) is passed through or reflected off the sample. The radiation excites molecular bonds to higher vibrational levels absorbing energy. The absorbed energy corresponds to paiiicular vibrational frequencies characteristic of a molecule or molecular group Therefore the IR technique is useful in determining unknown materials, determining bond orientation and quantitative and qualitative analysis of bond types [Mar86]. Attenuated total reflectance (ATR) spectroscopy also called internal reflection spectroscopy (IRS) may be the most widely utilized adaptation of IR to study inorganic material surfaces. A TR is performed by coupling the incident electromagnetic radiation into an IR transparent crystal of high refractive index as shown in Figure 2.17. When the beam reaches the interface between the crystal and the materials to be analyzed it is

PAGE 61

l/1 = p SAMP L E Figure 2.17 Multipl e internal reflection within the internal reflection element which is coupled to the sample 44

PAGE 62

45 internally reflected. However 100 % internal reflection does not occur and some part of the wave travels into the sample material to a depth governed b y the wavelength of radiation the IR crystal and the refractive index properties of the sample [Knu85]. Figure 2.18 is a plot of wavenumber vs. sampling depth for a thallium bromide-thallium iodide (KRS-5 ) IR crystal having a 45 incident angle coupled to a sample assumed to have an index ofrefraction of 1.5. A s can be seen from Figure 2 .18, the sampling depth increases as function of increasing wavelength ( decreasing w avenumber ) o ve r the mid infrared region. 2 5 5 Particle Size Determination Many techniques exist to measure particle size. These techniques include microscopy sedimentation methods electrical pulse counting lig ht scattering methods hydrodynamic methods and electroacoustics Each of these techniques has its advantages and limitations which may be dependent on size density optical properties or other parameters. Electrical pulse counting is able to count the number of particles in a known amount of solution by drawing the suspension through a very small orifice that has an electrode on either side of it as shown schematically in Figure 2 .19 [Hun93]. When a particle passes through the orifice interference with the current flowing between the two electrodes occurs and the resistance changes. The number of changes in current and the magnitude of change are recorded. Because the change in current is proportional to the volume of the particle passing through the orifice the result is a value of particle size when calibrated against a dispersion of known particle size The number of current changes acts as a particle counter generating a particle size distribution.

PAGE 63

2.80 2.40 ~2.00 0 IJ e 1.00 :I: t w 0 C, 1.20 z :i G. :,; 0.80 0.40 KRS-5 0 +---r--,.--r---,---.--~--.---.--, 4000 3200 2 400 1600 800 WAVENUMBER$ (cm Figure 2 .18. Sampling depth as a function of wavenumber for the KRS-5 internal reflection element having a 45 incident angle [Knu85]. 46

PAGE 64

t T o vac u um A mp T hr eshold Pu lse c urr e nt Amplifi e r CRO Co unt e r dri v e r St a rt pul se Di g it a l dis p l ay t St o p pulses Figure 2 .19. A schamatic diagram of a Coulter counter electrical pulse counting instrument [ Hun93 ]. 47

PAGE 65

48 2.5.6 X-Ray Diffraction X-Ray diffraction (XRD) is a tool used to investigate the arrangement of atoms in a crystal. XRD is useful for identifying compounds in crystalline materials determining crystallographic information of metals ceramics and polymers and performing quantitative phase analysis A monochromatic or near monochromatic beam of x-rays is focused on a specimen. The incident x-rays are diffracted by the lattice atoms in the crystal according to Bragg s Law [Cul78] nA = 2dsin0, (2.9) where n is the order number A is the wavelength of the incident x-rays dis the interplanar spacing and e is half of the angle of diffraction A scintillation detector is moved through some angle at a given rate and the number of counts or intensity is recorded The resulting pattern is a plot of x-ray intensity vs. angle (20) measured with respect to the incident beam Because the crystal structure is periodic the in-phase scattered electrons create an increase in intensity.

PAGE 66

CHAPTER 3 DEVELOPMENT AND CHARACTERIZATION OF MODEL COM PARTICLES 3 1 Introduction Calcium oxalate monohydrate (COM) is the most thermodynamically stable form of calcium oxalate and is a major component of human calcium oxalate calculi [Pri47 Deg8 la]. Stone formation in the urinary tract takes place opportunistically with an unfortunate, simultaneous convergence of factors that may be achieved by several pathways for a stone to form [Bro92 Ada95]. Many theories have been proposed to explain the critical factor(s) responsible for the formation of urinary stones [Bro92 Ada95 Fin78a Fin84 Rob86]. Kidney stone disease may take place as a consequence of injury supersaturation (particularly hyperoxaluria) nucleation and growth mechanisms which may include adhesion on epithelial cell surfaces and aggregation All of the previously mentioned factors may play a role in contributing to stone formation. Nonetheless stone formation remains an idiopathic disease with no one risk factor providing an obvious cause for stones in most cases. However both the in vi v o and in vitro evidence indicates that the presence of particles is a necessary but insufficient condition for stone formation. From this perspective the formation and presence of crystallites may be a typical renal function directed at concentrating oxalate for more effective irreversible elimination from the body. The abnormal stone-forming condition results from the a g gregation and further growth of such pre-existing crystallites [Fle78, Rya8 la, Ada8 l Deg91 Edy86 Ed y 87a Har86a Har86b Kok90 Rob85 Scu86a, Scu86b]. Indeed it has been found that while the solid phase in the urine of normal individuals is made up mainly of individual crystals 49

PAGE 67

50 large aggregates are often found in stone formers [Fle78]. Therefore both crystal growth and aggregation are generally regarded as important steps in the formation of calcium oxalate renal stones [Fle78 Rya81a]. Numerous studies have shown that the aggregation and / or adhesion of COM particles may potentially lead to the ultra-structures typical of kidney stones [Ada81 Deg91 Edy86, Edy87a Har86a Har86b Kok90 Rob85 Scu86a Scu86b]. Kok et al. [Kok90] reported that precipitation alone of COM particles cannot account for the large size particles necessary to cause blockage of the urinary tract. Aggregation is therefore the only possible mechanism whereby a stone may reach a size as large as 0.64 cm the maximum size that can pass through an average adult male urethra [Fin78b]. Aggregation is the process whereby crystals bind one to another to produce large clusters. The actual mechanism(s) of aggregation within the biophysical environment of the human kidney is uncertain; however, Adair et al. [Ada95] have classified six aggregation mechanisms acting alone or in concert which may contribute to stone formation: (a) coagulation in simple electrolyte solutions; (b) secondary minimum coagulation ; (c) heterocoagulation of particles with different surface charge polarity ; (d) polymer bridging flocculation; (e) secondary nucleation and growth; and (f) flocculation via immiscible amphiphilic molecules. The previously mentioned mechanisms which are summarized in Figure 3.1, stress the importance of knowing the environmental conditions in the urinary s y stem for formation of calcium oxalate calculi For example, citrate can retard the crystallization of stone-forming calcium salts by two broad means [Cha91]. First citrate complexes calcium and reduces the ionic calcium concentration available for precipitation in urine thereby lowering the supersaturation Second, citrate can directly inhibit crystallization of both calcium oxalate and calcium phosphate at the growing crystal-solution interface Thus citrate has been shown to inhibit spontaneous precipitation of calcium oxalate and to retard agglomeration of pre-formed calcium oxalate crystals Citrate is believed to also have a modest inhibitory effect against crystal growth

PAGE 68

+ % .~ Separation Distance Coagulation in Simple Electrolyte Solutions i3> + Separatbn Distance Secondary Minimum Coagulation + + 01 >t ::ti< / Electrolyte Concentratbn Separation Distance Heterocoagulation of Particles with Different Surface Charge Polarity Polymer Bridging Flocculation Primary Particle Secondary ~Particles Secondary Nucleation and Growth Flocculation via an Immiscible Amphiphilic Molecule Figure 3 .1. Possible aggregation mechanisms for particles in the urinary environment (from Adair et al. [ Ada95]) Vt

PAGE 69

of calcium oxalate Cody and Cody [Cod94] reported habit modification of COM crystals grown with citrate Interpenetrating crystals grown in distilled water without citrate have sharply-angled tips and are flattened parallel to the (010) face whereas crystals grown with citrate have rounded tips and are flattened on the (I 01) face 52 Few papers have dealt with the theoretical equilibrium shapes of COM crystals synthesized w1der different conditions or the atomic structure at specific faces [Deg91 Cod94]. Furthermore little if anything is known at the atomic-molecular level about the interaction between COM particles and important species such Ca2 \ C20/-, citrate dicitrate, and urea which are common in the urinary system and known to play a role in stone disease Deganello's work [Deg91] is an exception to the lack of data on the interaction between some species and COM crystals. Adsorption of nephrocalcin a protein found b y Nakagawa and co-workers [Nak83 Nak87] in human urine even at the lowest concentration tested affects the habit while also inhibiting the growth of COM. The nephrocalcin protein promotes the preferential development of the (101) faces to such an extent that the length to width ratio of the crystals decreases by approximatel y a factor of three In the course of this process, the apical planes eventually disappear and the size of the crystals decrease. Diminution of crystal size eventually becomes extreme when nephrocalcin reaches a concentration between 1 to 2 x 10-5 M. Mandel [Man94] reported that the molecular surface structures on the prominent crystal growth faces of COM are very important in determining potential long-range molecular bonding interactions between the crystals and the lipid-rich regions in the inner medullary collecting duct (IMCD) cell plasma membrane Mandel [Man94] also reported on epitaxial matching calculations for the high temperature form of COM calcium oxalate dihydrate (COD ), hydroxyapatite (HA) and uric acid (UA) crystal lattices against the dimensional repeating lattice of phospholipid headgroup structures. The Mandel study suggests that COD-COM epitaxy may be important relative to COD-lipid interactions in crystal attachment and stone development. Therefore the molecular modeling between

PAGE 70

53 specific surfaces of the COM crystal and these additives provide important insight for specific adsorption sites on COM surfaces and the growth and aggregation mechanisms of COM crystals in the urinary system. Other research suggesting the crystallographic importance of COM was performed by Habeger et al. [Hab97]. A crystallographic dependence of COM crystals on adhesion to fused quartz glass substrates was observed A variation in electrostatic charge as a function of crystallographic habit on the theoretical COM surface has been used to justify the differences in adhesive strength observed [Hab97]. In the current work morphological forms of COM crystals synthesized in different conditions have been studied by using the commercial computer programs SHAPE and A TOMS The computer program SHAPE has been used to determine the face indices of the synthesized COM crystals and generate the equilibrium shape as a function of central distance. The computer program A TOMS has been used to generate the surface atomic structure of COM crystals as a function of habit plane Certain characteristics in a particle system are desirable to rigorously study theoretical growth and aggregation and adhesion mechanisms Ideally, the particles should be spherical. If spheres cannot be produced a crystallographically definable morphology should be produced so that the habit planes can be identified. The theoretical study of growth and aggregation depends on well-defined particle morphologies. Unfortunately COM has a monoclinic crystal structure whose inherent asymmetry does not lend itself to the formation of spherical particles. Therefore the individual COM particles should be at least uniform with narrow particle size and shape distributions. Several different particle size ranges is also desirable Thus the objective in the current work is to prepare COM particles by several different methods followed by careful characterization to deduce the habit planes and atomic surface structure. Lastly characterization of the particles having the most consistent COM morpholo gy will be conducted

PAGE 71

54 3 .2 Preparation and Characterization of Calcium Oxalate Monohydrate Particles under Different Conditions Reagent grade chemicals were used in all precipitation studies without further purification All solutions were prepared with deionized water (specific conductivity less than 10 mhocm-1 ) which was provided by a Milli-Q high purity s y stem (Millipore Corporation Bedford MA) and were passed through a 0.2 rn filter. For identification purposes precipitated COM particles were designated by the negati v e logarithms o f the initial Ca +2 and c2O4 2 molar concentrations respectivel y (e.g. particles prepared from 10-3 M Ca +2 and 10-3 M c2O4 2 were designated "33" particles). 3.2.1 COM particles ( "33" particles) without seeds Reagent grade potassium oxalate monohydrate (K2C2O4H2O) (Fisher Scientific Inc., Fair Lawn NJ) was used to prepare stock 1 molar (M) K2C2O4 solution at 25 C After equilibration stock solutions were passed through a 0.22 m filter (MAGNA nylon supported plain MSI Westboro MA). Reagent grade calcium chloride dih y drate (CaC12 2H2O) (Fisher Scientific Inc., Fair Lawn NJ) was used to pr e pare 0.01 M CaCli solution at 25C. The CaC12 solution was prepared the da y o f use to prevent CaCO3 formation through reaction with CO2 from the atmosphere. Diluted solutions of K2C2O4 and CaC12 solutions were volumetrically prepared from the stock solutions. A schematic representation of the processing steps for the preparation of COM crystals ( "33" particles) without seeds is given in Figure 3 .2. Precipitation studies were performed by mi x ing equal volumes (10 liter ( f )) of 10-3 M CaC}i and 10-3 M K2C2O4 solutions at 25 C Crystals of COM were aged for at least 24 hours without stirring. Solutions containing precipitated crystals were 0.22 m filtered. The filtrate was washed with a solution saturated with COM to prevent dissolution of the precipitate during washing. After washing the recovered powders were freeze dried (Freeze Drier 4.5 Labconco Corp., Kansas City MO) and stored in a desiccator.

PAGE 72

10-3 M CaCii Solution (101) 10-3 M CaCl2 Solution (101) Precipitation Aging (fo r a t l east 24 hours without stirring) Recovery of COM particles (corcentra t e t h e pa1ticles i n the cen trifu ge) Washing & Filtering (was hing by sat urated COM solution with filtering through 0 .22 m) FreezeDrying (store the particles in a d es i cca t or) Characterization ( XRD SEM, and p a rticl e s i ze a nalysi s) Figure 3.2. Schematic representation of processing steps for the preparation of COM crystals ("33 particles) without seeds. 55

PAGE 73

56 3 .2.2 COM particles with seeds Seed crystals ("00" particles) of COM were prepared by mixing equal volumes (50 ml) of 1 M CaCh and 1 M K2C204 solutions at 25C. These seed crystals were aged for at least 24 hours without stirring. Crystals were concentrated by centrifugation and then washed with saturated COM solution. After washing, the recovered powders were freeze dried and stored in a desiccator. The seed suspension was prepared by dispersing 0.1 g of seed crystals in 5 ml isopropanol for a crystal growth experiment. The dispersion was treated by ultrasonication for 3 minutes to break up agglomerates. A processing schematic for the preparation of COM crystals ( 33" particles) with seeds is given in Figure 3 .3. Precipitation studies were performed by mixing equal volumes (500 ml) of 10-3 M CaCl2 and 10-3 M K2C204 solutions at 25 C with various amounts of seed crystals. Crystals of COM were aged for at least 24 hours without stirring. Crystals were filtered 0.22 m and then washed with a solution saturated with COM. The recovered powders were freeze dried and stored in a desiccator. 3.2.3 COM particles precipitated from homogeneous solution at 90 C Gordon Salutsky and Willard have reported a recipe for the production of COM by the thermal decomposition of dimethyl oxalate [Gor59]. The goal of this section was to duplicate this procedure and analyze the nature of the particles. Ammonium acetate (CH3COONH4 ) (Fisher Scientific Inc., Fair Lawn, NJ) and acetic acid (CH3COOH) (Fisher Scientific Inc. Fair Lawn, NJ) were used to prepare stock 2.5 M CH3COONH4 (aq) and 2 5 M CH3COOH (aq) solution at 25C. A buffer solution at pH 2.7 was prepared by mixing equal volumes (500 ml) of the 2.5 M CH3COONH4 (aq) and CH3COOH (aq) stock solutions at 25C. After equilibration stock solutions were passed through a 0.22 m filter

PAGE 74

103 M CaCI2 Soh.tion (2 50ml ) ------Seed Crystals ("00" particles ) Precipitation Aging (for at least 24 hours without stirring) Recovery of COM particles (concentrate the particles in the centrifuge) Washing & Filtering (was hing by saturated COM solution with filtering through 0 .22 m) Freeze-Drying (st.ere the particles in a desiccator) Characterization (XRD, SEM and partic l e size analysis) Figure 3 .3. Schematic representation of processing steps for the preparation of COM crystals ( 33" particles) with seeds. 57

PAGE 75

58 A schematic representation of processing steps for the preparation of COM crystals aged at 90 C is given in Figure 3.4 Reagent grade CaC12 2H2O was used to prepare 0 .01 M CaC12 solution at 25 C The CaC12 solution was prepared the da y of use to prevent CaCO3 formation through CO2 from the atmosphere. The dilute CaC12 solution were volumetrically prepared to obtain 2.5x10-3 M CaC12 solution. The pH of the dilute CaC12 solution was adjusted to 4.7 by adding 0 .01 M HCl solution. After equilibration the dilute CaC12 solution was filtered (0.22 rn) The 2.5x10-3 M CaC12 solution (150 ml) was rapidly mixed with buffer solution (100 ml) in a flask and 10 g of dimethyl oxalate (CH3OCOCOOCH3 ) (Fisher Scientific Inc. Fair Lawn NJ) was added. The flask was tightly closed and heated at 90 C for 1 hour in an oven to equilibrate. The holding time at 90 C was 2.5 hours After thermal treatment the solution was rapidly cooled to room temperature in an ice bath. Crystals were collected via centrifugation followed by washing with saturated COM solution previously passed through a 0.22 m filter. After washing, the recovered particles were freeze dried and stored in a desiccator. 3 2.4 Characterization The dried recovered powders were analyzed for phase composition using x-ray diffraction (XRD) (AP D3720 CuKa, fine tube 40kV-20mA Philips Electronics, Mahwah NJ) over a 28 range from 10-70 at rate of2.4/ min The morphology of the synthesized crystals were observed using scanning electron microscopy (SEM) (JSM 6400 JEOL Boston MA). Particle size analysis of typical COM particles without seeds was performed using an electrical sensing zone technique (ELZONE 80XY 95 m aperture Particle Data Incorporated Elhurst, IL). Zeta potential determination was performed using Rank Particle Microelectrophoresis (Apparatus Mark II Rank Brothers Cambridge, ENGLAND) as a

PAGE 76

0 0 1M CaClz Solu tion ( 100ml ) I I 2 5M C H3COONH4 Sohtio n ( S 0 m 0 I .. 2 SM CH3COOH Solu tio n ( 50m l ) 1 2. 5 103 M CaC I2 Solut i o n (15 0m 0 I I Bt.ifer Solu tio n (100ml) I A
PAGE 77

function of pH in COM saturated solution and in an artificial urine ion solution (AUIS) buffered to pH 6.0 at 37C. The composition of the AUIS is given in Table 3.1. 60 Scanning probe microscopy (SPM) (MultiMode, Digital Instruments Santa Barbara CA) was performed on the two major crystallographic surfaces of COM to determine surface roughness using a Nanoscope III controller. The J type scanner was operated in contact mode using a Si3N4 cantilever having a spring constant of 0 .12 N i m in saturated COM solution at 25 C 3.2 5 Crystal and Atomic Structure Modeling Using the Computer Programs SHAPE and ATOMS Based on crystallographic data including space groups and lattice parameters the computer program SHAPE (Macintosh version 4 .0, SHAPE Software, Kingsport TN) was used to determine the face indices of the COM particles synthesized under various conditions and to generate the equilibrium shape of the crystals as a function of the central distance [Coc61] The central distance is the perpendicular distance from the center of the crystal to the faces of the corresponding form. The greater the distance the less prominent the form (the smaller the area of the faces of that form in the final shape). Form factors in the program are the least number of face indices needed to generate the desired equilibrium shape. Twins of COM particles were drawn using the "twin option" and "epitaxy option" in SHAPE The theoretical atomic structure of COM was displayed using the computer program ATOMS (Macintosh version 2 .0, SHAPE Software Kingsport TN). The slice option of the computer program A TOMS was used in the process of analyzing and predicting surface structure as a function of habit plane. To draw an individual surface atomic structure using this option, it is necessary to enter the crystal class the space group the corresponding unit cell parameters and the face indices and atomic coordinates From the crystal class and space group, the program determines what symmetry

PAGE 78

Table 3.1. Composition of artificial urine ion solution Compound NaCl NaHPO 4 2H 2 O Na3C6HsO7 2H 2 O MgSO 4 Na2 SO 4 KCl CaCl 2 2H 2 O Na2C2O4 NH4Cl H4OH NaOH Solution Concentration ( M ) 0.10554 0 03654 0.00321 0 0038476 0 016952 0.06374 0.001 0 00131 0.03632 0 00062 0 005 61

PAGE 79

62 operators to use in the calculations. In addition to the indices of the face and input atoms, several other parameters were specified The thickness of the slice is in fractions of the d spacing Thus atoms present to a certain depth of unit cell were drawn with the "slice option" as a function of habit plane based on the atomic coordinates of the high and low temperature structures. 3.3 Results and Discussion First differing reports regarding lattice parameters for COM need to be reconciled. Table 3 .2 summarizes crystallographic data proposed by various authors [Deg81 a Taz80 Coc61 Coc62 Deg80] From Table 3 .2, the unit cell parameters reported b y Cocco [Coc61] and Tozzoli and Domeneghetti [Taz80] can be seen to have similar values in contrast to the unit cell parameters of the a and c axes reported by Deganello and Piro [Deg81 a]. The equilibrium structures of COM were generated the crystallographic data proposed by Deganello and Piro [Deg81a] and Tozzoli and Domeneghetti [Taz80] as shown in Figure 3 .5. Based on comparisons made between the theoretical equilibrium shape and the morphological form of the experimentally derived crystals the equilibrium shape of COM crystals modeled using the lattice parameters of Deganello and Piro [Deg81 a] can be more easily reconciled to the morphology of the experimentally synthesized COM crystals. It is also necessary to mention conflicting papers regarding COM atomic coordinates. Table 3 .3 shows atomic coordinates proposed by various authors [Deg81 a Taz80 Coc62]. From this table, the atomic coordinates reported by Cocco and Sabelli [Coc62] and Tozzoli and Domeneghetti [Taz80] have similar positions but only the x coordinates for carbon(l), carbon(2), oxygen(l), oxygen(3) and calcium(!) are different. The atomic coordinates reported by Deganello and Piro [Deg81a] are quite different from those of the previously mentioned authors. The equilibrium atomic structures of COM were generated using the three different atomic coordinates. Based on the comparison

PAGE 80

Table 3 .2. Values of the lattice parameters of calcium oxalate monohydrate Authors a(A) b (A) C (A) Cocco and Sabelli [Coc62] 6.24 14.58 9.89 Tazzoli and Domeneghetti [Taz80] 6.290(1) 14. 583(1) 10. 116(1) Deganello and Piro [Deg81 a] 9.9763(3) 14. 5884(4) 6.2913(32 B ( o ) 107 109.46(2) 107.05(2) Space Group P2i/c P2i/c P21/ n 0\ w

PAGE 81

64 (B) (C) ill ill C C A: ( 011) B : ( 0 11) y) C : ( 010 ) C I!. s;;; D : (0lj) ( E : (lQ.L) E F : ( 0 I I ) E Figure 3.5 (A) SEM photomicrograph and (B) and (C) theoretical equilibrium shap e s of the COM crystals precipitated without seeds (B) The equilibrium shape of COM crystals based on the crystallographic data of Deganello and Piro [Deg81a] and (C) the equilibrium shape of COM crystals based on the crystallographic data of Tazzoli and Domeneghetti [Taz80].

PAGE 82

Table 3 .3. Atomic coordinates of Whewellite (CaC2O4 H2O) Cocco and Sabelli [Coc62] Tazzo l i and D ome n eghetti [Taz80] Atom X y z X y z C(l) 0 0079 0 .32 20 0.2513 0.9832 0.3201 0.2452 C(2) 0.0 012 0.4296 0 .2540 1.0 009 0.427 0 0.2492 C(3) 0.5421 0.1103 0.1852 0.5189 0.1266 0 1812 C(4) 0.4822 0 1279 0.3200 0.4505 0.1173 0.3131 0(1) 0 0279 0.2811 0.1486 0.9756 0 2826 0.1322 0(2) 0.9981 0.4671 0 1389 1.0066 0.4659 0 1395 0(3) 0.0301 0.2844 0 3727 0 9799 0.2819 0.3550 0(4) 0.9911 0.4669 0.3605 1.0073 0 .4658 0 3614 0(5) 0 3827 0.1212 0.0727 0.3614 0 1418 0.0690 0(6) 0.7546 0.1261 0.1995 0.7245 0.1227 0.1974 0(7) 0.2690 0.1275 0 3017 0.2438 0.1229 0.2957 0(8) 0.6398 0.1170 0.4376 0 6073 0.1068 0.4264 Ca(l) 0. 0 049 0.1227 0.0639 0.9676 0.1243 0.0546 Ca(2) 0.0305 0.1263 0.4438 0.9968 0.1236 0.4357 W(l) 0.4032 0.4013 0.1059 0.3932 0.3459 0.1022 W(2) 0.6084 0 .3461 0.3977 0.5913 0.3829 0.3908 H(l) 0.487 0.372 0.051 H(2) 0.510 0.364 0.426 H(3) 0.530 0.367 0.320 Deganello and Piro [Deg81 a ] X y z -0 2452 0.3201 -0.2620 -0 2493 0.4270 -0.2481 -0.1812 0.1264 0.3377 -0.3130 0.1175 0.1375 -0.1322 -0.2826 -0 1566 0.1359 0.4660 0 1329 0.3555 0.2819 0.3751 -0.3614 -0.4658 -0 3541 0 0690 0.1418 0.2924 -0.1975 -0.1226 0.5271 -0.2958 -0.1229 -0 0519 -0.4263 -0 1066 0 1809 -0.0546 -0 1243 -0 0870 -0.4357 -0.1236 -0.4389 -0.6023 -0.1542 -0.2090 0 1093 -0.1172 -0.2940 0\ Vo

PAGE 83

66 between the theoretical atomic structures generated using ATOMS and the atomic structures from the papers by Deganello and Piro [Deg81 a] and Tazzoli and Domeneghetti [Taz80] the (010) and (100) planes generated with the atomic coordinates published by Cocco and Sabelli [Coc62] are exactly reconciled to the layering sequence reported by Deganello and Piro [Deg81a] and Tozzoli and Domeneghetti [Taz80] as shown in Figure 3.6(A) and (B). However, the (010) and (100) planes generated with the atomic coordinates by Tozzoli and Domeneghetti [Taz80] are not reconciled to the layering sequence reported by Deganello and Piro [Deg81a] and Tozzoli and Domeneghetti [Taz80]. The atomic structures of some oxalate ions are incomplete as shown in Figure 3.6(C) and (D). Furthermore, the (010) and (100) planes generated with the atomic coordinates ofDeganello and Piro [Deg81a] are completely different from the layering sequence reported by Deganello and Piro [Deg81a] and Tozzoli and Domeneghetti [Taz80] as shown in Figure 3 6(E) and (F). Thus the atomic coordinates of Cocco and Sabelli [Coc62] were used in computer simulations of atomic structure as a function of habit plane. 3 .3.1 COM particles ("33" particles) without seeds The COM crystals grown in distilled water without seeds are shown in 3. 5(A). The interpenetrating crystals have sharply-angled tips and a distinctive shape Figure 3.5(B) shows the equilibrium shapes of these interpenetrating twins with labeled face indices. For the twin operation, the individual crystal is reproduced according to the reflection on (101). The crystallographic form factors and central distances for the interpenetrating twins used in SHAPE calculations are listed in Table 3.4. Determinations made from observations of the computer calculation and the morphological form of the experimentally derived particles the { 010} and { 101} faces appear to be dominant.

PAGE 84

(A) (B) (C) ( D ) (E) (F) 0 : Hydrogen Q : Water : O xygen : Carbon O : Calcium Figure 3 .6. Theoretical atomic structures of COM crystals as a function of habit plane: (A) and (B) are the (010) and (100) planes respectively drawn using Cocco and Sabelli [Coc62] atomic coordinates ; (C) and (D) are the (010) and (100) planes respectively drawn using Tazzoli and Domeneghetti [Taz80] atomic coordinates ; (E) and (F) are the (010) and (100) planes respectively drawn using Deganello and Piro [Deg81a] atomic coordinates. 67

PAGE 85

Table 3.4. Crystal forms and corresponding central distance values used to generate the theoretical shape of COM crystals 33 COM Crystals* 00 COM Seed Crystals 32 COM Crystalst ( Aged at 90 C) Forms I 2 3 l 2 3 2 3 Face Indices 010 011 101 010/001 011 101 010/001 011 101 Central Distance (shape with (010) plane) 4.0 6.0 3 0 4.0 2.0 0.5 2.0 4.0 1.0 Central Distance (shape with (001) plane) 2 0 2 0 0.5 2 0 4.0 1.0 The equilibrium shape of the inte1-penetrating twin is generated according to the reflection on the ( l O 1). The equilibrium shape of the contact twin is generated according to the epitaxial operation. The ho~t crystal is specified by the face (001) and the vector (0 lO]and the guest crystal is specified by the face ( 00 l ) and the vector [ 0 TO]. The central distance for the epitactic face (001) is 2 0. 00

PAGE 86

69 3 3.2 COM particles ( 33" particles) with seeds The seed crystals ("00" particles) precipitated by mixing equal volumes (50 ml) of 1 M CaCl2 and 1 M K2C2O4 solutions are shown in Figure 3.7(A) The particle size varies between 100-500 nm. Although the shape and size are not uniform the crystals are composed of submicron size particles suitable for seed materials Figure 3. 7(B) shows the equilibrium shapes of the seed crystals with face indices as calculated by SHAPE with the crystallographic form factors and central distances for COM seed crystals listed in Table 3.4 The {l oi} faces appear to be dominant based on the computer calculation and the morphological form of the experimentally derived particles Also only the equilibrium shape generated with the (001) face was reconciled with the morphology of experimentally prepared seed crystals as shown in Figure 3.7(B). Figure 3.7(C) shows that the influence of seed particles on the size of COM crystals grown in distilled water. These interpenetrating crystals produced with 0.1 ml of seeding suspension ha v e similar shape to the COM crystals grown without seeds but are smaller in size. The maximum size of the particles is about 1 m. Thus it is demonstrated that the presence of seed particles may provide a low energy epitaxial surface in solution to lower the overall surface energy contribution to the nucleation barrier increasing nucleation frequency and reducing the particle size of COM crystals without changing the particle morphology Also the seeding studies suggest the possibility of controlling the size of the COM crystals ( "33" particles) by controlling the amount of seed materials. From the morphological difference between seed crystals and COM crystals synthesized with seeds the effect of seed crystals can be gauged from two aspects: (1) the presence of seed crystals in the solution system probably lowers the surface energy barrier of the system to the nucleation and growth of particles and ( 2 ) the transformation in this case is not only influenced by the number of added seed crystals

PAGE 87

(B) ( D ) C C D f A : (101) B : ( 011) C: ( 00 1) B : (Ol l) A : (10 1 ) B : ( 0 I I ) C: ( 001 ) D : (010) EE:H (e : 'j} A: (011) B: (Ol I) C : (010) D: (OIT) E: (101) F : ( OTT) Figure 3.7. SEM photomicrographs and theoretical equilibrium shapes of seed crystals and the COM crystals precipitated using see ds : (A) seed cry s tals (B) equilibrium shapes of seed crystals generated with the (001) and (010) faces respectively (C) COM crystals grown using seeds and (D) equi l ibrium shape of COM crystals. ----:i 0

PAGE 88

but also by forming larger numbers of nuclei due to the seeds and possibly secondary nucleation [Nyv85 Ran88]. 3 3 3 COM particles precipitated from homogeneous solution at 90 C 71 Habit modification of COM crystals grown with structurally specific additives gives further demonstration of specific preferential binding The COM crystals which were made using dimethyl oxalate decomposition from 2.5x103 M CaC12 solution at 90 C are shown in Figure 3.8(A) Habit modification is obvious for the COM crystals grown by dimethyl oxalate decomposition. Figure 3 .8(8) shows the equilibrium shapes of an individual COM crystal and contact twin with face indices The crystallographic form factors and central distances for COM crystals are listed in Table 3.4 For the contact twins the individual crystal is reproduced according to the epitaxial operation in SHAPE The host crystal is specified by the (001) face and the [O 1 OJ vector and the guest crystal is specified by the ( 00 I) face and the [010] vector. The central distance for the epitactic face ( 001) is 2.0 The { 101} faces appear to be dominant when comparing the computer calculation to the morphological form of the experimentally derived particles. Also the equilibrium shape generated with the (001) face more closely resembles the experimentally prepared COM crystals in contrast to the equilibrium shape generated with the (010) face as shown in Figure 3.8(8) and (C). 3 3.4 Atomic structures of COM particles as a function of habit plane for the high temperature form Equilibrium atomic structures of COM particles were generated as a function of habit plane to predict the atomic structures of specific surfaces. In whewellite there exist two different crystal structures : a high temperature structure (stability range 318-415 K ) and a low temperature structure (stability range 293-318 K) [Deg81 b Deg80]. Table 3 .5 and

PAGE 89

(C) A f-+ti A: (101) B A: ( !OT) B: (011) ( } c B : (010) C: (OlT) C: (Oll) D:(011) A D : (OT 1) E: (001) E : (011) F : (OOT) F: (011) Figure 3 .8. SEM photomicrograph and theoretical equilibrium shapes of the COM crystals precipitated from homogeneous solution at 90 C : (A) COM crystals (B) equilibrium shapes of an individual COM crystal and a crystal with a twin generated on the (001) face and (C) equilibrium shapes of COM crystals generated using the (010) face as the dominant face. ---..) N

PAGE 90

Table 3.5. Values of the crystallographic data for the high temperature form [Deg8 lb, Deg80] of Whewellite (CaC204 H20) a (A) b (A) C (A) Space Group Stability Range (K) Ca/Ox Ratio (010) (101) High Temperature Form 9.978(1) 7.295(1) 6.292(1) 12/ m 318-425 0 .9 58 1.658 73

PAGE 91

74 Table 3.6 summarize the crystallographic data and atomic coordinates of the high temperature form. Based on the crystallographic data and atomic coordinates, Figure 3.9 shows the equilibrium atomic structures of specific habit planes for the high temperature form. Habeger et al. [Hab97] has shown that the ratio of calcium atoms to oxalate molecules may have an important implication in the adhesion properties of the different crystal faces in a model environment. The ratio of calcium ions to oxalate ions is 0.958 in the (010) plane whereas the ratio of calcium ions to oxalate ions is 1.658 in the (101) plane for the high temperature form. Although the temperature in the human body cannot produce the high temperature phase of COM, the adsorbtion of ions or molecules may stabilize the high temperature structure at lower temperatures. The stabilization of the high temperature structure of COM in the urinary system may have a large implication in that the ( 101) has a Ca +2/ C204 2> 1 making the crystal face positively charged and able to electrostatically interact with predominately negatively charged biosurfaces found in the human body. 3.3.5 Characterization of COM crystals The morphologies reported herein were determined to be phase pure COM by XRD. A typical XRD pattern is shown in Figure 3 .10. As shown by SEM of the various COM morphologies, the "33 particle synthesizes without seeds were the most consistent morphology produced. Visually, they had very few differently shaped particles and a relatively narrow size distribution relative to the other particle morphologies synthesized. Therefore further characterization will only involve the 33" COM particles synthesized without seeds and further mention of COM particles refers only to this particular morphology. Peakfit software was used to fit a log normal curve to the experimentally produced particle size distribution. The particle size distribution was determined using an electrical zone sensing technique which results in an equivalent spherical diameter vs. frequency

PAGE 92

75 Table 3.6. Atomic coordinates for high temperature form [Deg81 b] of Whewellite (CaC2O4 H2O) High Temperature Form Atom X y z C(l) -0.0652 0.0 0.3984 C(2) 0 0 0.3937 0 0 0(1) 0.1105 -0.3157 0.1096 0(2) -0 0493 0 0 0.2119 0(3) 0 1783 0.0 0.4454 Ca(l) 0.1903 0.0 0 1751 Ca (2) W(l) -0 .35 52 0.0 0 0438 H(l) 0.45 0.0 0 04 H(2) 0 .33 0 .05 0.2

PAGE 93

76 (A) ( B ) O : Hydrogen Q : Water : O xygen : Carbon O : Calcium Figure 3.9. Theoretical atomic high temperature structure [Deg81b] (stability range: 318415 K) of COM crystals as a function of habit plane (A ) (010 ) plane and (B) ( 101 ).

PAGE 94

0.81 0.64 0.49 0.36 0.25 0.16 0.09 0.04 10.0 20.0 30.0 80.0 40.0 50.0 60.0 70.0 C2Ca04.H20 WHEWELLITE, SYN 20-231 100.0l I 60.0 I ~g: g __ ..y.J_.........,__ ........... ~.-.-.+,l ..... ,_~, ...... 1 ........... I....,~,......__.~, ...... ..... ............... \LL __._ ........,_~---.-~ ............. 10.0 20 0 30.0 40.0 50.0 60.0 70.0 Figure 3.10 A (top) typical experimental COM x-ray diffraction pattern and (bottom) the JCPDS file for the mineral whewellite CaC204H20. 77

PAGE 95

78 distribution as shown in Figure 3.11. The mean and standard deviation in particle size is 13. 04 m and 1.27 m respectively. The least square regression coefficient of the curve fit is 0 98. For purposes of calculation the relative dimensions of the COM particles were measured by analyzing multiple particles from SEM photomicrographs. The relative sizes illustrated in Figure 3.12 were used calculating surface areas of the two major crystallographic faces given the equivalent spherical diameter refer to Appendix A. Zeta potential of the COM crystals was determined as a function of pH in saturated COM solution as shown in Figure 3.13. Also on Figure 3.13 one data point representing the zeta potential of COM crystals in 10% AUIS at the buffered pH of 6.0 The zeta potential values for COM in saturated COM solution and in 10% AUIS at pH 6 are-32 mV and -21 mV respectively. As can be seen from Figure 3.13 the lower ionic strength 5x10-6 M COM saturated solution has a higher zeta potential than the higher ionic strength 3 .7lxl0-2 M 10% artificial urine at pH 6. This effect is expected and is due to compression of the electrical double layer. As the ionic strength of the electrolyte the COM crystals are suspended in increases (i.e. 10% AUIS to 50% AUIS to 100% AUIS), the value of zeta potential will decrease. When the ionic strength becomes high the measurement of zeta potential using Rank Particle Microelectrophoresis becomes impossible. Compression of the electrical double layer is extreme and the potential distribution near the particle surface drops rapidly as a function of distance into the solvent Figure 2.5 Therefore the particles do not move under the applied electric field and measurements of electrophoretic mobility cannot be made. Contact mode scanning probe microscopy (SPM) was performed on both the (010) and (101) crystallographic faces of a COM particle in saturated COM solution at 25 C to determine the surface roughness. The r a values or the mean roughness values for

PAGE 96

79 250 Experimental D ata 200 Fitted Curve >. () C 150 Q) ::J 0Q) lL cij 100 ~ Q) 0 50 0 0 10 2 0 30 40 Equ i valent Sph e rical D iam e ter (m) Figure 3 .11. The differential frequency vs equivalent spherical diameter particle size distribution of the experimentally produced COM crystals grown without seeds fit to a log-normal probability distribution

PAGE 97

80 ..,.,----A----~ A : B :C=1. 0 : 0.43 : 0 35 Figure 3 .12. The relative linear dimensions of experimentally produced COM crystals.

PAGE 98

0 -10 -> -20 s -ai ~ 30 Q) 0 0... ell a5 -40 N .... -50 .... -60 3 I CO M Saturated Solut i on & 10 % Artificial Urine -A. -\.,_.....--+\ f r I I 4 5 I I 6 7 pH !,f-, I I 8 9 10 81 Figure 3.13 Zeta potential as a function of pH for COM in saturated COM solution and zeta pot e ntial at pH 6 for COM in 10% AUIS.

PAGE 99

82 the COM (010) Figure 3.14 and ( 101), Figure 3.15 are 91nm and 81 nm respectively over a scan area of 5 m by 5 m. The high surface roughness of the COM crystal may be due to the growth mechanism of the twinned COM crystal. The crystals have a growth ledge at the twin boundary where the crystal may grow more rapidly on one half of the twin boundary but not on the other causing the roughness to be large across this boundary. 3 .4 Conclusions The equilibrium shapes with face indices of calcium oxalate monohydrate (COM) particles synthesized in different conditions have been successfully generated using the computer program SHAPE Computer calculations of theoretical crystallographic shapes have been reconciled to the observed shapes of experimentally synthesized COM particles under various conditions. Also, the seeding studies suggest that it is possible to control the size of COM crystals ("33" particles) by controlling the amount of seed materials. The computer program SHAPE was demonstrated to be a useful tool to determine the face indices of COM crystals The equilibrium shape of COM crystals based on the lattice parameters of Deganello and Piro [Deg8 la] is reconciled to the morphology of the experimentally synthesized COM crystals from the comparison between the calculated shape and the morphological form of the experimentally derived particles. Based on the face indices of the equilibrium shapes and atomic coordinates the atomic structures of COM particles as a function of habit plane have also been generated by using the computer program A TOMS From the comparison between the theoretical atomic structures generated by ATOMS the (010) and (100) planes generated with the atomic coordinates published by Cocco and Sabelli [Coc62] are reconciled to the layering sequence reported by Deganello and Piro [Deg8 la] and Tozzoli and Domeneghetti [Taz80].

PAGE 100

0 5 0 10.0 15.0 10.0 5.0 0 15, 0 JJM 8 3 Box Statistics 2 range 507.09 nM Hean 5 .475 nM RMS (Rq) 91.527 nM Hean roughness CRa) 31 ,592 nM Surface area Box x diMension 4 .638 JJM Boxy diMension 4.374 JJM Figure 3 1 4 Contact mode SPM scan of the COM (010) crystallographic face showing the surface roughness.

PAGE 101

0 2.5 5.0 7.5 10.0 7.5 5 0 2.5 0 10,0J,JM 8 4 Box Statistics 2 range 1 ,025 JJM Hean -6,614 nM RMS CRq) 175.45 nM Hean roughness (Ra) 81,202 nM Surface area Box x diMens i on 4 951 JJM Boxy diMension 4.834 JJM Figure 3 .15 Contact mode SPM scan of the COM ( l 01) crystallographic face showing the surface roughness.

PAGE 102

CHAPTER4 EVALUATION OF THE CALCIUM OXALATE MOMOHYDRATE HAMAKER CONSTANT BASED ON STATIC DIELECTRIC CONSTANT DETERMINATION AND ELECTRONIC POLARIZATION 4.1 Introduction The ability to predict calcium oxalate monohydrate (COM) stability with respect to aggregation and adhesion for patients suffering from a kidney stone is of great importance. Since COM crystals are a major component of many types of urinary stones the accurate predictions for COM aggregation in the urinary environment can aid in the treatment of such stone occurrences Many of the interactions that take place in the prediction and urinary environment are dominated by van der Waals forces due to the relatively high ionic strength in the urinary environment which tends to collapse the electrical double layer and hence diminishes e l ectrostatic effects. In fact Boeve et al. [Boe94] have measured the zeta potential for COM in an artificial urinary ion environment (ionic strength 0.33 M) to be < 5 mV Both aggregation of like particles and adhesion of particles to unlike surfaces can potentially lead to the production of a kidney stone [Fin84]. Therefore an accurate value of the Hamaker constant for the material COM is important in predicting the van der Waals interactions with l ike and un like materials COM crystals have been shown previously to adhere to unlike materials such as fused quartz [Chapter 5] and various protein substrates [Chapter 6]. The adhesion of COM crystals to the fused quartz and protein substrates occurred in a relatively low ionic strength ( -2x 10-4 M) environment in spite of an electrostatic repulsive interaction between the two materials This 85

PAGE 103

86 demonstrates the importance of understanding the nature of van der Waals interactions and of knowing the value of the Hamaker constant. Other aspects of particle dispersion such as those arising from surface energy which may contribute to the total interaction energy are beyond the scope of this study and will not be considered herein. The origin of the van der Waals attractive forces and the calculation of the Hamaker constant have been discussed in several reviews and, therefore will only be briefly summarized [Hou80 Isr92]. The attractive van der Waals forces are composed of multiple intermolecular interactions between the ions molecules and electrons that make up the particles interacting across a dielectric medium such as water. The mutual attraction between interacting particles arise from harmonic oscillations at the molecular atomic or subatomic level. There are three primary sources of such intermolecular interactions depending upon the nature of the interacting species. Some of the more important specific interactions are known as the Keesom Debye and London interactions The Keesom interactions are due to molecular dipoles in the particles. Debye interactions occur when molecular dipoles in one particle induce electronic polarization in the other interacting particle. As such Keesom and Debye interactions only take place if a material has one or more dipoles present within its structure In contrast London interactions are more ubiquitous because London interactions are due to mutual electronic polarization with all atoms of course composed of electrons. Lifshitz and co-workers demonstrated that an understanding of the dielectric constant over as broad a frequency range as possible is required to reconcile all of the various possible intermolecular interactions between particles in order to calculate the most accurate Hamaker constant [Hou80 Isr92]. The objective of the present work is to evaluate and increase the accuracy of the value of the Hamaker constant of COM previously calculated by Adair et al. [ Ada91] taking into account only electronic polarization. The re-evaluation will use the Tabor Winterton relationship incorporating new dielectric data for COM [Hou80 Isr92] The

PAGE 104

TaborWinterton relationship for dissimilar materials interacting across a medium is [Hou80 Isr92], 87 where Kand n refer to the static dielectric constant and the index of refraction at the characteristic frequency, Ve, for electronic polarization., respectively The subscripts 1 and 2 represent the interacting materials and the subscript 3 represents the medium. The TaborWinterton relationship is based on Lifshitz theory and assumes that the electronic relaxation frequency V e, is the same for all materials This assumption will not generally lead to too large an error since the value of V e for many materials is 3x1015 Hz A simplified version of the Tabor-Winterton relationship [Tab69] is shown in equation (4 2) This equation may be used to calculate the Hamaker constant for the interaction of identical materials across a medium such as for the COM-water-COM system [Hou80 Isr92], (4.2) where A 1 3 1 is the Hamaker constant for the interaction of material 1 (COM) with material 1 across a dielectric medium material 3 (water). Table 4.1 shows the index of refraction values as a function of optical direction and wavelength for COM which were used in a prior Hamaker constant calculation based solely on electronic polarization [ Ada91]. In

PAGE 105

88 Table 4.1. Index of refraction for COM as a function of optical direction and wa v elength [Pal51 ] Optical Direction a y 535 0 1.4939 1.5599 1.6567 Wavelength (run) 589.0 1.4909 1.554 1.6502 670 8 1.4878 1.5513 1.645

PAGE 106

89 equation ( 4.2), the second term on the right hand side accounts for electronic polarization while the first term accommodates static polarization terms. Before using the Tabor-Winterton relationship, absorption data in the microwave and infrared (IR) region of the electromagnetic spectrum is needed if the static dielectric constant is much larger than the dielectric constant (i.e. n2 ) for electronic polarization. The absorption in the IR often can be extrapolated to zero frequency and hence the static dielectric constant of the material may be determined The dielectric constant of a material can be deconvoluted from the dielectric properties of a composite using mixing rules, which are shown in Table 4.2 [8]. In the current work, the dielectric constant of a COM/silicone composite was measured and the COM dielectric constant was determined from dielectric mixing rules. Because the particles could not be grown to a large enough size to measure the single crystal properties and the production of large amounts of well crystallized particles is inefficient and time consuming composites as a function of solids loading of the dispersed COM par1icles in a matrix of known dielectric constant Eccosil 5019 were produced to measure the dielectric properties of COM. A number of silane coupling agents were examined to determine which material best dispersed COM in silicone. The dielectric constant was then used to calculate the Hamaker constant of COM using equations (4.1) or (4.2) 4.2 Materials and Methods COM particles were produced by the procedure described in Chapter 3.2 1.1. COM powder was dispersed in silicone (Eccosil 5019 Emerson and CurningTM, Inc. Woburn MA) using the silane coupling agents (Petrarch Systems Inc. Bristol PA. The COM particles must be homogeneously dispersed in the matrix to be properly described by the composite mixing rules. In order to achieve good COM particle dispersion in the silicone matrix a number of silane coupling agents were tested : 3aminopropyltriethoxysilane (APTES) ; 2-(3,4 epoxycyclohexyl) ethyltrimethoxysilane

PAGE 107

Table 4.2. The composite mixing rules evaluated where K is the is the static dielectric constant of the composite and V e and vd are the volume fractions of the continuous and dispersed phases respectively and K c and K d are the static dielectric constants for the continuous and dispersed phases respectively. Mixing Rule Series Parallel Maxwell Lichtenecker Mathematic Relationship 90

PAGE 108

91 (ECHEMS); (3-glycidoxypropyl) trimethoxysilane (GPTMS); isobutyltrimethoxysilane (IBTMS) ; 3-methacryloxypropyltrimethoxysilane (MPS); 3-(n-styrylmethyl-2aminoethylamino) propyltrimethoxysilane hydrochloride (SMAEPS) ; and vinyltriacetoxysilane (VAS). Table 4 3 shows the molecular structures of the silane coupling agents. A typical grafting solution to graft the coupling agent to COM surfaces consisted of 0 6 M coupling agent 0.33 M methanol 0.002 M acetic acid 0.00285 M deionized water, and 9.4 M toluene was prepared and equilibrated for 1 hour at 25 C 0 .05 g of COM crystals was added to the solution containing the coupling agent and sonicated for 1 minute. One hour after sonication the particles were filtered from solution using 0.22 m filter paper. The crystals were dried overnight under vacuum. It is important that the COM particles are well-dispersed in the silicone rubber matrix. Therefore low solids loading ( 0.01 v / o solids) samples were prepared with COM particles with each of the grafted coupling agents in the silicone rubber by hand mixing. A drop of the low solids loading mixture was placed on a glass serological ring slide, a cover slip placed over the ring and the composite mixture allowed to cure in place. The relatively transparent composites were then examined with optical microscopy to evaluate which silane coupling agent was most effective in dispersing the COM particles in the silicon rubber. COM crystals with the silane coupling agent which gave the best dispersion in the preliminary experiments were mixed with Eccosil 5019 in concentrations of 10, 20 25, and 30 volume percent and cast into 22 mm diameter pellets approximately 4 mm thick. The samples were cured at room temperature under a vacuum to remove any entrapped air. Final samples were cut from the 22 mm pellets using a 5 mm diameter brass cork borer. The samples were electroded using a conductive silver paint. A multi-frequency LCR meter (4274A, Hewlett Packard Rockville, MD) was used in conjunction with a computer controlled temperature chamber (MK 2300 Delta Design Inc. San Diego CA) to measure the capacitance and dissipation factor as a function of temperature and

PAGE 109

Table 4.3. Molecular structures of the silane coupling agents used to disperse COM in Eccosil 5019 silicone Product Acronym APTES ECHEMS GPTMS IBTMS Molecular Name 3-aminopropyltriethoxysilane 2-(3,4-epoxycyclohexyl) ethy ltrimethox ysilane 3-(gl ycidopropy l)trimethoxysilane isobutyltrimethoxysilane Molecular Structure ~I I _J-\,,, "1 I ~t,o 0 <\_\_ / "---1-\ 0 / _/ 0/ \_l_o I \ 0 / \0 N

PAGE 110

Table 4.3 (cont.). Product Acronym MPS SMAEPS VAS Molecular Name 3-methacryloxypropyltrimethoxysilane 3-(n-styrylmethyl-2-aminoethylamino) propyltrimethoxy s ilane hydrochloride vinyltriacetoxy s ilane Molecular Structure i-~(o J \ / ~~, I \_~~/ -0 0 I i---0\ 0~ \_J_ o r I o y o 0 0 / \0 uJ

PAGE 111

94 frequency (-100 Hz to 1 MHz) The equipment calculates dielectric permittivity from the capacitance sample geometry, and permittivity of air. All measurements followed the American Standards and Testing Materials (ASTM) designation D 150-81. An extrapolation to zero frequency gave the value of the dielectric constant at the static frequency for the composite K 1 The dielectric constant of Eccosil 5019 at 1 MHz is 2.8. The dielectric constant of water at 37C is 73.4 [Arc90]. Based on the solids loading of COM crystals in the polymer a suitable mixing rule was used to deconvolute the composite properties into the particle and polymer dielectric properties [Yar87] Four different composite mixing rules series parallel Maxwell s and Lichtenecker s (or logarithmic) were used to determine the static dielectric constant for COM. The mixing rule giving the best fit to the data was used in the final calculation of the COM Hamaker constant. The fit was performed using the Math Curve Fitter in SigmaPlot version 5 for Macintosh software (Jandel Scientific San Rafael CA). Values of refractive index as a function of frequency were used to construct Cauchy plots for each optical direction of COM and for water [Pal51]. A straight line fit to the data yielded a linear equation from which values of the main electronic absorption frequency typically in the UV V11v, the dielectric constant at the UV characteristic frequency Eim and the index of refraction at the characteristic frequency n0 were determined. The values of refractive index and dielectric constant were substituted into equation ( 4.2) to determine the Hamaker constants of COM crystals interacting with like optical directions (i.e., for interactions such as aCOMH2O -aCOM). The index of refraction of water at 37 C is 1.322 [Tho85]. Equation (4.1) was used to calculate Hamaker constants of COM crystals interacting with unlike optical directions (i.e., for interactions such as aCOM-H2O-PCOM).

PAGE 112

95 4.3 Results and Discussion Figure 4.1 shows SEM micro graphs of some of the COM particle dispersions in Eccosil 5019. Relatively poor dispersion was exhibited by COM in Eccosil in the absence of a coupling agent as shown in Figure 4 .1 a. The COM particles were unevenly distributed throughout the matrix and a large number of small agglomerates were present. Figure 4.1 b shows the sample containing SMAEPS coupling agent. Very poor dispersion was obtained in this case as indicated by tl1e gray area on the right of the photomicrograph which is a large mass of agglomerated COM particles. This is because the probably styryl group reacts poorly with the silicone rubber matrix as well as the charged groups ( Table 4.3). This composite was obviously inhomogeneous and not suitable for dielectric measurement. The best dispersion of COM in E ccosil resulted from the use of GPTMS as shown in Figure 4.1 c. This coupling agent has a glycido group which reacts favorably with the silicon rubber matrix. The GPTMS sample was composed of a relatively high number concentration of particles evenly distributed throughout the sample. All other COM particles treated with the coupling agents in Table 4.3 did not provide as good dispersion as the GPTMS. Therefore the coupling agent GPTMS was used in the remainder of the experiments for evaluation of the composite dielectric properties at highe1 solids loading. The four composite rules of mixing used in the determination of the static dielectric constant are shown in Table 4.2. The rules were evaluated for theii. ability to deconvolute the dielectric constant of COM from COM/silicone composite properties. Figure 4.2 shows the fits of all of the mixing rules to the experimental data. Based on the ~ ization of the sum of squared differences between the measured and predicted values generated using the SigmaPlot software, the series relationship gave the best fit overall. The dielectric value generated for the parallel mixing rule 7.8x106 is not physically relevant therefore this particular mixing rule was discarded. Table 4.4 gives

PAGE 113

96 (A) (B) (C) Figure 4.1. Optical photomicrographs of COM particles dispersed in Eccosil 5019 using (A) no coupling agent (B) SMAEPS coupling agent and (C) GPTMS coupling agent (bar-=100 m).

PAGE 114

97 60 Experimenta l 50 Lichte necker Maxwell c 40 S eries co UJ P arallel C 0 30 0 u ~ 20 ~ 0 10 0 0. Volu m e Fraction C O M Figure 4.2. The composite mixing rules fit to experimental dielectric data.

PAGE 115

Table 4.4 Dielectric constant val ues of COM d etermined by fitting the mixing rules to the experimental data Dielectric Constant Regression Coefficient NA-not applicable Lichtenecker 28 9 0.75 Composite Mixing Rule Maxwell 46. 1 0.77 parallel 7800000 NA senes 1 1.8 0.80 98

PAGE 116

99 the values of the dielectric constants and the value of the nonlinear regression coefficient determined using a best fit of each of the above equations to the experimental data. The series model although designed to represent the extreme case of alternating phases of a composite which are perpendicular to the electrode materials gave the best description of the data with a nonlinear regression coefficient of 0.80. The composite shown in Figure 4 .lc, obviously should not be represented by the series mixing model. Maxwell s mixing model which was theoretically derived to describe a matrix containing small spherical inclusions [Von54] may be a closer representation of the COM/silicone composite than the series model ; although COM particles are not spherical. Although the Lichtenecker model has been previously invoked to describe systems of irregularly shaped inclusions [Mon47] similar to the COM/silicone composite in the present work it gave the lowest regression coefficient 0.75 The series, Maxwell and Lichtenecker models describe similar dielectric behavior up to 35 volume percent (v/o) dispersed phase Above this concentration the differences between the models become discernible however in this work none of the composites were prepared with a content exceeding 30 v / o mixtures of COM with silicone rubber were too viscous to cast samples above 30 v / o COM. A composite having a dispersed phase of 50 v / o may be used to distinguish the series model from the Maxwell and Lichtenecker models. Furthermore the Maxwell and Lichtenecker models may be discernible at 80 v / o dispersed phase where the deviation between models becomes larger. However a homogeneous composite with more than a 40 v / o dispersed phase is very difficult to prepare and such a large content of dispersed phase may exceed the percolation limit of the composite. A large discrepancy in dielectric constant of 75% resulted from the series Lichtenecker and Maxwell mixing rules however this translates to a difference of only 10% (1.6xl0-2 1 J) in a calculated Hamaker constant. Since the Lichtenecker model has been shown to describe composites of the COM/ silicone type [Mon4 7] and given the

PAGE 117

100 difficulty in preparing a composite with a large second phase content the Lichtenecker mixing model was used to describe the composite dielectric properties. In order to calculate the Hamaker constant using the Tabor-Winterton relationship experimental data involving both the static dielectric response and the absorption spectra in the ultraviolet frequency regime ( m> 101 6 rad/ s ) must be known [Hou80] Plots of [n(w)2-1] vs. [(n(w)2-l)*w], known as Cauchy plots [Hou80] were const1ucted for each optical direction of COM and for water, as shown in Figure 4.3. The Cauchy plot gives a straight line with slope equal to 1 / m:iv and intercept equal to ci,v, where m uv (rad/s) is the characteristic frequency in the ultraviolet region of the electromagnetic spectrum and C uv is the oscillator strength. The corresponding refractive index and dielectric constant at tl1e UV characteristic frequency were determined using the equations [Hou80] ( 4.3 ) and ( 4.4 ) respectively. All of the linear regression coefficients of the line fit were between 0.98 and 1. Table 4.5 gives the values of UV characteristic frequency the dielectric constant at the UV characteristic frequency and the index of refraction at the UV characteristic frequency for water and the three optical directions of COM. The dielectric data determined for water compares very well with the values determined b y Hougl1 and White [Hou80]. Because large absorbance exist at frequencies lower than 101 6 rad/ s for COM, the value of uv determined from optical data will estimate the constant (i.e. the value of ( e0 -n 2 ) is significantl y larger than zero). When the static dielectric constant is significantly larger than the dielectric constant in the visible the i11frared term, n2 o f the absorption

PAGE 118

1.5 i I N 5 1.0 '-./ C 0.5 0 a 9 6 Alpha Beta Gamma Water 0.0 --------------Oe+OO 1 e+31 2e+31 3e+31 (n(w) 2 -1 )*w2 Figure 4.3. Cauchy plots for water and COM as a function of optical direction. 101

PAGE 119

Table 4.5. Calculated values of the UV characteristic frequency and corresponding dielectric constants and refractive indices as a function of crystallographic direction determined from Cauchy plots. Material a-COM ~-COM y-COM water (xl 01 6 rad/s) 1.77 1.58 1.46 1.87 (xl 01 5 Hz) 2.8 2.51 2.32 2.98 2.183 2.361 2.642 1.7475 1.477 1.5376 1.625 1.322 102

PAGE 120

103 spectra cannot be ignored [ 4]. In the current calculations the value of the Hamaker constant due to contributions at zero frequency ( 1.68x10-21 J) is less than one order of magnitude lower than the Hamaker constant due to contributions at higher frequencies ( 8-24)x10-21 J. Due to the equally strong dependence on both the zero and non-zero frequency terms of the Tabor-Winterton relationship, measurement of the dielectric properties of the COM/ silicone composite and subsequent deconvolution to the COM static dielectric constant is important. The calculations were performed using the assumption made by Tabor and Winterton that the UV characteristic frequency of most materials is approximately 3x101 5 Hz. The values of the characteristic frequency for COM and water shown in Table 4 5 are slightly lower than the assumed value of Tabor and Winterton at (2.32-2.98)x1015 Hz. This assumption led to an increase of 11 % in the Hamaker constant over the value of the Hamaker constant calculated if the actual determined values of characteristic frequency for the interacting materials were averaged together. Table 4.6 compares previously published values by Adair et al. [Ada91] to the current values calculated using the TaborWinterton relationship The average current value of, 13.7 (.90)x10-2 1 J compares favorably to the previously published value of, 11 .4 ( .42)x 10-2 1 J. Thus the addition of the static dielectric constant increases the Hamaker constant by about 15 percent. The improved accuracy of the COM Hamaker constant will allow increased reliability and confidence in calculating COM aggregation stability in the urinary environment. 4.4 Conclusions The Hamaker constant of COM was determined using the Tabor-Winterton relationship. The dielectric constant of COM wa s deconvoluted from a COM/silicone composite using various mixing rules. The series mixing rule yielded the highest nonlinear regression coefficient however the composites were clearly not representative of the

PAGE 121

104 series model. Therefore the Lichtenecker model previously shown to describe irre g ularly shaped inclusions in a matrix [Mon47] was used and a value of 28.9 for the COM dielectric constant was determined Plots of refractive index as a function of optical direction and wavelength were used to generate the ultraviolet absorption frequenc y and the dielectric and refractive index at the UV absorption frequency for use in the Hamaker constant calculation. The resulting value of the Hamaker constant 13.7 (.90)x10-2 1 J compares well with previously published values of 11.4 (.42)x10-2 1 J determined using the Gregory approximation.

PAGE 122

105 Table 4 .6. Comparison of A131 calculated for COM using Gregory's approximation vs. the Tabor-Winterton relationship oata taken from Adair et al. [ Ada91]. Type of Interaction a-H 20-a ~ H 20-~ y-H20-y a-H20-~ a-H20-y ~-H2 0 y Average 95% Confidence Interval A131 from Gregory s Approximation (x102 1 J) 7.13 11.3 16.3 8.99 10.8 13.6 11.4.421 A131 from TaborWinterton Relationship (xl 0 2 1 J) 6 .98 12. 6 23. 5 9.33 12.7 17.1 13. 7 5 90 J

PAGE 123

CHAPTER 5 DEVELOPMENT OF THE DYNAMIC WET CELL AND STREAMING POTENTIAL MEASUREMENTS 5 .1 Introduction The study of adhering materials is fundamental and very broad in application. For example the adhesion of Fe203 to a glass substrate with and without a layer of gelatin [Ryd95] of red blood cells to glass [Moh74] of human fibroblasts to glass [ v an92] of submicrometer particles to silicon [Bus93] and of bacteria to glass [Bus92] have all been measured using a hydrodynamic technique. Adhesion is particularly important in human kidney stone disease and other pathological biomineralization processes. Numerous studies have shown that the adhesion of calcium oxalate monohydrate (COM) particles to the wall of a kidney tubule may eventually lead to stone formation [Fin78a Fin84 Kok90]. In v ivo calcium oxalate particle attachment to a nephron wall has also been exhibited in a rat animal model [Kha91]. However the actual mechanism ( s ) of adhesion within the biophysical environment of the human kidney is not well understood. In past studies researchers have used many techniques to evaluate adhesion [Vis76 Cor66, Zim82]. These include rotating discs [Kri94] packed columns [Kal87] centrifugation methods [Kor60] vibrating methods [Der61] the surface force apparatus [Isr78] and a h y drod y namic method utilizing parallel plates [Pel82 Pel84]. The mos t recent technique for measuring the force with which a particle adheres to a surface is the scanning probe microscope (SPM) [Duc91]. These and other adhesion techniques are best suited to specific material systems. Therefore further discussion will be limited to only the hydrodynamic parallel-plate adhesion measuring technique which is considered to be most suitable for the measurement of COM crystals in model biological systems. 106

PAGE 124

107 Many researchers have studied COM crystals interacting with kidne y cells using in vitro experiments [Big96 Man91 Man94, Lei96 Ebi95 Ver96]. However these studies were static in nature lacking the hydrodynamic conditions that are found in the human kidney. The current adhesion investigation employs a measuring technique which is applicable to the study of kidney stone disease because the flow can easily be changed from laminar to peristaltic and nearly any eluent can be used for the interacting medium Also a number of different substrates may be used ranging from individual protein la y ers to actual cells under aseptic conditions The result of this measuring technique is a quantitative value of the force per unit area with which a crystal adheres to a substrate designated as the adhesive strength In the current study an apparatus was developed based on the principles of the capillary adhesion technique reported by Pelton et al. [Pel84 Pel82] to measure the adhesive strength between polystyrene spheres and a Pyrex substrate. A similar device to the current dynamic wet cell was developed by Hochnrnth et al. [Hoc73] and has been used in many investigations to study both attachment and removal of cellular material to specific surfaces [Moh74 van92 Dor77 For84 Tru90] However in the current investigation the adhesion of COM particles to a model fused silica surface was measured The apparatus shown in Figure 5.1 consists of an adhesion chamber incorporating two fused quartz slides separated by a spacer. The eluent flows through the dynamic wet cell at a specified rate which is calibrated against the pump drive. A camera mounted above the adhesion cell which is connected to a monitor and video cassette recorder records activity in the cell during each experiment. The linear particle velocity v, at a distance of one particle radius from the cell wall is calculated using the Poiseuille equation 2Q(R2 -r2 ) v=-----nR4 (5.1)

PAGE 125

108 (A) 0 0 0 0 0 0 0 r, 0 LI 0 0 0 ~o _____ o ___ o _____ o_,)'1PMMA I L-c .... -_ ........ ___,.__. _r::..L..---..L.._:::,....__,, I (B) Figure 5.1. (A) A picture and (B) a schema t ic diagram of the dynamic wet cell developed to measure adhesion of particu l ate to surfaces. In (B), structural thru-holes have been omitte d from the drawing to improve clarity.

PAGE 126

where Q is the volumetric flow rate, r is the particle radius and R is half the distance between the fused quartz slides. The corresponding displacement force acting at one particle radius F d is calculated using the fluid properties particle dimensions and the calculated linear velocity from equation (5.1) using the formula [Gol67] 109 = 1.7(6JTTJrv) (5.2) where Tl is the eluent viscosity r is the particle radius, and v is the linear velocity. An instrument to determine the zeta potential on a substrate by measuring the streaming potential was also developed to study the contribution of electrostatic interactions to the adhesion of COM to various substrates The instrument was similar to the rectangular capillary used by van Wagenen and Andrade [van80] and was used to determine the charge on the surface of the substrate material fused quartz in saturated COM solution conditions 5 .2 Materials and Methods 5 2. l Adhesion Measurements using the Dynamic Wet Cell COM crystals prepared as described in Chapter 3 were used throughout the following adhesion experiments. A 0 2 volume percent (V/ o) suspension of COM crystals in COM saturated solution was prepared and equilibrated in solution for 24 hours. Prior to assembly of the adhesion cell the glass substrates were surface treated and a uniform silanol surface ensured [Ile79] by soaking in 1 M KOH, 1 M HN03 and 1 M KOH consecutively for 30 seconds each. Before and after each acid or base wash the slides were washed with copious quantities of deionized water ( > 16 M.Q cm). The treated slides were stored in deionized water until being used in the adhesion measurements.

PAGE 127

110 Before adhesion measurements were performed the COM suspension and the dynamic wet cell were equilibrated at 37 C using two temperature controlled water baths. During the adhesion experiment the dynamic wet cell and the eluent were maintained at 3 7 C. The suspension was sonicated for 3 minutes and shaken using a vortex mixer for 1 minute to minimize the presence of agglomerates. The suspension was then injected into the dynamic wet cell using an injection port adjacent to the entrance of the cell and the particles settled out of solution onto the spectroscopic grade, fused quartz substrate (C O Quartz Esco Products Oak Ridge NJ) for 10 minutes. Given the size and density of the COM particles 10 minutes is more than sufficient time for the COM particles to settle from solution based on the Stokes settling equation. Also preliminary experiments showed that adhesion of COM to the fused quartz substrate did not change with increased residence time of contact between particle and glass before flow up to 60 minutes. The number of particles adhering to the fused quartz as a function of flow rate was determined by videotaping the particles adhering to the lower substrate in the dynamic wet cell. A low light CCD camera (JE-3662HR Javelin Electronics Torrance, CA) was mounted to the eyepiece of the microscope head of a Rank Brothers electrophoresis device normally configured to measure zeta potential in a capillary tube In this case, the dynamic wet cell replaces the capillary tube. The total magnification of the microscope was 250x given by a 20x water immersion objective lens l 25x intermediate lens and a 1 Ox eyepiece. The camera output was connected to a video cassette recorder (Magnavox VR9342 Philips Consumer Electronics Co., Knox vi lle TN) and to a real time video monitor (CVM14, Javelin Electronics, Torrance, CA). Taping proceeded continuously for the length of each experimental run. The flow rate was manually incrementally increased from Oto 136 ml/min using a variable speed pump dri v e (Ismatec SA Glattbrugg Switzerland) with a gear pump (Micropump Corporation Vancouver WA) The high flow rates where necessar y to produce shear stresses on the

PAGE 128

111 COM particles between 3000 and 5000 Pa. An audio connection consisting of a microphone (Radio Shack 33-3003 Tandy Corporation Fort Worth, TX) and microphone mixer (Radio Shack 32-1105A, Tandy Corporation Fort Wortl1 TX) attached to the VCR and monitor allowed the operator to record the flow rate for subsequent analysis. The duration for eacl1 particular flow rate was approximately 1 minute. A picture and a schematic diagram of the dynamic wet cell supporting equipment are shown in Figure 5.2. After completion of an adhesion experiment the videotape was analyzed for the number of particles adhering to the substrate at each flow rate. An example of an experiment is illustrated in Figure 5.3. Also the crystallographic orientation of each particle remaining after each particular flow rate was recorded. The force acting at one particle radius at each flow rate was calculated using equations (5.1) and (5.2) for particles -in both the (010) and ( 101) particle orientation. The corresponding average strength on each particle was calculated using the displacement force and the average area of the crystal faces. The COM particles used in the current study have the specific ratio of crystallographic dimensions shown in Figure 5.4c. These ratios were determined from -SEM analysis and used to calculate the average areas of the (010) and (101) crystallographic faces of the COM crystal which were both r--J 111 m2 The COM crystallographic faces shown in Figure 5.4b were determined using x ray data in the computer program SHAPE (Shape Software Kingsport, TN). Cody and Cody [Cod94] had previously indexed the faces correctly for the morphology of COM particles used in the experiments which aided in the modeling. The fractional coordinates published by Cocco [Coc61] were input to the computer program ATOMS (Shape Software, Kingsport TN), to display the atomic structure as a function of crystallographic habit as shown in Figure 5.5.

PAGE 129

(A) (B) CONSTANT TEMPERATURE BATH PUMP INPUT CONSTANT TEMPERATURE BATH 112 TV VCR Figure 5.2. (A) A picture and (B) a schematic diagram of the dynamic wet cell supporting eq u ipment.

PAGE 130

(A) (B) (C) Figure 5 3. A time lapse sequence of events during an adhesion experiment. The flow rates are (A) 12 ml/min, (B) 53 ml / min, and (C) 108ml/min. (bar = 40 m) 113

PAGE 131

114 (A) (011) (010) (011) << (101) >-> 1(i 1(101) (B) (011) (C) A:B: C =1. 0 : 0.43:0 35 Figure 5.4. COM particles of controlled morphology shown in a (A) scanning electron micrograph (B) modeled using SHAPE software showing the two dominant crystallographic faces (010) and ( 101 ), and (C) demonstrating the relative crystallographic size ratios.

PAGE 132

(010) (101) Ca/Ox=0 909 Ca /Ox=1 ) ) ) ) I I 11 I I I I I I I Calcium Oxygen 20 A 0 Carbon 0 Water Molecule Figure 5.5 Theoretical (010) and ( I 01) crystallographic planes of calcium oxalate monohydrate The Ca2+;c20/ ratio is given above each theoretical atomic structure 115

PAGE 133

116 5.2.2 Streaming Potential Measurement The zeta potential of the fused quartz substrates was determined using a streaming potential instrument which was built in the laboratory The streaming capillary consisted of two blocks of polymethylmethacrylate machine into the configuration shown in Figure 5.6. The cross sectional area of the rectangular capillary is 0.448 cm2 The streaming cell was designed to use the same size substrates that are used in the dynamic wet cell. The only difference being the streaming cell used two more slides to increase the surface area interacting with the solution. A platinum electrode is placed on each side of the streaming capillary shown in Figure 5 6 to measure the potential difference across the streaming channel. The electrodes consisted of woven platinum mesh. The leads to the electrodes are platinum wire and are woven into the mesh to eliminate any potentials which might arise due to soldering with another metal causing a galvanic cell. The platinum electrodes were blackened using a solution of 2% chloroplatanic acid (YSI 3140 YSI Incorporated Yellow Springs OH). Blackening is the process of depositing colloidal platinum on the surface of the platinum electrodes to increase the overall surface area and increase the sensitivity. The electrodes were connected to a high impedance electrometer 5xl01 3 n (614 Electrometer, Keithly Instruments Inc., Cleveland OH). The output of the electrometer was connected to a resistance-capacitance (RC) circuit shown in Figure 5.7. The circuit developed by Hom and Onoda [Hor77] is used to subtract potentials due to asymmetrical electrodes and / or electrode polarization of the electrolyte. The output of the RC circuit is connected to a strip chart recorder to record the streaming potential. Figure 5.8 is a picture the actual instrument and a schematic diagram of the streaming potential instrument.

PAGE 134

(B) (A) TOP VIEW ~-----------~ v---------v I I I I I I I I I I I I I I I I I I I I I L _Q I L _Q_ _I I ~-----------J I ~---------------J FRONT VIEW I I I ___ ...J I --c:-':: = = :J ~----SIDE V IEW Figure 5 6 A (A) picture and a (B) schema t ic d iagram of the streaming potential cell. 117

PAGE 135

Electrometer R(out)=10 kn flow rest 100 k Q 118 670Q 100 k Q chart recorder 330 Q Figure 5.7. A diagram of the R-C circuit used to eliminate asymmetry and electrode polarizations [Hor77].

PAGE 136

(B) (A) Const ant Temperatu r e Bath P u mp RC C ircuit Thermocouple Solutio n Reservior Chart Recorder I 0.0122 I E l ectromete r ~ v 119 Pressure j n Output LJ Reserv i o r St r e aming Pot e n t ial C ell Figure 5 8 (A) A picture and (B) a schematic diagram of the streaming potential instrument including all of the supporting equipment: solution reservoir streaming cell or capillary solution collection reservoir electrometer R-C circuit strip chart recorder peristaltic pump constant temperature bath recirculating vessel and thermocouple

PAGE 137

120 Streaming potentials Vs, were recorded for various differences in pressure M across the streaming capillary given by the difference in the pressure head of the holding reservoir and the output of the tubing as shown in Figure 5 .8. The different pressures were calculated using the equation [Esk68] M = pgt:.h (5. 3) where pis the density of H20 at 37 C, 0.99375 g / cm3 [Lid90] g is the acceleration due to gravity and !ih is the measured height difference of pressure head and outlet. The only pressure drop in the system was assumed to be across the streaming capillary because the cross sectional area of that region was 65% smaller than the next smallest cross section in the system The pressures were varied from 0.33 to 3.56 cm Hg The resulting streaming potentials generated from the electrolyte flow were recorded. The streaming potential measurements were recorded at the unadjusted pH of the eluent (pH:::::5.7) in COM saturated solution at a temperature of 37 C. To determine the effect of surface conduction a conductance meter (YSI Model 32 YSI Incorporated Yell ow Springs OH) was connected across the streaming capillary. The bulk solution conductivity was measured using the same conductance meter and a conductivity probe (YSI 8417 YSI Incorporated Yellow Springs OH) having a cell constant of 1/cm. From experimentally determined values of !1P and V s and material and liquid parameters the zeta potential of the substrate was calculated using equation 2 .6. Constants used in the calculation are shown in Table 5.1. Streaming potential was also performed on the same fused quartz substrates using a commercial apparatus (PAAR EK.A-Electro Kinetic Analyzer RV 4.0 Brookhaven Instruments Corporation Holtsville NY) yielding similar results.

PAGE 138

Table 5 .1 A list of the physical parameters necessary to calculate zeta potential from streaming potential. Physical Parameters Temperature (0C) Viscosity (Poise) [Lid90] Density (g/cm3 ) [Lid90] Dielectric Constant [Lid90] Specific Conductivity (1/Q cm) Values 37 0.006917 0.993075 73.2 2.48x106 121

PAGE 139

122 5.3 Results and Discussion 5.3.1 Dynamic Wet Cell The dynamic wet cell was machined out of two pieces of poly (methyl methacrylate) (PMMA) and is sealed with a rubber o-ring between the top and bottom of the cell. The entire dynamic wet cell was assembled as if an experiment was being performed and the volumetric flow rate as a function of dial number on the metered gear pump was calibrated. Figure 5.9 is a plot of dial setting on the metering pump vs volumetric flow rate. The error bars represent the standard deviation for 5 individual experiments. A line was fit to the data to simplify the calculation of force and stress a t a given flow rate or pump dial setting. The equation of the line is y = 75 99 x 4 35 and the least squared linear regression coefficient r2, is 0.9983 The fact that the intercept of the line is negative leads to a small underestimation in the applied h y drodynamic force of a few Pascals. The dynamic wet cell was fashioned to promote laminar flow across most of the length of the cell. A reservoir was placed before and after the actual adhesion section of the cell to allow flow to enter the adhesion area with as little perturbation as possible. Reynolds numbers Re, were calculated for the range of flow rate s and are shown in Figure 5 .10 using the equation [ van92a] R =p Q (w+2h) ( 5.4 ) where Q is the volumetric flow rate p is the fluid density is the viscosity w is the width of the flow chamber and h is the half the distance between the parallel plates. As can be seen in Figure 5.10 the Reynolds numbers climb as high as 200 for the highest flow

PAGE 140

200 150 '2 E .._ (I) 100 ro a: 0 LL 50 0 Dial Setting Figure 5.9. A plot of dial setting vs. volumetric flow rate used to calibrate the pump. The error bars represent the standard deviation of five individual experiments. 123

PAGE 141

124 300 250 Q) a: .... -Q) 200 .0 E ::::, z 150 (/) "Cl 0 C: 100 >. Q) a: 50 Flow Rate (ml / m i n ) Figure 5 .10 A plot of flow rate versus Reyno l ds number indicating the stable flow inside the dynamic wet cell at all experimenta l flow rates

PAGE 142

125 rates achieved 150 ml / min Since the critical Reynolds number for laminar flow is approximately 2 300 every experimental flow rate produced laminar flow [Esk68]. Further calculations were performed to determine the distance into the cell for the flow to become laminar Le, at increasing flow rates using the equation [ van92] L = 0.026hR (5.5) At the highest experimental flow rates 1 50 m l /min, the calcu l ations showed that a distance of 2.1 mm was necessary for the flow to become laminar. No particles within 4 mm of the inlet side of the flow chamber were measured 5.3.2 Model COM Particles COM particles with an average equivalent spherical diameter of 13.0 m were produced. The particle size standard deviation was re l atively narrow at 1.3 m. To simplify the calculations of adhesion strength the particle size was assumed to be monodisperse The experimental zeta potentials shown in Figure 5.11 demonstrate that the surface charge of the COM crystals is negative over the pH range investigated. This finding is not consistent with the results of Curreri Onoda and Finlayson [Cur79b] who showed the surface charge to be positive across most of the pH range at saturated conditions. The differences in surface charge may be due to the differences in crystal morphology The particle morphology on which Curreri et al. [Cur79b] performed zeta potential measurements was an aggregate of poorly defined particles while the particle morphology used in the current study is well defined Boeve et al. [Boe94] also reported similar values of COM zeta potential to those determined here. The zeta potential of COM determined by Boeve et al. [Boe94] was -5 m Vat a pH of 6.1 and an ionic strength of 0.33 M The differences in magnitude of the zeta potentials between

PAGE 143

126 0 10 ,,..--._ > 20 E '-../ co t .;:::::; C -3 0 Q) \,_' _,,..-+\ l 0 CL co a5 4 0 ,r T N 50 "'f-! -60 3 4 5 6 7 8 9 1 0 pH Figure 5.11. Experimentally determined zeta potential of COM particles in saturated calcium oxa l ate monohydrate sol ution. Each data point is the mean+/ -95% confidence interval. The ionic strength of the saturated COM solution was 8x 106 M.

PAGE 144

127 those currently measured and the potentials which Boeve et al. measured may be attributed to the differences in ionic strength In the current investigation the ionic strength of the COM saturated solution at 37 C is approximately 8xl06 M determined using atomic absorption to measure the [Ca2+ ]. 5.3. 3 Streaming Potential Just as flow dynamics were important in the development of the dynamic wet cell they are also important for the measurement of streaming potential. Streaming potential theory utilizes the Poiseuille equations to describe laminar flow in a capillary. To determine if the flow in the currently developed streaming potential cell was laminar Reynolds numbers were calculated for each driving pressure using the equations [Esk68] h1 M u =--ave 3 Lix (5.6) and R = phuaveM e (5.7) where U ave is the space average velocity Figure 5.12 is a plot of the driving pressure vs Reynolds number which shows that the flow is laminar even at the highest driving pressures, which is indicated by the Reynolds number being less than half of the critical Reynolds number of 2300. A plot of M vs streaming potential was used to determine the experimental zeta potential of the fused quartz substrate as shown in Figure 5 .13. A line fit to the data in Figure 5.13 has a least squared regression coefficient of 0 .98 and an intercept of -3.88 mV. The good fit of the line is important because streaming potential is directly proportional

PAGE 145

128 800 Q) a: .... -Q) 600 .0 E ::, z (/) "O 400 0 C >, Q) a: 200 Pressure Drop (cm Hg) Figure 5.12. A plot of pressure drop or driving pressure across the streaming capillary indicating that the flow is laminar under all experimental flow conditions.

PAGE 146

129 > -300 5 ci1 c Q) 0 -200 a.. O') C E Ct1 Q) .... tn -100 Driving Pressure (cm Hg) Figure 5.13. A plot of streaming potential vs. driving pressure for fused quartz in COM saturated solution showing a linear regression with an r2=0.98 and 95% confidence limits.

PAGE 147

130 to driving pressure (i.e the pressure across the streaming potential capillary) as can be seen from equation 2.6. The fact that the intercept is low in magnitude is an indication of how well the background potentials are being eliminated using the circuitry of Horn and Onoda [Hor77 Bal 73]. The zeta potential of the fused quartz substrate was measured to be -2 2 mV (mean% confidence interval) at 37C and pH 5 7 using the previously described streaming potential instrument. The value of zeta potential would appear to be low when compared with other previously published values ; however the Brookhaven Instruments commercial streaming potential analyzer measured a zeta potential of -16 .8.3 mV (meanstandard deviation) at 37 C and pH 6.8 on the same fused quartz material. Ranges of zeta potentials have been reported by various authors such as -120 to -150 m V [Jon45] and -90 m V [Sca92] for similar ionic strength solution conditions on vitreous silica. Although the electrolyte concentration is similar to the experimentally used value other authors used electrolytes which were 1: 1 in va l ence of cation:valence to anion. Interestingly, many of the reports of zeta potential are determined at a single pressure. From the reported data one cannot determine if the experimental streaming potential is directly proportional with driving pressure or if the intercept is zero. Both of which are a must for accurate calculation of zeta potential according to Ball and Fuerstenau [Bal 73] Fairbrother and Mastin [Fai24] showed the zeta potential -48 mV to drop when using a 1 x 10-5 M 2: 1 electrolyte. In the current experiments the electrolyte CaC204 is 2:2 in which both the co-ion and the counterion have a valence of 2 In coagulation theory the electrolyte generally becomes a more effective coagulating agent when it contains multivalent ions Also the coagulation concentration is most importantly determined by the valency of the counterion [Hun93]. In the current case the counterion is Ca2+ That is as the valency of the electrolyte increases the solution cannot support the charge and a decay of charge into the bulk solution occurs more rapidly which is demonstrated in Figure 2.5

PAGE 148

131 As mentioned in Chapter 2.4.1.2 the effect of surface conduction at low ion strength conditions may underestimate the value of zeta potential. The method of Briggs [Bri28] was used to determine if surface conduction was effecting the measurement. The excess of the conductance (1/Q) measured in the capillary (i e., with the substrate in place ) over the conductance of the bulk (i. e., just the electrolyte solution) is called the surface conductance [W 0046]. The experimentally measured conductance of the capillary or streaming cell was never larger than the conductance of the bulk electrolyte indicating that surface conduction is not effecting the experimental measurement of streaming potential. 5.3.4 Adhesion The negative surface charge of the COM crystal is related to the atomic structure of the (010) and (101) crystallographic planes These two planes make up much of the surface of the COM crystal. Both the (010) and (101) crystallographic planes are associated with more C20/" than Ca2+ (Figure 5.4) giving the particles a net negative theoretical surface charge which corresponds to the experimentally determined zeta potential values However the (101) face should be more positive than the (010) face because of the larger Ca2+/C20/" ratio for the fo1mer crystallographic face The adhesive strength of each of these two surfaces to the negatively charged fused quartz surface are not equivalent. The value of adhesion strength is taken as the value of applied stress at 50% of the initial particles adhering. This method is used to be consistent with previously published adhesion studies on other materials. The strength acting on the (010) surface at 50% adhering is 81 Pa while the strength acting on the (101) surface at 50% adhering is 170 Pa, as shown in Figure 5 .14. Both atomic surfaces predominately contain oxalate ions ; however the (101) surface has a higher Ca2+;c20 / ratio per unit area than does the (010) crystal plane making the (101) more positive than the (010). Electro statically a greater repulsive force should exist between the (010) and fused quartz than between the (101) and fused quartz. The greater repulsive force between the (010)

PAGE 149

132 100 0 ( 010), n=255 a (101 ) n=380 80 8 ..60 x 0 z -.. 40 z 0 200 4 00 600 800 1000 Applied Stress (Pa) Figure 5.14 The probability of an ini t ia l particle on the (010) 0 and on the (101), D adhering versus the applied stress acting on the particle at fai l ure as determined using the dynamic wet cell. The total number of particles counted that were l y ing on the (010) was n = 255 and on the ( 101) was n = 380.

PAGE 150

133 and the fused quartz will enable the particles lying on the (010) surface to be removed at a lower hydrodynamic shear force than those lying on the (101), which is confirmed experimentally Due to the relative crystallographic dimensions of COM particles having the ( 101) in contact with the substrate will have a greater displacement force applied at a given flow rate than particles with the (010) in contact with the substrate. Although contact areas of the two crystallographic planes are approximately equal particles with the ( 101) in contact with the substrate have a higher probability of adhering at a particular flow rate than those in the (010) orientation This verifies that the (101) planes have a specific interaction probability mediated by the greater incidence of Ca2+ in this crystallographic habit interacting with negatively charged sites on the SiO2 [Ile79]. The curve of the percent of adhered particles vs. the applied stress shown in Figure 5.14 has an interesting shape. Many natural processes exhibit a normal distribution which would give a sigmoidal cumulative normal distribution [Ebd77]. Since the minimum stress that can be measured using the dynamic wet cell and COM particles is approximately 10 Pa the cumulative normal distribution is truncated at the lower stresses and the sigmoidal shape may not be apparent. For materials having a higher adhesive strength the cumulative distribution of percent adhering vs. applied stress appear to be sigmoidal in nature 5.3.5 Theoretical Modeling oflnteractions The adhesion of the two negatively charged surfaces cannot be entirely attributed to electrostatic interactions. Adhesion between glass and COM particles is also due at least in part to van der Waals interactions and the force due to gravity. The van der Waals interactions between macroscopic bodies may be described using the Hamaker constant as described by the Tabor-Winterton relationship if both the index of refraction and the dielectric constant of the interacting materials are known [Tab80]. The van der Waals forces were estimated using the Hamaker constants for COM and fused quartz

PAGE 151

134 The A131 value for COM in water is l.37x10-20 J [Chapter 4] while the A131 value for fused silica in water is 0.849x10-2 0 J [Hou80]. The Hamaker value for COM-H2O-Glass was determined using the geometric mean [Isr92] (5.8) resu l ting in a value of A132 for the COM-H2O fused quartz system of l.08x10 -20 J. The van der Waals interaction energy is given by [Isr92] V=-~ A }2mt2 (5. 9) where V A is the energy due to van der Waals interactions A 132 is the Hamaker constant given above and dis the separation distance between the interacting materials which is taken as the surface roughness of the two materials .. The average separation distance was estimated from the surface roughness of COM particles and the fused quartz substrates measured using contact mode SPM in a COM saturated solution environment. The ra value or mean roughness value of the fused quartz substrate is 0.87 nm over a 5x5 m2 area, Figure 5.15. The r a values for the COM (010) Figure 3.14 and (101) Figure 3 .15, are 32 nm and 81 nm respectively over a scan area of 5x5 m2 The high surface roughness of the COM crystal may be due to the growth mechanism of the twinned COM crystal. The crystals may have a growth ledge at the twin boundary where the atoms grow more rapidly on one half of the twin boundary but not on the other causing the roughness to be l arge across this boundary However if a section analysis is performed on the SPM images to determine roughness across the particle surface in an area without a twin boundary values for roughness between 16 nm and 21 nm are measured.

PAGE 152

135 5,00 IMage Statistics IMg, Z range 14,770 nM IMg, Hean -0,0004 nM 2.50 IMg, RMS CRq) 1.154 nM IMg, Ra 0.871 nM IMg, Srf. area IMg, Srf. area diff 0 0 2.50 5, 00 JJM Figure 5.15 A scanning probe microscopy image of the fused quartz surface under a COM saturated solution liquid environment. Also given is the image roughness statistics which include the r a or mean surface roughness of 0.871 nm.

PAGE 153

136 To model the interactions due to the electrical double layers of the COM particle and fused quartz substrate a simplification of Derjaguin-Landau-Verwey-Overbeek (DL VO ) [Der4 l Ver48] theory was used The simplified theory presented by Hogg, Healy and Fuerstenau (HHF) [Hog66] uses the Debye-Hilckel approximation for low surface potential and does not account for dipole and specific chemical interactions. The Debye Hilckel reciprocal length parameter K:, for small surface potentials is given by [Isr92] 2 2 "" ce z "/( = L.. -'--' ; ek.T (5.10) where z is the v alence of each ionic species i in solution c is the concentration is the dielectric constant of the suspending medium e is the charge on a electron k is Boltzmann s constant and Tis the absolute temperature. Hogg et al. [Hog66] give the potential energy VR, between interacting flat double layers to be equal to the change in free energy of the double layer system when the plates are brought together from infinity. Thus ( 5 11) where G2d is the free energy of the double layer system when the plates are se parated b y a distance 2 d and Goo is the free energy of the double layer system when the plates are separated by a distance of infinity The components of the total energy due to the interacting double layers are given as [Hog66] (5.12) and

PAGE 154

G __ ( ( 2 + 2 ) 00 -Sn l/101 l/10 1 where l{/01 and l{/0 2 can be taken as the va l ues of zeta potential for COM and fused quartz, respectively. Substituting for the free energy terms in eq. ( 5 .11 ) 137 (5.13) (5.14) Equation (5.14) describes the potential energy of two interacting infinite parallel flat double layers as a function of the surface potential of each plate and the separation distance between the two plates The total interaction between COM and the substrate is given by ( 5 15) where V r is the total interaction potential V A is the potential due to van der Waals forces and VR is the potential due to the interacting electrical double layers The van der Waals potential VA, for the (010) crystallographic face of COM interacting with fused silica across water at a separation of 32 nm is thus -30x 10-8 J / m2 calculated using eq. (5.11 ). The potential due to the interaction of the electrical double layers VR, at the same distance of separation 32 nm is 1058x10-10 J / m2 calculated using eq. (5.14) Using eq. (5.15) the total interaction potential was calculated to be -1919xl 0-10 J / m2 which corresponds to an adhesive strength or adhesive pressure of 29 Pa The theoretical adhesive strengths were determined by a simple differentiation of eq. (5.9) and eq. (5. 14) with respect to separation distance. Similar calculations for the (101) face of COM interacting with fused silica across water were performed using the sw-face roughness as the separation distance and yielded a V A =-426x10-10 J / m2 VR=6x10-1 0 J / m2

PAGE 155

138 and Vr=-419x10-10 J/m2 The theoretical total adhesive pressure for (101) COM interacting with fused silica is 1 Pa. The theoretical values of 29 Pa and 1 Pa are much lower than the experimentally measured adhesion values of 81 Pa and 170 Pa for the interaction of fused quartz with the (010) and ( 101) COM faces respectively. If the separation distance used in the theoretical calculation is 20 nm, the value of adhesion strength is 98 Pa, which compares favorably with experimental results Krupp [Kru67] determined that the adhesive strength or adhesive pressure for various interacting materials is between (2-30)xl07 Pa for a separation of 0.4 nm. Israelachvili [Isr92] has shown comparable results for similar separation distances However for separation distances on the order of 20 nm the adhesive stress values are considerably lower [Isr92] and comparable to the values of adhesive strength determined herein. Gravitational forces were also calculated to determine their contribution to COM crystal adhesion using the average COM particle size and a particle density of, 2.2 g / cm3 resulting in a strength equal to 1.2 Pa. This represents the force due to gravity of l. lxl0-10 N acting on one particle of mean particle size divided by the average area -111 m2 of the (010) and ( 101) particle faces. Thus gravitational effects are insignificant when compared to the effects due to van der Waals and electrostatic interactions 5.4 Conclusions A device was designed and built and the methodology devoloped to measure the adhesion strength between a particle and substrate A second device was designed and built and the methodology developed to measure the zeta potential on a flat substrate. The parallel plate adhesion technique is a suitable method for determining the adhesive strength between a particle and a substrate in a dynamic liquid environment. The van der Waals interactions were estimated and related to the adhesion between COM crystals and a fused quartz substrate. Correlations were made between the theoretical atomic surface structure of COM and the crystallographic dependence on adhesive strength. Theoretical

PAGE 156

139 electrostatic considerations were invoked and are used to explain the crystallographic dependence. The application of this adhesion measuring technique to the study of kidney stone disease seems to be promising in that the importance of specific biological materials can be ranked with respect to their adhesive properties to COM crystals. Which in turn may allow researchers to focus on materials that have the greatest adhesion to COM.

PAGE 157

CHAPTER 6 MEASUREMENT OF COM ADHESION TO MACROMOLECULAR SUBSTRATES 6.1 Introduction Kidney stone disease has been described as an opportunistic disease which can rely on many causative mechanisms acting in concert [Fin78a Fin84]. One of the possible mechanisms is the growth of an adherent particle, called a fixed particle in the lumen of a nephron. Such an adherent particle can subsequently grow by secondary nucleation and growth and/or aggregation of crystallites to a size large enough to occlude a renal tubule which is the basis of the fixed stone mechanism of kidney stone development. Finlayson and Reid [Fin78b] calculated the time for a COM crystal to grow to a large enough size to cause an occlusion and determined that the crystal would not have enough time to grow to such a size. Therefore crystal attachment is a necessary process in kidney stone formation [Man 91, Man94]. Riese Mandel and coworkers [Man94 Rie92 Rie88] have shown in vitro attachment of COM crystals to inner medullary collecting duct epithelial cells of the rat animal model in the static case (i.e., without the pres e nce of flow) They also proposed that perturbations in the cell membrane structure with a loss in membrane polarity can enhance crystal attachment [Rie92]. In vivo crystal attachment to epithelial cells in the rat animal has been demonstrated by Khan et al. [Kha82]. Renal injury has also been implicated in fixed stone formation [Man91 Gil79 Kha84]. Khan [Kha82 Kha95b] has discussed that after attachment of a crystal to the epithelial cell surface that injury to the cell may occur but is not necessary to occur. 140

PAGE 158

141 A few different results may ensue due to crystal attachment to the epithelial cell. The cell may die and be washed away in the urine flow taking with it the crystal. The cell may envelope the crystal and continue to function normally [Lie92 Lie93 Lie94] with no further growth of the crystal. However, if epithelial cell injury has occurred basement membrane proteins may be accessible to adhere to COM crystals by sloughing of the cells leaving behind the exposed underlying architecture the basement membrane. The basement membrane is a layer of proteins which underlies the epithelial sheet and will be discussed subsequently in more detail. Depending on the extent of injury other proteins common to the extracellular matrix found below the basement membrane may also be accessible for attachment to COM crystals. Khan et al. [Kha84] demonstrated the association of crystals to fibrillar macromolecular structures by scanning electron microscopy (SEM) Khan has also shown crystals passing through the basement membrane into the extracellular space near the papillary tip [Kha95b ]. Researchers have shown the attachment of COM crystals to epithelial cells and fibrous proteins characteristic of those found in the basement membrane and extracellular matrix, yet the importance of these individual materials in kidney stone formation with respect to COM crystal adhesion is not known. The necessity to determine the important materials most responsible for COM crystal adhesion will allow researchers to focus their interests to those more important materials in the search for a possible cause and therefore relief from kidney stone formation. In the current study, both structural extracellular matrix (ECM) proteins collagen type I and collagen type IV and adhesive extracellular matrix glycoproteins fibronectin and laminin were examined. A positively charged, synthetic polymer PEI was also investigated. The collagen types of molecules are the most abundant of the ECM proteins Type I is fibrillar in nature and can be found in a number of physiological sites These and

PAGE 159

142 other types of collagen molecules may assemble into collagen fibrils which create mechanical stability for the skin tendon bone and internal organs [Ktih87a]. Collagen type IV is a nonfibril forming molecule which creates mechanical stability for the basement membrane and overlying cell architecture [Ktih87b]. Laminin and fibronectin are two glycoproteins found in the basement membrane underlying the epithelial sheet. These two proteins aid in the mediation of cell attachment to the extracellular matrix [Alb89]. The collagen molecules fibronectin and laminin are not normally accessible to COM crystals in the urine unless injury to the renal epithelial wall has occurred. In the current study, the adhesion of COM particles to different macromolecules was measured using a variation of the technique reported by Pelton et al. [Pel82 Pel84] to measure the adhesion between polystyrene spheres and a Pyrex substrate. A similar device was developed by Hochmuth et al. [Hoc73] and has been used extensively to study the adhesion of cellular material to surfaces [Dor77 For84 Moh74 Tru90 van92]. Details pertaining to the specific adhesion apparatus are described in Chapter 5 and will only be briefly be described herein. The apparatus consists of an adhesion chamber which incorporates two fused quartz slides separated by a teflon spacer. The eluent flows through the adhesion cell at a specified rate calibrated against the pump drive The eluent is forced through the adhesion chamber using a variable speed pump drive connected to a gear pump A camera mounted above the adhesion cell, which is connected to a monitor and video cassette recorder, records activity in the cell during an experiment. The method utilizes the volumetric flow rate adhesion cell and particle dimensions in Poiseuille's equation to calculate the linear velocity at one particle radius from the cell wall such that 2Q(R2 r2 ) V = ---'----~ nR4 ( 6.1)

PAGE 160

143 where v is the linear velocity and Q is the volumetric flow rate. Half the distance between the fused quartz slides is R and the particle radius is r The corresponding displacement force acting at one particle radius F d is calculated using the fluid properties particle dimensions and the calculated linear velocity from equation ( 6.1) such that [Gol67] = 1.7( 61r71rv) (6.2) where 7J is the eluent viscosity r the particle radius and v the linear velocity. 6.2 Materials and Methods 6.2.1 Particle Synthesis and Characterization COM particles were produced and characterized as described in Chapter 3. 6 2 2 Substrate Coating Optical grade fused silica slides (spectroscopic grade ESCO Products Oak Ridge NJ) were coated with a variety of macromolecules: polyethyleneimine (PEI) (Eastman Kodak Co., Rochester NY) collagen type I (C7661 Sigma Chemical Co. St. Louis MO) fibronectin (Fl 141, Sigma Chemical Co. St. Louis MO) and MATRIGEL Basement Membrane Matrix ( 40234C Collaborative Biomedical Products Bedford MA). MATRIGEL Basement Membrane Matrix consists of laminin collagen type IV, heparan sulfate proteoglycans entactin, and nidogen listed in descending order of quantity [Kle82] Each slide was coated with the desired macromolecule to be used as the substrate on which the adhesion strength to COM would be determined. The procedures employed for coating the glass with all of the macromolecules with the exception of PEI are available from the protein suppliers.

PAGE 161

144 The collagen I used was from rat tail and was in solid form from the vendor. A stock solution of 1 mg of solid collagen type I was d issolved in 1 ml of 0.1 M acetic acid (HC2H302). The stock solution was further diluted to make a working solution to coat at a coverage of 10 g /cm2 The dilution occurred by adding 645 16 l of the stock solution to 9.355 ml of 0 1 M acetic acid Each 2.54 cm x 2.54 cm fused quartz slide was covered by 1 ml of the collagen type I working solution. The slide was stored 24 hr at 4 C with the protein solution on it. After the 24 hr the remaining solution was aspirated or removed by suction. The slides then dried for 24 hr at 4 C before being used in adhesion experiments. The fibronectin was 0 1 % solution (1 mg protein/ml in 0 5 M NaCl, 0.05 M Tris buffer at pH 7.5) from bovine plasma. A working solution was prepared from the above solution by adding 32 08 l of the fibronectin stock solution to 967.9 l of Hank' s Balanced Salt Solution (14175 095 Gibco BRL, Gaithersburg MD) Each 2 54 cm x 2.54 cm fused quartz slide was covered by 2 ml of the fibronectin working solution to make the protein coverage on each slide 10 g /cm2 The slides sat for 1 hr under the protein solution the remaining solution was aspirated and the slides were stored at 4 C until used. The material MATRJGEL Basement Membrane Matrix is a solubilized basement membrane extracted from the Engelbreth-Holrn-Swarm mouse sarcoma or tumor. The protein mixture arrives as a stock solution at a concentration of 13.8 mg protein/ml. The protein mixture must remain frozen until use The MATRJGEL Basement Membrane Matrix was thawed at 4 C overnight. All the slides and pipettes were precooled because the MATRJGEL Basement Membrane Matrix will ge l rapidly and irreversib l y at 22 C to 35 C. A 1:10 dilution of the stock protein was prepared using serum-free medium (11039-021 Dulbecco Modified Eagle Medium: Nutrient Mixture F12, Gibco BRL, Gaithersburg MD) to make a working solution having a concentration of 1.35 mg/ ml.

PAGE 162

145 The fused quartz slides were coated with 1 m l of the working solution to give a coverage of200 g/ cm2 To coat the glass with PEI a 2 weight percent suspension of PEI was prepared The glass surfaces were prepared for coating by soaking in acetone and isopropanol for 3 min. respectively The slides were then liberally washed with deionized water. The slides were soaked and agitated in a 2 w / o suspension of PEI for 2 min and turned over and soaked for an additional 2 min. Finally the s l ides were again liberally washed with deionized water and allowed to dry To verify that the proteins were being retained on the fused quartz surface attenuated total reflection (A TR) fourier transform infrared spectroscopy (F TIR ) (20SXB FT -IR Spectrometer Nicolet Instrument Co. Madison, WI) was performed on select substrates coated using the previously described procedures A 45 thallium bromide-thallium iodide (KRS -5) crysta l having an index ofrefraction of 2.35 was used to couple the infrared radiation to achieve internal reflection. 6 2 3 Adhesion Measurements Adhesion measurements were performed on all the macromolecular substrates (i.e. collagen type I fibronectin MA TRIG EL Basement Membrane Matrix and polyeth y leneimine) as described in Chapter 5.2.1. The measurements were performed in a low ionic strength en v ironment COM saturated solution and a hig h ionic strength environment artificial urine ion solution (AUIS). The ionic strengths of COM saturated solution a nd AUIS are 5xl o-6 Mand 0.323 M respectively. The pH of COM saturated solution is 5.7 while that of AUIS is 6.0. Analysis of the experimental data was performed in the same manner as the data in Chapter 5 The data was plotted as applied stress vs (N(F)/N0)xlO0 where N0 is the original number of particles adhering at no applied stress and N(F ) is the number of particles adhering at each applied str e ss. A curve was fit to the e x perimental data using SigmaPlot Software ( Version 5.0 Jandel

PAGE 163

146 Corporation San Rafael CA) The nonlinear regression was of the form y = a exp (-bx), where a and b were variables to be solved for. After determining the values of a and b the value of applied stress was calculated at y=50 and was used for the mean applied stress value. 6 2.4 Streaming Potential Measurements Streaming potential measurements were performed on all the macromolecule coated substrates (i.e. collagen type I fibronectin MATRIGEL Basement Membrane Matrix and polyethyleneimine) using the instrument and the procedure described in Chapter 5.2.2. The measurements were performed in two different electrolyte solutions COM saturated solution and AUIS 6.3 Results and Discussion 6 3 1 Substrate Coverage Attenuated total reflection FTIR showed that after the fused quart z s lides were coated with the macromolecules that the macromolecules were retained Fig ure 6.1 is a plot of wavenumber vs percent absorbance for all of the substrates. The peaks below 1200 cm-1 are from the absorption of the fused quartz substrate (i e., Si-O-Si vibrations). At lower wavenumbers (higher wavelengths) sampling depth increases as seen in Figure 2.18 Since the majority of material being sampled is fused quartz the majority o f the absorbance will be due to the fused quartz Figures 6 2 is the same data shown in Figure 6.1; however Figure 6.2 shows a much smaller range in wavenumber. Figure 6 2 focuses on higher wa v enumbers 21003600 cm-1 In this range the fused quartz gives no distinct absorbance peaks. Howev e r the macromolecular substrates do exhibit absorbance peaks between 2800-3000 cm -1 du e to C-H stretch or N-H stretch. Because the macromolecular substrates exhibit absorbance

PAGE 164

4000 PEI on Fused Quartz MATRIGEL on Fused Quartz F i bronectin on Fused Quartz Collagen Type I on Fused Quartz Fused Quartz 3000 2000 Wavenumbers (cm 1 ) Absorbance (%) 1000 F i gure 6.1 A p l ot o f wavenum b er vs a b sor b ance for fuse d q u artz collagen type I fib ronectin MA TRI GEL and P EI prod u ce d u sing AT R -FTIR. All of the macro m o l ecu l ar su b stra t es were coate d on fuse d q u artz 147

PAGE 165

Fused Quartz 3400 C-H stretch or N-H stretch 3000 2600 Wavenumbers (cm1 ) 2200 Figure 6.2. A plot of wavenumber vs. absorbance for fused quartz collagen type I fibronectin MATRIGEL and PEI produced using ATR-FTIR for the wavenumbers 2100-3600 cm-1 All of the macromolecular substrates were coated on fused quartz 148

PAGE 166

149 peaks not found in the FTIR spectra of fused quartz alone the conclusion that the coating procedure is applicable and that the macromolecules are retained on the fused quartz even after five days at 4 C is made. 6 3.2 Zeta Potential Determinations The experimental zeta potential of COM particles at pH 6 in saturated COM solution and 10% AUIS at 37 C is -32 mV and -21 mV respectively Figure 6.3. Under the solution conditions of saturated COM at a pH of 5 7, the zeta potentials of the different macromolecular substrates varied over a range of values as shown in Table 6 .1. The values of zeta potential shown in Table 6.1 are the mean and 95% confidence interval for at least three separate experiments on the same substrate. Collagen type I fibronectin, and MA TRIG EL had negative zeta potential values. Figure 6.4 shows driving pressure vs. streaming potential for collagen type I, fibronectin MA TRIG EL and PEI. The scatter in the data shown in Figure 6.4 for PEI may be due to the fact that the streaming potential for this substrate was measured over a period of 3 days The equilibrium surface structure may age with time depending on solution conditions subsequently effecting the zeta potential. Zeta potential determination of the macromolecular substrates in AUIS was attempted using the streaming potential technique. The values of streaming potential in the high ionic strength complex ion solution were on the order of 1-2 m V at the highest driving pressures and undetectable at lower driving pressures. The asymmetry potentials were also about 1-2 m V Because the potentials we were trying to measure (i. e., streaming) and the potentials we were trying to negate (i.e. asymmetry) were approximately the same valid determinations of the actual streaming potentials could not be made even using the Hom and Onoda [Hor77] circuit.

PAGE 167

150 0 COM Saturated So luti o n 10 .. 10 % Art ificial U r ine ,......_ > 20 s A cu \, T ~ 30 Q) 0 ,~'\ ()_ cu -4 0 Q) r,r N 50 t,f_, -60 10 p H Figure 6.3. Zeta potential as a function of pH for COM in saturated COM solution and zeta potential at a pH of 6 for COM in 10% AUIS

PAGE 168

151 Table 6 1 The values of zeta potentia l of the macromolecular substrates determined using streaming potential measurements in saturated COM solution Macromolecular Substrate Collagen type I Fibronectin MATRIGEL Polyethy leneimine Zeta Potential (m V) Mean Standard De v iation -9 6 0 3 -12.8 2.5 -10.9 1.4 7 8 0 4

PAGE 169

152 300.------.----,---r-----r---.--~-T"""""---. 200 0 > .s 0 c 100 Q) 0 0.. C'l 0 C E eel Q) .... U) 100 200 ..._ __ __._ ___ ......_ ___ .....__ __ ___. 0 2 3 4 D riving Pressure (cm H g ) Figure 6.4. A plot of driving pressure vs streaming potential for the substrates collagen type I(~), fibronectin (V) MA TRI GEL ( D ) and PEI ( 0).

PAGE 170

153 6 3.3 Adhesion Measurements The measurement of adhesion of COM particles to the fused quartz and macromolecular substrates generated the curves labeled Figure 6.5 to Figure 6 .11. The plots of adhesion of COM particles to the MATRIGEL and PEI substrates in AUIS are not included because no decrease in the number of particles as the applied stress increased was observed In fact an increase in the number of particles adhering was observed throughout the experiments indicating an adhesive strength well above that measurable with the current instrument between particle and substrate in AUIS. The adhesion between glass and COM crystals was demonstrated in Chapter 5 and a COM crystallographic dependence on adhesion to glass was observed. In much of the results involving macromolecular substrates and COM saturated solution the same correlation between COM crystallographic habit and adhesive stress was observed as shown in Figure 6.12. In every case a COM particle lying on the ( 101) will have a higher adhesive stress than will a COM particle lying on the (010) when in contact with a negatively charged substrate with the exception of the fibronectin substrate. The fibronectin substrate had a slightly lower adhesive strength to the ( 101) than to the (010) -35 Pa Such trends can be explained in terms of electrostatic interactions The negatively charged proteins exhibit a greater repulsive force against the more highly negative 010 planes than the less negative 101 planes of the COM crystal resulting in a lower adhesive stress to the COM (010) versus the COM ( 101) due to electrostatic interactions. The COM crystals obtained the highest adhesive stress to the positivel y charged macromolecule PEI. The opposite crystallographic orientation effect of COM was observed on the adhesive stress to PEI. The more negatively charged (010) crystal plane had a larger adhesive strength to the positively charged PEI than did the less

PAGE 171

154 100 80 0 0 0 60 .,X 0 z -40 z 20 1000 Applied Stress (Pa) Figure 6.5. The probability of an initial COM particle on the (010) 0 and the (101 ), D adhering to fused quartz in AUIS versus the applied stress acting on a particle at failure as determined using the dynamic wet cell. The total number of particles counted lying on the (010) was n=l 13 and on the ( 101) was n=84

PAGE 172

155 1oow.-----,---"T"'"""----.----.----.------80 0 0 60 ,X 0 z -4 0 z 20 Do Do 0 0 500 1000 1 500 Appl i ed St r ess ( Pa) Figure 6.6. The probability of an initia l COM particle on the (010) 0 and the (101) D adhering to collagen type I in COM saturated so l ution versus the applied stress acting on a particle at fai l ure as determined using the dynamic wet cell. The total number of particles counted lying on the (010) was n=82 and on the (101) was n = l33.

PAGE 173

8 T""" X 0 z -LL z 1000 2000 3000 4000 5000 6000 Applied Stress (Pa) 156 Figure 6.7. The probability of an initial COM particle on the (010) 0 and the (101) D adhering to collagen type I in AUIS versus the applied stress acting on a particle at failure as determined using the dynamic wet ce ll. The total number of particles counted lying on the (010) was n=84 and on the (101) was n=72.

PAGE 174

100 a------..--~--.---"""T""--~-80 8 60 ,-x z o ...... 40 z 20 500 1000 1500 Applied Stress (Pa) 157 Figure 6.8. The probability ofan initial COM particle on the (010) 0 and the (101) D adhering to fibronectin in COM saturated solution versus the applied stress acting on a particle at failure as determined using the dynamic wet cell. The total number of particles counted lying on the (010) was n=87 and on the (101) was n = l04.

PAGE 175

0 0 T""" X 80 60 40 20 158 1000 2000 3000 4000 Applied Stress (Pa) Figure 6.9. The probability of an initial COM particle on the (010) 0 and the (101) D adhering to fibronectin in AUIS versus the applied stress acting on a particle at failure as determined using the dynamic wet cell. The total number of particles counted lying on the (010) was n = 95 and on the ( 101) was n=143.

PAGE 176

159 0 60 0 .,.... X 0 z -4 0 z 20 0 Applied St r ess ( P a) Figure 6 10. The probabi l ity of an initial COM particle on the (010) 0 and the (101), D adhering to MATRIGEL in COM saturated solution versus the applied str e ss actin g on a particle at failure as determ i ned u sing the dynamic wet c~ll. The total number of particles counted lying on the (010) was n = 57 and on the (101) was n=54

PAGE 177

160 0 60 0 .... D X z o 0 --D 40 z 20 Applied Stress (Pa) Figure 6.11. The probability of an initial COM particle on the (010) 0 and the ( l 01 ) D adhering to PEI in COM saturated solution versus the applied stress acting on a particle at failure as determined using the dynamic ~et cell. The tota l number of particles counted lying on the (010) was n=61 and on the ( 101) was n = 69

PAGE 178

161 1500 D (010) 8 --ro D (101) 0.. .. ......... ..c 1000 ..LJ O'> C t ... (/) ~ -10 ... CJ) 500 ... ..c ... - C :.::::; w C ro <..> CJ ~ ::, O'> 0 ro C a: 0 2 IC 0 ,.Q -<{ CJ) LL 2 :>-.. ::, ..c LL ..LJ :>-.. 0 0.. Figure 6.12 A bar chart summarizing the a dh esion data experimentally determined in COM saturated solution ( i e., low ionic strength) The numbers above each set o f bar s i s the value of the zeta potential for each substrate

PAGE 179

162 negatively charged ( 101) of the COM crystal. Although crystallographic orientation did not always effect the adhesive strength (e.g., fibronectin); experimentally determined adhesion results suggest that electrostatic interactions play a large role in COM adhesion at low ionic strength. Under high ionic strength conditions the adhesion of COM particles to the macromolecular substrates was higher than adhesion of COM to the same macromolecules under low ionic strength conditions, with the exception of the (010) crystal face of COM in contact with fibronectin, as shown in Figure 6.13. The increase may be due to the collapse of the ionic double layer surrounding both the particle and substrate. The double layer collapse leads to reduced electrostatic repulsion therefore allowing for shorter separation distances between the particle and the substrate where van der Waals attractive forces will be stronger. Because the ionic strength of the AUIS is so high the solution will not be able to support charge and the effect of electrostatic interactions will be negligible Calculations performed using the HHF theory for heterocoagulation as discussed in Chapter 5 for the materials COM and fused quartz in AUIS determined that a separation distance of less than 10 nm was necessary to produce adhesive pressures on the order of 5000 Pa. The two materials cannot come into such close proximity because the minimum surface roughness of COM is approximately 20-30 nm. When the fused quartz is coated with a macromolecule the van der Waals forces decrease due to the lower Hamaker constant of the macromolecules therefore limiting the total van der Waals pressure to below 5000 Pa. However, a macromolecule has the ability to rearrange itself to optimize its orientation and lower its free energy. Reorientation of the macromolecule may bring it into much closer proximity to the COM crystal allowing very short range interactions to transpire. Because both MA TRIG EL an d PEI had adhesive strengths above 5000 Pa (i.e., off the scale), and electrostatic and van der Waals forces cannot account for the high adhesion some other specific adhesion mechanism must be occurring such as covalent bonding

PAGE 180

163 6000 D ( 010) > 5 <5 ..----.. 5000 ro (101) 0.. D ......,,, .c 4000 .J...J 0) C: t (/) 3000 Q> ~ > 5 (/) cu cu cu cu Q> 2000 a.. a.. a.. a.. .c 0 0 0 0 -0 0 0 0 0 -<( l!) T"" l!) T"" C') l!) C') l!) 1000 >-5 > 5 A A A A . .. 0 . C: __J Q> C: :.c:; w C: ro Q> V ('.J .; ::::s 0) Q> 0 ro C: a: Q> -0 0 e IC: Q> (_) ,Q -4: Q> (/) LL 2 ::>,. ::::s .c LL ..L.J Q> ::>,. 0 0.. Figure 6.13. A bar chart summarizing the a d hesion data experimentally determined in COM saturated solution ( i.e., low ionic strength). The numbers above each set of bars is the value of the zeta potential for each substrate

PAGE 181

between the N atoms in the PEI and MATRIGEL and the Ca atoms in the COM crystal. 164 As shown in Table 6.2 generally at low ionic strength the adhesive strength between COM and substrates qualitatively depends on the difference in polarity and the magnitude of the difference in zeta potential values of the two adhering materials. That is, the adhesive strength of negatively charged COM particles to a particular macromolecule is larger if the zeta potential of that macromolecule is positively charged and lower if the macromolecule is negatively charged. However a negatively charged protein substrate has not produced an adhesive strength equal to zero indicating that forces other than purely electrostatic interactions are present. For proteins this may be due to conformational changes of the protein on the glass surface under slight variations in solution conditions Many of the proteins have both positive and negative sites due to the amino and carboxyl groups along the chain length. If a positive area of the protein is extended into the solution, electrostatic attraction will occur to a negative rich area on the COM crystal and vice-versa, increasing the adhesive force Another explanation may be the fact that the proteins are not a hard surface but a pliable surface capable to some extent of accommodating the crystal possibly allowing for optimum electrostatic interaction [Big96]. At high ionic strength another mechanism of adhesion is occurring The values of adhesion strength are given in Table 6 .3. The adhesion values of 3500 and 5100 Pa listed are the maximum adhesive strengths measurable for the (010) and ( 10 I) faces respectively. At lower values of adhesion van der Waals forces are dominant ; however at the higher adhesion values specific bonding interactions are occurring. Trommler et al. [Tro85] measured the adhesion between glass and red blood cells and concluded that molecular contacts must be made between the two negatively charged surfaces for adhesion to occur

PAGE 182

Table 6.2 Values of adhesion strength for COM particles adhering to macromolecular substrates as a function of crystallographic habit plane in COM saturated solution. Substrate Zeta Adhesion Strength Material Potential (Pa) (mV) (010) (101) Fused Quartz -22 81 170 Collagen type I -9. 6 208 475 Fibronectin -12.8 238 203 MATRIGEL -10.9 222 225 Polyethy leneimine 7.8 1305 1230 165

PAGE 183

166 6.3 Values of adhesion strength for COM particles adhering to macromolecular substrates as a function of crystallographic habit plane in AUIS Substrate Zeta Adhesion Strength Material Potential (Pa) (mV) (010) (101 ) Fused Quartz > -5 304 457 Collagen type I > -5 1497 2047 Fibronectin > -5 220 441 MATRIGEL > -5 > 3500 > 5100 Polyethy leneimine < 5 > 3500 > 5100

PAGE 184

167 In both solution conditions some amount of mechanical bonding will also be occurring due to the pliable nature of the macromolecular surfaces. Further experimental study in the area of COM crystal adhesion to protein substrates is needed to fully understand the mechanism(s) of attachment. SEM photomicrographs showed that in a few instances the COM particles were coated with a macromolecular layer after adhesion experiments Figure 6.14(A) In other samples the macromolecular coating was not observed yet adhesion took place as shown in Figure 6 l 4(B). The differences in coating may be due to the solubility of the protein in COM saturated solution at pH=5.7 and / or the position of the particle in the adhesion cell. If the particle is lying toward the center or beyond the first few millimeters of the entrance of the dynamic wet cell, the flow is established, the flow rate is higher and the protein in solution may be washed off by the high flow rate such that the particle is not coated. In an area of the cell where the flow rate is lower near the eluent inlet and very close to the side walls the protein in solution is more easily coated onto the surface of the COM crystals. The etching observed on the COM particles shown in Figure 6.14(B) is probably due to some modest dissolution of the particles synthesized at 25 C then equilibrated at 37 C in COM saturated solution a l so prepared at 25 C The increase in temperature slightly increases the solubility of COM in aqueous solution and dissolution may occur. This argument is also consistent with the greater etching observed on the Ca2+ -rich (101) than the more C2o/-rich (010) face Such binding of Ca2+ by insoluble urinary macromolecules has been previously discussed by Finlayson and Dubois [Fin73). 6 3 .4 Hydrodynamic Model of the Human Kidney Kok and Khan [Kok94] calculated the volumetric flow rates in the kidney to be 0 2 m l / min to 39 1 ml/min as shown in Table 6.4. These flow rates correspond to forces of 8 .6xl08 N to 2.3x104 N for a particle of 1 m diameter calculated using the volumetric flow rates and tubule dimensions from Kok and Khan [Kok94] in the Poiseuille equation

PAGE 185

168 (A) (B) Figure 6 .14. COM particles (A) coated during an adhesion experiment using fibronectin and (B) uncoated COM particles used in an adhesion experiment. Both photomicrographs were taken after flow. (bar = 10 m)

PAGE 186

169 Table 6.4. Human kidney tubule dimensions and volumetric flow rates reported by Kok and Khan [Kok94]. Proximal Tubule Descending Loop of Henle Ascending Loop of Henle Distal Tubule Outer Medullary Collecting Duct Inner Medullary Collecting Duct Duct of Bellini Tubule Diameter (m) 25-35 14-37 19-29 18-30 30-35 35-60 60-100 Volumetric Flow Rate (ml/min) 39.1 13. 1 5.8 5 8 3.7 0.9 0.2

PAGE 187

170 (Appendix B). In the current research the force is acting on the particles at a radius of 3.5 m for a particle on the (010) and at a radius of 4.2 m for a particle on the ( 101) For a particle ofradius 5 m, the forces acting on the particle range from 8.6x10-8 N to 2.3xl 04 N. The stress acting on a model COM particles having a 5 m and 8 m radius resulting from the hydrodynamic flow in each section of the kidney is shown in Figure 6.15. The current measured values of COM crystal adhesion to the macromolecular substrates are represented on Figure 6.15 by the shaded area. The experimentally measured values of adhesion correspond to the duct of Bellini and the inner medullary collecting duct of the human kidney where the shear stresses are lower due to larger tubule sizes and lower flow rates. Many kidney stones are in fact found in these two regions of the human kidney near the papillary tip (Kha91 Kha95b]. 6.4 Conclusions Adhesion of COM crystals to different extracellular matrix proteins substrates were measured experimentally and found to adhere with varying strengths. The adhesion strength of the macromolecules increased as the ionic strength of the solution conditions increased. In the low ionic strength solution conditions a COM crystallographic dependence was observed. Trends existing in the adhesion results demonstrate that electrostatic interactions play a significant role in COM crystal adhesion to surfaces at low ionic strength. At high ionic strength another mechanism of adhesion is occurring. At lower values of adhesion van der Waals forces are dominant; however at the higher adhesion values, specific bonding interactions are occurring Of the biologically significant materials, collagen type I exhibited the largest adhesive strength to COM at low ionic strength. However at high ionic strength the material MA TRJGEL a mixture of extracellular materials exhibited the highest adhesive strength to COM.

PAGE 188

171 1E+07 ..---.. 1E+06 ro a. ..._... (/) (/) i 1E+05 (/) I.... ro ..c 1E+04 (/) lJ ~ a.. 1E+03 a.. -<( 1E+02 .,_, .,_, <.> <.> C ::l C C ::l ::l ::l ..0 ..0 0 0 ::l I I ::l O'> O'> (D I"-"IC C "-ro 0 0 JS D D 0 ~ a.. a.. <.> <.> .,_, 0 0 ~ <.> ::l X 0 0 0 0 0 0 e ....J ....J 0 0 a. O'> O'> C C c c "O -a ro ro s:::: C ::l ::l <.> <.> "O "O (/) (/) -<( 2 2 0 I.... I.... 2 ::l C 0 C Figure 6 .15. A plot of the stress on a 5 m ( 0 ) to 8 m ( 0 ) radius mode l COM particle under the hydrodynamic conditions found in t h e different areas in the human kidney. The shaded region represents the range of experimentally measured values of adhesion of COM to biological materials.

PAGE 189

172 A hydrodynamic model was used to calculate the force applied on particles various radius in the different regions of the human kidney The values of stress which correspond to the calculated forces on the particles correspond to the experimentally measured adhesive strength values between COM and the macromolecules reported herein.

PAGE 190

CHAPTER 7 SUMMARY AND FUTURE WORK 7.1 Summary COM crystals having a distinct shape and narrow particle size where synthesized. The equilibrium shapes with face indices of experimentally produced COM particles synthesized in different conditions have been successfully generated using the computer program SHAPE Computer calculations of theoretical crystallographic shapes have been reconciled to the observed shapes of experimentally synthesized COM particles under various conditions. Also the seeding studies suggest that it is possible to control the size of COM crystals ("33" particles) by controlling the amount of seed materials. The computer program SHAPE was demonstrated to be a useful tool to determine the face indices of COM crystals The equilibriun1 shape of COM crystals based on the lattice parameters of Deganello and Piro [Deg81 a] is reconciled to the morphology of the experimentally synthesized COM crystals from the comparison between the calculated shape and the morphological form of the experimentally derived particles. Based on the face indices of the equilibrium shapes and atomic coordinates the atomic structures of COM particles as a function of habit plane have also been generated by using the computer program A TOMS From the comparison between the theoretical atomic structures generated by ATOMS the (010) and (100) planes generated with the atomic coordinates published by Cocco and Sabelli [Coc62] are reconciled to the layering sequence reported by Deganello and Piro [Deg81 a] and Tozzoli and Domeneghetti [Taz80]. 173

PAGE 191

174 The Hamaker constant of COM was determined using the Tabor-Winterton relationship. The dielectric constant of COM was deconvoluted from a COM/silicone composite using various mixing rules. The series mixing rule yielded the highest nonlinear regression coefficient however the composites were clearly not representative of the series model. Therefore the Lichtenecker model, previously shown to describe irregularly shaped inclusions in a matrix [Mon47] was used and a value of 28 9 for the COM dielectric constant was determined. Plots of refractive index as a function of optical direction and wavelength were used to generate the ultraviolet absorption frequency and the dielectric and refractive index at the UV absorption frequency for use in the Hamaker constant calculation. The resulting value of the Hamaker constant 13.7 (.90)xl0-21 J compares well with previously published values of 11.4 (.42)x10-21 J determined using the Gregory approximation. A device was designed and built and the methodology developed to measure the adhesion strength between a particle and substrate. A second device was designed and built and the methodology developed to measure the zeta potential on a flat substrate The parallel plate adhesion technique is a suitable method for determining the adhesive strength between a particle and a substrate in a dynamic liquid environment. The van der Waals interactions were estimated and related to the adhesion between COM crystals and a fused quartz substrate. Correlations were made between the theoretical atomic surface structure of COM and the crystallographic dependence on adhesive strength. Theoretical electrostatic considerations were invoked and are used to explain the crystallographic dependence. Adhesion of COM crystals to different extracellular matrix protein substrates were measured experimentally and found to adhere with varying strengths. The adhesion strength of the macromolecules increased as the ionic strength of the solution conditions increased indicating a decrease in electrostatic repulsion due to compression of the electrical double layer at elevated ionic strengths. In the low ionic strength solution

PAGE 192

175 conditions a COM crystallographic dependence was observed. A trend in the adhesion results demonstrates that electrostatic interaction plays a significant role in COM crystal adhesion to surfaces At high ionic strength another mechanism of adhesion is occurring. At lower values of adhesion van der Waals forces are dominant; however at the higher adhesion values specific bonding interactions are occurring which may be covalent in nature. Of the biologically significant materials collagen type I exhibited the largest adhesive stress to COM at low ionic strength. However at high ionic strength the material MATRIGEL a mixture of extracellular materials exhibited the highest adhesive strength to COM. A hydrodynamic model was used to demonstrate that COM crystals may adhere to areas in the kidney where the flow rate is low and the corresponding shear stress is low. The application of the adhesion measuring technique reported in this work is promising for the study of kidney stone disease in that the importance of specific biological materials can be ranked with respect to their adhesive properties to COM crystals. This method of measuring adhesion may be an accurate predictor of COM kidney stone formation by a fixed particle mechanism. Also the adhesion measuring technique may be used as a tool to evaluate biological materials implicated in kidney stone formation. Which in turn may allow researchers to focus on materials that have the greatest adhesion to COM. 7 2 Future Work A surface energy study of COM with respect to the two dominant crystallographic planes should be performed. Surface and interfacial energies may be measured and the work of adhesion determined. From this information the free energy change upon adhesion can be determined to investigate if adhesion of COM to a particular substrate is thermodynamically stable

PAGE 193

176 A re-evaluation of the video taped adhesion experiments to analyze the effect of orientation of the anisotropic COM particles with respect to the fluid flow direction should be performed. The error associated with the values of adhesive strength should decrease when only one particle orientation is analyzed at a time. The effect of COM surface roughness on adhesion would also be an interesting study. The effect of mechanical bonding between a COM crystal and macromolecule which has reoriented itself near a COM particle may be observed. The orientation of the macromolecular substrates should be determined to determine which molecular sites are available to interact with the COM crystal. Infrared spectroscopy can aid in this analysis. Also Surface Potential Microscopy and Non contact AFM may be performed on the substrates to map the surface potential and van der Waals forces respectively. Finally in an effort to minimize adhesion between COM and macromolecular substrates the effect of solute additions on adhesion ( e g., adhesion inhibitory molecules) should be measured The aspect of initial testing of therapeutic drugs for kidney stone disease is promising.

PAGE 194

APPEN D IX A DETERMINATION OF THE AREA OF COM CRYSTALLOGRAPHIC FACES USING EQUIVALENT SPHERICAL DIAMETER Average Equivalent Spherical Diameter (m) e sd= 13.01 n 3 V = -(esd) 6 [ ] a= (0.43~0 35) For COM "33" particles having the ratio of lengths of the crystal sides A:B : C = 1 0:0.43:0 .35, A:= 1a B : = Q .43 a c:=Q.3 5 a The dimensions of the crystal (m) A = 19. 714 B = 8.477 C=6.9 177 .... ~----A .... D :=0.19a E : = 0.33 a

PAGE 195

Determining the area of the two major crystallographic faces (m2 ) Area(0IO) = (A-2 D) C + 2-( C} D Area(l01) = AB-2-( B }E Area(0I0)= 110.175 Area( 101) = 111.963 178

PAGE 196

APPENDIX B DETERMINATION OF THE STRESS IN THE KIDNEY AS A FUNCTION OF FLOW AND DISTANCE FROM THE TUBULE WALL D ; = 25 14 19 18 30 35 60 0;= 39.l 13 1 5.8 5.8 3 7 0 9 0.2 Input: i=l, 2 .. 7 values taken from Kok and Khan [Kok90] Diameter of Proximal Tubule (25-30 m) Diameter of Descending Loop of Hen l e (14 37 m) Diameter of Ascending Loop of Henl e (19 29 m) Diameter of Distal Tubule ( 18-3 0 m) Diameter of Outer Medullar y Collecting Duct (30 -35 m) Diameter oflnner Medullary Collecting Duct (35 60 m) Diameter of Duct of Bellini ( 60 100 m) values taken from Kok and Khan [Kok90 ] Volumetric flow through the Proximal Tubu l e (ml/min) Volumetric flow through the Descending Loop of Henle (ml/min ) Volumetric flow through the Ascending Loop of Henle (ml/min Volumetric flow throu g h the Distal Tubule (ml/min) Volumetric flow throu g h the Outer Medullary Collecting Duct ( ml / min ) Volumetric flow through the Inner Medullary Collecting Duct ( ml / min ) Volumetric flow through the Duct of Bellini (ml/min) 179

PAGE 197

rs tone = 3x 10-6 Radius of a kidney stone undergoing the calculated force ( m ) T/ = 0.6915 Viscosity of water at 37 C (cPoise) a = l l lxl0-1 2 Area of crystal face (m2 ) Conversions and Calculations Determination of radius from diameter D -10-6 R .=-'--' 2 Viscosity conversioncPoise to Poise 71m = 0.1 77 Distance from the center of tubule at which the velocity is calculated r; = R ; rston e Flow rate conversionml / min to ml/s Q i Qsec -I 60 Flow rate conversionml/s to m3/ s H. = Q s ec ; I 1003 Calculation of the velosity at one stone radius from the tubule wall (mis ) V -2 H ; (R2 2 ) ; R 4 ; r; ; n Calculation of the force acting at one stone radius from the tubule wall (N ) F = 1.7 6 n 11m rstone V I '/ I Calculation of the stress acting at one stone radius from the tubule w all wi th an interaction area of a (Pa) F Stress. = _2.. I a 180

PAGE 198

181 Plot of kidney tubule section vs. stress I I 0 6 Stre ss. I 5 I 0 5 2 3 4 5 6 7

PAGE 199

Ada81 Ada88 Ada91 Ada95 Add85 Alb89 All90 Arc90 Bar78 Ber90 Big96 Boe94 Boy68 Bri28 REFERENCES Adair J.H. Ph. D. Dissertation, University of Florida Gainesville Adair, J.H. Aylmore L.A.G. Brockis J.G. and Bowyer R.C., J. Colloid Interface Sci 124 1. Adair J.H. Onoda G.Y., and Finlayson B., Am. J. Kidney Dis. lliil, 3396. Adair J.H. Linhart R.V. Habeger C.F. and Khan S.R. in: Calcium Oxalate in Biological Systems S.R Khan ed., CRC Press Boca Raton 37. Addadi L and Weiner, S., Proc. Natl. Acad. Sci USA 82 4110. Alberts B. Bray D., Lewis, J. Raff, M. Roberts K ., and Watson J.D. Molecular Biol;ogy of the Cell, 2nd ed Garland New York 787 Allen T., Particle Size Measurement,Chapman and Hall New York 153. Archer D.G. and Wang P., J. Phys. Chem. Ref. Data 19 371. Barouch E., Matijevic E., Ring T.A. and Finlan J.M. J. Colloid Interface Sci .lill, 1. Bergethon P .R. and Simons E.R. Biophysical Chemistry: Molecules to Membranes Springer-Verlag New York. Bigalow M.W Wiessner J.H. Kleinman J.G ., and Mandel N.S. J. Urol. 155 1094. Boeve E. R., Cao L. C., DeBruijn, W. C., Robertson W G., Romijn J. C., and Schroder, F. H., J. Urol. 152, 531. Boyce W.H. J. Med. 45, 673. Briggs D.K. J. Phys. Chem. 32 641. 182

PAGE 200

Bro92 Bul35 Bus84 Bus92 Bus93 Cam89 Cha84 Cha91 Cla95 Coc61 Coc62 Cod94 Coe91 Con90 Cor66 Cul78 Cur79a 183 Brown C.M and Purich, D L. in : Disorders of Bone and Mineral Metabolism F .L. Coe and M.J. Favus eds. Ra ven Press New York 1. Bull H B., J. Amer. Chem. Soc. 57 259. Busscher H.J., Weerkamp A.H. van der Mei, H.C., van Pelt A.W.J. de Jong H.P., and Arends J., J. Appl. Enviro. Microbio 48[5], 980. Busscher H.J., Doombusch 0.1. and Van der Mei H.C. J. Dent. Res IillJ. 4 91. Busnaina A., Taylor J., and Kashkoush I. J Adhesion Sci. Technol. ml, 441. Campbell A.A. Ebrahimpour, A. Perez L., Smesko S.A. and Nancollas G .H., CalcifTissue Int 45, 122 Chan D Y.C. and Halle B., J. Colloid Interface Sci. 102[2] 400 Charles Y.C. Amer. J. Kidney Diseases 17 4], 420. Clark J.Y. J. Urol. 154 2020. Cocco G., Atti. Accad. Naz. Lincei Rend Cl. Sci. Fis Mat. Nat. l,1, 292. Cocco G., and Sabelli C., Atti Soc. Tascana Sci. Nat. Pisa, Proc Verbali Mem. Ser. A 69[2] 289. Cody A.M. and Cody, R.D. J. Crystal Growth 135 235 Coe F.L. Nakagawa Y., and Parks, J.H. Am. J. Kidney Diseases .!llil 407. Conyers R.A.J. Bals, R., and Rofe A.M., Clin. Chem. 36[10], 1717. Com, M., in Aerosol Science ed C.N Davies Academic Press New York. Cullity B.D. Elements of X-Ray Diffraction, 2nd ed. Addison-Wesley, Reading MA. Curreri P.A. Ph.D. Dissertation University of Florida Gainesville FL.

PAGE 201

Cur79b Cur87 Deg80 De g81a Deg81b Deg91 Der41 Der54 Der61 Dit82 Dor77 Dow92 Dow93 Duc91 Ebd77 Ebi95 Edy86 Edy87a Curreri P., Onoda Y., Jr., and Finlayson B., J. Colloid Interface Sci. 22.ill, 170. 184 Curreri P.A., Onoda, G.Y., Jr., an d Finlayson B., in Proteins at Interfaces, Physicochemical and Biochemical Studies J.L. Brash and T .A. Horbett Eds., American Chemical Society Washington, D.C. 278 Deganello S., Zeitschrift fur Kristallographie 152,247. Deganello, S. and Piro 0., Neues Jahrbuch fur Mineralogie Monatshefte 2, 81. Deganello S., Acta Cryst. B37 826. Deganello S., Calcif. Tissue Int. 48, 421. Derjaguin B.V. and Landau L., Acta Physicochim URSS ,!i, 633 Derjaguin B.V., Disc Faraday Soc 18, 85. Derjaguin B.V. and Zimon, A.D. Kolloidn. Zh. 23[5] 544 Ditter W., Eisenlauer J., and Horn, D., in The Effect of Polymers on Dispersion Properties Academic Press, New York, 323. Doroszewski J., Skierski J., and Przadka L., Ex p Cell Res 104 335. Dowty E., and Richards, R .P., SHAPE A Computer Program for Drawing Crystals, v 3 .0. Dowty E. A TOMS A Computer Program for Displaying Atomic Structures, v.2.0. Ducker W.A. Senden T .J., and Pashley R.M. Nature 353 239. Ebdon D., Statistics in Geography : A Practical Approach Billing & Sons Great Britain Ebisuno S., Kohjimoto Y., Tamura M., Ohkawa T., Eur. Urol. 28 68. Edyvane K.A. Ryall R.L. and Marshall, V.R. Clin. Chim. Acta 156 81. Edyvane, K.A. Ryall R .L., and Marshall V .R. Urol. Res 12, 87.

PAGE 202

Ed y 87b Edy87c Eis85 Esk68 Fai25 Fin78a Fin78b Fin84 Fle78 For84 Gil79 Gol67 Gor59 Gre81 Gre85 Gro90 H a b97 Har86a 185 Ed y vane K .A., R y all R.L. Mazzaachi R.D., and Marshall V.R. U rol. Res .15, 87 Edyvane K.A. Hibberd C.M. Hamett, R.M. Marshall V.R,. and R y all R.L. Clin. Chim. Acta 167 329. Eisenlauer J. and Hom, D., Colloids Surf. 14, 121. Eskinazi S., Principles of Fluid Mechanics, 2nd ed ., All y n and Bacon Boston. Fairbrother F. and Mastin H. J. Chem Soc 127 323. Finla y son B., Kidney Int. 13, 344 Finlayson B. and Reid F., Invest. Urol.121il, 442. Finlayson B., Khan S.R. and Hackett R.L. Scan Elec Micro s c III 1419. Fleisch H., Kidney Int. .Ll,, 361 Forrester J.V and Lackie J.M., J. Cell Sci. 70 93 Gill W B. Ruggiero K., and Strauss F.H., II, In v est. Urol. 17, 257 Goldm a n A.J. Co x R.J. and Bre nner H. Chem E ng Sci 22, 653. Gordon L.,. Salut s ky M L and Willard H.H., Precipit a tion from Homogeneous Solution John Wile y and Son s, New York Gregor y, J Chem. Engr. Sci 36[11] 1789. Gregor y, J., J. Colloid Interface Sci. 105[ 2 ] 357 Gro v er P.K. R y all R.L. and Marshall V.R. Clin Chim Acta 190,223 H a b ege r C.F., Condon C., Khan S .R., Ad a ir J.H., in print Colloid Surf.B. --Hartel R.W., Gottung B.E. Randolph A.D., and Drach G .W., AICH E J. 32[7] 1176

PAGE 203

Har86b Hoc73 Hog66 Hor77 Hor90 Hou80 Hun62 Hun81 Hun93 Ile79 Isr78 Isr92 Jon45 Ka187 Kha82 Kha83a Kha83b 186 Hartel R W. and Randolph, A.D AIChE J. 32[7], 1186 Hochmuth R M., Mohandas N., and Blackshear Jr. P.L. Biophys. J. .!l 747. Hogg R. Healy T .W., and Fuerstenau D.W. Trans. Faraday Soc. 62 1638. Horn J.M. Jr. and Onoda G.Y. J. Colloid and Interface Sci 61, 272. Horn R.G. J. Am. Ceram. Soc 73[5 ] 1117 Hough D.B and White L.R., Adv. Colloid Interface Sci 14 3 Hunter R.J and Alexander A.E. J. Colloid Sci.11, 781. Hunter R.J. Zeta Potential in Colloid Science: Principles and Applications Academic Press, New York Hunter R.J. Introduction to Modern Colloid Science Oxford Science, New York Iler R.K. The Chemistry of Silica: Solubility, Polymerization, Colloid and Surface Properties and Biochemistry John Wiley and Sons New York. Israelachvilli J.N and Adams G.J J. Chem. Soc., Faraday Trans. 1 74 975. Israelachvili J.N. Intermolecular and Surface Forces, 2nd ed. Academic Press New York Jones G and Wood L.A. J. Chem. Phys. 13, 106. Kallay N., Barouch E., and Matijevic, E Advan Colloid Interface Sci 27 1. Khan S R Finlayson, B., and Hackett R.L Am. J. Pathol. 107 59. Khan S.R. Finlayson B. and Hackett R.L. J. Urol. 130 992 Khan S.R. Finlayson B. and Hackett R.L. Scan. Elec Microsc .P.L.l, 379

PAGE 204

Kha84 Kha87 Kha88 Kha91 Kha95a Kha95b Kle82 Knu85 Koh96 Kok90 Kok94 Kor60 Kri94 Kuc92 Kru67 Kiih87a Kilh87b Kuo80 Khan S R. Cockrell C.A. J. Urol. 132 153 Khan, S R. and Hackett R.L. Scan Microsc. 1[3] 1405 Khan S R. Shevock P.N., and Hackett, R.L. J. Urol. 139(2] 418. Khan S .R. and Hackett, R.L Scanning Microsc 2QJ 707 Khan S.R. U rol. Res 23 71. Khan S .R., Scanning Microsc .2I.!l, 89. Kleinman H.K. McGarvey M.L. Liotta, L.A. Try ggv ason K., and Martin G R., Biochemistry 21[24] 6188 Knutson K and L yman, D.J., Surface and Interfacial Asp e ct s o f Biomedical Polymers Vol. 1: Surface Chernitry and Physics J .D. Andrade ed., Plenum New York. 187 Kohjimoto Y., E bisuno S. Tamura, M and Ohkawa, T. Urol. Res 24 193. Kok, D J ., Papoulos S E., and Bijvoet O .L.M, Kidney Int. 37 51. Kok, D.J and Khan S.R. Kidney Int. 46 847. Kordecki M C. and Orr C., AMA Arch. Environ Health .!l!l 1. Krishnan S Busnaina A.A. Rimai D.S., and Demejo L.P., J. Adhesion Sci. Technol. fil.!.!l, 1357. Kucharz E.J. The Collagens : Biochemistry and Pathophysiology Sprin g er-Verlag New York. Krupp H., Ad v an Colloid Interface Sci .!., 111. Kuhn K. in Structure and Function of Collagen Types R May ne and R .E. Burgeson eds., Academic Press Orlando 1 Kuhn, K. in Structure and Function of Collagen Types, R May ne a nd R.E Burgeson eds. Academic Press Orlando 43 Kuo R.J. and Matijevic E. J. Colloid Interface Sci 78(2] 407

PAGE 205

Lan88 Lea77 Lid90 Lie92 Lie93 Lie94 Lie95 Lie96a Lie96b Man81 Man87 Man91 Man94 Mar30 Mar86 Mat81 Mey71 Mey82 188 Lanzalaco A.C. Singh R.P., Smesko, S A., Nancollas G.H. Sufrin G., Binette M., and Binette J.P. J. Ural. 139 190. Leal J. and Finlayson B., Invest. Ural. 14[4], 278. Lide, D.R. CRC Handbook of Chemistry and Physics, 71st ed. CRC Press Boca Raton. Lieske J.C. Walsh-Reitz M.M. and Toback, F.G., Am. J. Physiol. 262 F622. Lieske J.C. and Toback, F.G, Am J. Physiol. 264 F800. Lieske J.C. Swift H., Martin T., Patterson B., and Toback F.G Proc. Nat. Acad Sci. USA 21., 6987. Lieske J.C., Leonard R., and Toback, F.G., Am. J. Physiol. 268 F612. Lieske J.C. Leonard R., Swift, H., and Toback, F.G. Am. J. Physiol. 270 F192 Leiske J.C. Toback F.G. and Deganello S., Calcif. Tissue Int. 58 195. Mandel N.S. and Mandel, G.S. in Urolithiasis: Clinical and Basic Research Plenum Press New York. Mandel N.S., Mandel G.S. and Hasegawa A.T. J Ural. 138, 557. Mandel, N. and Riese, R., Am. J. Kidney Dis. 17[4], 402 Mandel N. J. Am. Soc. Nephrol. ~ 537. Martin W.McK. and Gartner, R.A., J. Phys. Chem 34 1509 Marcott C., ASM Handbook, Vol. 10: Materials Characterization ASM International, USA. Matijevic E., Pure Appl. Chem. 53, 2167 Meyer A.S. Finlayson B., and DuBois L., Br. J. Ural. 43, 154. Meyer J.L. and Thomas, W.C., Jr., J. Ural. 128 1372.

PAGE 206

Moh74 Mon47 Nak83 Nak87 Nat95 NIH97 Nyv85 Ols78 Ove52 Owe87 Pal51 Pel82 Pel84 Pri47 Pri86 Ran88 Ray87 189 Mohandas N Hochmuth R.M. and Spaeth E.E J. Biomed. Mater. Res ~ 119. Montgomery C. G., Techniques in Microwave Measurement McGraw Hill New York. Nakagawa Y., Abran1 V., Kezdy F.J., Kaiser E.T. and Coe F.L. J. Biol. Chem 258 12594. Nakagawa Y., Ahmed M., Hall, S.L., Deganello S and Coe F.L. J. Clin. Invest. 79 1782. National Hospital Discharge Survey: Annual Summary, 1993 CDC HHS. www.niddk nih gov/KU Stats / kustats htm. N y vlt J., Sohnel 0., Matuchova M ., and Broul M. The Kinetics of Industrial Crystallization Elsevier New York. Olsson J., Ph.D. Dissertation, University of Goteborg Sweden. Overbeek J.Th.G., Colloid Science, Vol. 1, H.R. Kruyt ed. Elsevier New York 115. Owens N.F. Gingell D., and Rutter P .R., J. Cell Sci 87 667 Palache C., Berman H., Frondel C., The System of Minerology, Vol. 2. Wiley New York. Pelton R.H. Nugent H. and Allen, L. H. Colloids and Surf ,1, 397 Pelton R.H. and Allen L.H., J. Colloid Interface Sci 99[2], 387 Prien E.L. and Frondel, C J. Urol. 57 946. Prien E.L. and Prien, E .L. Am. J. Med. 45 654. Randolph A.D., and Larson, M.A., Theory of Particulate Processess : Analysis and Techniques of Continous Crystallization Academic Press San Diego Ray D .T. and Hogg R., J. Colloid Interface Sci. 116[1] 3256,.

PAGE 207

Rie88 Rie92 Rob69 Rob85 Rob86 Rut47 Rya8la Rya81b Rya81c Rya84 Rya86 Ryd95 Sca92 Sch90 Scu86a Scu86b Son72 Ste90 Riese R.J., Riese R W., Kleinman J.G., Wiessner J H., Mandel G .S., and Mandel N.S. Am. J Physiol. 225 F1025 190 Riese R.J. Mandel N.S., Weissner, J.H., Mandel G.S., Becker C.G. and Kleinman J.G Am. J. Physiol. 262 Fl 77. Robertson W G Clin. Chim. Acta. 26 105. Robertson, W.G. and Peacock, M., in Pathogenesis of Urolithiasis in Urolithiasis Etiology Diagnosis Springer-Verlag New York 185. Robertson W .G., Urol. Int. 41, 329. Rutgers A J and DeSmet M., Trans. Faraday Soc 43, 102. Ryall, R L., Harnett, R.M. and Marshall V.R Clin. Chim. Acta 112 349. Ryall R., Ryall R.G., and Marshall V .R., Invest. Urol. .!.fil.21, 396 Ryall R.L. Bagley C.J., and Marshall, V R., Invest. Urol. .!.fil.21, 401, 1981. R y all R.L. and Marshall V .R., Clin Chim Acta 141, 197 Ryall R., Harnett R.M. and Marshall V.R. J Urol. 135, 174 Ryde, N., and Matijevic E., J. Colloid Interface Sci. 169 468 Scales P.J., Grieser F., and Healy T W., Langmuir 8 965 Scheaffer R.L. and McClave J.T. Probability and Statistics for Engineers PWS-KENT Boston. Scurr D S., and Robertson W G., J. Urol. 136 128 Scurr D.S,. and Robertson W.G. J. Urol. 136 505. Sonntag H. and Strenge K., Coagulation and Stability of Disperse Systems (Engl. trans.) Halsted Press New York. Stevens M.P. Polymer Chemistry: An Introduction, 2nd ed. Oxford New York

PAGE 208

Tab69 Tab80 Taz80 Tho85 Tru90 van80 van92 Ver48 Ver96 Vis76 Von54 Wie87 Woo46 Yam96 Yar87 Zim82 Tabor D ., and Winterton R.H.S., Proc Roy. Soc A 312 435. Tabor F.R.S and Winterton R.H S., Proc Roy. Soc. A 312 3 27. Tazzoli V., and Domeneghetti C., Arner. Min. 65, 327. Thormahlen I., Straub J., and Grigul U. J. Phys. Chem. Ref. Data 14 933 Truske y, G.A. and Pirone J.S. J. Biomed. Mater. Res. 24 1333 191 v an Wagenen R.A. and Andrade J.D. J. Colloid Interface Sci 76[2] 305 van Kooten T.G. Schakenraad J.M., Van der Mei H.C., and Bus s cher H.J. J. Biomed Mater. Res. 26 725. Verwe y and Overbeek Theory of the Stability of Lyophobic Colloids Else v ier New York Verkoelen C.F., Romijn J.C. Cao L.C. Boeve E.R. DeBruin W.C. and Schroder F.H., J. Urol. 155 749. Visser J., in Surface and Colloid Science, Vol.8 E Matije v ic ed. Wile y Interscience New York Von Hipple A. Dielectrics and Waves Wiley New York Wiessner J.H., Kleinman J.G. Blumenthal S S., Garancis J.C. Mandel G.S. and Mandel N.S. J. Urol. 138 640. Wood L.A. J. Arn Chem Soc. 68 437. Yamate T. Kohri K., Umekawa T. Arnasaki N. lsakawa Y., lguchi M., Kurita T. E ur. Urol. 30 388 Yarbrough W.A. Gururaja T.R. and Cross L.E ., Arn Ceram Soc Bull. 66[4] 692. Ziman, A.O. Adhesion of Dust and Powder, 2nd ed ., Consultants Bureau New York.

PAGE 209

BIOGRAPHICAL SKETCH Craig Fredrick Habeger was born to James and Sandra Habeger on June 10, 1969 in Youngstown Ohio. He was the youngest of three children Growing up in a rural area around Youngstown, he honed his common sense skills He often helped or watched his grandfather father brother and uncles fix farm and other machinery which gave him a good mechanical understanding and an interest in engineering. During his childhood his parents instilled in him a strong sense of ethics morality, honesty, and integrity. After high school, Craig attended the University of Cincinnati to earn his Bachelor of Science degree where his focus of study was materials science and engineering. During his undergraduate years at the UC, he groomed his social skills enabling him to work well in a team setting Also during that time he began to weight train. This activity lead to greater self confidence and perseverance. He graduated from UC in June of 1992 Craig moved onto graduate school at the University of Florida in the Materials Science and Engineering Department. At UF he pursued a Doctor of Philosophy degree in Materials Science and Engineering specializing in colloidal processing of ceramic materials. During his time at Florida Craig has grown tremendously He is viewed as a manager and a leader whose valued opinions are often sought. On June 24 1995 Craig Habeger married Nicole Schindler. On October 1 1997 Craig successfully defended his Ph.D. dissertation, becoming the first Ph.D. in his family (hopefully not the last) After graduating Craig, Nicole and their first addition (estimated arrival March 20 1998) will ride off into the sunset and live happily e v er after. 192

PAGE 210

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate in scope and quality as a dissertation for the degree of Doctor of Philosophy J---r--_.U{_ es H. Adair Chair ssociate Professor of Materials Science and Engineering I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate in scope and quality as a dissertation for the degree of Doctor of Philosophy. Christopher D. Batich Professor of Materials Science and Engineering I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate in scope and quality as a dissertation for the degree of Doctor of Philosophy. Anthony B. Bre a Associate Profe sor of Materials Science and Engineering I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate in scope and quality as a dissertation for the degree of Doctor of Philosophy ~Moudgil Professor of Materials Science and Engineering

PAGE 211

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarl y presentation and is fully adequate in s cope and quality as a dissertation for the degree of Doctor of Philisophy Raymond L. Hackett Professor of Pathology and Laboratory Medicine This dissertation was submitted to the Graduate Faculty of the College of Engineering and to the Graduate School and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy December 1997 Winfred M. Phillips Dean College of Engineering Karen A Holbrook Dean Graduate School

PAGE 212

LD 1780 199 'y -1 HI\~ UNIVERSITY OF FLORIDA IIIIIIIIIII IIIII II IIIIIIIII II llllll 11111111111111111111111111111 3 1262 08554 2438


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID ER33V5MCU_493Q1C INGEST_TIME 2011-09-29T20:18:46Z PACKAGE AA00004707_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES